Mr Justice Chan PJ : 1. I agree with the judgment of Mr Justice Hartmann NPJ. Mr Justice Ribeiro PJ: 2. I agree with the Judgment of Mr Justice Hartmann NPJ. Mr Justice Tang PJ: 3. I have read the judgment of Mr Justice Hartmann NPJ in draft. I respectfully agree with it and have nothing to add. Mr Justice Hartmann NPJ: The Issue in this Appeal 4. Article 24(3) of the Basic Law provides that permanent residents of the Hong Kong Special Administrative Region shall have right of abode. It is a right of considerable significance, including the right to live in Hong Kong without any condition of stay and free of any threat of deportation or removal[1]. 5. As to who qualifies to be a permanent resident, article 24(2)(4) provides that permanent residents shall include persons not of Chinese nationality “who have entered Hong Kong with valid travel documents, have ordinarily resided in Hong Kong for a continuous period of not less than seven years and have taken Hong Kong as their place of permanent residence”. 6. Recognition of permanent residence status pursuant to article 24(2)(4) is subject to a number of statutory qualifications contained in the Immigration Ordinance[2] ('the Ordinance'). Two of the qualifications are relevant to this appeal: (1) Paragraph 1(4)(b) of Schedule 1 to the Ordinance provides that, for the purpose of calculating the seven years of continuous ordinary residence stated in article 24(2)(4), the period of seven years immediately preceding the date of an application to the Director of Immigration (‘the Director’) for verification of permanent residence status shall be the basis of calculation. (2) S 2(4)(b) of the Ordinance provides that, for the purpose of assessing whether an applicant has over the required period of seven years “ordinarily resided” in Hong Kong, he shall not be treated as having been ordinarily resident: “during any period… of imprisonment or detention pursuant to the sentence or order of any court.” 7. Whether these two statutory qualifications are consistent with article 24(2)(4) of the Basic Law and, if so, their true nature and extent has been considered by this Court on two previous occasions, first, in Fateh Muhammad v Commissioner of Registration & Another[3]; second, in Prem Singh v Director of Immigration[4]. The two judgments confirm that the statutory qualifications are consistent with article 24(2)(4) and therefore constitutional. 8. Concerning paragraph 1(4)(b) of Schedule 1 to the Ordinance, the judgments held that an application for verification of permanent residence is to be judged at the time when an application is made by reference to the period immediately preceding that application. It is not therefore to be judged in the light of events which take place after the date of application. 9. Concerning s 2(4)(b) of the Ordinance, that is, the provision excluding any period of imprisonment or detention pursuant to the sentence or order of any court from the calculation of a person’s term of ordinary residence, the judgments concluded that, to be consistent with article 24(2)(4), “detention” was to be read ejusdem generis with “imprisonment” so that only periods of imprisonment or detention imposed in consequence of a breach of the criminal law were to fall within the ambit of the section. It was held that, in the context of an application to be recognised as a permanent resident, it is consonant with the ordinary and natural meaning of the phrase “ordinary residence” to exclude periods of imprisonment and detention which are punitive but not other forms of detention which do not evidence punishment for criminal wrongdoing, for example, an order of detention made by a court by reason of a subject’s mental illness. 10. It is in this context that the single issue that is the subject of this appeal arises. The issue is whether, consistent with the existing jurisprudence of this Court, a period of time spent in custody pending trial, when the trial itself results in a conviction and sentence of imprisonment, is to be read, in terms of s. 2(4)(b) of the Ordinance, as a period of “ imprisonment or detention pursuant to the sentence or order of any court”. Background 11. The factual circumstances which give rise to the issue may be summarised as follows. 12. The respondent, a national of Pakistan, came to Hong Kong in 1997 when he was 16 years of age to visit his father and other family members. Shortly thereafter he was granted permission to remain in Hong Kong as a dependent of his father. 13. In September 2005, the respondent was arrested and charged with two counts of wounding with intent. The charges arose out of a street fight between two groups of young men, the respondent being a member of one of the groups. Initially, the respondent was granted bail. However, in November 2005, when his case was transferred to the District Court for trial, his bail was rescinded and he was remanded in custody pending trial. 14. During the time that he was held in custody pending trial, the respondent entered into written communications with the Director concerning his entitlement to be verified as a permanent resident pursuant to article 24(2)(4). For reasons which are no longer in issue, the Court of Appeal held that this exchange between the respondent and the Director was to be read as resulting in a claim made by the respondent to the Director in mid-March 2006 for verification of his status as a permanent resident. 15. It is accepted that, at the time when he made his application to the Director, if the fact that he was held in custody pending trial was not taken into account, the respondent had been ordinarily resident in Hong Kong on a continuing basis for at least the previous seven years. 16. Some two weeks after the respondent was found to have made his application to the Director, he was brought to trial, convicted of both counts of wounding with intent and sentenced to 3 years imprisonment. 17. While serving his sentence, the respondent was informed by the Director that consideration was being given to his deportation from Hong Kong upon the completion of his sentence. Clearly, the primary reason for that considered course was the fact that the respondent had been found guilty of serious criminal offences. As it was, the day before his release from prison the respondent was served with an order of deportation for life issued by the Secretary for Security. 18. It was the service of this order which precipitated the respondent’s judicial review proceedings. The proceedings below: 19. In his judgment at first instance[5], Andrew Cheung J (as he then was) contrasted (i) a period of custody pending trial when the person in custody is then released before trial or acquitted with (ii) a period of detention pending trial, followed by conviction and a sentence of imprisonment. He accepted that plainly the latter period of remand in custody is not a period of “imprisonment pursuant to the sentence” of a court in terms of s 2(4)(b). The question therefore was whether it falls within the meaning of “detention pursuant to the order of any court”. 20. It was his finding that the concept of a period of custody pending trial, when the trial itself results in a conviction and sentence of imprisonment, while not on all fours with the more limited concept of imprisonment pursuant to sentence, was nevertheless of the “same nature”. In this regard, he looked to Bokhary PJ’s conclusion in Fateh Muhammad v Commissioner of Registration & Another[6] that: “In a provision like s 2(4)(b) “detention” and “order” must, in my view, be read as being of the same nature as “imprisonment” and “sentence” respectively.” [emphasis added][7] 21. A principal basis for the finding lay in the operation of s 67A of the Criminal Procedure Ordinance[8]. That section provides that, if a person has been held in custody pending his trial, the length of any consequent sentence of imprisonment is to be treated as reduced by that pre-trial period of detention. Cheung J accepted that s 67A does not act to turn a period of remand into a period of imprisonment but acts only to reduce any sentence of imprisonment imposed after trial. Nevertheless, in his view, when a period in custody pending trial results in a conviction and imprisonment, s 67A operates to render the period in custody pending trial of “the same nature” as the sentence of imprisonment imposed after trial. 22. Cheung J further took into account the ordinary and natural meaning in common law of the phrase “ordinarily resident”, placing reliance on Lord Scarman’s definition in R v Barnet L.B.C., ex parte Shah[9]. “Unless, therefore, it can be shown that the statutory framework or the legal context in which the words are used requires a different meaning, I unhesitatingly subscribe to the view that “ordinarily resident” refers to a man’s abode in a particular place or country which he has adopted voluntarily and for settled purposes as part of the regular order of his life for the time being, whether of short or of long duration.” Adopting a purposive approach to the construction of s 2(4)(b), the judge considered that detention pending trial, when the trial results in a conviction and sentence of imprisonment, was not consonant with the common law concept of “ordinarily resident”. The element of voluntariness on the part of the individual was missing. Far from being “ordinary”, such imprisonment was “out of the ordinary”. 23. Cheung J further took into account that – “… as a matter of general principle, no man should be allowed to profit from his own wrongdoing. It would lie ill in the mouth of the prisoner to assert the status of a permanent resident based on seven continuous years’ residence by taking advantage of his time of imprisonment in Hong Kong. It would simply offend one’s sense of what is right and fair.”[10] 24. In the result, Cheung J held that at the time when the respondent made his application to the Director for verification of his permanent resident status he was not ordinarily resident in Hong Kong. 25. The Court of Appeal[11] (Stock VP, Fok JA and Lam J) did not accept this analysis. 26. Giving the leading judgment of the court, Stock VP saw no warrant for widening the ambit of the term “detention pursuant to an order of any court” beyond the parameters first set by this Court in Fateh Muhammad and endorsed in its later judgment in Prem Singh, both holding that within the context of s 2(4)(b) “detention” was to be read ejusdem generis with “imprisonment”, both limited to imprisonment or detention as a consequence of a breach of the criminal law. 27. An order that a person be held in custody pending trial is not imposed for any breach of the criminal law; such an order is not in any way equivalent to a sentence. It is made solely to ensure the integrity of the criminal justice process. The Vice President was of the view that S 67A of the Criminal Procedure Ordinance does not ex post facto alter this fact: “The sentence commences on the date it is imposed. Section 67A… does not deem the sentence of imprisonment to have started from an earlier date, namely, the date of remand. And a judge does not have power to order a sentence to commence on some earlier date. Section 67A operates merely to treat the sentence passed as reduced: see Chan Hung v Commissioner of Correctional Services [2000] 3 HKC 767.[12] 28. The Vice President continued: “I am far from asserting that a true state of affairs on a given date may not be ascertained by reference to matters coming to light after that date. Nonetheless, the subsequent conviction and s 67A do not turn the remand order into something it is not, namely, a sentence of imprisonment.”[13] 29. Looking to the consequences of an interpretation in accordance with that made at first instance, the Vice President observed: “… the argument which permits ex post facto characterisation of the nature of the detention would enable, possibly require, the decision-maker in the face of an extant verification application, to delay his decision in order to await trial and its outcome and consequential appeals (and then, if there were an acquittal, to discount that period in custody); a scenario which is unlikely to have been in the legislature’s contemplation.”[14] 30. The Vice President further expressed concern that an interpretation in accordance with that made at first instance would give rise to a potential for injustice. This was on the basis that too much would turn on whether an applicant was or was not granted bail. If granted bail, he would remain ordinarily resident and able to seek verification of his status as a permanent resident; if refused bail, he would no longer be ordinarily resident. 31. For these reasons, the Vice President held that the provisions of s 2(4)(b) do not shut out an applicant by reason only of a period of remand in custody pending trial. The respondent was therefore held to be “ordinarily resident” at the time when he made his application to the Director. Submissions before this Court 32. Lord Pannick QC, who appeared as leading counsel for the appellants[15],submitted that the interpretation sought by the appellants - that a period in custody pending trial, when the trial resulted in conviction and imprisonment, fell within the meaning of “detention pursuant to the… order of any court” - was properly accommodated within the jurisprudence of this Court set out in Fateh Muhammad and Prem Singh. The following matters of law were common ground: (1) Within the context of s 2(4)(b), “detention pursuant to the… order of any court” was of the same genus as “imprisonment pursuant to the sentence… of any court”. Both had to be punitive in nature. (2) Detention pending trial by the order of any court was not punitive in nature. It was designed to ensure the integrity of the criminal trial process not to inflict punishment. (3) A sentence of imprisonment commenced when it was imposed; it was not retrospective. (4) For the purposes of article 24(2)(4) of the Basic Law, an application for verification of permanent residence is to be judged by reference to the seven-year period immediately preceding the date of application. 33. Within these confines, Lord Pannick’s submission was succinct, being founded upon the uncontentious proposition that, as s 2(4)(b) sought to qualify a constitutional provision, it was to be given a purposive construction. 34. Central to his submission was the operation and effect of s 67A of the Criminal Procedure Ordinance. While Lord Pannick did not seek to conflate the concept of detention pending trial with that of imprisonment imposed after conviction by sentence of a court, he submitted that it was permissible in seeking a purposive construction to look to the entirety of the process. That process consists, first, of a period of detention in custody pending trial, second, the subsequent conviction and sentence to a term of imprisonment and, third, the fact that by operation of s 67A the length of that sentence is to be treated as being reduced by the time spent in custody pending trial. It was submitted that, viewed as such, the period of custody pending trial becomes integral to the sentence of imprisonment, defining its true term. The causative link is s 67A. That link, it was submitted, drew the two concepts into such close proximity that in such circumstances a period of custody pending trial falls within the ambit of “detention pursuant to the… order of any court” in s 2(4)(b). 35. The following matters, it was suggested, must weigh in support of such a purposive construction. (1) By way of general observation, detention pending trial, by its nature, does not accord with the ordinary and natural meaning of the term “ordinarily resided” as defined by Lord Scarman in ex parte Shah[16]. Even if not consequent upon sentence, it is nevertheless an enforced presence which cannot be regarded as voluntarily adopted nor does it qualify as being for a settled purpose. It is a condition that is out of the ordinary. (2) Seven years ordinary and continuous residence is a qualification prescribed by the Basic Law for attaining a valuable status and right, namely, the status of a permanent resident and the right of abode[17]. For this reason it is appropriate to give article 24(2)(4) a restrictive meaning, limiting the scope and application of the statutory machinery allowing for the attainment of the right. Relevant to this, when looking to a period of detention pending trial, when the trial results in conviction and imprisonment, is the observation of Cheung J at first instance that “as a matter of general principle, no man should be allowed to profit from his own wrongdoing”. 36. It was further submitted that this construction would avoid one ‘stark anomaly’. Absent this construction, persons who are held in detention pending trial and persons who are on bail are both considered for the duration of that period to be ordinarily resident. If however both are convicted and sentenced to terms of imprisonment, the operation of s 67A reduces the sentence of the one who has been held in custody pending trial, enabling him to begin tolling afresh his seven years ordinary residence earlier than the one who has been on bail pending his trial. Conclusions 37. In my judgment, Lord Pannick’s submission must fail on the basis that s 67A does not constitute the compelling link that his submission requires it to be. 38. In Vallejos and Domingo v Commissioner of Registration[18], which is to be handed down on the same day as the present judgment, it is pointed out that Ex parte Shah[19] serves only as a starting-point when approaching the meaning of “ordinarily resided” in article 24(2)(4) of the Basic Law. It is also emphasised that the quality of residence relied on is important to determining whether it constitutes “ordinary residence” in the particular case. There is a fundamental qualitative difference between detention pending trial and imprisonment pursuant to the sentence of the court after trial. The first, even though it imposes hardship on the individual, is not punitive. The second is designed for the very purpose of being punitive. In this qualitative sense, an order of detention pending trial is more akin to an order of detention made by a court by reason of a subject’s mental illness, neither having any punitive element, and it is undisputed that the latter order is excluded from the ambit of s 2(4)(b). 39. The legislative intent behind s 67A is plain. It does not seek to reduce a sentence of imprisonment, the sentence remains the same. It does not seek to conflate the different regimes of detention prior to trial and imprisonment consequent upon sentence after trial. They remain separate regimes. It seeks to do no more than treat the individual prisoner fairly by recognising the hardship imposed upon him if, prior to his trial, he was kept in custody for the purpose of ensuring the integrity of the criminal trial process. 40. If, as agreed, a sentence of imprisonment takes effect when it is imposed, having no retrospective effect, I do not see that a limited statutory measure such as s 67A, which is only triggered if a sentence of imprisonment is imposed after trial, is capable of reaching back and giving to a period of detention pending trial the necessary punitive character so that, as a part of the full process of detention, conviction and imprisonment but nevertheless a distinct part, it is to be treated, for the purposes of s 2(4)(b), as being of the same genus as a sentence of imprisonment. 41. Lord Pannick has very correctly said that a supporting element of any purposive construction is the avoidance of unfairness. In my view, however, the far greater potential for unfairness lies in his construction. I say that because its consequence is to place pivotal emphasis on the uncertainties of seeking bail, a matter to which Stock VP alluded in his judgment in the Court of Appeal. A simple example suffices. Two persons charged with the same offence seek bail. One is able to obtain the provision of a surety which secures his liberty, the other is not. On the construction proposed, the one who secures bail is not constrained by s 2(4)(b) and remains ordinarily resident in Hong Kong while the one who fails to secure bail is constrained by the section and from the time of the order placing him into detention pending trial is no longer ordinarily resident. This is, and remains, so even if both are convicted after trial and sentenced to terms of imprisonment. 42. Finally, it seems to me that in a purposive construction of s 2(4)(b) there is much to be said for the need to seek finality in the administrative scheme managed by the Director in verifying claims for permanent resident status. The temporal parameters set by paragraph 1(4)(b) of Schedule 1 to the Ordinance are unambiguous and consistent with nature and purpose of article 24 (2)(4). 43. For the reasons given, I would dismiss the appeal, costs following the event. Sir Anthony Mason NPJ: 44. I agree with the judgment of Mr Justice Hartmann NPJ. Mr Justice Chan PJ : 45. The Court unanimously dismisses the appeal with costs. Lord Pannick QC, Mr Anderson Chow SC and Ms Eva Sit, instructed by Department of Justice for the appellants Mr Hectar Pun and Mr Newman Lam, instructed by Yip & Liu for the respondent [1] See s 2A of the Immigration Ordinance. [2] Cap 115. [3] (2001) 4 HKCFAR 278. [4] (2003) 6 HKCFAR 26. [5] [2010] 3 HKLRD 594. [6] p.283F. [7] Bokhary PJ went on to qualify this interpretation by saying: “Accordingly the only kind of detention covered by s 2(4)(b) is detention in a training centre or in a detention centre. (The word “order” in s 2(4)(b) is needed because, although s.4 of the Training Centres Ordinance (Cap 280) speaks of a “sentence of detention”, s 4 of the Detention Centres Ordinance (Cap 239) speaks of a “detention order”.)” Before us, other statutory provisions imposing “orders of detention” for breaches of the criminal law were identified, more particularly s 17 of the Reformatory Schools Ordinance, Cap 225; s 14 of the Juvenile Offenders Ordinance, Cap 226, and s 4 of the Drug Addiction Treatment Centres Ordinance, Cap 244. [8] Cap 221. [9] [1983] 2 AC 309, at 343G. [10] Paragraph 28 of the judgment. [11] Unreported, CACV 87/2010 dated 28 June 2011. [12] Paragraph 38. [13] Paragraph 40. [14] Paragraph 42. [15] Together with Mr Anderson Chow SC and Ms Eva Sit. [16] Supra, paragraph 22. [17] See the judgment of Bokhary PJ in Fateh Muhammad, p.284E. [18] FACV Nos. 19 and 20 of 2012. [19] Supra, paragraph 22. 
Chief Justice Ma: 1. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Tang PJ: 3. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Fok PJ: 4. This appeal involves the following questions of law, namely: “In the context of the regulatory regime under the Road Traffic Ordinance Cap.374 and its provisions governing road traffic offences and, in particular, in relation to the offence of using an unlicensed vehicle contrary to section 52(1)(a) and 52(10)(a) of the Road Traffic Ordinance, Cap.374: (1) What is the meaning of “use”? (2) Whether the meaning of “use” is in any way qualified by “drive”? (3) Whether, in relation to a person having the immediate “use” of a vehicle, the meaning of “use” is in any way qualified by the presence in a provision of the alternative of permitting or suffering the vehicle to be used or otherwise by the capacity of such a person?” A. Introduction 5. The respondent to this appeal, Mr Cheung Wai Kwong (“Cheung”), was prosecuted for the offence of possession of an offensive weapon in a public place[1] and, relevantly for this appeal, for the offence of using an unlicensed vehicle contrary to sections 52(1)(a) and 52(10)(a) of the Road Traffic Ordinance (Cap.374) (“the RTO”) (set out below). It was the prosecution case that, on 18 August 2015, in Ma Tong Road, Yuen Long, Cheung had with him a baseball bat and metal rod in a public place without lawful authority or reasonable excuse and that he used a private car (with the registration mark KG8680) on a road when the car was not registered and licensed in accordance with the RTO. 6. On 25 April 2016, Cheung was convicted of both offences and sentenced to 4 months’ imprisonment on the offensive weapons charge and 2 months’ imprisonment on the charge of using an unlicensed vehicle, both sentences to run consecutively.[2] 7. Cheung initially appealed his conviction and sentence in respect of both offences to the Court of First Instance but, shortly before the hearing of his appeal, he indicated through his counsel that he would only appeal in relation to the offence of using an unlicensed vehicle. By his judgment dated 23 September 2016,[3] the judge allowed Cheung’s appeal (limited to that conviction) and accordingly quashed his conviction and set aside his sentence for that offence. On 19 October 2016, on the application of the prosecution, the judge granted a certificate that the questions of law set out in paragraph [4] above were involved in the appeal and were reasonably arguable. Leave to appeal to this Court was granted by the Appeal Committee on 26 April 2017.[4] B. The facts relating to the s.52(1) charge 8. The prosecution case was based on the evidence of a police officer who observed a private car bearing registration number KG8680 parked at a bus stop in Ma Tong Road at 3.15am on 18 August 2015. A man (Cheung, whose identity was not in dispute) was seen sitting in the driver’s seat of the car, which had its headlights as well as the reading light inside the vehicle compartment on. No one else was inside the car. Cheung was observed for about five minutes and was seen to be bending down inside the car and looking outside from time to time. At 3.20am, Cheung switched off the headlights, alighted from the car and locked it. He then walked towards the rear of the car. The police suspected him of theft from the car and arrested him. The car key was found on Cheung and, when the car was searched, the baseball bat and metal rod were found in its boot. On investigation, it was discovered that the car was registered to a Mr Chan Lap-lun (“Chan”) but that its vehicle licence had expired as at 18 August 2015. 9. Cheung’s case was that he ran a car repair business and that he had bought the car from the registered owner, Chan. Chan gave Cheung a key to the car but it was arranged that delivery of the car would take place after the vehicle licence had expired. Cheung said that, on the evening of 17 August 2015, Chan called to tell Cheung that he had left the car in Ma Tong Road for him to collect but reminded him that it had some minor problems. After work that day, Cheung went to the car’s location in Ma Tong Road with the intention of collecting the vehicle registration book and then arranging for it to be towed to his garage for repairs to be carried out. C. The decisions below and the parties’ contentions on this appeal 10. The magistrate did not believe Cheung’s version of events. She believed instead that he drove the car to Ma Tong Road. However, she acknowledged that this was not the charge he faced under sections 52(1) and 52(10) of the RTO. Proceeding on the premise that a parked car could be in use on a road, she considered that the evidence relied upon by the prosecution (summarised in paragraph [8] above) was sufficient to prove that Cheung had used the car and that it was unlicensed. On this basis, she convicted Cheung of the offence. 11. On intermediate appeal, the judge rightly focused on the meaning of the word “use” in section 52(1) of the RTO. Applying a line of authorities considering that word in other road traffic legislation in England and which had been followed in Hong Kong (which I shall address below), he held that the word “use” in section 52(1) applied only to the driver of the vehicle or his employer where the driver was employed to drive for the employer’s affairs. On this basis, he held that, although the prosecution evidence showed that Cheung did use the car, it was insufficient to support the charge since the prosecution did not allege he drove the car to Ma Tong Road nor was there sufficient evidence to prove that fact. Accordingly, he allowed Cheung’s appeal. 12. On this appeal, Ms Vinci Lam SADPP, for the prosecution,[5] contended that the judge’s construction of the word “use” in section 52(1) of the RTO was too narrow and that a wider construction should be adopted on the basis of which this Court should conclude that Cheung did use the vehicle within the meaning of the section. At the beginning of her submissions, Ms Lam indicated that, in the event of the appeal being allowed, the Secretary for Justice did not seek the reinstatement of the conviction for using an unlicensed vehicle and that the appeal was pursued to provide guidance for future prosecutions for the offence. 13. For the respondent, Mr Hectar Pun SC,[6] contended that, adopting a purposive construction, the word “use” in section 52(1) means the use of a vehicle as a means of transport or locomotion and that, on the facts of this case, there was no use of a kind within the mischief to which section 52(1) was directed. Accordingly, he sought to uphold the judge’s construction of the word “use” in section 52(1) of the RTO. D. The construction of s.52(1) 14. The parties’ respective contentions raise the questions involved in this appeal and the answers to those questions depend on the proper construction of section 52(1) of the RTO. Applying the well-settled approach to statutory construction, this requires the Court to construe the relevant language used in the light of its context and purpose.[7] 15. Before embarking on this construction exercise, it should be noted, by way of caveat, that the particular language used in section 52(1) of the RTO differs, to a greater or lesser extent, from that under consideration in a number of the authorities cited in this appeal and considered below. For this reason, caution must be exercised in seeking to apply some of those cases directly to the present appeal. D.1 The statutory language and structure of section 52(1) 16. Section 52 consists of a series of prohibitory statements relating to different types of vehicle (sub-sections (1) to (9) inclusive) and then a provision creating offences setting out the maximum penalties (sub-section (10)). Sections 52(1) and 52(10) of the RTO provide as follows: “(1) Subject to this Ordinance, no person shall – (a) drive or use a vehicle of a class specified in Schedule 1, other than a rickshaw; or (b) suffer or permit such a vehicle to be driven or used, on a road unless the vehicle is registered and licensed in accordance with this Ordinance. … (10) Any person who contravenes – (a) this section, other than subsection (2), commits an offence and is liable in the case of a first conviction for that offence to a fine of $5,000 and to imprisonment for 3 months, and in the case of a second or subsequent conviction for that offence to a fine of $10,000 and to imprisonment for 6 months; (b) subsection (2) commits an offence and is liable in the case of a first conviction for that offence to a fine of $1,000 and to imprisonment for 3 months, and in the case of a second or subsequent conviction for that offence to a fine of $2,000 and to imprisonment for 6 months.”[8] 17. As a matter of language, section 52(1) is drafted to impose a prohibition on any person from carrying out certain activities, namely (a) driving or using, or (b) suffering or permitting the driving or using, of a relevant vehicle on a road unless registered and licensed as prescribed. 18. The nature of the activities that can constitute the offence is addressed in Section D.3 below but it should be noted at this stage that a difference between the activities in (a) and (b) is that the former activities of driving or using create what have been described as strict or absolute offences, in the sense that no mens rea other than driving or user need be proved. On the other hand, the latter activities of suffering or permitting the driving or user of a vehicle involve the offender having the state of mind of suffering or permitting another to drive or use the vehicle. For these propositions, see the decisions of the English Court of Appeal in James & Son Ld. v Smee and Green v Burnett & Anor [1955] 1 QB 78[9] per Parker J (as he then was) at p.90, Lord Goddard CJ at pp.93-94 and Slade J (as he then was) at p.95; and, in Hong Kong, Cheung So & Anor v The Queen [1967] HKLR 704[10] per Briggs J (as he then was) at pp.705-707, and Wong Yun-fai v The Queen [1977] HKLR 404[11] per Leonard J (as he then was) at p.405. On either basis, however, it is unnecessary for the prosecution to prove knowledge on the part of the defendant that the vehicle is not duly registered or licensed. D.2 The purpose of s.52 and its context within the scheme of the RTO 19. The Long Title to the RTO states its purpose as being “To provide for the regulation of road traffic and the use of vehicles and roads (including private roads) and for other purposes connected therewith”. Within the RTO, section 52 is the first section in Part 6 of that ordinance which is headed “Use, Sale and Hire of Vehicles”. Section 52 itself is headed “Restriction on the use of vehicles”. 20. The aim of section 52(1) is broad and, in requiring every vehicle driven or used on a road to be registered and licensed under the statutory registration and licensing systems,[12] includes at least three different purposes: (1) taxation and revenue generation (by the requirement to have paid registration and licence fees before a vehicle can be used on a road); (2) administrative regulation (by the registration and licensing systems which enable vehicles and their owners to be traced and identified); and also (3) road safety and protection of the public (by the registration and licensing systems that respectively ensure mechanical soundness and the requirement that all registered vehicles be covered by third party insurance). 21. Section 52(1) must also be construed in the context of the other provisions of the RTO itself and cognate ordinances and regulations which include the offences relating to driving a motor vehicle under the influence of alcohol or drugs (RTO s.39A et seq.), the Motor Vehicles Insurance (Third Party Risks) Ordinance (Cap.272),[13] and the Road Traffic (Registration and Licensing of Vehicles) Regulations (Cap.374E). 22. In relation to the offences of driving under the influence of alcohol or drugs, the RTO makes it an offence for a person who “drives or attempts to drive a motor vehicle, or is in charge of a motor vehicle, on any road”[14] with an alcohol concentration above the prescribed limit or under the influence of drugs. These sections therefore introduce a further activity in relation to a motor vehicle, namely that of being “in charge of” a vehicle, which must be considered as part of the context of the word “use” in section 52(1). As will be seen, being “in charge of” a vehicle is a broad concept. D.3 The activities that can constitute the offence under s.52(1) 23. Under section 52(1) of the RTO, three types of activity are prohibited in relation to an unregistered or unlicensed vehicle, namely no person shall (i) “drive”, (ii) “use” or (iii) “suffer or permit [it] to be driven or used”. 24. I shall address the words “drive” and “use” below. It is unnecessary in this appeal to address the acts of suffering or permitting a vehicle to be driven or used. For present purposes, there is no material difference between “suffer” or “permit” and they can be regarded as indicating different degrees of causation by a person charged with enabling the driving or using. 25. It is to be noted, however, that the prohibitions in section 52(1) are addressed to the person doing the particular activity. Thus, the prohibition against driving is directed to the driver, that against using to the user, and that against suffering or permitting the driving or using to the person so suffering or permitting. Properly defined, each of these acts, in relation to an unregistered or unlicensed motor vehicle, is within the mischief to which section 52(1) is directed. D.3a To “drive” 26. Applying the ordinary meaning of the word as a transitive verb as it is used in section 52(1), to “drive” a motor vehicle means to “operate and control the course of” that vehicle.[15] This is an activity that most people will easily be able to recognise most of the time when it is happening. The person doing the activity will usually be sitting in the driver’s seat of the car and directly operating the vehicle, usually (but not necessarily) with the engine running, so as to cause it to move, change directions and stop. 27. As Lord Widgery CJ said in R v MacDonagh[16]: “The essence of driving is the use of the driver’s controls in order to direct the movement, however that movement is produced.” Thus, the English Court of Appeal held that a man with both feet on the road pushing a car by putting his shoulder against the door pillar and one hand on the steering wheel to control its movement was not driving it. 28. The present appeal is not concerned with the meaning of “drive” in section 52(1) but the word forms part of the context of “use” in that provision. In this regard, it is instructive to note in passing cases on the meaning of driving which show that the courts have concluded that the act of driving may comprise times when the vehicle is stopped and that there is no principle of law that, as soon as a person alights from a vehicle, he can no longer be said to be driving or attempting to drive: see, in this context, R v Reid (Philip) [1973] 1 WLR 1283 at p.1288C-D referring to Pinner v Everett [1969] 1 WLR 1266[17] and Sakhuja v Allen [1973] AC 152.[18] In Edkins v Knowles [1973] 1 QB 748, Griffiths J (as he then was) summarised the collective effect of the earlier authorities as follows: “(1) The vehicle does not have to be in motion; there will always be a brief interval of time after the vehicle has been brought to rest and before the motorist has completed those operations necessarily connected with driving, such as applying the handbrake, switching off the ignition and securing this vehicle, during which he must still be considered to be driving. (2) When a motorist stops before he has completed his journey he may still be driving; an obvious example is when he is halted at traffic lights. Each case will depend upon its own facts, but generally the following questions will be relevant: (a) What was the purpose of the stop? If it is connected with the driving, and not for some purpose unconnected with the driving, the facts may justify a finding that the driving is continuing although the vehicle is stationary. (b) How long was he stopped? The longer he is stopped the more difficult it becomes to regard him as still driving. (c) Did he get out of the vehicle? If he remains in the vehicle it is some though not a conclusive indication that he is still driving.”[19] 29. Those cases did not concern statutory provisions that included the alternative of “use” as well as “drive” but the principle that emerges from them is that there may be temporary periods during which a person may have stopped a motor vehicle yet still properly be said to be driving it. D.3b The meaning of “use” 30. Whilst the activity of driving a motor vehicle is relatively straight forward to identify, that of using a vehicle is less so because of the inherent breadth of the verb “use”. 31. In Hickman v Chichester District Council,[20] McCullough J observed: “The verb ‘to use’ is one of the most common in the language. It has different shades of meaning. It is found, in different contexts, in different statutes. Like any other word it must always be construed in the context in which it appears.” And, in the same case, Neill LJ said: “The verb ‘to use’ is capable of bearing many different shades of meaning. When the object used is a motor vehicle the verb will usually connote that the vehicle is being used for its primary purpose of being driven or moved. It may then be relevant to consider the precise relationship between the alleged user and the vehicle to determine whether that person has control of the vehicle.”[21] 32. The inclusion of the alternatives of driving and using in section 52(1) indicates that the Legislature recognised and intended a distinction between the two activities. As Rigby CJ stated in R v Lau Ping: “… where the Legislature, in the course of the same Act or Ordinance and in subsidiary legislation made thereunder, has deliberately chosen to use different words ‘use’ and ‘drive’ in different sections and in different contexts and, indeed, in the Road Traffic (Construction and Use) Regulations, to use the words ‘drive’ or ‘use’ in the same regulation, it must be assumed that it intended different meanings to be attached to the use of those distinct words or expressions.”[22] 33. There can be little doubt that, as a matter of ordinary language, to drive a motor vehicle is to use it: see Gifford v Whittaker [1942] 1 KB 501 per Viscount Caldecote CJ at p.505; and R v Lau Ping [1970] HKLR 343 per Rigby CJ at p.351. In Elliott v Grey, a case concerning the then English equivalent of section 4(1) of the Motor Vehicles Insurance (Third Party Risks) Ordinance (Cap.272) (see FN 13 above), Lord Parker CJ noted:[23] “One thing seems to be clear, and that is that ‘use’ there is used in contradistinction to the word ‘drive,’ which appears in other sections … Prima facie, it is a wider term and includes something more than driving and certainly would include moving.” 34. As a matter of construction, therefore, the separate use of “drive” and “use” in section 52(1) indicates that, although it may include the activity of driving, to “use” a vehicle encompasses different and wider activities than driving. The critical question is what, more than driving, is included in the activity of using a motor vehicle? I shall return to address this question after first dealing with the principle applied by the judge below in determining that Cheung could not relevantly have been using the vehicle in question at the time of the alleged offence. D.3c The Windle v Dunning & Son Ltd line of cases 35. Given the breadth of the verb “use”, the activity of using an object can be carried out vicariously. In relation to a motor vehicle, although one person may be driving the vehicle, another may, as a matter of ordinary language, be said to be using the vehicle. In ordinary language, one uses a bus or taxi to go from one place to another. Similarly, a person may, by hiring a vehicle and its driver, have a degree of control over the movement and course of the vehicle such that he may, in general terms, be said to be using that vehicle. The width of the word “use” is sufficient to embrace the use of a thing by a person through the medium of an agent and so does not restrict the actor who may use the vehicle to the person driving it. 36. However, use of a vehicle in those situations is far removed from the purposes of section 52(1) or the person upon whom the responsibility for registering, licensing and insuring the vehicle ought to fall. There is therefore a need to restrict the breadth of the word “use” so as not to cast the net of liability under the section too widely. 37. Consistent with that consideration, a line of authorities has developed in England and Wales (and has been followed in Hong Kong) construing the word “use”, in road traffic legislation directed to a person using a vehicle which includes the alternatives of causing or permitting such use, narrowly. So construed, a person uses the vehicle only if he is the driver of the vehicle or its owner, where the driver is employed by the owner under a contract of service and at the material time was driving on his employer’s business. In this way, vicarious liability for using a vehicle which another person is driving is limited. 38. The first case in this line of authorities is Windle v Dunning & Son Ltd [1968] 1 WLR 552. The defendants there had hired three lorries from haulage contractors, who operated the vehicles and employed and supplied the drivers. The lorries were loaded under the defendants’ supervision. They were found to have exceeded the permitted weight limit and the defendants were prosecuted under a provision prohibiting use of a vehicle which did not comply with the relevant regulations.[24] Lord Parker CJ held: “In my judgment, as was said by Lord Clyde in giving the judgment in MacLeod v Penman[25]: ‘The presence in the section of the alternatives of causing or permitting the use must limit the scope of what is ‘using.’ Normally, ‘using’ is applicable to the actual driver.’ I entirely agree with that, and in my judgment ‘using’ when used in connection with causing and permitting has a restricted meaning. It certainly covers the driver; it may also cover the driver’s employer if he, the driver, is about his master’s business …”.[26] 39. Next, in Crawford v Haughton [1972] 1 WLR 572, the defendant, who was the owner of a vehicle which did not comply with the relevant construction and use regulations, had been convicted of using the vehicle in that condition on the basis that the driver of the vehicle, although not his employee, was driving with the defendant’s authority and knowledge and for the purposes of his business. Although noting that the evidence could have supported a charge of causing or permitting the vehicle to be used, the English Court of Appeal, applying Windle v Dunning & Son Ltd, quashed the conviction for using the vehicle, Lord Widgery CJ holding: “I have not found this a particularly easy case because I find it difficult to accept that, if a man can use a vehicle through the hands of his servant, he cannot be said to use it at the hands of someone who at his specific request drives it on a journey at the express orders and with the full knowledge of the owner. No doubt the line must be drawn somewhere, and the judgments of Lord Parker CJ to which I have referred show a tendency to restrict the capacity of persons using in cases where the alternatives of permitting or causing to be used are provided. I have thought for some time that it might be right for us to say in the present case that there is yet another category of user for present purposes, not merely the actual driver or his employer, but someone who by specific and immediate direction causes a vehicle to be driven in the manner in which it was driven in this case. But in the end I have come to the conclusion that it would not be right, in view of the authorities, to strive to extend the meaning of ‘use’ for the present purposes, and that only confusion may follow in subsequent cases if we endeavour so to do.”[27] 40. In Howard v G T Jones & Co Ltd [1975] RTR 150, Lord Widgery CJ acknowledged that drawing the line as it had been drawn might be criticised but held: “The position up to and including the views of this court in Crawford v Haughton really can be summarised in this way: if the legislation provides as alternatives using, causing, or permitting, then prima facie using relates to the man who actually drives; using is the driver’s offence, whereas permitting and causing are, as it were, the owner’s offences. However, the authorities have accepted, as my reference to Crawford v Haughton demonstrates, that where the vehicle is being driven by a servant of the owner and on the owner’s business, that is equivalent to driving by the owner himself and thus can come within the scope of the phrase ‘using a vehicle’. But up to now although the drawing of the line at that point is not wholly logical, there has been no extension of the categories of persons who can be said to use a vehicle beyond first the driver, and secondly the driver’s employer if the driver is working under a contract of service on behalf of the owner of the vehicle.”[28] 41. In West Yorkshire Trading Standards Service v Lex Vehicle Leasing Ltd [1996] RTR 70, Dyson J (as he then was), with whom McCowan LJ agreed, described this narrow approach to the meaning of the word “use” or “uses” in the context of a statute containing the alternatives of “causes or permits” as having “a long pedigree”[29] and held: “Thus the line has been clearly and consistently drawn by this court. A person is a user only if he is the driver or the owner of the vehicle, but it applies to the owner only if the driver is employed by the owner under a contract of service and at the material time he is driving on his employer’s business. The line has been described variously as not wholly logical and as somewhat artificial, but it has been drawn by this court after due consideration has been given to those criticisms, to some extent, for pragmatic reasons and to avoid confusion. … For my part, I am wholly unpersuaded, despite the criticisms of illogicality and artificiality that this well-established line of earlier cases, to some of which I have referred, was erroneous in law. It is quite plain that, in particular, Lord Widgery CJ on several occasions considered the matter very carefully, alive to the criticisms and decided for the reasons that I have mentioned to maintain the principle.”[30] 42. The same approach, developed in cases involving the construction and use regulations, has been followed in relation to the statutory prohibition against use of a vehicle without third party insurance[31]: see Jones v Director of Public Prosecutions [1999] RTR 1 per Sullivan J at p.9B-D. 43. The Windle v Dunning & Son Ltd line of authorities has been followed in Hong Kong in Wong Yun-fai v The Queen [1977] HKLR 404[32] per Leonard J (as he then was) at pp.406-407 and, more recently, in HKSAR v Law Man Bun, unrep., HCMA 367/2003, Judgment dated 1 March 2004, per Deputy High Court Judge Toh (as she then was) at [15]-[16].[33] And, as noted above, the judge in the present case applied the principle in this line of authorities in reaching the conclusion that Cheung was not using the vehicle in question. 44. Two important points should be noted in relation to the Windle v Dunning & Son Ltd line of authorities which are material distinctions from the present case. (1) First, the defendants in those cases were at one remove from the person carrying out the relevant activity, i.e. driving the vehicle, and liability was sought to be attached vicariously to the respective defendants for their user of the vehicle. Here, in contrast, Cheung was the person actually carrying out the activity said to constitute using the vehicle in question and there is no question of his having used the vehicle vicariously through another person. (2) Secondly, the statutory prohibition in each case, although including the alternative “causes or permits” (not materially different to the alternative of “suffer or permit” in section 52(1) of the RTO), did not include a prohibition on both driving and using as in section 52(1) under consideration here. 45. For completeness, it should be noted that the limitation on vicarious user of a vehicle in the Windle v Dunning & Son Ltd line of authorities does not preclude a prosecution on the basis that the defendant has suffered or permitted the vehicle to be driven or used: see Crawford v Haughton [1972] 1 WLR 572 per Lord Widgery CJ at p.576H. However, as noted in paragraph [18] above, this will require the prosecution to prove the defendant’s state of mind in relation to the suffering or permitting the driving or user as the case may be. D.3d Other cases on the meaning of “use” 46. Before dealing with the critical question of what, more than driving, constitutes using a vehicle, it is also convenient to consider other cases on the meaning of “use”. Some of the decisions may be described as surprising but it is important to keep in mind the context and purpose of the relevant statutory provisions under consideration. 47. I have already referred to Elliott v Grey [1960] 1 QB 367, which concerned a car parked outside its owner’s house. It had broken down and the defendant owner left it in the road until it could be repaired. Insurance was suspended in the meantime. The car could not be propelled mechanically since its engine did not work, the battery had been removed and there was no petrol in the tank. The defendant had no intention of driving the car or moving it from its position in the road. He was convicted of unlawfully using the car without third party insurance contrary to section 35(1) of the Road Traffic Act 1930[34] and his conviction was upheld by the Court of Appeal. Lord Parker CJ (with whom Cassels J and Edmund Davies J (as he then was) agreed) construed the word “use” as being equivalent to “have the use of a motor-vehicle on the road” and held: “In the present case, as I have said, though this car could not be driven, there is nothing to suggest that it could not be moved. As I pointed out in argument, for all we know it was on the top of a hill and a little boy could release the brake and the car could go careering down the hill. In the absence, at any rate, of a finding that it was immovable as, for instance, the wheels were removed or something of that sort, I cannot bring myself to think that this car was not fairly and squarely within the words which I have used, ‘have the use of a motor-vehicle on the road.’ … Here was a car which could be moved, albeit not driven, and, in those circumstances, I think the owner had the use of it on a road within the meaning of the section.”[35] 48. The distinction between a vehicle that is moveable and immoveable as the difference between using and not using a vehicle was applied in the cases of Hewer v Cutler [1974] RTR 155 (a case concerning prosecution for use of a vehicle without a test certificate) and Eden v Mitchell [1975] RTR 425 (a case concerning use of a vehicle with defective tyres). In Hewer v Cutler, the vehicle could not be driven and could not be moved and so its being parked on a road was held not to be user. The court there declined to take the step of accepting the argument that, even if the car was completely immobilised, there would still be an offending user. In Eden v Mitchell, the defendant’s parked car was not immobilised and was capable of being used. Notwithstanding that the defendant did not intend to use it, since the vehicle had not been completely immobilised, the prosecutor’s appeal against the dismissal of the informations was allowed. 49. Some care needs to be taken, in my view, in reading “use” in section 52(1) as being the equivalent of “have the use of a motor-vehicle on the road” as in Elliott v Grey. In so reading “use”, this approach distracts from the transitive verb “use” which focuses on the activity of using and the actor who is carrying out that activity and instead looks to the state of the vehicle alone and to the passive state of the owner, without any regard to any activity by any person. 50. User being an activity that can be carried out vicariously (as recognised by the Windle v Dunning & Son Ltd line of cases), the question arose in Brown v Roberts [1965] 1 QB 1 as to whether a passenger could be said to use a vehicle. In that case, a passenger opened the door of a van, in which she had been given a lift by its owner, and negligently struck and injured the plaintiff. The issue in the case was whether the owner had been in breach of his statutory duty in that he permitted the passenger to use the vehicle on the road without third party insurance as required under section 35(1) of the Road Traffic Act 1930. 51. On the question of use, Megaw J (as he then was) accepted (following Elliott v Grey) that the element of driving a vehicle is not an essential element of “using”, nor that “use” was confined to the owner of the vehicle.[36] However, he rejected the plaintiff’s submission that “use” should be given a wide meaning and that it was enough that the defendant was using the vehicle on the road as a means of transport. Instead, he accepted the submissions for the defendant that, first, for the purposes of use of a vehicle on the road, there must be at least an element, on the part of the user, of controlling, managing or operating the vehicle as a vehicle and, secondly, that it was inconsistent with the legislation as a whole to hold that use of a vehicle on a road within the meaning of section 35(1) of the 1930 Act included the mere relationship of a passenger to the motor vehicle or to a particular part of the motor vehicle.[37] He concluded: “[Counsel for the 2nd defendant owner] is right in his contention that a person does not ‘use … a motor vehicle on a road’ for the purposes of section 35(1) of the Act unless there is present, in the person alleged to be the user, an element of controlling, managing or operating the vehicle at the relevant time. Precisely what the extent of that element may be, it is unnecessary to seek to define. There was no such element present in the relationship between [the 1st defendant passenger] and the second defendant’s van. I do not accept that the control or management or operation of a door of the vehicle by the passenger entering or alighting amounts to the necessary control or management or operating of the vehicle.”[38] 52. The meaning given to user by Megaw J was endorsed by the English Court of Appeal in Hatton v Hall & Anor [1997] RTR 212. In that case, it was sought to be argued that the pillion passenger of a motorcycle was using the motorcycle. The Court of Appeal rejected this argument and applied Brown v Roberts in holding that, given the serious legal consequences of a finding that a person was using a vehicle under section 143(1) of the Road Traffic Act 1988,[39] “use” had to be given the restricted meaning imposed by Megaw J.[40] 53. In the course of his judgment for the court, Henry LJ referred to cases in which passengers had, nevertheless, been held to be using a vehicle, namely Leathley v Tatton [1980] RTR 21 and Stinton v Stinton [1995] RTR 167. These cases (which respectively arose out of somewhat unusual facts) were, however, reached on the basis that the passengers had been engaged in a joint enterprise or joint venture with the actual driver and so were jointly using the vehicle. The principle applied in such cases is whether the passenger was exercising a sufficient degree of control or management over the vehicle to be jointly using it. Such cases will depend on the evidence and Henry LJ upheld the first instance judge’s distinguishing of the joint enterprise approach in Stinton v Stinton and his application of the test of user in Brown v Roberts. He said: “Fundamental to the distinction he made was his assumption that not all plans shared between driver and passenger give the passenger sufficient management of the vehicle to make him a user of the vehicle. That conclusion seems to us to be good sense as well as good law. If a passenger employs a driver to drive against the clock to catch an aeroplane he clearly has a greater power to control and/or manage the vehicle than would be the case if a father agreed to drive his 18-year-old son and 80-year-old mother-in-law, neither able to drive, to visit his wife in hospital. Similarly, if a bank-robber is a pillion passenger on a get-away motorcycle, it may be inferred that he has a greater degree of management and control over the driver than the pillion passenger in the instant case. It is a question of fact and degree in each case, but in all cases user must be given the restricted meaning found by Megaw J.”[41] 54. In the recent case of UK Insurance Ltd v Holden & Anor [2017] 3 WLR 450, the English Court of Appeal considered the question of whether an owner’s carrying out of repairs to a vehicle to put it into a safe and good working condition fell within “use” of a vehicle for the purposes of section 145(3) of the Road Traffic Act 1988.[42] In his judgment, Sir Terence Etherton MR, having reviewed cases including Elliott v Grey, summarised a number of propositions including the following, namely that: “(1) ‘Use’ is not confined to the actual operation of the car in the sense of being driven. (2) There may be ‘use’ of a car when it is parked or even immobilised and incapable of being driven in the immediate future. (3) ‘Use’ of a vehicle includes anything which is consistent with the normal function of the vehicle.”[43] Hence, he concluded: “… it follows that the repair of a car, which the owner was driving but due to disrepair cannot be lawfully and safely driven, and which the owner wishes to effect as soon as possible in order to be able to drive the car lawfully and safely, is ‘use’ of the car within section 145(3)(a) of the RTA, being an activity consistent with its normal function for the purpose of that statutory provision.”[44] D.3e Being “in charge of” a motor vehicle 55. Although not an activity that falls within section 52(1), the concept of being in charge of a vehicle is one which is found in the RTO in the sections creating the offences of driving under the influence of alcohol or drugs and forms part of the context of section 52(1). This activity is described in the legislation in a descriptive sense in relation to the vehicle (“A person who … is in charge of a motor vehicle”) and is used in contrast to the activities of driving or attempting to drive. In terms of a spectrum of activities, being in charge of a vehicle is therefore a broad concept and, as a matter of ordinary language, requires less control, management or operation of a vehicle than does driving or using it. 56. The leading authority on the concept of being “in charge of” a motor vehicle is the decision of the English Divisional Court in DPP v Watkins [1989] 1 QB 821.[45] In that case, the defendant, who was drunk, was found in the early hours of the morning in the driver’s seat of a Mini car, not owned by him and parked in the street. He had a bunch of keys in his hand and one of these, marked “Honda”, was found between his thumb and forefinger. The key could be inserted in the ignition but there was no evidence it could start the Mini. On the prosecution’s appeal following the dismissal of the informations against the defendant, the Divisional Court allowed the appeal and remitted the case to the justices holding that a person, who was not the owner of a car, was in charge of the vehicle if he acted in a manner that showed that he had assumed control or intended to assume control of the vehicle preparatory to driving it. If the prosecution showed such assumption of control, the burden then fell on the defendant to prove that there was no likelihood of the vehicle being driven.[46] 57. In his judgment, Taylor LJ (as he then was) acknowledged that no hard and fast all-embracing test could be propounded as to the meaning of the phrase “in charge” but said: “The circumstances to be taken into account will vary infinitely but the following will usually be relevant: (i) Whether and where he is in the vehicle or how far he is from it. (ii) What he is doing at the relevant time. (iii) Whether he is in possession of a key that fits the ignition. (iv) Whether there is evidence of an intention to take or assert control of the car by driving or otherwise. (v) Whether any other person is in, at or near the vehicle and if so, the like particulars in respect of that person. It will be for the court to consider all the above factors with any others which may be relevant and reach its decision as a question of fact and degree.”[47] 58. Importantly, Taylor LJ drew a distinction between two classes of case. One is where the defendant is the owner or lawful possessor of the vehicle or has recently driven it. In such a case, he will have been in charge of it and the question for the court will be whether he is still in charge or whether he has relinquished such charge. The other is where the defendant is not the owner, lawful possessor or recent driver but is sitting in the vehicle or is otherwise involved with it. Here, the question for the court will be whether he has assumed being in charge of it.[48] 59. This distinction highlights the need, in the context of “in charge of” cases, of examining the status of the defendant and his connection or relationship with the vehicle. This is a consideration which, in my view, applies also to the question of whether a person is carrying out an activity that amounts to use of a vehicle and is reflected in the passage from Megaw J’s judgment in Brown v Roberts quoted in paragraph [51] above. D.3f What, more than driving, constitutes using a vehicle? 60. Pausing here, in terms of relevant prohibitions in the RTO, to “drive” a vehicle is an easily identifiable activity and a specific form of use (i.e. “use” embraces driving), but to “use” a vehicle can also involve an activity less specific than that. At the same time, to “use” a vehicle imports something more active than merely being “in charge of” it, which is a broadly defined act and depends on a range of circumstances and the person’s status and his connection or relation to the vehicle in question. 61. I now turn to the critical question in this case: what, more than driving, constitutes using a vehicle within section 52(1) of the RTO? 62. Having regard to the language of section 52(1), which includes both “drive” and “use”, and the context and purpose of the section, I would conclude that the proper construction of “use” is the activity of exercising a measure of control, management or operation over the motor vehicle as a means of transport, including any period of time between journeys. 63. This definition of “use” provides, in my view, a workable definition but it must be understood that, like the cases concerning being “in charge of” a vehicle, it can apply to different types of person. There may be some cases where the defendant is the owner or lawful possessor of the vehicle or has recently driven it. There may be other cases where the defendant is not the owner, lawful possessor or recent driver but is sitting in the vehicle or otherwise involved with it. The status of the person and his connection or relationship to the vehicle may influence the outcome of the decision as to whether any particular activity does or does not constitute a relevant user of the vehicle. 64. Further, since the Windle v Dunning & Son Ltd line of authorities is now well-established and has been consistently followed in Hong Kong, the extent to which a person may vicariously use a vehicle by another person’s driving is limited to the employer of the person driving or actively using it if the driving or use is for the business of the employer. But that principle does not preclude someone other than the driver of the vehicle being held to “use” the vehicle if he is the actor carrying out the relevant activity. 65. Just as Taylor LJ in DPP v Watkins noted that the circumstances to be taken into account for being “in charge of” a vehicle will vary infinitely and that ultimately each case will be one of fact and degree depending on various factors in the evidence, so too will each case where “use” of a vehicle is alleged. Numerous examples of varying degrees of activity can be posited (and some were canvassed in the course of the hearing of this appeal). It will not, in my view, assist to set out specific examples of activities falling within the definition of “use” and of others falling outside that definition. However, by focusing on the activity in question and the status of the actor and his connection and relationship to the vehicle, insofar as that informs the nature of the activity in which he is engaged, the proper scope of “use” in section 52(1) will be achieved. 66. I digress to comment on a case referred to by the prosecution in the Appellant’s Printed Case,[49] which was not cited by the parties below (nor in the case of HKSAR v Law Man Bun (supra)), namely The Queen v Poon Jing [1985] HKLR 341. That case concerned convictions for driving whilst disqualified under section 44(1) of the RTO and for driving (sic) a vehicle without third party insurance contrary to section 4(1) of the Motor Vehicles Insurance (Third Party Risks) Ordinance (Cap.272). On the facts, Deputy High Court Judge Cruden held that there was no act of driving and so quashed the convictions. The prosecution invited him to amend the second charge to using a vehicle without third party insurance, since the prohibition in section 4(1) (set out in FN13 above) is on using a vehicle without such insurance. The deputy judge refused to allow this amendment and, following cases including Wong Yun-fai v The Queen (supra) (which applied the principle laid down in Windle v Dunning & Son Ltd), held that where a person was charged with use of a vehicle, they were either the non-driver owners of the vehicles or the drivers themselves. Since there was no evidence the appellant was the owner, and on the facts he had not driven, the deputy judge held that the charge would not be made out even if amended. 67. The prosecution challenged such a restricted approach to the concept of using a vehicle. As already noted, that case did not concern section 52(1) of the RTO but I see no reason why the definition of “use” proposed above (or a broadly similar definition) should not also apply to “use” in section 4(1) of the Motor Vehicles Insurance (Third Party Risks) Ordinance (Cap.272). Insofar as the deputy judge in The Queen v Poon Jing held that no one other than the driver or his employer could use a vehicle, this holding cannot be supported since it is not consistent with the language of the prohibition which applies to all persons and is dependent on the application of a line of authorities dealing with the extent to which user can be vicarious. As will be seen, this error was also made by the judge in the present case. It is neither necessary nor appropriate, however, to examine whether the facts in The Queen v Poon Jing (which were disputed and not determined by the magistrate or deputy judge) would have been sufficient to sustain a charge of using a vehicle without third party insurance. E. Answers to the certified questions of law and summary on the construction of section 52(1) 68. In light of the above discussion, I would answer the certified questions of law as follows. The proper construction of “use” in section 52(1) is set out in Section D.3f above. The activity of using a vehicle is different to, and wider than, that of driving a vehicle: see Sections D.3a and D.3b above. Where a person is the immediate user of a vehicle and is directly carrying out the activity said to constitute the use, rather than vicariously through the act of another person driving the vehicle, the Windle v Dunning & Son Ltd line of authorities restricting the meaning of “use” does not apply. 69. To summarise my views set out above as to the proper construction of section 52(1) of the RTO:- (1) Section 52(1) prohibits the activities of driving, using and suffering or permitting the driving or using of an unregistered or unlicensed motor vehicle and is to be construed in the light of the three-fold purposes of: taxation and revenue generation; administrative regulation; and road safety and protection of the public. The prohibitions are addressed to the person doing the particular activity, each of which, properly defined, in relation to an unregistered or unlicensed motor vehicle, is within the mischief to which section 52(1) is directed. (2) By using the two verbs “drive” and “use”, section 52(1) contemplates different meanings to be attached to the two words. The latter term “use” is a wider term and includes something more than driving and would certainly include moving. (3) The activity of driving means to operate and control the course of a motor vehicle usually by sitting in the driver’s seat and directly operating the vehicle, usually (but not necessarily) with the engine running, so as to cause it to move, change directions and stop. The vehicle need not be in motion since the activity of driving will extend to those operations necessarily connected with driving and, depending on the reason for stopping, a person may still be driving the vehicle when it is stopped. (4) To “use” a motor vehicle includes driving it but also embraces a wider range of activities in relation to the vehicle less specific than driving. Having regard to the context and purpose of section 52(1), to “use” a vehicle means the activity of exercising a measure of control, management or operation over the vehicle as a means of transport, including any period of time between journeys. (5) To “use” a vehicle imports something more active than merely being “in charge of” it, which is a broadly defined act and depends on a range of circumstances (including those identified by Taylor LJ in DPP v Watkins [1989] 1 QB 821 at p.831F-G) and the person’s status and his connection or relation to the vehicle in question. (6) This definition of “use” can apply to different types of person, including the owner or lawful possessor of the vehicle or recent driver of it, or someone who is none of those but is sitting in the vehicle or otherwise involved with it. The status of the person and his connection or relationship to the vehicle may influence the outcome of the decision (e.g. excluding passengers) as to whether any particular activity constitutes prohibited user of the vehicle by him in the light of the statutory purposes mentioned above. (7) A person may “use” a vehicle vicariously through the agency of another person who is driving or using it but, in such a case, a restricted construction of “use” will be applied so that the user is limited to the employer of the person driving or actively using the vehicle if the driving or use is for the business of the employer. This construction does not preclude someone other than the driver of the vehicle being held to “use” it if he is the actor carrying out the relevant activity. (8) Whether any particular activity is user of a vehicle will depend on the evidence and the circumstances to be taken into account will vary infinitely. Each case will be one of fact and degree depending on various factors in the evidence. By focusing on the activity in question and the status of the actor and his connection and relationship to the vehicle, insofar as that informs the nature of the activity in which he is engaged, the proper scope of “use” in section 52(1) will be achieved. F. The Magistrate’s analysis 70. The magistrate’s analysis in her Statement of Findings was as follows: “18. Although I totally believe that it is the accused who drove the car to the scene, this is not the charge he faces. Certainly, this finding of mine is sufficient to convict the accused of the second charge. Paragraph 10.44 of Wilkinson states that ‘use’ must involve an element of controlling, managing or operating a vehicle: ‘‘Use’, under section 143 means that there must be an element of controlling, managing or operating the vehicle as a vehicle.’ Paragraph 10.55 reads: ‘A vehicle is in use on the road when it is stationary and unattended and it must be insured.’ A vehicle parked on the road without anyone inside is also in ‘use’. In the present incident, the headlights and the reading light of KG8680 were on. The accused was seen alighting from the car and locking the doors. He re-opened the car after the police requested for a car search. He was in possession of the car key. These pieces of evidence alone are sufficient to prove that the accused had used KG8680 which was unlicensed. The accused is convicted as charged.” 71. It was not open to the magistrate to convict Cheung on the basis that he had driven the car to Ma Tong Road since that was not the charge he faced. Moreover, as the judge noted, there was insufficient evidence to support a finding of fact that he had done so.[50] Even if Cheung’s evidence was disbelieved, there was simply no evidence that he had driven the car to the place where he was seen to be sitting in it. 72. The magistrate correctly focused on the activities of Cheung when interacting with the car. Her conclusion that the evidence was sufficient to constitute user within section 52(1) will be examined in Section H below. G. The Judge’s analysis 73. The judge framed the issue before him in the following terms in his judgment: “29. … The issue in this appeal is whether the proper interpretation of section 52(1)(a), as submitted by the Appellant, is that the word ‘use’ only applies to the actual driver of the subject vehicle or his employer.” But, with respect, it was wrong to focus on this narrow issue since it ignored the question of whether the actions of Cheung himself were sufficient to come within the proper definition of “use” in section 52(1). 74. Having correctly noted (in paragraph [31] of the CFI Judgment) that the addition of “use” as well as “drive” in the section, the Legislature must have intended that a vehicle can be used otherwise than by driving, the judge wrongly concluded that the word “use” could not apply to anyone other than the driver or his employer. His error in paragraph [31] was to regard the verb “use” as referring to the identity of the actor and this appears to have caused him to lose focus on what really matters which is the quality of action by any person which may or may not constitute “use” within the section. To this extent, his concentration (in paragraph [32] of the CFI Judgment) on the registered owner may have misled him into construing “use” too narrowly by reference to the principle in Windle v Dunning & Son Ltd. As I have noted in paragraph [44] above, there are two distinguishing features between this case and the Windle v Dunning & Son Ltd line of authorities. 75. Ultimately, in paragraph [34] of the CFI Judgment, having found that the appellant did in fact “use” the car, he nevertheless allowed the appeal on the basis that the appellant was not the driver or the employer of the (unidentified) driver. Although he was correct that the evidence did not support a finding that the appellant had driven the car, he wrongly decided the case on the simplistic basis that he did not fall within the definition of “driver or his employer” (paragraph [35] of the CFI Judgment). 76. What the judge should have done was to analyse the actions of Cheung to determine if that activity amounted to using the vehicle in question within the meaning of section 52(1), to which ultimate question I now turn. H. Was the respondent using the vehicle within the meaning of s.52(1)? 77. The findings of fact in the present case are limited to these: Cheung was sitting in the driver’s seat of the car with the headlights and reading light on; he bent down inside the car and looked outside from time to time; he switched off the headlights and alighted from and locked the car; and he walked towards the rear of the car. 78. Although the judge said this evidence “clearly shows” the defendant used the car (see the CFI Judgment at paragraph [34]), it is far from clear how alighting from the car and walking to its rear could constitute such use. All that the evidence shows is that Cheung sat in the driver’s seat when the headlights and reading light were on, that he turned those lights off and alighted and locked the car. He was not the registered owner, there was no evidence he had driven the car to the location on the road in question and no evidence as to the likelihood of his driving it away from that place. The prosecution did not invite the magistrate to find that the placing of his personal property in the boot of the car and other circumstances led to an inference that Cheung had either driven the car to Ma Tong Road or was intending to drive it away from there. 79. Accordingly, applying the definition of “use” set out in Section D.3f above, I would conclude that Cheung’s actions in the context of his status towards the car did not amount to exercising a measure of control, management or operation over the motor vehicle as a means of transport, including any period of time between journeys, so as to fall within the meaning of section 52(1) of the RTO. This case may perhaps be seen as falling close to the line and it is a matter of speculation as to whether, had there been additional evidence of how the vehicle came to be in Ma Tong Road and how it was intended that it be moved from that location, a contrary conclusion might have been reached. Nevertheless, on the evidence as it stands, I would not conclude that Cheung used the car contrary to section 52(1). 80. I would add that the evidence would appear to support the conclusion that Cheung was certainly “in charge of” the vehicle within the meaning of that phrase so that he would bear the burden of establishing the statutory defence: he had the ignition key of the car, the key was inserted in the car at least to the point where the lights were operational, Cheung turned off the lights and locked the car. However, the evidence does not ultimately support a conclusion that he used the car. I. Disposition 81. To the extent that the prosecution’s appeal was limited to seeking guidance on the certified questions of law, the answers that I would give to those questions are set out in Section E above. The appeal against the judge’s order below quashing the conviction in question should otherwise be dismissed. Mr Justice Gummow NPJ: 82. I agree with the judgment of Mr Justice Fok PJ. (Joseph Fok)				Permanent Judge (William Gummow)				Non-Permanent Judge Ms Vinci Lam SADPP and Mr Ivan Cheung PP, of the Department of Justice,	for the Appellant Mr Hectar Pun SC, Mr Peter Yu and Mr Anson Wong Yu Yat, instructed by 	Y.H. Yeung & Associates, assigned by the Director of Legal Aid, for the Respondent [1] Contrary to s.33(1) of the Public Order Ordinance (Cap.245). [2] In TMCC 442/2016 before Magistrate Kelly Shui. [3] In HCMA 271/2016 before Deputy High Court Judge Johnny Chan (“the CFI Judgment”). [4] In FAMC 51/2016 (Ma CJ, Tang PJ and Fok PJ). [5] Appearing with Mr Ivan Cheung PP. [6] Appearing with Mr Peter Yu and Mr Anson Wong Yu Yat. [7] See the recent decision of the Court in HKSAR v Tse Man Fei (2016) 19 HKCFAR 492 at [35] and the references cited there in FN 26. [8] The prohibition in sub-s.(2) is in the same format as sub-s.(1) and relates to the driving or user of a rickshaw on a road. [9] Concerning reg.101 of the Motor Vehicles (Construction and Use) Regulations 1951 which was drafted in terms of “If any person uses or causes or permits to be used on any road …” a vehicle contravening or failing to comply with the relevant regulations. [10] Concerning s.8 of the Road Traffic Ordinance (Cap.220) which, read with reg.100(1) of the Road Traffic (Construction and Use) Regulations (Cap.220), imposed a prohibition against “Any person who uses or causes or permits to be used on any road any vehicle …” not complying with the relevant regulation. [11] Concerning s.8 of the Road Traffic Ordinance (Cap.220) (see FN 10 supra) and reg.101 of the Road Traffic (Construction and Use) Regulations (Cap.220). [12] In Parts II and III of the Road Traffic (Registration and Licensing of Vehicles) Regulations (Cap.374E), made under section 6 of the RTO. [13] S.4(1) of Cap.272 provides that “Subject to the provisions of this Ordinance it shall not be lawful for any person to use, or to cause or permit any other person to use, a motor vehicle on a road unless there is in force in relation to the user of the vehicle by that person or that other person, as the case may be, such a policy of insurance or such a security in respect of third party risks as complies with the requirements of this Ordinance.” [14] RTO s.39A (driving, attempting to drive or being in charge of a motor vehicle with alcohol concentration above prescribed limit), and see also s.39J (driving motor vehicle without proper control under influence of specified illicit drug) and s.39L (driving motor vehicle without proper control under influence of drug other than specified illicit drug). [15] Shorter Oxford English Dictionary (6th Ed.) Vol.1, p.764. The RTO s.2 defines the word “driver” by reference to any person who is in charge of or assisting in the control of a vehicle. [16] [1974] RTR 372 at p.374E, concerning a prosecution under s.99 of the Road Traffic Act 1972 for driving while disqualified. [17] See esp. per Lord Reid at p.1273G-H. [18] See esp. per Lord Hailsham at p.164G-H, Viscount Dilhorne at p.174E-G, Lord Pearson at p.183F-G, Lord Cross at pp.191H-192A and Lord Salmon at p.201B-C (where some of their Lordships questioned the breadth of Lord Reid’s dictum in Pinner v Everett). [19] [1973] 1 QB 748 at p.756F-H. [20] [1992] RTR 121 (a case concerning the question of whether a leaflet placed under the windscreen wiper of a parked car constituted use of a vehicle for a purpose in connection with a trade or business) at pp.123L-124A. [21] Ibid. at p.127D-E. [22] [1970] HKLR 343 at p.352. [23] [1960] 1 QB 367 at pp.371-372. [24] Road Traffic Act 1960, s.64(2), which materially provided “… it shall not be lawful to use on a road a motor vehicle or trailer which does not comply with [the relevant] regulations … and a person who uses a motor vehicle or trailer in contravention of this subsection, or causes or permits the vehicle to be so used, shall be liable on summary conviction …”. [25] (1962) S.C. (J.) 31 at p.46; (1962) SLT 69 at p.75. [26] [1968] 1 WLR 552 at p.556; applied by Lord Parker in Carmichael & Sons Ltd v Cottle [1971] RTR 11 at p.14B-D. [27] [1972] 1 WLR 572 at p.576F-H. [28] [1975] RTR 150 at p.154D-G. [29] [1996] RTR 70 at p.74H. [30] Ibid. at p.76D-F and J-K. [31] Road Traffic Act 1988, s.143(1), equivalent to Motor Vehicles Insurance (Third Party Risks) Ordinance (Cap.272) s.4(1). [32] Concerning r.101 of the Road Traffic (Construction and Use) Regulations, read with s.8 of the Road Traffic Ordinance (Cap.220). [33] In relation to a charge of using a vehicle without third party insurance. [34] This provided inter alia that “it shall not be lawful for any person to use, or to cause or permit any other person to use, a motor vehicle on a road unless there is in force in relation to the user of the vehicle by that person or that other person, as the case may be, such a policy of insurance or such a security in respect of third-party risks as complies with the requirements of this Part of this Act.” [35] [1960] 1 QB 367 at pp.372-373. [36] [1965] 1 QB 1 at p.11F. [37] Ibid. at pp.13E-14E. [38] Ibid. at p.15B-C. [39] Equivalent to the Road Traffic Act 1930, s.35(1). [40] [1997] RTR 212 at pp.217F-218D. [41] Ibid. at pp.224G-225B. [42] Requiring insurance coverage in respect of liability caused by or arising out of “the use of the vehicle on a road or other public place …”. [43] [2017] 3 WLR 450 at [68]. [44] Ibid. at [69]. [45] Followed in Hong Kong in, e.g., HCMA 507/2014, unrep., 8 Dec 2014 at [14]-[16] and HCMA 661/2013, unrep., 14 Feb 2014 at [11]. [46] Under the deeming provision in s.5(3) of the Road Traffic Act 1972; that provision is in similar terms to s.39A(4) of the RTO which provides: “A person is deemed not to have been in charge of a motor vehicle if he proves that at the material time the circumstances were such that there was no likelihood of his driving the motor vehicle while the proportion of alcohol in his breath, blood or urine remained likely to exceed the prescribed limit.” [47] [1989] 1 QB 821 at p.831F-G. [48] Ibid. at p.831B-E. [49] At paragraphs [60]-[61]. [50] CFI Judgment at [35]. 
Mr Justice Ribeiro PJ : 1. At the hearing, the appeal was dismissed without entering into the merits and with the appellants ordered to pay the respondent’s costs on an indemnity basis, for reasons to be provided. These are the Court’s reasons. 2. This appeal arises out of proceedings brought by petition under section 168A of the Companies Ordinance[1] in relation to the conduct of the affairs of C Y Foundation Group Ltd (“CYF”), a Bermudan company registered in Hong Kong and listed on the Hong Kong Stock Exchange. 3. The respondent is the petitioner, Luck Continent Ltd (“LCL”), a BVI company owned and controlled by Dato Poh Po Lian (“Dato Poh”). It was the largest shareholder in CYF, holding 46.58% of the company’s issued shares at the date of the trial before Barma J[2] which began on 28 March 2011. The two individual appellants, Cheng Chee Tock Theodore (“Mr Cheng”) and his wife Leonora Yung (“Mdm Yung”), own and control the 3rd to 8th corporate appellants which together held 25.211% of CYF’s issued shares at the material time. The 9th respondent, CYF, was formally joined as a party but was unrepresented before us and can be ignored. The course of events 4. Dato Poh and Mr Cheng had both been members of CYF’s board, but they fell out during 2009 and Dato Poh resigned as director. He then sought to convene a Special General Meeting of CYF with a view to voting Mr Cheng’s nominees off the board and appointing his own nominees in their place. 5. However, Dato Poh was confronted with bye-law 86(4) of CYF’s Bermudan articles which required any removal of directors before expiry of their term of office to be by way of a special resolution carried by a majority of 75% of the shareholders present and voting. With Mr Cheng and Mdm Yung controlling just over 25% of CYF’s shares, they were in a position to block any such resolutions aimed at removing their nominees from the board or at altering the bye-laws so as to permit removal by an ordinary resolution. 6. The LCL camp noted that bye-law 86(4) was inconsistent with the Stock Exchange’s Listing Rules which stipulated that under the articles of listed companies, an ordinary resolution should suffice for the removal of directors. They expressed concern that failure to bring CYF’s bye-laws into line with that requirement might endanger CYF’s listed status, a concern given some substance by communications from the Stock Exchange. Dato Poh made three attempts during 2010 to have bye-law 86(4) amended by special resolution, but they were all blocked by the votes of Mr Cheng’s camp preventing attainment of the requisite 75% majority. 7. Pursuing a parallel strategy, Dato Poh managed on 26 February 2010 to have the bye-laws changed to increase the permitted number of directors on CYF’s board to 25, his intention being to take control of the board by LCL causing 13 of his nominees to be appointed as directors. However, that plan could only be implemented at the following AGM when the period of office of the existing directors expired. 8. In the meantime, on 14 April 2010, LCL presented the section 168A petition, alleging that by voting their shares so as to block amendment to CYF’s articles and thus preventing it from bringing bye-law 86(4) into line with the Listing Rules’ requirement, the appellants were putting the listed status of CYF at risk and were thereby causing CYF’s affairs to be conducted in a manner unfairly prejudicial to the interests of LCL as a member of the company. 9. On 30 August 2010, CYF’s offices were raided by officers of the Independent Commission Against Corruption and Mr Cheng was subsequently charged with an offence of fraud in relation to a transaction involving CYF. On the following day, 31 August 2010, trading in CYF’s shares was suspended. 10. As CYF’s AGM approached, Dato Poh’s plan to appoint the majority of directors on the board met obstacles involving an attempt to postpone the AGM indefinitely and an attempt to disqualify LCL from voting. After those obstacles were overcome, the AGM was held on 8 April 2011, at which Dato Poh succeeded in gaining control of the board. 11. After the trial which took place between 28 March and 18 April 2011, Barma J delivered judgment on 25 April 2012, holding in favour of the petitioner. He ordered bye-law 86(4) to be amended by replacing the word “special” with the word “ordinary” so that directors could at any time be removed by ordinary resolution. No stay of that order was sought and, on 17 July 2012, the Stock Exchange allowed trading in CYF shares to resume. 12. On 19 October 2012, Barma J ordered costs in favour of LCL save in respect of certain transactions and applications. 13. The Court of Appeal[3] dismissed the appellants’ appeal on 2 August 2013 and ordered them to pay LCL’s costs. 14. On 26 August 2013, the appellants issued a motion seeking leave to appeal which was granted by the Court of Appeal on 27 January 2014 on the basis that two questions of great general or public importance ought to be submitted to this Court. 15. Those questions raise the issue of whether, to what extent and on what basis a shareholder’s exercise of his voting rights in his own interests and in accordance with the provisions of a public company’s articles may be regarded as constituting or resulting in unfairly prejudicial conduct under section 168A. Late revelations 16. On 24 November 2014, two days before the start of this appeal, the Court received a letter from counsel for the respondent[4] indicating that by 21 September 2012, the 1st to 8th appellants had drastically reduced their shareholding in CYF from 25.211% to 1.57% and that while their current shareholding was unknown, it was clearly less than a 5% holding since a disclosure requirement under the Securities and Futures Ordinance would otherwise have been triggered. 17. It was only on 25 November 2014, the day before the start of the hearing of this appeal, that the Court was informed by Mr Victor Joffe, counsel for the appellants, that he had been told by his solicitors late on the previous day that (i) save for a small parcel of 10,000 shares held by the 4th appellant, the appellants had disposed of their shares in CYF in or by September 2012; and (ii) that the bye-laws were amended pursuant to the judgment of Barma J and that trading in the shares had resumed by about 17 July 2012. An academic appeal 18. In the light of those revelations, it is plain that this appeal has become completely academic. As we have seen, the dispute is between two blocks of shareholders struggling for board control. The appellants’ contention has throughout been that they have every right to vote their shares in their own interests, even if that resulted in defeating a resolution which left CYF with a bye-law inconsistent with the requirements of the Listing Rules, thereby possibly endangering its listed status. The respondent’s contention has been that voting their shares with that result justifies relief under section 168A. 19. Dato Poh had gained control of the board on 8 April 2011 and, by Barma J’s order dated 25 April 2012, bye-law 86(4) had been amended to permit removal of directors at any time by ordinary resolution. So at that stage, the legal battle for board control had apparently been won by the Dato Poh camp. And we now know that after trading resumed on 17 July 2012, Mr Cheng’s camp sold off almost all their CYF shares so that by September 2012, they retained no meaningful interest in the company. They effectively no longer had any CYF shares to vote and, anyone would have thought, must have given up the struggle over amendment of bye-law 86(4) and board control. 20. But somewhat astonishingly, without disclosing the loss of their status as holders of more than 25% of the shares, on 30 and 31 May 2013, some eight months after they had disposed of their shareholding, the appellants proceeded to argue the appeal before the Court of Appeal. Then, having lost that appeal, they lodged an application dated 26 August 2013 for leave to appeal to this Court. That application was granted by the Court of Appeal on 27 January 2014, and the hearing of the present appeal was set down for 26 November 2014. 21. It is hard to discern any reason for the appellants to have persisted in the appeals when they no longer had any meaningful interest as shareholders in CYF. Even if the Court of Appeal or this Court had been persuaded to overturn Barma J’s Order and to restore bye-law 86(4) to its original form, it could make no difference to the appellants since they no longer had the shares to block a special resolution and had in any case lost board control. Failure to disclose to the Court 22. It is reprehensible that the position of the appellants and the academic nature of the appeal was not disclosed to the Court of Appeal or, until the last minute, to this Court. While counsel were unaware of the change of status, we were told by Mr Joffe that the solicitors acting for the appellants knew from September 2012 that the appellants had sold their shares. No explanation has been provided for the solicitors failing to ensure that the Court of Appeal was so informed at the hearing of the appeal and when leave to appeal to this Court was sought. We were merely told that non-disclosure was “an oversight”, which is not an acceptable excuse. 23. If the true position had been disclosed, there is every likelihood that the Court of Appeal would have refused to entertain the appeal. It would certainly have refused to give leave for the appeal to progress to this Court. Very significant resources have consequently been expended by the Courts and by the respondent on a matter which is of no practical consequence to the appellants. 24. It is now over five years since the Civil Justice Reforms came into effect. Every legal practitioner must be expected now to be familiar with the underlying objectives listed in Order 1A r 1 of the Rules of the High Court and of the duty imposed by Order 1A r 3 on the parties and their legal representatives to assist the Court to further those objectives. Of particular relevance here are the objectives of ensuring that a case is dealt with as expeditiously as is reasonably possible; of ensuring fairness between the parties; and of ensuring that the resources of the Court are fairly distributed. Non-disclosure of the centrally important change of shareholder status constituted a failure by the appellants and their solicitors in their duty to the Court. Discretion to decide academic appeals 25. Counsel for the appellants acknowledged that the appeal is academic (save for the question of costs) but invited the Court nevertheless to deal with the merits of the appeal. 26. In Po Fun Chan v Winnie Cheung (2007) 10 HKCFAR 676 at §18, the Court’s approach to academic appeals was stated as follows: “Rarely and exceptionally, the public interest in having a particular point of law decided can be so great as to warrant leave to pursue an application or appeal even though the case has become academic as between the immediate parties save perhaps as to costs. Normally, however, such a case should proceed no further. And any issue as to costs should be dealt with on such an appreciation as can be formed on a broad view of the matter. This broad approach avoids further costs. It represents practical justice.” 27. Mr Joffe submitted that great public interest in having the relevant points of law decided exists. He suggested that in finding that the present facts were capable of constituting unfair prejudice, the courts below had erroneously gone beyond well-established limits of the section 168A jurisdiction, making it important for such error to be corrected. He also pointed to the respondent having invited the Court to re-formulate the test for or scope of section 168A relief as a reason for proceeding. 28. We are not persuaded that this is one of the rare and exceptional cases where, though academic, the Court should decide the merits of the appeal. 29. We accept that difficult and interesting questions are raised by the appeal. However, the context is a fact-specific private dispute between shareholders involving an unusual combination of facts, namely, that the company is a publicly listed company; that it happens to have a Bermudan bye-law which is at odds with the Listing Rules’ requirement relating to the removal of directors with possible consequences for the company’s listed status; that the respective shareholdings of the contending parties enabled special resolutions to be blocked by one faction; and that the petition is brought by the largest shareholder in a publicly listed company complaining of unfair prejudice instigated by parties holding a smaller parcel of shares. As Barma J noted,[5] these features make this a highly unusual section 168A case. 30. For these reasons, we dismissed the appeal without entering into the merits. We considered indemnity costs appropriate since leave to appeal would not have been granted if the true position had properly been disclosed. Mr Victor Joffe, Mr Kestrel Lam & Ms Theresa Chow, instructed by Peter K S Chan & Co, for the 1st to 8th respondents Mr. Charles Sussex, SC & Mr. Douglas Lam, instructed by Henry Wai & Co, for the petitioner Attendance of the 9th respondent is excused. [1] Cap 32. [2] HCMP 702/2010 (25 April 2012). [3] Stock VP, Lunn and Lam JJA, CACV 107/2012 (2 August 2013). [4] Mr Charles Sussex SC and Mr Douglas Lam. [5] Judgment §3. 
Chief Justice Ma and Mr Justice Cheung PJ: 1. This appeal raises a question of statutory interpretation, namely of s5(1)(b) of the Country Parks Ordinance[1] and its application in relation to the designation of enclaves as part of country parks in Hong Kong. A. BACKGROUND 2. In 1976, the Ordinance was enacted to provide a legal framework for the designation as well as the control and management of country parks and special areas. The Director of Agriculture, Fisheries and Conservation is designated as the Country and Marine Parks Authority (“the Authority”) in whom is vested the control and management of country parks and special areas.[2] Among other things, it is the duty of the Authority to make recommendations to the Chief Executive for the designation of areas as country parks.[3] The Authority is the respondent in this appeal. 3. The Ordinance also establishes the Country and Marine Parks Board, which is a consultative body to advise the Authority (“the Board”).[4] It also plays an adjudicative role in the statutory designation process of country parks[5] and in the control of the use of leased land within a country park.[6] The Board comprises the Authority and not less than 5 members who are public officers,[7] as well as members who are not public officers, including those representing conservationist interests. Its importance in the scheme of country parks can be seen in the specific reference to this body in the long title of the Ordinance, quite apart from the detailed provisions in it dealing with its operations.[8] 4. Many of the country parks in Hong Kong were designated in the 1970s and over the years, the number of country parks has been increased to 24 and they now cover some 40% of the land areas in Hong Kong. 5. Until 2010, the policy on the designation of country parks was to prefer government land, as designation of land in private ownership could give rise to objections and management problems. As a result, private land together with some surrounding government land were usually excluded from country park boundaries, and thus became “country park enclaves”. As at 2010, there were a total of 77 country park enclaves with a total area of 2,076 hectares. 6. By 2010, owing to the quickened pace of urbanisation, some country park enclaves were facing increasing development pressure. In June that year, unauthorised excavation works were detected within the country park enclave of Sai Wan,[9] triggering significant public concerns on the protection of country park enclaves in Hong Kong. This eventually led to a change of government policy as announced by the Chief Executive in his 2010-11 policy address,[10] in which the government undertook to either incorporate the enclaves into existing country parks, or determine their proper uses through statutory planning under the Town Planning Ordinance[11] to meet conservation and social development needs. This policy change involved 54 of the 77 country park enclaves, the others having already been covered by outline zoning plans made under the Town Planning Ordinance. 7. Pursuant to the 2010 change of government policy, the Authority prepared a working paper dated May 2011 entitled “Review of the Criteria for Designating Country Parks and Proposed Measures for Protecting Country Park Enclaves”[12] for the consultation and advice of the Board. As paragraph 1 of the Working Paper explained, the paper sought the advice of the Board on “the updated principles and criteria for designating new country parks or extending existing country parks”. The paper also proposed “comprehensive measures for protecting country park enclaves.”[13] The paper then discussed in detail the updated principles and criteria proposed for the Board’s advice, noting that, if adopted, “they would be used to assess the merits, justifications and implications of incorporating country park enclaves into the country park boundaries”.[14] The paper also discussed alternative measures to protect country park enclaves under the Town Planning Ordinance or under some existing schemes and policy.[15] Paragraph 5 of the paper was headed “The proposed way forward”. Paragraph 5.1 pointed out that of the existing 77 country park enclaves, 54 were not protected under the Town Planning Ordinance. Paragraph 5.2 went on to say that it was considered necessary to protect the 54 enclaves against incompatible uses by applying appropriate protection measures discussed in the paper. Paragraph 5.3 then stated:- “As regards the appropriate protection measures to be applied, we will, in conjunction with departments concerned, consider whether the enclave is suitable for designation as part of a country park according to the updated principles and criteria for designating new country parks or extending existing country parks set out at Annex 2, or it is more suitable to be protected by including the area into a DPA [Development Permission Area] plan followed by an OZP [Outline Zoning Plan] under the TPO [Town Planning Ordinance]. Actions will then be taken to initiate the respective procedures for applying the relevant protection measures as soon as practicable.” (Emphasis added) 8. The paper ended with paragraph 7, entitled “Advice sought”:- “Members[16] are invited to offer views and comments on the updated principles and criteria for designating new country parks or extending existing country parks, and the proposed measures and way forward for protecting country park enclaves set out in this paper.” The Working Paper is of some importance and is further considered below.[17] 9. At a meeting on 24 May 2011, the Board endorsed in principle the updated principles and criteria for designating new country parks or extending existing country parks set out in the Working Paper. On top of that, it was stated by the Authority that it “would submit the proposed measures for each enclave to the Board for deliberation” and “would take account of members’ views when it assessed the situation of individual enclaves in details [sic] and devised the most appropriate measures”.[18] Following such endorsement, the Authority proceeded to assess in batches the 54 enclaves as regards their suitability for inclusion into the surrounding country parks or alternatively, for protection by other measures. 10. Of those assessed,[19] 6 enclaves[20] were decided by the Authority as not suitable for inclusion into their surrounding country parks, who therefore decided not to recommend them for designation as country park. However, notwithstanding what had been stated at the 24 May 2011 meeting, unlike those enclaves which the Authority decided to recommend for designation, the decisions not to recommend the 6 enclaves were made by the Authority without consulting the Board. The Authority maintained the view that it was not under a duty to do so nor was it necessary to do so. B. PROCEEDINGS BELOW 11. The decisions not to include the 6 enclaves into their neighbouring country parks eventually led to the commencement by the applicant, an environmental activist, of the judicial review proceedings from which this appeal arose. Before the Court of First Instance,[21] the challenge against the Authority’s decision not to consult the Board under s5(1)(b) of the Ordinance failed, but the challenge against the Authority’s assessments and decisions not to recommend designation of the enclaves succeeded on public law grounds with which we are not concerned. In this appeal, we are concerned only with the earlier question of the true construction of s5(1)(b). 12. Section 5(1) of the Ordinance reads:- “(1) There is hereby established a Country and Marine Parks Board which shall — (a) act as a consultative body to advise the Authority upon any matter referred to it by the Authority; (b) consider and to advise the Authority on, the policy and programmes prepared by the Authority in respect ofcountry parks and special areas, including proposed country parks and special areas; and (c) consider any objections that may be lodged under section 11 or section 17.” 13. Au J, who heard the application for judicial review, accepted that under s5(1)(b), the Authority is under a duty to consult the Board so as to enable the Board to consider and advise the Authority on its “policy and programmes in respect of country parks …… .”[22] However, the judge disagreed that the assessments and decisions concerning the 6 enclaves were “policy” or “programmes” in respect of country parks:- “38. First, I accept that the usual meaning of ‘policy’ and ‘programme’ refers to or denotes respectively a course or set of general principles to guide the conducts and plan or scheme that is of general and broad nature. Such meaning is also consistent respectively with the usual meaning of the Chinese words ‘政策’ and ‘計劃’. These words therefore are not intended to include or cover individual one-off decision made in respect of country parks. … 40. However, the individual specific assessment conducted under the 2011 Criteria and decision made not to designate an area as country park are not and cannot be regarded as a ‘policy’ or ‘programme’. 41. Second, I agree that the assessment and decision not to designate an enclave as a country park cannot be a matter ‘in respect of country parks’ or ‘in respect of proposed country parks’. The assessment and decision do not involve a country park or a proposed country park as the enclave involved is by definition not a country park or a proposed park.” 14. On appeal by the applicant, the Court of Appeal[23] essentially agreed with the judge’s reasoning and dismissed the appeal.[24] The court explained what it understood to be the meaning of the “policy and programmes” of the Authority under s5(1)(b):- “In the statutory context of the [Ordinance], we agree with the judge that in Section 5(1)(b) ‘policy’ and ‘programmes’ refers [sic] to formulation of principles (as policy) and the making of plan or scheme (as programme) on a high level of generality. There is a clear distinction between the formulation of policy and programme on the one hand and their actual execution and implementation on the other. Whilst Section 5(1)(b) provides that the Board should consider and give advice on the former, it could not be the legislative intent that the Board, playing an advisory role, should be involved in the execution or implementation of policy or programme.”[25] (Emphasis added) 15. The Court of Appeal concluded that the application of the policy and implementation of the programme set out in the Working Paper in the assessment of each individual enclave was “a pure executive act instead of a policy making or programme planning stage.”[26] 16. The court rejected the applicant’s argument that the assessments should be viewed together to form a programme. It said:- “64. Likewise, the submission of Mr Kat [for the applicant] that each assessment should be viewed together with other assessments to form a programme cannot assist his case. Notwithstanding that these assessments were conducted with the same approach and with the same set of criteria (as they had to be since they were all done pursuant to the criteria set out in the 2011 paper), we do not regard such nexus as providing a basis for asserting that they were a programme within the meaning of Section 5(1)(b). If Mr Kat were correct, all decisions or actions or other executive steps taken under the 2011 Paper can be regarded as programme for which the Authority must seek the advice of the Board under that subsection.” 17. It is to be noted in relation to the proceedings below that both Au J and the Court of Appeal saw the issue raised by the applicant as being whether the individual assessments by the Authority regarding the 6 enclaves could themselves amount to a policy or programme within the meaning of s5(1)(b) of the Ordinance. This was regarded by the Court of Appeal as being “the central issue” in the appeal.[27] As we shall see presently, this was not the correct question to address on the facts of the present case. C. LEAVE TO APPEAL TO THE COURT OF FINAL APPEAL 18. On 12 May 2020, the Appeal Committee[28] gave leave to the applicant to appeal to this court on the following questions, namely:- “(a) On the true construction of section 5(1)(b) of the Country Parks Ordinance (Cap 208), in what circumstances does the Authority come under a duty to consult the Board, and thus to enable the Board to consider and advise the Authority, on the policy and programmes prepared by the Authority in respect of country parks ... including proposed country parks...? (b) In particular, to what extent, if any, does the Authority come under a duty to consult the Board, and thus to enable the Board to consider and advise the Authority, regarding the Authority’s assessments and decisions regarding the suitability or otherwise of designating existing enclaves as country park pursuant to the Authority’s Working Paper WP/CMPB/6/2011 published in May 2011?” These questions require an analysis of the meaning of s5(1)(b) of the Ordinance and its applications to the facts of the present case. D. PARTIES’ ARGUMENTS 19. For the applicant, Mr Nigel Kat SC[29] argued it was unhelpful simply to look at the definitions of “policy and programmes” in s5(1)(b) and then apply them to the facts, this being the approach adopted by Au J and the Court of Appeal. The difficulty with this approach, so it was essentially contended, was that it made the duty on the Board to consider and advise under that provision potentially too wide. Mr Kat pointed to the fact that the Authority has the responsibility for a wide variety of matters in country parks (see the duties on the Authority enumerated in s4 of the Ordinance) and its policies and programmes can accordingly vary greatly in their nature, scope and degree depending on their subject matter. For example, if “programme” were to be construed as a series of planned activities, events or future actions, even if such involved what may appear to be matters of relatively low importance such as country park patrol rotas, this would require the Board to be consulted. On the other hand, a one-off decision over an important topic such as a newly‑discovered coral reef or a marine park might not come within the ambit of s5(1)(b). According to Mr Kat, it was impossible to devise a satisfactory definition for the words “policy and programmes” that would be applicable and workable in all circumstances. 20. Instead, it was contended that rather than construing the specific words contained in s5(1)(b), a different approach should be adopted. Mr Kat argued that one should apply Wednesbury unreasonableness principles to determine whether a proposed measure mandates consultation. A decision maker, in determining whether a s5(1)(b) consultation arises, must ask whether any proposed measure or decision would amount to a policy or programme of “sufficient significance or importance” on which the Board ought to consider and advise. In determining whether any matter was of sufficient significance or importance, the Authority would have to take into account all relevant matters, whilst disregarding all those which are irrelevant; a “multi‑factorial” approach should be adopted. In this regard, reference was made to what Mr Kat called a “Tameside inquiry”.[30] 21. A multi‑factorial approach meant that the Authority had to take the following factors into account: the nature and subject of the proposed measures; the scale or degree of the proposed measures; the likely effect of the proposed measures on the country parks concerned; the relevant expertise available in the Board; the urgency and degree of necessity of the proposed measures; and the public interest in the proposed measures. 22. Adopting this approach, Mr Kat submitted that in relation to the designation of areas as country parks or as part of country parks as in the present case, it was difficult to conceive of circumstances “short of de minimis” in which it would not be sufficiently significant or important to have to consult the Board under s5(1)(b). Accordingly, he submitted that there was a duty in the present case to consult the Board on the suitability of the said 6 country park enclaves for inclusion into their neighbouring country parks pursuant to the Working Paper, subject only to de minimis. 23. Mr Abraham Chan SC,[31] for the Authority, submitted that the words “policy and programmes” in s5(1)(b) had to be construed, but he contended for a narrow construction. Like Mr Kat, Mr Chan’s approach was to try to avoid giving too wide a scope for the Board’s duty to consider and advise under that provision. As we understood them, Mr Chan’s submissions amounted to the following:- (1) The natural meaning of the words “policy” and “programme” was clear. In the respondent’s printed Case, it was accepted that the word “policy” referred to a course or set of general principles, while “programme” referred to a plan or planned series of intended activities, events or future actions.[32] In his oral submissions, however, Mr Chan narrowed his definition of programme to merely a plan or outline of intended activities,[33] rather than the activities themselves. (2) Accordingly, and this was the crux of the Authority’s case, these terms operated on a “high level of generality”. It is to be observed that this was the way the Court of Appeal had construed the relevant words. It was said by Lam VP,[34] “In the statutory context of [the Ordinance], we agree with the judge that in s5(1)(b) ‘policy’ and ‘programmes’ refers [sic] to formulation of principles (as policy) and the making of plan or scheme (as programme) on a high level of generality.” In other words, as Mr Chan puts it, the two terms refer to “framework or schematic matters” or “directive principles”. (3) The Authority argues that this must be so, otherwise the Board will be concerning itself with every detail, however small (as Mr Chan puts it “on nearly every aspect of a given authority’s work”). Reference was made to the factor of time and resources (although there was no evidence in relation to these facets). This would be unworkable. Accordingly, Mr Chan submitted that the Board’s role under s5(1)(b) was restricted to considering and advising on “framework or schematic matters”, certainly as far as programmes were concerned, and not on the intended activities, events or future actions themselves even if they were identified within a programme. (4) A workable balance therefore had to be reached when examining the legislative scheme envisaged by s5(1)(b). (5) Accordingly, in the application of the above analysis, the Authority contended that the individual assessments in relation to the 6 enclaves could not be either a policy or programme or a part of such within the meaning of s5(1)(b) of the Ordinance. In further support of this point, Mr Chan made the following submissions:- (a) In any event, decisions such as the said assessments individually could not linguistically be characterised as a policy or programme “prepared” by the Authority in respect of country parks. (b) The language of s5(1)(b) was to be contrasted with that used for s24(2).[35] While this latter provision specifically required the advice of the Board to be sought before any area could be designated as a special area, this was to be contrasted with the absence of such requirement in s5(1)(b) or anywhere else in the Ordinance. 24. Further on the construction of s5(1)(b), Mr Chan made the following point (a point which was accepted by Au J and which the Court of Appeal did not find it necessary to decide). He contended that in any event, the assessments of the 6 enclaves were not “in respect of” country parks; rather they were in respect of the enclaves. The words “in respect of” require that the principal practice or subject matter of the relevant policy or programme directly concerned country parks or special areas or areas proposed as such. 25. Before addressing these specific issues on the construction of s5(1)(b), we should first restate some relevant general principles and outline the respective roles of the Authority and the Board under the Ordinance. E. PRINCIPLES OF STATUTORY CONSTRUCTION 26. The principles of statutory construction are well established. Words are construed in their context and purpose. They are given their natural and ordinary meaning with context and purpose to be considered alongside the express wording from the start, and not merely at some later stage when an ambiguity is thought to arise.[36] 27. It is, however, important to emphasise that a purposive and contextual interpretation does not mean that one can disregard the actual words used in a statute. To the contrary, the court is to ascertain the intention of the legislature as expressed in the language of the statute. One cannot give a provision a meaning which the language of the statute, understood in the light of its context and purpose, cannot bear.[37] F. RESPECTIVE ROLES OF THE AUTHORITY AND THE BOARD 28. As the long title of the Ordinance tells us, the Ordinance was enacted to provide for the designation, control and management of country parks and special areas, the establishment of the Board, and for purposes connected therewith. The respective roles of the Authority and the Board are clearly set out. 29. The Authority is responsible for the control and management of country parks and its responsibilities and function are set out in ss3 and 4, and further expanded on in other parts of the Ordinance. The Board’s duties are different and are set out in summary in s5 of the Ordinance, again expanded on in other parts of the Ordinance. One of its principal functions is to be a consultative body. The Authority can consult the Board on any matter (s5(1)(a)), but accepts it must consult the Board to consider and advise on policies and programmes prepared by the Authority (s5(1)(b)). One of the questions for determination in this appeal[38] of course focuses on the circumstances when such consultation is mandatory. 30. Before resolving the two Questions in this appeal, we draw attention to an important area of overlap between the function of the Authority and that of the Board in relation to the designation of areas as country parks, this being the particular subject matter of the present appeal. Although Part III of the Ordinance is headed “DESIGNATION OF COUNTRY PARKS”, this is somewhat misleading in that there are relevant provisions to be found in other parts of the Ordinance. One such provision is s4(a), which places the primary responsibility for making recommendations to the Chief Executive on designation on the Authority. The presence of this provision in Part II of the Ordinance (“DUTIES OF COUNTRY AND MARINE PARKS AUTHORITY AND ESTABLISHMENT OF COUNTRY AND MARINE PARKS BOARD”) is significant in that this particular duty of the Authority, alongside the other duties set out in s4, is immediately followed by the role of the Board in s5, including the said duties of consultation set out in ss5(1)(a) and (b). There is a clearly discernible link between the responsibilities of the Authority in relation to the recommendation which may be made in designating (or not designating) areas as country parks and the participation of the Board in this subject matter. 31. The critical question is of course when the participation of the Board becomes compulsory in this regard, but for the time being the only point of emphasis is that the Board can and does have a potentially important role to play as far as the designation of areas as country parks is concerned. There are other relevant provisions. Section 8(4), in Part III of the Ordinance, provides specifically that the Authority shall consult the Board on the preparation of a draft map showing a proposed country park that is recommended by the Authority to the Chief Executive for designation. Further, the Board also has an adjudicative role under s11. However, the participation of the Board under Part III of the Ordinance (as well as that of other entities referred to therein such as the Chief Executive and members of the public) only arises when there has already been a designation of an area as a country park. At the earlier stage of whether or not an area should be designated as a country park, ss4 and 5 are relevant. At this stage, the only relevant persons who are involved are the Authority and the Board. 32. We now address the Questions in the appeal. As mentioned earlier, Question (a) calls for the construction of s5(1)(b), Question (b) is the application of the law to the facts of the case. G. CONSTRUCTION OF s5(1)(b) OF THE ORDINANCE 33. The words “policy and programmes” fall to be determined in particular in this appeal. In our view, these words, whether in English or in Chinese, unqualified and not defined elsewhere in the Ordinance, are everyday words and relatively straightforward to understand. The parties have provided many references to dictionary meanings but it is unnecessary to refer to them. 34. Basically, the word “policy” means a course or set of general principles offering guidance towards an objective or pointing the way towards an objective. It is general or broad in nature. In other words, the word “policy” suggests that it is something standing at a higher level of generality than specific decisions or acts. It functions to guide or direct future actions. The decision that country park enclaves may be considered for inclusion into existing country parks notwithstanding that they involve privately-owned land and the further decision to adopt the updated criteria set out in the Working Paper as assessment criteria for inclusion of country park enclaves into existing country parks exemplify what can amount to a policy in this context. 35. The word “programme” on the other hand is less general or broad in nature, being more specific in terms of what is sought to be achieved. It connotes a plan of action, a project or scheme or, as referred to earlier,[39] a series of planned activities, events or future actions. A programme is a suitable term to employ if it seeks to implement the effect of a policy by a series of intended activities, events or actions. There is no justification artificially to cut down the scope of what a programme can contain and exclude “executive acts” from a programme. For the same reason, it is unhelpful to say categorically that an act to carry out or implement a programme does not form part of the programme, for it begs the question as to what the programme comprises in the first place. Some reliance was placed by Mr Chan on s4(c) of the Ordinance,[40] on the word “measures” to indicate that acts of implementation might not necessarily fall within the words “policy and programmes”. The short answer is that such measures may fall within these words: it depends on what the relevant policy or programme comprises and this will of course depend on the facts. 36. In the case of s5(1)(b), the two words juxtaposed in this way, are not only to be construed as having different meanings but also cumulatively intended to give an indication of the spectrum of matters on which the Board must be consulted. This is the purpose of the provision and it arises in the present case in the context of the designation of areas as country parks. That said, it is important to bear in mind that specific words are used: only policies and programmes prepared by the Authority are targeted, not just any matter that is handled by the Authority. This disposes partly with the concern that the Board’s role is not to micro‑manage the functions of the Authority. However, the operation of s5(1)(b) is clear: as long as a policy or programme prepared by the Authority is involved, the Board must be consulted for its consideration of and advice on that policy or programme. The critical question is always whether the particular matter involved comes within the words policy or programme. 37. We now deal with the parties’ submissions, to which we have earlier referred. 38. As mentioned earlier, Mr Kat argued that the way s5(1)(b) operated and ought to be construed was that the Authority must consult the Board on any matters that have a significant impact or are otherwise important in respect of country parks. A multi‑factorial approach should be adopted. We reject these arguments for the following reasons:- (1) First, it is not a construction based on the wording of s5(1)(b). On the contrary, it is a construction which largely ignores the words “policy and programmes” in s5(1)(b), and is chiefly arrived at by reference to what the applicant perceives to be the purpose of the legislation and intent of the legislature. Under this construction, any “measure” that may have a significant impact or is otherwise important is covered, whereas a “policy” or “plan” that has no significant impact or is not otherwise important is excluded. Such a construction does violence to the words “policy and programmes”, indeed the whole of s5(1)(b). As mentioned, a provision cannot be given a meaning which its language cannot bear. If the legislation wanted to qualify words, it would have done so expressly. (2) Secondly, the suggested criteria of significance and importance must in practice be almost impossible to operate. In the context of country parks, rhetorically speaking, how is a court going to assess whether a matter is significant or important? What may be insignificant or unimportant to some people may be extremely significant or important to another group. The applicant herself makes the point that the Board comprises members from a diverse background and the membership has commonly included naturalists, conservationists and ecologists. There may be members from other backgrounds. (3) Thirdly, the applicant’s construction borrows from the concept of Wednesbury unreasonableness under general public law. There is simply no justification to import this concept into the construction of s5(1)(b). In fact, it confuses the mandatory duty to consult under s5(1)(b) with the discretion to consult under s5(1)(a). Under s5(1)(a), the Authority may, in its discretion, decide to refer any matter to the Board for consultation. The exercise of this statutory discretion is controlled by general public law principles such as those relied on by Mr Kat. For example, if a decision in respect of a country park, including a decision to designate or not designate an area as country park, is likely to have a significant impact or is otherwise important, public law principles may well require the Authority to exercise his discretion to consult the Board, particularly if the subject matter is one within the expertise of the Board,[41] or one on which the Board as a valuable consultative resource is intended by the legislature to have a say. In this way, the concern of Mr Kat that the Authority can bypass the Board and decide on matters of significant impact or otherwise of importance in respect of country parks is taken care of by the general law. However, where the matter comes within s5(1)(b), in other words involving a policy or programme prepared by the Authority, consultation is mandatory. And whether it does come within this provision depends on its true construction, not on concepts such as Wednesbury unreasonableness or a so‑called Tameside inquiry. 39. For his part, Mr Chan is of course correct in his general approach that in determining the effect of s5(1)(b), it is important to construe the words in that provision. However, we cannot accept the limitations that he, following the Court of Appeal, has sought to place particularly on the meaning of the word “programme”. As stated earlier, while the word “policy” is general or broad in nature (and can conceivably come within the words of the Court of Appeal “a high level of generality”), the term “programme” is quite different. It should not be limited to meaning only a plan or outline. There is no support from the purpose and context of s5(1)(b) to justify such a narrow interpretation. On the contrary, the purpose and context of that provision suggests, as we have seen, the importance of the role played by the Board under the Ordinance. It makes no sense to confine the Board’s consideration and advice only to a plan or outline and not to the substance of the activities, events or actions thereunder. 40. Nor are we persuaded that for reasons of time and resources, a so‑called workable balance needs to be reached in the construction of s5(1)(b). Like the submission that a high level of generality should be added as a qualification, this factor of time and resources finds no support or hint in the context of the Ordinance, whether in the body of the statute itself or even in any of the legislative materials. 41. Like Mr Kat, Mr Chan was essentially concerned about the whole system being unworkable in that the Board may become involved in many matters of fine detail. The ready answer to this is that it all depends on whether s5(1)(b) applies. Quite simply, if a policy or programme prepared by the Authority in respect of country parks is involved, the Board must be consulted. This is dependent on whether there is in existence a policy or programme within their definitions. This may or may not involve matters of detail and depends on the facts of any given case. We would also add that although the present case does not involve going into the construction of the words “consider and to advise” in s5(1)(b), it seems that the extent to which the Board may wish to deal with any particular policy or programme within the meaning of that provision would be entirely in its discretion subject to their acting within acknowledged administrative law principles. It is up to the Board and its members to determine the extent of their involvement and their dealing with detail in the course of considering and advising on policies and programmes. 42. Finally in this context, as regards the Authority’s argument based on the words “in respect of”, we see no reason to limit the broad meaning of “in respect of” in the way submitted by the Authority. As Fok PJ explained in Securities and Futures Commission v Pacific Sun Advisors Ltd:- [42] “… The words ‘in respect of’ mean having a relation, connection, reference or regard to something. In Cunard’s Trustees v Commissioners of Inland Revenue, they were described by Lord Greene MR as ‘colourless’. Mann CJ in Trustees Executors & Agency Co Ltd v Reilly said of them: ‘The words ‘in respect of’ are difficult of definition, but they have the widest possible meaning of any expression intended to convey some connection or relation between the two subject-matters to which the words refer.’” 43. The words “in respect of country parks” mean exactly what they say, that the relevant policy or programme prepared by the Authority have something to do with country parks, whether actual or proposed. The term “proposed” envisages a situation that a policy or programme may involve an area which is not yet a country park. H. APPLICATION TO FACTS OF PRESENT CASE 44. Turning to the facts of the present case, the Working Paper,[43] which was prepared by the Authority, clearly sets out a policy using the updated criteria to assess enclaves for the purposes of designation of country parks. But apart from that, it also sets out an action plan to assess all the 54 enclaves not protected under the Town Planning Ordinance for designation as part of a country park by reference to the updated criteria. This was the effect of paragraph 5.3 of the Working Paper. The individual assessments of the 54 enclaves that were to be made, formed part of a plan to implement the policy protecting the 54 enclaves. 45. In our view, this action plan of individual assessments fell within the meaning of programme under s5(1)(b) of the Ordinance. Earlier,[44] we referred to the approach of Au J and the Court of Appeal to inquire whether the various assessments could themselves be regarded individually as policies or programmes. Obviously, they were not policies (and this argument was not maintained by the applicant before us). They were not programmes prepared by the Authority either: they were simply assessments made by the Authority following the said policy. As stated earlier however, the assessments were a part of a programme but they were not programmes individually. 46. As for the “in respect of” argument,[45] the assessment programme was clearly in respect of the neighbouring country parks in that the purpose of the assessments was to decide whether the enclaves should be absorbed into these country parks. Looked at from another angle, the assessments were also in respect of proposed country parks in the sense that the enclaves were to be assessed for designation as country parks. 47. Accordingly, the assessments of the enclaves (which included the relevant 6 enclaves) were a part of a programme within the meaning of s5(1)(b) of the Ordinance and therefore required the Board to be consulted on them. 48. For completeness, we should just finally deal with the point made by Mr Chan in relation to s24(2) of the Ordinance.[46] The fact that there exists an express provision dealing with the role of the Board in relation to the designation of special areas cannot preclude its role in the designation of country parks if this otherwise comes within the terms of s5(1)(b) as properly construed. I. OUTCOME 49. For these reasons, we would allow the appeal, set aside the relevant orders below and make:- (1) an order of certiorari to bring up and quash the Authority’s decision not to consult the Board on the Authority’s assessments and decisions regarding the suitability or otherwise of designating the enclaves at Hoi Ha, Pak Lap, To Kwa Peng, So Lo Pun, Tin Fu Tsai and Pak Tam Au as country park pursuant to the Working Paper; and (2) an order of mandamus requiring the Authority to put before the Board an assessment of the suitability of incorporating each of the 6 enclaves into country parks in accordance with law for its consideration and advice, and to consider the advice of the Board thereon in formulating its recommendations to the Chief Executive for the designation of areas as country parks pursuant to s4(a) of the Ordinance. 50. As for costs, we would make an order nisi that the respondent do pay the appellant’s costs both here and below, such costs to be taxed if not agreed. The appellant’s costs should also be taxed in accordance with the Legal Aid Regulations. If a different order for costs is sought, written submissions should be lodged with the Registrar and served on the other party within two weeks of the handing down of this judgment, with liberty to the other party to serve written submissions in reply within two weeks thereafter. The matter will then be dealt with on paper. If no submissions are lodged or served within the initial two week period, the order nisi will stand absolute. Mr Justice Ribeiro PJ: 51. I agree with the joint judgment of Chief Justice Ma and Mr Justice Cheung PJ. Mr Justice Fok PJ: 52. I agree with the joint judgment of Chief Justice Ma and Mr Justice Cheung PJ. Lord Sumption NPJ: 53. I agree with the joint judgment of Chief Justice Ma and Mr Justice Cheung PJ. Chief Justice Ma: 54. Accordingly, we unanimously allow the appeal, set aside the relevant orders below and make the orders and directions indicated in paragraphs 49 and 50. Mr Nigel Kat SC and Mr Earl Deng, instructed by Vidler & Co., assigned by the Director of Legal Aid, for the Applicant (Appellant) Mr Abraham Chan SC, Mr Keith Lam and Mr John Leung, instructed by the Department of Justice, for the Respondent (Respondent) [1] Cap 208 (“the Ordinance”). [2] Section 3. [3] Section 4(a). [4] Sections 5(1)(a) and (b). [5] Sections 5(1)(c) and 11. [6] Sections 5(1)(c) and 17. [7] Section 5(2). [8] See Sections 5 and 6. [9] An enclave situated on the eastern coast of the Sai Kung peninsula. [10] Dated 13 October 2010 (“Sharing Prosperity for a Caring Society”). [11] Cap 131. [12] Working Paper WP/CMPB/6/2011 (“the Working Paper”). [13] Para 1. [14] Para 3.1. [15] Paras 4.5-4.8. [16] Members of the Board. [17] See para 44 below. [18] Paras 102/11 and 124/11 of the Minutes of Meeting. [19] As at June 2013, 26 enclaves had been assessed. [20] Hoi Ha, Pak Lap, To Kwa Peng, Pak Tam Au, So Lo Pun and Tin Fu Tsai. [21] HCAL 54/2014 (Au J as he then was). [22] Judgment dated 27 April 2017 (CFI Judgment), para 31. [23] Lam VP, Barma and Poon JJA. [24] Judgment of Lam VP dated 14 May 2019 (CA Judgment) [2019] 2 HKLRD 1198. [25] CA Judgment, para 57. [26] CA Judgment, para 61. [27] CA Judgment, para 50. [28] Ma CJ, Ribeiro PJ and Cheung PJ [2020] HKCFA 16. [29] Mr Earl Deng with him. [30] Referring to the principles established in Secretary of State for Education and Science v Tameside Metropolitan Borough Council [1977] AC 1014. [31] Leading Mr Keith Lam and Mr John Leung. [32] Paras 44(1) and 45(3). [33] Relying on a definition in The Shorter Oxford English Dictionary (6th ed, 2007) and on the Chinese version of “programme” in s5(1)(b) (“計劃”). [34] CA Judgment, para 57. [35] Section 24(2) in Part VI of the Ordinance deals with the designation of special areas. It states:- “(2) The Authority may, on the advice of the Board, by order in the Gazette, designate any area of Government land within a country park as a special area for the purposes of this Ordinance.” [36] Town Planning Board v Town Planning Appeal Board (2017) 20 HKCFAR 196, paras 29, 75; HKSAR v Lam Kwong Wai and Anr (2006) 9 HKCFAR 574, para 63. [37] Lam Kwong Wai, para 63. [38] See Question (a) as set out in para 18 above. [39] Paras 19 and 23(1) above. [40] “It shall be the duty of the Authority – … (c) to take such measures in respect of country parks and special areas as he thinks necessary – (i) to encourage their use and development for the purposes of recreation and tourism; (ii) to protect the vegetation and wild life inside country parks and special areas; (iii) to preserve and maintain buildings and sites of historic or cultural significance within country parks and special areas but without prejudice to the Antiquities and Monuments Ordinance (Cap. 53); and (iv) to provide facilities and services for the public enjoyment of country parks and special areas” [41] See Secretary of State for Education and Science v Tameside Metropolitan Borough Council [1977] AC 1014. [42] (2015) 18 HKCFAR 138, para 23. [43] See paras 7 and 8 above. [44] See para 17 above. [45] See paras 24, 42 and 43 above. [46] See para 23(5)(b) above. 
Mr Justice Ribeiro PJ: 1. The Court’s reasons for the result announced at the conclusion of the hearing are now given by Mr Justice Bokhary NPJ. Mr Justice Bokhary NPJ: 2. For the reasons given in Sky Heart Ltd v. Lee Hysan Co. Ltd (1997-98) 1 HKCFAR 318 and repeated in Chinachem Charitable Foundation Ltd v. Chan Chun Chuen (2011) 14 HKCFAR 798, this Court’s practice, in common with that of the courts of last resort of comparable legal systems, is not to review concurrent findings of fact save in rare and exceptional circumstances. This practice springs from the nature of a court of last resort’s role, which is primarily of resolving points of law of the highest importance. It is to be added that the practice also serves to spare litigants from the oppression of unduly protracted litigation. 3. When asked to review such findings, we may begin by hearing argument on the question of whether it would be purposeful to embark upon such a review. That is a procedure which we have employed in the past, and it was employed on the present occasion. Each of the present appeals was dependent for its success on the reversal of a concurrent finding of fact that 83.1% of the shares in New Asia International (Group) Ltd, a BVI company (“the Shares”) belong, as the wife said, to the husband rather than, as the husband and the woman with whom he is cohabiting said, to the cohabitee. At the conclusion of the argument on whether it would be purposeful to embark upon a review of this concurrent finding of fact, we decided that it would not. 4. That left these appeals (FACV No. 11 of 2013 brought by the cohabitee and FACV No. 19 of 2013 brought by the husband) without any ground on which to proceed. Both were therefore dismissed with costs, it being ordered that the husband and the cohabitee be jointly and severally liable to the wife for her costs of each appeal. A stay ordered by the Appeal Committee when granting the husband leave to appeal to this Court was lifted. 5. These are the Court's reasons for not considering it purposeful to embark upon a review of the concurrent finding of fact that the Shares belong to the husband. 6. Both of these appeals arose out of the wife's application for ancillary relief. If the Shares belong to the husband they would be taken into account in determining the award to the wife of ancillary relief, increasing it. So the ownership of the Shares was tried as a preliminary issue in the wife's application for ancillary relief, in which application the cohabitee became an intervener for the purposes of the preliminary issue. At first instance, Saunders J found as a fact that the Shares belong to the husband, thus deciding the preliminary issue in favour of the wife and against the husband and the cohabitee. On appeal to the Court of Appeal by the husband and the cohabitee, the Court of Appeal (Cheung, Yuen and Chu JJA) affirmed the finding of fact that the Shares belong to the husband, thus making that finding of fact a concurrent one. 7. This concurrent finding of fact was attacked by counsel for the cohabitee and counsel for the husband. 8. It was suggested that there was no evidence on which to find that the Shares belong to the husband. As to that, suffice it to say that there was evidence on which to so find. Indeed, it could be said, although it is unnecessary to say, that there was ample evidence on which to so find. 9. Then the second line of attack ran along the following lines. The Shares represent 83.1% of the shares in the company concerned. Saunders J thought that the wife’s stance was that only that percentage of the shares in the company concerned belonged to the husband. In fact, her stance was that a higher percentage of the shares in the company concerned belonged to him. As to that, all that has to be said is this. Whether a misapprehension of that sort vitiates a finding such as the one here in question depends on the circumstances. And in the present circumstances, it is plain that the misapprehension does not vitiate the finding. 10. Finally, it was said that references made by Saunders J to the proceedings before him being of an inquisitorial nature indicates that he had approached the question of share ownership as a matter of discretion. Those references do not indicate anything of the sort. Moreover, Saunders J said in terms that the question of whether the Shares belong to the husband was “a simple issue of fact to be determined on the balance of probabilities, with the wife, who seeks to assert that the shares are owned by the husband, carrying the burden of proof to establish that fact”. The appellants were unable to point to any finding in the judgment which was based on any other approach. Throughout his consideration of the evidence, Saunders J never approached the question of share ownership otherwise than as a question of fact. Mr Russell Coleman SC and Mr Robin Egerton, instructed by Mayer Brown JSM, for the Respondent (Appellant in FACV 19/2013 and 2nd Respondent in FACV 11/2013) Mr Paul Lam, SC and Mr Vincent Lung, instructed by Hom & Associates, for the Intervening Party (Appellant in FACV 11/2013 and 2nd Respondent in FACV 19/2013) Mr Charles Sussex SC and Mr Neal Clough, instructed by King & Company, for the Petitioner (1st Respondent in FACV 11 & 19/2013) 
Mr Justice Ribeiro PJ: 1. This appeal concerns an application by a litigant in matrimonial proceedings to be allowed to disclose to the Director of Public Prosecutions (“the DPP”) documents obtained by her on discovery, against objections based on legal professional privilege (“LPP”) raised by the parties who had given the discovery. At the hearing, the Court allowed the appeals by the parties objecting, with reasons to be given later. These are our reasons. A. The course of events 2. Florence Tsang Chiu-wing (“the Wife” or “W”) and Samarthur Li Kin-Kan (“the Husband” or “H”)[1] were married on 8 January 2000. The proceedings with which we are concerned ensued after their marriage broke down in 2008. 3. Samuel Tak Lee (“STL”) is H’s father and a man of great wealth with vast property holdings. STL assisted H, both financially and by using his business expertise, to acquire a portfolio of valuable real estate developments in Tokyo. The first property, known as the V28 building, was acquired by H on 19 March 1999, funded by a gift of ¥3.6 billion from STL, but the gift was made subject to an agreement referred to below as “the Framework Agreement” which gave STL the option to acquire the shares in the companies holding the property for a nominal sum. The Framework Agreement also required “Special Articles” to be written into the corporate constitutions restricting disposals, borrowings and other dealings with the corporate assets without STL’s prior written consent. Between 1999 and 2008, a total of 13 Tokyo properties were acquired by H with STL’s help. 4. H and W separated in January 2008 and, on W’s petition, they were divorced with the decree nisi issued on 25 August 2008 and the decree absolute granted on 5 April 2013. 5. In the ancillary relief proceedings which followed, a major issue between them concerned the extent of H’s assets available for distribution. This raised questions regarding the nature and extent of STL’s interest in the Japanese assets acquired by H. 6. On 4 November 2009, H and STL executed certain documents prepared by solicitors for STL, purporting to transfer the legal and beneficial interest in the Japanese assets from H to STL which, if valid, would have depleted the matrimonial assets available for distribution in the ancillary relief proceedings. The November 4 transactions were purportedly effected pursuant to the Framework Agreement and a convertible loan agreement dated 9 February 2006 (referred to below as “the 2006 CLA”). B. The relevant proceedings below 7. W, however, alleged that the 2006 CLA was a recent forgery. She instituted proceedings under section 17 of the Matrimonial Proceedings and Property Ordinance[2] to set aside the dispositions effected by the November 4 transactions. 8. It was in aid of those section 17 proceedings that, on 20 December 2010, W obtained orders for discovery from Saunders J in relation to the November 4 transactions and the 2006 CLA. H and STL resisted disclosure of some of those documents on the basis that they were protected by LPP but, after hearing full argument, Saunders J held that they fell within the rule in Cox and Railton[3] and were therefore not so protected. 9. The rule in Cox and Railton, as Stephen J put it, is that communications protected by legal professional privilege “cannot include the case of communications, criminal in themselves, or intended to further any criminal purpose”.[4] Explaining the principle that “confidential communications involving fraud are not privileged from disclosure”, Stephen J cited Lord Hatherley in Gartside v Outram[5] who stated: “The true doctrine is, that there is no confidence as to the disclosure of iniquity. You cannot make me the confidant of a crime or a fraud and be entitled to close up my lips upon any secret which you have the audacity to disclose to me relating to any fraudulent intention on your part; such a confidence cannot exist.” 10. Thus, the documents for which LPP had been claimed by H and STL came into W’s hands as a result of Saunders J’s Cox and Railton ruling, enabling her to use them for the purposes of the section 17 proceedings. On 23 February 2011, the section 17 proceedings (and certain other proceedings) were settled. While not admitting forgery, H and STL essentially capitulated. They did not oppose an order setting aside the documents executed on 4 November as well as the 2006 CLA. W was awarded indemnity costs. 11. W then applied to Saunders J to be allowed to use the discovered documents at the substantive ancillary relief hearing with a view to contending that the 2006 CLA was a forgery and that H’s reliance on it amounted to “conduct” to be taken into account by the Court in exercising its ancillary relief powers under MPPO section 7. Her application was refused, but Saunders J ruled that: “... the conduct of the husband and the father leading to, and including, the execution of the 2006 CLA was relevant to their credibility in the trial of the ancillary relief proceedings and that the husband and the father could be cross-examined on those matters ‘with the usual limitations that are imposed upon cross-examination on credit’.”[6] 12. In what has been referred to as his “Main Judgment” dated 1 December 2011[7], Saunders J made financial provision awards, subsequently varied on appeal, which do not presently concern us.[8] However, despite his earlier ruling limiting the use of the relevant documents to cross-examination as to credit, Saunders J in his judgment: “... relied on and drew inferences from contemporaneous documents in Attachment 2 and bundle SP1 in reaching his conclusion on the allegation of forgery . He found on the civil standard of proof that the husband and the father had forged the 2006 CLA which they had then used to attempt to defeat the wife’s claim for ancillary relief. He held that this document, which was dated 9 February 2006, was not signed by them on that day but was executed on a date subsequent to the breakdown in the marriage and was backdated and used as though it were executed on that day. He further held that the husband and the father were likely to have committed perjury in relation to the 2006 CLA in numerous affidavits in the ancillary relief proceedings.”[9] 13. Saunders J held that for the purposes of the ancillary relief proceedings, H and STL were persons whose evidence was not capable of belief. He directed the Registrar to forward a copy of his Main Judgment to the DPP.[10] 14. Following issue of the Main Judgment, H and STL applied to Saunders J to alter that judgment pursuant to what was called the Barrell jurisdiction whereby the court in exceptional circumstances, may agree to reconsider and reverse its own conclusions before its order is perfected.[11] They applied to the Judge to delete his findings of forgery against H and STL and to rescind his direction that the judgment be forwarded to the DPP. In what has been referred to as his Barrell Judgment,[12] Saunders J refused that application. 15. In the light of the findings his Lordship had made as to forgery, perverting the course of justice and perjury, the Secretary for Justice (“the SJ”) issued an Originating Summons applying to the Court[13] for the DPP to be given access to the relevant documents in aid of a criminal investigation. On 7 June 2012, on the application of STL and H, Saunders J recused himself from hearing the SJ’s application and the hearing proceeded before Ng J who, on 14 May 2013, granted the SJ access to the documents sought and released W from her implied undertaking in relation to the relevant documents for the purpose of: “ (a) disclosing to the [SJ] and the Hong Kong Police such of the Documents as may be required in the course of the criminal investigation, and, specifically, all documents and correspondence relating to the [2006 CLA]; and (b) discussing [those] documents ... with the [SJ] and the Hong Kong Police in the course of the criminal investigation.” 16. Those orders were stayed pending appeal and on 10 January 2014, the Court of Appeal handed down two judgments, one regarding Saunders J’s ancillary relief orders[14] and the other on the disclosure orders made by Ng J.[15] In the ancillary relief judgment, the Court of Appeal set aside the Judge’s findings on forgery and perjury, holding that they had not been properly made since the documents that he had relied on were not in evidence before him.[16] 17. Relevantly for present purposes, the Court of Appeal reversed Ng J’s order granting the SJ access to the documents and held that, in respect of documents in respect of which LPP was being claimed by H and STL, the DPP could not simply rely on Saunders J’s Cox and Railton ruling obtained by W, but would himself have to establish the applicability of the rule before he could be granted access to such documents. The question of LPP was accordingly remitted to be decided by a Judge of the Court of First Instance. That order has not been appealed and the remitter will be heard in due course. 18. However, the Court of Appeal considered W to be in a different position since: “...vis-à-vis [W], [H] and STL are bound by the rulings of Saunders J and they could no longer assert any LPP against her in respect of documents disclosed pursuant to the order of Saunders J.”[17] 19. Taking the view (i) that the public interest in upholding the integrity of the administration of justice was a discretionary factor of great weight; (ii) that release from the undertaking would not deter litigants from giving full and frank discovery since there is “no nexus between the duty to give full and frank discovery and the commission of perjury and using forged documents for the purpose of legal proceedings”; and (iii) that there was no injustice to H and STL since they had fully been able to argue for LPP before Saunders J; the Court of Appeal concluded that the balance overwhelmingly favoured upholding Ng J’s order releasing W from her implied undertaking.[18] C. Leave to appeal 20. The Court of Appeal having refused leave, the Appeal Committee granted leave to appeal confined to the following question: “Whether the documents in relation to which the [SJ] was granted access and in respect of which [W] was released from her implied undertaking by Order of Ng J dated 14 May 2013, to the extent of the documents for which legal professional privilege is claimed as identified by [H] and STL in affidavits to be filed within 21 days of today pursuant to the direction in paragraph 121(b) of the Court of Appeal’s Disclosure judgment dated 10 January 2014, may be disclosed to the [SJ] in accordance with the Order of Ng J without first determining the question whether they are protected by legal professional privilege.” D. Determination of this appeal 21. It was common ground at the hearing that as between W on the one hand and H and STL on the other, LPP could not be asserted in respect of the disputed documents since Saunders J had held that they were not privileged and his Cox and Railton ruling was not subject to any appeal. 22. However, that did not mean that W was free to use those documents as she wished. It has long been established that a party who obtains documents on discovery gives an implied undertaking to the Court “that he will make use of them only for the purposes of that action, and no other purpose.”[19] In Crest Homes PLC v Marks,[20] Lord Oliver of Aylmerton, speaking in the context of materials obtained by solicitors in Anton Piller raids, explained the scope of the rule in the following terms: “It is clearly established and has recently been affirmed in this House that a solicitor who, in the course of discovery in an action, obtains possession of copies of documents belonging to his client's adversary gives an implied undertaking to the court not to use that material nor to allow it to be used for any purpose other than the proper conduct of that action on behalf of his client: see Home Office v Harman [1983] 1 AC 280. It must not be used for any ‘collateral or ulterior’ purpose, to use the words of Jenkins J in Alterskye v Scott [1948] 1 All ER 469, approved and adopted by Lord Diplock in Harman's case, at p 302. Thus, for instance, to use a document obtained on discovery in one action as the foundation for a claim in a different and wholly unrelated proceeding would be a clear breach of the implied undertaking: see Riddick v Thames Board Mills Ltd [1977] QB 881. It has recently been held by Scott J in Sybron Corporation v Barclays Bank Plc [1985] Ch 299 - and this must, in my judgment, clearly be right - that the implied undertaking applies not merely to the documents discovered themselves but also to information derived from those documents whether it be embodied in a copy or stored in the mind. But the implied undertaking is one which is given to the court ordering discovery and it is clear and is not disputed by the appellants that it can, in appropriate circumstances, be released or modified by the court.” 23. Where release from the undertaking is sought, it is for the person who obtained the documents “to demonstrate cogent and persuasive reasons why it should be released.”[21] Each case turns on its own facts and no general principle can be formulated beyond stating: “...that the court will not release or modify the implied undertaking given on discovery save in special circumstances and where the release or modification will not occasion injustice to the person giving discovery.”[22] 24. W argued that in the present case, a cogent and persuasive reason to release her from her implied undertaking is the public interest in facilitating the reporting of crime by an alleged victim and the public interest in the proper investigation of suspected crime. Mr Charles Howard QC,[23] submitted that her position was quite separate from that of the SJ so that she should be permitted independently to supply the DPP with the documents and to discuss them with him without further delay. He submitted that the SJ’s application was irrelevant so far as his client’s undertaking was concerned. 25. We do not accept that the two applications can or should be viewed as wholly separate. They both concern the DPP’s access to documents originating from the same source: from discovery by H and STL. The DPP has had sight of Saunders J’s Main Judgment[24] but he has not got the documents referred to, although some excerpts from them appear in the body of the judgment. The Court of Appeal has held and the DPP rightly recognizes that some of the documents sought are prima facie protected by LPP and that the DPP must therefore bring himself within the Cox and Railton principle at the pending remitter hearing if he is to gain access to them. It makes no difference that the DPP may have read extracts from the documents said to be protected. As Lord Millett pointed out in B v Auckland District Law Society: “… unless waived ‘once privileged, always privileged’”. [25] Rejecting the Law Society’s metaphor of losing LPP because of “letting the cat out of the bag”, Lord Millett stated: “...The policy which protected them from unauthorised disclosure is the same. The cat is still a cat. It can be put back in the bag.”[26] 26. It is impossible to give proper consideration to the Court’s exercise of discretion regarding release of W from her implied undertaking without taking into account what the effect of doing so would be on H’s and STL’s claim to LPP as against the DPP. 27. It has repeatedly been stressed that LPP is a fundamental right which the courts will jealously protect. As Lord Hoffmann reiterated in R (Morgan Grenfell & Co Ltd) v Special Commissioner of Income Tax:[27] “...LPP is a fundamental human right long established in the common law. It is a necessary corollary of the right of any person to obtain skilled advice about the law. Such advice cannot be effectively obtained unless the client is able to put all the facts before the adviser without fear that they may afterwards be disclosed and used to his prejudice. The cases establishing this principle are collected in the speech of Lord Taylor of Gosforth CJ in R v Derby Magistrates' Court, Ex p B [1996] AC 487. It has been held by the European Court of Human Rights to be part of the right of privacy guaranteed by article 8 of the Convention (Campbell v United Kingdom (1992) 15 EHRR 137 ; Foxley v United Kingdom (2000) 31 EHRR 637 ) and held by the European Court of Justice to be a part of Community law: A M & S Europe Ltd v Commission of the European Communities (Case 155/79) [1983] QB 878 .” 28. In Hong Kong, LPP is constitutionally guaranteed by the confidential legal advice clause in Article 35 of the Basic Law,[28] as this Court noted in Akai Holdings Ltd v Ernst & Young (a Hong Kong firm).[29] 29. While an exercise of balancing competing interests is required in deciding whether someone should be released from the implied undertaking, it is well established that “LPP does not involve such a balancing of interests. It is absolute and is based not merely upon the general right to privacy but also upon the right of access to justice”.[30] 30. As between the SJ and H and STL, the issue of whether the documents sought are protected by LPP has not yet been determined. As Viscount Findlay made plain in O’Rourke v Darbishire:[31] “... it is not enough to allege fraud. If the communications to the solicitor were for the purpose of obtaining professional advice, there must be, in order to get rid of privilege, not merely an allegation that they were made for the purpose of getting advice for the commission of a fraud, but there must be something to give colour to the charge. The statement must be made in clear and definite terms, and there must further be some prima facie evidence that it has some foundation in fact.” 31. In our view, it would be wrong in principle to release W from her implied undertaking with the consequence of enabling her to provide to the SJ and DPP, documents for which LPP is claimed and not yet ruled upon by the Court. But it is not simply a question of directing W to await the final outcome of the SJ’s application for access to the disputed documents. Of course, if the SJ were to succeed in obtaining access to the same documents as W obtained, no practical difficulty would arise. However, it is plainly possible that the outcome of his pending application might involve the DPP failing to obtain access at all or obtaining a narrower range of documents than those obtained by W. In such a case, it would be highly incongruous for W to be released from her undertaking and thus permitted to provide to the DPP documents which the Court has held were protected as against him by LPP. To allow the Court’s ruling to be side-stepped in this way would be to undermine the Court’s decision and to bring the administration of justice into disrepute. 32. We should add that although Mr Howard QC submitted that there was nothing objectionable about W supplying the DPP with the documents and discussing them with him whatever the outcome of the pending Cox and Railton application, Mr Keith Yeung SC, the DPP, dissociated himself from that submission. He told the Court that he would have to think long and hard before accepting any documents from W to which the Court had denied him access, but he did not altogether rule out the possibility that he might do so. 33. In our view, it would not be proper for the DPP to gain access to any documents denied him by the Court by the simple expedient of receiving them from W. There can be no question of releasing W from her implied undertaking to enable her to achieve that purpose. Nor is there any question of the DPP being permitted to make derivative use of documents which are protected by LPP in his hands, the privilege being absolute. 34. In our view, no cogent argument has been advanced for releasing W from her undertaking at all. Holding W to her undertaking in the present case does not mean that the reporting of an alleged crime will be stifled. The Judge has made such a report by passing a copy of the Main Judgment to the DPP. If and to the extent that the DPP succeeds in bringing the documents concerned within the Cox and Railton rule, there would be no impediment to him using those documents in the criminal investigation. He could discuss them with W without her having to be released from her undertaking. It is appropriate that the availability of such documents for use in a criminal investigation should depend on the DPP establishing his entitlement to such use and not on W being relieved of her implied undertaking, regardless of the outcome of the DPP’s application. 35. For the foregoing reasons, we allowed the appeal. We make the following orders (those relating to continued stays being made at the parties’ requests), namely: (a) That the stay imposed by paragraph 9 of the Order of the Appeal Committee dated 23 April 2014 regarding implementation of the directions of the Court of Appeal in paragraph 121 of their judgment in CACV 101 and 107/2013 dated 10 January 2014, be continued until 14 days after the date of the handing down of this judgment or further order, such continued stay to expire unless otherwise directed. (b) That paragraph 2 of the Order of Ng J dated 14 May 2013 releasing W from her implied undertaking generally and paragraph 2 of the Order of the Court of Appeal in CACV 101/2013 dated 10 January 2014 dismissing the appeal against Ng J’s aforesaid Order be set aside. (c) That, subject to sub-paragraph (d) below, W be released from her implied undertaking only in respect of those documents obtained by her on discovery other than those referred to in the affidavits claiming LPP filed by H (exhibited as an undated affidavit to the affidavits of Por Keng Guan Catherine dated 14 May 2014 and 26 May 2014); and by STL dated 12 May 2014 (exhibited to the Affidavit of Jamie John Stranger dated 26 May 2014) in accordance with paragraph 1 b. of the Order of the Court of Appeal in CACV 101/2013 dated 10 January 2014 and paragraph 9 of the Order of the Appeal Committee dated 23 April 2014. (d) That the stay imposed by paragraph 10 of the Order of the Appeal Committee dated 23 April 2014 on implementation of paragraph 2 of Ng J’s aforesaid order in relation to the release of W from her implied undertaking be continued until 14 days after the date of the handing down of this judgment or further order, such continued stay to expire unless otherwise directed. (e) The stay granted by the Appeal Committee by paragraph 11 of their aforesaid Order dated 23 April 2014 be continued until 14 days after the date of the handing down of this judgment or further order, such continued stay to expire unless otherwise directed. 36. We make the following order nisi as to costs, namely, That W pay to H and STL :- (a) the costs of and occasioned by this appeal and by the application for leave to appeal; and (b) the costs in the Courts below in relation to W’s application to be released from her implied undertaking in respect of documents for which LPP is claimed and which are subject to determination by the Court of First Instance. (c) That there be no Order as to costs in relation to the SJ. 37. The parties have liberty to make submissions in writing, within 14 days from the handing down of this judgment, in relation to the terms and implementation of the aforesaid orders. In the absence of any such submissions, the order nisi as to costs shall stand as an order absolute without further direction. In the event that a hearing is required in respect of the aforesaid orders, we direct that such hearing be before a single Permanent Judge. Mr Edward Fitzgerald QC, Mr Peter Duncan SC and Mr Derek C.L. Chan, instructed by Stephenson Harwood for the appellants in FACV 5/2014 (3rd and 4th defendants) Mr Collingwood Thompson QC, Mr Russell Coleman SC and Mr Jeremy S.K. Chan, instructed by Stevenson, Wong & Co., for the appellant in FACV 6/2014 (2nd defendant) Mr Charles Howard QC and Ms Lorinda Lau, instructed by Florence Tsang & Co. for the 1st respondent in FACV5/2014, 2nd respondent in FACV 6/2014 (1st defendant) Mr Keith Yeung SC, DPP, and Mr Derek Wong, SPP, of Department of Justice for the 2nd respondent in FACV5/2014, 1st respondent in FACV 6/2014 (plaintiff) [1] The parties informed the Court that they have no objection to being referred to by their names. W was referred to below as the 1st defendant or “FTCW”, while H was referred to as the 2nd defendant or “SLKK”. H’s father was referred to as the 3rd defendant or “STL”. [2] “MPPO” Cap 192. Section 17(1)(b): Where proceedings for relief under any of the relevant provisions of this Ordinance ... are brought by a person ... against any other person ... , the court may, on an application by the applicant- if it is satisfied that the other party has, with the intention [of defeating the claim for financial provision], made a disposition to which this paragraph applies and that if the disposition were set aside financial provision or different financial provision would be granted to the applicant, make an order setting aside the disposition and give such consequential directions as it thinks fit for giving effect to the order (including directions requiring the making of any payment or the disposal of any property). [3] (1884) 14 QBD 153. [4] At 167. [5] 26 L J (Ch) 113, cited at (1884) 14 QBD 153, 169. [6] Court of Appeal Lam VP, Kwan and Barma JJ (ancillary relief judgment) CACV 154 and 166/2012 (10 January 2014), §26, referring to Saunders J, Main Judgment §§76 and 77. [7] Saunders J, Main Judgment, HCMC 5/2008 (1 December 2011). [8] We were informed that applications for leave to appeal to this Court are pending. [9] Court of Appeal (ancillary relief judgment) §27, referring to Saunders J, Main Judgment §§87, 90-127, 130-33, 136 and 137. [10] Saunders J, Main Judgment, §514. [11] Referring to Re Barrell Enterprises [1972] 3 All ER 631; considered in Stewart v Engel [2000] 3 All ER 518; and applied in Hong Kong Sun Jianqiang v Trans-Island Limousine Service Ltd [2004] 1 HKC 533. [12] Saunders J, Barrell Judgment, HCMC 5/2008 (6 July 2012) §§37-92. [13] HCMP No 188 of 2012. [14] Lam VP, Kwan and Barma JJ, CACV 154 & 166/2012 (10 January 2014). [15] Lam VP, Kwan and Barma JJ, CACV 101 & 107/2013 (10 January 2014). [16] At §§71-77. [17] CACV 101 & 107/2013 (10 January 2014), §139. [18] Ibid, §§141-146. [19] Riddick v Thames Board Mills Ltd [1977] QB 881 at 896. [20] [1987] 1 AC 829 at 853-854. [21] Ibid at 859. [22] Ibid at 860. [23] Appearing with Ms Lorinda Lau for W. [24] We were informed that a challenge has been mounted to the correctness of an un-redacted version having been provided to the DPP. [25] [2003] 2 AC 736 at §44. [26] Ibid at §69. [27] [2003] 1 AC 563 at §7. [28] Basic Law Article 35: “Hong Kong residents shall have the right to confidential legal advice, access to the courts, choice of lawyers for timely protection of their lawful rights and interests or for representation in the courts, and to judicial remedies. ...” [29] [2009] 12 HKCFA 649 at §67. [30] Lord Hoffmann in R (Morgan Grenfell & Co Ltd) at §16, citing R v Derby Magistrates' Court, Ex p B [1996] AC 487. See also Lord Millett in B v Auckland District Law Society [2003] 2 AC 736 at §§48 and 50, referring also to New Zealand and Australian authority: R v Uljee [1982] 1 NZLR 561, 576; and Carter v Northmore Hale Davy & Leake (1995) 183 CLR 121 at 128. [31] [1920] AC 581 at 604. 
Mr Justice Chan PJ: 1. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Ribeiro PJ: 2. The appellant, a barrister, was charged with attempting to pervert the course of justice on account of his having drafted three letters and caused them to be sent to a prosecution witness in the course of the trial at which he was appearing for the defence. The witness was a solicitor and the letters asserted that she was bound by legal professional privilege to refuse to provide the evidence which the prosecution sought to adduce from her. The letters threatened civil proceedings and a complaint to the Law Society if she failed to comply with the asserted duty. 3. The appellant (“Mr Wong”) was convicted after trial before H H Judge Tallentire in the District Court[1] and sentenced to six months’ imprisonment. His conviction and sentence were upheld by the Court of Appeal,[2] although on different grounds. The Appeal Committee granted leave to appeal, this being a case which raises important questions concerning the interaction between the offence of attempting to pervert the course of justice and a claim of legal professional privilege asserted against an intended witness. A. The course of events A.1 The proceedings against Ms Lau 4. In September 2001, KBC Bank NV, Hong Kong Branch (“the bank”) agreed to lend money to two “Maruyama” companies[3] secured on floating charges provided by them and also secured on a fixed and floating charge over the undertaking and property of E C Textiles Ltd (“ECT”), a company in the same group. ECT had a corporate director which was Brightway International Group Ltd (“Brightway”) and Ms Lau Wing Sun (“Ms Lau”) was a director of Brightway. It was Ms Lau who executed the deed of charge (“the deed”) on behalf Brightway and thus of ECT. It was this act which led to her being prosecuted for theft, it being alleged that she had purported to pledge to the bank property which she had no right to pledge and so had dishonestly appropriated such assets. 5. Ms Lau signed the deed at the offices of Messrs Christine M Koo & Co (“CMK&Co”) a firm of solicitors at which Ms Fanny Mak Siu-fan (“Ms Mak”) was employed as a consultant. Ms Mak was instructed by the bank in the transaction and witnessed Ms Lau’s signing of the deed. The letters subsequently drafted by Mr Wong were addressed to Ms Mak and an important issue in the case is whether she was impliedly acting also as solicitor for ECT when the deed was executed. 6. Ms Lau was arrested in February 2004 and, in November of that year, Mr Wong was instructed by Messrs Lau, Chan & Ko (“LC&K”) to act on her behalf in connection with the criminal investigation and proceedings. Two other barristers were also instructed on Ms Lau’s behalf, namely, Mr Lawrence Ngai (“Mr Ngai”) and Mr Paul Tse (“Mr Tse”). The solicitor on the record for LC&K was Mr Birney Yeung Wai (“Mr Yeung”) although it was a solicitor’s clerk, Mr Jacob Cheung, who actually handled the case. 7. The police sought evidence that it was Ms Lau who had signed the deed and in February 2005, they obtained two witness statements from Ms Mak in which she acknowledged that she had acted for the bank in the transaction. She stated that she had prepared the documents, arranged for their signature; had them registered in the Companies Registry and then delivered the originals to the bank. Ms Mak identified her own signature on the deed and stated that, following her usual practice, she would have verified the identity of the person signing on behalf of ECT – confirmed to be Ms Lau – before signing as having witnessed Ms Lau signing the document in her presence. 8. On 12 April 2005, Ms Lau was charged with theft. Her trial began in the District Court before Deputy District Judge Yiu on 15 November 2005.[4] It lasted 24 days, running initially until the first week of December and then resuming on 23 January 2006, with Ms Mak due to give evidence on the following day, 24 January. Ms Lau was eventually acquitted on 1 April 2006 for reasons unconnected with the case against Mr Wong. A.2 Mr Wong’s conduct 9. About three weeks before the start of Ms Lau’s trial, Mr Wong saw Ms Mak’s witness statements and in his “Note of possible defences” dated 24 October 2005 (circulated to the defence team), he first made the suggestion that Ms Mak may have been acting as Ms Lau’s solicitor and that she might infringe Ms Lau’s legal professional privilege (“LPP”) if she were to testify in court. The Note stated: “[Ms Mak] who is expected to testify against [Ms Lau] was her solicitor at the material time, so as her intended act of testifying in court is likely a breach of the privilege between solicitor/client which belongs to the client and not the solicitor. Further research needs to be done to find out the relevant authorities … and whether Defendant is entitled to enjoin the said solicitor not to do so.” 10. Mr Ngai, who was acting as Mr Wong’s junior, evidently had a stab at such further research and, in a memorandum to Mr Wong dated 28 October 2005, he cited China National Petroleum Corp v Fenwick Elliot,[5] an English decision mentioned in Phipson on Evidence, as authority for the proposition that “where a solicitor sought to interview and take a proof of evidence from a potential witness, privilege attached not merely to the proof and what was said during interview, but also to the identity of the witness”. However, Mr Ngai commented that that case “may not be of much assistance” since it dealt with a “factual scenario ... so different from our present case.” 11. Mr Wong was not put off the scent. In his “Further Note on Preparation of Defence” dated 1 November 2005, he asked for instructions from Ms Lau as to “the exact circumstances she was under (ie the communication between she and [Ms Mak]) leading to her signing [the deed]”. 12. A fortnight later, on 15 November 2005, the trial of Ms Lau commenced. Some 10 days into the trial, Mr Wong drafted and caused LC&K to send a letter dated 24 November 2005 to Ms Mak (“the 1st Letter”).[6] It is set out in full in the Court of Appeal’s judgment.[7] It enclosed a copy of a bill of costs dated 22 September 2001 issued by Ms Mak on behalf of CMK&Co in connection with the bank transaction (“the bill of costs”). The 1st Letter stated that LC&K were acting for ECT and asserted that Ms Mak had acted for ECT in that transaction, owing it a continuing duty to protect its privileged communications. It complained that by giving the witness statements to the police, Ms Mak had breached that duty and urged her “not to aggravate the aforesaid breaches and stop doing anything that would aggravate” them. It ended with the statement that ECT reserved “the right to take out civil action against you as well as refer the matter to the Law Society of Hong Kong for taking disciplinary action against you if the need exists.” 13. The enclosed bill of costs was addressed to ECT and captioned as being in the matter of the fixed and floating charges provided to the bank by the Maruyama companies and by ECT. CMK&Co invoiced ECT for $31,430.00 by way of costs and disbursements with the costs particularised as being “To our professional charges for preparing engrossing and attending execution of the [security documents] in favour of [the bank]”; “to our professional charge for preparing the supporting minutes in relation to the above”; and “to our professional charges for attending companies filing of the above documents with the Companies Registry”. 14. On 30 November 2005, Ms Mak replied.[8] She was guarded as to whether ECT was her client, stating: “... whether I acted for your client on the above matter is a matter to be determined on the facts of the case”, but in any event she denied the alleged duties and breaches. She stated that she was subject to a witness summons and “As such I am required by law to serve as a witness” at the trial, adding: “It would be for you to take out proceedings to set aside such witness summons if you so wish. You should appreciate that unless and until you have successfully obtained such order to set aside the said witness summons served on me, I would be obliged by law to be a witness in the court unless otherwise ordered.” Ms Mak informed the police that she had received the 1st Letter and they passed on that information to the prosecution. 15. Upon receiving her reply, Mr Wong drafted a letter which was sent out by LC&K to the Law Society dated 30 November 2005.[9] It asked in general terms (without giving names or other details) about a client’s rights and a solicitor’s duties regarding privilege, confidentiality and privacy arising out of advice given “in a property transaction” where the solicitor is summonsed to testify against the client. 16. Not surprisingly, the Law Society’s response dated 9 December 2005 stated that it was unable to advise given the limited facts provided. It sent LC&K a copy of the chapter on confidentiality in the Hong Kong Solicitors’ Guide to Professional Conduct. 17. On 12 December 2005, about a month into the trial, Mr Wong drafted and caused LC&K to send to Ms Mak the 2nd Letter upon which the charge is based.[10] It enclosed the correspondence with the Law Society (said to be “self-explanatory”) and, in reply to Ms Mak’s letter of 30 November, it drew attention to section 34(2) of the Criminal Procedure Ordinance[11] stating that such provision “dictates that it is for the witness who is served with the said Summons to apply to the issuing court to nullify [it]”. It cited R v Derby Magistrates’ Court, ex p B[12] for the proposition that evidence covered by LPP is not “material evidence” for the purposes of the section. 18. Ms Mak replied on 19 December 2005.[13] She repeated her guarded statement as to whether ECT had been her client; denied the relevance of section 34(2) and reiterated that it was for LC&K to take out proceedings to set aside the witness summons if they so desired, stating once more that unless that was done “I would be obliged by law to be a witness in the court unless otherwise ordered.” 19. As previously noted, the trial was adjourned part-heard in the first week of December and was due to resume on 23 January 2006. On 16 January, Mr Wong wrote to prosecuting counsel suggesting that Ms Mak be called on 24 January to accommodate Mr Wong’s other commitments. Neither by that letter nor by any prior communication did Mr Wong inform the prosecution that the defence had been writing to Ms Mak or that they had any concerns about LPP. 20. On the same day, 16 January, Mr Wong drafted a letter to be sent to the Law Society by LC&K. However, the fax coversheet contained a request by Mr Wong’s secretary that LC&K should “discuss with Counsel before issuing the letters.” 21. The letter to the Law Society was only sent out by LC&K and copied to Ms Mak on 20 January 2006 (“the 3rd Letter”). Since Ms Mak was due to give evidence on Tuesday, 24 January, this meant that those letters were only sent out on the preceding Friday, with a weekend intervening. The prosecution attaches importance to that timing as indicating that the letter was a pure threat and not seriously intended to evoke a response from the Law Society. 22. The enclosure to the 3rd Letter[14] provided the Law Society with copies of LC&K’s correspondence with Ms Mak concerning her intended giving of evidence. It complained that “despite repeated reminders, Ms Mak has adopted an attitude of defiance towards complying with her duty of upholding client’s privilege/confidentiality as well as privacy...” and invited the Law Society “to look into the matter and take whatever preventive and/or remedial measures necessary to uphold the professional ethics applicable to a member...” 23. As with the 1st and 2nd Letters, Ms Mak informed the police of her receipt of the 3rd Letter. A conference with prosecuting counsel was held with her on 23 January. A.3 The hearing 24. On 24 January 2006, just as the prosecution was about to call Ms Mak as a witness, Mr Wong rose to object to her giving evidence on the grounds of LPP. He told the Court that “she was the solicitor acting for ECT and its officers” in the transaction, pointing to the bill of costs. In support of his argument, he handed to the Judge a photocopy of a single page from the then current edition of Blackstone’s Criminal Practice which referred to a legal adviser’s duty, subject to the client’s waiver, to refuse to give evidence relating to communications between the adviser and the client for the purpose of obtaining or giving legal advice. The prosecution places substantial weight on the fact that the extract produced by Mr Wong was incomplete, misleadingly excluding – say the prosecution – passages identifying as matters not covered by the privilege, the identity of a solicitor’s client and matters falling within the crime and fraud exception.[15] I return in Sections E.2a to E.2c below to examine more closely what occurred at the hearing. 25. After hearing submissions, the trial Judge ruled against the privilege claim and Ms Mak testified that she had been instructed by the bank to prepare the deed; that she had verified the identity of Ms Lau as the person who had signed it in her presence on behalf of Brightway; and that she had then caused the deed to be registered in the Companies Registry. Mr Wong’s cross-examination sought to suggest that Ms Mak was not in fact sure that it was her own signature on the document as she had asked to see the originals before confirming that this was the case. A.4 The present charge 26. As the LC&K letters addressed to Ms Mak were signed by Mr Yeung, he was charged with perverting the course of justice. On 22 July 2009, after trial before Deputy Judge M Chow in the District Court, he was convicted and sentenced to four months’ imprisonment suspended for 12 months. However, his conviction was quashed by the Court of Appeal since, on the evidence, Mr Yeung had little real involvement in the conduct of Ms Lau’s case and the findings as to the state of his knowledge were unsafe.[16] 27. The fact that Mr Wong had drafted the letters and the extent of his involvement emerged in the course of Mr Yeung’s trial and in September 2009, Mr Wong was charged with attempting to pervert the course of justice, initiating the present proceedings. On 29 January 2010, Messrs Haldanes, his solicitors, wrote to Ms Lau asking whether she would waive privilege in respect of communications made in the course of Mr Wong acting for her. No waiver was forthcoming. B. The applicable principles 28. Two sets of principles are relevant: first, those which determine how the offence of attempting to pervert the course of justice is constituted, particularly (on the present facts) in relation to allegedly improper approaches made to a witness; and secondly, those which determine when legal professional privilege arises and the scope and effect of the privilege. B.1 Attempting to pervert the course of justice 29. In HKSAR v Egan,[17] this Court had occasion to consider the elements of a conspiracy to pervert the course of justice in relation to an alleged attempt to influence a potential witness. That review and the authorities there cited are closely relevant to the present case. They provide the basis for the following propositions. B.1a The offence generally 30. An attempt to pervert the course of justice is “the doing of some act which has a tendency and is intended to pervert the administration of public justice.”[18] The actus reus is thus the doing of the act with the prohibited tendency and the mens rea is the intention to pervert the course of justice. 31. With regard to the actus reus generally, the following propositions may be stated: (a) Although it is called an “attempt”, it is a substantive offence. But like an inchoate offence, it is committed even where the act with the prohibited tendency does not actually result in interference with the course of justice.[19] (b) A person’s conduct has a tendency to pervert the course of justice if it has a tendency towards “impairing (or preventing the exercise of) the capacity of a court or competent judicial authority to do justice”.[20] (c) The tendency must involve impairing the administration of justice in curial proceedings. Conduct with a tendency to result in unlawfulness which does not interfere with curial proceedings does not constitute the offence.[21] 32. The following propositions may be stated in respect of the mens rea of the offence: (a) To prove that a defendant intended to pervert the course of justice, it is necessary to prove that he knew that his conduct would have or that he intended it to have a tendency to pervert the course of justice in relation to the curial proceedings in question.[22] (b) Where his conduct has a manifest tendency to pervert the course of justice, the required intention may readily be inferred from proof that the defendant intended to perform the relevant act. But where the act does not manifestly have the prohibited tendency, it is necessary to prove the abovementioned specific intent on the defendant’s part.[23] B.1b The offence in connection with attempts to influence a witness 33. All kinds of conduct may have the prohibited tendency. The focus in the present case is on an approach made to a witness with a view to influencing her in relation to the evidence she is expected to give. In that context, the following propositions may be stated: (a) There is no property in a witness and in principle neither party can prevent the other side from seeing a witness. [24] Of course, the approach made must not involve doing any act with the prohibited tendency and intention vis-à-vis the witness. (b) Whether an approach does or does not have such a tendency depends on the circumstances. In many cases, the tendency will be clear, but more difficult questions may arise in the penumbra of the offence, such as where the approach is made believing that the intended evidence is false, or where threats are made to exercise a legal right.[25] In the present case, an approach made to assert LPP must be addressed. (c) In determining whether the prohibited tendency is established, the courts have focussed on the purpose of the approach and the means used to try to influence the witness. The purpose or objective of a defendant’s approach to a witness is thus usually treated as part of the actus reus, the prohibited tendency. However, it necessarily involves examining the accused’s mental state. So the concept of “purpose” overlaps with the mens rea requirements of the offence. (d) Both the purpose of the approach and the means used must be lawful. If either is unlawful or improper (in the sense of putting the proper administration of justice at risk), the prohibited tendency is proved. Thus, as stated in Egan: “...great care must be taken if an approach is to be made with a view to influencing a witness in relation to his or her anticipated evidence. Such an approach can only properly be made if its purpose is lawful and if no unlawful or improper means are used to secure that purpose.”[26] (e) An example of a lawful approach involves the case where a person who believes the intended evidence to be false approaches the witness with the object of persuading him or her not to give such false evidence (a lawful purpose) using reasoned argument supported by material facts and documents (lawful means).[27] (f) In contrast, if the object of the approach is to get the witness to give false evidence or to refrain from telling the truth, the offence is committed even if the means used involve no more than exercising or threatening to exercise a legal right (such as threatening to give an employee lawful notice terminating the employment).[28] (g) Equally, even if the object of the approach is lawful, for example, to persuade the witness not to give evidence believed to be false, the offence would be committed if unlawful or improper means (such as improper pressure or bribes) are used to achieve that aim.[29] (h) Whether conduct constitutes improper pressure may raise questions of fact and degree. It may be important to consider when, by whom and upon whom and for what purpose the pressure is applied.[30] (i) In some cases, in deciding whether the pressure acquired the prohibited tendency, it may be useful to ask whether it had the tendency to prevent the witness making a free and voluntary choice as to whether to give evidence and what evidence to give.[31] B.2 Legal professional privilege B.2a LPP generally 34. The rationale of legal professional privilege was stated by Lord Taylor of Gosforth in R (ex p B) v Derby Magistrates’ Court[32] as follows: “The principle which runs through all these cases ... is that a man must be able to consult his lawyer in confidence, since otherwise he might hold back half the truth. The client must be sure that what he tells his lawyer in confidence will never be revealed without his consent. Legal professional privilege is thus much more than an ordinary rule of evidence, limited in its application to the facts of a particular case. It is a fundamental condition on which the administration of justice as a whole rests.” 35. His Lordship also reiterated the well-established rules that a communication protected by privilege continues to be protected so long as the privilege is not waived by the client; that the privilege exists whether disclosure is sought for the purpose of civil or criminal proceedings, and whether by the prosecution or the defence; and that the refusal of the client to waive his privilege, for whatever reason, or for no reason, cannot be questioned or investigated by the court.[33] Referring to the position of the legal adviser, Lord Taylor stated: “...the privilege is that of the client, which he alone can waive, and ... the court will not permit, let alone order, the attorney to reveal the confidential communications which have passed between him and his former client. His mouth is shut forever.”[34] 36. The two categories of LPP are usually referred to as “legal advice privilege” and “litigation privilege”, described by Lord Scott of Foscote in Three Rivers District Council v Governor and Company of the Bank of England (No 6)[35] as follows: “The modern case law on legal professional privilege has divided the privilege into two categories, legal advice privilege and litigation privilege. Litigation privilege covers all documents brought into being for the purposes of litigation. Legal advice privilege covers communications between lawyers and their clients whereby legal advice is sought or given.” 37. Once established, legal advice privilege “gives the person entitled to it the right to decline to disclose or to allow to be disclosed the confidential communication or document in question.”[36] Moreover, unlike other classes of confidential communications, LPP cannot be overridden by the court: “... if a communication or document qualifies for legal professional privilege, the privilege is absolute. It cannot be overridden by some supposedly greater public interest. It can be waived by the person, the client, entitled to it and it can be overridden by statute (cf R (Morgan Grenfell & Co Ltd) v Special Comr of Income Tax [2003] 1 AC 563), but it is otherwise absolute. There is no balancing exercise that has to be carried out: see B v Auckland District Law Society [2003] 2 AC 736, 756-759, paras 46-54). ... legal professional privilege, if it is attracted by a particular communication between lawyer and client or attaches to a particular document, cannot be set aside on the ground that some other higher public interest requires that to be done.”[37] B.2b Implied retainers 38. In the present case, there was no litigation pending or in contemplation at the time when Ms Lau signed the deed. Any LPP arising could therefore only involve legal advice privilege. For legal advice privilege to arise, Ms Mak had to be acting as solicitor for ECT (for whom Ms Lau, a director, was acting) in the relevant transaction. Ms Mak was plainly instructed by the bank but was never expressly retained to act simultaneously as ECT’s solicitor. So it is necessary to consider the principles for determining when a solicitor is held to have impliedly accepted a retainer to act for the person in question. 39. It is clear that a solicitor and client relationship may arise by implication. Thus, as Scott LJ stated in Groom v Crocker:[38] “The relationship is normally started by a retainer, but the retainer will be presumed if the conduct of the two parties shows that the relationship of solicitor and client has in fact been established between them.” 40. A case where the court found an implied retainer clearly established is Pegrum v Fatharly[39] where, in order to save costs, both the lender and borrower in a loan transaction went together to consult the same solicitor who prepared all the transaction documents. Anderson J stated: “When both parties to a transaction consult the same solicitor and together give him the information needed to prepare the documents in which their respective rights and obligations are to be set out and the solicitor accepts responsibility to prepare the documents without any indication that he cannot fully discharge his professional duties to them both there is a strong bias towards finding that the solicitor tacitly agrees to act for both parties and to undertake the usual professional responsibilities to them both: see Midland Bank Trust Co Ltd v Hett, Stubbs & Kemp [1979] Ch 384 esp at 396. In the absence of a clear indication by the solicitor that the solicitor does not accept one of the parties as his client it is natural in such a case to assume both are relying on him for professional advice and assistance. This follows from the mere fact that both have consulted him. There may be other circumstances which show that there is no reliance by one or other of the parties on the solicitor, but, if not, reliance should be inferred as a fact.”[40] 41. His Honour added: “This does not mean a solicitor whose services are sought by both parties is bound to accept that he is to serve both parties. He can refuse to do so and elect to act for one party only. This requires a very clear statement by the solicitor that this is to be his position. It has even been held that he is duty bound in such a case to positively recommend that the other party get another solicitor and take independent advice before entering into the transaction, and, in the event that recommendation is not followed, to give him proper advice as to the risks in signing the documents: see Irvine v Shaw [1992] ANZ Conv R 83.”[41] 42. Whether an implied retainer arises therefore depends on whether a solicitor and client relationship can properly be inferred on the facts. In Watson v Ebsworth & Ebsworth,[42] the New South Wales Court of Appeal cited a helpful passage from Professor G E Dal Pont’s book Lawyers’ Professional Responsibility,[43] indicating the sorts of circumstances that might be relevant: “Retainers, like other contracts, need not be created by express words, whether written or oral; their existence can be inferred or implied from the circumstances. In this context it is the existence of the retainer that is implied, not its terms ... Proof of an implied retainer rests on proof of facts and circumstances sufficient to establish a tacit agreement to provide legal services. Its existence is determined by inference from objective facts, not merely by the lawyer's belief as to which clients he or she was acting for. The reasonable expectations of the alleged client carry significant weight here, as the lawyer may always take steps to dissuade any person from a belief that the lawyer acts for that person. The lawyer's file (including letters and other correspondence) and diary notes, specifically regarding how he or she has referred to and dealt with the claimant, may prove useful. Aspects that may impact upon a client's reasonable expectations of a retainer include the capacity in which the lawyer acted (say, giving advice in a non-legal capacity), who instructed the lawyer, who is liable for the lawyer's charges, and whether a contractual relationship existed with the claimant in the past (as the court may be readier to assume that the parties intended to resume that relationship in such a case).” 43. Anderson J[44] also cited Beach Petroleum NL v Kennedy,[45] where the Court emphasised the limited relevance of the solicitor’s understanding of his own retainer: “His own understanding of the relationship is not conclusive. The question of a retainer is not determined by the belief of the solicitor as to the company or companies for whom he or she is acting. It is determined by the objective facts. Mr Bateman's evidence of his understanding of the relationship is little more than evidence of inferences he drew from the circumstances. In the absence of plain language from any of the directors of the various companies, the matter remains to be determined by inference.” 44. In Hong Kong, an implied retainer was found by Tang J (as Tang PJ then was) to have arisen in International Trading Co Ltd v Lai Kam Man,[46] a property sale and purchase case in which the purchaser, in accordance with the memorandum of sale, had to bear all solicitors’ fees in connection with the purchase. Relevant to the inference being drawn was the fact that the solicitors, who had been nominated by the developers, had written to the plaintiff : “... saying that if the purchaser had any question or that further assistance was required, the purchaser should contact [the solicitors] at any time. The letter did not say that [the solicitors] were acting as [attesting agent only] only nor that they were the Vendor's solicitors only. It also did not say that such further assistance would be given without any responsibility and as the Vendor's solicitors only, and hence on behalf of a person with a conflicting interest. Nor did it say that the purchaser was under no direct obligation to pay any fees to [the solicitors] and that any fees payable would be payable by the purchaser pursuant to its contract with the Vendor.”[47] C. The Issues 45. Mr Wong’s case at trial was that he genuinely believed that the evidence Ms Mak was intended to give was covered by LPP and that the letters written to her were a legitimate assertion of ECT’s legal right to demand that she protect that privilege by declining to give the evidence sought. 46. To decide whether Mr Wong’s acts amounted to conduct with a tendency to pervert the course of justice and whether he knew or intended that his acts would have the prohibited tendency, a series of questions arise: (a) Was there a solicitor and client relationship between Ms Mak and ECT in connection with Ms Lau’s execution of the deed? What did Mr Wong believe the position to be? (b) If there was or may have been such a solicitor and client relationship, was the evidence which Ms Mak was intended to give within the scope of ECT’s LPP? What did Mr Wong believe the position to be? (c) What was Mr Wong’s purpose in causing the letters to be sent to Ms Mak? Was it an unlawful purpose? (d) Was Mr Wong’s conduct of sending the letters to Ms Mak without informing the prosecution, rather than directly raising the matter with the Court, conduct with the prohibited tendency? Did such conduct constitute unlawful or improper means of achieving his purpose (even if he may have been pursuing a lawful purpose)? (e) Did Mr Wong know that such conduct was or intend such conduct to be conduct with a tendency to pervert the course of justice? D. The Judge’s decision and the Court of Appeal’s views thereon 47. It may be noted that the Judge’s decision was issued after the Court of Appeal’s judgment in Egan,[48]but about six weeks before this Court’s judgment in Egan was published. The Court of Appeal’s judgment below was issued afterwards. D.1 Whether a solicitor and client relationship was created 48. The Judge did not consider it necessary to decide whether a solicitor and client relationship had arisen, saying that it was “not conclusive of guilt or innocence; merely a cogent milestone on the journey”.[49] However, he apparently inclined towards the view that no such relationship existed, placing reliance on the evidence of Ms Mak. 49. The Court of Appeal concluded[50] that the Judge had decided that no retainer existed and that his decision was erroneously based on Ms Mak’s understanding of who her client was. She had repeatedly denied that she was acting for anyone other than the bank,[51] saying that she was “certain” that ECT was not her client.[52] Without itself deciding whether an implied retainer had actually been created, Stock VP (with whom the other members of the Court agreed) stated: “It is merely necessary to say that Ms Mak’s ipse dixit was not conclusive of the matter and that the foundation existed of an arguable point, albeit not a strong foundation. It is not even necessary to go that far; for what suffices for part of the applicants’ purposes is some footing from which to accept that Wong genuinely thought he had the makings of an argument, an argument which he thought possible once he had sight of the solicitor’s invoice.”[53] D.2 Whether covered by LPP and what Mr Wong believed 50. The Judge concluded that objectively, there was no LPP “given the scope of the evidence that was required by the prosecution in the trial, and even if it did, it was covered by the exceptions I have referred to”.[54] The exceptions in question were “as to identity and to fraud”.[55] 51. As to Mr Wong’s belief, his Honour stated: “I am also satisfied that as far as you are concerned you had little or no belief in the actual existence of legal professional privilege and being alerted to the problems by your fellow counsel and the exceptions referred to in Blackstone. Your reliance on the bill of costs and instructions from your client constitutes a deficiency of basis which I neither believe nor accept. In short, if you did hold such a belief, it was a self-induced delusion contrary to the facts, the law and procedure.”[56] 52. The finding is somewhat equivocal (with references to Mr Wong having had “little or no belief” and possibly having that belief as a “self-induced delusion”), but I take it that the finding was that Mr Wong had no genuine belief in the existence of LPP. 53. It seems that the Judge’s thought that finding did not really matter since the offence was in any event established because improper means had been used: “Whether you believed in its existence or not seems to me to be of limited relevance subservient to what you did and intended to do. If you had a genuine belief in the true existence of legal professional privilege, there were at least two proper lines of enforcement, probably three.” [57] [Which he subsequently held had not properly been pursued] 54. The Court of Appeal agreed that objectively, there was no basis for an LPP claim. Stock VP gave three reasons for that conclusion: (i) that the evidence sought from Ms Mak did not involve legal advice;[58] (ii) that LPP does not cover the identity of the client except where disclosure may incriminate the client or indirectly reveal privileged communications;[59] and (iii) that in the present case, the signatory’s identity was not intended to be kept confidential, the bank having an interest in ensuring that the person concerned was properly signing on behalf of ECT.[60] On that basis, Stock VP held that “The argument was a nonsense”[61] and that “...Wong was ... patently wrong in asserting that the limited intended evidence by Mak was covered by legal professional privilege.”[62] 55. However, his Lordship was uneasy about the Judge’s finding that Mr Wong had no belief that an arguable LPP claim existed. Stock VP thought that there were grounds for doubting the genuineness of Mr Wong’s stated belief, including his partial citation of Blackstone, his oblique correspondence with the Law Society and his keeping of the LPP claim “up his sleeve” until the last moment.[63] But his Lordship noted that the Judge had not mentioned Mr Wong’s “Note of possible defences” and “Further Note on preparation of the defence”.[64] He also drew attention to the evidence of Mr Wong’s “aggressive tenacity on behalf of his clients” and “the evidence which suggests that he is a facts advocate, not much at ease with arguments of law”.[65] Stressing that “One has in a case such as this to take the greatest care to distinguish between misguided professional enthusiasm or even incompetence, on the one hand and, on the other, dishonesty”,[66] Stock VP held that there were sufficient residual doubts about the Judge’s conclusion as to lack of belief.[67] 56. The Court of Appeal held, however, that those doubts did not affect the safety of the conviction because “the Judge made it clear that these factors did not ultimately dictate his decision...” resting his conclusion on his finding that improper means were used.[68] D.3 Mr Wong’s purpose D.3a The prosecution’s case 57. In opening for the prosecution, Ms Charlotte Draycott SC submitted that Mr Wong’s acts were “an attempt to prevent or dissuade [Ms Mak] from testifying in court as a prosecution witness...”[69] This was put to him in cross-examination: “Q: So how can you explain what you did, except in terms of an attempt to dissuade her from coming to court.”[70] 58. However, the prosecution then introduced an alternative purpose. Ms Draycott suggested to Mr Wong that he knew full well that his privilege argument did not hold water and added: “And that was why you had been trying to get [Ms Mak] either not to come to court or to claim privilege or not give evidence because you knew if it was argued in court you were bound to lose.”[71] 59. Mr Graham Harris SC, then leading for the defence, objected that this was a change of case by the prosecution and, in closing submissions, Ms Draycott stated: “The third matter which they raise as an apparent change of ground is that in opening, it was said that the intent was both by threats and more subtle pressure to unlawfully persuade [Ms Mak] from giving evidence, and the letters were designed with intent to improperly influence [Ms Mak] not to give evidence. And it’s suggested there’s a change of ground now because in my closing submission I deal with it not only on that basis but also on the basis that it was an attempt to make [Ms Mak] do something which she did not want to do, which she did not feel was justified, in other words, to set aside her witness summons.”[72] 60. Ms Draycott argued that it was “splitting hairs” to suggest that the two ways Mr Wong’s purpose had been put were different: “... the immediate aim may have been to persuade [Ms Mak] to make an application to set aside her witness summons on the basis of legal professional privilege and not being able to give material evidence. But the ultimate aim was to stop that evidence going before the court, and as I said in closing, it’s splitting hairs to say they are different because the purpose was to stop the court having the evidence, and that perversion happened because pressure was brought to bear on her to do something which he did not want to do with that end in mind.”[73] D.3b The Judge’s findings 61. The Judge appears to have accepted that there was no difference in the two formulations of Mr Wong’s purpose. His Honour found that Mr Wong “decided to try to derail the witness prior to giving evidence”.[74] However, in the very next paragraph, he stated: “Your intention was to persuade her to make an application to set aside the witness summons she had received [on the basis of LPP].”[75] 62. The latter approach seems ultimately to have been favoured: “... you wilfully and consciously set aside legitimate lines of defence in order to try to force by threats and intimidation [Ms Mak], prosecution witness, to adopt and embrace a position she did not accept or believe in and take a course of action she did not wish to take.”[76] “They are threats simpliciter, threats intended to deflect her from her legal duty and to force her to make an application contrary to her beliefs and wishes”.[77] 63. His Honour suggested a motive for such threats: “It is clear to me, and it was clear to you, defendant, that if [Ms Mak] herself was to assert she was bound by legal professional privilege, as a solicitor, the court would be more likely to accept that position.”[78] D.3c The Court of Appeal’s approach 64. The Court of Appeal also viewed the case as proceeding on the latter basis. Stock VP described the essence of the case against Mr Wong[79] as seeking: “... by improper pressure to force a prosecution witness, herself a solicitor, to claim, on a basis in which Wong ... had no faith, that by reason of legal professional privilege she was precluded from testifying.”[80] 65. His Lordship also noted that, as a last resort, Mr Wong’s intention was to apply to the Court himself. His case: “... as no doubt the judge recognized, was that persuading Ms Mak to seek to set aside the summons – persuading her by the letters to her and to the Law Society – was the principal course, failing the success of which an application would be made ‘when the witness came to testify’”.[81] 66. The Court of Appeal did not consider the prosecution’s changed direction material, either substantively or procedurally: “It is true that the prosecutor’s closing submission was the first time that the objective was put in that particular way (to make an application, contrary to her wishes, to set aside) and, in that sense, it evidenced a change. But I am satisfied that the change was not material and, in any event, occasioned no prejudice.”[82] 67. Stock VP echoed the Judge’s view of Wong’s possible motive, namely, that a claim made by a solicitor “is likely to carry considerable weight; more weight on its face than if the claim were made by the client”.[83] D.4 Whether the conduct had the prohibited tendency D.4a The Judge’s decision 68. The Judge decided that even if a legitimate claim of LPP existed, the offence was committed because the means adopted to enforce that claim – his causing threatening letters to be sent – were improper and had the tendency to pervert the course of justice. 69. The Judge found that prohibited tendency arose out of (i) the fact of Mr Wong’s approach to a prosecution witness; (ii) his failure to inform the prosecution of what he was doing; and (iii) his failure to raise the issue of privilege directly with the Judge at Ms Lau’s trial, Judge Tallentire’s view being that that was the only proper avenue for such a claim. 70. Thus, his Honour took the view that was it was a breach of professional standards for Mr Wong to instigate: “...communications with a prosecution witness which contained clear and unambiguous threats, without informing the prosecution and in circumstances where you had, and you knew you would have, the opportunity to properly pursue this application when you appeared on behalf of your client in the District Court.”[84] 71. The fact that Mr Wong had approached a witness on the other side was said to be a major hurdle he had to face. Such an approach, his Honour held: “... should happen only in the most exceptional circumstances and be attended by the utmost circumspection and probably transparency.”[85] 72. He held that such an approach was improper because: “... the parties must perforce be the prosecution and the defence; not defence and a prosecution witness. Therefore, if the matter was to be resolved between the parties, it was the prosecution who should have been approached. You, however, did not do this; nor did you even inform the prosecution of your direct approach to the prosecution witness.”[86] 73. The failure to inform the prosecution was also regarded as highly significant: “What weighs heavily with me is why you did not seek to clarify the situation via the prosecution. This is unexplained and in my opinion is not capable of explanation.”[87] 74. The other major “hurdle” facing Mr Wong was identified as follows: “The second hurdle, one of great difficulty, is that the judge in the trial was the person to settle the matter. Your evidence that you intended this to be done, albeit by [Ms Mak’s] raising it, does not ring true....” 75. The Judge also found that the threatening tone of the letters, coupled with their timing, showed that they were purely intended as threats and could not be regarded as lawful attempts at reasoned persuasion.[88] D.4b The Court of Appeal’s disagreement with the Judge 76. The Court of Appeal disagreed with the Judge’s central reasons for holding that the conduct had the prohibited tendency. 77. Their Lordships held[89] that it was wrong to regard an approach made by the defence to a prosecution witness, without more, as objectionable and tending to pervert the course of justice. There was no basis for holding that an approach could only be made in “the most exceptional circumstances”. The established rule is that there is no property in a witness, as this Court reiterated in Egan[90]: “... as Lord Denning MR pointed out, neither party can ‘prohibit the other side from seeing a witness of fact, from getting the facts from him and from calling him to give evidence or from issuing him with a subpoena.’ Pointing to the Law Society’s guidance to its members, his Lordship stated: ‘... the Council have always held the view that there is no property in a witness and that so long as there is no question of tampering with the evidence of witnesses it is open to the solicitor for either party to civil or criminal proceedings to interview and take a statement from any witness or prospective witness at any stage in the proceedings, whether or not that witness has been interviewed or called as a witness by the other party.’”[91] 78. Given that there is no property in a witness, the Court of Appeal held that there was nothing improper in not telling the prosecution of the approach made.[92] 79. The Court of Appeal also disagreed with the Judge’s view that the criminal court was the only proper forum for raising an LPP objection. It rejected the defence submission that section 34(2) of the Criminal Procedure Ordinance[93] precluded ECT (and anyone other than the witness summonsed) from applying to the criminal court to discharge the summons, holding that jurisdiction to entertain a third person’s application existed at common law.[94] However, it noted that in Rockefeller & Co Inc v Secretary for Justice,[95] Godfrey VP had held that where documents said to be privileged had already come into the hands of a third party and it was sought to restrain their use court, the proper course was to bring civil proceedings to restrain such use on behalf of the person whose privilege it was. Godfrey VP also said in passing that if a privilege objection was taken to production of documents pursuant to a subpoena duces tecum, the party claiming the privilege “could no doubt have applied to the trial judge to have the subpoena set aside”. 80. The Court of Appeal therefore concluded that ECT could have chosen either to apply to the criminal court or to take the civil avenue.[96] It held nevertheless that the point was academic since no civil proceedings were intended and since Mr Wong had intended – he said as his last resort – to raise the point himself before the trial Judge in the criminal proceedings.[97] D.4c The Court of Appeal’s basis for upholding the conviction 81. Having rejected the basis upon which the Judge’s finding of a prohibited tendency had rested, the Court of Appeal nevertheless upheld the conviction, substituting its own basis for that finding, as set out in three key paragraphs of the judgment. 82. Having cited an important passage from R v Rogerson,[98] which will require closer examination, Stock VP held at paragraph 74 that “the course of justice is deflected by causing applications to be made to a court which, but for improper pressure, would not be made.” His Lordship elaborated upon that conclusion at paragraphs 168 and 169 of his judgment which are examined in Section H.2 below. D.5 Mr Wong’s mens rea 83. As mentioned above,[99] in cases involving approaches made to witnesses, the courts have examined the purpose of the approach and the means used to try to influence the witness. And, as I have noted, a finding of what the defendant’s “purpose” was inevitably bears upon a finding of whether he had the requisite mens rea. 84. In the present case, there is no doubt that Mr Wong intended to do the acts complained of: he drafted letters which contained threats of civil action and professional complaint and caused them to be sent to Ms Mak; and he did so without informing the prosecution and rather than taking the initiative of raising the matter directly with the Court. The Courts below found that this pressure had the purpose of compelling Ms Mak to apply to the Court to set aside the witness summons on the ground of LPP, with his own possible application as a last resort. The means employed by Mr Wong were held to constitute improper means with the prohibited tendency. Given those findings, it was implicitly found that mens rea was proved. E. Analysis of the issues E.1 Whether there was an implied retainer 85. It is not in dispute that Ms Mak acted for the bank. There was, in my view, a reasonably arguable case that an implied retainer had arisen for Ms Mak also to act for ECT in relation to the deed. More importantly, there is ample ground for accepting that Mr Wong may have believed that there was such an implied retainer. 86. As the Court of Appeal rightly held, whether or not such a relationship arose did not depend on Ms Mak’s subjective view. The legal principles are discussed in Section B.2 above. They establish that an implied retainer may be inferred if the objective facts indicate a tacit agreement to provide legal services. The relevant facts in the present case are as follows. 87. No other solicitor was involved. Ms Mak prepared all the documents, including ECT’s board resolutions. Those resolutions, together with her firm’s bill of costs, were sent to ECT in advance of their meeting for ECT’s approval.[100] The bill of costs was addressed directly to ECT, debiting it for the firm’s professional charges for preparing, etc, the charges and the board minutes, without suggesting that ECT was merely being asked to bear the bank’s legal costs. Ms Mak accepted in evidence that she may well have (and if asked would have) explained the parties’ rights and obligations arising under the deed. While she says that her usual practice included telling the bank’s customers that they might consider separate legal advice,[101] Ms Mak did not tell Ms Lau that Ms Mak was unable to act for ECT, nor did she positively recommend that ECT should be separately represented. In her written response to the 1st Letter which enclosed the bill of costs and alleged that Ms Mak had acted as solicitor for ECT, Ms Mak did not deny that relationship, merely stating: “... whether I acted for your client on the above matter is a matter to be determined on the facts of the case...” This was repeated in her response to the 2nd Letter. 88. It is possible that the precise position of Ms Mak acting solely as the bank’s solicitor might have been spelt out in the deed which was a 50 page document. It was also a document which Mr Wong had access to. However, it was not explored in the evidence. In opening, Ms Draycott handed up only a few extracted pages.[102] E.2 Whether the intended evidence was protected by LPP 89. As pointed out in Section D.2, Stock VP gave three reasons for concluding that the proposed evidence was not privileged, namely, that: (i) the intended evidence did not involve legal advice; (ii) the identity of a client is not privileged; and (iii) the identity of the person signing the deed was not confidential. I respectfully agree with the first and third reasons. However, the second reason requires to be addressed in some detail since it has had important ramifications for assessing Mr Wong’s state of knowledge and the propriety of his conduct. I will later deal more briefly with the first and the third reasons. E.2a Identity of the client – prosecution’s case 90. The prosecution deployed the “identity of the client” argument not only as a basis for arguing that the intended evidence was not privileged, but more importantly, for calling into question the genuineness of Mr Wong’s purported reliance on LPP. The prosecution’s argument ran as follows: (a) Mr Wong was seeking to suggest that the intended evidence was privileged because it would reveal the identity of Ms Mak’s client. (b) That was a bad point since in law, save for immaterial exceptions, the identity of the client is not privileged. (c) Mr Wong must have known this because, when taking the privilege objection at Ms Lau’s trial, he misleadingly handed to the Judge an incomplete citation of Blackstone, deliberately suppressing a passage that refuted the suggestion that the identity of the client is privileged. (d) This shows that the purported claim to LPP was a pretext and that the threats made against Ms Mak were wholly improper and conduct with a tendency to pervert the course of justice. 91. As Stock VP noted: “Much cross-examination was directed at the contention that he deliberately misled the court at the Lau trial by not revealing the passage in Blackstone which stood against his argument; and at other suggested indicia against his belief in the point. And in the closing submissions too, counsel for the prosecution asserted in terms that Wong ‘must have realised that if the claim was brought before a judge, it was likely to fail’.”[103] 92. The prosecution’s argument bore fruit. The Judge stated: “... it seems your research extended to extracts from Blackstone, choosing to ignore the exceptions, especially as to identity and to fraud, and that scenario involving a barrister of over 20 years experience and one whose own witnesses say is well capable of looking up the law is bizarre.”[104] 93. His Honour concluded: “... you had little or no belief in the actual existence of legal professional privilege and [sic] being alerted to the problems by your fellow counsel and the exceptions referred to in Blackstone.”[105] 94. The argument was also given some weight by Stock VP: “Wong’s omission to draw the Lau court’s attention to the relevant passage in Blackstone is supportive of the judge’s conclusion that Wong was well aware that the privilege argument was a non-starter.”[106] E.2b Identity of the client – irrelevant 95. There is indeed a body of jurisprudence on whether and in what circumstances a legal adviser may refuse to disclose his client’s identity. Someone may, for a whole variety of reasons,[107] want to know the identity of a certain other person and, knowing that a particular solicitor is acting for that person, seek to find out from that solicitor who that other person is. Such demands have often been met by the solicitor’s claim that his client’s identity is privileged and cannot be disclosed without the client’s consent. An extensive review of the authorities across several common law jurisdictions by the Australian Federal Court can be found in Commissioner of Taxation v Coombes.[108] The Court helpfully summarised its conclusions as follows: “The following propositions, among others, can be distilled from the cases we have examined: o Privilege attaches to communications, and not to facts which a lawyer observes while acting in the course of a retainer. o Privilege does not attach to everything a client says to the lawyer, but only to communications made by the client for the purpose of obtaining the lawyer's professional assistance. It will not attach to ``mere collateral facts''. The address and identity of a client will usually be ``collateral facts''. o Privilege attaches to communications only if they are confidential. In almost all cases the client's name and address will not have been communicated confidentially. o Instructions to a lawyer to do a particular thing, for example to prepare a legal document such as a will, are generally not privileged, because instructions to do something do not necessarily amount to a request for advice. o As a general rule, the identity of a client will not be privileged, as the privilege belongs to the client, and the retainer between the lawyer and the client must be demonstrated in order to establish the privilege. This requires disclosure of the client's identity. o Disclosure of the client's identity is necessary before the privilege can arise even if the client's name was given in confidence, and it was a condition of the lawyer's retainer that the client's identity be kept confidential. The client cannot by contract extend the area of privilege. o Some of the cases support an exception to this general rule when so much of the actual communication has already been disclosed that identification of the client amounts to disclosure of a confidential communication. This will be the case when the client's identity is so intertwined with the confidential communication that to disclose the identity would be to disclose the communication.”[109] 96. Those principles are plainly irrelevant in the present case. There was no question of anyone – Mr Wong or Ms Mak – seeking to discover or to withhold the identity of Ms Mak’s client. The very first sentence of the 1st Letter contained the assertion that ECT was Ms Mak’s client, relying on the bill of costs addressed to ECT, to which Ms Mak had made a guarded response.[110] Rather, the question was whether the identity of the person who had signed the deed was privileged. E.2c Identity of the client – the prosecution’s misapprehension of Mr Wong’s case 97. In fact, the suggestion that Mr Wong was trying to exclude Ms Mak’s intended evidence on the argument that it would offend LPP by disclosing her client’s identity originated from the prosecution. It was based on a misapprehension of Mr Wong’s case. The transcript of the application made on 24 and 25 January 2006 shows that it was not Mr Wong’s case that LPP prevented disclosure of the identity of Ms Mak’s client. His failure to hand up the passage in Blackstone saying that a client’s identity is not privileged, was therefore not an attempt to conceal an authority against an argument he was running. 98. Stock VP accurately summarises what occurred at the hearing and nowhere records that Mr Wong was seeking to argue that the intended evidence should be excluded as privileged information about the identity of Ms Mak’s client.[111] On the contrary, Mr Wong tendered the bill of costs and asserted that ECT and its officers were Ms Mak’s clients. As Stock VP notes, the identity of the client point arose when: “Counsel for the prosecution referred to authority the effect of which was that the identity of the client was not privileged”.[112] 99. On the first day, 24 January, the initial thrust of Mr Wong’s argument (which was not easy to follow) was that a breach of LPP was likely if Ms Mak were to testify because she would necessarily go beyond merely identifying the person who signed the deed and would stray into privileged territory. His submission was that: “At the time of the execution, attending a preparation of the deed, there would be discussion, there would be advice, explanation between [Ms Mak] and the client. And those information ... is privileged.”[113] 100. Prosecution counsel, Mr H L Wong, was at pains to assure the Court that he had no intention of asking Ms Mak about any communications between herself and Ms Lau. He submitted that asking Ms Mak whether and how she had verified Ms Lau’s identity; whether she had witnessed Ms Lau’s signing of the document; and whether she had caused the document to be registered in a public registry did not come within the scope of the privilege. 101. Mr Wong, then sought to argue that the privilege would be infringed because the document’s execution was somehow intrinsically bound up with supposedly privileged communications: “... The whole occasion of attending the office of the solicitor is in order to have the advice and then execution of document. One cannot dissect that occasion in different pieces.”[114] 102. The Judge asked for citation of authority and adjourned his ruling to the following day, 25 January 2006. When the hearing resumed, Mr Wong cited the China National Petroleum case and the extract from Phipson which Mr Ngai had previously mentioned,[115] for the proposition that privilege attaches “not merely to ... what was said in the interview but also to the identity of the witness”. That authority was, of course, irrelevant since, as Mr H L Wong submitted, it was a case about litigation privilege concerning the identity of a witness being proofed by a solicitor preparing a litigation brief, where different considerations apply.[116] However, it was not a case about privilege and the identity of a client. Mr Wong was trying to rely on that decision in support of his argument that a person’s identity could not be severed from privileged communications which had taken place with that person, repeating what he had endeavoured to argue on the previous day: “Now, my understanding of the reason behind this judgment covering the confidentiality of the identity of the person, is that this arising out of same matter, ie, during the course of taking proof of evidence, it would be artificial to try to sever this into this part is not proof read, this part is proof read. ... Our submission is, if it is arising out of the same matter, ie, the advice, preparation and execution of the deed, then it must be treated part and parcel of the same matter, ie, the previous communication at that time – under the law at that time …” 103. It was only on the second day, 25 January, that privilege and the client’s identity was first mentioned. Prosecution counsel, Mr H L Wong, read out a passage in Archbold stating that the identity of a client is not privileged. Mr Wong did not make any submissions in reply to that point. It had been on 24 January, before the client identity point was even mentioned, that Mr Wong had handed up the truncated extract from Blackstone. 104. The Judge ruled against Mr Wong’s application and permitted the prosecution to proceed with calling Ms Mak for the limited purposes indicated. 105. When Mr Wong came to be cross-examined by Ms Draycott at the trial in the present case, he tried to explain the submissions he made at Ms Lau’s trial along the lines indicated above: “... what I believed at that time is the communication that is the subject-matter of privilege and anything arose out of that could be covered by legal professional privilege, not simply the content of the communication.”[117] 106. There was the following exchange: “Q: Can you explain why you limited your extract of Blackstone simply to that single page? A: Looking at the transcript, what I remember what happened was that I relied on the general principle. That’s why I did not put forward the other pages ...”[118] 107. Mr Wong appears to have been saying (with justification) that he did not think the matters dealt with in the subsequent pages were relevant to his argument. The prosecution however pressed its view of what had happened: “Q: I suggest you were putting forward an incomplete and misleading picture. A: I disagree. Q: You knew perfectly well when you read out what you call the ‘general principle’ that there were exceptions and real issues in what you were saying, didn’t you? A: There has always been exceptions, right? Yes. Q: You chose not to bring those to the court’s attention and to limit the material you put in, even though on the very next page the answer to the question the judge was asking was shown. A: I disagree. Q: And that was why you have been trying to get [Ms Mak] either not to come to court or to claim privilege or not to give evidence because you knew if it was argued in court you were bound to lose. A: I disagree.” 108. In closing, , Ms Draycott submitted that the subsequent page in Blackstone had deliberately been withheld: “... we have copied the chapter of Blackstone on privilege into your bundle … and even the most cursory reading of it shows that the identity of a client is not covered by privilege. I invite you to read that and imagine yourself, as a lawyer, being paid to do your best for somebody and ask yourself, ‘Could I have missed that?’ and the answer is of course you couldn’t.”[119] 109. The suggestion that Mr Wong had deliberately misled the Court, knowing full well that his privilege argument was untenable was a grave allegation and, having carried significant weight with the Courts below, must have materially coloured their view of Mr Wong’s motives and conduct. It was, as appears from the foregoing, an unjustified misapprehension on the prosecution’s part. E.2d Lack of confidentiality 110. I turn to deal briefly with Stock VP’s third reason which is expressed as follows: “There could in the present case be no viable suggestion that any circumstance of confidence attached to the identity of the person who signed the deed of charge. The argument that privilege attached to the identity of Lau was a palpably unmeritorious one even if Lau rather than EC Textiles was the client. If EC Textiles was the client, it is even more difficult to see what possible interest the company could have in confidentiality of the identity of Lau. Further, EC Textiles and Lau must be taken to have known that the bank had a direct interest in being assured that the person signing the deed of charge was a bona fide signatory on behalf of EC Textiles; in other words, they must be taken to have known that that identity could not be kept confidential. The argument was a nonsense.”[120] 111. As Lord Taylor of Gosforth explained in R (ex p B) v Derby Magistrates’ Court,[121] LPP protects confidentialcommunications between a person and his legal advisers: “The client must be sure that what he tells his lawyer in confidence will never be revealed without his consent.” 112. It follows that a communication which is not made in confidence is not protected. As Lord Scott of Foscote stated in the Three Rivers case:[122] “... legal advice privilege arises out of a relationship of confidence between lawyer and client. Unless the communication or document for which privilege is sought is a confidential one, there can be no question of legal advice privilege arising. The confidential character of the communication or document is not by itself enough to enable privilege to be claimed but is an essential requirement.” 113. In the present case, there is nothing to suggest that anyone asked Ms Mak to keep the identity of Ms Lau as signatory of the deed confidential. The deed created a fixed and floating charge on ECT’s property and undertaking and it was clear, as stated in the bill of costs, that it was going to be registered in the Companies Registry and thus made a public document. The deed named Ms Lau as the person who signed it and stated that she had signed in the presence of a named solicitor who also signed as a witness. The evidence was that a member of the bank’s staff had accompanied Ms Lau to Ms Mak’s office to sign the deed. Ms Lau was asked for her identification document to enable Ms Mak to verify her identity, which Ms Lau evidently provided. And Ms Mak duly signed as witness to Ms Lau’s act of signing in Ms Mak’s presence, without anything arising to suggest that the identity of Ms Lau should be kept confidential. On the contrary, it must have been self-evident to everyone that a person asked to sign as a witness may later be asked to confirm the identity of the signatory. 114. Ms Clare Montgomery QC submitted on Mr Wong’s behalf that Stock VP should be understood as saying that confidentiality could not be maintained because it was a transaction involving the bank which necessarily knew and wished to be assured of the signatory’s identity. Her submission was that this was erroneous since the effect of such an arrangement was that while the information was not confidential or privileged as between ECT and the bank, it was confidential and privileged against the rest of the world, with the solicitor remaining obliged to protect the privilege unless it was waived by both ECT and the bank.[123] I do not read Stock VP’s judgment as being so limited. The point is simply that there was nothing to show that Ms Lau’s identity as signatory was confidential. E.2e Not legal advice 115. Returning to Stock VP’s first reason, in the Three Rivers case, Lord Scott suggested as a test for determining “whether the seeking of advice from or the giving of advice by lawyers does or does not take place in a relevant legal context so as to attract legal advice privilege” asking: “... whether the advice relates to the rights, liabilities, obligations or remedies of the client either under private law or under public law. If it does not, then, in my opinion, legal advice privilege would not apply. ...” 116. As indicated above, I agree with the Court of Appeal that Ms Mak’s intended evidence was not covered by LPP. Evidence that Ms Lau’s identity was verified and that she had signed the deed in Ms Mak’s presence does not touch on the obtaining or giving of legal advice as to anyone’s rights, liabilities, obligations or remedies. As Fulford J stated in R (Miller Gardner Solicitors) v Minshull St Crown Court:[124] “…the provision of an individual's name, address and contact number cannot, without more, be regarded as being made in connection with legal advice. It records nothing which passes between the solicitor and client in relation to the obtaining of or giving legal advice. ...” F. Whether Mr Wong may genuinely have believed that LPP existed 117. As noted above, the Judge found that Mr Wong had “little or no belief in the actual existence of legal professional privilege”, pointing to his having been “alerted to the problems by your fellow counsel and the exceptions referred to in Blackstone”. I have dealt with the Blackstone extract. It is with respect difficult to see any basis for suggesting that Mr Wong had been “alerted to the problems” by his fellow counsel. All that Mr Ngai had done was to cite an irrelevant case while expressing doubts as to its applicability.[125] There is no evidence that either Mr Ngai or Mr Tse had ever pointed out the real difficulties with the LPP argument. 118. Stock VP was uneasy about the Judge’s finding of lack of belief mainly because of the Note and Further Note written by Mr Wong containing his opinion that there was a privilege point to be taken or at least to be further explored. His Lordship decided that Mr Wong should receive the benefit of the doubt. I respectfully share that view. 119. The evidence provides a good picture of how the idea of an LPP claim developed. It will be recalled that having raised the LPP idea in his original Note, Mr Wong asked, in his Further Note dated 1 November 2006 for instructions from Ms Lau as to “the exact circumstances she was under (ie the communication between she and [Ms Mak]) leading to her signing [the deed]”. That resulted in him being shown the bill of costs which he evidently considered a breakthrough. This passage is from his re-examination: “Q: .... Try and tell us, if you can, when you first were given a copy of the fee note. A: Yes, that would be to the best of my memory a few days before the sending out or the drafting of the letter on 24 November. Q: So way after the notes that were passing between counsel. A: Yes, yes. It’s never mentioned in the notes there was a bill of costs. Q: What difference, if any, to the strength of your LPP argument did the revelation of the fee note make? A: That would. Substantially, I believe. It’s not simply an assertion by [Ms Lau] which she might have to substantiate by giving evidence on a voir dire, but that would be documentary evidence which I believe [Ms Mak] will accept as a genuine fee note from her firm. Q: Can you remember how the news of the discovery of this fee note was greeted by you[r] team of counsel? A: I remember that everybody was thrilled about the production of a fee note by [Ms Lau]. Q: Did it in your view increase or decrease the chances of you mounting a reasonable argument? A: Increase. Substantially increase.”[126] 120. It appears that, buoyed by discovery of the bill of costs, little thought was given as to whether the intended evidence itself was covered by LPP. As we have seen, in his evidence, Mr Wong adhered to an incoherent theory that there must have been prior privileged communications between Ms Mak and Ms Lau and that the execution of the deed, including the identity of the person signing, was somehow inextricably bound up with those supposedly privileged communications. 121. There can be no doubt that Mr Wong’s level of competence as a lawyer was low. In particular, he had a poor grasp of LPP as a legal doctrine.[127] In cross-examination, Mr Wong stated that he could see no difference between confidentiality and LPP and thought that a court could override them both: “Q: No, this is confidentiality not LPP? A: Well I see no difference. ... Q: No, LPP is different, isn’t it, because LPP the court cannot order that the matters be disclosed if they are coved by privilege, confidentiality the court can, that’s the difference, isn’t it? A: Well I don’t think so, I don’t think so.” .... Q: .... there are two things there, confidentiality and privilege, they’re different? A: Yes, yes different but the effect of a court order is the same. The court, the order can override the privilege as well as the confidentiality, that’s my understanding.”[128] 122. Mr Clive Grossman SC, called as a character witness for Mr Wong described him as “a man of integrity who worked hard for his client, but ... not of the highest intellect.”[129] Another character witness, Mr Philip Dykes SC said he was “a man of integrity whose style was enthusiastic and combative” adding that “he had spoken to [Mr Wong] on occasion and advised [him] to moderate [his] approach”.[130] 123. Stock VP pointed out that: “One has in a case such as this to take the greatest care to distinguish between misguided professional enthusiasm or even incompetence, on the one hand and, on the other, dishonesty.”[131] 124. His Lordship continued: “...although it is clear enough that Wong wished, if possible, to avoid a contested argument in court on the issue of privilege, and although I have not had the advantage, as did the trial judge of hearing the evidence, I still retain some doubt if the suggestion be that Wong had no belief at all in the point. That doubt arises from the evidence of Wong’s aggressive tenacity on behalf of his clients and of the evidence which suggests that he is a facts advocate, not much at ease with arguments of law.”[132] 125. The opinion of Mr Dykes SC that Mr Wong was “a competent lawyer, well able to look up the law”[133] does not appear to be borne out by the evidence. Mr Wong does not appear to have done more than look at Blackstone and, when asked by the Judge to produce authority overnight, only managed to re-cycle the case which had been mentioned, with reservations, by Mr Ngai in an earlier note. 126. The picture that emerges is therefore of a barrister of low competence with a poor understanding of the relevant concepts; doing no effective research; “thrilled” to have discovered the bill of costs, which was regarded as a justification for pursuing the LPP argument; coupled with an aggressive tenacity reflected in the three letters sent to Ms Mak – an unedifying vision, but distinctly more plausible, in my view, than the prosecution’s theory of a barrister well aware of the law but cynically using LPP as a pretext for what in truth was what the Judge had called “a threat simpliciter” aimed at deflecting Ms Mak from her duty. 127. I pause here to mention an argument advanced by Ms Montgomery QC concerning LPP as it affected Mr Wong’s defence. As we have seen,[134] before commencement of the trial in the present case, Mr Wong’s solicitors wrote to Ms Lau asking whether she would waive privilege regarding privileged communications made in the course of Mr Wong acting for her and no waiver was forthcoming. 128. In such circumstances, Ms Montgomery prayed in aid Medcalf v Mardell,[135]applied in Yau Chiu Wah v Gold Chief Investment Ltd,[136] for the proposition that: “...in the absence of the full facts due to the lay client's refusal to waive privilege, the Court is not entitled to speculate and infer that there could not have been any material upon which the legal representatives could have been justified in taking their course of action, that the benefit of the doubt has to accrue to the legal advisers and that ‘only rarely will the court be able to make “full allowance” for the inability of the practitioner to tell the whole story or to conclude that there is no room for doubt.’: [Medcalf v Mardell ] at 135D-136B.”[137] 129. I do not consider that principle applicable in the present case. True it is that Ms Lau has not waived her privilege. But it is perfectly clear that Mr Wong did not feel in any way constrained by the absence of his client’s waiver. He did not hesitate to disclose, directly and indirectly, the instructions sought and received, the discussions with his legal team and the advice which he rendered. He was not stopped by the Judge. Whether or not his disclosures put him in breach of his duties to his client, he certainly has no basis for complaining that the non-waiver of privilege has somehow constricted his defence. It is tempting to regard Mr Wong’s apparently cavalier approach to his own LPP obligations as another manifestation of his lack of understanding of what LPP involves. G. The purpose of Mr Wong’s threats 130. The evolution of the prosecution’s case regarding Mr Wong’s purpose in sending the threatening letters has been traced in Section D.3 above. It is not surprising that the initial allegation that his object was to get Ms Mak not to come to court was effectively abandoned and replaced for all practical purposes by the allegation that his objective was to get her to raise an LPP objection in an application to the Court since this would improve the chances of success. 131. The original allegation was unsustainable. As we have seen, the decision to write the letters evidently hinged on discovery of the bill of costs a few days before the 1st Letter was sent. By that time, the trial had been going for perhaps a week. Mr Wong’s evidence was that he believed that a witness summons would already have been served on Ms Mak, his experience being that such summonses are usually served a month to six weeks before a trial.[138] 132. Upon receiving the 1st and 2nd Letters, Ms Mak’s response was to point to the witness summons and to say: “... unless and until you have successfully obtained such order to set aside the said witness summons served on me, I would be obliged by law to be a witness in the court unless otherwise ordered.”[139] 133. Mr Wong testified in chief that he did not ask for a copy of the summons: “... because I had no doubt that the summons would have been served so – and I’ve no doubt about [Ms Mak’s] explanation that she had the summons to obey, I have no doubt that she must be telling the truth for that and also in the penultimate paragraph, my understanding is that [Ms Mak] is inviting us to maybe to voice out our complaint or our allegation in court ....”[140] 134. The following exchanges are also from Mr Wong’s examination in chief: “Q: .... when you drafted and caused to be sent out the letter of 24 November, what do you think [Ms Mak] might do? A: .... knowing that she was a solicitor, she might, I think she most likely will inform the prosecution about receiving the letter and most probably she would invite the view of the prosecution as to whether she would have to testify and of course she as a lawyer may form her own view whether she would be breaching the [LPP].” ... Q: Was it your intention to try to stop her from going to court? A: No, well my understanding of the purpose of the witness summons the same as [Ms Lau] that she’s obliged as per the law to attend court.”[141] .... “Q: When you wrote those letters to [Ms Mak], ... who did you think would be the final judge or arbiter of whether or not [LPP] was the subject of a proper claim? A: As always is the trial judge. ... Q: Did you ever intend, as is suggested in the charge, that [Ms Mak] should not go to court to testify? A: No.”[142] 135. Mr Wong gave the same evidence in cross-examination: “Q: So how can you explain what you did, except in terms of an attempt to dissuade her from coming to court. That’s what it was, wasn’t it? It was an attempt to dissuade her from coming to court. A: I disagree. ... I never attempted to persuade her not to come to court. My understanding of what likely would happen is: if she accepted our view on the LPP, she would still have to attend court to explain why she’s not able to testify. She has already been issued, according to her, well as far as to my understanding, the witness summons. She had no choice. I think she also made it clear in her reply that she had to attend court. I am referring, your Honour, to her letter of 30 November ...”[143] 136. Unlike the Court of Appeal, I consider the prosecution’s change of case and the evidence underlying that change to be crucial to the finding of a prohibited tendency and thus to the outcome of this appeal. H. Whether it was conduct with the prohibited tendency H.1 The Judge’s decision 137. As discussed in Section D.4 above, the Court of Appeal rejected the three grounds upon which the Judge founded his conclusion that Mr Wong’s conduct had the tendency to pervert the course of justice. Those grounds were (i) Mr Wong’s approach to a prosecution witness per se; (ii) his failure to inform the prosecution; and (iii) his failure himself to raise the issue with the Judge at the criminal trial. I respectfully agree that for the reasons given by Stock VP,[144] the Judge’s conclusions could not stand. 138. As to point (iii), I would add that I do not agree with the Judge’s view that in relation to the privilege claim “the parties must perforce be the prosecution and the defence; not defence and a prosecution witness” and that it was therefore improper to have communicated with Ms Mak. The “parties” to a legal advice privilege claim are the client and the solicitor. On the basis of what Mr Wong may genuinely have believed, the parties were ECT and Ms Mak, hence the debate in the impugned letters about who should make the application. Of course, given that Ms Mak was (as she had pointed out) bound to obey the witness summons, a challenge to her giving evidence on LPP grounds would inevitably have had to be brought at some point before the Court which had issued the summons. But that is not a reason for finding that communication between Mr Wong acting for ECT and Ms Mak was improper because “the parties must perforce be the prosecution and the defence”. H.2 The Court of Appeal’s decision 139. The critical question for the outcome of this appeal is whether the Court of Appeal’s substituted basis (set out in paragraphs 74, 168 and 169 of its judgment) for deciding that Mr Wong’s conduct had the prohibited tendency is correct. The basis contended for by the respondent overlaps with and is comprehended within the Court of Appeal’s substituted basis and can be dealt with simultaneously. It is appropriate to begin with paragraph 74, read together with paragraph 73. H.2a The substituted basis as put in paragraph 74 140. In paragraph 73, Stock VP cites the following passage from the joint judgment of Brennan and Toohey JJ in R v Rogerson:[145] “The course of justice consists in the due exercise by a court ... of its jurisdiction to enforce, adjust or declare the rights and liabilities of persons subject to the law in accordance with the law and the actual circumstances of the case. The course of justice is perverted (or obstructed) by impairing (or preventing the exercise of) the capacity of a court ... to do justice. The ways in which a court ... may be impaired in (or prevented from exercising) its capacity to do justice are various. Those ways comprehend, in our opinion, erosion of the integrity of the court..., hindering of access to it, deflecting applications that would be made to it, denying it knowledge of the relevant law or of the true circumstances of the case, and impeding the free exercise of its jurisdiction and powers including the powers of executing its decisions. An act which has a tendency to effect any such impairment is the actus reus of an attempt to pervert the course of justice.” 141. His Lordship states at paragraph 74: “It must also follow that the course of justice is deflected by causing applications to be made to a court which, but for improper pressure, would not be made. Once legal proceedings are in motion they should be permitted ‘to flow unobstructed and undiverted… justice should be administered in the way which is ordinarily pursued’: R v Taffs [1991] 1 NZLR 69. ” The analysis was further developed in paragraphs 168 and 169, but it is instructive to begin by taking paragraph 74 on its face. 142. The first two sentences in the abovementioned citation from Rogerson are important. For conduct to have a tendency to pervert the course of justice, it must be conduct which tends to cause a miscarriage of justice, in other words, a tendency towards “impairing (or preventing the exercise of) the capacity of a court or competent judicial authority to do justice”.[146] And it must involve a tendency to cause a miscarriage of justice in curial proceedings. In their Honours’ words, it must tend to interfere with “the due exercise by a court ... of its jurisdiction to enforce, adjust or declare the rights and liabilities of persons subject to the law in accordance with the law and the actual circumstances of the case”. Conduct with a tendency to cause unlawfulness outside of curial proceedings does not constitute the offence.[147] 143. In the second part of the Rogerson citation, Brennan and Toohey JJ gave as examples of interference with the due administration of justice: “...erosion of the integrity of the court ..., hindering of access to it, deflecting applications that would be made to it, denying it knowledge of the relevant law or of the true circumstances of the case, and impeding the free exercise of its jurisdiction and powers including the powers of executing its decisions.” 144. With respect, I do not accept Stock VP’s suggestion that “It must also follow” from the passage cited that “the course of justice is deflected by causing applications to be made to a court which, but for improper pressure, would not be made”. Where an application is made to the Court, it is determined by the Court “in the due exercise ... of its jurisdiction to enforce, adjust or declare the rights and liabilities of persons subject to the law in accordance with the law and the actual circumstances of the case.” 145. Conduct “deflecting applications that would be made to [the court]” aims to prevent the Court from exercising such jurisdiction and so may deny the would-be applicant justice. That is why it may rank as a perversion of the course of justice. But causing an application to be made to and duly determined by the Court plainly does not have the same effect. The Court’s capacity to do justice is not in any way impaired. Pressure on the intended applicant to bring the application may be a troubling nuisance to the person subjected to that pressure. It may even, in some circumstances, involve a form of unlawful intimidation. But such pressure does not interfere with the capacity of the Court to administer justice in the curial proceedings. 146. Stock VP refers in paragraph 74 to R v Taffs,[148] for the proposition that legal proceedings, once in motion should be permitted “to flow unobstructed and undiverted… justice should be administered in the way which is ordinarily pursued”. Taffs was however a case of a wholly different character. A barrister and solicitor instructed to defend a person charged with robbing a boy at knifepoint threatened the boy’s mother, saying that if the boy persisted in giving his evidence, he would, amongst other things, “mince the boy up in Court tomorrow”, “crucify him”; and cause him to be “publicly humiliated as a liar and a homosexual”. It was thus a case where the defendant was seeking to obstruct justice by dissuading a witness from coming to testify before the Court. 147. The defendant in Taffs claimed that he made those threats because he genuinely believed that the boy’s intended evidence was false. That was held to be no excuse, Cooke P pointing out that: “It would be dangerous to allow a lawyer, perhaps uncritically espousing his client's case, to threaten to use legal proceedings to publicly humiliate the adversary. To leave the lawyer free to utter such threats, provided only that he genuinely believes his client to be in the right, would savour of transferring the responsibility of judging the case from the Court to the legal representatives of the parties.”[149] 148. In the present case, as the prosecution belatedly accepted, there was no intention to dissuade Ms Mak from coming to court or to disobey the witness summons. Neither did anyone suggest that if she were to testify, she should not tell the truth or should mislead the Court. Nor did anyone suggest that she should conceal the fact that a demand had been made that she take an LPP objection. Mr Wong testified that he expected Ms Mak to inform the prosecution which would obviously be able to resist the application if it saw fit. Mr Wong’s purpose was to get Ms Mak to raise the LPP objection with the Court and he acknowledged that it was the Court which would be the final arbiter of whether the objection was valid. This case therefore did not pose a risk of allowing a lawyer to arrogate to himself instead of the Court responsibility of judging whether the privilege objection was good. If, on hearing the application, the Court had upheld the privilege, it would have been deprived of Ms Mak’s evidence by virtue of the law of privilege and not because the course of justice had been perverted. H.2b How the respondent puts its case for the prohibited tendency 149. Before going on to consider paragraphs 168 and 169 in the Court of Appeal’s judgment, it is convenient to consider the way that the respondent puts its case on the prohibited tendency. 150. Mr Jonathan Caplan QC encapsulated the respondent’s case concisely, submitting that the letters possessed the prohibited tendency because of the pressure engendered by their “timing, tone, purpose and content”. He argued that: “Even if the Appellant sought to pursue a legitimate aim by an appropriate avenue it did not entitle him to write a series of letters which by their tone, timing, enclosures and threatening content brought improper pressure to bear upon the witness.”[150] 151. Such pressure, he submitted, was exerted to compel Ms Mak to make a setting aside application which she did not believe in: “... the prosecution case was based on the content of the letters, their timing, and the fact that the Appellant intended thereby to pressure the witness to claim privilege and set aside her witness summons when she had no belief in such a course of action...”[151] 152. It will be apparent that this line of argument suffers from the deficiency identified above in the discussion of paragraph 74 of the Court of Appeal’s judgment. The timing, tone, purpose and content of the impugned letters and the pressure they exerted do not constitute the offence unless they had a tendency to impair the capacity of the Court to administer justice in the case. Pressure directed at compelling Ms Mak to place the privilege point before the Court for its ruling does not have that tendency. H.2c The substituted basis developed in paragraphs168 and 169 153. Stock VP acknowledged the aforesaid argument which he encapsulated as follows: “...to seek to preclude the giving of evidence, when its provision as evidence would breach legal professional privilege, can hardly constitute impermissible conduct; it could hardly have a tendency to pervert the course of justice since the preclusion of that evidence would be a product of the proper application of the doctrine of legal professional privilege and an attempt to preclude it on that ground is therefore incapable of constituting an attempt to divert the course of justice.”[152] 154. His Lordship, however, thought it was “off the mark”, stating: “A claim for privilege is indeed consistent with the course of justice and that is so even where the court’s search for the facts relevant to determination of the issue at hand may be impeded by the withholding of evidence on that basis. But – and this is the point – the machinery, the route, by which that claim comes before a court is itself also part of the course of justice.”[153] 155. The point was elaborated in paragraphs168 and 169 as follows: “168. To seek to force a third-party witness by threats, intimidation, oppression or harassment (as opposed to advice or persuasion or court order) to apply to set aside a summons on the grounds of legal professional privilege when such an application is against that person’s wishes or that person’s belief in the existence of the privilege is – if the attempt were to succeed – to divert the course of justice, because it is to force the presentation of an application as if it were voluntarily made, to force the putative witness to assert that she has no material evidence which she may legitimately provide, when patently the witness believes she has material evidence the provision of which is not precluded by law and when, furthermore, the claim of privilege is advanced by a solicitor, for such a claim from such a quarter is likely to carry considerable weight; more weight on its face than if the claim were made by the client. 169. It matters not for present purposes whether the application is one to set aside a witness summons or is an application for an injunction. Both are part of the course of justice. But the course of justice requires that neither application should be made as a result of unlawful or improper pressure or harassment. Therein lies the heart of this case against Wong.” 156. It was by means of this argument that the Court of Appeal sought to enhance and transform the impact of the letters. From the letters having a tendency limited to placing unwelcome but non-curial pressure on Ms Mak to make an application, the argument seeks to translate the pressure into conduct having a tendency to impair the Court’s capacity to do justice within the curial proceedings. It seeks to achieve this by postulating what, with respect, seem to be some highly artificial reactions on Ms Mak’s part to the demands made in the letters. 157. The argument postulates first, that the letters had the tendency “… to force the presentation of an application as if it were voluntarily made”. The suggestion is therefore that the pressure was likely to result in the Court being given a misleading impression. However, I am with respect unable to see why the pressure should be thought by Mr Wong or anyone else to be likely to have that result. Ms Mak had made her unwillingness to make the application clear in her responses to the 1st and 2nd Letters. She had insisted that it was up to Ms Lau’s lawyers to make the application if they thought fit. There was no reason to think that her attitude would change so that she would present herself to the Court as voluntarily making the application. We know that she had in fact complained about the letters to the police and to prosecuting counsel and that as soon as she entered the witness-box, she complained to the trial Judge about having received the letters, saying that she was troubled by their contents and by the complaints made to the Law Society against her integrity. Prosecution counsel made submissions condemning the conduct complained of and, as is obvious, it led to proceedings being instituted against Mr Wong. 158. The Court of Appeal’s substituted argument next postulates that Ms Mak would have been forced “to assert that she has no material evidence which she may legitimately provide, when patently the witness believes she has material evidence the provision of which is not precluded by law”. This is, with respect, a somewhat loaded way of saying she would have been “forced to assert LPP which she does not believe in”. The expression “no material evidence” comes from section 34(2) of the Criminal Procedure Ordinance which enables a person served with a witness summons to apply for its discharge by satisfying the court “that he cannot give any material evidence or, as the case may be, produce any document or thing likely to be of material evidence”. In the 2nd Letter, Mr Wong had suggested to Ms Mak that section 34(2) made her the appropriate person to make the application and that in so doing, she would be claiming that since her proposed evidence was privileged, she would have no material evidence to give, satisfying the requirements of the section. 159. So formulated, this second postulation takes the case no further than the proposition contained in paragraph 74 discussed above. Difficulty with the realism of the second scenario depicted also arises. It seeks once again to suggest that the course of justice would be perverted because the letters would have the effect of giving the Court a false impression because it envisages Ms Mak acting as an advocate of the privilege objection, contrary to her own beliefs. For the reasons stated above, there is no reason to assume that Ms Mak would fall into line with Mr Wong’s views on privilege rather than make it clear that she had been pressured to make the application against her own better judgment. 160. Thirdly, the Court of Appeal postulates that the course of justice was likely to be perverted since a claim of privilege advanced by a solicitor “is likely to carry considerable weight; more weight on its face than if the claim were made by the client”. That again presupposes that Ms Mak would supinely present the appearance of believing in the application rather than complain about having been pressured to make an application that she does not believe in. The latter course was far more likely and upon such a complaint being made, the application’s chances of success would, if anything, be likely to have been diminished. 161. Quite apart from the factual implausibility of the Court of Appeal’s substituted basis, it contains, in my view, an assumption that is in principle inappropriate in a case like the present. An essential aspect of the prohibited tendency as found by the Court of Appeal involves deprivation of the witness’s freedom to choose to give the intended evidence. 162. It is true that in Egan, after citing R v Meissner[154] which saw the prohibited tendency in that case as a tendency to “prevent the accused from making a free and voluntary choice concerning his or her plea to the charge”, I stated: “That was a case involving pressure on an accused person to plead guilty, but the propriety of the means used to influence a witness may also usefully be tested by asking whether the methods used leave the witness genuinely ‘free to make the choice’ as to whether to give evidence and what evidence to give.”[155] 163. That approach necessarily can only apply where the witness enjoys a free choice of whether or not to give evidence. I agree with Ms Montgomery that the “free choice” or “free will” test cannot appropriately be used as an indicator of the prohibited tendency where there is, or the defendant believes that there is, a bona fide arguable claim that the witness is under a legal duty by virtue of LPP not to give the proposed evidence. Ms Montgomery puts the point as follows: “A witness (particularly a legal professional) who is bound by LPP is both legally and professionally required to assert that privilege and should not proceed according to his preference, this is an intrinsic legal consequence of LPP enforcement. The expert witness in Protec Pacific[156] was clearly unhappy with the letters before action and the proceedings against him, they were clearly ‘intended to undermine [his] free will and choice’ as to how to proceed, but the proceedings were a legitimate means of enforcing a right to confidence and could not give rise to a criminal perversion of the course of justice. In this regard, the Court of Appeal erred in concluding that conduct pertaining to the enforcement of LPP which seeks to compel a witness to comply with her duty, is unlawful and improper unless the conduct consists only of advice, persuasion or a court order [CA§168].”[157] 164. For the foregoing reasons, I respectfully disagree with the Court of Appeal’s view that, on their substituted basis, the threats made in the letters constitute improper means with the prohibited tendency. In my judgment, the threats (both of civil proceedings and a complaint to the Law Society) made with the intention of compelling Ms Mak to raise LPP as a ground for not giving the intended evidence did not have the prohibited tendency and were incapable of constituting the actus reus of the offence. 165. In the light of my conclusions and the general acceptance that Mr Wong’s purpose was to press Ms Mak to make a court application, the question of mens rea requires no further discussion. Conclusions 166. I have accordingly reached the following conclusions: (a) There was ample basis for Mr Wong to believe that an implied retainer had arisen between Ms Mak and ECT. (b) While there was no objective basis for holding that the intended evidence was covered by LPP, the evidence justified the Court of Appeal in reversing the Judge and giving Mr Wong the benefit of the doubt as to his belief in the existence of LPP. (c) Acting with such belief, he caused the threatening letters to be sent for the purpose of pressuring Ms Mak to raise the LPP claim by making an application to the Court. (d) Applying pressure for that purpose did not constitute conduct with a tendency to interfere with the trial Court’s capacity to administer justice and therefore did not have the prohibited tendency. (e) I agree with the Court of Appeal’s reasons for reversing the Judge’s finding of a prohibited tendency but do not accept its substituted basis for such finding. (f) Likewise, I do not accept the respondent’s case on the prohibited tendency. 167. I would therefore allow the appeal and quash Mr Wong’s conviction. I would also direct that any submissions as to costs be lodged with the Registrar in writing within 14 days from the date of this judgment. Postscript 168. Before leaving this judgment, it is worthwhile re-iterating what I said in Egan about approaching witnesses or potential witnesses: “... great care must be taken if an approach is to be made with a view to influencing a witness in relation to his or her anticipated evidence. Such an approach can only properly be made if its purpose is lawful and if no unlawful or improper means are used to secure that purpose.”[158] 169. As the Court of Appeal pointed out: “The guidance provided by the Law Society of Hong Kong ... warns, at Chapter 10.12, that: ‘A solicitor should be aware that in seeking to exercise his right to interview a witness who has already been called by the other side or who to his knowledge is likely to be called by them, he may be exposed to the suggestion that he has tampered with the evidence. This may be so particularly where the witness subsequently changes his evidence. It may be wise in the circumstances for the solicitor to offer to interview the witness in the presence of a representative of the other side.’” 170. That is sound advice. Even if, as the Court of Appeal and I have held, there is no legal requirement that the other side be informed of an approach made to a witness or potential Wong, an unannounced approach to the other side’s witnesses in an adversarial system is bound to arouse suspicion. Not having been given notice of the approach and not being aware of what was said to the witness, concern may naturally arise as to whether some form of impropriety could have been involved. 171. Mr Wong has had occasion 	repeatedly to acknowledge that it would have been far better for him to have 	informed the prosecution of his LPP concerns and to have told them that he 	had written to Ms Mak on behalf of ECT to get her to assert the privilege 	which he believed existed in ECT’s favour. Not having taken such 	elementary precautions, Mr Wong found himself, over a period of some four 	years, having to cope with serious criminal charges carrying potentially 	disastrous professional consequences and having guilty verdicts entered 	against him in two courts. This judgment should therefore certainly 	not be taken as encouragement for practitioners to adopt the kind of course 	followed by Mr Wong. Mr Justice Tang PJ: 172. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Mortimer NPJ: 173. For the reasons given by Mr Justice Ribeiro PJ, I also would allow this appeal and quash the appellant’s conviction for attempting to pervert the course of justice. Sir Thomas Gault NPJ: 174. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Chan PJ: 175. The Court unanimously allows the appeal and quashes Mr Wong’s conviction. We direct that costs should be dealt with as set out in paragraph 167 of this judgment. Ms Clare Montgomery QC, Mr Graham Harris SC and Mr Benson Tsoi, instructed by Haldanes for the Appellant Mr Jonathan Caplan QC and Ms Charlotte Draycott SC instructed by the Department of Justice for the Respondent [1] DCCC 960/2009 (19 May 2010). [2] CACC 203/2010, Stock VP, Kwan and Lunn JJA (2 March 2012). [3] Maruyama Texknit (Hong Kong) Co Ltd and Maruyama (Asia) Co Ltd. [4] DCCC 320/2005. [5] [2002] EWHC 60. [6] Addressed to her c/o Messrs Lily Fenn & Partners (“LF&P”), where she then worked as a consultant. Her former firm of CMK&Co had split up and Ms Mak stayed on with the firm which became known as LF&P. [7] At §21. [8] Set out in full in the Court of Appeal’s judgment at §25. [9] See the full text at §26 of the Court of Appeal’s judgment. [10] See the Court of Appeal’s judgment §29 for the full text. [11] Cap 221. Section 34(2): “If any person in respect of whom a witness summons has been issued applies to the court out of which the summons was issued and satisfies it that he cannot give any material evidence or, as the case may be, produce any document or thing likely to be of material evidence the court may direct that the summons shall be of no effect.” [12] [1996] AC 487. [13] See Court of Appeal §31 for the full text. [14] See §37 of the Court of Appeal for the full text. [15] No submissions were made as to the crime and fraud exception, so further mention of it is unnecessary. [16] CACC 176/2010, Stock VP, Kwan and Lunn JJA (2 March 2012) at §§212-213, 217-220. [17] (2010) 13 HKCFAR 314 at §§122-139. [18] Egan §122, citing R v Vreones [1891] 1 QB 360. [19] Egan §123, citing R v Machin [1980] 1 WLR 763 at 767; R v Toney [1993] 1 WLR 364 at 367; R v Rogerson (1992) 174 CLR 268 at 279, 297; R v Meissner (1995) 184 CLR 132 at 141. [20] Egan §124, citing R v Machin [1980] 1 WLR 763; HKSAR v Wong Shing Yim Peter [2003] 3 HKLRD 1046 at §19; and R v Rogerson (1992) 174 CLR 268 at 280. [21] Egan §126, citing R v Vreones [1891] QB 360 at 369; and R v Rogerson (1992) 174 CLR 268 at 275-276. [22] Egan §126. [23] Egan §125, citing R v Rogerson (1992) 174 CLR 268 at 280-283; HKSAR v Wong Shing Yim Peter [2003] 3 HKLRD 1046 at §29. [24] Egan §132, citing Harmony Shipping Co SA v Saudi Europe Line Ltd [1979] 1 WLR 1380 at 1384-1385; Kwan Fung Kam v AG [1990] 2 HKC 577; and Connolly v Dale [1996] QB 120. [25] Egan §§128-129, citing [1976] 1 QB 372 at 383-384. [26] At §134. [27] Egan §135, citing R v Kellett [1976] 1 QB 372 at 386; R v Silverman (1908) 14 CCC 79 at 81-82. [28] Egan §137, citing R v Kellett [1976] 1 QB 372 at 391-392; Librizzi v State of Western Australia [2006] WAR 104 at §80. [29] Egan §138, citing R v Silverman (1908) 14 CCC 79 (Ontario Court of Appeal) and R v Taffs [1991] 1 NZLR 69 at 73. [30] Egan §139, citing R v Kellett [1976] 1 QB 372 at 392-393. [31] Egan §§135-136, citing R v Meissner (1995) 184 CLR 132 at 143. [32] [1996] AC 487 at 507. See also Three Rivers District Council v Governor and Company of the Bank of England (No 6) [2005] 1 AC 610 at §§30-33, citing leading authorities. [33] At 503. [34] At 504-505. [35] [2005] 1 AC 610 at §10. [36] Ibid, at §26. [37] Ibid, at §25. In Hong Kong, the extent to which it can be overridden by statute may raise constitutional issues in the light of Article 35 of the Basic Law. [38] [1939] 1 KB 194 at 222. This decision was overtaken and effectively overruled by later authority on the question of whether a client has a cause of action in tort against his solicitor. However, Scott LJ’s recognition of implied retainers is unaffected. [39] (1996) 14 WAR 92 (Supreme Court of Western Australia). [40] At 102. [41] Ibid. [42] [2010] VSCA 335 at §111. [43] §3.50 (4th ed, 2010). [44] At §117. [45] (1999) 48 NSWLR 1 at 53 per Spigelman CJ, Sheller and Stein JJA. [46] [2004] 2 HKLRD 937. [47] At §19. [48] Cited as HKSAR v Kanjanapas, Chong Kwong Derek and others,Ma CJHC, Tang VP and Wright J, CACC 248/2006 (12 February 2009). [49] Reasons for Verdict RV86. I have to confess that I do not understand what is intended by the phrase “a cogent milestone on the journey”. If the existence or otherwise of an implied retainer is a “cogent milestone”, does that not suggest that it has some unspecified legal significance? [50] Court of Appeal §106. [51] Transcript T176, T185, T188, T192. [52] T185. [53] Court of Appeal §110. [54] RV103. [55] RV94. [56] RV103. [57] RV104. [58] Court of Appeal §116, citing R (Miller Gardner Solicitors) v Minshull St Crown Court [2002] EWCH 3077 at §20; and Pascall v Galinksi [1970] 1 QB 38 at 44. [59] Court of Appeal §§117-120, citing Miley v Flood [2001] IEHC 9; Federal Commissioner of Taxation v Coombs (1999) 164 ALR 131 at §31; Re Ontario Securities Commission (1983) 146 DLR (3d) 73; Rosenberg v Jaine [1983] NZLR 1 at 6; Police v Mills [1993] 2 NZLR 592 at 597-8; and United States of America v Hodge and Zweig 548 F 2d 1347 (1977) at 1353. [60] Court of Appeal §122. [61] Ibid. [62] Court of Appeal §171. [63] Court of Appeal §172. [64] See Section A.2 of this judgment. [65] Court of Appeal §173. [66] Court of Appeal §172. [67] Court of Appeal §§173-174. [68] Court of Appeal §§174-177. [69] T148. [70] T284. [71] T308. [72] T363. [73] Ibid. [74] RV95. [75] RV96. [76] RV105. [77] RV106. [78] RV107. [79] He was also simultaneously dealing with Mr Yeung’s appeal. [80] Court of Appeal §7. See also §87 where the prosecution case was stated to be that Mr Wong “exerted improper pressure upon her to take a course which, but for that pressure, she had no intention of taking; namely, to claim privilege on behalf of [Ms Lau] or [ECT].” [81] Court of Appeal §162. [82] Court of Appeal §201. [83] Court of Appeal §168. [84] RV82. [85] RV97. [86] RV89. [87] RV94. [88] RV100-102, 106 and 109. [89] Court of Appeal §§138-142. [90] At §132. [91] Harmony Shipping Co SA v Saudi Europe Line Ltd [1979] 1 WLR 1380 at 1384-1385. [92] Court of Appeal §142. [93] Cap 221. Section 34(2): “If any person in respect of whom a witness summons has been issued applies to the court out of which the summons was issued and satisfies it that he cannot give any material evidence or, as the case may be, produce any document or thing likely to be of material evidence the court may direct that the summons shall be of no effect.” [94] Court of Appeal §129. [95] [2000] 3 HKLRD 351 at 355. [96] Court of Appeal §130. [97] Court of Appeal §§132-133. [98] (1992) 174 CLR 268 at 280. [99] In Section B.1b. [100] T182-183. [101] “I would also remind him or her that he or she can seek independent legal advice ... if he or she so wishes.” T181 [102] T148. [103] Court of Appeal §197. [104] RV94. [105] RV103. [106] Court of Appeal §172. [107] See for instance, Bursill v Tanner (1885) 16 QBD 1, where a judgment creditor wanted to know the identity of the trustees of a marriage settlement in the hope of levying execution against a married woman who had been successfully sued. Or, for instance, a desire to know who is really on the other side in litigation: Levy v Pope (1829) M & M 410 ; 173 ER 1206. Such information would obviously be important to be able to enforce costs orders. [108] [1999] 92 FCR 240, Sundberg, Merkel and Kenny JJ. [109] At §31. [110] See Section A.2 above. [111] Court of Appeal §§40-50. [112] At §49, italics supplied. [113] 24.1.06 Transcript p 3 (T3). [114] 24.1.06, T13. [115] See Section A.2 above. [116] Often linked to the adversarial nature of litigation, as explained in Waugh v British Railways Board [1980] AC 521 at 536. See also In re L (A Minor) (Police Investigation: Privilege) [1997] AC 16 at 26; and Three Rivers District Council v Governor and Company of the Bank of England (No 6) [2005] 1 AC 610 at §52. [117] T261. [118] T308. [119] T350. [120] Court of Appeal §122. [121] [1996] AC 487 at 507. [122] Three Rivers District Council v Governor and Company of the Bank of England (No 6 [2005] 1 AC 610 at §24. [123] Citing Hellenic Mutual War Risks Association (Bermuda) Ltd v Harrison (The Sagheera) [1997] 1 Lloyd’s Rep 160; and The TAG Group Litigation Winterthur Swiss Insurance Company v AG (Manchester) Ltd (in liquidation) [2006] EWHC 839. [124] [2002] EWCH 3077 (Divisional Court) at §20. [125] See Section A.2 above. [126] T330. [127] See Section B.2a above. [128] T278-279. [129] RV26. [130] RV26. [131] Court of Appeal §172. [132] Court of Appeal §173. [133] RV26. [134] Section A.4 above. [135] [2003] 1 AC 120. [136] [2003] 3 HKLRD 553. [137] Appellant’s printed case §2.140. [138] T235. [139] See Section A.2 above. [140] T235. [141] T236. [142] T250. [143] T284-285. [144] See Section D.4b above. [145] (1992) 174 CLR 268 at 280. [146] Egan §124, citing R v Machin [1980] 1 WLR 763; HKSAR v Wong Shing Yim Peter [2003] 3 HKLRD 1046 at §19; and R v Rogerson (1992) 174 CLR 268 at 280. [147] Egan §126, citing R v Vreones [1891] QB 360 at 369; and R v Rogerson (1992) 174 CLR 268 at 275-276. [148] [1991] 1 NZLR 69. [149] At 73. [150] Respondent’s printed case §25. [151] Respondent’s printed case §27(a). [152] Court of Appeal §164. [153] Court of Appeal §166. [154] (1995) 184 CLR 132 at 143. [155] Egan at §136. [156] A reference to Protec Pacific Pty v Brian Cherry [2008] VSC 76, where the court restrained a party’s former expert witness from discussing the case with the other side. [157] Appellant’s printed case §2.73. [158] At §134. 
Mr Justice Ribeiro PJ: 1. In these appeals, the proper approach to appointing a guardian ad litem in respect of a mentally incapacitated person falls to be considered. A. The background 2. The plaintiff is a licensed money lender. It lent the defendant (“Mr Law”) $30,000 (with interest at 47.4% per annum). When he defaulted in making repayments, it brought proceedings in the District Court,[1] obtaining judgment against him, a charging order on his flat and an order for vacant possession and sale,[2] all in default of acknowledgment of service by Mr Law. After the plaintiff had obtained vacant possession and entered into a provisional agreement for the sale of the flat, Madam Cheung, Mr Law’s wife, approached the Court contending that Mr Law was a mentally incapacitated person. With the help of solicitors, she issued a summons seeking orders that she be appointed his guardian ad litem and that the judgment and consequential orders be set aside, claiming that Mr Law did not understand the loan transaction. Medical reports by a psychiatrist and a clinical psychologist were filed in support. Pending resolution of the dispute, the plaintiff agreed not to proceed with the provisional sale. B. The decisions in the Courts below 3. Deputy District Judge R Yu[3] summarised the effect of the medical evidence filed by Madam Cheung as follows: “It is the opinion of Dr Ng Fung Shing that the Defendant has a full scale IQ score of 62 which indicates he is a mild grade mentally handicapped person. He has been a slow learner since birth and has studied in special school before. Because of his impaired intelligence, he has poor ability in comprehension. He cannot do even simply calculation. He fails to tell the nature of a contract even in broad terms. Dr Ng is of the opinion that the Defendant failed to understand the exact nature of a loan agreement and lack the mental capacity to enter into any contract in the form of loan agreements. The psychologist Dr Ng Kee On also commented that the Defendant is suffering from a significant intellectual dysfunction, functioning in the mildly handicapped spectrum of intellectual ability. The Defendant is likely to encounter difficulty in his daily activity including the handling of simple financial transaction.”[4] 4. Subsequently, his Honour directed that Mr Law be examined by two psychiatrists, one (Dr Chung See Yuen) appointed by the plaintiff and the other (Dr Ng Fung Shing) by Madam Cheung, and that they produce a joint report. The effect of the joint report was summarised by the Judge as follows: “In essence, both doctors agreed that the Defendant has a full scale intelligence score of 62 and the Defendant has been functioning in the mild grade mentally retarded range. Dr Ng maintains his opinion that the Defendant is incapable of instructing lawyers in relation to the present proceedings because of his mental incapacity. Dr Chung opines that the Defendant is not incapable of managing and administering his property and affairs. Dr Chung believes the Defendant would understand the nature of the loan agreement and he is capable of instructing lawyers now in relation to the present proceedings.”[5] 5. If Mr Law was indeed a mentally incapacitated person, Order 80 r 2(1) of the Rules of the District Court[6] would have precluded him from acknowledging service or defending the proceedings except by his guardian ad litem. However, in line with the plaintiff’s resistance to her appointment, the Judge refused to appoint Madam Cheung as guardian ad litem. In reaching that decision, he had regard to the joint psychiatric report and also viewed two videos showing communications between Mr Law and someone described as a referral agent who would assist him in getting loans. He rejected the view expressed in the report of Dr Ng in favour of that of Dr Chung and held that it had not been established that Mr Law was incapable of managing and administering his property and affairs or giving instructions to his legal advisors.[7] 6. Since Madam Cheung was denied locus as Mr Law’s guardian ad litem, the Judge dismissed her application to set aside the default judgment and consequential orders.[8] 7. On 23 April 2013,[9] the Judge refused the plaintiff’s application to vary the costs order made and Madam Cheung’s application for leave to appeal, elaborating on the basis upon which he had refused to appoint her Mr Law’s guardian ad litem, to which I shall return. 8. The Court of Appeal (which subsequently granted leave) upheld the Judge’s decision. [10] Cheung JA,[11] reviewed certain authorities and supported the Judge’s approach which involved considering whether Mr Law was capable of understanding the issues on which his consent or decision were likely to be necessary in the course of the proceedings, in the light of the medical and other evidence placed before the Court.[12] His Lordship rejected the argument that the Judge had adopted an inappropriately high standard more apposite for other types of inquiry into mental capacity.[13] 9. Leave to appeal to this Court was granted by the Appeal Committee on the basis that questions of the requisite importance arise on the appeal concerning: (a) the correct approach in law to deciding whether a person should be allowed to act as guardian ad litem with a view to protecting the interests of a person sued as defendant, who is alleged to be a mentally incapacitated person; and (b) the appropriateness of permitting the plaintiff in such cases to oppose such intervention by the proposed guardian ad litem. C. Errors in the judgments below 10. In my view, with respect, the Courts below: (a) adopted the wrong approach to, and the incorrect test of, mental incapacity in refusing to allow Madam Cheung to act as Mr Law’s guardian ad litem; and (b) erroneously allowed the plaintiff, whose interests are obviously adverse to those of Mr Law, to resist her appointment to defend the proceedings it had brought, so that the default judgment was upheld without the plaintiff’s claim ever being tested on its merits. D. The correct approach to the appointment of a guardian ad litem 11. The appointment of Madam Cheung as guardian ad litem for Mr Law ought to have been simple. 12. RDC Order 80 provides for how a “person under disability” participates in legal proceedings. As we have seen, it precludes a person under disability from defending proceedings save by a guardian ad litem.[14] It defines a “person under disability” as “a person who is a minor or a mentally incapacitated person”.[15] 13. Order 80 r 1 then defines “mentally incapacitated person” to mean: “... a mentally disordered person or a mentally handicapped person (within the meaning of the Mental Health Ordinance (Cap 136)) who, by reason of mental disorder or mental handicap, as the case may be, is incapable of managing and administering his property and affairs ...” 14. By section 2 of the Mental Health Ordinance (“MHO”), “mentally handicapped person” is defined as “a person who is or appears to be mentally handicapped”, and “mental handicap” is defined to mean: “... sub-average general intellectual functioning with deficiencies in adaptive behaviour, and ‘mentally handicapped’ shall be construed accordingly;” 15. MHO section 2 goes on to define “sub-average general intellectual functioning” to mean: “... an IQ of 70 or below according to the Wechsler Intelligence Scales for Children or an equivalent scale in a standardized intelligence test”. 16. It was common ground that D had an IQ of 62.[16] So Mr Law plainly fell within the statutory definition of “mentally handicapped person” [17] and thus within the definition of “mentally incapacitated person” in Order 80 r 1. 17. Madam Cheung (who was unrepresented before the Judge) took the step of issuing a summons seeking an Order that she be appointed Mr Law’s guardian ad litem. However, such an Order was not needed. In their judgments, the Courts below do not mention Order 80 r 3(2) which provides: “Except as provided by paragraph (4) or (5) or by rule 6,[18] an order appointing a person next friend or guardian ad litem of a person under disability is not necessary.” 18. By the combined effect of Order 80 r 3(6)(c)[19] and Order 80 r 3(8),[20] a person under disability is entitled to appear by his guardian ad litem if specified documents have been filed in court. Most importantly, there must be a certificate made by the solicitor representing the person under disability stating that the solicitor knows or believes that the person in question is a mentally incapacitated person, giving the grounds of his knowledge or belief; and stating that the proposed guardian ad litem has no interest in the matter adverse to that of the person under disability. 19. As we have seen, Mr Law’s IQ of 62 being common ground, he was plainly a “mentally handicapped person”. The requirement that he was incapable of managing and administering his property and affairs by reason of such mental handicap, did not require an inquiry by the Court provided that the solicitor instructed by Madam Cheung was prepared to certify that he or she knew or believed Mr Law to be so incapable.[21] As we have seen, the solicitor would have had ample grounds for so certifying on the basis of the opinions of Dr Ng Fung Shing and Dr Ng Kee On which were relied on in making Madam Cheung’s application to the Court. Madam Cheung had given her written consent to act as guardian ad litem and the solicitor would have been able to certify that she had no interest in the matter adverse to that of her husband, Mr Law. 20. Upon filing such documents in the Registry, Madam Cheung would, in accordance with Order 80 r 3(6)(c), have been entitled to appear as Mr Law’s guardian ad litem without any application to the Court. The statutory procedure is obviously designed to avoid unnecessary satellite litigation over mental capacity in ordinary cases involving persons under disability. 21. I might add in passing that the assumption by a next friend or guardian ad litem of that role without prior inquiry by the Court may of course not be final. If, for instance, it is subsequently discovered that the person assuming the role is not acting in good faith or is otherwise unsuitable, or indeed, for any other good reason, the Court may appoint someone else in substitution for the person previously assuming the role.[22] Moreover, if the person said to be under disability disputes any mental incapacity on his part, an inquiry may then be justified and the Court may order removal of the next friend or guardian ad litem.[23] E. Where the Courts below fell into error E.1 Distinction between debarring orders and protective appointments 22. The principal error made by the Judge and upheld by the Court of Appeal involved a misapplication of the Court of Appeal’s judgment in Ho Po Chu v Tung Chee Wah.[24] That was a case where the defendant, having been sued by the plaintiff, obtained a court order staying the plaintiff’s action unless and until she was represented by a next friend and solicitors were instructed. This was done on the basis that under Order 80 r 2, a person under disability may not bring a claim except by his next friend. When considering what, if any, evidence was needed to trigger such prohibition, Yuen JA drew an important distinction between subjecting someone to an adversarial debarring order (such as that sought by Mr Tung Chee Wah) and allowing someone (like Madam Cheung) to act in the role of a protective next friend or guardian ad litem: “It is therefore clear in my view that the Court should adopt a flexible approach to the evidence of mental incapacity under O 80. Where mental incapacity is relevant because a debarring order is sought, the court would expect clear medical evidence to prove the mental incapacity because the effect of a debarring order is to restrict a person's direct access to the court. However, where mental incapacity is relevant because an action is started by the next friend (or defended by the guardian ad litem) of a person said to be suffering mental incapacity and that act is obviously for the person's benefit, the court would not initially require formal medical evidence, although it would of course be prudent for the next friend or guardian ad litem (and the solicitor acting under his instructions) to have medical evidence available before the event in case of challenge.”[25] 23. Recognizing that a debarring order has the serious consequence of denying someone direct access to the courts, her Ladyship considered it appropriate to adopt a high evidential threshold for the making of such orders. She decided that it was proper to adopt the stringent requirements laid down in Part II of the MHO for applications to the court for an inquiry into “whether any person ... who is alleged to be mentally incapacitated is incapable, by reason of mental incapacity, of managing and administering his property and affairs”,[26] being a procedure that could lead to a decision under MHO, s10 which would “place the estate of the mentally incapacitated person out of his reach”.[27] Her Ladyship stated: “Although a debarring order only stays the proceedings until a next friend is appointed and solicitors are instructed, and so has a less draconian effect than a s 10 decision, given the effect on the person's direct access to the court and the stigma involved, in my view the court should require medical evidence of the same quality (even if not with the same formality) before it would make a debarring order. Certainly I cannot envisage a court making a debarring order without any medical evidence.”[28] 24. Yuen JA recognized that, in contrast, the Rules do not require a judicial inquiry before appointment of a protective next friend or guardian ad litem: “...O 80 r 3(8)(c)(i) shows that an inquiry is not needed before a mentally incapacitated person can sue by his next friend (or defend by his guardian ad litem). That rule sets out the documents that should be filed in the Registry before a person can appear by his next friend or guardian ad litem. It provides that if a person proposing to be next friend or guardian ad litem has not been authorised to act as such under Pt II MHO, there must be filed in the Registry a certificate made by the solicitor for the mentally incapacitated person certifying (amongst other things) that he believes the person to whom the certificate relates is a mentally incapacitated person, giving the grounds of his belief.”[29] 25. It is easy to see the basis for the distinction drawn by Yuen JA which, with respect, was clearly sound. A defendant sued by someone whom he considers to be mentally incapacitated seeks protection from being vexed by that plaintiff by asking the Court to debar him or her from proceeding save by a next friend acting through a solicitor. A defendant seeking such an order is obviously in an adversarial position vis-à-vis the plaintiff and, unsurprisingly, is required to provide “clear medical evidence to prove the mental incapacity because the effect of a debarring order is to restrict a person's direct access to the court.” On the other hand, a person in Madam Cheung’s position, seeks to be allowed to act as Mr Law’s guardian ad litem to protect Mr Law’s interests in the face of the plaintiff’s suit. Madam Cheung is not acting in an adversarial position vis-à-vis Mr Law, but protectively. As Yuen JA points out, in such cases, the court does not require medical evidence, a solicitor’s certificate being sufficient. 26. The Judge failed to appreciate this distinction. Madam Cheung was unrepresented, but counsel for the plaintiff submitted that before appointing her as guardian ad litem, “[the] medical evidence required should be of a comparable quality to the requirement for the purposes of a Part II inquiry under the Mental Health Ordinance”.[30] The Judge then referred to Ho Po Chu,[31] and noted that Yuen JA said “that the debarring order is less draconian than a section 10 decision. But the court should require medical evidence of the same quality before it would make a debarring order.”[32] His Honour concluded: “In this particular case, the finding that the Defendant is a mentally incapacitated person is a bar to the Plaintiff from pursuing this action against the Defendant until the appointment of guardian ad litem and appointment of solicitors. I agree with Mr Cheung that the medical evidence that should be adduced before such a finding should be made should be of the same quality as for a debarring order.”[33] 27. The Judge appears to have considered Madam Cheung’s application to be allowed to represent Mr Law an application to “bar the plaintiff from pursuing the action”, which was plainly erroneous. There was no question of the plaintiff being denied direct access to the courts. The question was whether Mr Law should be allowed a guardian ad litem to defend the plaintiff’s claim. It was quite inappropriate to require Madam Cheung to satisfy evidential requirements equivalent to those laid down by Part II of the MHO in relation to a process which could result in a third party taking over control of the mentally incapacitated person’s estate. 28. In the Court of Appeal, the complaint was made that the Judge had misapplied Ho Po Chu v Tung Chee Wah and so had “chosen to adopt a higher standard that is required for other types of inquiry”.[34] However, Cheung JA’s response fails to address the crucial distinction which Yuen JA had emphasised in Ho Po Chu[35] between an adversarial application for a debarring order and the protective appointment of a guardian ad litem under O 80. His Lordship merely stated: “This is not a case where the applicant intervened at an early stage of the proceedings. She only appeared after the judgment had been obtained and enforcement proceedings had been taken. Raising the issue of the defendant’s disability at such a late stage clearly required cogent evidence to be given. In any event, the provision of two experts was agreed to by the parties who were legally represented. I cannot see how the Judge’s decision can be faulted.”[36] E.2 Equating the appointment of a guardian ad litem with exercise of the Court’s paternal and administrative jurisdiction 29. On 23 April 2013,[37] the Judge handed down a further judgment dealing with the plaintiff’s application for a variation in the costs order and Madam Cheung’s application for leave to appeal. In the course of that judgment, with respect, his Honour appears to have further confused the role of the Court in relation to the appointment of a guardian ad litem in a case like the present, relying on a dictum of Viscount Haldane LC in Scott v Scott.[38] That was a case where the House of Lords was concerned with the question whether the Probate Divorce and Admiralty Court had power to hear a nullity petition in camera, being an issue relevant to deciding whether publication of the details of the hearing by a party constituted a contempt. 30. It was held that there was no basis for ordering the hearing to be in camera and, in the course of his speech, Viscount Haldane LC contrasted the position regarding in camera hearings obtaining in different courts and contexts, including in the Court of Protection, stating: “The case of wards of Court and lunatics stands on a different footing. There the judge who is administering their affairs, in the exercise of what has been called a paternal jurisdiction delegated to him from the Crown through the Lord Chancellor, is not sitting merely to decide a contested question. His position as an administrator as well as judge may require the application of another and overriding principle to regulate his procedure in the interest of those whose affairs are in his charge. In order to make my meaning distinct, I will put the proposition in another form. While the broad principle is that the Courts of this country must, as between parties, administer justice in public, this principle is subject to apparent exceptions, such as those to which I have referred. But the exceptions are themselves the outcome of a yet more fundamental principle that the chief object of Courts of justice must be to secure that justice is done. In the two cases of wards of Court and of lunatics the Court is really sitting primarily to guard the interests of the ward or the lunatic. Its jurisdiction is in this respect parental and administrative, and the disposal of controverted questions is an incident only in the jurisdiction. It may often be necessary, in order to attain its primary object, that the Court should exclude the public. The broad principle which ordinarily governs it therefore yields to the paramount duty, which is the care of the ward or the lunatic.”[39] 31. His Lordship was referring to the Court’s paternal and administrative jurisdiction over wards of court and persons then referred to as “lunatics”. Plainly, this has nothing to do with a defendant in ordinary civil proceedings being permitted or required to defend adversarial claims through a guardian ad litem. However, citing the two passages set out above, the Judge erroneously perceived the Court to be exercising the aforesaid jurisdiction, stating: “Therefore, when Madam Cheung made the application pursuant to Order 80, she meant to request that the Court exercise judicial power on the guardianship of the Defendant. The Court has no need to consider whether she has obtained the consent of the Defendant to this application. On the contrary, the Court has to consider and find, under independent medical advice, whether the Defendant can manage his own legal issues, and then determine who would be the most suitable one to be his guardian ad litem if he is found to be a mentally incapacitated person.”[40] 32. The inappropriateness of importing Court of Protection inquiry procedures into O 80 next friend or guardian ad litem decisions was recognized by the English Court of Appeal in Masterman-Lister v Brutton Co (Nos 1 and 2).[41] Kennedy LJ stated: “... it has to be recognised that someone who is treated as a patient for the purposes of Ord 80, r 1, who litigates by a next friend, is not necessarily and may never become accepted by the Court of Protection as a patient pursuant to section 94(2) of the 1983 Act. As is clear from the wording of section 94(2) the jurisdiction of the Court of Protection is only exercised when, after considering medical evidence, a nominated judge issatisfied as to the person's incapacity. Under Ord 80, r 1 no judicial officer has to consider medical evidence or be satisfied as to incapacity before a person can be treated as a patient.”[42] E.3 Reliance on Ng Hong Ki v Leung Fong Kiu[43] 33. As I have already pointed out, the present case is one where a judicial inquiry into Mr Law’s mental incapacity was not needed provided the solicitor’s certificate and other documents referred to in Section D above were filed. However, both the Judge[44] and the Court of Appeal[45] considered it necessary to conduct an inquiry, examining the medical and other evidence. They did so relying on the Court of Appeal’s earlier decision in Ng Hong Ki v Leung Fong Kiu and in particular on the following passage in the judgment of Kwan JA: “For the purpose of an appointment of a guardian ad litem under O 80, the test of mental incapacity is ‘issue-specific’ and the test that has to be applied is ‘whether the party to legal proceedings is capable of understanding, with the assistance of such proper explanation from legal advisers … the issues on which his consent or decision is likely to be necessary in the course of the proceedings’. The focus of the enquiry is on the litigation under consideration rather than the whole of that person's property and affairs. Complete incapacity is not required to be established (Masterman-Lister v Brutton & Co (Nos 1 and 2) [2003] 1 WLR 1511, paras18, 27, 62, 75). These features were identified in a person that would lead the Court to conclude that he was a person under disability for the purpose of O 80: ‘… he had no insight at all into his own mental state. He was not capable of instructing a solicitor properly. He certainly was not capable of exercising any reasonable judgment on a possible settlement.’ (Kirby v Leather [1965] 2 QB 367, 384; see also Tang Kam Sheung v Tang Kit Yee (unrep, HCA 677/2007, [2009] HKEC 1599), Carlye Chu J, para 16).”[46] 34. As is apparent, Kwan JA was basing herself essentially on Masterman-Lister v Brutton & Co (Nos 1 and 2),[47] a case in which the Court of Appeal affirmed the need for any inquiry into the alleged patient’s mental incapacity to be “issue-specific”, applying the test “in relation to the particular transaction (its nature and complexity) in respect of which the question whether a party has capacity falls to be decided.”[48] The context of the English Court of Appeal’s decision is therefore important. 35. Masterman-Lister was a case in which the plaintiff suffered head injuries in a traffic accident in September 1980 and issued a Writ against the defendant in December 1980. In 1987, on the advice of counsel and solicitors, the action was settled for £76,000 plus costs. In 1993 the plaintiff sought, notwithstanding the usual limitation periods, (i) to re-open the settlement entered into in 1987 and (ii) to sue his former solicitors for negligence. He could only succeed as to (i) if he was a person under disability when settling the action in 1987 since O 80 r 10 provided that a settlement with such a person was not valid without the approval of the court (which had not been obtained). And he could only succeed as to (ii) if he was shown to be under a disability for the purposes of the Limitation Act 1980 so that time for bringing proceedings was extended. In other words, it was undoubtedly the case that the plaintiff’s mental capacity was being debated in an adversarial context: he was seeking to deprive the original defendant of the benefit and finality of a settlement agreement; and to deprive his former solicitors of a limitation defence. It is therefore unsurprising that he had the burden of showing his mental incapacity after an adversarial judicial inquiry. 36. Nothing in Masterman-Lister throws doubt on the distinction drawn by Yuen JA in Ho Po Chu v Tung Chee Wah. Indeed, at §66 of Masterman-Lister, Chadwick LJ acknowledged the effect of O 80 r 3 (which was in identical terms to our rule): “Order 80, r 3(2) provided that (save in particular cases) an order appointing a person as next friend or guardian ad litem was not necessary. That, as it seems to me, is of some significance. The rule making body plainly contemplated, and intended, that the question whether a party was required to act through a next friend or guardian ad litem (as the case might be) should, in the ordinary case, be determined by the party himself or by those caring for him; perhaps with the advice of a solicitor but without the need for inquiry by the court. Order 80, r 2(3) required that a next friend or guardian ad litem must act by a solicitor; and rule 3(8)(c)(i) required that, in such a case, the solicitor was to file a certificate certifying that he believed the party to be a patient, with his grounds of belief. But there was no requirement, as such, in the rules for the filing or consideration of medical evidence. If the rule is to work in practice, the test of mental capacity should be such that, in the ordinary case, the need for a next friend or guardian ad litem should be readily recognised by an experienced solicitor.”(Italics supplied) 37. Kirby v Leather,[49] referred to in the passage from Ng Hong Ki cited above, was likewise a personal injuries case involving the adversarial issue of whether the plaintiff ought to be allowed to start proceedings after expiry of the usual limitation period on the ground that he was under a disability since the date when the cause of action accrued. 38. The other decision mentioned, Tang Kam Sheung v Tang Kit Yee,[50] was likewise concerned with ascertaining mental capacity in an adversarial situation. The plaintiff, an elderly lady, brought proceedings alleging that the defendant had misappropriated monies belonging to her and was met with a defence (inter alia) questioning her mental capacity to commence and carry on the proceedings on the ground that she was suffering from dementia. The Court was concerned with an attempt by the defendant to obtain a debarring order, staying the proceedings until a next friend was appointed.[51] Chu J accordingly held that, as in Ho Po Chu, such a challenge called for an inquiry into the plaintiff’s mental capacity before denying her direct access to the Court. 39. It follows that none of the authorities drawn upon by Kwan JA were concerned with the appointment of a guardian ad litem to protect the interests of the defendant in circumstances where no inquiry is needed. 40. The true focus of Ng Hong Ki lies elsewhere. The plaintiff landlord in that case served notice of termination on the defendant and brought summary judgment proceedings for possession of the premises. The defendant’s brother filed an affirmation on the defendant’s behalf stating that the defendant was suffering from schizophrenia and had been a patient under psychiatric hospital care since 1998. The brother was allowed to make oral submissions to the Master and so was informally treated like a guardian ad litem. As no substantive defence was raised, the Master gave summary judgment for the landlord. On appeal, it was sought to suggest by way of defence that the defendant was not the tenant, but the Judge considered that suggestion incredible and dismissed the appeal. However, he granted the brother’s application to be appointed guardian ad litem, being satisfied on the basis of a medical report that the defendant was a mentally incapacitated person for O 80 purposes. 41. On appeal to the Court of Appeal, it was sought to be argued on the defendant’s behalf that the Judge should have set aside the order for summary judgment since the proceedings had been conducted without a guardian ad litem and so were in breach of O 80 r 2(1) of the RDC, it being submitted that the substantive steps taken before the appointment of the guardian ad litem were nullities. This was the focus of Kwan JA’s judgment. Her Ladyship was not concerned with deciding whether or on what evidential basis a guardian ad litem should be appointed as one was already in place. Her Ladyship upheld the appointment made by the Judge although he had done so on “scanty” medical evidence.[52] The issue for the Court of Appeal was whether the preceding proceedings were a nullity. Her Ladyship answered in the negative, holding that the Court had power to regularise the position retrospectively.[53] It is in this context that the passage relied on in the Courts below should be placed. It does not provide a basis for supporting the approach adopted below. E.4 Permitting the plaintiff to oppose appointment of a guardian ad litem for Mr Law 42. It was, in my view, inappropriate to allow the plaintiff, acting in furtherance of its own interests, to seek to deprive Mr Law, a mentally handicapped person, of the protection of a guardian ad litem, especially since the plaintiff’s success meant that its default judgment would not be challenged on the merits. 43. I have already quoted Chadwick LJ’s comment that: “...the question whether a party was required to act through a next friend or guardian ad litem (as the case might be) should, in the ordinary case, be determined by the party himself or by those caring for him; perhaps with the advice of a solicitor but without the need for inquiry by the court.”[54] One might add a fortiori, not determined by the court on the basis of opposition from the other side. 44. In Folks v Faizey,[55] after referring to the passage just cited, Pill LJ criticised opposition by the respondent to appointment of a litigation friend (or next friend) for the protection of the appellant as “intermeddling for no sound reason”: “In this case, those advising the respondent, without any plausible reason in terms of protecting the respondent's own position, have sought to interfere in a procedure with which they were only minimally concerned. Indeed, the appointment of a litigation friend would give [...] protection to them as well as to the appellant and his advisors. I should not wish to describe the opposition as an abuse of the process of the court but in my judgment it is an intermeddling, for no sound reason, which the judge, on the evidence available, ought not to have tolerated.”[56] 45. The same criticism applies in this case. It would plainly have been appropriate and much fairer to allow Madam Cheung to assume the role of guardian ad litem and then to have her application to set the default judgment aside properly tried. If it was suggested that Mr Law lacked mental capacity to enter into the loan contracts, psychiatric and other evidence bearing on that issue could then be fully canvassed, the “issue-specific” questions then relating to his understanding of the transaction. There is no question of any issue estoppel being raised against the plaintiff in relation to the substantive action merely by reason of a guardian ad litem coming onto the record for the defendant. That is indeed a good reason for the defendant not to “intermeddle”. By taking the course followed in the Courts below, the issue of Mr Law’s mental capacity to enter into the moneylending contracts was never properly reached. 46. Moreover, it is suggested in the appellant’s printed case[57] (and was suggested in affirmations filed by Madam Cheung) that appointment of a guardian ad litem would enable other possible lines of defence to be explored: “...it is the Applicant’s position that in addition to the point about the Defendant’s capacity to enter into the Loan Agreement, the Defendant has other arguments to support the application for setting aside the default judgment, eg (a) the Loan Agreement was in contravention of the Money Lenders Ordinance; (b) it was entered into as a result of the undue influence exercised by the Plaintiff over the Defendant; and (c) it was in any event an unconscionable bargain (see Applicant’s 3rd Affirmation dated 16 March 2012).” 47. There may or may not be substance in these suggestions, but refusing Mr Law a guardian ad litem and dismissing the application to set aside the default judgment meant that those issues have not been aired. 48. As is often remarked, where a defendant is mentally incapacitated, it is usually in the interests of a plaintiff to be able to deal with a guardian ad litem who is able to ensure that the litigation is properly conducted and to engage in meaningful settlement negotiations. Since settlements by or on behalf of a person under disability are invalid without the Court’s approval,[58] the input of a guardian ad litem in obtaining such approval will generally be crucial. F. Disposal 49. For the foregoing reasons, I would order that: (a) the applicant’s appeals be allowed; (b) the Orders of Deputy District Judge R Yu dated 25 October 2012 (dismissing the applicant’s application to act as guardian ad litem and ordering costs against her) and dated 23 April 2013 in DCCJ 736/2011 and DCMP 2062/2011 (dismissing the applicant’s application for leave to appeal and a stay pending appeal, and ordering costs against her) and the Orders of the Court of Appeal dated 28 January 2014 (dismissing the applicant’s appeal and ordering costs against her) be set aside; (c) the applicant be appointed guardian ad litem of the defendant and that the applications to set aside the default judgment and consequential orders obtained by the plaintiff against the defendant be restored to the list on a date and subject to such directions as the Registrar of the High Court may determine, with the title of such proceedings amended to refer to the applicant, acting as the defendant’s guardian ad litem; (d) that there be an order nisi that costs of these appeals, including the applications for leave to appeal, and of the proceedings below referred to in sub-paragraph (b) above be paid by the plaintiff to the applicant, with liberty to the parties to lodge written submissions on the question of costs within 14 days of the date of this judgment and, in default of such submissions, that this order as to costs do stand as an order absolute without further direction. Mr Justice Tang PJ: 50. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Fok PJ: 51. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Stock NPJ: 52. I agree with the judgment of Mr Justice Ribeiro PJ. Sir Anthony Mason NPJ: 53. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Ribeiro PJ: 54. The Court unanimously allows the appeal and makes the Orders set out in paragraph 49 above. Mr Anson Wong SC, Mr Albert Luk & Mr Billy NP Ma, instructed by Ha & Ho, for the Applicant/Appellant Mr Wallace Cheung, Mr Leon Ho, instructed by Philip T.F. Wong & Co., for the Plaintiff/Respondent [1] DCCJ 736/2011 (25 October 2012). [2] DCMP 2062/2011 (18 October 2011). The aggregate sum owing amounting to $101,754.25 as at that date. [3] Now HH Judge R Yu. [4] Judgment §9. [5] Judgment §13. [6] “A person under disability may not bring, or make a claim in, any proceedings except by his next friend and may not acknowledge service, defend, make a counterclaim or intervene in any proceedings, or appear in any proceedings under a judgment or order notice of which has been served on him, except by his guardian ad litem.” All references to Orders and Rules in this judgment are to the Rules of the District Court (“RDC”). They are in the same terms as Order 80 of the Rules of the High Court (Cap 4). [7] Judgment §§74 and 75. [8] Judgment §79. [9] DCMP 2062/2011 and DCCJ 736/2011. [10] Cheung, Yuen and Chu JJA, CACV 127 & 128/2013 (17 February 2014). [11] With whom the other Judges agreed. [12] Court of Appeal §§7.1-14. [13] Judgment §23. [14] Order 80 r 2. [15] Order 80 r 1. [16] As appears from the Report of Dr Ng Kee-on dated 25 April 2009, this was the score arrived at using the Weschler Adult Intelligence Scale, Revised Version. The same applies to the Joint Report of Dr Chung See Yuen and Dr Ng Fung Shing dated 17 July 2012 (§35). [17] Contrary to the submissions of counsel for the plaintiff, no basis exists for suggesting that there may have been doubts as to whether Mr Law suffered from “deficiencies in adaptive behaviour” for the purposes of the definition of “mental handicap”. [18] The exceptions are not presently relevant. [19] “...a person under disability shall not be entitled to appear by his guardian ad litem on the hearing of a petition, summons or motion which, or notice of which, has been served on him, unless and until the documents listed in paragraph (8) have been filed in the Registry.” [20] “The documents referred to in paragraph (6) are the following- (a) a written consent to be next friend or guardian ad litem, as the case may be, of the person under disability in the cause or matter in question given by the person proposing to be such friend or guardian; ... (c) ... a certificate made by the solicitor for the person under disability certifying - (i) that he knows or believes, as the case may be, that the person to whom the certificate relates is a minor or a mentally incapacitated person, giving (in the case of a mentally incapacitated person) the grounds of his knowledge or belief; and ... (iii) except where the person named in the certificate as next friend or guardian ad litem, as the case may be, is the Official Solicitor, that the person so named has no interest in the cause or matter in question adverse to that of the person under disability.” [21] As recognized in Ho Po Chu v Tung Chee Wah [2006] 3 HKLRD 553 at §30. [22] Order 80 r 3(4): “Where a person has been or is next friend or guardian ad litem of a person under disability in any proceedings, no other person shall be entitled to act as such friend or guardian, as the case may be, of the person under disability in those proceedings unless the Court makes an order appointing him such friend or guardian in substitution for the person previously acting in that capacity.” (Italics supplied) See Hong Kong Civil Procedure 80/3/10. [23] Beall v Smith (1873) LR 9 Ch App 85; Howell v Lewis 65 LT 672; Fry v Fry (1890) 15 PD 50. A challenge as to mental capacity is envisaged in Ho Po Chu v Tung Chee Wah [2006] 3 HKLRD 553 at §52. [24] [2006] 3 HKLRD 553. [25] At §52. [26] MHO section 7(1). [27] [2006] 3 HKLRD 553 at §36. [28] Ibid at §39. [29] Ibid at §30. [30] DCCJ 736/2011 (25 October 2012) at §24. [31] Ibid at §§37 to 39. [32] Ibid at §26. [33] Ibid at §27. [34] Court of Appeal at §§22-23. [35] And to which, unfortunately, her Ladyship does not appear to have drawn attention in the Court of Appeal below. [36] Court of Appeal at §24. [37] DCCJ 736/2011 and DCMP 2062/2011. [38] [1913] AC 417. [39] At 437. [40] DCCJ 736/2011 and DCMP 2062/2011 (23 April 2013) at §42. [41] [2003] 1 WLR 1511. [42] At §12. [43] [2012] 1 HKLRD 435. [44] DCCJ 736/2011 (25 October 2012) at §§21, 74 and 75; DCMP 2062/2011 (23 April 2013) at §15. Although the Judge stated that “all parties to the litigation agree” as to the test, Madam Cheung was at the time unrepresented. [45] Court of Appeal §7.3, 8-9. [46] [2012] 1 HKLRD 435 at §34. [47] [2003] 1 WLR 1511. [48] At §66, per Chadwick LJ. [49] [1965] 2 QB 367. [50] Chu J, HCA 677/2007 (25 September 2009). [51] At §16. Chu J also considered it potentially necessary to consider proceedings under Part II of the MHO. [52] At §35. [53] At §§37-40. [54] Masterman-Lister v Brutton & Co. (Nos 1 and 2) [2003] 1 WLR 1511 at §66. [55] [2006] EWCA Civ 381. [56] At §20. [57] At §72. [58] Order 80 r 10. 
Chief Justice Ma: 1. In applications for ancillary relief in matrimonial proceedings, the court is given wide powers to make orders for financial provision. The jurisdiction is contained in ss 4, 6, 6A and 7 of the Matrimonial Proceedings and Property Ordinance (“the MPPO”),[1] where the court is able to make orders for financial provision (by way of periodic payments and lump sums etc) and for property adjustment. These orders can take effect immediately or, sometimes, they may take effect only in the future. We were in the present appeal concerned with a type of order that takes effect in the future in which the Court has ordered that upon the occurrence of a contingency, there will accrue to one of the parties to a dissolved marriage a stated or perceived benefit. 2. The particular order with which this appeal was concerned was in the following terms:- “Contingent upon the commencement of the redevelopment of Severn Villa by ADHL within the lifetime of the wife, the husband do pay the wife a further sum representing one half of the notional increase in the value of the 418,945,640 ADHL shares attributable to the redevelopment (less HK$230 million and all costs and expenses [of] or relating to the redevelopment), such value to be agreed, failing which the same to be assessed by an independent court expert to be appointed by the Court and such payment to be paid within one month of the final determination of such valuation.” 3. This type of order has been called a clawback order.[2] The nature of such an order and the type of factors that ought to be considered by a court before such orders are made, fell to be considered in the context of the present case. Three questions, which I shall presently set out, were identified by the Appeal Committee[3] as being questions of great general or public importance, and leave to appeal was accordingly given to the Respondent Appellant, Mr Raymond Kin Sang Hung (“H”). The Appeal Committee also granted leave to appeal to the Petitioner Respondent in this appeal, Ms Mimi Kar Kee Wong Hung (“W”) in relation to the form of the clawback provision. I shall also elaborate on this presently. 4. After hearing counsel,[4] we allowed H’s appeal (FACV 10 of 2014) and dismissed W’s appeal (FACV 11 of 2014). The clawback order referred to earlier was ordered to be set aside. 5. Before going into the reasons for the Court’s decision, I should first set out the relevant background. A BACKGROUND A.1 Facts 6. H and W married in 1973. They had two daughters and a son. The parties separated in 1999 and W petitioned for divorce on 25 February 1999 citing unreasonable behaviour. The decree absolute was not granted until 18 March 2010. Neither the reasons for this nor the history of the divorce proceedings are relevant to the present appeal and it is therefore unnecessary to say more about this aspect. 7. The context of the present appeal was W’s application for ancillary relief following the couple’s divorce. At stake were combined assets in excess of HK$1 billion, comprising a collection of antique furniture worth about $749 million,[5] some jewellery worth about $1.3 million, paintings and artefacts of the value of over $42 million and shares in the company, Applied Development Holdings Limited (“ADHL”) valued at over $330 million. In the lower courts, issues arose in relation to all these assets, but in this appeal we were concerned only with the ADHL shares. 8. In 1975, the parties formed a company manufacturing electronic products, Applied Electronic Limited (“AEL”). The majority shareholders were initially W’s uncle and aunt, who had provided most of the company’s startup capital. Their shareholding was later transferred to H and W. In 1986, AEL went public and was listed on the Hong Kong Stock Exchange as ADHL. The parties retained 50% of the shareholding. 9. Notwithstanding their separation in 1999, the parties continued to work within the company and its related and subsidiary companies. W was up until 2010 an executive director of the company and at some stage also its executive chairman. In March 2010, after the divorce proceedings became active again, W was removed as executive director of ADHL.[6] Her entitlement hitherto to housing also ceased. This housing consisted of an apartment at Severn Villa, located in Severn Road on The Peak. 10. Severn Villa is a four storey residential building comprising six units. Four of these were owned by Severn Villa Limited (“SVL”), a wholly owned subsidiary of ADHL. It is in relation these four units (“the Severn Villa Property”) with which the present appeal was concerned. The remaining two units in Severn Villa were owned by other persons. 11. Apart from the Severn Villa Property, ADHL also owned two properties located in the British Virgin Islands and Panama, held with a view to the development of hotel resorts. ADHL was essentially a property investment company. 12. At this point, it is important to recognise that neither H nor W at any stage owned the Severn Villa Property themselves: they were merely shareholders in a company (ADHL) which had a wholly held subsidiary company (SVL), and it was this subsidiary company which held the property. ADHL was not even some sort of corporate vehicle which held the property for them; the company was a public company, had other shareholders and also had business interests other than the Severn Villa Property. As recorded by the judge who tried the matter at first instance,[7] at one stage W claimed that the Severn Villa Property belonged beneficially to her and that SVL held the property on trust for her, since she had provided the purchase money through the sale of some properties in the United States. Following her removal as a director of ADHL, W instituted proceedings[8] making the claim that SVL held the legal title to the Severn Villa Property as a matter of “corporate convenience”. This claim came on for trial in April 2011 and on the first day of trial, W accepted she had no sustainable claim. The action was dismissed and she paid costs on an indemnity basis. A.2 The proceedings below 13. The application for ancillary relief was tried in the Court of First Instance before Deputy High Court Judge Carlson in a hearing lasting 14 days. At the beginning of his judgment, the learned judge had this to say:- “2. It is very easy to be wise after the event, but I am bound to say that with more than enough assets to go round between them and with a modicum of common sense and good judgment this is a dispute that could have been settled quickly and, for a case of this size, relatively cheaply. 3. I am afraid that this has become an object lesson in how matrimonial litigation should not be conducted. I have not been able to add up the total precisely, but the parties joint costs for this and the related action which I had consolidated with these matrimonial proceedings must exceed $50 million. The parties have fought each other before me for over two years. I have lost count of, and I don’t care to count, how many written rulings and judgments I have had to deliver or hand down. They have gone to the Court of Appeal twice before trial and before the Companies’ Judge on one occasion. The proceedings have seen me into retirement from the bench and I have returned as a retired judge to try this matter and its related action.” 14. These remarks have some relevance to a point which I shall develop below regarding the need to consider the desirability of finality, particularly in matrimonial litigation, when a court makes orders taking effect in the future in the form of the clawback order in the present case. 15. In the Court of First Instance, W did not seek a clawback order at all. Following the approach laid down by this Court in LKW v DD[9], the parties and the judge proceeded to look into and assess the value of the matrimonial assets I have identified earlier.[10] The evidential focus of the trial was not so much on the value of the assets (as to which there was much agreement between the parties) but mainly over an issue which was no longer a live one in the appeal before us, namely, an allegation by H that W had taken away 27 items (worth over $78 million) in the Hung Collection.[11] Much of the learned judge’s judgment dealt with this issue. There were also a number of legal issues, one of which was the appropriate proportion to be applied in the division of assets: W sought a 60/40 division in her favour, H submitted that an equal division was appropriate. 16. As regards the value of ADHL’s shares, this was agreed between the parties’ experts.[12] There was a minor difference regarding the shell value of the company, but this was resolved in H’s favour.[13] Despite the difference in shareholding in their names, it was also held that H and W had an equal interest in the shares held between them.[14] At trial, H sought an order that the total shares held by them be divided equally, but this was resisted by W on the basis that this would take away much of the value of the shares. According to her, it was more valuable for the shares to be held by one person.[15] W sought an order that the whole of shareholding in H’s name be transferred to her and for his part as an alternative, he sought an order that W’s shareholding be transferred to him. The judge ordered that W transfer her shares to H and that H had to pay W $161,120,620.09 for her half share in the shares held between them. 17. At trial, the extent of the differences between the parties regarding the shares in ADHL was as stated in the previous paragraph. No issue was raised by either party regarding the value of any redevelopment of the Severn Villa Property and certainly, no order remotely resembling the clawback order was sought by W. Before us, Mr Fung SC submitted that the issue of the redevelopment of the Severn Villa Property was raised before the trial judge. He was correct only to the extent that W did as part of the evidence before the court produce an expert report placing a redevelopment value on the Severn Villa Property at $665 million[16] and that in his Opening, Leading Counsel for W[17] did vaguely mention the aspect of redevelopment. However, as Mr Yu SC submitted in the written Case for H, it was at no stage pressed by W that the value of the ADHL shares ought to take into account the redevelopment potential of Severn Villa, there was no cross examination by W’s counsel of the expert valuer produced by H at trial on this aspect and in the closing submissions of counsel, no mention was made of it either. It is important also to remember that the value of the shares in ADHL was actually agreed between the parties at trial without any consideration of redevelopment, nor was this facet somehow left open. It was perhaps not surprising that W did not insist that the redevelopment potential of Severn Villa ought to be taken into account in valuing ADHL shares: at the trial stage, she was submitting that the shares in H’s name ought to be transferred to her and obviously she would not have wanted to pay more than was necessary for those shares. Neither the maker of the report on redevelopment potential nor her own expert on the valuation of the shares was called at trial. 18. W appealed. Again, it is not necessary to go into any of the grounds of appeal except in relation to the question of redevelopment. Though not raised at first instance, W sought to obtain an uplift in relation to the ADHL shares on the basis that the judge’s valuation did not reflect the redevelopment potential of the Severn Villa (being, as indicated earlier, $665 million).[18] Notwithstanding the fact that W had not sought an order below taking into account the redevelopment value, the Court of Appeal was of the view that a clawback order should be made, essentially for the following reasons:-[19] (1) Even where the parties had not raised a point in the court below, the courts generally had a quasi-inquisitorial role in matrimonial litigation. Reference was made to a passage in Parra v Parra[20] at para 22. I shall deal with this passage later in this judgment. (2) Reference was made to H’s evidence in which H expressed a hope that Severn Villa might be redeveloped in order to help ADHL out of its financial difficulties, to develop existing projects and look into future investment opportunities. H did, however, also go into the difficulties of this, given in particular the fact that the owner of one of the units in Severn Villa was holding out. (3) The clawback order was a type of order that had “withstood the test of time” and was one based on principle (specific reference was made to ss 7(1)(a) and (b) of the MPPO[21]). The value of the redevelopment of the Severn Villa to ADHL was regarded as something which H was “likely to have in the foreseeable future”. The redevelopment of the Severn Villa was seen as “not an unlikely eventuality”. Even if the prospect of redevelopment was remote, this was not a bar to a clawback order being made. In his judgment,[22] Cheung JA said this:- “4) The suggestion that the chance of redevelopment is remote must be considered in the proper perspective. The property is located on the Peak which is the choicest real estate in Hong Kong. Further, even in Parra, notwithstanding that the chance of redevelopment was remote, a claw back was nonetheless imposed, no doubt because of the Court’s wish to achieve fairness between the [parties].” (4) While the clawback order was inconsistent with a clean break,[23] the facts of the case justified such an order. (5) A clawback order did not mean that W would only get the upside of any redevelopment but no downside: “She is only asking for an uplift in the event of redevelopment of this property only and not for other purposes.” [24] 19. The terms of the clawback order made by the Court of Appeal differed from the order that had been suggested by W at the appeal. The main difference between the order as made and that prepared by W was that the stated contingency in the former was the “commencement of the redevelopment of Severn Villa by ADHL” (my emphasis). W’s proposed order merely referred to a redevelopment of the property without any restriction as to who would redevelop. This formed the subject matter of W’s appeal before us. 20. Further, the Court of Appeal gave liberty to apply to H to discharge the clawback order in the event of a sale of his shares in ADHL or if the company was delisted, provided that the sale was bona fide or that there were legitimate reasons to delist. 21. Apart from making the clawback order in relation to the ADHL shares, the Court of Appeal increased the amount that H had to pay to W for the transfer of the shares in her name to him to $167,159,310, this taking into account the shell value of the company which the Court of Appeal assessed at $258 million. B THIS APPEAL B.1 Leave to appeal 22. Both parties sought leave to appeal to this Court. The Appeal Committee granted leave to H on the following questions said to be of great general or public importance (all related to the clawback order):- “(1) Whether it is wrong in principle to grant the “clawback” order in a case where the property in question is not owned by the parties but by a company which is a subsidiary of a publicly listed company in which the parties are majority shareholders, especially in circumstances where the company would require further injection of capital; (2) Whether it is open to the Court of Appeal to impose a clawback provision when the same was not sought at the trial and where the parties agreed the valuation of the matrimonial asset in question; (3) Whether a clawback provision is contrary to the clean break principle and/or the principle of fairness which is fundamental in ancillary relief, and, if so, what are the circumstances and/or conditions in which it should be invoked and applied.” 23. As indicated earlier, leave was also granted to W on one issue regarding the clawback order, namely, in relation to the insertion of the words “redevelopment by ADHL”.[25] W had submitted at the hearing for leave that the addition of these words made the order nugatory in practical terms. In her written Case, however, W submitted a revised clawback order in the following terms:- “Proposed Claw Back Provision 1. Contingent upon the occurrence within W’s lifetime of a Specified Event defined in (2) below, H do pay W a sum representing one half of the increase in the market value of H’s ADHL shares attributable to the Specified Event, such value to be agreed by the parties, failing which the same to be assessed by an independent expert to be appointed by the Court and such payment to be paid within one month of the final determination of such valuation. 2. A Specified Event means any one of the following events: (1) A sale by ADHL of its shareholding in SVL at a price which takes into account the redevelopment value of Severn Villa, or the passing of a resolution by ADHL’s board of directors authorizing such sale; (2) A sale by SVL of its interest in Severn Villa at a price which takes into account the redevelopment value of Severn Villa, or the passing of a resolution by SVL’s board of directors authorizing such sale; (3) The passing of a resolution by ADHL’s board of directors or SVL’s board of directors to participate in the redevelopment of Severn Villa. 3. Liberty to either party to apply.” This proposed order was completely new; there had at no previous stage been any discussion about it at all. B.2 The issues to be resolved 24. Within the context of the type of order with which this Court was concerned in the present appeal,[26] the following issues fell to be determined within the three questions identified above:- (1) The general approach. (2) Fairness. (3) Finality and clean break. (4) The quasi-inquisitorial function of the courts and Flywin.[27] 25. These issues can conveniently be dealt with in turn. C THE GENERAL APPROACH 26. I have already made reference to the ancillary relief jurisdiction of the court contained in ss 4, 6, 6A and 7 of the MPPO. The opening words to s 7(1) of the Ordinance are important because the seven particular matters to which specific reference is made, are merely examples of “all the circumstances of the case”. In other words, while in most cases, the court will largely focus on one or more of the seven facets, it should not be restricted to them. This is of some relevance in the present case where the clawback order was made notwithstanding the remoteness of the possibility of the redevelopment of Severn Villa. 27. Authoritative guidance on the court’s approach in ancillary relief proceedings is contained in the decision of this court in LKW, to which reference has already been made.[28] It is obvious of course that the overall objective in ancillary relief proceedings is to achieve a fair outcome between the parties.[29] No two cases ever being the same, the application of the yardstick of fairness, although underlying the exercise of powers in ancillary relief applications, is a matter of judgment[30] and is primarily, if not mainly, a matter for the fact determining tribunal. The use of case law as precedents when what is relied on are not so much legal principles but essentially the application of principles to facts, is to be avoided, particularly in applications for ancillary relief where the primary yardstick is fairness. Apart from fairness, in exercising its powers regarding ancillary relief, the court ought also to bear firmly in mind the question of finality and the clean break principle. So how do these principles apply in the present case, in the context of clawback orders? D FAIRNESS 28. In the present case, the Court was concerned with an equal division of the assets and this ought to have been a relatively straightforward exercise. The particular asset to be considered were the shares in a company (ADHL). The trial judge was entirely correct in his approach, urged on him by the parties at the time, of valuing the shares in ADHL and then dividing the interest of each party in the shares (50% each). In most cases, after the appropriate division of the asset in question is ascertained (50/50 in the present instance), in the case of shares, the court will then determine the best way of dividing the asset between the parties. Sometimes, the simplest method will be to split the shares equally between the parties; in other situations, the court may order that one party should transfer his or her shareholding to the other party in exchange for a payment or equivalent, as happened in the present case. 29. The Court of Appeal, however, instead of leaving intact the order made by the trial judge, made an additional order in terms of the clawback order set out above.[31] The fundamental question to be asked in relation to this type of order was the same as in relation to any other order: is it fair to make it? The Court of Appeal recognised that this was the critical question.[32] 30. In answering this question, a court must evaluate those elements which are relevant to the exercise of discretion and in identifying factors which indicate fairness towards one party, the court should of course also look at any elements of unfairness, being the reverse side of fairness. In considering what is fair, a court must adopt a holistic approach, meaning that it must look at things in the round and not just focus on isolated aspects or on the point of view of only one party. In the present appeal, I believe this is where the Court of Appeal has erred in its approach: it looked at the position of W and the factors favouring her without taking adequately into account the factors going the other way. 31. From W’s point of view, the possibility of the redevelopment of Severn Villa and that this may result in an increase in the value of ADHL shares, could be said to be a relevant factor to be evaluated. The order as made by the judge did not reflect this possibility although he could not be faulted since neither party raised this point as we have seen. That W might not benefit following a possible redevelopment of Severn Villa heavily influenced the Court of Appeal to make the clawback order. Reliance was placed on Parra v Parra[33] where a similar order was made by the court. In that case, a clawback order was made relating to the possibility that planning permission for residential development might be given to the site in question (a brownfield site[34] in Checkendon). Notwithstanding that such an order was not consistent with a clean break and that the terms of the order dealing with its operation were potentially difficult and might result in further litigation,[35] the Court of Appeal in England upheld the making of the order on the basis that on the facts of that case, there would be a “terrible unfairness” if such an order was not made.[36] Particular emphasis was placed on the fact that the scale of any potential windfall, if the land were to be redeveloped, would be great (it would more than treble in value) and on the husband’s evidence at trial that it would indeed be unfair on the wife if she did not share the fruits of any future redevelopment.[37] 32. The fairness of the order on the facts in Parra did not of course make it a fair or appropriate one to be made in the present case. Cheung JA, in his judgment,[38] emphasised the fact that Parra v Parra had “withstood the test of time” and he also made reference to another English case in which a clawback order had been made.[39] With respect, I think that Cheung JA, who has considerable experience in matrimonial matters, has on this occasion rather overestimated the impact of Parra. That case was one in which, on the facts, it was appropriate to make a clawback order and the facts were, as I see them, quite exceptional.[40] 33. The facts of the present case were not exceptional by comparison, would not result in anywhere near a “terrible unfairness” and indeed militated against the making of a clawback order:- (1) The prospect of the redevelopment of Severn Villa was by no means a likely event. In his judgment, Cheung JA made reference to ss 7(1)(a) and (b) of the MPPO but there was no evidence that a redevelopment was “likely” at all. No evidence existed that it was and Mr Fung SC pointed to none, although he asserted the contrary; in fact the evidence pointed the other way. If there had been any credible evidence of the likelihood of redevelopment, it was inconceivable that the redevelopment value of Severn Villa would not have been factored into the valuation of the shares of ADHL. In fact, as we have seen, the NAV of the shares was agreed and no account was taken of the potential redevelopment of Severn Villa. (2) More important, there was nothing to suggest that the making of the clawback order, focusing on the potential redevelopment of Severn Villa, was right or fair in the first place. In Parra, the brownfield site was a property which had a special significance to the husband and wife: they had jointly bought it and it had been intended to be a part of their pension.[41] In the present case, the Severn Villa Property was owned by SVL, a wholly-owned subsidiary of a public company (ADHL) in which the parties held shares. Mr Fung SC submitted that the Severn Villa Property was purchased by W using the proceeds from the sale of properties in the US, but this submission was factually incorrect: see para 12 above. (3) The fact that the Severn Villa Property was held by a wholly-owned subsidiary of a public company was an important consideration of the court to take into account. ADHL was not merely a company which had as its only asset and business the Severn Villa Property. I have already referred to the other properties in the BVI and Panama.[42] In addition, the company had at least 12 other subsidiary companies apart from SVL.[43] Although Mr Fung described the company as “stagnant”, it simply could not be said that the company had no business prospects. There was evidence before the court by the time the appeal was heard of the fact that in May 2013, H had invested $31.08 million in convertible bonds issued by ADHL in order to protect his majority shareholding in the company. There was also evidence of a rights issue of between $100 million to $150 million needed by the company to further its business. All this suggested (and this was not disputed) that ADHL had an on-going business or at least might have future business prospects. It is difficult to see why it was fair to the parties to concentrate only on the Severn Villa Property and ignore the rest of the company’s business. As far as W was concerned, it was probable that the Severn Villa Property represented the best asset of the company but no reason was given as to why this asset should be isolated for consideration without any account being taken of the company’s other businesses. It seemed that the effect of the clawback order in the present case was that W would get the benefit of what was likely to be the company’s best asset without taking the burden of the further investment necessitated by the company’s financial condition and attendant risks. I would reiterate here that the relevant matrimonial asset with which we were concerned, were the parties’ shares in ADHL. As Mostyn J asked rhetorically in FZ v SZ[44] (referring to his own judgment in GW v RW (Financial Provision: Departure from Equality),[45] “why should one party receive most of the plums leaving the other with most of the duff?” (4) On this theme, it may further be pointed out that as ordered by the Court of Appeal, the relevant event was the redevelopment by ADHL of Severn Villa. If ADHL were to redevelop the property, there would be substantial expenses involved, without any certainty that there would be a profit to the company. W was not going to contribute to the expenses or share in any of the risks. Moreover, whatever might otherwise be the state of the company, as long as there was a “notional” increase in the value of ADHL shares attributable to any redevelopment (less $230 million[46] and expenses), W would stand to gain. (5) The points made above apply with even greater force to the orders advanced by W in this appeal. If the words “by ADHL” were deleted, then even if ADHL did not itself redevelop, H would still be liable to W under the terms of the clawback order. On the new form of the clawback order proposed by W in this appeal,[47] quite apart from the practical workability of it,[48] it is abundantly clear that W stood to gain even if ADHL did not; she would certainly not share in any losses, expenses or any of the risks. It will be observed that the new draft proposed by W did not make any allowance for either the $230 million or the costs and expenses of any redevelopment . (6) On the point that W was only taking the upside without the downside, the Court of Appeal accepted W’s submission[49] (maintained in this Court by Mr Fung SC) that there was no unfairness since she was only asking for an uplift in the event of redevelopment and not an uplift on other grounds. With respect, this provides no answer at all to the point. 34. On the aspect of fairness alone, it was inappropriate to make a clawback order. However, there is more. It was inappropriate to make the order when one came to consider the desirability of adhering to the clean break principle and the Court of Appeal ought not in any event to have embarked on considering the making of a clawback order in the first place. I now consider these two aspects. E FINALITY AND CLEAN BREAK 35. The desirability of finality in litigation has a somewhat special application in matrimonial matters in that while it is desirable to have a clean break between parties, the law recognises that this may not always be possible in the interests of fairness. Thus, for example, particularly where children are concerned, orders are frequently made for periodic payments which look to the future.[50] Various orders which a court can make under ss 4, 6 or 6A of the Ordinance also look to the future.[51] 36. Nevertheless, in accepting the above, one must as far as possible try to achieve a clean break between parties who, following the dissolution of a marriage, can often be severely at odds with one another. The desirability of a clean break should be at the top of a judge’s mind when considering issues of ancillary relief. While there is no statutory backing for this approach,[52] it is a common law principle of long standing. In LKW, reference was made to the clean break principle.[53] In Miller v Miller,[54] to which reference was made in LKW, the desirability of a clean break was emphasised in a number of passages.[55] In one, Lord Nicholls of Birkenhead referred to a familiar passage in the speech of Lord Scarman in Minton v Minton[56]: “An object of the modern law is to encourage [the parties] to put the past behind them and to begin a new life which is not overshadowed by the relationship which has broken down.” I would also echo the sentiments of Stock JA in L v C[57]:- “…That policy consideration is one that itself recognizes the considerable trauma that attends prolonged and costly matrimonial disputes, and the obvious advantage in settlement rather than in the litigious cauldron that stirs bitterness, prolongs uncertainty, and is the enemy of the clean break that enures to the advantage of husband, wife and children.” 37. As stated earlier, it may on occasion be desirable or fair to make orders which do have the effect of looking to the future, but not so as far as the clawback order was concerned in the present case. The terms of the clawback order made by the Court of Appeal were fraught with difficulties and would inevitably have given rise to substantial disputes between the parties in the future. I need only refer to a few of these difficulties:- (1) The term “commencement of the redevelopment” itself causes difficulties. It is evident that this term could possibly cover a number of scenarios. It cannot, as counsel submitted, be restricted only to “breaking ground” on the relevant site. (2) Further, a question will also arise as to the correct time (whether a point in time or a period) to ascertain the “notional increase” in the value of 418,945,640 ADHL shares attributable to the redevelopment. The fact that this notional value was to be net of $230 million, and the costs and expenses relating to the redevelopment, would seem to indicate that one should look at the position after the redevelopment has been completed. This, however, is by no means free from difficulties either. (3) Although an independent court expert is to be utilized, no doubt the parties would have much to put to this expert regarding any valuation. 38. These difficulties would remain if the words “by ADHL” are omitted from the draft. 39. On the new draft proposed by W on her appeal, quite apart from aspects giving to the fairness of such an order,[58] it was also full of potential difficulties; again I highlight just some of them:- (1) Instead of a notional increase in the value of the ADHL shares attributable to a redevelopment, reference is now made to the “market value” of the shares attributable to any of the Specified Events. At its lowest, this is an extremely difficult exercise, requiring the expertise not just of accountants, but perhaps also stockbrokers and market analysts. In these areas, it is likely that there would be differing opinions. Valuations are often a matter of opinion on which experts will differ.[59] (2) Two of the Specified Events refer to sales (either by ADHL of its shareholding in SVL or by SVL of its interest in the Severn Villa Property) which take “into account the redevelopment value of Severn Villa”. This, again, is an extremely controversial, not to mention difficult, exercise. (3) The third Specified Event refers to the passing of a resolution by ADHL’s board of directors “to participate in the redevelopment of Severn Villa”. This is vague: the word “participation” covers a number of different situations ranging from a redevelopment on its own to merely joining in some aspect of it. 40. I have earlier referred to the trial judge’s assessment of the conduct of the parties in the litigation.[60] It reflects the worst characteristics of matrimonial litigation. But the court should not make orders which will have the potential of prolonging the bitterness that may exist between former spouses, unless this is an inevitable consequence of what is seen to be fair and just. There was simply no necessity or justification to make an order having this effect in the present case. F THE QUASI-INQUISITOIRAL FUTURE OF THE COURT AND FLYWIN 41. I have already made the point that, at trial, the aspect of the redevelopment of Severn Villa, although vaguely raised, was never pursued by W and that W certainly made no submissions regarding a clawback order.[61] The matter was only raised at the appeal stage in spite of H’s objections. 42. In his judgment, Cheung JA was of the view that where the parties had not raised an issue, this did not prevent the court from doing so of its own motion. Reliance was placed[62] on the decision of the English Court of Appeal in Parra v Parra where Thorpe LJ mentioned the quasi-inquisitorial role of the court in ancillary relief litigation.[63] 43. As H pointed out in his written Case, there are limits as to how far this point can be stretched, particularly in the present case. What it does not mean is that matrimonial litigation is some form of “free for all” where procedural rules and practice, coupled with their rationale,[64] have no application. In matrimonial litigation, I daresay that the court would exercise its quasi-inquisitorial role most acutely where children are concerned, or if one or the other party or both parties suffered from some disadvantage. Where, however, both parties are properly represented (as in the present case where at every stage, H and W have been represented by leading counsel), the court can in almost every case, unless some exceptional circumstances exist, rely on the parties to look after their own interests. It would be quite wrong in such circumstances for the court to second-guess, much less override, the parties’ approach to the litigation, much of it being driven by tactical considerations. 44. There was no justification in the present case for the Court of Appeal to entertain new submissions based on the redevelopment of Severn Villa at the appeal stage. The parties had not raised this issue at the trial stage and probably had their reasons not to do so (as we have seen, they were certainly aware of the possibility of redevelopment).[65] 45. There was also a compelling reason for the Court of Appeal not to entertain any arguments relating to a clawback order. If the question of a clawback order had been properly raised at trial, the evidence adduced by both parties would have addressed this issue in full. I have dealt extensively with aspects of fairness and the foreseeable problems that a clawback order would have caused.[66] There is no doubt that the evidence of the parties, particularly that of H, would have dealt with at least some of these aspects. 46. In Flywin, this Court laid down a salutary rule that where a new point is taken on appeal by a party, that party will be barred from doing so “unless there is no reasonable possibility that the state of the evidence relevant to the point would have been materially more favourable to the other side if the point had been taken at the trial.”[67] As Mr Justice Bokhary PJ stated,[68] this rule is based on fairness. In the present appeal, this ought to have been a decisive consideration. There is no doubt that at least on H’s part, he would have wished to adduce additional evidence had the clawback order been a live issue then. Apart from anything else, he might not have been quite so willing to agree the NAV of the ADHL shares. G CONCLUSION 47. For these reasons, H’s appeal was allowed and W’s appeal dismissed. 48. As to costs, the parties wished to have sight of the reasons for our decision before making their submissions. The parties should within 14 days of the handing down of this judgment serve on each other and lodge with the Registrar of the Court any written submissions, with liberty on the parties to serve and lodge written submissions in reply within 14 days thereafter. Mr Justice Ribeiro PJ: 49. I agree with the judgment of the Chief Justice. Mr Justice Chan NPJ: 50. I agree with the judgment of the Chief Justice. Mr Justice Stock NPJ: 51. I agree with the judgment of the Chief Justice. Lord Walker of Gestingthorpe NPJ: 52. I agree with the judgment of the Chief Justice. Chief Justice Ma: 53. On costs, the Court makes the order set out in para 48 above. Mr Benjamin Yu, SC and Ms Bonnie Y.K. Cheng, instructed by Stevenson Wong & Co., for the Respondent (Appellant in FACV10/2014 and Respondent in FACV 11/2014) Mr. Daniel R. Fung, SC and Mr. David Chen, instructed by Vivien Chan & Co., for the Petitioner (Respondent in FACV 10/2014 and Appellant in FACV 11/2014) [1] Cap 192. [2] See Parra v Parra [2003] 1 FLR 942, at para [4]. This case is further discussed below. [3] In a Determination dated 23 October 2014 (CJ Ma, Ribeiro PJ and Chan NPJ). [4] W was represented by Mr Daniel Fung SC and Mr David Chen. H was represented by Mr Benjamin Yu SC and Ms Bonnie Cheng. [5] This was the parties’ agreed value based on an average of the valuations of three well-known valuers. This collection, famous in the Chinese antique furniture world, was known as the “Hung Collection”. [6] She became a non-executive director. She was removed from this post in July 2010. [7] Deputy High Court Judge Carlson, in a judgment dated 10 August 2012, at para 16. [8] In HCMP 243 of 2011. [9] [2010] 13 HKCFAR 537. [10] In para 7 above. [11] These and other issues, both legal and factual, were spent by the time the appeal was heard before us. [12] The net asset value of the total shares in the company was agreed at $410,568,000. [13] The judge valued this at $250 million. [14] The number of shares in ADHL held between H and W was 418,945,640 shares. [15] The main reason for this being that if the shares were held by one person, the holder would enjoy control of the company. The agreed NAV of the shares, higher than the market price, reflected this. [16] This was on the basis of the redevelopment of all six houses comprising Severn Villa, not just the four owned by SVL. [17] Not Mr Fung. [18] By the appeal stage, as indicated earlier (see para 16 above), it had been ordered that W transfer her shareholding in ADHL to H. [19] The reasons are contained in the judgment of Cheung JA, with which the other members of the Court (Lam VP and Fok JA) agreed. [20] See para 3 footnote 2 above. [21] Sections 7(1)(a) and (b) state:- “7 Matters to which court is to have regard in deciding what orders to make under sections 4, 5 and 6 (1) It shall be the duty of the court in deciding whether to exercise its powers under section 4, 6 or 6A in relation to a party to the marriage and, if so, in what manner, to have regard to the conduct of the parties and all the circumstances of the case including the following matters, that is to say- (a) the income, earning capacity, property and other financial resources which each of the parties to the marriage has or is likely to have in the foreseeable future; (b) the financial needs, obligations and responsibilities which each of the parties to the marriage has or is likely to have in the foreseeable future;” [22] CA judgment para 81(4). [23] The clean break principle is discussed below. [24] CA judgment para 81(3) . [25] See para 19 above. [26] The nature of the clawback order has been stated in para 1 above. [27] Flywin Company Limited v Strong & Associates Limited (2002) 5 HKCFAR 356. [28] See para 15 above. [29] Paras 24 and 56 of LKW. [30] See the remarks of Sir Mark Potter P in B v B [2008] 1 FCR 613, at para [54]. [31] Para 2 above. [32] CA judgment para 74. [33] Citation is given above in para 3 footnote 2. [34] Brownfield sites are properties which have limited development potential owing to the fact that they may contain industrial or chemical waste or residue. In Hong Kong, this term usually relates to “damaged” agricultural or industrial land. [35] See Parra at para [29]. [36] This term was used by the trial judge in that case: see Parra at para [18]. [37] See Parra at para [31]. [38] CA judgment para 79. [39] B v B, referred to earlier at para 27 footnote 30 above. [40] The Court of Appeal in that case regarded the clawback order as being “highly exceptional”: see Parra at para [29]. [41] See the decision of the trial judge: P v P [2002] 2 FLR 1075, at para [14]. [42] Para 11 above. [43] According to ADHL’s Annual Report for 2011. [44] [2011] 1 FLR 64 at para [109]. [45] [2003] 2 FLR 108 at para [64]. [46] This was the value of ADHL’s interest in SVP. [47] Para 23 above. [48] As to which, see section E below. [49] CA judgment para 81(3); see para 18(5) above. [50] See s 5 of the MPPO. [51] Again, such as periodic payments (s 4(1)(a)) or any order for the sale of property taking place in the future (s 6A(1)). [52] In contrast, say, to the position in England: see s 25A of the Matrimonial Causes Act 1973. [53] At para 74. [54] [2006] 2 AC 618. [55] At para 35 (Lord Nicholls of Birkenhead) and 144 (Baroness Hale of Richmond). [56] [1979] AC 593, at 608. [57] [2007] 3 HKLRD 819, at para 38. [58] See para 33(5) above. [59] See, for example, Miller v Miller at para 26, in a passage referred to in LKW at para 88. [60] See para 13 above. [61] See para 17 above. [62] CA judgment para 74. [63] At para [22]. This passage was referred to in LKW at para 69. [64] See, for example, the underlying objectives of the Civil Justice Reform. [65] See para 17 above. [66] See Sections D and E above. [67] At para 38. [68] At para 39. 
Mr Justice Ribeiro PJ: 1. This is the judgment of the Court. At the hearing, we dismissed the respondent’s application with reasons to be provided later. These are our reasons. 2. This is an unusual case in which the respondent (the prosecution) seeks to re-open a judgment of this Court and to have it set aside on the ground that it was obtained by the fraud of the appellant (the defendant). A. The relevant proceedings 3. While serving a sentence of 18 years’ imprisonment imposed on 23 June 1998 for drug trafficking, the defendant was convicted on 3 June 2005 for common assault on a Correctional Services Department officer after trial in the Magistrates’ Court.[1] He was sentenced to six months’ imprisonment consecutive to his existing 18 year sentence. His appeal to the Court of First Instance[2] against conviction and sentence was dismissed on 7 February 2006. 4. The defendant then sought leave out of time to appeal to this Court, relying on an affirmation dated 13 October 2008 which exhibited two statements purportedly made by the two CSD officers who had testified against him at the assault trial. The statements, which the defendant said he had come across by chance and had then obtained from the CSD, differed materially from the officers’ witness statements and their testimony at the trial. They had not been disclosed to the defence. Leave to appeal was granted and, on the hearing of the appeal, counsel for the prosecution conceded that the conviction could not be supported, given the existence of those undisclosed statements. Accordingly, on 30 June 2009, the Court quashed the defendant’s common assault conviction.[3] 5. The Court did not know that doubts had been cast on the genuineness of the documents tendered by the defendant and that an investigation was then underway. The defendant’s similar use of certain documents in connection with three other cases – two of common assault on prison personnel and one involving offences against prison discipline – was also under investigation. Although the prosecutor knew that the CSD officers concerned had denied making the statements being relied on by the defendant before this Court and that an investigation was in progress, he did not consider it necessary to inform the Court of the position and did not apply for the appeal to be adjourned, as one might have expected. 6. Those investigations led to the defendant being charged with and (after a 96 day trial) convicted of doing acts tending and intended to pervert the course of public justice in relation to the appeal heard by this Court.[4] The Judge found it proved beyond reasonable doubt that the defendant had instructed someone outside the prison to help forge the documents concerned (including those exhibited to his abovementioned affirmation), his purpose being to use them to mislead the judges with a view to gaining an unfair advantage for himself in the proceedings.[5] 7. The Judge also convicted the defendant on two other charges of perverting the course of justice and one charge of using copies of false instruments which he knew or believed to be false. Those convictions related to the three other cases mentioned above. They involved his tendering similar documents in attempting to overturn two convictions for assault for which he had been sentenced respectively to a fine of $1,500 and two months in prison. They also related to his use of similar documents with a view to mounting a judicial review challenge to his cumulative loss of 440 days of remission of sentence resulting from 60 adverse disciplinary decisions. The Judge sentenced him to four years’ imprisonment on each of the abovementioned charges, to run concurrently. 8. His appeal against conviction was dismissed by the Court of Appeal on 5 February 2013.[6] On the prosecution’s application for a review of the Judge’s sentences, the Court of Appeal held that the four-year sentence for deceiving this Court was manifestly inadequate. It noted that the offence was carefully planned, involving outsiders and multiple forged documents. It involved accusing CSD officers of framing him and committing perjury, putting their reputations and careers at risk. His scheme took up much public expenditure and judicial time, “immersing the judicial and prison systems in a stew of deception”.[7] It was an offence which “strikes at the very heart of the system by which law and order is enforced in society” threatening to “destroy public confidence in that system”.[8] Their Lordships therefore increased the sentence to one of six years’ imprisonment. 9. Their Lordships also held that the Judge should not have ordered all the sentences to run concurrently: “Each offence added to the culpability of the offender. So consecutive sentences were, in principle, warranted. But the totality of wholly consecutive sentences for the four offences would be dramatically excessive. We adjudge a totality of 8 years’ imprisonment to be the least that would properly be warranted; but given the jurisdictional sentencing limit of the District Court, that totality must be reduced to one of 7 years’ imprisonment. ... Since that sentence is greatly in excess of the original sentence, we shall reduce it further by three months and secure a total of 6 years 9 months’ imprisonment.”[9] B. The present application 10. The prosecution has very properly brought the present application[10] to enable the Court to consider what, if any, steps should be taken in the light of the developments subsequent to the defendant’s successful appeal. 11. There is no doubt that the Court has a discretionary power to re-open an appeal where a reasonable basis exists for alleging that its judgment was obtained by fraud. There is no express provision concerning the re-opening of appeals in the Court’s statute,[11] but it is clear that the Court has an implied power exceptionally to order an appeal to be re-opened where justice so demands, such power being reasonably required for the effective exercise of the judicial power granted by the Basic Law.[12] 12. In deciding whether that discretion should be exercised, the Court should obviously take all relevant considerations into account. It is necessary first to consider what purpose would be served by re-opening the appeal. In a case like the present, a question of central importance is whether it would be the Court’s object to reinstate the original sentence of six months’ imprisonment for the defendant’s assault on the CSD officer if it were to be found after inquiry that its judgment was obtained by fraud. If that sentence would not be reinstated even on a finding of such fraud, that would be a powerful argument against re-opening the appeal. 13. In our view, in the light of the successful prosecution of the defendant for perverting the course of justice and of the very substantial sentence – increased by the Court of Appeal from four to six years – imposed to punish him for his deception of this Court, it would not be appropriate to reinstate the original six-month sentence. The six-year sentence reflects the defendant’s overall criminality, including his attempted evasion of the original six-month sentence. It sufficiently demonstrates that perverting the course of justice does not pay, but results instead in a severely increased sentence. Moreover, the Court of Appeal’s capping of the overall sentence at six years and nine months, applying the totality principle and avoiding excessive disparity with the previous sentence, provides a further reason for concluding that the original six-month sentence should not, in the present case, be reinstated as an addition to that overall sentence. 14. Once that conclusion is reached, it becomes clear that the re-opening of the appeal and, if fraud is established, the setting aside of the Court’s judgment would have only symbolic value and no practical consequences. It would serve to mark the Court’s condemnation of the deception practised upon it and to remove from the Court’s record a judgment fraudulently obtained. That is, however, a relatively inconsequential benefit. No one could possibly doubt the condemnation of the defendant’s attempted perversion of the course of justice reflected in his being ordered to serve an additional six years and nine months in prison. Mr Wesley Wong SC submitted on the prosecution’s behalf that the Court should aim to reverse the fraudulent expunging of the defendant’s conviction for assault. However, while that conviction has been expunged, it has been replaced by a conviction recording the defendant’s fraudulent conduct and carrying a far heavier sentence. 15. While there may be some room for argument, in our view, the present application for the Court to re-open its appeal and to set aside its judgment should be approached on the basis that it falls within the category of civil proceedings, separate from the underlying criminal proceedings concerning the assault. Section 62 of the Evidence Ordinance[13] therefore applies, materially providing as follows: “(1) In any civil proceedings the fact that a person has been convicted of an offence by or before any court in Hong Kong shall ... be admissible in evidence for the purpose of proving, where to do so is relevant to any issue in those proceedings, that he committed that offence, whether he was so convicted upon a plea of guilty or otherwise and whether or not he is a party to the civil proceedings; ... (2) In any civil proceedings in which by virtue of this section a person is proved to have been convicted of an offence by or before any court in Hong Kong – (a) he shall be taken to have committed that offence, unless the contrary is proved; and (b) without prejudice to the reception of any other admissible evidence for the purpose of identifying the facts on which the conviction was based, the contents of any document which is admissible as evidence of the conviction, and the contents of the information, complaint, indictment or charge on which the person in question was convicted, shall be admissible in evidence for that purpose.” 16. It follows that while reliance can be placed by the prosecution on the defendant’s District Court conviction as evidence that he did obtain the judgment by fraud, the defendant would be entitled under section 62 to seek to prove the contrary.[14] 17. It seems likely that if the Court were to re-open the judgment to determine whether its judgment was in fact obtained by fraud, it would have to remit that issue to a first instance court for trial and a report. Bearing in mind the 96-day duration of the trial which has already taken place, the risk of such remitter proceedings being protracted must be recognized. The judicial and public resources which would have to be employed are likely to be wholly disproportionate to the merely symbolic potential value of the exercise. The Court would not be deterred by such considerations if the offence in question was serious and the public interest demanded determination of the fraud issue. But the present case does not bear those characteristics. 18. In our view, the abovementioned considerations weigh decisively against the re-opening of the appeal. We also think it relevant to take account of the fact that the prosecution contributed to the defendant obtaining the Court’s judgment by conceding the appeal without mentioning the ongoing investigation and without seeking an adjournment. 19. For the aforesaid reasons, the Court declined to exercise its discretion to re-open the appeal and dismissed the application while making it clear that no criticism of the prosecution is thereby intended. 20. We wish to express our gratitude to Mr Jeffrey Chau who acted as Amicus Curiae and provided the Court with thoughtful and comprehensively researched submissions which were of great assistance. Mr Wesley WC Wong SC and Mr Andrew HC Li of the Department of Justice for the Respondent (Applicant) Mr Gerard McCoy SC and Mr Timothy Parker instructed by Tso, Au, Yim & Yeung and assigned by the Legal Aid Department for the Appellant (Respondent) Mr Jeffrey Chau, Amicus Curiae [1] Before T H Casewell Esq, ESCC 3093/2004. [2] Deputy High Court Judge Leong, HCMA 621/2005. [3] FACC No 12 of 2008 (Reasons 8 July 2009). [4] Before H H Judge E Yip in DCCC 1431/2009 (6 January 2012). [5] At §§68 and 70. [6] Stock VP, Fok JA and Beeson J, CACC 18/2012 (Reasons 6 March 2013). [7] At §50. [8] At §52. [9] At §§60-61. [10] By Notice of Motion dated 4 September 2013, FAMP No 3/2013. [11] Hong Kong Court of Final Appeal Ordinance, Cap 484. [12] Government of HKSAR v Scott (2006) 9 HKCFAR 221, §4; PCCW-HKT Telephone Ltd v Telecommunications Authority (2005) 8 HKCFAR 337, §§40 and 60; Joint & Several Liquidators of Kong Wah Holdings Ltd v Grande Holdings Ltd (2006) 9 HKCFAR 795, §13. [13] Cap 8. [14] See Hunter v Chief Constable of the West Midlands Police 		[1982] AC 529 at 544. 
Chief Justice Ma: 1. I agree with the judgment of Mr Justice Spigelman NPJ. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Mr Justice Spigelman NPJ. Mr Justice Tang PJ: 3. I agree with the judgment of Mr Justice Spigelman NPJ. Mr Justice Fok PJ: 4. I agree with the judgment of Mr Justice Spigelman NPJ. Mr Justice SpigelmanNPJ: Procedural History 5. After trial before HH Judge Douglas TH Yau,[1] the Appellant was found guilty of an offence under s 25(1) and (3) of the Organized and Serious Crimes Ordinance, Cap 455 (“OSCO”). He was sentenced to two and a half year’s imprisonment. The Court of Appeal[2] dismissed the Appellant’s appeal against conviction[3] and refused leave to appeal to this Court.[4] 6. On the same day, 31 May 2013, the Court of Appeal granted bail to the Appellant, on his undertaking to file a Notice of Application for Leave to Appeal to the Court of Final Appeal within 7 days. The Appellant filed the Notice on the same day. 7. On 5 December 2013, this Court granted leave to the Applicant to appeal, limited to the following questions: (a) In relation to offences under section 25(1) of OSCO, when seeking to determine whether a person has reasonable grounds to believe that the property dealt with represents the proceeds of an indictable offence, what facts and matters are the judge or jury entitled to take into account as constituting such reasonable grounds? In particular, in addition to the objective facts relating to the Appellant's dealing with the property and known by the Appellant to exist, are they entitled to take into account of the Appellant’s perception and evaluation of those facts as constituting or contributing to reasonable grounds for believing that the property does not represent such proceeds? (b) In the determination of the objective element of the offence, what is the appropriate standard to be applied in evaluating the content of the reasonable man’s belief? (c) On the true construction of section 25(1) of OSCO, is it a defence to a charge under that section, the burden of which is on the Appellant to discharge, to show that notwithstanding the establishment by the prosecution of reasonable grounds to believe that the property dealt with represents the proceeds of an indictable offence, he nonetheless honestly and reasonably did not suspect the property to represent such proceeds? 8. On 11 December 2013, the Appellant filed a Notice of Appeal in this Court. The Statutory Scheme 9. The terms of s. 25 and s. 25A, of OSCO are: “25. Dealing with property known or believed to represent proceeds of indictable offence (1) Subject to section 25A, a person commits an offence if, knowing or having reasonable grounds to believe that any property in whole or in part directly or indirectly represents any person's proceeds of an indictable offence, he deals with that property. (2) In proceedings against a person for an offence under subsection (1), it is a defence to prove that- (a) he intended to disclose to an authorized officer such knowledge, suspicion or matter as is mentioned in section 25A(1) in relation to the act in contravention of subsection (1) concerned; and (b) there is reasonable excuse for his failure to make disclosure in accordance with section 25A(2). (3) A person who commits an offence under subsection (1) is liable- (a) on conviction upon indictment to a fine of $5000000 and to imprisonment for 14 years; or (b) on summary conviction to a fine of $500000 and to imprisonment for 3 years. (4) In this section and section 25A, references to an indictable offence include a reference to conduct which would constitute an indictable offence if it had occurred in Hong Kong. 25A. Disclosure of knowledge or suspicion that property represents proceeds, etc. of indictable offence (1) Where a person knows or suspects that any property- (a) in whole or in part directly or indirectly represents any person's proceeds of; (b) was used in connection with; or (c) is intended to be used in connection with, an indictable offence, he shall as soon as it is reasonable for him to do so disclose that knowledge or suspicion, together with any matter on which that knowledge or suspicion is based, to an authorized officer. (2) If a person who has made a disclosure referred to in subsection (1) does any act in contravention of section 25(1) (whether before or after such disclosure), and the disclosure relates to that act, he does not commit an offence under that section if- (a) that disclosure is made before he does that act and he does that act with the consent of an authorized officer; or (b) that disclosure is made- (i) after he does that act; (ii) on his initiative; and (iii) as soon as it is reasonable for him to make it. (3) A disclosure referred to in subsection (1)- (a) shall not be treated as a breach of any restriction upon the disclosure of information imposed by contract or by any enactment, rule of conduct or other provision; (b) shall not render the person who made it liable in damages for any loss arising out of- (i) the disclosure; (ii) any act done or omitted to be done in relation to the property concerned in consequence of the disclosure. (4) In the case of a person who was in employment at the relevant time, this section shall have effect in relation to disclosures to the appropriate person in accordance with the procedure established by his employer for the making of such disclosures as it has effect in relation to disclosures to an authorized officer. (5) A person commits an offence if, knowing or suspecting that a disclosure has been made under subsection (1) or (4), he discloses to any other person any matter which is likely to prejudice any investigation which might be conducted following that first-mentioned disclosure. (6) In proceedings against a person for an offence under subsection (5), it is a defence to prove- (a) that he did not know or suspect that the disclosure concerned was likely to be prejudicial in the way referred to in that subsection; or (b) that he had lawful authority or reasonable excuse for making that disclosure. (7) A person who contravenes subsection (1) commits an offence and is liable on conviction to a fine at level 5 and to imprisonment for 3 months. (8) A person who commits an offence under subsection (5) is liable- (a) on conviction upon indictment to a fine of $500000 and to imprisonment for 3 years; or (b) on summary conviction to a fine at level 6 and to imprisonment for 1 year. …” The persons involved 10. The Appellant came to Hong Kong from the Mainland in 1964 and established a successful garment manufacturing business. At the time of the trial at the end of 2011, he owned two factories on the Mainland, each occupying 40 acre sites, together employing a total of 1,000 workers. He also had a profitable business in Bangladesh where over 1,000 workers were employed. In Hong Kong, the Appellant had two premises, each covering more than 10,000 sq ft, in which 50 employees managed his overseas business. He was aged 66 at the time of the trial and was married, with three grown-up children. He had a clear record. 11. Kwok Wing (“Kwok”) and the Appellant had been close friends for over 30 years. Kwok was the chairman and a major shareholder in a Hong Kong listed company called Tack Fat Group International Limited (“Tack Fat”). He was also in the garments business, operating on a very large scale, with businesses in various countries including Cambodia, where he had a factory employing 20,000 workers. Among the sites he operated on the Mainland, were two factories with a combined area of 100,000 sq ft which he leased from the Appellant. 12. They both lived in Mei Foo Sun Chun and their families were close, frequently socialising together over the years. Kwok and the Appellant had also helped each other with unsecured, interest-free loans on occasion when cash-flow difficulties arose. Thus, in 1998, Kwok lent the Appellant $2 million and in 2005 or 2006, the Appellant reciprocated twice, lending Kwok $5 million and $2 million respectively. All the loans were repaid after a short time. The transaction leading to the money laundering charge 13. At the end of July 2008, Kwok called the Appellant, saying that two of his friends on the Mainland were going to return some money and asked the Appellant to let Kwok have the Appellant’s bank account details so that the money could be remitted to that account and kept for Kwok. The Appellant agreed without asking any questions about Kwok’s reasons for wishing to use his account, rather than one of his own. He instructed his bookkeeper to let Kwok have the bank account details of one of his companies called Mickles International Ltd (“Mickles”). 14. A “receipt” dated 29 July 2008, issued by Kwok to the Appellant, stated that Kwok would make payments into the Mickles account and that the sums would be returned to him. 15. On 1 August 2008, remittances of HK$7,582,150 and HK$6,467,230, made upon Kwok’s instructions by Guo Jinsheng (“Guo”) and Pan Duanfeng (“Pan”) respectively, were received in the Mickles account. 16. At Kwok’s request, on 27th August 2008 the Appellant caused the funds so received to be remitted in the total amount of HK$14,049,380.00 to Kwok’s company in Cambodia called Asia World Agricultural Development (Cambodia) Limited. Kwok’s fraud 17. On the basis of authority, it was unnecessary for the prosecution to prove that these monies were, as a matter of fact, the proceeds of an indictable offence, because the conduct element of the offence is dealing with property (see Oei Hengky Wiryo v HKSAR (No 2) (2007) 10 HKCFAR 98). Nevertheless, in this case, the prosecution did so. 18. The Appellant’s evidence was that he was unaware of anything amiss in connection with Kwok or Tack Fat until September or October 2008 when he was told that the workers in one of the factories that Kwok had leased from him had not been paid and that the machinery had had to be auctioned off to pay them. It became obvious that Tack Fat was in dire financial straits. It was fortunate for the Appellant that the auction proceeds covered the unpaid wages because he would, it seems, otherwise have been liable for the wages as lessor. The Appellant had not seen Kwok since late July 2008 and attempts to contact him after his failure to pay the wages failed. Kwok has not been heard from since. 19. It subsequently emerged, as established by the prosecution at the trial, that Kwok had conspired with Guo and Pan to perpetrate a fraud on Tack Fat. At a board meeting on 4 June 2008, Kwok had caused 20 million Tack Fat share options to be granted to each of Guo and Pan. Then at a board meeting on 6 June 2008, Kwok produced two cheques of HK$13 million each which purported to represent payment by Guo and Pan for the shares purchased by them in exercising those share options. Forty million Tack Fat shares were duly issued to Guo and Pan, but the company never received any payment for them. The two $13 million cheques produced by Kwok were never presented for payment. 20. Guo and Pan sold off the shares in the market between 16 June and 29 July 2008, the sales yielding proceeds of HK$9,562,424.00 and HK$10,873,478.53 respectively. The 1 August 2008 remittances of HK$7,582,150 and HK$6,467,230 received in the Mickles account, and the sum of HK$14,049,380 was transferred on 27 August to Kwok’s company in Cambodia, all derived from the proceeds of those share sales and thus from Kwok’s conspiracy to defraud, an indictable offence. The Charge 21. The Appellant was charged with dealing with property known or believed to represent proceeds of an indictable offence, contrary to sections 25(1) and 25(3) of OSCO, particularised by reference to the Appellant causing the funds to be remitted to Cambodia, as follows: “... on or about the 27th day of August 2008, in Hong Kong, knowing or having reasonable grounds to believe that property, namely the sum of $14,049,380.00 Hong Kong currency, in whole or in part directly or indirectly represented any person’s proceeds of an indictable offence, dealt with the said property”. 22. Notwithstanding the terms of the charge, the prosecution did not pursue a case of “knowledge”. The case was directed only to the alternative basis of liability, namely, “reasonable grounds to believe”. The prosecution’s case 23. The prosecution asserted that the Appellant had reasonable grounds to believe that the property represented the proceeds of an indictable offence, relying, in this Court, on the following: 1. Quantum of the money transferred: $14M+. 2. Manner and origins of transfer: 2 deposits from 2 HK accounts held by 2 unknown mainlanders. 3. Transnational dimension of the transfer: mainlanders paying $14M to Appellant’s HK bank account for onward transmission to Cambodian bank account. 4. $14M transferred to avoid deposit in Kwok’s HK business or personal bank accounts. 5. Appellant received and held $14M for 26 days in an inactive business bank account in Hong Kong. 6. Neither the Appellant nor Kwok gave any reason for why it was necessary for the $14M to be deposited into the Appellant’s company account before it was transmitted to Cambodia. 7. In late July 2008, Kwok told the Appellant to hold money temporarily for him. The next time they would speak was in late August 2008 when Kwok told the Appellant to transfer $14M to Kwok’s business account in Cambodia. The Appellant executed all of Kwok’s instructions without delay, question or report to the authorities. 8. When the Appellant agreed to hold the money, he heard from Kwok that two mainland friends were returning two payments of several million dollars to Kwok, that the Appellant was to hold it temporarily in the Appellant’s Hong Kong bank account. No mention was made of the names of the mainlanders, the precise amount of the money, the length of time to hold the money, or the nature of the transaction with Kwok. The Appellant made no inquiry of these or any other matter and arranged for his accounting manager to facilitate the transfer. 9. When told about the incoming money from Kwok, the accounting manager’s first reaction was to ask about the nature of the money. The Appellant said he did not know. 10. Kwok and the Appellant knew each other for over 30 years and were very close friends who regularly socialized and played sports together. They would go diving every weekend and play ping-pong on Tuesdays. The last time they saw each other was while playing ping-pong in late July 2008. After Kwok called the Appellant to ask him to hold the money, they stopped seeing each other. 11. Without having seen Kwok for four weeks, the Appellant in late August received Kwok’s call and request to transfer $14M to Kwok’s designated bank account in Cambodia. By this time, the Appellant knew the names of the mainlanders and the total sum of the amount transferred. Without making any inquiry, the Appellant promptly arranged to have this instruction executed by his accounting manager. 12. Never before had Kwok asked the Appellant to hold money temporarily for Kwok. Both had loaned money to each other in the past and paid back the loans. 13. The Appellant’s mainland manufacturing business had an annual turnover of $400M but a net profit of only $8M. The Appellant had no business in Cambodia. 14. Kwok was renting two factory premises of about 100,000 square feet on the mainland from the Appellant at the material time. 15. The Appellant knew that Kwok was the chairman of a Hong Kong publicly-listed company, Kwok had factories in the mainland, and Kwok’s business was 10 times bigger than his own. He also knew that Kwok had a factory building in Cambodia that employed 20,000 workers[5]. The Appellant’s case 24. The Appellant’s case was fundamentally that, on entirely reasonable grounds, he had trusted Kwok implicitly and had had no reason to suspect that the money remitted to his account had anything to do with any criminal offence. 25. He and Kwok had been close friends for over 30 years. He knew that Kwok was the chairman of a listed company and had always found him to be scrupulous, never having known him to have done anything dishonest or dishonourable nor having known his integrity or honesty to have been questioned.[6] They had implicitly trusted each other with unsecured and interest-free loans which had unfailingly been repaid.[7] 26. He knew that Kwok was in business in a very big way, with factories on the Mainland employing hundreds of thousands of people[8] and large-scale operations in various different countries, describing Kwok’s business as ten times the size of his own.[9] Kwok had invited him to meals where he had met some of Kwok’s customers who were very substantial and owned hundreds of stores.[10] He knew that Kwok had purchased a teenage garment brand at the cost of US$100 million.[11] He therefore did not consider the sum of HK$14 million a large amount, regarding it as “practically small change”, for someone as wealthy as he believed Kwok to be.[12] 27. On 29 July 2008, just before the money was transferred to the Mickles account, there had been a public announcement that trading in the shares of Tack Fat was to be suspended pending release of Tack Fat’s results for the year ending 31 March 2008. However, the Appellant was unaware of that announcement[13] and had believed throughout, until he heard about the non-payment of wages, that Kwok’s business was in good health, never having heard any suggestion that it was having any problems.[14] 28. The Appellant had never previously received a request from Kwok for money to be remitted into his account and, when questioned about this at the trial, the Appellant testified that he did not ask Kwok the reason and that he had agreed to allow his account to be used without giving it a second thought because he trusted Kwok.[15] 29. He stated that it never crossed his mind to ask Kwok why he did not deal with the money through his own companies. This piece of evidence was specifically disbelieved by the trial Judge (who had otherwise found the Appellant to be “a mostly ... honest and reliable witness”).[16] Interpreting Section 25 (1) 30. The directly relevant words that fall to be interpreted are: “having reasonable grounds to believe”. This is not a complicated formulation. The words are readily understandable. Most significantly, they occur in a provision which creates a criminal offence and which, with a maximum penalty of 14 years imprisonment, must be understood to be a serious offence. 31. In the criminal justice system of Hong Kong, the trial of such offences is conducted, pursuant to the exercise of the prosecutorial discretion to choose a court in which proceedings are commenced, either by a judge alone in the District Court, as happened in this case, or by a jury in the High Court. I will refer below to aspects of the case law, primarily in the Court of Appeal, which appears to me to adopt an unnecessarily elaborate analysis of the words that fall to be interpreted. In that regard, insufficient attention may have been given to the requirements of instructing a jury, particularly with respect to words that are not, in themselves, complex. 32. The context of the clause under consideration must be taken into account in the first instance and it is unnecessary to identify any verbal or grammatical ambiguity prior to so doing (See e.g. HKSAR v Cheung Kwun Yin (2009) 12 HKCFAR 568 at [12] – [13]; Vallejos v Commissioner of Registration (2013) 16 HKCFAR 45 at [75] – [77]). The immediate statutory context is of particular significance in this case. 33. The clause constitutes the mental element of a criminal offence in the alternative to actual knowledge. No doubt recklessness would be equated to knowledge. Reasonable grounds for belief are something less than either knowledge or recklessness. Nevertheless, the maximum penalty is the same for knowledge and for the “reasonable grounds” alternative. This is of critical significance when determining the nature and strength of the mental element that must be established by the prosecution beyond reasonable doubt. 34. The subsection under consideration also needs to be read in its complete statutory context, including the whole of the provisions of section 25, particularly subsection (2), and the provisions of section 25A, particularly subsections (1), (2) & (5). Those provisions emphasise the purpose of the legislative scheme to encourage reporting of the possible existence of the principal offence, based on either knowledge or suspicion. 35. The broader context includes the origins of these provisions of OSCO, in the, now, identical provisions in a scheme of longer duration, namely, the Drug Trafficking (Recovery of Proceeds) Ordinance (Cap. 405) (“DTROP”). The very subject matter of these original provisions, as now adopted with respect to organised and serious crime of any kind, highlights the significant public purposes to be served by these parallel regimes. 36. Two purposes stand out. First, by the identification of what has come to be called money-laundering, the legislative scheme deprives perpetrators of crime of the proceeds of their conduct. Secondly, the scheme ensures that, under pain of penalty, those who know, or even suspect that, relevantly, monies constitute such proceeds, will report that knowledge or suspicion to the authorities, to facilitate further investigation. 37. The Court was taken at some length to the legislative history. As is often the case, that proved of little value. The Respondent sought to put the legislative history to two principal uses. First, to emphasise the public purpose to ensure reporting. As I have noted, this appears on the face of the legislative provisions. Secondly, to highlight that the original proposal, to make the formulation “reasonable grounds to suspect”, was not pursued. The difference between “belief” and “suspicion” is well-known and, in any event, the distinction appears clearly in the text, by the use of “suspects” in section 25A. 38. The judgment from which this appeal is brought contains a detailed analysis of prior case law, primarily in the Court of Appeal. As the judgment notes, the principles consistently applied in that case law are derived from the judgment of Mayo JA in HKSAR v Shing Siu Ming [1999] 2 HKC 818. In the present case the trial judge applied that approach, as he was bound to do. 39. Subject only to the sequence in which questions are to be asked, a matter to which I will return below, the Court of Appeal adopted this approach, consistently with a long line of authority in that Court. The relevant passages from Shing Siu Ming are: 40. At page 825 H-I: “Knowledge if proved would simply resolve the matter. Difficulty, however, arises from the use of the words ‘having reasonable grounds to believe’. This phrase, we are satisfied, contains subjective and objective elements. In our view it requires proof that there were grounds that a common sense, right-thinking member of the community would consider were sufficient to lead a person to believe that the person being assisted was a drug trafficker or had benefited therefrom. That is the objective element. It must also be proved that those grounds were known to the defendant. That is the subjective element.” 41. And at page 829 A-B and H-I: “The prosecution is not called upon to prove actual belief. It would be sufficient to prove reasonable grounds for such a belief and that the defendant knew of those grounds … He (the judge) has similarly placed too high a burden on the prosecution as regards the objective element, i.e. the belief of the reasonable man. The jury did not have to be satisfied that a reasonable person would have held such a belief but only that such a person would be satisfied that there were grounds sufficient to sustain such a belief.” 42. As McWalters J, who delivered the principal judgment in the Court of Appeal, noted, this reasoning did not require that the two “elements” – “objective” and “subjective” be considered in the order in which they were stated at page 825 quoted. However, that had become the practice. The trial judge followed that practice in the present case. 43. The Court of Appeal concluded that the order should be reversed. The judgment of McWalters J said: “107. … [I]t seems to me that the risk that might be encountered by employing this order is that the first question, the objective question, might be answered by reference to what the prosecution has proven rather than by what the prosecution has proven was known by the defendant. By reversing the order this risk is eliminated.” “110. The first step in determining whether a defendant had reasonable grounds to believe is to identify all the facts known to the defendant that relate to the dealing with property that is the subject of the charge. These facts may, depending on the circumstances of each case, extend beyond those relating to just the dealing with the property and include facts known to the defendant about other persons or circumstances linked in some way to the dealing.” “112. The second step is to process these facts through the mind of the common sense, right-thinking member of the community and determine whether this person, possessed of these facts, objectively would consider them sufficient to lead a person to believe that the property in question constitutes the proceeds of an indictable offence. When this reasonable person considers these facts objectively it means he does so uninfluenced by the personal beliefs, perceptions or prejudices of the defendant.” (See also at [195] per Lunn JA and [218] per Stock VP). 44. There are four aspects of the approach derived from Shing Siu Ming which distract the decision maker -- whether a judge or jury -- from the terminology of the offence. 45. First, the test replaces the statutory word “grounds” with the word “facts”. The latter term is narrower. While all “facts” may be “grounds”, not all “grounds” are “facts”. Accordingly, the range of relevant circumstances is restricted. 46. Secondly, the introduction of a distinction between “objective” and “subjective” elements diverts attention away from the purpose of the statutory words to create the mental element of a criminal offence, as an alternative to a state of “knowledge”. 47. Thirdly, the personification of the “objective” element, in terms of a “reasonable person” or “right-thinking member of the community”, is a distinction that diverts attention away from the fact that it is the accused, rather than an abstract legal concept of a person, who must ‘have’ reasonable grounds for the requisite ‘belief’. 48. Fourthly, the language of “first step” and “second step”, in whatever order, directs the decision maker – whether judge or jury – as to how s/he or they must think. This is something which an appellate court should rarely do, if at all. Furthermore, these “steps” combine the two distractions referred to in the second and third points. 49. As I have said above, the statutory words “reasonable grounds” are perfectly understandable. There is no need for an abstract personification for purposes of their application. Similarly, these words can be applied directly and do not need further characterisation as “objective”. 50. Furthermore, by the use of the word “having”, the decision-maker’s attention is directed expressly, by the terms of the section, to the grounds available to the accused. There is no need to further characterise this element as “subjective”. Indeed, such language is apt to create confusion, particularly in a jury so instructed, as to why the mental element of the offence, created by the entire clause, is only in one part “subjective”. 51. The importance of maintaining the focus of inquiry upon the accused is emphasised in this Court’s reasoning in HKSAR v Yan Suiling (2012) 15 HKCFAR 146. Chan PJ, who delivered the judgment of the Court, accepted the trial judge’s opinion about the risks involved in the underground banking operation revealed in that case, namely, that no ordinary reasonable person who is law abiding would have taken part in such underground activities. (At [28]) However, his Lordship added at [29], referring to the trial judge: “But the real question he had to resolve in this case was whether it was true or might be true that the appellant was prepared to and did take these risks in engaging such services.” (Emphasis added) 52. As the Appellant submitted in this Court, the better and, I add, less complicated test was propounded in the Appeal Committee of this Court in Seng Yuet Fong v HKSAR [1999] 2 HKC 833at 836E-F: “To convict, the jury had to find that the accused had grounds for believing; and there was the additional requirement that the grounds must be reasonable: That is, that anyone looking at those grounds objectively would so believe.” 53. Unlike the Shing Siu Ming line of authority, this test: 1. Retains the statutory word “grounds”; 2. Avoids “subjective” terminology and any contrast with “objective” terminology; 3. Focuses attention on the accused, rather than on an abstract legal construct; 4. Avoids the language of “steps” and, like the section itself, propounds a compound test. 54. The unnecessary complications in the line of authority in the Court of Appeal have led to the somewhat convoluted submissions which the parties have been driven to make in the present appeal. Considerable attention was directed to the implications of the sequence in which the two elements needed to be addressed by the trial judge. Much ink was also spilt over whether there was any room in the relevant discourse for more than one “reasonable person” and, if so, what a number of “reasonable persons” would think. Such debates are not helpful and are too far removed from the language of the statute. 55. Although it is usually undesirable to substitute the words of a statute with an equivalent formulation, it is sometimes appropriate to do so, for purposes of clarification. For example, when an alternative formulation may assist a jury in its deliberations. On most such occasions, the Seng Yuet Fong formulation will be all that is required. 56. However, another formulation may sometimes assist. In s 25(1), the word “believe” is used in the sense of “know”. The two mental elements in the subsection should be understood as if they read: “knew or ought to have known”. 57. On the above analysis, there is a significant mens rea element in the second limb of the offence under s 25(1). This requirement disposes of the various scenarios put to this Court on behalf of the Appellant in which a morally blameless person could be convicted of an offence. Such submissions concerned Stock VP in the Court of Appeal. (See at [211] – [214]). On the above analysis, there is a strong element of moral blame. 58. Furthermore, on this basis, the “halfway house” defence does not arise. (See Hin Lin Yee v HKSAR (2010) 13 HKCFAR 142). It is, accordingly, unnecessary to consider the third ground on which leave to appeal was granted. The Applicable Standard 59. I find it convenient to deal first with the second ground on which leave to appeal was granted. The ground concerns the standard to be applied when evaluating the content of a “reasonable person’s belief”. The Appellant contends that the Court of Appeal applied an inappropriately low standard by adopting the word “could”, rather than “would”, when determining the causal link between the facts and the requisite belief. 60. This issue arises from the following passage in the judgment of the Court of Appeal: “130. …. As I have said the mens rea is not concerned with what the reasonable man might or would believe but rather with what he could believe.” 61. The reference to a prior statement, which commences this sentence, appears to me to be a reference to the analysis and application of Shing Siu Ming earlier in the judgment. Although Mayo JA in the earlier case had used the word “would” he had done so in a particular way. To repeat part of the relevant sentence from Page 825, fully set out in paragraph 40 above: “... there were grounds that a commonsense, right-thinking member of the community would consider were sufficient to lead a person to believe ...” (Emphasis added). It appears that it was because the reference to “would” was qualified by a reference to sufficiency, that the Court of Appeal in this case adopted the word “could”. 62. This occurred in the context of dealing with a submission by counsel appearing in the Court of Appeal for the Appellant, Ms Clare Montgomery QC, which referred to the judgment of the Court of Appeal in HKSAR v Lung Yun Ngan CACC 482/2010; unreported, 24 May 2011 at 52. In that passage to the Court had repeated the Shing Siu Ming formulation with the references to both “would” and “sufficient”. The Court of Appeal in the judgment under appeal had similarly adopted this formulation (at [112]). 63. When rejecting Ms Montgomery’s QC’s submission, the Court said: “125……this test focuses, and, with respect, rightly so, on the sufficiency of the grounds to form the requisite belief and not, as Ms Montgomery contends, on whether the grounds must inevitably lead to the formation of that belief. What the court there said was that the grounds had to be such that a commonsense, right-thinking member of the community ‘would consider sufficient’ and not ‘would lead a person to believe’. The two are quite separate issues and should not be confused.” 64. This reasoning is consistent with the Court’s earlier reference (at [70]) to the following statement from the judgment of the High Court of Australia in George v Rockett (1990) 170 CLR 104 at 112: “When a statute prescribes that there must be ‘reasonable grounds’ for a state of mind-including suspicion and belief-it requires the existence of facts which are sufficient to induce that state of mind in a reasonable person”. 65. McWalters J went on to say at [70]: “For an accurate and succinct expression of the legal position it is, in my view, difficult to improve upon this statement”. 66. Indeed, in this Court on appeal, the Respondent submitted that this formulation of the High Court of Australia should be adopted as the determinative test for the “reasonable grounds” alternative in section 25(1) of OSCO. 67. In my opinion, this submission suffers from a fundamental defect. George v Rockett involved a statutory test for the issue of a search warrant. Indeed, most of the case law on the concept of “reasonable grounds”, and cognate statutory formulations, is similarly concerned with the exercise of executive authority pursuant to statutes e.g. as grounds for arrest. Case law of this character, even if it deploys the same terminology, provides little, if any, guidance when the same words fall to be construed as an element of a criminal offence. 68. This differentiation is a manifestation of the principle of statutory interpretation which focuses on the significance of context, rather than adopting a “natural and ordinary meaning” of particular words. The formulation used to state the mental element of a criminal offence will not necessarily have the same meaning as the same formulation expressed as a description of the state of mind required for the exercise of an executive power. Case law of the latter character, where no issue of mens rea or proof beyond reasonable doubt arises, must be used with considerable circumspection in proceedings of the former character. 69. That is so in this case. The language of “sufficiency” – as found in Shing Siu Ming and George v Rockett – is inappropriate when determining the mental element of a criminal offence. In my opinion, the Court of Appeal erred in applying that language, admittedly based on a long line of earlier authority in that Court, to conclude that a test of “could believe” was applicable to section 25 (1), rather than a test of “would believe”. 70. In my opinion, the Court of Appeal has overlooked the significance of the formulation in the first judgment on this issue in this Court. I refer again to the judgment of the Appeal Committee of this Court, albeit a judgment of three judges on a leave application, in Seng Yuet Fong v HKSAR [1999] 2 HKC 833 at 836 E-F, where Litton PJ said: “To convict, the jury had to find that the accused had grounds for believing; and there was the additional requirement that the grounds must be reasonable: That is, that anyone looking at those grounds objectively would so believe.” (Emphasis added). 71. In Seng, Litton PJ relied on a passage from the judgment of Lord Woolf in the Privy Council case of Attorney General of Hong Kong v Lee Kwong Kut [1993] AC 951 at 964 D-E. It is noteworthy that this passage referred, at two places, to an actual rather than a probable state of affairs: “Once a person has knowledge or has reasonable grounds to believe that a relevant person carries on or has carried on drug trafficking….” (At D) (Emphasis added). “…[S]omeone whom you know or have reasonable grounds to believe is a person who carries on or has carried on…. drug trafficking”. (At F) (Emphasis added). 72. The application before the Appeal Committee in Seng was an appeal from the judgment of the Court of Appeal in the Shing Siu Ming case. The Appeal Committee had quoted that judgment -- with its reference to both “would” and “sufficient” -- without express disapproval. Nevertheless, after noting at page 837 F, that the Court of Appeal had not applied the analysis of the Privy Council, Litton PJ added at 837 H that “We see no real variance therefrom”. That was enough for the purpose of refusing leave. It was not however an express endorsement. It appears likely that, in a leave context, the possibility that the word “sufficient” constituted a significant qualification of the word “would” was not argued or readily apparent. 73. Finally, the test propounded by the Court of Appeal cannot, in my opinion, stand with the reasoning of this Court in HKSAR v Yan Suiling earlier referred to[17]. In the judgment of Chan PJ, he referred to the accused’s explanation that the payment, said to the constitute the “dealing”, contrary to s 25(1), was part of an underground banking operation. His Lordship further observed that, on the evidence: “…her explanation…at least might be true and that would have created a reasonable doubt” [44] (Emphasis added). Chan PJ went on to say of the trial judge: “It follows that he could not have concluded that there was an irresistible inference that she had reasonable ground to believe the cheque was the proceeds of an indictable offence” [44]. (Emphasis added) 74. The Court repeated the “irresistible inference” formulation and, significantly for the point presently under consideration, combined it with the words “must have had” in describing the link to “reasonable grounds to believe”. The passage is: “48. In this case, even if the appellant's evidence were properly rejected, what was left in the prosecution case was the mere fact that a large sum of money was deposited into the appellant's bank account by someone unknown to her and that she did not make any enquiry. The situation was certainly unusual but there could be a number of explanations and possible consequences of her inaction. The money might have come to her by mistake and she might be held answerable in a civil action for its return. She might even be liable for prosecution in respect of some other offence(s). However, we do not think that without more, an unexplained receipt points irresistibly to money laundering. We are not satisfied that on the facts as proved, there is sufficient evidence to draw the irresistible inference that she must have had reasonable grounds to believe the money represented the proceeds of crime.” (Emphasis added) 75. The terminology of “irresistible inference” is often deployed in the context of circumstantial evidence. (See e.g. Lee Chun Tat v HKSAR unrep. FAMC 52/2006, 14 December 2006 at [7]). The idea that an inference must be “irresistible” is an application of the principle that crime must be proven beyond reasonable doubt. This reinforces my above conclusion that cases on statutes which use the phrase “reasonable grounds to believe” as a statutory precondition for the exercise of executive power are inapplicable in a criminal offence context. 76. In my opinion, the use of the word “would” or equivalent terminology in this line of authority is inconsistent with the adoption by the Court of Appeal in this case of language of possibility, such as “could”. 77. These authorities reinforce the conclusion I have reached, based on the analysis of the words of the statute in their context and of the purpose of the legislative scheme. By the imposition of the same penalty, the mental element of the “reasonable grounds” alternative is regarded as being at the same level of moral obloquy as actual knowledge. A test that propounds a relationship between the existence of “grounds” and a state of ‘belief’ in terms of possibility does not do that. Only a test which states that those “grounds” would lead to the ‘belief’ does so. 78. It is of significance in this case that the trial judge, although purporting to apply Shing Siu Ming, did not adopt a ‘could believe’ formulation. He twice referred to “would believe”: “86. The prosecution has firstly to prove all the circumstances that they consider would lead a ‘common sense, right-thinking member of the community’ to have reasonable grounds to believe that the property was the proceeds of an indictable offence. If the prosecution fails to prove the circumstances or that the court finds that the circumstances would not so lead, then the defendant must be acquitted. 87. If the circumstances are proved and that the court finds that they would so lead, it is then for the prosecution to prove that those circumstances were known to the defendant at the time of the offence. The prosecution however do not have to prove that the defendant did in fact hold such a belief, all that is required is that the defendant knew of the circumstances.” (Emphasis added) 79. This Court would not interfere with the order of the Court of Appeal to dismiss the appeal, if it adopted an inappropriate test, if the trial judge committed no legal error. I will return to this issue below. The perception and evaluation issue 80. With respect to the first ground on which leave to appeal was granted, the submissions in this Court focused entirely on the second sentence: Whether a Defendant’s “perception and evaluation” of relevant facts “constituting or contributing to reasonable grounds for” the requisite belief can be taken into account. In the Court of Appeal, McWalters J concluded that, if it was permissible to take into account facts “as processed and assessed by the defendant”, it would turn an objective test into a subjective one. (At [99]). 81. His Lordship’s extended analysis on this issue was: “102. Identifying the relevant facts, however, is quite distinct from processing them. The applicant and the reasonable person may process them quite differently because the reasonable person though possessed of all of the relevant facts, is not, in processing them, influenced by the personal beliefs, perceptions or prejudices of the applicant. 103. Here, because the applicant is also a personal friend of Kwok and regularly socialized with him the applicant would, quite naturally, not want to think ill of him and not want to suspect him of being involved in anything dishonest or improper. A desire not to think ill of a friend or a wish to view his conduct through rose-coloured glasses may operate as blinkers on the applicant’s assessment of the conduct or facts known to him. But these blinkers are not worn by the reasonable man who, unconstrained by emotion, stands back from the facts and considers them dispassionately. This does not involve him in removing the facts from their context and analysing them in a meaningless isolation which inevitably and unfairly skews his assessment of them. It simply means that the personal beliefs, perceptions or prejudices of the applicant are removed from the assessment process. 104. Thus, in respect of the HK$14 million the applicant said that, in his perception of Kwok, such an amount would have been of no consequence to him. As I have already said the facts that Kwok was believed to be a successful and honourable businessman of many years standing who would be accustomed to dealing in large amounts of money would be relevant facts to take into account. But the applicant’s own perception of this amount or his belief of its significance to Kwok would not. Those subjective elements are removed so that they do not adulterate what has to be an objective assessment.” 82. This passage reflects the difficulties I have already mentioned that arise when words of the statute have labels such as “objective” attached to them for purposes of categorisation. The principal difficulty is that it is the label, rather than the words of the statute, that are then applied. In the present case, this effect is magnified by the artificiality of personification, which directs attention away from the person whose conduct needs to be assessed, namely, the accused. As I have emphasised above, it is the accused who must be shown to “have had reasonable grounds to believe”. 83. I can see no reason why, when applying the words of the statutory formulation, what McWalters J called: “the personal beliefs, perceptions and prejudices” of the accused should be entirely omitted from consideration. That does not mean that any such “perception or evaluation”, to use the words of the ground of appeal, is entitled to weight, let alone determinative weight. 84. The error, I believe, is in the reasoning process by which the statutory word “grounds” has been replaced by the word “facts”. As I have said above, “facts” is a narrower concept than “grounds”. It may be that a “belief, perception or prejudice” is not a “fact”. However, such matters fit quite readily within the concept of a “ground”, which a particular person can be said to have ‘had’. 85. When assessing the whole of the evidence, the judge or jury can give such weight to an accused’s belief, perception or prejudice as s/he believes is warranted. No doubt, in many cases, that decision maker will entirely discount such evidence of the accused. Nevertheless, they are “grounds” which stand or fall by the test of reasonableness. 86. The Respondent contended that taking into account the “perceptions and evaluations” of an accused was contrary to authorities which rejected, in various contexts, the importation into the standard of a reasonable person, the personal characteristics of a particular defendant. (Relying on R v B [2013] 1 Cr App R 36 at [28], Jersey v Holley [2005] AC 580 at [12], [22]; HKSAR v Siu Kam Lin CACC 186/2009, 15 Dec 2012; Cheung Kam Sing v HKSAR CACC 380/2010, 9 Dec 2011; Oblach v Regina [2005] NSWCCA 440; (2005) 195 FLR 212 at[57] – [59]). 87. However, statements made in the context of different statutory offences, even where “reasonable belief” is an element of the offence, cannot be transposed to another context. For example, some of these cases are concerned with “characteristics” rather than “grounds”. Others did not use a cognate formulation, as an alternative to “knowingly” with an identical maximum penalty. I do not find these authorities helpful. The two Court of Appeal cases in the list, however, involve s 25(1) of OSCO. They stand in the line of authority in that Court which is under review in this appeal. 88. For present purposes, Lord Sumption JSC put it precisely in Hayes v Willoughby [2013] UKSC 17; (2013) WLR 935 at [14]: “Reasonableness is an external objective standard applied to the outcome of a person’s thoughts or intentions”. 89. In my opinion, contrary to the conclusion of the Court of Appeal, this approach does not convert an objective test into a subjective test. That Court was in error in its analysis in this respect. In the District Court, the trial judge did not expressly exclude matters of “belief, perception and prejudice”. It is not entirely clear how he dealt with such matters. 90. When applying the reasoning in Shing Siu Ming, the judge pointed out that the “subjective belief” of the Appellant, about whether the money was the proceeds of an indictable offence, was “not the issue”. ([90]). He went on to explicitly reject the Appellant’s evidence that “he did not give any thought” to whether it was. His Honour said: “it is impossible for anyone to not give any thought to the matter given the huge amount of money. Not even for someone as well off as the defendant.” ([91]). 91. He further held: “[92].... Kwok never told the defendant how the money came about and why Kwok needed the defendant to receive it for him, I find that it is impossible that the defendant did not give any thought as to the propriety of the money when he was told the amount. [93] Even if we suppose the defendant was not lying about it, for the same reasons, I find that the defendant was deliberately turning a blind eye as to the question of propriety latest when he was informed that the amount received from Kwok’s friends was $14 million.” 92. His Honour went on to quote from the judgement in HKSAR v Wan Yet Kwai CACC 372/2008, a case concerned with an accused who turned a blind eye to a “reasonably held belief” that certain property did represent the proceeds of an indictable offence. ([94]). His Honour concluded with the observation: “[95] Therefore, whether the defendant was lying or turning a blind eye, the question is still whether a commonsense, right-thinking member of the community with the facts available would have reasonable grounds to believe that the money represented proceeds of an indictable offence.” 93. It appears to me that this analysis differs from that propounded in the Court of Appeal. The trial judge did take into account mental processes of the Appellant, but rejected his evidence in relevant respects and drew contrary inferences. 94. Nevertheless, when making his ultimate findings, in accordance with the test in Shing Siu Ming, his Honour first asked the question, identified as the first step: “Would a commonsense right-thinking member of the community have reasonable grounds to believe?” With respect to the list of relevant facts, set out in his Reasons for Verdict, the trial judge drew the conclusions that a “reasonable person” would have been “put on immediate alert”, or words to similar effect, such as “would be really worried” or “definitely be very doubtful”. ([101]-[108]). 95. It is not clear to me what role the findings of fact about the mental state of the Appellant played in the ultimate conclusion. His Honour dealt with the Shing Siu Ming second step in short form. He simply stated that the “facts” set out in the first step were known to the accused. 96. In this Court, the Appellant’s submissions focused on the effects of following the two stage test in the sequence set down in Shing Siu Ming. The trial judge himself said: “there is a subtle but real difference between the two approaches”. ([100]). The Court of Appeal expressed a preference for reversing the two stages, but concluded that there was no prejudice caused by proceeding in the established sequence. ([105], [218]). 97. Lord MacDonald QC, who appeared in this Court for the Appellant, submitted that a failure to consider relevant circumstances in the mind of the Appellant was inherent in the adoption by the trial judge of the Shing Siu Ming sequence of steps[18]. By first asking whether there were facts which would give rise to the requisite belief, from the perspective of a reasonable person, the scope of the matters considered was necessarily narrower than they would be if the questions were asked in the reverse order. Relevantly, they excluded from the scope of consideration any grounds in the “perception and evaluation” of the Appellant. He submitted that the wrong order must make a difference to the outcome, because it directly impacts, in the sense of narrowing, the circumstances to which the test of reasonableness is applied. 98. Lord MacDonald QC referred to the emphasis placed by the Appellant, before the trial judge, on the fact that the Appellant trusted Kwok. Although the facts which form the foundation of that trust were mentioned in the Reasons for Verdict, the existence of trust was not. This is a specific example, in this case, of the error that can arise if, as Lord MacDonald put it[19]: “the accused person only comes into it once the objective man has already made up his mind that there are reasonable grounds to believe”. 99. He submitted that the absence of any reference, to the existence of trust in the trial judge’s reasons, contrasts with his finding on the issue of whether the Appellant ‘turned a blind eye’ to the nature of the transaction. As I understood the submission, that finding could not be made without dealing with the Appellant's case of trust. 100. Furthermore, he submitted that the trigger for the findings that the appellant was either lying, when he said that he gave the matter no thought, or turned a blind eye to the matter, was the trial judge’s assertion that the amount of money was such that anyone would have been put on notice. This failed to take into account the Appellant’s evaluation of Kwok as an extremely wealthy and honest man. 101. I agree that the Shing Siu Ming sequence of the two step test can result in impermissibly restricting the range of grounds which an accused ‘had’, to which the test of ‘reasonableness’ must be applied. I am also of the view that that occurred in this case at trial. 102. It is clear that his Honour, when rejecting the Appellant’s evidence gave significant weight to the amount of $14 million, which he described as “huge…even for someone as well off as the defendant”. This was the trigger for the rejection of his evidence that he had not turned his mind to the provenance of the funds (at [91]) and, it appears to me, his alternative finding that the Appellant turned a blind eye to that provenance (at [93]). The focus on the Appellant’s wealth, rather than Kwok’s, was misplaced. 103. Subsequently, at [102] his Honour did refer to the Appellant’s evidence that this amount would be small “in the eyes of Kwok”. However, he immediately dismissed this evidence on the basis that “that ... is not the case with the reasonable person”. This could be considered a finding that, although this was a “ground” in the mind of the Appellant, it was not a “reasonable ground”. Even if so, the reasoning takes the Appellant’s evaluation of the amount in the context of Kwok’s wealth alone. It is disconnected from the rest of the lengthy personal and business relationship, to which there is no reference in this section of the Reasons for Verdict. 104. At paragraphs [101] to [107] of the Reasons for Verdict, the trial judge sets out, as the “first step”, the facts which a “commonsense right-thinking member of the community” would have reasonable grounds to believe. This is the list of facts which support the conclusion at paragraph [108] that such a person would have reasonable grounds to believe that the money in question represented proceeds of an indictable offence. The only matters about the Appellant’s attitude to Kwok listed are the reference at [102], to which I have referred in the immediately preceding paragraph, and perhaps references to the appellant’s knowledge that Kwok was “a very successful businessman” (at [103]) and controlled significant assets (at [104]). These were the paragraphs replied upon by the DPP in his submissions in support of the proposition that the trial judge did not err in this respect[20]. However, there is no reference to other “perceptions and evaluations” of the Appellant in this part of the Reasons. 105. Some of the Appellant’s evidence of this character was, of course, rejected, as I have indicated. However, as Lord MacDonald submitted in this Court, there was no reference to the Appellant’s case based on the trust that existed between himself and Kwok. 106. Furthermore, such references as exist in the relevant part of the Reasons, are not put in the broader context, which includes matters that are “facts” on any view. Although earlier in his judgement, the trial judge sets out the close personal and business relationship between the two men, extending as it did over decades, none of these factors are contained in the list of facts available for consideration by the “commonsense, right-thinking member of the community” at paras [101] to [107]. Accordingly, even the matters referred to in paragraphs [102], [103] and [104] are out of context. More importantly, none of these other aspects of the relationship are listed as facts which the “reasonable person” took into account. This error is probably a result of asking the Shing Siu Ming questions in the wrong order. 107. In my opinion the Appellant’s contention that the list of grounds considered by the trial judge was too narrow should be upheld. A new trial? 108. The appeal should be allowed and the conviction set aside. 109. In that event counsel for the Appellant submitted that there are a number of factors to be taken into account as to whether a new trial should be awarded. (Relying on James Henry Ting v HKSAR (2007) 10 HKCFAR 632). He placed reliance on the fact that this was a long and expensive and stressful process for the Appellant. He noted that the events the subject of the charge occurred in 2008 and a second trial would be held more than six years later. Further, the Appellant has already served 4.5 months in custody. Finally, the Appellant is now 69 years old. 110. It was also submitted that he had suffered because his right to travel was restricted, in a context where manufacturing facilities were located abroad. There were other adverse effects on his business which, the Respondent submits, are not the subject of evidence. Nevertheless an inference that the effects of conviction and a new trial on his business activities have been, and are substantial, can readily be drawn. 111. Counsel also submitted that the prosecution case is at the low end of the scale. However, the list of factors upon which the Respondent relied in this Court could be sufficient to support the requisite inference, particularly as filled out by the trial judge in his list of objective factors to which I referred. 112. Indeed, the fact that the Appellant was offered no explanation, and made no enquiry, as to why Kwok could not use one of his own accounts could form the basis of an “irresistible inference” that something untoward was going on, which required Kwok to hide the flow of funds. However, it is not so apparent that what was untoward involved proceeds of an indictable offence. 113. The prosecution case is not sufficiently strong, when taken into account with the other circumstances referred to at [109] – [110] above, to satisfy me that the interests of justice require that there be a new trial. I would not so order. Orders 114. 1. Appeal allowed. 2. Conviction quashed. 3. We make an order nisi that the appellant be paid the costs of and occasioned by the appeal to the Court of Appeal and to this Court with liberty to the parties, if so advised, to lodge with the Registrar written submissions as to costs within 14 days from the date of this judgment and direct that in default of such submissions, that the order nisi do stand as an order absolute without further direction. Lord Ken MacDonald QC, Mr Andrew Bruce SC and Mr B.K. Ho, instructed by King & Co, for the Appellant Mr Keith Yeung SC, DPP, of the Department of Justice, Mr Simon Young, on fiat and Ms Maggie Yang SADPP, of the Department, for the Respondent [1] Reasons for Verdict (“RFV”), DCCC 895/2011 (16 January 2012). [2] Stock VP, Lunn JA and McWalters J, CACC 34/2012 (31 May 2013). [3] His appeal against sentence having been abandoned. [4] Stock VP, Lunn JA and McWalters J, CACC 34/2012 (31 May 2013), at an expedited hearing held on the day that judgment was handed down. [5] Annex, Respondent’s Skeleton Submissions, 13 October 2014. [6] RFV§§22, 54 and 65. [7] RFV§§47 and 48. [8] RFV§104. [9] RFV§40. [10] RFV§41. [11] RFV§53. [12] RFV§102. [13] RFV§52. [14] RFV§§65-66. [15] RFV §§55-56; Stock VP, Lunn JA, McWalters J, CACC34/2012 (31 May 2013) §§15, 18, 19 and 20. [16] The Judge found that “the defendant is not telling the truth when he said that he did not give any thought as to whether the money that Kwok had asked him to receive might have represented proceeds from an indictable offence. The reason I so find is that it is impossible for anyone to not give any thought to the matter given the huge amount of money. Not even for someone as well off as the defendant.” RFV§91. [17] Para 51 above [18] Transcript, 15 October 2014, at pages 5-11, 28, 35-38 and 94. [19] Transcript, 15 October 2014 at page 36. [20] Transcript, 15 October 2014 at pages 89-91. 
Mr Justice Fok PJ (giving the Judgment of the Court) : Background 1. This uncontested appeal arises as a result of the Court’s decision, handed down on 5 June 2014, in HKSAR v Li Kwok Cheung George & Ors[1]. 2. That case involved four of six defendants tried for various offences. The four defendants in question were: Wayland Tsang Wai Lun (Tsang); his wife Nancy Kwok Wai Man (Kwok); Charles Cheng Kai Ming (Cheng); and George Li Kwok Cheung (Li). 3. Tsang and Kwok were directors and shareholders of a public company who were convicted of having conspired with an immunised prosecution witness, named Au Yeung, to defraud the Stock Exchange of Hong Kong and the public company’s shareholders by means of what was referred to as “the Charge 3 conspiracy”. 4. The facts giving rise to the Charge 3 conspiracy were summarised in the Court’s judgment in HKSAR v Li Kwok Cheung George & Ors as follows: “6. In mid-March 2002, the price of Grand Field’s shares had fallen from $0.68 to $0.12. In an effort to bolster that price, Tsang and Kwok hatched a scheme to give a false impression of profitable activity involving a Mainland joint venture to construct a natural gas pipeline near Chongqing. A company called Sino Richest Limited (‘Sino Richest’) was formed with shareholders, including a company called Logistic China Enterprises Limited (‘Logistic China’), who were ostensibly unconnected with Grand Field. On 25 May 2002, Sino Richest purported to sign a joint venture agreement with the Mainland party and shortly thereafter, Grand Field purported to acquire 75% of Sino Richest in consideration of Grand Field issuing to the shareholders of Sino Richest 315 million new Grand Field shares. On 4 June 2002, a bullish public announcement was made about this joint venture which was said to be subject to government licences and approvals. 7. The public announcement was false, it never having been the intention of Tsang and Kwok to proceed with the joint venture. However, Grand Field came under increasing pressure from the Stock Exchange to provide further details relating to the joint venture and the needed government approvals. As Tsang and Kwok were unable to comply, hoping to put an end to such pressure, they sought to pretend with Au Yeung’s help that Grand Field had sold its interest in the joint venture (held via Sino Richest) back to Logistic China. The purported sale was announced on 12 August 2003. This pretence led to their conviction of the Charge 3 conspiracy. Charge 3 alleged that Tsang, Kwok and Au Yeung had conspired to defraud the Stock Exchange and the shareholders of Grand Field by dishonestly ‘... concealing that there had been no genuine acquisition of a business to develop and construct a gas pipeline in Chongqing, the Mainland; and falsely representing that there was a genuine disposal of the aforesaid business in Chongqing’.” 5. Cheng and Li were directors of another listed company which was the holding company of a group providing financial and securities broking services. They, together with Tsang, Kwok, the appellant in this appeal[2] and one other, were alleged to have conspired to implement a scheme involving circular payments designed to conceal the Charge 3 conspiracy and charged with having conspired to commit a money laundering offence contrary to sections 25(1) and (3) of the Organized and Serious Crimes Ordinance[3]. This offence was the subject of Charge 4 of the indictment at trial. 6. The facts giving rise to Charge 4, as summarised in HKSAR v Li Kwok Cheung George & Ors, were as follows: “8. To give credence to the re-acquisition by Logistic China, the four appellants (and others) set up a scheme with a view to providing evidence that Logistic China had duly paid Grand Field (through its subsidiary called Ka Fong Industrial Limited (‘Ka Fong’)) $32 million to re-acquire the Sino Richest shares. This was where Cheng and Li came into the picture. On 31 July 2003, advancing funds which originated from Upbest, they caused a series of payments to be made (with Cheng signing a series of Upbest cheques) which resulted in a circular flow of funds bringing the money back to Upbest on the same day, after having passed through a web of local and offshore companies. Payments to and by Logistic China, Ka Fong and Tsang formed part of these circular payments. Logistic China received $32 million from a company upstream in the circular flow and then, purporting to pay for the re-acquired Sino Richest shares, made a payment of $32 million to Ka Fong (Grand Field’s subsidiary). Ka Fong (by a cheque signed by Kwok) then paid the $32 million to Tsang who then passed the money onwards so that it was eventually returned, after certain disguising complications, to Upbest. On the following day, 1 August 2003, Grand Field faxed to the Stock Exchange a copy of the payment-in slip showing that Logistic China had paid $32 million to Ka Fong on the previous day. 9. These circular payments were the subject of Charge 4 which alleged a ‘conspiracy to deal with property known or believed to represent the proceeds of an indictable offence contrary to ... sections 25(1) and (3) of [OSCO]’, particularised as follows: ‘[Tsang, Kwok, Cheng, Li (and others) [viz. the appellant and Au Yeung]] ... knowing or having reasonable grounds to believe that property, namely $32 million ... in whole or in part directly or indirectly represented proceeds of an indictable offence, conspired together with [Au Yeung] to deal with the said property.’” The appellant’s conviction and appeal to the Court of Appeal 7. The sole charge against the appellant was Charge 4. After trial before HH Judge A. Wong[4], the appellant, together with Tsang, Kwok, Cheng and Li were convicted of this charge on 5 March 2010. The appellant was sentenced to 6 months’ imprisonment for the offence and has since served that sentence. 8. On appeal to the Court of Appeal, the appeals of the appellant and his co-defendants in respect of Charge 4 were dismissed on 28 November 2012.[5] HKSAR v Li Kwok Cheung George & Ors 9. Following the dismissal of their appeals by the Court of Appeal, Tsang, Kwok, Cheng and Li applied for leave to appeal to this Court against their convictions under Charge 4. On 15 August 2013, the Appeal Committee granted them leave to appeal against their Charge 4 convictions and certified the following question of law, namely: “Whether the expression ‘proceeds of an indictable offence’ in s 25(1) of the Organized and Serious Crimes Ordinance, Cap 455, is confined to money gained from the commission of an indictable offence or, instead, extends to money used in the furtherance of such an offence.” 10. The Court heard the appeals of Tsang, Kwok, Cheng and Li[6] on 19 and 20 May 2014. The appeal raised the question of whether, by virtue of the statutory definition of “a person’s proceeds of an offence” in section 2(6)(a) of OSCO, the offence of money laundering in section 25(1) of OSCO could be committed where a person deals with funds that are known not to derive from any offence but which are intended to be used as part of a fraudulent conspiracy. 11. In its judgment dated 5 June 2014, the Court concluded[7] that: “… the Courts below were wrong to construe the expression ‘proceeds of an indictable offence’ in s 25(1) of OSCO as extending to cover money used as an instrument furthering the Charge 3 conspiracy where there was no evidence that the appellants knew or had reasonable grounds to believe that such money represented the proceeds of an indictable offence. It was on the contrary clear that all concerned knew that the payments and receipts of the sum of $32 million relied upon by the prosecution involved dealing with funds provided by Upbest, there being no evidence that such funds were other than legitimate. It was erroneous to construe section 2(6)(a) as widening the concept of ‘proceeds of an indictable offence’ to cover such payments.” 12. The Court accordingly held[8] that the Charge 4 conspiracy was not established and the appeals of Tsang, Kwok, Li and Cheng in respect of that charge were allowed and their convictions thereunder quashed. The appellant’s application for leave to appeal out of time 13. By notice dated 24 June 2014, the appellant applied to the Appeal Committee for leave to appeal against his conviction in respect of the Charge 4 conspiracy and, due to its lateness, for an extension of time in which to make such application. In respect of the latter application, the appellant explained that he did not apply for leave to appeal within the prescribed time after the dismissal of his appeal against conviction by the Court of Appeal due to his financial difficulties and personal circumstances. In short, he had exhausted his savings in defending himself at trial; he had borrowed funds from his wife to pursue his appeal to the Court of Appeal but had since separated from her; he therefore could not rely on her for further funds and had to repay what she had already lent him. 14. The appellant’s application was not opposed by the respondent and, by a consent order dated 11 September 2014, the Appeal Committee granted him leave to appeal and an extension of time in which to do so. 15. Although now water under the bridge, it is right to point out that the grant of such an extension was, of course, a rare and exceptional course since there is a well-recognised practical necessity for finality in the criminal process and an extension of time to appeal against a conviction out of time is not normally granted only on the ground that an authoritative judgment subsequent to the conviction has held the previous understanding of the law to be incorrect.[9] 16. In the present instance, the extension was justified on the basis of the very exceptional circumstances of the appellant’s case. The conviction sought to be appealed was not, as is more usually the case in this situation, unrelated to the subsequent authoritative judgment holding the previous understanding of the law as incorrect. On the contrary, the appellant’s conviction was in the same case, was for the same offence as Tsang, Kwok, Li and Cheng and was based on the same facts and evidence. The question of law determined by the Court in HKSAR v Li Kwok Cheung George was directly applicable to the appellant’s proposed appeal and, if applied, would fully determine that appeal. He had provided a reasonable explanation for his failure to apply for leave to appeal within the prescribed time limit. For the appellant otherwise to remain convicted of the Charge 4 offence, alone amongst the various parties charged with that offence and on an erroneous view of the law, would plainly subject him to substantial and grave injustice. The appeal 17. The Appeal Committee having granted the requisite extension of time to appeal together with leave to appeal, it now falls to consider the merits of the appeal itself. 18. The respondent having indicated that it would consent to the appeal being allowed, the Registrar directed the filing of a Joint Printed Case by the parties in accordance with the practice and procedure for uncontested appeals.[10] 19. On 11 November 2014, the parties duly filed such a case jointly submitting that the appeal should be allowed and proposing an agreed order as to costs. In support of the appeal, it was jointly submitted that, given the decision of the Court in HKSAR v Li Kwok Cheung George, substantial and grave injustice has been done to the appellant such that his appeal should be allowed. 20. In the light of the Court’s judgment in HKSAR v Li Kwok Cheung George, the appellant’s conviction in respect of Charge 4 is unsustainable as a matter of law for precisely the same reasons that led to the appeals of Tsang, Kwok, Li and Cheng being allowed in that case. There was no evidence that the appellant knew or had reasonable grounds to believe that the money used to further the Charge 3 conspiracy represented the proceeds of an indictable offence. The $32 million constituting the subject of the money laundering offence charged were funds provided by Upbest and there was no evidence that these were other than legitimate. They were therefore not “proceeds of an indictable offence” within section 25(1) of OSCO and it follows that Charge 4 could not be established against any of the parties to the alleged conspiracy. 21. In the circumstances, the appellant’s conviction of the Charge 4 offence, alone amongst the various parties charged with that offence and on an erroneous view of the law, involves a substantial and grave injustice. Disposition 22. For the reasons set out above, the appellant’s appeal is allowed and his conviction under Charge 4 quashed. 23. In accordance with the parties’ agreement, we make the following costs orders by consent: (1) The appellant be entitled to payment of his costs of trial in DCCC 24/2008, to be taxed if not agreed. (2) The appellant be entitled to payment of one-half of his costs in CACC 96/2010, with a certificate for two counsel, to be taxed if not agreed. (3) The appellant be entitled to payment of his costs of applying to the Appeal Committee for leave to appeal in FAMC 46/2014, to be taxed if not agreed. (4) The appellant be entitled to payment of his costs of the substantive appeal in FACC 9/2014, to be taxed if not agreed. Joint Written Submissions by: Mr Joseph Lee and Ms Emily Yu, instructed by Wat & Co., for the Appellant Mr Anthony Chau SPP, of the Department of Justice, for the Respondent [1] Reported in (2014) 17 HKCFAR 319 [2] Wong Wai Kwong, David [3] (Cap.455) (“OSCO”) [4] In DCCC No 24 of 2008, where the appellant was the 6th defendant [5] CACC 96/2010, Judgment handed down on 28 November 2012 [6] In FACC 4, 5 & 6 of 2013 [7] At §87 [8] At §88 [9] See HKSAR v Hung Chan Wa & Anor (2006) 9 HKCFAR 614 at §§24-25 [10] See Mok Kin Kau v HKSAR (2008) 11 HKCFAR 1 and HKSAR v Shum Wan Foon (2014) 17 HKCFAR 303 
Chief Justice Ma: 1. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Ribeiro PJ: 2. The appellant brought proceedings against the Director of Immigration for damages for false imprisonment claiming that he had been unlawfully detained purportedly under section 32 of the Immigration Ordinance (“IO”).[1] His claim was dismissed by HH Judge Leung in the District Court[2] and his appeal against that judgment was dismissed by the Court of Appeal.[3] Leave to appeal to this Court was given by the Appeal Committee[4] on the basis that a point of law of the requisite importance arises on the appeal, namely: “To what extent are the detention powers contained in section 32(2A) of the IO subject to an applicable principle of law barring arbitrary or unlawful detention?” A. The factual background 3. The appellant may be described as a serial over-stayer. He is a Pakistani national and first came to Hong Kong in 1992 using a passport bearing the name “Ghulam Rubbani” and stating his date of birth as 15 April 1971. He was allowed to stay for three months but overstayed for about 10 months. He was convicted of breaching a condition of stay, fined $1,000 and repatriated to Pakistan in August 1993. 4. Using another passport, this time in the name of “Mian Ghulam Rabani”, giving 15 April 1970 as his date of birth, the appellant returned to Hong Kong in 1994. He was again permitted to stay for 3 months and overstayed for some 5 months. He was again convicted of breaching a condition of stay, fined $1,200 and repatriated to Pakistan in July 1995. 5. He returned in May 1999, this time using a passport in the name of “Ghulam Rabbani” with a 1967 date of birth. After overstaying for about 4 months, he was convicted of breaching his condition of stay and of making a false representation to an officer (during his previous visit regarding his date of birth). He received a sentence of 3 months’ imprisonment suspended for two years and was sent back to Pakistan in October 1999. 6. The appellant last entered Hong Kong on 24 September 2000 (this time using a passport bearing the name “Ghulam Rbani”, with a 1971 date of birth). He was permitted to remain as a visitor until 15 October 2000 but overstayed for some 4½ years before he was arrested by the police on 1 April 2005 for a gambling offence. He was convicted of that offence as well as of breaching his condition of stay. The suspended sentence was activated and he was sentenced to a total of 7 months’ imprisonment. 7. After serving just under five months of his sentence, he was discharged on 23 August 2005. The Director immediately placed him under administrative detention pursuant to IO section 32(2A)(a) and subsequently issued a removal order against him on 10 September 2005. That order was not served because the appellant had meanwhile lodged a claim under the Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment (“CAT”) against refoulement which came to the attention of the Immigration Department’s Removal Sub-Division on 12 September 2005. 8. The removal order was revoked on 15 September 2005 and the appellant was eventually released on recognizance on 7 October 2005, approximately six weeks since he was first placed under administrative detention and three weeks since revocation of the removal order. The appellant’s claim for damages for false imprisonment was brought on 11 February 2010, some 4½ years after his release. I shall return[5] to consider in greater detail the circumstances of his detention. B. The appellant’s claim 9. As Lord Bridge of Harwich stated in R v Deputy Governor of Parkhurst Prison, Ex p Hague: “The tort of false imprisonment has two ingredients: the fact of imprisonment and the absence of lawful authority to justify it.”[6] Citing Atkin LJ for the proposition that “any restraint within defined bounds which is a restraint in fact may be an imprisonment”,[7] Lord Bridge continued: “Thus if A imposes on B a restraint within defined bounds and is sued by B for false imprisonment, the action will succeed or fail according to whether or not A can justify the restraint imposed on B as lawful.”[8] 10. There is no dispute that the period of the appellant’s confinement under executive detention constituted a period of imprisonment. The question is therefore whether the Director (who bears the burden[9]) can justify such detention as lawful. 11. Mr Philip Dykes SC[10] advances five arguments for contending that the Director is unable to provide such justification. Two involve reliance on constitutional guarantees and the other three bear on the scope and proper exercise of the statutory power of detention as a matter of public law forming part of the common law. 12. In making his constitutional arguments, Mr Dykes submits that the detention was unlawful in that it violated Article 5(1) of the Bill of Rights (“BOR Art 5(1)”) and/or Article 28 of the Basic Law (“BL Art 28”) made applicable to the appellant by Article 41 of the Basic Law (“BL Art 41”). Those Articles are set out below.[11] 13. The powers of detention relied on by the Director are contained in IO section 32,[12] being powers to detain pending a decision on whether to make a removal order (section 32(2A)); and pending removal where a removal order has been made (section 32(3A)). 14. In his public law argument, Mr Dykes submits that the Director’s purported exercise of the powers conferred by those sections in the present case was flawed and unlawful for three inter-related reasons, namely: (a) that applying the Hardial Singh principles,[13] the Director[14] could not lawfully continue to detain the appellant after it had become clear that his intended removal from Hong Kong could not be achieved within the time limits for detention laid down in section 32(2A) or otherwise within a reasonable time (“the Hardial Singh ground”); (b) that once the removal order was revoked on 15 September 2005, there was no legal basis for the continued exercise of the power (“the no legal basis ground”); and (c) that at the time of exercising the power, the Director had not published any statement of policy identifying criteria justifying detention under section 32 (“the lack of a policy ground”). 15. The first two of the abovementioned grounds involve the complaint that the duration of the appellant’s detention was excessive and cannot fully be justified, while the third ground involves the contention that the Director’s lack of a published policy tainted the entire period of detention. 16. It is Mr Dykes’s argument that on one or more of the aforesaid grounds, the detention was rendered unlawful, not merely as a matter of public law, but so as to constitute the tort of false imprisonment entitling the appellant to damages, this being a proposition which will also require examination.[15] 17. As Lord Brown of Eaton-Under-Heywood JSC pointed out,[16] the Hardial Singh principles, as applied in a succession of later cases, may be regarded as more favourable to detainees than the Strasbourg jurisprudence[17] requires. In like vein, Mr Dykes accepts that if his argument based on public law principles succeeds, the constitutional grounds would add nothing to the result. The focus of the argument has therefore been on the common law and public law principles and I shall deal with them first, before considering the constitutional arguments raised. C. The right at common law 18. The right to personal freedom and the concomitant right to be protected from arbitrary arrest or detention form part of the bedrock of the common law. Referring to an argument advanced by one of the parties in A andOthers v Secretary of State for the Home Department,[18]Lord Bingham of Cornhill commented: “In urging the fundamental importance of the right to personal freedom ... the appellants were able to draw on the long libertarian tradition of English law, dating back to chapter 39 of Magna Carta 1215, given effect in the ancient remedy of habeas corpus, declared in the Petition of Right 1628, upheld in a series of landmark decisions down the centuries and embodied in the substance and procedure of the law to our own day.” 19. As Lord Bridge emphasised in Ex p Khawaja, this right is of particular importance where an individual is subjected to executive detention: “My Lords, we should, I submit, regard with extreme jealousy any claim by the executive to imprison a citizen without trial and allow it only if it is clearly justified by the statutory language relied on. The fact that, in the case we are considering, detention is preliminary and incidental to expulsion from the country in my view strengthens rather than weakens the case for a robust exercise of the judicial function in safeguarding the citizen's rights.”[19] 20. These fundamental values apply with equal vigour in Hong Kong, safeguarded by, among other means, the writ of habeas corpus.[20] Giving the advice of the Privy Council in an appeal from Hong Kong,[21] Lord Browne-Wilkinson endorsed Lord Bridge’s statement just cited and emphasised that: “Such an approach is equally applicable to everyone within the jurisdiction of the court, whether or not he is a citizen of the country...”[22] D. Hardial Singh principles 21. In R v Governor of Durham Prison, ex p Hardial Singh,[23] Woolf J (as Lord Woolf of Barnes then was) had to deal with a habeas corpus application by a person against whom a deportation order had been made and who was detained under statutory powers authorising his detention pending removal.[24] The powers were not subject to any express limitation of time but, given the vital importance of the right to personal freedom, Woolf J construed them as subject to certain implied limitations relating to the purpose of the powers, the duration of the detention permitted and the Secretary of State’s obligation to act with due diligence.[25] 22. Those limitations became known as the Hardial Singh principles and have consistently been adopted in Hong Kong, [26] including by this Court which described them as representing “the proper approach to the statutory construction of any statutory power of administrative detention”.[27] 23. The principles have been conveniently summarised by Dyson LJ (as Lord Dyson then was) in R (I) v Secretary of State for the Home Department,[28] as follows: “(i) The Secretary of State must intend to deport the person and can only use the power to detain for that purpose; (ii) The deportee may only be detained for a period that is reasonable in all the circumstances; (iii) If, before the expiry of the reasonable period, it becomes apparent that the Secretary of State will not be able to effect deportation within that reasonable period, he should not seek to exercise the power of detention; (iv) The Secretary of State should act with the reasonable diligence and expedition to effect removal.” I shall refer to them as the 1st, 2nd, 3rd and 4th Hardial Singh principles, respectively. 24. His Lordship further explained: “Principles (ii) and (iii) are conceptually distinct. Principle (ii) is that the Secretary of State may not lawfully detain a person ‘pending removal’ for longer than a reasonable period. Once a reasonable period has expired, the detained person must be released. But there may be circumstances where, although a reasonable period has not yet expired, it becomes clear that the Secretary of State will not be able to deport the detained person within a reasonable period. In that event, principle (iii) applies. Thus, once it becomes apparent that the Secretary of State will not be able to effect the deportation within a reasonable period, the detention becomes unlawful even if the reasonable period has not yet expired.”[29] 25. The position at common law was summed up by Lord Hope in R (Kambadzi) v Secretary of State for the Home Department (Bail for Immigration Detainees intervening) in terms applicable to the claim in the present case:[30] “I would start therefore with principle that must lie at the heart of any discussion as to whether a person's detention can be justified. The liberty of the subject can be interfered with only upon grounds that the court will uphold as lawful.[31] In Ex p Evans (No 2) Lord Hobhouse of Woodborough said:[32] ‘Imprisonment involves the infringement of a legally protected right and therefore must be justified. If it cannot be lawfully justified, it is no defence for the defendant to say that he believed that he could justify it.’ We are dealing in this case with the power of executive detention under the 1971 Act. It depends on the exercise of a discretion, not on a warrant for detention issued by any court. That is why the manner of its exercise was so carefully qualified by Woolf J in Ex p Hardial Singh. The power to detain must be exercised reasonably and in a manner which is not arbitrary. If it is not, the detention cannot be lawfully justified.” E. IO section 32 26. The Director submits that the appellant’s detention was lawfully justified pursuant to a proper exercise of the powers conferred by IO section 32. The provisions of principal relevance provide as follows: Section 32 ... (2A) A person may be detained pending the decision of the Director of Immigration, the Deputy Director of Immigration or any assistant director of immigration as to whether or not a removal order should be made under section 19(1)(b)[33] in respect of that person- (a) for not more than 7 days under the authority of the Director of Immigration, the Deputy Director of Immigration or any assistant director of immigration; (b) for not more than a further 21 days under the authority of the Secretary for Security; and (c) where inquiries for the purpose of such decision have not been completed, for a further period of 21 days under the authority of the Secretary for Security, in addition to the periods provided under paragraphs (a) and (b). ... (3A) A person in respect of whom a removal order under section 19(1)(b) is in force may be detained under the authority of the Director of Immigration, the Deputy Director of Immigration or any assistant director of immigration pending his removal from Hong Kong under section 25. F. The applicability of the Hardial Singh principles to section 32 27. Section 32 confers discretionary executive detention powers which attract the application of Hardial Singh principles. However, it is important to note that the Director is not given an unlimited discretion but can only exercise his powers subject to certain built-in statutory limitations. 28. Thus, the powers are only exercisable against persons who can be made subject to a removal order under IO section 19(1)(b) which materially provides: Section 19(1)(b) “A removal order may be made against a person requiring him to leave Hong Kong by the Director if it appears to him that that person ... (i) might have been removed from Hong Kong under section 18(1) if the time limited by section 18(2) had not passed; (ii) has ... landed in Hong Kong unlawfully or is contravening or has contravened a condition of stay in respect of him; or ... (iia) not being a person who enjoys a right of abode ... or has the right to land ... has contravened section 42; (iii) being a person who ... may not remain in Hong Kong without the permission of an immigration officer or immigration assistant, has remained in Hong Kong without such permission.” 29. In other words, the powers of detention under sections 32(2A) and 32(3A) can only be used against persons who (i) could have been refused permission to land within two months of their arrival;[34] (ii) landed unlawfully or are here in breach of a condition of stay; (iia) not having a right of abode or right to land, committed immigration offences involving making false statements or the use or possession of false documents; or (iii) are here without necessary permission to remain. 30. One might add that although section 19(1)(b) identifies a person as a potential subject of a removal order (and therefore of detention) if “it appears to [the Director]” that such person comes within one of the specified categories, membership of one such category is what the Privy Council has called “a jurisdictional fact”[35] – a fact which goes to the Director’s jurisdiction to detain. Such membership must be capable of being objectively established by the Director on the balance of probabilities as a pre-condition of any detention and is not merely a matter of the Director’s subjective opinion. 31. The section 32(2A) power is also subject to specified time-limits for its exercise, with a requirement for successive reviews and higher-level authorisation where detention is to continue beyond the first seven-day period. 32. Furthermore, the power to detain under section 32(2A) is implicitly limited to detention for the purpose of taking steps, such as making relevant inquiries, needed to arrive at the decision whether to make a removal order. This appears from 32(2A)(c) which states that second 21 day extension may be authorised “where inquiries for the purpose of such decision have not been completed”. 33. While section 32(3A) does not lay down time-limits for effecting removal where there is a removal order in force, subsections (3D) and (4A)[36] make it clear that the section recognizes that detention pending removal can only be justified for a period that is “is reasonable having regard to all the circumstances affecting [the detainee’s] detention” and that reasonableness is a matter to be determined by the court. 34. Within the framework of the aforesaid statutory conditions, the Director has a discretion to decide whether to detain or continue the detention of a particular individual who falls within a class of persons eligible for detention; and the Secretary has a discretion whether to extend and continue a particular detainee’s detention. 35. Each of the Hardial Singh principles is potentially relevant to those decisions. It must throughout be borne in mind that it does not follow that a person who may be made the subject of a removal order should be locked up or should continue to be detained while the decision whether to order removal is taken or while awaiting removal. Such persons are often allowed to remain at liberty on their own recognizance or with the support of sureties or guarantors. 36. I pause to note that the Hardial Singh principles as formulated above address only detention pending removal – the position which arose under the pertinent UK legislation in that case and which arises under IO section 32(3A). It is however clear that, suitably adapted, the principles are also applicable to the earlier phase of detention pending a decision whether or not to make a removal order under section 32(2A). G. The circumstances of the appellant’s detention 37. Mr Dykes’ Hardial Singh ground of complaint makes it necessary to examine in some detail the circumstances of the appellant’s detention. 38. As we have seen, the appellant was due to be discharged from prison on 23 August 2005 after serving just under five months of his seven-month sentence. As his release date approached, he made several requests[37] to the Director to be repatriated to Lahore in Pakistan in order to be with his sick son and aged mother. 39. Upon his release from prison, he was immediately detained at the Castle Peak Bay Immigration Centre. In a subsequent memo,[38] the Director stated that such detention was “for enquiries under section 32(2A)(a) ... pending the decision as to whether or not a removal order should be made”. 40. On 24 August, he repeated his request to be returned to Pakistan, saying that his passport had been lost and that he needed help with the purchase of an air ticket. This led to the Director writing the next day to the Consulate General of Pakistan, asking for an emergency passport to be issued to the appellant for his passage home. Since the initial seven-day period of his detention was about to expire, authority was sought from the Secretary for Security to detain him for a further period of not more than 21 days from 30 August 2005 under section 32(2A)(b) on the basis that “enquiries are continuing and are unlikely to be concluded before the expiry” of the initial period. Authority for an extension “pending a decision as to whether or not a removal order should be made under section 19(1)(b)” was given on 29 August. 41. On 2 September, the Pakistan Consulate General issued the appellant a passport and the Director procured him an air ticket. On 5 September, a recommendation was made to process a removal order, noting that the appellant “was under the detention authority section 32(2A)(b) which will lapse on 19.9.05”. On 7 September, a file minute reviewing the appellant’s case recorded that he had assumed four different identities, been in breach of a condition of stay and had refused to answer questions about his different identities. It recommended issue of a removal order and was endorsed by more a senior officer on 8 September. On 10 September, a removal order was made. 42. That was done in ignorance of the fact that the appellant had meanwhile lodged a CAT claim which arrived at the Immigration Department on 8 September and reached the Department’s Removal Sub-division on 12 September. A file minute dated 13 September recited the appellant’s immigration and criminal record, noted that the removal order had not been served and recommended that the case be passed on to the CAT Section. 43. A file minute dated 15 September again recounted the appellant’s record; noted that he had kept silent when asked about his multiple identities; and noted that he had not indicated any fear of torture in Pakistan before 5 September 2005. The minute added that in issuing the extant removal order, “the ... subject torture claim had not been taken into consideration” and recommended that the removal order be withdrawn and that the appellant’s detention be continued. That recommendation was accepted up the chain of command and the removal order was revoked on the same day, 15 September, with one superior officer adding that the Secretary’s authority would be sought to detain the appellant under section 32(2A)(c), “KIV [keeping in view] the development of the CAT assessment” and that the Department would “review subject’s detention IDC [in due course]”. 44. Detention for “a further period of twenty one days pending the decision of the Director ... as to whether or not a removal order should be made under section 19(1)(b)” was authorised under 32(2A)(c) on 16 September 2005. 45. The first mention of the appellant’s possible release from detention appears in a memo dated 21 September 2005 from the Director to the Commissioner of Police, which, after setting out the appellant’s history, stated: “Currently we are considering to release [the appellant] on immigration recognizance pending [the Director’s] decision as to whether a removal order should be made against him. In this connection, please kindly advise whether you have any objection [to the appellant’s release on recognizance]”. On the same day, the appellant’s CAT assessment began, with an interview by the Department’s CAT Section. Two days later, on 23 September, the Commissioner replied that he had no comment. 46. On 26 September 2005, the appellant was asked to nominate a guarantor for his release, but he was unable to do so until 28 September, when the guarantor gave certain undertakings to the Director. The appellant went for a further CAT interview on 29 September and, on 3 October, a file minute containing a detailed review of his case ended with a recommendation stating: “Considering that the removal order ... has been withdrawn, it is recommended that he be released on recognizance” on stated terms as to a surety, weekly reporting, details of his address and his not taking employment. Release was authorized on 5 October and on 6 October, the appellant was told that he would be released on the following day when an Urdu interpreter would be available to explain the terms of his release. He was released on recognizance on 7 October 2005. H. The Hardial Singh ground 47. As noted above, the Hardial Singh ground involves the complaint that the duration of the appellant’s detention cannot be justified. Mr Dykes’ main submission is that revocation of the removal order on 15 September 2005 due to the lodging of the CAT claim, shows that the Director must have realised that a decision on a removal order could not be reached within the maximum time allowed for detention under section 32(2A) given the time obviously needed to determine the CAT claim. It was, he argues, thereafter unlawful to keep the appellant in detention. So put, it is an argument based mainly on the 3rd Hardial Singh principle. 48. However, Mr Dykes also seeks to argue (on the basis of the 2nd Hardial Singh principle) that the duration overall was in any event unlawful, being unreasonable in all the circumstances. I therefore turn to consider the successive phases of the appellant’s detention. H.1 Detention from 23 August to 15 September 2005 49. Although there was no direct statement of the reason for detaining the appellant immediately upon his release from prison, the reviews of the appellant’s history in the case file demonstrate that he was detained because of his repeated offences of over-staying. The file minutes point to his having used four different identities, to his refusing to answer questions about those identities and to the fact that he had been convicted of breaching his conditions of stay – the last matter being a jurisdictional fact qualifying the appellant for detention. It is obvious that the Director considered it proper to detain the appellant because he presented a real risk of absconding and going to ground. In my view, the initial decision to detain the appellant was plainly lawful. 50. Keeping him in detention pursuant to section 32(2A)(a) for the first seven days (which ended on 29 August) was also clearly lawful. Throughout that period, he was pressing the Director to repatriate him to Pakistan as soon as possible. The Director acted diligently and expeditiously to enable his return. The appellant needed a travel document so on 25 August, the Director wrote to the Pakistan Consulate requesting that he be issued with an emergency passport. A passport was duly issued on 2 September, by which time, the seven day period of detention authorized under section 32(2A)(a) had expired. 51. On 29 August, the Director obtained from the Secretary of Security authority to continue the detention for not more than 21 days commencing on 30 August pursuant to section 32(2A)(b). The grant of that extension cannot be criticised. The reasons for detaining the appellant continued to hold good and the appellant’s voluntary repatriation was being processed. The Director plainly expected to make a removal order when the arrangements for his return to Pakistan were in place. Thus, when the emergency passport was issued on 2 September, the Director took steps to secure the appellant an air ticket for his return to Pakistan and on 10 September, pursuant to a recommendation made on 7 September, a removal order was duly made. 52. It was at that point that a new development arose, involving the appellant’s change of mind about returning to Pakistan and his lodging of a torture claim. It is somewhat odd that, having been received at the Immigration Department on 8 September, it took four days for news of the new development to reach the Removal Sub-division, raising a question as to whether the Director had acted with reasonable diligence and expedition as required by the 4th Hardial Singh principle. However, that does not appear to have been explored in the evidence. No doubt, from the viewpoint of the officers dealing with his removal, his repatriation which would end the detention was being reasonably and diligently processed. 53. Once it was realised by those officers that the appellant had changed his mind and now resisted being returned to Pakistan allegedly for fear of being tortured there, it was quite properly decided to refer his claim to be assessed by the CAT Section. It was also perfectly proper to revoke the existing removal order on the basis that it could not be proceeded with, having been issued without taking account of the CAT claim. It was furthermore quite reasonable for the Removal Section to adopt the stance that they should keep in view the CAT Section’s assessment and review the appellant’s detention in due course. Given the fact that the appellant had been back and forth between Hong Kong and Pakistan and that he had never previously suggested a fear of torture, but on the contrary, had been pressing for his rapid return to that country, one cannot fault the officers dealing with his removal for wishing to see what the CAT Section would make of the appellant’s volte-face before deciding whether he had to be released. 54. I therefore do not accept that the Court should infer from the revocation on 15 September of the removal order that the Director had by then already concluded that the appellant could not be removed within the detention time limits specified in section 32. It was revoked because it could not be proceeded with without consideration given to the CAT claim and it was not out of the question that removal might be possible within the time remaining. H.2 Detention from 16 September to 7 October 55. What is not so readily acceptable, is the time taken for the process to unfold after 15 September. After a recommendation was made on 13 September to pass the file to the CAT Section and after revocation of the removal order on 15 September, it was envisaged that a decision on removal could not be reached before the then authorized period of detention expired on 19 September. Authority was therefore obtained on 16 September for a further 21 day extension under 32(2A)(c) beginning on 20 September. However, the CAT assessment process did not start until the appellant was first interviewed on 21 September, nine days after the claim reached the Removal Section and 13 days after it had first been received by the Immigration Department. It has to be borne in mind that in such cases, while the file is being passed from one officer to another, the appellant is being held in detention, deprived of his personal freedom, recognized both at common law and under our constitutional guarantees as a fundamental right. It is incumbent on the Director to process the case with a sense of urgency and to come promptly to a decision whether detention should continue. There is no explanation as to why the assessment could not have taken place much sooner. 56. The pace of dealing with the case did not improve. As we have seen, by 21 September, the appellant’s release was under consideration and the Commissioner of Police was being asked whether he had any objections. That point should have been reached some days sooner, in tandem with an earlier CAT assessment, if the process had been pursued with greater diligence and expedition as required by the 4th Hardial Singh principle. 57. It is furthermore difficult to understand why it was not until 26 September that the appellant was asked to nominate a guarantor. That occurred five days after his possible release was mooted and three days after it was clear that the Commissioner of Police had no interest in the appellant. The appellant managed to secure a guarantor two days later, on 28 September. It is evident from the events which followed that once the Commissioner’s lack of objection was known, it only remained to process the mechanics of release. There is no apparent reason why the appellant was not asked to secure a guarantor say, on 24 September, the day after the Commissioner’s response was obtained. More damagingly, there is no evident reason why it was not until 3 October that a recommendation for his release was made (5 days after a guarantor had been found) and not until 5 October that his release was authorized. There was then a further delay to secure an Urdu interpreter to explain the terms on which he was to be released before he finally emerged from detention on 7 October. One assumes that the terms of the guarantee needed were explained to the appellant by an Urdu interpreter and it is difficult to see why the terms on which he might be released if a guarantor could be secured could not have been explained at the same time, or at any rate, at a time which did not delay his actual release. H.3 Conclusion as to the Hardial Singh ground 58. For the foregoing reasons, I conclude that the appellant’s detention up to 15 September was lawful. I also conclude that while a period of detention after 15 September can lawfully be justified, the actual period of detention was excessive and inconsistent with the 2nd, 3rd and 4th Hardial Singh principles. 59. When it became known that the appellant was making a CAT claim which would interrupt the removal process, that claim should rapidly have received a first assessment by the CAT Section to see whether it was manifestly untenable. Once it became clear that the CAT claim had to run its course, it would have been obvious that no decision to make a removal order could have been arrived at within the maximum period of detention permitted under section 32. Applying the 3rd Hardial Singh principle, steps should then have been taken without delay to effect the appellant’s release. The steps actually taken, involving communication with the Commissioner of Police, the finding of a guarantor and explaining the terms of his release, ought to have been taken with far greater urgency. 60. In my view, looking at the position broadly, it is reasonable to conclude that using the reasonable diligence and expedition required by the 4th Hardial Singh principle, the entire process ought to have been completed some ten days sooner. I therefore conclude that the appellant’s detention for six weeks was 10 days more than what was justifiable in all the circumstances. Accordingly, I hold that on the Hardial Singh ground, the appellant is entitled to damages for false imprisonment for 10 days. I. The “no legal basis ground” 61. The “no legal basis” argument is advanced on the premise that once the removal order was revoked on 15 September 2005, the Director no longer had in view the making of a removal order within the time limited and so had no legal basis for detaining the appellant under section 32. On the factual analysis conducted above, this ground fails. The removal order was revoked because it had been overtaken by the CAT claim. The continued detention was in contemplation of a fresh removal order possibly being made and, when it became clear that that was not going to happen in the time limited, detention was continued while arrangements were made for the appellant’s release on recognizance. The unlawfulness I have found is based on dilatoriness and the unreasonableness of the detention’s overall duration. The suggestion that the Director lacked power to detain because he had abandoned all intention to make a removal order once the issued order was revoked is not made out. J. The “lack of a policy” ground J.1 The appellant’s argument and the decisions below 62. The appellant’s argument on this ground is that the entire period of his detention was rendered unlawful because in August 2005 when he was detained, the Director had not published any statement of policy identifying the criteria to be adopted in exercising his powers under section 32.[39] The appellant’s contention is that the Director was under a public law duty to exercise his powers of detention only in a certain and accessible manner and that this required him to publish policies identifying such criteria. Support for this argument is sought from two cases involving immigration detention recently decided by the United Kingdom Supreme Court, namely, R (WL (Congo)) v Home Secretary[40] and R (Kambadzi) v Secretary of State for the Home Department (Bail for Immigration Detainees intervening),[41] to which I shall return. 63. A similar argument was advanced in the Court of Appeal in A (Torture Claimant) v Director of Immigration[42] under the rubric of BOR Art 5(1) (and not as a domestic public law duty presently under consideration). The Court held that the grounds and procedure for detention under IO section 32 did not meet the requirements of certainty and accessibility laid down by BOR Art 5(1). It should however be noted that Tang VP (as he then was) emphasised that the question concerned the requirements of certainty and accessibility under BOR Art 5(1) and that his Lordship did not regard the Article as laying down an obligation to make and publish policies: “... we are here concerned with situations where art.5 of the HKBOR applies. Here the critical question is not whether there are policies as to the circumstances under which the power to detain might be exercised, the question is whether the grounds and procedure for detention are sufficiently certain and accessible, as is required by art.5. ...”[43] “Article 5 also requires that the grounds and procedure for the exercise of the power to detain must be certain and accessible. They could be made certain by a policy and accessible by publication. But the making of a policy is not the only way. Legislation, whether substantive or subsidiary, may do as well. But the question in every case must be, whether the grounds and procedure for detention are sufficiently certain and accessible.”[44] 64. A difficulty with the decision in the A (Torture Claimant) Case is that the effect of section 11 of the Hong Kong Bill of Rights Ordinance (“section 11”) [45] was not considered.[46] In the District Court below, applying Ubamaka as decided in the Court of Appeal,[47] Judge Leung held that section 11 precluded the appellant’s reliance on BOR Art 5(1)[48] and consequently held that the “lack of a policy” ground was not open to him.[49] However, His Honour indicated that if BOR Art 5(1) had been available, he would have found that the Director’s powers as set out in the legislation alone were not sufficiently certain and accessible;[50] that the “power to detain”[51] was in breach of BOR Art 5(1) and thus unlawful;[52] that such unlawfulness was causally linked to the appellant’s detention;[53] and that he would have assessed damages at $30,000.[54] 65. The Court of Appeal agreed with the Judge that reliance on BOR Art 5(1) was excluded by section 11 and that he was not bound to follow A (Torture Claimant).[55] The “lack of a policy” argument was not directly dealt with, but the Court of Appeal evidently agreed with the Judge since it dismissed the Director’s cross-appeal on the issue of damages. It upheld His Honour’s posited breach of BOR Art 5(1) on the footing that there was “simply no evidence of a published policy concerning the power of detention under section 32(2A)”[56] and that a causal link existed between such breach and the detention.[57] J.2 Published policies and public law duties 66. I accept the submission of Mr Anderson Chow SC[58] that there is no public law duty generally requiring decision-makers to publish policies setting out their criteria for exercising statutory discretionary powers. The absence of such published policies does not of itself constitute a breach of a public law duty. However, especially where executive detention is involved, if the powers concerned are so broad and so lacking in specificity that legitimate doubts may arise as to the basis on which they are exercised in a particular case, published policies setting out relevant criteria may introduce the transparency needed as a safeguard against arbitrariness. This is of particular importance as a safeguard against arbitrary detention, but whether such a concern arises depends on case-specific considerations. 67. That there is no general duty to publish policies on discretionary powers is clear when one considers the place of such policies in the overall public law scheme. The legislature no doubt confers the relevant discretion on the decision-maker because flexibility is desired in the context. Where the power is expressed in broad terms, the legislature often authorizes the making of rules or regulations or codes of practice from which guidance and limitations on the exercise of the power will appear. Where no such rule-making powers are expressly provided for, the courts have recognized that the decision-maker may properly publish policies to give guidance and promote consistency, certainty and fairness.[59] Where this is done, those affected may have a legitimate expectation that the policies will be adhered to, although the decision-maker may be able to justify departing from those policies for good reason. It is necessary as a matter of public law to strike a balance between on the one hand, preserving a beneficial discretion which is not excessively fettered; and providing transparency to promote non-arbitrary decision-making on the other. Whether a public law duty to publish policies arises therefore depends very much on the nature of the discretionary power concerned and how it is exercised. As Sir Anthony Mason NPJ recently stated in C and Others v Director of Immigration:[60] “The exercise of a general statutory power may give rise to judicial review by reason of the way in which the power is exercised. One such example, which is relevant for present purposes, is where the decision-maker adopts a policy in order to provide guidance as to the way in which the power will be exercised and to promote consistency in its exercise. Or when the decision-maker makes a representation as to the way in which the power is exercised, including cases where the representation creates a legitimate expectation. The adoption of a policy by a decision-maker exercising a very general discretion has the advantages of promoting certainty, consistency and administrative efficiency. It is, however, important that the policy adopted, whether general in character or confined to a class of persons, is not so rigid as to exclude the exercise of discretion by the decision-maker to consider the merits of the particular case and a willingness to depart from the policy, if need be, in a particular case, at least in the general run of cases. This is because the exclusion of a residual discretion as a result of a decision-maker applying a rigid policy might well, depending on the circumstances, be at variance with the very discretion created by the statute. However, where the decision-maker purports to exercise that discretion in accordance with the stated policy the manner of that exercise may be reviewed by the courts.” J.3 The UK Supreme Court decisions relied on by the appellant 68. WL (Congo)[61] and Kambadzi[62] were both cases concerned with powers of detention exercisable by the United Kingdom Home Secretary in an immigration context. They were powers contained in the Immigration Act 1971, Schedule 3, paras 2(2) and 2(3) which provided as follows: “(2) Where notice has been given to a person in accordance with regulations under section 105 of the Nationality, Immigration and Asylum Act 2002 of a decision to make a deportation order against him, and he is not detained in pursuance of the sentence or order of a court, he may be detained under the authority of the Secretary of State pending the making of the deportation order. (3) Where a deportation order is in force against any person, he may be detained under the authority of the Secretary of State pending his removal or departure from the United Kingdom (and if already detained by virtue of sub-paragraph (1) or (2) above when the order is made, shall continue to be detained unless he is released on bail or the Secretary of State directs otherwise).” 69. Those provisions do not lay down any conditions or criteria to be satisfied before detaining the individuals concerned. In para 2(2), all that is required is that the person (who is not the subject of a court order) be given notice that a decision has been made to deport him. That is the only condition specified for detaining him until the deportation order is made. And in para 2(3), if a deportation order is in force, he may be detained until his removal. 70. It is self-evident that the section 32(2A) and section 32(3A) powers of detention in our legislation are far less open-ended. As previously noted,[63] in the first place (by virtue of section 19(1)(b)), they are powers only available against persons who could have been refused permission to land; who landed unlawfully or have breached a condition of stay; who have no right of abode or right to land and have committed immigration offences involving making false statements; or who are here without permission to remain. Those are pre-conditions for detention rationally connected to a policy of detaining persons whose situation or past record indicates a risk of evading immigration control. Secondly, section 32(2A), unlike para 2(2) in the UK Schedule, sets a 49 day maximum for detention pending the making of a removal order; and furthermore breaks up that period into three parts, requiring each of the two possible extensions beyond the initial 7 days’ detention to be subject to a review and a fresh authorization at the higher, Secretary for Security, level. Thirdly, section 32(2A) impliedly identifies as the limited purpose of the detention the conducting of inquiries[64] relevant to deciding whether to make a removal order. And fourthly, subsections (3D) and (4A) of section 32 limit the period of detention pending removal under section 32(3A) to what is reasonable in all the circumstances, expressly acknowledging that “reasonableness” is subject to judicial scrutiny. 71. It is not surprising that when referring to the ostensibly unlimited detention powers in the UK legislation, Lord Dyson JSC stated: “The rule of law calls for a transparent statement by the executive of the circumstances in which the broad statutory criteria will be exercised. Just as arrest and surveillance powers need to be transparently identified through codes of practice and immigration powers need to be transparently identified through the immigration rules, so too the immigration detention powers need to be transparently identified through formulated policy statements.”[65] 72. The problem in WL (Congo) was not that there were no published policies. Given the breadth and unspecific nature of the UK detention powers, policies designed to give guidance as to how such powers would be exercised had unsurprisingly been published. The unlawfulness castigated by the members of the Court in WL (Congo) arose out of the fact that it had been misleadingly suggested by the Secretary of State that the published policies were being applied whereas in truth a contrary, unpublished policy was being operated. Lord Dyson JSC explained this as follows: “Between April 2006 and 9 September 2008, the Secretary of State's published policy on detention of FNPs[66] under her immigration powers was that there was a ‘presumption’ in favour of release, although detention could be justified in some circumstances. In fact, during this period the Secretary of State applied a quite different unpublished policy which was described as a ‘near blanket ban’ by the Secretary of State, Ms Jacqui Smith, to the Prime Minister, Gordon Brown, on 19 September 2007 in a document entitled ‘Bail Proposal for Foreign National Prisoners’ in which she said: ‘Since April 2006, the BIA [the Border and Immigration Agency] has been applying a near blanket ban on release, regardless of whether removal can be achieved and the level of risk to the public linked to the nature of the FNP's original offence. By currently having no discretion to grant bail, the BIA has to regularly transfer FNPs around the Estate.’”[67] 73. Similarly, Kambadzi was not about the lack of a policy. It concerned a breakdown in the system whereby the policy published in the Home Office’s manual prescribing periodic reviews of the detention of a person pending deportation was not adhered to. Baroness Hale of Richmond JSC described the unlawfulness arising in the following terms: “It is not statute, but the common law, indeed the rule of law itself, which imposes upon the Secretary of State the duty to comply with his own stated policy, unless he has a good reason to depart from it in the particular case at the particular time. Some parts of the policy in question are not directly concerned with the justification and procedure for the detention and have more to do with its quality or conditions. But the whole point of the regular reviews is to ensure that the detention is lawful. That is not surprising. It was held in Tan Te Lam, above, that the substantive limits on the power to detain were jurisdictional facts, so the Secretary of State has to be in a position to prove these if need be. He will not be able to do so unless he has kept the case under review. He himself has decided how often this needs to be done. Unless and until he changes his mind, the detainees are entitled to hold him to that. ... In my view, Munby J was right to hold that the reviews were 'fundamental to the propriety of the continuing detention' and 'a necessary prerequisite to the continuing legality of the detention' ...” 74. WL (Congo) and Kambadzi,properly understood, provide no support for the contention that in the present case, the absence of formulated policies to supplement operation of section 32 of the IO constituted a breach of a public law duty on the Director’s part. Those were decisions turning on broad and unspecific detention powers in legislation which is qualitatively different from the legislation we are concerned with. They were also decisions concerning a public law duty not presently engaged, namely, the duty to adhere to a published policy promulgated as supplementing and limiting a broad discretion conferred by statute. J.4 When a public law duty to publish policies may arise 75. There are, however, helpful statements[68] in the judgment of Lord Dyson JSC in WL (Congo) indicating when, as a matter of principle, it may be a public law requirement that a decision-maker should disclose or publish policies as to how a discretion (especially one involving executive detention) is generally to be exercised: “The individual has a basic public law right to have his or her case considered under whatever policy the executive sees fit to adopt provided that the adopted policy is a lawful exercise of the discretion conferred by the statute: see In re Findlay [1985] AC 318, 338e. There is a correlative right to know what that currently existing policy is, so that the individual can make relevant representations in relation to it. In R (Anufrijeva) v Secretary of State for the Home Department [2004] 1 AC 604 , para 26 Lord Steyn said: ‘Notice of a decision is required before it can have the character of a determination with legal effect because the individual concerned must be in a position to challenge the decision in the courts if he or she wishes to do so. This is not a technical rule. It is simply an application of the right of access to justice.’ Precisely the same is true of a detention policy. Notice is required so that the individual knows the criteria that are being applied and is able to challenge an adverse decision.”[69] 76. His Lordship added: “The precise extent of how much detail of a policy is required to be disclosed was the subject of some debate before us. It is not practicable to attempt an exhaustive definition. It is common ground that there is no obligation to publish drafts when a policy is evolving and that there might be compelling reasons not to publish some policies, for example, where national security issues are in play. Nor is it necessary to publish details which are irrelevant to the substance of decisions made pursuant to the policy. What must, however, be published is that which a person who is affected by the operation of the policy needs to know in order to make informed and meaningful representations to the decision-maker before a decision is made.”[70] 77. I would respectfully adopt that approach which focuses on the ability of the person affected to make informed and meaningful representations regarding exercise of the power. There is no general duty to publish relevant policies but where the power is conferred in terms capable of giving rise to genuine doubts as to the basis of its exercise and thus preventing or significantly hampering the making of such representations in particular cases, a public law duty may arise requiring the publication or disclosure of relevant criteria or reasons. Where the power involves interference with personal freedom, recognition of such a public law duty is necessary to avoid executive detentions lacking transparency and possibly involving arbitrariness. J.5 Applied in the present case 78. It is possible that in given circumstances, a public law duty could arise on the aforesaid basis in connection with the exercise of section 32 powers of detention. However, in the present case, the appellant could not have been in any doubt as to why and on what basis he was detained pending possible repatriation to Pakistan. The appellant’s history of evading immigration control and repeated immigration offences[71] must have left him in no doubt that he was under detention because he presented an obvious risk as a person likely to abscond and evade immigration control. The course of events involving his initial request to be returned, followed by his change of mind and CAT claim have been examined above. The criticism which is in my view properly levelled at the Director is that he failed to act with reasonable diligence and expedition and kept the appellant detained for an unreasonably long period, not that he had left the appellant in the dark as to why he was being held in detention. I therefore conclude that the “lack of a policy” ground is not made out. K. The constitutional arguments K.1 BOR Art 5(1) 79. The constitutional arguments can be disposed of more briefly. BOR Art 5(1), provides as follows: “Everyone has the right to liberty and security of person. No one shall be subjected to arbitrary arrest or detention. No one shall be deprived of his liberty except on such grounds and in accordance with such procedure as are established by law.” 80. The right which it proclaims mirrors the common law right against arbitrary detention discussed in Section C above.[72] The requirement that any deprivation of the right be on grounds and in accordance with procedures established by law has been interpreted in the jurisprudence of the European Court of Human Rights to encompass requirements which find an echo in the principles relating to transparency and non-arbitrariness discussed above. Thus, in Al-Nashif v Bulgaria[73], a case involving an individual held in detention pending deportation, the Strasbourg Court stated:[74] “The Court reiterates that the phrase ‘in accordance with the law’ implies that the legal basis must be ‘accessible’ and ‘foreseeable’. A rule's effects are ‘foreseeable’ if it is formulated with sufficient precision to enable any individual—if need be with appropriate advice—to regulate his conduct. In addition, there must be a measure of legal protection in domestic law against arbitrary interferences by public authorities with the rights safeguarded by the Convention. It would be contrary to the rule of law for the legal discretion granted to the executive in areas affecting fundamental rights to be expressed in terms of an unfettered power. Consequently, the law must indicate the scope of any such discretion conferred on the competent authorities and the manner of its exercise with sufficient clarity, having regard to the legitimate aim of the measure in question, to give the individual adequate protection against arbitrary interference.”[75] 81. The content of the BOR Art 5(1) right is therefore, as previously pointed out, no different from the content of the common law right and the appellant’s position is not improved by bringing himself within its terms. If one were faced with the argument that the common law freedom from arbitrary detention had in some way been curtailed by statute, it would be important to determine whether a constitutional challenge to such a statute could be mounted on the basis of BOR Art 5(1) or BL Art 28. However, there is no suggestion that the relevant IO provisions have such an effect. 82. In any event, it is clear that by reason of section 11 the appellant cannot bring himself within BOR Art 5(1). Section 11 provides: “As regards persons not having the right to enter and remain in Hong Kong, this Ordinance does not affect any immigration legislation governing entry into, stay in and departure from Hong Kong, or the application of any such legislation.” 83. The words “this Ordinance” in section 11 refer of course to the Hong Kong Bill of Rights Ordinance[76] which implements the International Covenant on Civil and Political Rights as applied to Hong Kong and is given constitutional status by Article 39[77] of the Basic Law (“BL Art 39”). BOR Art 5(1) is a provision of that Ordinance. 84. It is not disputed that the appellant is a person who does not have “the right to enter and remain in Hong Kong”. It follows that if, in exercising the detention powers contained in IO section 32, the Director is “applying legislation governing entry into, stay in and departure from Hong Kong”, section 11 precludes reliance by the appellant on BOR Art 5(1). 85. Mr Dykes seeks to argue that section 11 does not apply because, on its true construction, legislation conferring power to detain a person pending a decision on his or her removal and pending removal itself, is not “immigration legislation governing entry into, stay in and departure from Hong Kong”. I do not agree. Such powers are inextricably part of the legislative scheme of immigration control which regulates stay in and departure from Hong Kong. They are part of the statutory machinery designed to regulate termination of a person’s stay here and to ensure his or her enforced departure where a decision to remove is taken. 86. An argument similar to Mr Dykes’ submission was mounted and rejected in GA and Others v Director of Immigration[78] in the context of a claim by mandated refugees and a screened-in torture claimant to a right to work. As Chief Justice Ma pointed out, section 11 is concerned with immigration control, as reflected in Article 154(2) of the Basic Law.[79] The section “is intended to except immigration legislation that deals with each stage of a person’s stay in Hong Kong, ... from entry through his or her stay in Hong Kong, to departure.”[80] While it was recognized[81] that possible arguments exist as to the precise limits of what is meant by “immigration legislation governing entry into, stay in and departure from Hong Kong”, it was held that the power to permit or prohibit working while in Hong Kong was clearly within that phrase. In my view, the same applies to the detention powers under discussion. K.2 BL Art 28 and BL Art 41 87. The provisions of the Basic Law relied on provide as follows: BL Art 28 The freedom of the person of Hong Kong residents shall be inviolable. No Hong Kong resident shall be subjected to arbitrary or unlawful arrest, detention or imprisonment. ... BL Art 41 Persons in the Hong Kong Special Administrative Region other than Hong Kong residents shall, in accordance with law, enjoy the rights and freedoms of Hong Kong residents prescribed in this Chapter.[82] 88. As with BOR Art 5(1), the content of the right to personal freedom guaranteed by BL Art 28 is for present purposes no different from the common law right as applied in the preceding paragraphs of this judgment. It is therefore unnecessary further to consider the application of BL28. However, as there was some discussion at the hearing and in the Courts below regarding the relationship between BL Art 41 and BL Art 28, I will say something on that topic. 89. The appellant is not a Hong Kong resident. He can therefore only bring himself within BL Art 28 if he can rely on BL Art 41. The question which arose was whether BL Art 41 operates to permit such reliance. In particular, the question was whether the words “in accordance with law” in BL Art 41 bring section 11 to bear in a manner which precludes persons who, like the appellant, do not have “the right to enter and remain in Hong Kong”, from relying on BL Art 41. 90. The Courts below answered that question positively. HH Judge Leung stated: “The qualification is ‘in accordance with law’. The availability to non-Hong Kong residents of the protection of the rights contained in Chapter III of the Basic Law is subject to the law in force in Hong Kong. ‘Law’ must include the immigration reservation to the ICCPR, now reflected in 11 of the HKBORO, which upon the above analysis is consistent with Art.39 of the Basic Law.”[83] 91. In the Court of Appeal,[84] citing that Court’s earlier decision in MA & Ors v Director of Immigration,[85] Fok JA stated: “...the phrase ‘in accordance with law’ in BL41 includes the IO so that the provisions of that ordinance would be relevant in considering the extent of a Basic Law right for a non-resident.” 92. I respectfully agree that in the present case, the effect of section 11 is to preclude reliance by the appellant on BL Art 28. However, it is important to be clear why that is so. 93. One should at once dispel any notion that the words “in accordance with law” in BL Art 41 mean that it is open to the legislature, simply by passing legislation, to deprive persons other than Hong Kong residents who are present in the HKSAR of Chapter III rights. That would plainly be a wrong reading of BL Art 41. If that had been the Basic Law’s intention, BL Art 41 would not have been included at all. The object of BL41 is to extend the Chapter III constitutional guarantees to persons in Hong Kong who are not residents. Such guarantees take effect at the level of the constitution and are in principle not merely in the legislature’s gift, to be permitted or excluded at any stage if it sees fit. 94. The reason why section 11 results in excluding the appellant’s reliance on BL Art 41 is that, by virtue of BL Art 39, section 11 operates at the constitutional level and qualifies the scope and effect of BL Art 41. BL Art 39 states: “The provisions of the International Covenant on Civil and Political Rights, ... as applied to Hong Kong shall remain in force and shall be implemented through the laws of the Hong Kong Special Administrative Region. The rights and freedoms enjoyed by Hong Kong residents shall not be restricted unless as prescribed by law. Such restrictions shall not contravene the provisions of the preceding paragraph of this Article.” 95. It is well established that the ICCPR is implemented through the Hong Kong Bill of Rights Ordinance and that such implementation includes the immigration exception contained in section 11. Thus, in Gurung Kesh Bahadur v Director of Immigration,[86] Li CJ (with whom the other members of the Court agreed) stated: “The provisions of the International Covenant on Civil and Political Rights (the ICCPR) as applied to Hong Kong were implemented through the Hong Kong Bill of Rights Ordinance (Cap 383), which contains the Hong Kong Bill of Rights (the Bill). That Ordinance effects the incorporation of the ICCPR as applied to Hong Kong into our laws. See Shum Kwok Sher v HKSAR [2002] 2 HKLRD 793, (10 July 2002) para 53, HKSAR v Ng Kung Siu & Another (1999) 2 HKCFAR 442 at 455. The ICCPR as applied to Hong Kong was subject to the reservation, originally made by the United Kingdom, that immigration legislation as regards persons not having the right to enter and remain could continue to apply. It is unnecessary to set out the terms of the reservation in full since it is reflected in s 11 of the Hong Kong Bill of Rights Ordinance.” 96. As was pointed out in Ubamaka:[87] “... in adopting the ICCPR as applied to Hong Kong, Article 39 applied the Covenant subject to the immigration reservation made at the time of the United Kingdom’s ratification. ... section 11 is consistent with that reservation. It follows that section 11 has the blessing of Article 39 and cannot be unconstitutional. Moreover, Article 154(2) of the Basic Law expressly authorises the HKSAR Government to ‘apply immigration controls on entry into, stay in and departure from the Region by persons from foreign states and regions’.” 97. In other words, in giving constitutional status to the Bill of Rights Ordinance including section 11, BL Art 39 gives constitutional status to a specific exception to relevant provisions of the Bill of Rights in relation to persons not having the right to enter and remain in Hong Kong and in respect of immigration legislation governing entry into, stay in and departure from Hong Kong, or the application of such legislation. For the reasons discussed above,[88] that exception precludes resort to the right to liberty and security of person under BOR Art 5(1) by the aforesaid class of persons in relation to the application of the specified categories of immigration legislation. Giving effect to the words “in accordance with law” in BL Art 41, it is necessary to read BL Art 28, which is concerned with freedom of the person in general, as subject to that specific exception provided for by section 11, given constitutional status by BL Art 39. 98. As we have seen, section 11 does not preclude persons who come within its terms from relying on the common law right to personal freedom and to protection against arbitrary arrest or detention. Nor does section 11 preclude such persons from relying on Chapter III and Bill of Rights guarantees in respect of legislation other than “immigration legislation governing entry into, stay in and departure from Hong Kong”. 99. It is unclear to what extent, if at all, the words “in accordance with law” in BL Art 41 are capable of taking effect to exclude Chapter III rights in situations other than the section 11 case just discussed. That is a matter which is of no present concern. L. Damages 100. The tort of false imprisonment is actionable per se without proof of loss or damage and entitles the detainee to nominal damages in any event.[89] The question arises as to whether and when breach of a public law duty is such as to make the detention unlawful for the purposes of supporting a claim for substantial damages for false imprisonment. 101. In Holgate-Mohammed v Duke,[90] Lord Diplock decided that conduct held to be unlawful on Wednesbury principles rendered such conduct unlawful not merely for judicial review purposes, but also for the purpose of founding a cause of action in damages for false imprisonment. But it is clear that not every breach of a public law duty has that result.[91] 102. Breach of a public law duty, at least in relation to interference with the common law right to personal freedom, does not require causation to be proved. Thus, in Kambadzi, where there was a failure to comply with a public law duty to carry out periodic reviews of a person’s detention, it was no defence for the Secretary of State to say that there were good grounds for detaining the appellant anyway.[92] 103. However, when it comes to damages for false imprisonment, causation is important if the detainee is to be awarded more than nominal damages. Thus, in Kambadzi, Baroness Hale JSC stated:[93] “However, the result of any review, had it been held, cannot be irrelevant to the quantum of damages to which the detainee may be entitled. False imprisonment is a trespass to the person and therefore actionable per se, without proof of loss or damage. But that does not affect the principle that the defendant is only liable to pay substantial damages for the loss and damage which his wrongful act has caused. The amount of compensation to which a person is entitled must be affected by whether he would have suffered the loss and damage had things been done as they should have been done.” 104. Breach of the public law duty has the necessary causal connection with the detention and so is capable of constituting tortious unlawfulness if it “bears directly on the discretionary power that the executive is purporting to exercise”.[94] Lord Kerr of Tonaghmore JSC put it thus: “...The public law error in the present case bore directly on the decision to detain in that it was made without the necessary review of the justification for detention. As the majority in R (on the application of Lumba) v Secretary of State for the Home Dept also held, however, causation is relevant to the question of the recoverability of damages. For the reasons that I gave in my judgment in [WL (Congo)], I consider that if it can be shown that the claimant would not have been released if a proper review had been carried out, this must have an impact on the quantum of compensation and that nominal damages only will be recoverable.”[95] 105. As we have seen,[96] HH Judge Leung held that there had been no breach of duty and awarded no damages. He indicated, however, that if BOR Art 5(1) had been available, he would have found that that Article had been breached on the basis that the appellant’s “lack of a policy” argument succeeded, rendering the detention unlawful. He found that such unlawfulness was causally linked to the appellant’s detention and that he would have assessed damages at $30,000. The Judge’s hypothetical assessment was thus in respect of the entire six-week period of detention. Neither side has any quarrel with the quantum of that assessment. 106. I have held that the “lack of a policy argument” is not made out and accordingly do not accept that the entire period of detention was unlawful. Instead, I have held that the appellant’s detention was excessive in its duration and inconsistent with the 2nd, 3rd and 4th Hardial Singh principles. Breach of those public law principles plainly had a direct bearing on the continuation of the appellant’s detention since their observance would have resulted in his release some ten days sooner. Such public law unlawfulness therefore constitutes unlawfulness for the purposes of founding a claim for substantive damages for false imprisonment over a period of ten days. 107. Taking the Judge’s assessment of $30,000 for the entire period as a base and looking at the position broadly and robustly, I would assess damages for unlawful detention over 10 days in the sum of $10,000 and accordingly allow the appeal. I would also make an order nisi that the Director pay the appellant’s costs here and below, with liberty to the parties to lodge written submissions as to costs within 14 days of the date of this judgment if so advised and, in default of such submissions, that the order nisi stand as an order absolute without further order. I would also direct that the appellant’s own costs be taxed in accordance with the Legal Aid Regulations. Mr Justice Tang PJ: 108. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Bokhary NPJ: 109. I would allow this appeal to the extent proposed by Mr Justice Ribeiro PJ and for the reasons which he gives. All that I would add is a word on art. 41 of the Basic Law which is to be found in Chapter III thereof. Article 41 provides that “[p]ersons in the Hong Kong Special Administrative Region other than Hong Kong residents shall, in accordance with law, enjoy the rights and freedoms of Hong Kong residents prescribed in this Chapter”. 110. The rights and freedoms of all residents prescribed in Chapter III of the Basic Law are those which are not restricted to permanent residents (being the right of abode under art. 24 and the right to vote or stand for election under art. 26) or to indigenous inhabitants of the New Territories (under art. 40 which protects the lawful traditional rights and interests of such inhabitants). Article 39 refers to “Hong Kong residents” without restriction. But I do not propose to say anything about this article in the present context since it operates not by enumerating rights and freedoms but by referring to other instruments that do so. 111. Article 41’s obvious purpose is to make Hong Kong a place where human rights are protected for the benefit of everyone within its territory. The Basic Law stands above ordinary law. And whatever else may be the meaning and effect of the phrase “in accordance with law” to be found in art.41, its effect is most emphatically not that non-residents are exposed to deprivation by ordinary law of the constitutional rights and freedoms which are conferred upon all residents by Chapter III. The present case does not call for a full and final decision on the meaning and effect of the phrase “in accordance with law” to be found in art. 41 and in many other articles of Chapter III. Such a decision is for another day. But it should be said here and now that the meaning and effect of the phrase cannot be to permit the undermining by ordinary law of the fundamental rights and freedoms offered by a law higher than ordinary law. The meaning and effect of the phrase would seem to be to mandate ordinary laws that play their part in delivering the fundamentals offered by a higher law, namely our constitution the Basic Law. Lord Walker of Gestingthorpe NPJ : 112. I agree with the judgment of Mr Justice Ribeiro PJ. Chief Justice Ma : 113. The Court unanimously allows the appeal, awards the appellant the sum of HK$10,000.00 by way of damages for false imprisonment and makes the orders as to costs set out in the final paragraph of the judgment of Mr Justice Ribeiro PJ. Mr Philip Dykes SC, Mr Hectar Pun and Ms Christine Yu instructed by Yip & Liu and assigned by the Legal Aid Department for the Appellant Mr Anderson Chow SC and Ms Grace Chow instructed by the Department of Justice for the Respondent [1] Cap 115. [2] DCCJ 531/2010 (13 October 2011). [3] Cheung CJHC, Stock VP and Fok JA, CACV 267/2011 (4 December 2012). [4] Ma CJ, Chan and Tang PJJ, FAMV 20/2013 (9 August 2013). [5] In Section G of this judgment. [6] [1992] 1 AC 58 at 162. Adopted in R (Kambadzi) v Secretary of State for the Home Department (Bail for Immigration Detainees intervening) [2011] 1 WLR 1299 at §40; and R (WL (Congo)) v Home Secretary (SC (E)) [2012] 1 AC 245 at §65. [7] Meering v Grahame-White Aviation Co Ltd (1919) LT 44 at 54. [8] [1992] 1 AC 58 at 162. [9] R v Home Secretary ex p Khawaja [1984] 1 AC 74 at 110; R (Abbasi) v Secretary of State for Foreign and Commonwealth Affairs and Secretary of State for the Home Department [2003] UKHRR 76 at §60. [10] Appearing with Mr Hectar Pun for the appellant. [11] In Section K of this judgment. [12] Set out in Section D.1 below. [13] A reference to the principles laid down in the judgment of Woolf J in R v Governor of Durham Prison, ex p Hardial Singh [1984] 1 WLR 704; discussed in Section D of this judgment. [14] And, where applicable under section 32(2A), the Secretary for Security. For brevity, I will refer simply to the Director. [15] In Section L below. [16] R (Kambadzi) v Secretary of State for the Home Department (Bail for Immigration Detainees intervening) [2011] 1 WLR 1299 at §94. See also Lord Hope of Craighead DPSC at §59. [17] Regarding the protection of the right to liberty and security of the person under Article 5 of the European Convention on Human Rights. [18] [2005] 2 AC 68 at §36. [19] R v Home Secretary, ex p Khawaja [1984] 1 AC 74 at 122. [20] As provided for by section 22A of the High Court Ordinance (Cap 4). [21] Tan Te Lam and Others v Superintendent of Tai A Chau Detention Centre [1997] AC 97 at 113-114. [22] Referring to Lord Scarman in Ex p Khawaja at 111-112. [23] [1984] 1 WLR 704. [24] Immigration Act 1971, Sch 3, paragraphs 2(2) and 2(3) set out in Section J.3 below. [25] [1984] 1 WLR 704 at 706. [26] Tan Te Lam and Others v Superintendent of Tai A Chau Detention Centre [1997] AC 97 at 111; Chieng A Lac & Others v The Director of Immigration & Others [1997] HKLRD 271 at 277-278; A (Torture Claimant) v Director of Immigration [2008] 4 HKLRD 752 at §27-28. [27] Thang Thieu Quyen v Director of Immigration & Another (1997-98) 1 HKCFAR 167 at 185-186. [28] [2003] INLR 196 at §46. The summary was approved by the UK Supreme Court in R (WL (Congo)) v Home Secretary [2012] 1 AC 245 at §22. [29] At §47. [30] [2011] 1 WLR 1299 at §49. [31] Citing R v Governor of Brockhill Prison, ex p Evans (No 2) [2001] 2 AC 19 at 35; and Tan Te Lam v Superintendent of Tai A Chau Detention Centre [1997] AC 97 at 111. [32] [2001] 2 AC 19 at 42. [33] Section 19(1)(b) identifies the classes of persons who may be detained. Its terms are set out in Section F below. [34] IO section 18 relevantly states: “(1) Subject to subsection (2), an immigration officer or a chief immigration assistant may remove from Hong Kong (a) ... a person who ... is under section 11(1) refused permission to land; ... (2) A person who is refused permission to land in Hong Kong may not be removed from Hong Kong under subsection (1)(a) after the expiry of 2 months beginning with the date on which he landed.” [35] Tan Te Lam and Others v Superintendent of Tai A Chau Detention Centre [1997] AC 97 at 113. [36] Section 32(3D): “For the further avoidance of doubt, nothing in subsection (3B) shall prevent a court, in applying subsection (4A), from determining that a person has been detained for an unreasonable period.” Section 32(4A): “The detention of a person under this section shall not be unlawful by reason of the period of the detention if that period is reasonable having regard to all the circumstances affecting that person's detention including, in the case of a person being detained pending his removal from Hong Kong- (a) the extent to which it is possible to make arrangements to effect his removal; and (b) whether or not the person has declined arrangements made or proposed for his removal.” [37] At an Immigration Department interview on 12 August 2005; by a handwritten letter dated 18 August 2005 and at a further interview on 23 August 2005. [38] Dated 29 August 2005. [39] In 2008, policies identifying situations in which the section 32 powers are and are not likely to be exercised were published on the Immigration Department’s website at http://www.sb.gov.hk/eng/special/pdfs/Detention%20policy-e.pdf. [40] [2012] 1 AC 245. [41] [2011] 1 WLR 1299. [42] [2008] 4 HKLRD 752. [43] At §40. [44] At §41. [45] Cap 383. Section 11: “As regards persons not having the right to enter and remain in Hong Kong, this Ordinance does not affect any immigration legislation governing entry into, stay in and departure from Hong Kong, or the application of any such legislation.” [46] As pointed out by the Court of Appeal below, §41. [47] Ubamaka v Secretary for Security [2011] 1 HKLRD 359. [48] Discussed in Section K.1 below. [49] Judgment §51. [50] Judgment §77. [51] The Judge may have meant “the detention” rather than “the power to detain”. [52] Judgment §§83 and 85. [53] Judgment §109. [54] Judgment §158(3) and (4). [55] Court of Appeal §55, per Fok JA (as he then was). [56] Court of Appeal §87. [57] Court of Appeal §89. [58] Appearing for the Director with Ms Grace Chow. [59] See the authorities cited in De Smith’s Judicial Review, (Sweet & Maxwell, 7th Ed) §9-10. [60] Chan, Ribeiro and Tang PJJ, Bokhary NPJ and Sir Anthony Mason NPJ, FACV Nos 18, 19 & 20 of 2011 (25 March 2013) at §§73-74. [61] R (WL (Congo)) v Home Secretary [2012] 1 AC 245. [62] R (Kambadzi) v Secretary of State for the Home Department (Bail for Immigration Detainees intervening) [2011] 1 WLR 1299. [63] In Section F of this judgment. [64] And in my view, taking other reasonably necessary steps. [65] WL (Congo) at §34. [66] “FNPs” were “Foreign National Prisoners” under consideration for detention upon completion of their sentences of imprisonment: WL (Congo) at §1. [67] WL (Congo) at §5. [68] Albeit made in the different context of criticising the application of unpublished policies while ostensibly adopting inconsistent published policies. [69] At §§35 and 36. [70] At §38. See also Lord Walker of Gestingthorpe JSC at §190. [71] Discussed in Section H above. [72] See eg, Winterwerp v Netherlands (1979-80) 2 EHRR 387 at §37. [73] (2003) 36 EHRR 37. [74] In the context of Article 8 of the Convention dealing with right to respect for family life. [75] (2003) 36 EHRR 37 at §119. Another example is Amuur v France (1996) 22 EHRR 533 at §50. [76] Cap 383. [77] Article 39: “The provisions of the International Covenant on Civil and Political Rights, ... as applied to Hong Kong shall remain in force and shall be implemented through the laws of the Hong Kong Special Administrative Region. The rights and freedoms enjoyed by Hong Kong residents shall not be restricted unless as prescribed by law. Such restrictions shall not contravene the provisions of the preceding paragraph of this Article.” [78] Ma CJ, Ribeiro and Tang PJJ, Chan NPJ and Lord Clarke of Stone-cum-Ebony NPJ, FACV 7, 8, 9 & 10/2013 (18 February 2014). [79] At §29(2) and (3). BL Article 154(2): “The Government of the Hong Kong Special Administrative Region may apply immigration controls on entry into, stay in and departure from the Region by persons from foreign states and regions.” [80] At §30. [81] At §35. [82] Including BL Art 28. [83] At §56, citing Santosh Thewe & Anor v Director of Immigration [2000] 1 HKLRD 717 at 721D-722H; and Gurung Ganga Devi v Director of Immigration, HCAL 131/2008 (23 September 2009) at §§20-23; 27. [84] Court of Appeal at §63. [85] Stock VP, Kwan and Fok JJA, CACV 44-48/2011 (27 November 2012) at §145. [86] (2002) 5 HKCFAR 480 at §§21-22. [87] Ubamaka Edward Wilson v Secretary for Security (2012) 15 HKCFAR 743 at §95 [88] In Section K.1. [89] R (WL (Congo)) v Home Secretary (SC (E)) [2012] 1 AC 245 at §64 per Lord Dyson SC; R (Kambadzi) v Secretary of State for the Home Department (Bail for Immigration Detainees intervening) [2011] 1 WLR 1299 at §74 per Baroness Hale JSC. [90] [1984] 1 AC 437 at 443. [91] R (Kambadzi) v Secretary of State for the Home Department (Bail for Immigration Detainees intervening) [2011] 1 WLR 1299 at §43 and §65. [92] R (Kambadzi) v Secretary of State for the Home Department (Bail for Immigration Detainees intervening) [2011] 1 WLR 1299 at §54 and §74. [93] At §74. See also Lord Kerr of Tonaghmore JSC at §89. [94] R (Kambadzi) v Secretary of State for the Home Department (Bail for Immigration Detainees intervening) [2011] 1 WLR 1299 at §41 per Lord Hope of Craighead DPSC. [95] Kambadzi at §§88 and 89. [96] In Section J.1 above. 
Chief Justice Ma: 1. I agree with the joint judgment of Mr Justice Ribeiro PJ and Mr Justice Cheung PJ. Mr Justice Ribeiro PJ and Mr Justice Cheung PJ: 2. The 1st appellant, Chen Keen (alias Jack Chen), the 2nd appellant, Hao May (alias May Wang) and the 3rd appellant, Yee Wenjye (alias Eric Yee), “A1”, “A2” and “A3” respectively, were convicted on 29 April 2016 on two counts of conspiracy to defraud contrary to common law[1] after trial before Anthea Pang J and a jury.[2] The first count alleged a conspiracy to defraud the Stock Exchange of Hong Kong Limited (“SEHK”) and the second, a conspiracy to defraud China Jin Hui Mining Corporation Limited,[3] a company listed on the SEHK (referred to by its stock code number “462”) and its existing shareholders. The appellants were alleged to have conspired by using dishonest means to cause the SEHK and 462 to permit the acquisition by 462 of farms in New Zealand, held by companies owned by A2, to proceed. A1 was also convicted on a count of money laundering[4] in respect of his dealing with the alleged proceeds of such conspiracies to defraud. 3. Their appeals to the Court of Appeal[5] were unsuccessful and they were granted leave to appeal to this Court by the Appeal Committee[6] on two questions of law and in respect of two issues raised on the “substantial and grave injustice” (“SGI”) basis, to which we shall return. A. The factual background A.1 Planning to acquire farms in New Zealand 4. A2 controlled three companies which are relevant for present purposes. She was the owner and sole director of UBNZ Trustee Limited (“UTCL”[7]) which had, as its wholly-owned direct subsidiaries, UBNZ Funds Management Limited (“UBFM”) and UBNZ Assets Holdings Limited (“UBAH”). 5. In early 2008, A2 asked one Barry Fraser (“Fraser”) to help her look for New Zealand farms to purchase. Fraser identified certain cattle and dairy cattle farms (“the Properties”) owned by companies known as the “CraFarms Group” and A2 authorised him to enter into negotiations to acquire them. A1 then signed a commission sharing agreement (“CSA”) dated 8 October 2008 with A2 and Fraser (the latter signing on behalf of Latitude Asia Limited, a company of which his two sons were directors). The CSA provided as follows: “All parties agree that all commission received through dairy farm purchases with CraFarms Group (Incorporating Plateau Farms Ltd, Hillside Ltd, Taharua Ltd, Nugen Farms Ltd, Ferry View Farms Ltd, Windburn View Ltd) will be shared equally between the said three parties. All commissions negotiated and paid by vendors and investors (if any) will be shared equally between the said three parties. Commissions will be payable by the vendor and investors (if any) upon settlement of dairy farm purchasers. Commissions will be held by a nominated company trust account. The said parties will only be able to claim commission upon submitting an invoice to the nominated company. The Confidentiality Agreement between CraFarms Group and UBS (sic) Funds Management NZ Limited forms part of this agreement.” A.2 Arranging for 462 to acquire the farms 6. Some months later, in about April 2009, A1 approached Katherine Chan Wai-kay (referred to below as “Kathy Chan” or “PW1”[8]) who was a director and chairman of 462 which was, at that time, a moribund company with trading in its shares suspended. A1 told PW1 that he knew of farms in New Zealand “which were a very wealthy woman’s farms” (naming her as May Wang).[9] PW1 expressed an interest in having the farms injected into 462 and a few days later, A1 introduced PW1 to A2 when the supply of dairy products from New Zealand to the Mainland was discussed as a project with potential. A1 was then appointed a director and co-chairman of 462 to facilitate negotiations with A2. 7. On 18 May 2009, A2, through UBFM, entered into a series of agreements to purchase 22 farms owned by the six companies forming part of the CraFarms Group[10] for a total consideration of NZ$259 million.[11] 8. In the following week, on 22 May 2009, 462’s board, chaired by PW1 and A1, approved purchase by 462 of those farms with the minutes recording that the board was told that the ultimate beneficial owners of UTCL and UBFM were “third parties independent of the company and its connected persons.”[12] 9. Later that day, 462 entered into an Agreement dated 22 May 2009 with UTCL and UBFM for the purchase from UTCL of the shares of its subsidiary UBAH (“the Acquisition”), the vendors undertaking that UBAH would become the owner of the Properties (which its fellow subsidiary UBFM had contracted to purchase from CraFarms) before completion under the Agreement. 10. The purchase would initially be of 20% of the shares in UBAH for the HK Dollar equivalent of NZ$100 million with 462 having the option of acquiring the remaining 80% of UBAH for NZ$400 million. The consideration was to be provided by the issue of convertible notes, that is, debt instruments convertible into shares in 462, to UTCL. It was agreed that 462 would perform due diligence inquiries. Conditions precedent to the Acquisition included permission from the SEHK to list and deal in converted shares; approval by 462’s shareholders; and all necessary approvals from the relevant New Zealand authorities (in particular the Overseas Investment Office (“OIO”) of New Zealand) for the transfer of ownership of the farms. A.3 Due diligence and the Announcement 11. As the Acquisition qualified as a Very Substantial Acquisition (“VSA”) under the SEHK’s Listing Rules, trading in the shares of 462 was suspended between 25 May 2009 and 5 June 2009, pending the making of a public announcement.[13] On 3 June 2009, a firm of accountants, ShineWing HK (CPA) Ltd (“ShineWing”) was engaged by 462 to conduct a due diligence check on UBAH and the farms. At about the same time, A2 engaged A3 as a consultant to compile accounts for the Acquisition and to provide the financial information to ShineWing. 12. The public announcement was published on 4 June 2009 (“the Announcement”). It stated that 462 had agreed conditionally to purchase from UTCL the UBAH shares (describing the principal terms of the agreement) and contained the following declaration: “To the best of the information, knowledge and belief of the Directors having made all reasonable enquiries, UTCL and UBFM and their respective ultimate beneficial owners are third parties independent of the Company and its connected persons.” 13. In New Zealand, Stretton’s were the CraFarms Group’s accountants, Mark King (“King”) being the accountant in charge. In the course of the due diligence exercise, in June 2009 King provided unaudited management accounts for the years ended 31 May 2006, 2007 and 2008 to A3 and A2 who forwarded the same to ShineWing and PW1. 14. In e-mails examined in detail by the Court of Appeal,[14] A3 engaged in re-casting the figures in various ways under pressure from A2. By 24 June 2009, the overall deficit of NZ$5,420,975 shown in King’s figures had been altered to present an overall surplus of NZ$14,875,785.[15] After further manipulation, by 7 July 2009 the purported net profit figure for the 10 month period ended 31 March 2009 had been further revised upwards to NZ$17,789,715.[16] When the circular discussed below was published on 8 September 2009 (“the Circular”), it put the unaudited profit of UBAH for the year ended 31 May 2009 at NZ$20 million, equivalent to HK$95.95 million. 15. The prosecution case was that the manipulation by A3 (in conjunction with A2 and with the acquiescence of A1) of the historical figures provided by King represented the dishonest falsification of those figures. The defence argued that the adjusted figures did not represent and were known by ShineWing not to represent historical figures but reflected honest projections by A3 of achievable improvements brought about by achieving efficiencies and a “change of management focus”. A.4 Seeking SEHK’s approvals 16. During July and August, 462 was engaged in seeking the SEHK’s permission to issue the Circular required to be published for the Acquisition, being a VSA. This was referred to as a “pre-vetting” process and involved the SEHK raising a series of queries. In a letter dated 10 August 2009, the SEHK stated that it had received a complaint alleging that A2 was the sole shareholder and director of the vendor UTCL and that: “In respect of the Acquisition, Ms Wang is partnering with Mr Jack Keen Chen ... the executive director of the Company. They work together in New Zealand under the same office at Level 10, 175 Queen Street, Auckland, New Zealand, and the office is a property owned by Ms Wang ...” 17. The letter asked whether the allegations were accurate and went on as follows: “if Ms Wang is the beneficial owner of the Vendor, your confirmation whether there is any existing or prior relationship between Mr Chen and Ms Wang, or whether there is any prior arrangement(s) or understanding between Mr Chen and Ms Wang with respect to the Acquisition such that the Acquisition shall become a connected transaction under Chapter 14A ...” “We note that the circular on the Acquisition will be despatched by the Company in due course. Please ensure that all information relevant for shareholders to make an informed assessment on the Acquisition has been duly disclosed in such circular, including the information mentioned in the complaint above, where appropriate.” 18. A1 responded by saying that he had met A2 on non-business related events and that he had then been appointed as director of 462 to “assist and negotiate the acquisition”. He asserted that: “There were [sic] no existing or prior relationship between myself and Ms Wang with respect to the Acquisition”. A2 acknowledged that she was the sole shareholder of UTCL and likewise declared: “There was no existing or prior relationship between Mr Chen & me with respect to the Acquisition.” A.5 The Circular and the Acquisition 19. On 8 September 2009, 462 published the Circular which stated that: “To the best of the information, knowledge and belief of the Directors having made all reasonable enquiries, each of the Target Company, the Vendor, the Warrantor, the ultimate beneficial owner of the Target Company, the Vendor and the Warrantor, the registered owners and the ultimate beneficial owners of the Properties and Fixed Assets are Independent Third Parties to the Company and its connected persons and in particular there is no existing or prior relationship or understanding between Mr Jack Keen Chen and Ms May Wang with respect to the Acquisition such that the Acquisition shall become a connected transaction under Chapter 14A of the Listing Rules.” 20. As noted above, the Circular put the unaudited profit of UBAH for the year ended 31 May 2009 at NZ$20 million, equivalent to HK$95.95 million. The unaudited combined profit and loss statements attributable to the farms for the year ended 31 May 2009 annexed to the Circular also stated that the farms’ gross profits were HK$92,520,000 (“representing an improvement of 68.1% when compared to 31 May 2008”). The Circular further said that the statements had “been prepared based on the financial and other information provided by the management of the vendor”. 21. On 2 October 2009, at an Extraordinary General Meeting, 462 approved the Acquisition.[17] 22. On 4 December 2009, the SEHK granted listing approval for 462 to issue convertible notes to UTCL. 462 then appointed Sun Hung Kai Investment Services Ltd (“SHKI”) as placing agent to market the additional convertible notes.[18] A.6 The Convertible Notes, their proceeds and the incomplete Acquisition 23. Consequently, on 21 December 2009, 462 issued to UBTL NZ$50M worth of notes convertible into 110,431,200 shares in 462 as the Acquisition deposit and NZ$100M worth of notes convertible into 276,077,999 such shares for the acquisition of 20% of UBAH’s shares.[19] On the same day, the SEHK granted 462 listing approval to issue the additional convertible notes and an Optional Bond.[20] 24. Between 16 July and 1 September 2010, sales of some of the convertible notes took place and the 462 board approved conversions of various batches of convertible notes into shares. Because of the SEHK’s concerns in the light of various complaints made against A1 and A2, 462 froze and stopped transfer for any CN-A and CN-B shares released to UTCL.[21] A2 was declared bankrupt in New Zealand on 8 December 2010[22] and A1 and A2 were arrested on 16 December 2010.[23] 25. On 12 January 2011, OIO refused to approve transfer of the farms to UBAH and UBFM cancelled the sale and purchase agreements with the New Zealand farm companies.[24] B. The issues on these appeals 26. Multiple issues were canvassed at the trial where Anthea Pang J’s summing-up took over three days and covered 274 pages of transcript. The Court of Appeal likewise had to deal with numerous grounds of appeal and the judgment of the Court spans 260 pages. However, on the present appeals, the debate has been narrowed to focus on two main issues, namely, (i) whether there were certain misdirections regarding the relationship among the three appellants so as to invalidate their conviction (“the misdirection issue”); and (ii) whether, in the light of the Judge’s directions, the indictments for conspiracy to defraud were bad for duplicity (“the duplicity issue”). As it is common ground that A1’s conviction on the money laundering count stands or falls with his conviction for conspiracy to defraud, no independent issues in respect of that count require discussion in this judgment. 27. The two main issues are reflected in the Appeal Committee’s order granting leave to appeal in respect of two questions of law and two issues considered reasonably arguable on the SGI basis, as follows: In the context of charges of conspiracy to defraud the Stock Exchange and shareholders of a listed company, where false representations are said to have been made as to connections between the directors of the listed company and the vendor in a very substantial acquisition governed by the Listing Rules, for the purpose of the direction on falsity, is the Judge required to determine and direct the Jury, as a matter of law, on the meaning of a “connected transaction” and “connected persons” under the Listing Rules? (“Question 1”) Whether it is permissible in law to charge a single count of conspiracy to defraud, which identifies different agreed dishonest means and/or different objectives where the alleged co-conspirators have not all agreed to the use of the same dishonest means and/or the pursuit of these same objectives. (“Question 2”) Whether it was a misdirection for the Judge to instruct the jury that it was sufficient to convict on the basis of a finding that A1 and A2 had been “working together” in the specified context without examining whether A2 was a “connected person” so as to render the acquisition a “connected transaction” for the purposes of the Listing Rules. (“SGI-1”) Whether it was a misdirection for the Judge to instruct the jury that concealment or non-disclosure of the Commission Sharing Agreement of October 2008 could constitute a conspiracy to defraud. (“SGI-2”) 28. The misdirection issue is raised by Question 1 and SGI-1 and SGI-2 which substantially overlap. The duplicity issue arises from Question 2. It is convenient to deal first with Question 2 and the duplicity issue. C. Question 2 and the duplicity issue 29. The duplicity argument of Ms Clare Montgomery QC[25] on behalf of A1, adopted by Mr Ian Winter QC[26] for A2 and by Mr Tim Owen QC[27] for A3, focusses on the five particulars in each Count of the Indictment set out below, contending that her Ladyship’s direction on those particulars caused a latent duplicity to emerge, invalidating the jury’s verdict. C.1 The Indictment 30. The Indictment materially provides as follows: “ First Count (against all accused) STATEMENT OF OFFENCE Conspiracy to defraud, contrary to Common Law and punishable under section 159C(6) of the Crimes Ordinance, Cap 200 PARTICULARS OF OFFENCE CHEN Keen (alias Jack CHEN), HAO May … (alias May WANG) and Yee Wenjye … (alias Eric YEE), between the 7th day of May 2009 and the 8th day of September 2009, both dates inclusive, in Hong Kong, conspired together to defraud the [SEHK] by dishonestly: (a) falsely representing that [UTCL], [UBFM] and their respective ultimate beneficial owners were independent third parties to [462] and its connected persons; (b) falsely representing that there was no existing or prior relationship or understanding between [A1 and A2] with respect to the acquisition of the entire issued share capital of [UBAH] by [462] from [UTCL] (“the Acquisition”); (c) concealing or failing to declare that [A1] had an interest in the Acquisition, in that [A1 and A2] were parties to an agreement signed between them and Latitude Asia Limited dated the 8th day of October 2008 to share the commission arising from or received through the sale and purchase of the dairy farms owned by the CraFarms Group in New Zealand; (d) falsely representing that the gross profit of the properties and fixed assets relating to dairy farms owned by the CraFarms Group in New Zealand (“the Properties”) for the year ended the 31st day of May 2009 was approximately $92,520,000 Hong Kong currency; (e) concealing or failing to disclose the true financial position of the Properties.” 31. Particular (f) in Count 1 goes on to state: “(f) causing SEHK to allow the publication by [462] of the announcement dated the 4th day of June 2009 and the circular dated the 8th day of September 2009 in relation to the Acquisition which contained the false representations as aforesaid.” 32. The Statement of Offence and the parties to the conspiracy specified in Count 2 are the same as in Count 1. However, they are alleged, between 7 May 2009 and 19 July 2010, to have “conspired together to defraud [462] and its existing shareholders” by dishonestly making the same false representations and effecting the same concealments and non-disclosures as those set out in Particulars (a) to (e) in Count 1. 33. Count 2 has Particulars (f) and (g) which allege as follows: “(f) causing [462] and its existing shareholders to approve the agreement for the Acquisition; (g) causing [462] to issue and release convertible notes and an optional bond for the payment of the Acquisition.” C.2 The duplicity argument 34. The duplicity argument is that “each of Counts 1 and 2 improperly averred more than one conspiracy”[28] thus undermining the jury’s verdicts. Ms Montgomery QC submitted (i) that two quite different conspiracies were alleged by Particulars (a) to (c) and Particulars (d) and (e) respectively; (ii) that given the Judge’s direction that Particulars (a) to (c) were not relevant to A3 and that they could not find a conspiracy proved against him “based on any one or more of those particulars”[29]; and (iii) that given also the Judge’s direction that “at least two of the Accused ... intended that such a false representation be made or that there be such a concealment”[30]; it could not be excluded that the guilty verdicts returned were constituted by the jurors having been satisfied that the conspiracy constituted by Particulars (a) to (c) had been proved as against A1 and A2; while finding that the different conspiracy constituted by Particulars (d) and (e) was established as against A3 and either A2 or A1. Thus, it was argued, each of the conspiracy counts accommodated “two conspiracies with different objects and different parties charged under the same count”[31] and led to the possibility that the conviction of all three of the appellants on each count had been arrived at without the jury unanimously or by an acceptable majority having found either offence established against them. 35. It may be noted that the duplicity argument separates the particulars into two groups: Particulars (a) to (c) regarding false representations and concealment concerning the relationship between each of the appellants; and Particulars (d) and (e) regarding false representations and concealment concerning the financial position of the Properties to be acquired. It is crucial to the appellants’ argument that each of the two sets of particulars – Particulars (a) to (c) and Particulars (d) and (e) respectively – represents the agreed dishonest means under the conspiratorial agreement to which all the alleged conspirators must be proved to have been parties, reflecting their agreed common intention. 36. The correctness or otherwise of the duplicity argument therefore depends on a proper understanding of the applicable legal principles and on what the essential constituent elements of the conspiratorial agreements indicted consist of and what was the Judge’s direction thereon. C.3 The applicable legal principles C.3a Offence constituted on reaching agreement 37. The offence of conspiracy to defraud (like other conspiracies) is focussed upon and crystallised by the co-conspirators’ making of the unlawful agreement. As Thomas LJ explained in R v K: “The rationale for the retention of the offence of conspiracy to defraud is that the criminality aimed at is the agreement, not the carrying out of the agreement; if a sufficiently certain agreement is made to defraud, that is the criminal conduct encompassed within the offence and no more need be proved; provided there is that certainty in the agreement, it matters not how the participants individually intended to go about or actually went about defrauding the intended victims of their money.”[32] 38. Such an agreement is proved by showing that the co-conspirators agreed to act in concert to achieve a common object or purpose.[33] As Brennan J put it in Gerakiteys v The Queen:[34] “The identity of a conspiracy is to be found in what the conspirators commonly agree to or accept: a conspiracy is proved by evidence of the actual terms of the agreement made or accepted or by evidence from which an agreement to effect common objects or purposes is inferred.” 39. As Jordan CJ pointed out, “The prosecution is not called upon to define the exact moment at which the conspiracy began or the exact act which marked its inception.”[35] Nor is it necessary to prove that the conspirators all reached agreement at the same moment or have been in direct communication with one another. Thus an accused may join a pre-existing conspiracy, making it a “chain conspiracy”; or a number of conspirators may reach agreement with the same person at the hub of a “wheel conspiracy”. All would be part of the same conspiracy provided each agrees to give effect to a common object or purpose.[36] C.3b The essential ingredients of conspiracy to defraud 40. The agreement which is essential to a conspiracy to defraud is an agreement to use dishonest means to defraud someone. As Sir Anthony Mason NPJ (with whom the other members of the Court agreed) held in Mo Yuk Ping v HKSAR,[37] a conspiracy to defraud: “... is constituted by becoming a party to an agreement with another or others to use dishonest means: (a) with the purpose of causing economic loss to, or putting at risk the economic interests of, another; or (b) with the realization that the use of those means may cause such loss or put such interests at risk. The offence extends also to cases in which the dishonest means cause a person to act contrary to his public duty ...” 41. Turning to the mental element, his Lordship endorsed the statement of Lord Goff of Chieveley in Wai Yu-Tsang v The Queen,[38] to the following effect: “The question whether particular facts reveal a conspiracy to defraud depends upon what the conspirators have dishonestly agreed to do, and in particular whether they have agreed to practise a fraud on somebody. For this purpose it is enough for example that, as in Reg v Allsop [(1976) 64 Cr App R 29] and in the present case, the conspirators have dishonestly agreed to bring about a state of affairs which they realise will or may deceive the victim into so acting, or failing to act, that he will suffer economic loss or his economic interests will be put at risk.” 42. In HKSAR v Chan Wai Yip,[39] applying Mo Yuk Ping, Sir Anthony Mason NPJ stated: “In Mo Yuk Ping v HKSAR, this Court held that the essence of the offence of conspiracy to defraud was becoming a party to an agreement with another or others to use dishonest means (i) with the purpose of causing economic loss to, or putting at risk the economic interests of another; or (ii) with the realisation that the use of those means might cause such loss or put such interests at risk. In stating the elements of conspiracy to defraud in this way, this Court adopted earlier formulations of the offence in both the House of Lords (Welham v DPP [1961] AC 103; R v Scott [1975] AC 819) and the Privy Council on appeal from Hong Kong (Wai Yu Tsang v R [1992] 1 AC 269).” 43. It is therefore authoritatively established that the essence of a conspiracy to defraud is an agreement to use dishonest means by which the co-conspirators intend to bring about a state of affairs intending or in the realisation that someone will be defrauded either by having their economic interests injured or imperilled or by being deflected from performing their public duty. C.3c Object or purpose 44. In laying down the essential ingredients of conspiracy to defraud, Sir Anthony Mason NPJ emphasised the importance of distinguishing between the agreed object or purpose of the conspiracy and the dishonest means intended to be used. He endorsed Viscount Dilhorne’s statement that: “One must not confuse the object of a conspiracy with the means by which it is intended to be carried out.”[40] 45. This is important because (subject to Section 3e below) it is the object of the conspiracy which forms the essential ingredient of the offence upon which the co-conspirators must find consensus. 46. As Brennan J noted in Gerakiteys v The Queen, it is “the object or purpose of each conspiracy which gives the conspiracy its identity”.[41] And as Deane J said in the same case: “The gist of the crime of conspiracy lies in the making of the relevant agreement. It is the subject-matter and purpose of the agreement which determines whether it is criminally unlawful.”[42] 47. An important reason why an agreement to use unspecified dishonest means to defraud the victim is sufficient to constitute the offence is that co-conspirators, while agreeing on a common fraudulent purpose, may not yet have identified precisely what steps will be taken, and by which of them, to bring about the intended state of affairs. As Jordan CJ pointed out in R v Ongley,[43] the conspirators may not have identified their intended victims when they began their project, adding: “The fact that the evidence suggests that the accused probably met together from time to time to discuss, arrange and put in action methods for achieving their common purpose would be in no way inconsistent with their being guilty only of a single conspiracy, which had been arranged in general terms, and with their later agreements being merely directed to providing machinery for putting that conspiracy into operation.” 48. And as Thomas LJ explained in R v K,[44] “... if the conspirators agreed to make dishonest representations about the company to induce investors to buy shares, that was sufficient to constitute a certain agreement; it was not necessary that the conspirators agreed more specifically on the misrepresentations that were to be made; the precise nature of the representations to be made or made do not, in contrast to the statutory offence considered in Brown [(1984) 79 Cr App R 115], constitute ingredients of the offence of conspiracy to defraud.” C.3d Overt acts and particulars 49. It is accordingly essential not to confuse the facts and matters provided by way of particulars with the essential constituent elements of the conspiracy alleged. Particulars set out in the indictment or separately given by the prosecution function to inform the accused of the case they have to meet.[45] Particulars are frequently given of overt acts allegedly performed by the accused which the prosecution relies upon to invite the jury to infer the existence of the prior conspiratorial agreement and in particular, that it involved an agreement to use dishonest means. Subject to what is said below, such particulars should not be confused with the co-conspirators’ common purpose which forms the essence of the alleged conspiracy to defraud. 50. Thus, as to the relevance of overt acts, the Court in R v Ongley commented as follows: “If it is established that the accused did things which indicate that they were acting in concert to achieve a common purpose, this supplies all the evidence that is required to establish that they had agreed to achieve that purpose. Indeed, in a prosecution for conspiracy it is unusual for any other evidence of agreement to be tendered than is supplied by evidence of the respective overt acts.”[46] 51. In R v Hancock,[47] Stuart-Smith LJ explained the role of particulars and how they differ from the essential ingredients of the offence: “Since the case of Landy (1981) 72 Cr App R 237; [1981] 1 WLR 355, in a case where conspiracy to defraud is alleged, the Crown are required to set out sufficient particulars of the offence to enable the defence and the judge to know precisely, and on the face of the indictment itself, the nature of the prosecution case and to stop the prosecution shifting their ground during the course of the case. But simply because particulars of an offence are given does not mean that those particulars are an essential ingredient of the offence. In a case such as this the particulars do no more than specify the nature of the case the prosecution seek to prove and the principal overt acts upon which they rely to invite the jury to infer that there was a dishonest agreement and that a particular defendant was a party to it.” C.3e When dishonest means become a constituent ingredient of the conspiracy 52. It does not follow from the above, however, that the dishonest means agreed or used by the conspirators can never become a constituent ingredient of the conspiracy and need never be proven as such. 53. As mentioned, a conspiracy is capable of evolving. Specific dishonest means may be agreed upon by the conspirators after the original conspiratorial agreement was first made. New co-conspirators may join in the conspiracy. The same set of facts may, as a matter of law, give rise to more than one conspiracy, or several sub-conspiracies under an overarching conspiracy. Whilst the essence of a conspiracy is the making of the prohibited agreement such that, strictly speaking, the offence is committed when the agreement is made, the law regards a conspiracy as a continuing offence beginning when the conspiratorial agreement was first made and continuing until it ended.[48] Very often (and the present case is one such example), a conspiracy is charged as a continuing offence over a specified period, instead of an offence on a specified (or an unknown) day. During the continuation of the conspiratorial agreement, new or additional specific dishonest means may be agreed upon by the conspirators, new co-conspirators may join in the conspiracy, and so forth. From a prosecution point of view, in a complicated case, the permutations and combinations can be many. It is important for the prosecution to ascertain the facts carefully and decide what conspiracy/conspiracies they wish to charge, and to plead and make clear their case accordingly. For at trial, the prosecution will be required to prove by evidence the conspiratorial agreement(s) they have chosen to indict. 54. In a prosecution case where it is alleged that some specific dishonest means have been agreed upon as part of the conspiratorial agreement indicted, the indictment and the particulars must clearly inform the court and the defendants what the agreed dishonest means are said to be. The agreed dishonest means so pleaded constitute an essential constituent element of the conspiratorial agreement alleged and must be proved as part of the agreed object of the conspiracy by the prosecution beyond reasonable doubt. C.3f How confusion may arise 55. It is here that the situation becomes complicated by a matter that has already been mentioned, namely the general requirement on the prosecution to reasonably inform the court and the defendants in the indictment and the particulars what the prosecution case is. As explained, the purpose of the requirement is to enable the defendants and the court to know precisely the nature of the prosecution case, and to stop the prosecution shifting their ground during the course of the trial without the leave of the trial judge and the making of an amendment.[49] In other words, even in a prosecution case where it is not alleged that there has been any agreed dishonest means as part of the conspiratorial agreement charged, this requirement would still mean that in many if not all cases, the prosecution have to set out the dishonest means that have actually been used to implement the agreement, which the prosecution will rely on as proof, not of a constituent element of the agreement alleged, but of the existence of the agreement itself. These particulars of dishonest means are the overt acts relied on to infer the existence of the conspiratorial agreement. However, unlike the case discussed in the preceding paragraph, a failure to prove any one or more of these overt acts beyond reasonable doubt may or may not be fatal to the prosecution case. Everything still depends on the totality of the evidence adduced at trial. 56. It is not difficult to see that these two superficially similar, but fundamentally different, situations may give rise to confusion, particularly when the indictment is not sufficiently carefully drafted and clarification is not provided in the prosecution’s opening speech. This important distinction and the potential confusion are borne out by the warning given in Archbold 2019, paragraph 33-49, regarding the drafting of an indictment: “Care should be taken to differentiate between particulars which define the agreement which the Crown seek to prove, as in Bennett, and those which set out overt acts from which the Crown invite the jury to infer the existence of an agreement. Some counsel have begun to limit the particulars of offence to the terms of the agreement to be proved, followed by voluntary particulars which set out the principal overt acts relied upon. This practice, it is submitted, has the advantage of clarify and precision and was approved by the Court of Appeal in R. v K., above.” (emphasis added) 57. Commenting on an identical passage appearing in Archbold 2005, Professor Ormerod explained (in his commentary to the report of R v K in the Criminal Law Review – see below):[50] “Whether the particulars are elevated to the status of ingredients depends on the charge involved. Thus, in cases such as Brown where the allegation was of fraudulently inducing investors, particulars setting out the false representations form part of that ingredient. The same would be true of a deception offence. With conspiracy, the essential ingredient of the offence is the agreement. If the particulars provide detail of the manner in which the agreement was implemented they do not form part of the ingredients of the offence. There may of course be cases where the particulars relate not to the implementation, but to the form of the agreement itself, in which case they become part of the ingredients of the offence alleged and must be proved by the prosecution (see, e.g. Bennett). This method of indicting conspiracy to defraud, distinguishing between the essential ingredient of the agreement and the other particulars of implementation from which jurors are invited to infer dishonesty, etc., is commended by Archbold (2005) 34-42. The court’s per curiam observations endorse the approach.” (emphasis added) 58. In R v Bennett,[51] the indictment alleged an agreement with three specific stages of involvement by the conspirators: “… to defraud computer companies by: (i) dishonestly obtaining the account details of credit card holders; (ii) falsely purporting to be in possession of those details; and (iii) dishonestly using such details to obtain goods and services.” 59. That being the conspiratorial agreement alleged in the indictment such that the dishonest means (ie (i) to (iii)) actually formed part of the conspiratorial agreement itself, the English Court of Appeal held it was incumbent upon the prosecution to prove that the conspirators agreed to the dishonest means as part of their burden to prove the existence of the alleged conspiratorial agreement. 60. Likewise, in HKSAR v Cheng Chee Tock Theodore (No 2),[52] the particulars of the charge alleged that the conspirators: “ … conspired together … to defraud the Board of Directors and shareholders of CYF by dishonestly: (i) falsely concealing from CYF that [Cheng, one of the conspirators] had a beneficiary [sic] or financial interest in a property … ; and (ii) falsely representing to CYF that the acquisition of the said property by CYF did not constitute a connected transaction for the purpose of [the Listing Rules], thereby inducing CYF to complete the purchase of said property without convening the required company meeting(s) or complying with the Listing Rules.”[53] 61. This Court, reading the matters alleged in (i) and (ii) as the agreed dishonest means under the conspiratorial agreement alleged, held that: “[t]he prosecution was therefore alleging and was required to prove beyond reasonable doubt that: (a) Cheng and his co-conspirators conspired to effect a connected transaction constituted by CYF's acquisition of the 17/F Property; (b) They knew that it was a connected transaction but conspired falsely to represent to CYF that it was not; (c) Cheng had a beneficial or financial interest in the 17/F Property but conspired with his co-conspirators falsely to conceal that fact; and (d) that by such concealment and false representation, Cheng and his co-conspirators induced CYF to complete the acquisition without the necessary company resolutions and without complying with the relevant Listing Rules.”[54] 62. In other words, the Court required the prosecution to prove the agreed dishonest means (i) and (ii) as part of their proof of the alleged conspiratorial agreement. As they failed to prove that the transaction was a “connected transaction” within the meaning of the Listing Rules (ie dishonest means (ii)), the prosecution failed to prove the conspiratorial agreement charged and the appellant’s conviction was quashed. 63. By way of contrast, in R v K, the English Court of Appeal, distinguishing Bennett, was able to say, by looking at the indictment in that case, that the conspiratorial agreement indicted there did not include as one of its constituent elements the dishonest means that were actually employed to carry out the conspiratorial agreement to achieve the prohibited object agreed. The dishonest means particularised in the indictment were only the overt acts based on which the prosecution sought to prove the existence of the conspiracy charged. They were supplied to give reasonable information to the defendants on what the prosecution case was and how the prosecution intended to prove the existence of the conspiratorial agreement. 64. Explaining the crucial distinction under discussion, Thomas LJ (as he then was) observed: “… we agree with the editors of Archbold that much greater care needs to be taken in framing the indictment and especially in the definition of the agreement alleged. There must be a clear distinction between the agreement alleged and the reasonable information given in respect of it. ... In our view therefore, the indictment should identify the agreement alleged with the specificity necessary in the circumstances of each case; if the agreement alleged is complex, then details of that may be needed and those details will as in Bennett form part of what must be proved. If this course is followed, it should then be clear what the prosecution must prove and the matters on which the jury must be unanimous: see Bennett.”[55] (emphasis added) This is to be contrasted with the situation where the particulars are simply provided to give reasonable information in respect of the agreement alleged. As Thomas LJ went on to explain: “Further particulars should be given where it is necessary for the defendants to have further general information as to the nature of the charge and for the other purposes identified by Lawton L.J. in Landy. Such further particulars form no part of the ingredients of the offence and on these the jury do not have to be unanimous, as this court correctly decided in Hancock.”[56] (emphasis added) C.3g When a count in an indictment is duplicitous 65. The importance of the rule against duplicity is not in doubt. In particular, where there is a concealed duplicity of charges the accused may be convicted despite lack among jurors of the necessary unanimity, or of the permitted degree of sufficient unanimity. In a conspiracy case some jurors may convict on conspiracy A not B, and others on B not A. 66. Normally, such duplicity only arises in cases where two separate conspiratorial objects or purposes are charged in a single count. The rule was elucidated by Jordan CJ in R v Ongley,[57] as follows (with footnotes omitted): “A count in conspiracy must comply with the general rule of charging one offence only … Where a count in conspiracy charges, as it should do, only one conspiracy to effect some one or more improper purposes, the only issue before the jury under that count is whether all or any of the accused are guilty of the conspiracy alleged. The jury may find that all or any two or more of the accused are guilty of a conspiracy to effect all the improper purposes alleged … , or to effect some only of the improper purposes alleged … but they cannot find some of the accused guilty of a conspiracy to effect only some of the improper purposes and other of the accused guilty only of a conspiracy to effect only other of the improper purposes; because to do so would be to find the existence of two conspiracies under a count which charges one only; and this, the law does not allow …” 67. The focus was thus on a single count alleging two separate conspiracies by reference to two separate improper purposes. 68. In R v K,[58] the duplicity argument failed because it had wrongly concentrated on the particulars rather than the essential ingredients, in particular, the object or purpose, of the conspiracy: “Bearing in mind the clear distinction which must be drawn between the ingredients of the offence and the particulars, we are satisfied that the agreement alleged in this indictment was that spelt out by the judge in his summing up. It was an agreement for a single specified purpose—to defraud potential shareholders to part with their money by making representations about the company that a certain state of affairs existed whereas in truth it did not. Particulars (a)–(l) were given only to provide reasonable information as to the nature of the charge and as to the principal matters upon which the prosecution would invite the jury to infer that there was an agreement to defraud and that each of the defendants was party to it; the particulars did not purport to define the agreement.” C.3h Complications when a conspiracy with agreed specific dishonest means is charged 69. However, in a case where the prosecution chooses to include some specific dishonest means in the indictment not as overt acts, but as agreed dishonest means forming part of the conspiratorial agreement indicted, those agreed dishonest means, together with the object or purpose of the conspiracy, would give the conspiracy its identity. If in a single count of conspiracy, more than one agreed dishonest means are alleged and they are alleged in the alternative, and if there are more than two conspirators alleged to have been involved, a duplicity problem will potentially arise, unless appropriate directions are given by the trial judge. D. Is there duplicity in the present case? D1. The indictment 70. In the present case, it is therefore important to determine whether the dishonest means pleaded in Particulars (a) to (e) were meant and understood to be simply overt acts proving or evidencing the existence of the conspiratorial agreements alleged under Count 1 and Count 2, or whether they were said to be the dishonest means, forming part of the respective conspiratorial agreements, agreed upon by the conspirators concerned. 71. Turning first to the indictment, the particulars of Count 1 say: “[A1, A2 and A3], between the 7th day of May 2009 and the 8th day of September 2009, … conspired together to defraud [SEHK] by dishonestly: (a) falsely representing … ; (b) falsely representing … ; (c) concealing or failing to declare … ; (d) falsely representing … ; (e) concealing or failing to disclose; (f) causing SEHK to allow the publication … of the announcement ... and the circular … which contained the false representations as aforesaid.” 72. Significantly, the particulars do not say, for instance, that the conspirators conspired to defraud the SEHK by dishonest means, as evidenced by Particulars (a)-(e). Rather, the use of the words “conspired together to defraud the [SEHK] by dishonestly: [(a)-(e)]”[59] suggests strongly that the dishonest means particularised, that is, (a) to (e), were the agreed dishonest means forming part of the conspiratorial agreements indicted. 73. The particulars of Count 2 follow a similar format and employ similar wording. 74. In fact, the pleading formula used in the particulars in the present case, that is, the defendants “conspired together to defraud [the victim] by dishonestly [(a), (b), (c) etc] …”, is similar to that used in Bennett and Cheng Chee Tock Theodore (No 2). The particulars strongly suggest that in the present case, Particulars (a) to (e) were the agreed dishonest means forming part of the conspiratorial agreement indicted under each count. D.2 The written directions 75. This reading is strongly supported by the Judge’s written directions, agreed to by all counsel, given to the jury, as part of her summing-up. Unfortunately, the importance of the written directions to the issue under discussion has apparently not been fully appreciated by the Court of Appeal. The written directions read: “1. Before you can convict any of the 1st, 2nd or 3rd Accused of the offence of conspiracy to defraud in counts 1 and 2, you must be sure, separately in respect of each count and each Accused that: (i) … (ii) … (iii) the agreement was to defraud, that is, they had agreed to use dishonest means to bring about a state of affairs which they intended to, or realized might, cause the party or one of the parties named in the count, into acting, or failing to act, (a) for count 1 so that the named party (the SEHK), would perform an act contrary to its public duty; and (b) for count 2 so that the named party/parties (CJHM (462), and/or its existing shareholders) would suffer economic loss, or so that its or their economic interests would be put at risk. The Dishonest Means Alleged 2. In this case, the prosecution allege that the 1st and the 2nd Accused had agreed to use the dishonest means, that is, the false representations (statements) and the concealment particularized in paragraphs (a) to (c) in each count of the conspiracy to bring about the state of affairs or the result in paragraph (f) for count 1; and the results in paragraphs (f) and (g) for count 2. 3. In addition, the prosecution allege that the 1st, 2nd and 3rd Accused had agreed to use the dishonest means, that is, the false representation and the concealment particularized in paragraphs (d) and (e) in each count of the conspiracy to bring about the state of affairs or the result in paragraph (f) for count 1; and the results in paragraphs (f) and (g) for count 2. …” (emphasis added) 76. It can be clearly seen from those paragraphs that as the Judge understood it, which understanding was shared by both the prosecution and the defence, the prosecution’s allegation was that “the 1st and the 2nd Accused had agreed to use the dishonest means, that is, [(a) to (c)]”[60] to achieve (f) for Count 1; and (f) and (g) for Count 2. Likewise, it was the prosecution case that “the 1st, 2nd and 3rd Accused had agreed to use the dishonest means, that is, [(d) and (e)]”[61] to achieve (f) for Count 1; and (f) and (g) for Count 2. 77. We are unable to read the written directions as directions treating Particulars (a) to (e) as merely overt acts proving or evidencing the existence of the conspiratorial agreements to defraud the SEHK and 462 and its shareholders respectively. The caption preceding paragraphs 2 to 6 of the written directions, namely, “The Dishonest Means Alleged”, and the wording of paragraphs 2 and 3, are simply too clear to admit of a different reading such that Particulars (a) to (e) may merely be treated as overt acts. D.3 The Kevin Brown direction 78. The giving of a Kevin Brown direction[62] by the Judge in paragraphs 4 to 6 of the written directions as well as in her oral summing-up[63] reinforces this reading of her directions. Paragraphs 4 to 6 read: “4. It is not necessary for the prosecution to prove that all these representations/concealments were made before you may convict. It is sufficient for the prosecution to prove so that you are sure one or more of these representations/concealments was/were made and that the representation or representations was/were false. 5. However, there should be at least one of such representations/concealments upon which you are agreed to have been a false representation or to have been a concealment, and one which you are sure that at least two of the Accused knew of its existence, its falsity and intended that such a false representation be made or that there be such a concealment pursuant to the conspiracy to which they were parties before you may convict. 6. Further, although one conspiracy is charged against all the three Accused, particulars (a) to (c) only concern the 1st and the 2nd Accused whereas particulars (d) and (e) concern all the three Accused. Therefore, if you were to find that the conspiracy is proved based on any of the particulars (a) to (c) and not on any one of the particulars (d) and (e), you must acquit the 3rd Accused. You may convict the 3rd Accused only if you were to find that either particular (d) or particular (e) is proved and that all the other elements of the conspiracy are also proved.” 79. A Kevin Brown direction is only required in relation to an ingredient of the offence in question where the indictment or charge specifies a number of matters as such ingredients, any of which would be sufficient. A Kevin Brown direction is meant to tell the jury that they need not be satisfied with all the matters specified and relied on to prove the offence in question and that any one of them will do, but that in doing so, the jury as a whole must be agreed on at least one and the same ingredient.[64] The giving of a Kevin Brown direction by the Judge in the present case clearly indicates that her Ladyship, with the agreement of all counsel who were shown the directions in draft beforehand, regarded Particulars (a) to (e) not as mere overt acts but as matters specified and relied on by the prosecution to prove an essential ingredient of the offence and so part of the conspiratorial agreement, and that it would be sufficient for any of those matters to be proved. In other words, the Judge and counsel all took the view that Particulars (a), (b), (c), (d) and/or (e) formed part of the conspiratorial agreement alleged by the prosecution under each count. 80. Had the Judge and counsel regarded Particulars (a) to (e) as simply overt acts, the giving of a Kevin Brown direction would have been totally uncalled for and indeed wrong. 81. With respect, we do not agree with the Court of Appeal that the Judge’s giving of a Kevin Brown direction can simply be discounted as something unnecessary.[65] Since the Judge and counsel all regarded the particulars as forming part of the conspiratorial agreement, and since it was the prosecution case that they were alternative particulars such that proof of any one of them would prove the conspiratorial agreement alleged, in order to do justice to the prosecution case, it was necessary to give a Kevin Brown direction so that the jury would not erroneously believe that they had to find all the particulars proven before they could convict. 82. For this reason also, we are unable to agree with the Court of Appeal’s observation that the Judge was being “favourable” to the appellants, particularly towards A3, in her summing-up, when she directed that if the jury were sure about Particulars (a) to (c) but not (d) to (e), they must acquit A3. Rather, the Judge’s direction followed logically from what she (and counsel) regarded as the true case of the prosecution, that is, that the dishonest means particularised under (a) to (e), whether together or in the alternative to each other, formed part of the conspiratorial agreement alleged against the three appellants under each count. The Judge was not being generous to the appellants but was simply doing justice to the prosecution case as presented at trial. D.4 The prosecution’s speeches and the Judge’s summing-up 83. Nowhere did either the prosecution or the Judge refer to Particulars (a) to (e) as mere overt acts providing evidence on which the jury might infer the existence of the alleged conspiratorial agreement. Had the Judge so regarded (a) to (e), she would (and should) have instructed the jury to weigh all other evidence together with any of the overt acts that they were sure had been established before finding any of the appellants guilty of the relevant count. Nor should the failure to establish any such particulars (if they were merely overt acts) necessarily be fatal to the prosecution case against any appellants, for after all, by definition, one would have been concerned with the drawing of inferences regarding the existence of the alleged conspiratorial agreement from primary facts, which was an exercise dependent on the evidence available. 84. Instead, paragraphs 4 to 6 of the written directions direct the jury that they may convict A1 and A2 if they find against them on any of (a) to (c), and they may convict any two or more of the three appellants if either (d) or (e) is found proven against the two (or three), provided “that all other elements of the conspiracy are also proved.” This categorical approach to conviction (and acquittal) in paragraphs 4 to 6 can only be explained by the Judge and counsel’s treating the particulars not as particulars of overt acts but as matters forming part of the alleged conspiratorial agreement under each count. 85. This categorical approach is also apparent from the way the Judge summed up the case for the jury. For instance, she said, in relation to Particulars (d) and (e): “Now, if you were sure that at least two of the accused were two parties to the conspiracy involving any one of these particulars, particular (d) or particular (e), and you were sure that all the elements of the offence of conspiracy, which I have explained to you, have been proved then you would have to convict the accused accordingly in relation to that particular and in relation to that conspiracy count which you are considering.”[66] (emphasis added) 86. Contrary to what the Court of Appeal apparently thought,[67] it is not open to us simply to say that on a true analysis of the indictment, Particulars (a) to (e) do not form part of the conspiratorial agreement alleged in Count 1 or Count 2, so that one can ignore what the Judge, with the agreement of counsel, has said in her written directions or the Kevin Brown direction that she has given in her summing-up. Her written directions and her summing-up made clear what the prosecution case as presented at the trial was, and as understood by the Judge and by defence counsel to be. Indeed, the written directions were agreed to by all counsel including prosecution counsel. If that was what prosecution counsel agreed to be their true case and that was indeed the case presented at trial it is not open to this Court, sitting on appeal, to say that on a true legal analysis, the prosecution case was (or more correctly, could or should have been) something else. It is too late to turn back the clock. D.5 The duplicity arguments advanced 87. As mentioned, it is argued that the jury, acting pursuant to the written directions, may well have found A1 and A2 guilty of a conspiracy comprising, as part of the conspiratorial agreement, the agreed dishonest means particularised under (a), (b) and/or (c), and, found A3 together with A2 (or A1) guilty of a different conspiracy, comprising as part of that conspiratorial agreement, the agreed dishonest means particularised under (d) and/or (e). So although all three appellants were found guilty, they may have been found guilty as parties to two separate conspiracies rolled up in a single count, and thus the latent duplicity objection. 88. A variation of the same theme advanced on behalf of A3 at the hearing is this: given the way the jury were directed under the written directions, one cannot exclude the possibility that for A3’s conviction for the conspiracy involving the dishonest means particularised under (d) and (e), some of the jurors may have found him guilty of having conspired with A1, whereas other jurors may have found him guilty of having conspired with A2. There may thus not have been unanimity or an acceptable majority of jurors reaching a verdict of guilty on each count. D.6 Our view 89. If the conspiratorial agreement alleged at trial were simply one to defraud the SEHK (taking Count 1 as an example), without alleging any particular dishonest means as forming part of the conspiratorial agreement, so that Particulars (a) to (e) were simply overt acts based on which the prosecution sought to prove by inference the existence of the agreement, this argument of duplicity would have no substance. For in such a case, the conspiratorial agreement would not comprise any of the overt acts as part of the agreement itself. Convicting the three appellants by reference to different particulars, or different overt acts, would not, therefore, mean that they were convicted of different conspiracies. No duplicity problem would have arisen. However, for the reasons explained above, one has to proceed on the basis that Particulars (a) to (e) were alleged by the prosecution to be the dishonest means agreed to by the conspirators as essential ingredients of the conspiratorial agreement, that is, Particulars (a), (b) and/or (c) by A1 and A2, and (d) and/or (e) by all three appellants. 90. This gave rise to a latent duplicity problem as now relied on by the appellants in the appeal (even though counsel on either side, after seeing the written directions, failed to alert the Judge to the potential problem). This problem cannot, however, be overcome by using the “purpose/object” versus “particulars/essential ingredients” distinction, for the simple reason that under this scenario, the dishonest means pleaded in particulars are not merely particulars of overt acts but actually form an essential part of the conspiratorial agreement alleged. The possibility that the jury may have convicted A1 and A2 under (a), (b) and/or (c), and either A1 or A2 together with A3 under (d) and/or (e), therefore gives rise to the possibility that the three appellants were convicted where two conspiracies with different agreed dishonest means were alleged in a single count. 91. This is fatal to the convictions of all three appellants. As Ms Montgomery QC accepted, without dissent from counsel for the other appellants, the consequence is that a retrial should be ordered. 92. It is unfortunate that the duplicity point was not raised at trial by counsel. Any argument on duplicity should be raised at an early stage, precisely to avoid the outcome as shown by these appeals. E. The misdirection issue 93. This conclusion makes it unnecessary and undesirable to deal with the “misdirection issue” under Question 1 and SGI-1 and SGI-2. F. Outcome 94. For the above reasons, we would order that the convictions of all three appellants be quashed and that there be a re-trial with the prosecution being at liberty to amend the indictment or to prefer a fresh indictment. Mr Justice Fok PJ: 95. I agree with the joint judgment of Mr Justice Ribeiro PJ and Mr Justice Cheung PJ. Mr Justice Gummow NPJ: 96. I agree with the joint judgment of Mr Justice Ribeiro PJ and Mr Justice Cheung PJ. Chief Justice Ma: 97. These appeals are unanimously allowed. We order that the convictions of all three appellants be quashed and that there be a re-trial with the prosecution being at liberty to amend the indictment or to prefer a fresh indictment. Any submissions as to costs should be lodged within 21 days of the date of the handing down of this judgment with any submissions in reply to be lodged within 14 days thereafter. Ms Clare Montgomery QC, Mr Clive Grossman SC and Mr Benson Tsoi, instructed by Boase, Cohen & Collins, for D1/the appellant in FACC 26/2018 (the 1st appellant) Mr Ian Winter QC, Mr Graham Harris SC and Ms Emily Yu, instructed by Haldanes, for D2/the appellant in FACC 27/2018 (the 2nd appellant) Mr Tim Owen QC, Mr David Khosa and Ms Yasmine Zahir, instructed by Hobson & Ma, for D3/the appellant in FACC 28/2018 (the 3rd appellant) Ms Charlotte Draycott SC, on fiat and Ms Joycelyn Ng, SPP (Ag), of the Department of Justice, for the respondent [1] Although the offence of conspiracy is now statutory based (s 159A of the Crimes Ordinance, Cap 200), the offence of conspiracy at common law is preserved (s 159E(2) of the Crimes Ordinance). [2] HCCC 83/2014. [3] Subsequently re-named Natural Dairy (NZ) Holdings Limited. [4] Contrary to ss 25(1) and 25(3) of the Organized and Serious Crimes Ordinance (Cap 455). [5] Lunn VP, Macrae and Pang JJA [2018] HKCA 121 (16 April 2018). [6] Ribeiro, Fok and Cheung PJJ [2018] HKCFA 60 (14 December 2018). [7] Otherwise known as “UBTL”. [8] As Kathy Chan gave evidence as the first Prosecution Witness, she has been referred to throughout at “PW1”. [9] Summing-up 25.4.16, p 72. [10] Named in the CSA referred to above. [11] Admitted Facts §§37-42; Court of Appeal §8. [12] Court of Appeal §7. [13] Admitted Facts §45. [14] Court of Appeal §§291-320. [15] Court of Appeal §301. [16] Court of Appeal §313. [17] Admitted Facts §55. [18] Admitted Facts §§67-69. [19] 462 announcement dated 2 February 2011; Admitted Facts §25. [20] Admitted Facts §69. [21] 462 announcement dated 2 February 2011. [22] Admitted Facts §77. [23] Admitted Facts §78. [24] 462 announcement dated 2 February 2011. [25] Appearing with Mr Clive Grossman SC and Mr Benson Tsoi. [26] Appearing with Mr Graham Harris SC and Ms Emily Yu. [27] Appearing with Mr David Khosa and Ms Yasmine Zahir. [28] Court of Appeal §58. [29] A1’s written case §4.1(3) referring to summing-up 25.4.16, p 60. [30] Written directions on Counts 1 and 2 §5. [31] A1’s written case §4.2. [32] [2005] 1 Cr App R 25 at §35. [33] R v Ongley (1940) 57 WN (NSW) 116 at 117 per Jordan CJ. [34] (1984) 153 CLR 317 at 327. Brennan J dissented on the appropriate order to be made, but not on the general principles. [35] R v Ongley (1940) 57 WN (NSW) 116 at 117. [36] See Gerakiteys v The Queen (1984) 153 CLR 317 at 327-328. See generally, D Ormerod and K Laird, Smith, Hogan, and Ormerod’s Criminal Law (15th Ed, OUP), pp 438-440. [37] (2007) 10 HKCFAR 386 at §40. [38] [1992] 1 AC 269 at 279-280. [39] (2010) 13 HKCFAR 842 at §51. Mo Yuk Ping was also applied by this Court in HKSAR v Cheng Chee Tock Theodore (No 2) (2016) 19 HKCFAR 86. [40] Scott v Metropolitan Police Commissioner [1975] AC 819 at 839; cited in Mo Yuk Ping at §32. [41] (1984) 153 CLR 317 at 329. [42] Ibid, at 334. [43] (1940) 57 WN (NSW) 116 at 118. [44] [2005] 1 Cr App R 25 at §28. [45] Indictment Rules (Cap 221C) r 3(1). [46] (1940) 57 WN (NSW) 116 (CCA) 117. [47] [1996] 2 Cr App R 554 at 559-560. [48] DPP v Doot [1973] AC 807. [49] R v Landy [1981] 1 WLR 355 at 362; Indictment Rules (Cap 221C) r 3(1). [50] [2005] Crim LR 298. [51] (unreported) 6 May 1999. [52] (2016) 19 HKCFAR 86. [53] Ibid, at §13. [54] Ibid, at §14. [55] [2005] 1 Cr App R 25 at §36. [56] Ibid. [57] (1940) 57 WN (NSW) 116 at 117. [58] [2005] 1 Cr App R 25 at §34. [59] Emphasis added. [60] Emphasis added. [61] Emphasis added. [62] R v Kevin Brown (1984) 79 Cr App R 115. [63] Summing-up 25.4.16, pp 59H-61B. [64] R v Kevin Brown at 119. [65] Court of Appeal §71. [66] Summing-up 26.4.16, pp 180T-181E. [67] See Court of Appeal §§ 62-64, 67, 71 and 82. 
Mr Justice Fok PJ: 1. This uncontested appeal concerns an order for costs made by a judge against an unsuccessful defendant in a magistracy appeal. It reinforces the need for courts to have regard to the principles and practice laid down in HKSAR v Chui Shu Shing (2017) 20 HKCFAR 333 before exercising powers to order costs against parties in criminal appeals under the Costs in Criminal Cases Ordinance.[1] 2. On 30 December 2016, the appellant was tried before a magistrate[2] for the offence of careless driving,[3] arising out of a collision between his car and a taxi on Lantau Island. On 17 January 2017, the magistrate convicted the appellant of the offence and, on 2 February 2017, imposed a fine of HK$3,000 by way of sentence. 3. The appellant’s appeal against conviction to the Court of First Instance[4] was heard before Deputy High Court Judge Stanley Chan on 16 June 2017. By his judgment dated 11 October 2017, the Judge dismissed the appeal and upheld the conviction and, notwithstanding the absence of any application for costs by the prosecution (respondent in this appeal), ordered the appellant to pay costs in the sum of HK$2,000 to the respondent. The Judge did so without first affording the appellant an opportunity to be heard as to whether a costs order should be made. 4. In the concluding paragraph ([12]) of his judgment, the Judge said: “I find that the appellant does not have a good chance of success in this appeal hearing. I have considered the general principles set out in section 15 of the Costs in Criminal Cases Ordinance and I firmly believe that the court has to take a clear stance regarding the spirit and intent of section 13B in order to avoid abuse. The appellant is represented by a [sic] counsel, and naturally (he) is aware of all the advice regarding the liabilities involved in the present appeal, which naturally includes the costs. However, having considered the facts of the present case, I assess the costs summarily and decide to order that the appellant is to pay the respondent the costs of $2,000. I find that this is merely a very small percentage of the costs incurred, but it has its substantive significance.”[5] 5. The reference to section 13B of the Costs in Criminal Cases Ordinance[6] may be a typographical error since that section deals with the prosecution’s costs on an unsuccessful application by a defendant for a certificate under section 32(2) of the Hong Kong Court of Final Appeal Ordinance,[7] which was not the situation before the Judge since the appellant had not applied for such a certificate from him. The correct provision dealing with prosecution costs on an unmeritorious magistracy appeal by a defendant is instead section 13 of the Costs in Criminal Cases Ordinance. However, for the purposes of this appeal, nothing turns on this error, whether typographical or otherwise. 6. On 5 December 2017, pursuant to the costs order, the appellant paid the sum of HK$2,000 to the respondent. 7. On 10 April 2018, the appellant applied for leave to appeal to the Court of Final Appeal against the Judge’s judgment dismissing his appeal and upholding his conviction and also against the costs order. The respondent indicated to the Appeal Committee that it was prepared to concede to an order granting leave to appeal on the basis that the costs order was wrongly made and that the proposed appeal against that order should be allowed. 8. On 6 November 2018, the Appeal Committee dismissed the application for leave to appeal against conviction but granted the appellant leave to appeal against the costs order on the ground that it was reasonably arguable that substantial and grave injustice had been done on the basis that the costs order was made contrary to the principles laid down in HKSAR v Chui Shu Shing (2017) 20 HKCFAR 333.[8] At the same time, the Appeal Committee directed the parties to file a joint case in respect of the appeal in order for the Court to determine, in accordance with the procedure laid down in HKSAR v Shum Wan Foon (2014) 17 HKCFAR 303, whether it would dispose of the appeal on the papers and without an oral hearing. 9. The parties duly filed a Joint Case dated 11 December 2018 on the question of the costs order. In it, both parties submit that, in making the costs order against the appellant when he handed down judgment without giving the appellant any prior notice or warning and without giving him an opportunity to be heard on the question of costs, the Judge departed from accepted norms and substantial and grave injustice was done to the appellant. 10. The relevant principles governing the making of a costs order in magistracy appeals are those set out in the judgment of Ribeiro PJ in HKSAR v Chui Shu Shing at paragraphs [9] to [11]: “9. In order that the right of appeal granted by s.113 be given full effect, the s.13 requirement that the appeal was ‘without merit’ must be understood to mean that the appeal was not reasonably arguable. It must also be noted that such a finding, though necessary, is not sufficient. The judge must go on to apply the principles laid down by s.15. This obliges the court to ensure that any sum awarded is not punitive but only compensatory of properly incurred expenses. It must also ensure that a contemplated award would be ‘just and reasonable’. 10. For the court to address these issues, it is obviously necessary for it to have some idea of the magnitude of costs incurred by the prosecutor and in practice, these issues are likely to arise only where the prosecution applies for costs. It would then be necessary for the appellant to be afforded a proper opportunity to be heard as to whether the order should be made. 11. As the Court of Appeal recognized in HKSAR v Kissel (No 2), the Court must have regard to the financial means of the defendant and satisfy itself that any costs order is one that the defendant is able to pay. It would not be just and reasonable to order the appellant to pay costs beyond his or her means. And such an award might well operate punitively.” (Footnotes omitted) 11. Here, it is unquestionably the case that the Judge did not adhere to those principles. This is somewhat surprising since the judgment in HKSAR v Chui Shu Shing was handed down three months before the Judge’s judgment in the present case and HKSAR v Chui Shu Shing was an appeal to this Court from the same Judge. 12. We are therefore satisfied that the appeal should be allowed and the costs order in question quashed. 13. The parties are agreed that the HK$2,000 paid by the appellant to the respondent on 5 December 2017 pursuant to the costs order should be refunded to him and we so order. 14. Finally, the parties being in agreement as to the costs of this appeal, we make an order that: (1) The costs of this appeal and the application for leave to appeal on the issue of the costs order be awarded to the appellant, such costs not to exceed the contribution payable by the appellant to the Director of Legal Aid; and (2) The appellant’s own costs be taxed in accordance with the Legal Aid Regulations.[9] Joint Written Submissions by: Mr Caesar Lo, instructed by Cheung, Chan & Chung, assigned by the Director of Legal Aid, for the Appellant Ms Vinci Lam, SADPP and Ms Human Lam, PP, of the Department of Justice, for the Respondent [1] (Cap.492). [2] In TWCC 2455/2016. [3] Contrary to section 38(1) of the Road Traffic Ordinance (Cap.374). [4] In HCMA 76/2017. [5] The judgment was written in Chinese and this citation is taken from the Court Translator’s translation. [6] (Cap.492). [7] (Cap.484). [8] FAMC 21/2018, [2018] HKCFA 47, Reasons for Determination dated 9 November 2018. [9] (Cap.91A). 
Mr Justice Fok PJ: 1. It is self-evident that undertaking certain activities in the proximity of electricity supply cables, be they overhead or underground, is inherently dangerous and poses risks of personal injury or death, damage to property and disruption of the electricity supply. In this case, employees of the respondent company hammered iron rods into the ground and, in doing so, damaged an underground 132kV electricity cable. The respondent was charged with breaching the Electricity Supply Lines (Protection) Regulation[1] and this appeal, arising from the Judge’s quashing of the respondent’s conviction by the Magistrate, raised questions of statutory construction of certain provisions in the Regulation. 2. At the conclusion of the hearing, the Court allowed the appeal and made costs orders as set out below, indicating that our reasons would be handed down in due course. These are our reasons. A. The facts 3. The respondent was engaged by the MTR Corporation Limited (MTR) under a contract to perform “Ground Investigation for Slope Improvement Works” at a site adjacent to the East Rail Line near Tai Po Tau. Before beginning the work under the contract, the respondent wrote to CLP Power Hong Kong Limited (CLP) who replied enclosing a scale drawing of the locations of their underground electricity cables in the area where the work was to be carried out. 4. On 10 January 2011, work commenced with the setting up of a boundary for the works site prior to the commencement of underground electricity cable detection work by a competent person[2] which was due to take place the following day. To establish this boundary, which ran alongside the railway track, two workers of the respondent drove iron rods into the ground around the designated area to which orange nets were then attached. They did so manually using hammers but without the use of machinery of any kind. Each of the rods was between 1 to 2 metres long and was driven into the ground to a depth of about one foot (200-300 mm). The area to be enclosed was about 100 metres long on each side. On the day in question, the concrete cover protecting an underground electricity cable and the 132kV cable itself, which was buried 500 to 600 mm below ground, were damaged by one of the iron rods driven into the ground by the respondent’s workers. As a result of the damage caused to the underground cable, there was a voltage drop of 96% in the supply of electricity at 2.26pm on 10 January 2011. There was evidence, on inspection, of burn marks on the cable and its concrete cover. Fortunately, no one was injured. B. The proceedings below 5. As a result of the incident, the respondent was charged with the offence of “carrying out works below ground level in the vicinity of an underground electricity cable without having taken all reasonable steps to ascertain its existence”, contrary to sections 10(1)(a) and 17(3) of the Regulation.[3] The particulars of the offence read as follows: “Information has been laid THAT YOU on or about 10/1/2011, at the location of a works site at the slope of MTR East Rail Line opposite to Tai Po Tau pumping station, Tai Po Road-Tai Wo Section, New Territories, carried out works in the vicinity of an underground electricity cable which were below ground level and, before the works were begun failed to take all reasonable steps to ascertain the existence within the proposed works site and its vicinity of any such underground electricity cable and its alignment and depth.” 6. On 19 January 2012, after trial,[4] the Magistrate[5] convicted the respondent of the offence charged. He addressed the issue of whether the act of the respondent fell within the definition of “ramming” in section 2 of the Regulation[6] and concluded that it did.[7] He then addressed the issue of whether the act of the respondent was work necessary to ascertain the alignment and depth of an underground electricity cable, and thus excluded work under section 2 of the Regulation, and concluded that it was not.[8] After addressing certain other issues not relevant to this appeal, the Magistrate found the respondent guilty of the offence and imposed a fine of HK$7,500. 7. The respondent appealed against the conviction.[9] By her judgment dated 20 May 2013, the Judge[10] allowed the appeal and quashed the respondent’s conviction and set aside the sentence. She held that (i) hammering by hand was not within the definition of works, specifically “ramming”, in section 2 of the Regulation,[11] and (ii) that the respondent’s conduct fell within the exclusion in the definition of works in section 2 of the Regulation.[12] C. The certified questions 8. By Notice of Motion dated 28 May 2013, the prosecution applied to the Judge for a certificate that two points of law of great and general importance were involved in the appeal. That application was refused by the Judge on 7 June 2013 and an application for leave to appeal was therefore made to the Appeal Committee of this Court which, on 28 January 2014, granted leave to appeal on the following two points of law of great and general importance, namely: “First question: Whether or not the word ‘ramming’ in section 2(a)(vi) of the Electricity Supply Lines (Protection) Regulation, Cap.406H, should be construed as ‘ramming involving at least the use of machinery’ and ‘excluding the act of hammering by hand’? Second question: What types of works fall within the exemption of ‘works necessary to ascertain the alignment and depth of an underground electricity cable’ as provided by section 2 of the Electricity Supply Lines (Protection) Regulation, Cap.406H – in particular, whether or not works that are preparatory in nature and not being part of the actual works to ascertain the alignment and depth of an underground electricity cable fall within the said exemption?” D. The statutory provisions 9. The Regulation was made under section 59 of the Electricity Ordinance[13]. 10. Section 10 of the Regulation, under the heading “Works in Vicinity of Electricity Supply Lines”, provides: “10. Requirements relating to works in vicinity of electricity supply lines (1) A person shall not – (a) carry out or cause or permit another to carry out in the vicinity of an underground electricity cable any works which are below ground level; or (b) carry out or cause or permit another to carry out in the vicinity of an overhead electricity line works of any kind, unless before the works are begun all reasonable steps have been taken to ascertain the existence within the proposed works site and its vicinity of any such underground electricity cable and its alignment and depth or of any such overhead electricity line and its alignment, distance from the ground and voltage, as the case may be.” 11. Subject to the exemption provided for, section 10(1)(a) prohibits the carrying out of any works which are below ground level in the vicinity of an underground electricity cable[14] and section 10(1)(b) prohibits the carrying out of works of any kind in the vicinity of an overhead electricity line.[15] The exemption applies when, before the works are begun, all reasonable steps have been taken to ascertain the existence within the proposed works site and its vicinity either of any such underground electricity cable and its alignment and depth, or of any such overhead electricity line and its alignment, distance from the ground and voltage. 12. In relation to the taking of reasonable steps in relation to works in the vicinity of an underground electricity cable, section 10(3) of the Regulation provides that these: “…shall not be regarded as having been taken unless a competent person has undertaken an investigation for the purpose of ascertaining the existence within the proposed works site and its vicinity of any such underground electricity cable and its alignment and depth and has provided a written report of his findings as to those matters.” 13. By section 10(4), the taking of all reasonable steps will be deemed where there has been compliance with the provisions of a code of practice for any of the requirements of sections 10(1)(a) or (b) approved by the Director of Electrical and Mechanical Services.[16] The Director has issued a Code of Practice on Working near Electricity Supply Lines,[17] which provides, among other things, a table defining the distance within which works below ground level will be considered to be “works in the vicinity of U/G cables”.[18] 14. Section 17(3) of the Regulation provides that a person who contravenes any requirement of section 10(1)(a) or (b) commits an offence and is liable to a fine at level 4 and to imprisonment for 6 months. 15. Section 2 of the Regulation defines “works site” as meaning “any area in which works are being carried out” and, crucially for this appeal, “works” as meaning: “(a) any kind of works involving or in connection with – (i) building works within the meaning assigned to that term in section 2(1) of the Buildings Ordinance (Cap 123), and for the purposes of this subparagraph reference in that section to ‘ground investigation in the scheduled areas’ shall be read as a reference to ‘ground investigation’; (ii) the laying out, construction, alteration or repair of any road (whether or not at ground level), footpath, cycle track, pedestrian subway, footbridge, tunnel, airport runway, canal, reservoir, pipeline, railway or tramway; (iii) trench works, including – (A) water mains, storm water drains and sewers; and (B) such works when carried out by or for any public utility; (iv) the extraction of material from land or the seabed; (v) landfill works, river training works, slope works or reclamation works; or (vi) levelling, piling, ramming, dredging, boring, tunnelling or blasting works; (b) the use of any crane or hoist (within the meaning assigned to those terms in regulation 2(1) of the Construction Sites (Safety) Regulations (Cap 59 sub. leg. I)) or other equipment for the purpose of lifting objects; (c) the use of any heavy machinery or other equipment for the purpose of earth moving, but does not include works necessary to ascertain the alignment and depth of an underground electricity cable, nor any electrical work carried out by a registered electrical worker or registered electrical contractor;”. E. The issues in this appeal 16. As will be apparent from the certified questions of law set out above, the issues in this appeal concern the proper construction of: (1) the word “ramming” in section 2(a)(vi) of the Regulation (Issue 1); and (2) the exclusion in section 2 of the Regulation relating to works “necessary to ascertain the alignment and depth of an underground electricity cable” (Issue 2). 17. No issue arises in the present case regarding the exemption in section 10(1)(a), namely that, before the works were begun, all reasonable steps had been taken to ascertain the existence within the proposed works site and its vicinity of any underground electricity cable and its alignment and depth. In other cases under section 10(1) of the Regulation, however, that defence may be raised and, in that event, compliance with the Code of Practice, shall be deemed to constitute the taking of all reasonable steps for the purposes of that requirement.[19] Chapter 3 of the Code of Practice, the introduction to which summarises its contents[20], sets out the reasonable steps for working near underground electricity cables. 18. In the present case, it was never the case of the respondent that it had taken all reasonable steps to ascertain the existence within the proposed works site and its vicinity of any underground electricity cable and its alignment and depth before the iron rods were hammered into the ground. On the evidence, the provisions of section 10(3) of the Regulation plainly could not be shown to have been complied with. Indeed, a competent person had yet to be appointed and the establishment of the boundary of the works was done as a precursor to cable detection work due to take place the following day, so cable detection was not even to begin until after the carrying out of the work subject to the charge in this case. There could therefore be no basis for suggesting that there was compliance with section 10(3) of the Regulation or the Code of Practice by the taking of all reasonable steps as required. 19. As to the approach to statutory construction, it is common ground that the Court should adopt a purposive approach, construing the statutory language having regard to its context and purpose.[21] F. Issue 1: the proper construction of “ramming” 20. For the appellant, Mr William Tam[22] contended that the respondent’s act of hammering by means of a hand-held tool fell within the definition of “ramming” and the use of machinery is not necessary for the act of hammering to fall within that definition. The appellant supported the Magistrate’s construction and contended that the Judge was in error in her construction. 21. For the respondent, Mr Clive Grossman SC[23] contended that, considering the purpose of the legislation and the context of the word, “ramming” involves at least the use of machinery and excludes the act of hammering by hand. The respondent therefore sought to support the Judge’s construction overturning the conclusion of the Magistrate on this issue. (a) The natural and ordinary meaning of “ramming” 22. When it is said that context is the starting point, together with purpose, in statutory interpretation,[24] that is not to say that one puts the words being construed to one side. On the contrary, since contextual and purposive construction is a tool or aid to assist a court in arriving at an interpretation that gives effect to the legislative intention, one must always have regard to the particular words used by the legislature in expressing its will. A court cannot attribute to a statutory provision a meaning which the language of the statute, understood in the light of its context and the statutory purpose, is incapable of bearing.[25] For that reason, one must necessarily look to the statutory language to see what meaning or meanings it is capable of bearing. 23. In the present case, the word “ramming” is one which does not, on its own, provide an indication as to whether it is to be read as requiring the use of machinery or not. It is a general word and its dictionary meaning[26] includes: “To force or drive down or in by heavy blows; to drive (piles, etc.) into the soil in this way”. It also means “To cram, stuff, thrust (a person or thing) into something” and, in this sense, it may be used literally, to refer to a physical act of pushing or thrusting, or figuratively, to refer to a non-physical act of pressing or emphasising (e.g. “ramming a point down someone’s throat”). Therefore, little assistance is to be derived in construing the word “ramming” from what may be said to be its natural or ordinary meaning. 24. That resort to the dictionary meaning of the word is of limited assistance is illustrated by the evidence given by the respondent’s expert witness at trial that, as an engineer, he would regard “ramming” as involving the hoisting of an object by heavy machinery so as to enable it to hit against another object for the purpose of ground penetration and that he believed this meant that heavy machinery of a large scale was inevitable in ramming. The appellant takes issue with the suggestion that “ramming” has a technical meaning but the debate itself demonstrates that one cannot derive an answer to this particular construction issue by reference to the natural or ordinary meaning of “ramming”. 25. What one can conclude from the natural or ordinary meaning is that “ramming” can be an act of forcing or driving an object into something else by means of a tool, such as a hammer. One must then look to context and purpose to determine if the act requires the hammering to be by machine or whether it can be performed by hand. 26. The respondent did not suggest that the word “ramming” be given its natural or ordinary meaning but that, instead, it should be given its technical meaning in the context of construction or engineering practices. As indicated above, reliance is placed on the evidence of its expert witness at trial. However, the Magistrate’s analysis of the respondent’s expert’s evidence at trial[27] indicates that the expert’s conclusion that ramming involved the use of machinery was in the nature of a personal opinion and not supported by any authoritative scientific or other materials. More importantly, the conclusions reached by reference to the contextual and purposive construction of “ramming”, discussed below, do not support a conclusion that the word is used in a technical sense as contended by the respondent and so the expert evidence would not, in any event, be admissible on this issue. (b) The contextual meaning of “ramming” 27. The definition of “works” within the Regulation provides the immediate context of the word “ramming”. An examination of that definition reveals the following significant features that bear on the meaning of the word. 28. First, the definition of “works” is wide and embraces a multitude of different concepts. Paragraph (a) of the definition begins by providing that “works” means “any kind of works involving or in connection with” various matters so the definition is wider than just the specific items identified in sub-paragraphs (i) to (vi) but extends to embrace additional or ancillary acts which are involved or in connection with them. There is nothing to suggest that these additional or ancillary acts have to be mechanical acts or require the use of machinery. 29. Secondly, of the items identified in sub-paragraphs (i) to (vi), at least five are projects or specific tasks or ends of work rather than the means or processes by which those ends are achieved. Thus, sub-paragraph (i) refers to works involving or in connection with “building works” specifically as defined in the Buildings Ordinance.[28] Sub-paragraph (ii) refers to works involving or in connection with the laying out (etc.) of any of various types of route or way. Sub-paragraph (iii) refers to works involving or in connection with trench works, (iv) to extractions from the land or seabed, and (v) to landfill works, river training works, slope works or reclamation works. All these matters are projects or specific tasks or ends of work and, as such, “any kind of works” which involve or are in connection with those projects must, by definition, include a very wide range of possible activities some of which need not involve the use of machinery. Indeed, the reference to “building works” exemplifies this. As defined in the Buildings Ordinance, “building works” include “any kind of building construction, … repairs, demolition, alteration, addition and every kind of building operation, and includes drainage works”. The latter, “drainage works”, is further defined to include “any work connected with the construction, repair, alteration, disconnexion, trapping and ventilation of drains or sewers”. 30. Thirdly, although the acts identified in sub-paragraph (vi) (“levelling, piling, ramming, dredging, boring, tunnelling or blasting works”) are, in contrast, all specific processes or means by which various projects or ends may be achieved, “any kind of works involving or in connection with” those processes is within the definition of “works”. This again suggests a very wide range of possible activities which may fall within the definition. More importantly, though, the various processes listed in sub-paragraph (vi) include means of work which do not necessarily require the use of machinery of any kind. Levelling and tunnelling works, and certainly works in connection with those processes, can be done by hand and blasting works do not necessarily require machinery. 31. Fourthly, paragraphs (b) and (c) of the definition of “works” expressly refer to “the use of any crane or hoist … or other equipment” and “the use of any heavy machinery or other equipment” respectively for the two processes there referred to, namely “lifting objects” and “earth moving”. This demonstrates that where the use of machinery is required for works to come within the definition, the Regulation makes express reference to such machinery. By contrast, in sub-paragraph (vi), there is no express reference to any machinery and this provides a strong indication, as a matter of context, that it is not a necessary part of the processes there enumerated. 32. The respondent contended that the items in sub-paragraph (vi) provide a genus of works, all of which are works of a large magnitude or scale, destructive in nature or involving major disruption to the earth and thus requiring the use of machinery. However, even if some of those processes may involve the use of machinery, it is equally clear that some of them, for example levelling, tunnelling or blasting works, may be carried out without machinery. This undermines the argument that ramming must therefore be construed consistently to mean an act done by machinery. There is simply no relevant genus with which to be consistent. 33. A further contextual point of relevance is the fact that, in Hong Kong, underground electricity cables are not necessarily buried so deeply underground that they cannot be interfered with by the use of hand-held tools or, as a corollary, that only ramming by the use of a machine might interfere with them. Table 2 of the Code of Practice provides information on typical buried depths of underground cables in Hong Kong and these depths range in some cases[29] from as shallow as 450 mm. That being the case, there is all the more reason to include non-mechanical ramming within the definition of “works” since even such ramming could result in contact with underground electricity cables at such shallow depths. (c) The purposive meaning of “ramming” 34. As indicated in the opening paragraph of this judgment, carrying out work in the vicinity of electricity cables is inherently dangerous and may be potentially highly disruptive. It is therefore of little surprise that the long title of the Ordinance includes as one of its objectives “… to provide for measures designed to ensure that activities carried out in the vicinity of electricity supply lines do not prejudice safety or the continuity of the electricity supply”. 35. To this end, section 59(1) of the Ordinance provides that the Chief Executive in Council may make regulations “for the general purposes of this Ordinance including regulations for all or any of the following matters”, including: “(ia) measures designed to ensure that activities performed in the vicinity of electricity supply lines are performed in a manner that minimizes causing an electrical accident or interruption to the supply of electricity;”. 36. The Regulation having been made pursuant to section 59 of the Ordinance, it is abundantly clear that its legislative purpose is safety and the prevention of interruption to the electricity supply. The following aspects of the Regulation firmly reinforce this. (1) The title of the Regulation itself refers to the “protection” of electricity supply lines. (2) The Regulation provides a detailed scheme for the grant by the Director of approval to a person as “a competent person for the purposes of this Regulation”.[30] This involves the person attending and passing a course in locating underground electricity cables, not less than 6 months’ practical experience in locating such cables and competence in locating them.[31] As already pointed out, section 10 of the Regulation sets out requirements relating to works in the vicinity of supply lines. For the purposes of demonstrating that reasonable steps have been taken for the purposes of the exemption in section 10(1), it is necessary that a competent person “has undertaken an investigation for the purpose of ascertaining the existence within the proposed works site and its vicinity of any such underground electricity cable and its alignment and depth and has provided a written report of his findings as to those matters.”[32] In doing so, the competent person is required to “carry out the investigation in a manner that does not cause damage to, or impair the operation of, the underground electricity cable”.[33] (3) The Director is empowered, if he thinks a contravention of a requirement under the Regulation gives rise to “an undue risk of an electrical accident or an interruption to the supply of electricity”, to issue a remedial notice containing instructions as to measures to be taken to remedy the contravention.[34] The Director is also given powers of entry and inspection to premises or places in which works are being undertaken in the vicinity of an electricity supply line “in the interests of safety or of ensuring the continuity of supply of electricity”.[35] 37. The Director is also empowered to approve and issue codes of practice for the purpose of providing practical guidance in respect of any requirement under the Regulation.[36] Reference to the Code of Practice further reinforces the purposes of the legislative provisions of the Regulation. It states, in its Introduction, that: “Incidents involving damage to electricity supply lines can have serious consequences for workers and adversely impact on the community. Various types of works have the potential to cause such damage but adequate precautions can dramatically reduce the risk of an incident. This Code outlines the dangers and provides advice on how to reduce risk. … It deals specifically with risks to persons carrying out works and the necessary precautions needed to reduce the risk of accidents due to damage and unsafe practices.”[37] 38. Given the abundantly clear legislative purpose of the Regulation as a whole and the requirements relating to works in the vicinity of electricity supply lines prescribed in section 10 of the Regulation, the word “ramming” should, so far as permissible as a matter of language, be construed consistently with that purpose and in a manner which will best achieve that purpose. (d) The proper construction of “ramming” in the light of its context and purpose 39. In the light of the above context and legislative purpose of sections 2 and 10(1)(a) of the Regulation, “ramming” should be construed to include penetration of the ground by an object by means of force applied by any means, including a hammer, and not just by means of machinery. It is the penetration of the ground that poses the risk of contact of the penetrating object with an underground electricity cable and it is that activity which, purposively construed, the word “ramming” is intended to regulate. 40. The Judge, however, held: “12. Taking into account the provision of regulation 2 and the nature of the works enumerated therein, I find that “ramming works” mean works involving at least the use of machinery (whether heavy or not). The use of machinery would without doubt likely to cause damage to cables. The purpose of the Regulation was to ensure that activities carried out in the vicinity of electricity supply lines do not prejudice safety or the continuity of the electricity supply. 13. However the act of hammering is obviously different from the use of machinery. Generally speaking, if the works are carried out merely by hammering, the depth of penetration would be limited (in this case about 1 foot or 200-300 mm) and the likelihood of causing damage to underground cables (according to the Code of Practice, buried depth of 132kV cables should be no less than 1 metre) would be relatively low. On the whole, I do not consider the act of hammering iron rods into the ground constitutes “ramming” under the Regulation.” 41. This reasoning cannot, with respect, be accepted. Given the Judge’s acknowledgment of the purpose of the Regulation, there is no good reason to think that the likelihood of damage to electricity cables from ramming works would be limited to such works involving the use of machinery. It is plain that a penetrative object could be forced into the ground by hand or by the use of a non-mechanical tool to a depth at which it could come into contact with an underground electricity cable (as happened here). Although hammering by hand is different from the use of a machine, we do not consider that there is a proper basis for the Judge’s conclusion that, if works are carried out by hammering (by hand), the depth of penetration would be limited and the likelihood of causing damage to underground cables would be relatively low. The fact the iron rods in this case were to be hammered into the ground to a depth of between 200 to 300mm does not preclude their being hammered deeper into the ground either by inadvertence or over-enthusiasm on the part of the workers. Equally, the fact that the guide depth for 132kV cables is 1000mm does not preclude their being found at a shallower depth, as was in fact the case here. In our view, the risk of a hand-hammered iron rod coming into contact with an underground electricity cable is not properly to be characterised as “relatively low”. 42. The respondent contended that it could not have been the legislative intent to apply a meaning to “ramming” which includes any act of piercing the ground by hand-held means. It submitted that this would catch even the most mundane and everyday tasks, such as gardening or pitching a tent, and this absurdity should favour its contrary construction.[38] 43. However, a conclusion that “ramming” means penetration of the ground by an object by means of force, which force may be applied by any type of hammering and not just hammering by means of a machine, is not to be equated with mere gardening or a recreational pursuit such as pitching a tent. Those examples are extreme ones and, depending on the specific facts of any particular case, would most likely be held[39] to fall outside the statutory prohibition in section 10(1)(a) of the Regulation. 44. The legislative intent of the word “ramming” being clear, there is no basis for the respondent’s submission that the presumption against doubtful penalisation[40] should operate to favour its construction of the word. 45. On Issue 1, therefore, we would, with respect, differ from the Judge and answer the certified question in the negative. G. Issue 2: whether the works were excluded works 46. It was the appellant’s case that the erection of the iron rods and setting up the safety nets to form the boundary of the works site did not fall within the exclusion in the definition of “works” in section 2 of the Regulation, namely “works necessary to ascertain the alignment and depth of an underground electricity cable”. The appellant supported the Magistrate’s construction of this exclusion and contended that the Judge was in error in her construction. 47. For its part, the respondent contended that the delineation of the area of work was absolutely necessary for the ascertainment of the alignment and depth of underground electricity cables and so the work was excluded and not subject to the offence under section 10 of the Regulation. The respondent therefore sought to support the Judge’s construction of the exclusion, reversing the Magistrate on this issue. (a) The proper construction of the exclusion 48. The critical word in the exclusion is “necessary”. There is no dispute between the parties that the usual dictionary meaning of this word is “indispensable, requisite, essential, needful; that cannot be done without …”.[41] 49. As a matter of both context and purpose, the threshold test of necessity for ascertaining the alignment and depth of an underground electricity cable is a high one. The clear legislative intention of the definition of “works” in section 2 of the Regulation is to cast a wide net over activities that may be prohibited or regulated by section 10 of the Regulation. 50. In addition to the need to focus on the word “necessary” in the exclusion, it is also critical to focus on what the work sought to be excluded must be necessary for. It must be work necessary to ascertain the alignment and depth of an underground electricity cable. Work which is not necessary for that purpose is not excluded. 51. The appellant also referred, as part of the context of the exclusion, to the strictness of the requirement in section 10(5) of the Regulation, which imposes on a competent person who is undertaking an investigation to ascertain the existence, alignment and depth of an underground electricity cable a requirement to “carry out the investigation in a manner that does not cause damage to, or impair the operation of, the underground electricity cable”.[42] A competent person, when carrying out an investigation, should therefore not do anything that might risk damage to an underground electricity cable. As a matter of context, this supports a restrictive construction of the exclusion and a narrow interpretation as to what may fall within the ambit of works necessary to ascertain the alignment and depth of an underground electricity cable. 52. On this footing, works which are purely preparatory and not part of the ascertainment of the alignment and depth of an underground electricity cable would not fall within the exclusion. However, to answer the fact-sensitive specific part of the second certified question in this appeal, one must have regard to the particular facts and to the respondent’s case as to why the ramming work was being done. (b) Was the work in the present case within the exclusion? 53. In the present case, the respondent’s case was that putting up the enclosure nets was done at the request of the MTR to mark the delineation of the works site. The orange net was to be positioned at least 1.8 metres away from the railway tracks. It was to assist in preventing materials, plant or tools from being placed outside the works site and misplaced so that they might fall onto the tracks, and also to prevent workers and other people from straying onto the tracks. This would also reduce the risk of disruption to train services, damage to MTR property and harm to the well-being of workers and train passengers. Without the delineation of the works site, it was suggested that even the competent person would be at risk of harm when conducting the cable inspection. All this, it was said by the respondent, was necessary because of the location of the works area adjacent to the railway track. 54. It cannot be doubted that the location of the works site adjacent to the railway track was an element adding to the risk of the works themselves and that the works site had to be carefully delineated. But it remains necessary to ask whether the ramming work in question was necessary to ascertain the alignment and depth of an underground electricity cable. 55. Clearly the ramming work was not in fact part of the ascertainment of the alignment and depth of an underground electricity cable. It was not the method by which that information was to be gathered. It has not been suggested, nor could it sensibly be suggested, that hammering iron bars into the ground until they come into contact with an underground electricity cable is a proper way to ascertain the alignment and depth of an underground electricity cable. One could hardly imagine a less safe way of doing so. Could the works site, though, be properly delineated by some other method that did not involve the risky act of penetrating the ground in the vicinity of an underground electricity cable with an iron bar? 56. In the present case, the Judge thought it could not. She held: “15. Taking into account all circumstances of the case, I agree with Mr Poon that the process of hanging up the net by erection of iron rods was a pre-construction survey for works necessary to ascertain the alignment and depth of an underground electricity cable. Of course, it is true, as the prosecution expert witness said, that both active and passive underground cable detection did not include the act of erection of iron rods, nor was erection of iron rods a necessary step in cable detection. However, on works sites involving railway tracks, if the site area had not been enclosed by hanging up nets, the safety of workers might be jeopardized and the operation of trains obstructed. Therefore I consider the erection of enclosure a necessary step for the works site in this case. The representative of the Appellant company had also testified that the cable detection would be conducted on the next day after the erection of iron rods and safety net. This was not disputed by the prosecution. 16. I notice that the magistrate said in his statement of findings that the Appellant company should have appointed a competent person for cable detection before erecting iron rods. This would imply that persons engaged in the detection had to work on a railway site without any enclosing net. In other words, the safety of the persons who carry out the cable detection would not be safeguarded. In case anything happened, the operation of trains would also be affected. Therefore taking into account the circumstances of the case, I am of the opinion that the erection of iron rods for enclosure was a necessary safety precaution to be taken before the commencement of detection work and a necessary working process in preparation for the detection work to be carried out. It was so closely related to and connected with detection work and therefore should be regarded as ‘work necessary to ascertain the alignment and depth of an underground electricity cable’. The act is exempted works under regulation 2 of the Regulation.” 57. Again, with respect, the Judge’s reasoning cannot be accepted. The delineation of the boundary of the works site was not necessary to the ascertainment of the alignment and depth of an underground electricity cable. 58. In any event, even on the footing that delineation of the works site was necessary, there is no good reason why this could not be done by some method other than the risky act of hammering iron rods into the ground in the vicinity of an underground electricity cable. The appellant referred in its printed case to the use of mills barriers, i.e. a free-standing non-penetrative form of barrier used to demarcate or close off an area, as shown in the Code of Practice.[43] The respondent countered that such barriers could be moved and would not have the permanence necessary to afford proper protection. That is a risk but it is not suggested that there is no risk that a boundary consisting of iron bars and orange netting might not also be moved or otherwise breached. Certainly, as between the competing risk of mills barriers being moved and the risk of electrocution from an iron bar driven into the ground striking an underground electricity cable, the risk of the latter is, in our view, far more significant than the former. 59. This conclusion does not present insuperable difficulties for contractors, even if unprepared to take the risk of mills barriers being moved. As the appellant pointed out, the Code of Practice indicates the vicinity of an underground electricity cable as being within 3 metres for the purpose of trench works or vertical penetration.[44] By means of passive cable detection, using a hand-held device, the contractor could identify an underground electricity cable and then erect iron rods 3 metres away from the cable. At that distance, on the basis of the Code of Practice, the contractor would not risk breaching section 10(1)(a) of the Regulation. At the same time, more importantly, the contractor would not risk the health and safety of its workers or the supply of electricity. 60. On Issue 2, therefore, we would, with respect, again differ from the Judge and, in answer to the certified question, hold that, in the present case, the preparatory acts of hammering iron rods into the ground in order to set up nets to delineate the boundary of the works site did not fall within the exclusion in the definition of “works” in section 2 of the Regulation. H. An alternative basis for liability on the facts 61. On the basis of the conclusions reached above, it followed that the appeal must be allowed and the respondent’s conviction and sentence restored. 62. Nevertheless, even if the appellant’s case failed on Issue 1, there would have been an alternative (and straight forward) basis for holding, subject to Issue 2, that the works in the present case fell within section 2 of the Regulation. 63. That alternative basis derives from the definition in section 2(a)(v) of the Regulation, specifically “slope works”. As will be recalled, the respondent was engaged under a contract to perform ground investigation for slope improvement works. Given the wide and inclusive definition of “works” in section 2(a) of the Regulation, especially the words “any kind of works involving or in connection with” the items specified in sub-paragraphs (i) to (vi), it would follow that the hammering of the iron rods into the ground, which was work done pursuant to the respondent’s contract, was a kind of work in connection with slope work and therefore subject to the requirements of section 10(1)(a) of the Regulation. Mr Grossman realistically accepted that this was correct. On this straight forward basis, the debate as to the meaning of “ramming” need not have been gone into. 64. However, had the respondent prevailed on Issue 1, it would not have been appropriate to restore the respondent’s conviction on this alternative basis. Since the prosecution case was based on the contention that the relevant work carried out by the respondent in the vicinity of an underground electricity cable was “ramming” within section 2(a)(vi), it would not have been fair to the respondent to permit criminal liability to attach when that case was never put for it to answer in the courts below. I. Conclusion 65. For these reasons, we allowed the appeal, restoring the respondent’s conviction and the sentence imposed by the Magistrate. 66. As to costs, the Court ordered that: (1) The costs order imposed by the Court of First Instance be quashed; (2) The appellant be awarded the costs of the appeal in this Court, including the costs of the application for leave to appeal; (3) The appellant be awarded the costs of its application to the Court of First Instance for a certificate of points of law of great and general importance. Mr William Tam DDPP and Mr Franco Kuan SPP, of the Department of Justice, for the Appellant Mr Clive Grossman SC and Mr Jose d'Almada Remedios, instructed by Ng, Au Yeung & Partners, for the Respondent [1] Cap.406H (“the Regulation”) [2] Defined in section 2 of the Regulation as “a person approved as a competent person under section 3” [3] See below [4] In FLS 8174/2011 [5] Eric Cheung Kwan-ming, Esq. [6] See below [7] Statement of Findings §§69 to 80 [8] Statement of Findings §§81 to 93 [9] In HCMA 110/2012 [10] Deputy Judge A. Pang (as she then was) [11] Judgment §§11 to 13 [12] Judgment §§14 to 16 [13] Cap.406 (“the Ordinance”) [14] Defined in section 2 of the Regulation as “an electricity supply line located below ground level” [15] Defined in section 2 of the Regulation as “an electricity supply line located at or above ground level” [16] Hereinafter referred to as “the Director” [17] 2005 Edition (“the Code of Practice”), approved by the Director of Electrical and Mechanical Services and issued in accordance with the provisions of section 15 of the Regulation [18] Code of Practice, §1.3.2.5, Table 1 [19] Regulation, section 10(4) [20] This introduction reads: “The reasonable steps to a safe system of work for works near U/G cables begin at the design stage and involve the early participation of the electricity supplier. Once cable plans have been received from the supplier, a competent person is appointed to detect the alignment of the U/G cables and propose trial hole locations. Working closely with site personnel, the competent person helps to open trial holes and expose cables to allow their alignment and depth to be accurately ascertained. He then prepares the ‘Competent Person Written Report’ and clearly marks up cable alignment and depth on site.” [21] See, HKSAR v Cheung Kwun-yin (2009) 12 HKCFAR 568 at §§11-14; HKSAR v Lam Kwong Wai (2006) 9 HKCFAR 574 at §63; and Fully Profit (Asia) Ltd v Secretary for Justice (2013) 16 HKCFAR 351 at §15. [22] DDPP, leading Mr Franco Kuan SPP [23] Leading Mr Jose d’Almada Remedios [24] Fully Profit (Asia) Ltd v Secretary for Justice (supra.) at §15; Vallejos v Commissioner of Registration (2013) 16 HKCFAR 45 at §§74-77; HKSAR v Lam Kwong Wai (2006) 9 HKCFAR 574 at §63 [25] China Field Ltd v Appeal Tribunal (Buildings) (No.2) (2009) 12 HKCFAR 342 at §36 [26] The Oxford English Dictionary, 2nd Ed., Vol.XIII, p.151 [27] Statement of Findings §§48 to 50 and 70 to 74 [28] Cap.123 [29] By way of example, 380/220V cables are typically at buried depths between 450-900 mm. The Table also notes that it indicates approximate information only. [30] Regulation, section 3(1) [31] Regulation, section 3(3) [32] Regulation, section 10(3) [33] Regulation, section 10(5)(c) [34] Regulation, section 11(3) [35] Regulation, section 14(1)(a) [36] Regulation, section 15(1) [37] Code of Practice, paras. 1 & 1.1.2 [38] The respondent relies, in this regard, on: R v Skeen (1859) 28 LJMC 91 at 94 per Lord Campbell CJ; Hatzl v XL Insurance Co Ltd [2009] 3 All ER 617 at 629F-G per Jacob LJ; and R (on the application of Noone) v Governor of HMP Drake Hall [2010] 4 All ER 463 at 476 per Lord Saville SCJ. [39] Applying the maxim de minimis non curat lex (“the law does not concern itself with trifles”) [40] Bennion on Statutory Interpretation, 6th Ed., Section 271, p.749 [41] The Oxford English Dictionary, Vol.VII, p.60 [42] Regulation, section 10(5)(c) [43] Ibid. at pp.16, 19, 20 & 21 [44] Code of Practice, §1.3.2.5 
Mr Justice Ribeiro, Acting CJ: 1. This appeal raises questions concerning the meaning and proper application of accounting standards relating to the treatment of an asset classified as an available-for-sale financial asset and the duties of auditors in respect thereof for the purposes of professional disciplinary proceedings. A. The audit and the disciplinary proceedings A.1 Background to the audit 2. The audit was of the financial statements for the year ended 30 June 2009 of Heng Tai Consumables Group Limited (“Heng Tai”), a company listed on the Hong Kong Stock Exchange. The auditors were the 2nd appellant firm (“RSM”) with the 1st appellant acting as the engagement partner. 3. The relevant financial asset was a parcel of 71,060,000 shares in China Zenith Chemical Group Limited (“CZ”), a company also listed on the Hong Kong Stock Exchange. Heng Tai first acquired shares in CZ in early 2004 and, in the following three years, increased its shareholding which it regarded as a long-term investment and part of its business expansion plan. Prior to Heng Tai’s disposal of 400 million shares in CZ in late 2007, Heng Tai accounted for its investment in CZ as in an associate. After such disposal, the remaining CZ shares held by Heng Tai were accounted for as an available-for-sale financial asset (“AFSFA”) shown as an investment in its balance sheet. 4. Standards for the accounting treatment of such financial assets[1] are contained in Hong Kong Accounting Standard 39 (“HKAS 39”)[2] which, in the version containing amendments as at May 2009, will require close examination. For present purposes, it suffices to note that HKAS 39 requires such assets to be measured at fair value[3] and requires gains and losses of such assets to be recognised directly in equity through the statement of changes in equity, until the asset is derecognised.[4] However, if the fair value of such financial assets should decline to an extent that provides objective evidence that impairment loss has been incurred, the standard requires the cumulative loss to be removed from equity and recognised in profit or loss even though the financial asset has not been derecognised.[5] Impairment losses recognised in profit or loss cannot be reversed through profit or loss.[6] 5. The HKAS 39 scheme for the treatment of AFSFAs was reflected in the Notes to Heng Tai’s financial statements setting out the Company’s Significant Accounting Policies.[7] 6. When the 71,060,000 shares (representing a 1.9% interest in CZ) were reclassified as an AFSFA in Heng Tai’s balance sheet, they were recorded as costing $37,690,100 and thus at $0.53 per share. Their quoted market price then proceeded to fall. In the year ended 30 June 2008, based on a closing bid price of $0.465 per share, the statement of changes in equity noted an unrealised fair value loss of $4,647,200, the balance sheet fair value of the CZ shares being stated as $33,042,900. 7. In the year ended 30 June 2009, a further unrealised fair value loss of $18,120,300 on a closing bid price of $0.21 per share was noted, bringing the cumulative AFSFA loss to $22,767,500 recognised directly as a deficit in an investment revaluation reserve, with the balance sheet fair value of the shares stated as $14,922,600. A.2 The audit 8. The cumulative loss of $22,767,500, as the auditors noted, was a decline that was “not small as compared to the investment cost of $37,690,100”. It represented more than 60% of such cost. The consolidated profit of the Group for the year (without reduction for the accumulated AFSFA loss) was $110,030,000. RSM accordingly conducted an impairment review. 9. After discussions, RSM agreed with Heng Tai’s management that it was not appropriate to transfer the accumulated fair value loss from equity to profit or loss, in the light of (i) their interpretation of HKAS 39 (to which I shall return[8]); (ii) the decline in fair value being due to the impact of the global financial crisis in 2008 and 2009, felt by many listed companies; (iii) the apparently sound financial condition of CZ[9] so that its share price was expected gradually to move upward; (iv) CZ’s net asset value at 30 June 2009 exceeding the cost of the shares; (v) the view that prices of CZ shares were quoted in a thin market and might not reflect their true fair value; and (vi) since CZ had not declared any dividend since 2006, the fact that no adverse impact on the estimated future cash flows was anticipated. 10. In the audit working papers, it was noted that in August 2009, when given the opportunity of purchasing additional CZ shares at $0.11 per share, Heng Tai purchased two parcels of 35,530,000 shares and 18,663,713 shares respectively. These post-balance sheet events were taken as evidence of management’s confidence in a likely upward movement of the CZ share price. 11. RSM issued an unqualified audit report dated 28 October 2009. A.3 The disciplinary proceedings 12. Following receipt of a complaint, the Financial Reporting Council (“FRC”) directed[10] the Audit Investigation Board (“AIB”) to conduct an investigation into a possible auditing irregularity involving non-recognition of impairment losses on the AFSFA in Heng Tai’s 2009 financial statements. 13. The AIB produced a detailed Investigation Report dated 19 October 2012 in which its main findings were stated as follows: “The AIB noted that paragraph 58 of HKAS 39 requires an entity to assess at each balance sheet date whether there is any objective evidence that a financial asset ... is impaired. If any such evidence exists, the entity is required to follow paragraph 67 of HKAS 39 to determine the amount of impairment losses for a financial asset classified as available-for-sale. Paragraph 61 of HKAS 39 specifies that a significant or prolonged decline in the fair value of an investment in an equity instrument below its cost is objective evidence of impairment. At 30 June 2009, the Company’s AFS Financial Assets, with an original cost of $37.7 million, were carried at the fair value of $14.9 million based on the quoted market price at that date. The AIB considers the cumulative decline in the fair value of the AFS Financial Assets of $22.8 million which represented over 60% of the original cost, was significant, and therefore considers it an objective evidence of impairment. Following the requirement in paragraph 67 of HKAS 39, the cumulative decline in the fair value of $22.8 million should have been removed from equity, ie, investment revaluation reserve, and recognized in profit or loss in the 2009 Financial Statements.” 14. The AIB concluded: “As the non-recognition of impairment losses represented approximately 21% of the consolidated profit of the Group for the year ended 30 June 2009, the AIB considers that RSM should have expressed a modified opinion on the 2009 Financial Statements in this respect. The AIB, therefore, considers that there is non-compliance with paragraphs 11 and 13 of HKSA 700.”[11] 15. The FRC adopted the Investigation Report and referred the complaint to the Hong Kong Institute of Certified Public Accountants (“the Institute”), conveying the AIB’s opinion that there had been non-compliance with HKSA 700.11 and HKSA 700.13. 16. The Institute’s Council then referred the complaint to its Disciplinary Committee (“the Committee”) pursuant to section 34(1)(a)(vi) of the Professional Accountants Ordinance (“PAO”),[12] charging the present appellants with having “failed or neglected to observe, maintain or otherwise apply a professional standard, namely HKAS 39”. 17. After a hearing on 1 June 2015, the Committee found the complaint substantiated but also held that there were “strong mitigating factors” causing it to take “a very lenient approach”, fixing the penalty at $10,000 for each respondent, plus costs. It also directed that no publicity should be given to the sanction imposed without the consent of the respondents. A.4 The Court of Appeal’s decision and the present appeal 18. The appellants lodged an appeal to the Court of Appeal[13] against the Committee’s ruling that they had breached section 34(1)(a)(vi). The present respondent, the Institute’s Registrar, lodged a cross-appeal against the Committee’s order prohibiting publicising of the sanction. Giving its reasons on 30 August 2016,[14] the Court of Appeal dismissed both the appeal and the cross-appeal. The appellants’ application to the Court of Appeal for leave to appeal to this Court was also dismissed.[15] 19. By its Determination dated 27 April 2017, the Appeal Committee[16] granted leave to appeal on the following questions of law, namely: (1) Do auditors who, without more, wrongly interpret or apply a professional standard in discharging their function as auditors thereby commit a breach of s.34(1)(a)(vi) of the Professional Accountants Ordinance, Cap.50 (“PAO”)? (2) Does s.34(1)(a)(vi) of the PAO import a standard of reasonableness or other similar considerations which are relevant to assessing whether an auditor has “failed or neglected to observe, maintain or otherwise apply a professional standard” within the meaning of such provision? (3) On the true construction of Hong Kong Accounting Standard 39, must an impairment adjustment be made once there has been a significant or prolonged decline in the fair value of an available-for-sale asset? 20. Question 3 is logically the first question since a negative answer would remove the basis of the disciplinary finding against the appellants, and since Questions 1 and 2 only arise if the answer to Question 3 is in the affirmative. I shall therefore consider Question 3 first. However, before doing so, a preliminary matter should be dealt with. B. Preliminary matter – the complaint before the Committee 21. Mr Johnny Mok SC[17] sought to argue that the complaint faced by the appellants was defective in that “it was wrong to transpose the alleged non-compliance of HKSA 700 (which was the professional standard relied upon by AIB) into a breach of HKAS 39”. Mr Mark Strachan SC[18] rightly objects on the ground that leave to appeal was never granted for this point to be argued. 22. In any event, the argument can be simply disposed of. It is, as the Committee held, merely a criticism as to the form of the complaint. True it is that the complaint alleged that the appellants had “failed or neglected to observe, maintain or otherwise apply a professional standard, namely HKAS 39” which was, on its face, inapt since HKAS 39 is an accounting standard which applies to the audited entity rather than the auditors. But as the appellants undoubtedly appreciated from numerous communications with, and the findings of, the AIB, the substance of the complaint against them was that they had failed to comply with standards which required them as auditors to form an opinion “as to whether the financial statements give a true and fair view or are presented fairly, in all material respects, in accordance with the applicable financial reporting framework” which “involves evaluating whether the financial statements have been prepared and presented in accordance with the specific requirements of the applicable financial reporting framework for particular classes of transactions, account balances and disclosures”.[19] 23. In the present case, the applicable reporting framework consisted of the paragraphs of HKAS 39 regarding the specific requirements governing the accounting treatment of AFSFAs. The appellants were required, as auditors, to evaluate whether Heng Tai’s treatment of the AFSFA comprising the CZ shares had been prepared and presented in accordance with those standards. The complaint might have been more accurately drawn if it had run along the lines that the appellants had “failed or neglected to observe, maintain or otherwise apply a professional standard, namely HKSA 700, in that they failed properly to evaluate Heng Tai’s compliance with HKAS 39 in respect of an AFSFA, before issuing an unmodified audit opinion”. However, no one was in any doubt that such was the substance of the complaint being investigated by the AIB and the Committee. As the Court of Appeal held,[20] the Committee was right to reject the “wrong charge” argument, adopting a “substance over form” approach, since the form of the charge gave rise to no procedural or other unfairness. C. Question 3: On the true construction of HKAS 39, must an impairment adjustment be made once there has been a significant or prolonged decline in the fair value of an available-for-sale asset? 24. The PAO defines “professional standards” to include “standards of accounting, auditing and assurance practices, issue or specified ... under section 18A”.[21] Section 18A(1) provides the statutory basis for requiring standards issued by the Institute’s Council to be adhered to by certified public accountants: “The Council may, in relation to the practice of accountancy, issue or specify any— (a) ... (b) standards of accounting, auditing and assurance practices, required to be observed, maintained or otherwise applied by any certified public accountant.” 25. PAO sections 34 and 35 provide for disciplinary proceedings and sanctions against a certified public accountant who “failed or neglected to observe, maintain or otherwise apply a professional standard”,[22] this being the provision presently invoked by the Institute in respect of the appellants’ failure properly to evaluate Heng Tai’s accounting treatment of the AFSFA. 26. Evaluation of an entity’s accounting treatment of an AFSFA must be premised on the auditors’ understanding of the applicable standard. The present discussion concerns the correctness or otherwise of the appellants’ understanding of whether an impairment loss had to be recognised, an issue turning on the true construction of the relevant paragraphs of HKAS 39. C.1 Financial assets measured by their fair value 27. The stated objective of HKAS 39 is relevantly “to establish principles for recognising and measuring financial assets”.[23] Where a financial asset is designated as an AFSFA, it is measured at its fair value both initially[24] and after its initial recognition.[25] 28. “Fair value” is defined as “the amount for which an asset could be exchanged, or a liability settled, between knowledgeable, willing parties in an arm’s length transaction”[26] and it is important for present purposes to note that the standard specifies that “[the] best evidence of fair value is quoted prices in an active market.”[27] Application Guidance (AG71) issued in December 2007 explains that: “A financial instrument is regarded as quoted in an active market if quoted prices are readily and regularly available from an exchange, dealer, broker, industry group, pricing service or regulatory agency, and those prices represent actual and regularly occurring market transactions on an arm’s length basis. Fair value is defined in terms of a price agreed by a willing buyer and a willing seller in an arm’s length transaction. The objective of determining fair value for a financial instrument that is traded in an active market is to arrive at the price at which a transaction would occur at the balance sheet date in that instrument (ie without modifying or repackaging the instrument) in the most advantageous active market to which the entity has immediate access...”[28] It reiterated that: “The existence of published price quotations in an active market is the best evidence of fair value and when they exist they are used to measure the financial asset or financial liability.”[29] 29. This is how fair value is measured in respect of equity instruments such as Heng Tai’s parcel of CZ shares. I might interject that at the disciplinary hearing and before the Court of Appeal, the appellants had sought to argue that the quoted prices of CZ’s shares were not an appropriate measure of the fair value of Heng Tai’s investment because they were too thinly traded for the prices to be a reliable indication of value. That argument was rejected by the Committee which found that the quoted price of CZ shares “remained relatively stable within the band of approximately $0.15 and $0.25 per share” and since there had been continued trading in such shares in the material period.[30] This was upheld by the Court of Appeal[31] and the argument against the reliability of quoted prices is not pursued before this Court. C.2. Measuring gains and losses 30. HKAS 39.55 materially provides: “A gain or loss arising from a change in the fair value of a financial asset ... shall be recognised, as follows. ... (b) A gain or loss on an available-for-sale financial asset shall be recognised directly in equity, through the statement of changes in equity..., except for impairment losses (see paragraphs 67-70) ... until the financial asset is derecognised, at which time the cumulative gain or loss previously recognised in equity shall be recognised in profit or loss. ... Dividends on an available-for-sale equity instrument are recognised in profit or loss when the entity’s right to receive payment is established ...” 31. The standard therefore recognises that there may be gains or losses in the fair value of an AFSFA and provides that such gains and losses should be recorded directly in equity, ie, in the balance sheet, “except for impairment losses”. This leads us to HKAS 39.58 which obliges the audited entity to assess whether that exception applies: “An entity shall assess at each balance sheet date whether there is any objective evidence that a financial asset or group of financial assets is impaired. If any such evidence exists, the entity shall apply ... paragraph 67 (for available-for-sale financial assets) to determine the amount of any impairment loss.” 32. HKAS 39.67 provides as follows: “When a decline in the fair value of an available-for-sale financial asset has been recognised directly in equity and there is objective evidence that the asset is impaired (see paragraph 59), the cumulative loss that had been recognised directly in equity shall be removed from equity and recognised in profit or loss even though the financial asset has not been derecognised.” 33. HKAS 39.68 quantifies the cumulative loss to be so removed from equity: “The amount of the cumulative loss that is removed from equity and recognised in profit or loss under paragraph 67 shall be the difference between the acquisition cost (net of any principal repayment and amortisation) and current fair value, less any impairment loss on that financial asset previously recognised in profit or loss.” 34. If such an impairment loss concerns an AFSFA which consists of an investment in an equity instrument, that loss is treated as realised since HKAS 39.69 provides: “Impairment losses recognised in profit or loss for an investment in an equity instrument classified as available for sale shall not be reversed through profit or loss.” C.3 When is there objective evidence of an impairment loss? 35. The crucial issue that divides the parties in the present case is whether the auditors should have concluded that Heng Tai ought to have recognised that objective evidence of an impairment loss existed in relation to the CZ shares, triggering HKAS 39.67. The resolution of that issue depends on the true construction of HKAS 39.59 and HKAS 39.61 in the context of the abovementioned provisions. 36. HKAS 39.59 provides as follows: “A financial asset or a group of financial assets is impaired and impairment losses are incurred if, and only if, there is objective evidence of impairment as a result of one or more events that occurred after the initial recognition of the asset (a ‘loss event’) and that loss event (or events) has an impact on the estimated future cash flows of the financial asset or group of financial assets that can be reliably estimated. It may not be possible to identify a single, discrete event that caused the impairment. Rather the combined effect of several events may have caused the impairment. Losses expected as a result of future events, no matter how likely, are not recognised. Objective evidence that a financial asset or group of assets is impaired includes observable data that comes to the attention of the holder of the asset about the following loss events: (a) significant financial difficulty of the issuer or obligor; (b) a breach of contract, such as a default or delinquency in interest or principal payments; (c) the lender, for economic or legal reasons relating to the borrower’s financial difficulty, granting to the borrower a concession that the lender would not otherwise consider; (d) it becoming probable that the borrower will enter bankruptcy or other financial reorganisation; (e) the disappearance of an active market for that financial asset because of financial difficulties; or (f) observable data indicating that there is a measurable decrease in the estimated future cash flows from a group of financial assets since the initial recognition of those assets, although the decrease cannot yet be identified with the individual financial assets in the group, including: (i) adverse changes in the payment status of borrowers in the group (e.g. an increased number of delayed payments or an increased number of credit card borrowers who have reached their credit limit and are paying the minimum monthly amount); or (ii) national or local economic conditions that correlate with defaults on the assets in the group (e.g. an increase in the unemployment rate in the geographical area of the borrowers, a decrease in property prices for mortgages in the relevant area, a decrease in oil prices for loan assets to oil producers, or adverse changes in industry conditions that affect the borrowers in the group).” 37. It may be noted that HKAS 39.59 has the following features: (a) It applies to financial assets generally and not (as we shall see in the case of HKAS 39.61) to any particular classes of financial assets. (b) It provides that objective evidence of impairment exists if (i) a loss event is identified which (ii) has an impact on future cash flows of the financial asset that can be reliably estimated. (c) It adopts an “incurred loss” approach and rejects recognition of “losses expected as a result of future events, no matter how likely”. (d) It gives examples of loss events, all of which impact upon estimated future cash flows of the financial asset. Thus, sub-paragraphs (a) to (f) refer to cases where the issuer of the financial asset develops serious financial difficulties, calling into question the ability of the financial asset to generate income for the holder and thus adversely affecting the holder’s estimated future cash flows from that asset. 38. It is important to bear in mind that while HKAS 39.59 focusses on specific loss events adversely affecting cash flows, what constitutes objective evidence of impairment loss is not the impact on cash flows per se, but a decline in the fair value of the AFSFA, fair value being the basis for measuring gain and loss in respect of financial assets throughout HKAS 39. The effect of a loss event with an adverse effect on cash flows generated by the financial asset is that the fair value of the asset decreases since the amount which a knowledgeable, willing party would be willing to pay for such an impaired asset in an arm’s length transaction must inevitably be discounted. 39. There may obviously be objective evidence of impairment – evidence of a decline in the AFSFA’s fair value – which does not depend on an adverse impact on cash flow. HKAS 39.61 identifies two categories of such objective evidence. It provides as follows: “In addition to the types of events in paragraph 59, objective evidence of impairment for an investment in an equity instrument includes information about significant changes with an adverse effect that have taken place in the technological, market, economic or legal environment in which the issuer operates, and indicates that the cost of the investment in the equity instrument may not be recovered. A significant or prolonged decline in the fair value of an investment in an equity instrument below its cost is also objective evidence of impairment.” 40. The following features of HKAS 39.61 may be noted: (a) It refers to objective evidence of impairment which is additional to the objective evidence provided by “the types of events in [HKAS 39.59]”, that is, loss events which impact on cash flows. (b) Unlike HKAS 39.59 which covers financial assets generally, HKAS 39.61 deals specifically with investments in equity instruments. (c) It does not mention cash flows and does not base impairment loss on adverse impacts on cash flows. Instead it focusses on the asset’s decline in value below its investment cost. (d) It identifies two situations as providing objective evidence of such impairment loss, the first involving information about significant adverse changes in the technological, market, economic or legal environment in which the issuer of the equity instrument operates, which information indicates that the cost of the investment in the equity instrument may not be recovered. (e) The second involves a significant or prolonged decline in the fair value of the investment in an equity instrument below its cost. Since, as we have seen, HKAS 39 states that the best evidence of fair value is quoted prices in an active market, this equates with a significant or prolonged decline in the quoted price of an AFSFA consisting of listed shares. C.4 The answer to Question 3 41. The foregoing analysis leads to the conclusion that the answer to Question 3 is in the affirmative with the addition of the following italicised words: On the true construction of HKAS 39, an impairment adjustment must be made in respect of an AFSFA consisting of an equity instrument once there has been a significant or prolonged decline in its fair value. 42. This is because, in its second sentence, HKAS 39.61 identifies as a head of objective evidence of impairment, a significant or prolonged decline in the fair value of an equity instrument below its cost. This is a category of objective evidence of impairment which is additional to, and independent of, the categories of objective evidence comprising loss events which impact upon cash flows, set out in HKAS 39.59. It is also independent of the head of objective evidence described in the first sentence of HKAS 39.61. 43. This interpretation of HKAS 39.61 was adopted by the AIB: “Paragraph 61 of HKAS 39 specifies that a significant or prolonged decline in the fair value of an investment in an equity instrument below its costs is also objective evidence of impairment ... Given this explicit statement, it is not appropriate to consider a significant decline in the fair value is [sic] only an indication of possible impairment or ‘an example of the Loss Event constituting objective evidence of impairment’,”[32] 44. It was also the approach of the Committee. It found HKAS 39.61 “very clear” and “that for listed/quoted equity instruments such as shares quoted in an active stock market ... a significant or prolonged decline in the quoted share price of a listed company will no doubt be the most objective evidence of impairment.”[33] Since the cumulative decline in the fair value of the CZ shares of $22.8 million which represented over 60% of the original cost, was significant, no other evidence was needed to trigger HKAS 39.67. 45. This was upheld by the Court of Appeal[34] which cited in support relevant paragraphs from the “Basis for Conclusions” issued by the International Accounting Standards Board (“IASB”) to which I shall return.[35] Cheung JA referred in particular to BC 106 which states: “The Board[36] agreed that for marketable investments in equity instruments any impairment trigger other than a decline in fair value below cost is likely to be arbitrary to some extent. ...” His Lordship commented: “It is precisely because of the risk of arbitrariness referred to in BC 106 that paragraph 61 expressly recognizes a significant or prolonged decline in the fair value of the equity instrument below its cost as objective evidence of impairment.”[37] C.5 The auditors’ approach 46. The auditors’ understanding of the applicable standard is revealed by this statement in their working papers: “Our analysis As per HKAS 39, a financial asset is impaired, if and only if: (a) Objective evidence of impairment resulted from loss event, eg prolonged or significant decline in the fair value of CZ, significant financial difficulty of CZ, significant changes with adverse effect in technological, market, economic or legal environment so that the costs of investment in CZ may not be recovered; and (b) the loss event impacted on the estimated future cash flows from CZ.” 47. The words I have italicised come from HKAS 39.61 while those that are not italicised come from HKAS 39.59. The auditors therefore elided the two paragraphs and failed to recognise that (i) those paragraphs deal respectively with different types of financial asset, HKAS 39.61 dealing specifically with equity instruments (with which we are concerned); (ii) that the two paragraphs refer to different heads of objective evidence of impairment; (iii) that while the objective evidence referred to in HKAS 39.59 involves loss events which have an adverse impact on future cash flows, HKAS 39.61 (in its second sentence) looks for objective evidence of a significant decline in the asset’s fair value below its investment cost. 48. Having indiscriminately mixed the two provisions together, the auditors erroneously treated an adverse impact on cash flows as an essential requirement for finding objective evidence of impairment in every case. They failed to appreciate that no such requirement attaches to either of the heads of objective evidence specified in HKAS 39.61. 49. The auditors noted that there had been a $22,767,500 decline in the AFSFA’s fair value that was “not small as compared to the investment cost of $37,690,100”, representing as it did, more than 60% of such cost. However, they failed to recognise that such a decline was “significant” and sufficient to trigger HKAS 39.61. Instead, they proceeded to consider irrelevant matters which did not detract from the plainly established objective evidence of impairment but which were relied on as somehow justifying the non-triggering of HKAS 39.61. Those matters included (i) the fact that the quoted share prices of many listed companies had also been adversely affected by the global financial crisis; (ii) the fact that CZ had recurring profits and a net asset value exceeding Heng Tai’s investment cost; (iii) the disclosure in Heng Tai’s financial statements that the directors had considered but excluded impairment of the Company’s listed AFSFA investment;[38] and (iv) management’s expectation that the share price would rise, as evidenced by the post-balance sheet acquisition of additional CZ shares. C.6 The appellants’ argument set out in their Written Case 50. The appellants’ Written Case seeks to defend the auditors’ interpretation of the relevant paragraphs of HKAS 39. The central proposition it states is that: “... impairment adjustment is not to be made unless and until impairment loss has occurred and the loss event has an impact on the estimated future cash flows of the financial asset that can be reliably estimated.”[39] 51. Thus, it argues that HKAS 39.59 “imposes the two conditions and provides that the financial asset is impaired only if both conditions are satisfied”.[40] 52. Accordingly, like the auditors, the appellants’ Written Case elides HKAS 39.61 with HKAS 39.59. It contends that “HKAS 39.61 is not a stand alone provision” and that it provides: “... two additional types of loss event (being one of the conditions set out in HKAS 39.59) applicable to equity instruments. One of these ‘loss events’ is ‘significant or prolonged decline in the fair value of an investment in an equity instrument below its cost’.”[41] 53. The Written Case seeks to support this construction of the standard by arguing that: “Such interpretation is supported by the opening phrase of HKAS 39.61 ‘[i]n addition to the types of events in paragraph 59’ [emphasis added]. The word ‘events’ clearly refers to ‘loss events’.”[42] 54. The argument therefore treats HKAS 39.61 as merely adding two “loss events” to the list found in HKAS 39.59 and proceeds to attach to them the requirement that they must have an adverse impact on future cash flows. 55. That construction cannot be accepted. HKAS 39.61 states: “In addition to the types of events in paragraph 59, objective evidence of impairment for an investment in an equity instrument includes information about significant changes with an adverse effect that have taken place in the technological, market, economic or legal environment in which the issuer operates, and indicates that the cost of the investment in the equity instrument may not be recovered. A significant or prolonged decline in the fair value of an investment in an equity instrument below its cost is also objective evidence of impairment.” 56. Looking at the first sentence: its subject is the phrase “objective evidence of impairment for an investment in an equity instrument”; its verb is “includes”; and its object is the phrase describing “information” about the relevant significant changes. The first sentence of HKAS 39.61 therefore sets out a category of objective evidence relating to equity instruments. The introductory words “In addition to the types of events in paragraph 59” indicate that this head of objective evidence of impairment is additional to the heads of objective evidence comprising the “types of events in paragraph 59”. 57. The second sentence of HKAS 39.61 identifies a further head (“also”) of objective evidence of impairment constituted by a significant or prolonged decline in the fair value of an investment in an equity instrument below its cost. 58. Paragraph 61 is thus concerned with setting out categories of objective evidence which are additional to the objective evidence constituted by “the types of events in paragraph 59”, consisting of loss events with adverse impacts on future cash flows of the financial asset. The words “the types of events in paragraph 59” in the opening phrase of HKAS 39.61 are not a reference to “loss events” – the word “loss” is not used. In the context of the words which follow, the words “the types of events in paragraph 59” are a reference to the types of events which constitute objective evidence of impairment under HKAS 39.59. They do not operate to convert the two additional heads of objective evidence of impairment in HKAS 39.61 into merely additional “loss events” to be tacked onto HKAS 39.59. C.6a Basis for conclusions (as at December 2007) 59. The conclusion that HKAS 39.61, particularly in its second sentence, sets out an independent or “stand-alone” head of objective evidence of impairment sufficient to trigger HKAS 39.67, is supported by published guidance. 60. Noting that HKAS 39 is based on IAS 39 issued by the IASB, the Institute’s Council stated that it “agreed with the IASB’s basis for conclusions on IAS 39” and that “there are no significant differences between HKAS 39 and IAS 39”.[43] It therefore reproduced the IASB’s “basis for conclusions” for reference. 61. In paragraph BC96, comments received by the IASB disagreeing with use of quoted prices for measuring the fair value of an instrument quoted in an active market were noted. In response, the IASB justified using quoted prices as follows: “BC97. However, the Board confirmed that a quoted price is the appropriate measure of fair value for an instrument quoted in an active market, notably because (a) in an active market, the quoted price is the best evidence of fair value, given that fair value is defined in terms of a price agreed by a knowledgeable, willing buyer and a knowledgeable, willing seller; (b) it results in consistent measurement across entities; and (c) fair value as defined in the Standard does not depend on entity-specific factors. ...” 62. Guidance was also relevantly given in respect of impairment and uncollectibility of financial assets as follows. “BC105. Under IAS 39, investments in equity instruments that are classified as available for sale ... are subject to an impairment assessment. The original IAS 39 did not include guidance about impairment indicators that are specific to investments in equity instruments. Questions were raised about when in practice such investments become impaired. BC106. The Board agreed that for marketable investments in equity instruments any impairment trigger other than a decline in fair value below cost is likely to be arbitrary to some extent. If markets are reasonably efficient, today's market price is the best estimate of the discounted value of the future market price. However, the Board also concluded that it is important to provide guidance to address the questions raised in practice. BC107. The revised IAS 39 includes impairment triggers that the Board concluded were reasonable in the case of investments in equity instruments (paragraph 61). They apply in addition to those specified in paragraph 59, which focus on the assessment of impairment in debt instruments.” 63. BC105 therefore indicates that the intention in IAS 39, reproduced in HKAS 39, is to provide “impairment indicators that are specific to investments in equity instruments”. BC106 explains that for investments in marketable equity instruments, any impairment trigger other than a decline in fair value below cost is excluded because of likely arbitrariness. It is thus approaching such a decline as a “stand-alone” impairment trigger. BC107 makes it clear that the impairment triggers being discussed are those set out in paragraph 61, which the IASB concluded “were reasonable in the case of investments in equity instruments” contrasting them with the impairment triggers in paragraph 59 “which focus on the assessment of impairment in debt instruments”. C.6b July 2009 Newsletter of the International Financial Reporting Interpretations Committee (“IFRIC”) 64. It is the policy of the Institute’s Council to develop financial reporting standards to achieve convergence with IFRSs issued by the IASB.[44] Thus, as previously noted, HKAS 39 is based on IAS 39. Guidance provided in the captioned newsletter (cited by the AIB in its Investigation Report) on IAS 39.61 is therefore relevant: “(ii) Paragraph 67 of IAS 39 requires an entity to recognise an impairment loss on available-for-sale equity instruments if there is objective evidence of impairment. Paragraph 61 of IAS 39 states: ‘A significant or prolonged decline in the fair value of an investment in an equity instrument below its cost is also objective evidence of impairment.’ [emphasis added] Consequently, the IFRIC concluded that when such a decline exists, recognition of an impairment loss is required”. 65. The IFRIC therefore regarded the second sentence of IAS 39.61 as a “stand-alone” impairment trigger. This is reinforced by the subsequent paragraphs in the Newsletter which stated: “(iii) The fact that the decline in the value of an investment is in line with the overall level of decline in the relevant market does not mean that an entity can conclude the investment is not impaired. (iv) The existence of a significant or prolonged decline cannot be overcome by forecasts of an expected recovery of market values, regardless of their expected timing. Consequently, the IFRIC concluded that an anticipated market recovery is not relevant to the assessment of ‘significant or prolonged’.” 66. This was a rejection of the relevance of two of the matters relied on by the auditors in the present case as justifying the non-triggering of HKAS 39.67. C.7 The appellants’ argument as modified by Counsel 67. At the outset of the hearing, Mr Mok SC indicated that he had modified his views regarding the central proposition in the Written Case[45] that an impairment adjustment “is not to be made unless and until impairment loss has occurred and the loss event has an impact on the estimated future cash flows of the financial asset that can be reliably estimated”. Asked to clarify his position in writing, Counsel provided “2nd Supplemental Submissions” stating as follows: §1 The 2nd condition of HKAS 39.59 refers to the effect of the loss event, namely, the value of the investment is impaired in the sense that it is expected that the estimated future cash flows back to the entity have been impacted upon – i.e. they would not be collected as originally agreed or expected. §2 The future cash flow can take the form of return of the principal on a debt instrument, but it may also take the form of the proceeds of sale of an equity instrument. In the latter case, if it is expected that the proceeds of a future sale of the instrument would fall below the cost of investment, such expected impact would constitute an impairment loss. §3 Such impairment loss in the case of equity instruments may be caused by one of the loss events set out in HKAS 39.59 (a)-(f), e.g. “significant financial difficulty of the issuer of the equity instrument”, or “the disappearance of an active market for that financial asset because of financial difficulties”. §4 HKAS 39.61 provides further examples of loss events in relation to an equity instrument. The 1st sentence is concerned with loss events in the form of “significant changes with an adverse effect that have taken place in the technological, market, economic, or legal environment in which the issuer operates”, but such loss events by themselves do not constitute impairment loss. §5 The parallel in HKAS 39.61 to the 2nd condition in HKAS 39.59 is that there must be an indication “that the cost of the investment in the equity instrument may not be recovered”. §6 The 2nd sentence in HKAS 39.61 likewise describes a loss event, which is parallel to the 1st condition in HKAS 39.59. The 2nd condition in HKAS 39.59 is applicable if it is expected that the estimated cash flows back the entity would be impacted upon in a future sale of the equity instrument in the sense that the proceeds would fall below the cost of investment, i.e. the same condition as set out in the 1st sentence of HKAS 39.61. 68. This represents an attempt to conjure up a second condition to the objective evidence of impairment identified in the second sentence of HKAS 39.61. The reasoning appears to run along the following lines: (a) HKAS 39.59 sets out heads of objective evidence of impairment which have two elements: loss events plus an impact on cash flows [§§1 and 3]. (b) “Cash flow” should be given a broad meaning to embrace the proceeds of a future sale of the equity instrument, and if such proceeds are less than the investment cost, this should be regarded as an adverse impact on cash flows constituting impairment loss [§2]. (c) The head of objective evidence contained in the first sentence of HKAS 39.61 has a “parallel” structure involving a “loss event” constituted by significant adverse changes in the issuer’s environment, plus a risk of non-recovery of the investment cost [§§4 and 5]. (d) The second sentence should be construed as having a “parallel” structure: The “significant or prolonged decline” is the first condition and an adverse impact on cash flows is the second. The latter condition should be interpreted to mean a projection that the proceeds on a future sale of the equity instrument would fall below the cost of investment [§6]. 69. This argument cannot be accepted. There is no warrant for introducing a gloss on the concept of “cash flow”. HKAS 39.59 straightforwardly requires objective evidence of a loss event which has an impact on the estimated future cash flows of the financial asset, which assumes the continued existence of the financial asset and an assessment of the income it is likely to generate. There is no reason to strain the notion of “cash flow” to include possible losses on a future sale of the asset when no such sale is presently contemplated (which is why the asset is not derecognized and remains accounted for in the balance sheet). 70. Secondly, there is no basis for suggesting that because the heads of objective evidence in HKAS 39.59 and in the first sentence of HKAS 39.61 may be seen as having two elements – a loss event plus a second condition – the head of objective evidence identified in the second sentence of HKAS 39.61 ought likewise to have a second condition attributed to it by some form of “parallel” reasoning, when no words importing any such condition exist in the second sentence. 71. Thirdly, even if one were to go along with the contrived reasoning to arrive at §6 quoted above, it is hard to see how the new formulation could possibly work. The proposition appears to be that, in addition to objective evidence of a significant or prolonged decline in the AFSFA’s fair value below investment cost, a second condition must be satisfied before HKAS 39.67 is triggered, namely, that: “... it is expected that the estimated cash flows back to the entity would be impacted upon in a future sale of the equity instrument in the sense that the proceeds would fall below the cost of investment ...” 72. A future sale when? How is the date for the sale to be selected? How is one to ascertain what the quoted price and hence what the sale proceeds will be at that future time so as to ascertain whether such proceeds fall below the investment cost? This involves a departure from the incurred loss model adopted in the applicable standard. As HKAS 39.59 specifies: “Losses expected as a result of future events, no matter how likely, are not recognised.” Moreover, the modified formulation invites subjective speculation and lacks observable data capable of providing reasonable assurance based on objective evidence. 73. In contrast, if one accepts, as do the AIB, the Disciplinary Committee, the IASB, the IFRIC and the Court of Appeal, that the second sentence of HKAS 39.61 is an independent and self-contained standard for assessing impairment loss, the entity and the auditors are, at the balance sheet date, able to rely on objective evidence of a significant or prolonged decline in the fair value of an equity instrument by examining the record of quoted prices and trading volumes. It would be a matter of judgment to decide whether the decline is “significant”. But that would be based on objective evidence. 74. The suggestion in §6 of the reformulation that the proposed gloss on the second sentence of HKAS 39.61 introduces a condition which is “the same condition as set out in the 1st sentence of HKAS 39.61” is plainly incorrect. The first sentence of HKAS 39.61 involves examining observable facts including “information about significant changes with an adverse effect that have taken place in the technological, market, economic or legal environment in which the issuer operates”, providing objective evidence for an indication that “the cost of the investment in the equity instrument may not be recovered”. This is very different from the speculative suggestion in §6 of the re-formulation. 75. For the foregoing reasons, I do not accept the appellants’ interpretation of the applicable standard and would uphold the construction adopted by the Disciplinary Committee and the Court of Appeal. D. Question 1: Do auditors who, without more, wrongly interpret or apply a professional standard in discharging their function as auditors thereby commit a breach of PAO s.34(1)(a)(vi)? Question 2: Does s.34(1)(a)(vi) import a standard of reasonableness or other similar considerations which are relevant to assessing whether an auditor has “failed or neglected to observe, maintain or otherwise apply a professional standard” within the meaning of such provision? 76. Since the answer to Question 3 is “Yes”, Questions 1 and 2 fall to be considered. They may be taken together as they both concern the true construction of section 34(1)(a)(vi). The essential issue is whether, given that there was a failure to apply an applicable standard, section 34(1)(a)(vi) should be construed as incorporating a standard of reasonableness capable of excusing the auditors notwithstanding their default in observing the standard. 77. As we have seen,[46] PAO section 18A authorises the Institute’s Council to issue accounting and auditing standards which certified public accountants are required to observe, maintain or otherwise apply. 78. PAO section 34(1)(a) sets out categories of complaints against a certified public accountant which, if made to the Registrar of the Institute’s Council, must be referred to the Institute’s Council which may in turn refer the complaint to its Disciplinary Panels. The categories of complaint are of three types. The first[47] covers complaints involving cases where the certified public accountant has been convicted of certain criminal offences or been punished under the FRC Ordinance and the third[48] involves breaches involving corporate practices. 79. The present complaint falls within the second category[49] which involves infringements of professional standards or duties. The relevant provisions are as follows: “A complaint that – (a) a certified public accountant – ... (iv) has been negligent in the conduct of his profession; (v) without reasonable excuse, failed or neglected to comply with any direction issued under section 32F(2) and with which he was required by the Practice Review Committee to comply; (vi) failed or neglected to observe, maintain or otherwise apply a professional standard; (vii) without reasonable excuse, failed or neglected to comply with any requirement made under section 42D in relation to him by an Investigation Committee; (viii) has been guilty of professional misconduct; (ix) refused or neglected to comply with the provisions of any bylaw or rule made or any direction lawfully given by the Council; (x) was guilty of dishonourable conduct; ... shall be made to the Registrar who shall submit the complaint to the Council which may, in its discretion but subject to section 32D(7),[50] refer the complaint to the Disciplinary Panels.” 80. The complaint against the appellants under section 34(1)(a)(vi) was that they had “failed or neglected to observe, maintain or otherwise apply a professional standard, namely HKAS 39” but, as noted above[51] the substance of that complaint was that the appellants had failed or neglected to observe, maintain or otherwise apply the auditing standard HKSA 700, in that they failed properly to evaluate Heng Tai management’s compliance with HKAS 39 in respect of an AFSFA, before issuing an unmodified audit opinion. D.1 The language of section 34(1)(a)(vi) 81. On its face, section 34(1)(a)(vi) does not provide for any form of reasonable excuse as a justification for failing or neglecting to apply a relevant standard. The legislature could of course have provided for such an excuse, as it did by use of the words “without reasonable excuse” in paragraphs (v) and (vii). But it does not do so. 82. Taking section 34(1)(a)(vi) in the context of the other paragraphs in section 34(1)(a), it is apparent that a range of complaints of varying seriousness and requiring proof of different constituent elements are specified. Thus, complaints in the first category (eg, involving conviction for a criminal offence, especially an offence involving dishonesty) would obviously be at the serious end of the spectrum. A range also exists within the second category of complaints. Thus, a complaint under paragraph (iv) requires negligence to be established. Different constituents have to be proved to establish professional misconduct under paragraph (viii) or dishonourable conduct under paragraph (x). With paragraphs (v) and (vii), the certified public accountant has to be shown to have failed or neglected to comply with specific directions or requirements, such failure being excusable on reasonable grounds. Reasonable excuse is, however, not mentioned in paragraph (ix) which involves refusing or neglecting to comply with the Council’s lawful bylaws, rules or directions. 83. It is natural to read section 34(1)(a)(vi) as specifying a ground of complaint at the least serious end of the spectrum, not aimed at punishment, but aimed simply at enforcing the application of published standards in the interests of uniform and predictable professional practice without implying any fault, moral blame or misconduct. Such a complaint would merit commensurately minor sanctions.[52] Indeed, that appears to have been the approach adopted by the Committee in the present case in deciding on “very lenient” treatment of the appellants after having found the complaint substantiated. 84. Professional standards operating in that manner are not uncommon. Mr Strachan SC cited several examples in the case-law where failures to comply with applicable standards without fault on the individual’s part were upheld as infringements of those standards. 85. Thus, for instance, in Sadler v General Medical Council,[53] a Privy Council case involving a doctor who, through no fault of his own, failed to complete a required period of retraining to demonstrate an acceptable standard of professional performance, Lord Walker of Gestingthorpe stated: “[Their Lordships] ... are prepared to accept that the appellant did all he could to achieve effective retraining, and that the very unfortunate delay in starting the retraining is not his fault. But (whether in legal language or in everyday speech) failure does not always imply fault, and their Lordships are satisfied that in this context it should not be taken as implying fault. The purpose of all the provisions is to protect the public from sub-standard medical care, not to punish practitioners for blameworthy acts or omissions. The appellant was most unfortunate in the long delay in the commencement, and the abrupt termination of his retraining. But the fact is that he had not complied with the statement of requirements.” D.2 Reasonable excuse via HKSA 700 86. As noted above,[54] the substance of the complaint against the appellants is that they failed or neglected to observe, maintain or otherwise apply the auditing standard HKSA 700 in that they failed properly to evaluate Heng Tai’s non-compliance with HKAS 39 in respect of an AFSFA, before issuing an unmodified audit opinion. Mr Mok SC submits that HKSA 700 requires no more than reasonable best endeavours by an auditor, so that its incorporation as a necessary element of section 34(1)(a)(vi) complaint provides a basis for holding that the complaint was not established. D.3 The relevant provisions of HKSA 700[55] 87. HKSA 700.5 provides that “... the objective of an audit of financial statements is to enable the auditor to express an opinion whether the financial statements are prepared, in all material respects, in accordance with the applicable financial reporting framework.” 88. The aim is to produce an audit opinion which states: “... whether the financial statements ‘give a true and fair view’ or ‘are presented fairly, in all material respects,’ in accordance with the applicable financial reporting framework.”[56] 89. HKSA 700.10 elaborates on the applicable framework, pointing out that: “... the financial reporting framework provides a context for the auditor’s evaluation of the fair presentation of the financial statements, including whether the financial statements have been prepared and presented in accordance with the specific requirements of the applicable financial reporting framework for particular classes of transactions, account balances and disclosures.” In the present case, the applicable financial reporting framework consists of the relevant paragraphs of HKAS 39. 90. HKSA 700.11 and HKSA 700.12 stress that the focus of the auditing exercise is on the audit evidence and requires such evidence to be capable of providing reasonable assurance that the financial statements are free from material mis-statement: HKSA 700.11 “The auditor should evaluate the conclusions drawn from the audit evidence obtained as the basis for forming an opinion on the financial statements.” HKSA 700.12 “When forming an opinion on the financial statements, the auditor evaluates whether, based on the audit evidence obtained, there is reasonable assurance about whether the financial statements taken as a whole are free from material misstatement. This involves concluding whether sufficient appropriate audit evidence has been obtained to reduce to an acceptably low level the risks of material misstatement of the financial statements and evaluating the effects of uncorrected misstatements identified.” 91. HKSA 700.13 addresses the process of evaluation in arriving at the audit opinion: HKSA 700.13 “Forming an opinion as to whether the financial statements give a true and fair view or are presented fairly, in all material respects, in accordance with the applicable financial reporting framework involves evaluating whether the financial statements have been prepared and presented in accordance with the specific requirements of the applicable financial reporting framework for particular classes of transactions, account balances and disclosures. This evaluation includes considering whether, in the context of the applicable financial reporting framework: (a) The accounting policies selected and applied are consistent with the financial reporting framework and are appropriate in the circumstances; (b) The accounting estimates made by management are reasonable in the circumstances; (c) The information presented in the financial statements, including accounting policies, is relevant, reliable, comparable and understandable; and (d) The financial statements provide sufficient disclosures to enable users to understand the effect of material transactions and events on the information conveyed in the financial statement, for example, in the case of financial statements prepared in accordance with Hong Kong Financial Reporting Standards (HKFRSs), the entity’s financial position, financial performance and cash flows.” 92. Mr Mok SC relied on HKSA 700.12 for his submission that a standard of reasonableness was introduced since the auditor was only expected to achieve “reasonable assurance”, stressing that auditors are not guarantors of the accuracy of the financial statements. 93. However, “reasonable assurance” in the context of HKSA 700.12 does not have any bearing on the issue being discussed. As explained in HKSA 200.17:[57] “Reasonable assurance is a concept relating to the accumulation of the audit evidence necessary for the auditor to conclude that there are no material misstatements in the financial statements taken as a whole. Reasonable assurance relates to the whole audit process.” 94. It is in that context that HKSA 200.18 states: “An auditor cannot obtain absolute assurance because there are inherent limitations in an audit that affect an auditor’s ability to detect material misstatements ...” 95. Thus, HKSA 700.12 specifies “reasonable assurance” as the standard of reliability required of the audit evidence to provide a proper foundation for the audit opinion. No evidential issue arises in the present case. The audit evidence enabling the auditors to evaluate Heng Tai’s failure to trigger HKAS 39.67 was objective evidence of a significant or prolonged decline in the fair value of the CZ shares below the investment cost. The Committee and the Court of Appeal held that there was reliable and pertinent evidence available, consisting of the quoted prices and trading volumes of the shares over the year ended 30 June 2009, enabling the cumulative loss to be compared with the known acquisition cost. The difficulty, as noted above, was that the auditors misinterpreted HKAS 39.61 and so failed to give effect to that standard. It had nothing to do with the accumulation of evidence necessary to form their unqualified audit opinion. D.4 A reasonable misinterpretation of HKAS 39.61? 96. Mr Mok SC submitted on behalf of the appellants that in so far as the appellants had misinterpreted the applicable standard, it was a reasonable error to make, given that the relevant provisions were very unclear and difficult to construe. Mr Mok SC sought to find support for this approach in the Preface to Hong Kong Financial Reporting Standards[58] which, in §§21 and 22 state as follows: “21. Council ... expects members of the HKICPA[59] who assume responsibilities in respect of financial statements to observe HKFRSs.”[60] “22. Where this responsibility is evidenced by the association of members’ names with such financial statements in the capacity of directors or other officers, the onus will be on them to ensure that the existence and purpose of HKFRSs are fully understood by non-member fellow directors and other officers. Members of the HKICPA should also use their best endeavours to ensure that HKFRSs are observed and that departures found to be necessary are adequately disclosed and explained in the financial statements.” 97. I am unable to see how these paragraphs assist the appellants’ argument. In the first place, paragraph 22 deals with the responsibility of members acting, not in the capacity of auditors, but as directors or other officers of a company. It requires them to ensure that their fellow corporate officers fully understand the applicable standard. Thus, far from giving members an excuse for reasonably failing to comprehend a standard, paragraph 22 presupposes that members have a proper understanding which enables them to give their fellow officers advice about the standard. The reference to “best endeavours” is made in the context of requiring members who are officers do their best to ensure “that the HKFRSs are observed and that departures found to be necessary are adequately disclosed and explained in the financial statements”. This again assumes that the member has a proper understanding of the applicable standard to be able to exercise such best endeavours in relation to the preparation of the financial statements. 98. Paragraph 23 of the Preface which follows does deal with members who act as auditors. It provides that they “should be in a position to justify departures, to the extent that their concurrence with the departures is stated or implied.” This is germane to the present case where the appellants concurred with the departure by Heng Tai from the requirements of HKAS 39.61. Far from excusing them for not understanding the relevant standard, the Preface requires them to be in a position to justify such concurrence. 99. I therefore can see no basis for reading into section 34(1)(a)(vi) some standard of reasonableness capable of exonerating the appellants. 100. Nor do I accept that the relevant accounting standards were ambiguous or unclear, as the appellants sought to submit. The AIB had no difficulty in applying HKAS 39.61, calling it an “explicit statement” and treating it as a self-contained standard. This view was evidently shared by the FRC and the Institute which adopted the AIB Report and referred the matter to the Disciplinary Committee which found that standard “very clear”. It is also a view shared by the IASB which considered the paragraph 61 impairment triggers reasonable and opined that any impairment trigger other than a decline in fair value below cost was likely to be arbitrary. 101. In my view, as the foregoing analysis shows, when properly construed in the context of other relevant paragraphs of HKAS 39, the meaning of HKAS 39.61 is straightforward and clear. As the authorities show, the meaning of a provision may emerge as unambiguous and clear at the end of what may appear to be a difficult process of construction. As Lord Wilberforce pointed out: “... ambiguity in this context is not to be equated with difficulty of construction, even difficulty to a point where judicial opinion as to meaning has differed. This is, I venture to think, elementary law.”[61] 102. And in Scammel v Ouston,[62] Lord Wright stated: “[The Court] will not be deterred by mere difficulties of interpretation. Difficulty is not synonymous with ambiguity so long as any definite meaning can be extracted.” 103. In the Written Case, the appellants also sought to argue that an exculpatory standard of reasonableness is introduced first, by construing section 34(1)(a)(vi) as taking effect in a manner reflecting auditors’ common law duties for the purposes of tortious liability in negligence;[63] and secondly, by construing it as a provision subject to a presumption of mens rea and thus importing an intermediate form of liability allowing for a defence of reasonable belief.[64] These submissions were (correctly) not pursued at the hearing and require no further discussion. D.5 Conclusion as to Questions 1 and 2 104. For the foregoing reasons, my answers to Questions 1 and 2 are respectively “Yes” and “No”. E. Disposal of the appeal 105. The auditors failed properly to evaluate Heng Tai’s failure to trigger HKAS 39.67 before issuing their unmodified audit opinion and thus failed or neglected to observe, maintain or otherwise apply the professional standards set out in HKSA 700. The Disciplinary Committee was entitled to find the complaint established while recognising that it merited only a token sanction. And the Court of Appeal was right to uphold the Disciplinary Committee’s ruling. 106. I would accordingly dismiss this appeal and make an order nisi that the appellants pay the respondent’s costs with liberty to the parties to lodge written submissions on costs within 21 days of the handing down of this judgment, the order nisi to stand as an order absolute without further direction in default of such written submissions. Mr Justice Tang PJ: 107. I agree with the judgment of Mr Justice Ribeiro Acting CJ. Mr Justice Fok PJ: 108. I agree with the judgment of Mr Justice Ribeiro Acting CJ. Mr Justice Bokhary NPJ: 109. I agree with the judgment of Mr Justice Ribeiro Acting CJ. Mr Justice Gummow NPJ: 110. I agree with the judgment of Mr Justice Ribeiro Acting CJ. Mr Justice Ribeiro, Acting CJ: 111. The Court unanimously dismisses the appeal and makes the orders referred to in paragraph 106 of this judgment. Mr Johnny Mok SC and Mr Adrian Lai, instructed by Wilkinson & Grist, for the 1st and 2nd Respondents (1st and 2nd Appellants) Mr Mark Strachan SC and Mr Jeffrey Chau, instructed by Reed Smith Richards Butler, for the Complainant (Respondent) [1] Defined as “... those non-derivative financial assets that are designated as available for sale or are not classified as (a) loans and receivables, (b) held-to-maturity investments or (c) financial assets at fair value through profit or loss.” (HKAS 39.9) [2] HKAS 39 “Financial Instruments: Recognition and Measurement”. [3] HKAS 39.43 and HKAS 39.46. [4] HKAS 39.55(b). [5] HKAS 39.55(b), HKAS 39.58 and HKAS 39.67. [6] HKAS 39.69. [7] Annual Report 2009, Note §3(k)(ii). [8] The relevant paragraphs of HKAS 39 are set out in Sections C.1 to C.3 below. [9] For the year ended 30 June 2009, CZ recorded a turnover of $1,095,614,000 and a net profit of $205,860,000. [10] Pursuant to section 23(3)(b) of the FRC Ordinance (Cap 588). [11] “HKSA 700” is the Hong Kong Standard on Auditing 700. Those paragraphs are set out in Section D.3 below. [12] Cap 50. The relevant sections are set out in Sections C and D below. [13] Under PAO section 41(1)(b)(iii). [14] Cheung CJHC, Cheung and Yuen JJA, [2016] 4 HKLRD 763. [15] Cheung CJHC, Cheung and Yuen JJA, CACV 233/2015 (24 October 2016). [16] Tang and Fok PJJ, Bokhary NPJ, FAMV 51/2016. [17] Appearing with Mr Adrian Lai for the appellants. [18] Appearing with Mr Jeffrey Chau for the respondent. [19] HKSA 700.13. HKSA 700.11 requires the auditor to “evaluate the conclusions drawn from the audit evidence obtained as the basis for forming an opinion on the financial statements”. [20] Judgment §§4.14 and 4.17-4.23. [21] PAO section 2. [22] PAO section 34(1)(a)(vi). [23] HKAS 39.1. [24] HKAS 39.43 relevantly provides: “When a financial asset or financial liability is recognised initially, an entity shall measure it at its fair value ...” [25] HKAS 39.45: “For the purpose of measuring a financial asset after initial recognition, this Standard classifies financial assets into the following four categories defined in paragraph 9 ... (d) available-for-sale financial assets. These categories apply to measurement and profit or loss recognition under this Standard.” HKAS 39.46: “After initial recognition, an entity shall measure financial assets ... at their fair values, without any deduction for transaction costs it may incur on sale or other disposal [subject to exceptions which are not presently relevant].” [26] HKAS 39.9. [27] HKAS 39.48A. [28] HKAS 39, AG71. [29] Ibid. [30] Reasons for Decision §§5.1-5.2. [31] Judgment §§4.34-4.36 and 4.39-4.42. [32] Investigation Report §5.2.4. [33] Reasons for Decision §§4.1-4.5. [34] Judgment §§4.31-4.33. [35] In Section C.6a below. These pronouncements are stated to be “Accompanying but not forming part of HKAS 39”. [36] The Board’s Conclusions are relevant because they discuss International Accounting Standards from which the HKAS are derived, generally verbatim. [37] Judgment §4.33. [38] Note 4(j) stated: “Impairment of available-for-sale financial assets. The Group’s available-for-sale financial assets are listed investment stated at fair value based on the quoted market price. In determining the impairment, the directors have reviewed the audited financial information of the listed investment and justified its business operation is prospective and profitable. The directors have exercised their judgement and are satisfied that no impairment is required for the investment”. [39] Appellants’ case §81. [40] Ibid, §86 (emphasis in the original). [41] Appellants’ case §93 (emphasis in the original). [42] Ibid, §94. [43] Introduction to Basis for Conclusions (amended as at December 2007). [44] Preface to Hong Kong Financial Reporting Standards (October 2006), Introduction §2. [45] At §81. [46] Section C above. [47] Set out in section 34(1)(a)(i), (ia), (ib), (ic), (ii) and (iii). [48] Section 34(1)(a)(xi) and (xii). [49] Section 34(1)(a)(iv), (v), (vi), (vii), (viii), (ix) and (x). [50] Not presently relevant. [51] In Section B of this judgment. [52] PAO section 35 provides a Disciplinary Committee with a range of possible orders on finding a complaint substantiated, running from permanent removal from the register down to a reprimand. [53] [2003] 1 WLR 2259 (PC) at §38. [54] In Section B. [55] Issued October 2006. [56] HKSA 700.6. [57] Hong Kong Standard on Auditing 200 (Revised): Objective and General Principles Governing an Audit of Financial Statements (October 2006). [58] Issued by the Institute in October 2006. [59] The Institute. [60] Hong Kong Financial Reporting Standards, which include all HKASs: p3. [61] L Schuler AG v Wickman Machine Tool Sales Ltd [1974] AC 235 at 261. [62] G Scammel and Nephew, Ltd v H C and J G Ouston [1941] AC 251 at 268. [63] As set out in cases like Re London and General Bank Ltd ex p Theobald (No 2) [1895-9] All ER Rep 953 (ChD); Re Kingston Cotton Mill Company (No 2) [1896] 2 Ch 279 (ChD); In re City Equitable Fire Insurance Company, Limited [1925] Ch 407 (ChD) and Caparo Industries Plc v Dickman [1990] 2 AC 605 (HL). [64] As discussed in Hin Lin Yee v HKSAR (2010) 13 HKCFAR 142. 
The Court: 1. On 10 September 2014 we dismissed this appeal at the end of the appellant’s argument, without calling on the putative respondent and the putative intervener and announced that our reasons for judgment would be published on a date to be notified. We also ordered that the parties file and serve written submissions as to costs within fourteen days of the publication of our reasons. The paragraphs which follow set out our reasons for judgment. The questions arising in the this appeal 2. This appeal raises two questions of fundamental importance. They relate to the amenability of the processes of the Legislative Council (“the LegCo”), in particular decisions made by its President in the course of the legislative process, to review by the courts of Hong Kong. 3. The Appeal Committee of this Court granted the appellant, who is a member of the Council, leave to appeal from a decision of the Court of Appeal on the basis that the following two questions of great general or public importance are involved in this appeal: (1) Having regard to the Basic Law (“BL”) and the Rules of Procedure (“the Rules”) of the LegCo, under what circumstances may a decision of the President of the LegCo made during the legislative process be judicially reviewed? (2) In light of the answer to the 1st Question above, is the decision of the President of the LegCo on 17 May 2012 to close the debate of the committee of the whole Council on the amendments to the Legislative Council (Amendment) Bill 2012 (“the Bill”), purportedly pursuant to Article 72(1) of the BL and Rule 92 of the Rules, amenable to judicial review? The facts 4. For the purpose of these reasons, it is sufficient to state the facts as they were recorded in the judgment of Cheung CJHC. 5. The Bill was introduced by the Administration into the LegCo for first reading on 8 February 2012. The object of the Bill was to disqualify a person who has resigned as a member of the LegCo from standing for a by‑election to be held within 6 months of his resignation. After the second reading of the Bill was moved, it was adjourned under rule 54(4) of the Rules of the LegCo. The Bill was then referred to the House Committee, which set up a Bills Committee to study the Bill. The debate on the Bill before the Council was scheduled to resume on 2 May 2012. In the meantime, two LegCo members had been given permission by the President of the LegCo (“the President”) to move respectively 1,232 and 74 committee stage amendments to the Bill at the resumed debate. The 1,232 amendments proposed by the first member dealt with six themes – five were concerned with situations where the disqualification would not apply and the sixth proposed a reduction of the disqualification period. As for the 74 amendments proposed by the second member, they sought to improve on the language of the Chinese text of the Bill. 6. The avowed intention of the two legislators and their ally, the appellant (also a legislator), for the introduction of these numerous amendments was to filibuster the Bill, which they opposed and which they apprehended would otherwise be passed by the majority in the Council. 7. The motion for the second reading of the Bill was passed on 2 May 2012 after a debate that lasted 8 hours 39 minutes. The Committee stage of the Bill before the whole Council commenced the next day at 9:00 am but was adjourned as the meeting lacked a quorum. The Committee of the whole Council resumed to deal with the Bill in the late afternoon of 9 May 2012. There was a motion to adjourn the proceedings of the Committee which was eventually negatived after a debate that took 4 hours and 29 minutes and straddled two days. The Committee then proceeded to debate on the clauses of the Bill and all the committee stage amendments. 8. The debate took place at meetings of the Committee of the whole Council which were, in accordance with the Rules, presided over by the President as chairman. 9. By 4:30 am on 17 May 2012, the debate had gone on for over 33 hours and still no end to the debate was in sight. The President had on numerous occasions considered the speeches made by the filibusters (the appellant was one of them) which were irrelevant to the clauses and amendments, and had made decisions and rulings accordingly. In this situation, a Legislative Councillor made reference to the procedure called “closure motion” in other legislative bodies and suggested the President should conclude the debate immediately. 10. The President reviewed the situation and indicated his inclination to allow the members and the Government official who had also proposed a committee stage amendment to give concluding speeches and then end the debate. After hearing views from members further, the President announced his decision to that effect at 9:00 am on the same day (17 May 2012), and gave all those involved until 12:00 noon to conclude the debate. He based his decision on rule 92 of the Rules which relevantly provides that “in any matter not provided for” in those rules, the practice and procedure to be followed shall be as decided by the President. The debate duly ended at noon time. Thereafter, the amendments to the Bill proposed by the legislators were put to vote. They were all defeated. The voting itself took several days to complete. The Bill eventually became law on 1 June 2012. The proceedings in the courts below 11. Aggrieved by the President’s decision to end the debate and thus the filibustering exercise, the appellant sought leave to apply for judicial review of the President’s decision on the same day. After an urgent oral hearing attended by all parties concerned (including the Secretary for Justice as interested party) on 17 and 18 May 2012, Lam J announced his decision to refuse leave on 19 May 2012. He gave his written reasons on 25 May 2012, before the passage of the Bill on 1 June 2012. 12. Lam J refused leave to apply for judicial review, giving comprehensive reasons for the refusal of leave. He rejected the appellant’s argument that art 73(1) of the BL conferred a constitutional right to participate in the legislative processes of LegCo on a member of LegCo and also required the courts to exercise jurisdiction to review the procedural regularity of those processes, and held that, even if the Appellant had a good case for constitutional challenge, there was no reason for the court to entertain it at the pre-enactment stage. After consideration of written submissions lodged by the parties, his Lordship ordered the appellant to pay the President’s costs of the application and made no order as to costs between the appellant and the Secretary for Justice. 13. The appellant appealed to the Court of Appeal against the orders made by Lam J. The appellant’s notice of appeal was filed on 2 June 2012, the day after the passage of the Bill on 1 June 2012. A consequence of these events was that the appellant’s challenge to the President’s decision to terminate the debate was treated as having beenconverted from a pre-enactment challenge to the President’s decision to close the debate under rule 92 to a post-enactment challenge to the validity of the Legislative Council (Amendment) Act 2012 based on the alleged invalidity of the President’s decision. 14. The Court of Appeal unanimously dismissed the appeal and ordered the appellant to pay the costs of both respondent and intervener. The principal judgment was delivered by Cheung CJHC, with Kwan JA agreeing and Poon J agreeing in a short separate judgment. In the Court of Appeal, the appellant submitted, as he had submitted before Lam J, that, as a member of LegCo, he had a right under the BL and the Rules to participate in the processes of LegCo, that the President’s decision to close the debate denied this right and that the superior courts of Hong Kong have jurisdiction to review the processes of LegCo. The Court of Appeal rejected this submission, as Lam J had done at first instance, rejecting the appellant’s arguments based on art 73(1) of the BL and applying the principle that the courts do not intervene to review the internal processes of the legislature. The Court of Appeal refrained from expressing any view on rule 92 and the other rules referred to in argument. The appellant’s argument in this Court 15. Mr Martin Lee, SC for the appellant submitted that art 73(1) confers on the appellant as a member of LegCo a right to participate in the legislative processes of LegCo and makes compliance with the provisions of the BL and the Rules a condition of the validity of legislation enacted by LegCo. The next step in his argument was to say that rule 92 conferred no power on the President to close the debate and that, accordingly, non-compliance with the Rules vitiated the amendments which were subsequently enacted because art 73(1) stipulates that laws be made “in accordance with the provisions of this Law and legal procedures”. 16. The appellant’s argument in this Court differs from the argument presented for the appellant in the courts below in that the case presented here draws on the case law of the Israeli High Court of Justice on the Israeli Basic Law: The Knesset (1958) and the Israeli Basic Law: The Judiciary (1984). According to the appellant, the Israeli cases show that Israel has recognized that a member of the Knesset, the Israeli legislature, has a right to participate in the legislative processes of the legislature and has rejected the English approach that parliamentary proceedings are excluded from the range of judicial review. Instead, the Israeli High Court of Justice has adopted as the test for judicial intervention in the legislative process the question whether the defect that has occurred in the legislative process is a defect that goes to the heart of the process. There was in this case, so the appellant submitted, such a defect because the President was not entitled to rely on rule 92. 17. On the other hand, according to the printed cases of the putative respondent (“respondent”) and the putative intervener (“intervener”) the appellant should not be permitted to rely in this Court for the first time on the Israeli jurisprudence and the recognition in that jurisprudence of the principle that the courts will intervene to redress a defect that has occurred in the legislative process which goes to the heart of the process. To permit the appellant to do so, so the respondent and the intervener contended, would be to allow the appellant to raise new matter which was not considered by the courts below. According to the printed cases of the respondent and the intervener, the Israeli decisions should not be followed in Hong Kong because they are at variance with accepted common law principle and consequently both Lam J and the Court of Appeal were correct in rejecting the appellant’s case. 18. Because we came to the conclusion at the end of the appellant’s argument that the appeal must fail, even if the appellant’s reliance on the Israeli jurisprudence were taken into account, we did not call on the respondent and the intervener. It follows that we did not hear them in support of their preliminary objection to the use of the Israeli materials by the appellant. It is therefore important that we make the point that we have assumed, without deciding, that those materials are properly before the court in support of the appellant’s case. Later in these reasons, we state the grounds for our conclusion that as a matter of common law principle and public policy the case for the adoption of the Israeli jurisprudence is not compelling. Article 73(1): Does it confer a right on the appellant to participate in the legislative processes of LegCo? 19. Art 73(1) provides : “The Legislative Council of the Hong Kong Special Administrative Region shall exercise the following powers and functions; (1) To enact, amend or repeal laws in accordance with the provisions of this Law and legal procedures” The expression “legal procedures” plainly includes the Rules. 20. In seeking to extract from this provision the conferring of a right on an individual member to participate in LegCo’s legislative processes, Mr Martin Lee, SC argued that a grant of law-making power to LegCo necessarily gives to its members an individual constitutional right to participate in its legislative processes “in accordance with the provisions” of the BL “and legal procedures”. This right, it was claimed, embraced the right to speak at LegCo meetings but it did not include the right to engage in a filibuster, as Mr Lee SC correctly conceded. 21. The problem with this argument lies in the nature and language of art 73 taken as a whole. The purpose of the article, as is apparent from its language, is to confer certain powers and functions on LegCo as a law-making body, that is, as an institution. The article is not directed to the powers, let alone the rights, of individual members of LegCo. There is no reference in art 73, as there is in art 74[1], to members in their individual capacities. 22. That the purpose of art 73 is not to confer rights on individual members of LegCo to participate in its processes is supported by art 75[2] which authorises LegCo to make its rules of procedure “on its own” and by the extensive powers conferred on the President by art 72. The two articles indicate that LegCo is to have exclusive authority in determining its procedure and that the President is to exercise his power to “preside over meetings” under art 72 so as to ensure the orderly, efficient and fair disposition of LegCo’s business. 23. The consequences of the interpretation of art 73(1) advanced on behalf of the appellant are so daunting as to invite, if not demand, its rejection. The appellant’s interpretation would open the door to the courts so that any member of LegCo who was dissatisfied with the way in which the Rules were applied to him, or with rulings of the President, could seek relief from the courts by way of judicial review, not only post-enactment, but more importantly, pre-enactment. This prospect would be extremely damaging to the orderly, efficient and fair deliberations and working of LegCo. Its proceedings would be liable to disruption, delays and uncertainties occasioned by applications for judicial review, judgments and appeals. 24. This point, which was well made by the Court of Appeal in this case, is supported by the judgment of Binnie J who delivered the judgment of the Supreme Court of Canada in Canada (House of Common) v Vaid[3] where a similar or related question arose. Binnie J had this to say: “20. …It would be intolerable, for example, if a member of the House of Commons who was overlooked by the Speaker at question period could invoke the investigatory powers of the Canadian Human Rights Commission with a complaint that the Speaker’s choice of another member of the House discriminated on some ground prohibited by the Canadian Human Rights Act, or to seek a ruling from the ordinary courts that the Speaker’s choice violated the member’s guarantee of free speech under the Charter. These are truly matters ‘internal to the House’ to be resolved by its own procedures. Quite apart from the potential interference by outsiders in the direction of the House, such external intervention would inevitably create delays, disruption, uncertainties and costs which would hold up the nation’s business and on that account would be unacceptable even if, in the end, the Speaker's rulings were vindicated as entirely proper.” 25. Accordingly, we concluded that for the reasons already stated, art 73(1) does not confer on a member of LegCo a constitutional right to participate in its legislative processes by speaking. We agree with the conclusion reached by the Court of Appeal on this point and with Hartmann J in Leung Kwok Hung v President of the Legislative Council who said: “The powers and functions described in art 73 are not given to members of LegCo as individuals but to LegCo itself sitting as a legislative body”[4] Article 73(1): Does it mandate the exercise of jurisdiction by the Hong Kong courts to ensure compliance with the Rules in the legislative processes of LegCo? 26. Mr Martin Lee SC’s submission on this point was that the grant of law-making power is conditioned by the words “in accordance with” on compliance with the Rules and with rulings by the President pursuant to his powers. The answer to this submission is to be found in the provisions of the BL and in the principles of the common law governing the independence and autonomy of legislatures and the jurisdiction of the courts to intervene in matters concerning the internal processes of such law-making bodies. 27. It is convenient to deal, first, with the relevant common law principles. Although these principles have their origin in the power, privileges and characteristics of the Parliament, in particular the House of Commons, in the United Kingdom and in the recognition by the courts of the exclusive authority of the Parliament to determine its own internal procedures, it is preferable now to regard the relationship between a legislature and the courts as an outcome of the application of the doctrine of the separation of powers. This doctrine is a common law doctrine which, in the case of Hong Kong, is reinforced by the constitutional separation of powers provided for by the BL. Sections 1 and 2 of Ch IV “The Political Structure” set out the powers and functions of the Executive and Executives Agencies while Sections 3 and 4 make similar provision for the Legislature and the Judiciary. Art 2 of the BL also recognises the separation of powers in providing: “The National People's Congress authorizes the Hong Kong Special Administrative Region to exercise a high degree of autonomy and enjoy executive, legislative and independent judicial power, including that of final adjudication, in accordance with the provisions of this Law.” 28. In construing and applying the provisions of the BL, it is necessary not only to apply common law principles of interpretation but also principles, doctrines, concepts and understandings which are embedded in the common law. They include the doctrine of the separation of powers and, within it, the established relationship between the legislature and the courts. This relationship includes the principle that the courts will recognise the exclusive authority of the legislature in managing its own internal processes in the conduct of its business, in particular its legislative processes. The corollary is the proposition that the courts will not intervene to rule on the regularity or irregularity of the internal processes of the legislature but will leave it to determine exclusively for itself matters of this kind (“the non-intervention principle”)[5]. 29. The strength of this proposition rests not only on principle and authority but also on public policy. In Hong Kong, LegCo has as its primary responsibility its law-making function. It also has vested in it other important powers and functions under art 73, for example: “ (2) To examine and approve budgets introduced by the government; (3) To approve taxation and public expenditure; (4) To receive and debate the policy addresses of the Chief Executive; (5) To raise questions on the work of the government; (6) To debate any issue concerning public interests;” 30. The important responsibilities of LegCo, notably its law-making function, require, as with other legislatures, that it should be left to manage and resolve its own internal affairs, free from intervention by the courts and from the possible disruption, delays and uncertainties which could result from such intervention. Freedom from these problems is both desirable and necessary in the interests of the orderly, efficient and fair disposition of LegCo’s business. 31. The adoption of the principle of non-intervention by the courts will reduce, if not eliminate, the prospect of pre-enactment challenge to proceedings in LegCo. It will also reduce, if not eliminate, post-enactment challenges to the validity of laws made by LegCo based on irregularity in its proceedings, unless such an irregularity amounts to non-compliance with a requirement on which the validity of a law depends. 32. In this respect it is important to recognise that the principle of non-intervention is necessarily subject to constitutional requirements. The provisions of a written constitution may make the validity of a law depend upon any fact, event or circumstance they identify, and if one so identified is a proceeding in, or compliance with, a procedure in the legislature the courts must take it under its cognizance in order to determine whether the supposed law is a valid law[6]. In Australia, Cormack v Cope was such a case. There s 57 of the Australian Constitution provided a means of resolving a deadlock between the two Houses of Parliament culminating in a joint sitting of the two Houses to deliberate and vote upon a proposed law. But the section prescribed a procedure to be followed and compliance with that procedure was a condition of the validity of the proposed law when enacted. 33. Although the principle of non-intervention is directed to pre-enactment judicial intervention in the legislative processes, the grounds on which the principle is based have generated a strong related principle of interpretation or presumption. That principle or presumption is that the courts will lean against an interpretation of a constitutional provision that makes compliance with procedural regularity in the law-making processes of a legislature a condition of the validity of an enacted law. 34. It follows that art 73(1) is to be interpreted in the light of the foregoing background of common law principle and policy considerations. In this respect, we refer to the judgment of Lord Cooke of Thorndon P in Ah Chong v Legislative Assembly of Western Samoa[7] where he said with reference to the principle of non-intervention: “…like all principles this one has its limits and they are not always easily discernible. One limit must be that a written constitution such as that of Western Samoa may place upon the Courts some duty of scrutinising Parliamentary proceedings for alleged breaches of constitutional requirements. Thus, while normally it is for a legislative assembly to determine the effect of its own standing orders and to depart from them if the assembly sees fit, a Constitution may displace that presumption by making compliance with the standing orders a condition of the validity of the legislation or, no doubt, of the validity of other steps taken by the assembly. But we agree with McLelland J in Namoi Shire Council v Attorney-General for New South Wales [1980] 2 NSWLR 639, 645, that the Court would lean against such an interpretation, an approach also to be seen as suggested by the Niue Court of Appeal in the judgment already cited. In the present case Sapolu CJ would have required ‘irresistible clarity’. Possibly, in our respectful opinion, that puts the test a little high, but certainly any real ambiguity would be resolved in favour of non-intervention.”[8] 35. We refer also to the decision of the Privy Council in The Bahamas Methodist Church v Symonette because it illustrates the force of the interpretive principle favouring an interpretation of a Constitution which does not make compliance with procedural requirements in the legislative process a pre-condition of the validity of a law. There art 59(1) of the Constitution of the Bahamas enabled any member of the House, subject to the Constitution and the rules of procedure to introduce a Bill, or propose a motion for debate or present a petition. The article also provided that any proceeding initiated by a member should be debated and disposed of “according to the rules of procedure of the House”. The Privy Council held that the first provision did not make compliance with the rules a violation of the Constitution and that the second provision did not deprive either House of the power given by art 55(1) to regulate its own affairs. The Privy Council further held that clear language would be required to justify an interpretation of art 59(1) which opened to scrutiny the legislative processes of the House. 36. The critical aspect of art 73(1) is that it makes no attempt to address the question whether non-compliance with “legal procedures” will result in invalidity of a law which is enacted after non-compliance with such procedures. Because non-compliance with provisions of the BL will result in invalidity, it could be suggested that a similar consequence follows from non-compliance with “legal procedures”. Such a consequence, however, could not have been intended because it would entail the invalidity of a law enacted after a trivial or relatively minor infringement of the relevant “legal procedures”. 37. One possible answer to this problem would be to interpret the article so that invalidity results only when the infringement of the “legal procedures” is substantial, or to use the Israeli terminology, goes to “the heart” of the legislative processes. In our view, neither of these criteria is sufficiently precise to define the basis on which the courts should exercise a jurisdiction to intervene in the legislative process or to determine the validity of a law the enactment of which proceeds from non-compliance with the relevant procedure. In any event, it is a matter of conjecture to suggest that the intent of art 73(1) was to mandate the application of any such criterion. 38. Accordingly, we concluded that the provisions of art 73(1) are ambiguous on the point under consideration and that they do not displace the principle of non-intervention. Indeed, art 75 by providing that Rules shall be made by LegCo “on its own” supports the application of that principle here. Our conclusion on art 73(1) is, however, subject to one important qualification. 39. This qualification arises from the circumstance that, in the case of a written constitution, which confers law-making powers and functions on the legislature, the courts will determine whether the legislature has a particular power, privilege or immunity. In R v Richards; Ex parte Fitzpatrick v Browne[9], Dixon CJ, speaking for the High Court of Australia and with reference to the two Houses of the Australian Parliament, said: “it is for the courts to judge of the existence in either House of Parliament of a privilege, but, given an undoubted privilege, it is for the House to judge of the occasion and of the manner of its exercise”.[10] 40. Adopting a similar approach, in a case concerning the privileges of the Nova Scotia House of Assembly, McLachlin J said: “It is for the courts to determine whether necessity sufficient to support a privilege is made out.”[11] McLachlin J went on to say: “The test of necessity is not applied as a standard for judging the content of a claimed privilege, but for the purpose of determining the necessary sphere of exclusive or absolute ‘parliamentary’ or ‘legislative’ jurisdiction. If a matter falls within this necessary sphere of matters without which the dignity and efficiency of the House cannot be upheld, courts will not inquire into questions concerning such privilege. All such questions will instead fall to the exclusive jurisdiction of the legislative body. Thus the test of necessity for privilege is a jurisdictional test.”[12] The learned judge also observed that, if the courts went on to examine particular exercises of privilege, “they would trump the exclusive jurisdiction of the legislative body” It followed that the area for court review is “only at the initial jurisdictional level”.[13] 41. In referring to necessity, McLachlin J was addressing a situation in which it was essential to show that the existence of a particular privilege was necessary for a legislature to exercise its functions. The foundation for the approach taken by McLachlin J was the famous decision in Stockdale v Hansard[14]. 42. In Egan v Willis[15], Gaudron, Gummow and Hayne JJ referred to the two judgments referred to in paragraphs 39 and 40 with evident approval and went on to apply the principle enunciated in the two cases to the powers of a House of the New South Wales Parliament[16]. McHugh J likewise accepted the principle stated in the two judgments and applied it to the powers of the New South Wales House[17]. Kirby J was of a similar, albeit not precisely the same opinion[18], while Callinan J seems to have been of the same opinion[19]. In the result, the High Court of Australia held in Egan v Willis that a court may judge the existence of a power, privilege or immunity in a House of Parliament but may not examine the occasion and manner of its exercise and that a House of the New South Wales Parliament has power to suspend for a limited time a member of the House who refuses to produce a non-privileged document called for by the House. 43. Accordingly, our conclusion on this point was that, although art 73(1) does not make compliance with the Rules essential to the validity of the enactment of a law by LegCo and that it is for LegCo itself to determine its own procedures and how they will be applied, the courts will exercise jurisdiction to determine the existence of a power, privilege or immunity of LegCo. We also arrived at the conclusion that the courts will exercise jurisdiction to determine the existence of a power, privilege or immunity of the President of LegCo. We arrived at this conclusion in the light, not only of art 73(1), but also of the provisions of art 72 of the BL and the important powers and functions which it confers on the President, particularly the power to “preside over meetings”. The courts, however, will not exercise jurisdiction to determine the occasion or the manner of exercise of any such powers, privileges or immunities either by LegCo or the President. The appellant’s argument that rule 92 did not authorise the President to put an end to the filibuster and close the debate. 44. The appellant’s case on this question was elusive, to say the least of it. The appellant’s principal argument was that rule 92 could not authorise the President’s action because other rules, had they been invoked or applied, would have justified the President’s decision to close the debate. Rule 92 provides: “In any matter not provided for in these Rules of Procedure, the practice and procedure to be followed in the Council shall be such as may be decided by the President who may, if he thinks fit, be guided by the practice and procedure of other legislatures.” According to the President’s printed case, there is no rule which deals with a filibuster or authorises him to take any action with respect to a filibuster, so rule 92 applies. 45. The appellant contended that the President could have achieved the outcome which he in fact achieved by recourse to rules 34, 38(1a), 41(1), 45(1) and (2) and 57(4) (a) and (b). The appellant’s contention on this point was remarkable because it suggested that the dispute in this case was academic in the sense that the only dispute was about the basis for the decision taken by the President, not about the making of the decision itself. 46. Be this as it may, it is clear that the President has power to set limits to and terminate a debate. The existence of the power is inherent in, or incidental to, the power granted by art 72(1) to the President to preside over meetings, quite apart from rule 92. The rules of procedure for which provision is made by art 75, as far as they relate to the President and his powers and functions, are necessarily subject to the provisions of art 72 setting out his powers and functions. It is not for this Court to consider whether or not the power was properly exercised. Nor is it for us to determine whether the President’s decision constituted an unauthorized making of a rule of procedure, although, in passing, we observe that the argument had nothing to commend it. As for the rules which the appellant suggested should have been applied, no doubt the President kept them in mind for possible application to the situations to which they are addressed. The Israeli jurisprudence 47. Mr Lee SC sought to support his interpretation of art 73(1) by reference to the Israeli High Court of Justice decision in Israel Poultry Farmers Association v Government of Israel[20] where Beinisch J said: “22. The principle of participation, according to which each Knesset member has a right to participate in the legislative process, is also a basic principle in the legislative process of democracies. The principle of participation is merely a development of representative democracy and its application in parliamentary law … the Knesset acts through the parties and through the Knesset members. Therefore, in order to enable the Knesset to carry out its functions by virtue of the principle of democratic representation, each Knesset member should be allowed to participate in the parliamentary proceedings that are required in order to carry out these functions.”[21] 48. The recognition of the principle of participation, which seems to be well-entrenched in Israeli jurisprudence, is inextricably linked with the exercise by the Israeli High Court of Justice of a jurisdiction to intervene in the legislative process at the pre-enactment stage when there is a “defect that goes to the heart of the process”. According to Mr Martin Lee, SC, the recognition of exercise of this jurisdiction has evolved in a series of a cases, the principal decisions being MK Sarid v Chairman of the Knesset[22], Litzman v Knesset Speaker[23] and the Poultry Farmers Case[24] to which we have already referred in connection with the principle of participation. These decisions have rejected the English principle that parliamentary proceedings are excluded from the range of judicial review and, instead, have recognised that the Israeli High Court of Justice has jurisdiction to intervene in the legislative process at the pre-enactment stage when there is “a defect that goes to the heart of the process”.[25] 49. In the Poultry Farmers Case, Beinisch J pointed out “What is a ‘defect that goes to the heart of the process’ is not decided in accordance with the classification of the defect as a defect of ultra vires or as a formal violation of a certain section in the Knesset Procedure Rules, but in accordance with the strength of the violation that this defect causes to ‘major values of our constitutional system’ or to basic values of our constitutional system that underlie the legislative process…”[26] Such an approach, according to the learned judge, would restrict pre-enactment judicial intervention, to “serious and rare defects”[27]. 50. There are other aspects of the Israeli jurisprudence which differ from the traditional common law principles relating to judicial intervention in the legislative processes. Breaches of ordinary law, quite apart from breaches of the Israeli Basic Law, may prompt judicial intervention. Further, a defect which goes to the heart of the legislative process may not entail invalidity of the statute. It is said that one question which the Court should examine is whether the defects would have been passed but for the defect. It is also said that the Court should take into account the degree of reliance on the legislation, the extent of the reasonable expectations that it created and the consequences that will arise from declaring it void[28]. 51. In the light of this brief and no doubt less than adequate summary of the Israeli jurisprudence, based on the materials in the appellant’s printed case, we concluded that the case for adoption by this Court of the Israeli jurisprudence is less than compelling. In the first place, the Israeli approach to judicial intervention in the legislative processes is entirely at odds with the relevant and traditional principles of common law constitutionalism and the public policy on which they are based. Secondly, the principles governing such intervention by the Israeli High Court of Justice seemed to us, with great respect, to be insufficiently precise to offer firm guidance and to involve the making of judicial assessments of a kind which common law courts do not usually make. And we note that the appellant’s case contained no reference to a decided case in which the Israeli High Court of Justice has exercised its jurisdiction to intervene in the legislative processes. 52. Accordingly, we declined to adopt the Israeli jurisprudence. Conclusion 53. For the foregoing reasons we made the order dismissing the appeal. Mr Martin Lee SC, Mr Hectar Pun and Mr Carter Chim, instructed by JCC Cheung & Co., for the Appellant Mr Benjamin Yu SC and Mr Anthony Chan, instructed by Lo & Lo, for the Putative Respondent Mr Stewart K.M. Wong SC and Mr Jin Pao, instructed by the Department of Justice, for the Putative Intervener [1] Article 74 provides: “Members of the Legislative Council of the Hong Kong Special Administrative Region may introduce bills in accordance with the provisions of this Law and legal procedures. Bills which do not relate to public expenditure or political structure or the operation of the government may be introduced individually or jointly by members of the Council. The written consent of the Chief Executive shall be required before bills relating to government policies are introduced.” [2] Article 75 provides: “The quorum for the meeting of the Legislative Council of the Hong Kong Special Administrative Region shall be not less than one half of all its members. The rules of procedure of the Legislative Council shall be made by the Council on its own, provided that they do not contravene this Law.” [3] [2005] 1 SCR 667 [4] [2007] 1 HKLRD 387 at §4 [5] See the Bahamas Methodist Church v Symonette [2000] 5 LRC 196 at 207h-211a; Prebble v Television New Zealand Ltd [1995] AC 321 at 332-333; Rediffusion (Hong Kong) Ltd v Attorney General of Hong Kong [1970] AC 1136 at 1157; Cormack v Cope (1974) 131 CLR 432 at 453-454 [6]Clayton v Heffron (1960) 105 CLR 214 at 235; Bribery Commissioner v Ranasinghe (1965) AC 172 at 197-198; Rediffusion (Hong Kong) Ltd v Attorney General Hong Kong [1970] AC 1136 at 1156-1157; Cormack v Cope (1974) 131 CLR 432 at 452, 473 [7] [2001] NZAR 418 [8] ibid at p 427 [9] (1955) 92 CLR 157 [10] ibid at 162 [11] New Brunswick Broadcasting Co v Nova Scotia [1993] 1 SCR 319 at 382 [12] ibid at 383 [13] ibid at 384 [14] (1839) 9 Ad & El, 112 ER 1112 [15] [1998] 195 CLR 424 [16] ibid at § 27 [17] ibid at §§ 65-67 [18] ibid at §§ 133-134 [19] At §179 [20] [2004] Isr LR 383 [21] ibid at p 413, §22 [22] HCJ 652/81 [23] [2004] Isr LR 363 [24] [2004] Isr LR 383 [25] ibid at §16 at pp 407-408 [26] ibid; see also Litzman v Knesset Speaker [2004] Isr LR 363 at §16 [27] [2004] Isr 383 at §16 at p 408 [28] Israel Poultry Farmers Association v Government of Israel [2004] Isr LR 383 at §17 
Chief Justice Cheung: 1. I agree with the joint judgment of Mr Justice Fok PJ and Mr Justice Lam PJ. Mr Justice Ribeiro PJ: 2. I agree with the joint judgment of Mr Justice Fok PJ and Mr Justice Lam PJ. Mr Justice Fok PJ and Mr Justice Lam PJ: A. Introduction 3. As explained in Kam Leung Sui Kwan v Kam Kwan Lai[1] (which we shall refer to as “Yung Kee”), the statutory jurisdiction to wind up a foreign-incorporated company in Hong Kong is subject to self-imposed restraints that have been articulated as three threshold, or so-called core, requirements which must be satisfied before the court will exercise that jurisdiction.[2] These are that: (1) There must be a sufficient connection with Hong Kong, but this did not necessarily have to consist in the presence of assets within the jurisdiction; (2) There must be a reasonable possibility that the winding-up order would benefit those applying for it; and (3) The court must be able to exercise jurisdiction over one or more persons in the distribution of the company’s assets. 4. This appeal is concerned with the second of those three core requirements, the Appeal Committee having granted leave to appeal on the following question of law:[3] “Should the Hong Kong Court exercise its winding-up jurisdiction over foreign companies on the basis that a ‘benefit’ is made out under the 2nd core requirement if such ‘benefit’ does not arise as a consequence of the winding-up order being made, but rather, would only ever be realized if the winding-up order is either avoided or discharged? Put another way, is the ‘leverage’ created by the prospect of a winding-up (as opposed to the making of a winding-up order) a legitimate form of ‘benefit’ under the 2nd core requirement?” A.1 Background facts 5. The background facts can conveniently be taken largely from the summary provided in the judgment of the Court of Appeal.[4] 6. The appellant is a company incorporated in the Mainland of the PRC and its shares, both A and B shares, are listed on the Shenzhen Stock Exchange. It also has a primary listing of H shares on the Stock Exchange of Hong Kong Limited. It is registered as a non-Hong Kong company under Part 16 of the Companies Ordinance (Cap.622), having a place of business in Hong Kong. Its main businesses include paper manufacturing, forestry, finance and real estate. According to its audited financial statements for the year ended 2015, it had net assets in excess of RMB16 billion, and operated profitably. 7. The appellant and the respondent established a joint venture in the Mainland pursuant to an agreement entered into in October 2005. After disputes arose between them, in October 2012, the respondent commenced an arbitration against the appellant in Hong Kong pursuant to the joint venture agreement alleging breach of that agreement. In November 2015, the arbitral tribunal rendered an award in the arbitration ordering the appellant to pay damages of RMB167,860,000 to the respondent. The respondent having obtained an order from the Court of First Instance granting it leave to enforce the award in Hong Kong on 7 December 2015, the appellant applied to set aside that order and the award. In October 2016, M. Chan J dismissed that application as being totally without merit and ordered the appellant to pay the respondent’s costs on an indemnity basis.[5] The appellant did not appeal that decision. 8. On 18 October 2016, the respondent served a statutory demand on the appellant for (i) RMB273,450,830.10 in respect of contractual damages (including interest); (ii) US$3,807,956.09 in respect of legal fees, costs and interest thereon; and (iii) HK$3,545,075.02 in respect of fees payable to the Hong Kong International Arbitration Centre and the arbitral tribunal and interest thereon. The appellant did not pay any part of the amounts so demanded. 9. Instead, on 7 November 2016, the appellant applied ex parte and obtained an interim injunction order from Harris J to prevent the respondent from presenting a petition to wind it up. An originating summons to continue the injunction was presented the following day. On 11 November 2016, leave was granted to allow the originating summons to be amended to seek a declaration that, since the appellant is an unregistered company, the respondent would not be able to satisfy the three core requirements for the Hong Kong court to exercise its jurisdiction to wind up the appellant pursuant to section 327(3) of the Companies (Winding Up and Miscellaneous Provisions) Ordinance (Cap.32) and the interim injunction continued until the hearing of the amended originating summons. A.2 The CFI decision 10. The three core requirements have already been set out at [3] above. At the hearing of the amended originating summons before Harris J, the appellant accepted that the first and third core requirements were met in the present case and so the argument proceeded on the question of whether the second requirement, namely whether there was a reasonable possibility that the winding-up order would benefit those applying for it, was met. 11. Before Harris J, the appellant maintained that it conducted no business in or from Hong Kong and that its only connection with Hong Kong was its listing here. It contended it had no assets here, so nothing a liquidator could realise for the benefit of the respondent petitioner. Moreover, a Hong Kong liquidator would not be recognised in the Mainland, so the presence of substantial assets there (of over RMB16 billion) would not avail the respondent. There was therefore no reasonable prospect that the respondent would benefit from the making of a winding-up order in Hong Kong and it should take steps to enforce the award in the Mainland instead.[6] 12. For its part, the respondent contended that the second core requirement was satisfied either by the appellant’s H share listed status being a valuable and realisable asset in Hong Kong, or alternatively, by steps a liquidator might take to investigate a restructuring of the appellant that occurred in 2015 by which a formerly directly held Hong Kong subsidiary of the appellant became an indirect subsidiary by the interposition of Mainland and BVI companies and to recover assets held by that subsidiary for the benefit of the respondent and other creditors.[7] 13. Harris J found against the respondent in respect of both those arguments.[8] However, the judge ruled against the appellant on its amended originating summons, holding that the second core requirement was nevertheless satisfied because the leverage created by the prospect of a winding-up petition constituted sufficient benefit for the respondent[9] and, alternatively, because the second core requirement could be moderated on the basis that the appellant’s failure to pay the award amounted, in a non-technical sense, to contempt for the High Court of Hong Kong.[10] A.3 The presentation of the petition and its adjournment 14. In the light of the CFI Judgment, the interim injunction granted by Harris J lapsed and the respondent presented a winding-up petition in respect of the appellant in HCCW 175/2017. At the first hearing of the petition, pending an appeal by the appellant against the CFI Judgment, Harris J adjourned the petition sine die upon the appellant undertaking to procure a third party to pay into court the sums of HK$355,141,100.06 (the HKD equivalent of the amounts claimed in the statutory demand) and HK$33,971,332.38 (being interest thereon). Those sums were subsequently duly paid into court by a third party with payment out to be made upon further court order. A.4 The Court of Appeal decision 15. Before the Court of Appeal, the appellant argued, amongst other points, that the benefit from the prospect of a winding-up order is not a benefit from the making of an order. It was also contended that Harris J had been wrong to hold that the second core requirement could be moderated and to moderate it as he did. 16. The Court of Appeal upheld the judge’s decision, for reasons that will be explored later in this judgment, and dismissed the appeal. In short, the Court of Appeal agreed with Harris J that there was a real possibility of benefit to the respondent in the making of a winding-up order against the appellant.[11] They were disinclined to agree with the judge that the second core requirement was capable of moderation but ultimately concluded it was not necessary to consider whether, had it been possible to dispense with the second core requirement, this would have been an appropriate case in which to do so.[12] B. The appellant’s contentions in this appeal 17. The appellant’s answer to the certified question set out at [4] above is “No”.[13] In summary, there are two broad bases in the appellant’s case for that negative answer. 18. First, the appellant contends that the core requirements are jurisdictional restraints, and that their rationale is comity and the concomitant presumption against extra-territoriality. Because of this, the argument proceeds, the proper interpretation of the core requirements must be informed by their limitations and the principle of non-interference, and in particular that comity militates against accepting leverage as a proper benefit. This strand of the appellant’s case will be addressed below as “the appellant’s comity argument”. 19. Next, in addition to the appellant’s comity argument, the appellant separately contends that the second core requirement has always insisted that the benefit referred to has to be a benefit resulting from the making of the winding-up order and that there is no justification for departing from this understanding of the second core requirement. The appellant also argues that the authorities demonstrate that the nature of the benefit contemplated is property which is either money or something convertible into money, such as assets or choses in action, rather than an intangible benefit in a loose or broad sense, and there is no warrant to create new categories of benefit. This strand of the appellant’s case will be addressed below as “the ‘need for the making of a winding-up order’ argument”. C. Benefit under the 2nd Core Requirement C.1 The purpose of the 3 Core Requirements and the limited scope of this appeal 20. The statutory jurisdiction of the Hong Kong courts to wind up a foreign-incorporated company[14] is to be found in sections 327(1) and (3) of the Companies (Winding Up and Miscellaneous Provisions) Ordinance (Cap.32) (“the Ordinance”) which materially provide: “(1) Subject to the provisions of this Part, any unregistered company may be wound up under this Ordinance, … … (3) The circumstances in which an unregistered company may be wound up are as follows – (a) if the company is dissolved, or has ceased to carry on business, or is carrying on business only for the purpose of winding up its affairs; (b) if the company is unable to pay its debts; (c) if the court is of opinion that it is just and equitable that the company should be wound up.” 21. As indicated at [3] above, the three threshold or core requirements were approved by this Court in Yung Kee. There is no controversy in this appeal as to the articulation of the three requirements and there is no cause now to revisit them. The issue in this appeal is a narrow one, limited to the nature of the benefit that will satisfy the second requirement so that, if all three requirements are satisfied, the court may exercise its jurisdiction to make a winding-up order in respect of the company concerned. 22. At the outset, it is important to keep in mind that the three requirements articulated in Yung Kee are not statutory provisions and their interpretation should not be approached as if it were an exercise of statutory construction.[15] Their purpose, it should be remembered, is as self-imposed restraints on the exercise of jurisdiction to set in motion the court’s winding-up procedures over a body which prima facie is beyond the limits of territoriality[16] and there are many examples of the courts exercising that jurisdiction. 23. The Appellant refers to the three requirements as jurisdictional restraints. But they do not go to the existence of the jurisdiction to wind up foreign companies (which is entirely statutory). Rather, they are a judicially fashioned threshold for the exercise of the statutory jurisdiction to wind up a foreign company.[17] It is not normally necessary to distinguish between the existence of jurisdiction and the discretion to exercise it.[18] But in a context where it is necessary to draw the distinction,[19] it has been held in a series of decisions commencing with In re Drax Holdings Ltd[20] that the three requirements go to the discretion of the court, and not to the existence of its jurisdiction. The discretionary core requirements play a role in the international context, where the court will not wind up a foreign company in cases in which it has no legitimate interest to do so, for that would be to exercise an exorbitant jurisdiction contrary to international comity.[21] It is therefore preferable to refer to these requirements as threshold requirements, to avoid their description as “core” requirements being misunderstood as suggesting they go to the existence of the jurisdiction itself. 24. The conclusion, in any given case, that the self-imposed restraints do not preclude the existence of jurisdiction to wind up is to be distinguished from the subsequent question as to whether a winding-up order should actually be made: see [36] and [77] below. This distinction is well illustrated by the case of Yung Kee itself.[22] It is also illustrated in the present case by the fact that the appellant’s application below was an attempt to obtain an order of the Court of First Instance preventing the presentation of a winding-up petition by the respondent. C.2 The “need for the making of a winding-up order” argument 25. It is convenient to address this argument first. The appellant contends that, under the second requirement, the benefit must arise from the making of a winding-up order and not from any pressure, or leverage, arising before such an order is actually made. Therefore, in a case such as the present, where the company has no assets in Hong Kong, the fact that its H share listing on the Stock Exchange of Hong Kong means that it is vulnerable to commercial pressure to pay or procure payment to a creditor in respect of a debt incurred outside Hong Kong should not be held to constitute a legitimate benefit for the purpose of invoking the winding-up jurisdiction in Hong Kong. In fact, the appellant contends, the commercial pressure exerted by the presentation of the winding-up petition demonstrates a logical conundrum: the petitioner is invoking the jurisdiction precisely in order to force a commercial result (payment of the debt) that will result in a winding-up order not being made. Accordingly, the argument goes, the commercial pressure on a solvent company to pay or procure payment of the debt from assets outside Hong Kong, insofar as it is a benefit to the petitioner because it may lead to actual payment being made, is an illegitimate form of benefit for the purposes of the second requirement. 26. The principal basis for the appellant’s argument that, for the purposes of the second requirement, the benefit must flow from the making of a winding-up order and not otherwise, is the articulation of that requirement in Yung Kee, namely: “… there is a reasonable prospect that the petitioner will derive a sufficient benefit from the making of a winding-up order, whether by the distribution of its assets or otherwise, will always be necessary and will often be sufficient.”[23] 27. It is the appellant’s case that the necessary benefit must be a sufficient and tangible benefit that results from the making of a winding-up order, usually through the recovery of assets, whether money or tangible or intangible property convertible into money. In support, the appellant cites: (1) In re Compania Merabello San Nicholas S.A.;[24] (2) In re Eloc Electro-Optieck and Communicatie B.V.;[25] (3) International Westminster Bank plc v Okeanos Maritime Corp;[26] and (4) Re Real Estate Development Co.[27] 28. Whilst these cases did involve the recognition of a benefit arising from the making of a winding-up order, it does not follow that benefit for the purposes of the second requirement is limited only to one arising from that event. Indeed, as will be seen, in the case of an undisputed debt due and owing from a company to a creditor, there is a well-recognised statutory mechanism to seek payment of that debt by the presentation of a winding-up petition. C.3 The statutory demand mechanism to prove a company’s inability to pay its debts 29. One of the circumstances in which a company may be wound up is where it is “unable to pay its debts”: see section 327(3)(b) of the Ordinance in the case of an unregistered company, set out above, and section 177(1)(d) of the Ordinance, in materially similar terms, in the case of a registered company. 30. There is a statutory mechanism that deems a company unable to pay its debts. Under section 327(4)(a) of the Ordinance, it is provided that: “An unregistered company shall, for the purposes of this Ordinance, be deemed to be unable to pay its debts – (a) if – (i) a creditor, by assignment or otherwise, to whom the company is indebted in a sum then due that equals or exceeds the specified amount, has served on the company a written demand in the prescribed form requiring the company to pay the sum so due – (A) by leaving the demand at the principal place of business of the company; (B) by delivering the demand to any officer of the company; or (C) by otherwise serving the demand in any manner that the court may approve or direct; and (ii) the company has for 3 weeks after the service of the demand neglected to pay the sum, or to secure or compound for it to the reasonable satisfaction of the creditor”. A like deeming provision applies in relation to the service of a statutory demand in respect of a debt due from a registered company: see section 178(1)(a) of the Ordinance. 31. The statutory demand mechanism that finds its current version in the Ordinance was first introduced in England and Wales, in respect of registered companies, by section 68(1) of the Joint Stock Companies Act 1856. It was then extended to unregistered companies as well when that act was replaced by the Companies Act 1862: see sections 80(1) (for registered companies) and 199(4)(a) (for unregistered companies). 32. In Hong Kong, the statutory demand mechanism was introduced by section 79(1) of the Companies Ordinance (Ord. No. 1 of 1865) and, being based on the Joint Stock Companies Act 1856, was limited to registered companies. The 1865 Ordinance was replaced by the Companies Ordinance (Ord. No. 58 of 1911), which introduced provisions for the winding up of unregistered companies and extended the statutory demand mechanism to both registered (section 131(i)) and unregistered (section 246(1)(iv)(a)) companies. 33. The structure and wording of the relevant provisions in Hong Kong have remained largely identical since 1911. By the Companies (Winding Up and Miscellaneous Provisions) (Amendment) Ordinance 2016, an amendment was made to introduce a prescribed form for the statutory demand. That amending ordinance implemented certain recommendations made by the Law Reform Commission in its 1999 report on “The Winding-up Provisions of the Companies Ordinance”. In that report, the Law Reform Commission considered whether the three circumstances in which a company is deemed to be unable to pay its debts under section 178(1), including the non-compliance with a statutory demand, were adequate and concluded: “We are broadly satisfied that these three circumstances are adequate for the purposes of petitioning to wind-up a company that cannot or will not pay its debts. We note that Hong Kong is not alone in providing for a demand that is not based on a judgment as both the Insolvency Act, section 123(1)(a), and the Australian Corporations Law, section 459F, make similar provision. We consider that to change this approach would add considerably to the costs and time involved in recovering debts and we would not support any change. We have made a number of recommendations that should assist in making the process easier to understand and operate.”[28] 34. The statutory demand mechanism is a “convenient” method of establishing that a company is unable to pay its debts.[29] It operates as conclusive proof of the company’s inability to pay its debts for the purpose of establishing the court’s jurisdiction to make a winding-up order. Where the company is so deemed to be unable to pay its debts, it is perfectly proper for a creditor to present a winding-up petition in order to seek an order from the court to wind up the company. If the company is in fact solvent or has realisable assets, its winding up will eventually result in the creditor receiving a dividend from the company’s liquidator in satisfaction of the proof of debt lodged in the winding up, which may or may not satisfy the whole of that debt. Most creditors will no doubt hope that the matter is resolved more expeditiously. If the company is indeed solvent, the creditor will hope that the presentation of the petition itself will prompt the company to pay the debt before the matter proceeds to a winding up. As will be seen, case law recognises the propriety of the use of a winding-up petition as a means of applying commercial pressure to seek payment of an undisputed debt. 35. In BNY Corporate Trustee Services Ltd v Eurosail-UK 2007-3BL plc, Lord Walker of Gestingthorpe JSC noted: “A company’s non-compliance with a statutory demand, or non-satisfaction of execution of a judgment debt, is a matter that can be proved quite simply, usually by a single short witness statement. If proved, it establishes the court’s jurisdiction to make a winding up order, even if the company is in fact well able to pay its debts.”[30] And similar practical reasons for using a statutory demand to prove insolvency and to prompt the debtor to satisfy a debt are summarised in Insolvency Litigation: A Practical Guide, 3rd Ed., (Sweet & Maxwell, 2021) at [3-005]: “It is not in fact necessary for a creditor to serve a statutory demand on a debtor company before presenting a winding-up petition to the court. … Creditors do however continue to use the statutory demand route as a precursor to the presentation of a winding-up petition. The receipt of such a demand can focus the debtor’s mind on having to deal with the debt or dispute because of the consequences if it fails to do so. It can also flush out whether there is really any valid dispute in relation to the debt, particularly where the debtor has been reluctant to deal with the issue. Statutory demands are relatively cheap and quick to prepare in relation to the impact of receiving one.” 36. Although failure to comply with a statutory demand establishes, without the need for further proof, an undisputed debt as the basis for petitioning for the winding up of a company, it does not follow that a winding-up order will necessarily be made. Whether or not the court will ultimately make a winding-up order on the petition is a separate question but, even if the court declines to make the order, it does not follow that the circumstances conferring the winding- up jurisdiction on the court do not exist. The separate question of whether the court will make a winding-up order arises because the right to an order on proof of a company’s inability to pay its debts is subject to the wishes of the company’s creditors and contributories generally. As Buckley J explained in In re Crigglestone Coal Company Limited: “But then comes another consideration, viz., that the order which the petitioner seeks is not an order for his benefit, but an order for the benefit of a class of which he is a member. The right ex debito justitiæ is not his individual right, but his representative right. If a majority of the class are opposed to his view, and consider that they have a better chance of getting payment by abstaining from seizing the assets, then, upon general grounds and upon s.91 of the Companies Act, 1862, the Court gives effect to such right as the majority of the class desire to exercise. This is no exception. It is a recognition of the right, but affirms that it is the right not of the individual, but of the class; that it is for the majority to seek or to decline the order as best serves the interest of their class. It is a matter upon which the majority of the unsecured creditors are entitled to prevail, but on which the debtor has no voice.” [31] 37. Failure to comply with a statutory demand therefore operates as conclusive proof of insolvency for the purposes of engaging the jurisdiction to wind up a company. That the jurisdiction may ultimately not be exercised in favour of the making of a winding-up order does not diminish the purpose and effect of the statutory demand mechanism. So understood, there is no tension between failure to comply with a statutory demand operating as conclusive proof of insolvency (regardless of whether the company is, in fact, solvent) and the principle against using winding-up proceedings as a means of debt collection for disputed debts. As Harman J observed in Cornhill Insurance plc v Improvement Services Ltd: “… in my judgment each defendant was entitled to say: ‘I am undoubtedly owed £1,154. If you don’t pay me I must suspect you can’t. Therefore I can properly swear that you are insolvent and I can properly present a winding up petition to the Companies Court’.”[32] 38. The statutory demand mechanism is accordingly of great importance as a basis for establishing the jurisdiction to petition for the winding up of a company. C.4 Invocation of the winding-up jurisdiction as a legitimate means to enforce payment of an undisputed debt 39. It is important to draw a crucial distinction between those debts which are undisputed and those which are disputed. The presentation of a winding-up petition on the basis of an undisputed debt gives rise ex debito justitiæ to a winding-up order: In re Crigglestone Coal Company Limited.[33] Where, however, a debt is disputed, the presentation of a winding-up petition may amount to an abuse of the process of the court. As Kwan JA (as she then was) observed in Re Leung Cherng Jiunn (debtor): “It is well established that petitions are not meant to be used for the purpose of debt collection and the winding-up or bankruptcy jurisdiction of the court would be exercised only in very clear cases. Where oral evidence is required to decide a real and substantial dispute of fact, the court will dismiss the petition. And if there is an abuse of process in invoking the jurisdiction of the court in an improper manner, the petitioning creditor may be ordered to pay indemnity costs.” [34] 40. There is “a long line of authority … which confirms the right of a creditor owed an undisputed debt to petition the court for winding-up”[35] and “the courts have emphasised that a petition presented in order to bring pressure on a company to pay a debt which is indisputably due is perfectly proper”.[36] Indeed, debt repayment has been described as “the essential basis for a creditor’s petition”.[37] 41. In Mann v Goldstein, Ungoed-Thomas J considered the propriety of the presentation of a winding-up petition where a debt was disputed, holding that to invoke the winding-up jurisdiction when a debt is disputed on substantial grounds after it has become clear that it is so disputed would be an abuse of the process of the court. In the course of his judgment, however, he also addressed the appropriate course for the court when the creditor’s debt is clearly established, holding: “When the creditor’s debt is clearly established it seems to me to follow that this court would not, in general at any rate, interfere even though the company would appear to be solvent, for the creditor would as such be entitled to present a petition and the debtor would have his own remedy in paying the undisputed debt which he should pay. So, to persist in non-payment of the debt in such circumstances would itself either suggest inability to pay or that the application was an application that the court should give the debtor relief which it itself could provide, but would not provide, by paying the debt. Further, the winding-up order on the ground of inability to pay debts would be the very matter which it would be for the Companies Court to decide after presentation of the petition: and validly to present a creditor’s petition which the company inexplicably would not pay could hardly, in general at any rate, be an abuse of the process of the court.”[38] 42. The above passage supports the proposition that it is entirely proper to seek to enforce payment of an undisputed debt by the presentation of a winding-up petition. More importantly, the reference to “relief … by paying the debt” supports the further proposition that pressure on a debtor to pay an undisputed debt is a proper benefit of allowing a winding-up petition to proceed. 43. That further proposition is even more clearly stated in Re a company (No 006273 of 1992), where Millett J (as he then was) refused an injunction to prevent the presentation of a winding-up petition and held: “In my judgment, it is impossible to describe the presentation of a petition by the respondent based on an undisputed debt as an abuse of the process of the court. It is not threatened for an improper purpose, that is to say, in order to bring pressure on the company to pay a disputed debt. It is threatened for a purpose which is entirely proper, that is to say in order to bring pressure upon the company to pay a debt which is indisputably due.”[39] 44. The “vast majority” of petitions to wind up a company are creditors’ petitions[40] and as the Court noted in Yung Kee: “Creditors seek a winding-up order against their debtor in order to obtain payment in or towards satisfaction of their debts. The presence in Hong Kong of significant assets which may be made available to the liquidator for distribution among the creditors will usually suffice. The claim itself is usually simple to establish; the petitioner need prove only that he is owed a debt which is due and unpaid. If he considers it worthwhile he may chase the assets wherever they may be found and seek winding-up orders in different jurisdictions until his debt is satisfied.”[41] C.5 The benefit need not flow from an administration of the assets after a winding-up order is made 45. In the context of winding up domestic companies, the courts have always paid regard to the utility of exercising the discretion to make an order for the winding up of the company concerned. In In re Crigglestone Coal Company Limited,[42] Buckley J (as he then was) neatly explained the rationale for this: “The Court has often refused an order upon [the ground that the company had no assets to wind up], but not because it lies in the debtor’s mouth to say that he is not amenable to the jurisdiction because he has no property, but because the Court does not make an order when no benefit can result. If the order will be useful (not necessarily fruitful) there is jurisdiction to make it.” 46. For the winding up of foreign companies, Megarry J in In re Compania Merabello San Nicholas S.A.[43] applied the Crigglestone approach in the winding up of a Panamanian one-ship company. The only asset of the company appeared to be a claim against a mutual insurance club Oceanus. Instead of the administration of company’s assets by a liquidator, the petitioner’s aim was to procure the vesting of the Oceanus claim under the Third Parties (Rights against Insurers) Act 1930. His Lordship held that this would not exclude the court’s jurisdiction to wind up the company: “… on any footing the making of a winding up order will be beneficial to the petitioning creditors; and it cannot be said, and has not been said, that this is a petition which can bring no benefit to the petitioning creditors and is presented simply for the purpose of making costs. The probability of the liquidator having no assets to administer is not in my judgment a sufficient ground for saying that there is no jurisdiction to make the order. … In the case of English companies there is no need to lay a foundation for jurisdiction by showing that the company has assets here. In the case of foreign companies, the purpose of demonstrating that there are assets here is to establish jurisdiction. Once that is established, then I do not see why the Crigglestone approach should not apply. In other words, in the case of a foreign company, once it is shown that there is jurisdiction, the jurisdiction exists unless it is taken away by demonstrating that no possible benefit can follow from making a winding up order.”[44] 47. Subsequent cases clearly laid down that sufficient connection with the jurisdiction can be established by other means. The presence of significant assets within the jurisdiction is not the only means to do so though in most of the cases it would be sufficient.[45] Benefit only entered into the equation after sufficient connection establishing jurisdiction was shown. 48. Further, Megarry J reiterated that the winding-up mechanism need not be the only benefit of which the court should take heed: “… I think that in deciding whether to make a winding up order, as a process to secure payment of debts so far as possible, it would be wrong to say that the order should be made only if making it will effect payment in one of the ways that Parliament has provided and not in another way. Words in a judgment must be construed in relation to the subject-matter of the case in question, and not as if they were Acts of Parliament.”[46] 49. In the subsequent case of In re Eloc Electro-Optieck and Communicatie B.V.,[47]the petitioners sought the winding up of the company in England so that they could obtain payment from a redundancy fund maintained by the Secretary of State under the Employment Protection (Consolidation) Act 1978. Nourse J acknowledged that normally the court would not wind up a foreign company with no assets within the jurisdiction as there was nothing that could be realized for the benefit of the creditors. However, that was only a normal working rule whilst the fundamental principle, His Lordship held, is that “the court will not wind up a company if there is no likelihood that any advantage will be achieved by the petitioner.”[48] 50. An attempt to challenge the decision in Eloc was rejected by Peter Gibson J (as he then was) in In re a Company (No 00359 of 1987)[49]also reported as International Westminster Bank plc v Okeanos Maritime Corp.[50] The petitioner presented a petition against a Liberian company which had no assets in England. Sufficient connection was established without reference to the assets of the company. The petitioner could rely on a reasonable possibility of the liquidator succeeding in claims for fraudulent or wrongful trading as the relevant benefit. His Lordship affirmed the proposition that it was not necessary to show that the company had assets within the jurisdiction provided that there was a sufficient connection with the jurisdiction and there was a reasonable possibility of benefit accruing to the creditors from the winding up.[51] 51. The negative formulation by way of exclusion of jurisdiction in sub-paragraph (6) of Megarry J’s summary[52] thus evolved into a requirement of reasonable possibility of benefit where the foreign company had no assets within the jurisdiction.[53] Peter Gibson J’s re-formulation subsequently became the second “core” requirement in Re Real Estate Development Co.[54] Knox J agreed that it is possible to establish sufficient connection under the first “core” requirement in the absence of assets and activities carried on by the company within the jurisdiction.[55] Once sufficient connection is established, the courts focus on examining if “persons exist who are likely to benefit from the making of the order and who qualify for one reason or another as persons on whose behalf it would be right to set in motion the winding-up petition over a foreign company.”[56] Since, as discussed above, it is entirely proper to seek repayment of an undisputed debt by a statutory demand and winding-up petition, there is no justification for narrowly confining the relevant benefit to a consequence materialising only upon the making of the winding-up order itself as opposed to the invocation of the winding-up machinery by the presentation of a petition. 52. In Re OJSC Ank Yugraneft,[57] the petitioners sought to obtain a winding-up order in England regarding a Russian company which had already been declared by a court in Moscow to be insolvent with the appointment of a Russian bankruptcy administrator. In the meantime, proceedings were commenced in the English Commercial Court in the name of the company seeking damages for fraud concerning the reduction of the company’s interest in a joint venture. Christopher Clarke J (as he then was) held that the second “core” requirement was satisfied by the advantage of having the interest of the company represented by an English liquidator in the Commercial Court proceedings. Although the Russian administrator could act on behalf of the company in the Commercial Court proceedings, it was advantageous to have an English liquidator so that the proceedings would be supervised by a licensed insolvency practitioner conversant with English litigation and accountable and subject to supervision by the English Companies Court. His Lordship held that the winding-up jurisdiction against a foreign company can be invoked even if the same, or a similar, result can be achieved by other means.[58] 53. The courts have rejected attempts to focus on one single criterion (be it carrying on business in the jurisdiction, the establishment of a place of business or the presence of assets) in determining whether sufficient connection with the jurisdiction is established. In Yung Kee, Ma CJ and Lord Millett NPJ highlighted that the requirements have to be applied contextually in light of the facts of the case: “It must be remembered that the so-called core requirements are not statutory but self-imposed constraints adopted by the courts. In elucidating their meaning no question of statutory interpretation arises. The question is whether the connection of a company with Hong Kong is sufficient to justify the Hong Kong court in exercising its jurisdiction to wind that company up, and that is a question of degree. The nature of the connection will vary from case to case and is always a matter for the court.” [59] 54. Likewise, a pragmatic approach should be applied in assessing whether it would be useful to entertain a winding-up petition in respect of a foreign company. The benefit that a petitioning creditor can rely upon to satisfy the second requirement will vary from case to case depending on its facts. From the authorities discussed above, the following observations can be made: (1) There is no doctrinal justification for confining the relevant benefit narrowly to the distribution of assets by the liquidator in the winding up of the company; (2) It is sufficient that the benefit would be enjoyed solely by the petitioners; (3) There is also no doctrinal justification requiring the relevant benefit to come from the assets of the company; (4) There are cases where even though there was nothing for the liquidator to administer the courts did not find any difficulty in holding that the second requirement was satisfied so long as some useful purpose serving the legitimate interest of the petitioner can be identified; (5) The benefit need not be monetary or tangible in nature; and (6) The fact that a similar result could be achieved by other means does not preclude a particular benefit from being relied upon for the purposes of fulfilling the second requirement. 55. Such an approach accords with the practice adopted in the winding up of domestic companies. In Crigglestone the petition was opposed by debenture-holders who had a floating charge on all the property of the company. They had already obtained the appointment of a receiver and it was contended on their behalf that there were no assets available for distribution to the unsecured creditors. Buckley J ordered the company to be wound up because it could be useful to the unsecured creditors to have a liquidator representing their interest in defending the action brought by the debenture-holders and investigating into the circumstances under which some further loans were granted on the security of further debentures.[60] On appeal, the winding-up order was upheld. Collins MR held that a reasonable possibility that the unsecured creditors will derive an advantage from the winding up would be enough to support the making of the order.[61] Romer LJ agreed and held that it was sufficient that some good may result to the petitioners, thus echoing Buckley J’s view that the winding up had to be useful but not necessarily fruitful.[62] 56. A modern statement on the low threshold for the benefit requirement can be found in the judgment of Neuberger J (as Lord Neuberger of Abbotsbury then was) in Re Demaglass Holdings Ltd: “… in the case of an opposed petition, the petitioning creditor has to establish the possibility of the prospect of some sort of benefit from a winding up. The test, however appears to be a low one.” [63] C.6 Commercial pressure as a legitimate benefit under the second requirement 57. Once it is accepted that commercial pressure to achieve the repayment of an undisputed debt is an entirely proper purpose for a creditor’s winding-up petition (see Section C.4 above), it is difficult to see any principled basis for excluding that commercial pressure as a relevant benefit for the purposes of the second requirement. 58. It is to be recalled that Yung Kee concerned a shareholder’s petition to wind up a company on the just and equitable basis. The focus of such a petition is different to that of a creditor’s petition. As the Court stated: “The case of a shareholders’ petition is different. In the first place, the parties are different, for while the company is a necessary defendant to every winding-up petition, in the case of a creditor’s petition the dispute is between the petitioner and the company; whereas in the case of a shareholders’ petition, the dispute is between the petitioner and the other shareholders, and the company is the subject of the dispute rather than a party to it. Accordingly the presence of the other shareholders within the jurisdiction is an extremely weighty factor in establishing the sufficiency of the connection between the company and Hong Kong. In the second place, the petitioner’s purpose in seeking a winding-up order is also different, for the object is not so much to obtain payment of a debt but rather to realise the petitioner’s investment in the company. This claim may involve a lengthy and detailed examination of the management of the company’s internal affairs which a petitioner is unlikely to be willing to undertake more than once.”[64] 59. It is therefore understandable that the Court in Yung Kee was concerned to examine the consequences of the making of a winding-up order when considering whether the three requirements were satisfied so that there was a proper basis for invoking the winding-up jurisdiction. But it is inappropriate to extrapolate from those references (and in particular the articulation of the second requirement on which the appellant relies in this appeal) a proposition that in creditors’ winding-up proceedings, sufficient benefit can only be found in specified consequences flowing from the making of a winding-up order. 60. Reaching such a conclusion would fly in the face of the recognition of the propriety of invoking the winding-up jurisdiction as a means to bring commercial pressure to bear to obtain repayment of an undisputed debt and would also be an unduly narrow reading of the second requirement. The winding-up procedure is a process and it makes little sense to characterise the initiating steps of that process as somehow qualitatively different from steps taken in its continuation for the purpose of deciding whether to assume jurisdiction. It is illuminating to note that, in ReReal Estate Development Co, when considering the need for a sufficient connection between the company and the winding-up jurisdiction, Knox J posed the question, “sufficient for what?” To this, he gave the answer: “… sufficient to justify the court setting in motion its winding-up procedures over a body which prima facie is beyond the limits of territoriality.”[65] 61. There is no basis for reading the Court’s articulation of the second requirement in Yung Kee as restricting sufficient benefit in cases involving creditors’ petitions only to possible benefits flowing from the making of a winding-up order. The essential consideration in such cases is whether a reasonable possibility exists of a sufficient benefit accruing to the petitioner from being permitted to set in motion winding-up procedures in Hong Kong in respect of a non-Hong Kong company. That benefit will exist where setting those procedures in motion results in the payment of an undisputed debt. 62. The appellant contended that since the success of leverage is predicated upon a winding up being avoided by payment of the debt before the making of the winding-up order, such benefit does not fall within the scope of benefit recognized under the second requirement. 63. That argument rests on a false premise. There is no reason why the leverage would disappear upon the making of the winding-up order. It should be borne in mind that in the present case the issue of the exercise of the winding-up jurisdiction was considered by Harris J in the context of an application for an injunction to restrain the presentation of the petition. After His Lordship refused to grant the injunction, the petition was presented. The appellant then pre-empted the substantive hearing of the petition by seeking its adjournment on the condition of procuring a payment into court. Against that background, it is somewhat self-serving for the appellant to suggest that the success of leverage is predicated upon the winding up being avoided. But for the payment into court, there is little doubt that the respondent would have proceeded with the petition to seek a winding-up order. In principle, it is difficult to see why the court should not take account of leverage as a potential benefit in deciding whether to make such order at the hearing of the petition. 64. As litigants are at liberty to resolve civil disputes by alternative means, leverage is always in the background of any instance of civil litigation and a winding-up petition is no exception. Even after the making of a winding-up order, leverage can continue to operate in the form of a stay of the winding-up proceedings upon the satisfaction of the debts of the petitioning and supporting creditors. Although Yung Kee was decided in the context of a contributory’s winding up, leverage was also in play as this Court granted a stay of the winding-up order to give the parties an opportunity to come to terms on a buy-out.[66] 65. The strength of the leverage obviously depends on the practical impact of a winding-up order to be made in this jurisdiction. Obviously, if a foreign company has no incentive to negotiate for the avoidance or stay of the winding-up order (whether on account of it being hopelessly insolvent or the lack of real impact of the winding-up order on its business as in Banco Nacional de Cuba v Cosmos Trading Corp[67]), there is not much leverage which the petitioner can rely upon in practice. In cases where the leverage is real, though, its significance ultimately depends on the potential impact of a winding-up order. 66. In the present context, the leverage stems from the adverse consequences on the listing status of the appellant in the Stock Exchange of Hong Kong. Any potential impact in terms of possible sanctions by the Listing Division is as much effective before as after the making of a winding-up order. Viewed in that light, the benefit derived from such leverage is incidental to the possibility of the making of a winding-up order. 67. As we have already stated, the formulation of the threshold requirements in the case law should not be construed as if they were statutory provisions.[68] Since the requirement of benefit is to ensure that some useful purpose will be served, there is no justification in principle to exclude significant leverage benefit arising as an incident to the presentation of a petition from the assessment of the second requirement even though such benefit may also have a useful impact prior to the making of the winding-up order. In the event that the appellant remained recalcitrant, the leverage would continue to have an effect after the making of a winding-up order. D. The appellant’s comity argument 68. We now turn to address the appellant’s comity argument leading to the contention that the proper interpretation of the requirements must be informed by their limitations and the principle of non-interference. 69. The expression “comity” is used in many senses, but in the present context it means, inter alia, that the court would not pay sufficient deference to the legislative and judicial jurisdiction of the State of incorporation if it were to exercise winding-up jurisdiction over a foreign company in the absence of a sufficient connection with the forum.[69] Here the sufficient connection is plainly provided by the fact that the appellant’s shares are listed on the Stock Exchange of Hong Kong and it has a registered place of business in Hong Kong. 70. As discussed above,[70] the second requirement only comes into play after sufficient connection is established and it addresses the utility of winding-up proceedings in respect of foreign as well as domestic companies. Whilst comity is obviously relevant in the assessment of the first requirement, it has less relevance under the second requirement. Whilst the issue of comity may arise if the benefit that a petitioner relies upon depends on the liquidator exercising his authority in a foreign jurisdiction, this is not the situation in the present appeal. The leverage in the present case hinges on the potential impact on the listing of the H shares in the Stock Exchange of Hong Kong. There is no possible issue of comity in such an assessment. 71. The appellant accepted in the courts below that the first and third requirements were satisfied in the present case. The appellant has not resiled from that acceptance before us. In any event, it is plain from the background set out above that the appellant had a sufficient connection with Hong Kong. In the absence of any dispute in this respect, there is no room for suggesting that the exercise of the winding-up jurisdiction in the present instance would be a breach of comity. 72. In the circumstances, we do not find it necessary to address the appellant’s comity argument at any greater length. At its heart, the argument amounted to a bold attempt to superimpose on the three requirements a principle of non-interference.[71] This, in substance, is a forum conveniens issue. In short, it was contended, winding up a foreign company is only justified when the jurisdiction of incorporation cannot fulfil its function so that it is then necessary to fill the lacuna.[72] 73. The appellant relied on the dicta of Lord Evershed MR in Banque des Marchands de Moscou (Koupetschesky) v Kindersley[73] to support this proposition. In that case, the foreign corporation was a Russian bank having substantial assets in England which had been dissolved by the Soviet Government after the Russian Revolution. The English court made a winding-up order in respect of the bank and appointed a liquidator. Eighteen years later, in the course of proceedings brought by the liquidator, the winding-up order was challenged. The issue that came before the court was whether it was necessary to show that the foreign company had established a place of business in England before it could be wound up by the English courts. The Court of Appeal held that this was not necessary and the presence of assets and persons subject to the English jurisdiction would be enough. In respect of a foreign company dissolved at its place of incorporation, the Master of the Rolls was of the view that prima facie English courts would not interfere with the administration of its assets as long as those assets were being dealt with by the law of its place of incorporation. On the facts of that case, an English winding-up order was necessary because of the absence of any machinery under Russian law for due distribution of the English assets of a dissolved Russian bank. 74. However, His Lordship did not address the situation where a foreign company has not been dissolved at its place of incorporation and a winding-up petition is only presented in an English court. Subsequently, the law governing the exercise of the winding-up of a foreign company was developed in the cases discussed above. The three requirements were adopted and, as held by Morritt LJ in Re Latreefers Inc,[74]if the “core” requirements are satisfied there is no need for the addition of a fourth to ensure that the jurisdiction is only exercised for good reason. In Yung Kee, this Court endorsed the three requirements as the self-imposed constraints in the exercise of this jurisdiction in Hong Kong notwithstanding that the most appropriate jurisdiction to wind up a company is the one where it was incorporated.[75] There is no room for adding a further requirement of forum conveniens. 75. Whilst there are references to appropriateness of forum as being relevant in the cases, such discussions mainly arose as part of the assessment of sufficient connection under the first requirement.[76] Thus, Lloyd J (as he then was) explained the relevance of the concept of appropriate forum in Latreefers as follows: “It seems to me that it may be relevant to consider what connections there may be with other jurisdictions in order to answer the rather ill-defined question as to the sufficiency of the connection with this jurisdiction. If there is a clear and substantial connection with some other jurisdiction, it may be that more is needed by way of a connection with England and Wales for it to suffice as the foundation of the exercise of extra-territorial jurisdiction. Thus, in the Okeanos case … Peter Gibson J considered, in this context, whether there was any more appropriate jurisdiction to wind up the company.”[77] 76. There are also cases[78] in which appropriate forum is discussed as a separate requirement for determining if the jurisdiction to wind up a foreign company should be exercised. These were decisions made at the hearing of the petition or on a preliminary issue in the petition. The Court was not referred to any case where the issue was considered in the context of an application for an injunction to restrain the presentation of a petition. In any event, the addition of the criterion of forum conveniens as a separate jurisdictional requirement is contrary to the law endorsed by this Court in Yung Kee. 77. For present purposes, it suffices for us to hold that, if sufficient connection is established under the first requirement, any issue related to forum conveniens should only be a factor (as opposed to an essential requirement) that the court would consider at the hearing of the petition in deciding if a winding-up order should be made.[79] The weight to be attached to that issue may depend on the practical effects of the winding up of the foreign company. In cases like Merabello and Eloc, where it is not envisaged that the liquidator will administer any assets (whether within or outside Hong Kong), the fact that the company can also be wound up in the place of its incorporation will not be particularly significant. 78. In the courts below, the appellant did not raise any discrete argument on appropriate forum and the submissions focused on the second requirement. Leave to appeal to this Court was restricted to the narrow issue of benefit under that requirement. In the circumstances, it would not be right to permit a backdoor challenge to whether the first requirement is satisfied in the present case. In any event, as already noted, the satisfaction of this requirement in the present case was rightly accepted below. E. Answering the certified question 79. The short answer to the question of law for which leave was granted is, in respect of both ways it is put, “Yes.” E.1 Application of the principles to this case 80. In light of our conclusions above, Harris J was correct to conclude that the second requirement was satisfied in this case. The crux of his reasoning is expressed thus: “A solvent company that fails to pay a judgment or an award out of intransigence is liable to be wound up. It seems to me that the defendant can properly say that this is such a case and that the benefit to be derived from its proposed course of action is the leverage created by the prospect of a winding-up petition or the appointment of a liquidator and the steps a liquidator may take to recover assets even if such steps are problematic. In my view there is a reasonable prospect that the defendant would derive a benefit from a winding-up order.”[80] 81. We respectfully agree, save that we think it preferable to refer to the respondent deriving a benefit from the invocation of the court’s winding-up procedures over the appellant by the presentation of the petition. There is no necessity, as demonstrated above, for the benefit to flow directly from the making of a winding-up order. The fact that the adjournment of the winding-up petition in the present case was ordered on terms, which were duly complied with, that the appellant procure the payment into court of the amount set out in the respondent’s statutory demand and interest thereon demonstrates, in stark terms, the efficacy of the commercial pressure that the presentation of the petition gave rise to. 82. It follows that we also respectfully agree with the conclusion of the Court of Appeal that the second requirement was satisfied in this case by the pressure placed on the appellant to pay the undisputed, or indisputable, debt by the presentation of a winding-up petition.[81] However, we do not agree that it was necessary for the Court of Appeal to read Harris J’s judgment as if it were limited to a benefit flowing from the making of a winding-up order, rather than its mere prospect.[82] As we have sought to demonstrate above, there is no logical difficulty in regarding the commercial pressure caused by the presentation of a winding-up petition as a qualifying benefit for the second requirement. 83. We do, however, also respectfully agree with Barma JA’s characterisation of the nature of the inquiry as to whether the second requirement is satisfied, namely “to ascertain whether it would be appropriate to put into motion the winding-up machinery in respect of a particular overseas company” and that this allows for “some flexibility as to the nature or extent of the likely benefit to the petitioner that should be shown in order to satisfy the second core requirement, as long as the benefit can be said to be a real possibility, rather than a merely theoretical one.”[83] 84. Harris J went on to hold that the appellant’s disregard for the integrity of the Hong Kong legal system and its at least non-technical contempt for the High Court, by its wish to take advantage of the Hong Kong financial and legal systems yet at the same time refuse to honour the arbitration award and consequential judgment, would “justify moderation of the requirement that a benefit to the defendant be shown”.[84] 85. We do not, with respect, agree that the second requirement is subject to moderation in this manner or for this reason. The Court of Appeal declined to consider this argument.[85] In the light of the flexibility that Barma JA identified in the nature of the inquiry (correctly, in our view), moderation of the second requirement is inappropriate. E.2 Disposition 86. For the reasons set out above, we would therefore answer the certified question as set out at [79] above and dismiss the appeal. 87. As described at [14] above, the winding-up petition against the appellant was adjourned on terms that the appellant procure a third party to make payment of the amount set out in the statutory demand plus interest and costs into court, with payment out to abide by the further order of the Court of First Instance, the Court of Appeal or this Court. The Court of Appeal declined to order payment out, despite dismissing the appellant’s appeal below. That was understandable since an appeal to this Court was pursued. 88. Since, in our judgment, the appeal stands to be dismissed, there would appear now to be no good reason to delay payment out. Mr Joffe, appearing for the appellant,[86] however, resisted this and submitted that it would be more appropriate to direct this issue to be determined by the Court of First Instance. He indicated, on instructions, that the appellant might wish to adduce further evidence and start a further arbitration asserting that the respondent had withheld joint venture documents and that this undermined the validity of the original arbitral award giving rise to the judgment debt which was the subject of the statutory demand served on the appellant (see [7] and [8] above). Mr Joffe went so far as to suggest that the unavailability of the allegedly withheld documents deprived the appellant of the opportunity of properly advancing its case in the original arbitration and that it intended to apply to set aside the ensuing award, if necessary on the basis of fraud. Moreover, he suggested that the third party, whom the appellant had procured to make the payment in, might wish to be heard on the question of the payment out. 89. In our view, this attempt to delay the payment out of the sum paid into court as a condition of the adjournment of the winding-up petition pending resolution of the issue now determined by this appeal has no merit. The sum was paid into court pursuant to an undertaking to do so in order to secure the adjournment of the winding-up petition, which adjournment was effective upon compliance with the undertaking. The payment in was calculated on the basis of the amounts claimed in the statutory demand, which was not otherwise disputed. But for the payment in, the hearing of the petition would have proceeded on the basis of an undisputed debt and the deemed insolvency of the appellant arising from its failure or refusal to meet the statutory demand. For present purposes, it is now too late to challenge the underlying indebtedness and, in any event, the allegations advanced in support of the suggested remission of this issue to the Court of First Instance are wholly vague and unparticularised. 90. As far as the third party is concerned, we do not see any reason why that party should be entitled to make representations on the question of payment out. It is not a party to these proceedings, and has not even been identified, nor has it ever sought to intervene. In any event, that party must be taken to be fully aware of the terms of the undertaking given by the appellant pursuant to which the sum was paid into court, which included a term that it would be paid out at the discretion of the Court of First Instance, Court of Appeal or this Court. As such, it must have been the common intention of all parties, including the third party, that the discretion was to be exercised in the light of the outcome of any appeal against Harris J’s decision. The payment in was plainly made to secure the respondent’s interest in obtaining repayment of the debt, being the subject of the statutory demand, pending the outcome of any such appeal. In these circumstances, any interest the third party funder might have had must be subordinated to the contingency that this Court might order payment out of the sum to the respondent in the event of the dismissal of this appeal. 91. Accordingly, we would direct that the sum paid into court on 8 September 2017 in HCCW 175/2017, pursuant to the order of Harris J dated 28 August 2017, together with the accrued interest thereon, be paid out to the respondent forthwith. 92. Further, we would make an order nisi that the appellant pay the costs of this appeal to the respondent and direct that, if any party wishes to vary this order, written submissions should be filed within 14 days of the handing down of this judgment and the Court will make a final order as to costs on the papers. Lord Collins of Mapesbury NPJ: 93. I agree with the joint judgment of Mr Justice Fok PJ and Mr Justice Lam PJ. Chief Justice Cheung: 94. The appeal is therefore unanimously dismissed and the Court makes the orders set out at [91] and [92] above. Mr Victor Joffe and Mr Alexander Tang, instructed by King & Wood Mallesons, for the Plaintiff (Appellant) Mr Laurence Li SC and Mr Chow Ho Kiu, instructed by CL Chow & Macksion Chan, for the Defendant (Respondent) [1] (2015) 18 HKCFAR 501. [2] Ibid. at [20]. [3] [2022] HKCFA 1, FAMV 410/2021, 21 January 2022. [4] [2020] HKCA 670, CACV 158/2017, 5 August 2020 (“CA Judgment”). [5] HCCT 53/2015, Decision (not open to the public), dated 12 October 2016. [6] [2017] 4 HKLRD 84 (“CFI Judgment”) at [14]. [7] Ibid. at [15]. [8] Ibid. at [16]-[19] in respect of the listing and at [20]-[26] in respect of the investigation of the restructuring. [9] Ibid. at [28]-[29]. [10] Ibid. at [30]-[31]. [11] CA Judgment at [31]-[34]. [12] Ibid. at [21]-[28]. [13] The Plaintiff/Appellant’s Case at [13]. [14] The effect of the definition of “unregistered company” in s.326 is that any foreign company is subject to the winding-up jurisdiction. [15] Yung Kee at [36]. [16] Re Real Estate Development Co [1991] BCLC 210 per Knox J at 217d. [17] Yung Kee at [19] and [21]. [18] See Re Latreefers Inc [2001] BCC 174 at [30]. [19] Jurisdiction to approve schemes of arrangement, which applies only to companies “liable to be wound up” (now UK Companies Act 2006, s.895(2)). [20] [2004] 1 WLR 1049 at [23]-[26], per Lawrence Collins J (as Lord Collins of Mapesbury then was); Drax has been followed in several decisions, including Re OJSC Ank Yugraneft v Sibir Energy plc [2009] 1 BCLC 298; Re Rodenstock GmbH [2011] EWHC 1104 (Ch), [2011] Bus LR 1245; JSC Bank of Moscow v Kekhman [2015] EWHC 396 (Ch), [2015] 1 WLR 3737. [21] [2004] 1 WLR 1049 at [24]. [22] See Yung Kee at Section D (concerned with the question of whether the exercise of jurisdiction was proper) and compare Section E (concerned with the question of whether to make a winding-up order). [23] The Plaintiff/Appellant’s Case at [49], quoting Yung Kee at [24] (with appellant’s emphasis). [24] [1973] 1 Ch 75. [25] [1982] 1 Ch 43. [26] [1988] 1 Ch 210. [27] [1991] BCLC 210. [28] Law Reform Commission of Hong Kong, Report on The Winding-Up Provisions of the Companies Ordinance (1999) at [9.5] (emphasis added). [29] Clarke and Walker Pty Ltd v Thew (1967) 116 CLR 465 at 467; Derek French, Applications to Wind Up Companies (4th Ed.) at [7-281]. [30] [2013] 1 WLR 1408 at [24]. [31] [1906] 2 Ch 327 at 331-332. [32] [1986] 1 WLR 114 at 118G to H. [33] [1906] 2 Ch 327 per Buckley J at 331. [34] [2016] 1 HKLRD 850 at [27(5)]. [35] Sell Your Car With Us Ltd v Sareen [2019] EWHC 2332 (Ch), [2019] BCC 1211 at [53]. [36] Ibid. citing Goode on Principles of Corporate Insolvency Law (5th Ed.) at p.195. [37] In re Millennium Advanced Technology Ltd [2004] 1 WLR 2177 per DHCJ Briggs QC (now Lord Briggs of Westbourne JSC) at [36]. [38] [1968] 1 WLR 1091 at 1096C-F. [39] [1993] BCLC 131 at 132g (emphasis added). [40] Goode on Principles of Corporate Insolvency Law (5th Ed.) at [5-13]. [41] (2015) 18 HKCFAR 501 at [26] (emphasis added). [42] [1906] 2 Ch 327 at 332. [43] [1973] 1 Ch 75. [44] Ibid. at 90C to F (emphasis added). [45] See Yung Kee, supra at [26]. In Re OJSC Ank Yugraneft [2009] 1 BCLC 298 at [58], Christopher Clarke J made the observation that sufficient connection would not be established if the asset was too small or of such a character that the link with the jurisdiction was too tenuous and cited the facts of Re Real Estate Development Co [1991] BCLC 210 as an example. [46] [1973] 1 Ch 75 at 91F to G. [47] [1982] 1 Ch 43. [48] Ibid. at 47C to D and 48G. [49] [1988] 1 Ch 210. [50] [1987] BCLC 450. [51] [1988] 1 Ch 210 at 222H to 223B; 223H to 226A. [52] Merabello, supra at 92B. [53] The approach of Peter Gibson J was followed by Harman J in Ex parte Nyckeln Finance Co Ltd [1991] BCLC 539 and Lloyd J in Re Latreefers Inc [1999] 1 BCLC 271 where provisional liquidators were appointed in respect of foreign companies without assets within England. Lloyd J’s decision was upheld by the English Court of Appeal, [2001] BCC 174. [54] [1991] BCLC 210. [55] Ibid. at 217e to f. [56] Ibid. at 217g to h. [57] [2009] 1 BCLC 298. [58] Ibid. at [60]. [59] Supra at [36]. [60] Supra at 334-335. [61] Ibid. at 337-338. [62] Ibid. at 338. [63] [2001] 2 BCLC 633 at 638a to b. [64] (2015) 18 HKCFAR 501 at [27]. [65] [1991] BCLC 210 at 217d-e (emphasis added). [66] Yung Kee at [64]. [67] [2000] 1 BCLC 813. [68] Yung Kee at [36]. [69] In re Drax Holdings Ltd [2004] 1 WLR 1049 at [29]. [70] See [50] to [52] above. [71] The Plaintiff/Appellant’s Case at [41]. [72] Ibid. at [16.1]. [73] [1951] 1 Ch 112 at 125-126. [74] [2001] BCC 174 at [32]. [75] Supra at [19] to [21]. [76] Okeanos, Ex p Nyckeln Finance, Latreefers. [77] [1999] 1 BCLC 271 at 278c to d. [78] Atlantic & General Investment Trust Ltd v Richbell Information Services Inc [2000] 2 BCLC 778, Re Buccament Bay Ltd [2015] 1 BCLC 646. [79] Re Wallace Smith & Co Ltd [1992] BCLC 970 at 985g to i; Dicey, Morris & Collins on The Conflict of Laws (15th Ed) at §30-055; Kate Dawson, The doctrine of forum non conveniens and the winding up of insolvent foreign companies [2005] JBL 28. [80] CFI Judgment at [29]. [81] CA Judgment at [33]. [82] Ibid. at [30]-[31]. [83] Ibid. at [27]. [84] CFI Judgment at [30]. [85] CA Judgment at [27]-[28]. [86] With Mr Alexander Tang. 
Chief Justice Cheung: 1. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Fok PJ: A. Introduction 3. The certified question of law[1] for which leave to appeal to this Court was granted by the Appeal Committee[2] is: “Which alternative set out in Kulemesin v HKSAR (2013) 16 HKCFAR 195 shall be applicable in relation to the ‘likely to cause any person reasonably to fear’ limb of the offence created by section 18 of the Public Order Ordinance (Cap. 245)?” 4. The question arises out of the appellant’s conviction on a charge of taking part in an unlawful assembly, contrary to sections 18(1) and (3) of the Public Order Ordinance (Cap. 245) (“the POO”). The particulars of the charge materially read: “On 2nd November 2016, [the appellant], Yau Wai Ching, Yeung Lai Hong, Chung Suet Ying and Cheung Tsz Lung, together with other persons took part in an unlawful assembly outside Conference Room 1, 2nd Floor, Legislative Council Complex, No. 1 Legislative Council Road, Central, Hong Kong, i.e. they assembled together and conducted themselves in a disorderly, intimidating, insulting or provocative manner intended or likely to cause any person reasonably to fear that the persons so assembled would commit a breach of the peace, or would by such conduct provoke other persons to commit a breach of the peace.” 5. The appellant was convicted of the charge before a magistrate on 11 May 2018[3] and, on 4 June 2018, he was sentenced to four weeks’ imprisonment for the offence. The appellant appealed against his conviction and sentence to the Court of First Instance. On 2 September 2020, Wilson Chan J dismissed the appeal against both conviction and sentence.[4] The appellant sought certification of questions of law for the Court of Final Appeal and, as already stated, His Lordship certified the question of law set out at [3] above. 6. The question of law raises an important point of law concerning the identification of a gathering as an unlawful assembly and, specifically, as to the mental element of the offence of taking part in such an assembly. The status of an assembly as unlawful is also an element in various other offences under Part IV (Unlawful Assemblies, Riots and Similar Offences) of the POO and relates to other important questions of law that will fall to be determined in the future. [5] This Judgment only addresses the specific question raised in the certified question in this appeal. 7. As will be seen later in this Judgment, because of the magistrate’s findings of fact, upheld by the judge, the appellant’s appeal against his conviction must be dismissed regardless of the answer to the certified question of law. It is convenient to begin, however, with an analysis of the proper construction of the relevant offence creating section. B. An analysis of the statutory offence under section 18 of the POO 8. The Court of Final Appeal has previously dealt with the proper approach to the construction of offence creating provisions, and specifically the mental element of such offences, in Hin Lin Yee v HKSAR[6] (“Hin Lin Yee”) and Kulemesin v HKSAR[7] (“Kulemesin”). In determining the certified question in this appeal, these authorities are to be applied. B.1 The principles laid down by Hin Lin Yee and Kulemesin 9. The required mental state of any given statutory offence is a matter of statutory construction.[8] In that exercise, the principle of the presumption of mens rea may be engaged but it is important to note, as was pointed out in Hin Lin Yee, that not every case raises that presumption or involves consideration of the alternative categories identified in that decision (as later refined in Kulemesin). As Ribeiro PJ observed in Hin Lin Yee: “What, if any, mental state is required is a matter of statutory construction. The statute may of course be specific, saying for instance that the act must be done ‘wilfully’, ‘knowingly’, ‘negligently’, ‘without due care and attention’ and the like. It may go further and lay down a requirement not merely of a basic intent but also a specific intent: the alleged burglar, for example, must be shown to have (intentionally) entered a building as a trespasser with the specific intent of stealing or committing one of the other named offences when inside. Such provisions pose no problems beyond having to resolve possible arguments as to the scope of the words used and their proper application to the facts.”[9] 10. The principles discussed in Hin Lin Yee and Kulemesin are only applicable where “the provision which creates the offence is silent or ambiguous as to the state of mind required.”[10] It is in those cases that the presumption of mens rea arises, with the starting point being “that the statute must be construed adopting the presumption that it is incumbent on the prosecution to prove mens rea in relation to each element of the offence” and, to that extent, supplementing the text of the statutory language.[11] 11. However, the presumption is merely a starting point because it is “equally firmly established that a statute may, on its proper construction, displace the presumption of mens rea expressly or by necessary implication.”[12] So the first question of construction arising is whether the presumption is to be maintained or displaced: Hin Lin Yee at [45] and [98]; Kulemesin at [40]. And if it is determined to be so displaced, Hin Lin Yee requires that a second question be considered in tandem, namely: “By what, if any, mental requirement is the supplanted requirement of mens rea to be replaced?”[13] 12. As reformulated in Kulemesin, there are five constructional choices that present themselves as possible alternatives when asking, in tandem, the questions “has the presumption of mens rea been displaced” and “if so, by what”. Those five alternatives are: “(a) First, that the mens rea presumption persists and the prosecution must prove knowledge, intention or recklessness as to every element of the offence (‘the first alternative’); (b) Second, that the prosecution need not set out to prove mens rea, but if there is evidence capable of raising a reasonable doubt that the defendant may have acted or omitted to act in the honest and reasonable belief that the circumstances or likely consequences of his conduct were such that, if true, liability would not attach, he must be acquitted unless the prosecution proves beyond reasonable doubt the absence of such exculpatory belief or that there were no reasonable grounds for such belief (‘the second alternative’); (c) Third, that the presumption has been displaced so that the prosecution need not prove mens rea but that the accused has a good defence if he can prove on the balance of probabilities that he acted or omitted to act in the honest and reasonable belief that the circumstances or likely consequences of his conduct were such that, if true, he would not be guilty of the offence (‘the third alternative’); (d) Fourth, that the presumption has been displaced and that the accused is confined to relying on the statutory defences expressly provided for, the existence of such defences being inconsistent with the second and third alternatives mentioned above (‘the fourth alternative’); and (e) Fifth, that the presumption is displaced and the offence is one of absolute liability so that the prosecution succeeds if the prohibited act or omission is proved against the accused, regardless of his state of mind regarding the relevant elements of the offence in question (‘the fifth alternative’).”[14] 13. Thus, where the Court holds that the presumption of mens rea, in terms of intention, knowledge or recklessness, is displaced on the proper construction of the statute in a given case (i.e. that the first Kulemesin alternative does not continue to apply), it will then go on to decide as a further matter of statutory construction, which of the second, third, fourth or fifth Kulemesin alternatives applies. That construction will have regard to the nature and subject-matter of the offence, its seriousness in terms of penalty and social obloquy, the need to ensure protection of the public, the practicalities of prosecution and conviction, the extent to which the defendant might solely have access to relevant information, and so on. In Hin Lin Yee[15] and Kulemesin[16], there is discussion of these considerations as they related to the particular statutory offences in those cases. B.2 The text of section 18 of the POO 14. Section 18 reads: “(1) When 3 or more persons, assembled together, conduct themselves in a disorderly, intimidating, insulting or provocative manner intended or likely to cause any person reasonably to fear that the persons so assembled will commit a breach of the peace, or will by such conduct provoke other persons to commit a breach of the peace, they are an unlawful assembly. (2) It is immaterial that the original assembly was lawful if being assembled, they conduct themselves in such a manner as aforesaid. (3) Any person who takes part in an assembly which is an unlawful assembly by virtue of subsection (1) shall be guilty of the offence of unlawful assembly and shall be liable – (a) on conviction on indictment, to imprisonment for 5 years; and (b) on summary conviction, to a fine at level 2 and to imprisonment for 3 years.” 15. As will be apparent, section 18 begins by defining, in sub-section (1), what constitutes an unlawful assembly. Sub-section (2) clarifies that what determines the status of the assembly as unlawful is the conduct of the persons who have assembled, even if the assembly originated from a lawful assembly. If the assembly is unlawful as so defined, any person who takes part in the assembly commits an offence (section 18(3)). 16. The status of a gathering as an unlawful assembly is also material to the offence of riot, provided for in section 19 of the POO. However, as already noted at [4] above, this appeal is not concerned with that offence and it is unnecessary to address the provisions of that section in this Judgment. 17. The elements of the offence of unlawful assembly as statutorily set out in section 18 of the POO can conveniently be enumerated in the following seven separate steps: [1] “When 3 or more persons,” [2] “assembled together,” [3] “conduct themselves in a disorderly, intimidating, insulting or provocative manner” [4] “intended or likely to cause any person reasonably to fear that the persons so assembled will commit a breach of the peace, or will by such conduct provoke other persons to commit a breach of the peace,” [5] “they are an unlawful assembly.” [Section 18(1)] [6] “It is immaterial that the original assembly was lawful if being assembled, they conduct themselves in such a manner as aforesaid.” [Section 18(2)] [7] “Any person who takes part in an assembly which is an unlawful assembly by virtue of subsection (1) shall be guilty of the offence of unlawful assembly …”. [Section 18(3)] 18. So separated, the statutory language provides that, by [5], an unlawful assembly exists if elements [1] to [4] are established. [6] provides that an initially unlawful assembly may become an unlawful assembly if [1] to [4] should occur. [7] is the offence-creating provision which makes it an offence for any person to take part in an unlawful assembly. B.3 The two limbs of section 18(1) and the presumption of mens rea 19. Section 18(1) has two limbs, divided by the words “intended or likely”, which create two forms of the offence committed by taking part in the unlawful assembly as provided for in section 18(3). Although these must be read together, they differ significantly from each other. (1) The first is “the intended limb”, which provides that the offence is committed (by taking part): “When 3 or more persons, assembled together, conduct themselves in a disorderly, intimidating, insulting or provocative manner intended … to cause any person reasonably to fear that the persons so assembled will commit a breach of the peace …” (emphasis added). (2) The second is “the likely limb”, which provides that the offence is committed (by taking part): “When 3 or more persons, assembled together, conduct themselves in a disorderly, intimidating, insulting or provocative manner … likely to cause any person reasonably to fear that the persons so assembled will commit a breach of the peace …” (emphasis added).[17] 20. The offence under the intended limb expressly requires mens rea, specifying explicitly that it requires proof of intention. The actus reus requirements in section 18(1) are: (1) for there to be 3 or more persons; (2) assembled together; and (3) who conduct themselves in the prohibited manner, i.e. in a disorderly, intimidating, insulting or provocative manner. Since “intended” here can only be understood to mean “intended by the assembled persons”, the intended limb lays down as part of the mens rea requirements of the offence that the assembled persons conducting themselves in the prohibited ways must intend to “cause any person reasonably to fear that the persons so assembled will commit a breach of the peace”. Thus, construing the intended limb, there is no need to embark on the exercise envisaged in Hin Lin Yee or Kulemesin since the statute is neither silent nor ambiguous as to the state of mind required. The offence created by the intended limb expressly spells out the required mens rea in relation to the consequence of causing a reasonable fear of a breach of the peace. 21. The likely limb stipulates the same three actus reus elements, namely: (1) for there to be 3 or more persons; (2) assembled together; and (3) who conduct themselves in the prohibited manner, i.e. in a disorderly, intimidating, insulting or provocative manner. But at this point, a crucial difference from the intended limb arises. The words which then follow, “likely to cause any person reasonably to fear that the persons so assembled will commit a breach of the peace …”, do not address the mental state of the assembled persons. Instead, they refer to the quality or likely consequences of the prohibited conduct by the assembled persons as observed externally. Thus, the prohibited conduct must be of such a nature as to be likely to result in any person reasonably apprehending a breach of the peace by the assembled persons. 22. The phrase in section 18(1), “likely to cause [etc.]”, is properly to be read and understood purely as an additional actus reus element. The prohibited conduct must be of such a nature as to be likely to carry the consequence of causing any person reasonably to apprehend a breach of the peace by the assembled persons. 23. So understood, the offence created by the likely limb is silent as to what, if any, mental state must be proved against the defendants in relation to that likely consequence of their conduct. One must then embark on carrying out the Hin Lin Yee/Kulemsesin exercise. Doing so, one starts with the presumption that, normally, mens rea is required (see [10] above). 24. For the reasons that I shall develop, including consideration of the question of what alternative to mens rea might apply (see Section B.4 below), it is to be concluded that the presumption of mens rea has been displaced regarding the offence created by the likely limb. Since intention and thus mens rea is expressly required under the intended limb but deliberately omitted from the likely limb, the legislative intent cannot have been for the two offences to carry the same mental requirement, one introduced expressly and the other, redundantly, introduced by application of the presumption of mens rea. If it had been the legislative intent that the prohibited conduct must always be accompanied by an intention to cause the apprehension of a breach of the peace, the words “or likely” would simply have been left out. By inserting those words into section 18(1), and thereby creating the likely limb, the legislative intent must have been to create an offence which can be committed without proof of an intention to cause, or recklessness as to causing, the apprehension in question. Obviously, if such proof were available, liability would be established under the intended limb. But absence of proof of such intention clearly does not mean that liability cannot arise under the likely limb. 25. The question, then, is what, if any, mental state short of intention is needed for conviction under the likely limb. This will be addressed in the next section (B.4) below. B.4 The mental element, if any, relevant to the likely limb 26. The approach to determining what, if any, mental element applies, as reformulated in Kulemesin, is set out at [12] above. From the parties’ printed cases and submissions in this appeal, the contest is between the first, second and fifth alternatives. The appellant’s contention is that either the first or second Kulemesin alternatives apply to the likely limb but that, in any event, the fifth should not apply.[18] For the respondent’s part, it is contended that the likely limb of the offence falls within the fifth Kulemesin category.[19] 27. The fourth Kulemesin category, which neither party has suggested applies, can be eliminated immediately, since there is no relevant statutory defence against liability for unlawful assembly. 28. In submitting that the first category (full mens rea embracing intention, knowledge and recklessness) should apply to the likely limb, the appellant relies on essentially two main arguments, which are also relied upon in support of the second, and against the fifth, categories. Before addressing those arguments, the contention that the first category is supportable has already been disposed of. As explained at [24] above, the first category can be ignored since that is provided for by the intended limb of the offence and it can be assumed that it was not intended by the Legislature to be duplicated in the likely limb. As a matter purely of context, the statutory language cannot sensibly be construed as imposing full mens rea on the likely limb. Nor, as will be seen, does the legislative purpose support that conclusion. 29. Accordingly, the appellant’s reliance on the first instance decision of Albert Wong J in HKSAR v Leung Tin Kei[20] is misplaced. In that case, His Lordship held that the presumption of mens rea is not displaced. He held: “Upon careful consideration, I am of the view that as far as the ‘causing the stipulated fear by an objective standard’ limb is concerned, the presumption of mens rea ought not to be displaced. I am also of the view that ‘knowledge’ or ‘recklessness’ is the appropriate mens rea in relation to the ‘causing the stipulated fear by an objective standard’ limb of the elements.”[21] “Upon careful consideration, I rule that in respect of the ‘causing the stipulated fear by an objective standard’ limb, the prosecution has to prove that: the defendant knew of or was reckless as to the ‘prescribed act or acts’ being likely to cause ‘the stipulated fear’.”[22] That construction, as already explained at [24] above, cannot be accepted. Moreover, it overlooks the evident purpose of the likely limb reflected in the word “likely” and “reasonably”, namely the prevention of behaviour which creates an objective risk of a breach of the peace. His Lordship’s construction, therefore, cannot be right, and for reasons which follow, the judgment in HKSAR v Leung Tin Kei as regards the mental element of the likely limb of the offence of unlawful assembly,[23] must be overruled. 30. Turning to the contest between the second and fifth Kulemesin categories, the two main arguments advanced by the appellant as to the appropriate mental element for the likely limb of the offence are, in summary, that: (1) It is necessary to construe section 18(1) consistently with the constitutional protection afforded to the freedoms of assembly and expression and accordingly one can discard absolute liability as a conclusion; and (2) A culpable state of mind is required, and absolute liability should therefore not be imposed, because of the seriousness of the offence which is punishable on conviction on indictment by imprisonment for five years and on summary conviction by a fine and imprisonment for three years. 31. Neither of these arguments is convincing and both should be rejected. 32. As to the constitutional right of peaceful assembly, it is to be remembered that the Court held in HKSAR v Chow Nok Hang:[24] “Once a demonstrator becomes involved in violence or the threat of violence – somewhat archaically referred to as a ‘breach of the peace’ – that demonstrator crosses the line separating constitutionally protected peaceful demonstration from unlawful activity which is subject to legal sanctions and constraints. The same applies where the demonstrator crosses the line by unlawfully interfering with the rights and freedoms of others.” 33. The significance of crossing the line separating constitutionally protected behaviour and conduct that may be visited by sanction under the criminal law has been emphasised recently in this Court’s judgment in Kwok Wing Hang v Chief Executive in Council.[25] In the case of section 18(1), where the assembled persons, acting in concert, conduct themselves in the prohibited manner (i.e. in a disorderly, intimidating, insulting or provocative manner) with the objectively likely consequence of causing any person to have reasonable apprehension of a breach of the peace, their behaviour would inevitably involve the threat of a breach of the peace, taking them outside the scope of constitutional protection, whichever mental state may be required regarding the consequences of their conduct. 34. As to the argument based on the seriousness of the offence, this is undoubtedly a factor to be taken into account in the construction exercise, to be assessed together with all other relevant factors: “… all five of the reformulated alternatives … should be considered as possible conclusions when construing statutory criminal offences – both serious and regulatory – which are silent or ambiguous as to the state of mind required in respect of the element of the offence under consideration.”[26] 35. And, as pointed out in Hin Lin Yee,[27] the Court “recoils”, especially in relation to an offence carrying substantial penalties, from imposing absolute liability which results in a person’s conviction regardless of the mental state accompanying that person’s conduct where he or she may have “acted in a reasonable, diligent and socially unblameworthy manner” so that the criminal law ends up “snaring the diligent and socially responsible”. 36. In the present case, however, other factors, including the nature and subject-matter of the offence and the statutory objectives, including the need to ensure protection of the public and the maintenance of public order, are critical. 37. Where defendants charged with unlawful assembly have acted in a manner constituting the actus reus elements of the likely limb of section 18(1), they will, in a group of at least three persons, have assembled together and conducted themselves in a disorderly, intimidating, insulting or provocative manner which, viewed objectively, will have been likely to cause any person reasonably to fear that the assembled persons would commit a breach of the peace. This Court has previously endorsed the view that a breach of the peace occurs including: “… whenever harm is actually done or is likely to be done to a person or in his presence to his property or a person is in fear of being so harmed through an assault, an affray, a riot, unlawful assembly or other disturbance.” [28] 38. Persons who so conduct themselves in concert thereby causing any other person reasonably to fear that the assembled persons are likely to cause such harm are far removed from persons who have acted in a reasonable, diligent and socially unblameworthy manner or as socially responsible persons. In the circumstances, it is appropriate that section 18(1) be construed as prohibiting such conduct under the criminal law without requiring an additional mens rea element to be established. 39. Such a construction would also reflect the purpose of the offence. As the Attorney-General stated, when moving the first reading of the Public Order Bill 1967 (which became the POO): “The Bill seeks to confer adequate powers for the prevention and control of disorders at all stages, with particular emphasis upon dealing with them as early as possible”.[29] (emphasis added) And, as Lam JA (sitting at first instance) pointed out in Secretary for Justice v Leung Kwok Wah:[30] “It must be emphasised that s.18 is very much a preventive measure. There is no need for a breach of the peace to have occurred.” 40. An offence that is designed to deter conduct likely to cause any person reasonably to apprehend a breach of the peace, and thus to prevent breaches of the peace from happening, is not logically linked to whether the assembled persons do or do not foresee such reasonable apprehension as the consequence of their acts, but is focused on responding to the objectionable nature and quality of those acts. Unlawful assembly is a public order offence whose purpose is to protect the public from the harm of public disorder. Given this statutory purpose, the alternative of recklessness (as distinct from intention) contended for on behalf of the appellant can also be rejected as inappropriate in relation to the likely limb. 41. I would therefore conclude, as a matter of statutory construction, that the likely limb of the offence falls within the fifth Kulemesin category and so dispenses with any mens rea requirement in respect of the likely consequence of causing reasonable apprehension of a breach of the peace. Contrary to the appellant’s contentions inviting the Court to reject a construction that “overcriminalises” behaviour, the offence in fact has built-in requirements providing a firm basis for treating the conduct as culpable and deserving of criminal sanction. In this regard, the likely limb offence requires proof of mens rea in relation to the defendants assembling and acting together in performing the prohibited acts. It also requires their conduct, objectively assessed, to be of such a nature that it is likely to cause any person reasonably to apprehend that they will commit a breach of the peace. That such apprehension must be objectively reasonable on the part of any person, and not just a few overly-sensitive persons, underlines the requirement that their conduct must provide a substantial basis for regarding a breach of the peace as likely. In addition, the offence is complete only when, as stipulated in section 18(3), a person “takes part in” an unlawful assembly so defined. The taking part is an additional requirement which additionally militates against the possibility of an innocent person being inadvertently caught up in the assembly and thereby unfairly convicted. 42. Support for this construction of the likely limb of the section 18 offence can be found in the similarly structured offence of disorder in public places in section 17B of the POO. Section 17B(2) provides: “Any person who in any public place behaves in a noisy or disorderly manner, or uses, or distributes or displays any writing containing, threatening, abusive or insulting words, with intent to provoke a breach of the peace, or whereby a breach of the peace is likely to be caused, shall be guilty of an offence and shall be liable on conviction to a fine at level 2 and to imprisonment for 12 months.” Like section 18(1), therefore, section 17B(2) has two forms, concerning prohibited disorderly conduct either (1) intended to provoke a breach of the peace, or (2) whereby a breach of the peace is likely to be caused. 43. Section 17B(2) was considered by this Court in HKSAR v Chow Nok Hang and, although the focus of that case was not on the mental requirements of the offence, the judgments proceed on the evident assumption that the likely limb of section 17B(2) involves a purely objective assessment of the likely consequences of the impugned behaviour without reference to any subjective foresight or mental state on the part of the defendant. [31] 44. The appellant’s contention that this comparison was inapt because section 17B(2) is concerned with the likelihood to cause breach of the peace, whereas the offence under section 18 is concerned with the likelihood to cause a stipulated fear (“… likely to cause any person reasonably to fear that the persons so assembled will commit a breach of the peace …”) is, in this context, a distinction without a difference. The exercise in respect of both likely limbs of the two offences still requires an external objective assessment of the likelihood of an anticipated outcome that will not in fact have yet happened. 45. As discussed at [41] above, there is a significant degree of blameworthiness incorporated in the likely limb of the offence. For that reason, and notwithstanding the conclusion that on its construction the likely limb falls within the fifth category of Kulemesin, it is inapt to use the term “strict or absolute liability” to describe the likely limb of the section 18 offence. Although some regulatory offences, which on their proper construction are held to be within the fifth category of Kulemesin, are truly strict or absolute liability offences, that description is not appropriate in a case like the present where only one element of the offence is construed as falling within that category. 46. It therefore follows, from the discussion above, that the appellant’s contention that the first or second, and not fifth, Kulemesin category should apply must be rejected. 47. Although neither party has argued in favour of the third Kulemesin category, the same reasoning applies to it. As a matter of construction, the likely limb focuses on the likely effect of the conduct objectively assessed and not on the effect contemplated by the assembled persons. C. The charge against the appellant 48. The incident giving rise to the charge against the appellant arose from a melee involving the appellant and security staff of the Legislative Council (“LegCo”) when the appellant was seeking to force his way into a conference room where LegCo was meeting and from which he had been excluded. 49. The appellant had been elected to LegCo in September 2016 but he declined or neglected to take his oath on 12 October 2016.[32] The appellant wanted to re-take his oath at the LegCo meeting to be held on 2 November 2016 but he was not permitted to take part in that meeting. A notice to this effect was posted outside the LegCo chamber. Notwithstanding this, the appellant attended the LegCo meeting but was ordered to leave. When he refused to do so, the meeting was adjourned to a conference room on the second floor. At about 11.55am, the appellant and others tried to enter the conference room but were prevented from doing so by security guards. The appellant left but returned at about 1pm with about 14 others. They attempted to enter the conference room but were blocked by a cordon of security guards. 50. That attempted forced entry gave rise to the charge against the appellant. It was the prosecution case that the appellant and his accompanying group rushed at the cordon and pushed against the guards in an attempt to enter the conference room. There was a concerted effort to coordinate a surge forward to breach the cordon. The appellant was described as being very agitated and having lost control of his emotions. He held onto the frame of the door to the conference room and, on several occasions, tried to haul himself over the guards to get past them. Other members of the group supported him by pushing and shoving. The incident lasted about 20 minutes. The LegCo meeting was then adjourned and the group left. There were various injuries to some of the guards, requiring them to be taken to hospital, although fortunately none of the injuries were major. 51. The appellant testified in his own defence. He claimed not to have seen the notice excluding him from the LegCo meeting on 2 November 2016 but admitted he was asked to leave after attempting to re-take his oath. He was aware the meeting had been adjourned to the second floor conference room. When he went there at about noon, there were members of the media outside and he went elsewhere to be interviewed. He returned at 1pm thinking there would be fewer people present and intending to enter the conference room to take his oath. He said he did not think the security guards would stop him. When he arrived outside the entrance to the conference room, the guards pushed him away. He told the guards it was an offence to obstruct a LegCo member from attending a LegCo meeting. He lost his balance but he did not lean forward or charge. There was some force behind him which pushed him forward towards the guards who pushed him back. He claimed he wanted to leave the scene but there had been no way out. C.1 The findings in the courts below 52. The magistrate believed the prosecution evidence and rejected the appellant’s contention that he did not expect to see the security guards there and the claim that he wanted to leave but could not do so, and also his assertion that he did not charge the guards. The magistrate specifically found that: the appellant and the other defendants shared a common purpose to get past the cordon; they assembled together in the corridor outside the conference room; the appellant and the other four defendants and others charged the cordon and caused a disturbance outside the conference room for about 20 minutes in a confined area; the appellant tried to climb over the cordon and, each time he did so, his arm struck one of the guards, even if unintentionally; in the melee, several security guards were injured; and the conduct of the appellant and the other defendants was disorderly. 53. The magistrate was satisfied that the actions of the appellant and the other defendants were likely to cause any person reasonably to fear that they would commit a breach of the peace. She held that “whether such behaviour was likely to cause any person reasonably to fear that they (the persons so assembled) would commit a breach of the peace, the subjective intentions of each defendant do not matter. It is an objective criterion.”[33] 54. The magistrate also addressed the defence of honest and reasonable belief that was advanced. She held that, even if the defence were available, it could not, on the evidence, assist the appellant and the other defendants because the circumstances were such that they must have known that their conduct was likely to cause apprehension of a breach of the peace[34] and, in any event, their suggested honest and reasonable belief in their right in law to assemble and behave as they had was at odds with their contention that they had not behaved in that manner.[35] 55. In the Court of First Instance, the main issue was the mens rea issue with which this appeal has been concerned. The judge held that the legislative intention was “to impose an absolute liability” and he construed the likely limb as falling within the fifth category of Kulemesin.[36] In so deciding, the judge relied on the judgment of Lam JA in Secretary for Justice v Leung Kwok Wah (supra.) and declined to follow the judgment of Albert Wong J in HKSAR v Leung Tin Kei (supra.). However, the judge held that, even if the correct approach was that adopted in the latter case, it would not avail the appellant since the evidence was such that he must have known at the time of the melee that the group’s conduct was such as was likely to cause fear of a breach of the peace. 56. As will be apparent from Section B.4 of this Judgment, I respectfully agree with the conclusions of the magistrate and judge in respect of the mental element of the likely limb. However, for the reasons noted at [45] above, it would be preferable not to refer to the offence as being one of strict or absolute liability since it carries significant mens rea requirements in relation to important actus reus elements. C.2 No difference in outcome however certified question answered 57. Given the findings of the magistrate, upheld by the judge, the appellant’s guilt in respect of the offence is inevitable regardless of the answer to the certified question. As the judge rightly held: “Even if the Court is to adopt the Judgment of Hon Wong J in Leung Tin Kei, namely that the presumption of mens rea in Ingredient (3)(b) had not been displaced (the first alternative in Kulemesin), and that the Prosecution needed to prove the knowledge or recklessness as to the prescribed manner being likely to cause any person reasonably to have the stipulated fear, the Appellant would still have committed the offence given that the Magistrate ruled the Appellant must know the nature of their conduct and the circumstances at the time of the offence were likely to cause any person reasonably to fear that they would commit a breach of the peace.”[37] D. Disposition of the appeal and conclusions 58. For the above reasons, I would dismiss the appeal. 59. In answer to the certified question of law: (1) The actus reus elements of the likely limb of section 18(1) of the POO are: (i) there must be “3 or more persons”; (ii) they must be “assembled together”; (iii) they must “conduct themselves in a disorderly, intimidating, insulting or provocative manner”; and (iv) their conduct, viewed objectively, must “cause any person reasonably to fear that [they] will commit a breach of the peace … or provoke other persons to commit a breach of the peace”. (2) The prosecution will need to prove full mens rea on the part of the defendant in respect of each of elements (i) to (iii) above. No mens rea is required in respect of element (iv). (3) The prosecution must also prove the defendant took part in the unlawful assembly within section 18(3) of the POO. 60. HKSAR v Leung Tin Kei as regards the mental element of the likely limb of the offence of unlawful assembly[38] is overruled. Mr Justice Stock NPJ: 61. I agree with the judgment of Mr Justice Fok PJ. Mr Justice French NPJ: 62. I agree with the judgment of Mr Justice Fok PJ. Chief Justice Cheung: 63. The appeal is unanimously dismissed and we set out answers to the certified question of law in paragraph 59 above. Mr Douglas Kwok, Mr Law Ka Sing and Mr Howard Tang, instructed by Cedric & Co., for the Appellant Mr Jonathan Man DDPP (Ag.) and Mr Derek Lau SADPP (Ag.), of the Department of Justice, for the Respondent [1] [2020] HKCFI 2401, HCMA 303/2018, 11 September 2020 (Wilson Chan J). [2] [2021] HKCFA 6, FAMC 36/2020, 4 March 2021 (Ribeiro PJ, Fok PJ and Stock NPJ). [3] [2018] HKMagC 1, KCCC 2035/2017, 11 May 2018 (Ms Wong Sze Lai). [4] [2020] HKCFI 2152, HCMA 303/2018, 2 September 2020 (Wilson Chan J). [5] See, e.g., HKSAR v Lo Kin Man [2021] HKCFA 17, FAMC 12/2020, Determination, 17 May 2021, to be heard on 5 October 2021, together with the appeal in HKSAR v Tong Wai Hung [2021] HKCA 807, CASJ 1/2020 (in which leave to appeal to the Court of Final Appeal was granted in FAMC 21/2021). [6] (2010) 13 HKCFAR 142. [7] (2013) 16 HKCFAR 195. [8] (2010) 13 HKCFAR 142 at [38]-[39]. [9] Ibid. at [39] (footnotes omitted). [10] Ibid. at [40]. [11] Ibid. at [41]. [12] Ibid. at [43]. [13] Ibid. at [45]. [14] (2013) 16 HKCFAR 195 at [83] (emphasis in original). [15] (2010) 13 HKCFAR 142 at [140]-[166]. [16] (2013) 16 HKCFAR 195 at [88]-[90]. [17] The further alternative of provoking others to commit a breach of the peace (“or will by such conduct provoke other persons to commit a breach of the peace”) can be ignored for the purposes of the present analysis in respect of both “the intended limb” and “the likely limb”. [18] The Appellant’s Printed Case at [5], [71]-[78], [79]-[81] and [82]-[83]. [19] Case for the Respondent at [3], [60]-[63] and [72]. [20] [2018] HKCFI 2715, [2020] 1 HKLRD 1263. [21] Ibid. at [66]. [22] Ibid. at [73] (emphasis in original). [23] Ibid. at [78(4)]. [24] (2013) 16 HKCFAR 837 at [39]. [25] [2020] HKCFA 42, (2020) 23 HKCFAR 518 at [107] and [110]. [26] (2013) 16 HKCFAR 195 at [90]. [27] (2010) 13 HKCFAR 142 at [105], [110]. [28] R v Howell [1982] QB 416 at 427, referred to in HKSAR v Chow Nok Hang (2013) 16 HKCFAR 837 at [77]-[82]. [29] Hong Kong Legislative Council, Official Report of Proceedings (1 November 1967) at p.442. [30] [2012] 5 HKLRD 556 at [40]. [31] (2013) 16 HKCFAR 837 per Chan ACJ at [14] and Ribeiro PJ at [86], [93]-100]. [32] Yau Wai Ching v Chief Executive of HKSAR (2017) 20 HKCFAR 390 at [8]-[9], [15] and [28]. [33] Statement of Findings at [158] and [162]. [34] Ibid. at [166]. [35] Ibid. at [165]. [36] [2020] HKCFI 2152 at [36]-[44]. [37] Ibid. at [48]. [38] [2018] HKCFI 2715, [2020] 1 HKLRD 1263 at [78(4)]. 
Mr Justice Ribeiro and Mr Justice Fok PJJ: 1. We have had the advantage of reading in draft the judgment of Mr Justice Spigelman NPJ and respectfully agree with his analysis and reasoning and would likewise dismiss the appeal. 2. The starting-point in this appeal is that Ricky Cheng (“Ricky”) owed the usual fiduciary duties of a director as Smart Wave Limited’s (“Smart Wave”) sole director. The central question was whether, in the events that occurred, those duties were limited or cut down so as to enable Ricky to open further restaurants without breaching those fiduciary duties. The argument advanced in favour of such limitation of those duties rests on an agreement of the shareholders having the effect of authorizing or acquiescing in Ricky’s conduct and so preventing enforcement on behalf of Smart Wave of such duties by way of derivative action. 3. For the reasons given by Mr Justice Spigelman NPJ, we are not persuaded that any such agreement has been established. In particular, we do not consider that the 2004 Agreement was of continuing validity after it was discharged by mutual consent in the parties arriving at a settlement agreement reflected in the Hero Elegant Agreement. Even if the 2004 Agreement continued to operate, its content involved an agreement for the establishment of chain restaurants in which its parties would be able to participate as shareholders, rather than an agreement simply allowing Ricky to open further restaurants on his own. Furthermore, the parties to the 2004 Agreement and the Hero Elegant Agreement did not comprise all the shareholders in Smart Wave so that no informal authorization along the lines envisaged in Re Duomatic Ltd,[1] arose. Mr Justice Tang PJ: 4. Itamae sushi is a service mark. Smart Wave Limited, the 2nd defendant, owned and operated the first Itamae sushi restaurant. Ricky was the only director of Smart Wave. This is a derivative action (HCA 304/2011) brought by Jason Poon (“Jason”) on behalf of himself and all the other shareholders in Smart Wave except Ricky, the 1st defendant. The complaint is that Ricky was in breach of fiduciary duties as a director of Smart Wave in that he had opened and operated other Itamae sushi restaurants.[2] The derivative action was heard together before Madam Justice Mimmie Chan with what I will call the Hero Elegant Action[3] (HCA 1269/2008). 5. The events which led to Smart Wave operating the first Itamae sushi restaurant are important for a proper understanding of the complaint. Jason and Ricky told largely the same story. 6. Briefly stated, it was Jason’s case[4] that, he, his sister Daisy Poon (“Daisy”),[5] Ricky and Shigemitsu Katsuaki (“Katsuaki”) were shareholders in a group of companies which were engaged in operating and managing a chain of Japanese style noodle restaurants under the trade name of Ajisen Ramen. They held shares in the Ajisen companies in these proportions: Daisy 48%, Jason 15%, Ricky 23% and Shigemitsu Katsuaki 14% (“the Ajisen shareholders”). The Ajisen shareholders agreed to develop a chain of sushi restaurants using the same model so: “10. … that separate corporate vehicles would be formed to hold the interest of the said chain sushi restaurant business to be established by us. We … would then be allotted shares of and in the said corporate vehicles (‘the 2004 Agreement’).”[6] “11. … Ricky Cheng was tasked to coordinate the establishment and operation of the chain sushi restaurant … [Jason] was tasked to coordinate and supervise the decoration works of the new sushi restaurants.” “23. … Under the 2004 Agreement, [the Ajisen shareholders] agreed that they would each be allotted shares in the corporate vehicles formed to hold the interest of the chain sushi restaurant business. This was never confined to the 1st Itamae Sushi Restaurant, and would equally apply to the subsequent Itamae Sushi Restaurants established as a chain.” 7. The first Itamae restaurant was held in the name of Smart Wave. In due course, Daisy and Jason were allotted 24%[7] and 10% of the shares in Smart Wave respectively, Ricky 23% and Katsuaki[8] 15%, totalling 72%. The remaining 28% were allotted to 5 other shareholders (the minority shareholders).[9] I will deal with the significance of the minority shareholders later. Jason said Ricky’s explanation for the allotment to the minority shareholders was that: “… these people were either suppliers or key staff who would be working in the Itamae Sushi Restaurant and he needed them to ensure a smooth and successful operation for the new business.”[10] 8. Jason also said: “20. … The first Itamae Sushi Restaurant in Tsimshatsui proved to be a huge success. We therefore proceeded to open more Itamae Sushi Restaurants and Ricky Cheng was again tasked to coordinate their establishment.” 9. The second and third were opened in March 2005 and February 2006. Jason was responsible for the supervision of their decoration and he said, as was the practice with the Ajisen restaurants, he only charged a nominal fee for supervising such work. Jason said he asked Ricky from time to time when he would be allotted shares in these two restaurants but Ricky kept saying he was working on it. Later, in May/June 2006, he received a fax from Ricky proposing that in respect of these two Itamae sushi restaurants, Daisy would be allotted 23% and Jason allotted only 7% of the shares in them. When Jason complained to Ricky about the substantial reduction in his shareholdings from 10% to 7%, Ricky again said that he would need to allot some shares to other parties to facilitate the operation of the Itamae sushi business. 10. Cutting a long story short, the dispute over the allotment of shares in the subsequent Itamae restaurants led to the Hero Elegant Agreement. Another background detail which should be mentioned is that around the end of 2005, in order to go public, the Ajisen shareholders were advised to bring the Ajisen companies under one umbrella company,[11] and to facilitate the floatation, it was decided to segregate the management of the Ajisen Ramen business from the Itamae sushi business. At that time, Ricky was the managing director of the Ajisen Ramen companies in Hong Kong[12] group and the sole director of the companies operating Itamae sushi restaurants. The Ajisen shareholders were advised that this might not be acceptable to the listing Committee.[13] So Ricky resigned from Ajisen companies to “focus on the Itamae Sushi business”[14] and Daisy and Jason decided to focus on the Ajisen Raman business.[15] Fine Elite Group Limited (“Fine Elite”) was chosen to be the corporate vehicle to hold Jason and Daisy’s “shares in the Itamae Sushi business.”[16] Fine Elite entered into the Hero Elegant Agreement. Jason added: “36. As I mentioned earlier, Ricky Cheng agreed to allot 23% shares in the other Itamae Sushi Restaurants to Daisy and 8 % shares to me. … 38. … To prevent Ricky Cheng from reneging on this agreement again by failing to allot shares in future Itamae Sushi Restaurant, we decided that Ricky Cheng should be required to vest the interests in all existing (save for Smart Wave) and future Itamae Sushi Restaurants in a holding company and Daisy and I were to be allotted, through Fine Elite, the shares in the holding company…” 11. The Hero Elegant Agreement was dated 16 September 2006. At the time there were a total of 4 Itamae restaurants each held by a separate company, namely, D2 (Smart Wave), D3, D4 and D6. Three others were opened shortly afterwards. D5 was opened in October 2006, D8 November 2006 and D7 December 2006. The detailed provisions of the Hero Elegant Agreement are unimportant. What is important for present purpose is that, by HCA 1269/2008 Fine Elite sought specific performance of the Hero Elegant Agreement against Ricky so that Fine Elite should hold 31% of the shares in a holding company in which shares in all the Itamae companies subsequent to Smart Wave should be held. Ricky’s defence in HCA 1269/2008 was that Fine Elite had repudiated the Hero Elegant Agreement and the repudiation had been accepted by him such that he was discharged from further performance.[17] After trial, Mimmie Chan J held in favour of Ricky and dismissed Fine Elite’s claim based on the Hero Elegant Agreement. There was no appeal. The Hero Elegant Action was heard at the same time as the derivative action which was commenced by Jason in 2011 (HCA 304/2011). It was Jason’s case that the derivative claim represented his fall-back position and was made in case the Hero Elegant Action failed.[18] The learned trial judge dismissed the derivative claim against Ricky in respect of the subsequent Itamae restaurants but ordered Ricky to pay damages to be assessed in respect of the Itacho restaurants. The Court of Appeal reversed the judge in respect of the Itamae restaurant and ordered that in respect of the Itacho restaurants, Jason could elect on behalf of Smart Wave for an enquiry into damages/equitable compensation or an account of profits. 12. The appeal committee granted leave to appeal on the following two questions: (1) Whether the “no conflict rule” applies to a director of a chain business where the agreed modus operandi was to have one company for one agreed operation. Where the company was of a “limited nature” as found by the Trial Judge, with the agreed modus operandi of only operating one restaurant, whether the principle set out in In re Duomatic Ltd. [1969] 2 Ch 365 at 373 and EIC Services Ltd & Anor. v. Phipps & Ors [2003] BCC 931 at para 122 (“the Duomatic principle”) applies as found by the Court of Appeal; (2) In the context of a derivative action, where the only nominal representative has himself agreed to the modus operandi of one restaurant per company, and where he has acted and taken benefit on that basis by signing and suing on the Hero Elegant Agreement, whether he is entitled to rely on a contrary or inconsistent stance at all [Nurcombe v. Nurcombe [1985] WLR 370 at 376G-377D] or without the company calling contrary evidence from other shareholders. 13. In the derivative action, the principal complaint was that Ricky was in breach of fiduciary duties as a director of Smart Wave in establishing the later Itamae sushi restaurants. An additional complaint was that Ricky had started another chain of sushi restaurants under a confusingly similar name, namely, Itacho. I will not deal separately with the Complaint about Itacho. The resolution of this complaint turns also on the scope of Ricky’s fiduciary duties and whether Smart Wave must be taken to have agreed with what the learned judge described as “the limited nature of Smart Wave’s rights in the operation of one Itamae restaurant”.[19] In other words, that Smart Wave was established to operate a single Itamae restaurant as part of a chain of Itamae restaurants which would be owned by the Ajisen shareholders.[20] 14. As Ricky was a director (indeed, the only) of Smart Wave, I have no doubt that he was not entitled to make a profit or put himself in a position where his interest and duty conflict, unless Smart Wave could be taken to have agreed otherwise. That a director’s fiduciary duties may vary or be limited depending on the circumstances of the particular case is made clear by Lord Upjohn in Phipps v Boardman [1967] 2 AC 46 at 123 where he said: “Rules of equity have to be applied to such a great diversity of circumstances that they can be stated only in the most general terms and applied with particular attention to the exact circumstances of each case. The relevant rule for the decision of this case is the fundamental rule of equity that a person in a fiduciary capacity must not make a profit out of his trust which is part of the wider rule that a trustee must not place himself in a position where his duty and his interest may conflict. I believe the rule is best stated in Bray v Ford by Lord Herschell, who plainly recognised its limitations: ‘It is an inflexible rule of a Court of Equity that a person in a fiduciary position, such as the respondent’s, is not, unless otherwise expressly provided, entitled to make a profit; he is not allowed to put himself in a position where his interest and duty conflict.’” (my emphasis) 15. Mason J said to similar effect in Hospital Products Ltd v United States Surgical Corporation (1984) 156 CLR 41 at 97: “That contractual and fiduciary relationships may co-exist between the same parties has never been doubted. Indeed, the existence of a basic contractual relationship has in many situations provided a foundation for the erection of a fiduciary relationship. In these situations it is the contractual foundation which is all important because it is the contract that regulates the basic rights and liabilities of the parties. The fiduciary relationship, if it is to exist at all, must accommodate itself to the terms of the contract so that it is consistent with, and conforms to, them. The fiduciary relationship cannot be superimposed upon the contract in such a way as to alter the operation which the contract was intended to have according to its true construction.” 16. So the pertinent question is whether Smart Wave could be taken to have agreed that Ricky[21] could operate other Itamae restaurants which are not owned by Smart Wave. In para 6 above, I have set out what I regard as the clearest description of the 2004 Agreement, and in para 8 above, the role which Ricky was to play in the other Itamae restaurants and I will not repeat them. Such evidence, if accepted by the learned trial judge, would provide a firm foundation for the finding that Smart Wave was established pursuant to the 2004 Agreement. If the Ajisen shareholders were the only shareholders in Smart Wave, I have no doubt that Smart Wave could not assert that Ricky was not entitled to establish the other Itamae restaurants. That is so notwithstanding that Smart Wave was not itself a party to the 2004 Agreement. Re Duomatic[22] is accepted authority that: “(1) that where it could be shown that all the shareholders with the right to attend and vote at a general meeting had assented to some matter which a general meeting of the company could carry into effect, the assent was as binding as a resolution in general meeting …” [23] 17. Any fiduciary duties which Ricky might owe to Smart Wave must be consistent with and conform to, the terms of the 2004 Agreement. But there were the minority shareholders and I turn to consider their significance. 18. In para 63 the learned trial judge said: “Hence, as one of the restaurants in the chain, Smart Wave was never intended by its shareholders to have the exclusive right to carry on the sushi restaurant business using the Itamae name or Service Marks. Even on Jason’s and Daisy’s case, the shareholders of Smart Wave knew, intended and agreed that other companies would be set up to operate other sushi restaurants, using the Itamae name and Service Marks. …” 19. The learned judge then concluded that: “Ricky, as a director of Smart Wave, cannot be said to have acted in breach of his duties to Smart Wave by operating other sushi restaurants using the Service Marks.”[24] 20. The Court of Appeal reversed the learned trial judge because they thought she had overlooked the minority shareholders.[25] Cheung JA in a judgment which was agreed to by the other members of the court said: “Mr Paul Shieh SC and Miss Linda Chan SC on behalf of Jason took the short point that the shareholders of the Ajisen Group and Smart Wave are different and there was no evidential basis for the Judge to come to the view that the shareholders of Smart Wave had come to such an understanding or agreement. Further Ricky’s evidence is contrary to any such agreement or understanding by the shareholders of Smart Wave because he had accepted in cross-examination that in 2004, at the time of the alleged agreement, he had not told Jason and Daisy that he could open further Itamae restaurants on his own.”[26] 21. With respect, I cannot agree with the Court of Appeal. At para 62 of her judgment the learned trial judge referred to the shareholders of the Ajisen group when she was considering the 2004 Agreement and its offshoot the Hero Elegant Agreement. In para 63, she referred expressly (twice) to the Smart Wave shareholders. There is no reason to think that the learned judge had confused the Ajisen shareholders with the Smart Wave shareholders. The learned judge was fully aware of Re Duomatic Ltd and said the Re Duomatic principle[27] required the assent of “all shareholders of Smart Wave” and that was why the Hero Elegant Agreementwas not binding on Smart Wave.[28] That the learned judge was plainly aware that the Ajisen and Smart Wave shareholders were not identical is supported by her observation during the cross-examination of Ricky: “Judge: hold on. We need to make this clear, because your previous question you were talking about shareholders of Itamae, and now you were talking different shareholders of Ajisen or what, Mr Cheng?” [29] 22. The Court of Appeal also said that Ricky in his evidence admitted that he had not told Jason or Daisy that he could open further Itamae restaurants on his own. We have been given the relevant transcript. But I agree with Ms Eu SC for Ricky that the question was directed to Ricky’s claim that Jason and Daisy were not entitled to any shares in the other Itamae companies and that Ricky himself was solely entitled. In other words, the emphasis was “on his own”. That this was the point of the question and how it was understood by the learned trial judge is clear from para 5 of her judgment when she said: “On his part, Ricky denies that the Itamae restaurants were an extension of the Ajisen ramen restaurants. He claims that he had planned the operation of the Itamae sushi restaurants as his own chain, and not as a mere franchisee of a Japanese corporation, to bear his own style and his personal stamp. According to Ricky, the trademarks “ITAMAE”, “板前” and the associated logo were his own creation, and he had offered shares in Smart Wave to Daisy and Jason with the intention that Smart Wave was to operate the first Itamae restaurant only. Ricky claims that it was all along his intention to remain the sole owner of the Itamae name and to retain the right to use the “ITAMAE” trade marks for other sushi restaurants in the chain to be owned and operated by him. However, he admits in his testimony in court that he had not informed Ricky and Daisy of this in 2004.”[30] 23. In para 5, I said Jason and Ricky told largely the same story. When Jason’s evidence[31] is compared with para 15 of Ricky’s defence in the derivative action,[32] it is clear that the essential difference between the two versions is that according to Jason, pursuant to the 2004 Agreement, the Ajisen shareholders[33] would be entitled to be allotted shares in the subsequent Itamae sushi restaurants,[34] and Ricky’s case was that he alone was entitled to any share in the later Itamae restaurants. What Jason and Ricky's case had in common was that Smart Wave would be one of a chain of separately owned sushi restaurants. Nor does it appear that Ricky’s claim that he had reached an agreement or understanding with the shareholders of Smart Wave that Smart Wave was only entitled to run one Itamae sushi restaurant in a chain of such restaurants[35] was disputed. There is no indication that Jason put forward a different case. Indeed any such different case would have been inconsistent with his stance in the Hero Elegant Action. In the course of the submissions, I asked Mr Jat SC who together with Ms Linda Chan SC, appeared for Jason in this appeal, whether Ricky was cross-examined on his claim that he had reached an agreement or understanding with the shareholders of Smart Wave that Smart Wave was only entitled to operate one Itamae sushi restaurant in a chain of such restaurants. I was not told that there was such cross-examination. 24. I have had the advantage of reading Mr Justice Spigelman’s judgment in draft. His Lordship is of the view that the first question should be answered against Ricky and his appeal should be dismissed. Ribeiro and Fok PJJ agreed with his Lordship. With respect, I am unable to agree. 25. In paras 110-115, of Mr Justice Spigelman’s judgment, his Lordship discussed the single purpose evidence and concluded at para 115 that Ricky’s and Daisy’s: “… expectation or agreement was expressly interconnected with an expectation and agreement that they would be substantial shareholders in each such vehicle. These two elements were so closely interconnected, that it cannot be concluded that they had agreed to the restriction on Smart Wave on its own. In a case involving the equitable jurisdiction of the Court, the Appellant cannot rely only on that part of the promoters’ agreement that suits his case.” 26. With respect, this is a mischaracterization of Ricky’s conduct. It was not for Ricky to choose what suited him. He entered into a contract and he was bound by it. If he was in breach, as he clearly was, the other parties to the contract could hold him to it. Their remedy was to enforce the contract. This is not a case of discharge by accepted repudiation. Ricky’s breach was resolved by the Hero Elegant Agreement. There is no difference in principle between this settlement and Ricky paying damages for the breach. The fact that the Hero Elegant Agreement miscarried through no fault of Ricky makes no difference. Smart Wave was not a party to the 2004 Agreement, the significance of the 2004 Agreement to Smart Wave is Ricky’s case that to the knowledge and agreement of all the shareholders of Smart Wave it was established pursuant to the 2004 Agreement to be the first in a chain of independently owned Itamae restaurants (limited purpose company). 27. In paras 10 and 11 of Jason’s witness statement he made clear that it was his case that Smart Wave was established pursuant to the 2004 Agreement, and he expressly stated that Daisy was party to the 2004 Agreement. Daisy had not denied that. If Ricky was in breach of the 2004 Agreement, the other parties to the 2004 Agreement could seek enforcement of it. That was their remedy. It seems to be accepted that if Ricky had not been in breach of the agreement. Smart Wave being a limited purpose company would have no claim against Ricky, but because Ricky was in breach of the 2004 Agreement, Smart Wave had a claim for breach of fiduciary duties. It is not clear whether in that case both the other parties to the 2004 Agreement and Smart Wave would have a claim, the former in contract and the latter for breach of fiduciary duties. Nor, whether, upon breach, Smart Wave’s claim would immediately arise and the contractual claim disappears. Or that they co-exist. Presumably, Ricky would not have to pay two sets of damages. If so, does it mean if the contractual claim was brought first, Smart Wave would have no claim. It is not clear whether that would depend on whether the contractual claim is compromised or failed. 28. Here, Ricky’s breach predated the Hero Elegant Agreement. And the breach was compromised by the Hero Elegant Agreement. What is the effect of this compromise to Smart Wave’s claim? Suppose the Hero Elegant Agreement was duly performed, how would that affect Smart Wave’s right? Indeed, what if the Hero Elegant Action was settled, how would that leave Smart Wave? It may be pertinent to ask what was Smart Wave’s right vis-a-vis Ricky from the time of the initial breach of the 2004 Agreement to the discharge of the Hero Elegant Agreement? This is not an idle question, the answer should throw light on whether Smart Wave’s right on fiduciary duty could depend on Ricky’s breach of either the 2004 Agreement or the Hero Elegant Agreement. In my opinion, Ricky’s breach of these agreements (e.g. whether by cutting the other parties out of their shares or by giving less than what they were entitled to) can have no relevant impact on Smart Wave’s right against Ricky. The fact that Smart Wave’s shareholder had agreed that it should be a limited purpose company means that Ricky’s fiduciary duties towards Smart Wave must conform to and be consistent with that reality. I do not believe the fact that Smart Wave’s claim is based on breach of statutory duties makes any difference. The relevant question is what were those duties and that depends on whether it was agreed by its shareholders that Smart Wave should be a limited purpose company. 29. Also, in my view, on the facts, the overwhelming probability was that the minority shareholders had been allotted shares on the understanding that Smart Wave was to be one company in a chain of separately owned companies, what the learned trial judge called “the limited nature of Smart Wave’s rights in the operation of one Itamae restaurant”. There was simply no reason not to tell them. Nor any advantage in not doing so. Nor on the facts, was it likely that they would have objected or that if they had, it could not be resolved by not admitting that person(s) as a shareholder. It is common sense that promoters (and the Ajisen shareholders could be regarded as promoters) would inform persons who might be interested the nature of the investment they were promoting. It is just good business sense. Also, one would be asking for trouble otherwise. Nor would any elaborate explanation be needed, it follows from its limited purpose Smart Wave could not object to the establishment of independently owned subsequent restaurants. Nor was the limited nature of Smart Wave something the promoters were likely to have overlooked. That was the raison d’etre of the establishment of Smart Wave. As Jason’s evidence showed Ricky was the prime mover in the enterprise and they were following the Ajisen Ramen restaurant model. And it was Ricky who recruited the minority shareholders. I think it is inconceivable that he would not have made the minority shareholders aware of the limited nature of Smart Wave. The evidence was that the minority shareholders were Ricky’s friends and that it was Ricky who paid for the shares for all the minority shareholders[36] except those which were allotted to Kong.[37] Indeed, they were shareholders because, Jason said as Ricky explained, they were suppliers or key staff and Ricky needed them for the running of the Itamae restaurants. If the plan was for separately owned chain companies to run Itamae restaurants, and the evidence of Jason, Daisy and Ricky left no doubt that it was indeed so and the minority shareholders were admitted or given shares because they were key staff or suppliers, one would naturally expect that there would be different minority shareholders in different restaurants (such as managers). This should also militate against Smart Wave having any right over the later restaurants. Also, Jason relied on the fact that “at least Mak, Teraguchi and Wong”[38] would vote with Ricky such that Ricky controlled more than 50%[39] of the Smart Wave shares to justify a derivative action. It is a fact that none of the minority shareholder gave evidence, but given the narrow disagreement between Ricky’s and Jason’s case, i.e. whether Ricky was solely entitled to run other Itamae restaurants, that does not surprise me. Given the fact that the Itamae chain was a success, one would be surprised that the minority shareholders would have been so passive if they were not aware of the limited nature of Smart Wave. There is also the question of the ownership of the service marks. Smart Wave had not claimed that it owned the service marks[40] and as the judgment on the Hero Elegant Agreement showed, clause 14 of the Hero Elegant Agreement: “is clearly an acknowledgment that Ricky had the right to grant a licence for the use of the Service Marks, a right which only the owner of a trade mark can have.”[41] 30. It does not matter that the learned trial judge made no finding on the actual ownership of the service marks, it is significant that Smart Wave made no claim to ownership because it is supportive of the view that Smart Wave was established to the knowledge of its shareholders to be a single restaurant company in a chain, and Smart Wave was not entitled to open other Itamae restaurants. Lastly, a trial judge has the great advantage of seeing the witnesses and observing the unfolding of the parties’ cases before him. They provide the trial judge with an insight into the case which an appellate court does not share. Here the learned trial judge dealt with the knowledge of the shareholders of Smart Wave briefly but in no uncertain terms. What arguments were raised and to what extent they were pressed may explain the learned trial judge’s brief but firm response. How much evidence is required to tip the balance is a matter of judgment. With respect, there is no reason to think that the learned trial judge’s finding was plainly wrong. 31. For the above reasons, I do not believe that Court of Appeal was entitled to overturn the learned trial judge’s clear conclusion that Ricky had not acted in breach of his duties to Smart Wave by operating other Itamae restaurants. I would allow Ricky’s appeal. 32. I turn now to the second question which would only arise if I was wrong on the first question and that Ricky was in breach of fiduciary duties and it is sought to deny relief because of Jason’s conduct. Like Mr Justice Spigelman, with respect, I am also troubled by the fact that this question is raised for the first time in this court. Leave was granted on the second question, no doubt because it was felt that on the basis of Nurcombe v Nurcombe [1985] WLR 370, Jason might have been refused relief. But for the reasons given by his Lordship, I agree that the court should not entertain the second question. Mr Justice Bokhary NPJ: 33. On the primary facts found at first instance, which findings are justified by the evidence, it is to be concluded that all the individuals concerned always understood that the restaurant operated by the company in this derivative action would be the pioneer in a chain of restaurants, each operated by a different company. So there was no question of the pioneering company operating any further restaurant. And in so far as customers who would otherwise have patronized the pioneering restaurant are diverted to a restaurant further along the chain, such diversion must be taken to be within what the pioneering company had acquiesced in through the human agency by which it functioned. If any shareholder in the pioneering company is aggrieved by reason of not being allotted shares in a company operating a restaurant further along the chain, redress for such grievance is to be sought in an ordinary action based on an allegation of breach of a contractual obligation owed to the aggrieved shareholder, not in a derivative action based on an allegation of a breach of a fiduciary duty owed to the pioneering company. 34. Accordingly, on the basis of the argument advanced under the first of the two questions on which leave to appeal was granted, I would allow the appeal so as to dismiss the derivative action entirely. On the first question, therefore, I agree with Mr Justice Tang PJ, and on the second question, I agree with him and all the other members of the court. 35. Before parting with the case, I wish, first, to thank counsel on both sides for their typically able arguments. Secondly, I wish to acknowledge that I have found the judgments of the learned judges in the courts below helpful on all points - even those on which I have felt unable to share their views. Mr Justice Spigelman NPJ: 36. This appeal arises from a derivative action brought by the 1st Respondent, Poon Ka Man Jason (“Jason” or “the Respondent”), as a shareholder of the 2nd Respondent, Smart Wave Ltd (“Smart Wave”), against the 1st Appellant, Cheng Wai Tao (“Ricky”), with respect to his conduct in relation to the 2nd to 30th Appellants. Jason is a 10% shareholder of Smart Wave. Ricky is the registered owner of 38% of the shares in Smart Wave. 37. Smart Wave operated a sushi restaurant under the name “Itamae”. The 2nd to 9th Appellants each operated a sushi restaurant under the name “Itamae” (the “Itamae restaurants”). The 10th to 30th each operated a sushi restaurant under the name “Itacho” (the “Itacho restaurants”). 38. On 24 May 2013 the Trial Judge, Mimmie Chan J, held that Ricky had breached his fiduciary duties as a director of Smart Wave by operating the Itacho restaurants. Her Ladyship awarded damages to Smart Wave up to 2010, when Smart Wave ceased to operate its sushi restaurant. The Trial Judge dismissed an identical claim with respect to the Itamae restaurants. 39. Jason applied for variation of the judgment of the trial judge so as to allow Smart Wave to elect for an account of profits. On 6 December 2013, the trial judge dismissed the application on discretionary grounds. 40. Jason appealed and Ricky (and all the Appellants) cross appealed. The Court of Appeal rejected the cross appeal, which sought to overturn the First Instance judgment with respect to the Itacho restaurants. The Court of Appeal allowed Jason’s appeal and held that Ricky’s operation of both the Itamae and Itacho restaurants constituted breaches of Ricky’s fiduciary duties to Smart Wave. The Court of Appeal further held that Jason was entitled to elect between an account of profits and damages. The Court dismissed Jason’s appeal with respect to the 2010 temporal limit on damages or an account. Background Facts 41. Ricky, Jason and Daisy, together with a business associate Shigemitsu Katsuaki (“Shigemitsu”) had, since 1996, conducted a successful chain of Japanese noodle restaurants under the trade name “Ajisen Ramen”, under licence from Japan. For present purposes, it is an important feature of this business venture that each of these restaurants was operated by a separate corporate entity, in which the natural persons mentioned were the principal shareholders, of whom Daisy was the largest. Some of the companies had minority shareholders, whose identity varied from one company vehicle to another. 42. It was Jason’s case throughout, a case accepted by the trial judge and the Court of Appeal that, in 2004, the shareholders of the Ajisen Group decided to go into the sushi restaurant business on a similar basis to their existing arrangement. The first restaurant of an intended chain was opened by Smart Wave, using the trade name Itamae, in August 2004. 43. Smart Wave’s shareholders were – Ricky (23%), Shigemitsu (15%), Jason (10%) and his sister Daisy (24%) (who were all Ajisen shareholders), Kong Yiu Wai (10%), Mak Kin Shing (8%), Sato Akira (4%), Teraguchi Tadayoshi (4%) and Wong Yiu To (2%). By 12 March 2007 this had changed to Ricky having (38%), Shigemitsu having transferred his 15% to Ricky. 44. After the opening of the first Itamae restaurant, Ricky proceeded to act on the basis that he was entitled to develop the chain of sushi restaurants in his own right as the sole owner of the chain, subject to him deciding that additional shareholders should be introduced. Three additional Itamae restaurants were opened within two years of the first. Three more were in an advanced stage of development. Ricky, who was the sole director of Smart Wave, was the sole shareholder of each of the separate companies that operated each of these restaurants. That was also the case for all subsequent restaurants with, the Court was informed, one exception where there was a minority shareholder, unrelated to these proceedings. 45. Jason and Daisy regarded the expansion of the sushi chain as a contravention of the agreement that had been made by the shareholders of the Ajisen Group. A legal dispute developed. That dispute coincided with a dispute about the conduct of the Ajisen business and with the need to change the structure and shareholding of the Ajisen Group, in preparation for a public listing of that business. 46. The resolution of the parallel disputes was carried into effect in September 2006 by the parties entering into two written contracts: the “Favour Will Agreement”, with respect to be Ajisen restaurants, and the “Hero Elegant Agreement”, with respect to the Itamae restaurants. The latter is relevant for present purposes. That Agreement did not extend to the Itacho restaurants, which were not then in existence. 47. Whatever the rights Jason and Daisy might have had in the Itamae restaurant business, they agreed, by the entry of their company, Fine Elite Group Ltd (“Fine Elite”), into the Hero Elegant Agreement, to a new structure. Fine Elite accepted a 31% shareholding in Hero Elegant, whereas Ricky held 69%. The three new Itamae restaurants, and future restaurants, would be operated by subsidiaries of Hero Elegant. Smart Wave was not brought under this new arrangement, although it was agreed that that may happen in the future. 48. Ricky had applied for registration of the name “Itamae Sushi” as Service Marks, in his own name, in the PRC. He was obliged, however, under the Hero Elegant Agreement, to grant an exclusive licence of the Service Marks to the Hero restaurants. Fine Elite attempted to register “Itamae” as a Service Mark in Hong Kong, after Jason found out that Ricky had applied for registration in the PRC. Ricky asserted that this constituted a repudiation of the Hero Elegant Agreement and, in December 2006, he accepted the repudiation. 49. Ricky continued to open new Itamae restaurants in separate companies, of which he was sole shareholder. From July 2007 he commenced to open sushi restaurants using the trade name Itacho again in separate corporate vehicles. Jason and Daisy asserted that this conduct was a breach of the Hero Elegant Agreement. On 9 July 2008, Fine Elite commenced proceedings for specific performance of that Agreement, (the “Hero Elegant Action”). On 23 February 2011, Jason commenced the Derivative Action, expressed to be an alternative to the Hero Elegant Action. The First Instance Decision 50. The trial judge heard the two proceedings together. There is no appeal from that part of her judgment of 24 May 2013, concerning the Hero Elegant Action. 51. As indicated above, Daisy and Jason claimed they had entered into an agreement in 2004 with Ricky to develop a chain of sushi restaurants, each of which would be conducted by a separate corporate vehicle and in which Daisy and Jason, as well as Ricky and Shigemitsu, would be shareholders. This appeal proceeded on the basis that the Hero Elegant Agreement constituted a comprehensive resolution of the disputes about the alleged breach of the 2004 Agreement. 52. The Hero Elegant Action was instituted by Fine Elite on the basis that Ricky continued to manage and operate the Itamae restaurants, and subsequently the Itacho restaurants, on his own account. He did not honour the terms of the Hero Elegant Agreement in that he held all the shares in his own name. Fine Elite sought specific performance of the Agreement and damages or, alternatively, an account of profits. In his Defence, Ricky claimed that the Hero Elegant Agreement had been repudiated by Fine Elite and that he had accepted the repudiation. Accordingly, he was discharged from performance. 53. Her Ladyship held that Ricky was entitled to registration of the marks; that Fine Elite had repudiated the Hero Elegant Agreement and that Ricky had accepted the repudiation. Accordingly the Hero Elegant Action failed and the alternative derivative action on behalf of Smart Wave had to be determined. 54. Her Ladyship’s key findings with respect to the Itamae restaurants were: “61. Irrespective of whether Ricky is the owner of the Service Marks, it has never been pleaded that Smart Wave is the owner. On Jason’s or Fine Elite’s case in HCA 1269/2008, the shareholders of the Ajisen Group (Daisy, Jason, Ricky and Shigemitsu) agreed to develop Japanese-style chain restaurants using corporate vehicles in which all of them would be interested as shareholders, and it was pursuant to such agreement that Smart Wave was incorporated in 2004 to manage and operate the Itamae restaurant at Granville Road only. Under the Hero Elegant Agreement, which was signed to resolve the disputes over Daisy’s and Jason’s claims to the Itamae restaurants, Ricky and Fine Elite were to become shareholders of Hero, which with its subsidiaries were to continue and carry on the business of operating chain restaurants using the Service Marks. 62. On the case advanced by Jason and Daisy, the shareholders of the Ajisen Group had agreed ‘to develop chain sushi restaurants in furtherance of the then existing business’ (of the Ajisen Group), and had further agreed that ‘separate corporate vehicles would be formed to hold the interest of the said chain sushi restaurant business to be established’ (emphasis added), following the same shareholding of the Ajisen Group (paragraph 12 of Jason’s witness statement made on 24 October 2011). There is no doubt that, even on the Poons’ case, the intention was for a chain of Itamae restaurants to be established, using different corporate vehicles to hold and to operate different restaurants in the chain, all using the Service Marks. The Hero Elegant Agreement also envisaged this. Clause 14.5 provides for all chain restaurants conducting the business defined in the agreement to be owned and operated by ‘a Group Company’, with Hero being the holding company of all such group companies owning and operating the business of the sushi restaurants. 63. Hence, as one of the restaurants in the chain, Smart Wave was never intended by its shareholders to have the exclusive right to carry on the sushi restaurant business using the Itamae name or Service Marks. Even on Jason’s and Daisy’s case, the shareholders of Smart Wave knew, intended and agreed that other companies would be set up to operate other sushi restaurants, using the Itamae name and Service Marks. I agree with Leading Counsel for Ricky that Smart Wave is not in a position to complain about the use of the Service Marks by other companies, or of the existence and operation of other sushi restaurants in the chain. Ricky, as a director of Smart Wave, cannot be said to have acted in breach of his duties to Smart Wave by operating other sushi restaurants using the Service Marks.” 55. Her Ladyship’s key findings with respect to the Itacho restaurants were: “65. Whereas I accept that the shareholders of Smart Wave knew and hence had consented to the operation of other sushi restaurants in the chain by use of the Service Marks and the name ‘Itamae’, there is no evidence in this case on which I can find that the shareholders of Smart Wave had consented to the operation of sushi restaurants under the ‘ITACHO’ name. As a director of Smart Wave, it is undisputable that Ricky was under a fiduciary duty to act in the best interests of Smart Wave, and prima facie, it is against the interests of Smart Wave for its director Ricky to operate and consent to the operation of a competing sushi restaurant, under the name of ‘ITACHO 板長’ which I consider to be confusingly similar to the name ‘ITAMAE板前’. Due to the similarity between the names ‘ITAMAE 板前’ and ‘ITACHO板長’, I consider that the inevitable inference is that customers of the Itamae restaurants would patronize the Itacho restaurants in the belief that they are connected or related, such that the Itacho restaurants benefit from the goodwill in and reputation of the Itamae name. Leading Counsel for Ricky argued on his behalf that the Itacho restaurants complement, rather than compete with, the Itamae restaurants and the Itamae business, but there is no evidence whatsoever from Ricky to support such claim.” 56. In the subsequent judgment, on the account of profits issue, her Ladyship referred at [12], to “the limited nature of Smart Wave’s rights in the creation of one Itamae restaurant”. The Appellant relied on this formulation in this Court. The Judgment of the Court of Appeal 57. There was no appeal to the Court of Appeal from the First Instance decision on the Hero Elegant Action. 58. With respect to the appeal from the dismissal at First Instance of the case on the Itamae restaurants, Cheung JA noted at para 4.5: “In brief, the Judge held that the shareholders of Smart Wave knew, intended and agreed that Smart Wave would not have the exclusive right to operate another Itamae restaurant. On the contrary, other companies would be set up by Ricky to operate other Itamae restaurants of which the shareholders would not and could not complain.” 59. His Lordship went on to note at para 4.10: “The success of Ricky’s case really hinges on the evidence in support of the Judge’s finding on the agreement by the shareholders of Smart Wave. This is important because while there might be discussion, understanding or agreement amongst the shareholders of the Ajisen Group, the Ajisen Group and Smart Wave have different shareholders.” 60. His Lordship set out an extract from Jason’s witness statement and concluded: “4.11 Reading Jason’s witness statement in its context, he was referring to the shareholders of the Ajisen Group, namely, himself, Daisy, Ricky and Shigemitsu. He did not say there was an identical understanding of the shareholders of Smart Wave. For example, there was no evidence that Mr Sato Akira had anything to do with the Ajisen Group. 4.12 Likewise, Daisy’s witness statement said that Ricky informed her and the other shareholders of the Ajisen Group, as distinct from the shareholders of Smart Wave, when he opened the second Itamae restaurant”. 61. On this issue Cheung JA concluded: “4.18 In my view, in the absence of evidence in support of the Judge’s finding on the understanding or agreement by the shareholders of Smart Wave in respect of the operation of the other Itamae restaurants, Ricky was clearly in breach of his fiduciary duty towards Smart Wave under the well-established principle reaffirmed in Regal (Hastings). He, as the sole director of Smart Wave, was under a fiduciary duty to act in its best interests, and it was against the interests of Smart Wave for him to operate a competing sushi restaurant business where the business opportunity is diverted from Smart Wave to himself. The burden is on Ricky to demonstrate the propriety of the impugned transaction : Bishopsgate Investment Management Ltd [1993] BCC 120 at 139-140 and 143. 4.19 Much was said about Ricky’s exclusive rights to the Itamae service marks. But this must be a separate issue from the one concerning Ricky putting himself in a position of conflict with that of Smart Wave. 62. With respect to the cross-appeal on the Itacho restaurants findings, his Lordship held: “5.2 As to Ricky’s cross-appeal on the Itacho restaurants, Ms Eu rightly did not, and in my view could not in the absence of evidence, rely on the defence of the understanding or agreement of the shareholders towards the opening of the Itacho restaurants. MsEu, however, referred to the difference of the Itamae and Itacho service marks. While the two designs and their differences were raised in the Court below, they are clearly irrelevant to the issue of Ricky’s breach of fiduciary duty towards Smart Wave in opening the Itacho restaurants because Smart Wave’s claim is not based on the cause of action of passing off where it is relevant to consider the similarity of designs and names. The issue here is really whether Ricky had put himself in a position of conflict by the operation of other sushi restaurants. The other Itamae restaurants and the Itacho restaurants could well be operated by any other names without the slightest difference to Ricky’s liability. In my view the Judge had correctly found Ricky liable for his operation of the Itacho restaurants.” 63. The third matter before the Court of Appeal, namely, whether Respondent is entitled to elect an account for profits, is not before this Court. Leave to Appeal 64. On 5 August 2015 the Appeal Committee of the Court of Final Appeal granted leave to appeal from the Court of Appeal’s Judgment on the basis of the following questions of great, general and public importance, namely:- “(1) Whether the ‘no conflict rule’ applies to a director of a chain business where the agreed modus operandi was to have one company for one agreed operation. Where the company was of a ‘limited nature’ as found by the Trial Judge, with the agreed modus operandi of only operating one restaurant, whether the principle set out in In re Duomatic Ltd. [1969] 2 Ch 365 at 373 and EIC Services Ltd & Anor. v. Phipps & Ors [2003] BCC 931 at para 122 (‘the Duomatic principle’) applies as found by the Court of Appeal; (2) In the context of a derivative action, where the only nominal representative has himself agreed to the modus operandi of one restaurant per company, and where he has acted and taken benefit on that basis by signing and suing on the Hero Elegant Agreement, whether he is entitled to rely on a contrary or inconsistent stance at all [Nurcombe v. Nurcombe [1985] WLR 370 at 376G-377D] or without the company calling contrary evidence from other shareholders” The Pleadings 65. As some aspects of the pleadings were relied on in submissions to this Court, it is appropriate to set out relevant sections from both the Hero Elegant Action and the Derivative Action. 66. The “2004 Agreement” was pleaded in the two actions as follows: Derivative Action “6. Poons, Shigemitsu Katsuaki (‘Katsuaki’) and Ricky Cheng were business partners in a group of companies engaging in the business of operating and managing Japanese style noodle restaurants under the trade names of ‘AJISEN RAMEN’ and 味千拉麵’ (‘Ajisen Group’). 7. In around early 2004, the shareholders of Ajisen Group, namely, Jason Poon, Daisy Poon, Katsuaki and Ricky Cheng, agreed:- 7.1 to develop a chain of Japanese-style sushi restaurants business using corporate vehicle or vehicles in which all of them would be interested as shareholders; and 7.2 the new sushi restaurants would be managed and operated by Ricky Cheng (‘2004 Agreement’). 8. Pursuant to the 2004 Agreement- 8.1 Smart Wave was incorporated on 19th April 2004 as the vehicle to operate the first Itamae Sushi restaurant at Ground Floor, No.14 Granville Road, Tsimshatsui, Kowloon, Hong Kong and other Itamae Sushi restaurants to be established;” Hero Elegant Action “4. Jason Poon, Daisy Poon, Shigemitsu Katsuaki and Ricky Cheng used to be business partners in a group of companies engaged in the business of operating and managing Japanese-style noodle restaurants under the trade name of ‘AJISEN RAMEN’ and 味千拉麵’ (‘Ajisen Group’). 5. In around 2004, the shareholders of Ajisen Group agreed to develop Japanese-style chain restaurants in furtherance of the Ajisen Group’s business using corporate vehicles in which all of them would be interested as shareholders (‘the Agreement’). 6. In pursuance of the Agreement, Smart Wave was incorporated on 19th April 2004 to manage and operate the first Itamae restaurant at Ground Floor, No.14 Granville Road, Tsimshatsui, Kowloon, Hong Kong.” 67. The Appellant drew attention to the inconsistency between para 6 of the Hero Elegant Action and the words “and other Itamae sushi restaurants to be established” in para 8.1 of the Derivative Action. 68. Ricky’s Defence to both actions was in virtually identical terms. In the Derivative Action he pleaded: “9. In or around the beginning of 2004 and prior to the existence of Smart Wave, Ricky Cheng decided to establish and operate sushi business, beginning with one sushi restaurant. 10. Prior to the existence of Smart Wave or the first restaurant, Ricky Cheng had already designed the Service Marks bearing the name ‘ITAMAE SUSHI’, both in English and Chinese. The ‘ITAMAE SUSHI’ Service Marks and concept originated from Ricky Cheng and belonged to Ricky Cheng. 11. In about early 2004 Ricky Cheng intended to create a chain of Japanese sushi restaurants and created the Chinese and English names for ‘ITAMAE SUSHI’.” … 15. Ricky Cheng approached Kong Yiu Wai; Mak Kin Shing; Sato Akira; Teraguchi Tadayoshi; Wong Yui To; Jason Poon and Daisy Poon and reached the following Agreement/Understanding:- a. the sushi business would consist of a sushi restaurant to be operated in the name of Smart Wave to be called or known as ‘ITAMAE SUSHI’; b. the sushi restaurant would be managed by Ricky Cheng at the premises at Ground Floor, No.14 Granville Road, Tsimshatsui, Kowloon, Hong Kong; c. the shareholdings of Smart Wave would be allocated and held in the manner and proportions as pleaded in paragraph 8.3 of the Statement of Claim; d. the Service Marks bearing the name of ‘ITAMAE SUSHI’ was originated from, designed by and solely owned by and belonged to Ricky Cheng. In particular Ricky Cheng did not transfer, assign nor gift the Service Marks to Smart Wave nor any other person(s); e. Ricky Cheng would allow Smart Wave to use the Service Marks to operate one ‘ITAMAE SUSHI’ restaurant; f. the parties had only discussed that Smart Wave would be permitted and authorised to operate one sushi restaurant using the Service Marks, namely that at Ground Floor, No.14 Granville Road, Tsimshatsui, Kowloon, Hong Kong, and Ricky Cheng did not permit, authorise nor licence Smart Wave to establish, engage in or operate any other restaurants or sushi restaurants using the Service Marks; g. Ricky Cheng is and remains at liberty to establish, engage in and operate further or other sushi businesses or sushi restaurants, solely or in association with other parties or person and using the Service Marks including the names, arts and logos bearing the Chinese and English names ‘ITAMAE SUSHI’ h.the other parties (including Smart Wave, save as above agreed) would not and would not attempt to set up, deal with, use or purport to lay claim to the Service Marks which originated from, were designed by and solely owned by and belonged to Ricky Cheng, alternatively, this was an implied term by reason of inter alia obviousness, necessity and/or business efficacy. 16. Further, it is averred that in any event, the Service Marks including the names arts and logos bearing the Chinese and English names ‘ITAMAE SUSHI’ all originated from, were designed by and solely owned by and belonged to Ricky Cheng. 17. Further, it is averred that in any event, Ricky Cheng never purported to nor represented that he would nor did he in fact transfer, assign nor gift those Service Marks to Smart Wave nor any other person(s), save for permitting the one sushi restaurants to be operated by Smart Wave. 18. By reason of Ricky Cheng’s ownership of the Service Marks and his permission for Smart Wave to operate one restaurant using the same, the first ‘ITAMAE SUSHI’ Japanese restaurants was established. 19. It is averred that neither the Plaintiff nor Smart Wave in any event have any locus standi or authority to complain or bring action in respect of any of the Service Marks nor the Itacho Service Marks or any part or portion thereof. 20. Ricky Cheng alone all along had and continues to have the right and entitlement to establish, engage in and operate further or other sushi businesses or sushi restaurants, solely or in association with other parties or persons and using the Service Marks including the names, arts and logos bearing the Chinese and English names ‘ITAMAE SUSHI’” Question 1 69. The determination of the first question for which leave to appeal was granted, as set out at para 33 above, turns on both factual and legal issues. The Legal Issues 70. The Respondent contends that this is a case in which breach of a director’s fiduciary duties is clear, in the absence of the informed consent of the beneficiary, relevantly, by the shareholders of Smart Wave, which did not occur. The Appellant’s response is one of confession and avoidance. The issue of breach does not arise because the conduct complained of was beyond the scope of any such duty. 71. In the Derivative Action, the Respondent alleged that Ricky breached his fiduciary duties to Smart Wave by establishing, managing and operating the Itamae and Itacho restaurants. The allegation of breach in each case was pleaded in the following terms: that Ricky “…failed to act in the best interests of Smart Wave and acted solely for his personal benefit and the benefit of his companies.” (see paragraph 19, 28, 35). 72. The duty of a director to act in the best interests of the company is a statement of the positive duty of loyalty which is broader than, but encompasses, the conflict rule. See Bristol and West Building Society v Mothew [1998] Ch 1 at 18 per Millett LJ (as he then was). Perhaps it was the reference in the pleading to acting for his actual personal interests by way of contrast with the company’s interests, in the interests of the company, that led to the assumption in these proceedings, at all levels, that the pleading was a statement of the conflict rule. 73. Notwithstanding the fact that, in business opportunity cases, the conflict rule generally overlaps with the profit rule, in this case the Respondent relied only on the former. That was the rule applied by the trial judge and by the Court of Appeal. It was not suggested in this Court that we should do otherwise. Indeed, the fact that the trial judge limited recovery until the date in 2010, when Smart Wave ceased to operate its restaurant – upheld by the Court of Appeal and from which there is no appeal – is more consistent with a conflict rule finding, than it is with a profit rule finding. 74. The conflict rule is generally stated in the form that a fiduciary may not put himself or herself in a position where his or her interest and duty conflict (see Stuart Bridge et al Snell’s Equity Thirty Third Edition Sweet & Maxwell, Thomson Reuters, London, 2015 para7-018, relying on Bray v Ford [1896] AC 44 at 51 and Aberdeen Railway Co v Blaikie Bros (1854) 1 Macq 461 at 471; 23 LT 315 at 316 at 5. However, it is well established that there must be a “real sensible possibility of conflict” (Boardman v Phipps [1967] 2 AC 46 at 124 per Lord Upjohn). 75. The Appellant’s principal submission was that Smart Wave was a single purpose company i.e. a single restaurant company. Accordingly the scope of the fiduciary duty of Ricky, as the sole director of Smart Wave, did not extend to Ricky’s conduct in establishing and operating other sushi restaurants. By reason of the limited scope, there was no position of conflict. 76. The Respondent submitted that the Appellant was subject to the fiduciary duties of a director. Accordingly, it was for him to establish that the scope of those duties was restricted in the manner for which he contended. See Bishopsgate Investment v Maxwell (No. 2) [1994] All ER 261 at 265 per Hoffmann LJ, as he then was. He had not done so. 77. The boundary of fiduciary duty is variously expressed in the authorities “the company has no concern” or “no interest” in the impugned conduct (Bell v Lever Brothers Ltd [1932] AC 161 at p 194); “the fiduciary has no duties to perform in respect thereof” Boardman v Phipps [1967] 2 AC 46 at 130; “wholly without the scope of the firm’s business” Aas v Benham [1891] 2 Ch 244 at 256 per Lindley LJ; the company was “not in the business” Canberra Residential Developments Pty Limited v Brendas [2010] FCAFC 125, (2010) 80 ACSR 270 at [38]. 78. In a frequently cited passage in Birtchnell v Equity Trustees Executors and Agency Co Ltd (1929) 42 CLR 384 at 408, Dixon J said: “The subject matter over which the fiduciary obligations extend is determined by the character of the venture or undertaking for which the partnership exists.” 79. As Lord Wilberforce pointed out in New Zealand Netherlands Society “Oranje” Inc v Kuys [1973] 1 WLR 1126 at 1130, the principle stated by Dixon J is not limited to partnership, but is of general application. 80. To the same effect are the observations of Sir Anthony Mason in Hospital Products Ltd v United States Surgical Corporation (1984) 156 CLR 41 at 102, relying inter alia on Kuys: “[T]he scope of the fiduciary duty must be moulded according to the nature of the relationship and the facts of the case” 81. The Appellant adopted the terminology of Mason J in Hospital Products supra at 97, to submit that the contractual context, indeed the “contractual foundation”, of the fiduciary duty in this case was the “limited nature” of Smart Wave (Appellants’ Written Case at [48], [49] and [52]) namely the sole purpose or single restaurant restriction. The Court was referred to a line of authorities in which the scope of fiduciary duties was found to arise from a contractual relationship. 82. These authorities are of little assistance in this case. The fact that a director has fiduciary duties is a well established category, with well established incidents, arising from the nature of the office. The fiduciary relationship between a director and the company does not have a “foundation” in contract. 83. The Respondent relied on the reasoning of the English Court of Appeal in Re Allied Business and Financial Consultants Ltd: O’Donnell v Shanahan [2009] EWCA Civ 751; [2009] 2 BCLC 666 at [55]–[72] for the proposition that, in a business opportunity case, it was no answer that the opportunity fell outside the company’s business. [Respondent’s Written Case para 39(7)]. That Court expressly applied its analysis to both the profit rule and the conflict rule at [55], [67], [73]. 84. The decision distinguished the case of directors from that of partnership. It did not apply the reasoning in Aas v Benham supra on the basis that, unlike partners “director’s fiduciary duties are not circumscribed by the terms of a contract”,at [68] and [69]. Accordingly there was no room for a “scope of business inquiry” in the case of directors at [70]. 85. In my opinion, the reasoning in O’Donnell v Shanahan may go too far. In particular the attempt to distinguish Aas v Benhamis not compelling. Indeed it would be surprising if a different rule applied to directors and partners – the latter being an equally well established category of fiduciary relationship. 86. I do not understand Lindley LJ in Aas v Benham to have determined the scope of business in that case on the basis of the contract of partnership. He determined it on the basis of the facts of the case, namely what the two companies actually did. 87. In my opinion, a “scope of business test” may be applicable to a company director. The facts and circumstances of a particular case may be such as to modify the subject matter to which the fiduciary duties of a director apply. However, such modification must be binding in the corporate context. Such modification does not need to be formal – as in a provision in the constitutive documents or a shareholders’ resolution – as long as it is, in substance, equivalent to a formal modification (e.g. In ReDuomatic Ltd[1962] 2 Ch 365). 88. The submissions in this court proceeded on the basis that the answer to the scope of duty issue was to be determined on fiduciary principles. Relevantly, the beneficiaries, in this case all the shareholders, had to be unanimous and to give a fully informed consent. 89. It is necessary to commence, in the case of an established category, with the usual scope and incidents of a well recognised fiduciary relationship. In this case, that means that the Appellant has the onus of establishing that the single purpose restraint was, or should be regarded as, binding on Smart Wave, on whose behalf the Derivative Action is brought. 90. The Appellant's case was that the agreement amongst the promoters to establish the business on the same model as the Ajisen business, together with the concurrence of the minority shareholders in Smart Wave, was effective to constitute a restriction on the activities of Smart Wave. That was a determination, and therefore a limitation, of the scope of the duty owed by the sole director, Ricky. 91. In my opinion, it is self evident that, unless the scope of duty was limited in the way for which the Appellant contends, Ricky was in breach of the conflict rule. Although the trial judge expressed this conclusion in term of “competition” by the Itacho restaurants with the single restaurant conducted by Smart Wave (at para 65), the scope of the rule is not limited to competition. As the Court of Appeal stated, in addition to competition there is the possibility that a business opportunity was diverted from Smart Wave (at para 4.18). Furthermore, Smart Wave, as the first restaurant in what was to become a chain of restaurants, had an interest in the establishment and operation of the chain as it developed. The strength of the brand, and its popularity, would redound to its advantage. Mashonaland 92. The Appellant placed particular reliance on the reasoning in Bell v Lever Brothers Ltd [1932] AC 161, especially at 194-195, including the references by Lord Blanesburgh to the decision of Chitty J in London and Mashonaland Exploration Co Ltd v New Mashonaland [1891] WN 165. The statement of principle by his Lordship, in the passage relied on, that a director has no fiduciary duty in a matter in which “the company has no concern”, is an orthodox reflection of the scope of duty principle, discussed above. 93. However, the Appellant also relied on what Lord Blanesburgh called an “essential qualification” to the effect that it must appear that the fiduciary “has made use either of the property of the company or of some confidential information which has come to him as a director of the company”. The Appellant submitted there is no such finding here (Appellants’ Written Case at [56]). Further, the Appellant also relied on the proposition that there is no rule against being a director of competing companies (Appellants’ Written Case at [67]). 94. The formulation of an “essential qualification” by Lord Blanesburgh was a reference to the reasoning in Mashonaland. His Lordship’s reliance was, in my opinion, based on a misunderstanding of the earlier case. 95. Mashonaland is usually cited for the proposition that there is no rigid rule that a director cannot be involved in the business of a company in competition with the company of which he is a director. That proposition is either trite or is also based on a misconception of the case. 96. As Lord Blanesburgh made clear, he relied solely on the short report of Mashonaland in the Weekly Notes. All subsequent references to that case have been similarly confined. However, there is a fuller report of the case in The Times of 10 August 1891 at page 3. That report puts the reasoning of Chitty J in a different light. 97. The Earl of Mayo was appointed to both Boards for the purpose of his name appearing as a director in each prospectus. He never attended a Board Meeting or acted in the capacity of a director in any way. In that era, it was unnecessary and, no doubt, undignified, for Chitty J to point out that the Earl never would. He was there only to enhance the reputability of the investment, by allowing his name to be used on both prospectuses. 98. The duty of diligence expected from contemporary directors – even a Lord – makes Mashonaland an anachronistic authority on which reliance cannot now be placed. Note the observations of Hoffmann LJ, as he then was, in Bishopsgate Investment Management Ltd (in liq) v Maxwell (No. 2) [1994] All ER 261 at 264 on how “older authorities” on directors duties have been superceded by the interaction of statutory provision and the natural development of the common law. See also Gower and Davies, Principles of Modern Company Law Ninth Edition Sweet & Maxwell, Thomson Reuters, London, 2012 at [16-167]. 99. The Times report contains reference to two authorities on which Chitty J relied, neither of which are mentioned in the Weekly Notes report. It is clear that when Chitty J described the application as “unprecedented” – an observation only found in The Times –he was referring to the circumstances in which an injunction would lie. He relied on the judgment of Lindley LJ in Whitwood Chemical Co vHardman[1891] 2 Ch 416, where the Court refused an injunction to prevent an executive from working for a rival, there being no evidence he had actually done so. In the absence of a negative stipulation in the contract, no injunction was granted. 100. As Chitty J noted, there was no suggestion in that case that the executive was about to disclose confidential information, when an injunction could have been given. He added that, in the case before him, Lord Mayo had acquired no confidential information. The Times report makes it quite clear that Chitty J was not purporting to set out a comprehensive statement of breaches by a director who become involved with a competitor company. 101. Before concluding that “The company had the most appropriate remedy in its own hands”, namely removing the director, Chitty J had said: “if a director acted contrary to his duty the company could call upon him to resign, or could remove him. If his conduct was of a grosser kind, such as betraying secrets, they could get an injunction. In other cases of that class a remedy by damages seemed more appropriate. The extraordinary remedy by injunction – although as to this he must be considered as speaking with caution - would be rarely appropriate.” The word “such as”, before “betraying secrets”, makes it clear that the reference to confidential information was merely illustrative. 102. I set these passages out at some length because the case has been treated as a binding authority, in view of its approval by the House of Lords, based on an inadequate report. This has been so despite frequent criticism (see e.g. In Plus Group Ltd v Pyke [2002] EWCA Civ 370; [2002] 2 BCLC 201 at [72], [75], [79], [81]–[89]; British Midland Tool Ltd v Midland International Tooling Ltd [2003] EWHC 466 Ch, [2003] 2 BCLC 523 at [81]-[85]; Foster Bryant Surveying Ltd v Bryant[2007] EWCA Civ 200; [2007] 2 BCLC 239 at [70]; Eastland Technology Australia Pty Ltd v Whisson [2005] WASCA 144; (2005) 223 ALR 123 at [69]; Links Golf Tasmania Pty Ltd v Sattler [2012] FCA 634, (2012) 292 ALR 382 at [545]–[547], [555]–[556], [559]–[563]. 103. Perhaps the most thorough critical analysis of Mashonaland and its adoption in Bell v Lever Bros, is by the Scottish Court of Session, first by the Outer House in Commonwealth Oil and Gas Company Ltd v Baxter [2007] CSOH 198 at [175]–[176] and then by the Inner House in Commonwealth Oil and Gas Company v Baxter [2009] CSIH 75; [2010] SC 156 at [4]-[5], [75]-[78]. 104. In the light of this considerable body of criticism, Mashonaland and its adoption in Bell v Lever Bros, should be regarded as standing for no wider a proposition than the trite statement that the law will not interfere in the absence of evidence of a real possibility of breach of fiduciary duty, including of the conflict rule. 105. As Upjohn LJ said, the conflict rule: “…must be applied with common sense … and applied realistically to a state of affairs which discloses a real conflict of duty and interest and not to some theoretical or rhetorical conflict.” [Boulting v Association of Cinematograph Television and Allied Technicians [1963] 2 QB 606 at 637-638] 106. The Appellant’s reliance on the comments in Mashonaland about breach of confidence (Appellants’ Written Case [56]), should be rejected. The Appellant’s reliance on the proposition that there is no rigid rule against being a director of competitive companies (Appellants’ Written Case [67]) does not advance its case. The Factual Issues 107. The parties made conflicting submissions on two, inter-related, factual issues. First, did the shareholders of Smart Wave agree that the company would have no interest in any other sushi restaurants? Secondly, was the intention that Smart Wave would only operate one restaurant, inextricably interrelated with an understanding that the original Ajisen shareholders (until the departure of Shigemitsu) would become shareholders in each company established to operate a sushi restaurant? It is convenient to refer to these investors as “the promoters”, to distinguish them from the smaller shareholders in Smart Wave. 108. The Appellant contends that the answer to the first issue is “Yes”. On the whole of the evidence, the Court should infer such an agreement. The Respondent contends that the answer is “No”, particularly in the absence of any positive evidence to that effect from Ricky himself, let alone from the minority shareholders. 109. The Appellant contends, with respect to the second factual issue, that the Court of Appeal erred in failing to separate the single purpose restaurant understanding from the promoters’ arrangement for future shareholdings. The Respondent contends that the two matters cannot be separated. The Single Purpose Evidence 110. As noted above, the modus operandi of the Ajisen Group was that each restaurant was conducted by a separate corporate vehicle. This was agreed by the promoters to be the model for the chain of sushi restaurants. Jason said in his witness statement: “Eventually, the shareholders of the Hong Kong Ajisen Ramen Group, including Mr Shigemitsu, Daisy, Ricky Cheng and I, all agreed to develop chain sushi restaurants in furtherance of the then existing business. We further agreed that separate corporate vehicles would be formed to hold the interest of the said chain restaurant business to be established by us. We as shareholders of Hong Kong Ajisen Ramen Group would then be allotted shares of and in the said corporate vehicles (‘the 2004 Agreement’).” 111. He also said: “Among us, Ricky Cheng was tasked to coordinate the establishment and operation of the chain sushi restaurant and to look for a suitable venue for the restaurant, negotiating the lease with the prospective landlord and applying for the necessary licenses for the opening of the new sushi restaurant. … It has always been the understanding of the shareholders of the Hong Kong Ajisen Ramen Group including Mr Shigemitsu, Daisy, Ricky Cheng and I, that the said 1st Itamae Sushi Restaurant was established in pursuance of the 2004 Agreement.” 112. Furthermore, he said: “… Under the 2004 Agreement, the shareholders of the Hong Kong Ajisen Ramen Group agreed that they would each be allotted shares in the corporate vehicles formed to hold the interest of the chain sushi restaurant business. This was never confined to the 1st Itamae Sushi Restaurant, and would equally apply to the subsequent Itamae Sushi Restaurants established as a chain. Further, as mentioned above, under the 2004 Agreement the shareholdings in the said corporate vehicles should follow those of the Hong Kong Ajisen Ramen Group as set out in paragraph 5 above.” 113. In the materials provided to this Court, Daisy does not state, in terms, that the separate corporate vehicle modus operandi was agreed. She did however give evidence of discussions with Ricky about share allocations in the first two companies established after Smart Wave. Similarly, Jason gave evidence that he discussed with Ricky the allocation of shares in those companies to him. On their own evidence, it is clear that both Ricky and Daisy expected that each sushi restaurant would be operated within a separate corporate vehicle, but that they would be shareholders in each such vehicle. 114. Of course, Smart Wave itself reflects both aspects of the agreement. However, unless at least the majority of shareholders of Smart Wave - as the promoters are - can be said to be bound by the agreement, there is no proper basis on which Smart Wave itself can be found to be so bound. 115. Daisy and Jason expected, indeed agreed, with Ricky that Smart Wave would be the first of a number of corporate vehicles, each operating one restaurant. However, that expectation or agreement was expressly interconnected with an expectation and agreement that they would be substantial shareholders in each such vehicle. These two elements were so closely interconnected, that it cannot be concluded that they had agreed to the restriction on Smart Wave on its own. In a case involving the equitable jurisdiction of the Court, the Appellant cannot rely only on that part of the promoters’ agreement that suits his case. The Minority Shareholders Evidence 116. I turn to consider the Appellant’s contention that the minority shareholders agreed to, or at least acquiesced in, the sole purpose restriction. 117. Evidence about the background of the minority shareholders was set out in Jason’s witness statement: “As far as I know, the parties who were not the shareholders of the Hong Kong Ajisen Ramen Group either worked under Ricky Cheng or were his business partners or suppliers. For example, Mak Kin Shing and Wong Yui To were at the time operation managers of the Hong Kong Ajisen Ramen Group and were assigned to also manage the first Itamae Sushi Restaurant when it was opened. They later followed Ricky Cheng to the Itamae Sushi Group after Ricky Cheng’s departure from the Hong Kong Ajisen Ramen Group in August 2006. Kong Yiu Wai used to be the kitchenware supplier of the Hong Kong Ajisen Ramen Group. He was a good friend of Ricky Cheng and was also his business partner in a Chinese restaurant named ‘御品軒’ (now closed). Sato Akira was a friend of Ricky Cheng who was his business partner in a Japanese-style restaurant named ‘Sesson雪村爐端燒’. Teraguchi Tadayoshi was a supplier for the Itamae Sushi business.” 118. Jason also said at para 76 of his witness statement that “most of the current shareholders of Smart Wave are either nominees or business associates of Ricky Cheng”. Ricky generally confirmed this in his evidence at trial (see Part B Tab 11). 119. Para 15(g) of the Appellant’s Defence (in the derivative action) asserted that Ricky had informed all the shareholders in Smart Wave that he, Ricky, “is and remains at liberty to establish, engage in and operate further or other sushi business or sushi restaurants, solely or in association with other parties or person.” (see also Defence in the Hero Elegant Action, at para 9e). This proposition is repeated at para 20. Both are set out at para 68 above. In cross examination, however, he accepted that he did not tell any of the shareholders that. (Part B Tabs 13 and 18). 120. Para 15(f) of the Defence (in the derivative action) asserted, in the context of the chapeau referring to the “Agreement/Understanding” of the Smart Wave shareholders. “[T]he parties had only discussed that Smart Wave would be permitted and authorised to operate one sushi restaurant using the Service Marks … and Ricky Cheng did not permit, authorise nor license Smart Wave to establish, engage in or operate any other restaurants or sushi restaurants using the Service Marks.” 121. On the materials before this Court, there is no evidence in support of any such “discussion” amongst the shareholders of Smart Wave. Nevertheless, the trial judge held at [62], set out at para 54 above, that “the shareholders of Smart Wave knew, intended and agreed that other companies would be set up to operate other sushi restaurants, using the Itamae name and Service Marks.” 122. Other than a reference, in the previous paragraph, to the fact that the shareholders of the Ajisen Group had an intention to operate in that way, the trial judge referred to no evidence in support of that conclusion. The Court of Appeal correctly pointed out that the Ajisen shareholders and the Smart Wave shareholders were not the same. On the materials and submissions in this Court, there is no direct evidence of any such acquiescence, let alone agreement, by the minority shareholders in Smart Wave. 123. Counsel for the Appellant asked us to infer such from the whole of the evidence. She drew attention to the close relationship, both business and personal, between Ricky and those shareholders. She also relied on the absence of any complaint by them about the opening of competing restaurants, of which they must have been aware. She also relied on their failure to join the derivative action. 124. In my opinion, the fact that these shareholders had such a close relationship with Ricky, is a basis for drawing the opposite conclusion. First, the Appellant did not call them in support of the alleged “agreement” or “understanding”. That is a foundation for the usual inference that nothing they could say would support his case. Secondly, the more probable inference is that they were reluctant to jeopardise that relationship by complaining, let alone joining in hostile legal proceedings. As to the absence of complaint, none of them, after the departure of Shigemitsu, had been Ajisen shareholders. Accordingly, they did not have the incentive of a promise of a shareholding in future corporate vehicles. As to the legal proceedings, they could sit back and take the benefit of the proceedings, without risk of any kind. 125. For the above reasons, there is no evidence on which it can be concluded that the minority shareholders had acquiesced in or agreed to any such restriction. They may not have had any basis for an expectation that Smart Wave would operate other restaurants. However, that is not the same as acquiescing or agreeing to such a restriction. The failure to call them, indeed the failure of Ricky to give direct evidence of their alleged concurrence, leads to the conclusion that no inference of their concurrence is open. The Service Marks 126. The Appellant’s submissions made frequent reference to the fact that Ricky owned the Service Marks and had only licensed Smart Wave to operate a single restaurant. It appeared that counsel was relying on that fact as a distinct basis for the single restaurant restriction, irrespective of any agreement of the shareholders. 127. In these proceedings, there was no issue as to whether Ricky was using his position as a director of Smart Wave, or the skill and knowledge acquired in that capacity, in the course of establishing and operating the Itamae and Itacho sushi restaurants. The only such matter raised in the evidence and submissions was the proposition that Smart Wave could have no interest in those restaurants by reason of Ricky’s ownership of the Service Marks. 128. In the derivative action, the Court is not concerned with any contractual obligation which Ricky may have had to license the Service Marks. The issue before us, in this respect, is whether he was entitled to establish and operate the other restaurants at all, even though Smart Wave, absent any contractual license, could not itself have done so. 129. It is well established that a fiduciary is liable to account, even if the beneficiary was not otherwise able to take advantage of the opportunity. The foundational authority for this proposition is Keech v Sandford (1726) Sel Cas Ch 61, 25 ER 223. The most frequently cited application of the principle to directors is Regal (Hastings) Ltd v Gulliver[1967] 2 AC 134 especially at 143, 144, 145, 149, 153 and 154. 130. On the basis of this line of authority, the Appellant’s submission should be rejected. No doubt in any account for profits, the value of any license can be part of an assessment of just allowances. Question 2 131. I have set out at the second question at para 64 above. The Appellant submits that, if it should fail on Question 1, the Court should, in the exercise of its discretion, refuse relief. He relies in particular on the decision of the English Court of Appeal in Nurcombe v Nurcombe [1985] 1 WLR 370, where Lawton LJ said at 377: “…[T]he court is entitled to look at the conduct of a plaintiff in a minority shareholder’s action in order to satisfy itself that he is a proper person to bring the action on behalf of the company and that the company itself will benefit. A particular plaintiff may not be a proper person because his conduct is tainted in some way which under the rules of equity may bar relief. He may not have come with’ clean hands or he may have been guilty of delay”. 132. The Respondent contends that the reference to “clean hands” in this quotation contravenes the rejection, by the Appeal Committee of two grounds of appeal involving the terminology of clean hands. However, the sole focus of the actual submissions of the Appellant was on the question of inconsistency. There is no reliance on clean hands. The fact that there is such a reference in an authority relied on, is in no way inconsistent with the decision to reject leave in certain respects. 133. The Respondent also contends that various references to the Respondents receiving a “benefit” from the allegedly disentitling conduct is also inconsistent with the rejection of those two grounds of appeal. However, the terminology of ‘benefit’ appears in Question 2. 134. The Respondent makes a compelling case that the Appellant is not entitled to raise this discretionary matter for the first time in this Court. Discretionary refusal of relief must be fact based. As the Respondent contends, no such discretionary ground for refusing relief was pleaded and, accordingly, evidence was not directed to it. Furthermore, neither the trial judge nor the Court of Appeal made, or indeed were asked to make, any relevant findings of fact or law. 135. What we are left with is an abstract question, whether, without the broader context usually relevant to the grant of equitable relief, inconsistent conduct or, the failure of an alleged obligation to call “contrary evidence”, is sufficient to deny the equitable relief which the Respondent seeks on behalf of Smart Wave. In my opinion these matters should not be permitted to be agitated for the first time in this Court. 136. At the trial, the parties did not address themselves to this unpleaded case. We do not have the benefit of any findings of fact. This Court should not entertain an argument that relief should be denied on equitable grounds without taking the full range of relevant considerations into account, particularly in a case where both sides may be said to have acted in an inappropriate manner. This is not to say that the conduct of Jason and Ricky must be balanced. However, Ricky’s conduct is relevant to the denial of equitable relief to the company in the derivative action brought by Jason. 137. As the Privy Council said in Sang Lee Investment Co Ltd v Wing Kwai Investment Co Ltd [1983] HKLR 197 at 209: “…[T]heir Lordships do not accept that in a case of this sort, where there are alleged improprieties on each side, the proper approach of the court in exercising its discretion is to compare the misconduct on the one side with the misconduct on the other side. The court should first decide whether there has been any relevant want of faith, honesty or righteous dealing on the part of the person seeking relief, and the court should then decide whether, as a matter of discretion and in all the circumstances, which may include any relevant misconduct on the part of the person resisting equitable relief, it is right to grant or refuse specific performance. There is no balancing exercise which falls to be performed.” 138. It is not appropriate to speculate on the range of evidence that the Respondent may have wished to adduce if this issue had been pleaded, or even raised, at first instance. Three examples will suffice. 139. In his submission to this Court, the Appellant places emphasis on the alleged failure of the Respondent to call the minority shareholders (Appellants’ Written Case at [26], [27], [28], [91], [92] and especially [93].) This argument is picked up in the last clause of Question 2. It is by no means clear that the Respondent had any onus to call the minority shareholders on this discretionary issue. However, wherever the onus lies, the fact that this issue was not raised at first instance meant that neither side adduced evidence from those shareholders relevant to this issue. 140. The second example arises from the fact the principal act of inconsistency alleged by the Appellant is the entry into the Hero Elegant Agreement. This, of course, applies only to the Itamae restaurants, not to the Itacho restaurants. There is no appeal from her Ladyship’s finding at [66] that the proposition that the Hero Elegant Agreement somehow extended to the Itacho restaurants was never pleaded. No evidence, it appears, was adduced about Ricky’s conduct with respect to the Itacho restaurants, which could provide the context for the conduct of the Respondent said to disentitle him from relief in that respect. 141. Thirdly, it is apparent that the Hero Elegant Agreement did not stand alone. The parties also compromised Ricky’s claims with respect to the Ajisen restaurants in the Favor Will Agreement. Any appraisal of the Respondent’s conduct in entering the Hero Elegant Agreement, and any related conduct by Ricky, must be assessed in the full context. However, there was no inquiry into the Ajisen dispute leading to the Favor Will Agreement. 142. Finally, it is pertinent to point out that, Ricky also entered into the Hero Elegant Agreement. On the findings above, he was in breach of his fiduciary duty to Smart Wave. He is in an identical position as Jason in this respect. Although this is not a balancing exercise, this fact would be entitled to weight when assessing the conduct of Jason, for purposes of exercising the discretion to deny the company relief. 143. This Court should not entertain Question 2. The Appeal should be dismissed without the exercise of the discretion to refuse relief which the Appellant seeks. Mr Justice Ribeiro PJ: 144. By a majority, Mr Justice Tang PJ and Mr Justice Bokhary NPJ dissenting, the appeal is dismissed and an order for costs nisi is made in favour of the respondent with liberty to the parties, if so advised, to lodge submissions as to costs in writing within 14 days from the handing down of this judgment. (Kemal Bokhary)				Non-Permanent Judge (James Spigelman)				Non-Permanent Judge Ms Audrey Eu SC, Mr. Jeremy S.K. Chan and Mr Anson Wong Yu Yat, instructed by Tang, Lai & Leung, for the 1st defendant (1st appellant) and 3rd to 31st defendants (2nd to 30th appellants) Mr Jat Sew-tong SC, Ms Linda Chan SC, instructed by T. H. Koo & Associates, for the plaintiff (1st respondent) The 2nd defendant (2nd respondent) was not represented and did not appear [1] [1969] 2 Ch 365. [2] Another complaint relates to Ricky operating a chain of sushi restaurant called Itacho. See para 13 below. [3] See para 10 below. [4] Taken from his witness statement. [5] There is no material difference between Daisy and Jason’s evidence. [6] Para 10 of Jason’s witness statement. Unless otherwise stated, all paragraph references are to his statement. [7] It was raised to 24% from an initial proposed 23% after Daisy complained, so that Daisy should be the largest shareholder. Para 18. [8] Katsuaki’s shares have been transferred to Ricky. [9] They were Kong Yiu Wai (“Kong”) 10%, Mak Kin Shing (“Mak”) 8%, Sato Akira (“Sato”) 4%, Teraguchi Tadayoshi (“Teraguchi”) 4%, Wong Yui To (“Wong”) 2%. [10] Para 18, two of the minority shareholders became the managers of the first Itamae restaurant. [11] Para 26. [12] It is unimportant detail but there were Ajisen Ramen restaurants in the mainland which were part of the listed company. [13] Para 27. [14] Para 29. [15] Para 30. [16] Para 31. [17] See for example, para 15 Defence. [18] Section D3, Statement of Claim. [19] Decision, 6 December 2013, at para 12. [20] And other minor shareholders who could facilitate the operation of these Itamae restaurants. [21] Indeed the other Ajisen shareholders too. [22] [1969] 2 Ch 365. [23] The headnotes at 366. [24] Para 63 CFI judgment. I should add that Smart Wave never claimed that it had the exclusive right to the use of the service marks. [25] Paras 4.16 and 4.21, CA Judgment. [26] Para 4.6, CA Judgment. [27] Which can be. [28] Para 67. [29] This was used in the context where Smart Wave was referred to as the “first Itamae”. [30] See also para 25 of the CFI judgment. [31] See, for example, para 6 above. [32] Where he claimed that he had reached an agreement or understanding with all the shareholders of Smart Wave naming them individually). See para 68 below, where Mr Justice Spigelman reproduced the entire paragraph. [33] He and Daisy seemed to have accepted that there would be some minority shareholders when they accepted the percentage of shares allotted to them in Smart Wave. Similar percentage is reflected in the Hero Elegant Agreement. [34] Strictly speaking in the corporate vehicles holding those restaurants. [35] See para 15(f). [36] Mak and Wong became the managers of the first Itamae restaurant and it seems clear that managers or staff of other Itamae restaurants might be given shares in other Itamae restaurants. [37] 10%. [38] Collectively holding 14% of the Smart Waves’ shares. [39] Together with Ricky’s 23% and Katsuaki’s 15% transferred to Ricky. [40] Para 61, CFI Judgment. [41] Para 31, CFI Judgment. 
15. As mentioned earlier, a number of grounds of appeal were argued before the Court of Appeal but we are concerned only with what were argued as Grounds 3 and 4 in the application for leave to appeal. In relation to these grounds, it was contended that it was wrong for the judge to have directed on an all or nothing basis, there being evidence to suggest to the jury that at least the drugs in the black bag belonged not to him, but to Wai Wai. Accordingly, the judge ought to have required the prosecution to amend the indictment by splitting it so as to raise a different count in relation to the drugs found in the sling bag. This then would have enabled the judge to direct the jury to return a verdict, if they thought fit, confined to the sling bag. It was also argued that a separate charge ought to have been directed in relation to the ketamine found in the black bag. This was said to be such a small amount that a charge of simple possession ought to have been left to the jury to decide. 16. Separate judgments were given by the Court of Appeal. Macrae VP was of the overall view that the Appellant had received a fair direction. It was no part of a judge’s duty, he said, to “canvass all permutations of fact in his summing‑up, including those not even put forward by the defendant in evidence.”[12] The Vice President noted the defence case being that the Appellant was not in possession of any of the drugs. As he put it:- “5. The applicant elected to give evidence and it is significant to note that it was never his case that he was in possession of only the 1st batch of drugs but not the 2nd batch of drugs. His case was that he never had possession of either. Neither defence counsel nor prosecuting counsel sought to distinguish between his possession of the two batches of drugs during examination‑in‑chief or cross-examination of the applicant. The approach of the prosecution was that the applicant was in possession of all the drugs: the defence account was that he was in possession of none of them.” Later in his judgment, Macrae VP emphasised the need always to remember that defence counsel had the conduct of the defence at trial.[13] 17. As for the ketamine part, Macrae VP again referred to the fact that the defence approach was that the Appellant had nothing to do with the drugs, including the ketamine. He saw no reason to remove the ketamine part into a separate count when there was no evidence to suggest a case of simple possession, this forming of course no part of either the defence or the prosecution case.[14] 18. Pang JA who arrived at the same conclusion as the Vice President, emphasised the fact that the Appellant got exactly the summing‑up that he had sought from the judge and the direction was one that was “entirely favourable to the defence”.[15] 19. McWalters JA arrived at the opposite view. He placed reliance on the decision of this Court in Ho Hoi Shing v HKSAR.[16] The material part of his judgment is as follows:- “55. The relevant legal principles governing the duty of the judge in such a situation are set out in the judgment of Chan PJ in giving the judgment of the Court of Final Appeal in Ho Hoi Shing v HKSAR. Those principles make clear that even though a case may be fought on a particular basis, here the primary defence strategy, if the evidence throws up other possibilities, such as the secondary defence strategy, on which a reasonable jury might return an alternative verdict, then the judge is obliged to direct the jury on their right to return these alternative verdicts. This is so even if the alternative is inconsistent with the defence run by the defendant’s counsel. The threshold requirement is that “the evidence which may be relied on to support an alternative verdict must not be so incredible or tenuous or uncertain that no reasonable jury could have accepted it.” This threshold, the Court of Final Appeal said, is a low one.” Curiously, it is to be noted that Ho Hoi Shing was neither relied on nor addressed by either party in the appeal.[17] 20. Applying the principles as he understood them, McWalters JA was of the view that there was sufficient evidence for an alternative verdict in relation to the sling bag to have been placed before the jury. The indictment should accordingly have been amended. 21. The Appellant supports the view taken by McWalters JA and it is to his submissions before us I now turn. D. THE PARTIES’ SUBMISSIONS 22. Although the Appellant’s written Case dealt at some length with duplicity (both patent and latent), it became clear from leading counsel’s[18] submissions that the Appellant’s principal proposition was this:- that the judge in a criminal trial is under a duty to place before the jury “all possible alternatives which are open to it on the evidence, that is where there is evidence upon which a reasonable jury might return a verdict on that alternative”. This duty is incumbent on a judge even if such alternatives are not raised by the parties or even if they are inconsistent with the defence run by the defence. Reliance was placed on Ho Hoi Shing at paras 12 to 13 where Chan PJ said:- “12. It is not disputed that the judge in a criminal trial is under a duty to place before the jury all possible alternatives which are open to them on the evidence. He is not absolved from this duty even if a particular alternative is not raised by the parties or is inconsistent with the defence run by counsel for the accused. As Lord Clyde in Alexander Von Starck v The Queen[19] put it at p.1275: It is his responsibility not only to see that the trial is conducted with all due regard to the principle of fairness, but to place before the jury all the possible conclusions which may be open to them on the evidence which has been presented in the trial whether or not they have all been canvassed by either of the parties in their submissions. It is the duty of the judge to secure that the overall interests of justice are served in the resolution of the matter and that the jury is enabled to reach a sound conclusion on the facts in light of a complete understanding of the law applicable to them. 13. The reason for imposing such a duty on the trial judge is obvious: it is in the public interest that an accused charged with an offence is properly dealt with according to the law. He should be convicted of, but only of, an offence which is proved to have been committed by him. As explained by Lord Bingham of Cornhill in R v Coutts[20] at p.2159 para.12: The public interest is that, following a fairly conducted trial, defendants should be convicted of offences which they are proved to have committed and should not be convicted of offences which they are not proved to have committed. The interests of justice are not served if a defendant who has committed a lesser offence is either convicted of a greater offence, exposing him to greater punishment than his crime deserves, or acquitted altogether, enabling him to escape the measure of punishment which his crime deserves. The objective must be that defendants are neither over-convicted nor under-convicted, nor acquitted when they have committed a lesser offence of the type charged … It is the ultimate responsibility of the trial judge …” 23. It was submitted that the meaning of “possible alternatives” was, according to the way the first of the certified questions was drafted, “all possible defence scenarios in trial (i.e. primary and secondary defence strategy).” This followed the way McWalters JA put it in those parts of his judgment set out above.[21] 24. By failing to require the prosecution to amend the indictment, the judge could not therefore direct the jury on the possible defence scenarios said to arise on the evidence. These included the Appellant having been in possession and control of only the sling bag but not the black bag; and as far as the ketamine was concerned, this ought to have been made the subject of a separate charge of mere possession. 25. Accordingly, by reason of the jury having been deprived of the opportunity to consider verdicts based on the possible defence scenarios, the Appellant was deprived of a fair trial. This was notwithstanding that the judge summed up on the all or nothing basis that defence counsel had insisted on. There was, it was so submitted, a real danger of the jury having over‑convicted the Appellant. Reliance was placed on the following part of Chan PJ’s judgment in Ho Hoi Shing:- “23. It is also impossible to say that the jury's choice would not have been affected by the offers of alternative verdicts open to them. See Callinan J in Gilbert v R (2000) 201 CLR 414, p.441: It is contrary to human experience that in situations in which a choice of decisions may be made, what is chosen will be unaffected by the variety of the choices offered.” 26. The Respondent’s position was that the all or nothing direction to the jury was generous and entirely favourable to the Appellant. In any event, there was no need to provide the jury with alternatives because there was simply no evidential basis to support such. As for the scenario that the Appellant was in possession of the sling bag alone but not the black bag, this was never a part of the Appellant’s case nor did the evidence suggest such a scenario. In relation to the ketamine, there was again no evidence whatsoever to support the scenario that the Appellant possessed this for his own consumption. E. ANALYSIS OF THE LAW 27. There are a number of important connected features of a criminal trial that ought to be reiterated in the present case. First, most important of all, is the public interest in ensuring a fair trial and the primary responsibility for this rests on the trial judge. Fairness applies to both the defence and the prosecution. Secondly, the defence must be free to conduct its defence as it sees fit in the way the evidence is treated and the way the defence is presented to the court, provided this is done reasonably and properly. Thirdly, the prosecution must also similarly be free to conduct the prosecution as it sees fit. All this is of course subject to professional obligations and the overriding duties owed to the court and the administration of justice. 28. We are in this case particularly concerned with that aspect of fairness being the public interest in the fair outcome of a trial. This was set out in clear terms by Lord Bingham of Cornhill in R v Coutts.[22] I have already set out an extract contained in the judgment of Chan PJ in Ho Hoi Shing[23] but it is useful to set out the whole of that part of Lord Bingham’s speech:- “12. In any criminal prosecution for a serious offence there is an important public interest in the outcome: R v Fairbanks [1986] 1 WLR 1202, 1206. The public interest is that, following a fairly conducted trial, defendants should be convicted of offences which they are proved to have committed and should not be convicted of offences which they are not proved to have committed. The interests of justice are not served if a defendant who has committed a lesser offence is either convicted of a greater offence, exposing him to greater punishment than his crime deserves, or acquitted altogether, enabling him to escape the measure of punishment which his crime deserves. The objective must be that defendants are neither over-convicted nor under-convicted, nor acquitted when they have committed a lesser offence of the type charged. The human instrument relied on to achieve this objective in cases of serious crime is of course the jury. But to achieve it in some cases the jury must be alerted to the options open to it. This is not ultimately the responsibility of the prosecutor, important though his role as a minister of justice undoubtedly is. Nor is it the responsibility of defence counsel, whose proper professional concern is to serve what he and his client judge to be the best interests of the client. It is the ultimate responsibility of the trial judge: Von Starck v The Queen [2000] 1 WLR 1270, 1275; Hunter v The Queen [2003] UKPC 69, para 27.” 29. It is clear from this general statement of principle, much relied on by the Appellant in the present case, that Lord Bingham was referring to the context of alternative verdicts in relation to lesser offences. I will in due course return to this aspect. For the time being, one of the important questions that arises is to ask in what circumstances would it be incumbent on a judge (in a trial by jury) to direct the jury as to the alternative options open to it? 30. As set out above, it was the contention of the Appellant in the present case that all possible alternatives which are open on the evidence or all possible scenarios submitted by the defence in the trial should be placed before the jury, and in order to enable a jury to be so directed, a judge may have to order the indictment to be amended. This, it will be recalled, was essentially the view taken by McWalters JA, referring to “secondary” defence strategies. The only limitation, it was submitted, was that there must be some evidence in support of such possibilities that was not “incredible or tenuous or uncertain”.[24] 31. In my view, some care needs to be exercised when defining the obligation to direct alternative verdicts. It is unhelpful, not to say confusing, for a judge to have to direct a jury to alternative options based on vague expressions such as “possible alternatives”, “possible scenarios”, “alternative defence scenarios” or “secondary defence scenarios”. This was the type of expression used by McWalters JA in the passages set out above.[25] Apart from being imprecise and uncertain, it can give rise to considerable practical difficulties in application because it can lead to a real potential for confusion on the part of the jury. Where, for instance, the defence’s evidence and approach to the evidence is contrary to such alternative or possible or secondary factual scenarios, it would be confusing and wrong for a judge to have to direct a jury on those alternative factual scenarios. Were it otherwise, this would invite ingenious attempts to identify alternative scenarios, particularly after trial, in order to impugn a summing‑up, these allegations bearing little or no resemblance to what was the reality at trial. In the passage from the judgment of McWalters JA, he refers to the Appellant’s “primary defence” and “secondary defence”. This was, with respect, confusing. There was no alternative defence run by the Appellant: the only defence he ran was that he had nothing to do with drugs and that he was framed. The other scenarios referred to by defence counsel in his final submissions and by McWalters JA were not alternative defences at all but were mere speculations. 32. The answer to the question in what circumstances it would be incumbent on a judge to direct a jury as to the alternative options open to it is, I believe, to ask further whether there is an obvious alternative verdict which is supported by the evidence of that alternative. The requirement of an obvious alternative supported by evidence was discussed by Lord Bingham in the following passage in Coutts:- “23. The public interest in the administration of justice is, in my opinion, best served if in any trial on indictment the trial judge leaves to the jury, subject to any appropriate caution or warning, but irrespective of the wishes of trial counsel, any obvious alternative offence which there is evidence to support. I would not extend the rule to summary proceedings since, for all their potential importance to individuals, they do not engage the public interest to the same degree. I would also confine the rule to alternative verdicts obviously raised by the evidence: by that I refer to alternatives which should suggest themselves to the mind of any ordinarily knowledgeable and alert criminal judge, excluding alternatives which ingenious counsel may identify through diligent research after the trial. Application of this rule may in some cases benefit the defendant, protecting him against an excessive conviction. In other cases it may benefit the public, by providing for the conviction of a lawbreaker who deserves punishment. A defendant may, quite reasonably from his point of view, choose to roll the dice. But the interests of society should not depend on such a contingency.” (emphasis added) Lord Rodger of Earlsferry referred to a “viable” alternative verdict.[26] 33. The reference to alternative verdicts obviously raised by the evidence is instructive. In the main cases relied on by the Appellant in the present appeal, the alternative verdict (although not raised in the way the defence was run as a matter of law) was nevertheless a clear and obvious alternative on the evidence. Further, all these cases involved the alternative offence being lesser offences to the one charged:- (1) Coutts was a murder case in which it was held that manslaughter should have been left to the jury, even though defence counsel at trial had not asked for this course to be adopted. Though this course was inconsistent with the defence run at trial as a matter of law,[27] it arose from the evidence of the accused in that case. His evidence was to the effect that he had no intention to kill the deceased and that it was an accident. There was therefore evidence to suggest a reasonable possibility that the accused lacked the necessary intent to kill or cause grievous bodily harm.[28] (2) Ho Hoi Shing was similarly a case in which the obvious alternative manslaughter arose from the defendant’s own evidence to the effect that his mind was blank and that he did not intend to hurt the deceased. (3) Reference was made in Ho Hoi Shing (and in Coutts) to Von Starck,[29] which was again a case in which the alternative verdict of manslaughter arose from the evidence to the effect that the accused did not have the requisite intent for murder. Such evidence in support of a manslaughter alternative arose from the accused’s own case and evidence.[30] 34. What is the position when the alternative scenario relied on is contrary to the way the defence has conducted or presented its case on the facts? In the passage from Ho Hoi Shing quoted above,[31] Chan PJ referred to a judge not being absolved of the duty to direct on an alternative verdict “even if a particular alternative is not raised by the parties or is inconsistent with the defence run by counsel for the accused”. However, it is relevant to read the extract from the judgment of Lord Clyde in Von Starck which follows immediately after. This passage refers to conclusions open to the jury “on the evidence which has been presented in the trial whether or not they have all been canvassed by either of the parties in their submissions”. (emphasis added) 35. The way that the defence case is run on the facts is obviously relevant in determining whether there is sufficient evidence in support of an obvious alternative verdict. If a factual alternative does not arise in the way the defence has dealt with the facts and presented the case on the evidence (as opposed to the legal approach) this will in most cases be decisive. In R v Fairbanks,[32] another lesser offence case, the question was whether a jury ought to have been directed on the alternative lesser charge of careless driving, the main charge being dangerous driving causing death. The English Court of Appeal held that the lesser alternative charge ought to have been put to the jury given the evidence and the way the evidence was approached by the defence. Mustill LJ, who gave the judgment of the court, however, said this:-[33] “There appear to be three relevant authorities. The first two cases were Rex v. Vaughan (1908) 1 Cr.App.R. 25 and Rex v. Naylor (1910) 5 Cr.App.R. 19. In each case the appellant had been convicted of a serious offence of violence, and complained that the trial judge had not left to the jury the possibility of convicting on the lesser offence of unlawful wounding. The court dismissed each appeal. The reports are brief, but they demonstrate that the judge is not always obliged to leave all the alternative verdicts theoretically comprised in the charge of the offence, but need not (and indeed should not) do so unless the alternatives really arise on the issues as presented at the trial. … These cases bear out the conclusion, which we should in any event have reached, that the judge is obliged to leave the lesser alternative only if this is necessary in the interests of justice. Such interests will never be served in a situation where the lesser verdict simply does not arise on the way in which the case had been presented to the court: for example if the defence has never sought to deny that the full offence charged has been committed, but challenges that it was committed by the defendant. Again there may be instances where there was at one stage a question which would, if pursued, have left open the possibility of a lesser verdict, but which, in the light of the way the trial has developed, has simply ceased to be a live issue. In these and other situations it would only be harmful to confuse the jury by advising them of the possibility of a verdict which could make no sense.” 36. The quoted passage assists in determining whether an obvious alternative verdict ought to be placed before the jury in emphasising the need to look at the way the case on the facts has been run. It is to be noted that a part of this passage was referred to by Lord Bingham in Coutts.[34] The decision in Fairbanks was approved by the House of Lords in R v Maxwell,[35] and although Maxwell was said to be not “unproblematical” in Coutts,[36] the problem did not lie with the passage just quoted from Fairbanks. 37. The trial judge has a discretion in deciding on the most appropriate course to take when faced with the possibility of an alternative verdict. Orders under s 23 of the Criminal Procedure Ordinance[37] involve a judicial discretion. If such a situation arises or may reasonably arise, it will be good practice for the judge to elicit the views of the parties. Such views may not necessarily be decisive but they are certainly relevant. Counsel of course have the duty to assist the court in putting forward any relevant considerations, particularly those relating to a fair trial, for the judge to consider. 38. Lastly, before returning to the facts, I should mention this. The application of the approach in cases like Coutts, Ho Hoi Shing and Von Starck seems only to have been in cases involving alternative verdicts in relation to lesser offences. I have earlier alluded to this.[38] We have not been shown any authority where the approach has been applied in relation not to an alternative lesser offence, but to a situation (as in the present case) where it is said an indictment should be split into different counts. 39. It is conceivable that the general principle that an accused should not be over‑convicted or under‑convicted may in some exceptional cases require a judge to direct on possible alternative verdicts where it is said an indictment should be amended to be split into different counts because the evidence may be said to support alternative factual scenarios. Both parties have assumed this principle to apply in the present case, as did the Court of Appeal. While I do not consider that any such principle has been established on the authorities, I am content to examine the circumstances of the present case on the hypothesis that such a rule exists. I note, however, that there may be other considerations that could conceivably be relevant in this situation on which we have not heard argument.[39] 40. I now return to the facts. F. APPLICATION TO THE FACTS 41. I have already set out the essence of the Appellant’s submissions in this appeal. In brief, it was submitted that the judge ought to have amended the indictment to split it into two, if not three, separate charges involving trafficking in the drugs found in the sling bag, trafficking in the drugs found in the black bag and mere possession of the ketamine. 42. The ketamine aspect can be quickly disposed of. Even though the amount of ketamine was small, it was found together with the other drugs in the black bag on a table on which there were also found electronic scales. There was no evidence whatsoever to indicate that the ketamine was for the Appellant’s own consumption, nor was this a part of his case on the facts. This was clearly not an alternative scenario that was obvious. 43. In relation to the possible splitting up of the indictment into the drugs found in the sling bag and the black bag, it was not the Appellant’s factual case that he was in possession of one of the bags but not the other. Although defence counsel submitted that the drugs in the black bag belonged to Wai Wai, this did not involve the Appellant accepting he was in possession and control of the sling bag. His position on the facts was simple: he was not in possession of any of the drugs specified in the indictment. However, it was prosecuting counsel who first raised the possibility of this alternative scenario both before the judge and then in his final submissions.[40] Given that this was not the factual position adopted by the defence, it may have been somewhat odd for the prosecution to make reference to this. Pang JA described this as “enigmatic”.[41] 44. The scenario that the Appellant was in possession of only the drugs in the sling bag was not an obvious alternative. Be that as it may, the reality was that the jury heard submissions on the alternative scenario that the Appellant may have been in possession of only one of the bags containing drugs. The judge, however, dealt with this alternative in clear terms. He told counsel that it would be very dangerous to put alternative scenarios to the jury when the case had been conducted by the prosecution on an all or nothing basis.[42] More important, he alerted the jury in his summing‑up that although it was not a live issue that the Appellant was in possession of only one of the bags, nevertheless if the jury thought that, he directed them to acquit.[43] Thus, the jury were directed on the specific factual situation that the Appellant now focuses on and he was given what was undoubtedly a favourable direction in the present case. 45. The Appellant’s complaint, based on the cases discussed above, was that the judge gave an all or nothing direction to the jury and accordingly there was a danger, given this limited choice, that they may have over‑convicted. There are three reasons why this complaint should be rejected. First, this was the very direction sought by the defence. Secondly, if there was any potential injustice to the Appellant arising from such a direction, it was incumbent on both prosecuting and defence counsel to have alerted the judge to this. No such indication was given. Thirdly, the judge made reference to the specific factual scenario of the Appellant being in possession of the drugs in the sling bag only but not the black bag, directing the jury that in this contingency, they had to acquit. G. CONCLUSION 46. For the above reasons, the appeal should be dismissed. The two questions in respect of which leave to appeal was given are not sufficiently precise to be capable of being answered Yes or No. Much depends on the circumstances of any given case. Mr Justice Ribeiro PJ: 47. I agree with the judgment of Chief Justice Ma. Mr Justice Fok PJ: 48. I agree with the judgment of Chief Justice Ma. Mr Justice Cheung PJ: 49. I agree with the judgment of Chief Justice Ma. Lord Neuberger of Abbotsbury NPJ: 50. I agree with the judgment of Chief Justice Ma. Chief Justice Ma: 51. The appeal is accordingly dismissed. Mr Andrew Bruce SC and Mr Richard Donald, instructed by Cheung & Liu, assigned by the Director of Legal Aid, and Mr Adrian Lo, instructed by Cheung & Liu, on a pro bono basis, for the Appellant Mr David Leung SC, DDP, Ms Maggie Yang, DDPP and Mr Joe Hui, PP, of the Department of Justice, for the Respondent [1] By a jury before Deputy High Court Judge Lugar‑Mawson. [2] Cap 134. [3] Macrae VP, McWalters and Pang JJA. [4] A number of grounds of appeal were submitted, but we are only concerned with one aspect. [5] In a judgment dated 5 September 2018, McWalters JA dissenting (the CA judgment). [6] Under s 32(2) of the Hong Kong Court of Final Appeal Ordinance Cap 484. [7] Ma CJ, Fok and Cheung PJJ. [8] Not Mr David Leung SC, Ms Maggie Yang and Mr Joe Hui, who appeared before us in the appeal. [9] In fact, defence counsel in his closing submissions postulated two versions in this context: one was that he had just bought the drugs found in the sling bag from Wai Wai, the other that he had bought the drugs elsewhere and was taking them to Wai Wai when he was arrested. [10] It is not known what the judge had in mind but it was probably on the basis of some authorities to the effect that on a trafficking charge, the offence can be established even though only some of the drugs particularised in the indictment were proved to have been in the possession and control of the accused for that purpose. [11] In a judgment dated 11 January 2019. [12] CA judgment at para 19. [13] CA judgment at para 30. [14] CA judgment at para 21. [15] CA judgment at para 99. [16] (2008) 11 HKCFAR 354. [17] CA judgment at para 18. [18] Mr Andrew Bruce SC, with him Mr Richard Donald and Mr Adrian Lo (the written Case also included Mr Gerard McCoy SC). [19] [2000] 1 WLR 1270. [20] [2006] 1 WLR 2154. [21] See paras 11 and 19 above. [22] See fn 20 above. [23] Para 22 above. [24] Ho Hoi Shing at para 15. [25] See paras 11 and 19 above. [26] At paras 84, 85. [27] Meaning the defence submitted the accused should be guilty of murder or be acquitted completely, there being no middle course. [28] At para 36. [29] See fn 19 above. [30] Von Starck at 1273F-G. See also Coutts at paras 95 and 100. [31] Para 22 above. [32] [1986] 1 WLR 1202. [33] At 1205G to 1206B. [34] At para 16. [35] [1990] 1 WLR 401. [36] At para 19. [37] Cap 221, dealing with the amendment of indictments, separate trials and postponement of trials. [38] See para 29 above. [39] For example s 18 of the Criminal Procedure Ordinance dealing with the joinder of charges in the same indictment. [40] See paras 8 and 9 above. [41] CA judgment at para 97. [42] See para 12 above. [43] See para 13 above. 
Chief Justice Ma: 1. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Fok PJ: A. Introduction 3. The appellant, whose name has been anonymised in these proceedings as ZN, was brought to Hong Kong to work as a foreign domestic helper between 2007 and 2010. During that period, he was badly mistreated by his employer to an extent that constituted forced or compulsory labour within the meaning of Article 4(3) of the Hong Kong Bill of Rights.[1] After returning to Hong Kong in 2012, the appellant sought to report that treatment to various Government agencies but was treated in a way described to us by the respondents’ counsel as “disgraceful”. 4. As will be seen, arising from those basic facts, which are described in more detail below, this appeal broadly raises two important questions of law concerning BOR4:[2] (1) Does BOR4 include a prohibition against human trafficking and, if so, what is the scope of that prohibition? (2) Does BOR4 impose a positive duty on the Government of the Hong Kong Special Administrative Region (“HKSARG”) to maintain a specific offence criminalising the activities prohibited under that article? A.1 BOR4 5. Since it is central to this appeal, it is convenient to set out the terms of BOR4 at the outset of this judgment. BOR4 provides: “Article 4 No slavery or servitude (1) No one shall be held in slavery; slavery and the slave-trade in all their forms shall be prohibited. (2) No one shall be held in servitude. (3) (a) No one shall be required to perform forced or compulsory labour. (b) For the purpose of this paragraph the term ‘forced or compulsory labour’ shall not include – (i) any work or service normally required of a person who is under detention in consequence of a lawful order of a court, or of a person during conditional release from such detention; (ii) any service of a military character and, where conscientious objection is recognized, any national service required by law of conscientious objectors; (iii) any service exacted in cases of emergency or calamity threatening the life or well-being of the community; (iv) any work or service which forms part of normal civil obligations.” A.2 The phenomenon of human trafficking 6. Although the present case arises out of the facts concerning the appellant’s own case, the context of the case is wider. That wider context is the phenomenon of human trafficking, which is a worldwide problem and has also been recognised by the HKSARG as a Hong Kong problem. The accepted definition of human trafficking (which will be referred to in greater detail later in this judgment) is that set out in the Protocol to the United Nations Convention against Transnational Organized Crime, commonly referred to as the Palermo Protocol.[3] Human trafficking can rightly be called an evil scourge since it involves the movement of persons by deployment of certain underhand means for the purpose of exploiting them. The exploitation takes a number of serious forms including sexual exploitation, forced labour or services, slavery or practices similar to slavery, servitude or the removal of organs. 7. Although, as will be seen, the Palermo Protocol does not apply to Hong Kong, the problem of human trafficking nevertheless remains recognised as a substantive problem internationally as reflected by various measures taken in various parts of the world to combat it. In May 2005, the Council of Europe Convention on Action Against Trafficking in Human Beings (“the Anti-Trafficking Convention”) was adopted with one of its purposes being the prevention of human trafficking. In April 2011, the European Commission adopted a Directive on the trafficking in human beings.[4] In 2015, the United Nations adopted 17 Sustainable Development Goals, which included a target to end slavery, with a directive to “take immediate and effective measures to eradicate forced labour, end modern slavery and human trafficking and secure the prohibition and elimination of the worst forms of child labour”[5]. Also in 2015, the UK enacted the Modern Slavery Act 2015, which makes provision about slavery, servitude and forced or compulsory labour and about human trafficking. 8. With a total of approximately 370,000 foreign domestic helpers working in Hong Kong, there is potentially a sizeable number of persons at risk of exploitative practices that might constitute either forced or compulsory labour or human trafficking. That these matters are also a problem in Hong Kong, albeit the scale of that problem is a matter of contention between the parties, is reflected in the further evidence that was filed for the hearing of this case in the Court of Appeal. That evidence (discussed further below) refers to the fact that, in March 2018, the HKSARG established “a high-level, inter-bureaux/departmental Steering Committee chaired by the Chief Secretary for Administration to offer [a] strategic steer in respect of tackling [issues concerning human trafficking]” and that an “Action Plan to Tackle Trafficking in Persons and to Enhance Protection of Foreign Domestic Helpers in Hong Kong” was launched on 21 March 2018.[6] A.3 The Judicial Review 9. Following the treatment to which he was subjected by the Immigration Department, the Police and the Labour Department (described in Section A.6 below), the appellant commenced judicial review proceedings against the respondents seeking declaratory relief and damages for breach of his rights under BOR4. The proceedings were heard by Zervos J (as he then was) in January 2016 and the appellant gave oral evidence at the hearing which was subject to cross-examination.[7] The Judge noted that he was asked to determine whether the appellant was a victim of human trafficking for forced labour[8] and also whether the officers of the relevant authorities failed or neglected, in breach of BOR4, to take appropriate action in dealing with his case as one involving human trafficking or forced labour.[9] 10. Whilst acknowledging that he did not have evidence from the appellant’s employer or any other relevant witness concerning the events during 2007 and 2010, Zervos J made findings of fact based on the evidence presented to him. These findings, at [160] and [161] of the CFI Judgment, are set out in Section A.6 below. There was no appeal by the respondents against those findings of fact. 11. The central legal issue in the judicial review was the interpretation of BOR4.[10] The Judge held that implicit in each of the concepts prohibited by BOR4 is a prohibition against trafficking a person for each of slavery, servitude and forced or compulsory labour respectively.[11] He held that the prohibited concepts of slavery, servitude and forced or compulsory labour under BOR4 included cases where a person is trafficked for such purposes and that anyone involved in the transportation of a person for slavery, servitude or forced or compulsory labour is caught by BOR4.[12] 12. The Judge was satisfied that the evidence supported a case that the appellant was a victim of trafficking of a person for the purpose of forced labour.[13] 13. The Judge found that the current legal regime in Hong Kong does not address adequately or effectively the positive obligations under BOR4 to tackle forced labour and the trafficking of persons for forced labour.[14] He held that the critical flaw in the HKSARG’s obligations under BOR4 was the lack of a criminal offence and penalty that addresses the prohibited concept of forced or compulsory labour.[15] The Judge went on to hold that the HKSARG had positive obligations under BOR4 to enact measures to ensure the prohibition of forced or compulsory labour, including trafficking for that purpose.[16] He found that the HKSARG has not adequately fulfilled its positive obligations under BOR4 and that the appellant was denied his rights thereunder in not having his case recognised by the relevant authorities as one possibly involving human trafficking for forced labour.[17] 14. The Judge therefore granted a declaration in the following terms: “The application for judicial review be granted in so far as on the evidence presented, the evidence points to the Applicant having been a victim of human trafficking for forced labour, or forced labour, and that he, on various occasions from April 2012, approached officers of the relevant authorities and gave an account of his case that should have been sufficient to alert them, at least on some occasions, that this was a possible case of human trafficking for forced labour, or forced labour, and prompted them to take appropriate action; and that the Applicant was denied protection under Article 4 of the BOR, which in turn was due to the failure of the HKSARG to fulfil its obligations under Article 4 of the BOR”. He also directed that a further hearing be fixed for argument on the question of relief, including the question of costs and damages. A.4 The Court of Appeal 15. The respondents’ appeal against Zervos J’s judgment was allowed in part.[18] There were four issues on the appeal: (1) whether BOR4 covers human trafficking for forced labour; (2) whether the appellant was a victim of forced labour; (3) whether the HKSARG is in breach of its positive duties under BOR4 by not enacting a specific criminal offence to combat forced labour (or human trafficking for forced labour); and (4) whether the HKSARG breached its investigative duty under BOR4 in the case of the appellant.[19] 16. The main judgment was delivered by Cheung CJHC (as Cheung PJ then was), with concurring judgments from Lam VP and Poon JA. (1) On issue (1), the Court of Appeal held that BOR4 does not cover human trafficking (as a form of modern slavery) or human trafficking for forced labour.[20] (2) On issue (2), the Court held that a case of forced labour in contravention of BOR4(3) had clearly been established.[21] (3) On issue (3), the Court held that the case that the HKSARG had breached its positive obligations under BOR4 to provide practical and effective protection against forced labour by means of a specific criminal offence had not been made out and that Zervos J’s finding that the HKSARG had breached its positive duties under BOR4 must be disturbed.[22] (4) On issue (4), the Court upheld the Judge’s finding that the HKSARG had failed in its investigative duty under BOR4 in relation to the appellant’s complaints in the present case.[23] 17. By way of disposition of the appeal, the Court of Appeal allowed the appeal in part by amending the declaration made by Zervos J (set out at [14] above) to read: “The application for judicial review be granted on the basis that: (1) on the evidence presented, the evidence points to the applicant having been a victim of forced labour; (2) he, on various occasions from April 2012, approached officers of the relevant authorities and gave an account of his case that should have been sufficient to alert them, at least on some occasions, that this was a possible case of forced labour, and prompted them to take appropriate action which they failed to do; (3) the HKSARG was therefore in breach of its procedural obligation to investigate situations of potential forced labour under article 4 of the BOR; and (4) the applicant was denied protection under the same as a result”.[24] A.5 Leave to appeal to this Court 18. The Court of Appeal[25] granted the appellant leave to appeal to the Court of Final Appeal on two questions of law of great general or public importance, namely: “Does Article 4 BOR prohibit human trafficking: (a) for the purpose of exploitation; (b) for the purposes of slavery, servitude and forced labour; or (c) for the purpose of forced labour?” and “Does Article 4 BOR impose an absolute duty on the Hong Kong SARG to maintain an offence specifically criminalising (a) forced labour; or (b) human trafficking for the purposes of forced labour; or (c) human trafficking for the purposes of slavery, servitude and forced labour; or (d) human trafficking for the purpose of exploitation? If not, does Article 4 BOR impose (i) a contingent duty to maintain such an offence, and (ii) does the contingency arise (and is the duty triggered) when existing criminal law measures are ineffective, or (iii) is there a further requirement, such as the requirement, for example that maintaining such an offence is the only means of redressing the ineffectiveness?” A.6 The Judge’s findings of fact 19. The facts concerning the appellant’s employment and the exploitation to which he was subjected by his employer as well as the woefully ineffectual response of the respondents when he tried to report the mistreatment to the authorities were set out at [160] and [161] of the CFI Judgment. These tell a disturbing story and merit being set out in full. 20. The Judge recorded (at [160]): “160. On the evidence and materials before me, and for the purposes of these proceedings, I make the following relevant findings of fact: (1) The applicant is a Pakistani national who is a member of the Malik caste, which is considered in Pakistan, and accepted by the applicant, as being inferior to the Rana caste. (2) The employer and his family belong to the Rana caste. The employer comes from a prominent and well-connected family in Punjab, Pakistan, with extensive business interests in Pakistan, South Africa and Hong Kong. (3) The applicant worked for the employer and his family in Pakistan. Socio-economic and cultural norms led to the employer asserting considerable command and control over the applicant. (4) The employer and his family arranged for the applicant to work for them in Hong Kong. They sponsored his work permit and arranged his transportation to Hong Kong. The applicant had not previously travelled out of Pakistan. Because of his low education and low socio-economic status he was not familiar with the system and structures in Hong Kong. (5) The employer promised the applicant that he would have good working conditions and that he would receive a salary of $4,000 per month, although in the two contracts the monthly salary payment specified was respectively $3,400 plus $300 food allowance, and $3,580 plus $300 food allowance. I accept that documents may have been presented to him by the employer, but that he did not understand them and in some instances did not sign them. (6) The applicant was accompanied to Hong Kong in January 2007 by a member of the employer’s family who held the travel and identification documents of the applicant. These documents were kept by the employer while he was in Hong Kong. (7) While in Hong Kong the applicant was kept under the control of the employer and his family. The applicant was constrained and controlled both psychologically and economically by the employer. (8) All formal arrangements for the applicant’s employment and residence in Hong Kong were organised and arranged by the employer. The applicant had no knowledge of his rights and obligations or of those of his employer. He was placed in a state of dependency on the employer. (9) The applicant was employed as a foreign domestic helper. However, he was required to work in the employer’s trading company (which the applicant agreed to do) and reside at the office premises of the company. He slept on the carpeted floor of one of the offices at the premises. He was required to work long hours, seven days a week. He was given two meals a day and his movements were restricted to the office premises except for office errands. He was able to take breaks, but it is unclear how often these occurred and for how long they lasted. As the applicant resided at the office premises, he was under the direction and control of the employer; his movements were restricted; he had limited enjoyment of privacy; and was unable to live a normal life. (10) The applicant was regularly abused and beaten by the employer. Although he did not sustain serious injuries, he was nevertheless treated in a degrading and abusive manner. He worked under conditions of constant abuse, threats and beatings. (11) The employer and his family cajoled and deceived the applicant into taking up the employment in Hong Kong. The applicant was deceived about his working conditions (long hours of work and subjection to abuse and beatings) and payment of wages (unpaid for nearly 4 years). (12) The employer threatened the applicant that serious harm would result to him and his family if he left his employ, and claimed that he owed the employer a large sum of money for having been brought to Hong Kong. (13) The employer tricked the applicant into agreeing not to receive his monthly wage, and put off paying him the full remuneration due to him under the terms of his employment contracts. (14) At the end of 2010, the applicant asked the employer to give him an advance on his unpaid wages to assist his family in Pakistan. The employer deceived the applicant into returning to Pakistan, and then terminated his contract and sponsorship in order to avoid paying him the money that he owed him. This was also designed to prevent the applicant from returning to Hong Kong and claiming his unpaid wages against the employer. (15) The employer and members of his family and his associates, both in Hong Kong and in Pakistan, threatened the applicant and his family regarding the applicant’s pursuit of his claim for outstanding wages from the employer. (16) The employer, both directly and through his associates, made threats against the applicant in order to get him to withdraw or settle his unpaid salary claim. (17) During the period when the applicant worked for the employer, from May 2007 to December 2010, he made no report or complaint to the police or to any other authorities. He was not aware of his rights or remedies, and in particular he was unaware that his case could amount to one of human trafficking for forced labour. (18) The applicant in early December 2010 requested payment of the monies the employer owed him. The employer arranged the return of the applicant to Pakistan on the basis of him taking a holiday. While the applicant was in Pakistan, the employer terminated his contract and revoked his sponsorship in order to prevent the applicant from returning to Hong Kong to make a claim against him for the unpaid wages. (19) The applicant returned illegally to Hong Kong in April 2012 to claim his unpaid wages from his employer and to report the mistreatment that he had suffered from the employer. (20) The applicant attended the Immigration Department in early April 2012 to report that he had returned illegally to Hong Kong to claim his unpaid wages from his employer, and that he had been mistreated by him. (a) On the occasions where there are records of his attendances, these records do not disclose the detailed information that the applicant states he provided at the time. It is recorded that the applicant did mention that he had returned to claim unpaid wages for nearly 4 years[’] work, and this factor alone should have alerted the officer that his case may have involved a serious abuse of labour. (b) There was an occasion in early April 2012 when the applicant attended the office at Skyline Tower and spoke to an Immigration Department officer. He told the officer that he had come to Hong Kong illegally and was seeking help from the authorities to obtain his unpaid salary from his employer. He told the officer that he had worked in Hong Kong for almost 4 years and had never been paid. He gave details of the employer’s name and the place of work. He told the officer about the unfair treatment he had received from the employer, and mentioned that the employer had assaulted and abused him. The officer told him that this was a police case and that he should go to the police station. (21) From June to September 2013, the applicant was interviewed on several occasions by an officer of the Immigration Department in relation to his CAT claim. During the interviews, the applicant mentioned to the officer his entire story concerning the circumstances of his four-year employment in Hong Kong, including that he had been mistreated by the employer and that he had not been paid for the work that he had performed. (22) The applicant attended the Labour Department on 2 May 2012 and, with the assistance of a Labour Department officer, registered a claim for unpaid wages of over $200,000. He told the officer that he had worked for the employer for 4 years and was never paid. He said his employer had beaten him during his employment, and had forced him to return to Pakistan. (23) The applicant on various occasions in early April 2012, and on 17 May 2012, 28 May 2012, 27 July 2012 and 18 August 2012, attended various police stations and reported or complained to police officers about matters concerning threats by his employer in relation to his past employment with the employer and his claim for unpaid wages. (a) In early April 2012, the applicant attended the Tsim Sha Tsui Police Station, having been told to report his case to the police by an officer of the Immigration Department. He told a police officer that he had entered Hong Kong without a visa and that he had worked in Hong Kong for 4 years and had not been paid. The police officer told him to return to the Immigration Department. (b) On another occasion in early April 2012, the applicant attended Yau Ma Tei Police Station. He was spoken to by several police officers. He told them that he did not have a visa. He said he had worked for 4 years in Hong Kong and that he had not been paid. He said he returned to Hong Kong to obtain his unpaid salary. He said that during his employment he was beaten by his employer. He also told them that his employer wanted to kill him. The police officers said it was not a police matter and that he should go back to the Immigration Department. (c) On 16 July 2015, the applicant together with his legal representative attended Wan Chai Police Station. They reported to the police recent threats made to the applicant, and mentioned his human trafficking and forced labour case. A senior officer was informed about the applicant’s human trafficking and forced labour case. This officer registered the case, but suggested that the matter be referred to the Immigration Department. (24) The applicant stood trial in the District Court for an offence of robbery against associates of the employer, of which he was acquitted. There is a distinct possibility that he was wrongly accused of the crime.” 21. The Judge went on to record (at [161]): “161. On the evidence and materials before me, and for the purposes of these proceedings, I accept that: (1) The applicant was deceived by the employer and his family about his working conditions and payment of wages. (2) The applicant was regularly abused and beaten by the employer. Although this was not to a degree that required his hospitalisation, it was to such an extent that that the applicant was put in fear of the employer and placed under his control. (3) The applicant worked long hours and was taken advantage of by the employer. (4) Whilst the applicant was generally able to go out, he was nevertheless subject to restrictions by the employer that prevented him from living a normal life. His foreignness was exploited by the employer in order to restrict his movements and control him. He was also controlled due to the employer’s psychological and financial power over him. (5) The applicant lived in the office premises and relied on handouts, although he may have received some funds. (6) The applicant did not complain of the ill treatment he was receiving by the employer for the reasons he has given. He was afraid of the employer and his family and he expected he would receive the money owing to him. I should point out that the applicant, like many victims of abuse, tolerated and accepted more than he should have. Furthermore, he was unaware of his rights and remedies in Hong Kong and of the fact that the ill treatment he was suffering potentially constituted a case of human trafficking for forced labour. (7) The employer tricked the applicant into returning to Pakistan and then terminated his contract and sponsorship in order to prevent the applicant from making a claim against him for his unpaid salary. (8) The employer refused to pay the applicant his unpaid wages and used force and threats in Pakistan against him and his family to stop him from pursuing his claim. (9) The applicant returned to Hong Kong to press his claim for his unpaid salary, and to seek justice in relation to the treatment he had received from the employer. (10) The employer and his associates threatened the applicant and his family in relation to the applicant’s claim for the unpaid salary in Hong Kong. (11) The applicant attended the offices of the Immigration Department, the Labour Department and the Police on the various occasions that he has claimed, and revealed to the officers whom he saw information about the treatment he had received from the employer, and the fact that he had not been paid wages for a period of 4 years. The officers concerned should have been alerted that the applicant’s case involved a serious abuse of labour and that the applicant was potentially a victim of human trafficking or forced labour. (12) The applicant did not appreciate that he was a victim of human trafficking for forced labour.” B. The parties’ contentions on this appeal 22. The issues have narrowed since the hearing in the Court of Appeal. The appellant now seeks from this Court a declaration that the HKSARG is obliged to promote laws specifically criminalising human trafficking and forced or compulsory labour as specific offences. 23. It was the contention of Mr Raza Husain QC, leading counsel for the appellant,[26] that the two questions set out at [4] above should be answered in the affirmative and that, accordingly, the declaration he sought (at [25] below) should follow. On the first question, it was his contention that human trafficking was prohibited under BOR4 generally and as a separate and distinct concept. Alternatively, he contended that human trafficking was prohibited under BOR4 in respect of each of slavery, servitude and forced or compulsory labour. In the further alternative, he contended that BOR4 prohibited human trafficking for forced or compulsory labour. 24. On the second question, Mr Husain contended that there was a positive duty on the HKSARG to criminalise the activities prohibited under BOR4 by way of specific, or “bespoke”, offences. The scope of the duty to enact specific criminal legislation would depend on the answer to the first question as to the scope of BOR4. Mr Husain contended that this would in any event require a duty to criminalise forced or compulsory labour and also human trafficking for forced or compulsory labour. 25. Mr Husain did not seek an order of mandamus requiring the HKSARG to enact specific legislation but sought, instead, a declaration as to the duty upon the HKSARG. In this regard, he referred to the declaration that was sought in the judicial review, namely: “A declaration that the failure of the [HKSARG] to take steps to enact legislation specifically targeting human trafficking, including both adult and child trafficking and general labour trafficking as well as trafficking for sexual purposes is a breach of the Applicant’s rights under [BOR4] and of the rights of other trafficking victims in Hong Kong and of all persons entitled to the protection of the Basic Law who are or may be at risk of becoming trafficking victims in Hong Kong or elsewhere.” This declaration, to which he invited the Court to add “and forced or compulsory labour” after the words “as well as trafficking for sexual purposes” would, he contended, be a sufficient disposition of the appeal if the Court were to find in the appellant’s favour. 26. On the other hand, it was the contention of Lord Pannick QC, leading counsel for the respondents,[27] that the two questions should be answered in the negative. He contended that BOR4, properly construed, does not prohibit human trafficking either generally or for any of the prohibited activities of slavery, servitude or forced or compulsory labour. Nor does it impose a positive obligation on the HKSARG to enact specific legislation to criminalise any of the activities prohibited by BOR4. 27. In relation to the second question, Lord Pannick accepted that BOR4 does impose a positive obligation on the HKSARG to take steps to protect potential victims of slavery, servitude and forced or compulsory labour. However, while accepting that there had to be in place legislation to tackle these matters, he contended that there was no obligation to enact “bespoke” legislation for these purposes. He also contended that the new evidence placed by the respondents before the Court of Appeal showed that the HKSARG was seriously committed to combatting human trafficking and protecting vulnerable persons from exploitation. C. Question 1: The scope of BOR4 28. The HKBORO enacts the provisions of the International Covenant on Civil and Political Rights (“ICCPR”) as applied to Hong Kong and, by virtue of Article 39(1) of the Basic Law of the Hong Kong Special Administrative Region, is given constitutional status.[28] BOR4 (set out in Section A.1 above) is derived from and is in identical terms to Article 8 of the ICCPR (“ICCPR8”), save that the latter includes an additional sub-paragraph (3)(b) excluding imprisonment with hard labour from the definition of forced or compulsory labour. BOR4(3)(b) is therefore equivalent to ICCPR8(3)(c). BOR4/ICCPR8 are also similar to Article 4 of the European Convention on Human Rights (“ECHR4” and “ECHR” respectively), save that ECHR4 combines the prohibitions against slavery and servitude into one sub-paragraph and omits reference to the “slave-trade”.[29] C.1 Principles of construction 29. The answer to the question as to the scope of BOR4 is a matter of construction. BOR4 is a provision in the HKBORO, which is domestic legislation giving effect to the ICCPR, an international convention applicable to Hong Kong and having constitutional status by reason of Article 39(1) of the Basic Law. In the interpretation of constitutional instruments and statutes, the focus is on the context, purpose and actual words of the provision being construed. (1) In Ng Ka Ling & Others v Director of Immigration, the earliest case on interpretation of the Basic Law heard by this Court, the Court said: “… in ascertaining the true meaning of the instrument, the courts must consider the purpose of the instrument and its relevant provisions as well as the language of its text in the light of the context, context being of particular importance in the interpretation of a constitutional instrument.”[30] (2) Similarly, in Director of Immigration v Chong Fung Yuen, this Court held: “The exercise of interpretation requires the courts to identify the meaning borne by the language when considered in the light of its context and purpose. This is an objective exercise. Whilst the courts must avoid a literal, technical, narrow or rigid approach, they cannot give the language a meaning which the language cannot bear.”[31] (3) The importance of the language used is also a point emphasised in Lord Hoffmann’s judgment for the Privy Council in Matadeen v Pointu[32] and Lord Bingham’s speech in the House of Lords in the Roma Rights case.[33] In the latter case, Lord Bingham said (at [18]): “However generous and purposive its approach to interpretation, the court’s task remains one of interpreting the written document to which the contracting states have committed themselves. It must interpret what they have agreed. It has no warrant to give effect to what they might, or in an ideal world would, have agreed. This would violate the rule, also expressed in article 31(1) of the Vienna Convention, that a treaty should be interpreted in accordance with the ordinary meaning to be given to the terms of the treaty in their context. …” (4) The approach of the European Court of Human Rights (“ECtHR”) in respect of the ECHR is the same: “The Court is required to ascertain the ordinary meaning to be given to the words in their context and in the light of the object and purpose of the provision from which they are drawn.”[34] 30. Taking these principles into account, the Court should give a generous interpretation to provisions containing constitutional guarantees of freedoms and must keep in mind that constitutional provisions are living instruments intended to meet changing needs and circumstances: Ng Ka Ling & Others v Director of Immigration (supra.) at p.28D and p.29A. The principle that a statute is “always speaking” requires the courts to consider whether a new activity falls within the meaning of original statutory language: R (Quintavalle) v Health Secretary [2003] 2 AC 687 per Lord Bingham at [9]. C.2 The structure and context of BOR4 31. In terms of its structure, the provisions of BOR4 (as in the case of ICCPR8) are set out in three separate numbered paragraphs setting out prohibited forms of treatment. As a matter of immediate context within BOR4, therefore, the starting point is that the concepts of slavery (including all forms of slavery and the slave-trade), servitude and forced or compulsory labour are distinct. That the concepts are distinct can also be seen from the travaux préparatoires in relation to ICCPR8.[35] 32. These distinct concepts each have recognised definitions (and no issue is taken by the parties on these definitions): (1) “Slavery” and the “slave-trade” are defined in the Slavery Convention of 1926 in the following terms: “(1) Slavery is the status or condition of a person over whom any or all of the powers attaching to the right of ownership are exercised. (2) The slave trade includes all acts involved in the capture, acquisition or disposal of a person with intent to reduce him to slavery; all acts involved in the acquisition of a slave with a view to selling or exchanging him; all acts of disposal by sale or exchange of a slave acquired with a view to being sold or exchanged, and, in general, every act of trade or transport in slaves.”[36] (2) “Servitude” has been defined, by the ECtHR in relation to ECHR4, as “an obligation to provide one’s services that is imposed by the use of coercion” and therefore “to be linked with the concept of ‘slavery’”.[37] The ECtHR has also said: “… that servitude corresponds to a special type of forced or compulsory labour or, in other words, ‘aggravated’ forced or compulsory labour. As a matter of fact, the fundamental distinguishing feature between servitude and forced or compulsory labour within the meaning of Article 4 of the Convention lies in the victim’s feeling that their condition is permanent and that the situation is unlikely to change. It is sufficient that this feeling be based on the above-mentioned objective criteria or brought about or kept alive by those responsible for the situation.”[38] (3) The recognised definition of “forced or compulsory labour” is contained in Article 2(1) of the Forced Labour Convention 1930 (“FLC”), which states: “For the purposes of this Convention, the term forced or compulsory labour shall mean all work or service which is exacted from any person under the menace of any penalty and for which the said person has not offered himself voluntarily.” 33. The distinction between the concepts is reinforced by reference to section 5 of the HKBORO. (1) Section 5(1) provides: “In time of public emergency which threatens the life of the nation and the existence of which is officially proclaimed, measures may be taken derogating from the Bill of Rights to the extent strictly required by the exigencies of the situation, but these measures shall be taken in accordance with law.” (2) Section 5(2) then provides that: “No measure shall be taken under subsection (1) that – (a) is inconsistent with any obligation under international law that applies to Hong Kong (other than an obligation under the International Covenant on Civil and Political Rights); (b) involves discrimination solely on the ground of race, colour, sex, language, religion or social origin; or (c) derogates from articles 2, 3, 4(1) and (2), 7, 12, 13 and 15.” (3) Thus, the prohibition against slavery and servitude are non-derogable rights under the HKBOR, even in time of public emergency which threatens the life of the nation. In contrast, the prohibition of the requirement to perform forced or compulsory labour is derogable. This is plain both from: (i) the fact there are excluded forms of work or service expressly set out in BOR4(3)(b) itself (in contrast to the absence of any exclusions in BOR4(1) and (2)); and also, (ii) from the omission of BOR4(3)(a) from section 5(2)(c) of the HKBORO. C.3 The appellant’s construction arguments C.3a The primary argument: BOR4 prohibits human trafficking generally for exploitation 34. Mr Husain’s primary construction argument is that human trafficking generally for exploitation is included within the prohibition in BOR4. This argument is based on the wording in BOR4(1) stating that “slavery and the slave-trade in all their forms shall be prohibited” and relies on the definition of human trafficking in Article 3 of the Palermo Protocol (“PP3”). 35. PP3 defines the term “trafficking in persons” in the following terms: “(a) ‘Trafficking in persons’ shall mean the recruitment, transportation, transfer, harbouring or receipt of persons, by means of the threat or use of force or other forms of coercion, of abduction, of fraud, of deception, of the abuse of power or of a position of vulnerability or of the giving or receiving of payments or benefits to achieve the consent of a person having control over another person, for the purpose of exploitation. Exploitation shall include, at a minimum, the exploitation of the prostitution of others or other forms of sexual exploitation, forced labour or services, slavery or practices similar to slavery, servitude or the removal of organs; (b) The consent of a victim of trafficking in persons to the intended exploitation set forth in subparagraph (a) of this article shall be irrelevant where any of the means set forth in subparagraph (a) have been used; (c) The recruitment, transportation, transfer, harbouring or receipt of a child for the purpose of exploitation shall be considered ‘trafficking in persons’ even if this does not involve any of the means set forth in subparagraph (a) of this article; (d) ‘Child’ shall mean any person under eighteen years of age.” 36. There are three discrete parts to the definition of human trafficking in PP3. There must be an action involved, namely “the recruitment, transportation, transfer, harbouring or receipt of persons”. Next, certain wrongful means must be used, specifically “the threat or use of force or other forms of coercion, of abduction, of fraud, of deception, of the abuse of power or of a position of vulnerability or of the giving or receiving of payments or benefits to achieve the consent of a person having control over another person”. Finally, the action and means must be intended for a purpose, namely “exploitation”. It should be noted that: (1) The forms of exploitation are non-exhaustively described and include each of slavery, servitude and forced labour but also include other forms of exploitation such as prostitution and the removal of organs. (2) However, the exploitation need not have actually occurred for human trafficking to have taken place. It is the “recruitment, transportation” (etc.), by use of the specified means for the purpose of exploitation, that constitutes human trafficking. (3) For that reason, human trafficking is a process of, for example, recruiting (etc.) a person by the use of underhand means with a view to exploiting them in a particular manner. It is not directed to the outcome or substantive conduct, which is the carrying out of the actual exploitative purpose for which someone is, for example, deceptively recruited. 37. The prohibition in BOR4(1) on “slavery and the slave-trade in all their forms”, in particular the reference to “the slave-trade”, may be said to incorporate the process whereby a person is subjected to and brought into the state of slavery. Mr Husain therefore contended that, since BOR4(1) contemplates the prohibition of a process, when construed generously and evolutively, BOR4 must include a prohibition on human trafficking generally for exploitation and this would therefore include exploitation for the purpose of forced labour. 38. Mr Husain also relied on the concept of “modern slavery”, of which human trafficking is said to be a form in support of a generous and evolutive construction of BOR4. Paragraph [4] of the Explanatory Notes[39] to the Modern Slavery Act 2015, for example, states: “Modern slavery is a brutal form of organised crime in which people are treated as commodities and exploited for criminal gain. … Modern slavery, in particular human trafficking, is an international problem … Modern slavery takes a number of forms, including sexual exploitation, forced labour and domestic servitude, and victims come from all walks of life. …”[40] Judicially, recognition of contemporary forms of slavery also derives some support from the ECtHR decision in Rantsev v Cyprus and Russia,[41] to which it will be necessary to return later in this judgment. C.3b The alternative argument: BOR4 prohibits human trafficking for slavery, servitude and forced or compulsory labour 39. Mr Husain’s first alternative argument is a narrower version of his primary argument, relying on the same materials, to contend that the prohibition in BOR4, construed generously and evolutively, includes a prohibition on human trafficking for any of the three concepts in BOR4, namely slavery, servitude and forced or compulsory labour. C.3c The further alternative argument: BOR4(3)(a) prohibits human trafficking for forced labour 40. Mr Husain’s further alternative argument is based on the contention that BOR4(3)(a) not only prohibits forced or compulsory labour but also states that “[n]o one shall be required to perform” it. It is contended that the provision includes the prior process by which someone can be made to perform forced labour: a person may be “required to perform” forced labour at the point at which the trafficker recruits the victim because the trafficker knows that he requires the victim to perform forced labour. Hence, it is contended that BOR4(3)(a) includes a prohibition on human trafficking for forced labour. C.4 Rejecting the appellant’s construction arguments 41. For the following reasons, I cannot accept any of the appellant’s three construction arguments. C.4a Appellant’s arguments contrary to the language of BOR4 42. As a general proposition, when the Legislature has chosen, in the course of the same provision of an ordinance, to use different words to express concepts, it is to be assumed that the different words were intended to convey different meanings.[42] There is no reason to disapply that principle to an international treaty (the ICCPR) which a local ordinance (the HKBORO) has domesticated. On the contrary, Article 31(1) of the Vienna Convention on the Law of Treaties 1969 (“the Vienna Convention”) states that a treaty is to be interpreted “in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose”.[43] There would need to be good reason to assume that different terms employed in a single article of an international treaty were intended to convey the same meaning. 43. The appellant’s primary argument, however, blurs the distinctions between slavery, servitude and forced or compulsory labour in BOR4 to the point that they are no longer distinct concepts. The effect of the argument is to treat them all collectively under the rubric of human trafficking generally for exploitation. Although there is scope for suggesting that the concepts of slavery, servitude and forced or compulsory labour are cognate concepts with some overlap,[44] it is nevertheless clear that BOR4 is structured to distinguish between the three as separate and distinct concepts, with the latter concept of forced or compulsory labour being qualitatively different from slavery and servitude in that it is derogable (see Section C.2 above).[45] 44. Even assuming that human trafficking for the purposes of slavery falls within the phrase “slavery and the slave-trade in all their forms” in BOR4(1), the prohibitions in BOR4(2) and BOR4(3)(a) are prohibitions of substantive conduct rather than processes. The argument to the contrary in respect of BOR4(3)(a) is addressed below in Section C.4g. The prohibition on servitude in BOR4(2), however, is clearly a reference to the state of being held in servitude and not to any anterior process by which a person might be brought to that status. To expand the meaning of BOR4 to prohibit human trafficking for exploitation generally would be to ignore the language of BOR4 and impermissibly to alter the underlying concepts addressed in BOR4(2) and BOR4(3)(a). 45. Further, the reference in BOR4(1) to “slavery and the slave-trade in all their forms” is not a generic reference to servitude and forced or compulsory labour. The travaux préparatoires to the ICCPR show that ICCPR8(1) was intended to refer to slavery and the slave-trade in their narrow, traditional sense, i.e. as destruction of one’s juridical personality.[46] 46. The argument based on the expansion of the concept of slavery to include modern slavery (addressed further in Section C.4c below) might have some force if BOR4 only included a prohibition on “slavery and the slave-trade in all their forms”. However, BOR4 is not limited to that form of exploitation and includes, as separate prohibitions, servitude and forced or compulsory labour. There is no proper basis for blurring the distinctions to deprive those concepts of separate meanings by confining them to the meaning of slavery and the slave-trade in some wide generic sense of human trafficking generally for exploitation. 47. As Gleeson CJ observed, in R v Wei Tang: “It is important not to debase the currency of language, or to banalise crimes against humanity, by giving slavery a meaning that extends beyond the limits set by the text, context, and purpose of the 1926 Slavery Convention. In particular it is important to recognise that harsh and exploitative conditions of labour do not of themselves amount to slavery. The term ‘slave’ is sometimes used in a metaphorical sense to describe victims of such conditions, but that sense is not of present relevance. Some of the factors identified as relevant in Kunarac, such as control of movement and control of physical environment, involve questions of degree. An employer normally has some degree of control over the movements, or work environment, of an employee. Furthermore, geographical and other circumstances may limit an employee’s freedom of movement. Powers of control, in the context of an issue of slavery, are powers of the kind and degree that would attach to a right of ownership if such a right were legally possible, not powers of a kind that are no more than an incident of harsh employment, either generally or at a particular time or place.”[47] C.4b Impermissible application of Palermo Protocol to Hong Kong 48. The Palermo Protocol does not apply to Hong Kong. When the People’s Republic of China (“PRC”) acceded to the Palermo Protocol on 8 February 2010, the Central People’s Government (“CPG”) lodged a declaration in accordance with Article 153 of the Basic Law[48] to the effect that, unless notified by it, the Palermo Protocol should not apply to the Hong Kong Special Administrative Region.[49] 49. In the appellant’s written case, it was rightly accepted that this is an important matter.[50] Its significance lies in the fact that, under the common law dualist principle, an international treaty is not self-executing and that, until made part of Hong Kong domestic law by legislation, the provisions of such a treaty do not confer or impose any rights or obligations on individual citizens.[51] It also lies in the fact that the PRC’s reservation in respect of the Palermo Protocol was a deliberate choice not to apply its provisions to Hong Kong. As Lord Bingham observed, in the Roma Rights case: “The principle that pacta sunt servanda cannot require departure from what has been agreed. This is the more obviously true where a state or states very deliberately decided what they were and were not willing to undertake to do. …”[52] 50. However, the appellant’s argument that it is nevertheless permissible to use PP3(a) as an aid to construe ICCPR8, which does apply to Hong Kong and which BOR4 implements, so as to construe BOR4 as prohibiting human trafficking generally for exploitation, or as prohibiting human trafficking for servitude or forced or compulsory labour would be to apply the provisions of the Palermo Protocol to Hong Kong despite the clear intention of the CPG and the HKSARG that it should not apply. The argument is convoluted and not at all easy to follow. However, be that as it may, it would be inappropriate to give a “backdoor” application to a treaty which the PRC has expressly declared should not apply to Hong Kong. 51. The appellant’s written case referred to Article 31(3)(c) of the Vienna Convention[53] to support the contention that a treaty should so far as possible be interpreted in harmony with other rules of international law. Mr Husain also cited Demir and Baykara v Turkey[54] in support of the proposition that ratification by a state of an instrument is not necessary for it to be an aid to interpretation of a treaty to which that state is a party. However, the Vienna Convention is not relevant in the present context as the Court is not construing an unincorporated treaty which has not been applied to Hong Kong. Even if it were relevant, it would be to take the proposition in Demir and Baykara v Turkey too far to ignore the declaration by the PRC that the Palermo Protocol should not apply to Hong Kong at all and to conclude that, notwithstanding that express reservation, PP3(a) should be used to construe ICCPR8 and therefore BOR4. C.4c Modern slavery as a concept does not eliminate the distinctions between slavery, servitude and forced or compulsory labour 52. The term “modern slavery” is not a term of art and its genesis is not clear. The Explanatory Notes to the UK’s Modern Slavery Act 2015 have been referred to above. However, the Act does not seek to define the concept of “modern slavery” nor does it criminalise modern slavery as a separate concept. Instead, the principal offences under that Act are: (i) slavery and servitude (section 1(1)(a)); (ii) forced or compulsory labour (section 1(1)(b)); and (iii) human trafficking (section 2). The forms of exploitation which are the target of the offence of human trafficking are listed in section 3 and are: slavery, servitude and forced or compulsory labour; sexual exploitation; removal of organs etc. (under the Human Tissue Act 2004); securing services etc. by force, threats or deception; and securing services etc. from children and vulnerable persons. 53. The earliest judicial references in the authorities cited in this Court to contemporary forms of slavery are the judgments of the Trial Chamber of the International Criminal Tribunal for the Former Yugoslavia (“ICTY”) in Prosecutor v Kunarac and Others[55] and in the Appeals Chamber of that Tribunal in the same case.[56] That case was referred to in Rantsev v Cyprus and Russia,[57] where the ECtHR said: “280. The Court observes that the International Criminal Tribunal for the Former Yugoslavia concluded that the traditional concept of ‘slavery’ has evolved to encompass various contemporary forms of slavery based on the exercise of any or all of the powers attaching to the right of ownership. In assessing whether a situation amounts to a contemporary form of slavery, the Tribunal held that relevant factors included whether there was control of a person’s movement or physical environment, whether there was an element of psychological control, whether measures were taken to prevent or deter escape and whether there was control of sexuality and forced labour. 281. The Court considers that trafficking in human beings, by its very nature and aim of exploitation, is based on the exercise of powers attaching to the right of ownership. It treats human beings as commodities to be bought and sold and put to forced labour, often for little or no payment, usually in the sex industry but also elsewhere. It implies close surveillance of the activities of victims, whose movements are often circumscribed. It involves the use of violence and threats against victims, who live and work under poor conditions. It is described by Interights and in the explanatory report accompanying the Anti-Trafficking Convention as the modern form of the old worldwide slave trade. The Cypriot Ombudsman referred to sexual exploitation and trafficking taking place, ‘under a regime of modern slavery’.”[58] 54. However, there are a number of problematic features in this description of human trafficking as a form of modern slavery for the purposes of construing BOR4. (1) First, human trafficking is already defined internationally in PP3(a). That definition has been addressed above: it consists of three elements describing a process. The ECtHR’s definition of human trafficking (see the citation above to Rantsev v Cyprus and Russia at [281]), in contrast, looks to an outcome rather than the process. (2) Secondly, in limiting the aim of human trafficking to the exercise of ownership rights, the ECtHR’s definition of human trafficking unduly limits the forms of exploitation that may be involved in human trafficking too narrowly. Contrast instead the forms of exploitation that are included in the definition of human trafficking in PP3(a). (3) Thirdly, as will be seen in the discussion in Section C.4d below, the ECtHR’s approach to human trafficking is, at the same time, too broad as it ignores the separate concepts of slavery, servitude and forced or compulsory labour contained in ECHR4 (as they are in BOR4).[59] 55. These are cogent reasons for rejecting the construction of BOR4 that proceeds from the basis that human trafficking is modern slavery and, therefore, since (it is said) modern slavery must fall within the prohibition in BOR4(1), BOR4 should be construed as prohibiting human trafficking generally for exploitation. In addition to the objection noted at [44] above, if BOR4 is construed to prohibit human trafficking generally for exploitation, it would also prohibit other purposes than the three concepts of slavery, servitude and forced or compulsory labour (such as the removal of human organs) and would therefore ignore the specific concepts in BOR4. 56. On his first alternative construction argument that BOR4 should instead be construed to prohibit human trafficking for slavery, servitude and forced or compulsory labour, Mr Husain was constrained to accept in argument before us that his construction argument would mean that human trafficking for slavery and human trafficking for servitude are non-derogable prohibitions but that human trafficking for forced or compulsory labour is derogable (see [33] above). This is an artificial construction which has no discernable justification in logic or policy. C.4d No judicial support for the appellant’s construction arguments 57. The appellant’s construction arguments are not supported by any judicial decisions on ICCPR8 or BOR4 or similar provisions. In the appellant’s written case, however, it was contended that the decision of the ECtHR in Rantsev v Cyprus and Russia[60] is “valuable persuasive authority on the general principles underlying the protection” of BOR4(1).[61] This was the only judicial decision on ICCPR8 directly supporting the appellant’s arguments. 58. That case concerned the adequacy of measures taken by Cyprus to investigate the death of a young woman who had been allegedly trafficked from Russia to work in a cabaret in Cyprus. In relation to ECHR4, it was held, in that case, that there had been a violation by Cyprus in failing to afford the woman practical and effective protection against trafficking and exploitation in general and specific measures of protection. Of relevance are the following paragraphs of the ECtHR’s judgment in relation to ECHR4: “277. The absence of an express reference to trafficking in the Convention is unsurprising. The Convention was inspired by the Universal Declaration of Human Rights, proclaimed by the General Assembly of the United Nations in 1948, which itself made no express mention of trafficking. In its art.4, the Declaration prohibited ‘slavery and the slave trade in all their forms’. However, in assessing the scope of art.4 of the Convention, sight should not be lost of the Convention’s special features or of the fact that it is a living instrument which must be interpreted in the light of present-day conditions. The increasingly high standards required in the area of the protection of human rights and fundamental liberties correspondingly and inevitably require greater firmness in assessing breaches of the fundamental values of democratic societies. 278. The Court notes that trafficking in human beings as a global phenomenon has increased significantly in recent years. In Europe, its growth has been facilitated in part by the collapse of former Communist blocs. The conclusion of the Palermo Protocol in 2000 and the Anti-Trafficking Convention in 2005 demonstrate the increasing recognition at international level of the prevalence of trafficking and the need for measures to combat it. 279. The Court is not regularly called upon to consider the application of art.4 and, in particular, has had only one occasion to date to consider the extent to which treatment associated with trafficking fell within the scope of that article. In that case, the Court concluded that the treatment suffered by the applicant amounted to servitude and forced and compulsory labour, although it fell short of slavery. In light of the proliferation of both trafficking itself and of measures taken to combat it, the Court considers it appropriate in the present case to examine the extent to which trafficking itself may be considered to run counter to the spirit and purpose of art.4 of the Convention such as to fall within the scope of the guarantees offered by that article without the need to assess which of the three types of proscribed conduct are engaged by the particular treatment in the case in question.”[62] 59. The ECtHR adopted a broad brush approach in holding that human trafficking itself fell within the scope of ECHR4 and that it was not necessary to identify whether the treatment complained of was slavery, servitude or forced or compulsory labour. It held: “282. There can be no doubt that trafficking threatens the human dignity and fundamental freedoms of its victims and cannot be considered compatible with a democratic society and the values expounded in the Convention. In view of its obligation to interpret the Convention in light of present-day conditions, the Court considers it unnecessary to identify whether the treatment about which the applicant complains constitutes “slavery”, “servitude” or “forced and compulsory labour”. Instead, the Court concludes that trafficking itself, within the meaning of art.3(a) of the Palermo Protocol and art.4(a) of the Anti-Trafficking Convention, falls within the scope of art.4 of the Convention. The Russian Government’s objection of incompatibility ratione materiae is accordingly dismissed.” 60. The appropriateness of the Hong Kong courts taking account of established principles of international jurisprudence, including that of the ECtHR, in interpreting fundamental rights in the Basic Law and the BOR was acknowledged by this Court in Shum Kwok Sher v HKSAR.[63] The decisions of the ECtHR on provisions of the ECHR in the same or substantially the same terms as the BOR, though not binding on the courts of Hong Kong, are of high persuasive authority and have been so regarded by this Court.[64] However, the ECtHR plays a supra-national role under an international treaty in relation to member states of the Council of Europe and therefore functions quite differently to this Court which is a court of final adjudication in a domestic forum. Such differences may sometimes make it inappropriate to take the path followed by the ECtHR.[65] In addition, the ECtHR’s jurisprudence is fact sensitive so that it is hazardous to apply the decisions of that court to facts which are different.[66] 61. Regardless of that general caveat, the broad brush approach of the ECtHR in Rantsev v Cyprus and Russia is in any event unsatisfactory and should not be followed in respect of BOR4 because it expressly ignores the separate concepts of slavery, servitude and forced or compulsory labour. The appellant’s own case does not seek to ignore the distinctions between the different types of prohibited conduct in BOR4 (see FN8 above). That approach is correct since it would otherwise debase the language of BOR4 to treat the prohibited types of conduct in this broad brush manner. 62. As a matter of context, it should also be noted that it was clearly material, however, to the ECtHR’s decision in that case that both Cyprus and Russia were parties to the Palermo Protocol[67] and also that the Anti-Trafficking Convention[68] had been signed by 41 Member States of the Council of Europe and 26 had also ratified it (including Cyprus).[69] It is also relevant to note, as a matter of context when considering the European jurisprudence, that Article 5 of the Charter of Fundamental Rights of the European Union contains an express prohibition against trafficking in human beings.[70] 63. A similar contextual qualification also applies to Mr Husain’s reliance on a decision of the Inter-American Court of Human Rights in Case of the Hacienda Brasil Verde Workers v Brazil.[71] Mr Husain contended that the Inter-American Court’s interpretation in that case of the scope of the freedom from slavery in Article 6 of the American Convention on Human Rights supported his primary BOR4 construction argument. However, Article 6 of the American Convention is materially quite different to BOR4 in that it includes a specific prohibition on trafficking in women.[72] The principal issue in that case was whether the absence of protection in Brazil against the trafficking of male workers was discriminatory and not, as here, whether the concept of a prohibition of trafficking generally for exploitation can be implied into ICCPR8. C.4e HRC General Comment No.28 and HRC Concluding Observations provide no support for appellant’s construction of BOR4 64. In the appellant’s written case, it was contended that the United Nations Human Rights Committee (“HRC”) “has concluded that ICCPR Article 8 covers [human trafficking], despite the absence of any express reference to [human trafficking]”.[73] That contention was based on the statement by the HRC in its General Comment No.28 in respect of Article 3 of the ICCPR,[74] at [12], that: “Having regard to their obligations under article 8, States parties should inform the Committee of measures taken to eliminate trafficking of women and children, within the country or across borders, and forced prostitution. They must also provide information on measures taken to protect women and children, including foreign women and children, from slavery, disguised, inter alia, as domestic or other kinds of personal service. States parties where women and children are recruited, and from which they are taken, and States parties where they are received should provide information on measures, national or international, which have been taken in order to prevent the violation of women’s and children’s rights.” 65. It is, however, to read too much into that paragraph to conclude that ICCPR8 (and therefore BOR4) are to be construed as prohibiting human trafficking either generally for exploitation or for the specified purposes of slavery, servitude and forced or compulsory labour. There is no indication that the HRC was there stating definitively that ICCPR8 must, as a matter of its proper construction, be taken to prohibit human trafficking per se. The paragraph instead refers to the desirability of States Parties informing the HRC of measures taken to eliminate trafficking of women and children. Having regard to the prohibition on slavery and the slave-trade in all their forms in ICCPR8(1), that statement is understandable. This is reinforced by the statement of the desirability of the provision of information on measures to protect women and children “from slavery, disguised, inter alia, as domestic or other kinds of personal service”. 66. In any event, the weight to be attached to this particular paragraph in this General Comment is clearly a point open to argument. Although Sir Anthony Mason NPJ said, in this Court, that “[t]he General Comments are a valuable jurisprudential resource which is availed of by the Committee in its adjudicative role” and that “they provide influential guidance as to how the ICCPR is applied and will be applied by the [HRC] when sitting as a judicial body in making determinations”,[75] it is far from clear that those comments should apply to this particular General Comment. There is no indication in General Comment No.28 of the analysis by the HRC of ICCPR8 and its provisions, nor is there any indication of the scope of any representations made to the HRC regarding the obligations arising under ICCPR8. Absent a clear indication of the reasoning of the HRC for the statements in question, it would not be safe to conclude that the paragraph in question is intended to go as far as the proposition for which Mr Husain cites it. 67. This view is supported in particular by the travaux préparatoires which show that a proposal to replace the term “slave-trade” in ICCPR8(1) with “trade in human beings”, in order to cover traffic in women as well, was rejected on the basis that ICCPR8(1) should refer only to slavery and the slave-trade in their true sense.[76] 68. Similarly, Mr Husain’s reliance on the HRC’s Concluding Observations on the Third Periodic Report of Hong Kong, China,[77] as well as the Concluding Observations of the HRC in respect of the reports of various other States Parties in support of his construction of ICCPR8 as including a prohibition against human trafficking cannot be read as binding statements on the scope of ICCPR8. 69. In its observations on the Third Periodic Report of Hong Kong, for example, the HRC stated (at [20]): “The Committee is concerned about the persistence of the phenomenon of trafficking in persons in Hong Kong, China, and reports that Hong Kong, China, is a source, destination, and transit point for men, women, and teenage girls from Hong Kong, the mainland of China, and elsewhere in Southeast Asia, subjected to human trafficking and forced labour. The Committee is concerned about the reluctance of Hong Kong, China, to take steps which could lead to the extension of the Protocol to Prevent, Suppress and Punish Trafficking in Persons, Especially Women and Children, supplementing the United Nations Convention against Transnational Organized Crime (Palermo Protocol) to Hong Kong, China, (art. 8). Hong Kong, China, should intensify its efforts to identify victims of trafficking and ensure the systematic collection of data on trafficking flows to and in transit through the region, review its sentencing policy for perpetrators of trafficking-related crimes, support private shelters offering protection to victims, strengthen victim assistance by ensuring interpretation, medical care, counselling, legal support for claiming unpaid wages and compensation, long-term support for rehabilitation and stability of legal status to all victims of trafficking. The Committee recommends the inclusion of certain practices regarding foreign domestic workers in the definition of the crime of human trafficking. Hong Kong, China, should consider taking steps which could lead to the extension of the Palermo Protocol to Hong Kong, China, in order to strengthen its commitment to fight trafficking in persons in the region.” 70. The status of Concluding Observations of the HRC is ill-defined. They have no binding status and, although deserving of respect given the eminence of their authors, a distinction is to be drawn between pronouncements by the HRC on issues of violation of the ICCPR and where they otherwise purport to interpret treaty provisions, on the one hand, and where they provide general advice on strategies for enhanced implementation of a treaty and when they opine on matters extraneous to the actual treaty obligations of a State Party, on the other.[78] 71. The prescriptive nature of the HRC’s Concluding Observations in respect of Hong Kong in the passage quoted above are self-evident and neither those observations nor the other Concluding Observations referred to provide any reasoned analysis of why human trafficking falls within the protection afforded by ICCPR8. C.4f HKSAR policy does not govern the interpretation of BOR4 72. A point which Mr Husain repeatedly stressed in his submissions was that the respondents’ evidence in this case shows that the HKSARG “attaches great importance in combating human trafficking, as well as servitude and forced labour” and “spares no effort in combating human trafficking”.[79] 73. Whilst the respondents’ evidence as to its commitment to addressing human trafficking and the practical measures taken by the HKSARG is certainly relevant in the context of one aspect of Question 2 (see below), that evidence is not relevant to the construction of BOR4. There is a clear and obvious distinction between legal obligation and government policy. The decision to pursue a particular policy is a matter which may be based on administrative rather than legal considerations and the mere fact of pursuit of a policy to combat human trafficking does not require BOR4 to be construed as prohibiting that activity. C.4g BOR4(3)(a) prohibits the substantive conduct not the process 74. Mr Husain’s further alternative construction argument can be disposed of briefly. The argument is that the word “required” in BOR4(3)(a) must refer to a process, since a person can be required to perform forced or compulsory labour even before he actually does so and, therefore, that human trafficking for forced or compulsory labour would fall within the meaning of requiring someone to perform forced or compulsory labour. 75. This construction of BOR4(3)(a) is untenable. The word “required” is singularly inapt to convey a meaning of trafficking. If this was intended a much more suitable term would have been used. Additionally, the word “required” is also used twice in the descriptions in BOR4(3)(b) of excluded activities which do not fall within the prohibition on forced or compulsory labour: see BOR4(3)(b)(i) “any work or service normally required of a person who is under detention …” and BOR4(3)(b)(ii) “any service of a military character and, where conscientious objection is recognized, any national service required by law of conscientious objectors” (italics added). 76. Mr Husain accepted that the activities in BOR4(3)(b) referred to conduct rather than processes. That being so, it would mean that the same word used in BOR4(3) was to be given one meaning in BOR4(3)(a) and another in BOR4(3)(b). There is no good reason to do so. On the contrary, “it is a sound rule of construction to give the same meaning to the same words occurring in different parts of an Act of Parliament …”.[80] 77. The word “required” in BOR4(3)(a) is consistent with the definition of forced or compulsory labour in Article 2(1) of the FLC (see [32(3)] above). While BOR4(3)(a) uses the passive voice “required”, Article 2(1) of the FLC uses the active voice to define forced or compulsory labour as “work or service which is exacted from any person”. It is the substantive work or service that is prohibited. 78. Mr Husain referred, in the course of the argument, to the “Protocol of 2014 to the FLC 1930” adopted by the International Labour Organization (“ILO”), the Preamble of which included the statement: “Recognizing that the context and forms of forced or compulsory labour have changed and trafficking in persons for the purposes of forced or compulsory labour, which may involve sexual exploitation, is the subject of growing international concern and requires urgent action for its effective elimination”. He also referred to Article 1(3) of that Protocol which states: “The definition of forced or compulsory labour contained in the Convention [i.e. the FLC] is reaffirmed, and therefore the measures referred to in this Protocol shall include specific action against trafficking in persons for the purposes of forced or compulsory labour.” 79. The 2014 Protocol to the FLC does not support the appellant’s construction of BOR4(3)(a). The PRC, and therefore Hong Kong, is not a party to that Protocol and so its terms are not applicable. Nor does it purport to amend the definition of forced or compulsory labour in the FLC (at [32(3)] above). Instead, Article 1(3) of the Protocol “reaffirms” that definition and states that the measures in the Protocol are additional measures to be taken to combat human trafficking “for the purposes of forced or compulsory labour”. The expanded obligations under the 2014 Protocol, which does not apply to Hong Kong in any event, cannot be used to support the appellant’s construction of BOR4(3)(a).[81] C.5 Answering Question 1 80. I would therefore answer Question 1 as follows. (1) Although BOR4(1) may, by referring to “slavery and the slave-trade in all their forms”, prohibit trafficking for the purposes of slavery, BOR4(1) does not prohibit human trafficking generally for the purposes of exploitation. (2) Nor does BOR4(1) prohibit human trafficking for the purposes of servitude and forced or compulsory labour (even assuming it prohibits human trafficking for slavery). (3) BOR4(3)(a) does not prohibit human trafficking for forced or compulsory labour. 81. Although BOR4, as a constitutionally protected human right is to be construed generously, its language does not support the expansive constructions contended for on behalf of the appellant. Similarly, although the BOR is to be construed as a living instrument, there is no justification for giving an updated construction to the concepts of slavery, servitude and forced or compulsory labour which result in a meaning that is conceptually different: Boyce and another v The Queen [2005] 1 AC 400 at [28]-[29]; Birmingham City Council v Oakley [2001] 1 AC 617 at pp.631E-632C. 82. This is not, it should be emphasised, to endorse a literal, technical, narrow or rigid construction of the language of BOR4. That, it is clear, must be avoided.[82] Even so, for the reasons set out above, the appellant’s construction arguments cannot be accepted. D. Question 2: Does BOR4 require bespoke legislation? 83. Question 2 concerns the scope of the positive steps required to be taken by the HKSARG under BOR4. The respondents accept that there is an investigative duty imposed on the HKSARG by BOR4 and that the respondents failed to discharge that duty in respect of the appellant (and hence there is no appeal against the Court of Appeal’s decision on that issue). However, the question before us is whether BOR4 imposes an obligation on the HKSARG to enact specific or bespoke legislation to afford protection against the activities prohibited by BOR4 or whether reliance on a “patchwork” of statutory provisions is sufficient. D.1 The appellant’s case on Question 2 84. The appellant’s submissions on Question 2, as set out in his written case, are that: “a. There is an absolute (alternatively, a contingent) obligation to specifically criminalise all forms of [human trafficking]; b. There is a contingent obligation to specifically criminalise [forced or compulsory labour]; c. The contingency is that a patchwork of general offences, adaptively applied, is not shown by the Government to be effective – it is not necessary for the individual to show further that the enactment of a specific offence is the only means of redressing the ineffectiveness; d. That contingency is made out on the facts accepted by the courts below in the present case.”[83] 85. Having concluded, in answer to Question 1, that BOR4 does not contain a prohibition against human trafficking either generally for exploitation or specifically for forced or compulsory labour (see Section C.5 above), the premise on which the appellant’s contention on Question 2 that the HKSARG is under a duty to promote a specific offence criminalising human trafficking is undermined.[84] This conclusion also means that the appellant’s reliance on Article 5 of the Palermo Protocol[85] as an aid to interpreting BOR4 is misplaced. 86. However, the appellant’s contention that there is an obligation under BOR4 to criminalise forced or compulsory labour remains to be determined. In support of this contention, the appellant relies variously on: Article 25 of the FLC (“FLC25”); statements of the ILO Committee of Experts and Concluding Observations of the HRC; and the jurisprudence of the ECtHR on ECHR4. Mr Husain repeated, in this context, his submission that the respondents’ evidence showed that the HKSARG was committed, as a matter of policy, to combatting human trafficking and that this therefore showed that there was no policy objection to the enactment of a bespoke offence to criminalise forced or compulsory labour. He also submitted that, by analogy with torture, a patchwork of general criminal offences was not sufficient to protect the rights under BOR4. 87. The respondents accept that the authorities show that there is a positive duty on the HKSARG under BOR4 to have in place measures providing practical and effective protection against the activities prohibited under BOR4. However, they do not accept there is a duty to enact bespoke criminal legislation for that purpose. D.2 The HKSARG’s margin of discretion under BOR4 88. The rights protected under BOR4, like the right of peaceful assembly, involve a positive duty on the part of the HKSARG: see Leung Kwok Hung and Others v HKSAR (2005) 8 HKCFAR 229 at [22]. But, just as the authorities cannot guarantee that all lawful assemblies will always proceed peacefully, they cannot ensure that no person will ever be subjected by another to the treatment prohibited by BOR4. Since (as will be seen from the ECtHR decisions discussed in Section D.3a below) the touchstone is whether the protection of the rights under BOR4 is practical and effective, the decision as to how to achieve such protection must necessarily be a matter for the HKSARG, subject, of course, to the supervision of the courts to assess the practical efficacy of the measures adopted. 89. Part of the relevant context of ICCPR8, applied to Hong Kong through BOR4, is Article 2(2) of the ICCPR (“ICCPR2(2)”), which provides: “Where not already provided for by existing legislative or other measures, each State Party to the present Covenant undertakes to take the necessary steps, in accordance with its constitutional processes and with the provisions of the present Covenant, to adopt such laws or other measures as may be necessary to give effect to the rights recognized in the present Covenant.” 90. The wording of BOR4/ICCPR8 does not say in terms that specific criminal offences must be created to support the prohibition against slavery, servitude and forced or compulsory labour. On the contrary, as ICCPR2(2) shows, the obligation on States Parties presupposes a discretion as to the manner in which the rights, including those under BOR4/ICCPR8, are protected. 91. In the circumstances, it is appropriate that the discretion as to the choice of measures to be used should be wide as recognised by this Court and by the ECtHR in relation to the duty to protect the right to freedom of assembly.[86] Even in the context of the non-derogable rights not to be subject to torture or cruel, inhuman or degrading treatment, the ECtHR has recognised that the positive obligation should “be interpreted in such a way as not to impose an excessive burden on the authorities, bearing in mind, in particular, the unpredictability of human conduct and operational choices which must be made in terms of priorities and resources.”[87] 92. Inherent in this wide margin of discretion, also, is the need to consider whether an alleged breach of BOR4 rights has arisen because of the lack of a specific criminal offence. For the contention that a bespoke criminal offence is required in this case, there must be a causal connection between the absence of that specific offence against forced or compulsory labour and the breach of the appellant’s BOR4 rights: see, in this context, the reference to CN v UK (below).[88] To similar effect on the need for a causal connection between alleged failure to implement a particular measure and breach of an ECHR right, see: Beganović v Croatia,[89] O’Keeffe v Ireland[90] and Botta v Italy.[91] As will be seen (in Section D.5 below), that causal connection is not established in the present case. D.3 The appellant’s arguments for a bespoke criminal offence 93. For the following reasons, I would reject the appellant’s contention that BOR4 necessarily requires the HKSARG to enact a bespoke offence criminalising forced or compulsory labour (although, as will be seen, this may be demonstrated to be necessary if the measures adopted by the HKSARG are shown not to afford practical and effective protection of BOR4 rights). D.3a The European jurisprudence 94. Mr Husain relied on the decisions of the ECtHR in Siliadin v France,[92] CN and V v France,[93] CN v UK,[94] and Rantsev v Cyprus and Russia[95] in support of his contention that there is a positive obligation on the HKSARG under BOR4 to enact a bespoke offence criminalising forced or compulsory labour. 95. Siliadin v France concerned an applicant, aged 15 at the relevant time, who was found (at [129]) to have been held in servitude within the meaning of ECHR4. The ECtHR held (at [89]) that “governments have positive obligations … to adopt criminal law provisions which penalise the practices referred to in [ECHR4] and to apply them in practice” and (at [112]) that “Member States’ positive obligations under [ECHR4] must be seen as requiring the penalisation and effective prosecution of any act aimed at maintaining a person in such a situation”. 96. In that case, the French Government relied on various articles of the Criminal Code but the ECtHR observed (at [142]) that “those provisions do not deal specifically with the rights guaranteed under [ECHR4], but concern, in a much more restrictive way, exploitation through labour and subjection to working and living conditions that are incompatible with human dignity”. The ECtHR went on to say that “[i]t therefore needs to be determined whether, in the instant case, those articles provided effective penalties for the conduct to which the applicant had been subjected.” That question was answered in the negative on the facts as assessed by the ECtHR (at [148]). 97. The approach in Siliadin v France was followed in CN and V v France and in CN v UK respectively. (1) In CN and V v France, at [105]-[108], the ECtHR followed the reasoning and conclusion in Siliadin v France and found, again on the facts, that there had been a violation of ECHR4 in respect of the first applicant (who had been found to have been subjected to forced or compulsory labour) “as regards the State’s positive obligation to set in place a legislative and administrative framework to effectively combat servitude and forced labour”. (2) In CN v UK, at [66] and [76], the ECtHR followed Siliadin v France and found, on the facts of that case, that the investigation into the applicant’s complaints of domestic servitude was ineffective due to the absence of specific legislation criminalising such treatment. Consistent with Siliadin v France, the standard applied was whether “the legislative provisions in force in the United Kingdom at the relevant time were inadequate to afford practical and effective protection against treatment falling within the scope of [ECHR4]”. 98. The ECtHR followed the same approach in Rantsev v Cyprus and Russia. There, the ECtHR found, on the facts, that there had been a breach of the applicant’s ECHR4 rights by Cyprus to afford her “practical and effective protection against trafficking and exploitation” (at [293]) and in respect of the failures of the police authorities to take effective measures for her protection (at [298]), and by Russia in respect of its procedural obligation to investigate trafficking in relation to her case (at [308]). The ECtHR did not, however, hold that there was a breach of ECHR4 by reason of the failure of either Cyprus or Russia to enact specific legislation to criminalise any particular activity. 99. It is also to be noted, as a matter of context, that in Rantsev v Cyprus and Russia the ECtHR took into account the obligations under the Palermo Protocol and the Anti-Trafficking Convention in considering the scope of the obligations arising under ECHR4: “In its Siliadin judgment, the Court confirmed that art.4 entailed a specific positive obligation on Member States to penalise and prosecute effectively any act aimed at maintaining a person in a situation of slavery, servitude or forced or compulsory labour. In order to comply with this obligation, Member States are required to put in place a legislative and administrative framework to prohibit and punish trafficking. The Court observes that the Palermo Protocol and the Anti-Trafficking Convention refer to the need for a comprehensive approach to combat trafficking which includes measures to prevent trafficking and to protect victims, in addition to measures to punish traffickers. It is clear from the provisions of these two instruments that the contracting states, including almost all of the Member States of the Council of Europe, have formed the view that only a combination of measures addressing all three aspects can be effective in the fight against trafficking. Accordingly, the duty to penalise and prosecute trafficking is only one aspect of Member States’ general undertaking to combat trafficking. The extent of the positive obligations arising under art.4 must be considered within this broader context.”[96] 100. Given the different context of ECHR4, the European jurisprudence relied upon by the appellant in relation to BOR4 must be read with some caution. Even so, none of the ECtHR cases relied upon by the appellant supports the contention that BOR4 necessarily requires the HKSARG to enact specific criminal offences to penalise slavery, servitude and forced or compulsory labour. Instead, the relevant inquiry is whether, on the facts of any given case, the protection of the rights under the ECHR or, in our case, the BOR has been rendered “practical and effective”.[97] D.3b FLC25 and the views of the ILO Committee of Experts 101. FLC25 provides that: “The illegal exaction of forced or compulsory labour shall be punishable as a penal offence, and it shall be an obligation on any Member ratifying this Convention to ensure that the penalties imposed by law are really adequate and are strictly enforced.” 102. Whilst the FLC is applicable to Hong Kong, FLC25 has not been incorporated into Hong Kong law.[98] That notwithstanding, the Committee of Experts of the ILO has stated that: “Although desirable, general provisions making the use of forced labour a punishable offence with appropriate penalties are not always necessary to give effect to the provisions of Article 25 of the Convention. The Committee seeks assurance from governments, often by requesting supplementary information, that there are provisions allowing punishment of persons guilty of forced labour practices found to have occurred in their country. A wide range of provisions can thus be used in practice by the courts, especially where those provisions are interpreted together (with others such as those concerning coercion, the use of threats or violence, detention, exploitation of vulnerability, freedom of employment, and so on).”[99] 103. It is also clear from the Committee of Expert’s Report that, whilst “[i]t may in practice not be enough to adopt provisions making the use of forced labour an offence and establishing penalties for it in general terms”, by implication general provisions may be enough if the protection is sufficient for victims and to enable the authorities responsible for protecting their rights to ensure that they are enforced.[100] 104. This material therefore falls far short of supporting the contention that a bespoke offence criminalising forced or compulsory labour is a positive obligation of the HKSARG under BOR4. D.3c Concluding Observations of the HRC 105. In the appellant’s written case, reference is made to the Concluding Observations of the HRC in respect of Brazil in 1996 where the HRC said: “The Committee urges the State party to enforce laws prohibiting forced labour … and to implement programmes to prevent and combat such human rights abuses. … It is imperative that persons who are responsible for, or who directly profit from, forced labour … be severely punished under law.”[101] 106. This comment must be read in its proper context, as must also the HRC’s Concluding Observations on the Third Periodic Report of Hong Kong, China,[102] and the many other similar examples to which Mr Husain referred the Court during the hearing. The nature of the HRC’s Concluding Observations has been addressed above (in Section C.4e) and they are tailored to the particular circumstances of the individual States Parties concerned. There is no statement in any of those Concluding Observations of the existence of a positive legal obligation to enact a specific law to criminalise forced or compulsory labour, as opposed to general criminal legislation which may be used in practice to combat the exploitation of victims of forced or compulsory labour. D.3d HKSARG policy irrelevant to construction of BOR4 107. The respondents’ evidence as to the HKSARG’s commitment, as a matter of policy, to combat human trafficking may demonstrate that there is no policy objection to the enactment of a bespoke offence to criminalise forced or compulsory labour. However, for the reasons stated in Section C.4f above, that policy is not relevant to the construction of BOR4. D.3e Analogy with torture inapt 108. Mr Husain sought to draw an analogy with torture to support his contention that specific offences were required to criminalise slavery, servitude and forced or compulsory labour.[103] In Hong Kong, torture is the subject of the Crimes (Torture) Ordinance creating a specific offence of torture.[104] The analogy is not, however, apt. The Crimes (Torture) Ordinance was necessary to ensure compliance with the HKSARG’s obligation under the UN Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment (“the CAT”),[105] rather than to secure compliance with BOR3. Torture, unlike other forms of physical assault and abuse addressed by the general criminal law, is a separate type of pain and suffering specifically defined in Article 1 of the CAT and, under the Crimes (Torture) Ordinance, consists of the intentional infliction of severe pain or suffering by a public official or by another person acting at the instigation of a public official or with his consent or acquiescence. No general offence already existing under the criminal law of Hong Kong would criminalise that activity and so a specific ordinance was necessary. D.4 The measures taken by the HKSARG to protect BOR4 rights 109. The respondents’ evidence before the Judge as to the measures taken to combat human trafficking, exploitation and forced labour was principally set out in the affirmation of Leung Wendy, the Administrative Assistant to the Secretary for Security.[106] Her affirmation described the efforts of the HKSARG to address issues arising from human trafficking and forced labour by reference to inter-departmental collaboration, enforcement and prosecution, victim identification, victim protection and prevention and international cooperation. 110. It is unnecessary to summarise Ms Leung’s evidence at length. However, in respect of enforcement and prosecution, whilst acknowledging that Hong Kong “does not have a single unified piece of human trafficking or forced labour legislation”,[107] Ms Leung provided a list of the statutory offences that could be invoked to combat offences in relation to human trafficking and forced labour. They are directed to “a wide range of conduct commonly found in human trafficking and forced labour cases, including physical abuse, false imprisonment, criminal intimidation, unlawful custody of personal valuables, child abduction, child pornography, various trafficking activities for the purposes of prostitution and rape or other sexual offences”[108] and include provisions of: the Crimes Ordinance,[109] the Immigration Ordinance,[110] the Protection of Children and Juveniles Ordinance,[111] the Offences Against the Person Ordinance,[112] the Prevention of Child Pornography Ordinance,[113] the Employment Ordinance,[114] the Employment of Children Regulations,[115] and the Human Organ Transplant Ordinance.[116] 111. Ms Leung’s evidence also addressed the steps to improve the HKSARG’s capability to enforce offences related to human trafficking and forced labour. This includes the inter-departmental Joint Investigation Team of the Hong Kong Police Force, set up in 1998, to focus on human smuggling and whose terms of reference were expanded in 2014 to cover forced labour.[117] She also referred to the Prosecution Code published by the Prosecutions Division of the Department of Justice in September 2013, in which a new paragraph on “Human Exploitation Cases” was added in order to provide guidelines for prosecutors to assist them in identifying cases of human exploitation.[118] 112. Before the hearing in the Court of Appeal, the respondents were given leave to file further evidence to address steps taken by the HKSARG in tackling issues concerning human trafficking and forced labour since the hearing before Zervos J. In the affidavit of Mr Ng Hoi Ka,[119] Assistant Secretary for Security, it was explained that in March 2018 the HKSARG had established a high-level Steering Committee, chaired by the Chief Secretary for Administration, which had promulgated and launched, on 21 March 2018, an “Action Plan to Tackle Trafficking in Persons and to Enhance Protection of Foreign Domestic Helpers in Hong Kong”.[120] 113. Mr Ng also provided an updated list of statutory provisions that could be invoked to combat offences in relation to human trafficking or forced labour. Mr Ng explained that: in December 2016, the Security Bureau issued a “Guideline on Inter-departmental Cooperation for the Handling of Suspected Cases of Trafficking in Persons”;[121] in February 2018, the Employment Ordinance was amended to afford enhanced protection for foreign domestic helpers by increasing the maximum penalties for the offences of overcharging commission and for operating an unlicensed employment agency;[122] in 2015, the Immigration Department introduced a victim screening mechanism to identify victims of human trafficking.[123] 114. Given the wide margin of discretion that is to be afforded to the HKSARG in respect of its positive obligations under BOR4, the approach described in the respondents’ evidence is an acceptable method of compliance providing it affords practical and effective protection of the rights under BOR4. D.5 Absence of bespoke offence not shown to be cause of breach of appellant’s BOR4 rights 115. Having concluded, in answer to Question 1, that the scope of the protection in BOR4 does not extend to human trafficking for forced or compulsory labour, the approach and analysis of Zervos J on the nature of the obligation to protect potential victims under BOR4 can be disregarded. Instead, the question for this Court is whether it is shown that the breach of the appellant’s BOR4 rights was caused by the absence of a bespoke offence criminalising forced or compulsory labour. 116. Here, it is necessary to keep in focus the finding as to breach of BOR4 in this case. The Court of Appeal upheld the Judge’s finding that the HKSARG had failed in its investigative duty under BOR4 in relation to the appellant’s complaints in this case.[124] There is no appeal against that finding by the respondents and he will, in due course, receive a remedy for that breach in the further hearing directed by the Judge (see [14] above). However, there is no finding that the breach of the investigative duty under BOR4 was the result of the absence of a specific offence criminalising forced or compulsory labour. On the contrary, Cheung CJHC concluded: “From the evidence presented before the court, it is plain that the breach was due not to the absence of any specific criminal offence as such, but rather the lack of training of the officers of the various government authorities involved regarding article 4 violations, and the total lack of central supervision and coordination in terms of investigating and combating such violations.”[125] 117. There is no proper basis to disturb that finding of the Court of Appeal. Whilst it is true, as the appellant points out,[126] that his former employer has not been prosecuted for the misconduct alleged against him, it does not follow that this demonstrates that the patchwork of offences for which he might be prosecuted are inadequate to provide practical and effective protection against forced or compulsory labour. Given the range of offences of which the employer might have been charged, the failure by the respondents to investigate the appellant’s case could not, realistically, be said to have occurred because of the lack of a specific offence criminalising forced or compulsory labour. 118. Focusing on the obligation to protect against the requirement to perform forced or compulsory labour, it also cannot be said that the patchwork of offences available to the HKSARG is inadequate to provide practical and effective protection of that BOR4 right. The argument to the contrary advanced on behalf of the appellant was one of form over substance. It simply could not be shown that, had there been a bespoke offence prohibiting forced or compulsory labour, the appellant would have been better protected against the gross mistreatment to which he was subjected by his employer. 119. On the other hand, as the European jurisprudence in respect of ECHR4 shows, the fact sensitive nature of the inquiry as to the adequacy of measures taken in discharge of the positive obligation arising under BOR4 requires the HKSARG to be vigilant in ensuring its measures to prevent, identify and investigate cases of forced or compulsory labour, to enforce laws and prosecute offenders under the existing patchwork of criminal offences, and to provide protection for victims are practical and effective. 120. In his conclusion in the Court of Appeal on the issue of whether a specific criminal offence was required, Cheung CJHC said this: “From the latest evidence filed by the government with leave from us, it would appear that further and more sophisticated efforts have been and are being made by the government to strengthen the protection afforded to potential victims of article 4 with the aim to providing them with practical and effective protection. The effectiveness of those measures is doubted by the evidence filed on behalf of the applicant also with leave from us. It remains to be seen whether these or other further efforts by the government (absent specific criminal legislation) are sufficient to provide the requisite practical and effective protection. As I said, regardless of whether the learning curve is steep or not, the government should act fast. It may not take too many more cases to be brought to court before it will come to the ultimate conclusion that the enactment of specific criminal law is the only way out.”[127] 121. I would respectfully agree. It would be wrong for the respondents to rest on the laurels of their success on the appeal in this Court by relaxing the vigilance with which the measures to combat breaches of the rights protected by BOR4 are administered and enforced. Mr Husain was right to point out that human rights protections must not be “theoretical and illusory”.[128] Moreover, the respondents’ treatment of the appellant when he complained to the authorities of his mistreatment does not cast the respondents in a favourable light and was rightly described by Lord Pannick as disgraceful. Practical and effective protection of BOR4 rights should ensure that a case like the present is a rare and isolated event. D.6 Answering Question 2 122. I would therefore answer Question 2 as follows: (1) The HKSARG has a wide margin of discretion in the manner in which it complies with its positive obligations under BOR4 and there is no absolute duty on the HKSARG to maintain an offence specifically criminalising forced or compulsory labour. (2) To comply with its obligations in respect of BOR4, the HKSARG must take steps to afford practical and effective protection of those rights. Whether practical and effective protection has been provided will depend on the facts of any given case. (3) On the facts of this case, it has not been shown that a bespoke offence criminalising forced or compulsory labour is necessary in that the patchwork of offences already in existence failed to afford the appellant sufficient protection. (4) The determination that a bespoke offence is not required does not preclude a different conclusion being reached in a future case, in the event that the HKSARG is shown in future not to afford practical and effective protection of the rights under BOR4 by reason of the absence of such an offence. Nor should this judgment be taken to indicate that a patchwork of offences would necessarily be sufficient to address a prohibition on human trafficking, if the HKSARG were under a constitutional duty to prohibit that activity. 123. The appellant’s failure to establish a breach of his BOR4 rights by reason of the absence of a bespoke offence criminalising forced or compulsory labour is distinct from the question of whether the respondents were in breach of BOR4 in failing adequately to investigate the appellant’s complaints, for which a damages hearing will in due course be fixed in accordance with the Judge’s direction. E. Conclusion and disposition 124. For the reasons set out above, I would dismiss this appeal. 125. I would direct by way of order nisi that the parties bear their own costs of this appeal, with the appellant’s own costs to be taxed in accordance with the Legal Aid Regulations. I would also direct that there be liberty to the parties, if so advised, to file written submissions as to costs within 14 days of the date of the handing down of this judgment, and that, in the absence of such submissions, the order nisi stand as an order absolute without further direction. Mr Justice Chan NPJ: 126. I agree with the judgment of Mr Justice Fok PJ. Madam Justice McLachlin NPJ: 127. I agree with the judgment of Mr Justice Fok PJ. Mr Raza Husain QC and Mr Azan Marwah, instructed by Patricia Ho & Associates, assigned by the Director of Legal Aid, for the Appellant Lord Pannick QC, Mr Stewart Wong SC, Mr Jin Pao SC and Mr John Leung, instructed by the Department of Justice, for the 1st to 4th Respondents [1] The Hong Kong Bill of Rights (“HKBOR”) is set out in section 8 of the Hong Kong Bill of Rights Ordinance (Cap.383) (“HKBORO”). In this judgment, articles in the HKBOR will be referred to by the abbreviation “BOR” followed by the relevant numbered article. [2] The questions of law for which leave to appeal to this Court was granted by the Court of Appeal are set out below in Section A.5. [3] Its full name is the “Protocol to Prevent, Suppress and Punish Trafficking in Persons Especially Women and Children, supplementing the United Nations Convention against Transnational Organized Crime”; Adopted and opened for signature, ratification and accession by General Assembly resolution 55/25 of 15 November 2000. [4] Directive 2011/36/EU. [5] Resolution adopted by the General Assembly on 25 September 2015, A/RES/70/1, [8.7] at p.20. [6] Affidavit of Ng Hoi Ka, Assistant Secretary for Security of the Security Bureau, dated 4 May 2018, at [4]. [7] HCAL 15/2015, Judgment dated 23 December 2016 (“CFI Judgment”). [8] The appellant initially argued his case amounted one of servitude but ultimately put his case on the basis he was a victim of human trafficking for forced labour: CFI Judgment at [8]. [9] CFI Judgment at [10] and [18]. [10] Ibid. at [163]. [11] Ibid. at [260]. [12] Ibid. at [353]. [13] Ibid. at [337] and [343]. [14] Ibid. at [312]. [15] Ibid. at [351]. [16] Ibid. at [355]. [17] Ibid. at [356]. [18] CACV 14/2017, [2018] HKCA 473, Judgment dated 2 August 2018 (“CA Judgment”). [19] CA Judgment at [35]. [20] Ibid. at [132]. [21] Ibid. at [151]. [22] Ibid. at [187]. [23] Ibid. at [196]. [24] Ibid. at [198]. [25] Order of Lam VP, Barma and Poon JJA dated 21 May 2019 (Barma JA sitting in place of Cheung CJHC, who by then had been appointed a Permanent Judge of this Court). [26] Appearing with Mr Azan Marwah. [27] Appearing with Mr Stewart Wong SC, Mr Jin Pao SC and Mr John Leung. [28] See Comilang Milagros Tecson v Director of Immigration (2019) 22 HKCFAR 59 at [24]-[25] and the cases there cited. [29] ECHR4(1) provides: “No one shall be held in slavery or servitude.” [30] (1999) 2 HKCFAR 4 at p.28F. [31] (2001) 4 HKCFAR 211 at pp.223J-224B. [32] [1999] 1 AC 98 at p.108C-G. [33] R (European Roma Rights) v Prague Immigration Officer [2005] 2 AC 1 at [18]-[19]. [34] M v Italy (2013) 57 E.H.R.R. 29 at [147]. [35] Draft International Covenants on Human Rights, Annotation, A/2929, 1 July 1955, pp.91-92 at [17]; UN International Covenant on Civil and Political Rights, Nowak’s CCPR Commentary (3rd Revised Ed.) at pp.224-225, paras. [8]-[9]. [36] Slavery Convention 1926, Article 1. This definition has been accepted by the ECtHR, see Siliadin v France (2006) 43 E.H.R.R. 16 at [122] and M v Italy (2013) 57 E.H.R.R. 29 at [149]; by the High Court of Australia in R v Wei Tang [2008] HCA 39; (2008) 237 CLR 1 at [137]; and by the Court of Appeal of England and Wales in R v K(S) [2013] QB 82 at [39]. [37] Siliadin v France (2006) 43 E.H.R.R. 16 at [124]. [38] CN and V v France, ECtHR Application No.67724/09, 11 October 2012, at [91]. [39] It should be noted, however, that the Introduction to the Explanatory Notes states that “[t]hey do not form part of the Act and have not been endorsed by Parliament” [1] and “[t]hey are not, and are not meant to be, a comprehensive description of the Act” [2]. [40] See, also, UN International Covenant on Civil and Political Rights, Nowak’s CCPR Commentary (3rd Revised Ed.) at p.226, para. [11], expressing the view: “Modern or contemporary forms of slavery consist of human trafficking, forced labour and bonded labour.” [41] (2010) 51 E.H.R.R. 1 at [280]-[281]. [42] See, as an example, HKSAR v Cheung Wai Kwong (2017) 20 HKCFAR 524 at [32]. [43] See the reference to M v Italy (2013) 57 E.H.R.R. 29 at [29(4)] above. [44] See, in this context, R v Wei Tang [2008] HCA 39; (2008) 237 CLR 1 per Gleeson CJ at [29]. [45] See, also, UN International Covenant on Civil and Political Rights, Nowak’s CCPR Commentary (3rd Revised Ed.) at p.223, para. [5], stating: “Its location alongside the prohibition of slavery is justified in that the boundaries between slavery and servitude and other forms of forced or compulsory labour are not hard and fast. However, in light of the exceptions permitted for forced and compulsory labour, a precise delineation between the terms is necessary.” [46] UN International Covenant on Civil and Political Rights, Nowak’s CCPR Commentary (3rd Revised Ed.) at pp.224-225, para. [8]; see, also, [32(1)] above. [47] [2008] HCA 39; (2008) 237 CLR 1 at [32]; see also, per Kirby J at [112]. [48] Article 153(1) of the Basic Law provides: “The application to the Hong Kong Special Administrative Region of international agreements to which the People’s Republic of China is or becomes a party shall be decided by the Central People’s Government, in accordance with the circumstances and needs of the Region, and after seeking the views of the government of the Region.” [49] In contrast, the CPG declared that the Palermo Protocol would apply to the Macau Special Administrative Region. [50] Written Case of the Appellant at [38]. [51] Comilang Milagros Tecson v Director of Immigration (2019) 22 HKCFAR 59 at [74]. [52] [2005] 2 AC 1 at [19]. [53] Article 31(3) provides: “There shall be taken into account, together with the context: … (c) any relevant rules of international law applicable in the relations between the parties.” [54] (2009) 48 E.H.R.R. 54 at [85]-[86]. [55] Prosecutor v Kunarac and Others (2001) IT-96-23/1-T, Trial Chamber ICTY, at [539]-[542]. [56] Prosecutor v Kunarac and Others (2002) IT-96-23/1-A, Appeals Chamber ICTY, at [116]-[119]. [57] (2010) 51 E.H.R.R. 1. [58] Ibid. at [280]-[281] (footnotes omitted). [59] ECHR4 provides that “(1) No one shall be held in slavery or servitude”; and “(2) No one shall be required to perform forced or compulsory labour.” [60] (2010) 51 E.H.R.R. 1. [61] Written Case of the Appellant at [48]. [62] Footnotes in citation omitted. [63] (2002) 5 HKCFAR 381 at [59]. [64] Koon Wing Yee v Insider Dealing Tribunal (2008) 11 HKCFAR 170 at [27]. [65] See the decision of the Privy Council in Lendore and others v Attorney General of Trinidad and Tobago [2017] UKPC 25; [2017] 1 WLR 3369 at [60]-[61]. [66] Koon Wing Yee v Insider Dealing Tribunal (2008) 11 HKCFAR 170 at [28]. [67] (2010) 51 E.H.R.R. 1 at [149]. [68] CETS No.197, 16 May 2005. [69] (2010) 51 E.H.R.R. 1 at [160]. [70] Article 5(3) provides: “Trafficking in human beings is prohibited.” [71] Judgment of 20 October 2016 (Preliminary Objections, Merits, Reparations and Costs). [72] Article 6(1) of the American Convention provides: “No one shall be subject to slavery or to involuntary servitude, which are prohibited in all their forms, as are the slave trade and traffic in women.” [73] Written Case of the Appellant at [33]. [74] Adopted at the Sixty-eighth session of the HRC on 29 March 2000. [75] Koon Wing Yee v Insider Dealing Tribunal (2008) 11 HKCFAR 170 at [101]. [76] Draft International Covenants on Human Rights, Annotation, A/2929, 1 July 1955, pp.91-92 at [17]; UN International Covenant on Civil and Political Rights, Nowak’s CCPR Commentary (3rd Revised Ed.) at p.225, para. [9]. [77] Adopted by the HRC at its 107th session (11-28 March 2013). [78] The Concluding Observations of United Nations Human Rights Treaty Bodies, Michael O’Flaherty, Human Rights Law Review 6:1 (2006), 27-52 at p.36. [79] Affirmation of Leung Wendy, 10 September 2015, at [6] and [41]. [80] Courtauld v Legh (1868-69) LR 4 Ex 126, quoted by Lord Walker of Gestingthorpe in R v Islam [2009] 1 AC 1076 at [23]: see, HKSAR v Wan Thomas (2018) 21 HKCFAR 214 at [27]. [81] Similarly, where the ILO has linked forced or compulsory labour in the FLC with human trafficking, it has done so on the basis of the Palermo Protocol: ILO Report III (Part 1B), 96th Session, 2007, Eradication of Forced Labour at [77], p.41; ILO Report III (Part 1B), 101st Session, 2012, Giving Globalisation a Human Face at [297], p.128. [82] Ng Ka Ling & Others v Director of Immigration (1999) 2 HKCFAR 4 at p.28H. [83] Written Case of the Appellant at [67]. [84] Ibid. at [68]. [85] Article 5(1) provides under the heading “Criminalization”, that: “Each State Party shall adopt such legislative and other measures as may be necessary to establish as criminal offences the conduct set forth in article 3 of this Protocol, when committed intentionally.” Article 5(2) requires each State Party to adopt legislative and other measures to establish as criminal offences, “attempting” and “organizing or directing other persons” to commit, and “participating as an accomplice” in, an offence established in accordance with Article 5(1). [86] Leung Kwok Hung and Others v HKSAR (2005) 8 HKCFAR 229 at [22]; Plattform ‘Ärzte für das Leben’ v Austria (1988) 13 E.H.R.R. 204 at [34]. [87] O’Keeffe v Ireland (2014) 59 E.H.R.R. 15 at [144]. [88] (2013) 56 E.H.R.R. 24 at [78]-[81]. [89] ECtHR Application No.46423/06, 25 June 2009, at [71]. [90] (2014) 59 E.H.R.R. 15 at [149]. [91] (1998) 26 E.H.R.R. 241 at [34]. [92] (2006) 43 E.H.R.R. 16. [93] ECtHR Application No.67724/09, 11 October 2012. [94] (2013) 56 E.H.R.R. 24. [95] (2010) 51 E.H.R.R. 1. [96] (2010) 51 E.H.R.R. 1 at [285] (footnotes omitted). [97] Siliadin v France (2006) 43 E.H.R.R. 16 at [148]; Demir and Baykara v Turkey (2009) 48 E.H.R.R. 54 at [66]; Rantsev v Cyprus and Russia (2010) 51 E.H.R.R. 1 at [275]; CN and V v France, ECtHR Application No.67724/09, 11 October 2012, at [121]; CN v UK (2013) 56 E.H.R.R. 24 at [77]. [98] See the dualist principle referred to at [49] above. [99] ILO Report III (Part 1B), 96th Session, 2007, Eradication of Forced Labour at FN319, pp.73-74. [100] Ibid. at [138], p.74. [101] Written Case of the Appellant at [73] citing Concluding Observations of the HRC on Brazil, CCPR/C/79/Add.66, 24 July 1996 at [31]. [102] Adopted by the HRC at its 107th session (11-28 March 2013). [103] In support of his argument on Question 1 that BOR4 should be construed as including a protection against human trafficking as an anterior process to the prohibited conduct, Mr Husain also sought to draw on the analogy of torture by contending that the protection against torture in Article 3 of the BOR (“BOR3”) is also construed as including a prohibition on refoulement to another country where there is a real risk of torture. However, torture is distinguishable in this respect because the obligation of non-refoulement relates to the HKSARG taking a decision as to how to deal with a person within Hong Kong: see Ubamaka v Secretary for Security (2012) 15 HKCFAR 743 (at Section H). [104] (Cap.427). [105] Article 4 of the Convention provides: “(1) Each State Party shall ensure that all acts of torture are offences under its criminal law. The same shall apply to an attempt to commit torture and to an act by any person which constitutes complicity or participation in torture. (2) Each State Party shall make these offences punishable by appropriate penalties which take into account their grave nature.” [106] Affirmation dated 10 September 2015 (“Leung affirmation”). [107] Leung affirmation at [14]. [108] Ibid. at [13]. [109] (Cap.200). [110] (Cap.115). [111] (Cap.213). [112] (Cap.212). [113] (Cap.579). [114] (Cap.57). [115] (Cap.57B). [116] (Cap.465). [117] Leung affirmation at [16]. [118] Ibid. at [19]-[23]. [119] Affidavit dated 4 May 2018 (“Ng affidavit”). [120] Ng affidavit at [4]. [121] Ibid. at [10]. [122] Ibid. at [11]. [123] Ibid. at [12]. [124] CA Judgment at [196]. [125] Ibid. at [176]. [126] Written Case of the Appellant at [79]. [127] CA Judgment at [188]. [128] Demir and Baykara v Turkey (2009) 48 E.H.R.R. 54 at [66]. 
A. INTRODUCTION 1. These proceedings raise the important legal questions of whether the oath required by the constitution to be taken by all members‑elect of the Legislative Council (“the LegCo”) was validly taken by the four members-elect herein, and whether in law they should be regarded to have declined or neglected to take the oath, and thereby be disqualified from taking up the office of a LegCo member. 2. Mr Law Kwun Chung (“Mr Law”), Mr Leung Kwok Hung (“Mr Leung”), Ms Lau Siu Lai (“Ms Lau”) and Mr Yiu Chung Yim (“Mr Yiu”) were respectively elected in the general election held in September 2016 to be a member of the LegCo. For convenience, I will refer them collectively as “the Defendants”. 3. However, before the Defendants could validly assume the office (“the Office”) of a LegCo member, as members‑elect, they were mandatorily and constitutionally required under Article 104 of the Basic Law[1] (“BL104”) to take the legislative oath (“the LegCo Oath”) to swear to uphold the Basic Law (“the BL”) and to swear allegiance to the Hong Kong Special Administrative Region of the People’s Republic of China. 4. The content and form of the LegCo Oath is prescribed by section 16(d) and Schedule 2 of the Oaths and Declarations Ordinance (Cap 11) (“ODO”), which is as follows: “THE LEGISLATIVE COUNCIL OATH I swear that, being a member of the Legislative Council of the Hong Kong Special Administrative Region of the People’s Republic of China, I will uphold the Basic Law of the Hong Kong Special Administrative Region of the People’s Republic of China, bear allegiance to the Hong Kong Special Administrative Region of the People’s Republic of China and serve the Hong Kong Special Administrative Region conscientiously, dutifully, in full accordance with the law, honestly and with integrity. (name of person making the oath)” (emphasis added) 5. Under section 7 of the ODO, a member-elect could opt to make an affirmation instead of taking an oath for the purpose of the LegCo Oath. The form and content of the affirmation shall be the same as the prescribed LegCo Oath, save that an affirmation shall commence with “I, [name of the person] of [address], solemnly and sincerely affirm” and end with “Affirmed at [date] before [the administrator of the affirmation].” 6. Each of the Defendants proceeded to take the LegCo Oath on 12 October 2016 before the Clerk to the LegCo (“the Clerk”) by way of affirmation. For convenience, in this judgment, I would refer to the making of affirmation by each of the Defendants as taking the LegCo Oath. 7. The Clerk administered the oath-taking by Mr Law, Mr Leung and Ms Lau. But he declined the jurisdiction to administer Mr Yiu’s affirmation as he was of the view that Mr Yiu altered the form of the LegCo Oath. 8. On 18 October 2016, the President of the LegCo (“the President”) made a ruling on the validity of the LegCo Oath taken by, among others, Mr Law, Ms Lau and Mr Yiu. For the present purposes, the ruling (“the President’s Ruling”) is in substance that: (1) Mr Law’s manner in taking the affirmation was consistent with the ODO and thus valid. (2) Ms Lau’s affirmation was invalid in light of the way she purported to take the LegCo Oath. However, the President allowed her to make the affirmation afresh. (3) The Clerk was correct in declining jurisdiction to administer Mr Yiu’s affirmation as he had altered the contents of the LegCo Oath when he purported to take the affirmation. The purported affirmation was inconsistent with the ODO and thus invalid. The President however allowed Mr Yiu to retake the affirmation afresh. 9. Ms Lau and Mr Yiu purported to retake the LegCo Oath by way of affirmation respectively on 2 November and 19 October 2016, which have been treated by the President as valid. 10. By way of these proceedings, the Chief Executive (“the CE”) and the Secretary for Justice (“the SJ”) contend that the way and manner in which each of the Defendants purported to take the LegCo Oath on 12 October 2016 was not in compliance with the legal requirements of BL104 and the ODO and in law amounted to declining or neglecting to take the LegCo Oath when requested to do so. In the premises, as a matter of law, they have therefore been disqualified since 12 October 2016 from assuming or entering on the Office and could not be permitted to retake the oath thereafter. The CE and the SJ (collectively, “the Plaintiffs”) therefore also submit that (a) the Clerk’s decision (“the Clerk’s Decision”) in accepting the oath taken by Mr Leung as valid is wrong in law; and (b) the President’s Ruling in accepting that Mr Law’s affirmation is valid, and in allowing Ms Lau and Mr Yiu to retake the LegCo Oath is similarly wrong in law. 11. By way of the judicial reviews, the Plaintiffs seek (a) an order to quash the President’s Ruling and the Clerk’s Decision; (b) declarations that the LegCo Oath purportedly taken by each of the Defendants is invalid, and that the President had no power to administer the LegCo Oath purportedly retaken by Ms Lau and Mr Yiu on 19 October 2016; and (c) declarations that the Defendants have since been disqualified from assuming or entering on the Office. 12. By way of the respective Originating Summonses, in addition to the invalidity declarations, the SJ seeks additionally declarations that (a) each of the Defendants has been disqualified from assuming and entering on the Office or has vacated the same, and the Office purportedly held by each of the Defendants is now vacant; and (b) each of the Defendants is not entitled to act or claim to act as a member of the LegCo. The SJ also asks for an injunction to restrain each of them from acting or claiming to act as a LegCo member. 13. The Defendants oppose these proceedings. They effectively say that what they did in taking the LegCo Oath on 12 October 2016 was legally compliant. Alternatively, even if what they did was not in strict compliance with the legal requirements, they did not in law decline or neglect to take the LegCo Oath on 12 October 2016. They were therefore entitled to be allowed to retake the oath afresh. 14. In light of the above contentions, it is obviously necessary for the court to first identify the legal principles that govern the taking of the LegCo Oath. B. THE LAW GOVERNING THE TAKING OF THE LEGCO OATH 15. The law governing the taking of the LegCo Oath as prescribed under BL104 and the relevant provisions of the ODO has recently been reviewed by the Court of Appeal in CE & SJ v President of LegCo (CACV 224 - 227/2016, 30 November 2016, per Cheung CJHC, Lam VP and Poon JA) (“the CA Judgment”), upholding this court’s first instance judgment (HCAL 185/2016 & HCMP 2819/2016, 15 November 2016) (“the CFI Judgment”), and in the Court of Appeal’s judgment in refusing leave to appeal to the Court of Final Appeal (CACV 224 ‑ 227/2016, 16 January 2017, per Cheung CJHC, Lam VP and Poon JA) (“the CA Leave Judgment”). 16. Before I set out the relevant legal principles governing the taking of the LegCo Oath, it is useful to first remind ourselves the relevant constitutional and statutory provisions pertinent to the requirements for taking the oath. They are as follows. B1. BL104 and the Interpretation 17. First and foremost, the starting point must be BL104 which provides the constitutional requirement that, among others, a LegCo member-elect must take the LegCo Oath before he or she could assume the Office. 18. BL 104 provides as follows: “When assuming office, the Chief Executive, principal officials, members of the Executive Council and of the Legislative Council, judges of the courts at all levels and other members of the judiciary in the Hong Kong Special Administrative Region must, in accordance with law, swear to uphold the Basic Law of the Hong Kong Special Administrative Region of the People’s Republic of China and swear allegiance to the Hong Kong Special Administrative Region of the People’s Republic of China.” 19. Further, the meaning of the BL104 has also been set out at paragraphs 1 to 3 of the Interpretation (“the Interpretation”) issued by the Standing Committee of the National People’s Congress on 7 November 2016. They provide as follows: “1. ‘To uphold the Basic Law of the Hong Kong Special Administrative Region of the People’s Republic of China’ and to bear ‘allegiance to the Hong Kong Special Administrative Region of the People’s Republic of China’ as stipulated in Article 104 of the Basic Law of the Hong Kong Special Administrative Region of the People’s Republic of China, are not only the legal content which must be included in the oath prescribed by the Article, but also the legal requirements and preconditions for standing for election in respect of or taking up the public office specified in the Article. 2. The provisions in Article 104 of the Basic Law of the Hong Kong Special Administrative Region of the People’s Republic of China that ‘When assuming office’, the relevant public officers ‘must, in accordance with law, swear’ bear the following meaning: (1) Oath taking is the legal prerequisite and required procedure for public officers specified in the Article to assume office. No public office shall be assumed, no corresponding powers and functions shall be exercised, and no corresponding entitlements shall be enjoyed by anyone who fails to lawfully and validly take the oath or who declines to take the oath. (2) Oath taking must comply with the legal requirements in respect of its form and content. An oath taker must take the oath sincerely and solemnly, and must accurately, completely and solemnly read out the oath prescribed by law, the content of which includes ‘will uphold the Basic Law of the Hong Kong Special Administrative Region of the People’s Republic of China, bear allegiance to the Hong Kong Special Administrative Region of the People’s Republic of China’. (3) An oath taker is disqualified forthwith from assuming the public office specified in the Article if he or she declines to take the oath. An oath taker who intentionally reads out words which do not accord with the wording of the oath prescribed by law, or takes the oath in a manner which is not sincere or not solemn, shall be treated as declining to take the oath. The oath so taken is invalid and the oath taker is disqualified forthwith from assuming the public office specified in the Article. (4) The oath must be taken before the person authorized by law to administer the oath. The person administering the oath has the duty to ensure that the oath is taken in a lawful manner. He or she shall determine that an oath taken in compliance with this Interpretation and the requirements under the laws of the Hong Kong Special Administrative Region is valid, and that an oath which is not taken in compliance with this Interpretation and the requirements under the laws of the Hong Kong Special Administrative Region is invalid. If the oath taken is determined as invalid, no arrangement shall be made for retaking the oath. 3. The taking of the oath stipulated by Article 104 of the Basic Law of the Hong Kong Special Administrative Region of the People’s Republic of China is a legal pledge made by the public officers specified in the Article to the People’s Republic of China and its Hong Kong Special Administrative Region, and is legally binding. The oath taker must sincerely believe in and strictly abide by the relevant oath prescribed by law. An oath taker who makes a false oath, or, who, after taking the oath, engages in conduct in breach of the oath, shall bear legal responsibility in accordance with law.” (emphasis added) 20. The Interpretation as to the true and proper meaning of BL104 is binding on all the courts in Hong Kong. See: Director of Immigration v Chong Fong Yuen (2001) 4 HKCFAR 211 at 222G-H, per Li CJ; the CA Judgment, paragraphs 8, 29, 53 - 58; the CFI Judgment, paragraph 20. 21. In this respect, Mr Lee SC (for Mr Leung) seeks to contend in these proceedings that the Interpretation is not a true and proper interpretation of BL104 as envisaged under Article 158 of the BL, but amounts to an amendment of BL104. As such, it does not and cannot operate retrospectively. 22. However, the Court of Appeal in the CA Judgment at paragraphs 53 - 59, in rejecting a similar argument, has concluded that the Interpretation provides the true and proper meaning of BL104 and takes effect from 1 July 1997. This conclusion is binding on this court. This point is therefore simply not open to Mr Lee to argue.[2] B2. The relevant provisions of the ODO 23. The relevant provisions of the ODO are these. 24. Sections 16 and 19 of the ODO provide as follows: “Section: 16 Forms of Oaths The Oaths referred to in this Ordinance as- … (d) the Legislative Council Oath; … shall be in the respective forms set out in Schedule 2. … Section: 19 Oath of Legislative Councillors A member of the Legislative Council shall, as soon as possible after the commencement of his term of office, take the Legislative Council Oath which- (a) if taken at the first sitting of the session of the Legislative Council immediately after a general election of all members of the Council and before the election of the President of the Council, shall be administered by the Clerk to the Council; (b) if taken at any other sitting of the Council, shall be administered by the President of the Council or any member acting in his place.” (emphasis added) 25. Section 21 of the ODO further provides for the legal consequence of non-compliance by a person who is requested to take an oath as required under the ODO as follows: “Section: 21 Consequence of non-compliance Any person who declines or neglects to take an oath duly requested which he is required to take by this Part, shall- (a) if he has already entered on his office, vacate it, and (b) if he has not entered on his office, be disqualified from entering on it.” (emphasis added) B3. The legal requirements for taking the LegCo Oath 26. Upon considering the plain words of BL104, the Interpretation, the relevant provisions in the ODO, the CA Judgment, the CA Leave Judgment, and the CFI Judgment, together with the authorities cited in those judgments, the court has identified the following legal principles which govern the taking of the LegCo Oath as prescribed by BL104. 27. First, it is a constitutional and mandatory requirement that a member elect of the LegCo must properly and validly take the LegCo Oath both in form and in substance as required by the law before he could assume the Office. In other words, taking the LegCo Oath in accordance with the law is a prerequisite and precondition to the assumption of the Office. See: BL104, the Interpretation, paragraph 2(1) and the CA Judgment, paragraph 27. 28. Second, taking the LegCo Oath in form and in substance means, in law, that the oath taker must: (1) take the LegCo Oath in exactly the same form and content as prescribed under Schedule 2 of the ODO (“the Exact Form and Content Requirement”); (2) do it solemnly and sincerely (“the Solemnity Requirement”); and (3) sincerely believe in and strictly abide by the pledges in the oath at the time of taking the oath (“the Substantive Belief Requirement”). See: the Interpretation, paragraphs 2(2), (3) and 3; the CA Judgment, paragraph 27. 29. Third, under the Exact Form and Content Requirement, the oath taker must accurately and completely read out the oath as prescribed. Further, as set out expressly under paragraph 2(3) of the Interpretation, an oath taker who “intentionally reads out words which do not accord with the wording of the oath prescribed by law, …, shall be treated as declining to take the oath. The oath so taken is invalid and the oath taker is disqualified forthwith from assuming the public office specified in [BL104]” (emphasis added). The mischief of “reading out words which do not accord with the wording of the prescribed oath” provided in this paragraph as a matter of plain meaning must cover not only the failure to read out words of the prescribed form of the oath, but also the adding of words or worded messages to the taking of the oath. In the premises, if the oath taker fails to read out the prescribed oath completely and accurately, or seeks to add to the oath other words or worded messages, this would be regarded in law as altering the form and content of the prescribed oath and thus in breach of the Exact Form and Content Requirement. See also the Interpretation, paragraphs 2(1) and (3). 30. This is also consistent with the judgement in Leung Kwok Hung v Legislative Council Secretariat (HCAL 112/2004, 6 October 2004, per Hartmann J) at paragraph 35 - 37, as I will further explain at paragraphs 88 - 91 below. 31. Fourth, under the Solemnity Requirement, the oath taker must take the oath in a such solemn manner. The word “solemn” bears the commonly understood meaning of being dignified and formal.[3] In other words, understood in the context of oath taking, the oath taker has to take the oath in such dignified and formal way and manner which commensurate and is consistent with the respect that should be accorded to the constitutional importance of the oath taking requirement,and to reflect and underline the very serious and important commitment of the oath taker to bind himself or herself to bear true allegiance to the Hong Kong Special Administrative Region of the People’s Republic of China and to uphold the BL. See: the Interpretation, paragraphs 2(2) and (3); the CA Judgment, paragraph 26 and 27, per Cheung CJHC, paragraph 72, per Lam VP; the CFI Judgment, paragraphs 31 to 33; and AG v Bradlaugh (1885) 14 QBD 667 at 685, per Brett MR. 32. Fifth, under the Substantive Belief Requirement, the oath taker in taking the oath must at the time of the oath also faithfully and genuinely commit and bind himself or herself to uphold and abide by the obligations set out in the LegCo Oath. See: the Interpretation, paragraph 3; the CA Judgment, paragraph 26 and 27, per Cheung CJHC, paragraph 72, per Lam VP; the CFI Judgment, paragraphs 31 to 33. 33. Sixth, an oath taker who seeks to alter the form, manner or substance of the oath when taking it will offend BL104 and be unlawful and of no effect. See: the Interpretation, paragraph 2(2); the CFI Judgment, paragraph 31, adopting Leung Kwok Hung v Legislative Council Secretariat, supra, at paragraphs 36 - 40. 34. An oath taker would be in law disqualified from assuming or entering on the Office or must vacate the Office if he or she declines or neglects to take the LegCo Oath when requested to do so. In this respect, the oath taker shall be regarded in law to have declined or neglected to take the LegCo Oath if he commits any intentional acts or conducts, which are found not to be compliant with the oath taking legal requirements. See: the Interpretation, paragraphs 2(3) and (4); section 21 of the ODO; the CA Judgment, paragraph 43; the CFI Judgment, paragraphs 34 - 35 and 94 - 100. 35. Seventh, the court is the final arbiter in determining whether an oath taker has declined or omitted to take the LegCo Oath in failing to comply with the legal requirements. See: the CA Judgment, paragraphs 32 ‑ 33. 36. Eighth, the court adopts an objective test in determining this question. In other words, the court would determine whether the manner and way in which an oath taker takes the LegCo Oath when assessed and viewed objectively is compliant with the legal requirements and whether the said manner and way is carried out intentionally or wilfully. See: the CA Judgment, paragraphs 5, 27 and 41; the CFI Judgment, paragraphs 33, 35(1), 38 ‑ 40, 42, 45 and 46. 37. In this respect, given the objective assessment, the court would look at the conducts, manner and words adopted by an oath taker in taking the LegCo Oath with a view to deciding what meaning those conducts, manner and words convey to a reasonable person, and whether he or she intentionally acted in such a way. In this objective exercise, the court is not concerned with the subjective meaning of the conducts, manner and words adopted by the oath taker and the oath taker’s subjective intention or thinking in so doing is irrelevant. Cf: Ming Shiu Chung v Ming Shiu Sum (2006) 9 HKCFAR 334 at paragraph 72, per Ribeiro PJ. 38. The above principles are in my view clear in light of the authorities and the constitutional and statutory provisions. However, in support of their opposition in these proceedings, the Defendants’ respective leading counsel have advanced a number of submissions seeking to challenge or qualify the above legal principles. It is convenient for me to deal with these general legal submissions at this stage. 39. First, Mr Lee SC submits that both BL104 and sections 16 and 19 of the ODO have not prescribed expressly for the manner in which the oath should be taken. More importantly, says Mr Lee, these provisions have not provided expressly for the requirement that the oath should be taken “solemnly”. In the premises, Mr Lee argues that as a matter of law, there is simply no requirement that the LegCo Oath must be taken in a solemn and sincere manner, and an oath taker can adopt whatever manner and form he finds most appropriate to express his intention to bind his conscience with the obligations in the oath that he is to take. In further support of this contention, Mr Lee also submits that the modern attitude towards the taking of oaths and affirmation is to regard it as mere ritualistic formality. As such, coupled with the lack of any statutory provisions prescribing the manner for taking the oath, there could not be any legal requirement for taking the oath in a solemn manner. 40. There is nothing in this argument: (1) As set out above, paragraph 2 of the Interpretation has clearly provided that the prescribed oath must be taken solemnly and sincerely. It is therefore a constitutional and legal requirement that the oath taker must take the oath in a solemn and sincere manner. (2) In any event, as observed by this court in the CFI Judgment at paragraphs 31 and 32, as a matter of common law, given its importance, an oath must be taken solemnly and sincerely. The Court of Appeal in the CA Judgment at paragraph 27 also confirms that the prescribed oath must be taken solemnly and sincerely. 41. It is therefore clear and unarguable that a member-elect of the LegCo must take the LegCo Oath in a solemn and sincere manner as a matter of law. 42. Second, Mr Lee further submits that, even if the law requires the oath taker to take the oath solemnly, what manner can or cannot be regarded as “solemn” in a given circumstance has no clear and absolute definition. There must be a range of acceptable conducts and behavior which could be regarded as sufficiently solemn in taking the LegCo Oath. In the premises, Mr Lee says the court in assessing whether a member-elect has taken the LegCo Oath “solemnly” as required by the law, the court should have regard to LegCo’s accepted various rules and practices (collectively, “LegCo Practices”) relating to members’ attire and display of objects in the LegCo.[4] Premised on this, Mr Lee further says that, as a matter of principle, as long as the member-elect takes the LegCo Oath in a manner that accords with the LegCo Practices, the requirement of solemnity is fulfilled. 43. I am unable to accept these submissions. 44. When an act is required to be carried out solemnly, the degree of the requisite solemnity may vary according to the occasion and purpose of which the concerned act is to be carried out. For the present purposes, the requisite degree of solemnity should be measured against the constitutional importance of taking the LegCo Oath and the utmost seriousness of the obligations pledged under the oath. The oath is to swear allegiance to the Hong Kong Special Administrative Region of the People’s Republic of China, and to swear to uphold the BL. These are pledges of utmost constitutional importance and significance, which must be undertaken by the member-elect before he or she could assume the Office. In the premises, the solemnness required of must reflect and commensurate with the high constitutional importance of the occasion and the serious purpose and promise of the oath taking. 45. On the other hand, the LegCo Practices relate to members’ conducts in LegCo’s proceedings and debates in exercising their rights and duties as LegCo members. These are very different matters in nature and purposes when compared with the oath taking occasion as observed above. It is also pertinent to note that a member can only exercise those rights and powers as a member of LegCo after he or she has validly taken the LegCo Oath. Thus, as a matter of general proposition, one cannot necessarily and simply equate what can be regarded as acceptable conducts for a member in making speeches in a LegCo debate with the degree of solemnity required of for the constitutionally important and serious occasion of oath taking. 46. I would therefore reject as a matter of principle the contention that as long as the member-elect takes the LegCo Oath in a way and manner that accords with the LegCo Practices, the legal requirement of solemnity is fulfilled. Whether the concerned oath taking manner or conduct of a member-elect meets the solemnity requirement must be determined and assessed objectively in the individual circumstances of each case. 47. Related to the above contention is also the submissions made by Mr Lee SC and Mr Chan SC (for Ms Lau) that in determining whether a LegCo member-elect’s oath-taking accords with the requisite solemnity, the court should give a wide margin of deference to the relevant President’s decision. Mr Lee’s reasons for the submission are the same as his above submissions, that is that the LegCo’s Practices should constitute the benchmark for measuring the requisite solemnity. I have already rejected the contention above. 48. Mr Chan makes a separate submission in support of this argument. Senior counsel says, given the historical development of the taking of oath to swear allegiance in England, the taking of oaths is closely related to political suppression against non-Christians. Historically, it always served a political function and is always coercive in nature. The content of the oath, and its consequential pledge or allegiance, could change according to the prevailing political climate. He therefore submits that taking an oath is a political act and the purpose of an oath is not about what one believes. The purpose of the oath lies in the act of taking the oath itself, and it is an act to accept the consequences of a breach of the pledge. It is merely a ritual and a formality, imposed by law, the completion of which formality being a means to enforce what is prescribed in the pledge. 49. As such, Mr Chan submits that the determination of whether there is “decline” or “neglect” by an oath taker to take the LegCo Oath is a highly political decision, and in light of its political context which the court does not have the expertise to judge, the court should afford “great weight” to the decision of the oath administrator in determining whether there is “decline” or “neglect” to take an oath. Although as a matter of law, the court is the final arbiter of what constitutes “decline” or “neglect”, in applying the law, Mr Chan says the court has to duly give such great weight to the decision of the politicians. 50. Mr Dykes SC (for Mr Law) has also addressed the court in his skeleton in detail the history of the development of oath taking in England. Based on that, leading counsel similarly submits that the oath taking requirement in the Hong Kong Special Administrative Region is and should be regarded as a mere formality, and as along as the oath taker has complied with the form by reading out all the words of the oath, irrespective of his or her belief in the oath that has been taken, he or she has validly taken the oath. 51. With respect, I am unable to accept these submissions. 52. Paragraph 3 of the Interpretation provides expressly that “[t]he oath taker must sincerely believe in and strictly abide by the relevant oath prescribed by law. An oath taker who makes a false oath, or, who, after taking the oath, engages in conduct in breach of the oath, shall bear legal responsibility in accordance with the law” (emphasis added). The italicized words show clearly that it is our constitutional requirement that the oath taker when taking oath must sincerely and truly believe in it. I therefore cannot accept that the taking of the oath as required in our law under BL104 is only to fulfil a ritual or formality as contended. It is also not only to provide a legal basis to check and punish future breaches by the oath taker of the promise as effectively submitted by Mr Chan and Mr Dykes. It is a constitutional legal requirement that the oath taker in taking the oath must also sincerely and truly believe in the pledges under the oath that he or she is taking. In the premises, in determining whether an oath taker has validly taken the oath, the court is entitled to and should look into, among others, the question (when challenged) as to whether the oath taker has manifested objectively that he or she sincerely and truly believes in the oath at the time when he or she takes it. 53. This is in any event also consistent with the common law position that the law requires the oath taker to genuinely believe in the pledges of the oath as this court has observed at paragraphs 30 - 32 in the CFI Judgment with reference to the common law authorities therein. 54. Further, I also reject Mr Chan’s submission that the determination of whether an oath taker has “declined” or “neglected” to take the oath is a “highly political” decision, and hence the court should invariably give “great weight” to the oath administrator’s view and reasons on this. 55. The question as to what should be the appropriate form and contents of the oath and the allegiances enshrined therein that a State and Parliament seek to impose upon the high public office bearers through the constitution may well be a political one. But once a decision has been made by the State and Parliament on that, and the chosen form and content of the requisite oath have been enacted in the constitution and the relevant statutes to become part of the law of the land, it is solely a question of law whether an oath taken is compliant with the requirements prescribed by the constitution and the relevant statutes. The question of legal compliance is and should not bea political one. 56. In the present context, the issues as to whether a member‑elect’s oath taking is compliant with the legal requirements under BL104 (read together with the Interpretation) and the relevant provision of the ODO (underlying and reflecting the constitutional requirements) are strictly legal questions that the court has to determine. Politics or political arguments do not and should not feature at all in this determination. In this respect, it is pertinent to note that Lam VP has aptly warned against the involvement of politics or political arguments in this judicial process at paragraph 68 of the CA Judgment as follows: “68. As I have said, in adjudicating on these issues the courts only concern themselves with legal questions, and the issues before us in these appeals are whether the requirement in article 104 has been complied with and if not, what are the consequences. In our deliberations, we address legal arguments advanced before us and apply the law (including the Interpretation, which upon its pronouncement becomes part of our laws) strictly as we find them. It is for this very reason that the Chief Judge had to stop Mr Yu when counsel at one stage unwittingly treaded beyond the proper scope of legal arguments by quoting from Socrates on abuse of democracy. It is important that we keep politics out of the judicial process.” (emphasis added) 57. It is therefore incorrect to say that the court in determining whether there is “decline” or “neglect” by a member-elect to take the oath in failing to comply with the legal requirements must in general give a wide margin of deference and great weight to the President or the Clerk’s decision in this respect. The extent of the relevance, where appropriate, of the oath administrator’s decision to the court’s determination of the question of the strict legal compliance of a subject oath taking has already been addressed by the learned Chief Judge in the CA Judgment at paragraphs 39 and 40 as follows: “39. In the final analysis, what is at stake is the compliance of a constitutional requirement of great significance. In any given set of facts, this can admit of one correct answer only. There is no room for a court to simply sit back without correcting an answer given by the oath administrator which the court considers to be wrong, at the expense of the constitutional requirement. What is in issue is squarely a judicial matter which the courts alone are given the judicial power of the Special Administrative Region under the Basic Law to determine. What is involved is not an ordinary judicial review type of situation where the court only conducts a Wednesbury unreasonableness review. Rather, there can be only one right answer when the issue of compliance with the constitutional requirement is raised and nothing short of a full merit review will suffice. The court, according to the Basic Law, is the ordained organ to determine the question. 40. Of course, what I have said above does not prevent at all a court from, when hearing a dispute on the validity of an oath taken or one regarding whether the oath taker has declined or refused to take the oath when duly requested to do so, receiving evidence from the oath administrator on what his views are and the reasons for those views, insofar as they are relevant and admissible, and according them weight accordingly.” (emphasis added) 58. Thus, the administrator’s views and reasons may form part of the evidence (insofar as if they are relevant and admissible) for the court to take into account in determining the question of constitutional compliance of the subject oath taking. If and when they are relevant and admissible in an individual case, as the learned Chief Judge has pointed out, the court would then give such appropriate weight as it thinks fit to those views and reasons. What appropriate weight should be given would be dependent upon the nature and basis of that view and reasons, if any, and must be determined on a case by case basis. However, it is incorrect to say that, as a matter of principle or by default, the administrator’s views and reasons must be given “great weight” by the court as Mr Chan submits. This is particularly so as the learned Chief Judge has emphasized that when a question of an oath taking’s constitutional compliance is involved, there can only be “one correct answer” and the court has to be engaged in a full merits review instead of conducting a Wednesbury unreasonableness review. 59. Third, Mr Lee contends that the requirements on oath taking constitute restrictions on the member-elect’s constitutional rights (a) to stand for election, take part in the conduct of public affairs under Articles 26 and 39(2) of the BL (which provides the constitutional basis to incorporate Article 25 of the International Covenant on Civil and Political Rights (“ICCPR”) through Article 21 of the Hong Kong Bill of Rights (“HKBOR”)); and (b) of freedom of opinion and expression under Article 27 of the BL, and Article 16 of the HKBOR (incorporating Article 19 of the ICCPR). For convenience, I would refer these rights collectively as “the Purported Rights”. 60. As provided under Article 39(2) of the BL, these rights shall not be restricted unless “prescribed by law”. Mr Lee argues that the present requirements of oath taking are “so vague and uncertain” that they fail to meet the “prescribed by law” requirement. 61. Further, Mr Lee says the restrictions in any event cannot be justified under the four elements of the test of proportionality as now laid down in Hysan Development Co Ltd v Town Planning Board [2016] 6 HKC 58 (CFA) at paragraph 54. 62. In other words, Mr Lee contends that the oath taking requirement is unconstitutional since, as a restriction of the Purported Rights, it is not “prescribed by law” and in any event fails to meet the proportionality test. 63. With the greatest respect to Mr Lee, I am unable to accept these startling submissions. 64. Mr Lee’s submissions, put to its logical conclusion, amount to saying that the court could declare BL104, which itself is a provision of our constitution, to be “unconstitutional” if it does not meet the “prescribed by law” and proportionality requirements. 65. BL104 by itself is part of the constitution. There is no question that it can be said to be unconstitutional. This startling proposition that a court can declare BL104, a constitutional provision itself, to be “unconstitutional” simply cannot stand. 66. In any event, the court should seek to construe all the constitutional provisions within the BL to be consistent with each other. In this respect, it is clear as a matter of construction that, the Purported Rights, which are expressly provided not to be absolute, shall be read to be subject to BL104. 67. I have no hesitation to reject Mr Lee’s above submissions. 68. Fourth, Mr Lee argues that since a member-elect is in effect “penalised” by being prevented from taking up the Office if it is found that he has declined or neglected to take the LegCo Oath, the court must apply the criminal standard of proof of beyond reasonable doubt to find a member-elect has so declined or neglected to take the LegCo Oath. This is particular so as, says Mr Lee, the “penalty” is a very serious one when considered under the context that he or she is elected by the electorate to represent them at the LegCo. 69. I am not persuaded by these submissions. 70. As held by the Court of Appeal in the CA Judgment, taking the LegCo Oath is a constitutional prerequisite or precondition for a member-elect to assume the Office as provided under BL104. Further, it is clear law that a member-elect could not be so qualified to take up the Office if he or she declines or neglects to take the LegCo Oath as provided under the Interpretation and section 21 of the ODO. It is thus clear that the valid assumption of the Office after a successful election is still subject to validly taking the LegCo Oath. Persons running for the election must be taken to know this precondition in law. In the premises, it is incorrect to characterise, as Mr Lee seeks to do, the consequence that a member‑elect would not be qualified to assume the Office if it is found that he has declined or neglected to take the LegCo Oath as a “penalty”. It is not. 71. I would therefore reject the contention that the court should apply a beyond reasonable doubt standard of proof in determining on the evidence as to whether any of the Defendants declined or neglected to take the LegCo Oath in the circumstances of the present cases. 72. The court would adopt the civil standard of proof of balance of probabilities in determining whether the Defendants declined or neglected to take the LegCo. However, in doing so, I accept the submissions by the leading counsel for Mr Law, Ms Lau and Mr Yiu that, in light of the importance of the question, it requires cogent evidence to find on the balance of probabilities that objectively the Defendants did decline or neglect to take the LegCo Oath. 73. Bearing all the above principles in mind, I would now look at each of the Defendants case in turn to determine whether, viewed objectively, he or she declined or neglected to take the LegCo Oath. C. WHETHER EACH OF THE DEFENDANTS DECLINED OR NEGLECTED TO TAKE THE LEGCO OATH ON 12 OCTOBER 2016 C1. The undisputed underlying facts before 12 October 2016 74. It is not in dispute that the Clerk had reminded each of the Defendants (as with all the other members-elect) to take the LegCo Oath on 12 October 2016 by various circulars as follows. 75. On 20 September 2016, the Clerk issued a circular[5] to all members-elect regarding the taking of the LegCo Oath. The circular reminded members-elect that members of the LegCo must swear to uphold the BL and swear allegiance to the Hong Kong Special Administrative Region of the People’s Republic of China under BL104. It further informed all members‑elect that the first meeting of the Sixth LegCo would be held on 12 October 2016 and that under Rule 12(1) of the LegCo’s Rules of Procedure, members-elect should take the oath at the first meeting of a term of the LegCo. It also reminded members‑elect that under Rule 1 of the Rules of Procedure, no member shall attend or vote therein until he or she has made an oath in accordance with the provisions of the ODO. Members-elect were invited to indicate his or her choice among the three versions of the LegCo Oath, drawn up in accordance with sections 5 and 7 of and Part IV of Schedule 2 to the ODO. 76. On 7 October 2016, the Clerk issued another circular[6] to members, concerning the first meeting of the Sixth LegCo that would be held on 12 October 2016, explaining to members the detailed procedures and arrangements of oath-taking at the start of the meeting, as well as the form and manner of oath-taking. In “the points to note for oath-taking by Members”, enclosed to this circular, members were reminded to take the oath in the form and manner prescribed by the ODO, which include reading out all the words of the LegCo Oath prescribed by law. 77. On 11 October 2016, the Clerk issued the third circular[7] to all members of the LegCo. In this circular, members were once again reminded to attend the 1st LegCo meeting on 12 October 2016 on time and to “take the oath” in accordance with the order as set out in the Agenda for the meeting. It also reiterated the requirement under BL104 and that the LegCo must be in the form set out in Schedule 2 to the ODO, and members must say the words of the oath therein. The Government also published a statement on 11 October 2016 on oath‑taking by members-elect of the LegCo as this is a matter of great constitutional importance. 78. The taking of the LegCo Oath by members of the Sixth LegCo was held on 12 October 2016. The entire oath-taking proceeding on that day was captured as Webcast recording which is available on the LegCo website. C2. The oath taking on 12 October 2016 by Mr Law 79. The objective and undisputed evidence shows that Mr Law took the LegCo Oath in the following way and manner on 12 October 2016. 80. Mr Law made the following statement (“the Opening Statement”) upon being requested by the Clerk to take the LegCo Oath: “誓詞，英文係‘Affirmation’，佢拉丁文原意係使其更堅定更堅強。宣誓就係一個莊嚴嘅儀式，要我地向香港人承諾未來要知行合一，捍衛香港人嘅權利。但今日呢個神聖嘅儀式，已經淪為政權嘅工具，強行令民意代表屈服喺制度同埋極權之下。You can chain me, you can torture me, you can even destroy this body, but you will never imprison my mind. 我今日要完成必要嘅程序，但係唔代表我會屈服喺極權之下。香港市民永遠都係我地服務同埋團結嘅對象，我係絕對唔會效忠於殘殺人民嘅政權，我一定會堅持原則，用良知守護香港。希望在於人民，改變始於抗爭。”[8] 81. Around three seconds later, Mr Law purported to take the LegCo Oath, adopting an apparent and distinct rising tone whenever he spoke the word “國”: “本人羅冠聰，謹以至誠，據實聲明及確認，本人就任中華人民共和國香港特別行政區立法會議員，定當擁護《中華人民共和國香港特別行政區基本法》，效忠中華人民共和國香港特別行政區，盡忠職守，遵守法律，廉潔奉公，為香港特別行政區服務。” (emphasis added) 82. Around one to two seconds later, Mr Law shouted “權力歸於人民，暴政必亡，民主自決，抗爭到底。” (“the Closing Statement”). The Clerk remained silent throughout the purported oath-taking of Mr Law. 83. In the President’s Ruling made on 18 October 2016, the President concluded that the LegCo Oath as taken by Mr Law was valid. The President explained at paragraph 11: “11. Hon Nathan LAW Kwun-chung subscribed an affirmation in Cantonese when taking the LegCo Oath. I notice that Mr LAW used a different tone when pronouncing ‘China’ in his affirmation. Based on the manner that he took his oath, objectively assessed, and taking into account that he read out all the words prescribed by the [ODO] during his oath-taking, I am prepared to accept that the manner in which he subscribed his affirmation was not inconsistent with the [ODO].” 84. The Plaintiffs now submit that, objectively assessed, in purportedly taking the LegCo Oath in the above way and manner, Mr Law did not comply with the legal requirement to faithfully or truthfully commit himself to upholding and abiding the obligations set out in the LegCo Oath. 85. I agree. My reasons are as follows. 86. Objectively understood, Mr Law manifested to a reasonable person at the time of the oath taking that he was only forced by the system and the totalitarian authority to take the LegCo Oath, something which had become a “political tool” used by the political regime: (1) It is important to note that Mr Law made the Opening Statement after he had already been requested by the Clerk to take the LegCo Oath. From that point onwards, he had already commenced the process of taking the LegCo Oath. What therefore followed after that request must be objectively intended to be acts referable to the taking of the LegCo Oath. The court therefore is entitled to look at the Opening Statement in determining whether Mr Law had complied with the legal requirements in taking the LegCo Oath. (2) In pronouncing in “the Opening Statement” that “但今日呢個神聖嘅儀式，已經淪為政權嘅工具，強行令民意代表屈服喺制度同埋極權之下”, Mr Law must be objectively referring to the requirement of taking the LegCo Oath to have already been rendered (“淪為”) as a “political tool” (“政權嘅工具”), as the taking of the LegCo Oath was the very ceremony he was about to go through after making the Opening Statement. Moreover, the plain words “強行令民意代表屈服喺制度同埋極權之下” also conveyed the meaning that the requirement to take the LegCo Oath was something he as an elected representative was “forced” to do under the system and totalitarian authority. (3) Objectively understood, in the context of taking the LegCo Oath, the “system” he referred to could only be understood by a reasonable person to mean the legal requirements provided under BL104 and the ODO in requiring him to take the LegCo Oath. The “authority” he referred to therefore could also only mean either the Hong Kong Special Administrative Region or the People’s Republic of China, or both, since the oath taking requirement is stipulated under the BL which is enacted by the National People’s Congress of the People’s Republic of China, and that he is required to pledge allegiance to the Hong Kong Special Administrative Region of the People’s Republic of China and to swear to uphold the BL. In other words, he regarded the requirement to take the LegCo Oath was only a political tool employed by the People’s Republic of China and the Hong Kong Special Administrative Region of the People’s Republic of China through the legal requirements to force him to make those pledges. (4) When these were followed by the sentences “You can chain me, you can torture me, you can even destroy this body, but you will never imprison my mind 我今日要完成必要嘅程序，但係唔代表我會屈服喺極權之下”, objectively he was also clearly expressing the meaning that he was only to go through the motion in taking the LegCo Oath, and that exercise itself did not represent that he would subjugate himself to the totalitarian authority. (5) When all these are read as a whole together with the LegCo Oath, which he read out following the Opening Statement, it is objectively clear that Mr Law manifested to a reasonable person that he regarded the taking of the LegCo Oath as an involuntary and mechanical exercise which he was forced to go through, and he did not truthfully and sincerely believe in and commit himself to the pledges enshrined in that oath. (6) Moreover, when the adoption of a rising intonation by Mr Law whenever he read only the word “國” in all the three references to the phrase “the Hong Kong Special Administrative Region of the People’s Republic of China” (中華人民共和國香港特別行政區) was understood together with his references to “totalitarian authority” which he would not “subjugate” to in the Opening Statement, Mr Law was also objectively conveying the message that he did not respect or recognise the legitimacy of the People’s Republic of China as the sovereign of the Hong Kong Special Administrative Region. It must be noted that the words “中華人民共和國香港特別行政區” constitutes one complete phrase and meaning in the LegCo Oath, referring to the special constitutional arrangement that Hong Kong is a special administrative region of the People’s Republic of China. In targeting and focusing only on the word “國” in this phrase by adopting a rising intonation, Mr Law was objectively expressing a doubt on or disrespect of the status of the People’s Republic of China as Hong Kong’s legitimate sovereign country. (7) As this court has said in the CFI Judgment at paragraphs 43 ‑ 44, the recognition of the People’s Republic of China as Hong Kong’s sovereign is fundamental to the One Country, Two Systems principle, which in turn is fundamental to the enactment of the BL and the underlying constitutional model in establishing the Hong Kong Special Administrative Region. Refusing to recognise the legitimacy of the People’s Republic of China as the Hong Kong Special Administrative Region’s sovereign is inconsistent and contrary to swearing allegiance to the Hong Kong Special Administrative Region of the People’s Republic of China and to uphold the BL. In the premises, objectively viewed, Mr Law could not and did not manifest a genuine and truthful intention to commit to and abide by these obligations in the LegCo Oath when purportedly taking it. (8) There is also no question that he intended to make the Opening Statement before reading the LegCo Oath, and adopted a rising tone whenever he read the word “國” in the oath. In the premises, for all the above reasons, it is clear to me that Mr Law declined or neglected to take the LegCo Oath in substance as he did not faithfully and sincerely commit to and abide by the obligations provided in the oath. On this basis alone, he should be and has been so disqualified in law from assuming the Office since 12 October 2016. 87. Further, I agree with Mr Mok SC for the Plaintiffs that by making the Opening Statement and the Closing Statement in the way as Mr Law did, he also failed to comply with the Exact Form and Content Requirement for taking the LegCo Oath. I will explain why. 88. As held by Hartmann J in Leung Kwok Hung v the Legislative Council Secretariat, supra, it is a legal requirement that the LegCo member-elect in taking the LegCo Oath must, objectively viewed, comply with the exact form and substance of the LegCo Oath. Hartmann J emphasized that the law “allows only for limited differences as to how a person wishes to take the oath, it does not allow for any real difference in the form – and thereby the substance – of the oath itself.” In this respect, the learned judge also made it clear that altering the form of the LegCo Oath also alters the substance of the oath itself. See paragraph 35. 89. In that case, Mr Leung sought a declaration from the court in an intended judicial review that it was lawful for him to take the LegCo Oath in the following amended form: “I, Leung Kwok-hung, solemnly, sincerely, and truly declare and affirm that I swear by the people of China and the residents of Hong Kong, as well as the principles of democracy, justice, human rights and freedom that, being a member of the Legislative Council of the Hong Kong Special Administrative Region of the People’s Republic of China, I will uphold the Basic Law of the Hong Kong Special Administrative Region etc.” (emphasis added to highlight the proposed amendments to the LegCo Oath) 90. Although in that amended form, Mr Leung would still purport to swear to uphold the BL and to pledge allegiance to the Hong Kong Special Administrative Region of the People’s Republic of China as required under BL104, Hartmann J (in refusing leave) still concluded that taking the oath in the proposed amended form would offend BL104 and therefore be unlawful as it did not comply with the exact form and substance requirement. He explained this at paragraphs 36, 37 and 40 as follows: “36. In my judgment, the amended oath that the applicant wishes to take goes further than merely defining how he wishes to take his oath, it alters the form and thereby the substance of the oath itself. 37. The Ordinance does not permit any person, religious or not, to swear by a list of principles : religious, political, economic, ecological or philosophical. But viewed objectively, in my opinion, the applicant in the present case wishes to swear by such a list of principles. That they may be laudible principles, that they may perhaps be universal truths, does not alter the fact that they are a list of principles and no such list of whatever kind is permitted by the Ordinance. … 40. In summary, I have concluded that the applicant has not been able to demonstrate any form of prima facie arguable case that his intended form of oath may be consistent with art.104 of the Basic Law. In my judgment, it is manifest that the oath or affirmation he seeks to take when he assumes office, will offend art.104 and will therefore be unlawful and of no effect. In the circumstances, I am satisfied that the applicant has not demonstrated a prima facie arguable case for the declaration that he seeks.” (emphasis added) 91. In the premises, as a matter of law, an oath taker would be regarded as having failed to comply with the Exact Form and Content Requirement if he seeks to swear by a list of principles, beliefs or truths (whether religious, political, economic, ecological or philosophical, and whether they are controversial or not) not provided in the prescribed form. By the same token, it must similarly (if not a fortiori) be impermissible in law for an oath taker to incorporate or convey in the oath additional worded messages other than the oath itself, whether or not these messages are religious, political, economic, ecological or philosophical, and whether they are controversial or not. 92. This legal position is also consistent with and supported by paragraph 2 of the Interpretation as I have explained at paragraphs 29 and 30 above. 93. Bearing this in mind, when the fact that Mr Law made the Opening and Closing Statements respectively right before and after his reading of the LegCo Oath is viewed together with the contents of the Statements as a whole, objectively, it must appear to a reasonable person that Mr Law intended to convey the worded messages in the Opening and Closing Statements as part of the oath taking itself. In the premises, objectively, he sought to incorporate and convey additional worded messages in the oath taking. This offends the Exact Form and Content Requirement and therefore also BL104. 94. Mr Dykes SC (leading Mr Tam) have made various submissions to contend that Mr Law had properly complied with the legal requirements in taking the LegCo Oath. I will deal with these arguments in turn. 95. First, Mr Dykes submits that, by reference to the history of the development of oath taking in England, the taking of the LegCo Oath is a mere formality and the court should and could not inquire into the question of whether the oath taker genuinely and faithfully believe in the obligations pledged in the oath. Hence, Mr Law should be regarded as having complied with the legal requirements of taking the LegCo Oath as long as he has accurately and completely read out all the words of the oath as a matter of form, which Mr Dykes submits he did. 96. For the reasons I have explained at paragraphs 52 - 53 above, I reject this contention. Both as a matter of express requirement set out in the Interpretation, and as a matter of common law, it is a legal requirement in Hong Kong that the oath taker must take the oath prescribed under BL104 and the relevant provisions of the ODO both in form and in substance, and that the oath taker must faithfully and sincerely believe in and commit himself to the obligations provided in the oath at the time when taking it. 97. Second, linked with the above contentions, Mr Dykes appears also to submit that, given that the historical purpose of taking an oath of allegiance in England is to avert acts of sedition and subversion, as long as Mr Law is not advocating independence of Hong Kong, there would be no infringement of the allegiance obligations in the LegCo Oath. Objectively the court cannot and should not conclude that he did not faithfully and sincerely believe in and commit to those obligations as Mr Law, says Mr Dykes, does not advocate Hong Kong independence. 98. I must also reject this submission. The law requires, among others, the oath taker to faithfully and sincerely believe in and commit himself to the allegiance obligations in the LegCo Oath. This is a question that would be assessed objectively by looking at all the relevant matters pertaining to it. Someone who advocates and supports independence of Hong Kong would obviously be regarded as not having a genuine and sincere intention to commit himself to those allegiances for the reasons already explained in the CFI and CA Judgments. However, as a matter of general principle, this cannot be the only basis or circumstances where the court could conclude that the oath taker has failed to show objectively the requisite faithful and sincere belief and commitment. It must be open to the court to find the same when appropriate in other circumstances. 99. Third, Mr Law has filed an affirmation in this application seeking to explain what he now said to be the real meaning and intention behind making the Opening and Closing Statements, and in reading out the LegCo Oath in the way as he did. In gist, he explained that: (1) The words and sentences he used in the Opening Statement were not intended to criticize either the Hong Kong Special Administrative Region Government or the People’s Republic of China. He was only expressing his concerns about possibilities in the future. All that he intended to say was that, looking to the future, he was concerned that the oath taking ceremony and the subsequent possibility of an allegation of oath-breaking “could be used” by the authority as a tool to achieve political purpose. Further, when he used the words “我係絕對唔會效忠於殘殺人民嘅政權”, he was only expressing a message that, looking forward to the future (but not something of the present), he would not swear allegiance to a “regime” if she kills her own people. In particular, when he used the words “政權”, he was referring to a regime but not a country. He reiterated that it was important to distinguish between a regime and country, since swearing allegiance was to a country but not a regime. He therefore was not expressing any criticism or disrespect of the People’s Republic of China as a country or that he would not swear allegiance to the Hong Kong Special Administrative Region of the People’s Republic of China. He emphasized in the affirmation that he has “absolutely no intention to violently overthrow the political institutions of the HKSAR and PRC”. He also denied that he “was not sincere and did not genuinely pledge allegiance to the HKSAR or the PRC” as now alleged by the Plaintiffs. See paragraphs 29 - 37 of Mr Law’s affirmation. (2) Further, the adoption of a rising tone in reading out the word “國” was only to show his “love” to the People’s Republic of China as a country, while also at the same time to “criticise the current administration of the Communist Party”. This was, he said, “different from being disrespectful or not bearing allegiance to the Basic Law or ‘One Country, Two Systems’”. He emphasized that as a popularly elected politician, he had “the responsibility to express [his] political views.” There was again therefore no disrespect or refusal to recognise the People’s Republic of China as a country. See paragraphs 38 - 44 of Mr Law’s affirmation. (3) The words used in the Closing Statement were not to advocate independence of Hong Kong. In fact, he emphasized that he and his party (Demosistō) do not advocate or support the independence of Hong Kong. They only advocate the right of self-determination on matters which are within Hong Kong’s own autonomy as prescribed under the One Country, Two Systems principle. See: paragraphs 45 - 55 of Mr Law’s affirmation. 100. Based on these self-professed explanations, Mr Law therefore says he did intend to commit to and abide by the obligations pledged in the LegCo Oath. This is particularly so, Mr Law emphasized, as there were no good reasons why he would intentionally flout the requirements of taking the LegCo Oath and risk giving away the Office after spending substantial effort and expenditures to run for and win the election in the first place. 101. With respect, I will also reject this contention. 102. As the Court of Appeal has emphasized in the CA Judgment, the court adopts an objective assessment of the evidence relevant to the oath taking to determine whether the oath taken is compliant with the legal requirements. As I mentioned above, in adopting the objective assessment, the court is to determine what a reasonable person would objectively understand the meaning from the words, conducts and manner adopted by the oath taker in taking the oath. In that exercise, the court must also apply a degree of common sense. However, in the objective assessment, the court is not concerned with the subjective intention of the oath taker in adopting the subject words, conducts and manner or the subjective meaning the oath taker seeks to accord to those words, conducts and manner. Hence, evidence on such subjective intention, thought process or meaning are irrelevant to the object exercise. 103. Mr Law’s subjective explanations now set out in his affirmation and summarized above are in nature matters relating to his subjective thought process, intention and meaning in adopting the complained conducts in taking the LegCo Oath. As a matter of law, they are not relevant to the court’s objective assessment as to whether the way and manner in which Mr Law purported to take the LegCo Oath complied with the legal requirements. 104. Further and in any event, even if the court was to take into account these explanations, I do not find them to support Mr Law’s contentions that he had complied with the oath taking legal requirements. 105. First of all, these subjective explanations simply cannot be accepted to displace, in the objective assessment, the clear meaning of the words, conducts and manner adopted by Mr Law in taking the LegCo Oath. This is so as they are clearly inconsistent with those objective meanings found by the court above: (1) The explanations are highly convoluted and simply not borne out by the plain words of the Opening Statement. (2) Further, the words used in the Opening Statement do not in any way denote a reference to future possibilities only as now alleged by Mr Law. In particular, the use of the words “今日” twice in the Opening Statement clearly shows that Mr Law was addressing his involuntariness in going through the oath taking exercise on that day, and that expressed involuntariness would not bear any real meaning unless he was also referring to his present criticisms of the system and authority. (3) As such, the authority he was referring to can only be understood to be a reference to the present authority which has imposed the requirements of the oath taking, something which Mr Law expressed in the Opening Statement his involuntariness in going through it. Properly understood, as explained above, this could only be a reference to the Hong Kong Special Administrative Region of the People’s Republic of China, or the People’s Republic of China, or both. (4) Further, the word “國” used in the phrase “中華人民共和國香港特別行政區” in the context of the LegCo Oath can only mean the People’s Republic of China as a country. Hence, it is not open (let alone reasonably open) to an objective person to understand the word when read out could be capable of being dissected (as Mr Law now alleges to be the case in adopting a rising tone only on this word) to have a separate meaning of emphasizing the love for the People’s Republic of China as a country but a criticism of her present administration of the Communist Party. 106. Further and in any event, in the subjective explanations, Mr Law confirms that he seeks to convey other “political messages” in the oath-taking itself. This would clearly violate the Exact Form and Content Requirement in taking the LegCo Oath. 107. For these reasons, I also reject Mr Law’s contentions based on his subjective explanations. 108. Fourth, Mr Law says what he did in making the Opening and Closing Statements, and in adopting a rising tone in reading out the word “國” when taking the oath is consistent with, or no different from in nature with many other past practices adopted by various LegCo members-elect in taking the LegCo Oath in previous LegCo terms. All these previous oath takings had been accepted by the Clerk or the President and had not been challenged by the CE or SJ. In fact, it was only after he had researched and studied all these past practices (and knowing that they had not been declared invalid) that he adopted the manner and way in which he took the LegCo Oath. This, he says, further shows that he never had an intention to flout the legal requirements in taking the LegCo Oath. He says in the premises, even if as a matter of law the court finds him not to have complied with the legal requirements in taking the LegCo Oath, he did not intend to so flout the law, and therefore he did not “decline” or “neglect” to take the LegCo Oath. He should therefore not be disqualified from taking the Office but be permitted to retake the LegCo Oath again. 109. I am also unable to accept these submissions. 110. Given the objective test, in determining whether an oath taker has in law declined or neglected to take the oath, it would be sufficient to find that (a) the oath taker intended to adopt the very words, conducts and manner which are under challenge to take the oath; and (b) the said words, manner and way on an objective construction are found to be not in compliance with the oath taking legal requirements. It is not a necessary element for the court to be satisfied that the oath taker has a specific intent to flout or not to comply with the very legal requirements. Of course, if it can be established on an objective assessment that the oath taker indeed intended not to comply with those specific legal requirements in taking the oath, it would be a fortiori that he or she has declined or neglected to take the oath. 111. In any event, the fact that an oath administrator (ie, the Clerk or the President) on other occasions had not taken issues on certain allegedly similar manner and way in which the LegCo Oath was taken could not support Mr Law’s contentions that what he did in the present case should also similarly be regarded as having complied with the oath taking’s constitutional legal requirements. The court is not bound by any previous conclusions made by the oath administrator. As emphasized by the Court of Appeal in the CA Judgment, the court is the final arbiter on the question of whether an oath taker has properly complied with constitutional requirements for the oath taking and whether he or she has declined or neglected to take the oath. In any event, every case must be decided on its own facts. 112. In the premises, based on the indisputable facts in the present case, the court is satisfied that, as a matter of law, objectively Mr Law declined or neglected to take the LegCo Oath on 12 October 2016 as: (1) It is clear that he intended to adopt the very words, conducts and manner in which he purported to take the LegCo Oath by (a) making the Opening Statement; (b) adopting a rising tone in reading out the word “國” in all three references to the phrase “中華人民共和國香港特別行政區” in the LegCo Oath; and (c) making the Closing Statement. (2) The said words, conducts and manner adopted by Mr Law in purporting to take the LegCo Oath, on an objective construction, do not satisfy the Exact Form and Content Requirement and the Substantive Belief Requirement as explained above. 113. He has therefore been disqualified in law from assuming or entering the Office since 12 October 2016. The President therefore erred in law in his decision in ruling that Mr Law has validly taken the LegCo Oath, and that decision should be quashed. 114. Further, it follows that since 12 October 2016, Mr Law has claimed to be entitled to act as a LegCo member when he is not so entitled. The court should also grant the declarations and injunction under section 73 of the Legislative Council Ordinance (Cap 542) (“LCO”) as sought by the Plaintiffs under HCMP 3379/2016. C3. The oath taking on 12 October 2016 by Mr Leung 115. The objective and undisputed evidence shows that Mr Leung took the LegCo Oath in the following way and manner on 12 October 2016. 116. According to the video footage, after the Clerk called Mr Leung’s name requesting Mr Leung to walk up to the table at the centre of the chamber to take the LegCo Oath in accordance with the predetermined order, Mr Leung, who was wearing a black t-shirt with the words “公民抗命”[9] printed thereon, walked down the hallway to the table, carrying an opened yellow umbrella (with many words written thereon, including “結束一黨專政”[10]) in his right hand and a paper board showing the words “人大831決議”[11] (with a cross on it) in his left hand and shouted in Cantonese “雨傘運動！不屈不撓！公民抗命！無畏無懼！人民自主自決！無須中共批准！我要雙普選！梁振英下台！”.[12] 117. When Mr Leung reached the oath-taking table, he put the paper board on the table, continued holding the yellow umbrella in his right hand and said the following in Cantonese (“the Statement”): “得未？係用中文定英文定上海話？我揀左中文，唔好意思呀。聽住各位，係好莊嚴架。兩年之前人民喺出面，爭取雙普選，俾人拉俾人打。”[13] 118. Mr Leung then paused for around two seconds and started reading words of the LegCo Oath in Cantonese with a much louder voice and in a truncated manner as follows (with the opened yellow umbrella in his right hand): “本人 (pause) 梁國雄 (pause)，謹以至誠 (pause)，據實聲明(pause) 及確認 (pause)，本人 (pause) 就任 (pause) 中華人民(pause) 共和國 (pause) 香港 (pause) 特別 (pause) 行政區(pause) 立法會議員 (pause)，定當擁護 (pause)《中華人民共和國[said hurriedly] (pause) 香港 (pause) 特別行政區(pause) 基本法》(pause)，效忠 (pause) 中華人民 (pause) 共和國 (pause) 香港 (pause) 特別行政區 (pause)，盡忠職守(pause)，遵 (pause) 守法律[in lower voice] (pause)，廉潔奉公 (pause)，為香港 (pause) [in louder voice]特別行政區(pause) 服務。”[14] 119. Around one second after he had read out the words of the LegCo Oath, Mr Leung (a) shouted “撤銷人大831決議！我要雙普選！”;[15] (b) put down the yellow umbrella on the table, opened with the top of the umbrella pointing away from him; (c) tore a piece of paper with the words “人大831決議”[16] (without a cross on it) into pieces while shouting “撤銷人大831決議！撤銷人大831決議！我要雙普選！人民自主自決！無須中共批准！”;[17] (d) threw the pieces of paper away towards the ceiling; and (e) left where he had been standing with his other props. 120. The Clerk did not say anything about Mr Leung’s purported oath-taking immediately following his departure and thereafter. As such, the Clerk effectively regarded the oath purportedly taken by Mr Leung on 12 October 2016 as valid. 121. Mr Mok first submits that the way in which Mr Leung took the LegCo Oath through the means of conduct, words and paraphernalia as described above shows clearly and objectively that he did not take the oath in the requisite solemn and sincere manner. Mr Leung therefore failed to comply with the Solemnity Requirement and the LegCo Oath purportedly taken by him is therefore unlawful. 122. I agree. 123. As I have explained above, under the Solemnity Requirement, as a matter of law, an oath maker must take the LegCo Oath in a solemn and sincere manner which is objectively consistent with and commensurate with the constitutional importance of the oath taking ceremony and procedure, and the seriousness of the pledges of allegiance enshrined in the oath. 124. Moreover, given the fundamental constitutional importance of the requirement to take the LegCo Oath, and the utmost seriousness expected of the oath taker in making the constitutionally required pledges in the oath, the objective solemnness and sincerity required of in the procedure should be such that the oath taker must take the oath in such a formal and dignified manner that demonstrate to a reasonable person the oath taker’s high degree of respect and seriousness given to the oath taking procedure and the oath itself. This is also to demonstrate to the public the oath taker’s sincere and substantive belief in pledging his allegiance as required by the constitution. 125. In my view, the manner in which Mr Leung took the oath goes well outside an objective reasonable range of such requisite solemnity and sincerity: (1) The ceremony and procedure of the oath taking is to serve only one purpose, that is for the oath taker to comply with the constitutional requirements to take the oath in the form and substance as prescribed and required by BL104 and the ODO. Hence, the ceremony and procedure demands a solemnity and sincerity that underline this only purpose. (2) However, the holding of an umbrella, in particular with it opened, in the oath taking ceremony clearly simply does not accord in any reasonable way with the importance and seriousness of the taking of an oath as the ceremony’s only purpose. (3) Similarly, the chanting of slogans and the tearing of the paper with the message on it are also acts and conducts which were totally unrelated to the taking of the oath itself. (4) Clearly in my view, these theatrical acts viewed objectively had rendered the occasion without the requisite dignity and respect that was consistent with the constitutional importance and seriousness of the oath taking exercise. (5) Further, viewed against the above theatrical conducts, the Statement (with the words “係好莊嚴架”) uttered by Mr Leung just before he purportedly took the LegCo Oath was, as submitted by the Plaintiffs, plainly a sarcastic remark as objectively appeared to a reasonable person suggesting precisely the opposite. 126. In the premises, on any objective view, the above acts and conducts of Mr Leung, whether viewed independently or collectively, show that Mr Leung did not take the LegCo Oath with the requisite solemnity and sincerity as required by the law. He therefore failed to satisfy the Solemnity Requirement. 127. Moreover, Mr Leung’s chanted slogans were all carried out after he had been requested by the Clerk to take the LegCo Oath, and close to and after the reading of the LegCo Oath. In the circumstances, viewed objectively, I also agree with Mr Mok that it is clear that Mr Leung also sought to incorporate as part of the oath taking itself additional worded messages through his chanted slogans. For the reasons I have set out at paragraphs 91 and 92 above, he also therefore failed to comply with the Exact Form and Content Requirement. The oath taking is unlawful and invalid. 128. Again, there is no doubt that he intended to adopt and carry out those acts and conducts, which objectively viewed did not comply with the Solemnity Requirement and the Exact Form and Content Requirement. Mr Leung therefore declined or neglected to take the LegCo Oath when requested to do so on 12 October 2016. He should since 12 October 2016 be disqualified in law from assuming the Office, and there is no question of retaking the oath. 129. Mr Leung has raised a number of contentions as to why he should be regarded in law to have complied with the constitutional legal requirements in taking the LegCo Oath, or alternatively, he should be allowed to retake the oath. I would look at them in turn as follows. 130. First, Mr Lee SC (for Mr Leung) submits that there is no requirement in law prescribing an oath taker to take the LegCo Oath in a solemn manner. There is nothing in this. I have dealt with and rejected these submissions at paragraphs 39 - 41 above. 131. Second, Mr Lee contends that even if it is necessary in law to take the LegCo Oath solemnly, Mr Leung’s conducts and acts in the present case are entirely consistent with the LegCo Practices and similar conducts and acts adopted by Mr Leung himself and other LegCo members-elect in previous oath takings, which all had been in the past accepted to be valid by the Clerk or the President. The court should take these into account to conclude that Mr Leung’s conducts in the present case fall within the acceptable range of solemnity required for taking the LegCo Oath. 132. For the same reasons I have set out at paragraphs 44 ‑ 58 above, I reject this contention. 133. Third, Mr Leung (like Mr Law) also argues that he had no subjective intention in flouting the legal requirements for taking the LegCo Oath, as he was under the impression that what he did would be acceptable by reason of the past practices. He therefore did not intentionally breach the legal requirements for validly taking the oath, and should be allowed to re-take the oath even if the court finds that he did not comply with the legal requirements in taking the LegCo Oath on 12 October 2016. 134. For the same reasons I have explained at paragraph 110 ‑ 111 above, I cannot accept this argument. 135. Fourth, Mr Lee SC submits that, given the significant consequence of disqualification in being found to have failed to comply with the legal requirements for oath taking, as a matter of law, there should be a high degree of fairness in this context which requires that: (1) Before the oath-taking ceremony is held, members-elect should be informed of what can and cannot be done by clear guidelines or at least by some illustrative examples. (2) In some occasions, the Clerk or the President may have a lingering doubt on whether a legislature was sincere and solemn when taking the LegCo Oath. For instance, the Clerk or the President may not be sure whether the deviation from the statutory wordings of the LegCo Oath was deliberate or inadvertent. In these circumstances, the Clerk or the President should inform the member-elect what his concerns are and request the member-elect to address them. The member-elect should then be requested to read the LegCo Oath again in light of the concerns of the Clerk or the President so that he can confirm whether the member-elect is willing to take the LegCo Oath solemnly and sincerely. 136. There are no merits in these submissions. 137. It is a requirement of the law (whether under BL104 or the common law) that an oath taker must take the oath in a solemn and sincere manner. Members-elect are expected to know the law, and ignorance of the law is never a valid excuse for failing to comply with it. As a matter of fact, Mr Leung himself has confirmed that he has always understood that he has to take the LegCo Oath seriously, solemnly and sincerely.[18] 138. Solemnity and sincerity are commonly understood words, and need no specific guidelines. It is true that the degree and extent of solemnity and sincerity required of may vary in different circumstances and occasions. But in an objective test, what the law requires is that the oath taker’s conducts or acts in taking the oath would be regarded as falling within the range of the degree of solemnity regarded by an ordinary and reasonable person to be commensurate and consistent with the oath taking occasion, having regard to its importance and seriousness. This is a matter of objective common sense. It is neither necessary nor practicable to impose clear and definite guidelines as submitted by Mr Lee. 139. Finally, Mr Leung has deposed in his affirmation that he was indeed serious and solemn in taking the LegCo Oath on 12 October 2016 and why he genuinely commits himself to the pledge of allegiance under the oath notwithstanding the way and manner in which he took the oath. He explained in substance at paragraph 89 as follows (footnotes omitted): “89. In response to paragraphs 17 and 26 of the Form 86 and paragraphs 18 and 21 of the HCMP Affirmation, I say as follows: a. I did not merely go through the motion of taking the LegCo Affirmation as alleged. I took the 2016 Affirmation solemnly, sincerely and with every intention to be bound by it. My seriousness in reading the words of the LegCo Affirmation is demonstrated in the video recording; b. I did not read out the words of the LegCo Affirmation in a truncated or heavily truncated manner as alleged. I merely paused between words. There is no prescribed point(s) for pausing in the LegCo Affirmation prescribed by Part IV of Schedule 2 of the ODO. Indeed, there was not, and could not have been any motive for me to pause between the words 中華人民 and 共和國 for 中華人民共和國 or 香港 and 特別行政區 for 香港特別行政區; c. I was serious in stating that the oath-taking ceremony was very solemn, and the allegation that I ‘ridiculed and questioned the solemnity of oath taking’ by making this statement is totally unfounded; d. I made short statements before and after taking the 2016 Affirmation. The short statements I made before and after the taking of every one of my five LegCo oaths represented my political platforms,and I made them with the intention of telling the people of Hong Kong, in particular, my voters,that I would do my best to achieve those political goals during my term of office as a legislator. The language I used and the manner of my delivering those statements were consistent with my informal and direct way of communicating with my voters. In making those statements, I used the same informal language and adopted the same direct manner of speaking as I normally would in communicating with my voters. But those statements did not form any part of the LegCo oath. I did not change the wording of the LegCo oath or add or delete any word to or from it. I read out the prescribed LegCo oath completely and accurately; e. Carrying a yellow umbrella and displaying a paper-board concerning “人大831決議” whilst taking the LegCo oath did not mean that I was not serious, sincere or solemn in taking the oath; f. Making statements, carrying and displaying objects and tearing up a piece of paper with the words “人大831決議” after taking the LegCo oath likewise did not mean that I was not serious, sincere or solemn in taking the oath; and g. In supporting self-autonomy and/or self-determination, I do not advocate for or support the independence of Hong Kong. Self-autonomy and self-determination are different from,and should not be equated with independence of Hong Kong. Self-autonomy and self-determination are not inconsistent with the allegiance I owe to the HKSAR of the PRC. I believe that Hong Kong citizens have the freedom to discuss and decide on Hong Kong’s future after 2047 under the existing framework of ‘One Country, Two Systems’ in which Hong Kong is, as stipulated in the Basic Law, an inalienable part of the PRC. It is totally irresponsible and unfair to conclude that in supporting self-autonomy and/or self-determination, I do not genuinely bear allegiance (and do not intend to bear allegiance) to the HKSAR of the PRC and should therefore lose my seat in LegCo; and h. Likewise, saying ‘無須中共批准’ does not mean that I do not genuinely bear allegiance (and do not intend to bear allegiance) to the HKSAR of the PRC.” 140. As I have repeatedly said above, these subjective explanations relating to Mr Leung’s subjective thinking and thought process are not relevant to the objective assessment that the court adopts in determining whether Mr Leung’s oath taking complied with the legal requirements. 141. In any event, these explanations (which relate to his subjective intention to commit to the pledges in the LegCo Oath) do not assist Mr Leung. They do not impact in any way on the court’s conclusion that the way and manner in which Mr Leung took the LegCo Oath did not objectively comply with the Solemnity Requirement. Further, they indeed confirm that Mr Leung sought to incorporate in the oath taking itself additional worded messages. This by itself violates the Exact Form and Content Requirement as I have explained above. The oath taking by Mr Leung therefore did not comply with the legal requirements and is invalid in law. 142. For all the above reasons, I conclude that Mr Leung declined or neglected to take the LegCo Oath when requested to do so on 12 October 2016. He is therefore disqualified from assuming the Office as a matter of law and is not entitled to claim to act as a LegCo member. 143. The Clerk’s Decision to accept Mr Leung’s purported affirmation of the LegCo Oath as valid is therefore wrong in law and should be quashed. 144. Further, it follows that Mr Leung has claimed to be entitled to act as a LegCo member when he is not so entitled. The court should grant the declarations and injunction under section 73 of the LCO as sought by the Plaintiffs under HCMP 3382/2016. C4. The oath taking on 12 October 2016 by Ms Lau 145. The objective and undisputed evidence shows that Ms Lau took the LegCo Oath in the following way and manner on 12 October 2016. 146. According to the video footage, Ms Lau was requested by the Clerk to walk up to the table at the centre of the chamber to take the LegCo Oath in accordance with the predetermined order. Ms Lau purported to take the LegCo Oath by first saying the following in Cantonese (“Ms Lau’s Opening Statement”): “本人劉小麗謹此承諾，本人由街頭進入議會，定必秉承雨傘運動命運自主精神，與香港人同行，連結議會內外，對抗極權。我們要活在真誠磊落之中，打破冷漠犬儒，在黑暗中尋找希望，共同開創民主自決之路。推倒高牆，自決自強。”[19] 147. She then purported to read out the following words of the LegCo Oath at a slow pace with a clear pause lasting about six seconds or so from one character to the next following: “我……謹……此……宣……誓……，本……人……就……任……中……華……人……民……共……和……” 148. At this juncture, the Clerk interjected and informed Ms Lau that she had altered the contents of the LegCo Oath and if she did so he would not have jurisdiction to administer her oath. He requested Ms Lau to take the LegCo Oath afresh. In the course of the exchanges between the Clerk and Ms Lau, she also uttered the words “國”, “無呀” and “點樣改變呀？”. 149. Ms Lau then purportedly read out the words of the LegCo Oath again, also at a slow pace, uttering character by character, with a similar long pause from one word to the next following (except for her own name). Having spent around 10 minutes reading out the words of the LegCo Oath in this manner, she ended her oath taking with the statement (“Ms Lau’s Closing Statement”) “爭取全民退休保障，落實墟市政策，捍衛香港人生活尊嚴”,[20] this time at a normal pace. 150. The entire oath-taking lasted for a total of around 13 minutes. The Clerk did not say anything about Ms Lau’s purported oath-taking afterwards. 151. Further, Ms Lau made the following statements about her above oath taking on Facebook webpage and the media as follows. 152. On 13 October 2016, Ms Lau posted on the public Facebook webpage for her political affiliation “Democracy Groundwork” or “小麗民主教室” a statement (“the FB Statement”) in Chinese explaining her conduct during her purported oath-taking on 12 October 2016. The FB Statement started with the following remarks: “我所讀的，是九十多個沒有串連的獨立字句，毫無連貫性及意義可言。一切意義，純是觀眾自行分句，主觀判斷的憶測而已。”[21] 153. Following the opening remarks, the post was entitled 【劉小麗：慢讀是要彰顯誓詞的虛妄】,[22] within which Ms Lau expressed her views against the “core of the lies” that is the Hong Kong Special Administrative Region Government, and stated that the whole of the BL and the political system including the CE are not established through the will of the people and hence are illegal, the details of which are as follows: “【劉小麗：慢讀是要彰顯誓詞的虛妄】[23] 關於昨天的宣誓儀式，我想說一個故事。[24] 一個哈維爾說的故事。七十年代的捷克，一位菜檔老闆在檔口掛上一條橫額：「全世界工人團結起來！」任誰都察覺得到，菜檔老闆並不真心相信那句口號，他之所以要展示橫額，只是因為他害怕招惹麻煩，於是虛與委蛇，維持「社會和諧」。這句顯然是當權者授意的口號，最考妙之處在於，菜檔老闆無須明言對政權的恐懼，他甚至可以誑言掩飾：「『全世界工人團結起來』，有咩問題？」[25] 一個極權社會，依靠謊言維繫，人們活在謊言與虛偽之中。他們未必同意謊言，但只要假扮同意，或至少保持沉默，就可以在體制中生活。選擇順從的人，甚至可以主動迎合體制的要求去得到好處，以滿足自己的權力慾。[26] 昨天在會議廳，全體立法會議員作出宣誓。有些西裝骨骨、代表特權階級的愛國愛港人士，堂而皇之宣讀誓詞，完成行禮如儀的劇本，彷彿他們真的「謹以至誠」，許諾為人民服務。由此，他們成為尊貴議員，為權貴打手欺壓港人，還可以誑言掩飾：「做立法會議員服務市民，有咩問題？」[27] 謊言的核心，正是特區政府的整個政制，表面上有合法性，其實一點都無！整個基本法，整個政制連特首，都未經過港人民意授權，所以根本非法！[28] 一個公平公正的議會，本來是民主自由、人民意志的彰顯。但在香港，我們只有鳥籠式半民主；今次選舉期間的種種操控、取消候選人參選資格，踐踏僅餘的程序公正。立法會主席選舉上，梁君彥及秘書處在梁的國籍問題處理兒戲，竟然到投票日仍然疑點重重，最終梁君彥在混亂中「當選」，亦難以服眾。[29] 昨天，我將官方誓詞逐字宣讀。誓詞變成九十多句毫無連貫性的句子，沒有任何組合、連結及意義，令聽眾無法聽到任何句式及語氣。這樣，一切意義就純是觀眾自行分句，主觀判斷造成的憶測而已。這個做法是為了彰顯行禮如儀的虛偽。包括梁君彥在內的建制派，妄言任何阻撓梁君彥順利當選主席的反對派議員都是「搞事」、「破壞議會」，我卻要傳達一個訊息：流暢鏗鏘的宣誓是虛偽的，和諧的議會也是虛偽的。[30] 哈維爾說，要對抗極權，最根本的就是要瓦解維繫極權的謊言與虛偽。只有每個人都堅持不隨俗表態，說出自己真正所思所想，高牆才有被打破的可能。昨天只是一個開始，今後我將會在議會內外，與香港人同行，努力對抗各種體制要求的虛偽，做到哈維爾所說的活在真誠磊落之中（living in truth），共同開創民主自決之路。[31] 最後，我在宣讀官方誓詞前的那一段說話，是更真誠的版本： 本人 劉小麗 謹以至誠，向香港市民承諾： 本人由街頭進入議會，定必秉承雨傘運動命運自主精神，與香港人同行，連結議會內外，對抗極權。我們要活在真誠磊落之中，打破冷漠犬儒，在黑暗中尋找希望，共同開創民主自決之路。推倒高牆，自決自強！”[32] 154. Ms Lau had also explained to the media the rationale of her 6‑second pause in between each word of the LegCo Oath, which was reported in Apple Daily’s article (“the Apple Daily Article”) dated 13 October 2016, in which Ms Lau stated, among others, the following: “如果每個字都相隔六秒，你點可以覺得佢係一句句子嚟呢？可能係九十幾篇文章嚟！”[33] 155. In the President’s Ruling made on 18 October 2016, the President ruled that the oath purportedly taken by Ms Lau was invalid, but allowed her to retake the oath. In making this decision, the President stated the following (at paragraphs 8 and 9): “8. Hon LAU Siu-lai subscribed an affirmation in Cantonese when taking the LegCo Oath. I notice that in her first attempt to subscribe the affirmation, she read out the words of the oath at a normal speed but was stopped by the Clerk as she had altered the content of the affirmation. In her second attempt, she subscribed the affirmation at a speed very much slower than that in her first attempt. She just read out each word of the oath prescribed by the Ordinance individually, with a long pause between words, making it objectively difficult for anyone to understand her affirmation. The conduct of Ms LAU’s oath-taking, objectively assessed, has shown that she was not serious about the affirmation and had no intention to be bound by it. 9. Based on the above reasons, I rule that Ms LAU’s affirmation was invalid and consider that she should subscribe the affirmation afresh. I am prepared to allow her to do so at a Council meeting if she puts forward her requests in writing.” 156. The Plaintiffs now submit that this is a clear case that, objectively viewed, Ms Lau did not genuinely and faithfully accept and thus bind and commit herself to the obligations pledged in the LegCo Oath. 157. I agree. 158. The way in which Ms Lau read out LegCo Oath by breaking up each of the words therein with a some 6-second pause in between each of these words would objectively and unquestionably lead a reasonable person to conclude that she did not intend to convey any meaning of the contents and pledges of the LegCo Oath when reading it out. All the individual words of the LegCo Oath were read out in an incoherent and unlinked manner when she adopted a mechanical manner in reading out each of the words with an approximately 6-second break of silence. Objectively, in so doing, no useful meaning of the very contents of the oath (which is only established through the linking up of the individual words therein in a coherent way) was intended to be conveyed to a reasonable person. 159. On this basis alone, I have no hesitation in coming to the conclusion that, objectively looking at the manner and way in which Ms Lau purported to take the LegCo Oath, Ms Lau did not manifest an intention to genuinely and faithfully accept (and hence commit and bind herself to) the obligations of the pledges embodied in the LegCo Oath. She has clearly failed to comply with the Substantive Belief Requirement. 160. Moreover, the objective conclusion that Ms Lau did not intend to commit herself to the obligations therein is indeed confirmed by herself. As mentioned above, in the FB Statement, Ms Lau stated in no uncertain terms that what she purportedly read on the day of taking the LegCo Oath was ninety-odd unrelated and independent word-sentences, entirely devoid of coherence and meaning (“昨天，我將官方誓詞逐字宣讀。誓詞變成九十多句毫無連貫性的句子，沒有任何組合、連結及意義，令聽眾無法聽到任何句式及語氣”). 161. As Mr Mok submits, the court is entitled to take into account Ms Lau’s above admission made shortly after taking the purported oath on 12 October 2016. Such evidence would be admissible, and only limitation with regard to time is that it must not be beyond such time as would prevent reasonable men from acting upon it: AG v Bradlaugh (1885) 14 QBD 667 at 699, per Brett MR, and 711, per Cotton LJ. 162. There is therefore no question that Ms Lau did not comply with the Substantive Belief Requirement in taking the LegCo Oath. 163. Further, I accept Mr Mok’s further submissions that Ms Lau also violated the Exact Form and Content Requirement in incorporating additional messages in the oath itself. 164. Again, Ms Lau made the Opening Statement and Closing Statement respectively right before and after the reading out of the LegCo Oath, and after she had been requested to take the oath. In the circumstances, objectively viewed as a whole, these statements were made as part of the oath taking itself. Ms Lau therefore sought to convey additional worded messages (those two statements) in the oath taking, which is not permissible in law. The oath taking therefore also did not comply with the Exact Form and Content Requirement and is unlawful and invalid. 165. Finally, it is obvious that she intended to take the LegCo Oath in the slow-paced manner and made the Opening and Closing Statements in the way she did, which objectively viewed, did not comply respectively with the Substantive Belief Requirement and the Exact Form and Content Requirement. In law, she therefore declined or neglected to take the LegCo Oath when requested to do so on 12 October 2016. 166. Mr Chan SC for Ms Lau has put forward a number of contentions arguing that she has validly taken the LegCo Oath. They are as follows. 167. First, Ms Lau has deposed in her affirmation filed in these proceedings at length the reasons as to why she took the LegCo Oath in the way she did. In substance, she emphasized that she had always been interested in the study of linguistics, and the 6-second pause for each of the words when reading out the LegCo Oath was indeed a deliberate manner she had worked out as a result of her study and research in linguistics and past practices adopted by other LegCo members-elect in taking the LegCo. She intended to make use of the “effect” of “silence” as a linguistic skill to underline and highlight the message to the public the “hypocrisy” of some LegCo members (but who have not been identified by Ms Lau) who “treated the oath as going through the motion” and to convey the message that “their fluent oaths are hypocritical”. In particular, she now maintained that by taking the oath at a slow pace, she wished to “contrast with some LegCo members that [she] was indeed solemn and sincere and was taking [her] oath seriously.” She said in so doing, she was “inviting the public to reflect on the true meaning behind taking an oath through [her] solemn and sincere way of oath-taking”. She wished to “bring to the public attention the hypocrisy of some LegCo members behind their fluent but insincere oath-taking”. Ms Lau emphasised that she “did not intend [her] oath to be meaningless. To the contrary, [she] very much intended [her] oath to be meaningful”, as she “wished that [her] oath could allow the public to reflect on the true meaning behind taking an oath”. See: Ms Lau’s affirmation, paragraphs 56 - 84. 168. Ms Lau also said the FB Statement and the media interview as reported in the Apple Daily Article (which she said in any event did not fully reflect everything she had said in the interview) should also be understood in the above way. All that she was trying to say and convey by way of the FB Statement and her explanations to the media is to similarly highlight the hypocrisy of some LegCo members behind their fluent but insincere oath-taking and to arouse the public to reflect to this question. 169. With respect, I am unable to accept this contention. 170. Ms Lau’s self-explanations are in nature evidence and matters of her alleged subjective thought process and subjective intention in adopting the slow-paced reading. They also relate to what alleged subjective meaning she seeks to accord to that conduct. As I mentioned above, they are irrelevant to the objective assessment of whether the oath taking was legally compliant. The court would not take them into account in construing what meaning Ms Lau’s manner and way adopted in taking the LegCo Oath would be conveyed to a reasonable and objective person at the time of the oath taking. 171. In this respect, Mr Chan SC fairly recognises that the court is to apply an objective test in assessing Ms Lau’s conducts and words in taking the LegCo Oath, and that the subjective thinking of Ms Lau is strictly speaking irrelevant to the objective exercise. Leading counsel however argues that what Ms Lau has sought to explain in her affirmation should be taken as the relevant background and context against which the court should construe objectively what a reasonable person would have understood her slow-paced reading of the LegCo Oath to mean. In particular, Mr Chan says, in the context as explained by Ms Lau, if it is unequivocal as to what she could be reasonably and objectively understood to mean, the court should accept an interpretation favourable to her since there would be a lack of cogent evidence supporting a breach of the legal requirements. 172. With respect to Mr Chan, the objective context and background relevant to an objective construction should be ones that would have been reasonably known to a reasonable person present at the time of the oath taking. In this respect, I am unable to see how it could possibly be said that those subjective explanations I have summarised above as to Ms Lau’s interest and study in linguistics and her “intention” to convey the message to the public to underline the “hypocrisy” of other unidentified members of the LegCo could and would have been reasonably and objectively known to a reasonable person. The same applies to the construction of the FB Statement and Ms Lau’s explanations to the media made shortly after 12 October 2016. 173. I must therefore also reject this contention, and would not take into account Ms Lau’s above explanations in considering the objective meaning that would be conveyed to a reasonable person by her slow-paced reading of the LegCo Oath. 174. In any event, I also accept Mr Mok’s submissions that even if the court was to take into account Ms Lau’s purported explanations, they simply do not assist her. In my view, the objective meaning and effect of the peculiar way in which Ms Lau read out the LegCo Oath as this court has found above is so obvious and unequivocal that Ms Lau’s self-explanations as to its underlying meaning and message simply could not be accepted to have any bearing on the objective construction so as to displace it. 175. For these reasons, I reject this contention based on the self-explanations. 176. Second, Mr Chan submits that the court should give due respect and weight to the Clerk’s decision in not treating Ms Lau’s oath taking on 12 October 2016 to be invalid. Alternatively, Mr Chan similarly says the court should give due weight to the President’s decision to allow her to retake the oath. Mr Chan therefore submits that the court should be very slow in coming to a conclusion which is different from either the Clerk’s decision or the President’s Ruling. Mr Chan’s bases for these submissions are that a decision on the compliance of taking an oath of allegiance is in nature a political one and what constituted acceptable etiquette in so doing would be a matter that would be judged in light of the tradition, culture, history and values of the LegCo. The Clerk and the President who have administered so many LegCo Oaths and who are familiar with the LegCo practices are the best persons to make these decisions. 177. I have considered these submissions as a matter principle above at paragraphs 54 - 58 and rejected the same. For those same reasons, I reject this argument. In any event, I have explained above why Ms Lau is regarded clearly in law to have declined or neglected to take the LegCo Oath by adopting the slow-paced reading. There is nothing stated by the President in his decision that would persuade me that my conclusion is incorrect. 178. Third, Mr Chan submits that whether Ms Lau had declined or neglected to take the LegCo Oath must be determined at the time of taking the oath. Thus, the court should ignore the FB Statement and the Apple Daily Article. These are, Mr Chan says, ex post facto statements which are not relevant to the question that needs to be determined. 179. I am unable to accept this. 180. The FB Statement and the media explanations as objectively construed are in the nature of post event admission. These have always been admissible as evidence. 181. In any event, as I have said above at paragraphs 158 and 159, by objectively looking at Ms Lau’s slow-paced reading of the LegCo Oath alone without reference to the FB Statement and the Apple Daily Article, the court has already come to the clear conclusion that she manifested no intention to accept and commit herself to the LegCo Oath obligations. 182. For all the above reasons, I reject all of Ms Lau’s arguments to say that she had validly complied with the oath taking requirements in purportedly taking the LegCo Oath on 12 October 2016. 183. This is an obvious case that in law Ms Lau declined or neglected to take the LegCo Oath when requested to do so on 12 October 2016. She has since been disqualified as a matter of law from assuming or entering on the Office. 184. In the premises, the President’s Ruling in allowing her to retake to the LegCo Oath is wrong in law and should be quashed. 185. The court should also declare that the Office held by Ms Lau has become vacant since 12 October 2016 and that she has since wrongfully claimed to be entitled to act in the Office as sought by the Plaintiffs. There should also be an injunction restraining her from so acting or claiming to be so entitled to act. C5. The oath taking on 12 October 2016 by Mr Yiu 186. The objective and undisputed evidence shows that Mr Yiu took the LegCo Oath in the following way and manner on 12 October 2016. 187. According to the video footage, following the request made to Mr Yiu by the Clerk for him to take the LegCo Oath, Mr Yiu purported to take the LegCo Oath by reading out in Cantonese words which do not accord with the wording of the LegCo Oath as prescribed by the ODO in the following terms (the underlined words are not in the LegCo Oath as prescribed by the ODO): “本人姚松炎，謹以至誠，據實聲明及確認，本人就任中華人民共和國香港特別行政區立法會議員，定當守護香港制度公義，爭取真普選，為香港可持續發展服務，定當擁護《中華人民共和國香港特別行政區基本法》，效忠中華人民共和國香港特別行政區，盡忠職守，遵守法律，廉潔奉公，為香港特別行政區服務。”[34] 188. At this juncture, the Clerk told Mr Yiu that since Mr Yiu had added words to the oath and thereby amended the oath, he (ie, the Clerk) had no jurisdiction to administer the oath for Mr Yiu, and requested Mr Yiu to take the LegCo Oath again according to the prescribed form and content, otherwise, Mr Yiu would not be allowed to attend LegCo meetings according to Rule 1 of the Rules of Procedure of the LegCo. 189. Mr Yiu said in response, “好呀”, then said the following (again, the same underlined words are not in the prescribed form of the LegCo Oath): “本人姚松炎，謹以至誠，據實聲明及確認，本人就任中華人民共和國香港特別行政區立法會議員，定當擁護《中華人民共和國香港特別行政區基本法》，效忠中華人民共和國香港特別行政區，盡忠職守，遵守法律，廉潔奉公，為香港特別行政區服務，定當守護香港制度公義，爭取真普選，為香港的可持續發展服務。”[35] 190. At this point, the Clerk stated that Mr Yiu added words immediately after the oath which were not contained in the prescribed form and that he had no jurisdiction to administer the oath for Mr Yiu, and asked Mr Yiu to return to his seat. 191. In the President’s Ruling made on 18 October 2016, the President confirmed the Clerk’s above decision and ruled that Mr Yiu did not validly take the LegCo Oath on 12 October 2016, but the President allowed Mr Yiu to retake the oath. In making this decision, the President stated the following (at paragraphs 6 and 7): “6. After examining the oath/affirmations taken by Dr Hon YIU Chung-yim, and two other Members at the Council meeting of 12 October 2016, I affirm the Clerk’s decision to decline jurisdiction to administer their oath/affirmations as they altered the contents of their oath/affirmations. ... I notice that Dr YIU has added words to the end of his affirmation, and as such, altered its content as prescribed by the ODO. 7. By virtue of section 19 of the ODO, a Member shall, as soon as possible after the commencement of his term of office, take the LegCo Oath which, if taken at a Council meeting other than the first one after a general election, shall be administered by me. As such, I accede to Dr YIU’s written request for taking his affirmation afresh at the Council meeting of 19 October 2016...” 192. At the next LegCo meeting on 19 October 2016, Mr Yiu purportedly “retook” the LegCo Oath before the President at the LegCo meeting. 193. Mr Mok now submits that, objectively looked at, in taking the LegCo Oath on 12 October 2016, Mr Yiu had clearly breached the Exact Form and Content Requirement, by intentionally added the extra words (“the Extra Words”) “定當守護香港制度公義，爭取真普選，為香港可持續發展服務”[36] in the prescribed content of the oath. Mr Yiu therefore had declined or neglected to take the LegCo Oath on 12 October 2016. 194. I agree with Mr Mok. 195. As described above, Mr Yiu purported to take the LegCo Oath twice on 12 October 2016. 196. When requested by the Clerk to take the LegCo Oath on 12 October 2016, Mr Yiu first purported to take the LegCo Oath by uttering the following: “本人姚松炎，謹以至誠，據實聲明及確認，本人就任中華人民共和國香港特別行政區立法會議員，定當守護香港制度公義，爭取真普選，為香港可持續發展服務，定當擁護《中華人民共和國香港特別行政區基本法》，效忠中華人民共和國香港特別行政區，盡忠職守，遵守法律，廉潔奉公，為香港特別行政區服務。” (underlined are the Extra Words) 197. There can be no dispute that the Extra Words are not in the prescribed form of the LegCo Oath. By adding these words in the middle of the oath, plainly Mr Yiu had breached the Exact Form and Content Requirement which thereby rendered that oath taking invalid as a matter of law. 198. The only other question left is whether objectively he intended to add the Extra Words in the middle of the oath as he did. If so, he would be regarded in law as having declined or neglected to take the LegCo Oath, and be disqualified from taking the Office and should not be permitted to retake the oath again. 199. Objectively assessed, I agree with Mr Mok that Mr Yiu intended to do so: (1) As demonstrated in the video of the oath taking, Mr Yiu after reading out the Extra Words in the beginning of the LegCo Oath, and after pausing momentarily, continued to read on with the rest of the words in the LegCo Oath to complete it. (2) The Extra Words constituted a complete sentence of some 25 words. It would be objectively unreasonable for a person to have read and completed this entire sentence if it was not in the intended place of a passage. More so, when he indeed continued to finish reading thereafter the entire passage. Hence, the fact that Mr Yiu read out the entire sentence of the Extra Words in the middle of the LegCo Oath and continued thereafter to finish reading the rest of the LegCo Oath shows objectively that he intended to read out the Extra Words in the way he did. (3) Alternatively, objectively, if he had not intended to insert the Extra Words in the beginning of the oath as he did, and that was merely a mistake, a reasonable person in his position would have stopped and started over again without reading out the Extra Words in the same place again. This is so as: (a) Mr Yiu (and a reasonable person in his position) should reasonably have realised that he had made a mistake in reading out the Extra Words in the wrong place given that the Extra Words constituted a long complete sentence. (b) Mr Yiu (and a reasonable person in his position) had before taking the LegCo Oath been reminded repeatedly by the LegCo circulars that it would be unlawful to take the LegCo Oath by adding words in the oath. Thus, objectively, Mr Yiu (and a reasonable person in his position) would have realised at the least that reading out words not in conformity with the prescribed form of the LegCo Oath in the middle of oath would render the oath taking invalid. (4) In the premises, Mr Yiu in deciding to continue to read out the rest of the LegCo Oath even after reading out the Extra Words demonstrates objectively that at least by that time he intended to insert and read out the Extra Words in the way as he did in this first attempt. 200. As such, Mr Yiu did decline or neglect to take the LegCo Oath when requested to do so in failing to comply with the Exact Form and Content Requirement in the first attempt. He should have already been disqualified from taking the Office then and not be permitted to retake the oath. 201. In any event, there is no question in my view that Mr Yiu must be regarded in law to have declined or neglected to take the LegCo Oath in accordance with the law in the second time when requested by the Clerk to do it again. This is so because: (1) Although he was already warned by the Clerk that his first attempt oath taking was invalid by adding the Extra Words to the LegCo Oath, he still inserted and read out the Extra Words again the second time albeit at the end of the LegCo Oath. Viewed objectively, there is no doubt that he intended to add the Extra Words in the way as he did. This by itself is in breach of the Exact Form and Content Requirement, and the oath purportedly taken is invalid. (2) Moreover, as shown in the video, Mr Yiu read out the Extra Words immediately, continuously and smoothly following the last sentence of the LegCo Oath, as if they were of the same passage. Objectively viewed, the Extra Words were therefore intended to be read out and understood as part and parcel of the oath. (3) In the premises, Mr Yiu intended to read out the Extra Words and in the way as he did, which was not in compliance with the Exact Form and Content Requirement. Objectively assessed, he therefore declined or neglected to take the LegCo Oath when he was requested to do so. 202. In the circumstances, as a matter of law, Mr Yiu in any event should be disqualified from assuming the Office after the second attempt in purporting to take the LegCo Oath on 12 October 2016. 203. Mr Yiu raises a number of contentions to say why he did not decline or neglect to take the LegCo Oath on 12 October 2016. 204. First, he says it was not his intention to alter the form or substance of the LegCo Oath by adding the Extra Words. In support, Mr Yiu has filed an affirmation to explain why he took the LegCo Oath in the way as he did in the first and second attempts. In substance, he explained that: (1) From the LegCo circulars and the government press release, he himself understood the position of taking the LegCo Oath being, among others, that (a) he must say all the words of the LegCo Oath solemnly, but there was no prohibition on insertion of words before, in between, or after the LegCo Oath; (b) a LegCo Oath that is in the form and manner inconsistent with the form prescribed by the ODO, that in the sense that it “alters the substance of the oath itself”, would be unlawful and of no effect; but (c) for whatever reason if he failed to take the LegCo Oath in the form and manner as prescribed by the ODO, he would be permitted to take an oath again at any other LegCo meeting which would be administered by the President: paragraphs 21 - 26. (2) With the above (mis)understanding in mind, and having studied the practices of oath taking by other LegCo members-elect in previous occasions, he intended and decided to add the Extra Words to the end of his LegCo affirmation (ie, the LegCo Oath): paragraphs 27 - 38. (3) His original plan was to read out the Extra Words after finishing reading out the LegCo Oath solemnly and sincerely. He initially intended to memorise the entire LegCo Oath and the Extra Words by heart and he tried to do that whilst waiting for his turn to take the LegCo Oath. However, he found that to be too difficult and decided that he would read out from his mobile phone when taking the oath: paragraphs 39 - 41. (4) However, when it came to his turn in taking the LegCo Oath, he was a little nervous. Inadvertently and mistakenly, he read out the Extra Words (which began with the words “定當”) first as he was searching from the mobile phone screen for the beginning words “定當擁護” of the LegCo Oath. He realised it was a mistake at that time, but thinking that it should not invalidate his oath (given the past practices of other LegCo members-elect and his above understanding of the legal position), he decided on the spur of moment that he should carry on reading to finish the entire LegCo Oath: paragraph 42. (5) When he was asked by the Clerk to do it the second time, he read out the Extra Words at the end of the LegCo Oath as he had originally planned to do. He did not have any reasons to believe that this would invalidate the oath taking as what he did was consistent with what other LegCo members had done in previous years and this year: paragraph 44. (6) He did not decline or neglect to take the LegCo Oath. He in fact read out all the words in the LegCo oath solemnly and sincerely and genuinely agreed to be bound by the serious obligations in the LegCo Oath. He is not a secessionist and not one of those who “openly advocated the notion of ‘Hong Kong independence’, ‘Hong Kong national self‑determination’ or similar notions, condemned by the [Interpretation]”: paragraphs 42 - 44 and 51. 205. Again, Mr Yiu’s above explanations as to why he took the LegCo Oath in the way as he did are in nature matters and evidence of his subjective intention and thought process. They are irrelevant to the objective question of whether his oath taking complied with the relevant legal requirements, and whether he declined or neglected in law to take the LegCo Oath in so doing. 206. In any event, even if the court was to take these explanations into account, they indeed fortify the conclusion that, objectively, Mr Yiu declined or neglected to take the LegCo Oath on 12 October 2016. 207. In relation to the first attempt, Mr Yiu states that, when reading the LegCo Oath, he realised that he “accidentally picked up” the additional sentence which he had “intended to add at the end”. Nevertheless, he decided to “carry on reading”. In the premises, whether or not the original utterance was inadvertent, his own evidence confirms that he consciously deviated from the contents of the oath when he decided to “carry on reading”. Hence, even on his own case, he did at the material time intend to read out the Extra Words in the middle of the entire LegCo Oath as a whole, which objectively assessed in law offends the Exact Form and Content Requirement. As a matter of law, he had declined or neglected to take the LegCo Oath and should be disqualified from assuming the Office. 208. In relation to the second attempt, in his own evidence, it is clear that he indeed intended to add and read out the Extra Words at the end of the LegCo Oath. For the reasons I have explained above, objectively viewed, the way and manner in which he read out those words immediately and continuously after the LegCo Oath would lead a reasonable person to understand him as reading out the Extra Words by treating them as part of the LegCo Oath. This would be altering the form and content of the LegCo Oath and thereby unlawful. As such, objectively assessed as a matter of law, Mr Yiu declined or neglected to take the LegCo Oath at least in his second attempt in any event. He shall equally be disqualified from assuming the Office in law. 209. Mr Yiu claims in his affirmation[37] that it is his understanding that there is no prohibition on insertion of words before, in between, or after the LegCo Oath, and it is only unlawful to alter the substance of the LegCo Oath on the pledges. He also claims that although he was referred to the authority of Leung Kwok Hung v Legislative Council Secretariat, supra, in the LegCo circulars, he was not provided a copy. Presumably what he wanted to highlight by this claim is that he had not read the authority in forming his (mis)understanding of the legal position. 210. With respect, this could not assist Mr Yiu at all: (1) Ignorance or misunderstanding of the law is never a good excuse to justify an act which is legally invalid. (2) Further, whilst the court looks only at the objective effect of Mr Yiu’s words and conduct, it is worth noting that Mr Yiu was, at the material times, fully aware of the existence of Hartmann J’s judgment in Leung Kwok Hung v Legislative Council Secretariat, supra, and that he was reminded by the Clerk on 20 September 2016 that he should “read the judgment of Mr Justice Hartmann” if he had any doubt. The fact that “no copies of that judgment were handed out” to the members is no excuse for not reading it, since it is available online to everyone and Mr Yiu was intending to take the risk of adding words to his LegCo Oath. (3) There is, therefore, no justification whatsoever for his alleged understanding that there is “no prohibition on insertion of words before, in between, or after the LegCo Oath”. (4) Even the press release of 11 October 2016 (which Mr Yiu relies on, at Mr Yiu’s affirmation at paragraph 25) does not assist him. It refers expressly to Hartmann J’s judgment, stating that a member-elect “must take his or her oath in a manner and form that accords with the Ordinance” and that taking the oath in a manner or form that is inconsistent with that as prescribed by the ODO would thereby result in “altering the substance of the oath itself, the oath offends Article 104 of the Basic Law and will therefore be unlawful and of no legal effect”. 211. I therefore reject this contention. 212. Second, Ms Eu SC for Mr Yiu submits that in deciding whether Mr Yiu declined or neglected to take the LegCo Oath on 12 October 2016, the Plaintiffs have to establish that Mr Yiu “intentionally and deliberately flouted” the legal requirements under BL104 and the ODO when taking the LegCo Oath as he did. 213. I disagree. As I have explained in paragraph 110 above, given the objective test, all that the Plaintiffs have to satisfy the court is that (a) objectively, Mr Yiu intended to add the Extra Words in the way as he did; and (b) the Extra Words so added, objectively viewed, altered the form or substance of the LegCo Oath. It is not necessary to show that Mr Yiu intended to flout the specific legal requirement. 214. Third, Ms Eu submits that the substantive legal test of a valid oath is whether, viewed objectively, the oath taker can be seen to give his pledge in relation to the three aspects of the oath, namely, to uphold the BL, to bear allegiance to the Hong Kong Special Administrative Region and to serve the Hong Kong Special Administrative Region conscientiously. Hence, even though a valid oath has to be correct in form and substance, the oath taker is only to be disqualified from assuming the office if he intentionally refuses to be bound by or give the above three pledges of the oath. In relation to this question, Ms Eu submits that Mr Yiu’s evidence (as summarised above at paragraph 204) is clearly relevant, which negates any conclusion that he refused to be bound or give the pledges. 215. With respect, I am unable to accept these submissions. 216. As a matter of law, paragraph 2(3) of the Interpretation expressly provides that “An oath taker who intentionally reads out words which do not accord with the wording of the oath prescribed by law … shall be treated as declining to take the oath”. In other words, an oath taker must not intentionally do any acts to change the form of the oath when taking it. Compliance with the exact form of the oath is therefore a separate and distinct legal requirement for a valid oath taking. 217. In any event, in Leung Kwok Hung v Legislative Council Secretariat, supra, Hartmann J has already held at paragraphs 35 - 36 (as quoted above) that in law an alteration in the form of the oath is also a change in the substance of the oath itself. 218. Further, the court assesses objectively whether an oath taker intentionally change the form of the LegCo Oath when taking it. 219. Fourth, Ms Eu submits that since the Court of Appeal in the CA Judgment at paragraphs 39 - 40, 79 and 95 has emphasized that the court in hearing a challenge on the validity of an oath taking has to be engaged in a full merits review (which approach is binding on this court), the court must have had regard to all the relevant objective evidence put forward by Mr Yiu to determine objectively whether he declined or neglected to take the LegCo Oath. 220. The objective evidence put forward in Mr Yiu’s affirmation and relied on by Ms Eu is: (1) Yiu’s passion for and his past contribution to public affairs (paragraphs 4 - 7). (2) His mission in running for office (paragraphs 9 - 10). (3) His efforts in the election campaign (paragraphs 11 - 15). (4) His work since the election (paragraphs 14 - 15). (5) His meticulous attention to briefings, circulars, press release and past practices, as to the requirements for proper oath‑taking (paragraphs 16 - 37). (6) His spontaneous response “好呀” (which can mean alright, okay or good) to the Clerk when he was stopped and invited to retake the oath on 12 October 2016 (paragraph 44). (7) His taking of the oath afresh with the Extra Words after the oath-taking on 12 October 2016 (paragraph 44). (8) His confirmation that he is at all material times ready and willing to become a LegCo member to uphold the BL, bear allegiance to the Hong Kong Special Administrative Region and serve the Hong Kong Special Administrative Region (paragraph 56). (9) His questions to the Clerk at around 12:31 pm for explanation as to why he was not permitted to take the LegCo Oath (paragraph 47). (10) His letter to the President dated 13 October 2016 asking to take the oath afresh (paragraph 49). (11) His dedication to LegCo work during all this time (paragraph 57). (12) His confirmation that he is not a secessionist or an advocate of Hong Kong independence (paragraphs 8, 53(1) and (3)). (13) The total absence of any reason as to why Mr Yiu should decline or neglect to take the oath or refuse to be bound. 221. Ms Eu submits that the above objective evidence sheds light on the objective circumstances before, during and after Mr Yiu’s oath taking, and his subjective intention before, during and after the oath taking. Given all the above, it makes absolutely no sense for him to intentionally refuse to take the oath. Further, says Ms Eu, objectively speaking, Mr Yiu read out the three core aspects of the LegCo Oath accurately and solemnly, and thus gave a pledge intending to be bound by those three aspects of the oath. Having regard to such objective evidence, Ms Eu submits that, objectively assessed, the Plaintiffs simply cannot show that Mr Yiu intended to flout the law and attract the serious consequence of disqualification. 222. Alternatively, Ms Eu points out that Mr Yiu has also provided in his affirmation the “innocent explanations” as to why he took the LegCo Oath in the way he did. These explanations are that: (1) They would enjoy immunity from suit for words spoken before the LegCo, or written in a report (paragraph 17). (2) An oath that is in form and manner “inconsistent with” the ODO, in the sense that it “[alters] the substance of the oath itself” will be unlawful and of no legal effect (paragraph 26(2)). (3) The validity of the oath does not mechanically depend upon the utterance of the exact number of words, without more or less. Adding word or sentence to the oath would not invalidate the oaths taken (paragraphs 18, 24(3), 27 - 33, 35, 42 and 44 - 45). (4) If the oath is not taken in accordance with the relevant requirements, a LegCo member may seek to take the LegCo Oath afresh and have the oath administered by the President at another Council meeting (paragraphs 19, 24(4), 26(4)). (5) The many past cases of oath taking where words were added. It was never said that this amounted to an invalid oath and in the past the Hon Wong Yuk Man was allowed to retake the oath (paragraph 27 - 37). 223. Ms Eu argues that, given these explanations and Mr Yiu’s repeated statements in the affirmation that he never intended to decline or neglect to take the oath,[38] viewed objectively, there is simply no compelling evidence to show that he intentionally and deliberately flouted the requirements of the law, or he did not intend to be bound by the oath, or he refused to give the pledge. In the circumstances, at most Mr Yiu made an invalid and non-confirming oath and should be allowed to take it again. 224. With respect to Ms Eu, notwithstanding her forcible submissions, I am unable to accept them. 225. As I see it, the underlying plank of Ms Eu’s above submission is that in order to find that Mr Yiu had declined or neglected to take the LegCo Oath, the court must find, albeit objectively, that he had a specific intent to flout the legal requirements or that he intended not to be bound by the pledges in the oath. Hence, as Ms Eu further submits, given his innocent explanations, especially his mistaken understanding of the law, and the lack of any good reasons for him to want to risk losing his seat after winning it, there cannot be a finding based on compelling evidence that he did specifically intend to flout the law or not to commit to the pledges in the LegCo Oath, and thereby to attract the serious consequence of losing his hard won seat. 226. Ms Eu’s submissions would amount to the need to find that the oath taker, in taking the oath, intended to break the law relevant to oath taking and to lose to his or seat by doing so. However, this underlying contention is in my view incorrect. 227. As I have repeatedly explained above, given the objective test, it is not necessary as a matter of law to find a specific intention to break or not to comply with the specific legal requirements relevant to oath taking. It is sufficient to show that the oath taker intended to adopt a particular way, conduct or manner to take the LegCo Oath, and that particular way, conduct and manner of taking the oath objectively assessed do not comply with the relevant legal requirements. 228. Once understood the legal test in this way regarding the finding in law whether an oath taker has “declined” and “neglected” to take the LegCo Oath, for the reasons I have set out in paragraphs 204 ‑ 211 above, the evidence and innocent explanations put forward by Mr Yiu do not assist him. Quite to the contrary, as explained above, they support the clear conclusion that, objectively, Mr Yiu did decline or neglect to take the LegCo Oath whether at his first or second attempt on 12 October 2016. 229. In the premises and for the above reasons, I conclude that Mr Yiu failed to comply with the constitutional requirement under BL104 and declined or neglected to take the LegCo Oath when requested to do so on 12 October 2016. He has since been disqualified from assuming or entering on the Office. 230. The President thus erred in law in allowing Mr Yiu to retake his oath. His decision should be quashed. 231. Finally, since Mr Yiu has claimed to be entitled to act as a LegCo member since 19 October 2016, the court should grant a declaration and injunction under section 73 of the LCO. D. THE DEFENDANTS’ STAY APPLICATIONS 232. On the last day of the hearing, after Mr Mok had completed his reply submissions for the Plaintiffs, Mr Dykes (for Mr Law) made an oral application for a permanent stay of these proceedings.[39] In gist, Mr Dykes’ main reasons for making the application are these. In the absence of any explanations from the Plaintiffs as to why they brought these disqualification proceedings against only the Defendants but not some other LegCo members-elect (who took the LegCo Oath in similar manners and ways as the Defendants did), the present proceedings brought against Mr Law are an abuse of process, and the court should exercise its inherent jurisdiction and under Order 18, rule 19(1)(d) of the Rules of the High Court to stay permanently or dismiss the proceedings. 233. All the other Defendants through their respectively leading counsel then followed suit and made an oral application for a permanent stay or dismissal of the present proceedings for abuse. 234. Given the informal way that the stay applications were made, and that the Plaintiffs were not given a proper opportunity to deal with them, I gave directions for the parties to file written submissions respectively in support and in response to these applications. The parties have duly done so.[40] 235. Further, the court has also granted retrospective leave to the Plaintiffs to file the 2nd affirmation of Ms Wong Hwa Yih (a senior Assistant Law Officer) for this purpose. Ms Wong deposes that: (1) The CE and the Hong Kong Special Administrative Region Government have the constitutional responsibility to uphold and implement the BL, to execute the relevant laws under the BL and to safeguard public interest, and the SJ as the guardian of public interest has similar constitutional duties commensurate with his official position. (2) The Plaintiffs’ reasons for deciding to commence proceedings on the four cases and these four cases alone after the handing down of the CA Judgment was purely based on advice as to legal merits (after consulting external independent counsel and senior counsel), without any political consideration. (3) This indeed has all along been stated in public statements by the SJ, the Government and the CE (the public statements are exhibited in Mr Wong’s 2nd affirmation). 236. I will now turn to consider the merits of the stay applications. 237. The Defendants’ submissions in support of the stay applications are essentially these: (1) It is an abuse of legal power and process for political considerations: Judicial Review Handbook at paragraph 52.2. In particular, it is submitted that the court should not allow its process to be used by the CE and the SJto advance sectional interests so that a political advantage is gained by selective disqualification. Cf: Porter v Magill [2002] 2 AC 357, at paragraph 19(5), per Lord Bingham. (2) In the present context, there are many other members-elect (“the Other Members-Elect”) who took the LegCo Oath on 12 October 2016 in similar ways, conducts and manner as the Defendants did,[41] which ways, conducts and manner are now subject to the Plaintiffs’ complaint in these proceedings for legal non-compliance. In particular, a number of the Other Members-Elect read out messages and slogans not related to the oath before and after the oath itself. Notwithstanding the close similarities in the way in which the Other Members-Elect and the Defendants took the LegCo Oath, the Plaintiffs only selectively brought the present disqualification proceedings against the Defendants but not the Other Members-Elect. (3) The Plaintiffs have not explained why they have not brought similar proceedings against the Other Members‑Elect. In the circumstances, in particular given the Defendants’ identification of the Other Members-Elect, the court can infer that the Government has no good reason if she gives no explanation for taking a particular course if all the circumstances of the case seem to put in favour of taking a certain course. Cf: Padfield v Minister of Agriculture, Fisheries and Food [1968] AC 997 at 1053G-1054A, per Lord Pearce. (4) In this respect: (a) Mr Law further says it is obvious in these circumstances, in particular in the lack of any explanations from the Government, that the Government is “politically motivated” in only bringing the proceedings against the Defendants but not the others so as to disqualify so many pro‑democrats as would destroy their majority in the LegCo in Geographical Constituencies.[42] In support, Mr Law has filed his 3rd affirmation[43] disposing that the pro-democrats and non-establishment camp have a majority of 17:16 in the Geographical Constituencies, and that if he, Mr Leung, Ms Lau (excluding Mr Yiu who belongs to the functional constituency) are also disqualified, the pro-democracy and non-establishment camp would lose their 17:16 majority in the Geographical Constituency. (b) Mr Lau asserts it is inevitable to infer in the absence of an explanation that these proceedings were brought against only the four Defendants for an improper purpose, namely succumbing to undue political pressure. She says the inference of political pressure is fortified in light of the fact that the Interpretation was issued in the middle of the judicial trial of the disqualification proceedings which eventually resulted in the CFI Judgment, and the Explanations that accompanied the Interpretation have gone far beyond the Interpretation.[44] (c) Mr Yiu says that the Plaintiffs selectively targeted the four Defendants for purposes they (the CE and the SJ) do not want to make clear to the public. Mr Yiu further says the 2nd affirmation of Ms Wong Hwa Yih in fact shows that the Plaintiffs simply could not come up with any good reasons that they could disclose to say why only the Defendants have been targeted for disqualification, as the affirmation (including the press statements) contains nothing but bare denials and self‑serving assertions that they have relied on legal advice to sue only the Defendants. It must be noted (as emphasized by all the Defendants in their submissions) that the Plaintiffs have chosen not to waive the legal privilege (which they could have) and disclose the purported legal advice to support the explanations.[45] (d) Mr Leung effectively agrees and adopts the above submissions of the other Defendants and says, in the absence of an explanation, the Plaintiffs’ decision to sue only the four Defendants was tainted with political considerations or arbitrary. He further submits that this is particularly so as these proceedings are “unique” in that the executive (the CE in his personal capacity) and the SJ are suing the legislature viz the President and the Clerk and the four elected members of the LegCo.[46] (5) Mr Leung also argues that the Plaintiffs have extended their case in the hearing against the Defendants (which has not been set out in the Forms 86 and the Originating Summonses) by submitting that messages added before, during or after the reading of the LegCo Oath form part of the oath itself, amounting to a contravention of Exact Form and Content Requirement. This by itself is an abuse of process.[47] 238. I am not persuaded that the commencement of the present proceedings amount to an abuse of process as contended by the Defendants. My reasons are as follows. 239. As submitted by Mr Yu SC for the Plaintiffs, the burden on proving an abuse is on the Defendants, which burden is a heavy one, and the power to grant a permanent stay is one only to be exercised in the most exceptional circumstances: Williams v Spautz (1992) 174 CLR 509 at 529; Fox v Attorney-General [2002] NZLR 62 at 71. Hence, it has been observed by Sumption JSC in Crawford Adjusters (Cayman) Ltd v Sagicor General Insurance (Cayman) Ltd [2014] AC 366 at paragraph 149 that such cases are extremely rare. 240. Moreover, in considering an application for permanent stay based on using the proceedings for an improper purpose, the law draws a clear and important distinction based on the use of the process for obtaining judgment against the misuse of the process for a predominant purpose other than for which the proceedings are designed. This important distinction has been explained and illustrated in the following authorities. 241. Mason CJ, Dawson, Toohey and McHugh JJ in Williams v Spautz, supra, at 526 gave the example as follows: “Thus, to take an example mentioned in argument, an alderman prosecutes another alderman who is a political opponent for failure to disclose a relevant pecuniary interest when voting to approve a contract, intending to secure the opponent's conviction so that he or she will then be disqualified from office as an alderman by reason of that conviction, pursuant to local government legislation regulating the holding of such offices. The ultimate purpose of bringing about disqualification is not within the scope of the criminal process instituted by the prosecutor. But the immediate purpose of the prosecutor is within that scope. And the existence of the ultimate purpose cannot constitute an abuse of process when that purpose is to bring about a result for which the law provides in the event that the proceedings terminate in the prosecutor's favour.” 242. Similarly, Lord Wilson JSC and Sumption JSC respectively explained this in Crawford Adjusters (Cayman) Ltd, supra, at paragraphs 63 and 149: “63. What is an improper purpose? A helpful metaphor suggested by Isaacs J in the High Court of Australia in Varawa v Howard Smith Co Ltd (1911)13 CLR 35, 91, is that of a stalking-horse: ‘If the proceedings are merely a stalking-horse to coerce the defendant in some way entirely outside the ambit of the legal claim on which the court is asked to adjudicate they are regarded as an abuse of process for this purpose ...’ The metaphor aids resolution of the conundrum raised by the example of a claimant who intends that the result of the action will be the economic downfall of the defendant who may be a business rival or just an enemy. If the claimant’s intention is that the result of victory in the action will be the defendant’s downfall, then his purpose is not improper: for it is nothing other than to achieve victory in the action, with all such consequences as may flow from it. If, on the other hand, his intention is to secure the defendant’s downfall – or some other disadvantage to the defendant or advantage to himself – by use of the proceedings otherwise than for the purpose for which they are designed, then his purpose is improper.” “149 … The essence of the tort is the abuse of civil proceedings for a predominant purpose other than that for which they were designed. This means for the purpose of obtaining some wholly extraneous benefit other than the relief sought and not reasonably flowing from or connected with the relief sought. The paradigm case is the use of the processes of the court as a tool of extortion, by putting pressure on the defendant to do something wholly unconnected with the relief which he has no obligation to do.” (emphasis added) 243. Recently in Melbourne City Investments Pty Ltd v Treasury Wine Estates Ltd (2016) 243 FCR 474, Foster J at paragraphs 117 - 119 referred to Williams v Spautz for the principle that it is the use of the proceedings which must be examined and which may constitute an abuse of process. The learned judge, after considering a line of authorities, then concluded at paragraphs 147(d) and (f) that: (1) An improper purpose is a purpose to use a proceeding as a means of obtaining some advantage for which that proceeding is not designed. It is the use of the proceedings which must be examined and which may constitute an abuse of process. The purposes which legal proceedings are designed to serve are the protection or vindication of particular rights or immunities, the maintenance or affection of particular legal relationships and the imposition of particular legal penalties, liabilities and obligations. The pursuit of a legal remedy is not converted into an abuse of process merely by an unworthy or ulterior motive. (2) The onus of proving an abuse in any given case rests upon the party alleging abuse. That onus is a heavy one. 244. Bearing in mind the above relevant principles and fundamental distinction, I accept the Plaintiffs’ submissions that the Defendants have not discharged the burden to show that it is an abuse of process for the Plaintiffs to bring the present disqualification proceedings against the Defendants. 245. As emphasised by Mr Yu, it is for the Defendants to show that there had been an abuse as contended, and prima facie not for the Plaintiffs to explain why the commencement of the proceedings was not an abuse. 246. In any event, it is incorrect for the Defendants to say that the Plaintiffs have not explained why the proceedings were brought against the Defendants. As set out in Mr Wong’s 2nd affirmation, the SJ, the CE and the Government have issued statements as early as 1, 2 and 6 December 2016 repeatedly explaining and emphasizing that the only standard to be applied in taking action was based on legal advice including advice from independent counsel, and there will not be and has not been any political consideration. 247. As the Plaintiffs have submitted, a case to disqualify a LegCo member has to be clear and the court looks at the totality of evidence in each case. It is the Plaintiffs’ case that in their assessments, these four cases are clear because they involve a number of features which, taken cumulatively in each case, clearly establish a case for disqualification. They also contain features over and above and are different from other cases such as (as pointed out by Mr Yu): Ms Lau’s Opening Statement and slow reading, Mr Law’s Opening Statement demonstrating his insincerity and his intentional intonation, Mr Yiu’s repeated insertion of the Extra Words as part of his oath, Mr Leung’s theatrical performance showing his insincerity and his use of props and theatrical conduct as and when he was taking the oath forming part and parcel of the same. The Plaintiffs therefore submit they formed the view on legal advice that these four cases best satisfied the clear ones, and that is entirely reasonable for them to act on legal advice as to the merits of each case and to proceed only on cases which are regarded as clear. 248. Mr Yu further emphasizes that the pursuit of clear cases by the CE or the SJ meets or accords with the Defendants’ submissions on the cogency of evidence required. The CE or the SJ recognize the need for a high threshold and adopt a prudent approach. This is, says Mr Yu, the antithesis of an abuse. 249. In this respect, it must be noted that the court must consider each case on its own facts and against its own context. As made clear in the CA Judgment (at paragraph 39): “In the final analysis, what is at stake is the compliance of a constitutional requirement of great significance. In any given set of facts, this can admit of one correct answer only.” (emphasis added). The Defendants never contend that the court must necessarily reach the same “correct answer” in each of the four cases here. The case against each of them differs, and they retain different legal teams and advance different defences. It is not the Defendants’ case that either the court should rule against all four Defendants or else it should rule against none of them. 250. The court also notes that it is not the Defendants’ case that, if the CE or the SJ were to pursue the other LegCo members, the court must necessarily reach the same “correct answer” and either rule against all these LegCo members (including the Defendants) or else rule against none of them. 251. In the premises, when it is looked at in the above way, I am of the view that the Defendants cannot fault the Plaintiffs’ prima facie explanation that the decisions to bring the present proceedings against them but not the Other Members‑Elect were based on legal advice on merits. 252. But more importantly, the court is not persuaded by the Defendants’ contentions as summarized above to say why the present proceedings brought against them are an abuse of process. This is what I now turn to look at. 253. With respect, the Defendants’ principal basis for saying that the proceedings were brought for political purposes is at best speculative and in any event does not hold out: (1) Mr Law’s above contention (which is shared by Mr Leung and Ms Lau) on political motivation is, as submitted by Mr Yu, a non sequitur. If there was indeed a political motive in seeking disqualification of pro-democrats in order to secure a pro-establishment majority in the Geographic Constituency as alleged, Mr Law’s theory does not explain why the Plaintiffs would have to be selective and only proceeded against the four Defendants, but not the Other Members-Elect, who could generally also be regarded belonging to the “pro-democrats” and “non-establishment” camp. The theory also could not explain why the Plaintiffs at the same time also commenced the proceedings against Mr Yiu who, as pointed by Mr Law himself, belongs to the functional constituency. (2) The same applies to Ms Lau’s allegation that the proceedings were issued under “political pressure”. If this were correct, it is difficult to understand why the same alleged “political pressure” would not have caused the Plaintiffs to seek to disqualify the Other Members-Elect as well. (3) It must be remembered that the burden is on Defendants to prove abuse. Mere speculation based on a prima facie logically defective theory does not suffice. (4) Mr Yiu’s purely speculative, vague and unsupported submission that “it appears the Plaintiffs selectively targeted the four Defendants for purposes they (the CE and the SJ) do not want to make clear to the public” also simply could not discharge the burden to show abuse. (5) Mr Leung’s additional submission to say that the proceedings are brought for political considerations because it is “unique” for the executive (being the CE and the SJ) to sue the legislative (viz the President and the four elected members-elect) is, with the greatest respect, without any substance. As a matter of proper legal procedures and rights, the CE and the SJ are the proper parties to bring these proceedings concerning the constitutional compliance by members-elect of the LegCo to take the LegCo Oath. See: the CA Judgment, paragraphs 48 - 50, per Cheung CJHC. (6) Moreover, as emphasized by Poon JA in the CA Judgment at paragraph 87, elected members-elect of the LegCo must also comply with the constitutional requirements provided in the BL. When disputes arise as to whether individual LegCo members have breached the constitutional requirements mandated in the BL, the court has a duty to adjudicate and rule on the matters, and in so doing, the court does not seek to undermine the authority or function of the LegCo or diminish the mandate that the electors gave to the LegCo members concerned. Rather, the court ensures that the LegCo or the members concerned exercise their powers lawfully in accordance with the constitutional requirements to maintain the integrity of the LegCo and the confidence of the public in the institution. (7) Hence, the mere fact that disqualification proceedings are commenced by the CE and the SJ (as the proper parties) against elected members-elect of the LegCo and the President or the Clerk in relation to disputes concerning the important question of constitutional compliance in taking the LegCo Oath and in assuming the Office cannot by itself be said to be political in nature. 254. Further, in my view, the above identified falsity of the Defendants’ theories and submissions as to why they say the proceedings against them only are brought with political motivation or improper purpose rather reinforces the Plaintiffs’ explanation that these proceedings were taken out based on legal advice on merits. 255. As summarized at paragraph 237(5) above, the Defendants seek also to rely on the contention that the Plaintiffs had purportedly expanded the case against the Defendants at the hearing to support the submissions that the Plaintiffs had abused the process by targeting only the Defendants but not the Other Members‑Elect (since the “new case” is allegedly equally applicable to the circumstances of the Other Members‑Elect). Even if (which I do not accept) the Plaintiffs did expand their case against the Defendants by contending that an oath taker cannot add any worded messages to the LegCo Oath itself, given that it is the Defendants’ contention that the Plaintiffs only thought of and relied on the “new case” by the time these proceedings have come to the hearing in March 2017, by definition, the Plaintiffs simply could not have recognised that they could pursue against the Other Members-Elect based on the “new case” at the time the commencement of the present proceedings. By March 2017, as pointed out by Mr Dykes in his skeleton at paragraph 16, it would already be too late for the Plaintiffs to bring any disqualification proceedings against the Other Members-Elect on the new ground. This “new case” submission simply does not support an abuse of process as contended by the Defendants. 256. Finally, there is also no evidence to suggest that the Plaintiffs did not intend to bring the proceedings to their conclusion and seek the relief sought. To the contrary, it is indeed clear that the Plaintiffs have proceeded against the Defendants to trial. In the premises, bearing in mind the important distinction in law as emphasised above, purely for the sake of argument, even if (which is not established by the Defendants for the above reasons) the Plaintiffs did so with the object of seeking disqualification thereby achieving some extraneous purpose, this cannot constitute an abuse of process when that alleged purpose is to bring about a result for which the law provides in the event that the proceedings end in favour of the Plaintiffs. 257. In the premises, I am not satisfied that the Defendants have discharged the burden to show a clear case that the commencement of these proceedings against the Defendants is an abuse of process by the Plaintiffs to achieve an ulterior or political motive. 258. I would dismiss the Defendants’ respective applications for a permanent stay or dismissal of the present proceedings. E. CONCLUSION AND DISPOSITIONS 259. For all the above reasons, the Plaintiffs succeed in each of these proceedings respectively against the Defendants. As against each of the Defendants, I will make the following orders. 260. Against Mr Law: (1) Under the judicial review in HCAL 223/2016: (a) A declaration that the President’s Ruling is contrary to law and ultra vires. (b) A declaration that the LegCo Oath purportedly taken by Mr Law on 12 October 2016 is invalid. (c) A declaration that Mr Law has since 12 October 2016 been disqualified from assuming and entering on the Office or has vacated the same. (d) A declaration that the Office purportedly held by Mr Law is now vacant. (e) An order of certiorari to quash the President’s Ruling. (2) Under the Origination Summons in HCMP 3379/2016: (a) Declarations that: (i) The LegCo Oath purportedly taken or made by Mr Law on 12 October 2016 is invalid. (ii) Mr Law has since 12 October 2016 been disqualified from assuming and entering on the Office or has vacated the same. (iii) The Office purportedly held by Mr Law is now vacant. (iv) Mr Law acted as a member of the LegCo and/or claimed to be entitled to act as a member of the LegCo while disqualified from acting in such office. (v) Mr Law is not entitled to act as a member of the LegCo or claim to act as a member of the LegCo. (b) An injunction to restrain Mr Law from acting as a member of the LegCo and claiming to act as a member of the LegCo. 261. Against Mr Leung: (1) Under the judicial review in HCAL 224/2016: (a) A declaration that the Clerk’s Decision is contrary to law and ultra vires. (b) A declaration that the LegCo Oath purportedly taken by Mr Leung on 12 October 2016 is invalid. (c) A declaration that Mr Leung has since 12 October 2016 been disqualified from assuming and entering on the Office or has vacated the same. (d) A declaration that Mr Leung is disqualified as a member of the LegCo and the Office purportedly held by Mr Leung is now vacant. (e) An order of certiorari to quash the Clerk’s Decision. (2) Under the Origination Summons in HCMP 3382/2016: (a) Declarations that: (i) The LegCo Oath purportedly taken or made by Mr Leung on 12 October 2016 is invalid. (ii) Mr Leung has since 12 October 2016 been disqualified from assuming and entering on the Office or has vacated the same. (iii) The Office purportedly held by Mr Leung is now vacant. (iv) Mr Leung acted as a member of the LegCo and/or claimed to be entitled to act as a member of the LegCo while disqualified from acting in such office. (v) Mr Leung is not entitled to act as a member of the LegCo or claim to act as a member of the LegCo. (b) An injunction to restrain Mr Leung from acting as a member of the LegCo and claiming to act as a member of the LegCo. 262. Against Ms Lau: (1) Under the judicial review in HCAL 225/2016: (a) A declaration that the President’s Ruling is contrary to law and ultra vires. (b) A declaration that since 12 October 2016, Ms Lau has been disqualified from assuming and entering on the Office of a member of the LegCo or has vacated the same and is not entitled to take the LegCo Oath afresh. (c) A declaration that the President had no power to administer the LegCo Oath purportedly taken by Ms Lau on 2 November 2016. (d) A declaration that the LegCo Oath purportedly taken by Ms Lau on 2 November 2016 is invalid. (e) A declaration that the Office purportedly held by Ms Lau is now vacant. (f) An order of certiorari to quash the President’s Ruling. (2) Under the Origination Summons in HCMP 3381/2016: (a) Declarations that: (i) The LegCo Oath purportedly taken or made by Ms Lau on 12 October 2016 is invalid. (ii) Ms Lau has since been disqualified from assuming and entering on the Office or has vacated the same. (iii) Ms Lau is not entitled to retake the LegCo Oath. (iv) The LegCo Oath purportedly taken or made by Ms Lau on 2 November 2016 is invalid. (v) The Office purportedly held by Ms Lau is now vacant. (vi) Ms Lau acted as a member of LegCo and/or claimed to be entitled to act as a member of LegCo while disqualified from acting in such office. (vii) Ms Lau is not entitled to act as a member of the LegCo or claim to act as a member of the LegCo. (b) An injunction to restrain Ms Lau from acting as a member of the LegCo and claiming to act as a member of the LegCo. 263. Against Mr Yiu: (1) Under the judicial review in HCAL 226/2016: (a) A declaration that the President’s Ruling is contrary to law and ultra vires. (b) A declaration that Mr Yiu has since 12 October 2016 been disqualified from assuming and entering on the Office of a member of the LegCo or has vacated the same, and is not entitled to take the LegCo Oath afresh. (c) A declaration that the President had no power to administer the LegCo Oath purportedly taken by Mr Yiu on 19 October 2016. (d) A declaration that the LegCo Oath purportedly taken by Mr Yiu on 19 October 2016 is invalid. (e) A declaration that Mr Yiu is disqualified as a member of the LegCo and the Office purportedly held by Mr Yiu is now vacant. (f) An order of certiorari to quash the President’s Ruling. (2) Under the Origination Summons in HCMP 3378/2016: (a) Declarations that: (i) The LegCo Oath purportedly taken or made by Mr Yiu on 12 October 2016 is invalid. (ii) Mr Yiu has been disqualified from assuming and entering on the Office or has vacated the same. (iii) Mr Yiu is not entitled to retake the LegCo Oath. (iv) The LegCo Oath purportedly taken or made by Mr Yiu on 19 October 2016 is invalid. (v) The Office purportedly held by Mr Yiu is now vacant. (vi) Mr Yiu acted as a member of the LegCo and/or claimed to be entitled to act as a member of the LegCo while disqualified from acting in such office. (vii) Mr Yiu is not entitled to act as a member of the LegCo or claim to act as a member of the LegCo. (b) An injunction to restrain Mr Yiu from acting as a member of the LegCo and claiming to act as a member of the LegCo. 264. There are no reasons why costs should not follow the event. I therefore make an order nisi that costs of each of the judicial reviews be to the Plaintiffs and costs of each of the origination summonses be to the SJ, to be taxed if not agreed with certificate for three counsel. Mr Leung’s own costs be taxed in accordance with legal aid regulations. 265. Lastly, I thank counsel’s assistance in these matters. Mr Benjamin Yu, SC (only in opposing the Defendants’ stay application), Mr Johnny Mok, SC, Mr Jimmy Ma & Mr Jenkin Suen, instructed by Department of Justice, for the plaintiffs in HCAL 223-226/2016 and the plaintiffs in HCMP 3378, 3379, 3381, 3382/2016 Mr Philip Dykes, SC, & Mr Jeffrey Tam, instructed by Ho Tse Wai & Partners, for the interested party in HCAL 223/2016 and the defendant in HCMP 3379/2016 Ms Audrey Eu, SC, & Ms Christine Yu, instructed by Khoo & Co, for the interested party in HCAL 226/2016 and the defendant in HCMP 3378/2016 Mr Martin Lee, SC, Mr Carter Chim & Ms Senia Ng, instructed by JCC Cheung & Co, assigned by Director of Legal Aid for the interested party in HCAL 224/2016 and the defendant in HCMP 3382/2016 Mr Johannes Chan, SC (Hon), Mr Hectar Pun, SC, & Mr Anson Wong Yu-yat, instructed by Bond Ng Solicitors, for the interested party in HCAL 225/2016 and the defendant in HCMP 3381/2016 Lo & Lo for the respondent in HCAL 223 - 226/2016, attendance excused [1] The constitution for the Hong Kong Special Administrative Region. [2] Leading counsel for the other Defendants have rightly recognized this position and sought only to reserve their rights to advance similar arguments concerning the effect of the Interpretation on appeal. [3] See The New Shorter Oxford English Dictionary. [4] Mr Lee relies on the accepted practices that can be found in the LegCo’s Rules of Procedure, various past rulings by the President and the “Committee on Rules of Procedure of the LegCo for the Hong Kong Special Administrative Region – Progress Report for the period October 2013 to June 2014” dated 9 July 2014. Some of these have also been referred to and summarized in “A companion to the history, rules and practices of the Legislative Council of the Hong Kong Special Administrative Region” printed by the LegCo Commission and laid on the table of the LegCo on 18 June 2014. See also Leung Kwok Hung’s 2nd affirmation, paragraphs 62 - 79. [5] LC Paper No CB(3) 2/16-17 [6] LC Paper No CB(3) 17/16-17 [7] LC Paper No CB(3) 18/16-17 [8] English translation: Affirmation, English being “affirmation”, has the original meaning in Latin of rendering it more assertive or strengthened. Taking the oath is a solemn ceremony, requiring us to make a pledge to Hong Kong people that we will keep our words and actions as one, to staunchly protect the rights of Hong Kong people. However, this sacred ceremony has today been reduced to a political tool used by those in power to forcefully subject representatives of people’s will to the system and the totalitarian authority. You can chain me, you can torture me, you can even destroy this body, but you will never imprison my mind. Today I must complete the necessary procedure, but this does not mean I am subjugating myself under the totalitarian authority. Hong Kong citizens will forever be the ones whom we serve and unite, I will absolutely not bear allegiance to a political administration which brutally killed its people, I will maintain my principles, and protect Hong Kong people with my conscience. Hope is in the people, change is in resistance. [9] English translation: Civil Disobedience. [10] English translation: End One-Party Dictatorship. [11] English translation: National People’s Congress (“NPC”) 831 Decision. [12] English translation: Umbrella Movement! Indomitable! Civil Disobedience! Without Fear! Self-Autonomy and Self-Determination for People! No Approval from the Communist Party of China is Required! I Want Dual Universal Suffrage! Leung Chun Ying Step Down! [13] English translation: Okay? Use Chinese or English or Shanghainese? I chose Chinese, excuse me. Listen everybody, it is very solemn. Two years ago, people were out there, fighting for dual universal suffrage, and were arrested and beaten up. [14] English translation: I, Leung Kwok Hung, solemnly, sincerely, and truly declare and affirm that, being a member of the Legislative Council of the Hong Kong Special Administrative Region of the People’s Republic of China, I will uphold the Basic Law of the Hong Kong Special Administrative Region of the People’s Republic of China, bear allegiance to the Hong Kong Special Administrative Region of the People’s Republic of China and serve the Hong Kong Special Administrative Region conscientiously, dutifully, in full accordance with the law, honestly and with integrity. [15] English translation: Revoke NPC 831 Decision! I Want Dual Universal Suffrage! [16] English translation: NPC 831 Decision. [17] English translation: Revoke NPC 831 Decision! Revoke NPC 831 Decision! I Want Dual Universal Suffrage! Self-Autonomy and Self-Determination for People! No Approval from the Communist Party of China is Required! [18] Mr Leung deposed at paragraph 87 of his 2nd affirmation that “I have always understood that the LegCo oath has to be taken seriously, solemnly and sincerely…” [19] English translation: I, Lau Siu Lai, hereby make this pledge, that I, having entered the Council from the streets, will certainly carry forth the spirit of self-determination of destiny in the Umbrella Movement, walk with the Hong Kong people, connect the inside and outside of the Council, to fight against totalitarianism. We must live in truth, break the indifference and cynicism, look for hope in the darkness and blaze a trail to democratic self-determination together. Tear down the high wall [for] self-determination and self-reliance. [20] English translation: Fight for universal retirement protection, implement policy for bazaar and staunchly defend the dignity of Hong Kong people in their living. [21] English translation: What I read, was ninety-odd unrelated and independent word sentences, entirely devoid of coherence and meaning. All meaning was merely a guess from the subjective judgment of the audience by making up the clauses themselves. [22] English translation: 【Lau Siu Lai: Slow reading was to manifest the falsehood of the oath】. [23] English translation: 【Lau Siu Lai: Slow reading was to manifest the falsehood of the oath】. [24] English translation: With regard to the oath-taking ceremony yesterday, I would like to tell a story. [25] English translation: A story told by Vaclav Havel. In the Czech Republic in the seventies, a vegetable stall owner put up a banner in the stall: “Unite the workers in the whole world!” Anyone would notice that the vegetable stall owner did not truly believe in that slogan; the reason why he had to display the banner was only because he was afraid of inviting trouble, and so he acted perfunctorily, to maintain “social harmony”. This slogan was clearly hinted by the one in authority; the most artful point was that the vegetable stall owner need not expressly state his fear about the authorities; he could even hide it with a lie: “‘Unite the workers in the whole world’, what’s the problem with that?” [26] English translation: A totalitarian society is maintained by lies. People live in lies and hypocrisy. They may not agree to the lies, but as long as they pretend to agree, or at least remain silent, they can live in the system. Those who choose to obey may even take the initiative to meet the requirements of the system to obtain benefits, in order to satisfy their own desire to power. [27] English translation: In the Chamber yesterday, all members of the LegCo took their oaths. Some patriots in suits, representing the privileged class, read the oaths aloud and openly, and completed the ritual script, as if they were really to “solemnly and sincerely” promise to serve the people. As a result, they became honorable members of the Council, to act as hatchet men of the bigwigs to suppress Hong Kong people, and can even hide it with a he: “To serve the people as a Legislative Councillor, what’s the problem with that?” [28] English translation: The core of the lie is the whole political system of the SAR Government, which has legitimacy on the face of it, but actually there is none! The whole Basic Law, the whole political system together with the Chief Executive, have not been authorized by the public will of the Hong Kong people, and are therefore fundamentally illegal! [29] English translation: A fair and just Council was supposed to be the manifestation of democracy, freedom and people's will. But in Hong Kong, we only have birdcage-style semi-democracy; the various manipulations and disqualification of candidates in this election, trampled the only procedural fairness left. At the election of the President of the LegCo, Andrew Leung’s and the Secretariat’s handling of Leung’s nationality issue was like child’s play, as there were, surprisingly, still a lot of doubtful points on the day of polling. At the end Andrew Leung was “elected” in chaos, but it was difficult to convince the public. [30] English translation: Yesterday, I read out the official oath word-by-word. The oath became ninety-odd sentences devoid of any coherence, without any combination, connection and meaning, so that the audiences could not hear any sentence structure and tone. In this way, all meaning was merely a guess resulting from the subjective judgment of the audiences by making up the clauses themselves. This course of action was to manifest the hypocrisy of the ritual. The pro‑establishment camp including Andrew Leung, raved that any members of the opposition camp who tried to obstruct Andrew Leung from being elected as the President smoothly were “creating troubles” and “causing damage to the Council”, but I want to convey a message instead: fluent and sonorous oath is hypocritical, a harmonious Council is also hypocritical. [31] English translation: Vaclav Havel said, to fight totalitarianism, the most fundamental way is to smash the lies and hypocrisy that maintain the totalitarianism. Only when everyone insist on not declaring their stance in accordance with the conventions, and say their own true thoughts, would it be possible for the high wall to be broken. Yesterday was just a beginning, from now on I will walk with the Hong Kong people, inside and outside of the Council, to strive to fight against the hypocrisy of various institutional requirements, to accomplish what Vaclav Havel said – living in truth, and to blaze a trail to democratic self-determination together. [32] English translation: Lastly, the speech I made before reading out the official oath was a more sincere version: I, Lau Siu Lai, solemnly and sincerely, pledge to the Hong Kong people that: I, having entered the Council from the streets, will certainly carry forth the spirit of self-determination of destiny in the Umbrella Movement, walk with the Hong Kong people, connect the inside and outside of the Council, to fight against totalitarianism. We must live in truth, break the indifference and cynicism, look for hope in the darkness, and blaze a trail to democratic self-determination together. Tear down the high wall [for] self-determination and self-reliance! [33] English translation: Ifthere is a 6-second pause in between each word, how can you view them as one sentence? They could be ninety-odd pieces of articles! [34] The underlined words being words added to the content of the prescribed oath under section 16(d) of and Schedule 2to the ODO. The words, translated into English means, “I will protect the justice system in Hong Kong, fight for true democracy, and serve Hong Kong for its sustainable development”. [35] The underlined words being words added to the content ofthe prescribed oath, carrying the same translated meaning as that in Footnote 34 above. [36] Translated as: I will protect the justice system in Hong Kong, fight for true democracy, and serve Hong Kong for its sustainable development. [37] See paragraphs 20, 24(3) and 26(2). [38] See paragraphs 42 - 44, 53(6) and 56 of Mr Yiu’s affirmation. [39] Mr Dykes did in the course of the hearing indicate that he might consider making such an application after hearing Mr Mok’s reply submissions to see if Mr Mok would provide any explanations as to why the Plaintiffs had commenced the disqualification proceedings against only the Defendants. [40] It should be noted that despite the clear purpose that these supplemental submissions are only to deal with the stay applications, Mr Lee has at paragraphs 6 - 13 of the written submissions dated 17 March 2017 and paragraphs 5 - 7 of the reply submissions dated 12 April 2017 made substantial further and new submissions based on an argument of legitimate expectation to oppose the substantive applications against Mr Leung under the judicial review and the Originating Summons (in substance that the past practices of the President and the Clerk in accepting the way and manner in which Mr Leung and other LegCo members-elect had taken the LegCo Oath in the past have created a legitimate expectation on the part of Mr Leung that his present oath taking would not be treated as unlawful. In the premises, he should be given prior notice for any change in the benchmark for measuring the degree of solemnity required for taking the LegCo Oath). This is clearly in breach of the directions and should not be condoned. Mr Lee should not be allowed to have a second bite of the cherry in his submissions. The court therefore would disregard this part of the submissions. In any event, the court would accept Mr Yu SC’s submissions set out in the appendix to his written submissions dated 31 March 2017 to reject the argument based on legitimate expectation on principal ground that an expectation is not legitimate or cannot be substantively enforced if the substantive enforcement would be contrary to law. [41] See paragraph 3 of Mr Leung’s written submissions listing a total of 18 such Other Members‑Elect. Mr Law has identified seven (which overlap with the 18 identified by Mr Leung) at paragraphs 25 - 26 of his affirmation. Ms Lau relies on the same as identified by Mr Law (see paragraph 13 of Ms Lau’s written submissions). Mr Yiu has identified five such Other Members-Elect (which also overlap with the 18 identified by Mr Leung) at paragraph 35 of his affirmation. [42] See paragraph 17 of Mr Law’s written submissions dated 17 March 2017, and paragraph 22 of his reply submissions dated 12 April 2017. [43] The court would also grant retrospective leave for the filing of Mr Law’s 3rd affirmation dated 12 April 2017. [44] See paragraphs 13 - 14 of Ms Lau’s written submissions dated 17 March 2017, and paragraphs 5 - 13 of her written submissions dated 12 April 2017. [45] See paragraph 8 of Mr Yiu’s written submissions dated 17 March 2017, and paragraphs 12 and 13 of the reply submissions dated 12 April 2017. [46] See paragraphs 2 and 5 of Mr Leung’s written submissions dated 17 March 2017 and paragraph 3 of the reply submissions dated 12 April 2017. [47] See paragraph 1(a) of Mr Leung’s written submissions. 
Chief Justice Ma: 1. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Ribeiro PJ: 2. This appeal is concerned with the protection conferred by Article 6 and Article 105 of the Basic Law on private property rights in connection with planning restrictions laid down by the Town Planning Board (“the TPB” or “the Board”). It also falls to the Court to consider how the principle of proportionality is applied in that context. A. The proceedings in the courts below A.1 Hysan’s proceedings 3. The appellants in FACV 21 and 22 of 2015 are all companies in the Hysan Development Company Limited group and are collectively referred to in this judgment as “Hysan”. 4. Hysan owns extensive and substantial properties on various sites in Causeway Bay and Wanchai. It objected to a series of planning restrictions contained in two draft Outline Zoning Plans[1] (“OZPs”) gazetted by the Board. Those restrictions (to which the relevant sites had not previously been subjected) comprised: (a) Building height restrictions (“BHRs”) which limited the heights of buildings to 130 mPD in most cases;[2] (b) Podium height restrictions (“PHRs”) setting height limits of 32 mPD and 20 mPD on podiums at particular sites; (c) Non-building areas (“NBAs”) of specified widths (2m and 5m at certain sites) prohibiting construction above ground in the designated areas; and (d) Building setbacks of specified widths (1.5m and 0.5m in certain instances). 5. The Board stated that the purpose of the NBAs, PHRs and setbacks was to facilitate air ventilation and pedestrian traffic flow in what are crowded and densely built-up areas. It rejected all of Hysan’s representations against the restrictions with the exception of agreeing to increase the BHR limit from 100 mPD to 130 mPD for one site. Hysan brought judicial review proceedings to challenge the Board’s decisions rejecting its representations. 6. At first instance, Reyes J[3] dismissed Hysan’s application save in respect of its challenge to the imposition of a 5m NBA in one location. Hysan’s appeal was allowed by the Court of Appeal[4] on various administrative law grounds.[5] It granted Orders of Certiorari quashing the Board’s decisions and Orders of Mandamus directing it to reconsider its decisions in accordance with the Court of Appeal’s decision. A.2 OGL’s proceedings 7. Oriental Generation Limited (“OGL”) intended to redevelop a site occupied by dilapidated buildings known as Kai Tak Mansion, having entered into agreements to purchase over 80% of the undivided shares held by owners of units in those buildings. OGL submitted to the Building Authority building plans for two towers with a height of 203 mPD which were consistent with the OZP then applicable. However, the Building Authority rejected those plans because they were inconsistent with draft OZPs[6] gazetted by the TPB shortly afterwards imposing three new restrictions which comprised: (a) a 110 mPD BHR (subsequently raised by the Board to 130 mPD); (b) a 10m NBA on two boundaries of the site; and (c) a 20m wide Building Gap through the middle of the site. 8. In judicial review proceedings before Reyes J,[7] OGL succeeded in having the three new restrictions quashed on the ground that they were arbitrary. The Court of Appeal dismissed the Board’s appeal.[8] B. The issues on this Appeal 9. Hysan and OGL had both unsuccessfully contended below that the planning restrictions represented a disproportionate and therefore unconstitutional infringement of their property rights in contravention of Articles 6 and 105 of the Basic Law. Although they had succeeded in having the TPB’s decisions quashed on traditional judicial review grounds, they sought leave to appeal to this Court on the constitutional issues with a view to ensuring that, when reconsidering its decisions regarding the restrictions, the Board would have the guidance of this Court’s judgment as to the relevance and application of Articles 6 and 105. 10. On 18 November 2015, the Appeal Committee[9] granted Hysan leave to appeal on the following question:[10] “Whether in the determination of the lawfulness and validity of any restriction imposed by the Respondent by way of planning, Article 6 and/or Article 105 of the Basic Law are engaged, and if so, whether such restriction must satisfy the requirement of proportionality, and whether the Hong Kong Court should adopt the European jurisprudence on Article 1 of the First Protocol of the European Convention on Human Rights or some other test of proportionality, and if so, what.” 11. OGL was granted leave to intervene in Hysan’s appeal to enable it to be heard on the aforesaid question which is also relevant to OGL’s remitter. 12. The questions which fall to be answered on this Appeal are therefore as follows: (a) Are Articles 6 and 105 of the Basic Law engaged where landowners complain about planning restrictions imposed by the Board on the use of their land? (b) If so, must the restrictions be subjected to a proportionality analysis? (c) If so, what standards or tests should the Court apply in conducting a proportionality assessment in a case like the present, and in this context, is the jurisprudence of the European Court of Human Rights (“ECtHR”) on Article 1 of Protocol 1 (“A1P1”)[11] of the European Convention on Human Rights (“ECHR”) of assistance? C. The Basic Law Articles concerned 13. Article 6 states: “The Hong Kong Special Administrative Region shall protect the right of private ownership of property in accordance with law”. 14. Article 105 (so far as presently relevant) is in the following terms: “The Hong Kong Special Administrative Region shall, in accordance with law, protect the right of individuals and legal persons to the acquisition, use, disposal and inheritance of property and their right to compensation for lawful deprivation of their property. 15. Article 7 has also been cited. It provides: “The land and natural resources within the Hong Kong Special Administrative Region shall be State property. The Government of the Hong Kong Special Administrative Region shall be responsible for their management, use and development and for their lease or grant to individuals, legal persons or organizations for use or development. The revenues derived therefrom shall be exclusively at the disposal of the government of the Region.” D. Are Articles 6 and 105 engaged? D.1 Reyes J’s approach 16. In Hysan’s case,[12] Reyes J understood Hysan to be contending that the planning restrictions “constitute a deprivation of property” in violation of Articles 6 and 105 on the basis that the restrictions “would have a negative impact of billions of dollars on the value of those sites”.[13] He held in effect that Articles 6 and 105 were not engaged, stating: “... the mere fact that zoning restrictions imposed in the public interest will lead to a diminution of property values will not, without more, amount to an unlawful deprivation of property contrary to the Basic Law. A landowner takes property subject to an implied condition that, for the public good, the Government may by regulation (including OZPs) limit the uses to which such land can be put in the future.”[14] 17. While this was said in relation to “deprivation”, the point, if valid, would be equally applicable to “restriction of use”. The proposition is that statutory planning restrictions inherently form part of the property rights acquired by the landowner so that such restrictions, whenever imposed, do not (“without more”) infringe Articles 6 and 105. D.2 The Court of Appeal’s approach 18. In Hysan’s appeal,[15] the Court of Appeal recognized that property rights of landowners are not absolute[16] and that the protection conferred by Article 105 extends beyond cases of deprivation without compensation to cover the rights of individuals to acquire, use, dispose of and inherit property.[17] However, it held that Articles 6 and 105 were not engaged in the context of the Board’s planning restrictions. 19. The Court interpreted the words “in accordance with law” in Articles 6 and 105 as qualifying or restricting the protection conferred so that such protection “is only to the extent that such acquisition, use, disposal and inheritance is ‘in accordance with law’”.[18] Thus, in distinguishing the case-law of the ECtHR on A1P1, the Court of Appeal stated: “... unlike our Article 105, [in A1P1] there is no qualification of the relevant right as being ‘in accordance with law’ under the first rule...”[19] It added: “In contrast, the right to use one’s property which is protected under Article 105 is subject to the rider of ‘in accordance with law’.”[20] 20. The proposition is therefore that the words “in accordance with law” limit the protection conferred by Articles 6 and 105 exclusively to a requirement that property rights be protected by legally certain and accessible laws. 21. The Court of Appeal, echoing Reyes J in Hysan’s case, proceeded additionally to decide that the protections conferred by Articles 6 and 105 are not engaged in respect of planning restrictions imposed on Hysan and OGL because their rights as property owners are intrinsically defined by and subject to legal restrictions, including the power of the Board to impose such planning restrictions, as incidents of their ownership in accordance with the general law, so that the restrictions whenever imposed do not represent incursions into constitutionally guaranteed rights and thus do not need to be justified. 22. It cited in support Fine Tower Associates Ltd v Town Planning Board,[21]where Stock JA said, in the context of rejecting a deprivation claim: “Article 105 of the Basic Law does not sit alone. It is to be read in conjunction with art 7 … There can be no expectation upon the purchase of land that the use permitted by the lease will forever after match the use permitted by town planning regulation. It is an incident of ownership that the uses permitted by the authorities may change. Land is purchased with that knowledge, actual or imputed. The value of these lots upon acquisition were enjoyed under the limitation that is implied by this knowledge …” 23. By way of elaboration, the Court in the Hysan appeal added:[22] “... Apart from lease conditions, restrictions are imposed on the right to use one’s land by the Buildings Ordinance, the TPO[23] and other legislative provisions and common law. ... The same point can be made by reference to the obligations imposed upon a landowner by the law of tort (like the law of nuisance, the principle of Ryland v Fletcher) in terms of the use of the land. Likewise, there are common law and statutory restrictions on the disposal, acquisition and inheritance of land. These are all part and parcel of the general law which, in a sense, prescribes what one can do with one’s land. As a matter of law, an owner of a piece of land in Hong Kong holds a bundle of rights conferred by the general law upon him in relation to the land. These rights are subject to the restrictions imposed by the general law. It follows that a landowner does not have the right to use his land in any manner beyond that permitted by the general law.” D.3 The parties’ submissions on whether the Articles are engaged 24. Lord Pannick QC appearing for the Board,[24] supports the Court of Appeal’s conclusion. He submits that Article 105 is merely “a continuity provision” aimed at preserving Hong Kong’s capitalist economy and, like the Court of Appeal, lays emphasis on the words “in accordance with law”. Indeed, Counsel accepts that if that phrase had been absent, there would be no basis for contending that Articles 6 and 105 are not engaged. 25. The Board’s submission is thus that the protection of property rights conferred by the Articles is limited to a requirement that such rights be protected by legally certain provisions. The point as put in the Board’s printed case is that Article 105 only: “... confers a limited right: a right to compensation for deprivation of property, and a right that any interference with property must be set out in an Ordinance, regulations or some other legal instrument.”[25] 26. In aid of that interpretation, Lord Pannick QC points out that Article 105 is not in Chapter III of the Basic Law dealing with Fundamental Rights; that under Article 7, the HKSAR retains responsibility for the “management, use and development” of land in Hong Kong; and that no provision is made for compensation for interference with land short of expropriation. 27. Mr Benjamin Yu SC,[26] on the other hand, asserts that Articles 6 and 105 are self-evidently engaged and seeks to refute the arguments in support relied on by the Board. Mr Nigel Pleming QC,[27] adopts Mr Yu SC’s submissions on this point. D.4 Conclusions as to engagement of Articles 6 and 105 28. In my view, Articles 6 and 105 are plainly engaged and the Court of Appeal fell into error in holding otherwise. D.4a “In accordance with law” 29. The central feature of Articles 6 and 105 is that they impose an obligation on the HKSAR to protect private property rights. Thus, Article 6 states that “The Hong Kong Special Administrative Region shall protect the right of private ownership of property...” And Article 105 expands on the obligation, stating: “The Hong Kong Special Administrative Region shall ... protect the right of individuals and legal persons to the acquisition, use, disposal and inheritance of property and their right to compensation for lawful deprivation of their property.” 30. Articles 6 and 105 stipulate that the obligation is to be discharged by providing such protection “in accordance with law”. That phrase and similar phrases such as “prescribed by law” and “according to law”, appear in numerous Articles of the Basic Law and the Bill of Rights. It is well-established that they mandate the principle of legal certainty, requiring the subject-matter of the Article to be regulated by laws which are accessible and precisely defined.[28] It follows that the phrase introduces another aspect of protection: Property rights are to be guaranteed by clear and accessible laws, and not, for instance, left to uncharted administrative discretion. 31. In many other Articles of the Basic Law, the phrase functions in the same way, stipulating a requirement for legally certain means for protecting rights and freedoms. To take just one example, Article 4 provides: “The Hong Kong Special Administrative Region shall safeguard the rights and freedoms of the residents of the Hong Kong Special Administrative Region and of other persons in the Region in accordance with law.” This appears alongside Article 6 in Chapter I which lays down General Principles. It obliges the HKSAR to safeguard a broad spectrum of rights and freedoms in clear and accessible terms. 32. I can therefore see no basis for reading the words “in accordance with law” as qualifying or limiting the protection conferred by Articles 6 and 105. Such an interpretation of the Articles in effect involves the unwarranted insertion of the word “only” before the phrase in question. Far from diminishing the protection of those Articles, the phrase confers the added protection of legal certainty. 33. The ancillary points relied on by the Board do not impair this conclusion. Whether a Basic Law provision confers constitutional protection on any rights depends on its proper interpretation and not merely on the Chapter heading of the section in which it is located. Articles 6 and 105 expressly confer protection on private property rights and it is nothing to the point that they are not located in Chapter III dealing with Fundamental Rights. Various Articles conferring guaranteed rights can be found in other Chapters. For instance, the important constitutional rights to a fair trial and to the presumption of innocence are found in Article 87 in Chapter IV, Section 4 which deals with the Judiciary. And Article 121 conferring protection on land leases is found in Chapter V, Section 2. Article 105 is likewise found in Chapter V.[29] 34. Contrary to the view taken by the Court of Appeal in Fine Tower Associates Ltd v Town Planning Board,[30]Article 7 does not bear on the question of whether Articles 6 and 105 are engaged. It lays down the general principle that land and natural resources in the HKSAR are State property and allocates the responsibility and revenue from their management, use, development and leasing, etc, to the HKSAR Government. Where the Government leases or grants land to some person, Article 6 plays the complementary role of guaranteeing protection of the private property rights acquired. 35. Neither does the fact that Article 105 makes no provision for compensation for interference with land short of expropriation have any present relevance. Conferment of a right to compensation in deprivation cases does not diminish the protection conferred against other forms of interference with the right to acquire, use, dispose of and inherit property. D.4b Intrinsic restrictions 36. The Board’s other argument,[31] which I also do not accept, is that Hysan’s and OGL’s property rights do not attract the protection of Articles 6 and 105 against planning restrictions because their rights are intrinsically subject to the imposition of such constraints so that the restrictions should not be regarded as encroachments on rights guaranteed by those Articles. 37. In the present case, the developers’ challenge is not to the constitutionality of the TPO itself. The challenge is to the Board’s decisions taken in the purported exercise of statutory powers conferred by that Ordinance. The restrictions complained of are new, constraining the use by Hysan and OGL of the land in ways which did not previously apply while the sites were in their ownership. In these circumstances, I do not accept that the more restrictive BHRs, PHRs, NBAs, setbacks and building gap imposed by the Board can be disregarded as mere incidents of ownership which do not engage the Articles 6 and 105 rights. 38. An argument similar to the Board’s was not accepted by the ECtHR in JA Pye (Oxford) Ltd v United Kingdom.[32] The Chamber rejected the suggestion that loss of the applicant’s land to a neighbouring land owner by adverse possession was merely an incident of the land’s ownership so that A1P1 was not engaged. It stated: “... the pre-existing rules on adverse possession could not be said to be an incident of the applicant companies' property right at the time of its acquisition such that Art 1 ceased to be engaged when the relevant provisions took effect and the property right was lost after 12 years' adverse possession.”[33] 39. This was upheld by the Grand Chamber: “... Where there is a dispute as to whether an applicant has a property interest which is eligible for protection under Art.1 of Protocol No.1, the Court is required to determine the legal position of the applicant. In the present case, the applicant companies were the beneficial owners of the land in Berkshire, as they were successive registered proprietors. The land was not subject to a right of pre-emption, as in the case of Beyeler,[34] but it was subject to the ordinary law of the land, including, by way of example, town and country planning legislation, compulsory purchase legislation, and the various rules on adverse possession. The applicant companies' possessions were necessarily limited by the various rules of statute and common law applicable to real estate. It remains the case, however, that the applicant companies lost the beneficial ownership of 23ha of agricultural land as a result of the operation of the 1925 and 1980 Acts. The Court finds inescapable the Chamber's conclusion that Art.1 of Protocol No.1 is applicable.”[35] 40. A fortiori, in cases like the present, the fact that the statutory power to impose planning restrictions existed prior to the owner’s acquisition of the site does not mean that new and more intrusive constraints imposed by a TPB decision made after the land’s acquisition can be disregarded as mere incidents of ownership so as to exclude the protection of Articles 6 and 105. Interference with the owners’ protected rights occurs when the new restrictions take effect, derogating from those rights and thus engaging those Articles. 41. I should however make it clear that I am not excluding the possibility that it may in law be open to an owner to raise constitutional objections based on Articles 6 and 105 in respect of planning restrictions which were in place prior to his acquisition of the land. I wish to leave open such an argument since a similar question was the subject of disagreement in the House of Lords in Wilson v First County Trust Ltd (No 2).[36] There, the issue was whether A1P1 was engaged when a contract of loan made on the security of a car pawned to a pawnbroker was rendered unenforceable by a provision in the English Consumer Credit Act 1974. Lord Hope of Craighead,[37] Lord Hobhouse of Woodborough[38] and Lord Scott of Foscote[39] held that it was not engaged since the Act had invalidated the contract from the outset so that no existing contractual right had been interfered with when the Act operated to make the transaction unenforceable. Lord Nicholls of Birkenhead disagreed, stating: “... The proposition advanced by the Secretary of State [accepted by the three members of the panel referred to above] would mean that however arbitrary or discriminatory such legislation might be, if it was in existence when the transaction took place a court enforcing human rights values would be impotent. A Convention right guaranteeing a right of property would have nothing to say. That is not an attractive conclusion.”[40] 42. The present discussion relates only to challenges made to decisions taken by an executive authority pursuant to statutory machinery, not involving a challenge to the constitutionality of the empowering statute itself. Different types of legislation may give rise to different forms of interference having differing impacts upon property rights. I wish also to leave open the question whether, and if so, in what way Articles 6 and 105 are engaged on a challenge to the constitutionality of such legislation. E. Must the restrictions be subjected to a proportionality analysis? E.1 A criterion for assessing derogations from constitutional rights 43. This Court has recognized that certain constitutionally guaranteed rights, such as the prohibition against torture and cruel, inhuman or degrading treatment or punishment,[41] are absolute and that in such cases, there is no room for any proportionality analysis.[42] 44. Where the guaranteed right is not absolute, the law may validly create restrictions limiting such rights.[43] It is for the Court to determine the permissible extent of those restrictions and it does so by a process referred to as a proportionality analysis. 45. That lawful limitations of guaranteed rights may validly be created is acknowledged by Article 39 of the Basic Law which states: (1) The provisions of the International Covenant on Civil and Political Rights, the International Covenant on Economic, Social and Cultural Rights, and international labour conventions as applied to Hong Kong shall remain in force and shall be implemented through the laws of the Hong Kong Special Administrative Region. (2) The rights and freedoms enjoyed by Hong Kong residents shall not be restricted unless as prescribed by law. Such restrictions shall not contravene the provisions of the preceding paragraph of this Article. 46. Article 39(2) therefore makes it clear (although stating it in negative terms) that such restrictions are permissible but that they must be “prescribed by law”, satisfying the requirement of legal certainty, and must be consistent with the provisions of the specified international instruments as implemented through the laws of the HKSAR. Thus, certain limits are placed on the nature and extent of permissible restrictions by provisions of the Bill of Rights contained in the Bill of Rights Ordinance[44] which implements the International Covenant on Civil and Political Rights in Hong Kong law. 47. Leung Kwok Hung v HKSAR[45] provides an example. This was a case involving a challenge to the constitutionality of a statutory scheme giving powers to the Commissioner of Police to regulate public processions on the basis that the scheme violated the right of peaceful assembly guaranteed by Article 27 of the Basic Law[46] and Article 17 of the Bill of Rights. 48. Article 17 states: “The right of peaceful assembly shall be recognized. No restrictions may be placed on the exercise of this right other than those imposed in conformity with the law and which are necessary in a democratic society in the interests of national security or public safety, public order (ordre public), the protection of public health or morals or the protection of the rights and freedoms of others.” 49. It was held that in accordance with Article 39(2), the right to peaceful assembly both under the Basic Law and the Bill of Rights could be subjected to restrictions provided that two requirements were satisfied: (i) that the restriction must be prescribed by law and (ii) that it must be necessary in a democratic society, with the aim of protecting national security or public safety, etc.[47] The requirement that the measure be “necessary in a democratic society” was interpreted to require application of a proportionality test[48] and the purposes specified (ie, the protection of national security, etc) were held to represent an exhaustive list of the permissible aims of any limitation of the right.[49] 50. Where the constitutional right invoked is not absolute but no express guidance is given by the Basic Law or Bill of Rights as to the allowable limits of derogations from that right, principles have been evolved by the courts as to how the proportionality analysis is to be applied, drawing heavily on the jurisprudence of other jurisdictions.[50] 51. Thus, in HKSAR v Lam Kwong Wai,[51] a reverse onus provision regarding the offence of possession of an imitation firearm was challenged as violating the guarantee of the presumption of innocence in Article 87(2) of the Basic Law[52] and Article 11(1) of the Bill of Rights.[53] Neither of those Articles contains any express provisions regarding inroads into the protected right. Sir Anthony Mason NPJ, writing for the Court, explained the approach to proportionality in the following terms:[54] “In the context of contravention, the first question is whether s.20(3)(c) derogates from the presumption of innocence. If this question is answered in the affirmative, two further questions arise. Stated in accordance with the formulation in Leung Kwok Hung & Others v HKSAR (2005) 8 HKCFAR 229 at p 253I, they are: (1) is the derogation rationally connected with the pursuit of a legitimate societal aim (the rationality test); and (2) are the means employed, the imposition of the reverse persuasive onus, no more than is necessary to achieve that legitimate aim (the proportionality test)?” E.2 A structured, three-step inquiry 52. In subsequent authorities, after a finding that the invoked right is engaged, the two questions referred to by Sir Anthony Mason NPJ have tended to be expressed as a three-step inquiry. Thus, in Mok Charles v Tam Wai Ho,[55] Ma CJ formulated the approach as follows: “The proportionality test, which is a well known test in our courts, consists of the following analysis in respect of any restriction or limitation: (a) The restriction or limitation must pursue a legitimate aim. (b) The restriction or limitation must also be rationally connected to that legitimate aim. (c) The restriction or limitation must also be no more than is necessary to accomplish that legitimate aim.” 53. In their joint judgment in Official Receiver v Zhi Charles,[56] Fok PJ and Stock NPJ summarised the position as follows: “There is a well-established sequence of questions that must be addressed when an issue of constitutionality is raised before a court. The first question is concerned with the identification of a constitutional right and asks whether such a right is engaged. If not, the constitutional challenge fails in limine. The next question is whether the legislative provision or conduct complained of amounts to an interference with, or restriction of, that right. Again, if the answer is no, the challenge fails without further inquiry. If, on the other hand, the answer to that question is yes, then it is necessary to consider whether those rights are absolute, in which case no infringement or restriction is permitted and no question of proportionality arises, or, if not absolute, whether the relevant infringement or restriction can be justified on the proportionality analysis. The proportionality analysis in a case like the present involves asking, first, whether the infringement or restriction pursues a legitimate societal aim; secondly, whether the infringement or restriction is rationally connected with that legitimate aim; and thirdly, whether the infringement or restriction is no more than is necessary to accomplish that legitimate aim.” E.3 Proportionality in the present case 54. In the light of the foregoing, it is clear that a proportionality analysis of the planning restrictions is required in the present case. The right of private property guaranteed by Articles 6 and 105 is engaged. They are Articles which make no express provision regarding permissible restrictions so that the existing approach is for the Court to undertake a three-step inquiry asking (i) whether those restrictions pursue a legitimate aim; (ii) whether they are rationally connected with achieving that aim; and (iii) whether they represent a proportionate means of achieving that end. As discussed below, there is also substantial authority for including a fourth step which involves (iv) weighing the detrimental impact of the restrictions against the social benefit gained. 55. The first two stages of the inquiry have attracted little discussion in the present case. Hysan and OGL have tended to assume (without conceding) that the planning restrictions do pursue a legitimate aim and that they are rationally connected thereto. They are right to do so. There is no suggestion that the Board was acting other than in good faith in the discharge of its duties under the TPO. Its decisions have been successfully challenged on administrative law grounds, but it cannot be suggested that the objectives of lawfully imposed planning restrictions would not provide a legitimate basis for derogating from property rights. The purposes of such restrictions are stated in section 3 of TPO which identifies the Board’s function as the exercise of its town planning powers “with a view to the promotion of the health, safety, convenience and general welfare of the community” which obviously involve legitimate aims. The present planning restrictions aimed at facilitating air ventilation and pedestrian flow and at setting appropriate building heights, if properly arrived at, would clearly have been rationally connected to achieving those objectives. F. Assessing proportionality 56. It is at the third stage of the inquiry, assessing the proportionality of the impugned measure in relation to the legitimate aim, that issue has been joined. It is at that stage that different standards have been evolved by the courts. However, before embarking upon an examination of the different standards, it is appropriate to examine the nature of the structured approach to the assessment and to consider whether it should comprise a three or four-step inquiry. F.1 The adoption of a structured approach in Hong Kong 57. The first case on the Hong Kong Bill of Rights to reach the Privy Council was Attorney-General of Hong Kong v Lee Kwong-kut,[57] which concerned the presumption of innocence and certain reverse onus provisions. It was submitted by counsel that the structured proportionality approach of Dickson CJ in the Canadian Supreme Court in R v Oakes[58] (to which I shall return) should be endorsed. However, Lord Woolf, delivering the Privy Council’s advice, considered it “not necessary, at least in the vast majority of cases, to follow the somewhat complex process”[59] Dickson CJ had adopted. Instead, the Privy Council opted for a broad concept of “reasonableness”, holding that the Court should simply ask itself “whether, under the provision in question, the prosecution is required to prove the important elements of the offence; while the defendant is reasonably given the burden of establishing a proviso or an exemption or the like”[60] and if so, hold that no contravention has occurred. 58. That approach did not take root. In the next (and last) Hong Kong Bill of Rights case to get to the Privy Council, Ming Pao Newspapers Ltd v AG of Hong Kong,[61] Lord Jauncey of Tullichettle who delivered the advice, turned to the ECtHR’s jurisprudence and approved the proposition that “[any] restrictions on the guaranteed right of freedom of expression must be proportionate to the aims sought to be achieved thereby”,[62] noting that “[the ECtHR] accepts that contracting states enjoy a margin of appreciation in determining what is necessary to achieve a legitimate aim.”[63] The Privy Council proceeded to employ a proportionality analysis[64] and concluded that the enactment “cannot be described as ‘so unreasonable as to be outside the state’s margin of appreciation’” adding that their Lordships thought the legislator’s decision “eminently sensible and by no means disproportionate to the important objectives sought to be achieved”.[65] As we have seen, in subsequent cases in the HKSAR,[66] a three-step proportionality analysis has generally been adopted. 59. It is quite understandable that the more structured proportionality approach – what Baroness Hale of Richmond DPSC called “an orderly process of decision-making”[67] – should have been preferred over the relatively amorphous standard of reasonableness. In Bank Mellat v Her Majesty’s Treasury (No 2),[68] (where a four-step approach was adopted) Lord Reed JSC explained why: “Its attraction as a heuristic tool is that, by breaking down an assessment of proportionality into distinct elements, it can clarify different aspects of such an assessment, and make value judgments more explicit.” 60. However, the point to emphasise is that the three (or four) distinct elements of the analysis cannot be treated as existing in isolated airtight compartments, unaffected by each other. Recent authority has highlighted the fluidity and flexibility of the proportionality concept and the fact that its elements are conceptually inter-related and inter-dependent. As Lord Sumption JSC noted in the Bank Mellat case: “The requirements of rationality and proportionality, as applied to decisions engaging the human rights of applicants, inevitably overlap.”[69] 61. His Lordship described his approach to proportionality as one which: “... depends on an exacting analysis of the factual case advanced in defence of the measure, in order to determine (i) whether its objective is sufficiently important to justify the limitation of a fundamental right; (ii) whether it is rationally connected to the objective; (iii) whether a less intrusive measure could have been used; and (iv) whether, having regard to these matters and to the severity of the consequences, a fair balance has been struck between the rights of the individual and the interests of the community.”[70] 62. Lord Sumption JSC accepted that the four requirements are logically separate, but reiterated that “in practice they inevitably overlap because the same facts are likely to be relevant to more than one of them.” He pointed out that “... in the normal case where the effectiveness of the measure and the degree of interference are not absolute values but questions of degree”, those two elements are “inversely related to each other”. In other words, “The question is whether a less intrusive measure could have been used without unacceptably compromising the objective.”[71] 63. In Pham v Secretary of State for the Home Department (Open Society Justice Initiative Intervening),[72] his Lordship further underscored the inter-relationship among elements of the analysis, showing that the importance of the right encroached upon influences the substance of the proportionality assessment: “... I doubt whether it is either possible or desirable to distinguish categorically between ordinary and fundamental rights, applying different principles to the latter. There is in reality a sliding scale, in which the cogency of the justification required for interfering with a right will be proportionate to its perceived importance and the extent of the interference.” F.2 Three-step and four-step approaches 64. The tendency of the Hong Kong courts to adopt a three-step approach to proportionality is in line with the influential 1999 judgment of Lord Clyde in de Freitas v Permanent Secretary of Ministry of Agriculture, Fisheries, Lands and Housing,[73] where his Lordship stated a test which asks: “... whether: (i) the legislative objective is sufficiently important to justify limiting a fundamental right; (ii) the measures designed to meet the legislative objective are rationally connected to it; and (iii) the means used to impair the right or freedom are no more than is necessary to accomplish the objective.” 65. However, a substantial body of authority exists for adding a fourth step, as exemplified by Lord Sumption JSC’s judgment in the Bank Mellat case cited above.[74] Indeed, in common law jurisprudence, a four-step approach can be found dating back to the important judgment in 1986 of Dickson CJ in R v Oakes[75] dealing with proportionality in the context of section 1 of the Canadian Charter of Rights and Freedoms.[76] Section 1 limits derogations from guaranteed rights to those which “can be demonstrably justified in a free and democratic society”. Dickson CJ held that for such justification to be shown, the objective of the restriction had to be of sufficient importance to warrant overriding the guaranteed right; that the means chosen had to be proportionate, balancing the interests of society and individuals or groups; and that such means had to be rationally connected to the objective and such as to impair the constitutional right “as little as possible”. His Honour added that “there must be a proportionality between the effects of the measures which are responsible for limiting the Charter right or freedom, and the objective which has been identified as of ‘sufficient importance’”.[77] As McLachlin J explained in RJR-Macdonald Inc v The Attorney General of Canada,[78] the final stage of the proportionality analysis involves “balancing the negative effects of the infringement of rights against the positive benefits associated with the legislative goal”. And as her Honour there pointed out, it is only necessary to consider this fourth step if the first three requirements of legitimate aim, rational connection and minimal impairment are satisfied. 66. In Huang v Secretary of State for the Home Department,[79] Lord Bingham of Cornhill accepted counsel’s submission that the decision in de Freitas v Permanent Secretary of Ministry of Agriculture, Fisheries, Lands and Housing,[80] was deficient in omitting reference to the fourth step. Referring to “the overriding requirement” in Dickson CJ’s judgment in R v Oakes, his Lordship stated: “This feature is ... the need to balance the interests of society with those of individuals and groups. This is indeed an aspect which should never be overlooked or discounted. The House recognised as much in R (Razgar) v Secretary of State for the Home Department,[81] when, having suggested a series of questions which an adjudicator would have to ask and answer in deciding a Convention question, it said that the judgment on proportionality: ‘must always involve the striking of a fair balance between the rights of the individual and the interests of the community which is inherent in the whole of the Convention. The severity and consequences of the interference will call for careful assessment at this stage.’ (see para 20)” 67. A four-step approach appears now to be generally accepted in the United Kingdom.[82] As Lord Reed JSC puts it in the Bank Mellat case:[83] “... the question at step four is whether the impact of the rights infringement is disproportionate to the likely benefit of the impugned measure.” And as Baroness Hale of Richmond DPSC states in R (Lord Carlile of Berriew) v Secretary of State for the Home Department,[84] the fourth question “can be encapsulated as ‘do the ends justify the means’?” 68. The fourth stage inquiry has also featured prominently in Strasbourg and UK jurisprudence by reference to a concept of “fair balance” in cases on A1P1.[85] I should interject (since it is a matter raised in the question on which leave to appeal was granted) that, subject to the usual need to pay careful attention to any differences that may arise from the language of A1P1 and the context of decisions on that Article, A1P1 jurisprudence is generally of assistance when the protection of property rights comes to be considered. 69. It should be noted that “fair balance” is used in more than one sense in the case-law. It is often said generally to be an objective inherent in the whole of the ECHR[86] and in some A1P1 cases, “fair balance” has been equated with a standard of reasonable necessity in applying a proportionality test.[87] But relevantly for present purposes, “fair balance” has been given a meaning reflecting a fourth step which mandates striking a fair balance “between the demands of the general interest of the community and the requirements of the protection of the individual’s fundamental rights”.[88] In particular, it has been held that: “... the Court must assess whether, taken overall, the applicants can be said to have suffered an ‘individual and excessive burden’.”[89] F.3 Should Hong Kong adopt a four-step inquiry? 70. As we have seen, Lord Sumption JSC noted in Bank Mellat v Her Majesty’s Treasury (No 2),[90] that the elements of the four-step analysis inevitably overlap. There is force in the argument that where a measure has passed the tests set at the first three stages – it pursues a legitimate aim, is rationally connected thereto and is no more than necessary to achieve that aim (a standard which is further discussed below) – such a measure could be expected to reflect a reasonable balance between the general interest of the community and protection of the individual’s fundamental rights. One might therefore doubt the need for the additional step. 71. In Alberta v Hutterian Brethren of Wilson Colony[91] in the Canadian Supreme Court, that issue was addressed by McLachlin CJ: “It may be questioned how a law which has passed the rigours of the first three stages of the proportionality analysis — pressing goal, rational connection, and minimum impairment — could fail at the final inquiry of proportionality of effects. The answer lies in the fact that the first three stages of Oakes are anchored in an assessment of the law’s purpose. Only the fourth branch takes full account of the ‘severity of the deleterious effects of a measure on individuals or groups’. As President Barak explains: Whereas the rational connection test and the least harmful measure test are essentially determined against the background of the proper objective, and are derived from the need to realize it, the test of proportionality (stricto sensu) examines whether the realization of this proper objective is commensurate with the deleterious effect upon the human right. . . . It requires placing colliding values and interests side by side and balancing them according to their weight. [‘Proportional Effect: The Israeli Experience’ (2007), 57 UTLJ 369, at p 374.] In my view, the distinction drawn by Barak is a salutary one, though it has not always been strictly followed by Canadian courts. Because the minimal impairment and proportionality of effects analyses involve different kinds of balancing, analytical clarity and transparency are well served by distinguishing between them. Where no alternative means are reasonably capable of satisfying the government’s objective, the real issue is whether the impact of the rights infringement is disproportionate to the likely benefits of the impugned law. Rather than reading down the government’s objective within the minimal impairment analysis, the court should acknowledge that no less drastic means are available and proceed to the final stage of Oakes.” 72. Lord Reed JSC commented: “In relation to the fourth criterion, there is a meaningful distinction to be drawn (as was explained by McLachlin CJ in Alberta v Hutterian Brethren of Wilson Colony [2009] 2 SCR 567, para 76) between the question whether a particular objective is in principle sufficiently important to justify limiting a particular right (step one), and the question whether, having determined that no less drastic means of achieving the objective are available, the impact of the rights infringement is disproportionate to the likely benefits of the impugned measure (step four).”[92] 73. In my view, the case for accepting in principle the applicability of a fourth step in the proportionality analysis is logically compelling although in the great majority of cases, its application would not invalidate a restriction which has satisfied the requirements of the first three stages of the inquiry. One would hope and expect that most laws and governmental decisions at the sub-constitutional level internally reflect a reasonable balance between the public interest pursued by such laws and the rights of individuals or groups negatively affected by those laws. In such cases, where the law passes the first three tests, it would be unlikely to fail the test of proportionality “stricto sensu” (in the narrow, overall sense) at the fourth stage. But one may exceptionally be faced with a law whose content is such that its application produces extremely unbalanced and unfair results, oppressively imposing excessive burdens on the individuals affected. 74. In his book Proportionality - Constitutional Rights and their Limitations,[93]Professor Aharon Barak, the former President of the Supreme Court of Israel, provides some examples[94] and refers to a hypothetical instance given by Professor Dieter Grimm[95] which illustrates the nature of the test and its potentially moral content as well as its importance: “Assume a law that allows the police to shoot a person (even if this shooting would lead to that person’s death) if it is the only way to prevent that person from harming another’s property. This law is designed to protect private property, and therefore its purpose is proper. The means chosen by the legislator are rational, since it advances the proper purpose. Therefore, the law meets the necessity test as well. However, the provision is still unconstitutional because the protection of private property cannot justify the taking of human life.”[96] 75. As Professor Grimm points out (referring to what I have been calling the “fourth step” as the “third step”): “... the impact of an infringement of a fundamental right can be fully assessed only in the third step. The two previous steps can only reveal the failure of a law to reach its objective; they cannot evaluate the relative weight of the objective of the law, on the one hand, and the fundamental right, on the other, in the context of the legislation under review.”[97] 76. The fourth step therefore requires the Court to examine the overall impact of the impugned measure and to decide whether a fair balance has been struck between the general interest and the individual rights intruded upon, the requirement of such a fair balance being inherent in the protection of fundamental rights.[98] As the ECtHR pointed out in the context of the ECHR in Soering v United Kingdom[99]: “... inherent in the whole of the Convention is a search for a fair balance between the demands of the general interest of the community and the requirements of the protection of the individual’s fundamental rights.” 77. Although this Court has not explored in detail the appropriateness of adding a fourth step, it has on several occasions referred to the need to balance the general interest of the community against the rights of the individual encroached upon. Thus, Leung Kwok Hung v HKSAR,[100] the majority held that in applying a proportionality test, “a proper balance is struck between the interests of society on the one hand and the individual’s right of peaceful assembly on the other.” Similarly, in Koon Wing Yee v Insider Dealing Tribunal (2008),[101] a case concerning the privilege against self-incrimination, Sir Anthony Mason NPJ acknowledged “... the need for a fair balance between the general interest of the community and the personal rights of the individual”, citing Sporrong and Lonnroth v Sweden,[102] a leading case on A1P1, in which the ECtHR endorsed a fourth step test,[103] holding that the Swedish measures had upset the fair balance in that they had caused the applicants to bear “an individual and excessive burden”. His Lordship went on to contrast the case at hand with other cases where it had been held “that the legislation struck a fair balance between protection of an individual’s rights and the public interest...”[104] And in A v Commissioner of Independent Commission Against Corruption,[105] another self-incrimination case, the impugned measure was held to represent “a fair balance between the public interest in realising the legitimate aim of suppressing corruption and protection of the fundamental rights of the individual.” 78. While in the great majority of cases the result arrived at after undertaking the first three inquiries is unlikely to be changed by it, a four-step analysis should, in my view, be explicitly adopted in Hong Kong. Without its inclusion, the proportionality assessment would be confined to gauging the incursion in relation to its aim. The balancing of societal and individual interests against each other which lies at the heart of any system for the protection of human rights would not be addressed. This requires the Court to make a value judgment as to whether the impugned law or governmental decision, despite having satisfied the first three requirements, operates on particular individuals with such oppressive unfairness that it cannot be regarded as a proportionate means of achieving the legitimate aim in question. But that should not cause the Court to shy away from the fourth question since such a value judgment is inherent in the proportionality analysis. As Lord Nicholls of Birkenhead noted in Wilson v First County Trust Ltd (No 2):[106] “The court must decide whether the means employed by the statute to achieve the policy objective is appropriate and not disproportionate in its adverse effect. This involves a ‘value judgment’ by the court, made by reference to the circumstances prevailing when the issue has to be decided.” 79. And as Lord Reed JSC pointed out in Bank Mellat: “An assessment of proportionality inevitably involves a value judgment at the stage at which a balance has to be struck between the importance of the objective pursued and the value of the right intruded upon.”[107] 80. I hasten to add that in holding that a fourth step should be added, I must not be taken to be casting doubt on the correctness of any previous decision involving a three-step assessment. G. What standards should the Court adopt in assessing proportionality? 81. I turn then to the standards applicable at the third stage of the inquiry. Assuming that the Court has found that the impugned measures advance an aim that is legitimate and that they are rationally connected to achieving that aim, what standard (or standards) should it apply in deciding whether those measures are a proportionate means of achieving that aim? How does the Court decide that a particular measure encroaches upon a guaranteed right to an unacceptable extent in promoting a legitimate aim, rendering that measure unconstitutional? 82. The parties have championed two main standards. Hysan and OGL advocate the Court’s adoption of a “no more than necessary” standard while the Board argues for a “manifestly without reasonable foundation” standard. G.1 “No more than necessary” – a standard of reasonable necessity 83. As we have seen, in adopting the proportionality principle, the Hong Kong courts have referred to the third stage of the inquiry as posing the question whether the encroaching measure is “no more than necessary” for advancing the legitimate aim espoused. That is a natural way of expressing the test since it reflects the essential purpose of the exercise: the Court’s endeavour to accommodate acceptable limitations of constitutional rights in the pursuit of a legitimate societal interest while preserving to the maximum extent the guarantees laid down in the constitution. However, the words “no more than necessary” do not lay down a strict, bright line test. They lay down a test of reasonable, not strict, necessity. 84. Thus, in an A1P1 deprivation case, the ECtHR in James v United Kingdom,[108] held that there must be “a reasonable relationship of proportionality between the means employed and the aim sought to be realised”. It rejected the applicants’ argument that “only if there was no other less drastic remedy for the perceived injustice that the extreme remedy of expropriation could satisfy the requirements of Article 1”, commenting: “This amounts to reading a test of strict necessity into the Article, an interpretation which the Court does not find warranted. The availability of alternative solutions does not in itself render the leasehold reform legislation unjustified; it constitutes one factor, along with others, relevant for determining whether the means chosen could be regarded as reasonable and suited to achieving the legitimate aim being pursued, having regard to the need to strike a ‘fair balance’.” 85. The element of reasonableness was emphasised by McLachlin J in RJR-Macdonald Inc v The Attorney General of Canada”:[109] “As the second step in the proportionality analysis, the government must show that the measures at issue impair the right of free expression as little as reasonably possible in order to achieve the legislative objective. The impairment must be ‘minimal’, that is, the law must be carefully tailored so that rights are impaired no more than necessary. The tailoring process seldom admits of perfection and the courts must accord some leeway to the legislator. If the law falls within a range of reasonable alternatives, the courts will not find it overbroad merely because they can conceive of an alternative which might better tailor objective to infringement .... On the other hand, if the government fails to explain why a significantly less intrusive and equally effective measure was not chosen, the law may fail.” 86. In the United Kingdom, as explained by Lord Sumption JSC, the Court adopts a standard relative to the objective pursued, asking “whether a less intrusive measure could have been used without unacceptably compromising the objective”.[110] His Lordship also pointed out that the standard operates in relation to rights viewed as on a “sliding scale” in which “the cogency of the justification required for interfering with a right will be proportionate to its perceived importance and the extent of the interference”.[111] The test is one of reasonableness: “The court must consider whether some less onerous alternative would have been available without unreasonably impairing the objective.”[112] 87. In Hong Kong, in Fok Chun Wa v Hospital Authority,[113] Ma CJ recognized that a strict necessity test would cause problems in practice, citing Blackmun J who pointed out:[114] “... a judge would be unimaginative indeed if he could not come up with something a little less ‘drastic’ or a little less ‘restrictive’ in almost any situation, and thereby enable himself to vote to strike legislation down.” 88. That the test in Hong Kong is one of reasonable necessity was reiterated in Official Receiver v Zhi Charles,[115] a case involving the freedom to travel, where reference was made to the RJR-Macdonald Inc case and Fok PJ and Stock NPJ stated that a minimal impairment test did not mean “that the restriction must be the very least intrusive method of securing the objective which might be imagined or devised”. G.2 “Manifestly without reasonable foundation” G.2a A concept linked to the margin of appreciation 89. The phrase “manifestly without reasonable foundation” is familiar in the jurisprudence of the ECtHR. It is closely related to the concept of “margin of appreciation”. The Strasbourg Court being a supra-national Court, recognizes that in some cases, it is appropriate to regard the national courts of Member States[116] as being better placed to assess the legitimacy and importance of national policy objectives and to determine what encroachments on Convention rights are acceptable as the means of advancing those aims. In such cases, the ECtHR has afforded Member States a wide margin of appreciation and has not regarded encroaching legal measures as disproportionate unless they could be said to be “manifestly without reasonable foundation”. 90. In Buckley v United Kingdom,[117] the ECtHR acknowledged that the scope of the margin of appreciation will vary according to the context, and that relevant factors include the nature of the Convention right, its importance for the individual and the nature of the activities concerned. 91. The ECtHR has, for example, afforded a wide margin of appreciation and adopted a “manifestly without reasonable foundation” threshold in cases involving a Member State’s political and socio-economic policies. Thus, in James v United Kingdom,[118] a case concerned with leasehold enfranchisement legislation which the applicants claimed violated A1P1, the Court stated: “... the decision to enact laws expropriating property will commonly involve consideration of political, economic and social issues on which opinions within a democratic society may reasonably differ widely. The Court, finding it natural that the margin of appreciation available to the legislature in implementing social and economic policies should be a wide one, will respect the legislature’s judgment as to what is ‘in the public interest’ unless that judgment be manifestly without reasonable foundation.” 92. In Lithgow v United Kingdom,[119] the Court explained its approach: “Because of their direct knowledge of their society and its needs and resources, the national authorities are in principle better placed than the international judge to appreciate what measures are appropriate in this area and consequently the margin of appreciation available to them should be a wide one. ... Accordingly, the Court's power of review in the present case is limited to ascertaining whether the decisions regarding compensation fell outside the United Kingdom's wide margin of appreciation; it will respect the legislature's judgment in this connection unless that judgment was manifestly without reasonable foundation.” 93. Accordingly, in such cases, the Member State enjoys a broad margin of appreciation: “... both with regard to the existence of a problem of public concern warranting measures of control and as to the choice of the detailed rules for the implementation of such measures. The Court will respect the legislature's judgement as to what is in the general interest unless that judgement be manifestly without reasonable foundation.”[120] 94. It is relevant to note that the ECtHR has placed town planning decisions within this category. Thus, in Sporrong and Lonnroth v Sweden,[121] the Court: “[found] it natural that, in an area as complex and difficult as that of the development of large cities, the Contracting States should enjoy a wide margin of appreciation in order to implement their town-planning policy.” 95. This has become accepted doctrine, as the Court indicated in Phocas v France:[122] “Like the Government and the Commission, the Court reiterates that the Contracting States enjoy a wide margin of appreciation in order to implement their town-planning policy.” 96. The “manifest” standard has also been adopted in the United Kingdom. Thus, in R (Carson) v Secretary of State for Work and Pensions,[123] referring to the standard as marking out “the appropriate intensity of scrutiny” Lord Walker of Gestingthorpe held that withholding of cost of living increases from certain welfare recipients was not manifestly without reasonable foundation. Similarly, in R (SG and Others) Secretary of State for Work and Pensions (Child Poverty Action Group Intervening),[124] Lord Reed JSC applied the “manifest” standard where the question of proportionality involved “controversial issues of social and economic policy, with major implications for public expenditure”, the determination of such issues being regarded as “pre-eminently the function of democratically elected institutions”. G.2b The “manifest” threshold and margin of discretion in domestic cases 97. Obviously, some matters of concern to the Strasbourg Court, such as the relationship between itself as a supra-national court and Member States with differing political and socio-economic policies, have no relevance to issues in a domestic court. 98. Nonetheless, it is well-recognized that parallel considerations arise in a domestic context where the court is determining the proportionality of a measure taken by the legislature or executive. In R v Director of Public Prosecutions, Ex p Kebilene,[125] Lord Hope of Craighead, referring to the need for domestic courts to decide Convention-based questions involving issues of proportionality, stated: “In this area difficult choices may have to be made by the executive or the legislature between the rights of the individual and the needs of society. In some circumstances it will be appropriate for the courts to recognise that there is an area of judgment within which the judiciary will defer, on democratic grounds, to the considered opinion of the elected body or person whose act or decision is said to be incompatible with the Convention.” 99. Objection has been taken to the suggestion of “deference” by the judiciary partly because (as Lord Sumption JSC puts it) of “overtones of cringing abstention in the face of superior status”[126] conveyed by that word and, more substantially, because it is the Court and not the legislative and executive authorities that is “the ultimate arbiter of the appropriate balance between two incommensurate values: the Convention rights engaged and the interests of the community relied on to justify interfering with it.”[127] I agree that it would be better to use different terminology to describe the demarcation of functions between the judiciary and the other branches of government,[128] but, as Ma CJ has noted, the expression used is not of prime significance. The important consideration is the need in this context to recognize the different constitutional roles of the judiciary on the one hand and the legislative and executive authorities on the other.[129] G.2c Margin of discretion in Hong Kong 100. Along with domestic courts in the UK, this Court has recognized the existence of a domestic doctrine similar to that of the margin of appreciation at the supra-national level which I shall refer to as “the margin of discretion”.[130] It did so, for instance, in Lau Cheong v HKSAR,[131] a case in which particular weight was given to the legislature’s decision to impose a mandatory sentence of life imprisonment for murder, holding this not to be “manifestly disproportionate”.[132] And in HKSAR v Lam Kwong Wai,[133] Sir Anthony Mason NPJ commented: “The weight to be accorded to the legislative judgment by the Court will vary from case to case depending upon the nature of the problem, whether the executive and the legislature are better equipped than the courts to understand its ramifications and the means of dealing with it.” 101. A prominent example of this Court affording the government a wide margin of discretion and linking this to a “manifestly without reasonable foundation” standard for intervention is Fok Chun Wa v Hospital Authority,[134] where the applicants alleged that charging Mainland women married to Hong Kong residents higher obstetrics fees was discriminatory in violation of Article 25 of the Basic Law and Article 22 of the Bill of Rights. Ma CJ, with whom the other members of the Court agreed, noted how the ECtHR doctrine had been taken up in cases like ex p Kebilene and then in the Hong Kong courts; and how that doctrine was particularly relevant to challenges to the government’s socio-economic policies, especially involving the allocation of limited public resources. His Lordship stated:[135] “In the area of healthcare, where resources are also limited and the demands from many different interests heavy, the courts are not equipped (nor is it their role) to make the ‘difficult and agonizing judgments’ (in the words of Sir Thomas Bingham MR in R v Cambridge Health Authority, ex p B [1995] 1 WLR 898, 906E-F) that have to be made allocating funds to one sector or another. ... In this area where limited public funds are involved, the courts have recognised that lines have had to be drawn by the executive or the legislature. On the whole, save where the line has been drawn in contravention of core values (this will be further discussed below) or where it is shown to be manifestly without reasonable foundation, the courts have left it to the authorities to identify the relevant line to be drawn: see R (Animal Defenders International) v Secretary of State for Culture, Media and Sport [2008] 1 AC 1312 , 1348C-D ([33]) where Lord Bingham of Cornhill made the point that it was for the Legislature to decide where the line had to be drawn, even if it meant that hard cases would arise when persons would fall within the wrong side of the line; Mathews v Diaz 426 US 67, 83-84 (a decision of the US Supreme Court).” 102. In summarising the position “[in] the socio-economic context, where policy considerations are best left to the executive, legislative and other authority”, Ma CJ formulated the “manifest” standard as follows: “Where a number of alternative, but reasonable, solutions to a problem exist, the court will not put itself in a place of the executive or legislature or other authority to decide which is the best option. That is not its role. The court will only interfere where the option chosen is clearly beyond the spectrum of reasonable options; in other words, the option has clearly gone too far (or further than necessary) to deal with the problem. In this situation, the court will not have been satisfied under the third limb of the justification test.”[136] 103. A second prominent example is to be found in Kong Yunming v Director of Social Welfare,[137] where the issue was whether the government’s imposition of a requirement of seven years’ residence in place of the single year originally required as a condition of eligibility for a basic welfare benefit violated the right to social welfare under Articles 36 and 145 of the Basic Law. Following Fok Chun Wah,the “manifestly without reasonable foundation” threshold for intervention was adopted in the following terms: “Where the disputed measure involves implementation of the Government's socio-economic policy choices regarding the allocation of limited public funds without impinging upon fundamental rights or involving possible discrimination on inherently suspect grounds, the Court has held that it has a duty to intervene only where the impugned measure is ‘manifestly without reasonable justification’.”[138] 104. It was explained that this approach was linked to the ECtHR’s “margin of appreciation” adapted for application in our domestic context: “That is a test initially applied by the European Court of Human Rights while according a broad margin of appreciation to member States in setting and implementing their socio-economic policies. As the Chief Justice points out, the margin of appreciation principle has previously been adapted to apply in the context of our domestic law. It is appropriate similarly to apply the ‘manifestly without reasonable foundation’ test in our domestic context.” G.3 Factors relevant to choosing the basis for assessing the impugned planning restrictions 105. What principles should the Court apply in choosing between the competing standards? If it should become necessary for the Court to determine whether, after re-considering the developers’ objections, the Board has acted in accordance with their Article 6 and 105 rights, Hysan and OGL contend that the Court should ask whether the restrictions then imposed are no more than reasonably necessary to achieve the legitimate aims advocated. The Board, on the other hand, argues that in such eventuality, the Court ought only to intervene if satisfied that such restrictions are manifestly without reasonable foundation. 106. In principle, the choice of the standard for the Court’s intervention depends on the extent of the appropriate margin of discretion, determined by factors which affect the proportionality analysis in the circumstances of the particular case. In cases calling for a wide margin of discretion, the “manifest” threshold may well be apposite, whereas cases admitting of a narrow or no margin of discretion are more appropriately analysed on the basis of “reasonable necessity”. Which standard or threshold to choose therefore depends on the appropriate width of the margin. 107. As we have seen,[139] in the ECtHR context the scope of the margin of appreciation is held to vary according to the context with a number of factors being relevant. The same applies in domestic cases where such factors principally relate to (i) the significance of and degree of interference with the right in question; and (ii) the identity of the decision-maker as well as the nature and features of the encroaching measure relevant to setting the margin of discretion.[140] G.3a The significance of and extent of interference with the right 108. A theme of the foregoing discussion has been the inter-related and inter-dependent qualities of the various elements of a proportionality analysis. While there would be no point in attempting to construct a formal hierarchy of constitutional rights, a sliding scale has been recognized in which the cogency of the justification required for interfering with a particular right will be proportionate to the perceived importance of that right and the extent of the interference.[141] 109. The specific right invoked may have a low significance and the lower the significance, the broader the margin of discretion is likely to be. Thus, in Belfast City Council v Miss Behavin’ Ltd,[142] the applicant claimed that the city council’s refusal to licence use of premises as a sex shop selling pornography violated its right to free expression and its A1P1 property rights. Lord Hoffmann held as follows: “If article 10 and article 1 of the First Protocol are engaged at all, they operate at a very low level. The right to vend pornography is not the most important right of free expression in a democratic society and the licensing system does not prohibit anyone from exercising it. It only prevents him from using unlicensed premises for that purpose. ... This is an area of social control in which the Strasbourg court has always accorded a wide margin of appreciation to member states, which in terms of the domestic constitution translates into the broad power of judgment entrusted to local authorities by the legislature.” 110. This was also the view of Lord Neuberger of Abbotsbury PSC: “...when it comes to restrictions on the dissemination of pornographic material, the margin of appreciation afforded to member states must, it appears to me, be wide.”[143] 111. In contrast, it has often been said that where a restriction is discriminatory on one of the “suspect grounds” (identified in Article 22 of the Bill of Rights[144]) there would have to be “very weighty” reasons justifying the incursion,[145] obviously resulting in a much narrower margin of discretion. 112. The extent of interference with the right is also obviously relevant. A relatively trivial interference will be much easier to justify. Thus, in Koon Wing Yee v Insider Dealing Tribunal,[146] a case about infringement of the privilege against self-incrimination, Sir Anthony Mason NPJ distinguished the case before the Court on the one hand from Brown v Stott,[147] and O’Halloran v United Kingdom,[148]on the other, pointing out that the encroachment in those cases was minor, consisting of the individual being compelled to answer “a single, simple question” whereas in Koon Wing Yee, the questions and compulsory answers were not so limited and constituted a substantial intrusion into the privilege.[149] 113. In some cases, the interference may be so great that neither proportionality analysis nor margin of discretion are meaningful concepts. In such cases, a violation of a right may be held to have occurred on the footing that the impugned measure has destroyed “the essence of the right”. For example, the right to marry guaranteed by Article 37 of the Basic Law and Article 19(2) of the Bill of Rights may validly be limited to some degree by rules relating to marriageable age, consanguinity and so forth. But a law which imposes a blanket prohibition against a certain individual (a post-operative transsexual person) marrying was held to destroy the essence of the right in her case and ruled unconstitutional.[150] G.3b The measure’s content and features relevant to the margin of discretion 114. The content and features of the impugned measure, the identity and constitutional role of its originator and any special competence possessed by such person, are likely to be highly relevant to deciding how wide the margin of discretion should be. 115. If assessment of the proportionality of the measure calls for the application of purely legal principles and an assessment which the Court is the expert to make, the primary decision-maker having no special competence or expertise, it is likely that the margin of discretion will have little role to play and that the Court will simply adopt a standard of reasonable necessity. 116. On the other hand, a decision-maker’s views resulting in the promulgation of the impugned measure may be given much weight and thus afforded a wide margin of discretion reflected by use of a “manifest” standard where the decision-maker is likely to be better placed than the Court to assess what is needed in the public interest.[151] The Court may for instance, be satisfied that he had special access to information; special expertise in its assessment; or an overview enabling him to assess competing and possibly prior claims for scarce resources.[152] The Court might also refrain from intervening because the measure reflects a predictive or judgmental decision which it was the institutional role of the decision-maker to take and as to which no single “right answer” exists.[153] 117. The Court is likely to take such an approach, for instance, in relation to matters touching on national security.[154] In Hong Kong there are, additionally, limitations placed on the Court’s role on questions touching on defence and foreign policy.[155] 118. A broad margin of discretion might also be mandated by separation of powers principles and recognition of the different institutional roles played by the Court and the relevant decision-maker. Thus, a wide margin of discretion may be permitted to the legislature in respect of enactments allocating public resources on the footing that such distributive decisions are properly the responsibility of the legislature for which it makes itself politically accountable, rather than for the courts on a constitutional review. A similar view was taken by this Court in Lau Cheong v HKSAR,[156] where the legislature’s decision to retain mandatory life imprisonment for murder, accompanied by statutory machinery for reviewing the prisoner’s incarceration, was regarded as a reflection of the separation of powers principle and a measure that could not be held to be manifestly disproportionate. G.4 The meaning of “manifestly without reasonable foundation” and the relationship between the two standards 119. It should be noted that the difference between the two standards is one of degree. Once it is recognized that the former threshold is a standard of reasonable necessity, it becomes clear that it is located on the same “reasonableness” spectrum as the standard which asks whether a measure is “manifestly without reasonable foundation”. That may be why that phrase is sometimes referred to in the authorities as a measure of the “intensity” of judicial scrutiny rather than as a “standard”. 120. In practice, where a wide margin of discretion exists, in applying the “manifest” standard, the Court will allow the decision-maker latitude to adopt one of a relatively wide range of possible alternatives in fashioning the impugned measure which encroaches upon the protected right. Assuming that such measure pursues a legitimate aim and is rationally connected to achieving that aim, the Court will not be astute to insist on a potentially less intrusive measure but will only intervene to strike down the impugned provision as unconstitutional if, as Ma CJ put it in Fok Chun Wa,[157] “the option chosen is clearly beyond the spectrum of reasonable options” to deal with the problem. 121. Where, on the other hand, the standard of reasonable necessity is applicable, the “minimal impairment” approach explained by McLachlin J in RJR-Macdonald Inc v The Attorney General of Canada”[158] is apposite. The legislative or executive authority must show that the measure impairs the right as little as reasonably possible in order to achieve the legislative objective: “If the law falls within a range of reasonable alternatives, the courts will not find it overbroad merely because they can conceive of an alternative which might better tailor objective to infringement”. On the other hand, if the authority “fails to explain why a significantly less intrusive and equally effective measure was not chosen” the measure may fail. In these cases, the acceptable range of reasonable alternatives will depend on the factual context but one would expect such range to be significantly narrower than where the “manifest” threshold is applied. 122. It is perhaps worth re-iterating that while for the purposes of elucidation, two differently named standards are referred to: “reasonable necessity” and “manifestly without reasonable foundation”, they indicate positions on a continuous spectrum rather than wholly independent concepts. 123. One further point to make on the meaning of “manifestly without reasonable foundation” is that one should not be misled into thinking that the word “manifestly” suggests that the Court may dispense with a close examination of the circumstances of the case in considering whether the threshold is met. As Lords Reed and Toulson JJSC stated in R (Lumsdon) v Legal Services Board,[159] “It would however be a mistake to suppose that the ‘manifestly inappropriate’ test means that the court's scrutiny of the justification for the measure is cursory or perfunctory. While the court will be slow to substitute its own evaluative judgment for that of the primary decision-maker, and will not intervene merely because it would have struck a different balance between countervailing considerations, it will consider in some depth the factual foundation and reasoning underlying that judgment.” G.5 The factors to be applied in the present case 124. Looking to the future, there is some artificiality in the discussion since there are no extant decisions of the Board pending their re-consideration of representations made by Hysan and OGL. How a proportionality analysis, if necessary, would be approached would depend on what planning restrictions result from that process. What is stated here is therefore necessarily confined to remarks of a general nature. 125. In the present case, the developers’ rights as owners of the sites are obviously substantial and of high constitutional significance, reflecting the general principle of safeguarding private property rights guaranteed by the Basic Law. What the extent of interference to those rights may be after the Board reconsiders the developers’ case involves a question of fact to be examined in due course. 126. Leaving the factor of the extent of interference aside, the constitution and decision-making machinery of the Board as the originator of any potentially impugned planning restrictions would, in my view, strongly favour adoption of a broad margin of discretion near the “manifestly without reasonable foundation” end of the spectrum. As Reyes J pointed out in his OGL judgment,[160] “[p]lanning is a holistic process, involving balancing numerous factors”. Draft OZPs and individual planning restrictions are the product of a decision-making process carried out by machinery created by statute and designed to ensure consideration of competing views after consultation with the public and with the parties affected with the benefit of input from relevant experts. Planning decisions are made with entire districts, and not just the parties’ sites, in view. The system includes as part of its design, the possibility of an aggrieved party seeking administrative law remedies if material flaws occur in the way the Board arrives at its decisions, as occurred in the present cases. But if the statutory process is undertaken without judicially reviewable flaws, it is hard to see any reason for thinking that the planning restrictions imposed should be liable to be struck down as constitutionally invalid. As Lord Neuberger of Abbotsbury PSC noted in the Carlile case:[161] “...where, as here, the relevant decision maker has carried out the balancing exercise, and has not made any errors of primary fact or principle and has not reached an irrational conclusion, so that the only issue is the proportionality of the decision, the court cannot simply frank the decision, but it must give the decision appropriate weight, and that weight may be decisive.” 127. And in Belfast City Council v Miss Behavin’ Ltd,[162] Lord Hoffman held that: “If the local authority exercises that power rationally and in accordance with the purposes of the statute, it would require very unusual facts for it to amount to a disproportionate restriction on Convention rights.” In like vein, his Lordship remarked: “... I find it difficult to imagine a case in which a proper exercise by the council of its powers under the Order could be a breach of an applicant's Convention rights.” 128. The developers’ arguments raise similar questions. If, as in the present cases, planning restrictions are imposed which are arbitrary, based on inadequate inquiry or procedurally flawed, traditional judicial review remedies are available. But the developers’ proposition on this appeal is that even if the Board avoids falling into any error and decides on planning restrictions which are unimpeachable on traditional grounds, a basis ought nevertheless to exist for them to challenge those restrictions as disproportionate and unconstitutional. While I do not rule out the possibility of such an exceptional situation arising, like Lord Hoffmann, I find it difficult, at least in the abstract, to envisage the emergence of such a case. 129. I would therefore conclude that town planning restrictions, assuming them to be unassailable on traditional judicial review grounds would in general only be susceptible to constitutional review if the Court is satisfied that they are manifestly without reasonable foundation. 130. A final point might be mentioned by way of guidance to the Board. It should be emphasised that it is the Court which has the ultimate responsibility for determining whether any restriction imposed by the Board can be subjected to a successful constitutional challenge. The Board’s role is to carry out its duties and to exercise its powers in accordance with the TPO. To adapt what Lord Hoffmann said in R (SB) v Governors of Denbigh High School,[163] members of the TPB cannot be expected to make the Board’s planning decisions with textbooks on human rights law at their elbows. No doubt the Board will receive appropriate legal advice including advice regarding the property rights of others guaranteed by the Basic Law. But it is not the Board’s task to conduct a proportionality analysis, much less to mouth incantations about proportionality in rendering its decisions. H. Conclusions summarised 131. I would summarise the conclusions I have reached as follows. 132. Articles 6 and 105 are engaged in cases where it is factually established that planning restrictions imposed by the TPB encroach upon a landowner’s property rights. 133. Where such encroachment on the right is established, the extent, if any, of the encroaching measure’s validity is determined by a proportionality analysis. 134. In Hong Kong, such a proportionality assessment has been viewed as involving a three-step process of asking (i) whether the intrusive measure pursues a legitimate aim; (ii) if so, whether it is rationally connected with advancing that aim; and (iii) whether the measure is no more than necessary for that purpose. 135. A fourth step should be added. In line with a substantial body of authority, where an encroaching measure has passed the three-step test, the analysis should incorporate a fourth step asking whether a reasonable balance has been struck between the societal benefits of the encroachment and the inroads made into the constitutionally protected rights of the individual, asking in particular whether pursuit of the societal interest results in an unacceptably harsh burden on the individual. 136. At the third stage, assessing the permissible extent of the incursion into the protected right, two main standards have been applied. The first is the test of whether the intruding measure is “no more than necessary” to achieve the legitimate aim in question. This must be understood to be a test of reasonable necessity. If the Court is satisfied that a significantly less intrusive and equally effective measure is available, the impugned measure may be disallowed. 137. An alternative standard which may be applied at the third stage is one which asks whether the encroaching measure is “manifestly without reasonable foundation”, being a standard closely related to the concept of “margin of appreciation” in ECtHR jurisprudence. 138. At the supra-national level of the ECtHR, the margin of appreciation doctrine involves the recognition that on certain issues, the Court should allow Member State latitude to decide on the legitimacy of their societal aims and the means to achieve them since they are better placed to make the assessment. Similar considerations have led the Court at a domestic level to allow the legislative and executive authorities latitude or a “margin of discretion” to do the same, applying the “manifestly without reasonable foundation” standard in such cases. 139. The “manifest” standard has been used in cases where the Court recognizes that the originator of the impugned measure is better placed to assess the appropriate means to advance the legitimate aim espoused. This has occurred in cases involving implementation of the legislature’s or executive’s political, social or economic policies but the principle is not confined to such cases. 140. The location of the standard in the spectrum of reasonableness depends on many factors relating principally to the significance and degree of interference with the right; the identity of the decision-maker; and the nature and features of the encroaching measure relevant to setting the margin of discretion. 141. The difference between the two standards is one of degree, with the Court in both cases, scrutinising the circumstances of the case and the factual bases claimed for the incursion. 142. No extant planning restrictions fall to be considered since the Board is to reconsider its decisions. In general terms, where the Board reaches decisions which are not flawed on traditional judicially reviewable grounds, any imposed restrictions which encroach upon a landowner’s property rights should be subject to constitutional review applying the “manifestly without reasonable foundation” standard. It is considered to be highly unlikely that Board decisions imposing planning restrictions arrived at lawfully and in conformity with the principles of traditional judicial review, would be susceptible to constitutional review unless the measures are exceptionally unreasonable. I. Disposal of the appeal and costs 143. The Court of Appeal made orders directing the Board in each case to reconsider its decisions in accordance with the Court of Appeal’s judgment. The orders for remitter obviously remain. However, the orders should be varied to delete reference to such re-consideration being in accordance with the judgments of the Court of Appeal. 144. Given the nature of the issues and the conclusions that I have reached, I would make an order nisi that there be no order in respect of the costs of this appeal. I would direct that the parties have liberty, if so advised, to lodge written submissions as to costs within 21 days after delivery of this judgment and, in default of such submissions, that the order as to costs stand as an order absolute without further direction. Mr Justice Tang PJ: 145. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Fok PJ: 146. I agree with the judgment of Mr Justice Ribeiro PJ. Lord Neuberger of Abbotsbury NPJ: 147. I agree with the judgment of Mr Justice Ribeiro PJ. Chief Justice Ma: 148. For the above reasons, these appeals are determined in accordance with the orders set out in paragraphs 143 and 144 above. Mr Benjamin Yu SC and Mr Alexander Stock SC, instructed by Mayer Brown JSM, for the Appellants Lord Pannick QC, Mr Jat Sew-Tong SC and Mr Abraham Chan, instructed by the Department of Justice, for the Respondent Mr Nigel Pleming QC, Ms Audrey Eu SC and Mr Jonathan Lee, instructed by Philip T F Wong & Co., for the Intervener [1] Draft Causeway Bay Outline Zoning Plan No. S/H6/15 and draft Wanchai Outline Zoning Plan No. S/H5/26. [2] In two cases, 200 mPD BHRs were laid down. [3] HCAL 38/2011 and HCAL 57/2011 heard together (14 September 2012). [4] Lam VP, Chu JA and Au J, CACV 232/2012 and CACV 233/2012 heard together (13 November 2014). [5] That the TPB had failed to discharge its duty to make proper inquiry referred to in Secretary of State for Education and Science v Tameside Metropolitan Borough Council [1977] AC 1014 at 1065; that it had taken into account irrelevant considerations; and that there had been procedural unfairness in the way it reached its decisions. [6] Draft Ngau Tau Kok and Kowloon Bay Outline Zoning Plans Nos S/K13/26 and S/K13/27, gazetted on 19 November 2010 and 7 October 2011 respectively. [7] HCAL 34/2012 (11 May 2012). [8] Lam VP, Barma JA and Poon J, CACV 127/2012 and CACV 129/2012 (13 November 2014). [9] Tang and Fok PJJ and Chan NPJ, FAMV 28, 29/2015 (Reasons 18 November 2015). [10] Leave was initially granted on a second question raised by the Board in relation to Reyes J’s quashing of the 5m NBA, but that appeal has since been withdrawn. [11] A1P1 provides as follows: “Every natural or legal person is entitled to the peaceful enjoyment of his possessions. No one shall be deprived of his possessions except in the public interest and subject to the conditions provided for by law and by the general principles of international law. The preceding provisions shall not, however, in any way impair the right of a State to enforce such laws as it deems necessary to control the use of property in accordance with the general interest or to secure the payment of taxes or other contributions or penalties.” [12] A similar approach was taken in OGL’s case : see §§89 and 90. [13] Hysan Judgment at §§190 and 191. However, as Hysan pointed out in the Court of Appeal, its case has not been one of “deprivation”, but a complaint regarding a restriction on the use of its property: Court of Appeal at §45. [14] Hysan Judgment at §195. [15] OGL’s appeal does not require separate examination since the Court of Appeal there followed the decision handed down on the same day in the Hysan appeal on this point. [16] Hysan Court of Appeal at §56. [17] Ibid at §60. [18] Ibid. [19] Ibid at §82. [20] Ibid at §83. [21] [2008] 1 HKLRD 553 at §33. [22] Hysan Court of Appeal at §§61-62. [23] Town Planning Ordinance (Cap 131). [24] With Mr Jat Sew-tong SC and Mr Abraham Chan. [25] TPB’s printed case at §9(9). [26] Appearing with Mr Alexander Stock SC for Hysan. [27] Appearing for OGL with Ms Audrey Eu SC and Mr Jonathan Lee. [28] Shum Kwok Sher v HKSAR (2002) 5 HKCFAR 381 at §§62-65; Gurung Kesh Bahadur v Director of Immigration (2002) 5 HKCFAR 480 at §34; Mo Yuk Ping v HKSAR (2007) HKCFAR 386 at §§59 and 61. [29] Section 1: Public Finance, Monetary Affairs, Trade, Industry and Commerce. [30] [2008] 1 HKLRD 553 at §33. [31] See Sections D.1 and D.2 above, per Reyes J and the Court of Appeal, citing Fine Tower Associates Ltd v Town Planning Board [2008] 1 HKLRD 553 at §33. [32] (2008) 46 EHRR 45. [33] Ibid at §37. [34] Beyler v Italy (2001) 33 EHRR 52 at §106. [35] JA Pye (Oxford) Ltd v United Kingdom (2008) 46 EHRR 45 at §§61-63. [36] [2004] 1 AC 816. [37] Ibid, at §§106-107. [38] Ibid, at §137. [39] Ibid, at §168 [40] Ibid, at §41. [41] Basic Law, Article 28(2); Bill of Rights, Article 3. [42] Ubamaka v Secretary for Security (2012) 15 HKCFAR 743 at §§106-111; Kong Yunming v Director of Social Welfare (2013) 16 HKCFAR 950 at §38. [43] Catholic Diocese of Hong Kong v Secretary for Justice (2011) 14 HKCFAR 754 at §65; Kong Yunming v Director of Social Welfare (2013) 16 HKCFAR 950 at §38; Official Receiver v Zhi Charles (2015) 18 HKCFAR 467 at §22. [44] Cap 383. See Leung Kwok Hung v HKSAR (2005) 8 HKCFAR 229 at §19 on implementation. [45] (2005) 8 HKCFAR 229. [46] “Hong Kong residents shall have freedom of speech, of the press and of publication; freedom of association, of assembly, of procession and of demonstration; and the right and freedom to form and join trade unions, and to strike.” [47] Leung Kwok Hung v HKSAR (2005) 8 HKCFAR 229 at §§17 and 19. [48] Ibid at §§33-34. [49] Ibid at §35. [50] Including the ECtHR’s jurisprudence on the ECHR, the House of Lords and UK Supreme Court’s case-law on the United Kingdom’s Human Rights Act 1998 and the decisions of the Canadian Supreme Court. See HKSAR v Lam Kwong Wai (2006) 9 HKCFAR 574 at §37. [51] (2006) 9 HKCFAR 574. [52] “Anyone who is lawfully arrested shall have the right to a fair trial by the judicial organs without delay and shall be presumed innocent until convicted by the judicial organs.” [53] “Everyone charged with a criminal offence shall have the right to be presumed innocent until proved guilty according to law.” [54] (2006) 9 HKCFAR 574 at §40 [55] (2010) 13 HKCFAR 762 at §28. This was a case where a provision making the determination of the Court of First Instance on election petition matters final was challenged as a possible violation of Article 82 of the Basic Law which provides that the power of final adjudication shall be vested in the Court of Final Appeal. [56] (2015) 18 HKCFAR 467 at §§22-23; citing Mok Charles v Tam Wai Ho (2010) 13 HKCFAR 762 at §28; Catholic Diocese of Hong Kong v Secretary for Justice (2011) 14 HKCFAR 754 at §65; and Kong Yunming v Director of Social Welfare (2013) 16 HKCFAR 950 at §§39-40. [57] [1993] AC 951. [58] R v Oakes [1986] 1 SCR 103. [59] Ibid at p 972. [60] Ibid. [61] [1996] AC 907. This involved a newspaper’s contention that a provision making it an offence to disclose details of an investigation into a suspected offence under the Prevention of Bribery Ordinance violated its freedom of expression. [62] Ibid at p 917. [63] Ibid. [64] Ibid at pp 918-921. [65] Ibid at p 922. [66] Beginning with HKSAR v Ng Kung Siu (1999) 2 HKCFAR 442 (a case on flag desecration and freedom of expression) and leading up to Leung Kwok Hung v HKSAR (2005) 8 HKCFAR 229; HKSAR v Lam Kwong Wai (2006) 9 HKCFAR 574 and the other cases cited above. [67] R (Lord Carlile of Berriew) v Secretary of State for the Home Department [2015] AC 945 at §89. [68] [2014] AC 700 at §74. His Lordship was referring to the judgment of Dickson CJ in R v Oakes [1986] 1 SCR 103 at 139. [69] [2014] AC 700 at §20. [70] Ibid. [71] Ibid. [72] [2015] 1 WLR 1591 at §106. [73] [1999] 1 AC 69 at 80. [74] Bank Mellat v Her Majesty’s Treasury (No 2) [2014] AC 700 at §20. [75] [1986] 1 SCR 103. [76] “The Canadian Charter of Rights and Freedoms guarantees the rights and freedoms set out in it subject only to such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society.” [77] At pp 138-139. [78] [1995] 3 SCR 199 at §175. See also McLachlin CJ in Alberta v Hutterian Brethren of Wilson Colony [2009] 2 SCR 567 at §76. [79] [2007] 2 AC 167 at §19. [80] [1999] 1 AC 69. [81] [2004] 2 AC 368 at §§ 17–20, 26, 27, 60, 77. [82] See R (Aguilar Quila) v Secretary of State for the Home Department [2012] 1 AC 621 at §45; Bank Mellat v Her Majesty’s Treasury (No 2) [2014] AC 700 at §20 and §65; R (Lord Carlile of Berriew) v Secretary of State for the Home Department [2015] AC 945 at §§19, 98 and §148; and In re Recovery of Medical Costs for Asbestos Diseases (Wales) Bill [2015] AC 1016 at §45. It has, however, been accepted that the four-step analysis does not apply in European Union law: R (Lumsdon) v Legal Services Board [2016] AC 697 at §26. [83] [2014] AC 700 at §74. [84] [2015] AC 945 at §98. [85] Further discussed in Section G below. [86] Eg, Sporrong and Lonnroth v Sweden (1983) 5 EHRR 35 at §69: “The search for this balance is inherent in the whole of the Convention and is also reflected in the structure of Article 1.” [87] Eg, James v United Kingdom (1986) 8 EHRR 123 at §50: “Not only must a measure depriving a person of his property pursue, on the facts as well as in principle, a legitimate aim ‘in the public interest’, but there must also be a reasonable relationship of proportionality between the means employed and the aim sought to be realised. This latter requirement was expressed in other terms in the Sporrong and Lönnroth judgment by the notion of the ‘fair balance’ that must be struck between the demands of the general interest of the community and the requirements of the protection of the individual’s fundamental rights...” [88] Ibid. See also, eg, James v United Kingdom (1986) 8 EHRR 123 at §50 and Lithgow v United Kingdom (1986) 8 EHRR 329 at §120. [89] Denimark Ltd v United Kingdom (2000) 30 EHRR CD 144 at CD150. See also, James v United Kingdom (1986) 8 EHRR 123 at §50; AXA General Insurance Ltd v HM Advocate [2012] 1 AC 868 at §§37 and 126; Bank Mellat v Her Majesty’s Treasury (No 2) [2014] AC 700 at §70. [90] [2014] AC 700 at §20. [91] [2009] 2 SCR 567 at §76. [92] Bank Mellat v Her Majesty’s Treasury (No. 2) [2014] AC 700 at §76. [93] Cambridge UP, 2012. [94] In Chapter 12, including an Israeli law setting a blanket restriction on entry of spouses of Israeli citizens residing in the Occupied Territories on national security grounds; legislation authorizing seizure of land relating to the Security Fence in the West Bank and Ontario regulations restricting dentists’ advertisements (see pp 341-342). [95] Dieter Grimm, “Proportionality in Canadian and German Constitutional Jurisprudence” 57 U Toronto L J 383, 396 (2007). [96] Barak, op cit at pp 342-343. [97] Grimm, op cit at p 396. [98] Sporrong and Lonnroth v Sweden (1983) 5 EHRR 35 at §69; Pine Valley Developments Ltd v Ireland (1991) 14 EHRR 319 at §79; [99] (1989) 11 EHRR 439 at §89. For similar statements in cases relating to property rights, see for instance, Sporrong and Lonnroth v Sweden (1983) 5 EHRR 35 at §69; Pine Valley Developments Ltd v Ireland (1991) 14 EHRR 319 at §79; Lough v First Secretary of State [2004] 1 WLR 2557 at §31; and Bank Mellat v Her Majesty’s Treasury (No 2) [2014] AC 700 at §70. [100] Li CJ, Chan and Ribeiro PJJ and Sir Anthony Mason NPJ (2005) 8 HKCFAR 229 at §35. [101] (2008) 11 HKCFAR 170 at §62. [102] (1983) 5 EHRR 35. [103] At §69: “...the Court must determine whether a fair balance was struck between the demands of the general interest of the community and the requirements of the protection of the individual's fundamental rights. The search for this balance is inherent in the whole of the Convention and is also reflected in the structure of Article 1.” [104] (2008) 11 HKCFAR 170 at §79. The concept is referred to again at §84. [105] (2012) 15 HKCFAR 362 at §130. [106] [2004] 1 AC 816 at §62. [107] Bank Mellat v Her Majesty’s Treasury (No 2) [2014] AC 700 at §71. See also R (SB) v Governors of Denbigh High School [2007] 1 AC 100 at §30; Belfast City Council v Miss Behavin’ Ltd [2007] 1 WLR 1420 at §88; R (Lord Carlile of Berriew) v Secretary of State for the Home Department [2015] AC 945 at §29. [108] (1986) 8 EHRR 123 at §§50-51. [109] [1995] 3 SCR 199 at §160 (citations omitted). [110] Bank Mellat v Her Majesty’s Treasury (No 2) [2014] AC 700 at §20. [111] Pham v Secretary of State for the Home Department (Open Society Justice Initiative Intervening) [2015] 1 WLR 1591 at §106. [112] Per Lord Sumption JSC in R (Lord Carlile of Berriew) v Secretary of State for the Home Department [2015] AC 945 at §34. [113] (2012) 15 HKCFAR 409 at §75(6). [114] In Illinois State Board of Elections v Socialist Workers Party (1979) 440 US 173, 188–189. Blackmun J’s comment was also cited by Lord Reed JSC in Bank Mellat [2014] AC 700 at §75. [115] (2015) 18 HKCFAR 467 at §53. [116] Of the Council of Europe. [117] (1996) 23 EHRR 101 at §74. [118] (1986) 8 EHRR 123 at §46. [119] (1986) 8 EHRR 329 at §122. [120] Mellacher v Austria (1989) 12 EHRR 391 at §45. [121] (1983) 5 EHRR 35 at §69. [122] (2001) 32 EHRR 11 at §55. See also Pine Valley Developments Ltd v Ireland (1991) 14 EHRR 319 at §79; and Chapman v United Kingdom (2001) 33 EHRR 18 at §92. [123] [2006] 1 AC 173 at §§86-91. [124] [2015] 1 WLR 1449 at §§92-93. [125] [2000] 2 AC 326 at 381. [126] R (Lord Carlile of Berriew) v Secretary of State for the Home Department [2015] AC 945 at §22. [127] Ibid at §34. [128] Ibid at §33. [129] Fok Chun Wa v Hospital Authority (2012) 15 HKCFAR 409 at §64. [130] As noted by Ma CJ in Mok Charles v Tam Wai Ho (2010) 13 HKCFAR 762 at §55. [131] (2002) 5 HKCFAR 415 at §102-105, citing ex p Kebilene [2000] 2 AC 326 and Brown v Stott [2003] 1 AC 681 at 703, per Lord Bingham of Cornhill: “While a national court does not accord the margin of appreciation recognised by the European court as a supra-national court, it will give weight to the decisions of a representative legislature and a democratic government within the discretionary area of judgment accorded to those bodies...” [132] (2002) 5 HKCFAR 415 at §123. [133] (2006) 9 HKCFAR 574 at §45. [134] (2012) 15 HKCFAR 409. [135] Ibid at §§70 and 71. [136] Ibid at §75(3). [137] (2013) 16 HKCFAR 950. [138] Ibid at §41. The seven-year requirement was held to fail the rationality test and thus to be a violation of Articles 36 and 145. [139] Buckley v United Kingdom (1996) 23 EHRR 101 at §74 [140] See R (Lord Carlile of Berriew) v Secretary of State for the Home Department [2015] AC 945 at §34 where similar factors are discussed. [141] Pham v Secretary of State for the Home Department (Open Society Justice Initiative Intervening [2015] 1 WLR 1591 at §106. [142] [2007] 1 WLR 1420 at §16. [143] Ibid at §83. [144] Bill of Rights Art 22: Grounds “such as race, colour, sex, language, religion, political or other opinion, national or social origin, property, birth or other status”. [145] See eg, R (Carson) v Secretary of State for Work and Pensions [2006] 1 AC 173 at §58; Stec v United Kingdom (2006) 43 EHRR 47 at §52 and Humphreys v Revenue and Customs Commissions [2012] 1 WLR 1545 at §16. [146] (2008) 11 HKCFAR 170 at §§78 and 79. [147] [2003] 1 AC 681. [148] (2008) 46 EHHR 21. [149] A similar approach was adopted by Ma CJHC (as Ma CJ then was) in Secretary for Justice v Latker [2009] 2 HKC 100 at §§35-37. [150] W v Registrar of Marriages (2013) 16 HKCFAR 112. [151] R (Lord Carlile of Berriew) v Secretary of State for the Home Department [2015] AC 945 at §34. [152] See R (on the application of Rotherham Metropolitan Borough Council) v Secretary of State for Business, Innovation and Skills [2015] 3 All ER 1 at §23. [153] R (Lord Carlile of Berriew) v Secretary of State for the Home Department [2015] AC 945 at §32. [154] Bank Mellat v Her Majesty’s Treasury (No. 2) [2014] AC 700 at §21; R (Lord Carlile of Berriew) v Secretary of State for the Home Department [2015] AC 945 at §105. [155] Discussed in Democratic Republic of the Congo v FG Hemisphere Associates LLC (No 1)(2011) 14 HKCFAR 95. [156] (2002) 5 HKCFAR 415 at §§101-105 and 123. [157] Fok Chun Wa v Hospital Authority (2012) 15 HKCFAR 409 at §75(3). [158] [1995] 3 SCR 199 at §160 (citations omitted). [159] [2016] AC 697 at §44. [160] At §100. [161] R (Lord Carlile of Berriew) v Secretary of State for the Home Department [2015] AC 945 at §68. [162] [2007] 1 WLR 1420 at §§12 and 16. [163] [2007] 1 AC 100 at §68. 
CRIMINAL APPEAL NO. 418 OF 2014 Hon Yeung VP : The Facts 1. The appellant (Ngo Van-nam) was intercepted by a police officer at Kwai Fong MTR Station on 25 July 2012 and he failed to produce any identification document. Upon enquiry, the appellant admitted that he had sneaked into Hong Kong from the Mainland on 25 July 2012 and that he had committed a robbery case in Tuen Mun in March 2012. The appellant was arrested. 2. In the subsequent interviews by the police, the appellant further admitted the following: i. Two days before he committed the robbery in March 2012, he came to Hong Kong and met two Vietnamese friends “Ninh” and “Quang” and the three of them decided to rob a taxi. ii. They took a taxi from Tung Chung to Tuen Mun and he sat at the left-side back passenger seat. Upon arrival at Tuen Mun, “Ninh” grabbed the taxi-driver’s neck and declared “robbery” and “Quang” took his money, mobile phone, watch and ring. iii. He then pulled away the wire of the walkie-talkie of the taxi before he fled with the others. iv. “Ninh” gave him the mobile phone and some money. The proceeds were used for purchasing dangerous drugs. v. He then surrendered himself to the Police Station as an “Illegal Immigrant”, while his two friends had sneaked back to Vietnam. vi. He sneaked into China and then to Hong Kong by hiding underneath a truck on 25 July 2012. vii. He intended to come to Hong Kong for curing his addiction to dangerous drug. 3. The appellant was charged with (1) Robbery; and (2) Remaining in Hong Kong without the authority of the Director of Immigration after having landed unlawfully in Hong Kong. 4. On 9 November 2012, the appellant appeared before H.H. Judge Geiser in the District Court and pleaded guilty to both charges. In connection with the robbery charge, he admitted the following facts: “ At about 10:00 p.m. on 13 March 2012, three males (one of them was the appellant) boarded PW1’s taxi outside Tung Chung MTR Station, one of them sat next to PW1 while the other two sat at the back. PW1 was asked to go to Fu Tei Road, Tuen Mun, NewTerritories. Upon reaching Tuen Mun Fresh Water Primary Service Reservoir, Fu Tei Road, near lamppost No. DP 3572, the one sitting at PW1’s right-side back gripped PW1’s neck from behind, pointed at his neck with an 8 inches long knife and declared “Robbery”. The one sat at the left-side back of PW1 asked PW1 not to move. The one sat next to PW1 ordered PW1 to switch off all the lights of the taxi, used his hand to break the walkie-talkie and asked PW1 to take out all the money. PW1 handed in HK$1,200 banknotes to the male sitting at the front passenger seat, he then took away HK$200 coins (approximately) from the front seat; PW1 also handed in his watch (HK$500), mobile phone (HK$1,500) and silver ring (HK$100) to him. The three males then left. PW1 suffered no injuries and reported the case to police.” 5. The appellant sought leave to appeal against sentence out of time and his application was granted by McWalters JA sitting as a single judge of the Court of Appeal on 16 July 2015. The appellant’s background and mitigation 6. The appellant, aged 26 and single, used to live in Vietnam, and was educated up to Form 5. He had a sister who was married to a Hong Kong citizen, but his parents were living in Vietnam. The appellant had a clear record although he had entered Hong Kong illegally on two previous occasions. 7. It was submitted that the appellant had surrendered himself and had revealed voluntarily to the police the robbery offence when he was intercepted on 25 July 2012. The suggestion was that without his co-operation, the appellant could not be caught and would not be charged with robbery. The appellant claimed that he had followed his sister’s advice and decided to reveal everything to the police. 8. The undisputed facts showed that the appellant had surrendered himself to the police three days after committing the robbery in March 2012, but on that occasion he had only briefly revealed his illegal immigrant status without mentioning the robbery case. The appellant was not charged, but was repatriated to Vietnam directly. 9. The appellant suggested that he just followed two older people to commit robbery and that he was remorseful. It was argued that the offence was not the worst case of the type as the amount involved was not large and that only minimum force was used with no injury to the victim at all. The Sentence by the judge 10. The judge rejected the suggestion that the appellant had voluntarily surrendered to the police, pointing out that he was in fact intercepted by the police. In the course of mitigation, the following exchange between the judge and counsel for the appellant (Mr Li) took place: “Court: Because I do not consider this being a situation where he voluntarily surrendered himself to the police. Were it not for the fact of his interception outside the MTR station, he would not have been caught. Mr Li: Yes, indeed. Your Honour, I mean, I tend to agree with that, I mean, that observation. I mean, there is a difference between the two… Court: You can say that he co-operated by telling the police all about what had happened in relation to the robbery in March... But I don’t think it’s right to say that he surrendered himself, because he didn’t. Mr Li: …not in the traditional manner …as you walk into the police station… ‘I killed somebody,’ and then giving the details. I mean, that--I must say that is the normal or the standard way of doing things, sometimes accompanied by lawyers... …it is my submission that, in essence, it is the same thing, in the sense that but for his voluntary disclosure…there is no way he can be caught or even charged for this offence…” 11. The judge accepted that the appellant had volunteered the information regarding the robbery case to the police and that in the absence of such admission, his involvement would never have come to light. The judge also took into account the appellant’s guilty plea to the charges. 12. The judge, having referred to R v Tran Van Anh [1993] 2 HKCLR 122; and Secretary for Justice v Tso Tsz Kin [2004] 2 HKC 139, took the view that a starting point of 7 years’ imprisonment for the robbery charge should be adopted. The judge reduced the starting point by one-third on account of the guilty plea and gave a further discount of three months because the appellant had volunteered details of the robbery to the police. The judge sentenced the appellant to 53 months’ imprisonment on the robbery charge. 13. The judge sentenced the appellant to 15 months’ imprisonment on the unlawful remaining charge and ordered 10 months of the 15 months to run consecutively to the 53 months, making a total sentence of 5 years and 3 months’ imprisonment. The Appeal 14. In presenting his argument before McWalters JA, the appellant claimed that he came back to Hong Kong because he wanted to confess to the robbery offence. McWalters JA took the view that if the appellant’s claim was accepted, he would have a strong arguable case for saying that he should have received a greater discount and that no part of the unlawful remaining offence sentence should have been ordered to be served consecutively. McWalters JA reminded the appellant that if he wished to adduce further evidence to support his claim, he had to apply for leave to adduce further evidence before the Court of Appeal. There was no such application from the appellant. Grounds of Appeal 15. Mr Clive Grossman SC leading Mr Andy Hung for the appellant initially suggested that the appellant came back to Hong Kong because he wanted to surrender and confess to the robbery offence. When confronted with the judge’s findings that the appellant did not voluntarily surrender to the police, Mr Grossman said he was happy to proceed with the appeal only on the basis that but for the appellant’s volunteered confession, his involvement in the robbery would not have come to light. 16. Mr Grossman did not suggest that the starting points of 7 years and 15 months adopted by the judge for the two offences were manifestly excessive or wrong in principle. Mr Grossman’s only complaint was that the judge had not given a sufficient discount for the appellant’s volunteered confession. He emphasized that the appellant confessed to the robbery when his complicity was neither known to nor suspected by the police. 17. Mr Grossman suggested that the judge had not directed himself that it was in the public interest to encourage criminals such as the appellant to assist the law enforcement authorities to clear up crimes which would not otherwise have been brought to justice. Discussion 18. When he surrendered himself in March 2012, the appellant did not reveal to the police the robbery that he had committed shortly beforehand, but only revealed his illegal immigrant status and he was immediately repatriated. He then went to the Mainland and sneaked back to Hong Kong a few months later. 19. When he was in Hong Kong, the appellant did not on his own volition surrender himself to the police. Rather, he was intercepted by the police in the street. The appellant did not tell the police or the judge that he returned to Hong Kong from the Mainland in July 2012 because he wanted to surrender to the police in connection with the robbery that he had committed in March 2012. 20. The appellant’s assertion that he came back to Hong Kong because he wanted to surrender and confess to the robbery offence was highly improbable and most unconvincing. In any event, the applicant’s assertion was rejected by the judge. As such an assertion was no longer relied on by Mr Grossman, we would say no more about it. 21. Mr Grossman’s complaint that the judge had not given heed to the need of encouraging criminals to assist law enforcement authorities was misplaced. It was such an obvious matter that the judge could not have overlooked it. If the judge had not given heed to the appellant’s volunteered confession, he would not have given him the additional 3 months’ discount of sentence. 22. The only issue that concerned us was whether the judge’s decision of not giving the appellant a larger sentence discount than the 3 months for his volunteered confession was appealable. 23. Mr Grossman suggested that where a person confessed to a crime, the commission of which was unknown to the authorities or confessed his complicity in a known crime when his complicity was neither known nor suspected, he should be given an additional discount over and above the one-third discount for a guilty plea as an added element of leniency. Mr Grossman, in his written submissions, relied heavily on HKSAR v Hui Chi Tong (unreported CACC 414/2007), HKSAR v Tsang Kai On [2011] 2 HKLRD 340 and HKSAR v Choi Ka Kin Seraphim (unreported CACC 377/2012). 24. In para. 25 of the judgment of Hui Chi Tong (supra), the Court of Appeal said: “ But for the voluntary confession, the applicant would not have been found guilty of (the) 1st charge. Such voluntary admission, coupled with a plea later, showed genuine remorse. It is in the public interest to encourage an offender to ‘own up’ to his misdeed and to face the consequence. The most effective way of giving encouragement is to give a greater than normal discount upon a plea.” 25. In Tsang Kai On (supra), the Court of Appeal made similar comments in para. 14 of the judgment: “ That the applicant still voluntarily pleaded guilty notwithstanding a lack of sufficient evidence from the prosecution showed that he was genuinely remorseful. In these circumstances, the Court should give a further discount to the applicant as an encouragement. This is also in the public interest…” 26. In both Hui Chi Tong (supra) and Tsang Kai On (supra), the Court of Appeal had not examined the theoretical basis and the logical foundation of the further discount other than saying that it was in the public interest to do so. The Court of Appeal did not indicate the appropriate extent of the further discount either. 27. In Choi Ka Kin Seraphim (supra), the Court of Appeal suggested that a total sentence discount of 40% would be appropriate when the defendant not only pleaded guilty to the charge, but also gave useful information to the police resulting in the arrest and successful prosecution of another defendant as well as voluntarily admitted to other offences for which he had not been arrested. 28. However, it was the policy of the courts to take into account in mitigation of sentence useful assistance given by the defendant to the authorities. The well established extent of reduction for a defendant who pleaded guilty and provided assistance to the authorities is 40% to 45%. (See HKSAR v Ng Shek Yu CACC 178/2000) 29. The Court of Final Appeal in Z v HKSAR (2007) 10 HKCFAR 183 affirmed such an approach when Li CJ said at p 194 of the judgment of the court: “ The Court of Appeal is well placed to consider the range of reductions of sentence for co-operation with the authorities. It has used various percentages of discounts for different degrees of assistance. It has applied a usual discount of 40% (including the one-third reduction for a guilty plea) for an appellant who had provided assistance to the authorities without testifying against those about whom they had provided information. HKSAR v Y [2005] 3 HKC 337 at p.340. (Compare its earlier decision in HKSAR v Ng Shek Yu (unrep., CACC 178/2000, [2001] HKEC 243) referring to a discount of between 40% to 45% as usually appropriate in such circumstances.)… … In the present case, the relevant figure is the usual discount of about 40% as the appellant had provided useful information but had not given evidence. Its appropriateness had not been questioned in this appeal…” 30. If the Court of Appeal in Choi Ka Kin Seraphim (supra) had intended to give the defendant who pleaded guilty and gave useful information an additional sentence discount because he had voluntarily admitted to other offences, the discount should have been more than 40%. 31. Mr Grossman also relied on English authorities such as R v Claydon (1994) 15 Cr App R (S) 526, R v Hoult (1990) 12 Cr App R (S) 180, R v Ellis (1986) 6 NSWLR 603, R v Mitsuaki Richard Mirchandaney [2003] EWCA Crim 336and suggested that the appellant should be given a discount of sentence of more than one-third. 32. Quite apart from the fact that there were other relevant statutory provisions regulating sentences, the cases relied on by Mr Grossman all concerned defendants who voluntarily surrendered to the police and admitted the offences when the police had no evidence to connect them to the offences. 33. In Claydon (supra), the defendant surrendered to the police sometime after the offence and admitted the offence. In Hoult (supra), the defendant went to a police station and admitted his part in a robbery despite the fact that police inquiries were unsuccessful and the crime was treated as unsolved. 34. In Mirchandaney (supra), the defendant committed robbery and one week later, he made an anonymous call to the police seeking to confess his part in the robbery. He was told that the police would not take anonymous telephone calls. He then handed himself in to the police station one day after the telephone call and disclosed his part in the robbery. 35. In Ellis (supra), the defendant, having committed a series of robberies between September and November 1984, attended a minister of religion in December 1984 and confessed his involvement in those offences. He was advised to see a solicitor to make a clean breast of them. He then consulted a solicitor and was interviewed by the police following upon his solicitor having telephoned them and informed them that he had a client who wished to disclose his guilt. 36. We accept that in individual cases where a conscience-stricken defendant voluntarily surrendered to the authorities and admitted the offences despite a total lack of evidence against him, the court could grant him a sentence discount of more than one-third for such exceptional remorse. However, this was not such a case. 37. The judge had found and it was no longer disputed that the appellant did not voluntarily surrender himself to the police, but was intercepted and arrested for a separate offence before he confessed to the robbery offence. 38. The question of whether a defendant who voluntarily confessed and pleaded guilty to certain offences notwithstanding a lack of evidence from the prosecution to link the offences to him should be given a discount of more than one-third has recently been examined by this court in HKSAR v Ma Ming [2013] 1 HKLRD 813. 39. The Court disapproved of the suggestion in Hui Chi Tong (supra) and Tsang Kai On (supra) that a defendant who voluntarily confessed and pleaded guilty despite a lack of evidence against him would be entitled to a sentence discount of more than one-third. The Court wrote in its judgment of Ma Ming (supra) the following: “ 27. A discount of one-third is quite a substantial discount. One of the main purposes of the court giving this one-third discount to a defendant who pleads guilty is to encourage a guilty person to own up to the crimes he committed, so as to conserve the resources of the community and to ensure that justice can be administered more efficiently and matters can be concluded in the most expeditious manner. … 29. In view of this, when giving the one-third discount the approach taken by the court is a firm and broad-brush approach. It gives a defendant who timely pleads guilty the one-third discount, which is substantial, without regard for niceties, in order to discourage excessive arguments and to prevent wasteful use of resources of the community. 30. In Secretary for Justice v Lee Chun Ho Jeef [2009] 6 HKC 471, this court gave a clear exposition of this stance. At A-C on page 44 [sic] 7 of the judgment the Court of Appeal pointed out that: ‘ The Court of Appeal have repeatedly emphasized that the one-third discount is “usually to be regarded as the high watermark of the discount given to a defendant pleading guilty in good time.” The respondent’s co-operation with the police referred to by the judge was his admission of the offence to the police, both at the scene of the crime and at the police station. Such “cooperation” is just part and parcel of the respondent’s admission of his guilty, albeit at the first available opportunity. This mitigating factor should be subsumed within the one-third discount.’ 31. The applicant cooperated with the police and after he was arrested he frankly confessed. This led to his being charged with the 1st charge. We are of the view that this factor should also be subsumed within the one-third discount, and should not be treated as a reason for giving any discount of more than one-third. Otherwise, it would be in conflict with the policy and purpose of giving the one-third discount. Not only would it create uncertainty but also it would give rise to disputes, and so it is against public interest. 32. When dealing with individual cases and when considering the totality of the sentence, the court, in exercising its discretion, can take into account the fact that the frank confession of the defendant provides the only evidence which support the charge of charges and therefore make minor adjustments to the total sentence. To this we do not object. However, this factor does not support the argument that this kind of defendant must be given a further discount over and above the one-third discount. If the court does not give such defendants any discount in addition to the one-third discount, that does not constitute an arguable ground of appeal. (Emphasis added).” 40. The same point was made in HKSAR v Chu Kwok Chu [2013] 6 HKC 357 at p 362 F-H: “ The fact that a judge may, in the exercise of his discretion, give a defendant, who voluntarily confesses to the offence and without whose confession the prosecution would have no evidence against him or who voluntarily pleads guilty notwithstanding a lack of sufficient evidence, a further discount of sentence other than the usual one-third does not mean that where a judge refuses to do so, the defendant will have a legitimate complaint and can successfully appeal against such a refusal.” 41. We remain of the same view. Despite the fact that the appellant, after his arrest for failing to produce an identification document, had confessed to the robbery charge, the judge was not obliged to give him a discount over and above the one-third discount upon his pleading guilty to the charges. Such a mitigating factor was subsumed within the one-third discount. 42. In any event, the judge did give the appellant an additional three-month discount and ordered only part of the sentence on the unlawful remaining charge to run consecutively to the sentence on the robbery charge. The judge’s approach was in fact a lenient one and the appellant cannot have any legitimate complaint. The total sentence is neither manifestly excessive nor wrong in principle. 43. The only ground of appeal put forward on behalf of the appellant was without merit and the appeal against sentence is therefore dismissed. 44. I agree with the judgments of Lunn VP in CACC 418/2014 and CACC 327/2015 and of Macrae JA in CACC 327/2015. CRIMINAL APPEAL NO. 418 OF 2014 and CRIMINAL APPEAL NO. 327 OF 2015 Hon Lunn VP : 45. At the hearing of CACC 418/2014 on 21 January 2016, we adjourned proceedings to 19 and 20 April 2016 in order to receive submissions as to the general practice of affording a discount in sentence to a defendant for his plea(s) of guilty, directed the Registrar to brief leading counsel to act as an amicus curiae and invited the Director of Legal Aid and the Director of Public Prosecutions, to brief leading and junior counsel for the appellant and respondent respectively to assist the court. In addition, we directed that the application for leave to appeal against sentence in CACC 327/2015 be heard together with the appeal in CACC 418/2014. 46. The two cases before this Court afford the Court an opportunity generally to reconsider its policy of affording a discount of sentence to those defendants who plead guilty to the charges brought against them. In doing so, we have been assisted greatly by the submissions of the amici curiae, Mr Robert Pang SC and Ms Maggie Wong, together with the detailed and helpful submissions of the Director of Public Prosecutions, Mr Keith Yeung SC and those of Mr Michael Blanchflower SC. 47. In the judgment of this Court, similarly constituted, in HKSAR v Lo Kam Fai,[1] Yeung VP noted that, “A defendant who enters a timely plea of guilty is normally entitled to a sentence discount of one third from the starting point because it is in the public interest to do so.” Yeung VP went on to cite with approval the judgment of Kirby J [2] in the High Court of Australia in R v Cameron[3] in which under the rubric of ‘The consideration of the public interest’, he said: “ The main features of the public interest, relevant to the discount for a plea of guilty, are “purely utilitarian”. They include the fact that a plea of guilty saves the community the cost and inconvenience of the trial of the prisoner which must otherwise be undertaken. It also involves a saving in costs that must otherwise be expended upon the provision of judicial and court facilities; prosecutorial operations; the supply of legal aid to accused persons; witness fees; and the fees paid, and inconvenience caused, to any jurors summoned to perform jury service. Even a plea at a late stage, indeed even one offered on the day of trial or during a trial, may, to some extent, involve savings of all these kinds. … it is in the public interest to facilitate pleas of guilty by those who are guilty and to conserve the trial process substantially to cases where there is a real contest about guilt. Doing this helps ease the congestion in the courts that delay the hearing of such trials as must be held. It also encourages the clear-up rate for crime and so vindicates public confidence in the processes established to protect the community and uphold its laws. A plea of guilty may also help the victims of crime to put their experience behind them; to receive vindication and support from their families and friends and possibly assistance from the community for injuries they have suffered. Especially in cases of homicide and sexual offences, a plea of guilty may spare the victim or the victim’s family and friends the ordeal of having to give evidence.” 48. Having noted that this Court had observed in HKSAR v Ma Ming [4]that, “Difficulties and disputes may arise in carrying out this policy of giving a one-third discount and in deciding what a timely plea of guilty is”,[5] Yeung VP said:[6] “ In order to avoid uncertainty, delay and unnecessary expense arising out of arguments over niceties and subtleties, the court, as a matter of policy, would normally give a defendant a sentence discount of one-third from the starting point as long as he pleads guilty before the trial starts. Such a policy can be abused. Defendants who plead guilty at a very late stage when they should plead guilty at much earlier stages still get the full one-third discount and very often defendants deliberately wait until the last moment before pleading guilty, resulting in a wastage of huge public resources. The policy that a defendant will get the full one-third discount as long as he pleads guilty at any time before the trial starts may have to be adjusted to deal with such an abuse, but it is not a matter that should affect the outcome of this appeal.” [Italics added.] 49. Macrae JA also expressed concerns about the fairness of the policy of affording the same discount in sentence to defendants who pleaded guilty at different stages in the proceedings and articulated his misgivings of the blurring of the notion of a “timely plea”:[7] “ However, the notion of a “timely plea” has become somewhat distorted, and regrettably devalued, by the way our courts have over many years come to accord the one-third discount to any plea, however untimely, as long as it is entered before the trial is due to begin. Thus, the defendant on a rape charge, who indicates his intention of contesting his guilt, thereby obliging the case to be fully prepared with expert DNA and other scientific evidence and fixed for trial before the High Court, with the jury about to be empanelled and the traumatised victim poised to give evidence, will receive the same discount if he pleads guilty at arraignment as the drug trafficker who tenders a plea of guilty in the magistrate’s court and is committed to the High Court for sentence. The defendant who obliges the prosecution to prepare a complex fraud case over months, sometimes years, with flow charts, schedules, banker’s affirmations and expert accounting evidence, on the basis that he will plead not guilty but then changes his plea on the first day of a trial set down for several months, will generally receive the same discount for that plea as the defendant who acknowledges his guilt from the outset, cooperates with the authorities and demonstrates genuine remorse.” The Prison Rules 50. In his written submissions, Mr Yeung SC identified what he described as, “clear motivations” for a defendant to delay pleading guilty until the commencement of his trial if, nevertheless, he was to be afforded a one-third discount from the starting point taken for sentence. As he pointed out, the Prison Rules [8] differentiate between, “prisoners awaiting trial” and a person who is a “convicted prisoner”. Prisoners awaiting trial are defined in rule 188(1) as including persons committed to prison for safe custody in any of the following circumstances: “ (a) on their committal for trial for any indictable offence; (aa) on their detention pursuant to an order of transfer made under section 88 of the Magistrates Ordinance; (b) on their detention pending the hearing before a magistrate of a charge against them on an indictable offence; (c) on their detention pending the hearing of an information or complaint against them;” 51. The rules make for separate treatment for prisoners awaiting trial, so that he: (i) may procure for himself, or receive at proper hours, food and malt liquor;[9] (ii) shall not, during any period of 24 hours, receive or purchase more than one pint of malt liquor or cider, or more than half a pint of wine;[10] (iii) may wear his own clothes if sufficient and fit for use;[11] (iv) shall have the option of employment in the service or industries of the prison at his election;[12] (v) shall, subject to the order of the Superintendent, be permitted to be visited by one visitor, or if circumstances permit, by two at the same time, for a quarter of an hour on any week day, during such hour as may from time to time be appointed;[13] (vi) may send and receive letters at all reasonable times and shall be furnished by the Superintendent with a reasonable amount of paper and other writing materials for purpose of communicating with his friends or for preparing his defence.[14] 52. By contrast, in respect of a prisoner: • He is required to engage in useful work for not more than 10 hours a day, of which so far as practicable at least 8 hours shall be spent in associated or other work outside the cells, rooms, dormitories or wards.[15] • No persons, other than the relatives and friends of a prisoner, shall be allowed to visit him, except by special authority… (a) they shall be allowed to visit a prisoner twice a month and no more than 3 persons shall be allowed at one time; (b) … limited to 30 minutes on each occasion.[16] 53. Nevertheless, by operation of section 67A of the Criminal Procedure Ordinance, Cap. 221, the sentence imposed on a defendant is reduced by the period he was detained in custody as a prisoner awaiting trial by an order of a court.[17] Statistics 54. At the outset of the hearing, as intimated in a Notice of Motion filed by the Department of Justice, the Court received affirmations from Aaron Lee Man Kie, the Chief Court Prosecutor of the Prosecutions Division, and Chow Sze Yu, the Senior Law Clerk of the Prosecutions Division, in which they provided statistical information about the manner in which cases were dealt with in the Magistracy, District Court and Court of First Instance for the calendar years 2014 and 2015. Together with a Notice of Motion, filed by the Department of Justice on 15 July 2016, without objection from any of the parties, the Court received an affirmation of Kei Wing Cheong, a law clerk of the Court of First Instance unit of the Prosecutions Division, Department of Justice, which provided statistical information in relation to cases committed to the Court of First Instance. The Magistracy 55. In 2014, of 10,862 defendants whose trials were fixed upon their pleas of ‘Not guilty’, 4,232 (38.96%) pleaded guilty to some or all of the charges brought against them at or after commencement of the trial. In 2015, of 9,811 defendants whose trials were fixed upon their pleas of ‘Not guilty’, 3,657 defendants (37.27%) pleaded guilty to some or all of the charges brought against them at or after commencement of the trial. The District Court 56. In 2014, of the 556 cases fixed for trial, upon an indication of a plea of ‘Not guilty’ at the Plea Day hearing, in 411 cases (73.92%) some or all of the defendants pleaded guilty to some or all of the charges at the commencement of the trial. In 2015, of the 474 cases fixed for trial, upon an indication of a plea of ‘Not guilty’ at the Plea Day hearing, in 345 cases (72.78%) some or all of the defendants pleaded guilty to some or all of the charges at the commencement of the trial. The Court of First Instance Committals for sentence 57. In 2014 and 2015, 202 and 172 cases respectively were committed to the Court of First Instance from the Magistracy for sentence. Committals for trial: subsequent pleas of guilty 58. In 2014 and 2015, 284 and 301 cases respectively were committed for trial in the Court of First Instance from the Magistracy. Of the “defendants pleading guilty to some or all of the charges after the committal proceedings”, 41.15% and 46.30% did so in 2014 and 2015 respectively. The discount of sentence to be afforded to a defendant for plea of guilty 59. A review of the judgments of this Court over the past 40 years demonstrates changes in the policy of affording a discount in sentence to a defendant on his plea(s) of guilty to the charges laid against him. As is readily apparent, in the 1980s it was the practice to distinguish between those defendants who tendered a plea of guilty at the earliest “practical stage”[18], “earliest possible moment”[19], “first opportunity” [20] and those who pleaded guilty at a later stage in the proceedings, including the “first morning of trial”.[21] A discount from the starting point taken for sentence of up to 25% was afforded to the former and up to 20% for the latter.[22] The strength of the prosecution case was one of the factors taken into account in reducing the discount given to a defendant who pleaded guilty.[23] Later, it was said that a discount of one-third [24] from the starting point taken for sentence was to be afforded to the former and a reduced discount afforded to the latter. Then, the practice of reducing the discount given to a defendant, who pleaded guilty, to reflect the strength of the prosecution case was rejected.[25] Subsequently, it was stated that it was now the practice of the Courts, subject to limited exceptions, to afford the defendant a discount of one-third from the starting point taken for sentence, even if the defendant tendered his plea only on the first day of his trial.[26] Discounts in sentence for the timing of a plea of guilty: up to 20% and up to 25%. 60. In the Queen v Chan Chi Yuen [27], this Court allowed an application for review of sentence imposed on the respondent on his conviction on his pleas of guilty to four charges of robbery, in which the applicant was one of two or more robbers, one of whom was armed with a knife, and imposed sentences of 2 years and 9 months’ imprisonment on each of the four charges, which sentences were ordered to be served concurrently. In the judgment of this Court, McMullin VP said that the appropriate sentence for the offences after trial was in the range of 5 to 6 years’ imprisonment. Having noted that the judge had accepted that, “…responsibility for none of these offences would ever have been laid to the Respondent’s door but for his own confessions”, McMullin VP said that, “a considerable discount should be allowed for that.” However, he went on to say: “ Even in cases where there is an abundance of evidence available for trial and there has been a plea of guilty, it has been a familiar practice in these courts to allow as much as 25% for the plea. That of course depends upon the quality of the plea and the point of time at which it is offered.” [Italics added.] 61. In the judgment of this Court in the Queen v Leung Yiu Hung & Others,[28] Yang JA, as Yang CJ was then, declined “to lay down certain guidelines and discounts where pleas of guilty are entered.” He said: “ We do not think it would be helpful to lay down such guidelines because each case must be dealt with on its individual merits, depending on the circumstances of the case and the circumstances which led an accused plead guilty.” 62. In the judgment of this Court in the Queen v Wong Ping Yu & Another [29], Roberts CJ observed that it was appropriate for the Court to afford a greater discount to a defendant, who had pleaded guilty in the Magistracy and who had been committed for sentence to the High Court, than that afforded to a defendant who pleaded guilty on the first morning of trial: “ It is common for discounts of up to a fifth to be allowed in serious cases if there is a plea on the first morning of trial. In this instance, there was a plea at the earliest practical stage, that is to say, when the defendants appeared before the Magistrate, as a result of which they were merely committed to the High Court for sentence. This is a practice which is to be encouraged in the public interest, and we feel that it should be recognized by a greater discount than a plea on the first morning of a trial some weeks later would justify. We think that an appropriate discount would have been approximately 25%.” [Italics added.] 63. In the Attorney General v Han Man Fai & Another [30], on arraignment the two applicants had pleaded guilty to a count of unlawful trafficking in, and another count of possession of, dangerous drugs. Having stipulated a starting point for sentence of 9 years’ imprisonment for the former count, the judge afforded them a discount of one-third for their pleas of guilty and sentenced each of them to 6 years’ imprisonment. Having adverted tothe Queen v Chan Chi Yuen and the Queen v Leung Yiu Hung & Others,in the judgment of this Court, Cons VP said:[31] “ We do not wish to be drawn into any mathematical discussion. We would only say that judges do appear to us to give discounts, depending of course on the circumstances of the individual case, which are generally in the region of 25%.” 64. In the result, having determined that the appropriate starting point to be taken for sentence was 12 years’ imprisonment, and having noted that pleas had been indicated at the pre-trial hearing, Cons VP said, “… indications at that stage should be encouraged.” Of the starting point, Cons VP said that it [32]: “ ... could properly have been discounted to 9 years for their pleas of guilty.” Of course, that was a discount of 25% from the starting point taken for sentence. Having regard to the fact that the application was one of review of sentence, in quashing the original sentences of 6 years’ imprisonment the Court imposed concurrent sentences of 8 years’ imprisonment on each of the counts against each of the applicants. 65. In the Queen v Kwok Chi Kwan and Another [33],the two applicants had pleaded guilty at Plea Day in the District Court to four and five counts of robbery respectively. Nevertheless, the judge sentenced them to the maximum sentence, namely 7 years’ imprisonment, which can be imposed on the District Court. In allowing their applications for leave to appeal against sentence and imposing sentences of 6 years’ and 6½ years’ imprisonment respectively, in the judgment of the Court, Silke VP addressed the rationale for allowing discounts from otherwise appropriate sentences to defendants who plead guilty:[34] “ Pleas of guilty are to be encouraged for various well known reasons: to give allowance for the remorse indicated by such a course - though of course “remorse” can take many forms, from the genuine sorrow to an acceptance of the inevitable -; to assist in the saving of time - thus bringing more speedily to trial cases waiting in the lists and expense: to avoid the necessity for the bringing of witnesses to Court - thus avoiding disruption in the daily lives of those involved. Pleas made at the earliest possible moment deserve greater encouragement.” [Italics added.] 66. In the Queen v Leung Tin Man [35], this Court refused an application for leave to appeal against a sentence of 14 years’ imprisonment imposed on the applicant on his plea of guilty to a count of unlawful trafficking in dangerous drugs. The judge had stipulated a starting point for sentence of 16 years’ imprisonment, and afforded the applicant a discount of 2 years’ imprisonment only. Of the observation in the judgment of McMullin VP in the Queen v Chan Chi Yuen that, “…it has been a familiar practice in these courts to allow as much as 25% for the plea”, in the judgment of the Court Silke VP said: [36] “ …it has been made clear that the words “as much as” do not mean that in every single case that discount on a plea of guilty must be 25%. The quantum is a matter within the discretion of the trial judge and in exercising that discretion he would no doubt bear in mind the strength of the evidence and any other matter affecting either the offence or the offender in coming to the discount he thinks proper to give up on a plea of guilty. As we indicated this applicant was caught red-handed. ” 67. In the Attorney General v Wong Kwok Wai [37], this Court allowed an application for a review of a sentence of 10 years’ imprisonment, imposed on the respondent following his plea of guilty in the Magistracy and his committal for sentence in the High Court for an offence of unlawful trafficking in dangerous drugs, and said that a sentence of 15 years’ imprisonment would have been appropriate. However, given that the application was for a review of sentence, the Court substituted a sentence of 14 years’ imprisonment. The judge had taken a starting point for sentence of 17 years’ imprisonment, and said that he assessed the appropriate discount of sentence to be 40%, to reflect the plea of guilty and the assistance given to the authorities. 68. This Court determined that the appropriate starting point was 20 years’ imprisonment. Noting that the information provided to the authorities had not proved fruitful and that the applicant had merely offered to give evidence if there were arrests made in the future, this Court said that was “not something which should be considered in mitigation”. Silke VP in the judgment of the Court said:[38] “ This court does not encourage mathematical calculations of discount. Such a method can well lead to submissions in subsequent cases that one man got, say 35%, and “I got only 30%, please give me “the same” or to a form of fixed percentage which becomes immutable. Each case must depend entirely on its own facts and no doubt a sentencing judge in the exercise of his discretion will bear in mind all the many factors - or the lack of them - which permit him to reduce his starting point sentence.” 69. In the result, Silke VP concluded:[39] “ Had the trial judge, giving all due allowance for the plea at (the) first opportunity which, as we have said, was the major mitigation in this case coupled with the frank admissions from the outset, adopted a starting point of 20 years, and imposed a sentence of 15 years imprisonment he would not have been wrong.” [Italics added.] Of course, that discount of sentence was 25% from the starting point taken for sentence that this Court said was appropriate. 70. In giving the judgment of this Court in the Queen v Law Hon Chung [40], allowing the appeal against a sentence of 10 years’ imprisonment imposed for a conviction for robbery and imposing a sentence of 8 years’ imprisonment, Silke VP said that the Court was not prepared to say that the starting point of 12 years’ imprisonment stipulated by the judge was wrong. However, he went on to say that:[41] “ …the trial judge does not seem to have given full consideration to the effect of that which the applicant said after his arrest which led, as we have indicated earlier, to the arrest of his confederates and the recovery of the property.” 71. In that case, having been arrested, the defendant “gave a full confessional statement in the course of which he named his co-defendants and gave details of the planning of the robbery. Resulting from that information the other four persons were arrested and all the property was recovered.” Then, he pleaded guilty in the Magistracy and had been committed for sentence to the High Court. 72. Of those circumstances, Silke VP said:[42] “ People who plead guilty at the earliest possible moment, as did the applicant here, are entitled to a greater discount than that which normally attaches to a plea of guilty. In all the circumstances here, and with particular reference to the information which the applicant had given and his very early plea, we consider the sentence of eight years imprisonment would have been appropriate.” [Italics added.] 73. Clearly, the discount of one-third from the starting point that this Court afforded the appellant, reflected not only his early plea of guilty in the Magistracy but also the assistance that he had rendered the authorities, which had resulted in the arrest of others and the recovery of stolen property. 74. In the Queen v Lai Kwok Hung [43], this Court allowed an appeal against a sentenceof 9 years’ imprisonment imposed on the applicant in the High Court, on his committal to that court for sentence, following his plea of guilty in the Magistracy to a single count of unlawfully trafficking in dangerous drugs. The judge stipulated a starting point for sentence of 11 years’ imprisonment and, affording the applicant a discount of 2 years’ imprisonment for his “plea and his admission at an early stage, but having regard also to the fact that his plea ofguilty could be regarded as little more than a recognition of the inevitable.” 75. In the judgment of this Court, Mortimer JA noted that it was submitted on behalf of the applicant that:[44] “ it is disappointing to see a judge, giving credit for a plea of guilt at an early stage but taking away from the applicant much of the benefit by saying that it could be regarded as ‘little more than a recognition of the inevitable’. He submits that pleas of guilty are to be encouraged. He submits - and we accept - that there is no such thing as an inevitable plea nor an inevitable conviction.” [Italics added.] 76. Mortimer JA went on to say:[45] “ If courts do not recognise that an early plea is an expression of remorse and if those who plead guilty and save time and expense to the public are not given full credit, there will be little benefit for an accused to plead guilty. For our part, we also recognise that a failure to give proper weight to a plea puts counsel in difficulty in the advice he is to give to those who may be inclined to be remorseful and accept their guilt.” [Italics added.] 77. In the result, this Court substituted a sentence of 7 years’ imprisonment, “to reflect the plea that was tendered at an early stage.” That represented a discount from the starting point of 36.6%. A one-third discount for a plea of guilty 78. In the Queen v Chan Leung[46] this Court refused an application for leave to appeal against sentence by an appellant who had been sentenced to 2 years and 3 months’ imprisonment for an offence of attempting to export four motor cars on an unnumbered Mainland vessel from Hong Kong without lawful excuse. On being intercepted by the police, as motor cars were being loaded on to his vessel from a lighter, the applicant had attempted to escape in the vessel. Having taken a starting point for sentence of 3 years’ imprisonment, the judge discounted sentence, for the plea of guilty, by 9 months’ imprisonment only. That was a discount of 25%. Of the suggestion that the judge erred in not affording the applicant a discount of one-third from the starting point, in the judgment of this Court Liu JA said: “ …there is no inflexible rule that a discount of one-third should be granted where there is a plea of guilty.” 79. In the Queen v Ng Wing Kwong [47], this Court refused an application for leave to appeal against sentence by an applicant who had been sentenced to 15 years’ imprisonment, having pleaded guilty at his re-trial to a single count of unlawfully trafficking in dangerous drugs. In the first trial, the applicant had pleaded not guilty but, on being convicted, had been sentenced to 20 years’ imprisonment. The judge took 20 years as a starting point for sentence in the re-trial. So, the discount afforded to the applicant was 25% only. In the judgment of this Court, Mayo JA said: “ One of the matters complained of is that the trial judge failed to give the applicant sufficient discount for his guilty of plea. While the question of the amount of discount is very much a matter for the consideration of the trial judge there was a complication in the present case. The trial was a retrial. On the previous occasion the applicant had pleaded not guilty to the offence and had been sentenced to 20 years’ imprisonment. It is in these circumstances that the trial judge did not give a discount of 33% of the term. We do not think that the trial judge’s approach in this respect can be faulted.” 80. In the Queen v Wong Ngai Hung [48], this Court allowed an appeal against a sentence of 5½ years’ imprisonment and substituted a sentence of 4½ years’ imprisonment. The judge had taken a starting point of 6½ years’ imprisonment, following the appellant’s plea of guilty in the District Court to a charge of unlawfully trafficking in heroin, and reduced it by one year’s imprisonment only. He did so, first because the plea of guilty had not been made at the first possible opportunity, but only on the first day of trial and, secondly because it had been made in the face of overwhelming evidence. In the judgment of this Court, Mortimer JA said: “ This Court has said on many occasions that - other things being equal - credit for a plea of guilty ought to be in the range of one-third reduction but the matter is within the discretion of the sentencing judge when he takes into account all the surrounding circumstances. It is therefore not wrong for him to take into account the time when the plea is made.” [Italics added.] 81. Of the fact that the plea of guilty was tendered at the first day of trial, he said: “ … it appears that in the District Court an accused person will appear first on a plea day. It is an opportunity for him to plead guilty if he wishes. But in this particular case, had the judge made inquiries, he might have discovered that there was some confusion in the applicant’s mind at the time of that plea day and that the advice which was then available to him was limited. It may well be the point taken against the applicant by the judge that he had not made a plea on the first opportunity is not sound. But in any event, the merit of a plea of guilty, even just before the commencement of a trial, is considerable. Credit should be given even at (that ?) time because there is still saving of time and expense although obviously not as much as if a plea is made at an earlier stage.” [Italics added.] 82. Of the fact that the plea of guilty had been tendered in the face of overwhelming evidence, he went on to say: “ Again, that is a matter that the judge can take into account. What weight it should be given depends upon all the circumstances. But even if the evidence is overwhelming, it does not mean that there should be minimal credit. It is rightly said that there is no such thing as an open-and-shut case. This is particularly so in drugs cases. The entering of a plea — even with overwhelming evidence — is of more than minimal weight.” The discount from the starting point taken for sentence afforded by this Court was 30.8%. 83. In R v Lo Chi Man [49], this Court allowed an appeal against a sentence of 4 years and 9 months’ imprisonment imposed on the applicant following his plea of guilty to a charge of robbery and substituted a sentence of 4 years’ imprisonment. In November 1993, the victim had been threatened with a knife and tied up with tape by three robbers. Having stipulated a starting point for sentence of 6 years’ imprisonment and having noted that the applicant had absconded to Macau and remained at large until September 1995, the judge said: “ The only mitigation in your favour is your plea of guilty, although you cannot expect the same discount for plea as you would have received if you had pleaded at the same time and your companion in crime.” 84. In the judgment of this Court, Nazareth VP said: “ In recent times particularly, this Court has repeatedly stressed the importance of giving an adequate discount for pleas of guilty, it should be of the order of one-third. The discount not merely recognises remorse, but the considerable benefit that enures to the courts and the expeditious administration of justice. The latter is served no less if the plea comes to be made long after the crime.” [Italics added.] 85. In the result, Nazareth VP concluded: “ The appellant’s full admission to the police followed by his plea saved the police, the prosecution and the courts the work and time that would otherwise have been required. We are unable to see any good reason for denying the applicant the usual discount of one-third.” [Italics added.] 86. In the Queen v Lau Kin Hong[50], a Magisterial appeal, Chan J, as Chan NPJ was then, dismissed an appeal from a sentence of 14 months’ imprisonment imposed for an offence of theft, which the magistrate treated as a case of pickpocketing, and for which he had stipulated a starting point for the sentence of 15 months’ imprisonment. In doing so, the magistrate noted that the plea of guilty had been tendered at a late stage and there was no indication of remorse. Chan J said: “ The usual one-third discount is, in my view, reserved for those cases where the defendant pleads guilty at the earliest possible opportunity, has shown genuine remorse and has saved the court and everybody some time. In the present case, the appellant appeared to have pleaded guilty only at the very last moment - on two previous available occasions, he did not. It is also quite clear that there was no sign of any genuine remorse on the appellant.” [Italics added.] 87. In the Queen v Wu Yau Man [51] this Court set aside a sentence of 2 years and 6 months’ imprisonment imposed on the applicant following his plea of guilty in the District Court to a charge of unlawful possession of heroin hydrochloride and ordered that the applicant continue to be the subject of an existing Drug Addiction Treatment Centre Order. Having stipulated a starting point for sentence of 3 years’ imprisonment, the judge afforded the applicant a discount of 6 months’ imprisonment only. In the judgment of this Court, Nazareth VP said: “ The judge’s failure to accord the appellant the full one-third discount this Court has repeatedly affirmed in recent times as the usual discount, notwithstanding his reference to the appellant’s early admission of possession, only served to increase our disquiet.” 88. In the Queen v Lun Nai Kin [52]this Court allowed an appeal against a sentence of 2½ years’ imprisonment imposed on the applicant after his conviction after trial in the District Court of unlawful possession of heroin hydrochloride and substituted a sentence of 2 years’ imprisonment. The judge acquitted the applicant of the charge of unlawfully trafficking in that amount of dangerous drugs. At all times the appellant had been prepared to plead guilty to the charge of possession only. Having stipulated a starting point for sentence of 3 years’ imprisonment, the judge reduced that sentence to 2½ years’ imprisonment only, reflecting the fact that the applicant had admitted possession of dangerous drugs, but also having regard to the fact that he had been caught red-handed. 89. In the judgment of this Court, Power CJ (Ag.) said: “ We have said more than once that an appropriate discount for plea is a discount of one-third. We have also said that the question of discount is a matter for the discretion of the trial judge, but that this discretion must be exercised upon some rational basis which the trial judge should advert to in his Reasons for Sentence. The trial judge did not give any reason why he was not giving the full one-third discount in this case and we, having considered the facts, are unable to discover any reason why that discount should not have been given.” [Italics added.] 90. In the Queen v Guo Jun [53] this Court refused an application for leave to appeal against a total sentence of 16 years’ imprisonment imposed on the applicant following his pleas of guilty to an offence of attempted robbery, false imprisonment and possession of arms and ammunition without a licence. The judge had taken a global starting point for sentence of 20 years’ imprisonment. Of the discount afforded to the applicant, having noted that it was agreed that the applicant had “tried to plead guilty at an early stage”, in the judgment of the Court, Yeung J, as Yeung VP was then, said: “ It is true to say that there has been a practice in the courts in Hong Kong, in order to encourage accuseds in appropriate cases to plead guilty, that a guilty plea will reduce the sentence by one-third. But this practice is not a strait-jacket rule. In our view, the trial judge must be given some discretion on matters of this kind.” [Italics added.] 91. In HKSAR v Leung Kwai Sing [54], this Court allowed the applicant’s appeal against a sentence of 3 years’ imprisonment and substituted a sentence of 2 years and 8 months’ imprisonment for an offence of unlawful possession of heroin hydrochloride. On the first day of his trial the applicant pleaded not guilty to a charge of unlawfully trafficking in the dangerous drugs, but guilty to possession of those dangerous drugs. Having stipulated a starting point for sentence of 4 years’ imprisonment, the judge said that he afforded the applicant “a reduction of 25% for his plea on the day of trial”. 92. In the judgment of this Court, Power VP said: “ We appreciate the approach which the judge took. He was following the line, which is one not infrequently followed and is a quite proper one, that if a person delays his plea until the day of trial he is not entitled to the full one-third discount. However, in the present case he was only given the opportunity to plead to the charge of simple possession on the day of trial and he did, therefore, enter his plea at the earliest available opportunity. It is our view that, in such circumstances, he should receive the full one-third discount.” [Italics added.] [55] Overwhelming evidence: relevance to a plea of guilty 93. The substantial reservations expressed by this Court in the judgments of Mortimer JA in the Queen v Wong Ngai Hung andthe Queen v Lai Kwok Hung as to the validity of the practice of reducing the discount afforded to a defendant for his plea of guilty, for the reason that the evidence against him was overwhelming, were taken further by the judgments of this Court in HKSAR v Wong Ka Kuen [56]and in HKSAR v Yeung Kin Man [57], delivered respectively by Stuart-Moore VP and Leong JA. 94. In HKSAR v Wong Ka Kuen, this Court allowed an appeal against a total sentence of 19 years’ imprisonment, imposed after the 2nd appellant pleaded guilty, to a count of unlawful trafficking in heroin hydrochloride and of manufacturing heroin hydrochloride, and substituted a sentence of 18 years’ imprisonment. The 2nd appellant pleaded guilty “at the outset” [58], whereas his co-defendant stood trial on that count and another count, but was convicted of both. The judge had stipulated a starting point for sentence of 27 years’ imprisonment, but as Stuart-Moore VP noted in the judgment of the Court, the judge went on to say:[59] “ In (a) normal case, a full credit for plea of guilty would be a one‑third discount. In the present case, in view of the evidence there was against you, it would be difficult to see what other alternative you would have had apart from pleading guilty.” 95. Of that reasoning, Stuart-Moore VP said:[60] “ The judge had already indicated that he accepted the mitigation that it had always been D2’s intention to plead guilty. This Court has stressed on frequent occasions that in the absence of good reason where a timely plea has been entered, a defendant is entitled to his full one‑third discount. Being caught “red‑handed” is not, in the view of this Court, a sufficient reason to disallow the full discount. Accordingly, the sentences imposed on D2 were wrong in principle.” [Italics added.] 96. In HKSAR v Yeung Kin Ma, this Court allowed an appeal against a sentence of 20 years’ imprisonment imposed on the applicant on his conviction on his pleas of guilty to two counts of unlawfully trafficking in heroin hydrochloride, and substituted a sentence of 18 years’ imprisonment. The judge had taken a starting point for sentence of 27 years’ imprisonment. 97. In the judgment of this Court, Leong JA noted that the applicant had accepted that not only had the dangerous drugs in question been found in a car driven by the applicant and at his home but also his fingerprints had been found on the material wrapping the dangerous drugs. Further, on both occasions the applicant admitted that the dangerous drugs belonged to him. Of the judge’s reasons for sentence, Leong JA said that the judge explained that he, “…considered that substantial credit was due to the applicant for his guilty plea and full cooperation but, as the evidence was “overwhelming”, he would give a discount of 7 years only.” [61] 98. Having noted that, having regard to the judgment of this Court in HKSAR v Wong Ka Kuen, counsel for the respondent, “… concedes that the Deputy Judge was wrong not to have given the full one-third discount to the applicant” [62], Leong JA cited with approval the passage of the judgment of Stuart-Moore VP set out earlier in the judgment, including “Being caught “red-handed” is not in the view of this Court sufficient reason to disallow the full discount.” In the result, he concluded:[63] “ We are of the opinion that the settled practice is that in Wong Ka Kuen and Another… and the Deputy Judge was wrong to have given a lesser discount solely by reason of the overwhelming evidence against the applicant.” 99. In HKSAR v Chui Chi Chi & Another (No 2) [64], this Court allowed the appeals of two appellants against sentences of 9 years’ and 12 years’ imprisonment respectively, imposed on their pleas of guilty to one and two counts respectively of trafficking unlawfully in dangerous drugs and substituted sentences of 7½ years’ and 10 years’ imprisonment respectively. The former sentences were imposed at a re-trial, their earlier convictions after a trial having been quashed by this Court. The judge took a starting point for sentence of 12 years’ imprisonment and imposed sentences of 9 years’ imprisonment on each of the counts. In respect of the 2nd applicant, the judge ordered that 3 years’ imprisonment in the sentence imposed for the 2nd count be served consecutively to the sentence of imprisonment imposed for the 1st count. 100. This Court said that the starting points taken for sentence by the judge were too high. Of the discount afforded to the applicants by the judge, in the judgment of this Court, Wong JA said:[65] “ The Judge declined to give a full one-third discount to the applicants, despite the pleas of guilty, because he took the view that the pleas were tendered at the retrial and not at the first trial. He gave each of them a discount of one-quarter.” 101. Having cited the passage from the judgment of Mayo JA in R v Ng Wing Kwong quoted earlier, Wong JA went on to say:[66] “ Since the decision of Ng Wing-kwong in 1995, this Court has on a number of occasions in recent years pronounced that a full one-third discount is the norm rather than the exception for a timely plea of guilty and it is only in exceptional circumstances that a lesser discount than one-third should be given. It is advisable for sentencing judges and magistrates to take heed and bear this in mind when passing sentence in cases where pleas of guilty have been entered. The applicants pleaded guilty at the retrial and these cannot be said to be truly timely pleas. They were each given a discount of one-quarter which is not an insubstantial discount. In these circumstances, we are not prepared to interfere with the judge’s exercise of discretion.” [Italics added.] 102. In the result, having stipulated starting points for sentence for each count of 10 years’ imprisonment, Wong JA said, “A discount of one-quarter for each applicant will be taken into account to calculate the actual sentences.” 103. In HKSAR v Lee Man Ki [67], this Court allowed the applicant’s appeal against a total sentence of 4½ years’ imprisonment imposed for two charges of robbery and substituted a total sentence of 4 years’ imprisonment. The District Court judge had taken starting points for sentence for the two charges of 2 years’ and 4 years’ imprisonment respectively. Although the applicant had pleaded guilty “at the very outset of the trial”, nevertheless the judge afforded the applicant a discount of only 25%, imposing sentences of 1½ years’ and 3 years’ imprisonment respectively. He ordered the sentence of 3 years’ imprisonment to be served consecutively to the sentence of 1½ years’ imprisonment. In the judgment of the Court, Wong JA said:[68] “ … It is also correct that a discount of one third would normally be given for early pleas of guilty. There are, however, circumstances that may warrant a discount of less than one third, such as a late plea or that the accused had absconded from bail during trial. But this case is not one of them. The judge has not given any reason as to why he did this. The applicant had pleaded guilty at the very outset of the trial and there were no circumstances that he should not be accorded a full discount. In our view, the judge was wrong.” 104. In HKSAR v Lo Chi Yip [69], the 1st applicant pleaded guilty on the first day of his trial to a charge of wounding with intent, but not guilty to a charge of blackmail. After a trial, he was found guilty of the latter charge. He was sentenced to 6 years’ imprisonment for the charge of wounding with intent and 2 years’ imprisonment for the charge of blackmail. One year of the sentence of imprisonment imposed in respect of the blackmail charge was ordered to be served consecutively to the sentence of 6 years’ imprisonment imposed for the wounding charge. So, the 1st applicant was sentenced to a total of 7 years’ imprisonment. This Court allowed his appeal against the sentence of 6 years’ imprisonment imposed for the wounding charge and substituted a sentence of 4 years and 8 months’ imprisonment.[70] 105. The judge had stipulated a starting point for sentence of 7 years’ imprisonment and afforded the 1st applicant a discount of one year’s imprisonment only “because it was only tendered at the last minute” [71] and because he had been “caught ‘red-handed’.” [72] 106. Of the circumstances in which the 1st applicant had pleaded guilty, in the judgment of the Court, Leong JA said:[73] “ The first applicant had pleaded to the charge of wounding on the first available opportunity, that is, on the first day of the trial. Mr Tolliday-Wright submits that in the District Court, normally on a plea day, a defendant would not have much opportunity for the purpose of considering a plea in the sense that he has time to fully understand the implications of a plea and proceed to plead guilty.” 107. Leong JA went on to say:[74] “ This Court has indicated in HKSAR v Wong Ka Kuen & Another (unrep. Crim. App. No. 35 of 1998) that a full one-third discount should be given for a timely plea. The practice of giving one-third discount for a timely plea has now been settled in the recent case of HKSAR v Yeung Kin Man [2000] 2 HKLRD 821. We do not think that the first applicant’s plea was untimely. The fact that the Judge only gave him a one-year discount instead of the full one-third is not in accordance with this settled practice.” [Italics added.] 108. Of the judge’s statement that he afforded the applicant a reduced discount because he had been caught “red-handed”, in his judgment Stuart-Moore VP said:[75] “ Although this was once considered to be a good ground for refusing the discount that would otherwise have been given, this has not been so for quite some time …” 109. Stuart-Moore VP went on to say:[76] “ Apparently, the principle that being caught red-handed is not a sufficient reason to disallow the full discount where an offender pleads guilty needs to be stated again. Counsel must be able to advise their clients with full confidence as to this aspect of sentencing, and it is neither fair nor sensible that this principle should be dependent upon the whim of the judge before whom the offender appears as to whether the full discount or less than the full discount should be given merely because an offender has been caught red-handed or, putting it another way, has been caught in the act. Counsel cannot properly advise his client that he will receive a discount of one-third if, before one judge, it may be decided that being caught red-handed deserves a discount short of one-third and yet, before another judge, it may be decided that the full discount is appropriate. This is no more than a statement of common sense, but it is an important rule of practice in order to achieve fairness and parity towards all defendants.” 110. Stuart-Moore VP concluded:[77] “ The recent authorities on this aspect of sentencing are copious and the sentencing practice which used to allow for reduced discounts in cases where the offender was caught red-handed have long since been disapproved. It suffices to make reference to the case of HKSAR v Wong Ka-kuen & Another CACC 35/1998.” 111. Of the discount to be afforded to the applicant for pleading guilty at the first day of this trial, Stuart-Moore VP said:[78] “ The other ground for justifying a reduced discount was said to be the late plea entered on “the very first day of the trial”. In some circumstances, that might have been a justification for slightly reducing the discount although we do not seek to encourage that view. This will entirely depend on all the circumstances of individual cases, particularly when considering whether a plea at court on the first day of trial has been a technical manoeuvre resulting in frightened or reluctant witnesses being forced to come to court, or resulting in a prolongation of an investigation that could have been curtailed far earlier if a plea had been indicated at the first opportunity. These are examples and by no means do they provide an exhaustive list.” [Italics added.] 112. In HKSAR v Lee Kwok Chuen [79], the applicant was sentenced to a total of 4 years’ imprisonment for three charges of burglary of domestic premises. For the first two burglaries, which were committed on the same day in neighbouring flats, the judge took a starting point for sentence of 3½ years’ imprisonment. For the third burglary, committed nine months later, he took a starting point for sentence of 3 years’ imprisonment. He said that he afforded the applicant a discount of one-third from those starting points. However, in addition, he afforded the appellant a further discount of 4 months’ imprisonment to reflect his previously clear record. The sentence of 22 months’ imprisonment imposed for the third burglary was ordered to be served consecutively to the sentences of imprisonment imposed in respect of the other two burglaries. 113. Of the fact that the judge had afforded the applicant a further discount of 4 months’ imprisonment, in excess of the one-third discount afforded for the applicant’s pleas of guilty, in the judgment of the Court, Stuart-Moore VP said that the judge had been wrong to do so not only because he had described the applicant as having “made a career out of burglaries for a year” but also because, in any event:[80] “ … the judge had already given a discount of one-third which is usually to be regarded as the high watermark of the discount given to a defendant pleading guilty in good time.” [Italics added.] 114. In HKSAR v Li Tak Yin [81] this Court allowed the appeal against the total sentence of 5 years and 3 months’ imprisonment imposed on an applicant on his pleas of guilty on the first day of trial to offences of burglary, handling stolen goods and unlawful possession of dangerous drugs. This Court substituted a total sentence of 4 years and 8 months’ imprisonment. His plea of guilty to the offence of handling stolen goods was to a lesser offence than that charged, namely robbery. 115. The applicant had been found in possession of dangerous drugs and then made admissions which, together with property recovered from his home, formed the basis of the other two charges. The judge took 4 years’ imprisonment as the starting point for the first two offences and 8 months’ imprisonment for the third offence. Although the judge acknowledged that, “It has now become the practice to allow a one-third reduction where an accused enters such a plea at the earliest opportunity”, nevertheless he afforded the applicant a discount of 25% only from that taken as the starting points for sentence. 116. In the judgment of the Court, Stock JA noted that in not affording the applicant a one-third discount, the judge said:[82] “ …you did not enter your pleas until what may well be regarded as the last possible practical time. At a recent pre-trial review you indicated that you would plead guilty to the drugs charge but at the same time additional trial time was sought due to your indication that you would challenge the admissibility of certain statements made by you.” 117. Stock JA went on to say:[83] “ It is established that “in the absence of good reason where a timely plea has been entered, a defendant is entitled to a full one-third discount.” See HKSAR v Wong Ka Kuen (unrep. Crim App No 35 of 1998, [1999] HKEC 112); and see also HKSAR v Lo Chi Yip [2002] 3 HKLRD 270 at p. 275, and HKSAR v Yeung Kin Man [2002] 2 HKLRD 821at p. 823. The fact that a guilty plea has been tendered on the first day of trial where it could have been tendered earlier is generally not, of itself, without more, sound reason for reducing the discount normally given; although there will be cases and circumstances where a delay or last minute plea might well warrant such a reduction.” [Italics added.] 118. Having cited the passage from the judgment of Stuart-Moore, quoted earlier, in HKSAR v Lo Chi Yip, Stock JA said:[84] “ This is the current approach. In this particular case, the judge emphasised the fact that the plea was not entered at “the earliest opportunity”. We accept that there have been cases in the past in which that phrase has been used, but of late the courts have rather been using the phrase “a timely plea” and the judge’s approach is one which might suggest that it is exceptional rather than normal to deduct one-third if the plea of guilty is offered on the first day of trial. We emphasise that the one-third discount is not a rigid rule, for there are instances where a lesser discount will be, and as the cases show have been, warranted, and there are many examples provided by the cases. But there should, in our judgment, be some sound reason or circumstance, over and above the mere fact that the plea is only tendered on the first day of trial. We do not see that such sound reason existed in this case.” [Italics added.] 119. Stock JA went on to note that, although the history of the circumstances in which the applicant pleaded guilty to the alternative charge of handling stolen goods was, “not clear”, nevertheless it appeared that, “… following discussions with the prosecution before the date of arraignment … the applicant offered a plea to handling, which was accepted.” In the result, this Court quashed the sentences of imprisonment imposed by the judge and imposed sentences of imprisonment that afforded the applicant a discount of one-third from that taken as the starting point for sentence. 120. In HKSAR v Ting Chiu & Another [85],this Court quashed the sentences of 45 months’ imprisonment that the judge had imposed on each of the applicants on their pleas of guilty to an offence of robbery on the first day of the trial and, in substitution, imposed sentences of 32 months’ imprisonment. This Court determined that the judge erred in stipulating a starting point for sentence of 5 years’ imprisonment, rather than 4 years’ imprisonment, and in affording the applicants a discount of only 25%, “because it was only today that the plea of guilty was indicated to the first charge [robbery]. This was a plea not tendered at the first available opportunity.” [86] 121. In the judgment of this Court, Woo JA, as Woo VP was then, said:[87] “ A reduction of one-third from the starting point to give credit to a guilty plea has been a well-settled practice in recent years.” 122. Having cited a passage from the judgment of Stuart-Moore in HKSAR v Lo Chi Yip & Another, confirming that it was the practice of the Court “to accord a defendant the customary one-third discount for a plea of guilty where no good reason was shown for a departure from the practice”, Woo JA said:[88] “ Stuart-Moore VP intimated that the rationale behind the adherence to this general practice was that “counsel must be able to advise their clients with full confidence as to this aspect of sentencing” (at 277A), and “it is an important rule of practice in order to achieve fairness and parity towards all defendants” (at 277D). It is also pertinent to note that a failure to follow the practice would, save in most exceptional circumstances, inevitably lead to interference by the Court of Appeal, thus increasing costs and wasting resources for all concerned.” 123. Having cited other passages from that judgment, and passages from the judgment of Stock JA in HKSAR v Li Tak Yin, describing the limited circumstances in which a judge might decline to afford the full one-third discount to a defendant on his plea of guilty to a charge, Woo JA said: [89] “ … the judge merely said that the pleas by the applicants, which were made on the first day of the trial, were not tendered at the first available opportunity, as a basis for his not adopting the normal one-third discount. We cannot discern any reason or circumstance which would justify the normal reduction not being applied.” 124. In conclusion, Woo JA noted that, in any event, the pleas of the applicants were not “untimely”, given that in a Newtonhearing they had successfully contested the issue of whether or not a knife was produced in the course of therobbery:[90] “ However, the applicants’ pleas did not in any event qualify as “untimely”, because they made no difference to the requirement for the victims to attend court to give evidence as they had to give evidence on the question whether a knife was produced in the course of the robbery, which resulted in a Newton Inquiry. The applicants were well justified in raising that question since the Inquiry was resolved in their favour. If it had been otherwise, there might have been material for the judge to consider to reduce the normal discount for plea.” 125. In Yu Fai Tat v HKSAR [91] the magistrate had stipulated a starting point for sentence of 15 months’ imprisonment for an offence of possession of obscene articles for publication, contrary to section 21(1)(a) of the Control of Obscene and Indecent Articles Ordinance, Cap. 390. Although the Court of Final Appeal determined that the magistrate was in error in affording the defendant a discount of 2 months’ imprisonment, in addition to the one-third discount, “because he admitted the obscenity of the discs, without requiring their examination and determination by the Obscene Articles Tribunal”, nevertheless it determined that, at the hearing of the appeal, the single judge had erred in increasing the sentence of imprisonment by two months without having afforded the applicant an opportunity to address the Court on that issue. 126. In the judgment of Sir Derek Cons NPJ, with whom the other judges agreed, the policy of the courts of discounting sentences by reason of a plea of guilty was addressed:[92] “ … there are several factors that have influenced courts to the policy of discounting sentences by reason of a plea of guilty. But the most influential are in general the saving of judicial time and dispensing with the need for the attendance and examination of witnesses. These factors may of course vary greatly between individual cases, both in quantity and quality. A complex fraud may take far longer to try than a simple possession of drugs; the victim of a theft is unlikely to suffer the trauma that a rape victim may undergo in having to give evidence in open court. Nevertheless a standard one third discount is customary nowadays. It is by no means an absolute figure, and courts will vary it for good reason. Examples of reason to increase that we have been referred to are full and early restitution of large sums of stolen money: HKSAR v. Leung Shuk Man [2002] 3 HKC 424 and the giving of evidence by an accomplice: HKSAR v. Chan Sau Hing CACC 211/2001. There are also the well known “supergrass” cases.” [Italics added.] Factors subsumed in the one-third discount 127. In the Secretary for Justice v Lee Chun Ho, Jeff [93] this Court allowed an application by the Secretary for Justice for the review of sentences imposed on the respondent on his conviction on his pleas of guilty to offences of robbery and using an imitation firearm with intent to resist or prevent lawful arrest and substituted sentences of 6 years’ and 4 years’ imprisonment respectively, ordering that six months of the latter sentence of imprisonment was to be served consecutively to the former. The respondent was caught in a pursuit initiated by the victim, assisted by a passerby and then police officers. He admitted the offences, pleaded guilty in the Magistracy and was committed for sentence to the Court of First Instance. 128. In sentencing the respondent, the Recorder had taken 6 years’ imprisonment as the starting point for sentence for each offence, which he had discounted by one-third, “…on account of the guilty pleas and then by a further 6 months to 42 months for the appellant’s pleading guilty at the magistrates’ court and for his cooperation with the police, both at the scene of crime and in the police station.” [94] 129. Of the discount from the starting point taken for sentence, over and above the one-third discount, afforded to the respondent, in the judgment of this Court, Yeung JA said:[95] “ The Court of Appeal have repeatedly emphasized that the one-third discount “is usually to be regarded as the high watermark of the discount given to a defendant pleading guilty in good time.” The respondent’s co-operation with the police referred to by the judge was his admission of the offence to the police, both at the scene of the crime and at the police station. Such a “co-operation” is just part and parcel of the respondent’s admission of his guilt, albeit at the first available opportunity. This mitigating factor should be subsumed within the one-third discount. We agreed with Mr Lee’s submission that the judge should not have given a further six-month discount on top of the one-third discount for the guilty pleas.” 130. In HKSAR v Chow Yuen Fai [96] this Court dismissed an application for leave to appeal against sentences imposed for an array of sexual offences involving children of which the applicant had been convicted on his own pleas of guilty in the District Court. Of the issue of whether or not in such circumstances a defendant ought to be afforded a discount in addition to the one-third discount, to reflect the fact that the victims did not have to undergo the ordeal of giving evidence, in the judgment of this Court Ma CJHC, as Ma CJ was then, said:[97] “ Here, our attention has been drawn by Mr McGowan to a number of cases in which the courts appear to have given more than a one‑third discount for the plea of guilty. In HKSAR v See Tak Man [1998] 1 HKLRD 794, in relation to a charge of indecent assault (the victim was 12 years old; the offence involved oral sex and various other acts falling short of attempted buggery and also ejaculation into the boy’s mouth), the Court of Appeal reduced the starting point by “little more than one‑third” in view of the accused being a first time offender and also the victim being spared the ordeal of having to give evidence. We were also referred to Wong Ying Ho (see para 28 above [98]) where the Court of Appeal left undisturbed a discount of 37.5%. In our judgment, however, these and other cases which suggest that an additional discount may be given where a victim is spared the ordeal of giving evidence (and therefore having to recount the traumatic incident or incidents in question) must now be viewed against the general principle that the one‑third discount is usually to be regarded as the high watermark of the discount for pleading guilty in good time : see HKSAR v Wen Zelang [2006] 4 HKLRD 460; HKSAR v Ng Ngok Wai (吳岳威) [2008] 1 HKLRD 546 (this case involved an indecent assault on teenage children). Only in exceptional cases should a discount of more than one‑third be given for a timely plea.” [Italics added.] 131. In the judgment of this Court in HKSAR v Tsang Cho Kiu [99], D Pang J, as Pang JA was then, cited the passage quoted from the judgment of Ma CJHC with approval and as representing “the current position in law”. 132. In HKSAR v Ma Ming [100] this Court rejected the applicant’s submission that he was entitled to a discount, in addition to the one-third discount for his plea of guilty, to reflect his co-operation with the police and his admissions which had led to a charge of conspiracy to procure the transfer to another of travel documents, in addition to the substantive charge of the same nature. In the judgment of this Court, Yeung VP said:[101] “ Since the mid 1990s the courts in Hong Kong have adopted the sentencing policy that if a defendant enters a timely plea of guilty, as a rule he is entitled to one third discount on the sentence he would otherwise receive. In HKSAR v Chui Chi Wai & Another (No. 2) [2000] 1 HKLRD 704, the Court of Appeal had this to say on page 707 E-F: “Since the decision of R v Ng Wing Kwong (unrep CACC 62/1995) in 1995, this Court has on a number of occasions in recent years pronounced that a full one-third discount is the norm rather than the exception for a timely plea of guilty and it is only in exceptional circumstances that a lesser discount than one-third should be given.” ” 133. Of the rationale for the policy of affording a discount to a defendant for a plea of guilty, Yeung VP said:[102] “ If unfruitful contest, delay and unnecessary expense of resources can be avoided, so that the court’s time can be allocated to cases which warrant more of its attention, thus ensuring that judicial proceedings will proceed more expeditiously and more efficiently, then giving the defendant who timely pleads guilty a one third discount is a policy decision which serves the public interest. A discount of one third is quite a substantial discount. One of the main purposes of the court giving this one third discount to a defendant who pleads guilty is to encourage a guilty person to own up to the crimes he committed, so as to conserve the resources of the community and to ensure that justice can be administered more efficiently and matters can be concluded in the most expeditions manner.” 134. Of the difficulties for a court to determine “what a timely plea of guilty is”, Yeung VP said: “ Since there is a period of time between the moment when a suspect is put under arrest and the time when he appears in court to face the charge or charges, what happened during that period of time and the attitude taken by the suspect might influence the sentencing decision. If the court has to take into account all these factors before passing sentence, not a few disputes may arise and the sentence that will be passed ultimately would become uncertain. To give one example: ‘ Is a defendant who admits his guilt forthwith at the outset and cooperates with the prosecution entitled to a larger discount on his sentence when compared with a defendant who makes up his mind to plead guilty only when the trial begins?’ Many people will think that the answer is yes and, on the face of it, it seems reasonable. However, in order to put this kind of thinking into practice, the court may very likely have to go into fine distinctions and subtle differences between different cases and adjust the sentences according to such niceties and subtleties. Such an approach would lengthen and complicate the court proceedings, increase costs, adversely affect court efficiency and delay the handling of cases which genuinely require the court’s attention. It not only goes contrary to the policy and basic objective of giving a one third discount, but is also against public interest.” [Question indented.] 135. In the result, Yeung VP concluded that: “ …when giving the one third discount the approach taken by the court is a firm and broad-brush approach. It gives a defendant who timely pleads guilty the one third discount, which is substantial, without regard for niceties, in order to discourage excessive arguments and to prevent wasteful use of the resources of the community.” Other jurisdictions: approach to a discount in sentence for a plea of guilty 136. In their most helpful submissions, Mr Yeung and Mr Pang have provided the Court with an overview of the approach taken in other jurisdictions in affording a defendant a discount in sentence for a plea of guilty. In doing so, they have referred the Court to the sentencing regimes obtaining in England and Wales, Scotland, Australia and New Zealand. Two things stand out. First, all of them, save for the exception of Tasmania, have their own statutory sentencing regimes. Secondly, none of them afford a discount of one-third to a defendant who first intimates a plea of guilty as late as the first day of the trial. England and Wales 137. Section 144(1) of the Criminal Justice Act 2003 provides that: “ (1) In determining what sentence to pass on an offender who has pleaded guilty to an offence before the court or another court, a court must take into account- (a) the stage in the proceeding for the offence at which the offender indicated his intention to plead guilty, and (b) the circumstances in which this indication was given.” 138. Section 172 of the Act provides that every court must have regard to any sentencing guidelines issued by the Sentencing Guidelines Council, now Sentencing Council, when sentencing. Its definitive guideline ‘Reduction in Sentence for a Guilty Plea’ (Revised 2007) identifies the purpose of the practice of affording a defendant a discount in sentence for a plea of guilty:[103] “ A reduction in sentence is appropriate because a guilty plea avoids the need for a trial (thus enabling other cases to be disposed of more expeditiously), shortens the gap between charge and sentence, saves considerable cost, and, in the case of an early plea, saves victims and witnesses from the concern about having to give evidence. The reduction principle derives from the need for the effective administration of justice and not as an aspect of mitigation.” 139. In R v Caley[104],in the context of those Sentencing Guidelines, the Court of Appeal of England and Wales addressed a number of different questions associated with the practice in sentencing which recognises that, “a distinction should ordinarily be drawn between a defendant who admits his guilt and one who does not.” [105] 140. Of the purpose of the sentencing guidelines, articulated in paragraph 2.2, Hughes LJ, as Lord Hughes was then, said in the judgment of the Court:[106] “ In order of importance, plainly the first is the benefit for victims and witnesses. The impact of crime on its victims can be enormous or slight, but whether it is large or small the knowledge that a defendant has accepted his guilt and that punishment will follow normally reduces that impact substantially and thus brings significant benefit to the victim. It is generally worse for the victim when the offender, although guilty, is defiant. The same applies to the impact on those who may have to give evidence; they include, but are not confined to, the victim. A few may relish it, or think that they will, but for most the process is normally stressful and often unavoidably uncomfortable. Moreover the anticipation may often be painful, sometimes even more than the actuality. For both victims and witnesses the benefit from a plea of guilty remains even when it comes late, but generally speaking the later it is the less the benefit. The second major reason for the practice is a more pragmatic one but it is nevertheless vital in the public interest. The expenditure in public time and money on trials and on preparation for trials is considerable. The case must be thoroughly prepared so that the exacting standard of proof rightly required in a criminal case can be met. Further investigation is likely to be necessary, as may the assembly of a good deal more evidence, lay and expert. Such steps are necessary, but expensive. They are avoided or much reduced by an admission of guilt. The public’s limited resources can then be concentrated on those cases where a trial will really be necessary, and such cases will not be delayed, often with accused persons in custody. At present something of the order of 75% of all Crown Court cases result in pleas of guilty; if in all those cases the defendants were out of defiance or otherwise to insist on each detail of the case being proved to the hilt the administration of criminal justice would be in danger of collapse. ” [Italics added’] 141. Having acknowledged that a plea of guilty may be an indication of remorse, but equally might be borne out of a realisation that there was no way out of the consequences of the prosecution, Hughes LJ noted that the sentencing scheme afforded a discount to both categories of defendant and did so, “by reducing the sentence which would have been imposed for trial by a proportion, on a sliding scale depending on when the plea of guilty was indicated.” The Sentencing Guideline provides for a reduction of about one-third from the sentence that would have been passed after a trial for those who indicate a plea of guilty at the “first reasonable opportunity; when this occurs will vary from case to case”. Of that provision, Hughes LJ said:[107] “ The SGC Guideline rightly makes it clear that the question of when the defendant’s first reasonable opportunity arose is a matter for the sentencing judge. Individual cases may call for individual decisions about this. But it is obviously desirable for there to be a baseline of broadly consistent approach if justice is to be done between different offenders, in all parts of the country. Equally such consistency is necessary if proper advice is to be given to accused persons.” 142. Of the difficulties arising from an interpretation of that phrase, Hughes LJ said:[108] “ There is sometimes confusion in argument between (i) the first reasonable opportunity for the defendant to indicate his guilt and (ii) the opportunity for his lawyers to assess the strength of the case against him and to advise him on it. It is obvious that the second depends on the evidence being assembled and served. The first, however, frequently does not. There will certainly be cases where a defendant genuinely does not know whether he is guilty or not and needs advice and/or sight of the evidence in order to decide. We do not attempt to define them, and they do not arise in the present appeals. They might however include cases where even if the facts are known there is a need for legal advice as to whether an offence is constituted by them, or cases where a defendant genuinely has no recollection of events. There may be other cases in which a defendant cannot reasonably be expected to make any admission until he and his advisers have seen at least some of the evidence. Such cases aside, however, whilst it is perfectly proper for a defendant to require advice from his lawyers on the strength of the evidence (just as he is perfectly entitled to insist on putting the Crown to proof at trial), he does not require it in order to know whether he is guilty or not; he requires it in order to assess the prospects of conviction or acquittal, which is different. Moreover, even though a defendant may need advice on which charge he ought to plead guilty to, there is often no reason why uncertainty about this should inhibit him from admitting, if it is true, what acts he did. If he does so, normally the public benefits to which we have referred will flow.” [Italics added.] 143. In a Consultation conducted between 11 February and 5 May 2016, the Sentencing Council in the United Kingdom sought views on their Draft Guideline ‘Reduction in sentence for a guilty plea’. Of the draft guidelines, the Sentencing Council said:[109] “ The Council has designed the revised guideline for guilty plea reductions to clarify the levels of reduction appropriate for the different stages at which the plea is entered. The revised guideline seeks to encourage those defendants who are aware of their guilt to enter a plea as early in the court process as possible. When this occurs, victims and witnesses are spared having to appear at court to testify and the police and Crown Prosecution Service can apply their resources to the investigation and prosecution of other cases. Offenders who accept their responsibility in this way benefit from receiving a modest reduction in their sentence. ” [Italics added.] 144. Of the purpose of affording a reduction in sentence for pleas of guilty, the Sentencing Council said: [110] “ The purpose of reducing sentences when offenders plead guilty is to encourage them to admit their guilt as early as possible. By bringing forward the point at which some offenders plead guilty the proposed guideline will generate, to a greater or lesser degree, the following benefits: • Overall, victims and witnesses in many cases will be informed earlier than in the past that their testimony is not required as the defendant has pleaded guilty. The earlier the plea is entered, the sooner victims and witnesses can be reassured that the offender has accepted responsibility for the offence and that they will not have to worry about having to go to court. In addition, victims will also benefit from seeing a more consistent approach to determining sentence reductions; and • There will be resource savings for the police, the Crown Prosecution Service (CPS), the Legal Aid Agency and Her Majesty’s Courts and Tribunals Service. These savings in turn benefit victims and witnesses in that they allow more time and resources to be concentrated on investigating and prosecuting other cases. ” 145. The Sentencing Council noted that the Draft Guideline did not use the phrase “first reasonable opportunity”, rather it referred to “first stage of the proceedings”. Of the stages identified in the Draft Guideline, it went on to state:[111] “ Where a plea is indicated at the first stage of the proceedings a reduction of one-third (and not more than one-third) should be made (subject to the exceptions in section F). The first stage will be the first point at which the charge is put to the offender in court and a plea (or indication of plea) is sought. ” 146. Of the subsequent stages, it said:[112] “ After the first stage of the proceedings the maximum level of reduction is one-fifth (subject to the exceptions in section F).” Detailed provisions were stipulated according to the jurisdiction in which the trial was held and, where appropriate, where the proceedings began. 147. Of the sliding scale of reduction that was applicable thereafter, it said:[113] “ The reduction should be decreased from one-fifth to a maximum of one-tenth on the first day of trial proportionate to the time when the guilty plea is first indicated relative to the progress of the case and the trial date (subject to the exceptions in section F). The reduction may be decreased further, even to zero, if the guilty plea is entered during the course of the trial.” 148. Of factors of personal mitigation, the Sentencing Council stated:[114] “ The purpose of reducing the sentence for a guilty plea is to yield the benefits described above and the guilty plea should be considered by the court to be independent of the offender’s personal mitigation. Thus factors such as admissions at interview, co-operation with the investigation and demonstrations of remorse should not be taken into account in determining the level of reduction. Rather, they should be considered separately and prior to any guilty plea reduction, as potential mitigating factors.” 149. Furthermore, and contrary to the existing guidance[115], the Sentencing Council said:[116] “ The benefits apply regardless of the strength of the evidence against an offender. The strength of the evidence should not be taken into account when determining the level of reduction.” [Italics added.] Scotland 150. Section 196 of the Criminal Procedure (Scotland) Act 1995, as amended, first recognised in legislationthe principle of sentence discounting, providing that: “ In determining what sentence to pass on, or what other disposal or order to make in relation to, an offender who has pled guilty to an offence, a court may take into account- (a) the stage in the proceedings for the offence at which the offender indicated his intention to plead guilty, and (b) the circumstances in which that indication was given.” 151. In Gemmell v HM Advocate [117] the High Court of Justiciary considered that legislation and related principles of sentencing in seven appeals against sentence. In his judgment, Lord Justice Clerk Gill described the discounting process as involving three stages, namely: [118] “ (1) to decide what the sentence would be if no question of a discount arose; (2) to decide whether there should be a discount, and (3) if so, to decide what the amount of it should be.” 152. Lord Justice Clerk Gill described the legislative provision as, “statutory encouragement of early pleas.” Of the basis for the provision, he said:[119] “ It is based on the objective value of an early plea in the administrative and other costs, and the personal inconvenience, that it saves.” 153. He described those benefits as including:[120] “ In some cases, there is a saving of inconvenience to complainers and witnesses. In a small minority of cases there is a saving in jury costs. There is also a benefit to the criminal justice system in the avoidance of undue delay between arrest and sentencing. But the primary benefit that is realised in every case is the saving of administrative costs and the reduction of the court’s workload.” The court’s discretion 154. Of the approach to be taken in sentencing a defendant on his plea of guilty, Lord Justice Clerk Gill said:[121] “ … an accused is not entitled to any particular discount in return for a plea of guilty. The level of discount, if any, is and must always be a matter for the discretion of the sentencer … Moreover, even in a discretionary matter such as this, it is desirable that the court should exercise its discretion in accordance with some broad general principles.” 155. Of the level of discount to be afforded to such a defendant, Lord Justice Clerk Gill said:[122] “ In my opinion the sliding scale approach exemplified in the Definitive Guideline carries a risk of rigidity. That approach was favoured by this court in Spence v HM Advocate, but with the clear reservation that the amount of the discount, if any, would depend in every case on the circumstances (paragraph [15]).” He went on to say: “ In my view, the broad principle that, in general, the discount will be the greater the earlier the plea is probably a sufficient statement of guidance for most purposes.” 156. Of the approach to be taken to what he identified as the three stages of sentencing in those circumstances, Lord Justice Clerk Gill said:[123] “ the assessment of the headline sentence and the assessment of any discount are separate processes governed by separate criteria. When the headline sentence is assessed at the first stage of the sentencing process, the sentencer makes a judgment from a consideration of numerous sentencing objectives, such as retribution, denunciation, public protection and deterrence. But when he considers the matter of a discount, the only relevant consideration, in my view, is how far the so-called utilitarian benefits of the early plea have been achieved. That is an objective consideration unrelated, in my opinion, to the moral values on which the headline sentence is fixed.” 157. Of the approach to be taken to the discount to be afforded to the defendant, Lord Justice Clerk Gill said:[124] “ the two questions to be answered are how early the plea was tendered and the extent to which the tendering of it furthered the objective justifications set out in Du Plooy.[125] ” 158. Earlier in his judgment, Lord Justice Clerk Gill referred to the judgment of the High Court of Justiciary in Du Plooy as giving guidance, “as the basis of, and the scope for, an allowance in the sentencing of an accused in respect of the fact that he has pled guilty”. Of that, he said that an allowance was to be given:[126] “ … for the fact that the tendering of a plea of guilty was likely to save public money and court time, and in general to avoid inconvenience to witnesses; or, in certain types of cases, to avoid the additional distress that could be caused by their having to be precognosced or to give evidence.” 159. Lord Justice Clerk Gill noted that in reaching that determination, the Court adopted:[127] “ … the language of Spigelman CJ in a decision of the Court of Appeal of New South Wales (R v Thomson; R v Houlton (2000) 49 NSWLR 383) and Kirby J in the High Court of Australia (Cameron v R (2002) 187 ALR 65), the court said that these specific considerations gave “utilitarian value” to the plea (paragraph [16]).” 160. Of the obvious implied relevance of an early plea of guilty, given the statutory provision, Lord Justice Clerk Gill said:[128] “ the essential consideration is how early in the proceedings the accused has indicated his intention to plead guilty. In any given case, the discount will be greater the earlier the plea is tendered. ” What is an early plea of guilty? 161. Of what constituted an early plea of guilty, Lord Justice Clerk Gill said:[129] “ We have become familiar in this court with the argument that the accused is justified in withholding an early plea yet invoking s. 196 where there has been a delay in obtaining Crown disclosure, police statements, forensic reports and the like; or where investigations have been carried out by the defence. This is a specious argument. I repeat what I said in HM Advocate v Thomson (2006 S.C.C.CR. ,p.271, para 27): ‘ If an accused person has committed the crime charged, he can plead guilty to it at the outset and benefit from his plea by way of discount when the sentence is assessed; or he can defer pleading until he is sure that the Crown have a corroborated case, in the knowledge that a sentence discount may be reduced or refused altogether. That is the choice that he must make. He cannot have it both ways’ (paragraph [27]);” The strength of the prosecution case 162. Of the decisions of the Court that the strength of the prosecution case restricts the discount afforded to the defendant, Lord Justice Clerk Gill said:[130] “ I have come to the view that this approach is unsound. For the court to refuse or to minimise a discount on this basis is for the court to decide what the outcome would have been if the accused had gone to trial. In my view, there are dangers in that approach. It is the common experience of practitioners that criminal trials regularly produce the unexpected. Moreover, it is undesirable in my view that in determining the sentence the court should become involved in an appraisal of the strength of the Crown case based mainly on the Crown narrative. Experience shows that Crown witnesses do not always live up to their precognitions and that on occasions even the strongest cases come to grief. I also agree with the point made by Lord Eassie that in many cases the strength of the Crown case results from the accused’s frankness with the police. In some cases, without the accused’s own admission, the Crown would be in difficulty in finding corroboration. It is illogical, in my view, to withhold a discount from the accused in such circumstances. I conclude therefore that the strength of the Crown case ought not to be treated as a factor influencing the amount of the discount.” [Italics added.] 163. In agreeing with Lord Justice Clerk Gill, that the strength of the prosecution case was not a material factor to be taken into account in determining sentence, Lord Eassie said that not only did he have “reservations in principle” but also there were practical problems:[131] “ More generally, it is in my view wholly undesirable that on a plea of guilty the court should begin to entertain competing submissions at the respective strengths of the Crown and defence case. A narration of agreed facts tendered at a plea is no basis for assessment of actual strengths and weaknesses of the parties’ positions and to inquire beyond that, as a matter relevant to the giving or extent of any discount on account of the timing of the plea of guilty, would, in my view, put in jeopardy the practical working of our system of criminal procedure in disposing of cases in which an accused pleads guilty. Ultimately, the strength of the prosecution case can only be tested by trial.” Assistance to the authorities 164. Of the factor of assistance to the authorities given by a defendant, Lord Justice Clerk Gill described as “confused” the reasoning of the Court in Du Plooy that it was a matter to be taken into account in the discount of sentence. Rather, he said:[132] “ Any assistance that the accused may have given to the police or the Crown in the investigation and prosecution of the offence is properly a matter of mitigation and, as such, a matter to be taken into account in the assessment of the headline sentence.” Remorse 165. Lord Justice Clerk Gill said that, in his opinion, the remorse of a defendant, “cannot…be a proper justification for a sentence discount.” He went on to say:[133] “ My own view is that there is seldom any sure criterion for assessing whether the accused is truly remorseful; but where there is convincing evidence of remorse, the sentencer may make allowance for it, as an aspect of mitigation, in deciding on the starting figure (Sentencing Guidelines Council Revised Guideline, Reduction in Sentence for a Guilty Plea (2007), para 2.4; …” Australia 166. In the judgment of the majority [134] in the High Court of Australia in Cameron v The Queen [135], the basis on which it was permissible to afford a defendant a discount in sentence following his plea of guilty was addressed:[136] “ It is well established that the fact that an accused person has pleaded guilty is a matter properly to be taken into account in mitigation of his or her sentence. In Siganto v The Queen [137]it was said: ‘ a plea of guilty is ordinarily a matter to be taken into account in mitigation; first, because it is usually evidence of some remorse on the part of the offender, and second, on the pragmatic ground that the community is spared the expense of a contested trial. The extent of the mitigation may vary depending on the circumstances of the case.’ It should at once be noted that remorse is not necessarily the only subjective matter revealed by a plea of guilty. The plea may also indicate acceptance of responsibility and a willingness to facilitate the course of justice.” 167. Of the issue of discrimination against a defendant who availed himself of the right to trial, the majority said: [138] “ Although a plea of guilty may be taken into account in mitigation, a convicted person may not be penalised for having insisted on his or her right to trial [139]. The distinction between allowing a reduction for a plea of guilty and not penalising a convicted person for not pleading guilty is not without its subtleties, but it is, nonetheless, a real distinction, albeit one the rationale for which may need some refinement in expression if the distinction is to be seen as non-discriminatory. It is difficult to see that a person who has exercised his or her right to trial is not being discriminated against by reason of his or her exercising that right if, in otherwise comparable circumstances, another’s plea of guilty results in a reduction of the sentence that would otherwise have been imposed on the pragmatic and objective ground that the plea has saved the community the expense of a trial. However, the same is not true if the plea is seen, subjectively, as the willingness of the offender to facilitate the course of justice. Reconciliation of the requirement that a person not be penalised for pleading not guilty with the rule that a plea of guilty may be taken into account in mitigation requires that the rationale for that rule, so far as it depends on factors other than remorse and acceptance of responsibility, be expressed in terms of willingness to facilitate the course of justice and not on the basis that the plea has saved the community the expense of a contested hearing.” [Italics added.] 168. In that appeal, from the Court of Criminal Appeal in Western Australia, initially the dangerous drug had been described wrongly, by its chemical composition, as what is known commonly as ‘Ecstasy’, not ‘Speed’. In consequence, the appellant’s first intimation of a plea of guilty was more than six months after he had been arrested and charged and, thereafter, repeatedly remanded by the courts. The judge had reduced the appellant’s sentence by 10% to reflect his plea of guilty. At issue was whether or not the appellant’s plea was to be regarded as a “fast track” plea of guilty, for which the practice was to afford a discount in sentence of “between 20-25 per cent up to 30-35 per cent depending on the circumstances” of the plea. Of the statutory provision, namely the Sentencing Act 1995 (WA), the majority said:[140] “ The Sentencing Act, which sets out sentencing principles applicable to all persons convicted of an offence, specifies, in s 8, mitigating factors to be taken into account on sentence. One such factor is that the offender pleaded guilty. By s 8(2) it is provided that “the earlier in proceedings that [the guilty plea] is made, or indication is given that it will be made, the greater the mitigation.” Provision is also made in s 7 with respect to aggravating factors. The fact that the offender pleaded not guilty is expressly excluded from those aggravating factors by s 7(2)(a) of the Sentencing Act.” 169. Of the resolution of the reconciliation of the two provisions, the majority said:[141] “ s 8(2) must be read as allowing that a plea of guilty may be taken into account in mitigation for the reason that a guilty plea evidences a willingness to facilitate the course of justice and not simply because the plea saves the time and expense of those involved in the administration of criminal justice. That being so, the relevant question is not simply when the plea was entered but, as was accepted by the Court of Criminal Appeal in this matter, whether it was possible to enter a plea at an earlier time.” 170. In the result, the majority concluded:[142] “ ... the rationale for the rule a plea may be taken into account in mitigation, namely, that, being a sign of remorse and acceptance of responsibility, (is) the cooperative consideration of willingness to facilitate the course of justice. And once that rationale is accepted, the respondent’s suggestion that the extent to which a plea of guilty may be taken into account in mitigation may vary according to whether it was or was not a “fast-track” plea must be rejected. Rather, the issue is to what extent the plea is indicative of remorse, acceptance of responsibility and willingness to facilitate the course of justice. And a significant consideration on that issue is whether the plea was entered at the first reasonable opportunity.” [Italics added.] 171. As noted earlier in this judgment, in his judgment Kirby J said that “The main features of the public interest”, relevant to the discount for a plea of guilty, are “purely utilitarian”. Thereafter, he enumerated those features. Of the relevance of the stage in the proceedings at which the plea of guilty was the first intimated, Kirby J said:[143] “ Obviously, the timing of any plea of guilty has a large bearing on the credit that should be given to the prisoner. A plea of guilty at the last moment (as on the day set down for the trial) will ordinarily attract a smaller discount in sentence than one that is entered at the first reasonable opportunity. But even a belated plea will normally attract a discount.” The submissions of the amici curiae 172. Mr Pang submitted that most common law jurisdictions recognise that an early plea of guilty should give rise to a greater reduction in sentence than a late plea. It was suggested that the most persuasive reason advanced was the resulting utilitarian value. By contrast, the utilitarian value of a plea of guilty at trial was greatly reduced. 173. Of the issue of the identification of the time at which an early plea is to be recognised, the Court was invited to note that different jurisdictions approached the matter differently. England and Wales have adopted sliding scales, with very specific sentencing discounts applied at specific stages in the criminal justice process when a plea is indicated. Other jurisdictions leave the extent of reduction in discount for a plea, and the point in the criminal justice process at which a reduction in discount for plea should be applied, to the unfettered discretion of the sentencing judge. 174. Mr Pang submitted that a sliding scale, with specific defined discounts at defined ‘waypoints’ in the criminal justice process, has the benefit of certainty. However, such certainty carries with it a rigidity, which may create unfairness. A defendant’s plea may result in the same utilitarian benefit, but he might receive a lesser discount simply because he has missed a particular ‘waypoint’. In summary trials in the Magistrate’s courts, the time between first appearance and trial may be so short that a defendant may not have sufficient time to make a considered decision on plea and may therefore forfeit what would otherwise be an appropriate sentencing discount. 175. Mr Pang suggested that, by contrast, a broad discretion in determining the discount to be afforded to a defendant on his plea of guilty might lead to uncertainty, making it less attractive to an accused to enter a plea of guilty. Also, it might lead to an increase in appeals against sentence. 176. Rather than adopting a full sliding scale with a number of specific ‘waypoints’, Mr Pang submitted that the full sentencing discount ought to be afforded for a defendant’s “timely” plea, and lesser discounts afforded for pleas which are not “timely”. Timeliness should not be determined by reference only to a specific stage in the proceedings. Rather, a defendant should be accorded sufficient time to receive and consider appropriate legal advice before indicating his plea. If allowance is not made for a defendant to receive proper legal advice before tendering his plea, there may well be justifiable complaints of derogations to the right to trial and the right to legal advice. 177. Mr Pang contended that a “non-timely” plea occurs only when, having had the opportunity to receive and consider legal advice, a defendant does not then indicate or tender a plea. 178. It was contended that a discount of sentence based only on a consideration of the utilitarian value ignores other legitimate and relevant factors, such as remorse. Accordingly, Mr Pang submitted that a further discount, over and above that afforded for the utilitarian value of a plea, was to be afforded to reflect those factors. By contrast, Mr Pang acknowledged that the current practice in sentencing subsumes the defendant’s remorse, his acceptance of responsibility and administrative convenience of a timely plea within the current one-third discount. He suggested that in order to properly reflect such mitigating factors, either the effective “cap” of a discount of one-third of the starting point could be removed or the discount afforded for the purely utilitarian value of the plea could be reduced from one-third. However, he said that consideration ought to be given to whether adopting the latter process might prove unattractive to defendants and result in a reduction in those pleading guilty and an increase in those seeking trial. 179. Finally, Mr Pang submitted that there should not be a return to the practice of reducing the discount afforded to a defendant for his plea of guilty if it was made in the face of overwhelming evidence. Quite apart from the difficulties in assessing whether the available and unchallenged evidence was actually “overwhelming”, such an approach ignored the utilitarian value of a plea of guilty. General submissions : Abdou Maikido Abdoulkarim 180. In his submissions in respect of the general policy of affording a defendant a discount for a ‘timely’ plea of guilty, Mr Blanchflower SC invited the Court to note that the practice had developed that a plea tendered up and until the first day of trial was regarded as ‘timely’. The one-third discount was the “high-water mark” and subsumed all mitigation, other than exceptional factors such as full and early restitution, giving evidence against a co-defendant or the “supergrass” cases. 181. Mr Blanchflower submitted that separating the one-third discount into subjective and objective components is contrary to the current practice in Hong Kong. In the judgment of this Court in HKSAR v Abdallah Anwar Abbas, Stuart-Moore VP described the one-third as subsuming a “combination” of mitigating reasons.[144] 182. Furthermore, Mr Blanchflower contended that there was a long line of judgments of this Court in which it had been held that the strength of the prosecution case was irrelevant to the discount to be afforded to a defendant for his plea of guilty. 183. Mr Blanchflower suggested that this Court should not adopt the approach of affording a defendant a reduced discount for his plea of guilty taken in England and Wales and articulated in the Sentencing Guidelines Council’s Definitive Guidelines. To do so, would be to punish a defendant by reducing the discount afforded to him because he might have delayed exercising his right to waive his constitutional rights to a fair trial and presumption of innocence. In any event, reducing the incentive to plead guilty would encourage some defendants to plead not guilty and “try their luck”. 184. Finally, Mr Blanchflower submitted that before the Court should contemplate a material change in practice there must be strong policy reasons supported by clear, compelling evidence. They were absent. General submissions: Ngo Van Nam 185. In his submissions in respect of the general policy of affording a defendant a discount for a ‘timely’ plea of guilty, Mr Grossman SC invited the Court to note that the practice had developed that exceptional circumstances had to be shown to withhold a discount of one-third from the starting point for sentence for a defendant who pleaded guilty before or at the first day of trial. Such circumstances included the fact that it was a retrial or the need to hold a Newton hearing. 186. Of the practice in England and Wales, Mr Grossman said that it was clear that the one-third discount afforded in such circumstances was separate and distinct from all other mitigating factors. The Court should distinguish between artificial remorse, theoretically attributable to a plea of guilty, when the defendant was caught red-handed on the one hand and a genuine and demonstrable remorse evidenced by a volunteered confession to the offence. The submissions of the Department of Justice Discount of sentence upon a plea of guilty 187. In summarising his submissions, Mr Yeung said that the Courts’ practice of discounting a defendant’s sentence upon a plea of guilty accords with the public interest. It was an acknowledgement of the utilitarian value of a plea of guilty, the defendant’s remorse and his acceptance of responsibility. He contended that the full one-third discount should be afforded only to those defendants who entered a timely plea and are genuinely remorseful.[145] 188. Of what was to beregarded as a timely plea, he submitted that it was, subjectively, to what extent the plea was indicative of remorse, acceptance of responsibility and willingness to facilitate the course of justice and, objectively, the utilitarian value of the plea.[146] He contended that a significant factor was the timing of the plea. In that respect, the appropriatequestion to pose was when it was reasonable, in all the circumstances and as a matter of practicality, to have expected a plea of guilty to have been tendered or indicated. 189. Mr Yeung contended that the line of judgments of this Court to the effect that a full one-third discount should normally be afforded to a defendant, even for a plea of guilty made on the first day of trial, is at odds with the policy reasons for granting a discount for a plea and at variance with an earlier and even longer line of authorities of the Court, in which court-door pleas were not afforded the full discount. He submitted that court-door pleas ought not to be afforded the full one-third discount and invited this Court to clarify the conflict in the judgments of this Court. 190. Mr Yeung suggested that all mitigating factors that flow from the utilitarian value of a plea of guilty and the subjective factor of remorse, other than exceptional remorse, are subsumed in the one-third discount. 191. Of the approach taken by the then Sentencing Guidelines Council in England and Wales, namely the provision of a sliding scale from a full discount of one-third afforded to a defendant who pleaded guilty at the“first reasonable opportunity” to not more than one-tenth for a court-door plea, Mr Yeung said that whilst this Court has always regarded the timeliness of a plea as relevant, it had never set any guiding percentage to be applied mechanically by reference to the stage of the proceedings at which a plea was made. The question of discount has been a matter for the discretion of the trial judge. 192. Finally, Mr Yeung submitted that the approach in England and Wales of having pre-set percentage discounts to be applied mechanically depending principally upon the stage at which the plea was entered was not to be supported. The issue of what constituted the “first reasonable opportunity” had proved to be controversial and had led to many appeals against sentence. A consideration of the submissions Discount for a plea of guilty: strength of the prosecution case 193. As we indicated in the course of the hearing, the practice of not having regard to the strength of the prosecution case in determining the discount to be afforded to a defendant for his plea of guilty is not only well established and of long-standing in Hong Kong but also this Court has provided cogent reasons for the change from the earlier practice, where regard was had to that factor. 194. As noted earlier, powerful support for that approach is to be found in the judgments of Lord Justice Clerk Gill and Lord Eassie in Gemmell v HM Advocate[147] and, albeit necessarily to a lesser extent, given the sentencing guidelines, in the judgment of Hughes JA in R v Caley. In the latter case, Hughes JA said:[148] “ The various public benefits which underlie the practice of reducing sentence for plea of guilty apply just as much to overwhelming cases as to less strong ones. Next, judges ought to be wary of concluding that a case is “overwhelming” when all that is seen is evidence which is not contested. Thirdly, even when the case is very strong indeed, some defendants will elect to force the issue to trial, as indeed is their right. It cannot be assumed that defendants will make rational decisions or ones which are born of any inclination to co-operate with the system, but those who do merit recognition.” 195. Earlier in his judgment, Hughes JA had cited with approval an observation of Lord Judge CJ in his judgment in the Court of Appeal of England and Wales in R v Wilson [149], in which the Court held that the judge was wrong not to afford the appellant any discount in stipulating the minimum term of imprisonment to be served for sentences of life imprisonment for his pleas of guilty, notwithstanding the judge’s finding that the prosecution case was overwhelming. Lord Judge CJ said “Even in an overwhelming case the guilty plea has a distinct public benefit. The earlier that it was indicated, the better for everyone.” [150] 196. Furthermore, as noted earlier, in England and Wales the Sentencing Council has proposed in its Draft Guideline to remove the requirement that, in affording a defendant a discount of sentence for his plea of guilty, consideration be given to the factor of the strength of the prosecution case. In making that recommendation, the Sentencing Council said that the benefits flowing from a plea of guilty “apply regardless of the strength of the evidence against an offender”. Further, they noted that research indicated that “withholding the guilty plea reduction is not applied consistently”. Of that issue, it was said, “What amounts to an overwhelming case is necessarily a subjective judgment and courts have interpreted it differently.” 197. In contrast to the cogent reasons advanced in a number of judgments of this Court, which resulted in the eventual resolution of the Court that the strength of the prosecution case was not to be taken into account in reducing the discount afforded to a defendant for his plea of guilty, with great respect, nothing in the way of analytical reasoning has been advanced in the judgments of this Court to explain the change from a discount of about 25% for an early plea of guilty to the current situation in which a defendant is afforded a discount of one-third for his plea of guilty up to and until the commencement of the trial. As noted earlier, Hong Kong is the only jurisdiction to which we have been referred in which a discount of one-third is afforded to a defendant on the first day of trial. The stage of a plea of guilty: a range of discount 198. There is no doubt that the utilitarian value of a plea of guilty is greater the earlier the plea of guilty is intimated or tendered. The authorities to which we have referred speak with one voice in that respect. We are satisfied that it is not only logical, but also fair to reflect that factor in sentencing, so that a defendant who pleads guilty at an earlier stage is to be afforded a greater discount in sentence than a defendant who pleads guilty at a later stage. That was the practice to which Roberts CJ referred in the Queen v Wong Ping Yu & Another. There, it was said that the practice was that a discount of up to 20% from the starting point for sentence was afforded to a defendant who pleaded guilty on the first day of trial, whereas a discount of up to 25% was afforded to a defendant who pleaded in the Magistracy and was committed for sentence. 199. We are satisfied that, subject to the overriding discretion of the judge in sentence, a discount of 20% from that taken for the starting point for sentence is the appropriate discount to be afforded to a defendant who pleads guilty only on the first day of trial. In broad terms it reflects the reduced utilitarian value of the plea of guilty, in comparison to a plea of guilty intimated at an early stage. On the other hand, although it was the practice of times gone by to afford a discount of up to 25% from the starting point taken for sentence to a defendant who pleaded at an early stage in the proceedings, and although no reasons or explanations were articulated in the judgments of this Court in which it was said that the practice had changed to one affording a discount of one-third to those who pleaded guilty at that early stage, we acknowledge that the practice is not only very well established but also working effectively, in that it encourages those who are guilty to plead guilty. Accordingly, we are satisfied that, subject to the overriding discretion of the judge in sentence, a one-third discount is appropriate in those circumstances. That range of discount in sentence, reflecting the stage at which a plea of guilty is indicated, affords those who wish to plead guilty an appropriate incentive to intimate a plea of guilty at an early stage. Factors of mitigation subsumed in a one-third discount 200. In contrast to the approach in sentencing in Scotland and Australia, as evidenced in the judgments of the Courts of those jurisdictions, in Hong Kong to a considerable extent this Court has elided the difference between the utilitarian value of a plea of guilty and the subjective factor of remorse. So, other than in exceptional circumstances, the remorse of a defendant who pleads guilty is taken to be subsumed in the discount of one-third afforded for a plea of guilty. As noted earlier, this Court has used the phrase the “high water mark” in a number of judgments in reference to the one-third discount to describe the fact that remorse is subsumed in that discount. Notwithstanding that the practice does not reflect a pure, principled approach to sentencing, we are satisfied that it is not only well established but also working in practice. Accordingly, we are satisfied that it is appropriate to continue sentencing on that basis. Knowledge of the facts of the prosecution case and receipt of legal advice 201. As noted earlier, in the judgments of the Court of Appeal of England and Wales in R v Caley and that of Lord Justice Clerk Gill in Gemmell v HM Advocate in the High Court of Justiciary of Scotland, in determining what was the “first reasonable opportunity” for a defendant to indicate his guilt or what was “an early plea” respectively, a distinction was drawn between that stage of proceedings and the separate and different stage at which a defendant exercised his undoubted right to avail himself of the opportunity, “for his lawyers to assess the strength of the case against him and to advise him on it.” [151] Of the different circumstances, Hughes LJ said “… the second depends on the evidence being assembled and served. The first, however frequently does not.” However, he acknowledged that “… there will certainly be cases where a defendant genuinely does not know whether he is guilty or not and needs advice and/or sight of the evidence in order to decide.” Nevertheless, he went on to conclude:[152] “ Such cases aside, however, whilst it is perfectly proper for a defendant to require advice from his lawyers on the strength of the evidence (just as he is perfectly entitled to insist on putting the Crown to proof at trial), he does not require it in order to know whether he is guilty or not; he requires it in order to assess the prospects of conviction or acquittal, which is different.” [Italics added.] In our judgment, with respect, there is considerable force in those observations. An early plea (i) The Court of First Instance 202. In practice, almost all cases that are heard in the Court of First Instance are committed to the Court from the Magistracy, either for trial or for sentence. We understand from the most helpful submissions of Mr Pang that, on their first appearances in the Magistracy, those defendants who are not legally represented privately are usually represented by the Duty Lawyer Scheme. It is the practice in the Magistracy to inform those defendants whose cases are to be committed to the Court of First Instance, by service of a form entitled ‘Important Notice’, that the Duty Lawyer Service will not represent them in committal proceedings and that they may instruct lawyers privately or apply for legal aid. The appointment of the ‘Return Day’ 203. Section 80 A of the Magistrate’s Ordinance makes provision for the appointment of a Return Day by the Court for the continuation of committal proceedings, which is to be not less than 10 days and not more than 42 days from the date of the appointment of the Return Day. Upon reasonable cause being shown, the Court may appoint another Return Day. On the appointment of the Return Day, the Magistrate is required to inform the accused: “ (a) of his right to apply for legal aid; (b) that not less than 7 clear days before the return day he will receive a copy of the complaint made or information laid together with copies of witness statements and any documentary evidence in support thereof, being the statements and evidence upon which the prosecutor will seek the accused’s committal; (c) on the return day, he will have the right to require a preliminary inquiry and, if he does so require, he may, at the inquiry, call witnesses to give evidence on his behalf; (d) where there is more than one charge that, in the event of his requiring a preliminary inquiry on any charge, the inquiry will be held into all the charges against him and that only at the conclusion of the inquiry will he have the opportunity to plead guilty to any charge; (e) if he does not require a preliminary inquiry, he will be committed for trial without an inquiry unless he pleads guilty to the charge, in which case he will be committed for sentence on that charge.” 204. The Court was informed by Mr Pang that it was normally after the appointment of the Return Day that defendants apply for legal aid and that normally the Legal Aid Department process those applications within 10 days. Similarly, we were informed that it was commonplace for adjournments to be sought by the prosecution and granted by the Court to enable the prosecution to serve the requisite material on the defendant. Furthermore, that again it was commonplace that the Legal Aid Department received that material only shortly before the Return Day so that, on the application of the Legal Aid Department, proceedings were often adjourned yet again. Mr Pang said that adjournments of that kind were sought so that the Legal Aid Department could consider the papers with a view to determining whether there is a prima facie case disclosed in the papers, to advise the defendant on the mode of committal and to deal with other pre-committal notice - such as bail applications. 205. Mr Pang informed the Court that Legal Aid counsel “attended” defendants on the Return Day and before their committal. Helpfully, we were provided with a pro forma used in such interviews, entitled “Matters to be dealt with in Committal Interview”. The acronym ‘A/P’ used in the form appears to refer to the aided person. The form states, inter-alia: “ 1. when A/P received papers? Consent to abridgement of less than 7 days? If no papers, case will be adjourned and A/P consents to adjournments? Does AP require translation of papers other than English before/after committal? 2. A/P was given an opportunity to view non-documentary exhibits? Want it if not? 3. A/P read the papers? A/P understands charge he is facing? 4. A/P’s intended plea. In case of A/P pleading guilty, matters normally adjourned 4 weeks for Summary of facts to be agreed and plea. 5. Counsel’s advice given to A/P re whether Prima facie case & mode of Committal. 6. A/P understands the advice and takes counsel’s advice? 7. Yes cautioned statement given involuntarily, does A/P once had complained to be registered with CAPO? 8. Alibi warning explained. 9. Bail application. Instructions and advice. 10. Post Committal Procedure explained. 11. Any other matters.” Procedure on the Return Day 206. Section 80 C of the Magistrate’s Ordinance requires the Magistrate to inform the accused on the Return Day that, unless he elects to have the charge heard at a preliminary hearing, he will be committed without such an inquiry. Further, provision is made that, if the defendant does not make or is deemed to make such an election, the Magistrate shall inform the accused that:[153] “ (a) he is not obliged to say anything in respect of the charge but that he may plead guilty to the charge and that such a plea will result in his being committed for sentence on that charge; ” It is in those circumstances that a defendant is committed to the Court of First Instance, either for trial or for sentence. The Indictment 207. Section 14(1)(a) of the Criminal Procedure Ordinance requires the Secretary for Justice to file an indictment within seven days of the committal of the defendant to the Court of First Instance for trial.[154] The Listing Judge 208. After an indictment is filed with the Court of First Instance, the matter is fixed for hearing before the Listing Judge. Usually, that takes place within four to six weeks of the committal having occurred. 209. Mr Pang informed the Court that, after a hearing date is fixed before the Listing Judge, the Legal Aid Department arranges for one of its clerks to take further instructions from the defendants whom they represent: (i) to take listing instructions; and (ii) to ascertain whether he intends to plead guilty or not. Whether or not the defendant indicates that he wishes to plead guilty or proceed to have a trial, legal aid counsel represents the defendant at the Listing hearing. In the event that the defendant maintains his instructions to plead not guilty and only after trial dates are fixed does the Legal Aid Department take steps to assign solicitors and counsel to represent the defendant at trial. Solicitors and counsel assigned by the Legal Aid Department 210. As is to be expected, after they have been assigned by the Legal Aid Department to represent a defendant, arrangements are made by solicitors to take full instructions from the defendant, so that they are able to brief counsel. A one-third discount 211. We are satisfied that in cases committed for trial or sentence the stage at which a discount of a full one-third is to be afforded to the defendant is at the stage of committal described earlier. Usually, but subject to the overriding discretion of the judge in sentence, the opportunity to secure a one-third discount from the starting point for sentence occurs when the defendant is given the opportunity to plead guilty in the Magistracy, pleads guilty and is committed for sentence to the Court of First Instance. Obviously, that is an event of importance to a defendant. So, those representing a defendant who is to be committed to the Court of First Instance must advise a defendant of his options, so that he is in a position to make an informed choice. As noted earlier, that does not require that a defendant is given an assessment of “the prospects of conviction or acquittal”. At issue only, is whether or not in his instructions the defendant acknowledges that he performed the acts, with the accompanying mental element, proof of which is required to establish the offence. A 25% discount 212. We are satisfied that, if the defendant indicates to the Court or prosecution that he wishes to plead guilty after he has been committed to the Court of First Instance for trial, but up to and until dates are fixed for his trial by the Listing Judge, the appropriate discount for sentence from the starting point for sentence, subject to the overriding discretion of the judge in sentence, is 25%. Plea of guilty at the first day of trial: a 20% discount. 213. Further, we are satisfied that, subject to the overriding discretion of the judge in sentencing, it is appropriate that the Courts revert to the previous practice of affording a defendant who pleads guilty on the first day of his trial a discount of 20% from the starting point taken for sentence. An indication of a plea of guilty after the fixing of trial dates but before the first day of trial: 25% to 20% discount 214. Next, we are satisfied that, subject to the overriding discretion of the judge in sentencing, the range of discount to be afforded to a defendant who indicates to the Court or the prosecution that he intends to plead guilty after trial dates have been fixed but before the first day of trial, lies between the 25% discount to be afforded to the defendant who indicates that he wishes to plead guilty up to and until dates are fixed for his trial by the Listing Judge and the 20% discount to be afforded to the defendant for a plea of guilty on the first day of trial. In determining the appropriate discount to be afforded to the defendant in those circumstances, the judge will have regard to the time at which the indication was given and to all the other relevant circumstances. A plea of guilty after arraignment and during the trial 215. Finally, we are satisfied that, subject to the overriding discretion of the judge in sentencing, the discount to be afforded to a defendant who pleads guilty after arraignment but during the trial itself would usually be less than the 20% afforded to the defendant who pleads guilty on the first day of trial and will reflect the circumstances in which the plea was tendered. Often, it will follow the holding of a voir dire. In those circumstances, it may be that the nature of the challenge to the admissibility of the evidence will be relevant. In other circumstances, it may occur in circumstances where the defence has sought to test some other aspect of the prosecution case. The discount to be afforded to the defendant in those circumstances is pre-eminently one for the trial judge. (ii) The District Court 216. By contrast to the procedure obtaining for committal of a defendant to the Court of First Instance, a defendant who is transferred to the District Court from the Magistracy has no opportunity to tender a plea of guilty in the Magistracy. 217. Mr Pang informed the Court that, apart from those who are privately legally represented, other defendants, if they wish to be so represented, may be represented by the Duty Lawyer Scheme. On the day of their transfer to the District Court defendants are informed by the Duty Lawyer Service by service of a notice entitled ‘Important Notice’ that the Duty Lawyer Service will not be available to them in the District Court but that, if they are not to be represented privately, they may apply for legal aid, so that they may be represented in the District Court. The recipients of such notices are enjoined to take steps to be represented “well before” the date for their first appearance in the District Court. Mr Pang informed the Court that the processing of such applications for legal aid takes about 10 working days. Plea Day 218. Although the first day of hearing of cases transferred from the Magistracy to the District Court is uniformly known as the “Plea Day” no pleas are in fact tendered. Rather, it is a hearing at which an indication may be given of what pleas are to be tendered and at which dates may be fixed for trial. 219. Mr Pang informed the Court that it was estimated that 85% of defendants appearing in the District Court did so having been granted legal aid. However, he said that often legal aid counsel first saw those defendants who had been granted legal aid in the dock in the Plea Day Court. It appears that unsatisfactory situation arises because of a combination of factors: the time taken in processing the applications for legal aid following transfer to the District Court; the lack of manpower in the Legal Aid Department and the fact of the short time gap between transfer from the Magistracy to the Plea Day hearing. As a result, the Court was informed that in about 30% of the cases the proceedings were adjourned to a second Plea Day hearing. There was no dispute that in those circumstances adjournments were readily granted by the judge presiding in the Plea Day hearing. 220. In taking instructions from defendants who have been granted legal aid, the law clerks of the Legal Aid Department make use of a pro forma entitled “General Instructions” (Bail applications/Plea Day cases). Amongst the information sought is that in respect of the defence: “ (a) no knowledge of existence/substance (b) no intention (c) wrongly identified by witnesses (d) fabricated by witnesses (e) fingerprint/DNA (f) exhibits (g) possession of DD your own consumption (h) with consent” 221. Elsewhere the form addresses issues in respect of cautioned statements; identification evidence; trafficking in dangerous drugs; defence witnesses; medical evidence; fingerprint evidence; and the chain of evidence. An indication of a plea of guilty at Plea day: one-third discount 222. We are satisfied that, subject to the overriding discretion of the judge in sentencing, a one third-discount from the starting point taken for sentence is to be afforded to those defendants who indicate at the Plea Day that they intend pleading guilty. That is to be the case whether that is done on the first of such hearing dates or on a subsequent Plea Day, necessitated in order to obtain adequate instructions and provide a defendant with appropriate advice. A lesser discount is to be afforded to a defendant who pleads guilty thereafter. Plea of guilty at the first day of trial: 20% discount 223. If the plea of guilty is tendered at the first day of trial, subject to the judge’s overriding discretion in sentencing, the appropriate discount for sentence is 20% of that taken as the starting point. An indication of a plea of guilty after the fixing of trial dates but before the first day of trial 224. Subject to the overriding discretion of the judge in sentencing, a defendant who gives the Court or the prosecution an indication of a plea of not guilty at the Plea Day after which trial dates are fixed, who then indicates to the Court or the prosecution before the first day of trial that he wishes to plead guilty, is to be afforded a discount between 25% and 20% of that taken as the starting point for sentence. In determining the appropriate discount to be afforded to the defendant in those circumstances, the judge will have regard to the time at which the indication was given and to all the other relevant circumstances. A plea of guilty during the trial 225. We are satisfied that, subject to the overriding discretion of the judge in sentencing, the discount to be afforded to a defendant who pleads guilty after the plea is taken and during the trial itself would usually be less than the 20% afforded to the defendant who pleads guilty on the first day of trial and will reflect the circumstances in which the plea was tendered. The discount to be afforded to the defendant in those circumstances is pre-eminently one for the trial judge, having regard to all the circumstances including those mentioned in paragraph 215. The Magistracy 226. In the Magistrates Courts the Duty Lawyer Service covers some 300 statutory and common law offences. However, it does not encompass committal proceedings; hawking offences; traffic summonses and regulatory offences, such as summonses issued by the Environmental Protection Department, Inland Revenue Department and the Fire Services Department 227. Legal representation is provided to defendants for their first appearance in court without means testing. That representation is available by application to the Court Liaison office at each of the Magistrates Courts. The Court Liaison office makes contact with all defendants who are remanded in custody. Subsequent representation is subject to a means test, for which the financial eligibility limit is set at a gross annual income of $185,810. Having passed the means test, a defendant is required to pay a fixed handling fee of $540, which encompasses all hearings no matter how long the trial lasts. The Administrator of the Duty Lawyer Service has the discretion to waive the means test if she is of the view that it is in the interests of justice so to do. 228. Mr Pang informed the Court that at the first hearing of a case in the Magistrates court the prosecution make available to the magistrate and the Duty Lawyer a copy of the charge(s) and a document entitled ‘Brief Facts’ which, as its title suggests, gives a short summary of the prosecution case in respect of circumstances of the commission of the offence. Armed with that document the Duty Lawyer takes instructions from the defendant with the assistance of a pro forma. At the beginning, that document provides for information to be set out of the “Offence Charged” and the “Defendant’s Plea”. Also, it provides for the provision of further details under the headings “Personal Particulars”; “Any outstanding case?”; “Convictions”; “Bail” and “Mitigation”. 229. There is no dispute that often a Duty Lawyer is called upon to represent multiple defendants at any given hearing day. As a result, it is often the case that application is made by the Duty Lawyer for an adjournment to obtain material from a prosecution or to obtain better instructions, which applications are usually granted by the Magistrate. Often, the prosecution indicates to the Court that they are not ready for a plea to be taken from the defendant. That may occur, for example, because the prosecution requires a Government Chemist’s certificate in respect of the analysis of dangerous drugs. So, in many cases there is an adjourned hearing before the prosecution asks for a plea to be taken from the defendant, and the defendant indicates whether or not he pleads guilty or not guilty. In the latter case, trial dates are fixed. In the former case, it may be that the case can be dealt with that day or, if not, a date is fixed for sentencing. 230. In cases where a defendant is represented privately by a lawyer, it is to be expected that at the first hearing day, if the defence is not in a position to tender a plea to the charge, for whatever reason, an application would be made for an adjournment. On the other hand, if the defence is ready to proceed with tendering a plea to the charge, the case can proceed and, if the plea is one of not guilty, trial dates can be fixed. 231. It follows that, in common with the practice in the higher courts a defendant is afforded not only the opportunity to be informed of the charge and, by provision of the Brief Facts, made aware of the prosecution case in summary but also has the opportunity to seek legal advice before being called upon to tender a plea to the charge. As is to be expected it is clear that, on the request of the defence, the Courts grant adjournments to those defendants who are not in that position. Plea of guilty: the one-third discount 232. In those circumstances we are satisfied that, usually, but subject to the overriding discretion of the magistrate in sentencing, the opportunity to secure a one-third discount from the starting point for sentence occurs when the defendant is asked to tender a plea to the charge. If he pleads not guilty and trial dates are fixed that opportunity is lost. Plea of guilty at the first day of trial 233. If the defendant pleads guilty at the first day fixed for his trial usually, but subject to the overriding discretion of the magistrate in sentencing, the discount to be afforded to the defendant is 20% of the starting point for sentence. An indication of a plea of guilty after the fixing of trial dates but before the first day of trial 234. If the defendant indicates to the Court or to the prosecution in advance of the date fixed for trial that he wishes now to tender a plea of guilty, followed by a plea of guilty, the discount to be afforded tothe defendant, subject to the overriding discretion of the magistrate in sentencing, lies between 25% and 20% of the starting point for sentence. In determining the appropriate discount to be afforded to the defendant in those circumstances the magistrate will have regard to the time at which the indication was given and to all the other relevant circumstances. A plea of guilty during the trial 235. We are satisfied that, subject to the overriding discretion of the judge in sentencing, the discount to be afforded to a defendant who pleads guilty after the plea is taken and during the trial itself would usually be less than the 20% afforded to the defendant who pleads guilty on the first day of trial and will reflect the circumstances in which the plea was tendered. The discount to be afforded to the defendant in those circumstances is pre-eminently one for the trial magistrate, having regard to all the circumstances including those mentioned in paragraph 215. 236. The revised practice of affording discounts of sentence for pleas of guilty is to be applied only to those who, in future, reach the stages in criminal proceedings identified in this judgment at which revised discounts of sentence for pleas of guilty are identified. So, for example, the revised practice applies to a defendant currently in the magistracy who, in the future, is committed to the Court of First Instance for trial. On the other hand, for example, a defendant in respect of whom trial dates have been fixed in the Court of First Instance who pleads guilty on the first day of his trial is to benefit from the existing practice of affording a discount of one-third from that taken as a starting point for sentence. 237. A draft of this judgment was circulated to all other members of the Court of Appeal, all of whom have indicated that they support the revision of the general practice of affording a discount of one-third on a plea of guilty, set out in paragraphs 193-236 of this judgment. 238. I agree with the judgments of Yeung VP in CACC 418/2014 and Macrae JA in CACC 327/2015. CRIMINAL APPEAL NO. 327 OF 2015 Hon Macrae JA : 239. The applicant (Abdou Maikido Abdoulkarim), a Nigerian national aged 35 at the time of the offence, pleaded guilty on 17 September 2015 before Deputy Judge Beeson in the High Court to one count of trafficking unlawfully in a dangerous drug, namely 2,410.28 grammes of a crystalline solid containing 2,357.23 grammes of methamphetamine hydrochloride (commonly known as ‘Ice’). The facts 240. The applicant had arrived at 21:05 hours on 7 April 2014 at the Shenzhen Bay Control Point, New Territories, Hong Kong from the Mainland. He was escorted to the Customs departure hall for immigration clearance. Upon inspection and investigation of his luggage, a total of 24 packages of ‘Ice’ were found concealed in the lining of six ladies’ handbags, which, amongst other things, he was carrying inside his backpack. Following his arrest and caution, the applicant was asked who owned the drugs, to which he replied: “I don’t know, my friends give me to bring six handbags to his girlfriend in Hong Kong.” When asked whom he meant by “his friends”, he said he did not know and was waiting for someone to call him after he arrived. 241. In a subsequent video recorded interview, the applicant denied knowing that there were dangerous drugs inside the six handbags, or that there were handbags inside his backpack. However, the applicant admitted before the judge when he pleaded guilty that he had trafficked in the drugs particularised in the indictment. 242. The retail value of the dangerous drugs as at April 2014 was HK$980,983. The judge’s reasons for sentence 243. The judge adopted an initial starting point of 22 years’ imprisonment for the narcotic concerned after extrapolating the guidelines of this Court in Attorney-General v Ching Kwok Hung [155], the facts giving rise to his arrest having occurred about two months before the revised guidelines for ‘Ice’ in HKSAR v Tam Yi Chun [156] were handed down. She duly enhanced the starting point for sentence by 2 years’ imprisonment for the international element in trafficking more than 2 kilogrammes of ‘Ice’ across the border into Hong Kong. She then reduced the notional sentence of 24 years’ imprisonment after trial by 25%, resulting in a sentence of 18 years’ imprisonment. 244. In her reasons for sentence, the judge explained the reasons for her departure from the one-third discount for plea as follows: “ When considering the discount, I note that the defendant was arrested on 8 April and first appeared in the Magistracy on 9 April 2014. On 17 October 2014 he pleaded guilty at committal. On 14 January 2015 he appeared for sentence in the High Court and told the court that he did not know that the dangerous drugs were in the bag he carried. He also told the court that he wanted to be sentenced anyway for the offence on that day, but the deputy High Court judge who was presiding clearly could not accept that plea and explained why that was so to the defendant. The plea of guilty was vacated and conviction set aside. A ‘For Mention’ hearing date followed on 16 April 2015 and indicated an adjournment for plea. On 6 May 2015 a plea of not guilty was entered and the case was ordered to be re-listed. On the 27 July 2015, a pre-trial review proceeded and the trial dates were fixed from 7 September 2015 with eight days allotted. On 14 September, three days before the trial was due to start, the court was advised by letter that the defendant would plead guilty to the charge. I set out this timetable to illustrate the waste of court time and public money that has been caused by the defendant’s dilatory behaviour, and I have considered whether this would justify my reducing the usual one-third discount for plea. I note that the defendant has been legally advised throughout these proceedings. I note too that on all occasions after his arrest, an Igbo Interpreter has been made available to him. His initial plea was admitted as being made voluntarily and with a full understanding of the charge. On its face it was an unequivocal plea of guilty. I am quite aware of the one-third discount for plea being an important factor in court sentencing. However, the norm is that the one-third discount is given for a timely plea. It is not right that a discount remain available at its full range and in all cases, especially when there has been delay. A three day window before the trial advising his plea means that an eight-day trial has been wasted as it is not possible for a comparable case to be slotted into those days as would be desirable. He has also had a long gap between the arrest and the initial plea to have legal advice. There was further opportunity for legal advice between 14 January and the pre-trial review. Between the pre-trial review and the start of the trial that has been further time for him to get advice, or to consider giving information. I consider that some note should be taken of the fact of this delay. I take the 24-year sentence and discount it by one-quarter to 18 years.” The appeal 245. Mr Blanchflower SC takes a single ground of appeal on behalf of the applicant, namely that the judge erred in finding that the plea of guilty tendered by the applicant was not a “timely” one and that he was not therefore entitled to the full one-third discount, thereby rendering the sentence manifestly excessive. He argues that not only was the reduced discount of 25% inconsistent with prevailing sentencing principles applicable at the time of sentence, but the judge gave no warning of what she proposed to do, thus depriving the applicant’s counsel of the opportunity of making submissions on the issue and, if necessary, referring to relevant authorities. Discussion 246. This Court can well understand the frustration of the courts and those whose responsibility it is to list cases for trial when defendants prevaricate or change their minds and instructions about their pleas resulting in trials that are aborted and ensuing hearing dates that are rendered useless. The applicant’s conduct in this particular case exemplifies how the utilitarian value of a timely plea was minimal. One should not forget that when an 8-day trial in the High Court is aborted at short notice, not only is the time of the courts and witnesses wasted, but other defendants who are in custody patiently waiting for their own trials to take place are also disadvantaged, particularly so if ultimately they are acquitted. 247. Unfortunately, however, the judge did not draw her entirely proper concerns to defence counsel’s attention at any stage prior to passing sentence, and it seems evident from the transcript of proceedings that he was completely unaware of what she had in mind. Indeed, he had based his mitigation on an assumption, which he articulated at the outset of his remarks as follows: “ The defendant has obviously pleaded guilty and seeks the one-third discount on the sentence of imprisonment which must follow.” 248. The judge did not demur or engage counsel on his submission at all. No doubt counsel would have sat down at the conclusion of his plea in mitigation, satisfied that his assumption was correct and that his client would receive a full one-third discount from whatever the starting point was found to be. Had the judge voiced her misgivings about the timetable of the applicant’s plea and raised the possibility that she was considering adopting a reduced discount in view of it during mitigation, counsel may well have sought a brief adjournment and pointed to some of the authorities to which we have referred earlier in this judgment; in particular, HKSAR v Lo Chi Yip [157]; HKSAR v Li Tak Yin [158]. Whether his submissions on the issue were or were not ultimately accepted, the matter would at least have been canvassed and he would have had the chance to try and persuade the judge out of the course she was indicating. 249. It might be suggested from the timetable of the applicant’s plea that counsel should have anticipated that the judge might well have been considering a reduced discount for plea and that he should not have assumed, therefore, that his client would receive the full discount. There would be some force in that observation. However, here counsel raised his expectation that the full one-third discount would follow at the outset of his mitigation and issue was not joined with the proposition by the court. In those circumstances, we think counsel was entitled to assume from the judge’s silence when directly confronted with the submission, particularly given the state of the authorities which we have discussed, that she was not going to depart from the usual one-third discount for plea. 250. In the particular circumstances, therefore, we feel that we should allow the appeal and give the applicant what we think his counsel evidently anticipated and what he would no doubt have told his lay client to expect, namely the conventional one-third discount from the starting point for pleading guilty. In the circumstances, we are prepared to grant the application for leave, and treating the hearing of the application as the hearing of the appeal, we allow the appeal and reduce the starting point (about which no complaint is made) by one-third, resulting in a sentence of 16 years’ imprisonment in place of 18 years’ imprisonment. 251. To that extent, the appeal is allowed. 252. I agree with the judgments of Yeung VP in CACC 418/2014 and Lunn VP in CACC 418/2014 and 327/2015. Mr Keith Yeung, SC, DPP, Ms Christal Chan, SPP, Ms Rosa Lo, SPP and Mr Ivan Cheung PP, of the Department of Justice, for the respondent Mr Clive Grossman, SC and Mr Andy Hung, instructed by Augustine C.Y. Tong & Co., assigned by Director of Legal Aid, for the appellant (in CACC 418/2014) Mr Michael Blanchflower, SC and Mr Michael Arthur, instructed by Krishnan & Tsang, assigned by Director of Legal Aid, for the applicant (in CACC 327/2015) Mr Robert Pang, SC and Ms Maggie Wong, appointed as Amicus Curiae by the Court [1] HKSAR v Lo Kam Fai (CACC 374/2014; unreported, 2 February 2016). [2] R v Cameron, in which, for different reasons, Kirby J concurred with the joint judgment of Gaudron, Gummow and Callinan JJs. [3] R v Cameron [2002] 209 CLR 339. [4] HKSAR v Ma Ming [2013] 1 HKLRD 813, at 820. [5] HKSAR v Lo Kam Fai, paragraph 5. [6] HKSAR v Lo Kam Fai, paragraphs 6 and 7. [7] HKSAR v Lo Kam Fai, paragraphs 87-8. [8] Prison Rules, Cap. 234A, rule 188. [9] Prison Rules, rule 192. [10] Prison Rules, rule 195. [11] Prison Rules, rule 196. [12] Prison Rules, rule 201. [13] Prison Rules,rule 203. [14] Prison Rules, rule 206. [15] Prison Rules, rule 38. [16] Prison Rules, rule 48. [17] “The length of any sentence of imprisonment imposed on a person by a court shall be treated as reduced by any period during which he was in custody by reason only of having been committed to custody by an order of a court made in connection with any proceedings relating to the sentence or the offence for which it was passed, or with any proceedings from which those proceedings arose…” [18] The Queen v Wong Ping Yu & Another (CACC 16/1987; unreported, 12 March 1987). [19] The Queen v Kwok Chi Kwan and Another [1990] 1 HKLR 293. [20] Attorney General v Wong Kwok Wai [1991] 2 HKLR 384. [21] The Queen v Wong Ping Yu & Another (CACC 16/1987; unreported, 12 March 1987). [22] The Queen v Wong Ping Yu & Another. [23] The Queen v Leung Tin Man (CACC 411/1990, unreported, 19 April 1991). [24] The Queen v Wong Ngai Hung (CACC 596/1995; unreported, 3 April 1996). [25] HKSAR v Wong Ka Kuen (CACC 35/1998; unreported, 28 January 1999). [26] HKSAR v Wong Ka Kuen. [27] The Queen v Chan Chi Yuen (CAAR 8/1983; unreported, 8 July 1983). [28] The Queen v Leung Yiu Hung & Others (CACC 452/1984; unreported, 23 November 1984). [29] The Queen v Wong Ping Yu & Another (CACC 16/1987; unreported, 12 March 1987). [30] Attorney General v Han Man Fai & Another (CAAR 7/1988, unreported, 14 September 1988). [31] Attorney General v Han Man Fai & Another, page 5. [32] Attorney General v Han Man Fai & Another, pages 5 and 6. [33] The Queen v Kwok Chi Kwan [1990] 1 HKLR 293. [34] The Queen v Kwok Chi Kwan, page 296 A-C. [35] The Queen v Leung Tin Man (CACC 411/1990, unreported, 19 April 1991). [36] The Queen v Leung Tin Man, page 5. [37] Attorney General v Wong Kwok Wai [1991] 2 HKLR 384. [38] Attorney General v Wong Kwok Wai, page 387 C-D. [39] Attorney General v Wong Kwok Wai,page 387 E-F. [40] The Queen v Law Hon Chung (CACC 548/1990; unreported, 25 February 1992). [41] The Queen v Law Hon Chung, page 5. [42] The Queen v Law Hon Chung, page 5. [43] The Queen v Lai Kwok Hung [1994] 1 HKC 283 (4 May 1994). [44] The Queen v Lai Kwok Hung, page 285 A-B. [45] The Queen v Lai Kwok Hung,page 285 B-C. [46] The Queen v Chan Leung (CACC 347/1994; unreported, 17 January 1995). [47] The Queen v Ng Wing Kwong (CACC 62/1995; unreported, 1 September 1995). [48] The Queen v Wong Ngai Hung (CACC 596/1995; unreported, 3 April 1996). [49] R v Lo Chi Man [1996] 4 HKC 699 (22 May 1996). [50] The Queen v Lau Kin Hong (HCMA 355/1996; unreported, 11 July 1996). [51] The Queen v Wu Yau Man (CACC 27/1996; unreported, 29 October 1996). [52] The Queen v Lun Nai Kin (CACC 652/1996; unreported, 18 April 1997). [53] The Queen v Guo Jun (CACC 366/1996; unreported, 30 January 1997). [54] HKSAR v Leung Kwai Sing (CACC 484/1999; unreported, 11 February 1999). [55] HKSAR v Leung Kwai Sing, page 2. [56] HKSAR v Wong Ka Kuen (CACC 35/1998; unreported, 28 January 1999). [57] HKSAR v Yeung Kin Man [2000] 2 HKLRD 821. [58] HKSAR v Wong Ka Kuen, page 2. [59] HKSAR v Wong Ka Kuen,page 9. [60] HKSAR v Wong Ka Kuen,page 9. [61] HKSAR v Yeung Kin Man, page 822 G. [62] HKSAR v Yeung Kin Man, page 923 D-E. [63] HKSAR v Yeung Kin Man, page 823 G-H. [64] HKSAR v Chui Chi Chi & Another (No 2) [2000] 1 HKLRD 704. [65] HKSAR v Chui Chi Chi & Another (No 2), page 706 F-G. [66] HKSAR v Chui Chi Chi & Another (No 2),page 707 E-G. [67] HKSAR v Lee Man Ki (CACC 66/2000; unreported, 18 April 2000). [68] HKSAR v Lee Man Ki, pages 2 and 3. [69] HKSAR v Lo Chi Yip [2000] 3 HKLRD 274 (21 July 2000). [70] HKSAR v Lo Chi Yip, page 275 H-J. [71] HKSAR v Lo Chi Yip, page 274 G-H. [72] HKSAR v Lo Chi Yip,page 275 E-F. [73] HKSAR v Lo Chi Yip,page 275 F-G. [74] HKSAR v Lo Chi Yip, page 275 G-H. [75] HKSAR v Lo Chi Yip, page 276 E-F. [76] HKSAR v Lo Chi Yip,page 277 A-D. [77] HKSAR v Lo Chi Yip,page 277 F-G. [78] HKSAR v Lo Chi Yip,page 276 F-H. [79] HKSAR v Lee Kwok Chuen (CACC 445/2000; unreported, 10 August 2001). [80] HKSAR v Lee Kwok Chuen, paragraph 11. [81] HKSAR v Li Tak Yin [2003] HKLRD 519 (6 February 2003). [82] HKSAR v Li Tak Yin;page 521 G-K, paragraph 7. [83] HKSAR v Li Tak Yin;page 522 E-G, paragraphs 12 and 13. [84] HKSAR v Li Tak Yin; page 523 A-D, paragraph 14. [85] HKSAR v Ting Chiu & Another [2003] 3 HKLRD 378 (14 August 2003). [86] HKSAR v Ting Chiu & Another; page 383 J, paragraph 17. [87] HKSAR v Ting Chiu & Another; page 384 C, paragraph 18. [88] HKSAR v Ting Chiu & Another; page 384 E-G, paragraphs 19-20. [89] HKSAR v Ting Chiu & Another; page 385 I-J, paragraph 23. [90] HKSAR v Ting Chiu & Another; page 386 A-C, paragraph 23. [91] Yu Fai Tat v HKSAR (2004) 7 HKCFAR 293. [92] Yu Fai Tat v HKSAR; page 297 G-J, paragraph 9. [93] Secretary for Justice v Lee Chun Ho, Jeff [2010] 1 HKLRD 84 (3 November 2009). [94] Secretary for Justice v Lee Chun Ho, Jeff; page 87, paragraph 9. [95] Secretary for Justice v Lee Chun Ho, Jeff; pages 90-1, paragraph 37. [96] HKSAR v Chow Yuen Fai [2010] 1 HKLRD 354 (8 December 2009). [97] HKSAR v Chow Yuen Fai: page 367, paragraphs 30-1. [98] HKSAR v Wong Ying Ho [1999] 4 HKC 825. [99] HKSAR v Tsang Cho Kiu (CACC 42/2014; unreported, 27 October 2014). [100] HKSAR v Ma Ming [2013] 1 HKLRD 813. [101] HKSAR v Ma Ming, paragraph 23. [102] HKSAR v Ma Ming, paragraphs 26 and 27. [103] ‘Reduction in Sentence for a Guilty Plea’ (Revised 2007), at paragraph 2.2. [104] R v Caley (2013) 2 Cr App R (S) 305. [105] R v Caley; page 310, paragraph 1. [106] R v Caley; page 311, paragraphs 5 and 6. [107] R v Caley; page 312, paragraph 9. [108] R v Caley; page 313, paragraph 14. [109] Draft Guideline ‘Reduction in Sentence for a Guilty Plea Guideline’ Consultation, page 5. [110] Draft Guideline ‘Reduction in Sentence for a Guilty Plea Guideline’ Consultation, page 7. [111] Draft Guideline ‘Reduction in Sentence for a Guilty Plea Guideline’ Consultation, page 41. [112] Draft Guideline ‘Reduction in Sentence for a Guilty Plea Guideline’ Consultation, page 41. [113] Draft Guideline ‘Reduction in Sentence for a Guilty Plea Guideline’ Consultation, page 41. [114] Draft Guideline ‘Reduction in Sentence for a Guilty Plea Guideline’ Consultation, page 40. [115] ‘Reduction in Sentence for a Guilty Plea’ - Definitive Guideline Revised 2007, paragraphs 5.3 and 5.4. [116] Draft Guideline ‘Reduction in Sentence for a Guilty Plea Guideline’ Consultation, page 40. [117] Gemmell v HM Advocate [2012] S.C.L. 385; SLT 484. [118] Gemmell v HM Advocate, paragraph 27. [119] Gemmell v HM Advocate, paragraph 33. [120] Gemmell v HM Advocate, paragraph 34. [121] Gemmell v HM Advocate, paragraphs 31-2. [122] Gemmell v HM Advocate, paragraph 78. [123] Gemmell v HM Advocate, paragraph 37. [124] Gemmell v HM Advocate, paragraph 38. [125] Du Plooy v HM Advocate [2003] S.L.T. 1237. [126] Gemmell v HM Advocate, paragraph 8. [127] Gemmell v HM Advocate, paragraph 8. [128] Gemmell v HM Advocate, paragraph 41. [129] Gemmell v HM Advocate, paragraph 42. [130] Gemmell v HM Advocate, paragraph 48. [131] Gemmell v HM Advocate, paragraph 149. [132] Gemmell v HM Advocate, paragraph 50. [133] Gemmell v HM Advocate, paragraph 51. [134] Gaudron, Gummow and Callinan JJ [135] Cameron v The Queen (2002) 209 CLR 339. [136] Cameron v The Queen, paragraph 11. [137] Siganto v The Queen (1998) 194 CLR 656 at 663-664 [22], per Gleeson CJ, Gummow, Hayne and Callinan JJ. [138] Cameron v The Queen, paragraphs 12-14. [139] Siganto v The Queen (1998) 194 CLR 656 at 663 [22], per Gleeson CJ, Gummow, Hayne and Callinan JJ. See also R v Gray [1977] VR 225 at 231. [140] Cameron v The Queen, paragraph 17. [141] Cameron v The Queen, paragraph 19. [142] Cameron v The Queen, paragraph 22. [143] Cameron v The Queen, paragraph 65(4). [144] HKSAR v Abdallah Anwar Abbas [2009] 2 HKC197, paragraph 14. “It has sometimes been said that a full discount of one-third is given in order to reflect a defendant’s ‘remorsefulness’. Perhaps these days this is more to be regarded as a convenient, if not antiquated, label to describe the attitude of someone who, by pleading guilty, not only shows some degree of remorse but also provides for himself powerful mitigation by reason of the saving of court time and public funds which would otherwise be expended on a trial and by obviating the need on the part of witnesses to testify which may often represent for them a considerable ordeal. The combination of these reasons will almost inevitably lead to the defendant being rewarded with a substantial discount, normally by a reduction of a third from the starting point, unless there is proper justification for departing from this practice.” [Italics added.] [145] R v Lau Kin Hong (HCMA 355/1996), paragraph 8. [146] Cameron v The Queen, paragraphs 22, 74 & 75. [147] Gemmell v HM Advocate, paragraphs 48 and 148. [148] R v Caley,paragraph 24. [149] R v Wilson [2012] 2 Cr. App. R.(S) 440. [150] R v Wilson, paragraph 29. [151] R v Caley; page 313, paragraph 14. [152] R v Caley; page 313, paragraph 14. [153] Section 80C(a) of the Magistrates Ordinance. [154] (1) The Secretary for Justice, if he sees fit to institute criminal proceedings, shall institute such proceedings in the court against the accused person as to him may seem legal and proper- (a) in the case of a committal for trial under section 80C(4) of the Magistrates Ordinance (Cap 227), within 7 days of such committal; [155] [1991] 2 HKLR 125. [156] [2014] 3 HKLRD 691. [157] [2000] 3 HKLRD 274. [158] [2003] 1 HKLRD 519. 
Chief Justice Cheung: 1. I agree with the judgment of Lord Hoffmann NPJ. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Lord Hoffmann NPJ. Mr Justice Fok PJ: 3. I agree with the judgment of Lord Hoffmann NPJ. Mr Justice Lam PJ: 4. I agree with the judgment of Lord Hoffmann NPJ. Lord Hoffmann NPJ: 5. This is an appeal against a decision of the Court of Appeal (Kwan VP, Au and G Lam JJA) in a mortgage action, ordering enforcement of a legal charge and payment of the loan and other monies thereby secured. The court reversed the decision of the judge (Patrick Fung SC, sitting as a Deputy High Court Judge), who had dismissed the action on the ground that both the charge and the recovery of the loan were unenforceable because the parties had agreed an excessive rate of interest: section 24(1) and (2) of the Money Lenders Ordinance (Cap 163). 6. The borrower was Hong Kong Sai Kung Ngong Wo Resort Development Limited, a wholly-owned subsidiary of a PRC company. The lender was Totalcorp (Nominees) Limited, a BVI company. The parties were put in touch with each other by one Lau Hoi Tuen (“Mr Lau”) who combined employment as a solicitor’s clerk with acting as a loan broker. After introducing the parties Mr Lau acted on behalf of the lender, employing other solicitors for the conveyancing and subsequent litigation. 7. There is some confusion about how much was lent. The borrower’s letter of application for the loan says HK$45 million. The Legal Charge dated 5 February 2015 says HK$44 million. The judge found it was HK$41,004,220, apparently because he deducted some HK$4 million which the borrower had agreed to pay to the lender and others (interest payable in advance to the lender, commissions payable to Mr Lau and another intermediary, conveyancing costs) and which it had directed the lender’s solicitors to pay out of the HK$45 million. I find it difficult to see how the borrower can be said to have paid these debts without borrowing the money. But there is no appeal against the finding that the loan was HK$41,004,220 and I say no more about it. 8. Mr Lau’s practice appears to have been to document the terms of the loan by setting them out in a “Loan Application” for the borrower to sign and return to him. In this case, the letter of 2 February 2015 said that the loan was for three months and that the rate of interest was to be 2.5% per month. 9. The borrower could not repay the loan on the due date (4th May 2015) but the lender agreed to an extension until 4 June 2015 which was documented in an “Application for Extension of Loan Term” signed on behalf of the borrower on 5 May 2015. It said that during the extension period the rate of interest was to be 3.5% per month. Two further extensions were agreed on 4 June 2015 (for another month at 3.5% per month) and on 6 July 2015 (for up to 10 days at 4% per month). On 17 July 2015 there was a fourth extension, apparently sine die (the period of extension was left blank) at 4% per month. The interest was to be payable on the capital loan (said to have been HK$45 million) together with arrears of interest up to 5 July 2015. This amounted in all to HK$47,879,550 and was said to produce an interest liability of HK$63,839.40 a day. If one calculates this as an annual rate on a loan of HK$45 million, it is 51.8%. On a loan of HK$41 million it is 56.8%. 10. Section 24 of the Money Lenders Ordinance (Cap 163) provides as follows: (1) Any person (whether a money lender or not) who lends or offers to lend money at an effective rate of interest which exceeds 60 per cent per annum commits an offence. (2) No agreement for the repayment of any loan or for the payment of interest on any loan and no security given in respect of any such agreement or loan shall be enforceable in any case in which the effective rate of interest exceeds the rate specified in subsection (1). 11. It follows that up to and including the fourth extension, the rates of interest charged by the lender fell within the maximum prescribed by the Ordinance. But the judge found that after the fourth extension, there was a further agreement which provided for the borrower to pay an annual rate of 74%. This made the loan irrecoverable. 12. The evidence upon which the judge found that such an agreement must have been made was a demand letter dated 14 June 2016 and sent by the lender’s solicitors to the borrower, of which a copy was produced by Mr Lau during the course of his evidence. It set out the borrower’s indebtedness in accordance with the terms of the four extensions: 13. The letter then said: “Further interest continued to accrue on the sum of HK$63,241,006.00 at the rate of 4% per month from 1st June 2016 until the date of payment.” 14. The judge said that he inferred from the letter and this last sentence in particular that there must have been an agreement that from 1 June 2016 the borrower would pay interest at the monthly rate of 4% on HK$63,241,006 instead of the fourth extension rate of 4% on HK$47,879,550. Applied to a loan which the judge held to have been HK$41,004,220, that produced an annual rate of 74%. 15. In my opinion there was no basis for a finding that there had been such an agreement. Nothing in the demand letter supports it. First, the statement in the letter about further interest does not allege that there has been any such agreement. It is simply an assertion as to what the author thinks the position is after the fourth extension. Secondly, the letter says that further interest “continued” to accrue. There is, I think, some ambiguity about the words “on the sum of HK$63,241,006.00”. It can mean, and perhaps in most contexts would mean, that HK$63,241,006.00 is to be treated as the sum on which the monthly rate of 4% is to be calculated. But it can also mean, and I think in this context was intended to mean, simply that accruing interest at 4% upon an unspecified capital will be added to the HK$63,241,006.00 liability. That such a meaning was intended is, I think, shown by the use of the word “continued”. That must mean that interest has been and will be added in the same way as before. The imposition of an interest liability calculated by reference to a new and higher capital sum cannot be described as a continuation. 16. There was no other evidence of what the judge called an agreement “contained in or reflected by” the demand letter and the Court of Appeal called an “inferred agreement”. The invariable practice by which the parties entered into agreements was for Mr Lau to draft a letter setting out the new terms and for it to be signed on behalf of the borrower. But the borrower never asserted that there was such an agreement and Mr Lau himself denied it. The judge took a poor view of Mr Lau’s conduct in failing to disclose his retained copy of the demand letter at an earlier stage and said that Mr Lau had hoped to conceal it in order not to reveal that there had been an agreement to pay an unlawful rate of interest. In my opinion there was nothing to support such a finding. The demand letter was of no particular significance. It had been followed by the commencement of a mortgage action which was afterwards discontinued. When Mr Lau was asked in cross-examination to produce a copy, he did so. The original of the letter would have been in the possession of the borrower, which would make an attempt to conceal its existence by suppressing the copy a rather futile exercise. 17. The action started after the 14 June 2016 demand letter was, as I have said, discontinued. On 14 October 2016 the parties entered into a new agreement (“the Final Agreement”) by which the terms of the loan after the first three months (i.e. the rates stipulated by the four extension agreements) were retrospectively revised. The rate of interest over the whole period of extension was to be 3.8% per month. The Final Agreement listed “all documents previously signed concerning the payment of interest on the Loan” and declared them to have ceased to be effective. The last document on the list is the fourth extension. There is no mention of a subsequent agreement increasing the annual rate of interest to 74% or any other rate. In my opinion this document, signed by the borrower, is inconsistent with there having been an “inferred agreement”. 18. An appellate court is traditionally reluctant to question a judge’s findings of primary fact. But the existence of the inferred agreement was not such a finding. It was an inference from the language of the demand letter and the conduct of the proceedings by the lender which in my opinion could not properly be drawn. 19. If, as I consider, there was no inferred agreement, that is sufficient to dispose of the appeal. The lender’s claim was based upon the Final Agreement, which provided for interest at an annual rate of 45.6% (3.8% × 12) and therefore neither it nor any predecessor agreement infringed section 24 of the Money Lenders Ordinance. The Court of Appeal, however, was content to accept that there had been an inferred agreement and dealt with the appeal in an altogether different way. They said that according to authorities such as BS Lyle Ltd v. Chappell [1932] 1KB 691, any change in the terms of a loan was deemed to take the form of a fresh loan on the new terms. The capital which the inferred agreement deemed to have been lent was therefore a new loan equal to the borrower’s entire indebtedness at the time, capital and arrears of interest included. The interest of 4% per month would be payable on this amount. That made it 48% per year. 20. This ground for allowing the appeal gave rise to the first certified question: “Whether BS Lyle Ltd v. Chappell [1932] 1KB 691 (followed in New Japan Securities International (HK) Ltd v. Lim Yiong Lin [1987] 3 HKC 153 and Honip Credit Ltd v. Leung Tak Sing Paul [2020] HKCA 879) lay down a rule of law or a rule of construction that whenever there is an extension of a loan a new loan capitalising the accrued interest comes into existence?” 21. The judge thought there was such a rule. He answered in the affirmative the question which he had put to himself “whether every time when there was an application for extension which was agreed to by the plaintiff the same should be treated as a new loan.” The Court of Appeal agreed. They both described it as a rule of construction but since it was indifferent to the actual language of the extension, it looks more like a rule of law. The Court of Appeal remarked that the judge appeared to have forgotten to apply this rule to the deemed new loan created by his inferred agreement. 22. In my opinion, however, there is no such rule. All that BS Lyle Ltd v. Chappell and the cases following it decide is that if the parties agree that money advanced on a fresh loan on new terms will be used to pay off an old loan, they can give effect to the transaction by set-off without having to pay over the money and then take it back. The agreement can be taken at face value. But there is nothing in any of these cases to suggest that if the parties agree to vary the terms of a loan, by extending the term or changing the rate of interest, that must count as entering into a new loan. In either case, the law gives effect to what, as a matter of construction, appears to have been the intention of the parties evinced by the language they have used. If it is expressed to be a new loan, it is treated as a new loan, notwithstanding that the money was used to pay off an old loan. If it varies the terms of an old loan, the agreement will vary the terms without creating a new loan. If there had been, as the judge found, an “inferred agreement” to pay 4% per month on the total indebtedness at the time, it would in my opinion have taken effect as a variation of the previous rate of interest and produced an annual rate in excess of 60%. However, as the inferred agreement was imaginary, the point does not arise. 23. As there was in my opinion no agreement which contravened section 24 of the Money Lenders Ordinance and the Final Agreement was not deemed to create a loan used to pay off a previously unenforceable loan, I do not think it is necessary to answer the second certified question, which is based on the contrary hypotheses. 24. I would therefore dismiss the appeal. 25. Further, I would make an order nisi that the appellant pay the costs of this appeal to the respondent and direct that, if any party wishes to vary this order, written submissions should be filed within 14 days of the handing down of this judgment and the Court will make a final order as to costs on the papers. Chief Justice Cheung: 26. Accordingly, the appeal is unanimously dismissed and the Court makes the order and direction indicated in [25] above. Mr Kenneth C L Chan, Mr Roland Lau and Mr Dixon Co, instructed by Kong & Tang, for the defendant (appellant) Mr Victor Dawes SC, Mr Joshua Chan and Mr Cedric Yeung, instructed by MinterEllison LLP, for the plaintiff (respondent) 
Chief Justice Ma: 1. I agree with the judgment of Mr Justice Tang PJ and Lord Collins of Mapesbury NPJ. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Mr Justice Tang PJ and Lord Collins of Mapesbury NPJ. Mr Justice Tang PJ and Lord Collins of Mapesbury NPJ: Introduction 3. ENM Holdings Limited (“ENM”) operates a private members’ Club, the Hilltop Country club (“the Club”) in Tsuen Wan, the New Territories. The land on which the Club (Lot number 360 in DD 454 – “the Lot”) is situated is the subject of a grant of a lease to ENM (under its former name of Essential Enterprise Co Ltd) on November 12, 1976, and extensions granted on July 15, 1980 and December 28, 1985. The terms of the Grant were modified on May 9, 1984 and May 22, 1986. Nothing turns on the extensions or modifications. 4. ENM built structures and facilities on the Lot for operating the Club. The Club is situated on top of a hill and surrounded by Government land. The only vehicular access leading from the public street at the foot of the hill up to the Club is Hilltop Road, which is on Government land. Hilltop Road was constructed before 1963, and by 1976 it was also the main access road to other lots in the vicinity. It is not known who constructed Hilltop Road, but the Building Authority (“the BA”) accepts that the Government was responsible for the upholding, maintenance and repair of the whole of the original Hilltop Road prior to the Grant. 5. Hilltop Road was formed by the “cutting and tipping” method, i.e. cutting into the natural terrain and end tipping the loose soil fill (extracted from the cutting) onto the natural hillside, with the road being constructed on top of the cutting and end tipping. The end tipping created man-made soil fill slopes and embankments on the downside hill of the road, giving support to the relevant portions of the road. The BA takes the view that the slopes are an integral part of the embankment supporting Hilltop Road and failure of those parts would lead to collapse of the road. 6. A long section of the road is shown coloured brown on the plan annexed to the Grant (“the Brown Area”) and this appeal concerns ENM’s obligation to maintain slopes which are outside the Brown Area, but are in the view of BA essential for the support of that part of Hilltop Road situated within the Brown Area. 7. Two dangerous hillside orders were served by the BA on ENM under the Buildings Ordinance, Cap 123, section 27A, requiring ENM as “the person who under the terms of a Government lease is under an obligation to maintain the land” to submit proposals for remedial work to be done to two man-made slopes on the side of Hilltop Road. These orders relate to two areas outside but along the Brown Area called Feature 156(1) and Feature 33(1). 8. ENM appealed to the Appeal Tribunal (Buildings) (“the Tribunal”). Before the Tribunal the BA contended that under Special Condition 31 (“SC31”) of the Grant ENM was obliged to maintain Feature 156(1) and that under both SC 13 and SC 31 ENM was obliged to maintain Feature 33(1). The Tribunal, by a determination dated September 17, 2013 (“the Determination”) ordered that the dangerous hillside order in relation to Feature 156(1) be withdrawn. And in relation to the dangerous hillside order concerning Feature 33(1), the Tribunal held that ENM was liable under Special Condition 13 (“SC13”) of the Grant to maintain only the Northern End of Feature 33(1) because “construction works carried out by (ENM) … caused a risk of landslip, subsidence or falling away at the Northern End of Feature 33(1),” and ordered that the dangerous hillside order be amended so as to limit it to the Northern End of Feature 33(1) only. 9. On BA’s application for judicial review, with ENM joined as the interested party, Au J by judgment dated October 27, 2016, quashed the determination of the Tribunal relating to Feature 156(1) holding that on its proper construction SC31 obliged ENM to maintain Feature 156(1) and in relation to Feature 33(1) such obligation arose under both SC31 and SC13. Au J also held that the Tribunal was wrong to have proceeded on the basis that BA’s case relating to Feature 33(1) was confined to the Northern End only, and remitted it to the Tribunal to determine whether ENM was obliged to maintain the entire Feature 33(1), because the Tribunal had misunderstood the BA’s case or had failed to give adequate reasons. 10. On ENM’s appeal, the Court of Appeal reinstated the Determination in relation to Feature 156(1) holding that on SC31’s proper construction, ENM was not obliged to maintain Feature 156(1) or Feature 33(1), but dismissed the appeal in relation to Feature 33(1) because of SC13. By its order of December 1, 2017, the Court of Appeal gave permission to the BA to appeal on the basis that the following questions raised points of law of general importance: “Question 1 On the proper construction of SC31, whether the maintenance obligation under it is confined to maintaining the paved way and any incidental structures on the paved way such as the matters mentioned in sections 19 to 25 of the Building (Private Streets and Access Roads) Regulations, or whether, on a proper construction, the maintenance obligation encompasses the obligation to maintain everything that is (a) constructed in association with the paved way (b) adjoining the paved way and (c) essential to the supporting of the paved way (including supporting slopes or structures)? Question 2 Whether the construction of SC31 ought to be different where (a) the paved way had already been constructed when the Government lease was executed or (b) the paved way had not been constructed at the time the Government lease was executed? Question 3 Whether a grantee subject to SC31 or a lease condition that is substantially similar to it (i.e. containing the phrase ‘uphold, maintain and repair such paved way and everything forming portion of or pertaining to it…’ or words to that effect) is, for the purpose of section 27A of the Buildings Ordinance (Cap. 123), a person whom the BA may serve a dangerous hillside order on as someone who under the terms of a Government lease is under an obligation to maintain natural, formed or man-made land or any earth-retaining structure?” The Facts Feature 156(1) 11. This is on the southern side embankment of Hilltop Road. 12. Feature 156 is the name of a hillside slope which has been divided into three sub-divisions. Feature 156 is 85m long, and of a maximum height of 11m, and has a slope gradient of 30 degrees. Work was done in the vicinity on land sold to developers for house construction. 13. Feature 156(2) falls within private land and the responsibility for maintenance falls on the land owner. The maintenance responsibility of Feature 156(3) initially rested on the grantee of Lot 1217SA in DD 451 but as this lot was resumed by the Government, the responsibility became the Government’s. 14. Feature 156(1) lies on Government land to the south of the club. Hilltop Road is located immediately at the crest of this slope. Feature 33(1) 15. The second feature is on the west side of the end of Hilltop Road, near the entrance to the club. It is also on Government land. Feature 33 is the name of a hillside slope which has been divided into two sub-divisions, Feature 33(1) and Feature 33(2). The maintenance responsibility of Feature 33(2) rested initially with the holder of a Short Term Tenancy which had since been cancelled, and the responsibility became that of the Government. 16. Feature 33 is 24m long and attains a maximum height of 10m, with a slope gradient of 35 degrees. There has at all material times been an access road within Feature 33(1), cutting Feature 33(1) into two halves. There is a footpath leading from that access road to other areas; the footpath is not within Feature 33(1) and was constructed between 1963 and 1964, and was not related to ENM. There were human activities within Feature 33(1), including the presence of squatters and other structures; and a village house was constructed between 1964 and 1973. 17. In 1981, the Northern End of Feature 33(1) was modified by ENM for construction of a car park platform within the Hilltop Club, but there has been no change since then.[1] The works were not in connection with any work to upgrade or re-site Hilltop Road.[2] The Grant and SC 13 and 31 18. The Grant contained Special Conditions, of which SC 31 provided as follows: “The grantee shall construct a paved way to the standards laid down in the Building (Private Streets and Access Roads) Regulations over the area shown coloured brown on the plan annexed hereto and shall uphold, maintainand repair such paved way and everything forming portion of or pertaining to itto the satisfaction of the said Director, and the grantee shall be responsible for the whole as if he were absolute owner thereof. Any alteration to the public street from which the paved way is to be constructed, absorbing a portion of such paved way or affecting the gradient thereof, shall not give rise to any claim by the grantee, who shall carry out all consequent alterations to such paved way constructed by him. The Government reserves the right to grant rights-of-way over such paved way to the owners of any other lots in the vicinity now or in the future, or to take over the whole or any portion of the said paved way for the purposes of a public street without payment of any compensation to the grantee or to other owners to whom rights-of-way over the whole or any portion of the said paved way may have been granted.” 19. The construction of SC31 lies at the heart of this appeal, and BA’s case that ENM is obliged to maintain Feature 156(1) depends on it. 20. At this point it is also relevant to mention that SC13 provided: “Where any cutting away, removal or setting back of adjacent or nearby hillside or banks or any building up or filling in is required for the purpose of or in connection with the formation, levelling or development of the lot or any part thereof, the grantee shall construct or bear the cost of the construction of such retaining walls or other support as shall or may then or at any time thereafter be necessary to protect and support such hillside and banks and the lot itself and to obviate and prevent any falling away, landslip or subsidence occurring thereafter, and shall at all times maintain the said retaining walls or other support in good and substantial repair and condition. In the event than as a result or arising out of any such formation, levelling or development any landslip, subsidence or falling away occurs at any time, whether in or from the adjacent hillside or banks and whether the same be Crown or leased land, or in or from the lot itself, the grantee shall at his own expense reinstate and make good the same and shall indemnify the Government from and against all costs, charges, damages, demands and claims whatsoever which shall or may be made, suffered or incurred through or by reason of such landslip, subsidence or falling away. In addition to any other rights or remedies herein provided for breach of any of the conditions hereof the Director of Public Works shall be entitled by a notice in writing to call upon the grantee to carry out such construction and/or maintenance or to reinstate and make good any falling away, landslip or subsidence, and if the grantee shall neglect or fail to comply with such notice within the period specified therein the said Director may forthwith execute and carry out the work and the grantee shall on demand repay to the Government the cost thereof.” Certificate of Compliance 21. On April 26, 1984 the District Lands Officer issued to ENM a Certificate of Compliancecertifying: “all the positive obligations imposed on the grantee in respect of (the Lot) under the general and special conditions of exchange contained in the (Grant) under which the same is held from the Crown have been complied with to my satisfaction.” Regulation of dangerous slopes 22. At the time of the Grant, slopes were not regulated: evidence before the Tribunal of Mr Philip Chung.[3] 23. But in 1980 the Buildings Ordinance, Cap 123, was amended by the addition of section 27A which gave the BA power to declare any natural, formed or man-made land, or any earth-retaining structure as dangerous and serve a dangerous hillside order on the owner of the land or structure or the person who, under the terms of what was then a Crown lease, was under an obligation to maintain the land or structure, requiring that person to carry out an investigation and do necessary work to make the land or structure safe. 24. The relevant part of section 27A, as subsequently amended in 1990, 1993, 1998, and 2004, is as follows: “(1) Where in the opinion of the Building Authority any natural, formed or man-made land, or any earth-retaining structure, has due to any cause been rendered so dangerous or liable to become so dangerous that it will collapse, or be likely to collapse, either totally or partially, and thereby will cause, or will be likely to cause, a risk of injury to any person or damage to any property, the Building Authority may by order in writing served on the owner of the land or structure, or on the person who under the terms of a Government lease is under an obligation to maintain the land or structure, declare the land or structure to be dangerous or liable to become dangerous. (2) An order under subsection (1) may require the owner or person referred to in subsection (1) — (a) to do such work as may be specified in the order; (b) to appoint an authorized person, a registered structural engineer or a registered geotechnical engineer or any combination of them specified in the order to carry out such investigation in relation to the land or structure as may be so specified; and (c) to submit for approval by the Building Authority proposals for work to be done to make the land or structure safe based on the findings of the investigation, within such time or times as may be specified in the order.” 25. Consequently, since 1980 section 27A has applied not only to the owner of the land or structure, but also to “the person who under the terms of a Government lease is under an obligation to maintain the land or structure.” Dangerous hillside orders 26. In August 2005, a Stage 2 Study Report prepared by Maunsell Geotechnical Services Ltd for the GEO/CEDD recommended a proposed Dangerous Hillside Order for Feature 156(1), and on December 27, 2006, the BA served on ENM a dangerous hillside order (“the 1st DH Order”) relating to Feature 156(1). 27. In August 2007, a Stage 2 Study Report prepared by Scott Wilson Ltd for the GEO/CEDD recommended a dangerous hillside order for Feature 33(1), and on January 25, 2008, the BA served on ENM a dangerous hillside order (“the 2nd DH Order”) relating to Feature 33(1). 28. Under the 1st and 2nd DH Orders, ENM was required to investigate the two slopes and submit proposals for remedial/preventive work for the BA’s approval. 29. Mr Philip Chung’s evidence was that the slopes were an integral and essential part of the support for Hilltop Road. He explained that he would use only the expression “integral and essential support” in respect of man-made slopes formed by cutting or filling and not for the natural terrain. 30. ENM does not challenge the BA’s view that the Features are dangerous or liable to become so.[4] The Tribunal’s decision 31. Before the Tribunal the BA argued that ENM came within section 27A of the Building Ordinance because it was “the person who under the terms of a Government lease [was] under an obligation to maintain the land or structure” in respect of Feature 156(1) by virtue of SC31 and in respect of Feature 33(1) by virtue of both SC31 and SC13. The Tribunal decided that ENM was not obliged to maintain either Feature 156(1) or Feature 33(1) under SC 31 properly construed. Accordingly, it allowed ENM’s appeal against the 1st DH Order and ordered it to be withdrawn. In respect of the 2nd DH Order, the Tribunal was of the view that ENM was obliged to maintain only the northern end of Feature 33(1) under SC 13, and ordered the 2nd DH Order to be amended limited to the Northern End of Feature 33(1) only. 32. The Tribunal’s construction of SC31 was substantially influenced by the undisputed fact that “(at the time of the grant) both the Government and (ENM) knew very well that there was already in existence a paved way.” The Tribunal believed “that at the time, both the Government and (ENM) must have discovered that firstly the paved way was substandard and secondly the paved way had not been properly sited. So in substance, what the Government really required of (ENM) under SC (31) was not to construct a paved way afresh but to upgrade the paved way to the standards of a public street and to position it properly within (the Brown Area).”[5] The Tribunal then concluded that the expression “‘everything pertaining to’ in SC (31) should mean ‘everything within the Brown Area that pertains to the paved way.’”[6] And “that the common intention of the parties in respect of (ENM’s) maintenance responsibility under SC (31) was indeed intended by the parties to be limited to the surface of the right-of-way. That being the case, we conclude that the true meaning of the words ‘pertaining to’ in SC (31) in the context of this case should mean ‘everything that was constructed on the surface of the paved way.’”[7] That the words “pertaining to” in SC31 should mean “everything that was constructed on the surface of the paved way.”[8] Decision of Au J 33. Au J gave leave to apply for judicial review. ENM was joined in the proceedings as the interested party. By a judgment handed down on October 27, 2016, Au J quashed the Tribunal’s determination and remitted the appeals to the Tribunal, differently constituted if necessary, for reconsideration in light of his findings. 34. On the construction of SC31, Au J was unable to agree with the Tribunal because: “50. … The Tribunal came to this view based on what it believed to be what the parties ‘must have discovered’ at the time of the New Grant. This ‘belief’ is not supported by any evidence before it. The Tribunal also did not set out in the Determination what is the evidence or basis that it relied on to support such ‘belief’. Such ‘belief’ is also not something which in my view could be reasonably and objectively inferred.” 35. To the contrary, Au J believed: “42. … it must be objectively within the parties’ reasonable contemplation at the time of the New Grant that, in constructing the paved way on the original Hilltop Road to a standard and quality required under the Regulations, necessary corresponding construction works might have to be carried out at those hillside slopes along the original Hilltop road. 43. With this context in mind, it is clear to me that the objective intention of the parties is that it was the grantee who would have an obligation to maintain and repair not only the paved way constructed by it, but also other structures formed as a result of the construction works, and which are necessary for the construction and support of the paved way.” 36. He then concluded “the words ‘everything … pertaining to’ the paved way are intended to refer to everything that is formed as a result of the paving construction works, and which has become substantially connected to or in support of the ‘paved’ Hilltop Road.”[9] 37. Au J’s order was that, as regards SC31, the matter should be remitted to the Tribunal for reconsideration of whether Features 156(1) and 33(1) were man-made slopes which resulted from the grantee’s construction works and were substantially connected to or in support of Hilltop Road; and that, as regards SC13, the matter should be remitted to the Tribunal for reconsideration of whether the whole of Feature 33(1) was the responsibility of ENM, because the Tribunal had misunderstood the BA’s case or had failed to give adequate reasons. Court of Appeal 38. By its judgment of August 11, 2017, the Court of Appeal allowed ENM’s appeal as regards SC31, and dismissed the appeal as regards SC13. The appeal to this court is only against Au J’s decision on SC31. 39. The Court of Appeal noted that Au J’s construction of SC31 was not the construction advanced by BA before him.[10] BA’s construction was that ENM should be treated as if it had constructed the original Hilltop Road and the supporting soil filled slopes whereas the judge construed SC31 on the basis that it was within the reasonable anticipation of both parties that ENM might need to carry out construction work on the slopes to upgrade and re-site the original Hilltop road and hence ENM’s liability would depend on whether the features resulted from the construction work actually carried out by ENM, [11] and for this purpose he remitted the appeals to the tribunal for rehearing. Before the Court of Appeal, Mr Mok[12] submitted that SC 31 should be interpreted to mean that: “the grantee was treated as having assumed the responsibility to construct or to complete the construction of the original Hilltop Road and the supporting soil filled slopes, and has thereby also assumed maintenance responsibility for the road and the slopes, even though the construction work was not carried out by the grantee.”[13] (emphasis added) 40. He “acknowledged that everything would fall into place if at the time of the grant the original Hilltop Road had yet to be built.”[14] And “To overcome the fact there was an existing road and the man-made slopes had been formed, he submitted that SC31 should be interpreted to mean that the grantee was treated as having assumed the responsibility of constructing, completing the construction of, the original Hilltop road and the necessary supporting soil filled slopes, even though the grantee had not done so.”[15] Kwan JA[16] said this was a strained construction and there was nothing to show that the parties had treated or assumed a factual basis different from that which prevailed at the time.[17] 41. Kwan JA then concluded that the grantee’s obligation under SC31 was to construct the paved way situated in the Brown Area to the standards of the Regulations, which was the existing Hilltop Road that had already been constructed by cutting and tipping with man-made slopes and embankments. And that SC31 only obliges the grantee to maintain everything within the boundaries of the paved way being the Brown Area on the plan annexed to the Grant, and any incidental structures on the paved way such as stones, manhole covers and gratings, drains sewers and channels referred to in sections 19 to 25 of the Regulations.[18] The arguments Preliminary question 42. ENM has raised the question whether the BA’s principal argument, namely that the parties must be taken to have contracted on the basis that at the time of the Grant, the access road was required to be constructed, is properly before this court. 43. The Court of Appeal gave the BA leave to appeal to this Court on three questions, of which questions 1 and 2 set out, inter alia, the BA’s interpretation, and both Au J and the Court of Appeal considered the BA’s interpretation. The Court therefore accepts the BA’s argument that the question that requires the Court’s determination concerns the interpretation of SC31, which is a question of law. The BA’s arguments 44. BA’s principal argument can be taken from its printed case: “25. On the basis that the paved way had actually been constructed before the Grant, SC 31 therefore must be read on the assumption (which both parties presumably knew was not true) that the paved way had not been constructed and was required to be constructed.” This led to what was said to be an overarching submission that: “26.1 the contractual intention was clearly that the grantee would be required to maintain both the paved way and any man-made slopes that would have been created or come into existence for the construction or support of the paved way.” Mr Yu SC[19] submitted where parties, knowing the full facts, enter into a transaction by assuming a different state of facts or law as between themselves for the purposes of that transaction, the parties are estopped from denying or challenging the assumed state of facts or law: Prime Sight Ltd v Lavarello [2014] AC 436 (PC), para 47; Springwell Navigation Corporation v JP Morgan Chase Bank [2010] 2 CLC 705, paras 143-144, 156-169; Nokia Corporation v TCT Mobile (unrep, CACV 191/2014, 6 March 2017), paras 20-24; Spencer Bower, Reliance-based Estoppel (5th ed, 2017), paras 8.67-8.69; Chitty on Contracts (32nd ed, 2015), paras 4-116. ENM’s arguments 45. The approach advocated by the BA requires a strained reading of SC31 by attributing to the parties an intention to contract on an assumed state of affairs (i.e. that the access road had yet to be formed by the grantee) which they knew to be untrue when there is no evidence to suggest that the parties had intended to contract on that basis. 46. The construction of SC31 advocated by the BA is also difficult to reconcile with the other provisions in the Grant (e.g., SC13, SC18, SC22 and C28) where clear language was used whenever it was sought to impose upon the grantee responsibilities to carry out repair and maintenance on Government land outside the boundaries of the Lot. 47. The result contended for by the BA is also objectively unfair in that on any view of the matter, the responsibility for maintaining and repairing all of the man-made slopes created by the formation of Hilltop Road along its entire length would impose a massive financial burden of unknown extent upon the grantee which bears no relation to the benefit obtained under the Grant. 48. The Court should not accept a construction of SC31 which would lead to such an unfair result in the absence of clear words to show that such a result was indeed intended by the parties. This is especially so when at the date of the Grant, it was most unlikely that the parties had applied their minds to the criteria for making private individuals liable for repair and maintenance of hillside slopes situated on Government land given that there was not yet in existence a government department dedicated to ensuring the safety of slopes. 49. Even if slope maintenance was on the parties’ mind, they could not have envisaged that the obligation would in time become as onerous as it has become in recent years given that section 27A of the Buildings Ordinance was enacted only in 1980 and since then the factor of safety applied to determine whether a slope is dangerous for the purposes of section 27A has been progressively increased. 50. The obligation to “uphold, maintain and repair” only extended to “such paved way” as the grantee has “constructed” pursuant to the obligation imposed by the first half of the first sentence of SC31. 51. Since a road had already been formed in the Brown Area prior to 1963, the obligation imposed upon the grantee by SC31 to “construct a paved way” should be read as excluding the work required to form the pre-existing road which had already taken place more than a decade before the date of the Grant. 52. The principle in Prime Sight does not apply because the parties have not contracted on a false assumption. Conclusions 53. SC31 has to be construed in the proper context, with regard to the purpose of the contractual term and the background that is objectively or reasonably known to the parties at the time of the agreement. The overriding objective is to give effect to what a reasonable person would have understood the parties to mean. See Jumbo King Ltd v Faithful Properties Ltd (1999) 2 HKCFAR 279, 296; Fully Profit (Asia) Ltd v SJ (2013) 16 HKCFAR 351, para 15; SJ v Joseph Lo Kin Ching (2015) 18 HKCFAR 169, para 29; Yung Chi Keung v Protection of Wages on Insolvency Board (2016) 19 HKCFAR 469, para 22; Wood v Capita Insurance Services Ltd [2017] AC 1173, paras 8-15. 54. The more unreasonable the result the more unlikely it is that the parties can have intended it, and if they do intend it the more necessary it is that they shall make that intention abundantly clear: Schuler (L) AG v Wickman Machine Tool Sales Ltd[1974] AC 235, 251. But on the other hand the Court must be alive to the possibility that one side may have agreed to something which with the benefit of hindsight may not have served his interest: Arnold v Britton [2015] AC 1619, paras 20, 77; Wood v Capita Insurance Services Ltd [2017] AC 1173, para 11. 55. The background and factual matrix are uncontroversial. (1) Hilltop Road was built by cutting and tipping at least since 1963. Both Hilltop Road and the Features 33 and 156 (the “slopes”) predated the Grant. (2) The Brown Area has always been government land. (3) The slopes which were situated outside the Brown Area were also government land (4) Government accepted that it had the duty to maintain Hilltop Road prior to the Grant. (5) It is likely, if not certain, that the Government had the duty to maintain the slopes prior to the Grant. (6) At all material times other people have been using the road. (7) There was an access road splitting Feature 33(1). (8) At all material times there were other people and their structures on Feature 33(1). (9) There was no regulation of slopes in 1976 (and not until 1980) but there were Building (Private Streets and Access Roads) Regulations for access roads. 56. What does SC31 mean? It consists of 3 sentences but only the first need concern us. The first sentence imposes three obligations: first, the grantee (ENM) shall construct a paved way over the Brown Area to the standards laid down in the Regulations; second, the grantee (ENM) shall uphold, maintain and repair (a) the paved way and (b) everything forming portion of or pertaining to it; third, the grantee (ENM) shall be responsible for the whole as if it were the absolute owner thereof. 57. The BA relied strongly on the principle that where parties, knowing the full facts, enter into a transaction by assuming a different state of facts or law as between themselves for the purposes of that transaction, the parties are estopped from denying or challenging the assumed state of facts or law: Prime Sight Ltd v Lavarello [2014] AC 436 (PC), para 47. But this is not a case where the parties assumed as between themselves that there was no paved way, and in our view the principle has no application. 58. The words “construct a paved way to the standards laid down in the Building (Private Streets and Access Roads) Regulations” are apt to describe the work required to upgrade a (possibly unpaved or not paved to the standards of the regulations) road which had already been formed by carrying out additional work to bring it up to the standards laid down in the Building (Private Streets and Access Roads) Regulations (e.g. by paving its surface and kerbing its verges etc. as prescribed by the regulations which would not require any work to be carried out on the slopes). 59. If the obligation imposed on the grantee by the first half of the first sentence of SC31 did not include the work that had already been carried out prior to the date of the Grant to form the road by the “cutting and tipping” method, then it must follow that the obligation to “uphold, maintain and repair” imposed upon the grantee by the second half of the first sentence of SC31 could not be intended to extend to an obligation to maintain the man-made slopes that were created in the course of formation of the access road by the “cutting and tipping” method. 60. The statement that “the grantee shall be responsible for the whole as if it were the absolute owner thereof”, does not assist in the interpretation of the scope of the grantee’s obligation under SC31. 61. It follows that in our view, ENM is not under an obligation to maintain Feature 156(1). 62. This appeal does not require us to determine ENM’s maintenance obligation under SC31 in relation to the paved way and we do not do so.[20] The Certified Questions 63. We will deal with the second question first. Question 2 “Whether the construction of SC 31 ought to be different where (a) the paved way had already been constructed when the Government lease was executed or (b) the paved way had not been constructed at the time the Government lease was executed?” 64. As we have said SC31 has to be construed in the proper context, with regard to the purpose of the contractual term in the background that is objectively and reasonably known to the parties at the time of the agreement. When the facts are different, the interpretation may well be different. If at the time of the Grant there had been no Hilltop Road at the Brown Area and it was necessary to build a paved way from scratch, it may be that SC31 would have required a paved way to be built. And if it was built by the cutting and tipping method, it may be that the obligation to maintain the paved way would have included an obligation to maintain the man-made slope formed as a result. However, this is not a question we should answer. This appeal is concerned with the construction of SC31 in the context of this Grant and Hilltop Road was built before the Grant. Question 1 “On the proper construction of SC31, whether the maintenance obligation under it is confined to maintaining the paved way and any incidental structures on the paved way such as the matters mentioned in sections 19 to 25 of the Building (Private Streets and Access Roads) Regulations, or whether, on a proper construction, the maintenance obligation encompasses the obligation to maintain everything that is (a) constructed in association with the paved way (b) adjoining the paved way and (c) essential to the supporting of the paved way (including supporting slopes or structures)?” 65. In this appeal, we are concerned with ENM’s obligation under SC31, in particular, whether ENM is responsible to maintain Feature 156(1) or Feature 33(1). We have explained why in our opinion, ENM is not responsible. The question as framed stated that those features were “(a) constructed in association with the paved way.” As we have explained, ENM’s obligation under SC31 was to construct the existing paved way to the standard laid down by the Regulations. ENM was neither obliged to build a new paved way nor was it the common intention of the parties that ENM should be liable as if it had constructed the paved way and Hilltop Road by the cutting and tipping method. As for (b) obligation to maintain everything “adjoining the paved way”, Feature 156(1) does indeed adjoin the paved way, but this does not assist in the construction of SC 31. As for (c) “essential to the support of the paved way (including supporting slopes or structures),” we are concerned with the obligation to maintain the slope at Feature 156(1). As we have noted ENM does not challenge BA’s view that the features are dangerous or liable to become so.[21] We are prepared to proceed on the basis that unless they are attended to, these dangerous slopes will endanger Hilltop Road. But this does not answer the question who is responsible to maintain the slopes or in the words of section 27A, whether under the terms of the Grant, in this context, SC31, ENM is obliged to maintain the slopes. We have stated above why we do not think so. Question 3 “Whether a grantee subject to SC31 or a lease condition that is substantially similar to it (i.e. containing the phrase “uphold, maintain and repair such paved way and everything forming portion of or pertaining to it…” or words to that effect) is, for the purpose of section 27A of the Building Ordinance (Cap. 123), a person whom the BA may serve a dangerous hillside order on as someone who under the terms of a Government lease is under an obligation to maintain natural, formed or man-made land or any earth-retaining structure?” 66. It is said SC31 or lease conditions substantially the same are used in other government leases. Given that the answer must depend on the facts and circumstances of each Government lease, we do not think it is appropriate or possible to answer this question. Disposition 67. For the above reasons, we would dismiss BA’s appeal. We would also make a costs order nisi in favour of ENM which shall become absolute unless within 14 days of this judgment, either party applies for a different order. Mr Justice Fok PJ: 68. I agree with the judgment of Mr Justice Tang PJ and Lord Collins of Mapesbury NPJ. (Joseph Fok)				Permanent Judge (Lord Collins of Mapesbury)				Non-Permanent Judge Mr Benjamin Yu SC, Mr Mok Yeuk-Chi and Mr Anthony Chan, instructed by the Department of Justice, for the applicant (appellant) Mr Ambrose Ho SC, Mr Allen Lam and Mr Michael Yin, instructed by Ford, Kwan & Company, for the Interested Party (2nd respondent) Appeal Tribunal (Buildings), the 1st respondent, attendance excused [1] Determination, p 28. [2] Court of Appeal, para 23. [3] The Chief Geotechnical Engineer of the Geotechnical Engineering Office, Civil Engineering and Development Department (GEO/CEDD). [4] Determination, p 4. [5] Determination, p 14. [6] Determination, p 15. [7] Determination, pages 16-17. [8] Determination, p 17. [9] Para 44. [10] Court of Appeal, para 20. [11] Court of Appeal, para 22. However, with respect, it is not clear whether that was Au J’s view. The Declaration 3A in the Order made by Au J suggests otherwise. [12] Who appeared for BA together with Mr Anthony Chan in the Court of Appeal. [13] Court of Appeal, para 39. [14] Court of Appeal, para 38. [15] Court of Appeal, para 21(4). [16] Whose judgment was agreed to by Lam VP and McWalters J. [17] Court of Appeal, para 41. [18] Court of Appeal, paras 41 and 42. [19] Appearing for BA together with Mr Mok and Mr Anthony Chan. [20] We express no view, for example, on whether the obligation is confined, as the Tribunal believed, to everything that was constructed on the surface of the paved way. Determination, p 17 and Court of Appeal, para 42. [21] Para 30 above. Determination, p 4. 
Chief Justice Ma: 1. I agree with the joint Reasons for Judgment of Mr Justice Ribeiro PJ and Lord Collins of Mapesbury NPJ. Mr Justice Ribeiro PJ and Lord Collins of Mapesbury NPJ: 2. At the hearing, this appeal was dismissed with costs. These are our reasons. 3. This appeal concerns a claim by a financial adviser for a success fee in respect of a fundraising transaction entered into by its client after termination of the advisory agreement. It raises questions concerning the proper construction of that agreement. A. The background 4. The respondent DIO Corporation (“DIO”), a Korean company listed on the KOSDAQ,[1] was looking to raise additional capital to expand its overseas business and was seeking international advice on fundraising. It engaged the appellant Eminent Investments (Asia Pacific) Limited (“Eminent”), a Hong Kong company, to act as its financial adviser pursuant to a Financial Advisory Agreement (“FAA”) dated 10 October 2008. It was drafted by Eminent and supplemented by an Addendum dated 15 January 2009 (“The Addendum”). 5. In April 2009, Eminent introduced Dentsply International Inc (“Dentsply”), a US company listed on the NASDAQ, to DIO as a prospective investor. After Eminent provided Dentsply with some information on DIO, a telephone conference call involving the three parties plus UBS (an investment bank advising Dentsply) occurred on 9 April 2009. 6. The call was described by the trial Judge, Mr Recorder Shieh SC, as involving discussions which were “very general” and in which no particular deals or fundraising transactions were raised.[2] While the representatives of Eminent and DIO came away from that call in an optimistic mood, in mid-April or at the end of April Dentsply informed them that it was not interested in proceeding. 7. On 30 April 2009, Charlie Lee, who had worked for Eminent left his employment. And on 8 June 2009, Eminent terminated the employment of Christopher Song and Jung Lee. All three had participated in the telephone conference with Dentsply as part of Eminent’s team. 8. On 16 June 2009, Eminent asked DIO if it was still looking for capitalisation and intimated that Song and Lee had been laid off. DIO did not seek Eminent’s further assistance and on 30 December 2009, it asked Eminent to return USD$70,000 which had been paid as consulting fees. DIO unsuccessfully argued at the trial that by demanding repayment it had accepted Eminent’s prior repudiatory breach, terminating the FAA. However, Eminent asserts that the Agreement was terminated on or about 29 January 2010 after one month’s notice. 9. It appears that Eminent ceased business in March 2010 and that Eminent and its principal named Kane Yang later pleaded guilty to breaches of the Securities and Futures Ordinance.[3] 10. The Judge found that in May 2010, Friadent GmbH (“Friadent”), a subsidiary of Dentsply, approached DIO regarding a possible OEM (original equipment manufacturing) arrangement for the manufacture of dental implants.[4] Charlie Lee, who had previously worked for Eminent, was engaged by DIO to assist in communications with Friadent and Dentsply. 11. In September 2010, Bret Wise, the CEO and chairman of Dentsply, visited DIO in Pusan and, having been much impressed, wanted to explore a greater partnership with DIO. The Judge found that Charlie Lee did a great deal of work liaising with DIO, Friadent and Dentsply to that end.[5] 12. On 9 December 2010, a deal between Dentsply and DIO was announced. Dentsply acquired less than 20% of DIO’s outstanding shares while its German subsidiary, Dentsply Germany Investments GmbH, acquired KRW56.6 billion worth of convertible bonds issued by DIO.[6] The Judge found that Eminent had failed to show that any work done by it under the FAA was the effective cause of this deal.[7] B. The relevant provisions of the FAA and the Addendum 13. The following provisions of the FAA and Addendum (further discussed in Section H of these Reasons) are of present relevance. They contain numerous grammatical and spelling errors. B.1 The FAA 14. The First Recital: “WHEREAS, the Company hereto has retained the Financial Advisor on a sole and exclusive basis to advise the Company with respect to its Corporate Governance, Equity Valuation, Financial & Accounting Analysis, Corporate Financial Planning, Strategic Capital Raising and other Corporate Financing activities”. 15. The Second Recital: “WHEREAS, the Parties hereto desire to specifically state the services mentioned above and under this agreement to be provided to the Company by the Financial Advisor, the compensation to be received by Financial Advisor [sic] from the Company for providing such services, and the terms and conditions that shall govern the relationship between the Parties.” 16. The FAA was for a term of 18 months, terminable by either party giving 30 days prior written notice.[8] 17. Clause 2 is headed “ENGAGEMENT & FEE STRUCTURE” and, in Clause i) it lists the services to be provided by Eminent including the study of DIO’s business operations and financial performance; assistance in preparing DIO’s business plan to be presented to potential investors; assembling and discussing business, financial and legal information with accountants, lawyers and auditors; providing quarterly research reports for institutional investors and advising on investor relations and on a communications and public relations strategy. Clause 2 i)(4) in particular states that services which Eminent is to provide include: “On a best effort basis, [advising] the Company on Mergers & Acquisitions, Fund Raising, private placements or shareholder restructuring prior to any IPO or Secondary Listing”. 18. Provision is then made for the payment of three types of fees, namely retainer fees (USD$100,000 in two instalments), an independent equity research fee (USD$30,000) and importantly, in Clause 2 iv) a “TRANSACTION FEE”, also referred to as a “success fee”: “Upon completion of any transaction for the Company[,] [t]he Company agrees to pay the Financial Advisor a success fee including and not limited to a three percent (3%) [sic] of the total transactional amount tied to any financial transaction related to Fund Raising or Private Placement or Shareholder restructuring, or Mergers & Acquisition for the Company.” 19. Clause 2 iv) applies to transactions completed during the currency of the FAA. Centrally important is Clause 3 i) which has been referred to as a “tail gunner clause” [9] and which relates to success fees payable after its termination. Under the heading “OBLIGATIONS OF THE COMPANY” it provides: “The Company agrees that within a period of two (2) years after the termination of this Agreement, should the Company complete a transaction including and not limited to an [sic] secondary listing or fund raising with any third parties or receive funds from a financing source introduced by the Financial Advisor, the Company shall pay Financial Advisor its fees according to this Agreement or and [sic] any executed amendments thereof or a generally accepted market compensation in case the transaction compensation method is not explicitly stated in this agreement.” 20. Eminent also relies on Clause 4 a) which, under the heading “MISCELLANEOUS”, states as follows: “Financial Advisor acknowledges that the Company will be providing confidential information to Financial Advisor. The Parties hereto agree that they will cooperate with each other and provide full due diligence, and that all conversations, documentation, or work products will be kept in the utmost confidence. The Company acknowledges that Financial Advisor will be providing confidential information to the Company. Accordingly, the Company agrees that the information, including but not limited to, Financial Advisor’s database of investors, methods, systems and procedures, will be kept confidential and shall not, without the prior written consent of Financial Advisor, be disclosed by the Company, except to its attorneys, accountant and Board of Directors. Financial Advisor intends to introduce the Company to its contacts, including private and institutional financing sources or strategic alliances. The Company agrees that it will respect Financial Advisor’s relationships with these investors and other sources of financing and that the Company shall not participate in or permits [sic] the circumvention of any obligation to Financial Advisor created by this Agreement or any means.” B.2 The Addendum 21. The Addendum dated 15 January 2009 recites (Second Recital) that “to officially expand the role of the Financial Advisor, the Parties have agreed to amend certain terms and/or conditions of the [FAA] ...”. 22. Third Recital: “WHEREAS, the [sic] under this Agreement, the role of the Financial Advisor has been appointed as sole and exclusive financial advisor for Financial Scenario(s) including and not limited to Secondary Listings in an overseas stock market, backdoor listings, Public Placements, Private Placements, Mergers & Acquisitions (‘M&A’), Management Buy-outs (‘MBO’), Leverage [sic] Buy-outs (‘LBO’) in one or series of Transaction(s) in any form including and not limited to loans, bonds, common stock, preferred stock, convertible securities or equity linked debt instruments of any kind conducted by the Company and its subsidiaries and affiliates directly or from any registered shareholder(s) of the Company.” 23. Fourth Recital: “WHEREAS, the following fees are excluded from the foregoing Success Fee and/or Financial Advisory Fee under the terms and condition of this Agreement and the First Agreement: All legal, accounting and audit costs for transactional due diligence and any specific circumstances required by the third parties, by law or by government regulations to retain professional services to complete a proposed transaction.” 24. The Fifth Recital makes it clear that the Addendum is to be read as part of the FAA: “WHEREAS, this Agreement specifies the agreed terms and conditions which applies [sic] to certain roles and responsibilities that are not stated on the First Agreement [ie, the FAA] and are to be included as part of the ... First Agreement.” 25. Paragraph 1 adds to the ENGAGEMENT & FEE STRUCTURE provisions of Clause 2 of the FAA as follows. 26. Paragraph 1a: “Upon completion of the listing or Secondary Placement of the Company on the London Stock Exchange Alternative Investment Market or AIM (or such other stock exchange as agreed to by the Company), a success fee in cash equivalent to two percent (2%) of the transaction value will be paid to the Financial Advisor, within fourteen (14) days of the Transaction closing date. ...” 27. Paragraph 1b: “Upon successful completion of any mergers & acquisition transaction involving the Company and a third party, the Company agrees to pay the Financial Advisor within fourteen (14) days of the Transaction closing date, a success fee in the amount equal to three percent 3% of the total transaction value in cash calculated based on enterprise value of the acquired company including any assumed liabilities.” 28. Paragraph 1d:[10] “In addition to the advisory fee for the First Agreement, the Company agrees to pay the Financial Advisor another fixed retainer of ... (USD450,000), paid in four installments [sic] upon the execution of this agreement. First installment [sic], ... (USD 50,000) is due within seven (7) days upon this Agreement signed by both parties. Second installment [sic], ... (USD 85,000) is due before February 20th, 2009. Third installment [sic], ... (USD 65,000) is due upon the completion of any of the following events: (i) the Company signs an Engagement Letter or Letter of Proposal from a Nominated Advisor ...; (ii) the Company formally approves a potential mergers & acquisition transaction in the form of but not limited to an Official Letter of Intent ... or Memorandum of Understanding ... or an Official Merger Agreement related to a mergers & acquisition transaction with a third party; (iii) the Company received and accepted the term sheet for the Transaction(s) of any of the Financial Scenario(s) under the terms and conditions of this Agreement. Fourth installment [sic] or final payment, ... (USD 250,000) is due within seven (7) days upon the completion of any the Transaction(s) under the Financial Scenario(s).” 29. Paragraph 2 of the Addendum removed the entitlement to an Independent Equity Research Fee under the FAA. C. Eminent’s construction 30. The fundamental proposition advanced by Mr Wong Yan Lung SC[11] on Eminent’s behalf is that Eminent is entitled to fees under Clause 3 i) which should be construed as requiring introduction of a party as opposed to introduction of a transaction. 31. This is elaborated upon as follows: “... [Clause 3 i)] should be construed against its relevant business contexts, purposes, common sense and efficacy, which included the need to protect Eminent as the [Financial Advisor], and to hedge against the risk of DIO as Client to circumvent the obligation to pay Eminent by terminating the engagement of Eminent before closing a deal with a financial source like Dentsply who was introduced by Eminent. Plainly, [Clause 3 i)] should be construed as requiring ‘introduction of a party’ as opposed to ‘introduction of a transaction’.”[12] “... Eminent’s construction of [Clause 3 i)], viz. ‘introduced by the Financial Advisor’ qualifies the ‘counterparty’ (i.e. a secondary stock exchange, a third party, or a financing source) with whom DIO entered into a deal or from whom DIO received funds, rather than ‘the transaction’, is a far more natural, logical and grammatical construction (as further elaborated in Section D2.3 below). In any event, Eminent’s construction must at least be an equally plausible construction, which means [Clause 3 i)] is far from clear. Even on the CA’s flawed approach, the CA should have considered the business contexts in the construction exercise.”[13] “The words ‘introduced by the Financial Advisor’ qualify or describe ‘the counterparty’ (i.e. (a) ‘a secondary listing’ (i.e. a secondary stock exchange other than Hong Kong), or (b) ‘any third parties’, or (c) ‘a financing source’), rather than ‘the transaction’.”[14] 32. Eminent contends that it is entitled to a Transaction Fee consequent upon DIO and Dentsply entering into the above-mentioned transaction on 9 December 2010 which occurred well within the two-year post-termination period referred to in Clause 3 i): “In April 2009, Eminent introduced to DIO a manufacturer and distributor of professional dental products called Dentsply International Inc (‘Dentsply’). The FAA was terminated on or about 29 January 2010 after one month’s notice. On 9 December 2010, a deal was announced whereby Dentsply acquired shares and convertible bonds of DIO (‘the Dentsply Deal’), just over 10 months (i.e. within 2 years) after such termination.”[15] 33. At the hearing, Mr Wong SC ventured to suggest that there is an alternative basis for triggering an entitlement to a Transaction Fee, namely, by Eminent rendering the advisory and preparatory services referred to in Clause 2 i) as part of a process of “shaping up” DIO to be attractive to potential investors. He sought to argue that even without Eminent having introduced DIO to the financing source or otherwise contributed to completion of the eventual transaction, provision of the Clause 2 i) services would be sufficient to qualify Eminent for payment of the Transaction Fee if DIO completed a fundraising transaction. 34. This was a new argument since the “shaping up” process is only mentioned in Eminent’s Written Case as a common feature of financial advisory agreements generally, providing “important context” for the purposes of construction.[16] The importance of introducing or “matching” the client with the investor is there maintained in line with Mr Wong SC’s fundamental proposition.[17] It is fair to say that the new argument was not seriously pressed and, being singularly unconvincing, must be rejected and needs no further discussion. D. DIO’s construction 35. DIO’s argument, advanced by Mr Charles Sussex SC,[18] is that the FAA should be construed as requiring “introduction to the transaction which ultimately took place (involving receipt of funds from a financing source)”.[19] This, it is argued, “imputes a degree of causation and/or proximity between the introduction and the eventual transaction, and accords with the general rule that a commission agent is only entitled to commission on a transaction where his services are the effective cause of the transaction being brought about”.[20] 36. As the Judge found that Eminent had failed to show that any work done by it under the FAA was the effective cause of the Dentsply deal announced on 9 December 2010,[21] DIO contends that no entitlement to the Transaction Fee arises. Alternatively, if contrary to DIO’s primary case, Eminent’s construction is preferred, DIO submits that the Court should imply an “effective cause” requirement to give business efficacy to the FAA.[22] E. The reasoning of the Judge 37. The Judge’s reasoning[23] was that the starting point was to ascertain how, for the purposes of Clause 2 iv) of the FAA, Eminent was to earn its fee during the currency of the contract, and then to ascertain the effect of Clause 3 i) on Eminent's ability to earn the Transaction Fee. The answer was that under Clause 2 iv) the fee was earned “[u]pon completion of any transaction for the Company.” That connoted active involvement and participation by Eminent in the transaction. Mere introduction was not enough, and Eminent had to do something so as to enable it to say that it has completed a transaction for DIO. Consequently, Clause 2 iv) was to be interpreted as requiring Eminent to be an effective cause in completing the transaction; or alternatively a term importing such an effective cause requirement was to be implied. 38. He held that, as a matter of common sense, Eminent could not be in a better position if the transaction in question took place after termination rather than before. A number of features of Clause 3 i) supported that conclusion: (a) the expression “[DIO] shall pay [Eminent] its fees according to this Agreement” meant Eminent’s fees as claimable and calculated according to Clause 2 iv); (b) the defining phrase was “a transaction,” which could not mean any transaction, but harked back to the sort of transaction which, under the Transaction Fee clause (Clause 2 iv)), would have entitled Eminent to a fee had the FAA not been terminated; (c) two of the three scenarios (secondary listing and fundraising with any third parties) did not contain any limiting phrase linking them to Eminent, and they could not be taken literally because otherwise it would mean that if DIO entered into a secondary listing transaction which had absolutely nothing to do with Eminent or work done by it, Eminent would still be able to claim a Transaction Fee against it, or if DIO succeeded in raising funds with a third party that it identified without involving Eminent in any way, Eminent would still be able to claim a Transaction Fee. F. The reasoning of the Court of Appeal 39. Cheung JA, with whom the other members of the Court of Appeal agreed,[24] held that the language of Clause 3 i) was clear enough to construe it as requiring Eminent to have introduced to DIO a transaction which DIO completed before Eminent was entitled to receive the fee. To focus simply on the word “introduction” would require the Court to ignore the rest of the phrase, namely, DIO completing “a transaction .... introduced by” Eminent. The three transactions named in Clause 3 i) were descriptive and not exhaustive (“a transaction including and not limited to”). The contextual background was that DIO wanted to seek capital to expand its business, and the entitlement to a fee by Eminent was clearly not merely based on an introduction to DIO of a party but an introduction which led to a transaction. The fact that Clause 3 i) was only effective for two years after a termination pointed to a relation back to Clause 2 iv) of work required to be done by Eminent under the FAA, namely, the introduction of a transaction which raised funds for DIO and its completion which entitles it to receive fees. Clause 3 i) could not be looked at in isolation from the other parts of the FAA. G. Interpretation G.1 The arguments 40. On this appeal Eminent has argued for a “purposive” approach to interpretation and has suggested that the Court of Appeal was wrong to conclude[25] that: (1) there was still a choice between “textualism” and “contextualism”; (2) where the document was professionally prepared, sophisticated and complex, the textual approach may be adopted without the need to resort to the context or factual matrix of the case, and context was relevant only to the clarification of inconsistencies or gaps. The result, according to Eminent, was that the Court of Appeal wrongly construed Clause 3 i) in vacuo detached from the relevant contextual considerations, with the result that (i) it excluded the context of the FAA as a distinctive class of agreement whereby the financial adviser provided financial services to DIO in the venture capital market, which was very different from other superficially comparable agency contracts such as estate agency agreements; (ii) it wrongly leaned towards the “effective cause” requirement; (iii) it failed to consider or analyse[26] relevant “contextual background” matters, including the submission abbreviated as “conditional embargo”, and the line of cases relating to financial advisory agreements, etc. 41. DIO, on the other hand, has emphasised a “textual” approach, and has relied on the points that: (1) commercial common sense should not be invoked by the courts to undervalue the importance of the language of the provision which is to be construed; (2) where the parties have used unambiguous language, the court should apply it, and should be very slow to reject the natural meaning of a provision as correct simply because it appears to be a very imprudent term for one of the parties to have agreed, even ignoring the benefit of wisdom of hindsight. G.2 General approach to interpretation 42. The highest courts in Hong Kong and in England and Wales have returned often in recent years to the principles of the interpretation of contracts.[27] Reported cases deal of course with the difficult cases and it is easy to overlook the fact that the overwhelming majority of contracts are interpreted and performed without difficulty in accordance with their terms. 43. It is a truism that the starting point is the ordinary and natural meaning of the words of the contract, and of course in the vast majority of cases that is the ending point also. But, as Ma CJ pointed out in Fully Profit (Asia) Ltd v Secretary for Justice,[28] in the more difficult cases it is not particularly helpful to refer to the “ordinary and natural meaning” of words because in such cases there can be much debate over exactly what is the ordinary or natural meaning of words; and in those cases the surer guide to interpretation is context. 44. In Wood v Capita Insurance Services Ltd,[29] Lord Hodge JSC reviewed the many cases on interpretation and emphasised that interpretation was a unitary exercise. That is why, where there are conflicting interpretations, account should be taken of the natural and ordinary meaning of the provision in question, the purpose of the contract and of the provision, other relevant provisions, the facts and circumstances known or assumed by the parties at the time that the contract was executed, the quality of the drafting of the instrument, and commercial common sense. 45. The following points emerge from the judgment of Lord Hodge JSC: (a) it does not matter whether the more detailed analysis commences with the factual background and the implications of rival constructions or a close examination of the relevant language in the contract, so long as the court balances the indications given by each; (b) the court must be alive to the possibility that one side may have agreed to something which with hindsight did not serve its interest, or that a provision may be a negotiated compromise or that the negotiators were not able to agree more precise terms; (c) some agreements may be successfully interpreted principally by textual analysis, for example because of their sophistication and complexity and because they have been negotiated and prepared with the assistance of skilled professionals, whereas the correct interpretation of contracts which are marked by informality, brevity or the absence of skilled professional assistance may be achieved by a greater emphasis on the factual matrix; (d) but negotiators of complex formal contracts may often not achieve a logical and coherent text because of, for example, the conflicting aims of the parties, failures of communication, differing drafting practices, or deadlines which require the parties to compromise in order to reach agreement; and (e) commercial common sense and surrounding circumstances should not be used to undervalue the importance of the language of the provision which is to be construed, and the mere fact that a contractual arrangement, if interpreted according to its natural language, has worked out badly for one of the parties, is not a reason for departing from the natural language. 46. In our view, there is no substance to Eminent’s criticism of the Court of Appeal. The Court of Appeal plainly rejected[30] the notion that there was a conflict between a “textual” and a “contextual” approach, specifically relying on Lord Hodge’s judgment in Wood v Capita Insurance Services Ltd.[31] It is true that on one reading of paragraph 7.4, the Court of Appeal may be taken to have suggested that, in the case of professionally drafted documents, the context was relevant only to dealing with inconsistencies or gaps. There is no such limitation, but a fair reading of the judgment as a whole shows that the Court of Appeal was not excluding context in the whole process of interpretation. H. The proper construction of the FAA 47. The relevant provisions of the FAA and Addendum are set out in Section B above. H.1 The FAA viewed overall 48. Eminent’s claim rests on Clause 3 i) which must be construed in the context of the FAA as a whole. It is also to be construed on the undisputed footing that the FAA was entered into in order to provide DIO with international financial advice on fundraising with a view to DIO raising additional capital to expand its overseas business. 49. As the Second Recital declares, the FAA is an agreement appointing Eminent as Financial Advisor to render a range of services and providing for compensation to be paid to Eminent for furnishing the same. In the present context, advice on “Strategic Capital Raising and other Corporate Financing activities” are of primary relevance. 50. The FAA establishes a relationship between Eminent and DIO which is to span 18 months (unless terminated by 30 days’ written notice). It envisages Eminent delving into DIO’s business and performing the various tasks listed in Clause 2 i), being compensated for its work by different types of fees. We are of course focussing on a “Transaction Fee” and the discussion is as to what triggers an obligation to pay the same. However, under the FAA, Eminent is to receive compensation in the form of retainer and equity research fees whether or not a Transaction Fee is ever earned. Other fees potentially payable in stages are provided for by the Addendum. The FAA and the Addendum therefore do not constitute a simple agreement for payment of a commission upon an agent making a successful one-off business introduction. H.2 Entitlement to a Transaction Fee under the FAA during its currency 51. Entitlement to a Transaction Fee during the currency of the FAA is provided for by Clause 2 iv), which links such entitlement to services provided by Eminent of the nature referred to in Clause 2 i)(4), ie, advising DIO “on Mergers & Acquisitions, Fund Raising, private placements or shareholder restructuring prior to any IPO or Secondary Listing”. These are different forms of transaction whereby DIO might raise funds to expand its overseas business, as intended by the parties. A merger or acquisition would thus involve combining with a suitable partner enterprise. Fundraising is a more general concept, involving potential investors putting money into DIO’s business whether as equity or debt. An “IPO or Secondary Listing” and the work preparatory thereto (“private placements or shareholder restructuring prior to” such events) envisage an offering of shares in DIO to the public. Since DIO is already listed on the KOSDAQ, the possibility of a “Secondary Listing” on another stock exchange is contemplated. 52. Clause 2 iv), makes a completed transaction pivotal for entitlement to a Transaction Fee. Thus: (a) The clause is headed “TRANSACTION FEE”. (b) The fee is a “success fee” payable “[u]pon completion of [the relevant transaction]”. (c) The amount of the fee is 3% of the “total transactional amount”. (d) The relevant transaction (which supplies the transactional amount for calculating the 3%) is “any financial transaction related to Fund Raising or Private Placement or Shareholder restructuring, or Mergers & Acquisition for the Company”. The relevant transaction is thus one involving a form of fundraising for DIO, as seen in Clause 2 i)(4). (e) Accordingly, a Transaction Fee is earned during the currency of the FAA upon the completion of a fundraising transaction, calculated as 3% of the amount raised, and paid to Eminent as a success fee reflecting the services it provided towards achieving completion of the transaction. 53. On the foregoing interpretation of Clause 2 iv), to earn a Transaction Fee, Eminent is required not merely to introduce the third party concerned, but to put in work towards achieving the successful completion of the actual fundraising transaction. It is inconsistent with Eminent’s proposed construction that introduction of a third party (as opposed to introduction of a transaction) is all that is required. It is noteworthy that neither the word “introduce” nor any of its derivatives appear in Clause 2 iv) whereas the word “transaction” appears four times. H.3 Entitlement to a Transaction Fee under the Addendum 54. The provisions of the Addendum are relevant and admissible to interpret the FAA because the parties expressly agreed (Fifth Recital of the Addendum) that they were to be read together. 55. The construction which stresses the pivotal nature of the fundraising transaction and its completion is reinforced by the Addendum. Thus, the Third Recital identifies Eminent’s role in “Financial Scenarios” listing additional forms of fundraising and makes it plain that these are “Transaction(s) ... conducted by the Company, its subsidiaries, etc”.[32] And the Fourth Recital refers to “the foregoing Success Fee” and excludes from it all “legal, accounting and audit costs for transactional due diligence and any specific circumstances required by the third parties, by law or by government regulations to retain professional services to complete a proposed transaction”. 56. Paragraph 1a, provides for a success fee payable in relation to a specific potential form of fundraising involving a listing or placement on the London Stock Exchange AIM and, like Clause 2 iv), it conditions entitlement to that fee on “completion of the listing or Secondary Placement”. It lays down as the amount of the fee, 2% “of the transaction value”. While this transaction does not arise on the facts, the structure of this provision, fixing upon completion of the transaction, is a further indication of the contractual intent. 57. Paragraph 1b focusses on “mergers & acquisition transactions involving the Company and a third party” and promises payment to Eminent “upon successful completion” within 14 days of the transaction’s closing date, of a success fee representing 3% of “the total transaction value in cash calculated based on enterprise value of the acquired company including any assumed liabilities.” Payment is again of a “success fee” dependent on Eminent’s contribution to a successfully completed transaction. 58. Paragraph 1d is also important. Added in January 2009 in the period running up to their exploration of Dentsply’s possible interest in investing in DIO, Eminent negotiated an additional “fixed retainer” of USD$450,000 of which USD$200,000 was payable in stages (with instalments to be paid on signing; on approval of a potential M&A “transaction with a third party”; and on acceptance of “the term sheet for the Transaction(s) of any of the Financial Scenario(s)”). The final payment of USD$250,000 is conditional upon “completion of any [sic] the Transaction(s) under the Financial Scenario(s)”. 59. Paragraph 1d evidently caters for the eventuality that the intended interaction with the potential investor might come to nothing so that no success fee would be payable. It provides for compensation to be paid to Eminent in stages as work is done and progress is made. It demonstrates that Eminent is intended to be actively involved in bringing about completion of the transaction and compensated for its services at each stage. Consistently with the FAA’s provisions conditioning entitlement to the Transaction Fee on a completed transaction, paragraph 1d makes entitlement to the final instalment dependent on successful completion of the relevant financial transaction while affording Eminent recompense for work done along the way. There is no provision anywhere in the Addendum attaching a fee entitlement merely to “introduction” of a counterparty. H.4 Entitlement to a Transaction Fee after termination of the FAA 60. Does the position differ regarding entitlement to a Transaction Fee under Clause 3 i)? In our view, the answer is “No”. The situation differs to the extent that it deals with Eminent’s entitlement to fees during the two-year period following termination of the FAA, but the requirement which triggers that entitlement is the same as during the FAA’s currency. Indeed, it would be strange (as the Judge pointed out) if the entitlement to fees were different and in some way less demanding after its termination. 61. Under Clause 3 i), DIO agrees that within the two-year period, fees are payable “should the Company complete a transaction” which is described in terms echoing the description of transactions referred to in the First and Second Recitals and in Clause 2 iv). They are, as we have seen, fundraising transactions.[33] 62. The Transaction Fee payable is calculated on the same basis as during the FAA’s currency, Clause 3 i) providing for Eminent to be paid “its fees according to this Agreement [the FAA] or and [sic] any executed amendments thereof” [including the Addendum]. There is a fall-back provision to apply “a generally accepted market compensation in case the transaction compensation method is not explicitly stated in this agreement”. Thus, one would expect a Transaction Fee applicable to Clause 3 i) to represent 3% of the total transactional amount of the successful fundraising transaction, in accordance with Clause 2 iv) and paragraph 1b of the Addendum. 63. One difference between Clause 3 i) and Clause 2 iv) is that the former refers to an introduction by Eminent. It speaks of DIO completing a transaction “including and not limited to an [sic] secondary listing or fund raising with any third parties or receive funds from a financing source introduced by the Financial Advisor”. Eminent seizes on this and, as noted above, argues that in Clause 3 i), “introduced by the Financial Advisor” qualifies “the ‘counterparty’ (ie a secondary stock exchange, a third party, or a financing source) with whom DIO entered into a deal or from whom DIO received funds, rather than ‘the transaction’”.[34] 64. We do not agree. That argument disregards crucial words in the clause. The words “introduced by the Financial Advisor” qualify “a transaction including and not limited to [a] secondary listing or fund raising with any third parties or receive funds from a financing source”. It is the transaction which is successfully completed that Eminent has to introduce, being a transaction described as including but not limited to a secondary listing or fundraising, etc. If Eminent introduces the third party but then plays no part or an insignificant part in bringing about the fundraising transaction eventually completed, there is no entitlement to a Transaction Fee. This construction dovetails with the pivotal requirement that the services rendered by Eminent must contribute to completion of the relevant fundraising transaction. 65. Moreover, as a matter of language, when Clause 3 i) speaks of “[a] secondary listing ... introduced by the Financial Advisor”, it is not referring to Eminent introducing a counterparty. A secondary listing is a transaction – referred to as a “Financial Scenario” in the Addendum, not a person. This is reflected in the Addendum’s Third Recital which includes a secondary listing as one of the “Financial Scenarios” along with “backdoor listings, Public Placements, Private Placements, Mergers & Acquisitions (‘M&A’), Management Buy-outs (‘MBO’), Leverage [sic] Buy-outs (‘LBO’)”. H.5 The function of Clause 3 i) 66. There is some suggestion by Eminent that making a Transaction Fee entitlement dependent on its having introduced DIO to the successfully completed transaction as opposed merely to a third party would somehow render the tail gunner clause “meaningless”.[35] We do not agree. 67. Clause 3 i) guards against Eminent being unfairly deprived of a Transaction Fee which it has substantially earned. Since a completed transaction is the necessary trigger for the fee entitlement, a situation could arise where Eminent has done everything necessary to bring a fundraising deal to fruition but where, in the absence of Clause 3 i), it might be excluded from recompense by DIO deferring completion until after the FAA is terminated. Thus, Clause 3 i) provides the financial adviser with a safeguard, ensuring that the fee is payable “should the Company complete [the relevant] transaction ... introduced by [Eminent]” during the two-year post-termination period. Commercially, it is unlikely that a fundraising deal would be delayed by more than two years merely as an expedient for avoiding payment of the Transaction Fee. H.6 Irrelevance of Clause 4(a). 68. Eminent argues that Clause 4(a) supports its construction because it stipulates that Eminent “intends to introduce [DIO] to its contacts, including private and institutional financing sources or strategic alliances” and DIO agrees to respect Eminent’s relationships with such investors and sources of financing and to refrain from “the circumvention of any obligation to [Eminent] created by [the FAA] or any means”. 69. This adds nothing to the foregoing analysis. The relevant obligation which the FAA creates is for DIO to pay Eminent the Transaction Fee where it has introduced a successfully completed transaction. If that obligation has not been triggered, there is no question of circumvention. H.7 No presumption or implied term 70. Reliance has been placed in argument and in some of the decided cases on Article 57 in Bowstead & Reynolds on Agency[36] which states: “Subject to any special terms or other indications in the contract of agency, where the remuneration of an agent is a commission on a transaction to be brought about, he is not entitled to such commission unless his services were the effective cause of the transaction being brought about.” 71. However, as the commentary in Bowstead & Reynolds makes clear, the statement in Article 57 is no more than a particular example of the wider principle stated in the preceding Article,[37] that where an agent is entitled to remuneration upon the happening of a future event, the entitlement does not arise until that event has occurred; and the event upon which the agent's entitlement to remuneration arises is to be ascertained from the terms of the agency contract. Everything depends on the contract’s construction and it is inappropriate to regard Article 57 as stating a substantive legal rule as to the existence of either a presumption or an implied term in favour of an “effective cause” requirement. 72. It follows that there is no special approach to the construction of contractual terms governing post-termination payments to financial advisers. All depends on the application of the established rules on construction of contracts to the particular case. H.8 The authorities cited by the parties 73. The parties referred the court to many authorities (several involving the use of the term “introduction” or “introduce”) on the question whether it is necessary for the acts of an agent or adviser to be the effective cause of the contemplated transaction before the agent or adviser is entitled to a fee. 74. The basic principles emerged in connection with the activities of estate agents. In Millar, Son & Co v Radford,[38] Lord Collins MR said: “It was, therefore, important to point out that the right to commission did not arise out of the mere fact that agents had introduced a tenant or a purchaser. It was not sufficient to show that the introduction was a causa sine qua non. It was necessary to show that the introduction was an efficient cause in bringing about the letting or the sale.” That was because “[t]he owner is offering to the agent a reward if the agent’s activity helps to bring about an actual sale”: Luxor (Eastbourne), Limited v Cooper.[39] 75. In Foxtons Ltd v Pelkey Bicknell,[40] Mrs Bicknell, the vendor, agreed to terms with Foxtons which provided for a commission of 2.25% of the sale price achieved, on the basis of sole agency, payable if at any time unconditional contracts were exchanged with “a purchaser introduced by us during the period of our sole agency or with whom we have had negotiations about the property during that period; or with a purchaser introduced by or offering via another agent during that period.” Foxtons showed the house to a Mr Low, who was looking for a house on behalf of his former wife. He initially liked it but then lost interest when his former wife viewed it. The sole agency came to an end and Foxtons and Mrs Bicknell agreed a multiple agency agreement at a rate of 3%. Mrs Bicknell also appointed Hamptons on a multiple agency basis. Mr and Mrs Low visited the property some months later and Mrs Low purchased it. Mrs Bicknell paid Hamptons their commission. Foxtons then claimed theirs on the basis that contracts were exchanged with “a purchaser introduced by” them. The Court of Appeal decided that Foxtons were not entitled to the commission because, among other reasons (per Lord Neuberger of Abbotsbury): (1) the court will readily imply a term that the agent is not entitled to commission on a contract unless his services were the effective cause of the transaction being brought about, especially in the context of a residential consumer, unless the provisions of the particular contract or the facts of the case negative it; (2) the main reason for the implication of such a term is to minimise the risk of the seller having to pay two commissions; (3) the expression “a purchaser” in the phrase “a purchaser introduced by us” did not mean “a person who at some time in the future becomes a purchaser” regardless of whether his purchase or the interest which gave rise to it owed anything to Foxtons, but it meant “a person who becomes a purchaser as a result of our introduction”; and (4) the latter interpretation accorded better with the principles to be extracted from the cases, in particular the normal notion that an agent can only recover if he introduces someone who becomes a purchaser as a result of his introduction, and such an interpretation would also make it unlikely that the client would be liable for two commissions. 76. But the authorities make it clear that, although they lean towards the application of the “effective cause” principle either by interpretation of the express terms or by implication of a term to that effect, the principle will be displaced by language which points in a different direction: eg Brian Cooper & Co v Fairview Estates (Investments) Limited,[41] (“We confirm that we are pleased to offer a full scale letting fee to your company should you introduce a tenant by whom you are unable to be retained, and with whom we have not been in previous communication and who subsequently completes a lease”): “I can see no necessity in this case to imply a term. On the contrary, I regard the relevant language as being inconsistent with implication of a term imposing an additional implied requirement that the estate agent must be at least an effective cause of the lease being granted” (per Woolf LJ (as he then was)); The County Homesearch Co (Thames & Chilterns) Ltd v Cowham,[42] (deeming provision to the effect that a property was “introduced” if the defendant had either received particulars of the property from County Homesearch, directly or indirectly, or from any estate agent with which County Homesearch had regular contact): “express terms of the contract are inconsistent with any implied requirement that the agent be an effective cause of the transaction.” (per Longmore LJ). 77. By contrast, in MSM Consulting Limited v United Republic of Tanzania,[43] MSM was retained by Tanzania to search for new premises for its High Commission. A fee was “payable by the Client upon completion of the purchase of a property which we have introduced to the client or representative of the Client.” Christopher Clarke J applied Foxtons and held that MSM was not entitled to the fee because it had not been the effective cause of the transaction: the implication of an “effective cause” term arises as a result of the nature of the transaction; and where an agent is engaged upon terms that his commission is not payable unless a contract is concluded between his client and a purchaser and the commission is to be a percentage of the price, the nature of the transaction and the requirements of business efficiency will usually (subject to any special terms or other indications in the contract of agency) dictate that the agent should be remunerated if, but not unless, it is his efforts that have brought about the transaction in question. 78. In cases in which the transactions are closer to the present case, it has been held that the terms of the contract were inconsistent with an interpretation of the language, or the use of an implied term, to require the agent or adviser to have been the effective cause of the transaction. Thus in Watersheds v Christopher Simms,[44] a businessman who had entered into a seven-year contract with a company providing corporate services to help him finance, purchase, develop and then sell a business was liable to pay the company its agreed percentage of the sale proceeds even though the company had not done any work in connection with the sale. The contract was not analogous to an estate agent’s contract with a seller and it would be inconsistent with the purpose of the contract to construe it as containing an implied term allowing the company to be paid only if its work had been an effective cause of the sale. 79. In two cases involving a “tail gunner clause”, an adviser was held entitled to fees on transactions which were completed after termination of the advisory contract and which the adviser did not bring about. 80. In Seymour Pierce Limited v Grandtop International Holdings Limited,[45] a financial adviser gave advice to a company pursuant to a letter of engagement in connection with the acquisition of a football club and claimed a success fee. The contract provided that “in the event the engagement pursuant to this letter of engagement is terminated by the Company and an Offer for the Target is declared or becomes wholly unconditional as the result of any offer made by or in association with the Company within a period of 12 months after the effective date of termination the Company shall pay to Seymour Pierce the Success Fee in full.” One of the defences to the claim was that the adviser should not be able to recover the fee because it was not involved in the ultimate acquisition. Eady J held that there was no need to imply an effective cause term as a matter of business efficacy. The clause was entirely comprehensible without any such implication. 81. So also in Edmond de Rothschild Securities (UK) Ltd v Exillon Energy plc,[46] the defendant agreed that “[Rothschild] has the exclusive right to act as its financial advisor in connection with the Transaction. If, notwithstanding the termination of this agreement, the Company completes the Transaction within a period of eighteen months from the date of such termination, the appropriate fee ... shall be payable to [Rothschild].” It was held that a term that the success fee was payable in certain events was clear and that there was no need to imply a term. The natural meaning was supported by the consideration (inter alia) that it would generally be difficult for Rothschild to prove that its work constituted an effective cause of a sale. 82. On the other hand in Cavendish Corporate Finance LLP v KIMS Property Co Ltd,[47] (“In the event of a successful fundraising from the Cavendish exercise, a fee of 3.5 per cent of new monies raised (‘success fee’) from the investor ...”) the judge interpreted the expression to mean that the success fee was payable only if Cavendish’s work was an effective cause of the deal that was done. 83. Both parties relied on United States cases. Eminent cited five cases from the Southern District of New York (applying New York law) in which major investment banks sued successfully for fees notwithstanding that the client alleged that the financing envisaged under the agreement was not obtained through the investment bank. 84. In each of these cases, the court found that the language of the financing agreement clearly had the result that the investment bank was entitled to payment: PaineWebber Inc v Campeau Corp,[48] (“... there is nothing ambiguous .... The bonus payment ... is conditioned upon the occurrence of a well-defined event, that is, an Acquisition of Allied. There is nothing on the face of the agreement that suggests PaineWebber would have to perform additional services in order to receive the bonus payment of $5.75 million”); Chase Manhattan Bank NA v Remington Products Inc,[49] (“... Engagement Agreement provided that if a Transaction (or an agreement leading to a Transaction) was in place by the end of a one year ‘tail period’ …, Chase would receive a fee under the same terms and in the same manner as during the term of the Agreement ... Nor does [the Agreement] condition Chase’s receipt of a fee on its bringing about a Transaction or on the consummation of any particular type of Transaction. The sole prerequisite to payment listed ... is the consummation of a Transaction (or the signing of an agreement that results in a Transaction)”; Lazard Freres & Co v Crown Sterling Management[50] (“The Agreement specifically provided that Lazard would be entitled to its fees upon the occurrence of certain events, ‘regardless of whether [Lazard] is responsible for arranging said events.’”); CIBC World Markets Corp v TechTrader, Inc,[51] (“This clause creates an unambiguous duty on the part of TT to pay the fee if the stipulated Transaction occurs”). 85. In the most recent of the decisions cited by Eminent, Deutsche Bank Securities Inc v Rhodes,[52] (“During the Exclusivity Period, the Company agrees that DBSI shall have the right, but not the obligation, to act as sole book-running manager, and DBSI has the right but not the obligation, to act as exclusive underwriter, initial purchaser, placement agent or syndication agent in connection with any bank or loan financing consisting of institutional and/or pro rata loans of, the Company or its subsidiaries during the term of this Agreement on a best efforts basis”) the District Court referred to those four cases, and said (at 668-669): “... it is not uncommon in the investment banking field for investment banks to be contractually entitled to fees even when they do not arrange or facilitate the transactions. … The cases are different, as they involve different contracts and different language. But the concepts are the same. In these cases, the language of the agreements and the totality of the circumstances showed unambiguously that the investment banks were entitled to fees, under exclusivity provisions, even if the clients obtained financing through others. Comparable language and circumstances exist here.” 86. DIO does not accept that financial advisers are a well-established and specialised category of service providers, and says that in the United States authorities the clauses were wholly different from Clause 3 i), are not authority for any general proposition of industry practice of financial advisers in general, or in Hong Kong in particular. 87. Instead, DIO relies on two cases from the Third Circuit Court of Appeals (applying New Jersey law) in which it was re-affirmed that under New Jersey law there is a presumption that when a brokerage contract is silent as to the service required to earn a commission from a seller or buyer, a commission will be earned only if the broker was the “efficient producing cause” of the sale: Inventive Music Ltd v Cohen[53] (agent for sale of the defendant’s record company); Vanguard Telecommunications, Inc. v Southern New England Tel Co.[54] In the latter case (agent for the sale of the defendant’s fibre optic systems), the court said (at 651): New Jersey common law ... presumes that when a brokerage contract is silent as to the service required to earn a commission from a seller or buyer, a commission will be earned only if the broker was the “efficient producing cause” of the sale ... This presumption is based on public policy intended to effectuate justice between the parties and is not intended to rewrite an agreement which the parties deliberately executed ... To overcome this presumption, an agreement must contain language which explicitly negatives the presumption. The contract language presented must be “without qualification, and … emphatic and specific in statement” [citing an earlier case] ... Absent language which specifically abrogates the common law presumption, that presumption prevails; commissions will only be awarded to a broker who was the efficient cause of the sale. 88. The decisions in the Southern District of New York are in line with the English authorities in making it clear that the question is solely one of interpretation. It is plain from the language of presumption and the invocation of public policy that the principle applied in New Jersey does not represent the law in Hong Kong. 89. Outside the sphere of estate agents, where (especially in the residential consumer context) there is a common understanding of the agent’s duties and the consequences of the absence of a presumption or of an implied term, there is little to be said for a presumption, and nothing to be said for an implied term. All will depend on the construction of the term in question in the context of the agreement as a whole and the purpose of the transaction. H.9 Grey areas and an “effective cause” 90. A requirement that the financial adviser’s contribution be an “effective cause” of the transaction may possibly arise in a different but related context. Where parties agree to a “tail gunner clause”, they inevitably run the risk of creating grey areas which may give rise to controversy. Thus, the financial adviser may have gone beyond merely introducing the client to the potential investor and done a significant amount of work in promoting the deal which is eventually completed after termination of the FAA but within the run-off period. The client (or a replacement financial adviser) might however have taken up the running and made substantial contributions essential to achieving that completion. It may thus be arguable whether, on particular facts, the financial adviser has done enough to be entitled to the fee. Unless the agreement deals explicitly with that question, the issue would arise as to what the proper criterion is for determining the answer. 91. This issue does not presently arise since, on the Judge’s findings, Eminent did not introduce or contribute anything at all to the transaction entered into between DIO and Dentsply and announced on 9 December 2010. We simply mention that where such a grey area does arise, a possible approach might be to construe the contract as recognising an entitlement to a success fee where the financial adviser’s advice and assistance is found to be an “effective cause” of the transaction’s completion, a criterion adopted in some of the decided cases. Once again, all will depend on the interpretation of the agreement. I. Conclusion 92. For the foregoing reasons, in agreement with the Court of Appeal, we consider the meaning of Clause 3 i) on its true construction to be clear. There is no need to consider implication of any term into the FAA for present purposes. Eminent’s entitlement to the Transaction Fee claimed requires it to have introduced DIO to the completed transaction which was announced on 9 December 2010. It was insufficient for Eminent simply to have introduced DIO to Dentsply in April 2009. 93. There was accordingly no entitlement to payment under Clause 3 i) and we therefore dismissed this Appeal with costs. Mr Justice Fok PJ: 94. I agree with the joint Reasons for Judgment of Mr Justice Ribeiro PJ and Lord Collins of Mapesbury NPJ. Mr Justice Cheung PJ: 95. I agree with the joint Reasons for Judgment of Mr Justice Ribeiro PJ and Lord Collins of Mapesbury NPJ. Chief Justice Ma: 96. For the above-mentioned Reasons, the appeal was unanimously dismissed with costs. Mr Wong Yan Lung SC and Mr Jenkin Suen SC, instructed by Tang & So, for the Plaintiff (Appellant) Mr Charles Sussex SC and Mr Richard Zimmern, instructed by DLA Piper Hong Kong, for the Defendant (Respondent) [1] Korean Securities Dealers Automated Quotations board of the Korea Exchange, the Korean equivalent of the NASDAQ, the National Association of Securities Dealers Automated Quotations exchange, in the US. [2] HCA 1292/2011 (23 September 2016) at §229. [3] Ibid §50. Pleaded guilty on 28 April 2011. [4] Ibid §256. [5] Ibid §§253-254. [6] Ibid §33. [7] Ibid §257. [8] Clause 1). [9] The expression is used in financial circles in relation to fees for transactions which complete after the termination of an adviser’s engagement: see Kilometre Capital Management Cayman v Shanda Games Limited HCA 2009/2014 (13 July 2015), §9. [10] Paragraph 1 of the Addendum makes it clear that its sub-paragraphs, including a, b and d, are “an addendum to the existing ‘ENGAGEMENT & FEE STRUCTURE’, Section 2 on page two of the [FAA]”. [11] Appearing with Mr Jenkin Suen SC. [12] Eminent Written Case §58. [13] Ibid §66(6). [14] Ibid §71. [15] Ibid §13. [16] Ibid §§34-38. [17] Ibid §§40-41. [18] Appearing with Mr Richard Zimmern. [19] DIO Written Case §61. [20] Ibid. Article 57 of Peter Watts and F M B Reynolds, Bowstead & Reynolds on Agency (2018) 21st Ed, §7-027 is cited in support of the “general rule”, discussed below in Section H.7 of these Reasons. [21] CFI §257. [22] DIO Written Case §§71-78. [23] Summarised at CFI §§212-218. [24] Kwan VP, Cheung and Au JJA [2019] HKCA 606 at §§7.5-7.10. [25] CA§§7.1-7.4. [26] Ibid §§7.8-7.9. [27] Including Prenn v Simmonds [1971] 1 WLR 1381; Investors Compensation Scheme Ltd v West Bromwich Building Society [1998] 1 WLR 896; Jumbo King Ltd v Faithful Properties Ltd (1999) 2 HKCFAR 279; Bank of Credit and Commerce International SA v Ali [2002] 1 AC 251; River Trade Terminal Co Ltd v Secretary for Justice (2005) 8 HKCFAR 95; Re Sigma Finance Corp (in administrative receivership) [2009] UKSC 2, [2010] 1 All ER 571; Rainy Sky SA v Kookmin Bank [2011] UKSC 50, [2011] 1 WLR 2900; New World Harbourview Hotel Co Ltd v ACE Insurance Ltd (2012) 15 HKCFAR 120; Fully Profit (Asia) Ltd v Secretary for Justice (2013) 16 HKCFAR 351; Sinoearn International Ltd v Hyundai-CCECC Joint Venture (2013) 16 HKCFAR 632; Arnold v Britton [2015] UKSC 36, [2015] AC 1619; Wood v Capita Insurance Services Ltd [2017] UKSC 24, [2017] AC 1173. [28] (2013) 16 HKCFAR 351 at §15. [29] [2017] AC 1173. [30] CA§7.4. [31] [2017] AC 1173 at §13. [32] “... including but not limited to Secondary Listings in an overseas stock market, backdoor listings, Public Placements, Private Placements, Mergers & Acquisitions (‘M&A’), Management Buy-outs (‘MBO’), Leverage [sic] Buy-outs (‘LBO’) in one or series of Transaction(s) in any form including and not limited to loans, bonds, common stock, preferred stock, convertible securities or equity linked debt instruments of any kind conducted by the Company and its subsidiaries and affiliates directly or from any registered shareholder(s) of the Company.” [33] “... [A] transaction including and not limited to an [sic] secondary listing or fund raising with any third parties or receive funds from a financing source introduced by the Financial Advisor”. [34] Eminent Written Case §71. [35] Eminent Written Case §66(5). [36] Peter Watts and F M B Reynolds, (2018) 21st Ed, §7-027. [37] Ibid,Article 56, §7-013. [38] (1903) 19 TLR 575. [39] [1941] AC 108 (HL), 117 (Viscount Simon LC). [40] [2008] EWCA Civ 419, [2008] 2 EGLR 23 (CA). [41] [1987] 1 EGLR 18 (CA). [42] [2008] EWCA Civ 26, [2008] 1 WLR 909 (CA). [43] [2009] EWHC 121 (QB) (Christopher Clarke J). [44] [2009] EWHC 713 (QB) (Burnett J). [45] [2010] EWHC 676 (QB) (Eady J). [46] [2014] EWHC 2165 (Comm) (Males J). [47] [2014] EWHC 1282 (Ch) (Mark Anderson QC). [48] 670 F Supp 100 (SDNY 1987), at 105. [49] 865 F Supp 194 (SDNY 1994), at 197 and 199. [50] 901 F Supp 133 (SDNY 1995), at 137. [51] 183 F Supp 2d 605 (SDNY 2001), at 611. [52] 578 F Supp 2d 652 (SDNY 2008), at 657. [53] 617 F 2d 29, 32 (3d Cir 1980). [54] 900 F 2d 645 (3d Cir 1990). 
Chief Justice Ma: A. INTRODUCTION 1. After hearing submissions, the appeal was allowed and it was ordered that the appellant be released from custody. These are the Reasons for allowing the appeal. 2. The issue before the Court was whether it was in the interests of justice to order a second retrial for the appellant in relation to a charge of trafficking in dangerous drugs.[1] In the present case, after allowing her appeal, the Court of Appeal had made an order for a second retrial.[2] 3. Unusually, the Appeal Committee of the Court of Final Appeal[3] granted leave to appeal on the substantial and grave injustice limb[4] to enable the appellant to argue against the order for a second retrial. I say “unusually” because the discretion whether or not to order a retrial after the allowing of an appeal in criminal proceedings is usually best left to the Court of Appeal to determine based on the interests of justice.[5] 4. Leave was granted on the basis that it was reasonably arguable the Court of Appeal had been wrong in the exercise of its discretion by not sufficiently taking into account “the undue hardship and unfairness allegedly occasioned to the [appellant] by the second retrial order, particularly given her psychiatric condition and the further deterioration of her condition said to have resulted from the order”.[6] In support of this ground, leave was given by the Appeal Committee to adduce a medical report of the appellant’s mental condition, as well as to prepare a joint expert report on the appellant’s current psychiatric condition and also on her fitness to stand trial for a third time. The appellant’s mental condition had been raised in the Court of Appeal and there was medical evidence placed before the court. Essentially, it was said that she was suffering from “adjustment disorder”. The appellant was diagnosed with Major Depressive Disorder (MDD) after the hearing before the Court of Appeal. I shall go more into this aspect later. A.1 Facts and procedural history 5. The facts pertaining to the trafficking charge are relatively simple. On 15 November 2012, the appellant arrived in Hong Kong from Kuala Lumpur. At Hong Kong International Airport, her suitcase was inspected. After unzipping the lining of the suitcase, a customs officer found 2 packets containing 1.79 kilogrammes of a mixture having 0.8 kilogrammes of heroin hydrochloride. The appellant was accordingly charged with the trafficking offence. 6. The appellant was first tried before a jury in February 2014. After a 6-day trial, she was convicted on a 5:2 majority verdict and sentenced to a term of 21 years’ imprisonment. In February 2015, her first appeal to the Court of Appeal was dismissed. However, she was given leave to appeal to the Court of Final Appeal in May 2016 and in February 2017,[7] her appeal was allowed on the basis that inadequate directions had been given to the jury by the trial judge in relation to an equivocal statement that had been made by the appellant at the airport, this statement being alleged by the prosecution to constitute an admission of guilt. Her conviction was accordingly quashed. A retrial was ordered. 7. The second trial took place in July 2017, this time lasting some 14 days. The appellant was again convicted by another majority verdict (6:1). In August 2017, she was sentenced to a term of 21 years’ imprisonment. In August 2018, she appealed to the Court of Appeal. In November 2018, the Court of Appeal allowed her appeal[8] and quashed the conviction on the basis of another misdirection by the trial judge to the jury, this time in relation to the giving of adequate warnings to the jury, in particular as to whether lies allegedly told by the appellant or acts of concealment on her part could be relevant to proving the trafficking offence. As stated earlier,[9] in December 2018 the Court of Appeal ordered a second retrial and in November 2019, the Appeal Committee granted leave to the appellant to appeal to this Court. 8. Throughout this time, since November 2012, the appellant had remained in custody. Up to the hearing of the present appeal, she had therefore been in custody for about 7½ years. B. RETRIALS – THE RELEVANT PRINCIPLES 9. Section 83E of the Criminal Procedure Ordinance[10] states that the Court of Appeal, if it decides to allow an appeal, may order a retrial if “the interests of justice” so require. The jurisdiction to order a retrial is thus put in the widest possible terms. In HKSAR v Tam Ho Nam (No. 2),[11] Fok PJ restated the relevant principles regarding retrials by reference to Au Pui Kuen v Attorney‑General of Hong Kong,[12] Ting James Henry v HKSAR[13] and Kissel v HKSAR.[14] The following principles, distilled from the above and other cases, are relevant in the present case:- (1) Whether or not a retrial should be ordered is a matter of discretion. This discretion is usually exercised, as it should be, by the Court of Appeal, relying on their “collective sense of justice and common sense.”[15] And, as was put by Lord Bingham of Cornhill, there must be “an informed and dispassionate assessment of how the interests of justice in the widest sense are best served”; it is important to maintain “confidence in the efficacy of the criminal justice system.”[16] (2) The discretion whether or not to order a retrial depends entirely on what justice requires (this being the “critical question”[17]). (3) The interests of justice of course include a consideration of an accused’s interests and circumstances. The criminal justice system is there to bring matters to a conclusion without undue delay and without oppression; these are “accepted norms”.[18] It should be acknowledged that any criminal trial is to some degree an ordeal for the accused.[19] The interests of justice also include the interest of the public in seeing those who are guilty of serious crimes brought to justice and not escape merely because of a technical error in the conduct of a trial or in the summing up to a jury.[20] In Au Pui Kuen,[21] Lord Diplock referred to the following passage from the judgment of Gould Ag CJ in Ng Yuk-kin v The Crown:[22] that there may be cases where it “is in the interest of the public, the complainant, and the appellant himself that the question of guilt or otherwise be determined finally by the verdict of a jury, and not left as something which must remain undecided by reason of a defect in legal machinery”. In assessing the public interest, a court must take into account the views of the prosecution which is best qualified (and I would add has the duty) to present the views of the public[23], although it must ultimately be for the court to determine what is in the public interest. The strength of the prosecution case is also a relevant consideration. (4) The interests of justice require all relevant factors, both for and against a retrial, to be taken into account. Such factors will not only vary from case to case, but their relative importance and weight will also be different in any given case.[24] (5) The above said, one factor that must be given significant weight is the fact that the accused has already undergone a trial, in particular where the trial is long and complex.[25] This is all the more so when there is involved a second retrial, which means of course the possibility of a third trial for the same offence. In Mok Kin Kau,[26] the ordering of a second retrial after 2 concluded trials and appeals, and the serving of the whole sentence, was said to be an “unusual course” and in such a situation, in the absence of a special or compelling reason, this was a “departure from accepted norms” sufficient to constitute a substantial and grave injustice. Although it is not unprecedented for a second retrial to be ordered, the cases accept that this is an “unusual” course to take,[27] even where the accused has not served the whole of his or her sentence as was the case in Mok Kin Kau. Given that the ordering of a second retrial is an unusual course, a court would have to be persuaded by cogent and compelling reasons to make such an order. This is consistent with the interests of justice, but of course all relevant factors must be carefully weighed in this exercise of discretion. (6) Another factor that should also be taken into account is the time that an accused has spent in custody and in relation to this facet, the time that an accused has been in custody will have to be seen against the likely sentence that he or she might receive on a retrial.[28] 10. In common with the review of discretion in other areas, this Court will not disturb the exercise of discretion by the Court of Appeal in the determination whether or not to order a retrial unless there has been a serious error of law or principle or approach, such as where the court below has failed to take into account a relevant consideration or has taken into account an irrelevant one. Only in such a situation would the Court of Final Appeal be justified in exercising the discretion afresh. Furthermore, it is also important to emphasise that the relative weight which the Court of Appeal “ascribes to each relevant factor” is a matter within its discretion and it is not open to the Court of Final Appeal to seek to ascribe a different weight.[29] 11. I now deal with the application of these principles in the present case. C. APPLICATION OF PRINCIPLES 12. As mentioned earlier, the appellant’s mental condition was raised in the Court of Appeal. A Psychiatric Report from Dr Dorothy Tang[30] diagnosed the appellant to be suffering from adjustment reaction and situation reaction, although at the time of observation her mental condition was stable. However, as part of the psychiatric history, it was noted that her condition worsened when she had setbacks in legal proceedings. It was also noted that she harboured “fleeting suicidal ideation but she denied any concrete plan”. Reference was made to an occasion in December 2015 when she drank 2-3 mouthfuls of shampoo. An Incident Report Form detailed the incident. A letter dated 26 November 2018 from the Correctional Services Department to the appellant’s solicitors was also placed before the Court. The letter provided details of another incident (on 1 November 2018) when the appellant wounded herself with a comb. When seen by a clinical psychologist, she “admitted active self‑harm thought and counselling was provided to her”. 13. In ordering a retrial, the Court of Appeal made no reference at all to the evidence relating to her mental condition, merely stating: “The fact that the applicant has to face a third trial and that she has been remanded in custody for over 6 years, all of which might be causing her frustration but is insufficient to persuade this Court not to order the case for retrial.”[31] This was, with respect, at best a non‑appreciation and understatement of the appellant’s mental condition, particularly in the context of the lengthy time she had already been in custody and the fact she was facing a third trial. It was for this reason that leave to appeal was given that it was reasonably arguable a substantial and grave injustice had been done in making the order for a second retrial without adequate consideration of the hardship factor.[32] 14. In my view, the Court of Appeal was in serious error in its treatment (or non‑treatment) of the relevance of the appellant’s mental condition. It was a factor that had to be taken into account, particularly in the context of the time she had been in custody. If it had been properly taken into account, the Appeal Committee would not have been minded to grant leave to appeal. As it was not, this Court was accordingly free to consider afresh the exercise of the discretion whether or not to order a retrial. 15. The starting point in the present case was, since the prosecution was seeking a second retrial, that cogent and compelling reasons had to be advanced by it to justify this order. Furthermore, as stated earlier, up to the time her appeal was heard by this Court, the appellant had been in custody for 7½ years. It was not suggested that any part of this period was attributable to any fault on the part of the appellant. Even taking as optimistic a view as possible of the date for a retrial if one had been ordered (taking further into consideration the fact that a retrial may last 14 days[33]), it was likely that before the second retrial took place, the appellant would have been in custody for approximately 8 years without her guilt having been established. After both her previous trials, the appellant was sentenced to 21 years’ imprisonment. On the assumption that if she were convicted again and sentenced to the same term, after taking into account the allowance for good behaviour, the appellant would already have been in custody 8 years out of 14. This is over half of a long sentence, served even before she has been convicted and before the ordeal of yet another trial. This is the relevant context in which to evaluate the other relevant factors in the exercise of discretion in the present case. 16. Ms Maggie Yang for the respondent[34] did not dispute the relevant principles regarding retrials but emphasised two points in support of the order for a second retrial. First, she submitted it was in the public interest for the appellant to have to face trial for a serious offence. This is right: trafficking in dangerous drugs is without doubt a serious offence. The Court of Appeal said that the charge was “extremely serious”.[35] Secondly, Ms Yang also contended that the evidence against the appellant was strong. The Court of Appeal said that “the prosecution evidence supporting the charge is strong and compelling”.[36] This was correct inasmuch as the facts relied on by the prosecution are concerned: there was little dispute as to what occurred at the airport when the appellant was found in possession of a suitcase containing dangerous drugs. However, Ms Yang went further to enumerate a number of points to demonstrate that the appellant’s version of events – essentially to show that she had no knowledge of the fact that she was in possession of dangerous drugs – was incredible.[37] While I accept there is some force in these points, they cannot be taken too far. In relation to most, if not all, of these points, the Court of Appeal was of the view that they were open to question. It is also to be observed that the verdicts in the previous two trials were majority ones.[38] 17. I now turn to the medical evidence regarding the appellant’s mental condition. In granting leave to appeal, the Appeal Committee had given leave to adduce a medical report of Dr Tsang Fan Kwong as well as directed that a joint expert report be prepared on the appellant’s psychiatric condition to be considered in the present appeal. A Joint Psychiatric Report dated 7 January 2020 was prepared by Dr Tsang (the appellant’s expert) and Dr Tang (the respondent’s expert) and this was before us. As we shall see, the relevance of this evidence lay in putting into perspective the lengthy period the appellant had already been in custody. 18. There is a considerable amount of material before us including the said report of Dr Tsang and the Joint Psychiatric Report of Dr Tsang and Dr Tang. It is not necessary to deal with this material in detail but I would highlight the following:- (1) Although Dr Tsang and Dr Tang differed in their opinions on a number of matters, they were to varying degrees agreed that the appellant suffered from “[MDD], single episode, moderate to severe with psychotic‑symptoms”. This diagnosis followed the criteria laid down by the Diagnostic and Statistical Manual of Mental Disorder (5th edition) published by the American Psychiatric Association.[39] It is right to point out that Dr Tang was of the view that the appellant’s depression may not have been as severe as Dr Tsang diagnosed, that she may be in “partial remission” and that she appeared to be responding to medication she had not received prior to seeing the two doctors, but Dr Tang did agree with the diagnosis of MDD, albeit perhaps on a moderate scale. Dr Tang also said this: “[the appellant] is impressed to have inadequate stress coping skill, as evidenced [by] the discrepancy between her stable mental condition during the day in [Siu Lam Psychiatric Centre], and anxious‑look with slow response when she thought about the appeal and court hearing in clinical interviews”. (2) Notwithstanding the above diagnosis of MDD, both doctors were of the view that the appellant was mentally fit to stand trial. As was said in the Joint Psychiatric Report, Dr Tsang stated that “She understands the nature of the alleged offence. She is able to instruct a lawyer. She is able to follow court proceedings. She knows the difference between pleading guilty or not.” (3) That said, it is nevertheless clear from the Joint Report that the appellant was significantly stressed when reminded of the ongoing court proceedings. Dr Tsang was of the view that due to her depression, “her energy, confidence, attention, concentration and memory capacity were substantially impaired” and that it would “be a very difficult task for her to defend herself in a third trial”. Dr Tang was less pessimistic about the appellant’s ability to defend herself in a trial, but was nevertheless of the view: “Court hearing is an undoubtedly stressful experience, thus stress coping strategies and relaxation exercise would be continuously discussed with [the appellant].” See also her view as quoted in sub‑paragraph (1) above. 19. In view of the lack of agreement between the doctors as to whether or not the appellant would actually be disadvantaged in defending herself in another trial, Ms Yang submitted that this issue should not be determined by this Court and should instead be left to be decided by the Court of First Instance in an application for a stay of proceedings (a retrial being ordered in the meantime). Mr Khattak on the other hand argued that the issue should be determined by this Court. 20. I agree that the issue of whether the appellant would be unable properly to defend herself in another trial cannot be determined by this Court without hearing evidence from the doctors. However, the relevance of the medical evidence lay not so much in the resolution of this issue; rather, the relevance of the appellant’s medical condition lay instead in putting into proper perspective and context the lengthy period that the appellant had already been in custody in the present cases. 21. Taking all the above considerations into account, I am of the view that it was clearly in the interests of justice not to order a second retrial for the appellant. Perhaps individually, each factor may not necessarily be sufficient but cumulatively it was compelling to call a halt to the proceedings in the present case. Insufficiently cogent or compelling reasons existed to justify an order for a second retrial. D. CONCLUSION 22. For the above reasons, the appeal was allowed and the order for a retrial quashed, with the consequence that it was ordered the appellant should be released. Mr Justice Ribeiro PJ: 23. I agree with the judgment of Chief Justice Ma. Mr Justice Fok PJ: 24. I agree with the judgment of Chief Justice Ma. Mr Justice Cheung PJ: 25. I agree with the judgment of Chief Justice Ma. Mr Justice Stock NPJ: 26. I agree with the judgment of Chief Justice Ma. Mr Shahmim K Khattak, Mr Anthony Lai and Ms Holly Cheng, instructed by Annie Leung & Co., assigned by the Director of Legal Aid, for the Appellant Ms Maggie Yang, DDPP and Mr Franco Kuan, SPP, of the Department of Justice, for the Respondent [1] Contrary to sections 4(1)(a) and (3) of the Dangerous Drugs Ordinance Cap 134. [2] Yeung Ag CJHC, Poon and Pang JJA in a Determination dated 21 December 2018 [2018] HKCA 957 (the CA Determination). [3] Cheung Ag CJ, Ribeiro PJ and Fok PJ in a Determination dated 14 November 2019 [2019] HKCFA 41 (the AC Determination). [4] Section 32(2) of the Hong Kong Court of Final Appeal Ordinance Cap 484. [5] Section 83E of the Criminal Procedure Ordinance Cap 221 provides:- “83E. Power to order retrial (1) Where the Court of Appeal allows an appeal against conviction and it appears to the Court of Appeal that the interests of justice so require, it may order the appellant to be retried.” [6] AC Determination para 4. [7] (2017) 20 HKCFAR 71. [8] Yeung Ag CJHC, Poon and Pang JJA in a judgment dated 15 November 2018 [2018] HKCA 592. [9] See paras 2 and 3 above. [10] Set out in para 3 fn 5 above. [11] (2017) 20 HKCFAR 414. [12] [1980] AC 351. [13] (2007) 10 HKCFAR 632. [14] (2010) 13 HKCFAR 27. [15] Au Pui Kuen at 357D-E. [16] Forrester Bowe (Junior) v The Queen [2001] UKPC 19, at para 39. [17] Tam Ho Nam (No. 2) at para 21. [18] Mok Kin Kau v HKSAR (2008) 11 HKCFAR 1, at para 10. [19] Au Pui Kuen at 356H. [20] Au Pui Kuen at 357C-D. [21] At 359D-E. See also Dennis Reid v The Queen [1980] AC 343, at 350G-H. [22] [1955] 39 HKLR 49, at 60. [23] Ting James Henry at para 51. [24] Au Pui Kuen at 357D-E. [25] Ting James Henry at para 50. [26] At paras 12 and 14. [27] See Tam Ho Nam (No. 2) at para 24 referring to R v Chau Mei Ling [1981] HKC 542, at 545B-C; Mok Kin Kau at para 7; HKSAR v Li Yanhong (No. 2) [2016] 1 HKLRD 946, at para 14. [28] See, for example, Ting James Henry at para 52. [29] See, in the context of sentencing, Secretary for Justice v Wong Chi Fung (2018) 21 HKCFAR 35, at para 62. [30] A Visiting Psychiatrist at Siu Lam Psychiatric Centre. [31] CA Determination para 11. [32] AC Determination para 5. [33] Like the first retrial: see para 7 above. [34] With her, Mr Franco Kuan. Mr Shahmim Khattak, Mr Anthony Lai and Ms Holly Cheng appeared for the appellant. [35] CA Determination para 10. [36] CA Determination para 10. [37] These points are set out in para 38 of the respondent’s written Case. They largely consist of an assertion that the appellant was lying and attempting to cover up the truth. [38] See paras 6 and 7 above. [39] According to the criteria for a diagnosis of depression, in summary an individual must experience five or more of the following symptoms: 1. Depressed mood; 2. Diminished interest or loss of pleasure in almost all activities (anhedonia); 3. Significant weight change or appetite disturbance; 4. Sleep disturbance (insomnia or hypersomnia); 5. Psychomotor agitation or retardation; 6. Fatigue or loss of energy; 7. Feelings of worthlessness; 8. Diminished ability to think or concentrate; indecisiveness; 9. Recurrent thoughts of death, recurrent suicidal ideation without a specific plan, or a suicide attempt or specific plan for committing suicide. The appellant was found to have all 9 symptoms. 
Press Summary (English) Press Summary (Chinese) DCCC 854/2021 [2022] HKDC 981 IN THE DISTRICT COURT OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO 854 OF 2021 -------------------------- --------------------------- --------------------------------------- REASONS FOR VERDICT --------------------------------------- 1. At the material times, the General Union of the Hong Kong Speech Therapists [香港言語治療師總工會] (“GUHKST”) was a registered trade union in the Hong Kong Special Administrative Region (“HKSAR” or “Hong Kong”) of the People’s Republic of China (“PRC”), and the 5 defendants herein were the executives of the Executive Council of GUHKST. They face one charge of conspiracy to print, publish, distribute, display and/or reproduce seditious publications, contrary to sections 10(1)(c), 159A and 159C of the Crimes Ordinance, Cap 200 (“the Charge”). Particulars of the offence are as follows: “[D1 to D5], between the 4th day of June, 2020 and the 22nd day of July, 2021, both dates inclusive, in Hong Kong, conspired together and with WONG Hoi-ching and other persons, to print, publish, distribute, display and/or reproduce seditious publications, namely three books by the name of “羊村守衛者”, “羊村十二勇士” and “羊村清道夫”, having an intention: - (a) to bring into hatred or contempt or to excite disaffection against the Central Authorities and/or the Government of the Hong Kong Special Administrative Region; (b) to bring into hatred or contempt or to excite disaffection against the administration of justice in Hong Kong; (c) to raise discontent or disaffection amongst inhabitants of Hong Kong; (d) to incite persons to violence; and/or (e) to counsel disobedience to law or to any lawful order.” 2. All defendants plead not guilty to the charge. THE PROSECUTION’S CASE 3. It is alleged by the prosecution that D1 to D5 among themselves and together with Wong Hoi-ching and other persons had embarked upon an agreement to cause the three picture books particularized in the Charge, ie “羊村守衛者” (“Book 1”), “羊村十二勇士” (“Book 2”) and “羊村清道夫” (“Book 3”), to be printed, published, distributed, displayed and/or reproduced in the name of GUHKST from 4 June 2020 (ie the day when Book 1 was published) until 22 July 2021 when they were arrested, and that these three books had the seditious intention as particularized in the Charge. THE DEFENCE CASE 4. All defendants argue that the offence charged is unconstitutional on the ground that it is inconsistent with their freedom of expression, speech and publication, and/or freedom to engage in literary and artistic creation and other cultural activities guaranteed by the Basic Law, and the Hong Kong Bill of Rights Ordinance, Cap 383, and that the offence charged should therefore be struck down outright, or it should be given a remedial interpretation in the way suggested by counsel for some defendants to render it constitutionally compliant. It is also argued that the Charge should not cover “Central Authorities”, and that the facts proved by the prosecution are insufficient to prove the Charge beyond reasonable doubt against each of the defendants. THE EVIDENCE 5. Nearly the whole of the prosecution evidence is admitted by the defendants pursuant to section 65C of the Criminal Procedure Ordinance, Cap 221. The facts admitted are set out in the Admitted Facts (Exhibit P1), and the various exhibits (ie Exhibits P2 to P144) referred to in the Admitted Facts are produced into evidence with their consent. 6. A witness statement made by Miss Cheng Kin-yee, a senior manager of the Hong Kong and Shanghai Banking Corporation, is admitted into evidence pursuant to section 65B of the Criminal Procedure Ordinance. Miss Cheng explains the procedure to follow for a registered trade union to set up a business account with the bank. She is not called upon for cross examination. Her evidence is not in dispute. 7. Only one witness has testified in court. He is DPC12776 (“PW1”). He gives evidence on his investigation into the contents of the three books particularized in the Charge to establish their correlation with some real persons and some actual events in Hong Kong. He also gives evidence on the status of publication of the three books in social media platforms and the internet, and on the information that he has retrieved from the internet webpage of GUHKST. He produces the results of his investigation (Exhibits P146 to P150). He is cross-examined by counsel for some of the defendants with a view to clarifying or correcting some of his evidence. By and large, his evidence is also not in dispute. 8. At the close of the prosecution’s case, there is no submission of no case to answer from any of the defendants. After this court has ruled that each defendant has a case to answer, all of them elect not to give evidence and call no witnesses. On the other hand, certain facts have been agreed between D4 and the prosecution pursuant to section 65C of the Criminal Procedure Ordinance (Exhibit D4-1), and the exhibits referred to therein are produced with their consent (ie Exhibits D4-2 to D4-27). The facts so admitted are binding upon the prosecution and D4 only. However, if any of such admitted facts are beneficial to the defence of any other defendant, these facts will be taken into account by this court for the benefit of this other defendant even though he or she has not adopted these facts formally, but the facts admitted between D4 and the prosecution will not be used in any way to the detriment of these other defendants unless such facts are also proved upon another basis[1]. THE PRIMARY FACTS 9. At the close of the evidence, counsel for the prosecution and all defendants have made extensive submissions on the law and on the facts of this case. The legal arguments focus on the constitutionality of the offence charged, whereas the factual arguments focus on the sufficiency of the evidence to prove the guilt of each defendant. From the evidence and the submissions, it is clear that the primary facts of this case are not in dispute. 10. In its written closing submissions, the prosecution has attached a Chronology of Events (Annex-1) which is prepared partly on the basis of the evidence adduced at this trial, and partly on the basis that the court may take judicial notice of certain events. The accuracy of the events described in this Chronology is not challenged by any of the defendants, including those events that the prosecution says that this court may take judicial notice of. 11. I am satisfied that it is proper to take judicial notice of those events as described in the Chronology, bearing in mind not only that those events are known, but also that similar events had been described by the Court of Final Appeal in Kwok Wing Hang v Chief Executive in Council[2], and by Mr Wang Chen, the Vice Chairman of the Standing Committee of the National People’s Congress, on 22 May 2020 in his Explanation on “The Draft Decision of the National People’s Congress on Establishing and Improving the Legal System and Enforcement Mechanisms for the Hong Kong Special Administrative Region to Safeguard National Security”, which was adopted by the Court of Final Appeal and recited in its judgment in HKSAR v Lai Chee Ying[3]. 12. On the basis of judicial notice and/or the evidence given by PW1, this court finds that the following political and social events had occurred in HKSAR prior to the establishment of GUHKST: (a) In February 2019, the HKSAR Government proposed the Extradition Bill[4]. A section of the public took to the streets to voice out their opposition. (b) Since around mid-2019, street protests began to persist with some of them escalating into territorial-wide or large scale unlawful assemblies or riots. Other forms of protests began to take shape such as street propaganda[5], cyber abusing, doxing and consumer boycotting[6], industrial actions and strikes. Some of the protests had escalated to become vandalism[7], vigilantism[8], criminal damage, paralyzing public transports, etc. Parties supporting or believed to be supporting the establishment and persons who were or believed to be mainlanders became targets of attack. (c) On 12 June 2019, the LegCo was besieged by protesters, pressuring the Government to withdraw the Extradition Bill. (d) On 15 June 2019, the Chief Executive of HKSAR declared to suspend the Extradition Bill. (e) On 1 July 2019, a sizable number of the public took to the street to protest against the Extradition Bill, and after which, some of the protesters forcefully entered the LegCo Building. (f) On 4 September 2019, the Chief Executive of HKSAR declared to withdraw the Extradition Bill. (g) On 1 October 2019, a riot broke out in Tsuen Wan, during which a rioter was shot in the chest[9]. (h) On 4 October 2019, a riot broke out in Sai Wan Ho, during which a rioter was shot in the thigh[10]. (i) In November 2019, “three suspensions” (“大三罷”, ie stop work, stop classes, and stop markets) were undertaken by some members of the public as a means to pressurize the Government. 13. As to the establishment of GUHKST and the election of executives into the 1st Executive Council of GUHKST, the following facts have been proved by the Admitted Facts (Exhibit P1) and the various exhibits produced together with the Admitted Facts: (a) By an application dated 13 November 2019 (Exhibit P2), GUHKST applied to the Registrar of Trade Unions of the Labour Department for registration as a trade union under the Trade Unions Ordinance, Cap 332. (b) According to this application (Exhibit P2), GUHKST was established on 13 November 2019, and it had 7 members on that day. (c) This application (Exhibit P2) was signed by 7 persons who confirmed themselves to be the voting members of GUHKST. These 7 persons included D1 and Miss Wong Hoi-ching, but did not include D2 to D5. (d) By a Certificate dated 18 November 2019 (Exhibit P3), the Registrar of Trade Unions certified that it had received the application by GUHKST for registration as a trade union. (e) By a Certificate of Registration of a Trade Union dated 22 November 2019 (Exhibit P4), GUHKST was duly registered as a trade union in Hong Kong. (f) Since its establishment, GUHKST set up its website in the internet, and accounts in three social media platforms, ie Facebook, Instagram and Telegram, as channels for communication. (g) By a Notice dated 27 November 2019, the Preparatory Group of GUHKST (香港言語治療師總工會籌備小組) convened the 1st Extraordinary General Meeting (“EGM”) for the members of GUHKST to be held on 12 December 2019[11]. Two documents were attached to this Notice. Attachment I was the agenda of the 1st EGM. Attachment II was a 17-page document in Chinese with the title of “香港言語治療師總工會規則” (“The Rules of GUHKST”)[12]. (h) On 12 December 2019, GUHKST held its 1st EGM. In this meeting, members of GUHKST adopted the Rules of GUHKST[13], and elected D1 to D5 and Miss Wong Hoi-ching to act as the executives of the 1st Executive Council (第一屆理事會) of GUHKST[14]. (i) On 24 December 2019, the Registrar of Trade Unions received a List of Officials dated 12 December 2019 (Exhibit P5) from GUHKST. According to this List, D1 to D5 and Miss Wong Hoi-ching were elected into the Executive Council of GUHKST on 12 December 2019 for a term of two years, with D1 acting as the Chairperson, D2 and Miss Wong Hoi-ching as the two Vice-chairpersons, D3 as the Secretary, D4 as the Treasurer, and D5 as an Officer. (j) On 10 January 2020, D1, D3 and D4 attended HSBC to apply for a bank account for GUHKST which required the joint signatures of D1 and D4, or the joint signatures of D3 and D4 to operate. D1, D3, D4 and Miss Wong Hoi-ching each had 25% of the voting rights in relation to the account, while D1 and D4 were the key controllers, direct appointees and contact persons. D1 and D4 were the authorized persons to use the bank cards and operate online banking of the account[15]. (k) On 30 January 2020, D4 entered into a License Agreement with Newswood Ltd to rent Room A2, Flat 1011, Block B, Hung Hom Commercial Centre, Ma Tau Wai Road, Hung Hom (“the Hung Hom office”), which served as the registered office and correspondence address of GUHKST for a term of 6 months commencing on 12 February 2020 [Exhibit P6(a)]. D4 entered into another agreement on 7 August 2020 to rent the same premises for another 6 months commencing from 12 August 2020 [Exhibit P6(b)]. (l) In January 2020, GUHKST engaged the service of a registered accountant to act as the auditor of GUHKST’s account. Two Statements of Accounts dated 29 July 2020 [Exhibit P9(a)] and 12 October 2020 [Exhibit P9(b)] were signed by the accountant, D1 and D4, and submitted to the Labour Department. (m) Until 22 July 2021 when D1 to D5 were arrested, GUHKST remained in existence, D1 to D5 continued to hold the same respective posts in the Executive Council of GUHKST. GUHKST still used the same Hung Hom office, and used the same bank account. 14. On the basis of judicial notice and/or the evidence given by PW1, this court finds that the following political and social events had occurred in HKSAR after the establishment of GUHKST: (a) In late January 2020, Covid-19 broke out in the Mainland and Hong Kong, and then worldwide. (b) In February 2020, industrial strikes were organized by some members of the public in the hope of pressuring the HKSAR Government to close all borders with the Mainland. (c) The Decision of the National People’s Congress on Establishing and Improving the Legal System and Enforcement Mechanisms for the Hong Kong Special Administrative Region to Safeguard National Security was adopted at the Third Session of the Thirteenth National People’s Congress on 28 May 2020 leading to the enactment of the National Security Law. (d) On 30 June 2020, the National Security Law was promulgated in HKSAR. (e) On 11 and 12 July 2020, the Hong Kong pro-democracy 35+ primaries (which selected pro-democracy candidates to run for the subsequent LegCo election in 2020) were held in Hong Kong. (f) In August 2020, the media reported that 12 fugitives who had been charged for committing various serious offences in Hong Kong fled from the jurisdiction in a boat and entered the Chinese waters unlawfully. They were intercepted by the PRC authorities and escorted to the Mainland for disposal. PUBLICATION AND PROMOTION OF THE PICTURE BOOKS 15. According to the Admitted Facts (Exhibit P1) and the various exhibits referred to therein, the defendants admitted that GUHKST had published and promoted the three books as follows. “羊村守衛者” (Book 1) 16. On 4 June 2020, e-edition (電子版) of Book 1 was published on an internet platform “ISSUU” with its link (and list of shops distributing hardcopies for free) posted on GUHKST’s website (Exhibit P13), and its social media accounts in Facebook, Instagram and Telegram. 17. Hardcopies (實體版) of Book 1 were placed inside shops for collection by the general public free of charge. GUHKST published the list of shops where hardcopies of Book 1 were available for collection on its website and social media platforms. The list of shops published in GUHKST’s Instagram account [Exhibit P18(a)] was dated 5 June 2020, and the list of shops published in its Facebook account [Exhibit P18(b)] showed it was last updated on 30 September 2020. 18. On or before 15 June 2020, D1 and D2 attended an interview by inmediahk.net (香港獨立媒體) to promote Book 1. An article entitled “羊村團結抵抗灰狼入侵 言語治療師工會出繪本述香港故事 細路都睇得明” was published online by inmediahk.net on 15 June 2020 in relation to this interview [Exhibit P34(a)(ii)]. On 17 June 2020, a note of thanks in relation to this interview by inmediahk.net was posted in the Facebook account of GUHKST [Exhibit P34(a)(i)]. 19. On or before 26 June 2020, D1 and D2 attended an interview by Radio Free Asia (香港自由亞洲電台) to promote Book 1. An article entitled “言語治療師創「送中」兒童繪本 – 在黑暗時代播下自由種子” in relation to this interview was published online by Radio Free Asia on 26 June 2020 [Exhibit P34b(ii)], which included its video interview with D1 and D2 [Exhibit P34(b)(iii)(V)[16]]. On 26 June 2020, a note of thanks to Radio Free Asia was posted in the Facebook account of GUHKST [Exhibit P34(b)(i)] or [Exhibit D4-15]. 20. On or before 28 June 2020, D1 and D2 attended an interview by Apple Daily (蘋果日報) to promote Book 1. An article entitled “言語治療師工會播民主種子 狼逼羊反 繪本向兒童說抗爭” in relation to this interview was published online by Apple Daily on 28 June 2020 [Exhibit P34(c)(ii)]. On 28 June 2020, a note of thanks to Apple Daily was posted in GUHKST’s Facebook account [Exhibit P34(c)(i)]. 21. On 28 June 2020, the parent-child reading guide (“親子共讀指引”) of Book 1 was uploaded to the website of GUHKST (Exhibit P30). According to the facts admitted between the prosecution and D4, the parent-reading guide was also uploaded to GUHKST’s Facebook account (Exhibit D4-16), and Telegram account (Exhibit D4-17) on 28 June 2020. 22. On 25 July 2020, the video version (影視版) of Book 1 [Exhibit P16(a)] was uploaded to Youtube by Hang Ten Collage (杏壇放送學院)[17]. 23. On or before 30 September 2020, D1 attended an interview by inmediahk.net to explain why GUHKST was set up, and how Book 1 was received by the public and what they planned to do next. An article entitled “「不為殘酷事實擊退只希望一輩人多點努力」-- 專訪言語治療師總工會主席黎雯齡” relating to this interview was published online by inmediahk.net on 30 September 2020 [Exhibit P34(d)]. “羊村十二勇士” (Book 2) 24. On 19 December 2020, e-edition of Book 2 (Exhibit P14) was published on an internet platform “ISSUU” with its link (and list of shops distributing hardcopies for free) posted on GUHKST’s website, and its social media accounts in Facebook, Instagram and Telegram. 25. Hardcopies of Book 2 were placed inside shops for collection by the general public free of charge. GUHKST published the list of shops where hardcopies of Book 2 were available for collection on its website and social media platforms. The list of shops published in GUHKST’s Instagram account [Exhibit P18(c)] was dated 19 December 2020, and the list of shops published in its Facebook account [Exhibit P18(d)] showed that it was last updated on 16 January 2021. 26. On 23 December 2020, Stand News (立場新聞) published online an article entitled “屬於香港人嘅繪本 -- 《羊村守衛者》同《羊村十二勇士》” [Exhibit P34(e)(ii)]. On 24 December 2020, a note of thanks to Stand News was posted in GUHKST’s Facebook for providing it with a platform for promoting Book 1 and Book 2 [Exhibit P34(e)(i)]. 27. On 5 January 2021, the police obtained one hardcopy of Book 1 and one hardcopy of Book 2 (Exhibit P24), both of which were placed on the top of a table near the entrance of a shop which assisted in the distribution of the books. 28. On 10 January 2021, the video version of Book 2 [Exhibit P17(a)[18]] was uploaded to Youtube by Hang Ten Collage. 29. On or before 14 January 2021, D1 and D2 attended an interview by inmediahk.net to promote Book 2. An article entitled “【12港人】工會出版《羊村十二勇士》繪本 「希望十二港人知道仲有人記掛住佢哋」” was published online by inmediahk.net on 14 January 2021 in relation to this interview [Exhibit P34(f)(ii)]. On 15 January 2021, a note of thanks in relation to this interview by inmediahk.net was posted in the Facebook account of GUHKST which expressed its gratitude for having one more opportunity for its executives to explain why the picture books were created [Exhibit P34(f)(i)]. “羊村清道夫” (Book 3) 30. On 16 March 2021, e-edition of Book 3 (Exhibit P15) was published on an internet platform “ISSUU” with its link (and list of shops distributing hardcopies for free) posted on GUHKST’s website, and its social media accounts in Facebook, Instagram and Telegram. 31. Hardcopies of Book 3 were placed inside shops for collection by the general public free of charge. GUHKST published the list of shops where hardcopies of Book 3 were available for collection on its website and social media platforms. According to the list of shops published on 23 March 2021 in GUHKST’s Facebook account [Exhibit P18(e) or D4-27], as well as in its Instagram account and Telegram account [Exhibit D4-27 according to the facts admitted between the prosecution and D4], hardcopies of Book 3 were available for collection in those shops on the following day, ie 24 March 2021. 32. On 24 March 2021, D1 and D2 attended a radio interview with an online radio “D100” to promote Book 3. The interview was video-recorded [Exhibit P34(g)(ii)(V)[19]]. 33. On 24 March 2021, the police obtained from 9 shops a total of 9 hardcopies of Book 3 (ie one copy from each shop) [Exhibits P19 to P23, and P25 to P28]. On 25 March 2021, the police obtained one hardcopy of Book 3 (Exhibit P29) from another shop. The books were placed either on the cashier counters or on tables close to the entrances of these shops[20]. The representatives of two of these shops pointed out that D2 and Miss Wong Hoi-ching had approached them. One of the shops was the same shop from which the police obtained a hardcopy of Book 1 and a hardcopy of Book 2 on 5 January 2021 (see paragraph 27 above). 34. On 20 June 2021, D2 and D4 held the first “Parents-Children Reading Session” (親子讀書會)[21]. According to the promotion records posted in GUHKST’s Facebook and Instagram accounts before the event [Exhibit P31(a)], the activity target was children studying in kindergarten and junior primary school, and the book to be read together by parents and children was Book 1[22]. According to the photographs and note of thanks posted in GUHKST’s Facebook and Instagram accounts after the event [Exhibit P31(b)], children with adults attended the book reading session and the book read together was Book 1[23]. 35. On 1 July 2021, D1 and D2 distributed Book 3 at a street booth in Causeway Bay. Photographs of the event was posted in the GUHKST’s accounts in Facebook and Instagram (Exhibit P32). 36. Promotion materials had been posted in the Facebook, Instagram and Telegram accounts of GUHKST for another Parents-Children Reading Session to be held on 25 July 2021 with children studying in kindergarten and junior primary school as the activity target, and Book 1 to be the book to be read (Exhibit P33). This reading session had not taken place due to the arrest of the defendants. 37. On 22 July 2021, D1 to D5 were arrested. Miss Wong Hoi-ching left Hong Kong on 21 July 2021. 38. The residences of D1 to D5 and the Hung Hom office of GUHKST were searched and various exhibits were seized [see Section H of the Admitted Facts (Exhibit P1) for the details]. 39. Upon search, drafts of another picture book entitled《羊村投票日》(“Book 4”) which pertained to the 35+ Primaries were found respectively in the electronic devices of D1, D4 and D5. 40. D1 was confirmed to be the administrator of GUHKST’s accounts in Facebook, Instagram and Telegram. 41. According to the undisputed evidence of PW1, the Facebook, Instagram and Telegram accounts of GUHKST were last updated on 4 July 2021. His searches conducted on 22 July 2022 on the internet using the titles of Book 1, Book 2 and Book 3 as key words revealed three Youtube links to the video versions of Book 1 [Exhibits P148(1) to (3)], one Youtube link to the video version of Book 2 [Exhibit P148(4)], and two Youtube links to the video version of Book 3 [Exhibits P149(5) & (6)], but the video versions of the books uploaded onto the GUHKST’s website and published by Hang Tan Collage had been removed. OTHER PUBLICATIONS OF 1ST EXECUTIVE COUNCIL OF GUHKST 42. According to the Admitted Facts (Exhibit P1), in or about August 2020, publications of the following titles had been uploaded onto the website of GUHKST: (a) “The platform booklet of the candidates for election into the 2019 Executive Council of GUHKST[24]” (Exhibit P35C). (b) “The platform booklet of the 2020 – 2021 Executive Council of GUHKST[25]” (Exhibit P35B), and (c) “2019 – 2020 Annual Report[26]” (Exhibit P35A). 43. Between 2020 and 2021, 5 newsletters of GUHKST had been uploaded to the website of GUHKST. They were Newsletter Nos. 1 to 5, for the respective periods of November 2019 to February 2020 [Exhibit P36(a)], February 2019 to May 2020[27] [Exhibit P36(b)], May 2020 to August 2020 [Exhibit P36(c)], August 2020 to November 2020 [Exhibit P36(d)], and November 2020 to February 2021 [Exhibit P36(e)]. The defendants admitted that Newsletters Nos. 3 and 5 had announced news concerning publications of the picture books. SEARCH INSIDE THE HUNG HOM OFFICE OF GUHKST 44. A search was conducted by the police inside the Hung Hom office of GUHKST in the presence of D1 after her arrest on 22 July 2021. The exhibits seized are listed out in paragraph 31 of the Admitted Facts (Exhibit P1), including but not limited to the following: (a) 63 hardcopies of Book 1 (Exhibit P124), (b) 179 hardcopies of Book 2 (Exhibit P125), (c) one hardcopy of Book 2 in A4 size (Exhibit P126), (d) 178 hardcopies of Book 3 (Exhibit P127), and (e) 18 pieces of toy figures in usual protester’s gear and with the words “GUHKST” (Exhibit P128). 45. All defendants have a clear criminal record. I remind myself of the good character direction both in terms of propensity and credibility in favour of each of the defendants. 46. D1 and D2 were charged on 23 July 2021. D3 to D5 were charged on 30 August 2021. FACTS ADMITTED BETWEEN THE PROSECUTION AND D4 47. These facts are set out in D4’s Admitted Facts (Exhibit D4-1) and the attached exhibits (Exhibits D4-2 to D4-27). It is not necessary to recite them in detail. Suffice to say that these admitted facts have proven that GUHKST had undertaken a range of activities in the capacity of a registered trade union in Hong Kong both internally with its members and externally with other organizations apart from publishing Books 1 to 3. SECTIONS 9 AND 10(1)(C) OF THE CRIMES ORDINANCE 48. In simple terms, the prosecution alleges that the defendants had conspired together and with Wong Hoi-ching and other persons to commit an offence contrary to section 10(1)(c) of the Crimes Ordinance. 49. Section 10(1) of the Crimes Ordinance has created 4 different offences under each of its 4 subsections. This case concerns only with section 10(1)(c), which stipulates that any person who prints, publishes, sells, offers for sale, distributes, displays or reproduces any seditious publication shall be guilty of an offence. The maximum penalty stipulated is a fine of HK$5,000 and imprisonment for 2 years for a first offence, and 3 years’ imprisonment for a subsequent offence. 50. According to section 10(5) of the Crimes Ordinance, “seditious publication” means a publication having a seditious intention. 51. “Seditious intention” is defined in section 9 of the Crimes Ordinance. Section 9 in its current statutory form is as follows: “9. Seditious intention (1) A seditious intention is an intention— (a) to bring into hatred or contempt or to excite disaffection against the person of Her Majesty, or Her Heirs or Successors, or against the Government of Hong Kong, or the government of any other part of Her Majesty’s dominions or of any territory under Her Majesty’s protection as by law established; or (Replaced 28 of 1938 s. 2) (b) to excite Her Majesty’s subjects or inhabitants of Hong Kong to attempt to procure the alteration, otherwise than by lawful means, of any other matter in Hong Kong as by law established; or (c) to bring into hatred or contempt or to excite disaffection against the administration of justice in Hong Kong; or (d) to raise discontent or disaffection amongst Her Majesty’s subjects or inhabitants of Hong Kong; or (e) to promote feelings of ill-will and enmity between different classes of the population of Hong Kong; or (f) to incite persons to violence; or (Added 30 of 1970 s. 2) (g) to counsel disobedience to law or to any lawful order. (Added 30 of 1970 s. 2) (2) An act, speech or publication is not seditious by reason only that it intends— (Amended 28 of 1938 s. 2) (a) to show that Her Majesty has been misled or mistaken in any of Her measures; or (b) to point out errors or defects in the government or constitution of Hong Kong as by law established or in legislation or in the administration of justice with a view to the remedying of such errors or defects; or (c) to persuade Her Majesty’s subjects or inhabitants of Hong Kong to attempt to procure by lawful means the alteration of any matter in Hong Kong as by law established; or (d) to point out, with a view to their removal, any matters which are producing or have a tendency to produce feelings of ill-will and enmity between different classes of the population of Hong Kong. (Amended 28 of 1938 s. 2) (3) (Repealed 74 of 1992 s. 2) ISSUES 52. From the submissions of the parties, I find that the following issues fall for consideration: - (a) Does the offence charged cover the “Central Authorities”? (b) What are the elements of the offence, in particular whether it requires proof of “the common law intention”? (c) Is the offence charged unconstitutional? (d) Were the three picture books or any of them seditious publication? (e) Was there the alleged conspiracy, and if so, did D1 to D5 or any of them take part in it? (f) Is prosecution of the offence time-barred? Whether the offence charged cover “Central Authorities” 53. It is clear that the term “Central Authorities” does not appear in the written provisions of section 9 of the Crimes Ordinance as they are now in the statute book, but the term “Her Majesty” with other terms and expressions associated with that concept are still there. Section 9 was of course enacted at the time when Hong Kong was under the colonial rule, but section 9 in its written form has not been amended despite the PRC has already resumed exercising sovereignty over Hong Kong. 54. The Crimes Ordinance, including its sections 9 and 10, were part of the “laws previously in force in Hong Kong” before the Handover. It is adopted as laws of the HKSAR pursuant to Article 160 of the Basic Law (“BL”) and section 2A of the Interpretation and General Clauses Ordinance, Cap 1 (“IGCO”). To facilitate the modification and the adaptations of the previous laws so as to bring them into conformity with the status of Hong Kong as a Special Administrative Region of PRC, section 1 of Schedule 8 to the IGCO provides that: “Any reference in any provision to Her Majesty, the Crown, the British Government or the Secretary of State (or to similar names, terms or expressions) where the content of the provision— (a) relates to title to land in the Hong Kong Special Administrative Region; (b) involves affairs for which the Central People’s Government of the People’s Republic of China has responsibility; (c) involves the relationship between the Central Authorities and the Hong Kong Special Administrative Region, shall be construed as a reference to the Central People’s Government or other competent authorities of the People’s Republic of China.” 55. Section 2 of Schedule 8 to the IGCO further provides that: “Any reference in any provision to Her Majesty, the Crown, the British Government or the Secretary of State (or to similar names, terms or expressions) in contexts other than those specified in section 1 shall be construed as a reference to the Government of the Hong Kong Special Administrative Region.” 56. As to the meaning of “Central Authorities”, section 3 of IGCO when defining “State (國家)” provides that “State” includes “the Central Authorities of the People’s Republic of China that exercise functions for which the Central People’s Government has responsibility under the Basic Law”. It is therefore clear that the functions that fall within the responsibility of the Central People’s Government under the BL are to be exercised by the Central Authorities. In other words, the responsibility of the Central People’s Government under the BL can be regarded as the responsibility of the Central Authorities. 57. On the basis of these statutory provisions, counsel for D4 submits that the reference to “Her Majesty” in section 9 of the Crimes Ordinance cannot be construed as a reference to the “Central Authorities” pursuant to section 1 of Schedule 8 to the IGCO. His argument runs as follows. First, section 1(a) of the IGCO is irrelevant because this case is not related to title to land in HKSAR. Second, while the Central People’s Government or the Central Authorities are responsible under the BL for the foreign affairs relating to the HKSAR (Article 13), the defence of the HKSAR (Article 14), the addition or deletion of laws relating to the defence and foreign affairs and matters outside the autonomy of the HKSAR from Annex III to the BL [Article 18(3)], as well as the declaration of a state of emergency here [Article 18(4)], none of these affairs are referred to in section 9 of the Crimes Ordinance. Lastly, section 1(c) is not applicable because section 9 of the Crimes Ordinance does not concern with the relationship between the Central Authorities and HKSAR. Hence, counsel for D4 submits that any reference to “Her Majesty” in section 9 of the Crimes Ordinance should be construed as a reference to the Government of HKSAR according to section 2 (instead of section 1) of Schedule 8 to the IGCO. 58. Counsel for D4 further asks this court to distinguish between power and responsibility. He submits that it is the Government of HKSAR which has the responsibility to defend national security in the Region, whereas the Central People’s Government only has the power to oversee that this is done. He refers to Article 23 of the BL which stipulates that the HKSAR shall enact laws “on its own” to prohibit any act of treason, secession, sedition, subversion against the Central People’s Government, and against various other kinds of national security offences. He also refers to the two Explanations[28] given by the Standing Committee of the National People’s Congress when the National Security Law (“NSL”)[29] was going through its legislative process in the NPC, in which it was said that the HKSAR has a “constitutional responsibility” to safeguard national security in Hong Kong. He refers to Articles 7 to 11 of the NSL which stipulate the duties that HKSAR has to discharge in safeguarding national security, and refers to Articles 12, 16 and 18 which stipulate that certain specific government bodies have to be set up in HKSAR to perform the designated functions to safeguard national security. In summary, counsel for D4 submits that it is the Government of HKSAR which has the responsibility as prescribed by the BL to defend Hong Kong rather than the Central People’s Government, and that the responsibility lies solely upon the Government of HKSAR under the BL. In addition, counsel for D4 submits that at time when section 9 of the Crimes Ordinance was enacted, it could not be related to the foreign affairs or the defence of Hong Kong. Hence, the two criteria that have to be satisfied before section 1(b) of Schedule 8 to the IGCO can be invoked have not been satisfied, and any reference to “Her Majesty” in section 9 of the Crimes Ordinance can only be construed in accordance with section 2 of Schedule 8 to the IGCO, which means that any reference to “Her Majesty” in section 9 of the Crimes Ordinance can only be construed as a reference to the Government of the HKSAR rather than the Central People’s Government or the Central Authorities. Counsel for D3 also put forward similar submissions. 59. In my judgment, counsel’s submissions do not sit well with Article 14 of the BL. Article 14(1) provides that “the Central People’s Government shall be responsible for the defence of the Hong Kong Special Administrative Region”, whereas Article 14(2) provides that “the Government of the Hong Kong Special Administrative Region shall be responsible for maintenance of public order in the Region”. Since the Government of HKSAR will take care of public order in the Region under Article 14(2), what is left to be defended by the Central People’s Government under Article 14(1) can only be the defence of the HKSAR against any national security risk. 60. Furthermore, it is important to bear in mind that the HKSAR is an inalienable part of PRC under Article 1 of the BL. Since HKSAR is part and parcel of the PRC, it follows that safeguarding national security in HKSAR must be a component of the overall national security framework of the PRC, which must be a matter outside the limits of the HKSAR’s autonomy but within the purview of the Central Authorities. In other words, there is nothing called defending against national security risk of Hong Kong, but defending against national security risk in HKSAR. It has also been stated unequivocally in Article 3 of the NSL that the Central People’s Government has an overarching responsibility for national security affairs relating to the HKSAR. The fact that the HKSAR is under a constitutional responsibility to safeguard national security in the Region is not inconsistent with the Central People’s Government having the responsibility for safeguarding national security in HKSAR under the BL. On the contrary, it is only logical that the local government is under a constitutional duty to help the Central People’s Government in this regard. In my judgment, the Central People’s Government is responsible for the defence of HKSAR as prescribed by Article 14(1) of the BL. 61. As to whether the contents of section 9 of the Crimes Ordinance involve affairs for which the Central People’s Government has responsibility, the answer must be yes. Although section 9 does not say expressly that it was enacted for the defence of Hong Kong, it was clear from the provisions of section 9 that it was enacted to protect “the person of Her Majesty, or Her Heirs or Successors”, which must mean that section 9 was enacted to protect not only the Monarch of the British Empire but also her Monarchy. It was inconceivable that section 9 would not be used by the British ruler for the defence of her colony in Hong Kong if a sedition offence occurred here against her, her government or the local colonial government. Hence, it is incorrect to say that section 9 of the Crimes Ordinance was not related to the defence of Hong Kong. In addition, the Court of Final Appeal has already ruled definitively in HKSAR v Ng Hau Yi[30] that a prohibited act of sedition, including an offence contrary to section 10(1)(c) of the Crimes Ordinance, qualified as an offence endangering national security. In other words, the sedition offence created by sections 9 and 10 of the Crimes Ordinance must be one of the legislations to be used for the defence of HKSAR against offences endangering national security, which is one of the affairs for which the Central People’s Government has responsibility under the BL. 62. For these reasons, I am of the view that the criteria for triggering section 1(b) of Schedule 8 to the IGCO are satisfied, and that any reference to “Her Majesty” in section 9 of the Crimes Ordinance shall be construed as a reference to “the Central People’s Government or other competent authorities of the People’s Republic of China”. 63. Of course, the issue here is whether the Charge should cover the “Central Authorities” rather than the “Central People’s Government”. But since the Central Authorities exercise functions for which the Central People’s Government has responsibility under the BL, it will be proper to construe any reference to “Her Majesty” in section 9 of the Crimes Ordinance to be a reference to “the Central Authorities”, since “the Central Authorities” must fall within the limb of “other competent authorities of the People’s Republic of China” stated in section 1 of Schedule 8 to the IGCO. 64. Further or alternatively, reference can be made to the BL 160 Decision[31], which seeks to implement Article 160 of the BL to enable the laws in force in Hong Kong before 1 July 1997 to remain applicable after that date in HKSAR. By Articles 4 and 5 of this Decision, it is stipulated that the laws previously in force shall be adopted with such modifications, adaptations, restrictions and exceptions as may be necessary in accordance with the substitution rules stated in Annex III of that Decision. Paragraph 1 of that Annex states inter alia that in any provision that involves the affairs within the responsibilities of the Central Authorities (and the Central People’s Government is not mentioned there) as prescribed by the BL, any reference in these provisions to “Her Majesty”, etc should be construed as a reference to “the Central Authorities”. Since Schedule 8 to the IGCO was enacted to give effect to the BL 160 Decision, it must follow that the construction rules in Schedule 8 to the IGCO should operate in the same way as the substitution rules in Annex III of the BL 160 Decision. On this basis, reliance can be placed on section 22 of Schedule 8 to the IGCO, which provides that this Schedule applies unless the context otherwise requires, to construe any reference to “Her Majesty” in section 9 of the Crimes Ordinance to be a reference to “the Central Authorities”. 65. Another argument put forward by counsel for D4 is that the criminalization of sedition and its reach should be the concern of the Government of HKSAR because Article 23 of the BL has stipulated that HKSAR shall enact laws “on its own” to prohibit amongst others any act of sedition. On this basis, it is submitted that the matters referred to in section 9 of the Crimes Ordinance cannot be within the responsibility of the Central People’s Government. This submission is rejected. No offence of sedition has been enacted. What happens here is simply that the laws previously in force in Hong Kong before the Handover have been construed in this manner according to Schedule 8 to the IGCO. 66. Counsel for D4 also argues that if the Central People’s Government has responsibility under section 9 of the Crimes Ordinance, it would mean that everything done in relation to it, including prosecution of the relevant offence, would also be the responsibility of the Central People’s Government. The submission is also rejected. Article 14(1) only says that the Central People’s Government is responsible for the defence of HKSAR. Prosecution of offences in HKSAR is not covered. 67. For these reasons, I rule that the Central Authorities have been properly included in the Charge against the defendants. 68. The prosecution has submitted that the same ruling can be reached on the basis that the contents of section 9 of the Crimes Ordinance also involve the relationship between the Central Authorities and HKSAR, hence, section 1(c) of Schedule 8 to the IGCO also applies. In light of my ruling, it is not necessary to say anything further. 69. The prosecution further submits that, in the context of section 9(1)(a) of the Crimes Ordinance, “the Central Authorities” have to be considered to be “the body of central power under the constitutional order established by the Constitution of the People’s Republic of China”[32] under the leadership of the Communist Party, which, by reference to Chapter III “State Institutions” of the Constitution of the PRC, includes but is not limited to the National People’s Congress, the National People’s Congress Standing Committee, the President of the PRC, the State Council, the Central Military Commission. The reasons are set out in Part F.2 of its written submissions on the Interpretation and Human Rights issues. There are no submissions to the contrary. I accept the submissions of the prosecution. 70. According to the construction stated above, section 9(1)(a) of the Crimes Ordinance shall now read: “A seditious intention is an intention to bring into hatred or contempt or to excite disaffection against the Central Authorities, or against the Government of Hong Kong”. The other parts of the written section 9(1)(a) are to be disregarded as the types of colonial governments referred to therein do not exist now. References to “Her Majesty’s subjects” in sections 9(1)(b), 9(1)(d) and 9(2)(c) shall also be disregarded, but the other parts of each subsection shall remain. Hence, sections 9(1)(c), 9(1)(f) and 9(1)(g) shall remain unchanged, but section 9(1)(d) shall read: “A seditious intention is an intention to raise discontent or disaffection amongst inhabitants of Hong Kong.” 71. Likewise, section 9(2)(a) shall read: “An act, speech, or publication is not seditious by reason only that it intends to show that the Central Authorities or the Government of HKSAR has been misled or mistaken in any of their measures”. It is not necessary to change sections 9(2)(b) and 9(2)(d). For section 9(2)(c), the only change is to ignore the reference to “Her Majesty’s subjects”, leaving behind “to persuade inhabitants of HKSAR to attempt to procure any lawful means the alteration of any matter in HKSAR as by law established”. 72. The particulars of the Charge against the defendants are drafted according to these terms. Elements of the offence charged 73. In my judgment, from the plain reading of the statutory provision in question, a person will not commit an offence contrary to section 10(1)(c) of the Crimes Ordinance unless: (a) he prints, publishes, sells, offers for sale, distributes, displays or reproduces any publication (“the prescribed act”); (b) the publication is having a seditious intention; and (c) at the time when he performs the prescribed act, (i) he intends to perform the prescribed act, (ii) he knows that the publication is having a seditious intention, and (iii) he has a seditious intention. 74. There is no dispute between the parties that a defendant must intend to perform the prescribed act, and know that the publication is a seditious publication before he can be convicted of the offence. The only dispute is whether he must also have the seditious intention himself. The prosecution submits that while the publication must have a seditious intention, the defendant need not. Counsel for D2 submits that evidence of D2’s intention is neither a necessary nor a sufficient condition for proof of guilt. In my judgment, it is necessary to prove that the defendant also has a seditious intention for the following reasons. 75. First, where a statute laying down a criminal offence is silent on the relevant mental element, the starting point in interpreting the statute is that there is a common law presumption of mens rea, that this presumption is a strong one, and that it will only be rebutted by express words or by necessary implication. The prosecution has not demonstrated why this presumption of mens rea has been rebutted. 76. Second, at the time when the Sedition Ordinance 1970 was incorporated into the Crimes Ordinance, in addition to the present sections 9(1) and 9(2), there was a section 9(3) which provided that: “In determining whether the intention with which any act was done, any words were spoken, or any document was published, was or was not seditious, every person shall be deemed to intend the consequences which would naturally follow from his conduct at the time and under the circumstances in which he so conducted himself.” This deeming provision was clearly a kind of presumption assisting the prosecution to prove that the defendant had the requisite intention to do the act, utter the words, or publish the document. This section 9(3) was however repealed on 17 July 1992 by the Crimes (Amendment) (No. 2) Ordinance 1992 because it was considered to be inconsistent with the presumption of innocence provided by Article 11 of the Hong Kong Bill of Rights (“HKBOR”) and the spirit of section 65A of the Criminal Procedure Ordinance. The relevant official record of proceedings clearly showed that section 9(3) was repealed not because the Legislature considered that the prosecution had to prove the defendant’s intention unaided by any presumption. 77. Third, in Fei Yi Ming & Lee Tsung Ying v R[33], when dealing with one of the grounds of appeal in connection with section 3(2) of the Sedition Ordinance which was the same as section 9(3) of the Crimes Ordinance before it was repealed, the then Chief Justice recited in the judgment, at page 157, the following direction given by the trial judge to the jury: “You have been invited to look at the other articles in evidence. They are for your consideration in assisting you to throw light on the state of the publisher’s mind when he published the article of the 5th of March in the Ta Kung Pao, and articles published before and after the 5th of March are there in evidence before you – their weight, their value is entirely a matter for you.” (Italics for emphasis) It is therefore clear from this direction to jury that the defendant must have a seditious intention as an element of the offence. 78. Having said that, I am also of the opinion that, if the publication has a seditious intention involving more than one limb of section 9(1) of the Crimes Ordinance, it is not necessary for the defendant to have a seditious intention mirroring exactly in every respect with the publication so long as he shares some of the seditious intention of the publication. “Seditious intention” – the burden of proof 79. Section 9(1) of the Crimes Ordinance has stipulated that a seditious intention is an intention under one of the 7 limbs listed in sub-subsections (a) to (g). On the other hand, section 9(2) provides that any act, speech or publication is not seditious by reason only that it intends to achieve one of the purposes as stated in subsections (a) to (d). 80. In my judgment, section 9(2) of the Crimes Ordinance does not place any kind of burden of proof on the defendant. As a seditious intention is the fundamental core element of each of the seditious offences, the burden must be on the prosecution to prove this element of the offence, which will necessitate the prosecution to prove not only that the defendant has a seditious intention within one or more of the limbs stated in section 9(1)(a) to (g) of the Crimes Ordinance, but also that his act, speech or publication is not within any limb stated in section 9(2)(a) to (d). An authority supporting this approach is the case of Tong Yiu Wah v HKSAR[34] although that case was not a sedition case. “The Common Law Intention” 81. It is argued strongly by counsel for all defendants that, in addition to an intention within one or more of the 7 limbs stated in section 9(1) of the Crimes Ordinance, a seditious intention must also include “an intention to incite persons to violence or to create public disturbance or disorder for the purpose of disturbing constituted authority” formulated by case law in various common law jurisdiction (“the Common Law Intention”). In my judgment, there is no legal basis to incorporate the Common Law Intention into the statutory definition of “seditious intention” stipulated in section 9 of the Crimes Ordinance. 82. First, while the offence of sedition has its origin in the common law, sedition has been a statutory offence in Hong Kong since 1914 when the Seditious Publications Ordinance was enacted. There have been various legislative amendments thereafter, and the law now is that stated in sections 9 and 10 of the Crimes Ordinance. The meaning of “seditious intention” in Hong Kong is therefore defined by statute, including what is and what is not “seditious intention” according to the definition stipulated in the two subsections of section 9. Hence, it is the statutory meaning of “seditious intention” that matters. 83. Second, when moving the Sedition (Amendment) Bill 1970 to include an intention to incite other persons to violence, or to counsel disobedience to law or to any lawful order [which later became the present sections 9(1)(f) and 9(1)(g)] as another two limbs of seditious intention in the Sedition Ordinance, the then Attorney General said it was likely that, for a seditious intention in the other forms [which were equivalent to the present section 9(1)(a) to (e)], there would usually be an incitement to violence or to disobedience of the law. By using the words “likely” and “usually”, it shows that the Common Law Intention (which required an intention to incite other persons to violence) was not a necessary element of the seditious offences in Hong Kong. Likewise, the colonial government in Hong Kong pushed through the Crimes (Amendment) (No. 2) Bill on 24 June 1997 to add the words “with the intention of causing violence or creating public disorder or a public disturbance” into section 10(1) of the Crimes Ordinance. If seditious intention in section 9 had already included the Common Law Intention, there was clearly no need to push through this amendment. 84. Third, in Fei Yi Ming, it was held on appeal by the Full Court that incitement to violence was not a necessary element to be proved by the prosecution. This authority is binding on this court under the doctrine of stare decisis. 85. Fourth, defence counsel have cited many cases on the Common Law Intention, but the introduction of the Common Law Intention as a requirement constituting a seditious intention seemed to stem from the need perceived by the judges to limit the scope of the sedition offence so that it was only when the words uttered had created a real risk to upset the political and social order before the criminal law would step in to intervene. In those days, such real risk might be created only when violence or threat of violence was used. The situation has changed nowadays. Violence is not the only means to bring down a government or cripple its running. Spreading rumour, hatred and disinformation is clearly a readily available and may even be a more effective weapon without the need to incite people to violence. Since the common law is always developing, there is no reason now to stick to the Common Law Intention formulated decades ago to make incitement to violence a prerequisite component of a seditious intention. 86. Fifth, Article 20 of the NSL has made it clear that the offence of secession can be committed whether or not force or threat of force is used. Since the offence of sedition usually occurs as a prelude to secession, incorporating the Common Law Intention into the definition of seditious intention in the Crimes Ordinance does not conform with the intention that the NSL and the laws of HKSAR including sedition should work as a coherent whole to safeguard national security in the Region. 87. For these reasons, I hold that seditious intention as defined in section 9 of the Crimes Ordinance has never included the Common Law Intention as its necessary ingredient, and that there is no basis today to interpret seditious intention to include the Common Law Intention. Constitutional challenge to the offence charged 88. This prosecution is based on what the defendants had allegedly said and published. It raises the question whether the offence charged has infringed upon their rights to freedoms of expression, of speech, of publication, and of literary and artistic creation and other cultural activities. It is accepted by all parties that these freedoms are fundamental rights guaranteed and protected by Articles 27 and 34 of the BL, Article 19 of the International Covenant on Civil and Political Right (“ICCPR”) which is implemented by Article 39 of the BL, as well as Article 16 of the HKBOR. 89. Likewise, it is accepted that these are not absolute rights and may be subject to restrictions. However, such restrictions must be “prescribed by law” and must not contravene the provisions of ICCPR (Article 39(2) of the BL). 90. Article 19 of the ICCPR and Article 16 of HKBOR are in exactly the same terms. They provide the following: “(1) Everyone shall have the right to hold opinions without interference. (2) Everyone shall have the right to freedom of expression; this right shall include freedom to seek, receive and impart information and ideas of all kinds, regardless of frontiers, either orally, in writing or in print, in the form of art, or through any other media of his choice. (3) The exercise of the rights provided for in paragraph (2) of this article carries with it special duties and responsibilities. It may therefore be subject to certain restrictions, but these shall only be such as are provided by law and are necessary - (a) for respect of the rights or reputations of others; or (b) for the protection of national security or of public order (ordre public), or of public health or morals.” 91. At the outset, I accept that this court is under a duty to consider and determine constitutionality of the offence charged. In discharging this duty, I bear in mind that the court must give a generous interpretation to the fundamental rights guaranteed by the constitution, and that any restriction must be narrowly interpreted: HKSAR v Ng Kung Siu[35]. In the context of this particular case, it is particularly important to remember that: - “This freedom [of expression] includes the freedom to express ideas which the majority may find disagreeable or offensive and the freedom to criticize governmental institutions and the conduct of public officials.” 92. The first issue is whether the restriction, ie section 10(1)(c) of the Crimes Ordinance satisfies the “prescribed by law” requirement stipulated in Article 39(2) of the BL. The central requirement is that the offence must have a sufficiently clearly formulated core to enable a person, with advice if necessary, to regulate his conduct so as to avoid liability for that offence. At the same time, the principles recognize the need for both flexibility and development: Winnie Lo v HKSAR[36], and since a fundamental freedom of speech is involved, there must be an exceptionally high degree of certainty: Shum Kwok Sher v HKSAR[37]. 93. It is the defence submission that the offence requires proof of a seditious intention, but the words “hatred” (憎恨), “contempt” (藐視), “disaffection” (離叛), and “discontent” (不滿) in section 9 are too vague, imprecise and subjective so that an ordinary person will not be able to regulate his conduct to avoid criminal liability, and the fear of stepping onto the wrong side of the law will produce a chilling effect in derogation of the right to freedom of expression. For these reasons, it is submitted that section 10(1)(c) fails to satisfy the “prescribed by law” requirement. 94. In my view, “hatred”, “contempt”, “disaffection”, and “discontent” both in English and Chinese are just words with ordinary meaning, and these concepts are best left to the trial judge or jury to be applied in their ordinary meaning to the time, place and circumstances of the conduct in question. Besides, while these words are descriptive of a person’s subjective feelings or emotions, ordinary people in their daily lives must have personally experienced this kind of subjective feelings or emotions, and noticed other persons behave in that mental state. From these experiences and observations, it is clear that while the feelings or emotions are subjective, some objective facts will have to exist first in order to bring about that kind of subjective feelings or emotions. For instance, we hate a classmate because he has been bullying us for a long time; we look down on another person with contempt because he behaves in an immoral manner; we have no affection towards a leader who does not take care of our interest; and we feel discontented when we do not get what we deserve. From daily living experience, a judge or a jury is in a position to decide whether the acts done or the words uttered (which are proved by objective evidence) will lead to the kind of subjective feelings or emotions referred to in section 9(1). Hence, despite the end result being related to the subjective feelings or emotions of the targeted person of the seditious act or words, there are objective facts assisting a judge or a jury to make a determination whether the kind of subjective feelings or emotions have been produced, and whether acts were done or words were uttered with the intention to produce those subjective feelings or emotions. Likewise, anyone doing or intending to do the act or uttering the words may rely on his own experience, and if necessary obtain advice, to decide whether or not his act or words will be regarded as having a seditious intention. Hence, there is still a sufficiently clearly formulated core to determine whether or not his act or words may attract criminal liability. In the context of section 9(1) of the Crimes Ordinance, the acts or words that are prohibited are those acts or words that have the effect of demeaning the Central Authorities and/or the Government of HKSAR in the eyes of the general public, and/or estranging the relationship between these institutions and the people here, thereby damaging the legitimacy of the authorities and their relationship with the people, which in turn will or may endanger the political order and social tranquility of the nation. While it is not possible to list out each and every prohibited act, there is a “sufficiently clearly formulated core” to enable a person, with advice if necessary, to regulate his conduct so as to avoid criminal liability. 95. Furthermore, a seditious intention does not depend on the subjective feeling of the target institutions or persons, but depends on the subjective intention of the person uttering the words or printing the publication. He knows his intention of the words he used or the publication he printed. He can certainly refer to section 9(2) of the Crimes Ordinance to find out if his words or publications will or will not be regarded as seditious. For instance, if he is just criticizing the Central Authorities or the Government of HKSAR has been misled or mistaken in any of their measures, it will not even be necessary for him to worry about what “hatred”, “contempt”, “disaffection” or “discontent” in the context of section 9(1) may mean because his speech or publication will not be regarded as seditious in any event. In other words, it is only when he cannot rely on any of the 4 limbs in section 9(2), then it gives him a warning that he has to consider carefully whether what he is going to do or say may be prohibited by sections 9 and 10 of the Crimes Ordinance, and he may seek legal advice if necessary. From this perspective, section 9(2) has also assisted in delineating the area of risk, and this is sufficient to satisfy the requirement of certainty. 96. For these reasons, I am satisfied that despite the use of words like “hatred”, “contempt”, “disaffection”, and “discontent” in section 9 of the Crimes Ordinance, these concepts are not so vague or imprecise, and I am of the view that the degree of legal certainty satisfies the “prescribed by law” requirement stipulated in Article 19 of the ICCPR and Article 16 of HKBOR. 97. The next step is to consider whether the restriction or limitation is a proportionate measure. It is common ground that when the court considers whether a restriction imposed on a constitutionally guaranteed right is valid, it should go through the 4-step analysis as set out in Hysan Development Co Ltd v Town Planning Board[38]: (a) The restriction or limitation must pursue a legitimate aim; (b) The restriction or limitation must be rationally connected to that legitimate aim; (c) The restriction or limitation must be no more than necessary to accomplice that legitimate aim; (d) A reasonable balance must be struck between the societal benefits and the inroads made into the constitutionally protected rights of the individual, asking in particular whether the pursuit of the societal interest resulted in an unacceptably harsh burden on the individual. 98. On Step 1: criminalization of seditious acts clearly pursues a legitimate aim. The nature of the offence of sedition as described in R v Sullivan and Pigott [39] shows why criminalizing sedition is necessary. Fitzgerald J. said: “Sedition is a crime against society, nearly allied to that of treason, and it frequently precedes treason by a short interval. Sedition in itself is a comprehensive term, and it embraces all those practices, whether by word, deed or in writing, which are calculated to disturb the tranquility of the State, and lead ignorant persons to endeavor to subvert the Government and the laws of the empire. The objects of sedition are generally to induce discontent and insurrection and stir up opposition to the Government, and bring the administration of justice into contempt; and the very tendency of sedition is to incite the people to insurrection and rebellion. Sedition has been described as disloyalty in action, and the law considers as sedition all those practices which have for their object to excite discontent or dissatisfaction, to create public disturbance, or to lead to civil war; to bring into hatred or contempt the Sovereign or the Government, the laws or constitution of the realm, and generally all endeavours to promote public disorder …” 99. Step 2: It is also beyond argument that criminalization of seditious acts by sections 9 and 10 of the Crimes Ordinance is rationally connected with that legitimate aim. 100. The real arguments turn on Step 3, ie whether the restriction or limitation is no more than necessary to accomplice that legitimate aim. It is clear that the only possible ground justifying the restriction or limitation are national security and public order (ordre public) as stated in Article 19 of the ICCPR and Article 16 of HKBOR. 101. The dispute is to what extent restrictions can be imposed on the right to free speech in the name of national security. It is submitted that the concept of national security should be construed according to the Siracusa Principles on the Limitations and Derogation Provisions in the ICCPR so that the right to freedoms of expression, publication and literary and artistic creation will not be unduly restricted. According to the Siracusa Principles, a legitimate national security interest is one that aims “to protect the existence of the nation or its territorial integrity or political independence against force or threat of force. 102. The Siracusa Principles do not enjoy the status of rules of law in HKSAR. They were issued 38 years ago and likely to be outdated. Nowadays, the existence of a nation, its territorial integrity or political independence can be threatened not just by force or threat of force, but by propaganda spreading rumours, misinformation and disinformation that make the people no longer trust and even hate their government, resulting in serious social unrest and chaos. In this sense, making sedition an offence should be even a more important tool for protection of national security rather than holding it unconstitutional. In addition, its scope should not be unduly restricted, or else it will be ineffective in protecting national security. The issue is how to strike a balance. It must also be borne in mind that vague and arbitrary limitations on the right to freedoms of expression and publication cannot be accepted. 103. On this issue, I am of the view that the overseas statutes, case law, law commission working paper, academic commentaries are of little assistance. Their political background, social condition, culture and the availability of alternative legislations to deal with seditious situations to safeguard their national security are different to those in HKSAR. Hence, our focus should be on the unique political and social conditions of the Region, and the other laws in force in HKSAR. 104. In considering whether the offences created by sections 9 and 10 of the Crimes Ordinance are no more than necessary to accomplice the legitimate aim, it is important to understand the political and social condition in HKSAR at the time of the alleged offence against the defendants up till now: see paragraph 12 above. Since the start of the Anti-Extradition Movement in mid-2019, HKSAR went through a lengthy period of mass riots and civil unrests, some of them were extremely violent and weapons and petrol bombs were used. While the exact number of people participated in these riotous activities will never be known, it can be said with certainty that there were over tens of thousands of them. What these people had demonstrated during this period of time was that they did not recognize the sovereignty of the People’s Republic of China over HKSAR, and they did not support the policy of the “One Country, Two systems”. There were calls for Hong Kong independence or self-determination. Although the situation in HKSAR has more or less calmed down after the promulgation of the NSL, it is clear that these people have little change in their attitude. They just go underground and the seeds of unrest are still there. The political situation appears to be calm on the surface but very volatile underneath. 105. Under these circumstances, there is a strong pressing need to safeguard national security in HKSAR to prevent riots and civil unrests of any magnitude from happening again. It is also essential to protect the constitutional order of HKSAR under the “One Country, Two Systems” policy and to restore national unity as soon and as fullest as possible. It is also of fundamental importance that Hong Kong residents can have a prolonged period of living in peaceful environments after public tranquility had been destroyed or damaged for one year. It is therefore important to adopt measures to protect the general welfare and the interest of the collectivity as a whole under the concept of public order (ordre public). For these reasons, I find that the restrictions imposed by sections 9 and 10 of the Crimes Ordinance on the right to freedoms of expression and publication are necessary for the protection of national security and protection of public order (ordre public). From the decision of Ng Kung Siu, it is clear that a person’s right to freedom of expression can be restricted for the protection of public order (ordre public) for the benefit of the legitimate societal interests to consolidate the new constitutional order in Hong Kong and to implement the policy of “One Country, Two Systems”. The same principle should apply to the present case. At all material times of this alleged offence up till now, HKSAR has a wounded constitutional order to heal, and a detached national unity to restore. 106. Of course, there is still the question whether the restriction caused by these seditious offences is no more than necessary to limit the right to freedom of expression and of publication, etc. The present law does not prevent anyone to say and publish whatever they like, including criticisms of the Central Authorities and the Government of HKSAR in any form, provided that they do so without a seditious intention. Furthermore, on a proper construction of sections 9 and 10 of the Crimes Ordinance, the prosecution is required to prove that the defendant cannot benefit from the “defence” stated in section 9(2) of the Crimes Ordinance, and that the defendant had a seditious intention when he did the act complained of. 107. Furthermore, it is noteworthy that Article 2 of the NSL provides that the provisions of Article 1 and 12 of the BL on the legal status of the HKSAR are the fundamental provisions. No institution, organization or individual in the Region shall contravene these provisions in exercising their rights and freedom. Article 1 of BL provides that “the HKSAR is an inalienable part of the PRC”, and Article 12 of the BL provides that “the HKSAR shall be a local administrative region of the PRC, which shall enjoy a high degree of autonomy and come under the Central People’s Government”. The effect of Article 2 of the NSL is simply that when anyone exercises his rights and freedom, he cannot refuse to recognize HKSAR is an inalienable part of PRC, or that HKSAR enjoys only a high degree of autonomy rather than complete autonomy. When this simple rules are adhered to, it is unlikely for anyone to commit an offence of sedition unwittingly. 108. I am of the view that sections 9 and 10 of the Crimes Ordinance do not impose restriction more than necessary to limit the right to freedom of expression, publication, etc. for the protection of national security and of public order (ordre public). 109. Step 4: There is also nothing to suggest that a reasonable balance had not been struck between the restrictions and the societal benefits obtained as a result. 110. I rule that the constitutional challenge fails. Were the picture books seditious publication? 111. I now turn to the issue whether the three picture books or any of them are publication having a seditious intention. According to the Particulars of Charge, the prosecution alleges that the three books had a seditious intention in terms of section 9(1)(a), (c), (d), (f) and (g). 112. On the approach to be adopted to ascertain if all or any of the three picture books had a seditious intention, counsel for D2 and D3 submit that since a seditious publication is a publication having a seditious intention, the seditious intention must be found within the publication itself, and the mental state and the intention of a defendant are irrelevant. In other words, whether or not any of the books was a seditious publication has to be decided by simply reading the book itself. What the defendant had said about the book were irrelevant and prejudicial. The question is for the court to ascertain the intention of the publication, and not the intention of the defendant. 113. In my view, the publisher of a book is the best person to know what messages he intends the book to convey. His statement in this regard constitutes evidence of the intention of the book, and his statement in this regard may constitute an admission or a declaration against interest which is admissible in evidence subject to the usual criteria of voluntariness and fairness. Of course, it does not mean that whatever the publisher says about the intention of the book is the sole or decisive factor for deciding whether the book is having a seditious intention. It is just one piece of the evidence. If the book cannot convey the seditious intention that the publisher intends the book to convey, for instance, due to its poor production and quality, this book cannot be held to be a seditious publication. Furthermore, the statements made by the publisher may serve the purpose of refuting an otherwise appropriate inference that may be drawn from reading the book that it had a seditious intention. 114. It is also submitted by the defence that reliance should not be placed on Book 4 to prove whether or not Books 1 to 3 or any of them were seditious publication on the ground that Book 4 was not included in the Charge, and in any event, it had not been published, meaning that it could be amended at any time. I agree that the content of Book 4 should not be used as a tool to construe the intention of the other three books, though I am of the view that evidence relating to Book 4 is admissible to show the course of conduct pursued by the defendants concerned. 115. Counsel for prosecution points out that relevant factors for consideration include (i) nature of the audience, (ii) the state of the public feelings, and (iii) the time, place and mode of publication. 116. Counsel for D3 submits that the court should not read the books through the eyes of an extremely radical or cynical reader, but it must consider the likely effect of the books on ordinary people and on the audience to which the books are addressed. I agree. 117. Counsel for all defendants submit that none of the three picture books has seditious intention. In summary, the defence contends that: (a) the three books were simply children’s tales or fables; (b) the three books were advocating virtues that are universally celebrated; (c) the three books alone did not within itself provide any background of the setting but only relates a fairy tale at a high level of abstraction; (d) the comments made in the books were mere criticism of the government, or criticism of a political nature, even if some comments are vigorously and strongly worded, expressing disapprobation of actions of the Government of HKSAR without exciting or causing public disorders by acts of violence; (e) the books did not incite or impose any negative thought to the readers, and each book was concluded with an open-ended question, leaving it to the readers to decide what is best for them; (f) None of the words in Book 2 had lobbied for any foreign intervention or international sanction against HKSAR and PRC and their respective officials; (g) The books were an aid or medium to help parents and educational workers explain to children the happening of the social events in Hong Kong in the past year; and (h) The books keep a record of the happening of the social events in Hong Kong. 118. I have read the three books and considered if all or any of them are publication having a seditious intention. I bear in mind that the target reader can be as young as 4 years old. According to the recommendation printed at the back of each book, it is recommended that for children aged between 4 and 7, parents may read the book together with them. For children aged over 7, they can read it themselves. 119. After reading just the books themselves and without referring to any other materials including what was said by some defendants about the books, the broad brush but deep impression was that the wolves are evil and the sheep are kind. The wolves are aggressors and the sheep are the oppressed. The wolves are to be hated and the sheep are to be glorified. Reading just one of the books is sufficient to produce this impression. 120. Just as a story between the sheep and the wolves as depicted in the books, it is not wrong for the defence to say that it was just a fable teaching some universally celebrated virtues. The wolves represent the wicked. The Wolf-chairman is a tyrant. He directs the wolves under him to attempt occupying the sheep village and ruin the peaceful and happy life of the sheep. The Wolf-chairman and the wolves are the villains and must be hated. They are to be kept away. On the other hand, the sheep representing the virtues were to be glorified for their courage and relentless effort in protecting their home. They are to be supported. There is of course nothing wrong to teach children virtues like these. There is also nothing wrong to tell them they should protect their home if some evil people come to harm them and they should resist. 121. However, a fable only promotes universal virtues, or tells the moral of a story. It will not identify the real characters, so as not to accuse anyone in the wrong. But this is not the case in any of the books. In the foreword of Book 1, it was stated explicitly by the 1st Executive Council of GUHKST that “2019 was turning point in Hong Kong. The anti-legislation movement wakes us up to realize that we are just frogs being boiled in warm water, and that One Country, Two Systems, separation of power, and freedom of expression were just veils over eyes that are torn apart …”. The specific reference to the “anti-legislation movement” clearly referred to the Anti-Extradition Bill Movement. In the epilogue at the end of Book 1 (or Timeline as Counsel for D2 calls it), it was stated explicitly what was the particular event that certain pages of the book were talking about. The title of this epilogue was “羊村守衛戰 VS 反送中運動”, and in the next line “背景資料”. It is argued that “VS” stands for “versus”, so that there cannot be any substitution of the events or characters in the Timeline (or in real life) for any of the characters or events in the story. In my opinion, the “VS” simply meant “contrast with”, and this page made it clear which page of Book 1 was talking about which real life events as particularized in the Timeline. 122. Counsel for D4 submits that Book 1, as well as the other books, are just record of what had happened in Hong Kong, and it just represents the perspective of a section of the public, and the events may be exaggerated, but this form of writing is acceptable in all broadminded democratic societies, and that this is so even if some readers may find the contents disturbing, shocking or even offensive. I agree with counsel, but still, it depends on what has really been depicted. 123. The biggest problem about Book 1, indeed about Books 2 and 3 as well, is that after the story is told, the children are to be told that the story is real. They will be told that in fact, they are the sheep, and the wolves who are trying to harm them are the PRC Government and the Hong Kong Government. 124. I shall not repeat the plot of each book here. It is not necessary to put them onto the record. By identifying the PRC Government as the wolves, and the Chief Executive of HKSAR as the wolves masqueraded as a sheep at the direction of the Wolf-chairman, along the story line told in Book 1, the children will be led into belief that the PRC Government is coming to Hong Kong with the wicked intention of taking away their home and ruining their happy life with no right to do so at all. The publishers of the Book clearly do not recognize that the PRC has legitimately resumed exercising sovereignty over HKSAR, but the children will be led to hate and excite their disaffection against the Central Authorities. The children will also be led to believe that the Chief Executive of HKSAR is sent by the Central Authorities with the ulterior motive of hurting them. The publishers clearly refuse to accept the constitutional order of HKSAR after the Handover, and lead the children to look down on the Chief Executive of HKSAR with contempt. The children are also led to believe that new immigrants from PRC are sent here deliberately to use up their resources. The publishers refuse to accept the immigration arrangement agreed between the PRC and the HKSAR Governments and lead the children to feel discontent with the new immigrants. The children will also be led to believe that if they are not obedient, they will be sent to prison. The publishers therefore lead the children not to trust the administration of justice in Hong Kong and look down upon the police, the prosecution and the court with contempt. The children are also led to believe that the Extradition Bill is a tool to suppress dissenting Hong Kong residents and subject them to arbitrary arrest, and they may even be sent to prison in China. The children will be led to believe that the only way to protect their home is to resist and to use force if necessary against the authorities. In similar vein, by Book 2, the children will be led to believe that the 12 fugitives are victims of oppression and unfair prosecution, and that they are forced to leave their home in short notice, only to find that they had already been closely monitored by the wicked force to be taken to prison. The children will be led to believe that these 12 fugitives are unfairly detained in PRC. In Book 3, the children will be led to believe the Government deliberately allowed people coming from China to make their home dirty and spread the pandemic. The publishers incite discontent or disaffection amongst Hong Kong residents. 125. It is alleged by the defence that what the defendants intended to do was to leave a record of the event. It is submitted that the court should not decide what the defendants had said is true or not. I agree that this is the state of the law at this point of time, and the question of misinformation and disinformation may have to be addressed by other legislation. However, what has happened here is that the publishers of the books clearly refuse to recognize that PRC has resumed exercising sovereignty over HKSAR, nor do they recognize the new constitutional order in the Region, and lead the children to think that what the authorities both in PRC and HKSAR have done is wrong and illegitimate. 126. It is also submitted that the Books did not mean to indoctrinate any sort of ideas into the mind of the children. Great reliance is placed on the open question at the end of each book for the children to answer. It is said that children facing this kind of question came up with different answers, for instance, one said he would fight the wolves, and the other said he would hide out of fear. In my judgment, it is hypocritical to say that the children are allowed an open mind to decide the answer. It is patently clear from the structure of each book that the thinking of the children is to be guided in a particular way when the story is being told, and when their mindset is already confined in a particular mode, the so-called open answer from the child himself is no more but a guided answer. The danger of this type of answer is that the child may think that the idea is his, not knowing that it has been fed to him, rendering it extremely difficult to change the way of his thinking subsequently. 127. There is no submission that the publication is not seditious upon any of the section 9(2) ground. There is no evidence to this effect in any event, so that there is nothing for the prosecution to disprove. 128. Upon the analysis above, I hold that each of Books 1, 2 and 3 is a publication having seditious intention. The seditious intention stems not merely from the words, but from the words with the proscribed effects intended to result in the mind of children as stated in section 9(1)(a), (c), (d), (f) and (g). Was there the alleged conspiracy, and if so, who took part in it? 129. In order to establish the offence of conspiracy, the prosecution has to prove the following elements[40]: (a) An agreement between two or more persons. (b) The future course of conduct agreed under that agreement to be pursued. (c) The intentions of the parties under that agreement. (d) If the agreement is carried out in accordance with the parties’ intentions, this will necessarily amount to or involve the commission of an offence by one or more of the parties to the agreement or would do so but for the existence of facts which render the commission of the relevant offence impossible. 130. There is no dispute that Book 1, Book 2 and Book 3 are the publications of GUHKST. This per se provides direct evidence that all executives of the Executive Council of GUHKST had participated in this agreement to print, publish, distribute, display and reproduce the three picture books unless there is evidence that may prevent the inference to be drawn. 131. As to who were the executives at the material times, the evidence has proved beyond reasonable doubt that D1 to D5 and Wong Hoi-ching were the executives of the 1st Executive Council of GUHKST from 12 December 2019 until the date of arrest on 22 July 2021. 132. There is no evidence that after the defendants and Wong Hoi-ching had changed their duty posts after the election into the 1st Executive Council. It therefore means that, at all material times, D1 was the Chairperson, D2 and Wong Hoi-ching were the 2 Vice–chairpersons, D3 was the Secretary, D4 was the Treasurer, and D5 was the Officer of the 1st Executive Council of GUHKST. 133. There is also no evidence to show that any of the defendants had not participated in the functioning of the Executive Council since their election as executives. On the contrary, Clause 7.3 of the Rules of GUHKST provides that Executive Council shall hold meeting at least once every two months, and the quorum has to be made up by at least half of the executives[41]. The only reasonable inference is that the defendants must have been performing their function as the executives. 134. There is also clear evidence that GUHKST was intended to be used as a political platform and that each of these defendants agreed to it. Their intention was manifested in the declaration of political stance made by them before and after their election into the Executive Council of GUHKST that they would stand for the Powerless and the Voiceless, and pursue for the 5 Demands[42]. It is submitted that GUHKST was not a political platform because its executives had also participated in a lot of other work in the name of the trade union. Such work included communicating with government departments, Education Bureau, Social Welfare Department and employers about measures to be taken against Covid-19, and work involving its members. I accept that such work has been done, but it is clear and it is only common sense that all these work can be done at the same time together with publishing the picture books. In my view, GUHKST was clearly set up for political purposes, and D1 had said so in one of the radio interviews. D1 said, “工會是政治能量的載體。” (Exhibit P34(d)). 135. Because of this reason, it is not surprising that GUHKST will publish the picture books in question. In fact, there are many publications of GUHKST which must have been prepared and/or approved by the defendants as the executives of the Executive Council which revealed their connection with the publication of Books 1 to 3. 136. In the 2019 – 2020 Annual Report[43], it was reported that the Executive Council had created an Anti-Extradition Movement story and invited illustrator Rainection to draw pictures, and the books were printed in “yellow shops”, and the Executive Council had also created the related “Reading-together guide”. 137. In Newsletter No. 3 [Exhibit P36(c)], it was reported that “the executives had created a story book about the Anti-Extradition Movement, using metaphors to narrate the causes of the Movement, with a view to using the expertise of speech therapists to help parents to grasp the skill of explaining social events to children[44]”, and that the image of Book 1 was printed. Surrounding this image of Book 1 was the following summary about the publication and production of Book 1: (a) books were distributed in about 75 “yellow” shops, district counsellors’ offices and organizations; (b) illustrator Rainection was invited to assist in the drawing with printing then done in “yellow” shop; (c) expenses HK$14,385.53, donation received HK$18,546.8; (d) people reached: 217,365; (e) interviewed by 3 media (inmediahk, Apple Daily, and Radio Free Asia); (f) e-version and reading-together guide uploaded, and audio book produced in collaboration with Hang Tan Collage; (g) 3,000 copies printed after increasing printing twice; and (h) sharing 1,165 times[45]. 138. In Newsletter No. 5 [Exhibit P36(e)], its first page was printed with an image of Book 2, and its third page contained the following report: “producing [title of Book 2] children picture book, placed inside about 30 shops with good conscience for citizens to collect for free, and received interview by two media[46]”. 139. Pursuant to section 65C, Criminal Procedure Ordinance, the defendants admitted that Newsletters No. 3 and 5 had announced news concerning publications of the picture books. 140. From all these Annual Report and Newsletters about publication of Book 1 and 2, in the absence of any evidence to the contrary, the only reasonable and irresistible inference is that each one of them was aware of the publication of Books 1 and 2, and that they agreed to the publication of Books 1 and 2 by the Executive Council in which he or she was an executive. 141. In the platform booklet of the 2020 – 2021 Executive Council of GUHKST (Exhibit P35B), it was stated that Book 2 was published by the Trade Union as part of propaganda education. It was further stated under the column of “future work direction” that “produce new picture books”(製作新繪本)[47]. Since this platform booklet set out the working directions of the Executive Council for the year 2020 – 2021, i.e. the second year of the term of office of D1 to D5 and Wong Hoi-ching, in the absence of any evidence to the contrary, the only reasonable and irresistible inference to be drawn is that all of them had agreed that further picture books similar to Books 1 and 2 will be printed, published, distributed, displayed, etc. As a matter of fact, Book 3 was subsequently published on 16 March 2021, and Book 4 was in the process of being prepared when all the defendants were arrested. The truth of the statements made by the defendants in their working platform has been confirmed. 142. Other than being one of the executives in the Council, there are other evidence linking each of the defendants to the agreement for the printing, publication, distribution, display, etc. of Books 1, 2 and 3. 143. D1 had attended 4 media interview to promote Book 1, one media interview to promote Book 2, and one media interview to promote Book 3. D1 is also the administrator of the website and the social media accounts in Facebook, Instagram and Telegram of GUHKST by which the books were published online, D1 liaised with shops for distributing Books 1 to 3, D1 distributed Book 3 in the street, D1 at the time of her arrest possessed at her home 2 hardcopies of Book 1, 3 hardcopies of Book 2 and 3 hardcopies of Book 3 (and that she has more than one copy of each book leads to the irresistible inference that she possessed these books for the purpose of publication), D1 was in possession of the screen shot relating to draft content of Book 4 in her mobile phone and computer. 144. D2 had attended 3 media interview to promote Book 1, one media interview to promote Book 2, and one media interview to promote Book 3, D2 distributed Book 3 in the street, D2 approached shops for display or distribution of hardcopies of Book 3, D2 held one Parent-Children Reading Together Session with D4, D2’s phone number was given to the printer of the books and settled 5 invoices with monies from her bank account, D2 at the time of arrest possessed at her home 3 hardcopies of Book 1, 13 hardcopies of Book 2, and 47 hardcopies of Book 3 (and that she has more than one copy of each book leads to the irresistible inference that she possessed these books for the purpose of publication). 145. D3 at the time of arrest possessed at her home 12 hardcopies of Book 1 and 5 hardcopies of Book 3 (and that she has more than one copy of each book leads to the irresistible inference that she possessed these books for the purpose of publication). 146. As to D4, screen captures of Book 1 and the images of promotion banners of Book 1 and Book 3 were found in D4’s mobile phone. Softcopies of Book 4 were found in D4’s computer. D4’s phone number was provided to the printer of the books, digital images of 3 invoices for printing books were found in D4’s computer, D4 liaised with shops for distribution of books, D4 held the 1st Parent-Children Reading Together Session together with D2, D4 has at the time of arrest possessed at his home 3 hardcopies of Book 1, 6 hardcopies of Book 2, and 5 hardcopies of Book 3 (and that he has more than one copy of each book leads to the irresistible inference that he possessed these books for the purpose of publication). 147. From D5’s mobile phone, draft note of Book 3 was found. Screen captures of Book 4 were found via the logged in Google drive account of D5, and images of promotion banners of Book 2 and Book 3. D5 possessed a PowerPoint file of the draft presentation for the 1st Parent-Children Reading Together Session in his mobile phone, Instagram and WhatsApp messages in relation to the 1st and 2nd Parent-Children Reading Together Session (including confirmation messages to the applicants), and images in relation to promotion of the 2nd Reading session. D5 liaised with shops for distribution of books. D5 has at the time of arrest possessed at his home 17 hardcopies of Book 1, 6 hardcopies of Book 2, and 1 hardcopies of Book 3 (and that he has more than one copy of each book leads to the irresistible inference that he possessed these books for the purpose of publication). 148. Since it is admitted by all defendants pursuant to section 65C that at the time of their arrest, hardcopies of Books 1, 2 and 3 were found in their possession, and since each one could have access to the website and social media accounts of GUHKST, in the absence of any evidence to the contrary, the only reasonable and irresistible inference to be drawn is that they each knew the contents of each book, and agreed for them to be published. For this reason, I also find that each one of them had a seditious intention to publish the books. 149. I also find that it has been proved beyond reasonable doubt, from the roles played by each of the defendants as particularized above, each of them had between 4 June 2020 and 22 July 2021 entered into an agreement amongst themselves and with Wong Hoi-ching and others to print, publish, distribute, display, and/or reproduce Books 1, 2 and 3, knowing that the books printed or to be printed are publications with seditious intention. I find that they intended to have this agreement carried out, and in the course of carrying out the agreement, Books 1, 2 and 3 had already been published, and Book 4 was being prepared. 150. Subject to the time bar argument, I shall convict D1 to D5 of the Charge. Time bar 151. Section 11 of the Crimes Ordinance provides that “No prosecution for an offence under section 10 shall be begun except within 6 months after the offence is committed.” On the other hand, Section 159D provides in essence that if an offence has been committed pursuant to a conspiracy, and there is a time limitation for prosecuting that substantive offence, once prosecution of the substantive offence is already time-barred, the prosecution cannot circumvent the time bar by prosecuting the defendant for entering into a conspiracy to commit that offence. The legal principle is clear. 152. Counsel for the defendants submit that there was only one single agreement according to the prosecution, and the publication of the 3 books are simply different parts of that single transaction.It is pointed out that Book 1 was published on 4 June 2020. Book 2 was published on 19 December 2020. Book 3 was published on 16 March 2021. No charge was laid against D2 until 23 July 2021. It is then submitted that the Charge was out of time in so far as it related to the substantive publications of Book 1 and Book 2. 153. The defence points out that according to section 159A of the Crimes Ordinance, it is possible for a conspiracy to involve more than one offence. It is submitted that the prosecution has formulated the Charge as one single offence of conspiracy in relation to the three books. The offence of conspiracy was therefore complete when the agreement was made. If any single offence has been committed pursuant to the agreement alleged, the limitation period applies, and no section 159A proceedings for a conspiracy involving that underlying substantial offence should be initiated. 154. The submissions by the defence must be rejected. Section 159A of the Crimes Ordinance has made it clear that if a person agrees with another person or persons that a course of conduct shall be pursued which, if the agreement is carried out in accordance with their intention, will necessarily amount to or involve the commission of any offence or offences by one or more of the parties to the agreement, he is guilty of conspiracy to commit the offence or offences in question. In other words, if there is an agreement to pursue a course of conduct that involves the commission of a series of substantive offences on a continuous basis, that person is guilty of a conspiracy to commit the offences in question, and that conspiracy will only come to an end when that person and the others agree that they will no longer pursue that course of conduct, or he himself withdraws from the agreement. 155. In the present case, it is stated in the Particulars of the Charge that the alleged offence took place between 4 June 2020 and 22 July 2021. The prosecution is clearly alleging that the conspiracy in question was an offence continuing throughout this period of time. It is also clear from paragraph 3 of the prosecution’s amended opening that it is alleging that the defendants and the others “had embarked upon an agreement to cause Book 1 to Book 3 to be printed, published, distributed, displayed and/or produced”. The evidence has also shown that the three books were printed and published as a series of books, and there is also evidence that Book 4 was about to be printed and published as text image or e-edition of Book 4 were found in the electronic devices of D1, D4 and D5 at the time of their arrest. It is also a fact that when D2 and D4 were holding the Parents-Children Reading Together Session, they had arranged Book 1 as the book to read (which clearly proves that the three books form one series of publication), and D2 and D4 holding this kind of reading-together session with children and their parents in which they told the parents and children how to read the book was in fact an act amounting to publication of Book 1. Further Reading Session of this kind was arranged to be held on 25 July 2021, which did not take place just because the defendants were arrested. It is therefore beyond any shadow of doubt that the conspiracy entered into between the defendants had not come to an end before their arrest, and that the conspiracy was not just for printing of the book, but also for publishing, distributing and displaying the books as well. 156. Furthermore, even if the publication date of Book 3 were taken to be the date when the conspiracy ended (which should not be the case), prosecution of the offence would only be time-barred by 15 September 2021. For these reasons, I reject the defence submissions and hold that the prosecution is not time barred. 157. As the prosecution is not time barred, each of D1 to D5 is convicted as charged. [1] In the course of final submissions, counsel for D1 says that D1 will adopt the facts admitted between D4 and the prosecution, whereas counsel for D2, D3 and D5 say that their respective clients will not adopt such facts, but if the court finds any part of these facts is beneficial to the defence of their respective clients, the court should take those facts into account in its deliberation of verdicts. This court is prepared to proceed in the way submitted by counsel for D2, D3 and D5. This court will also extend the same beneficial treatment to D1. [2] FACV6, 7, 8 & 9/2020; [2020] HKCFA 42; (2020) 23 HKCFAR 518, at Section C.1 [3] FACC1/2021; [2021] HKCFA 3; (2021) 24 HKCFAR 33, para 12. [4] Full name of the Extradition Bill was “Fugitive Offenders and Mutual Legal Assistance in Criminal Matters Legislation (Amendment) Bill 2019”. [5] 連儂牆連儂隧道 [6] 黃色經濟圈 [7] In colloquial term: “裝修” [8] In colloquial term: “私了” [9] PW1 has also given evidence on this event. [10] PW1 has also given evidence on this event. [11] Part of Exhibit P133 (ie page 1705 in Trial Bundle Vol 3). Exhibit P133 were documents seized by the police on 22 July 2021 inside the Hung Hom office of GUHKST – see paragraph 31 of the Admitted Facts (Exhibit P1). [12] Part of Exhibit P133 (ie pages 1706 to 1714 in Trial Bundle Vol 3). A copy of the same set of Rules of GUHKST was also marked as Exhibit D4-2 in the Facts admitted between D4 and the prosecution (Exhibit D4-1). [13] The relevant motion and voting record is part of Exhibit P133 (ie page 1678 in Trial Bundle Vol 3). [14] The relevant motions and voting records are part of Exhibit P133 (ie pages 1679 to 1684 in Trial Bundle Vol 3). [15] See paragraph 4 of the Admitted Facts (Exhibit P1), the bank records (Exhibit P7a), and the banker’s affirmation (Exhibit P7b). [16] Transcripts of this video interview is Exhibit P34(b)(iii). [17] According to the screen capture, the English name of 杏壇放送學院was “Hang Ten Collage”, not “College”. Exhibit P16(b) is the transcripts of Exhibit P16(a). [18] Exhibit P17(b) is the transcripts of Exhibit P17(a). [19] Exhibit P34(g)(ii) is the transcripts of Exhibit P34(g)(ii)(V). [20] See paragraph 16 of the Admitted Facts (Exhibit P1) for the names of the shops. [21] See paragraph 19(i) of the Admitted Facts (Exhibit P1). [22] The promotion material had the title “《羊村守衛者》親子讀書會”, and the content reads “對象:幼稚園、初小小朋友”、“內容: 共讀繪本《羊村守衛者》”. [23] Photographs of the event showed that the words “《羊村守衛者》親子讀書會” appeared on the projector screen, and the note of thanks stated that “今日有幸能夠同一班小朋友一齊共讀《羊村守衛者》呢本書 …” (“it is a privilege to read [title of Book 1] together with a group of children today …”) [24] The original was in Chinese: “香港言語治療師總工會二零一九年度候選理事會政綱小冊子”. [25] The original was in Chinese: “香港言語治療師總工會 2020 - 2021年度理事會政綱小冊子”. [26] The original was in Chinese: “2019 – 2020年度周年報告”. Its cover page indicated that this report covered the period between December 2019 and July 2020 only. [27] The original in Chinese was: “第二期會訊 2019.02 – 2020.05”. Apparently, the title was incorrect because the period covered should be between February 2020 rather than February 2019. [28] “The 20 May 2020 Explanation” and “The 18 June 2020 Explanation” by the Standing Committee of the National People’s Congress referred to in the Prosecution’s List of Authorities. [29] The Law of the People’s Republic of China on Safeguarding National Security in the Hong Kong Special Administrative Region (“the NSL”) [30] (2021) 24 HKCFAR 417 [31] Decision of the Standing Committee of the National People’s Congress Concerning the Handling of the Laws Previously in Force in Hong Kong in Accordance with Article 160 of the Basic Law of the Hong Kong Special Administrative Region of the People’s Republic of China (adopted at the 24th Meeting of the Standing Committee of the Eighth National People’s Congress on February 23, 1997 (“BL 160 Decision”) [32] In Chinese: “《中華人民共和國憲法》確立的憲制秩序下的中央政權機關” [33] (1952) 36 HKLR 133 [34] (2007) 10 HKCFAR 324 [35] (1999) 2 HKCFAR 442 [36] (2012) 15 HKCFAR 16 [37] (2002) 5 HKCFAR 381, at para 4 [38] (2016) 19 HKCFAR 372 [39] (1868) 11 Cox CC 44, 45 [40] HKSAR v Harjani Haresh Marlidhar (2019) 22 HKCFAR 446 [41] Part of Exhibit P133 [p. 1710(a)] [42] Exhibit P35C, Exhibit P10 (page 1195 – 1196) [43] Exhibit P35A [44] The original Chinese text as appearing at page 1263 in Trial Bundle Vol. 2 was: “理事創作咗一個有關反送中運動嘅故事, 以比喻嘅方法敘述咗運動嘅因由, 期望用言語治療師嘅專業幫助家長掌握同小朋友解釋社會事件嘅技巧。” [45] The original Chinese text as appearing at page 1263 in Trial Bundle Vol. 2 was: “喺大約75間黃店/區議員辦事處/組織派發, 邀請咗插畫家Raineaction幫忙畫圖, 再喺黃店印繪本, 支出: HKD14,385.33 捐款收入 HKD18,546.8, 接觸對象217,365, 接受咗三個媒體嘅訪問(香港獨立媒體、蘋果日報同自由亞洲電台粵語部, 上載咗電子書、共讀指引, 交同《杏壇放送學院》合作製作有聲書, 加印咗兩次, 先後印咗3000本, 分享次數1165”。 [46] The original Chinese text as appearing at page 1263 in Trial Bundle Vol. 2 was: “製作《羊村十二勇士》兒童繪本, 放置喺約30間良心店舖畀市民免費索取, 接受咗兩間媒體嘅訪問”。 [47] Exhibit P35B (p. 756) 
Chief Justice Ma: A. INTRODUCTION 1. On 11 July 2016, the Court handed down judgment in HKSAR v Yeung Ka Sing, Carson[1] and HKSAR v Salim, Majed,[2] both appeals having been heard together. I will refer to the judgment in those appeals simply as “Yeung”. The four Questions of great and general importance heard in those appeals related to the offence of money laundering, a shorthand description for the offence contained in s 25(1) of the Organized and Serious Crimes Ordinance, Cap 455 (“OSCO”).[3] The present appeal, for which leave was given[4] before the appeals in Yeung were determined on the same four Questions, is the first appeal before this Court since the judgment in Yeung. 2. The four Questions for which leave was given in the present appeal were as follows:- “1. In the context of the offence of money laundering under section 25 of the Organized and Serious Crimes Ordinance, Cap 455 (‘the Ordinance’), how does the rule against duplicity operate? In particular, whether the offence of money laundering, capable of being committed in any of the modes of ‘dealing’ as included in its definition under section 2 of the Ordinance, is or could be a continuing offence so that the rule against duplicity does not apply; and how do the exceptions to the rule against duplicity (namely, ‘one transaction’ as in DPP v Merriman [1973] AC 584, ‘general deficiency’ as in R v Tomlin [1954] 2 QB 274 and ‘continuous course of conduct’ as in Barton v DPP [2001] 165 JP 779) apply to a charge of money laundering which alleges multiple dealings some of which involve money from known and different sources? 2. When considering whether a defendant had reasonable grounds to believe in the context of s 25(1) of the Ordinance, how does a trial judge reconcile the formulation set out in Seng Yuet Fong v HKSAR (1999) 2 HKC 833 and the formulation ‘knew or ought to have known’ set out in HKSAR v Pang Hung Fai (2014) 17 HKCFAR 778? Under what circumstances should the trial judge apply these two formulations? 3. On a charge of dealing with proceeds of crime contrary to s 25 (1) of the Organized and Serious Crimes Ordinance (Cap 455) (“OSCO”), is it necessary for the prosecution to prove, as an element of the offence, that the proceeds being dealt with were in fact proceeds of an indictable offence? Was Oei Hengky Wiryo v HKSAR (2007) 10 HKCFAR 98 wrongly decided on this issue? 4. In considering the mens rea element of a charge contrary to s 25(1) of OSCO, to what extent does a trial judge need to make positive findings as to a defendant’s belief, thoughts, intentions at the material time even though the judge rejects the defendant’s testimony? In particular, where the trial judge rejects the defendant’s testimony, to what extent can the judge remain oblivious to the defendant’s actual reason(s) for dealing with the specified proceeds in making the finding that the defendant had reasonable grounds to believe that the proceeds he dealt with were proceeds of crime?” 3. The legal arguments raised by the parties to the present appeal[5] in their original Cases have been dealt with and determined in Yeung. Notwithstanding this, the Appellant seeks to raise three arguments which I will presently identify. To dispose of some of these arguments, it will be necessary to go into the facts. B. THE FACTS RELIED ON BY THE PROSECUTION 4. The Appellant was convicted after trial[6] of three counts of dealing with property contrary to s 25(1) of OSCO, and sentenced to seven years’ imprisonment on each count to be served concurrently. His appeals against both conviction and sentence were dismissed by the Court of Appeal.[7] 5. The Appellant is a PRC citizen. In July 2010, he set up a company in Hong Kong, Qi Long Da Company (“QLD”).[8] In August that year, QLD opened a Hong Kong dollar and a foreign currency account with the Bank of China (Hong Kong) Limited; these were known as Accounts A and B at trial. In November 2010, he set up another company in Hong Kong, Ying Ge Sheng Company (“YGS”).[9] In December, YGS opened a Hong Kong dollar account with the Chiyu Bank; this was Account C at trial. The Appellant was the sole owner of these companies and was the sole signatory of the three accounts. 6. Accounts A and B were closed after seven months, Account C after only two months. However, during that time, the following activities took place in them:- (1) Account A: there were altogether 333 transactions consisting of 134 deposits and 199 outgoing remittances. The deposits totalled $54,653,131.03, the outward remittances $54,646,454.24. (2) Account B: there were altogether 617 transactions consisting of 289 deposits and 328 outgoing remittances. The deposits totalled US$51,674,795.89 and the equivalent of HK$402,029,912.02; the outward remittances amounted to US$51,672,487.79 and the equivalent of HK$402,011,955.01. (3) Account C: for this account, there were 56 transactions consisting of 20 deposits and 36 outgoing remittances. The deposits totalled $9,508,924.09, the outward remittances $9,508,476.82. Effectively the three accounts had deposits totalling about HK$450 million, with virtually the same amount being outwardly remitted all within a relatively short period of time. Neither the Appellant nor the two companies were licensed remittance agents or money changers. 7. The three counts with which the Appellant was charged were in relation to dealings in these accounts. Though it was alleged in each count that the Appellant knew or had reasonable grounds to believe that the monies in these accounts represented the proceeds of an indictable offence, like many other money laundering cases, the prosecution relied only on the “reasonable grounds” part of s 25(1) of OSCO. C. THE APPELLANT’S POSITION AT TRIAL AND THE JUDGE’S FINDINGS 8. The Appellant did not dispute the facts outlined above. His case was that he ran an underground banking operation whereby monies from clients would be deposited into the three accounts and from there, monies would then be remitted into accounts outside Hong Kong. The reason for operating an “underground” (meaning unlicensed, unregulated and secretive) bank was said to be exchange control restrictions. Monies were deposited into the Hong Kong accounts from various countries in Africa, Asia and the Mainland and then outwardly remitted mainly to the Mainland and occasionally to Taiwan. 9. The Appellant accepted that in the operation of an underground bank in this way, there was a risk that money laundering would take place. The Appellant’s case (and this really became the focus of the trial) was that he had all along performed “gatekeeping” precautions, meaning that by reason of the risk of money laundering, he was at pains to take precautionary measures in operating the three accounts. Accordingly, it was contended that he lacked the necessary mens rea in relation to the offences with which he was charged; he did not have reasonable grounds to believe that the monies in the three accounts “in whole or in part directly or indirectly” represented any person’s proceeds of an indictable offence. 10. The trial Judge heard evidence that the monies which were deposited into the three accounts came from essentially four different sources. In relation to each source, the Appellant gave evidence going to the precautions he took or which were taken to ascertain the source and veracity of the funds deposited into the accounts, to investigate the background of the clients making the deposits and to gather relevant documents in these respects. 11. In reaching his verdict, the Judge analysed the evidence put forward by the Appellant, including his own testimony. In relation to each of the four sources of funds which were deposited into the three accounts, the Judge was not satisfied that there was any credible or reliable testimony that the Appellant had performed any “gatekeeping” precautions. The Judge found that the Appellant could not have been satisfied of the integrity of the people with whom he dealt and no meaningful attempts were made to be satisfied of the identities of the depositors nor of the origins of the monies and such documents as there were were either inconsistent or otherwise unconvincing. In the course of his analysis, the Judge made references to the Appellant deliberately turning a blind eye, to his disbelieving the Appellant, fabrications, inconsistencies, documentary irregularities etc. The Judge concluded that the Appellant had failed to discharge any “gatekeeping” responsibilities and found that there were sufficient grounds for any “commonsense, right thinking” person to believe that the depositors of the monies were laundering money[10] and there was no basis for the Appellant to think otherwise. In paragraph 77 of the Reasons for Verdict, the Judge said this:- “The Defendant understood that the deposits might come from illegal dealings and he had to perform gatekeeping and supervision to prevent clients from laundering money through him. The Defendant did not know much about his clients. In respect of Shi Xubo personally, Shi Xubo’s clients, Yang Wendan personally, Yang Wendan’s clients, Ke Chengjie personally and other clients, I am of the view that there was no basis for him to believe that these people would not launder money through him.” 12. In the Court of Appeal, one of the Appellant’s arguments was that the Judge had erred in his analysis of the evidence. The Court concluded this point to be entirely without merit and ought not even to have been raised. D. ISSUES IN THIS APPEAL 13. I have earlier identified the four Questions of great and general importance for which leave was given in the present appeal[11] and mentioned the fact that these Questions have now been determined by this Court in Yeung. In spite of this, the Appellant however now wishes in this appeal to raise the following points said to arise from Yeung:- (1) Where the subject property in the charge under s 25(1) of OSCO has emanated from some activity which has taken place overseas, by reason of the operation of s 25(4) of the Ordinance, it must be proven as part of the mens rea of the offence that an accused was aware that the relevant property emanated from conduct that would constitute an indictable offence in Hong Kong if it had occurred here. (The s 25(4) point). (2) Where the relevant property in a charge under s 25(1) of OSCO consists of a chose in action (such as a deposit of money into a bank account which creates a debt as between the bank and the customer), each chose in action must be separately charged and it is not possible without offending the rule against duplicity to make a global charge aggregating a number of different items as in the present case with three charges faced by the Appellant. (The duplicity issue). (3) The trial Judge was wrong to regard the mere operation of an underground bank as being sufficient to establish the actus reus of the s 25(1) offence. Further, the Judge was wrong to regard the mens rea of the offence to be satisfied merely by the Appellant being aware of the risks of money laundering in the running of an underground banking operation. The Judge’s findings on mens rea were also to be invalidated on the ground that he applied an objective test. (Factual issues). E. THE s 25(4) POINT 14. The actus reus of the offence in s 25(1) consists of dealing with property. The mens rea of the offence in s 25(1) consists either of actual knowledge that the relevant property in the charge “in whole or in part directly or indirectly represents any person’s proceeds of an indictable offence” or that the accused has reasonable grounds so to believe. The mens rea requirements of the offence have been extensively considered by this Court, most recently in Yeung. It is unnecessary to go through this exercise again. Where a charge is made under the “knowing” limb of s 25(1), the requirements are clear. So too are the requirements under the “reasonable grounds” limb: here, as reiterated by Yeung, one is concerned with the question whether the accused had reasonable grounds to believe that the property dealt with represented the proceeds of crime. Under this second limb of s 25(1), there is no requirement that it must be proved that the accused was actually aware of the crime that produced the relevant property dealt with by the accused. Mr Lee submits, however, that it must be shown at least the accused knew of the nature of the activity that has brought about the property dealt with. With respect, that is not what the provision says. Ms Yang referred the Court to the following passage in Yeung:- “44. The current offence therefore focuses on the property – “any property” – and on the circumstances surrounding the defendant’s acts of dealing with that property. Obviously, if he can be proven to have known that the property represented someone’s proceeds of an indictable offence, the offence is established. But, if the defendant does not know for a fact that the property represented such proceeds (as is likely to be the case with professional money launderers who would make it their business not to know), it is sufficient for the prosecution to establish that, given the circumstances of which he was aware, surrounding his dealing with the relevant property, the defendant had reasonable grounds to believe that it represented the proceeds of someone’s indictable offence, whether committed in Hong Kong or abroad.” 15. S 25(4) takes the matter no further. It is a provision that is of relevance to overseas conduct. This provision qualifies the reference to “an indictable offence” in s 25(1) (and s 25A) by stating that this includes a reference to conduct which would constitute an indictable offence if it had occurred in Hong Kong. So, for example, where it is shown that the relevant property came from some identifiable act or acts committed overseas, it will be caught by s 25(1) if such conduct constituted an indictable offence in Hong Kong, irrespective of the legal position in the jurisdiction where such conduct took place. As the provision itself clearly implies, the focus is on the position under Hong Kong law, not on the legality of such conduct under any foreign law.[12] 16. The Appellant, however, relies on s 25(4) to support the argument that where, as in the present case, property has originated from overseas, the prosecution is required to identify the underlying conduct and then prove that an accused knew of both the conduct and the criminal nature of that conduct, and it has also to be further shown by the prosecution that such conduct amounted to an indictable offence in Hong Kong if it had occurred here. Accordingly, it was said this provision supported Mr Lee’s submission that it was necessary to show the accused knew of the nature of the activity that produced the relevant property dealt with by the accused. This argument, however, misunderstands the nature of the offence contained in s 25(1) and must be rejected. The mens rea of the s 25(1) offence is as stated above. Under that provision, it is not a requirement that the underlying predicate offence be identified and this is the position confirmed by this Court in Yeung.[13] The provision simply does not require proof of knowledge of the actual conduct that has produced the relevant property. F. THE DUPLICITY ISSUE 17. This issue can be disposed of shortly. In Yeung, as in the present case, there were involved substantial numbers of deposits of money into the five accounts forming the subject matter of the five counts which Mr Yeung faced. This Court dealt with the same arguments raised in this appeal, namely that given the different provenance of the multiple deposits of money in bank accounts, the charges were said to be duplicitous.[14] The Court held on the facts of Yeung, however, that it was proper for the transactions in the charges to be aggregated and that this did not fall foul of the rule against duplicity, given that the multiplicity of dealings had the common purpose of concealment.[15] In the present case, the three accounts which were the subject matter of the three charges faced by the Appellant were used as part of an underground banking operation and this was admitted by him. Numerous deposits were made followed by withdrawals, all taking place within a relatively short period of time. As we have seen, the total amount of deposits were more or less matched by a corresponding amount of withdrawals. Each account had these common features. The defence run by the Appellant was also the same in relation to each account; that he had performed “gatekeeping” precautions. In these circumstances, and also given the fact that no issue on duplicity and particularly no issue on any unfairness arising from duplicitous charges having been raised at or before trial, the three charges were unobjectionable. G. FACTUAL ISSUES 18. A number of points were made by the Appellant on the facts. Irrespective of whether, strictly speaking, the Appellant should have been allowed to raise these points since leave was only given in relation to the four points earlier set out and it was not clear that these points arose from the decision in Yeung, we nevertheless heard Mr Lee SC on them. They, however, have no merit. 19. First, it was simply wrong to contend that the Judge regarded the mere operation of an underground bank as being sufficient to establish the actus reus of the offence. The Judge, in the course of his Reasons for Verdict, analysed the activities in the three bank accounts in the context of the underground banking operations run by the Appellant. There can be no doubt the activities in the accounts constituted dealing for the purposes of s 25(1). The Judge also carefully examined his assertion that he had performed “gatekeeping” precautions. 20. Secondly, as seen earlier, the Judge was not satisfied by the Appellant’s evidence and concluded that there were sufficient grounds for any “commonsense, right thinking person” to believe that the persons who made deposits in the accounts were laundering money.[16] By reason of the reference to “commonsense, right thinking” persons, it was then submitted by Mr Lee that the Judge had applied the wrong test. The correct test is that as established by the decision of this Court in HKSAR v Pang Hung Fai[17] as explained in Yeung. The Court of Appeal was of the view, however, that despite the Judge having applied the wrong test[18], it made no difference to the result since the trial Judge did consider the offence from the point of view of the Appellant. As the Reasons for Verdict show, the Judge analysed and made findings on what facts were known to the Appellant and also considered his perception of the operation he was engaged in. The trial Judge having rejected the Appellant’s evidence that he had performed “gatekeeping” precautions, what was left were a number of highly suspicious activities in the three accounts, of which the Appellant was aware. As stated before, the Appellant accepted that his operation of an underground bank did involve the risk of it being used for money laundering purposes. I find no error with the Court of Appeal’s conclusions here. 21. Thirdly, in relation to the mens rea of the offence, the Judge did not, as Mr Lee submitted, simply regard the Appellant’s awareness of the risk of money laundering in the operation of his underground banking activities as being by itself enough to satisfy the mens rea of the offence. As shown earlier, the Judge analysed in some detail the underground banking operations in the present case as well as the Appellant’s evidence and contentions in relation to these operations. H. CONCLUSION 22. For the above reasons, this appeal is dismissed. Mr Justice Ribeiro PJ: 23. I agree with the judgment of the Chief Justice. Mr Justice Tang PJ: 24. I agree with the judgment of the Chief Justice. Mr Justice Fok PJ: 25. I agree with the judgment of the Chief Justice. Lord Neuberger of Abbotsbury NPJ: 26. I agree with the judgment of the Chief Justice. Mr Robert S K Lee SC, Mr David Iu and Ms Sharon Chan, instructed by Cheung & Liu, for the Appellant Ms Maggie Yang, SADPP of the Department of Justice, for the Respondent [1] FACC Nos. 5 and 6 of 2015. [2] FACC No. 1 of 2015. [3] Section 25(1) provides:- “Subject to section 25A, a person commits an offence if, knowing or having reasonable grounds to believe that any property in whole or in part directly or indirectly represents any person’s proceeds of an indictable offence, he deals with that property.” [4] By the Appeal Committee (Ma CJ, Ribeiro and Tang PJJ) on 19 August 2015. [5] The Appellant is represented by Mr Robert Lee SC, Mr David Iu and Ms Sharon Chan. The Respondent is represented by Ms Maggie Yang. [6] On 23 January 2014 before HH Judge E Yip. [7] On 24 March 2015 (Yeung VP, Poon and D Pang JJ). D Pang J (now Pang JA) gave the Judgment of the Court. [8] 奇隆達公司. [9] 英格盛公司. [10] Paragraphs 82, 84, 86 and 88 of the Reasons for Verdict. [11] See paragraph 2 above. [12] See the decision of the Appeal Committee in Lok Kar Win Kevin v HKSAR, FAMC 27 of 1999, 25 November 1999 at paragraph 9. [13] See Section C of Yeung. [14] See paragraph 162 of Yeung. [15] See paragraph 166 of Yeung. [16] See paragraph 11 above. [17] (2014) 17 HKCFAR 778. [18] Namely the test in the pre-Pang Hung Fai decision of the Court of Appeal in HKSAR v Shing Siu Ming [1999] 2 HKC 818. 
------------------------------ Chief Justice Ma: 1. I agree with the judgment of Lord Phillips of Worth Matravers NPJ. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Lord Phillips of Worth Matravers NPJ. Mr Justice Tang PJ: 3. I agree with the judgment of Lord Phillips of Worth Matravers NPJ. Mr Justice Bokhary NPJ: 4. I agree with the judgment of Lord Phillips of Worth Matravers NPJ. Lord Phillips of Worth Matravers NPJ: Introduction 5. On 12 July 2007 the first appellant (“Vand”) entered into a share purchase agreement (“the SPA”) with the respondent (“Pony”) under which it agreed to sell to Pony 370 million shares in a Hong Kong Company called Hans Energy Company Limited (“Hans”). Completion took place on the following day. The second appellant, Mr David An, guaranteed Vand’s obligations under the SPA. The SPA included a put option under which, on the occurrence of certain specified events, Pony would be entitled to sell the shares back to Vand. On 6 July 2009 Pony purported to exercise that option. This action raises the issue of whether the put option was validly exercised. On 30 November 2011 Reyes J gave a judgment in which he held that the put option had been validly exercised. He subsequently ordered Vand and Mr An to pay Pony the put option price of HK$510,708,649.03, together with interest. 6. The appellants appealed against the order of Reyes J. On 30 October 2012 the Court of Appeal dismissed their appeal. Kwan JA gave the only reasoned judgment, with which Fok JA and McWalters J agreed. Vand and Mr An now appeal as of right to this Court. Background to the SPA 7. One of the attractions to Pony of the SPA was an indirect interest that Hans enjoyed in a petro-chemical dock construction project (“the Dock Construction Project”) at the Guangdong Dongguan Humen Bay Shatian Bay Region in the People’s Republic of China (“PRC”). The Construction Project was to be carried out by Dongguan Dongzhou International Petrochemical Storage Company Limited (“Dongguan Dongzhou”), a company formed specifically for this purpose. 8. The Dock Construction Project required approval by, among others, the National Development and Reform Commission of the PRC (“NDRC”). That approval had been obtained on 5 September 2005. At that time, however, Dongguan Dongzhou had not yet been incorporated. The approval was granted on the basis that Dongguan Dongzhou would be a joint venture company in which the majority shareholding would be owned, indirectly, by Hans, but a 15% interest would be held by Dongguan Humen Port Holding Company (“DHPC”), a company owned by the local municipal government. 9. In circumstances that do not appear from the judgments below, and are not material, DHPC’s interest in the Dock Construction Contract was subsequently bought out by (Guangdong (Panyu) Petrochemical Storage & Transportation Limited (“Yuehai”). This was one of the Hans group of companies, which consequently became the owner of 15% of the shares in Dongguan Dongzhou. This change in shareholding had the approval of the local municipal government, but had not been approved by NDRC. 10. The lawyers advising Pony on the drafting of the SPA had two relevant concerns. It is reasonable to conclude that these were communicated to Vand and Mr An, so that it is legitimate to have regard to them as relevant background to the SPA. The first concern was that Yuehai had not obtained a necessary certificate of approval by NDRC for storing product oil in its facilities. 11. The second, and material, concern was expressed in a report to Pony by Clifford Chance. Privilege was waived in respect of the relevant passage, which commented as follows on the change in the ownership of Dongguan Dongzhou after the grant of approval by NDRC: “We understand that the failure to obtain NDRC’s approval for the changes in equity holders may cause the NDRC to suspend the project operations and impose administrative liability on persons who are responsible for the failure.” The relevant provisions of the SPA 12. The parties agreed to the insertion into the SPA of the following provisions in order to meet the two concerns referred to above. “6.2 Covenants by the Vendor (A)The Vendor hereby covenants with the Purchaser that it shall procure the Company [Hans] to carry out all reasonable steps (i) to obtain the final approval certificate for storage of product oil from the relevant government authority in respect of the facilities owned by (“Yuehai”) for the purpose of storing product oil in accordance with the applicable laws and regulations as soon as possible; and (ii) to obtain the necessary approval from, and to carry out the relevant filings with, the National Development and Reform Commission of the People’s Republic of China (中华人民共和国国家发展和改革委员会) in respect of the change in ownership of the petro-chemical dock construction project at the Guangdong Dongguan Humen Bay Shatian Bay Region as soon as possible ((i) and (ii) together the “Approvals”). (B) The Vendor further covenants with the Purchaser that it shall furnish to the Purchaser copies of all relevant official receipts, confirmations or such other documents from the relevant government authorities with respect to the Approvals. The Vendor shall keep the Purchaser informed with respect to the application procedures and provide the Purchaser with the right to access all relevant information and any communications with the relevant government authorities as may be reasonably required by the Purchaser and the right to enquire the progress of the applications from time to time. (C) The Vendor hereby grants an irrevocable put option (the “Put Option”) to the Purchaser to sell, and to require the Vendor to buy, all of the Put Option Interest subject to the terms and conditions of this Clause 6.2. (D)The Put Option may be exercised by the Purchaser upon the occurrence of any of the following events at the Put Option Price: (1) the failure of Yuehai to obtain the final approval certificate for storage of product oil (成品油仓储经营批准证书) from the relevant government authority in respect of the facilities owned by it for the purpose of storing product oil in accordance with the applicable laws and regulations, at the fault of, or arising from an act or omission of, the Vendor or Yuehai; or (2) the failure of Yuehai to obtain the necessary approval from or to carry out the necessary filing with, the National Development and Reform Commission of the People’s Republic of China (中华人民共和国国家发展和改革委员会) for the change in ownership of the petro-chemical dock construction project at the Guangdong Dongguan Humen Bay Shatian Bay Region, at the fault, or arising from an act or omission of, the Vendor or Yuehai. (E) Within 30 Business Days upon the occurrence of any of the triggering events as referred to in Clause 6.2 (D) above, the Purchaser may exercise the Put Option by delivering a put option notice (the “Put Option Notice”) to the Vendor … (F) Notwithstanding any provisions herein contained to the contrary, the Put Option granted under this Clause 6.2 shall lapse automatically upon the expiry of two years from the Completion Date.” 13. Clause 8.3 of the SPA provided that neither the vendor nor the guarantor should have any liability in respect of any claim under the SPA unless notice of the claim was given within 24 month of the date of completion. 14. Clause 8.6 provided that neither the vendor nor the guarantor should be liable for, inter alia: “anything expressly provided to be done but omitted to be done pursuant to this agreement.” 15. Pony pleaded that the SPA was subject to an implied term to the following effect: Vand would be able to take, and would succeed in taking, all the steps necessary to procure approval by NDRC in respect of the change in share ownership of Dongguan Dongzhou within two years of the completion date. At para 110 of his judgment Reyes J held that no such term was to be implied. He observed that permission might not have been obtained from NDRC within the two year period without this being attributable to any fault or omission on the part of Vand or Yuehai, whereupon the put option would simply lapse. I consider that Reyes J was correct to reject the implied term. Indeed, it was implicit from clause 6.2 F of the SPA that the parties envisaged that what would otherwise be a triggering event might occur after the expiry of the two year period. Formalities 16. To obtain NDRC’s consent to the change of ownership of Dongguan Dongzhou, the relevant application had to be passed up the following hierarchical chain. No short cuts could be taken: (i) Application for consent had to be made to the Dongguan Humen Bay Management Committee (“DHBMC”). (ii) Having approved the application DHBMC had to pass the application up to the Dongguan Municipal Development and Reform Bureau (“DMDRB”) (iii) Once approved by DMDRB the application had to be referred to successive higher levels of authority, including the Development and Reform Committee of Guangdong Province until, finally, it reached the NDRC. 17. Applying hindsight on the basis of what in fact occurred Reyes J concluded at paragraph 150 of his judgment that, had there been a prompt application for the approval, and had progression of the application up the hierarchical chain then proceeded smoothly, it would have taken between 10 and 15 months “give or take a week or so” for the approval of NDRC to be obtained. The application for approval was not, however, made promptly nor was progression of the application up the chain smooth. What in fact occurred 18. It fell to Dongguan Dongzhou to take the necessary steps to obtain approval of the application. This action has proceeded, reasonably, on the premise that Vand, in carrying out its obligations under clause 6.2 of the SPA, was in a position, through Hans and its chain of subsidiary companies, to procure that Dongguan Dongzhou acted with all reasonable diligence in seeking the approval of the application. This Dongguan Dongzhou signally failed to do. 19. Dongguan Dongzhou did not make the initial application to DHBMC until 14 November 2008, that is about 16 months after the date of completion of the SPA. Thereafter Dongguan Dongzhou’s conduct in chasing up the progress of the application was described by Reyes J at para 130 of his judgment as “lack-a-daisical”. In particular the company did nothing between 21 January 2009 and 20 April 2009, at which date it was found that DHBMC had lost the original application, so Dongguan Dongzhou had to submit a new one. Thereafter at para 134 Reyes J identified a further lack of diligence on the part of Dongguan Dongzhou in taking a month between 22 May 2009 and 23 June 2009 in making a submission to the Development and Reform Commission of Guangdgong Province when this could and should have been done in half that time. 20. On 6 July 2009 Pony wrote to Vand purporting to exercise the put option. The letter alleged that Yuehai had failed to obtain NDRC’s approval of the change in ownership of Dongguan Dongzhou and that this failure arose through the fault, or an act or omission, of Vand or Yuehai. This was said to constitute the occurrence of an “event” as provided for in clause 6.2(D)(2) of the SPA. 21. Vand replied on 10 July 2009 denying that any such event had occurred. 22. On 8 February 2010 NDRC wrote a letter approving the change in the ownership of Dongguan Dongzhou. The rival contentions 23. Vand advanced a number of reasons for contending that there was no valid exercise by Pony of the put option. These were dealt with carefully and at length by Reyes J, who rejected each one of them. The Court of Appeal affirmed his judgment in relation to each. Before this Court, however, Mr Warren Chan S.C. with admirable economy refined his oral argument to a single point. The principle underlying this point had always featured in Vand’s arguments below, but I believe that Mr Charles Sussex S.C. for Pony was correct to comment that Mr Chan’s formulation of the point before us was novel. Mr Sussex sensibly made no objection to this. 24. Mr Chan’s point was that, on the true construction of clause 6.2 of the SPA, the “failure” by Vand to procure the “necessary approval” had to be an “event” capable of triggering the running of the 30 day period within which the option had to be exercised. A letter from NDRC refusing to approve the change of ownership would have constituted such an event. As it was Pony was not able to point to the occurrence of any such event within 30 days of 6 July, when Pony purported to exercise the put option. All that Pony was able to point to was a continuous period during which Vand had not yet succeeded in obtaining NDRC’s approval. This did not constitute an “event”. 25. Mr Sussex accepted that a letter from NRDC refusing to approve the change of ownership of Dongguan Dongzhou would amount to a “failure” by Vand to procure the necessary approval. He contended, however, that a “failure” to procure the “necessary approval” could also be a “continuing failure”. Provided that the continuing failure was caused by the fault, or an act or omission, of Vand or Yuehai it could trigger the put option. On the facts Mr Sussex contended that there had been fault on the part of Vand and Yuehai in failing to make a timely application for approval of the change of ownership of Dongguan Dongzhou. This was a breach of Clause 6.2(A)(ii). Vand then came under a continuing duty to remedy this breach and its omission to do so amounted to a continuing failure to procure the necessary obligation, giving rise to a right to exercise the put option. 26. I have referred at paragraph 15 above to Pony’s unsuccessful attempt to imply a term into the SPA that Vand would take all the steps necessary to procure approval by NDRC of the change of ownership of Dongguan Dongzhou within two years of the completion date. In argument at first instance Mr Sussex built on this implied term by submitting that the put option could be exercised once a reasonable person would conclude that, by reason of Vand’s omission to remedy its breach of Clause 6.2(A)(ii), it would not be possible to obtain NDRC approval by 13 July 2009, that is within two years of completion. The findings of the Courts below 27. Mr Sussex’s submissions found favour with Reyes J, notwithstanding that he had rejected the implied term upon which it appeared to be founded. The critical finding of the judge appears at para 155 of his judgment: “The failure to remedy the breach of cl.6.2A(ii), by taking prompt action to advance the application …can amount to an ‘omission’ within the meaning of cl. 6.2D(2). I agree with Mr Sussex that by June 2009 Dongguan Dongzhou (and hence Vand) had by its lack of activity behaved in such a way as to lead a reasonable person to conclude that it would not be possible to obtain NDRC approval by 13 July 2009. There being a continuing omission to act promptly in June 2009 there would be a triggering event within cl.6.2D(2).” 28. In the Court of Appeal Kwan JA endorsed this conclusion in identical terms at para 54 of her judgment. Discussion 29. The result of this case turns, my view, on two vital words in clauses 6.2(D) and (E) of the SPA: “failure” and “events”. The 30 day period for exercising the put option had to be triggered by a specified “event”. That event was the “failure” of Yuehai to obtain the approval of NDRC to the change in ownership of Dongguan Dongzhou. 30. An “event”, giving that word its natural meaning, is something that happens at a specific moment in time. In answer to the question “when did that event occur” it is usually possible to identify a date. In a helpful passage in Axa Reinsurance (UK) PLC v Field [1996] 1 WLR 1026 at p 1035 Lord Mustill defined an “event” and distinguished its meaning from that of the word “cause”: “In ordinary speech, an event is something which happens at a particular time, at a particular place, in a particular way…A cause is to my mind altogether less constricted. It can be a continuing state of affairs; it can be the absence of something happening”. 31. That the parties to the SPA intended event to bear its natural meaning is confirmed by the qualification of the word “event” by the word “triggering”. The event was to be a happening that would be capable of triggering the start of a 30 day period. Thus it had to be possible to identify the date on which the event occurred. 32. “Failure” is a word that can have a number of different meanings, depending upon the context in which it is used. In some contexts it can be used to describe a specific event happening at a particular time: “heart failure”, “the failure of the electricity supply”, “failure to pass a driving test”. The word can, however, be used perfectly naturally in a different sense, to describe a non-event, a continuous state of affairs, a period during which an individual does not do something. It can be another word for an omission. Contrast “yesterday I failed to pass my driving test” with “throughout the seven years that he lived in England he failed to take a driving test”. The first describes an event, the second a non-event or omission. 33. Mr Chan and Mr Sussex were agreed that a refusal by NDRC to approve the change of ownership would properly have qualified as “a failure to obtain” NDRC’s approval of the change of ownership with the meaning of clause 6.2(D). That accorded to the word “failure” the first of the meanings that I have set out above. Had there been such a refusal there would have been no difficulty in treating it as an “event” that triggered the 30 day period for exercising the put option. Furthermore, it was to guard against the risk of such a refusal that the parties had inserted the put option into the SPA. For these reasons I consider that counsel were correct in agreeing that such a refusal by NDRC would have constituted a “failure to obtain the necessary approval” within the meaning of clause 6.2 (D). 34. Mr Sussex submitted, however, that the word “failure” in clause 6.2(D) was also capable of covering the second meaning of that word that I have identified above, namely the omission on the part of Vand to obtain the approval of NDRC to the change of ownership over a continuous period. On this argument the “failure to obtain approval” described not merely a refusal of approval but a continuous period during which Vand had not yet succeeded in obtaining approval. In my opinion this submission faced insuperable obstacles. 35. In the first place while the continuous omission to obtain approval could, if taken out of context, properly be described as a “failure” to obtain approval, it could not properly be described as an “event”. It was the antithesis of an “event”, it was a “non-event”. Thus, in the context of clause 6.2, “failure” could not properly bear the meaning of a continuous state of affairs. 36. Secondly, because it was not an “event”, the continuing omission to obtain approval was incapable of acting as a trigger. Pony has throughout this litigation been unable to point to a moment in time when the “failure” to obtain approval triggered the running of the 30 day option period. As I understand Pony’s case it is that there was a continuous triggering of the right to exercise the put option during the continuous omission to obtain NDRC’s approval to the change of ownership, at least during any period when the omission to obtain approval was itself caused by an omission on the part of Vand to rectify its earlier fault in not making a prompt application for approval. 37. This leads to the third obstacle. It is impossible to identify, in theory, let alone in practice, at what point in time, on Pony’s case, the fact that Vand had not yet succeeded in obtaining approval first triggered the right to exercise the put option. The continuous omission to obtain approval began on the day of completion of the SPA, but on Pony’s argument did not give rise to a right to exercise the put option until it was attributable to the fault, or an act or omission, of Vand or Yuehai. When, in theory, did that moment arrive? When NDRC should have responded to the application for approval had Vand and Yuehai taken all reasonable steps to obtain this? This would mean that the put option could have been exercised had Vand been responsible for even a week of delay in the process of obtaining approval. This, understandably, was never Pony’s case. I was not clear what Pony’s case was as to the moment when the put option first became exercisable. It seems to me that whatever it was in theory, it was impossible in practice to identify when that moment would arrive. This was because no one could forecast with any degree of accuracy how long the laborious process of obtaining approval would take if Vand exercised due diligence. 38. In short, it is neither possible to read, in its context in clause 6.2, the word “failure” as meaning a continuing omission, nor is it possible to give workable effect to clause 6.2 if “failure” is given that meaning. 39. It is no doubt these difficulties that led Pony in these proceedings to seek to link the right to exercise the put option to the expiry of the period of two years from the date of completion of the SPA. Reyes J and the Court of Appeal appear to have accepted this linkage in the passage that I have quoted at para 27 above. I have not found that passage easy to follow. It uses the language of repudiatory or anticipatory breach of an obligation to obtain NDRC approval by July 13th 2009. But there was no such obligation and concepts of repudiatory or anticipatory breach do not assist in resolving the critical issue in this case of the meaning of “failure to obtain the necessary approval”. 40. My conclusions can be shortly summarised. The “event” described as “failure to obtain the necessary approval of NDRC” in clause 6.2 of the SPA was a refusal by NDRC to give that approval. That event never occurred. On the contrary, NDRC finally gave its approval to the change of ownership of Dongguan Dongzhou. It follows the put option was never triggered. 41. Mr Sussex urged that such a result was not acceptable as it meant that Pony had no remedy if Vand took no steps to obtain approval from NDRC within the two year period. This may be true, but the terms of the SPA suggests that the concern of the parties was with events that occurred within the period of two years from completion. Not only did the put option lapse at the end of two years, but no other kind of claim could be advanced after two years had elapsed. Pony was concerned that NDRC would treat their lack of consent to the change of ownership of Dongguan Dongzhou as a ground for suspending the project operations. If that was going to happen it was likely to happen within two years, and Mr Sussex accepted in argument that the parties could reasonably have expected that NDRC would respond within two years to the application for approval of the change of ownership. 42. As things turned out, the concern that led to the insertion of the put option never came to pass. NDRC approved the change of ownership of Dongguan Dongzhou, albeit belatedly. In these circumstances I do not consider that Pony has reason to feel aggrieved at my conclusion that the put option was never triggered and that, accordingly, this appeal must be allowed. 43. I would make an order nisi that Pony pay the costs of Vand and Mr An in this court and in the courts below. Any submissions as to costs to be made in writing and lodged with the Registrar within 21 days from the date of this judgment, and any response to be made within 21 days thereafter; and in default of such written submissions, that the order nisi should stand as an order absolute without further order. Mr Warren Chan SC and Mr Liu Man Kin instructed by K & L Gates, for the Appellants Mr Charles Sussex SC and Ms Queenie Lau instructed by Herbert Smith Freehills, for the Respondent 
Judgment of the Court: 1. At the conclusion of the hearing of the appeal, we quashed the conviction, allowed the appeal and ordered a re-trial. These are our reasons. Introduction 2. No statement by an accused is admissible in evidence against him unless it is shown by the prosecution to have been a voluntary statement.[1] Admissibility is a matter for the judge and is normally resolved in a voir dire. If the evidence is held to be admissible, it will go before the jury whose concern is with the weight to be given to its contents.[2] At the trial, all evidence relating to the circumstances in which the alleged statement was made are admissible for the purpose of enabling the jury to decide upon the weight (if any) to be given to the statement; whether or not such evidence has been taken in the absence of the jury before the admission of the statement.[3] 3. In this appeal we are concerned with first, in what circumstances a direction should be given to a jury about the voluntariness of a confession notwithstanding the fact that an accused denies making the confession, and secondly, in the event that such a direction is required , whether it is: “… sufficient for a judge to warn the jury that they may consider that they should give no weight to a confession and disregard it if they conclude that it may have been obtained as the result of oppression; or whether the judge must go further and direct them that if they so conclude they are required to disregard it.” [4]emphasis supplied. 4. The answer provided by a majority of the House of Lords in R v Mushtaq and subsequent decisions of the Privy Council, was that the judge must direct the jury as to voluntariness where the circumstances of the case suggest that the confession was, or may have been, induced by oppression and, if they conclude that the confession may have been obtained as the result of oppression they must disregard it. This is commonly called the Mushtaq direction. In July 2009, the Hong Kong Judicial Institute (“Judicial Institute”) issued new specimen directions[5] based on the Mushtaq direction. In HKSAR v Ho Wing To (No 2) [2012] 1 HKLRD 1023, the Court of Appeal adopted the Mushtaq direction, although it recognized that Chan Wei Keung v R [1967] 2 AC 160, a decision of the Privy Council on appeal from Hong Kong, was to the opposite effect.[6] This court is not bound by Chan Wei Keung but must decide for ourselves what is appropriate for Hong Kong.[7] Mr William Tam, Deputy Director of Public Prosecutions, quite properly did not invite us to follow Chan Wei Keung. However, he submitted that “as in Wizzard, the appellant in the present case denied making any confession but alleged police misconduct.[8] The Privy Council held in Wizzard that a specific direction on voluntariness was not required.”[9] This submission required careful consideration. 5. We have considered the relevant authorities and are of the opinion that a Mushtaq direction is required in Hong Kong where the circumstances suggest that the statement was, or may have been, induced by oppression, including cases where the defendant denied making the confession. Some of these authorities show that the courts were, for many years, uncertain whether, when the defendant denied making the confession, the court was required, indeed permitted, to deal with the voluntariness of a denied statement. How the uncertainty was removed may help provide an answer to the instant appeal. The Background 6. The appellant was convicted after trial before a jury of the offence of trafficking in 1 kilogram of a mixture containing 0.86 kilogram of cocaine. 7. The prosecution case is simple. The appellant was arrested by a group of policemen, after a tip-off, at about 5:55 pm on 15 July 2011 in the corridor on the ground floor of King Wah House, Shan King Estate, Tuen Mun. He was searched on the spot by DPC 53274 (“PW1”), who found a paper bag inside the appellant’s shoulder bag. PW1 took out the paper bag and felt its weight. He opened it in front of the appellant. Inside the paper bag, PW1 saw a brick of suspected drugs (in transparent bag) inside. He then arrested the appellant for trafficking and cautioned him in Cantonese. The appellant replied in fluent Cantonese but with a Mandarin accent, “Ah Sir, I am caught. I carried the drugs for somebody else.”. This statement was the relevant statement to be considered in the present appeal: it was a confession of knowledge. A few minutes later, at about 6:05 pm, PW1 and his colleague took the appellant to Room C2003 in King Wah House, the home of the appellant’s ex-wife. They searched Room C2003 but nothing was found. There, at 6:40 pm, PW1 post-recorded the oral statement in his note book, which was then signed by the appellant. Later that night, in a video-recorded interview under caution, the appellant admitted, inter alia, that because he was in financial difficulties, he agreed to carry the drug for a person called “Ah Wong” for $6,000. The prosecution case was that all three statements were made by the appellant voluntarily. 8. The defence case[10] was that as the appellant was walking in the corridor, at least four to five people rushed toward him, two on each side who twisted his arms. At the same time, one of the policemen pressed his head down while another took out a white paper bag from his shoulder bag. He then turned the appellant’s shoulder bag upside down, pouring all the contents onto the ground. The appellant was then handcuffed but he was not cautioned. Nor was he formally arrested. He was made to stand against a wall for about 9 minutes. The appellant denied he made the oral statement. He was then taken to Room C2003, where his ex-wife was. Many policemen swore at him and told him to shut up. The appellant asked PW1 to loosen his handcuff a bit and he was told he should co-operate and sign a statement. He signed it although he couldn’t see what had been written because he did not have his reading glasses. He claimed that he had been threatened before the video-interview and he was taught what to say by the police. 9. As for the white paper bag inside his shoulder bag, the appellant’s case was that he thought it contained money. He said, his son, Pang Cheung was arrested in May 2010 for trafficking in dangerous drugs, and was eventually sentenced to about 9 years’ imprisonment. After his son was arrested, his son asked him to ask his boss “Ah Wong” for what he called “private meal money” as well as money for lawyers. But because the appellant’s ex-wife did not wish to contact the boss, he did so. On the first occasion, which was about one month after his son’s arrest, he contacted Ah Wong, and was told to wait at a taxi stand near King Wah House. A young man came in a car and gave him money saying that it came from the boss. That was the private meal money and it was $50,000. As for the money for the lawyers, he said he asked Ah Wong for the money on a number of occasions, but Ah Wong said the prospect of a successful defence was poor, and a lot of money was needed, and Ah Wong suggested a “family settlement allowance” instead. He and his ex-wife then lost contact with Ah Wong until 14 July 2011, when Ah Wong called his ex-wife and said that the syndicate decided to provide $300,000 as family settlement allowance. So the appellant went to his ex-wife’s home at 9:00 am on 15 July to await instructions. To cut a long story short, at 5:30 pm, outside King Wah House, the same young man came in a taxi and gave him a white paper bag through the window and said it was money from the boss. The appellant did not look at the content and put it into his shoulder bag and walked back to King Wah House. He was arrested in the corridor on the ground floor of King Wah House. Proceedings Below 10. At trial, a voir dire was held. In the appellant’s case[11] we were informed: “10. The learned trial judge held a voir dire to determine the admissibility of the alleged oral admission after which he ruled the appellant did make the alleged oral admission voluntarily.” 11. The respondent’s case[12], filed in response stated: “8. At the trial, the prosecution sought to adduce into evidence the appellant’s oral admission, the post-record of the oral admission and the video-recorded interview. Objection was taken to the admissibility of the evidence and a voir dire was conducted by the trial judge. Both the post-record and the video-recorded interview were excluded after the voir dire but the judge ruled that the prosecution was permitted to lead evidence before the jury of the oral admission at the scene of arrest.” 12. These submissions suggested that the oral statement was dealt with in the voir dire. Shortly before the hearing of this appeal, we asked for a transcript of the ruling on voir dire as well as the written grounds of objection (“the grounds”) produced at trial. The grounds made no mention of the oral statement. The transcript of the judge’s brief ruling stated expressly that he was not satisfied with the voluntariness of the two written statements, one of which, of course, purported to be a record of the oral statement. The judge said at the end of his ruling, after saying that he was not satisfied with the voluntariness of the written statements[13]: “For the oral narration, the verbal admission, you [the prosecution][14] still have the right to call this police constable to give evidence about it. You [the defence[15]] have the right to say that I did not say that. During that time, that may bring up other matters. I do not know. It is for you to control your witness or else it may be unfavourable to you. This is it.” 13. Before us, Mr Tam submitted that the voluntariness of the oral statement was not dealt with in the voir dire because there was no challenge to its voluntariness: the grounds did not refer to it, nor did the judge’s ruling mention voluntariness in respect of it. Mr Richard Wong for the appellant suggested that counsel then appearing for the defendant, not Mr Richard Wong, did not object to the admissibility of the oral statement, because counsel was not aware that a double-barrelled attack was possible. But no attempt had been made to confirm that with trial counsel. Be that as it may, in the Court of Appeal, the second of the two grounds of appeal was: “The trial judge only directed the jury to consider whether the applicant had in fact made the oral admission without directing them to consider whether the admission had been or might have been made under threat or inducement by the police.” 14. The submission of Mr Henry Ma, who appeared for the appellant in the Court of Appeal only, was recorded in the judgment of the Court of Appeal in these terms: “33. Mr Ma submitted that the applicant's evidence was that, upon interception and search by the police, he was subjected to violence and threat, and that the police neither formally arrested him nor cautioned him. Mr Ma submitted that the trial judge should therefore have directed the jury to consider whether the admission made by the applicant was (or might have been) rendered unreliable by oppression from someone with authority …” 15. We believe it does not matter, for the disposal of this appeal, whether a voir dire had been held to enquire into the voluntariness of the oral statement. Had a voir dire been held in respect of the oral statement it would have been clear its voluntariness was understood to be in issue. However, the critical question was whether the voluntariness of the oral statement was an issue before the jury and whether the jury ought to have been given directions as to this aspect notwithstanding the denial that he made the relevant statement. 16. Before the jury, the appellant gave evidence[16] about the circumstances of his arrest. He said four to five men rushed towards him, grabbed his arms and pressed his head down. He was then handcuffed and hooded, and after a few minutes taken to Room C2003. No one opened the white paper bag in front of him or formally arrested or cautioned him. He denied making the oral statement. Whether we proceed on the basis of the facts summarized in the summing up or the version quoted in the judgment of the Court of Appeal, we have no doubt that they raised an issue of voluntariness in relation to the oral statement which had to be considered by the jury when they were deciding what, if any, weight should be given to the oral statement. 17. It was not disputed at trial that the drugs were found in a white paper bag inside the appellant’s shoulder bag. His knowledge of the contents of the bag was the critical issue at trial. He gave evidence about how the bag came to be in his possession and his belief that the bag contained money for his son. In such circumstances, Pang J directed the jury in these terms: “… both prosecution witnesses said that the police did formally arrest the Defendant and [caution] him thereafter, and that under caution, the Defendant admitted that he carried the dangerous drugs for someone else. On the other hand, the Defendant denied all these. So what do you do? You decide where the truth lies. The most important thing is that if you think that the Defendant did not, or may not have said ‘Ah Sir, I am caught. I carry the dangerous drugs for someone else’, then you will have to forget it, as if you have never heard of it and rely on other testimonies and evidence to decide the case. On the contrary, if you are sure that the Defendant did say those words, then this sentence may prove that the Defendant knew that dangerous drugs were inside the white paper bag, it may even be a piece of strong evidence that he trafficked in the dangerous drugs.” 18. The Court of Appeal dismissed the defendant’s appeal. Chu JA, giving the judgment of the court, said: “34. In our view, although the applicant's evidence was that when he was intercepted on the ground floor of King Wah House, the police grabbed hold of his hands and pressed his head down, he also said that he did not say anything until his arrival at his ex-wife’s resident. This being the case, all that the jury had to decide was whether the applicant had made the oral admission under caution that he carried the dangerous drug for someone else. As to whether he made the oral admission as a result of threat or inducement by the police, this was simply not the defence case, nor did the applicant give evidence to that effect at the trial before the jury. It was not necessary for the trial judge to give directions in his summing-up as to whether the admission was made under threat or inducement by the police and whether its contents were true.” This Appeal 19. We gave leave to appeal so that we could consider whether on the evidence the jury should have been directed on the voluntariness of the oral statement. If so, the content of such directions would also be in issue. Double-Barrelled Attack 20. We mentioned earlier uncertainty about the treatment of confessions which the defendant denied making but where there was evidence of oppressive conduct at or before the alleged making. As late as 1997, the position in Hong Kong was that the court had no jurisdiction to consider the admissibility of an oral confession which the defendant denied making. 21. That was so, notwithstanding that in Ajodha and Others v The State [1982] AC 204, the Privy Council, on appeal from the Court of Appeal of Trinidad and Tobago, gave an affirmative answer to the question: “when the prosecution proposes to tender in evidence a written statement of confession signed by the accused and the accused denies that he is the author of the statement but admits that the signature or signatures on the document are his and claims that they were obtained from him by threat or inducement, does this raise a question of law for decision by the judge as to the admissibility of the statement?”[17] 22. Lord Bridge explained at page 220F: “The fallacy, ... is to suppose that a challenge by an accused person to a statement tendered in evidence against him on the ground that he never made it and a challenge on the ground that the statement was not voluntary are mutually exclusive, so as to force upon the judge a choice between leaving an issue of fact to the jury and deciding an issue of admissibility himself. In all cases where the accused denies authorship of the contents of a written statement but complains that the signature or signatures on the document which he admits to be his own were improperly obtained from him by threat or inducement, he is challenging the prosecution's evidence on both grounds and there is nothing in the least illogical or inconsistent in his doing so.” 23. In Thongjai, the Privy Council was concerned with two appeals from Hong Kong. The facts in the case of the appellant, Lee Chun Kong, are surprisingly similar to the case under appeal. Lee was charged with trafficking in dangerous drugs. The prosecution case was that upon arrest, with packets containing a large quantity of heroin being found on him, Lee said “Ah Sir, don’t need to say anything. You already know it is white powder.” Later, Lee signed the record of his oral statement in the constable’s notebook. Later still, at the police station, he made a further written statement. Lee denied making the oral statement. Lee’s case was that he was at the lift lobby to his home: “… when suddenly three men rushed up to him, one pushed him to the floor and immediately punched and kicked him, and his nose bled. He was made to squat facing the wall, his hands were handcuffed behind his back and he was hooded with a hood with eyeholes.” 24. A voir dire was held into the admissibility of these two written statements. They were ruled inadmissible by the trial judge. But the judge refused to hold a voir dire in respect of the oral statement on the ground that he had no jurisdiction to do so. The prosecution was permitted to adduce the evidence of his oral admission at the trial before the jury. Lee’s appeal to the Court of Appeal was unsuccessful. However, he succeeded before the Privy Council and a re-trial was ordered. 25. Lord Hutton, in delivering the judgment of the Privy Council said that the principles stated by Lord Bridge in Ajdoha also applied to oral admissions[18] and that it is erroneous to think that if the defendant denies making the oral statement, then there is no issue of admissibility for the trial judge to decide on a voir dire, even if the defendant alleges that he was improperly treated by the police before or at the time of making the oral admission.[19] 26. The position in Australia was the same. In MacPherson v The Queen [1981] 147 CLR 512 where the prosecution relied on an oral confession of the accused who complained of oppressive police conduct but whose evidence was that: “I didn’t say anything to the police except my name and address.” 27. Gibbs CJ and Wilson J said at 522: “… although the question whether any confession was made is one for the jury and not for the judge, a voir dire is required when the accused disputes that any confession was made, but also claims that there was such inducement or pressure that if a confession was made it was not voluntary. The recent decision of the Judicial Committee in (Ajodha) supports this conclusion.” 28. They also explained, if the court could not decide on admissibility first, the jury: “… will have before them evidence that has not been found to be admissible, and an important rule which exists to protect accused persons, and to maintain proper standards of police investigation, will have been subverted. Of course once the evidence of the confession is admitted the jury are not concerned with the question whether it was voluntary; they have to consider only whether it was made and whether it was true, although they are entitled to consider the circumstances surrounding the making of the statement in deciding upon its weight and value ...” 29. The above authorities show that, there is no difference between situations where the defendant admits making the admission (oral or in writing) but raises the issue that it was not voluntary, and where he denies making the statement but claims that, if he did make it, he made it involuntarily. In both situations, a voir dire should be held. Should the confessions be admitted by the judge, they will go before the jury to decide on the weight, if any, to be accorded to them. Although, in the case where an accused denied making the statement, the jury must also decide whether he had made the statement. Direction to the Jury 30. We turn to consider how the jury should approach a confession when there was evidence before them which suggest that the confession was or may have been obtained as the result of oppression. 31. In R v Bass [1953] 1 QB 680, a decision of the Court of Criminal Appeal, Byrne J, speaking for himself as well as for Lord Goddard CJ and Parker J said : “… when a statement has been admitted by the judge he should direct the jury to apply to their consideration of it the principle as stated by Lord Sumner, and he should further tell them that if they are not satisfied that it was made voluntarily, they should give it no weight at all and disregarded it.” 32. Lord Sumner’s principle was quoted by Byrne J in the immediately preceding paragraph of his judgment and read: “It has long been established as a positive rule of English criminal law, that no statement by an accused is admissible in evidence against him unless it is shown by the prosecution to have been a voluntary statement, in the sense that it has not been obtained from him either by fear of prejudice or hope of advantage exercised or held out by a person in authority. The principle is as old as Lord Hale.” 33. In Basto v the Queen (1954) 91 CLR 628, the High Court of Australia, said the passage from the judgment of Byrne J quoted above: “… cannot be right. The admissibility of evidence is not for the jury to decide, be it dependent on fact or law: and voluntariness is only a test of admissibility”[20] 34. The High Court explained : “… the question what probative value should be allowed to the statements made by the prisoner is not the same as the question whether they are voluntary statements nor at all dependent upon the answer to the latter question. A confessional statement may be voluntary and yet to act upon it might be quite unsafe; it may have no probative value. Or such a statement may be involuntary and yet carry with it the greatest assurance of its reliability or truth.” 35. Chan Wei Keung (1967) 2 AC 160, is a decision of the Privy Council on appeal from Hong Kong, where the majority in the Court of Appeal was of the view that it was necessary to give a separate direction to the jury that they must be satisfied beyond reasonable doubt as to the voluntariness of the confessions before giving them any consideration. The Privy Council after considering Basto as well as other Australian and Canadian decisions to similar effect, disapproved of R v Bass and concluded that no such direction was necessary. 36. There, the matter rested until Mushtaq. In Mushtaq, the certified point of law or general public importance before the House of Lords [21] was: “Whether in view of article 6 of the Convention for the Protection of Human Rights and Fundamental Freedoms[22], a judge, who had ruled pursuant to section 76(2)[23] of the Police and Criminal Evidence Act 1984 (PACE) that evidence of the alleged confession has not been obtained by oppression, nor has it been obtained in consequence of anything said or done which is likely to render unreliable any confession, is required to direct the jury, if they conclude that the alleged confession may have been so obtained, they must disregard it.” (1515) 37. The judge had directed the jury that if they are sure that the accused had made the statement and that it was true they may rely on it: “…even if it was or may have been made as a result of oppression or other improper circumstances.”[24] 38. Hence, Lord Rodger of Earlsferry pointed out in his judgment (which had the concurrence of Lord Steyn and Lord Phillips) it was important to note that: “the Crown's contention is that in cases where the jury have rightly concluded that the confession was obtained by oppression or other improper means, they are to be told that they can still rely on it if they think that it is true.” 39. This stark statement lays bare the importance of the issue. In our respectful view, the majority[25] in Mushtaq provided the only acceptable answer. 40. Lord Rodger rejected that contention in these words: “44. At various points during the hearing before the House, it was emphasised that under the present system if a jury think that a confession was, or may have been, obtained by oppression, they are unlikely to accord it any weight or value. So there was no real difference between a situation where a jury give no weight or value to a confession in this way and one where they are told that they must disregard it. For all practical purposes, the present system achieved what the appellant was seeking.” 45. That argument falls short if the rule against admitting an involuntary confession is not based simply on its potential unreliability. And it is indeed clear that, according to the more modern analysis at least, the rule rests on a rather wider basis. Giving the opinion of the Privy Council in Lam Chi-ming v the Queen [1991] AC 212, 220E-F, Lord Griffiths said: ‘their Lordships are of the view that the more recent English cases established that the rejection of an improperly obtained confession is not dependent only upon possible unreliability but also upon the principle that a man cannot be compelled to incriminate himself and upon the importance that attaches in a civilised society to proper behaviour by the police towards those in their custody. All three of these factors have combined to produce the rule of law applicable in Hong Kong as well as in England that a confession is not admissible in evidence unless the prosecution established that it was voluntary.’ [26] Therefore, even if a jury would be unlikely to rely on a confession which they considered had been obtained by compulsion, the question still remains whether, having regard to the principle that a man cannot be compelled to incriminate himself and having regard to the importance attached to proper behaviour by the police, the jury are entitled to relied on a confession which they consider was, or may have been, obtained by oppression or other improper means. 46. Since the three considerations mentioned by Lord Griffiths lie behind section 76(2), it respectfully seems to me that it is inconsistent with the very purpose of that provision to affirm that the jury are entitled to rely on a confession in such circumstances. Under section 76(2), if an objection to a confession is raised, the judge must exclude it unless he is satisfied beyond a reasonable doubt that it was not obtained by oppression or any other improper means. The evidence is excluded because, for all the kinds of reasons explained by Lord Griffiths, Parliament considers that it should not play any part in the jury’s verdict. It flies in the face of that policy to say that a jury are entitled to rely on a confession even though, as the ultimate arbiters of all matters of fact, they properly consider that it was, or may have been, obtained by oppression or any other improper means. 47. In my view, therefore, the logic of section 76(2) of PACE really requires that the jury should be directed that, if they consider that the confession was, or may have been, obtained by oppression or in consequence of anything said or done which was likely to render it unreliable, they should disregard it. In giving effect to the policy of Parliament in this way, your Lordships are merely reverting to the approach laid down by the Court of Criminal Appeal … in R v Bass.” 41. Lord Rodger then quoted the passages quoted in paras 31 and 32 above and continued: “It seems clear that the court saw the direction to disregard the confession in such circumstances as part and parcel of the jury’s exercise of attributing the appropriate weight to the confession: in circumstances where they found that it had not been voluntary, for reasons going back to the time of Lord Hale, they should give it no weight and should disregard it.” 42. Lord Rodger recognized that he was departing from Chan Wei Keung but considered that s 76(2) PACE required a different approach, and: “49. If there were any doubt about the proper approach, however, it would be removed by article 6(1) of the Convention … It is well known that among the rights implied into article 6(1) is a right against self-incrimination…” 43. Wizzard supports this new approach, there, Lord Phillips, rejected the argument that Mushtaq had departed from the common law and was not applicable in Jamaica[27], and said [28]: “The relevant principle derived both from section 76 (2) of PACE and article 6 of the ECHR is the principle against self-incrimination. That is a long-recognised principle of the common law. The approach in R v Bass accorded with that principle. The approach in Chan Wei Keung v The Queen did not. The latter decision was a false step in the development of the common law. Mushtaq has re-established the correct approach and is, in consequence, applicable in Jamaica.” 44. In Mushtaq, Lord Carswell was of the view, that the right not to incriminate oneself is a constituent element of a fair procedure under Article 6(1) and it would not be compatible with the Convention right to a fair hearing for the possibility to be left open that the jury may take into account a confession made by an accused person and rely upon it in assessing his guilt on a charge of a criminal offence if they consider that the content is true, notwithstanding the fact that they consider that it has been or may have been obtained as the result of oppression. 45. Lord Carswell also agreed that Chan Wei Keung and Basto should no longer be followed, buthe took the view that the Bass direction[29] was sufficient to accommodate s 76(2) and that s 76(2) did not require the judge to direct the jury to disregard a confession which was or may have been obtained by oppression. 46. Lord Hutton in his dissenting judgment, had in mind the division of function between the judge and the jury, the former being concerned with the admissibility of the confession and the latter, its reliability, and said the division of functions has significant advantages, for example, it ensures that if the judge rules a confession inadmissible, the jury will never become aware of the confession and because the evidence given on the voir dire by the defence is not revealed to the jury, the defendant is not inhibited in the voir dire from giving evidence. 47. With respect, these are good reasons for, and no doubt explain why, voluntariness should first be considered by the judge on a voir dire. On the voir dire the court is not concerned with the reliability of the confession. A true but involuntary confession may not be admitted.[30] After the voir dire, the issue of the voluntariness of the statement may be repeated before the jury. With respect, we cannot agree with Basto that voluntariness is only concerned with admissibility. The jury is as much concerned with the voluntariness of the confession as the judge. Though, prior to Mushtaq, only as far as its voluntariness might affect reliability.[31] 48. Lord Hutton approved the opinion stated in Mirfield on Silence, Confessions and Improperly Obtained Evidence (1997) pp 51, 52 that: “… There seems to be no good reason for the accused to have two bites of the cherry in relation to the issue of admissibility.” 49. If the voir dire is understood and justified as a protection for the defendant who would otherwise be put into an impossible position. It is not a first bite of the cherry but basic fairness to an accused. The jury are the ultimate arbiters of fact and but for the inherent injustice to the defendant identified by Lord Hutton, voluntariness would be decided by the same jury which have to determine guilt. 50. Moreover, as Hong Kong Archbold 15-116 noted “At the request of the defendant the issue of voluntariness may be heard before the jury.” Presumably, in that event, the jury would be directed to disregard the confession if they conclude that it was or may have been involuntary. If that is correct, since, even after a voir dire, the question of voluntariness of the confession remains a matter for the jury, and the jury may not be told about the voir dire, or the view of the judge on voluntariness, it is difficult to see why they should not be directed to disregard the confession if they conclude that the confession was or may have been involuntary. 51. Moreover, the temporal nature of the judge’s decision on admissibility is shown by the fact that he may change his mind after the voir dire and in the course of the trial.[32] The position now to apply in Hong Kong 52. In Hong Kong we do not have the equivalent of s 76(2) PACE. But, when commenting on the principles relating to voluntariness, Li CJ said: “The underlying rationale is based both on the need to ensure the reliability of confessions as well as the right of silence…The right of silence is deeply rooted in the common law.”[33] 53. However, we have the equivalent of Article 6(1) in Article of the Hong Kong Bill of Rights.[34] 54. In our opinion, Chan Wei Keung should no longer be followed in Hong Kong and the common law is as stated in Mushtaq and Wizzard. It follows that in our respectful view the Court of Appeal in Ho Wing To (No 2) and the Judicial Institute were right in adopting the Mushtaq direction. 55. We return to Mr Tam’s submission that : “43. … as in Wizzard, the Appellant in the present case denied making any confession but alleged police misconduct. The Privy Council held in Wizzard that a specific direction on voluntariness was not required.” 56. In Wizzard, Lord Phillips said: “35. A Mushtaq direction is only required where there is a possibility that the jury may conclude (i) that a statement was made by the defendant, (ii) the statement was true but (iii) the statement was, or may have been, induced by oppression.” 57. Mr Tam submitted that (iii) is only satisfied if there was actual evidence from the defence that, if the confession was made, it was induced by oppression. Mr Tam submitted that if the defendant had said he had been beaten up by the police at or before the time of the alleged confession but claimed that he had remained silent, (iii) would not be satisfied. We do not agree. The burden remains on the prosecution to prove before the jury that the confession is voluntarily made. Evidence relating to the circumstances in which the confession was allegedly made is relevant to the jury’s consideration. If circumstances[35] exist on which the jury could reasonably conclude that the confession was, or may have been, induced by oppression, the jury should be given a Mushtaq direction. 58. In a suitable case, the jury may reject the denial but are satisfied that it was voluntary. As Mason J pointed out in MacPherson, a defendant may deny making the statement because: “… he may have thought, justifiably, that he would damaged that case by maintaining strongly before the jury an alternative case that the admissions made were not voluntary.” (532) 59. Returning to the three conditions in Wizzard, we believe all three are arguably satisfied in this case. 60. In Wizzard, the Privy Council concluded on the facts, condition (iii) was not satisfied. However, Lord Phillips said: “36. … The appropriate direction (had there been evidence that the statement under caution had been forced out of the appellant by violence) would have been a Mushtaq direction.” 61. The facts in Wizzard are difficult and we will not go into them. It is unnecessary to do so. The clear statements in Wizzard quoted in paras 56 and 60 above are sufficient for the present purpose. 62. Moreover, in Benjamin v The State [2012] UKPC 8 on appeal from the Court of Appeal of the Republic of Trinidad and Tobago[36], when dealing with the argument that no Mushtaq direction was necessary because the appellants denied making the statements, Lord Kerr said in the judgment of the Privy Council : “A Mushtaq Direction 14. A Mushtaq direction is one which instructs the jury that if they consider that written or oral statements were, or may have been, obtained by oppression or in consequence of anything said or done which was likely to render it unreliable, they should disregard it... 15. On this appeal the respondent argues that the position in Wizzard is replicated by the circumstances of the present case. Both appellants had denied that they had made statements. Applying the reasoning in Wizzard, therefore, no Mushtaq direction was needed. The Board does not accept this submission. It was clearly open to the jury to conclude that the appellants had made the statements attributed to them. After all, it was emphatically the prosecution's case that they had done so – indeed, had made the statements in the presence of justices of the peace. Likewise, it was open to the jury to find that the statements were true; this was again the prosecution's categorical case. Finally there was evidence on which the jury could have concluded that the appellants' signatures were appended to the statements as a result of oppression. All three conditions necessary to activate a Mushtaq direction were therefore present. 16. The Board in Wizzard considered that the fact that the appellant in that case had made an unsworn statement from the dock, denying that he had made the confession which the police claimed he did, meant that a Mushtaq direction was not required. It is, with respect, somewhat difficult to understand why this should be so. Simply because the appellant had denied making the statement, it does not follow that the jury could not find that he had done so.” 63. With respect, we agree. To distinguish confessions according to whether the defendant denied making them harks back to the time when whether voluntariness could be in issue when the defendant denied the making of the confession was uncertain. That issue was put to rest in Thongjai. We would not permit it to be resurrected in any shape or form. 64. In the instant case, a Mushtaq direction should have been given. Although the respondent’s case originally asserted this in the alternative, Mr Tam rightly did not contend that the learned judge’s submission satisfied Mushtaq. For these reasons, we allowed the appeal, and quashed the conviction. Mr Tam asked for a re-trial, which Mr Wong, rightly did not oppose. A re-trial was ordered. Mr Richard Wong, instructed by Morley Chow Seto, assigned by Director of Legal Aid, for the appellant Mr William Tam, DDPP and Ms Annie Li, PP of the Department of Justice, for the respondent [1] See Ibrahim v the King [1914] AC 599, 609. [2] R v Murray [1951] 1 KB 391, where Lord Goddard CJ said, when delivering the judgment of the court: “it has always, as far as this court is aware, been the right of counsel for the defence to cross-examine again the witnesses who have already given evidence in the absence of jury; for if he can induce the jury to think that the confession was obtained through some threat or promise, its value will be enormously weakened. The weight and value of the evidence are always matters for the jury.” [3] Section 59 of the Criminal Procedure Ordinance Cap 221. R v Murray ibid. [4] Lord Carswell in R v Mushtaq [2005] 1WLR 1513, 1536. [5] The relevant specimen direction, reads: “In this case, the defendant alleges that (summarise the allegation). If you conclude that that allegation is or may be correct and that the admissions/answers were or may have been obtained as a result of that conduct, then you must disregard the admissions/answers.” [6] In Wizzard v The Queen [2007]LRC 457, Lord Phillips of Worth Matravers, delivering the decision of the Privy Council on appeal from Jamaica, described Chan Wei Keung as “a false step in the development of the common law” and said “Mushtaq has re-established the correct approach”. [7] Solicitor (24/07) v Law Society of Hong Kong (2008) 11 HKCFAR 117, paras 17 & 18. [8] A so-called “double-barrelled attack on a statement on the grounds that (a) it is not free and voluntary, and (b) it is not made by him.” per Bollers CJ in the Court of Appeal of Guyana in State v Fowler (1970) 16 WIR 452, 465 and quoted by Lord Hutton in Thongjai v The Queen [1997] HKLRD 678, 684. [9] Para 43 of the respondent’s case. [10] This is taken from the summing-up. The submission of counsel for the appellant in the Court of Appeal, quoted at para 14 below, recorded that the appellant claimed he also had been threatened upon arrest. We do not have a transcript of the appellant’s evidence. [11] Dated 30 May 2014. [12] Dated 7 July 2014. [13] In an agreed translation for this appeal. [14] Not in the original Chinese. [15] Ditto. [16] This is taken from the summing-up. [17] At 214F, Lord Bridge said: “This question has provoked the keenest judicial controversy in a number of Caribbean appellate courts and a great amount of erudition has been devoted to the lengthy judgments which have been written answering the main question in one way or the other and expressing different shades of opinion on a number of related questions.” [18] 684E. [19] 688G. [20] At p 641, with respect we cannot agree that voluntariness is only a test of admissibility. We shall endeavour to explain below. [21] Lord Steyn, Lord Hutton, Lord Phillips of Worth Matravers MR, Lord Rodger of Earlsferry and Lord Carswell. [22] “In the determination of his civil rights and obligations or of any criminal charge against him, everyone is entitled to a fair and public hearing within a reasonable time by an independent and impartial tribunal established by law.” [23] Section 76(2) provides: “If, in any proceedings where the prosecution proposes to give in evidence a confession made by an accused person, it is represented to the court that the confession was or may have been obtained (a) by oppression of the person who made it; or (b) in consequence of anything said or done which was likely, in the circumstances existing at the time, to render unreliable any confession which might be made by him in consequence thereof, the court shall not allow the confession to be given in evidence against him except in so far as the prosecution proves to the court beyond reasonable doubt that the confession (notwithstanding that it may be true) was not obtained as aforesaid.” [24] However, on the facts, there were no evidence of oppression to raise an issue of voluntariness before the jury. (1527F) And for this reason, the appeal was dismissed. [25] Lords Steyn, Phillips, Rodger and Carswell. [26] In A v Home Secretary (No 2) [2006] 2 AC 221 at 249 Lord Bingham, after quoting the same passage from Lord Griffiths’ judgment and noting that it has been endorsed by the House of Lords in Mushtaq, said that: “… in jurisdictions where the law is in general harmony with the English common law reliability has not been treated as the sole test of admissibility in this context.” See paras 14-17 of Lord Bingham’s judgment. [27] Jamaica did not have the equivalent of s 76(2) of PACE or Article 6(1) of European Convention of Human Rights (“ECHR”). [28] With the concurrence of Lord Bingham of Cornhill, Lord Hoffmann, Lord Carswell and Lord Brown of Eaton-under-Heywood. [29] Lord Carswell said: “69. One may observe that the Bass direction is not expressed in mandatory terms, and it cannot in my opinion be said that the common law authorities required a judge to go so far as to direct the jury that they MUST give such a confession no weight and disregard it.” Emphasized supplied. [30] Lam chi-ming see para 40 above. [31] To say that voluntariness is a test of admissibility does not fully answer the question. The reason involuntary confession is not admissible lies in the principle against self-incrimination, a long recognized principle of the common law, Lord Phillips in Wizzard. [32] R v Ajit Singh Sat-Bhambra (1989) 88 CR App R 55, 62 Lord Lane CJ said: “In Watson (1980) 70 CR App R 273 decided before the 1984 Act (PACE), it was held that a judge who has second thoughts about the voluntariness of a statement which he has earlier ruled admissible upon the voir dire may, where it is appropriate so to do, change his opinion as to its admissibility, and may take such steps as are necessary to put matters right, by, for example, directing the jury to disregard it or discharging the jury.” [33] With the concurrence of Litton, Ching, and Bokhary PJJ and Sir Anthony Mason NPJ, in Secretary for Justice v Lam Tat Ming and Another (2000) 3 HKCFAR 168, 178. [34] Which provides, where relevant: “in the determination of any criminal charge against him, or of his rights and obligations in a suit at law, everyone shall be entitled to a fair and public hearing by a competent, independent and impartial tribunal established by law.” [35] Whether such circumstances exist in a particular case is fact sensitive. [36] Lord Kerr, Lord Clarke, Lord Wilson, Dame Heather Hallett and Dame Janet Smith. 
Chief Justice Ma : 1. This appeal concerns the question of the Government rent that is payable on construction sites used for development or redevelopment. That such sites are subject to the payment of Government rent can be seen from Regulation 2 of the Government Rent (Assessment and Collection) Regulation, Cap 515A. 2. For the reasons contained in the judgment of Lord Walker of Gestingthorpe NPJ, I would dismiss the appeal. The decisions of this Court in Commissioner of Rating and Valuation v Agrila Ltd (2001) 4 HKCFAR 83 and in the present appeal are to be regarded as providing the necessary guidance as to the assessability and assessment of Government rent in relation to construction sites used for development or redevelopment Mr Justice Chan PJ : 3. I agree with the judgment of Lord Walker of Gestingthorpe NPJ. Mr Justice Ribeiro PJ : 4. I agree with the judgment of Lord Walker of Gestingthorpe NPJ. Mr Justice Bokhary NPJ : 5. I am in entire agreement with the judgment of Lord Walker of Gestingthorpe NPJ, and add only a word of my own. The nil or nominal rateable value for which Mr John Howell QC contends on the appellant’s behalf is obviously unrelated to reality. Among the ways in which Mr Howell skillfully endeavoured to cope with that difficulty in his path was to place reliance on a passage in Lord Asquith of Bishopstone’s speech in East End Dwellings Ltd v Finsbury Borough Council [1952] AC 109. That passage, which appears at pp 132- 133, reads: “If you are bidden to treat an imaginary state of affairs as real, you must surely, unless prohibited from doing so, also imagine as real the consequences and incidents which, if the putative state of affairs had in fact existed, must inevitably have flowed from or accompanied it. One of these in this case is emancipation from the 1939 level of rents. The statute says that you must imagine a certain state of affairs; it does not say that having done so, you must cause or permit your imagination to boggle when it comes to the inevitable corollaries of that state of affairs.” 6. Such a statement must nowadays be read subject to the imperative of purposiveness in statutory interpretation. Turning from the general to the particular, I begin by noting that in the present case, the “imaginary state of affairs”, to use Lord Asquith’s expression, is the one that arises whenever the nature of the tenement concerned is such that it would never in real life be taken on a year to year tenancy. It arises because of the statutory requirement that the rateable value of all tenements, even those of that nature, be ascertained in terms of the hypothetical year to year tenancy laid down by s 7(2) of the Rating Ordinance, Cap 116. 7. Needless to say, that rating hypothesis was not enacted for the purpose of creating unreality. Obviously, its objective is uniformity. And that means uniformity in the starting point from which to proceed when ascertaining rateable values. No such value is ultimately ascertained, as opposed to distorted, if purposiveness is laid aside and artificiality is carried to the point of a substantial value being artificially reduced to nothing or next to nothing. In my view, Lord Asquith’s statement quoted above does not compel anything of that kind. Lord Walker of Gestingthorpe NPJ : 8. This appeal is the latest stage in protracted litigation concerned with the rent payable under Government leases of land which is in course of development or redevelopment. The problem came before this Court in 2001 in two consolidated appeals, one brought by the Commissioner of Rating and Valuation (“the Commissioner”) and the other (by way of cross-appeal) by Agrila Ltd and 58 other companies having the same, or a similar interest to that of Agrila Ltd: Commissioner of Rating and Valuation v Agrila Ltd and others (2001) 4 HKCFAR 83 (“Agrila”)[1]. This Court allowed the Commissioner’s appeal and dismissed the cross-appeal from the decision of the Court of Appeal [2000] 1 HKC 175, which was on an appeal from a decision of the Lands Tribunal issued on 20 February 1998. The Tribunal had, with the consent of the parties, restricted its decision to answering nine preliminary questions of law (a tenth question had been formulated but was disposed of by agreement). Best Origin Ltd (“Best Origin”), the appellant in the present appeal, is one of the other 58 companies which were parties to Agrila. 9. The questions answered by the Tribunal, and subsequently considered by the Court of Appeal and this Court, related to the validity, interpretation and application of statutory provisions under which rents payable under Government leases of land in course of development or redevelopment are to be calculated by reference to rateable value. There is no longer any issue as to the validity of these provisions, but their correct interpretation and application is still contested. Agrila raised issues of great public importance and considerable difficulty. Several of those issues have been clearly and authoritatively answered, leaving no room for further dispute. But the preliminary nature of some of the questions raised in Agrila inevitably led to a degree of generality in the answers given by this Court. In the present appeal, by contrast, there is a detailed and comprehensive decision of the Tribunal given after it had considered a large volume of oral and documentary evidence. 10. It is best to start with a very general explanation of the problem. In Hong Kong the law of rating is based on the English law of rating, which has a long history, as Tang Ag CJHC remarked at the beginning of his judgment in the Court of Appeal. It has (in striking contrast to other forms of taxation, which are subject to frequent legislative intervention) shown remarkable stability over the years. Instead the courts have undertaken the task of interpreting and adapting the traditional statutory formula so as to apply it to modern conditions. All this was explained by Viscount Hailsham LC in Railway Assessment Authority v Southern Railway Company [1936] AC 266, 273, cited by Tang Ag CJHC in para 2 of his judgment. 11. On this occasion the legislature has intervened, and the courts have to work out the full implications of the intervention. It has for a long time been a fundamental proposition of rating law (as Sir Anthony Mason NPJ put it in Agrila, para 48 (99C), citing Lord Radcliffe in London County Council v Wilkins (Valuation Officer) [1957] AC 362, 380) that: “Building sites themselves are not treated as hereditaments [the English rating equivalent for tenements] while the work of building is in progress.” That is because a building site is not in rateable occupation, a term of art indicating occupation which is (i) actual, (ii) exclusive, (iii) beneficial in the sense of being of value to the occupier, and (iv) not too transient. These requirements apply under the Rating Ordinance: Yiu Lian Machinery Repairing Works v Commissioner for Rating and Valuation [1982] HKDCLR 32, 39, cited by Sir Anthony Mason NPJ in Agrila at para 47 (98I-99B). 12. It is now clear and no longer in dispute, as a result of this Court’s decision in Agrila, that s 2 of the Government Rent (Assessment and Collection) Regulation, Cap 515A (“the Rent Regulation”), made under s 34 of the Government Rent (Assessment and Collection) Ordinance, Cap 515 (“the Rent Ordinance”), is valid and changes (though only for the purposes of Government rent) the “fundamental proposition” that a building site is not a rateable tenement. Section 2 (set out in para 24 below) is a “deeming” provision under which a building site is to be treated, for the purposes of the Rent Ordinance, as if it were a rateable tenement. But, as Sir Anthony Mason NPJ observed in Agrila, para 49 (99F-G), s 2 says nothing about how the rateable value is to be ascertained. That question was raised as point 4 before the Tribunal in Agrila, and it received an answer, in general terms, in the judgment of Sir Anthony Mason NPJ, with which the other members of this Court agreed. But it is argued that the Tribunal made an error of law in applying that guidance in the present case. 13. As is apparent from the differences of opinion between the Court of Appeal and this Court in Agrila, it is not easy to see how the deeming provision in s 2 of the Rent Regulation is intended to fit in with the scheme of the rules as to the ascertainment of rateable value set out in ss 7 and 7A of the Rating Ordinance, Cap 116. Those rules themselves contain some provisions of a hypothetical character, in particular the requirement to ascertain “the rent at which the tenement might reasonably be expected to let, from year to year”. Lord Wilberforce commented on the corresponding words in English rating statutes, in Dawkins (Valuation Officer) v Ash Brothers and Heaton Ltd [1969] 2 AC 366, 385: “The interpretation of these words has come over the years to be invested with a good deal of learning, and even of mystery, but up to a point there is nothing very difficult about them. It is upon the delimination of the ‘actual,’ on the one hand, and the ‘hypothetical,’ on the other, that the argument in the present case takes its shape.” The same general comment might be made about the argument in this appeal. The Rating Ordinance 14. The Rent Ordinance and the Rent Regulation operate, to a large extent, by reference to the Rating Ordinance, and with a similar scheme for assessment. It is therefore best to start with the Rating Ordinance. The scheme of assessment for rating purposes is that under ss 11 to 14 of the Rating Ordinance the Commissioner must, at the direction of the Chief Executive, prepare a valuation list every year. The Chief Executive has power under s 11(1)(b) to designate a date (“the relevant date”) by reference to which the values of tenements are to be ascertained. The present practice is for the valuation list to come into force on 1st April based on valuations as at 1st July (the relevant date) in the preceding year. The list must set out the rateable value of each tenement (which is under s 2 the basic unit of property on which rates are to be levied). By s 18 rates are to be levied on each tenement as a prescribed percentage of its rateable value. Rating uses the concept of rateable value in order to set a standard as between one rateable tenement and another. In the Court of Appeal Tang Ag CJHC cited the observations on this point made by Lord Pearce in Dawkins [1969] 2 AC 366, 381: “Rating seeks a standard by which every hereditament in this country can be measured in relation to every other hereditament. It is not seeking to establish the true value of any particular hereditament, but rather its value in comparison with the respective values of the rest.” 15. Sections 7 and 7A are of central importance to this appeal. They direct how rateable value is to be assessed. Section 7 is as follows: “(1) Subject to sections 8, 8A and 9, the rateable value of a tenement shall be ascertained in accordance with this section and section 7A. (2) The rateable value of a tenement shall be an amount equal to the rent at which the tenement might reasonably be expected to let, from year to year, if – (a) the tenant undertook to pay all usual tenant’s rates and taxes; and (b) the landlord undertook to pay the Government rent, the costs of repairs and insurance and any other expenses necessary to maintain the tenement in a state to command that rent.” Section 8, 8A and 9 (relating to machinery, plant and advertising stations) are not relevant to this appeal, except for one point on s 8 taken by Mr Holgate QC (for the Commissioner) as part of a fallback submission. 16. Subsections (2) and (5) of s 7A are as follows: “(2) The rateable value of any tenement to be included in a list prepared under section 12 shall be ascertained by reference to the relevant date on the assumption that at that date- (a) the tenement was in the same state as at the time the list comes into force; (b) any relevant factors affecting the mode or character of occupation were those subsisting at the time the list comes into force; and (c) the locality in which the tenement is situated was in the same state, with regard to other premises situated in the locality, the occupation and use of those premises, the transport services and other facilities available in the locality and other matters affecting the amenities of the locality, as at the time the list comes into force. … (5) In this section ‘relevant date’ means the date designated by the Chief Executive under section 11(1)(b).” Subsection (1) was repealed in 1987 and subsections (3) and (4) are not relevant to this appeal. 17. Section 36 contains exemptions from liability to rates for various categories of land and buildings, including agricultural land. The effect of the exemption is that tenements in these categories are not rateable tenements. Sections 37 to 40 lay down the procedure for challenging decisions of the Commissioner. There is a right of appeal to the Lands Tribunal, with a possible further appeal on the ground of error of law. The Rent Ordinance and the Rent Regulation 18. The complex background to the enactment of the Rent Ordinance is described in the judgment of Sir Anthony Mason NPJ in Agrila, paras 52 to 64 (100A-102D). He referred to the Crown Leases Ordinance, Cap 40 (enacted in 1973), the Joint Declaration of 19 December 1984, the Rent Conditions uniformly applied between 1985 and 1997, the New Territories Leases (Extension) Ordinance, Cap 150 (enacted in 1988) and the Basic Law, promulgated on 4 April 1990 and coming into force on 1 July 1997. He also referred (para 74, 104A-C) to matters which are a legitimate guide to statutory interpretation under Pepper v Hart [1993] AC 593. It is not necessary to repeat all these background matters, but some reference will be made to them below. 19. Section 3 of the Rent Ordinance identifies the leases to which it applies. They are referred to as “applicable leases”. It is common ground that it applies to the lease of Inland Lot 8874, Electric Road (“the site”) which is the subject-matter of this appeal. 20. By s 6(1) the lessee under an applicable lease is liable to pay, as Government rent, “an annual rent of an amount equal to 3 per cent of the rateable value of the land leased”. By s 6(2)(a) the lessee is treated as having covenanted to pay the Government rent as assessed “under and in accordance with this Ordinance” – that is the Rent Ordinance, not the Rating Ordinance. As Sir Anthony Mason NPJ put it in Agrila, para 19 (94A-D), the two statutory regimes are distinct, but a concept originating in the Rating Ordinance is applied by the Rent Ordinance for a different purpose. 21. Sections 10 and 11 of the Rent Ordinance contain provisions similar to those of ss 11 to 14 of the Rating Ordinance. The Commissioner prepares the Government Rent Roll (“the Roll”) each year at the direction of the Chief Executive, who has power to designate the date (“the valuation date”) by reference to which rateable values are to be assessed for rent purposes. The present practice, matching that under the Rating Ordinance, is for the Roll to come into force on 1st April based on valuations as at 1st July (the valuation date) in the preceding year. This appeal relates to the first Roll, for which the interval between the dates was rather longer. The first Roll came into force on 28 June 1997, based on valuations as at 1 July 1996. 22. By s 7(1) the rateable value of land which is let under an applicable lease is the aggregate of the rateable values of the tenements comprised in that land. By s 7(2) rateable values are to be those set out in the Roll (or in an interim valuation, which is not relevant to this appeal). Section 8(1) gives the Commissioner power to value land held under an applicable lease in order to ascertain its rateable value for the purposes of s 7, and by s 8(2) the Rating Ordinance “applies to the ascertainment of rateable values under this Ordinance subject to any specific provisions of this Ordinance”. 23. Sections 16 to 27 contain a procedural code for challenges to decisions of the Commissioner. Proposals and appeals made by the lessee in respect of “identical tenements” (as defined in s 18 of the Rent Ordinance) may be made only under the Rating Ordinance. Proposals and appeals in respect of non-identical tenements are to be made under the Rent Ordinance. Building sites are non-identical tenements, because they are not included in valuation lists for the purposes of rating under the Rating Ordinance. Under s 17 of the Rent Ordinance a lessee may make a proposal for reduction of a rateable value shown on the Roll. The Commissioner then serves a decision notice under s 25. By s 26 a dissatisfied lessee may appeal to the Lands Tribunal, whose powers are laid down in s 27. There is a further appeal to the Court of Appeal, but only on a point of law, under s 11 of the Lands Tribunal Ordinance, Cap 17. 24. Section 34 of the Rent Ordinance gives the Chief Executive power to make regulations enabling rateable values to be ascertained for “land, including interests held under applicable leases”. The Rent Regulation was made under this power. Section 2 of the Rent Regulation provides as follows: “2. Rateable value of leased land before development Where any leased land has not been developed after the commencement of the term of the applicable lease under which it is leased, the rateable value of the leased land at any time before any part of it is developed shall be ascertained as if the leased land were a tenement liable for assessment to rates under the Rating Ordinance (Cap 116).” Section 4 applies to land before redevelopment (defined as development after a previous development of land subject to a lease, so that it applies where a building has been constructed and then demolished during the term of the same lease). The rateable value is to be taken as the rateable value of the site as if the demolished building were still standing. This is called the last available rateable value (“LARV”). Sections 3 and 5 contain corresponding provisions for land that is partially developed or partially redeveloped. In Agrila this Court upheld the validity not only of s 2 but also that of ss 4 and 5 and (by implication) s 3. This issues on this appeal 25. In Agrila this Court was concerned with all nine of the questions which the Tribunal had raised for preliminary rulings on points of law. On this appeal the Court is concerned with only one of the questions (the fourth), and with the full implications of the answer to that question given in the judgment of Sir Anthony Mason NPJ in Agrila. But it may be helpful to set out the first question and answer also; both these questions and answers are set out in para 117 (113A-C, 113I-114D) of his judgment: “Point 1: Whether the Commissioner is required or empowered by s.8 of the Rent Ordinance or regulation 2 of the Rent regulations to ascertain the rateable value of leased land before or during development otherwise than in accordance with sections 7 and 7A of the Rating Ordinance (and the rebus sic stantibus rule) whether or not it is liable for assessment to rates under the Rating Ordinance. Answer : (i) For the purposes of determining the amount of Government rent payable, regulation 2 deems that the leased land is ‘a tenement liable for assessment to rates under the Rating Ordinance’; (ii) Thus, on a true construction of regulation 2, the non-rateability of the leased land under the Rating Ordinance is to be disregarded and the rateable value is to be ascertained on that basis in accordance with sections 7 and 7A of the Rating Ordinance. … Point 4 : When making a valuation under s.8 of the Rent Ordinance of leased land before or during development what assumptions, whether using the contractors or another basis of valuation, the Commissioner is required or empowered by law to make as to :- (a) the terms of the hypothetical tenancy of the land; (b) the state of the land; and (c) the mode or character of occupation of the land. Answer : (a) The terms of the hypothetical tenancy are given by s.7(2) of the Rating Ordinance. (b) The state of each site should be taken as it was on the relevant date, having regard to all the intrinsic characteristics of each site. When determining those characteristics, the Lands Tribunal should take into account evidence as to the likelihood at the relevant date of development being carried out on each site by the hypothetical tenant amongst other relevant considerations. (c) The evidence referred to in (b) above should be taken into account in determining the mode or category of occupation for each site at the relevant date and in particular whether the site was being occupied as a development site.” 26. The general issue in this appeal is whether the Tribunal made an error of law in its understanding and application of the principles embodied in, or underlying, the answer that this Court gave to point 4 in Agrila. That general issue has been refined into five more particular issues set out in para 1 of the court’s order dated 1 December 2011 granting leave to appeal. That part of the order is set out (with some small clerical amendments) in the next paragraph. But this detailed formulation of particular issues, some of them (such as the rebus sic stantibus principle) of a technical character, must be read subject to an important caveat. The court’s task is essentially an exercise in statutory construction. The task is to determine the legislative intention as expressed in the text of the Rent Ordinance, the Rent Regulation and the parts of the Rating Ordinance which they incorporate by reference. Rating law has a long history, as already noted. But respect for intellectual freight from the past must not obstruct a purposive approach which looks for the aims and scheme of the legislation as a whole. The five particular issues identified in the order granting leave to appeal are interwoven strands of the one essential issue of statutory construction. 27. The order states the issues as follows: (a) issues formulated in paragraph 17.2 of the Amended Application for Leave to Appeal to the Court of Final Appeal, namely (i) what the function of the rebus sic stantibus principle is in rating law and, in particular, whether it operates to limit the assumptions which may be made about the state of the tenement and how it may be enjoyed, or whether it requires it to be assumed that the tenement will be enjoyed profitably for certain purposes, under the hypothetical yearly tenancy; (ii) what the content of the rebus sic stantibus principle is and, in particular, whether the rateable value of a tenement may be assessed by reference to what it may come to be developed and thereafter used for (even if it could not be so used for that purpose given the state of the tenement on the material date) and, if it may be, what content (if any) the principle has; (iii) whether the rateable value of a tenement may be assessed on the basis that the hypothetical tenant may enjoy it other than as the tenant under the hypothetical yearly tenancy or on the basis that, after the end of that tenancy, he may enjoy any building constructed on it; and (iv) whether the rateable value of a tenement may be assessed on the assumption that there will be a demand to occupy it under a tenancy from year to year (given that those are the only terms on which that tenement may be occupied) if that demand would not exist in the real world in such circumstances; and (b) the issue whether s 2 of the Rent Regulation creates a special regime for the ascertainment of the rateable value of development sites for the purpose of Government rent which requires the value of any development once completed to be taken into account in determining the leased land’s rateable value in a period before any development is completed. The facts 28. The site was bought by Best Origin, a subsidiary of Sino Land Company Ltd, at a Government land auction on 11 December 1996. Best Origin paid a premium of $760 m. The term of the Government lease expires on 30 June 2047. At the time of the sale there were some unoccupied buildings on the site, that is the old Causeway Bay police station and its accommodation block. The new owner was required to demolish these buildings within nine months and to erect a new building to be completed by the end of 2000. The new building was to have a total floor area of not less than 11,000 square metres and the maximum floor area (having regard to the Building (Planning) Regulations) was 18,339 square metres. 29. The completed development was described by the Tribunal in para 10 of its decision: “10. As a matter of fact, a 35-storey building with 2 floors of retail spaces and 5 floors of car-parking facilities, one refuge floor, one floor for plant rooms and 26 floors of office units was erected at the site and occupation permit was issued on 6 December 2000. It is now known as 148 Electric Road.” 30. Demolition of the buildings on the site began before 28 June 1997 and was completed very soon after that date. The agreed statement of facts before the Tribunal included (in paras 38 to 40) the following matters: “Valuation Assumptions 38. The Tenement is to be valued as a ‘development site’ as at 1 July 1996 (‘the Relevant Date’). Pursuant to section 7A(2) of the Rating Ordinance, the Tenement must be valued at the Relevant Date but on the assumption that the matters listed in (a) to (c) of section 7A(2) were as they were on 28th June 1997. The State of the Tenement at 28th June 1997 39. As at 28 June 1997, the state of the Tenement was a ‘development site’ on which then stood a disused police station which was in the process of being demolished. Demolition was completed on 9 July 1997. Factors affecting mode or character of occupation 40. As at 28 June 1997, demolition of the existing Police Station and Accommodation Block had commenced and comprised an early phase of the development of the Site by the construction of a new building. Therefore, the mode and character of occupation was that of a development site, which the Government lessee had the right to build a commercial building with a maximum gross floor area of 18,339 m2.” 31. The agreed statement of facts (dated 18 October 2006) also identified a number of factual issues, and matters of professional judgment as to valuation assumptions, on which the parties had not been able to agree as at that date. But since then agreement has been reached on further issues and matters of judgment. It is therefore unnecessary to summarize more of the statement. The parties are to be congratulated that they have been able to reduce the area of dispute in this way. 32. In the Roll for 1997-98, coming into force on 28 June 1997, the rateable value originally entered was $37,050,000. On the same day Best Origin made a proposal under ss 16 and 17 of the Rent Ordinance to alter the entry. By a notice of decision issued on 20 February 1998 under s 21 of the Rent Ordinance the Commissioner reduced the rateable value to $29,640,000. On 18 March 1998 Best Origin gave notice of appeal under s 26. 33. On 18 June 1998 the Civil Litigation Unit of the Department of Justice furnished the solicitors acting for Best Origin with a copy of a short valuation report on the site prepared by the Commissioner. It valued the site, as at the valuation date, at $741,000,000 and arrived at the rateable value of $29,640,000 by decapitalization at 4 per cent. A note to the valuation stated: “The decapitalization rate adopted is dependent on the type of development being undertaken. Generally the following rates have been adopted: - domestic 3.5%; office/commercial 4%; and industrial/godown 5%.” So the Commissioner was relying “from day 1” (as it was put in the Tribunal’s decision, para 136(3)) on the contractor’s basis (“CB”) method of valuation, as explained and considered further below. The proceedings before the Tribunal 34. The proceedings before the Tribunal (Lam J, President, and Mr W K Lo, Member) occupied a considerable time. There were six weeks of oral hearings during October and November 2006; four further days of oral hearings in December 2006 and March 2007; and then final written submissions, completed by August 2007. Each side put forward reports by several different experts, but not all of them were cross-examined at length. The principal valuation experts relied on by Best Origin were Mr George Doran and (after Mr Doran’s retirement) Mr Lynch. The Tribunal did not accept their conclusions but its rejection of their evidence was because of the faulty legal basis of their instructions. Mr John Charman, Mr David Lee and Professor Gordon Hughes also gave evidence on behalf of Best Origin; the latter’s evidence was quite severely criticized by the Tribunal. The principal expert for the Commissioner was Mr Tang Ping-kwong; Professor Andrew Baum, Mr Nicholas Brooke and Mr Laurence Hatchwell also gave evidence on her behalf. 35. The Tribunal’s decision was published on 25 February 2008. It runs to 324 paragraphs. Paragraphs 1 to 126 are the work of the President, with the first 86 paragraphs discussing and reaching conclusions about the legal principles to be applied, paras 87 to 121 commenting on some of the expert evidence, and paras 122 to 126 commenting on a strike-out summons issued by the Commissioner (on which no order has been made, either by the Tribunal or by the Court of Appeal). Paragraphs 127 to 316 are the work of Mr Lo, and are almost exclusively concerned with the technicalities of the valuation exercise. The final eight paragraphs are written by the President, and deal with the disposal of the appeal. The rateable value of the site for 1997-98 was to be reduced from $29,640,000 to $26,880,000, and the Commissioner was directed to amend the Roll accordingly. 36. There is not in the Tribunal’s decision any discernible difference of opinion between the President and Mr Lo. Their contributions to the decision are clearly intended to be read as a whole. But it is convenient, in considering such a lengthy written decision, to have a general outline of its arrangement. 37. In paras 7 and 8 of its decision the Tribunal cited the answer given by this Court to point 4 in Agrila (para 25 above) and observed that this was the first case in which the Tribunal had to apply the guidance in Agrila. The Tribunal went on to observe (para 18) that the first issue was to determine the implications of the agreed fact that the mode and character of the occupation of the site was that of a development site. 38. That was indeed the first and essential issue of law to be resolved, and the Tribunal’s discussion and resolution of it occupy paras 19 to 86 of the written decision. It is essentially the only issue on the present appeal, although the terms of the order granting leave to appeal to this Court (para 27 above) show that it can be separated out into a number of sub-issues. It would not be helpful to attempt any brief summary of the Tribunal’s reasoning in this part of its decision. It will be better to consider, point by point, the sustained attacks on its reasoning put forward in the written and oral submissions made on behalf of Best Origin. 39. There is no longer any dispute as to the more technical questions of valuation methods and judgment that were addressed mainly by Mr Lo. It is sufficient to record that the Tribunal approved the use of the CB method (described in professional guidance notes set out in para 137 of the decision). It rejected various arguments put forward by Best Origin in contending that the $760 m premium paid for the site, even after adjustment, was not the correct starting-point for the CB calculation. It took the correct starting-point to be an adjusted amount of $611 m (para 206). It then gave detailed consideration to the appropriate decapitalization rate and decided that the rate of 4.4 per cent put forward by Mr Tang was reasonable and appropriate (para 235). 4.4 per cent of $611 m is $26,884,000, but in the final disposal (para 322) it was rounded to four significant figures, that is $26,880,000. The proceedings in the Court of Appeal 40. Best Origin appealed to the Court of Appeal (Tang Ag CJHC, Kwan JA and A Cheung J) which on 19 November 2010 dismissed the appeal after a five-day hearing ([2011] 1 HKC 488). The main judgment was given by Tang Ag CJHC, with whom the other members of the court agreed. 41. As with the decision of the Lands Tribunal, it would not be helpful to attempt a brief summary of the judgment of Tang Ag CJHC. It will be better to consider it, so far as necessary, by reference to the written and oral submissions made on behalf of Best Origin. But before coming to those submissions point by point, it will be useful to take a broader look at two topics of central importance to the argument, that is the rebus sic stantibus principle and the rating hypothesis of a yearly tenancy. The rebus principle 42. “Rebus sic stantibus” is a Latin phrase with the literal meaning “things standing thus” or, in rather more colloquial English, “as things stand”. It has been familiar to lawyers for centuries, but originally in the context of the law of treaties, a branch of public international law. In that context it refers to the doctrine (well-recognized in principle, but controversial in its practical application) that any treaty must be taken to contain an implied term that it may cease to bind the parties after a fundamental change of circumstances. In that context it goes back to the founding fathers of international law, including Grotius in the 17th century and Vattel in the 18th century. It is still relevant today, and is reflected in Article 62 of the 1969 Vienna Convention on the Law of Treaties (see for instance Huang, The Doctrine of Rebus Sic Stantibus in International Law – a Dissertion (Shanghai 1935); McNair, Law of Treaties (1961) pp 681-691; Oppenheim, International Law, 9th ed (1996) pp 1304-1309). 43. The first known use of the expression in the context of rating is in the judgment of Cockburn CJ in R v Fletton Overseers (1861) 3 E&E 450, 465. In this context it is a convenient shorthand expression by which to identify the general effect of statutory provisions such as those found in the three paragraphs of s 7A(2) of the Rating Ordinance: that is assumptions as to (a) the state of the tenement being the same, (b) relevant factors affecting the mode or character of occupation of the tenement being the same, and (c) the state of the locality being the same. In Hong Kong the provisions have a further secondary function, that is to avoid the need for projection of values by bridging the gap (typically nine months) between the relevant date and the date when the valuation list comes into force. As was explained by the Lands Tribunal (President: HH Judge Cruden) in China Light & Power Co Ltd v Commissioner of Rating and Valuation [1996] RA 475, 482-483, this was the purpose of the introduction of s 7A in 1981. 44. In the context of rating the expression is therefore no more than a convenient shorthand for referring to statutory assumptions which are (as appears from many of the authorities cited in argument) applied in a fairly flexible way in order not to depart too far from the realities of the situation. (These assumptions, as well as the assumption of a yearly tenancy, are part of the rating hypothesis, so that there is some overlap with the next section of this judgment.) The fact that the shorthand is expressed in an ancient classical language should not create the impression that it embodies some fundamental principle of justice. Its function is more mundane, that is to provide some reasonably straightforward statutory assumptions in order to simplify the task of official valuers (and the task of tribunals reviewing their decisions) in administering a tax charged on an annual basis. The provisions set limits on the need for valuers and tribunals to inquire into the likely economic effect of future alterations in the structure of a tenement, or in the use to which it is put, or of possible future changes in the neighbourhood where the tenement is situated. 45. That is the purpose of the provisions. They do not require the tribunal to disregard what Lord Pearson, in Dawkins [1969] 2 AC 366, 393 referred to as “a present probability of a future happening”, because “the present probability affects the present value of the hereditament” (in that case part of the hereditament, a factory, was likely to be demolished for road-widening purposes within about a year). 46. The provisions must also be applied so as to take account of the intrinsic character of the tenement in question. Mr Howell QC (for Best Origin) mentioned as examples a tenement consisting of a gravel pit or a landfill site. If a gravel pit or a landfill site is in active use it is inevitable, and intrinsic to its character as a tenement, that it will change appreciably in the course of a year, as gravel is extracted or waste is put in. The same is true of an active development site, as the site is agreed to have been in 1997-98. Within the period of four years after the Government land auction (with the 1997-98 rental year starting shortly after the beginning of that four-year period) the site went from one with derelict police buildings awaiting demolition to a new 35-storey building for retail and office use. Throughout the period it must have been in a state of almost ceaseless change. The rating hypothesis: the problem of the yearly tenancy 47. “The rating hypothesis” is not a term of art, but it is often used to identify the hypothetical test laid down by statute as the basis for the ascertainment of rateable values. That test is laid down in Hong Kong by the joint effect of ss 7(2) and 7A(2) of the Rating Ordinance (and in England, by similar statutory provisions). The general legislative purpose of the test is to achieve uniformity in valuation for rating purposes, and fairness as between one ratepayer and another. As Hodson LJ said in Humber Ltd v Jones (Valuation Officer) (1960) 6 RRC 161, 166: “It is clear that it is necessary to set up some standard. The actual rent paid would not be an adequate standard for rating purposes, because it might be complicated by there being a small rent because of the payment of a premium, or perhaps because it was an old rent fixed a long time ago when the economic value of the hereditament was low.” 48. The part of the test laid down in s 7A(2), that is the rebus principle, has already been considered. It is now necessary to consider the part of the test laid down in s 7(2), and in particular the yearly nature of the tenancy which it postulates: “the rent at which the tenement might reasonably be expected to let, from year to year, …” on stated assumptions as to the obligations of the landlord and the tenant respectively under the tenancy. It is the statutory requirement of a hypothetical tenancy from year to year that has caused great difficulty in this case, as it has in numerous other reported and unreported decisions for more than a century. In the paragraphs that follow the expression “the rating hypothesis” is directed primarily to the yearly character of the tenancy postulated by s 7(2). 49. At common law a yearly tenancy can be terminated by either party on six months’ notice expiring at the end of a year of the tenancy. The rating hypothesis disregards statutory interventions in the common law of landlord and tenant, whether they take the form of restrictions on the level of rent or protection of security of tenure (Poplar Metropolitan Borough Assessment Committee v Roberts [1922] 2 AC 93). So in relation to premises used for heavy industry or for the purposes of a public utility such as a power station, involving large-scale investment in expensive infrastructure and plant, there is often great difficulty in the application of the hypothesis. For instance Willmer LJ said in Humber Ltd v Jones (Valuation Officer) (1960) 6 RRC 161, 171: “Bearing in mind that the statutory hypothesis is a tenancy from year to year, everybody agrees that, in the world as it is, no sane manufacturer would take a tenancy from year to year of a factory in which he is going to install valuable machinery and for which he is going to assemble a skilled labour force to work for him.” Lord Millett NPJ applied these observations to the Lamma Island power station in Hong Kong Electric Co Ltd v Commissioner of Rating and Valuation [2011] 4 HKC 509, para 146. 50. The same considerations apply to a development site on which major construction work is proceeding. In this case the Tribunal, echoing the observations just cited, stated (para 29): “Obviously, there are tenements that no tenant would take up a tenancy from year to year in the real world. Lord Pearce [in Dawkins [1969] 2 AC 366, 381-382] gave the examples of sewage works, portion of railway lines. Development sites fall within the same category. In the real world, no developer with his right mind will take up a site for development based on a yearly tenancy.” 51. Since the 19th century courts have struggled with this problem. One theme that emerges strongly is that although a yearly tenancy can be terminated at relatively short notice, it will unless terminated continue from year to year, indefinitely. One of the earliest observations on this point, repeatedly cited in later authorities, is that of Cockburn CJ in Great Eastern Railway Co v Haughley (1866) LR 1 QB 666, 679: “But I think it is one thing to start with the assumption that you are dealing with a tenancy from year to year, and another thing to say that the hypothetical tenant, in calculating what he can reasonably pay as rent for the premises, is necessarily to assume that his tenancy would not last beyond a year. I think the possibility of its longer duration is one of the surrounding circumstances which the tenant from year to year would take into account.” 52. It is unnecessary to multiply citations on this point. But there are two decisions of the House of Lords that are of particular interest. One is the case of Railway Assessment Authority v Southern Railway Company [1936] AC 266. Mention has already been made of the speech of Viscount Hailsham LC, part of which was quoted by Tang Ag CJHC in para 2 of his judgment in the Court of Appeal. But it bears repetition, not least because it goes on to refer to the CB method of valuation, adopted by the Tribunal in this case, as an appropriate solution to the problem. After describing the early history of rating in England Viscount Hailsham continued: “My Lords, this method of assessing liability for rates was no doubt very suitable for the simple conditions which prevailed in the reign of Queen Elizabeth; but when it was sought to apply it to modern conditions, and especially to the assessment of great public utility undertakings, such as railways, waterworks, gasworks and the like, whose operations might easily extend over great areas, stretching far beyond the limits of any individual parish, the greatest difficulty was found in adapting the statutory provisions to the facts of such a case. The Courts, more than once, appealed to the Legislature to relieve them from the burden thus cast upon them; see, for example, the judgment of Lord Campbell in Reg. v. Great Western Ry. Co. [(1851) 15 QB 379, 396, 397] and the judgment of Wightman J. in Reg. v. West Middlesex Waterworks Co. [(1859) 1 E&E 716, 728]. Unfortunately, the Legislature did not see its way to intervene and the Courts were compelled to evolve a system of calculation, with the assistance of the expert advisers to the rating authorities and to the undertakings, which involved a number of very difficult assumptions, but which received the approval of your Lordships’ House on a number of occasions, and which became recognized as the standard method of assessing the rateable value of hereditaments in these cases. So far as the railways were concerned, the method adopted was to divide the undertaking into those portions which were regarded as directly productive of profit, such as the permanent way, and those portions which were regarded as only indirectly productive, such as stations and other hereditaments of that kind. The latter were usually valued by taking the cost of erection and allowing a percentage upon that cost, on the theory that the companies would not have erected them unless they had regarded them as worth at least a fair average rate of interest upon the money invested in their construction. This method was commonly described as the ‘contractor’s basis.’” 53. The other case that calls for further mention is Dawkins [1969] 2 AC 366. The ratepayers owned a factory, part of which was compulsorily acquired for road-widening. It was then let to the ratepayers on a yearly tenancy, which was to be terminated when the road works started. So there was an actual yearly tenancy which really was likely to be terminated in the near future, as well as the hypothetical yearly tenancy postulated by the rating hypothesis. The House of Lords was split, but the majority decided that the prospect of early termination, in the real world, should be recognized in applying the hypothesis. 54. Lord Guest, who was one of the dissenting minority, said at p 380: “Some difficulty may have been caused by undue emphasis on the expression ‘year to year’ as if this was limited to a yearly tenancy. The expression ‘taking one year with another’ which appears in the Valuation (Metropolis) Act, 1869, quoted in Poplar Metropolitan Borough Assessment Committee v Roberts [1922] 2 AC 93 and treated as having the same effect as a tenancy ‘from year to year’, makes it clear that, although as a matter of valuation a yearly tenancy is to be assumed, the hypothesis is that it will be of indefinite duration.” 55. Lord Wilberforce (in the majority) made a thorough review of the old cases and referred to “the practical approach” of the judges. He also referred to “taking one year with another” as having always been regarded as an identical test. He concluded at p 387: “So we should regard the words ‘from year to year’ as meaning no more than that the tenancy is not a fixed or definite one; it is one of indefinite duration, determinable by notice, but not, I would think, according to the technicalities governing the giving of notice in tenancies of this kind.” 56. Lord Pearson (also in the majority), after citing the observations of Cockburn CJ in Great Eastern, saw no inconsistency with the rebus principle (at p 393): “It was also said on behalf of the appellant that equality of rating requires that each hereditament should be valued as it now is – rebus sic stantibus – and the prospect of a future partial destruction of it must be disregarded. But it seems to me that this point can be turned against the appellant. In the expression rebus sic stantibus which are the res? In other words, which are the factors to be taken into account in order to produce equality of rating? There is, in this case, a present probability of a future happening, and the present probability affects the present value of the hereditament.” 57. These and other authorities cited to the Court show that over the years the courts and tribunals have understood and applied the statutory hypothesis in a practical and flexible way so as to assess rateable values as fairly and uniformly as possible. There have been a few cases in which it has been held that a rateable tenement (or hereditament) has a nil rateable value. The court was shown two: Black v Oliver [1978] 1 QB 870 and Hoare (Valuation Officer) v National Trust [1998] RA 391. These were both cases on very unusual facts: at one extreme a house that was unfit for human habitation, but occupied by an unfortunate tenant who might lose her claim to be rehoused if she moved out; and at the other extreme historic mansions owned by the National Trust, which on the rating hypothesis was the only possible tenant for the buildings, which were very expensive to maintain. The extreme nature of these examples serves to emphasize that despite the difficulty of applying the hypothesis to some types of tenement, tribunals and courts do their best to apply it in a practical way, and it is very unusual for a rateable tenement or hereditament to be found to have no rateable value. The well-known case of Consett Iron Co v Assessment Committee for North Western Area of Durham [1931] AC 396 is a striking example of this. A loss-making coalmine, with no prospect of early economic recovery, was held to have a significant rateable value, despite the yearly tenancy postulated by the rating hypothesis. 58. Both leading counsel agreed that there is some difference between the Hong Kong case law and the English case law as to how far the rebus principle permits a valuer to contemplate physical changes in a tenement (on the one hand) and changes in its use (on the other hand), and the interaction between these matters. On the authorities as they stand that appears to be so, although the point has not yet been definitively considered by this Court. It is not necessary to go into it on this occasion. The point is of some general importance, but in the context of this appeal it is a matter of fine tuning. It cannot be determinative of the more fundamental issue now before the Court. So far as it has anything to contribute to the resolution of this appeal, I would proceed on the basis that the Hong Kong approach is rather more liberal. The rating hypothesis: how far should it be taken? 59. In Hoare (Valuation Officer) v National Trust [1998] RA 391, 408 Schiemann LJ stated, “The statutory hypothesis is only a mechanism for enabling one to arrive at a value for a particular hereditament for rating purposes. It does not entitle the valuer to depart from the real world further than the hypothesis compels.” Similarly in the same case Peter Gibson LJ drew attention to other areas of law, such as compulsory purchase and taxation, in which statutory hypotheses are regularly encountered. He cited what Hoffmann LJ said about valuation in a capital transfer tax case, Inland Revenue Commissioners v Gray [1994] RVR 129, 136: “It cannot be too strongly emphasized that although the sale is hypothetical, there is nothing hypothetical about the open market in which it is supposed to have taken place.” The same applies, equally emphatically, to the hypothetical yearly tenancy postulated by the Rating Ordinance. It must take account of real market conditions: Hong Kong Electric Co Ltd v Commissioner of Rating and Valuation [2011] 4 HKC 509, 549, para 147(d), where Lord Millett NPJ set out with approval the Tribunal’s summary. 60. So Hoare is authority for the proposition that the rating hypothesis should not be taken further than is necessary, and the decision of the House of Lords in Dawkins is to the same general effect. But how, in this context, is necessity to be judged? Mr Howell cited a passage from the speech of Lord Asquith in East End Dwellings Co Ltd v Finsbury Borough Council [1952] AC 109, 132-133: “ If you are bidden to treat an imaginary state of affairs as real, you must surely, unless prohibited from doing so, also imagine as real the consequences and incidents which, if the putative state of affairs had in fact existed, must inevitably have flowed from or accompanied it … The statute says that you must imagine a certain state of affairs; it does not say that having done so, you must cause or permit your imagination to boggle when it comes to the inevitable corollaries of that state of affairs.” 61. Lord Asquith’s epigrammatic saying has often been cited, one such citation being by the Lands Tribunal in China Light & Power Co Ltd v Commissioner of Rating and Valuation [1996] RA 475, 485. But I respectfully doubt whether it is entirely consistent with the purposive approach that courts now take on issues of statutory construction. Any statutory hypothesis has an element of “makebelieve” (the expression used by Willmer LJ in Humber (1960) 6 RRC 161,171), since it is treating some state of affairs as real regardless of whether it exists, or even if it is known not to exist, in the real world. Identifying the “inevitable corollaries” of a hypothetical state of affairs may be problematical. 62. The trend of more recent authorities can be seen in cases such as Inland Revenue Commissioners v Metrolands (Property Finance) Ltd [1981] 1 WLR 637, in which Nourse J said: “When considering the extent to which a deeming provision should be applied, the court is entitled and bound to ascertain for what purposes and between what persons the statutory fiction is to be resorted to.” This emphasis on statutory policy and purposes was approved by the Court of Appeal and the House of Lords in Marshall v Kerr (1993) 67 TC 56, [1995] 1 AC 148, although reference was also made to inevitable consequences and incidents, language clearly derived from East End Dwellings. 63. This reasoning was developed by Neuberger J in Jenks v Dickinson [1997] STC 853, 878: “It appears to me that the observations of Peter Gibson J, approved by Lord Browne-Wilkinson, in Marshall indicate that, when considering the extent to which one can ‘do some violence to the words’ and whether one can ‘discard the ordinary meaning’, one can, indeed one should, take into account the fact that one is construing a deeming provision. This is not to say that normal principles of construction somehow cease to apply when one is concerned with interpreting a deeming provision; there is no basis in principle or authority for such a proposition. It is more that, by its very nature, a deeming provision involves artificial assumptions. It will frequently be difficult or unrealistic to expect the legislature to be able satisfactorily to [prescribe] the precise limit to the circumstances in which, or the extent to which, the artificial assumptions are to be made.” This passage was approved by the Supreme Court of the United Kingdom in Commissioners for HM Revenue and Customs v DCC Holdings (UK) Ltd [2011] 1 WLR 44, para 39 (and para 40 gives an example of apparently inevitable consequences which could not sensibly have been intended). 64. The statutory purpose of the rating hypothesis is, as already noted, to provide a pattern or template for ascertaining rateable values, in a uniform and reasonably simple way, and on a yearly basis. It is important to note that it is not only the hypothetical tenancy, but also the actual process of assessment, that is on a yearly basis. Mr Holgate pointed out that even in a situation in which the hypothetical tenant could expect to remain as tenant for a long time (such as Humber (1960) 6 RRC 161, 169, where there was evidence that “if the landlord gave notice to the tenant in a case of this kind, he would know that he would never find another tenant to go into that factory”), the Rating Ordinance requires a rateable value to be fixed for every rateable tenement on a yearly basis. The prospective length of the hypothetical tenancy, and events that may or may not happen in the course of its duration, are relevant only if and so far as they affect the current rating year; similarly in the next rating year, and so on. The more remote such future happenings are, in time or in probability (or both), the less plausible is the argument for treating them as within the scope and legislative purpose of the rating hypothesis. It will be necessary to come back to this in considering various disputed points on the Tribunal’s decision. The Rent Regulation revisited 65. In Agrila the Court of Appeal [2000] 1 HKC 175 held that s 2 of the Rent Regulation was nugatory in its effect. This Court reversed that conclusion. The legislative purpose of s 2 is, as Sir Anthony Mason NPJ put it (para 49, 99E-F) “to overcome the problem that building sites are not rateable tenements for the purposes of the Rating Ordinance”. 66. Mr Howell, without seeking to challenge that statement of the section’s legislative purpose, pointed out that the section’s scope is not limited to development sites. In the section (the text of which is set out at para 24 above) the word “development” (defined in s 1) is not actually used at all (although the word does appear in the heading of that section), but the word “developed” (used twice) must be intended to have a meaning corresponding to the definition, and it is reasonably clear from the context that it refers to the completion of a development. The words of the deeming provision, “as if the leased land were a tenement liable for assessment to rates,” make sense only in relation to land which would, in the absence of the deeming provision, not be liable for assessment to rates. It could apply, at least in theory, to agricultural land and certain other categories of land or buildings which under s 36 of the Rating Ordinance are exempt from liability for assessment, and so are not rateable tenements. But it must be borne in mind that these provisions are relevant only in relation to land comprised in applicable leases (as defined in s 3 of the Rent Ordinance). In practice, the importance of s 2 is in relation to active development sites. 67. That is confirmed by the legislative history of the Rent Regulation, to which Sir Anthony Mason NPJ referred in paras 67 to 74 (102J-104C) of his judgment under the principle in Pepper v Hart [1993] AC 593. Each side sought to derive some advantage from this material. Mr Howell pointed to the closing remarks of the Secretary for Planning, Environment and Lands in opposing Mr Arculli’s amendment (Hansard for 25 June 1997, pp 812-813): “There is no question of the Administration redefining the rateable value. The purpose of section 2 is simply to enable the Administration to ascertain the rateable value for the purpose of determining the amount of Government rent payable as required by the Joint Declaration and the Basic Law, yet leaves room for the situation where a minimal rateable value or [nil-agreed correction] rateable value is ascertained in which case it is accepted that no Government rent would be payable.” Mr Holgate pointed out that the original wording of s 2 had provided for the rateable value of newly granted sites to be 5 per cent of the market value of the land (p 811) and that the current wording (which Mr Arculli’s amendment proposed to change) followed closely the general lease conditions (p 812 - these conditions are referred to in para 58 (101B-E) of the judgment of Sir Anthony Mason NPJ). 68. Sir Anthony Mason NPJ concluded on this point (para 74, 104A-B) that “the legislative history strongly confirms the meaning already placed on s 8(2) and regulation 2”. By that he did not of course refer to the judgment of the Court of Appeal, which was reversed, but to para 49 (99E-H) of his own judgment. Some reference has already been made to that paragraph but it is best to set it out in full: “ Viewed in the light of these well-established principles of rating law, the purpose of reg.2 seems to be reasonably clear. It is to overcome the problem that building sites are not rateable tenements for the purposes of the Rating Ordinance. The regulation achieves this purpose by providing that the rateable value of the leased land before any part of it is developed shall be ascertained ‘as if the leased land were a tenement liable for assessment to rates under the Rating Ordinance’. The regulation says nothing about how the rateable value is to be ascertained. That function remains to be dealt with, as s.8(2) of the Rent Ordinance prescribes, in accordance with ss.7 and 7A of the Rating Ordinance. The regulation makes no attempt to displace the operation of these sections. So the preferable meaning to be given to the words quoted above is that they require the rateable value of the leased land to be ascertained on the assumption that it is rateable tenement.” 69. There was also some discussion, in the course of argument, about the contrast between s 2 and s 4 of the Rent Regulation. Section 4 contains the LARV rule as described in para 24 above. Mr Howell suggested that as s 4 undoubtedly looks to the past, it would be odd if s 2 were to be construed as looking to the future in order to charge tax by reference to the completed development. But against that it might be said that what the sections have in common is that neither seeks to levy tax in respect of value added by expenditure on construction works during the development (or redevelopment). A route map for discussion of the issues 70. It will be apparent from the preceding sections of this judgment that the Court’s task, with the assistance of counsel’s very thorough written and oral submissions, is to decide questions of some difficulty and great public importance. The essential issue is one of statutory construction, but it is an area of law with a long history and a lot of authority. The best way forward may be to simplify and rearrange the issues set out in para 27 above, as follows: (1) Does s 2 of the Rent Regulation create a special regime? (2) What is the content of the rebus principle in relation to active development sites falling within s 2? (3) How does the yearly tenancy postulated by the rating hypothesis apply in relation to those sites? 71. Each of these questions (which inevitably overlap to some extent) must be taken in three stages: (a) Was it decided by this Court in Agrila? (b) Did the Tribunal make a significant error of law, either by misunderstanding or misapplying the principles and guidance to be found in Agrila, or by getting the law wrong on a point not covered by Agrila? (c) Did the Court of Appeal correct any part of the Tribunal’s reasoning which was unsound, or reinforce its reasoning when it was sound? As to what Agrila decided, valuable guidance is available in the judgment of Ribeiro PJ in Hong Kong Electric Co Ltd v Commissioner of Rating and Valuation [2011] 4 HKC 509, 521-528, paras 24-58, with which the other members of this Court agreed. As to the Court of Appeal’s decision in this case, it may be noted that although this appeal is in form an appeal from the judgment of the Court of Appeal, counsel’s arguments have largely focused on the decision of the Tribunal. But some references have been made to the judgment of Tang Ag CJHC in the Court of Appeal. In the discussion that follows “CFA” refers to the judgment of Sir Anthony Mason NPJ in Agrila, “LT” to the Tribunal’s decision in this appeal, and “CA” to the judgment of Tang Ag CJHC in the Court of Appeal. Does section 2 create a special regime? 72. This question, if it is to be answered shortly, admits of only one answer: No. Sir Anthony Mason NPJ made that clear (CFA para 49, 99E-H, set out in full in para 68 above). It is also spelled out in the answer to point 1, and reinforced by the answer to point 2 (“This point does not arise”), and by the reference to s 7(2) of the Rating Ordinance in the answer to point 4 (the terms of which, as originally formulated, were not satisfactory: CFA para 98, 109C-D). 73. This is however an area in which any one-word answer is incomplete, and may give rise to misunderstanding. Section 2 altered (for the purposes of Government rent) something which had until then been regarded (CFA para 48, 99B-E) as a fundamental proposition of rating law, without (CFA para 49, 99E-H) saying anything about how the rateable value of a development site was to be ascertained. This Court has held that the alteration cannot be regarded as nugatory. It was therefore necessary to go back to first principles in considering the implications for the ascertainment of rateable value of this novel enactment which requires development sites to be treated as rateable tenements (whereas previously they had been treated as not rateable, because they were regarded as not in actual and/or beneficial occupation). 74. Sir Anthony Mason NPJ touched on this point, rather inconclusively, in a passage emphasizing the distinction between rateability and rateable value (CFA paras 86 and 87, 106F-107B). Then after a reference (CFA para 89, 107F) to a well-known passage in the judgment of Scott LJ in Robinson Brothers (Brewers) Ltd v Houghton and Chester-le-Street Assessment Committee [1937] 2 KB 445, 468-469 as to the need for the value to take into account “every intrinsic quality and every intrinsic circumstance which tends to push the rental value either up or down”, Sir Anthony emphasized the importance, in ascertaining the rateable value of a development site, of recognizing its character as such a site (CFA para 94, 108E-F): “It follows that, in ascertaining the rateable value of the sites, it is permissible to have regard to their character as development sites for that is an intrinsic characteristic of each property. Having regard to that characteristic entails taking account of the likelihood of development taking place and proceeding to completion.” This is an important passage in which the use of the words “character” and “characteristic” is significant, since s 7A(2)(b) of the Rating Ordinance refers to “the mode or character of occupation”. Like a gravel pit or a landfill site in active use, an active development site has an intrinsic characteristic of change. In the case of a development site the direction of change is, in the normal case, towards completion of the development. 75. The Tribunal recognized the importance of this point. It is worth noting that all the passages in its decision relevant to this appeal come under the heading (immediately after LT para 18), “The implications of the site being a development site”. The point is developed in a passage starting at LT para 26: “In our view, it is almost self-evident that once it is concluded that the site is a development site, there must be a sufficient likelihood of change of use so as to affect the rent that a hypothetical tenant would pay.” The change of use referred to is of course the completion of the development, when (in this case) the new building would be available for letting as over a hundred retail and office units. The discussion of this point continues to LT para 29, a passage that has already been quoted (para 50 above). It is picked up again in LT paras 33 and 52 to 54. These are among the passages that Mr Howell has singled out for criticism. 76. Tang Ag CJHC (CA para 36) expressed his complete agreement with LT para 54, part of which is as follows: “It is difficult to see how the development potential of a development site can properly be taken into account if the rating hypothesis is to be construed in such a way to confine the expected duration of occupation to the period of construction or a short time afterwards for marketing of the units … The intrinsic quality in terms of the development potential of the site (which is the actual purpose of occupation in the real world) will not be taken into account. It would become the value of a bare construction site without any development potential. Such an approach is inconsistent with the judgment of the Court of Final Appeal in Agrila and it is also against the principles set out in the other authorities mentioned above [notably Robinson Brothers and Consett Iron].” Tang Ag CJHC went on (CA paras 36 and 38) to answer an objection raised by Mr Howell in relation to CFA para 94 (108E-G). He returned to mode and character of occupation in CA para 60. 77. On this question as a whole I would conclude that although s 2 of the Rent Regulation certainly has not introduced a new regime displacing ss 7 and 7A of the Rating Ordinance, it has posed a new problem as to how the new deeming provision in s 2 is intended to interact with the rating hypothesis in ss 7 and 7A. The solution to that problem has in my view been correctly identified as starting with the recognition of the intrinsic character of these sites as development sites. What is the content of the rebus principle in relation to active development sites? 78. Sir Anthony Mason NPJ referred (CFA paras 86-88, 106F-107D) to Dawkins [1969] 2 AC 366, including Lord Wilberforce’s reference (at p 386) to the hypothetical tenant taking into account “not only any immediately actual defects or disadvantages (such as planning restrictions) but disadvantages, or advantages, which he can see coming”, and Lord Pearson’s reference (at p 393) to “a present probability of a future happening, and the present probability affects the present value of the hereditament”. Sir Anthony continued (CFA para 89, 107E-G): “Dawkins is an answer to the main thrust of the respondents’ case that s 7A(2)(a) and the rebus principle require that the valuation must be based on an actual tenement in its existing state. The point is that, although the rebus principle requires the tenement to be valued as in fact it is, the valuer must consider ‘every intrinsic quality and every intrinsic circumstance which tends to push the rental value either up or down’ (Robinson Brothers (Brewers) Ltd v Houghton and Chester-le-Street Assessment Committee [1937] 2 KB 445 at 468-469, per Scott LJ.” 79. Sir Anthony Mason NPJ then applied these general principles to a tenement with the intrinsic character of a development site (CFA para 92, 108A-C): “The fact that occupation of a construction site does not enable a hypothetical tenant to make a profit during the construction period does not mean that the property has no rateable value. Occupation of the site may nevertheless be valuable and command a significant rent (London County Council v Erith [1893] AC 562 at 591).” Then after a reference to Consett Iron he continued (CFA paras 93 and 94, 108C-F): “If there is a sufficient likelihood of a change of use of the property as would affect the mind of a hypothetical tenant and alter the rent he would pay for it in its existing state, that is a matter to be taken into account in the valuation. And the current occupier is to be regarded as a party who might become the hypothetical tenant (London County Council v Erith). It follows that, in ascertaining the rateable value of the sites, it is permissible to have regard to their character as development sites for that is an intrinsic characteristic of each property. Having regard to that characteristic entails taking account of the likelihood of development taking place and proceeding to completion. But this does not mean that the sites should be valued as completed developments.” 80. These are very important paragraphs in the Agrila judgment. They are expressed in general terms, because (as Ribeiro PJ pointed out in Hong Kong Electric, para 29) this Court was in Agrila dealing with issues in principle and in relation to 59 development sites, on some of which (CFA para 13, 92I-93B) development had not advanced beyond the erection of boardings. The above passages emphasise the value of occupation (in rating terms, the value to the hypothetical tenant), this being the statutory intention behind s 7(2) of the Rating Ordinance: see Hong Kong Electric, para 65. 81. London County Council v Erith [1893] AC 562 was concerned with a tenement which was incapable of ever yielding a profit, but was nevertheless rateable because its occupation was of value to the statutory undertaker in the performance of its functions. The facts are summarized in the headnote: “The London County Council were owners of land and premises consisting of a pumping-station and works, which they occupied and used as a necessary part of the metropolitan sewage system and to enable them to perform statutory duties. So long as the land and premises were used as part of the sewage system they were incapable of yielding a profit and the London County Council were practically the only possible tenants.” There is a single speech by Lord Herschell LC with which the rest of the Appellate Committee of the House of Lords concurred. It makes clear, in a long passage at pp 588-592, that occupation can be of value even though no pecuniary profit is derived from the occupation. There is a citation from the judgment of Fry LJ in R v School Board for London (1886) 17 QBD 738: “The term ‘sterility’ has been introduced into the cases because, as a general rule, a profit is produced; but it does not by any means follow that because there is no profit there is no value.” There is also a reference, at p 593, to what can be recognized as the CB valuation method. 82. The Tribunal cited CFA paras 92 and 93 (108A-E) at the beginning of its discussion of the implications of the site being a development site (LT paras 20 to 23). But the fact that Sir Anthony Mason NPJ chose to refer (in CFA para 93, 108D) to a change of use (rather than the successful completion of the development) seems to have led the Tribunal to suppose that a change of use would by itself necessitate the coming to an end of the hypothetical tenancy. This was in my view an error of law (but not one which ultimately affects the outcome of the appeal). It was an error because the hypothetical tenancy is to be regarded as continuing indefinitely, unless it is terminated by notice given by the hypothetical landlord. It is true that in the real world the letting of the completed building as more than a hundred separate tenements would lead to a multiplicity of new hypothetical tenancies as part of a valuation exercise undertaken in or about the year 2000. That was foreseeable, but it was irrelevant to the valuation exercise carried out as at 28 June 1997, for reasons already considered (para 64 above). 83. This error led the Tribunal to the notion (which emerges in a clear form in LT para 37) of the developer continuing to be entitled to occupation (in substance, to enjoyment of profits from the completed development) after the end of the hypothetical yearly tenancy. This tends to distort some of the Tribunal’s reasoning. Nevertheless the Tribunal understood and followed the essential point of what Agrila has to say about how the rebus principle applies to development sites: that the prospect of successful and profitable development is already, during the course of construction, an intrinsic characteristic of a development site. That is clear from LT paras 41 and 46-49. 84. In those paragraphs the Tribunal used the expression “development potential” on more than one occasion. Mr Howell was critical of this, and Mr Holgate was prepared to accept that the expression is ambiguous. It is an imprecise expression, but its general meaning is reasonably clear. Developers develop sites because of their development potential, which in economic terms is the prospect of profiting from the development. In this case the clearest indication of the site’s development potential is the sum of $760 m that Best Origin paid at the keenly contested Government land auction in December 1996. Paragraphs 28 and 29 of the agreed statement of facts show that this was not a freak result, but a reflection of the strong demand and limited supply on the market at that time. Mr Howell’s submission, which is central to his case, that the Tribunal felt bound to conjure up a demand that could not have existed consistently with the rating hypothesis, depends almost entirely on the yearly nature of the hypothetical tenancy. That submission is considered in the next section of this judgment. 85. In the Court of Appeal Tang Ag CJHC referred (CA paras 20 and 50) to CFA paras 90 and 92 (107G-I and 108A-C), and to the prospect of continuation of the hypothetical tenancy. In parts of his judgment he seems, as I read it, to have come close to commenting on and perhaps disagreeing with the Tribunal’s view that the developer might remain in occupation after the end of the hypothetical tenancy. In CA para 61 he described the point (correctly in my view) as irrelevant to the ascertainment of the rateable value of the site as a development site. 86. It may be worth mentioning, but only as a footnote, that in his submissions Mr Holgate insisted that the site should during the construction period be regarded as a bare site, with the building in course of erection on it disregarded. This submission must have been made in order to assist his secondary, or fallback, case as to the severance of the building from the land. It does not assist his primary case, and so far as it is relevant at all I do not think it is correct. The intrinsic character of a development site is to grow, in its physical appearance and in its value. It is not immediately productive of profit, but its long-term growth and productivity is the antithesis of the “sterility” referred to by Fry LJ in R v School Board for London. 87. On the second question I conclude that in Agrila this Court recognized the importance, in the application of the rating hypothesis, of the intrinsic character of a development site. To answer the question posed by Lord Pearson in Dawkins [1969] 2 AC 366, 393, the carrying out of the development was part of (indeed, the heart of) the res. The Tribunal understood this and applied it, though the Tribunal fell into error over the termination of the hypothetical tenancy. The Court of Appeal followed what this Court had said in Agrila, and rightly regarded the Tribunal’s error as irrelevant. Mr Howell’s rather sanguinary submission that the rebus principle has been eviscerated cannot be accepted. But the principle has been adapted to make it apply to the wholly novel situation brought about by s 2 of the Rent Regulation. How does the hypothetical yearly tenancy apply in relation to development sites? 88. It is clear that in Agrila this Court was invited to give guidance as to how the yearly tenancy postulated by the rating hypothesis was to be applied by valuers to development sites falling with s 2 of the Rating Regulation. On this appeal counsel referred the Court at some length to the parties’ lists of assumptions and the responses to them filed in Agrila, and to extracts from the transcript of the oral argument in that case on 8 and 9 February 2001. However, in Agrila, this Court (wisely, in my respectful opinion) did not go very far in giving detailed guidance. 89. It was wise to do so because the appeal in Agrila was confined to preliminary issues of law affecting a large number of development sites in different areas and with differing prospects of construction work being undertaken in the immediate future (CFA paras 13-15, 92I-93D). Although there were apparently some agreed facts in relation to some specimen sites, there had been no Tribunal hearing to determine contested facts. The question posed at point 4 (CFA para 117, 113I-114A) was expressed in general terms that Sir Anthony Mason NPJ considered to be unsatisfactory (CFA para 98, 109C-D). 90. So far as this Court did deal with the point the relevant part of the judgment is CFA paras 79 to 95 (105B-108H). But this passage is, as the heading before para 79 (105B) indicates, concerned mainly with the likelihood of the development being commenced and completed. In other words, it is concerned mainly with the intrinsic character of a development site, and much of the passage has already been quoted or summarized in the last section of this judgment (paras 78 to 80 above). Sir Anthony Mason NPJ does in this part of his judgment refer to the hypothetical tenancy as being of indefinite duration, and there are references to Humber and China Light & Power (CFA paras 90 and 91, 107G-J). But apart from London County Council v Erith there is mention of the leading cases in which the House of Lords has confronted the problem of tenements whose character is such that in the real world no one would ever take them on a yearly tenancy. Probably these cases were not cited to the Court. 91. The relevant passage in the judgment ends as follows, in continuation of the second quotation at para 79 above (CFA paras 94 and 95, 108F-109H): “But this does not mean that the sites should be valued as completed developments. Nor does it mean that either of the Commissioner’s methods of valuation or what has been described as ‘the contractor’s method’ of valuation should be adopted. The appropriate mode of valuation, apart from what is prescribed by relevant principles of law, is a matter for the Lands Tribunal to determine. It is not for this Court to express an opinion about valuation or about the appropriateness of any method of valuation. In these cases much will depend on the estimated duration of a yearly tenancy which the hypothetical tenant might secure. It might be sufficiently long to allow for completion of the relevant development, so that the hypothetical tenancy would extend eventually to such a situation.” I read this as a considered indication that further and more detailed guidance would have been premature. It would have to wait until there was an appeal from the Tribunal after it had heard evidence and legal argument, and reached its own decision based on its special skill and experience in valuation matters. The Tribunal’s decision in the present case, upheld by the Court of Appeal, gives this Court the opportunity to give further guidance. 92. I have already considered much of the Tribunal’s decision. I concentrate now on the significance of the yearly tenancy hypothesis, without repeating what has already been said about the “change of use” point. LT para 29 states: “ Obviously, there are tenements that no tenant would take up a tenancy from year to year in the real world. Lord Pearce gave the examples of sewage works, portion of railway lines. Development sites fall within the same category. In the real world, no developer with his right mind will take up a site for development based on a yearly tenancy. In this sort of situation, the rating hypothesis of yearly tenancy cannot be permitted to operate in such a way to exclude a tenancy for that particular mode or character of occupation. Otherwise, it would be putting the cart before the horse and the rating hypothesis would frustrate instead of facilitating a proper valuation for rating purposes.” That paragraph, and para 33 which develops the same line of thought, were severely criticized by Mr Howell. His argument was that the Tribunal seemed to be under some compulsion to create a demand for a yearly tenancy for which there would be no demand in the real world. 93. I would accept some of this criticism, but only to a limited extent. I think that the Tribunal, lacking any detailed guidance from Agrila, started with a valuable insight. The Tribunal recognized that development sites, like sewage works and portions of railway lines, are tenements that no developer and no statutory undertaker would contemplate taking on a yearly tenancy. In that type of case the yearly nature of the hypothetical tenancy compelled “rating experts and the Courts to have recourse to hypotheses of a more or less violent character,” as Lord Cave LC put it in Kingston Union v Metropolitan Water Board [1926] AC 331, 338. These methods were the profits (or “R&E”) method and the CB method, both of which are discussed in the speeches of Lord Cave and Lord Atkinson. They are methods under which the possibility of the hypothetical landlord serving notice to quit becomes an irrelevance. 94. Having set off in the right direction the Tribunal then, in my respectful opinion, took a wrong turning. Instead of following the public utility cases, where the yearly nature of the hypothetical tenancy becomes irrelevant, the Tribunal were tempted into elaborating the hypothesis, a course warned against in Hoare (Valuation Officer) v National Trust and the other cases mentioned in paras 62 and 63 above. One elaboration, the hypothetical tenant’s continued enjoyment after the change of use on completion of the development, has been discussed already. The other (LT paras 44-46) was an extension of the assumption that the hypothetical landlord and the hypothetical tenant “both are reasonable people, the landlord not being extortionate, the tenant not being under pressure”: Lord Denning MR in R v Paddington Valuation Officer ex parte Peachey Property Corp Ltd [1966] 1 QB 380, 412. 95. There is no doubt but that the initial bargaining as to the rent payable under the hypothetical tenancy must be supposed to take place between two parties acting reasonably. But it is an unwarranted extension of this (as Mr Howell rightly submitted) to suppose that the hypothetical landlord would refrain from acting in his own best interests if he could, by serving notice to quit soon after completion of the development, lawfully obtain a great financial advantage. The solution to the puzzle is not to elaborate the hypothesis by supposing that the hypothetical landlord would act contrary to his interest, but to recognize that it goes beyond the statutory purpose and scope of the rating hypothesis. 96. These are not however fatal defects in the Tribunal’s reasoning. The essential point was that it recognized (LT para 29) the feature that a development site has in common with some tenements owned by public utilities or statutory undertakers: that it is not immediately productive of income or profits, but is nevertheless of real value to the occupier. After straying a little from the right path by over-elaboration of the statutory hypothesis, the Tribunal came back to the right path with another valuable insight (LT para 72) as to the novelty of the situation created by s 2 of the Rent Regulation. 97. In LT para 78 it is stated to have been common ground that the building constructed on the site should be treated “for these purposes” (which were not precisely defined) as the property of the hypothetical tenant. Mr Howell told the Court that this was a misunderstanding, and that that concession was made only for the limited purpose of assessing the rateable value of the site by the R&E method (which was not the method advocated by the Commissioner, or adopted by the Tribunal). I readily accept that there was some misunderstanding here. The point is in any event irrelevant to the CB method of valuation (which the Commissioner did advocate from the outset, and which the Tribunal accepted). The CB method takes account of the initial expenditure on the acquisition of the site (subject to appropriate adjustments) but does not take into account the cost of constructing the new building. The adoption of the CB method, and its application in this case, are comprehensively covered in the part of the decision prepared by Mr Lo, and are not a separate issue in this appeal. 98. In the Court of Appeal Tang Ag CJHC, in his summary of Agrila, cited (CA paras 16 and 17) Lord Herschell’s observation in London County Council v Erith [1893] AC 562, 595, that the words of the rating hypothesis: “only provide the means of arriving at what is the annual value of the premises” and the observations of Scott LJ in Townley Mill Company (1919) Ltd v Oldham Assessment Committee [1936] 1 KB 585, 643, that “the hypothesis of the imaginary tenant is intended to assist and not hinder that process.” He also cited (CA para 20) observations by Sir Anthony Mason NPJ (CFA paras 90 and 92, 107G-I and 108A-C) as to the indefinite duration of the yearly tenancy and the importance of occupation being of value to the hypothetical tenant, even if he receives no immediate rental income. 99. Tang Ag CJHC then addressed (CA para 25) Mr Howell’s reliance on the yearly nature of the hypothetical tenancy, noting that Best Origin’s primary case before the Tribunal had been that the only possible tenant would be a building contractor. He went on to refer to passages from the judgments of Scott LJ in Robinson Brothers (Brewers) Ltd v Houghton and Chester-le-Street Assessment Committee [1937] 2 KB 445, 475 and Lord Coleridge CJ in Smith v Birmingham (Churchwardens) (1889) 22 QBD 703, 705-706. These passages serve to reinforce the fundamental points that I take from the Tribunal’s decision, that is (i) that as a result of s 2 of the Rent Regulation, development sites present a novel problem, since they are potentially of great value to their developer, but do not produce income during the construction period; and (ii) that the solution to the problem lies in the public utility cases, which are analogous. Tang Ag CJHC also referred with approval (CA para 36) to a passage in the Tribunal’s decision (LT para 54) already cited (para 76 above). 100. Tang Ag CJHC then referred (CA para 39) to the prospect of profitable development (which was, of course, to be judged on evidence available in 1997). He observed that there was no evidence before the Tribunal that the development might be unprofitable. It is of course necessary to avoid hindsight. But the price paid by Best Origin in 1996 and the agreed facts about the property market at that time (para 84 above) indicate that there was every reason to expect that it would be profitable. CA paras 41 to 46 address the argument that the hypothetical landlord would be either unable or unwilling to obtain the completed building for himself by giving notice terminating the tenancy. I have already explained (para 97 above) why this point, and the further discussion (CA paras 50-58) about ownership of the completed building under the rating hypothesis, are in my view irrelevant. Conclusion 101. For these reasons I conclude that the Tribunal and the Court of Appeal were correct in recognizing that s 2 of the Rent Regulation produced a novel situation which called for some development in this area of the law. It is not however a novel or unprincipled development. It extends to development sites, during the period of active construction while they are not producing income, the treatment which has since the 19th century been accorded, regularly and uncontroversially, to the unproductive parts of undertakings owned by municipalities or statutory bodies, such as sewage treatment works, waterworks, gasworks, and so on (there is a comprehensive list of 28 types of tenement in Halsbury’s Laws 5th ed Vol 70 para 166). This extension is of course only for the purposes of assessing Government rent; for rating purposes a development site is still regarded as not being a rateable tenement. 102. Earlier in this judgment (para 52 above) I thought it unnecessary to multiply citation of authority. But in conclusion it may be useful, in order to make good the point that this is not a novel or unprincipled development, to set out some of the principal authorities which establish (i) that the yearly character of the hypothetical tenancy has been interpreted as indicating an indefinite duration and equivalent to “taking one year with another”; (ii) that in the absence of clarifying legislation, the courts have had no option but to adapt the rating hypothesis to make it workable in relation to some tenements of a special character; and (iii) that the CB method has often been approved by the court as appropriate for tenements which, although of value to their occupiers, were not income-producing (though the detailed application of that method is a matter for valuation experts). 103. Most of these authorities are decisions of the House of Lords. Several of them were presided over by the Lord Chancellor, indicating the importance that has always been accorded to this branch of the law. Most of them have been mentioned already, but it may be useful to bring them together, in chronological order: R v Coventry Canal Co (1859) 28 LJ (MC) 102, 104 (Lord Campbell CJ); Great Eastern Railway Co v Overseers of Haughley [1866] LR 1 QB 666, 679 (Cockburn CJ); Smith v Churchwardens and Overseers of Birmingham (1889) 22 QBD 703, 705-706 (Lord Coleridge CJ), 707 (Lindley LJ); London County Council v Erith [1893] AC 562, 585, 588-593, 596 (Lord Herschell LC); Kingston Union v Metropolitan Water Board [1926] AC 331, 338-339, 344-345 (Viscount Cave LC), 346-350 (Lord Atkinson); Railway Assessment Authority v Southern Railway Company [1936] AC 266, 273-274 (Viscount Hailsham LC); Robinson Brothers (Brewers) Ltd v Houghton and Chester-le-Street Assessment Committee [1937] 2 KB 445, 468-471, 474-475 (Scott LJ, affirmed by the House of Lords [1938] AC 321); and Dawkins (Valuation Officer) v Ash Brothers and Heaton Ltd [1969] 2 AC 366, 380 (Lord Guest), 382-384 (Lord Pearce), 386-388 (Lord Wilberforce) and 392-393 (Lord Pearson). 104. It is unnecessary to consider issue estoppel, or Mr Holgate’s fallback position on hypothetical ownership of the completed building. The striking-out application is no longer a live issue. 105. I would therefore dismiss Best Origin’s appeal and make an order nisi that the Commissioner’s costs should be taxed (if not agreed) and paid by Best Origin. In the event that either party contends for a different order, that party must lodge written submissions with the Registrar of this Court within 21 days of the handing down of this judgment. Any written submissions in reply should be lodged within 21 days thereafter. Failing any submissions to the contrary, the costs order nisi will become absolute. Chief Justice Ma : 106. The Court unanimously dismisses the appeal and makes an order for costs in terms of para 105. Mr John Howell QC and Mr Nigel Kat, instructed by Woo, Kwan, Lee & Lo, for the appellant Mr David Holgate QC and Mr John Litton, instructed by the Department of Justice, for the respondent [1] In the official published report the paragraphs of the judgment are not numbered. But most on-line versions have numbered paragraphs and paragraph numbers were used, for convenience, at the hearing. These paragraph numbers will be used in this judgment. The relevant page numbers in the official report are inserted in brackets. 
Mr Justice Chan, Acting CJ: 1. The Since Shum Kwok Sher v HKSAR (2002) 5 HKCFAR 381 and Sin Kam Wah & another v HKSAR (2005) 8 HKCFAR 192, officers in the public service who were accused of improper or criminal activities in the discharge of their public duty or the exercise of their discretion have sometimes been charged with the offence of misconduct in public office. In some cases, this offence was charged in addition to other substantive criminal offences which the public officers face. In this appeal, the appropriateness of such a charge in the circumstances of this case is questioned by the appellant. It is contended that his conduct was not so serious as to constitute this common law offence. The appellant’s conduct 2. The appellant is a doctor employed by the Hospital Authority. At the time of the alleged offences, he was working at the out-patient clinic in the Central Kowloon Health Centre (“the Clinic”). 3. The normal procedure for a patient to seek the services of the Clinic is as follows. He makes an appointment with the Clinic; he attends the Clinic personally to register his attendance; he pays $45 (which includes consultation fee and any prescribed medicine) to get a ticket and waits for his turn; he is called to see one of the doctors; he consults the doctor who diagnoses him and puts a consultation summary of his conditions into the computer; the doctor prescribes the medicine using the computer; the patient takes the prescription to the pharmacy; he gets a ticket and waits to take his medicine. Each doctor at the Clinic has a daily quota of 70 patients but he has the discretion whether to see any additional walk-in patient. 4. The appellant’s family members were entitled to public medical services when they were in Hong Kong and had previously made use of the services at the Clinic. 5. Between 22 June 2007 and 28 November 2008, on 16 occasions, the appellant booked appointments in the names of his parents and 6 year old son for consultation in the Clinic. On each of these occasions, he bought tickets for consultation, wrote consultation summaries and prescribed medicines on the computer and later obtained the medicine. The total cost of all the medicines to the Government was $247. At the time of these consultations, his parents were in Australia and his son was attending kindergarten. They did not attend the Clinic at the relevant times. 6. There is no express rule set by the Hospital Authority which requires that a doctor must diagnose his patients inside the Clinic, although he is generally expected to do so. The doctors and staff of the Clinic said that upon seeing the relevant consultation summaries in the present case, they would assume that the consultations in question were conducted inside the Clinic. The pharmacist also said that if he had known that the diagnoses were not conducted in the Clinic, he would not dispense the prescribed medicine. 7. The appellant explained that the diagnoses of his parents were conducted over long distance phone calls and the diagnoses of his son were done at home. He thought that in order to obtain medicines for his family members, he had to follow the procedure at the Clinic. Some of the medicines obtained were either mailed to his parents in Australia or delivered to them by his relatives. Other medicines were placed at home for future use. He conceded that he could not personally verify his parents’ symptoms without seeing them face to face or his son’s condition without the necessary equipment. The charges 8. The appellant was charged with two counts of misconduct in public office (charges 1 and 5), three counts of fraud, contrary to s.16A of the Theft Ordinance, Cap 210 (charges 2, 3 and 4) which were alternatives to charge 1, and 22 counts of gaining access to computers with criminal or dishonest intent, contrary to s.161(1) (b) of the Crimes Ordinance, Cap 200 (charges 6 to 27) which were alternatives to charge 5. 9. He was convicted by the Deputy Magistrate of charge 1 and acquitted of charge 5. The other charges were left on file and no verdict was given in respect of these charges. His conviction on charge 1 was upheld on appeal by Deputy High Court Judge A Wong. 10. With the leave of the Appeal Committee, the appellant now appeals on the substantial and grave injustice ground. Issues in this appeal 11. The common law offence of misconduct in public office was affirmed in Shum Kwon Sher and the ingredients of this offence were reformulated in Sin Kam Wah: the offence is committed where a public officer in the course of or in relation to his public office willfully and intentionally misconducts himself by act or omission, without reasonable excuse or justification and the misconduct is serious, not trivial, having regard to the responsibilities of the office and the officeholder, the importance of the public objects which they serve and the nature and extent of the departure from those responsibilities (para. 45). 12. The crucial question to be decided in this case was whether what he did amounted to misconduct which was so serious as to constitute the offence of misconduct in public office. The Magistrate and the Judge took the view that it was. 13. This is challenged by Mr Peter Duncan SC, leading M Derek Chan and Ms Fiona Nam for the appellant. He relies on two grounds in submitting that the conviction should be quashed: (1) the appellant was not guilty of the misconduct as alleged in that it was wrong to find as the Magistrate did that the appellant had falsified the records of the Hospital Authority in order to show that his parents and son had attended the Clinic when in fact they had not; and (2) the appellant’s conduct when considered in the proper context was not sufficiently serious as to amount to the offence of misconduct in public office. Was there any falsification? 14. In relation to the first ground, it is argued that the prosecution had failed to prove that the appellant had falsified any record and that the Magistrate was wrong to focus on certain records or acts and interpret them out of context when in fact none of the records said there was physical attendance by the appellant’s parents or his son; in short, the appellant did not make any false statement in the consultation summaries or any of the records. 15. The particulars in charge 1 stated relevantly that the appellant had wilfully and intentionally misconducted himself by: (a) falsifying records including consultation summaries and out-patient prescriptions in the computer system “purporting to show that his family members … had respectively on various occasions attended the [Clinic] when his family members had in fact not attended”; and (b) causing or intending to cause the Hospital Authority to issue prescribed medicines to his family members. The charge does not however specify how he had falsified the records. 16. It is not disputed that the appellant had followed the normal procedure on the occasions in question, except that his family members did not personally attend the Clinic at the relevant times. There is no evidence to suggest that the diagnoses of his family members as described by the appellant did not happen, that is, he had not diagnosed them on those occasions either over the phone or at home. There is also no evidence to show that he had made up the entries in the consultation summaries or made any false entry in any of the records. The Magistrate certainly made no finding to this effect. She only found that the appellant’s diagnoses as stated in the consultation summaries might not be accurate since he could not verify them face to face in relation to his parents’ conditions and did not have the necessary equipment at home to do so in relation to his son’s condition. 17. The Magistrate considered (and the Judge agreed) that the appellant had falsified the consultation summaries and computer records as a whole in that he had omitted to state in the records that his family members were not at the Clinic, knowing that such records represented to others that his family members had attended the Clinic. Thus the real complaint against the appellant was that he had failed to make it clear in the consultation summaries and prescription forms that his family members did not personally attend the Clinic on those occasions, hence giving rise to the impression that they had. 18. With respect, I find it difficult to accept the Magistrate’s conclusion. First, it must be noted that this offence is not a statutory offence and there is no definition of what amounts to “falsification”. This word should not be construed as if it bears any statutory meaning. The common understanding of this word (as the two Chinese characters adopted by the Magistrate also suggest) is that it involves creating a false document or record, or changing any entry or adding any false entry in a document or record. I cannot see how the appellant can be said to have done any of these in the present case. Nor was he alleged to have done so. 19. Secondly, there is no evidence of any requirement or obligation on the part of a doctor or staff member to state in the consultation summary or any part of the computer record whether a patient is present in the Clinic or not. This seems to be consistent with the absence of any rule requiring that a doctor must diagnose his patient inside the Clinic. While some may consider it prudent or proper to make this clear in the records, the failure to do so cannot be regarded as an act of falsification. 20. Thirdly, there is no evidence that the consultation summary or any other record is intended to record the personal attendance of a patient. That is not the function of a consultation summary or prescription form. None of the consultation summaries required the author to say where the diagnoses were conducted. Looking at these documents, it is also not clear how and where this information was to be put in the documents if it were so intended. 21. Fourthly, it is argued by the prosecution (and this appears to have been accepted by the Magistrate and the Judge) that if the appellant had not intended to represent to others that his family members had attended, he would not have gone through the normal procedure. However, there is no clear evidence as to any other way in which a doctor can get medicine for his family members who are entitled to public medical services and whom he has diagnosed outside the Clinic without going through the normal procedure. In the present case, there is nothing to challenge the appellant’s evidence that he went through the normal procedure with the intention of obtaining medicines for his family members whom he had diagnosed without asking them to attend the Clinic personally. Some might consider this unnecessary bearing in mind the close relationship between them. 22. The appellant could of course have stated in the records that his family members had not attended the Clinic on the occasions in question. Admittedly, his failure to do so did have the effect of misleading the reader of the relevant records that his family members had actually attended for consultation on those occasions. But such impression was the result of his following the normal procedure and not because of anything he had done to or written on the records. In my view, it would be difficult in the circumstances of this case to say that by failing to do so, he had created any false document or had made any change to any entry or added any false entry in any of the records. I am therefore inclined to think that there was no falsification of any record by the appellant. 23. Nor can it be said that the omission to state the absence of the patient in the consultation summary has caused the Hospital Authority to prescribe medicine to the patient. Notwithstanding what the pharmacist said, there is no evidence that the Hospital Authority would not have prescribed medicine to a patient who has not personally attended the Clinic. If there is no express requirement that an out-patient must be diagnosed inside the Clinic, I doubt if prescription could be declined. After all, the prescription was indeed given by a doctor of the Clinic and the appellant had paid the requisite fee which included any prescribed medicine. 24. Although this is not a statutory offence in which the prosecution have to set out and prove all the ingredients of the offence as provided in the statute, nevertheless, it is important to formulate the particulars of the charge correctly so that the matters alleged against the appellant as constituting this common law offence are clearly stated. This is required to enable the accused to know the case he has to meet at the trial. For the reasons I have discussed above, I do not think the prosecution have succeeded in proving their pleaded case against the appellant. Was the appellant’s conduct serious misconduct? 25. I shall now turn to the second ground. In Shum Kwok Sher, when Sir Anthony Mason NPJ was considering the offence of misconduct in public office, he pointed out that the difficulty in this offence stems “from the range of misconduct by officials which may fall within the reach of the offence” and he remarked that “what constitutes misconduct in a particular case will depend upon the nature of the relevant power or duty of the officer or of the office which is held and the nature of the conduct said to constitute the commission of the offence.” (para.69) 26. In considering this important question, one must not lose sight of the object of this offence. It is clear from a review of the authorities that this offence is aimed at punishing an abuse by a public officer of the power and duty entrusted to him for the public benefit or of his official position: R v Bembridge, 3 Doug KB 327, Lord Mansfield CJ; Professor P D Finn in his article “Public Officers: Some Personal Liabilities” (1977) 51 ALJ 313, 315 and in “Official Misconduct” (1978) 2 Crim LJ 307 at 308; R v Dytham [1979] QB 722, 727G to 728A, Lord Widgery CJ; Shum Kwok Sher, Sir Anthony Mason NPJ in para.76; and HKSAR v Wong Lin Kay (2012) 15 HKCFAR 185, Ribeiro PJ in para. 17 and Lord Millett NPJ in paras. 44 and 46. 27. In Shum Kwok Sher, when he was formulating the ingredients of the offence of misconduct in public office, Sir Anthony Mason NPJ considered that to constitute the offence, the conduct in question must be serious misconduct, not trivial and this is to be determined having regard to: (i) the responsibilities of the office and the office holder, (ii) the importance of the public objects which they serve, and (iii) the nature and extent of the departure from those responsibilities (para.86). He regarded that this is consistent with the concept of abuse of office. But he acknowledged that there would be borderline cases between this common law offence and disciplinary offences (para.87). 28. These factors were adopted by the English Court of Appeal in Attorney General’s Reference (No.3 of 2003) [2005] QB 73. However, Pill LJ added in para.46 the following remarks: “ … Having considered the authorities, we agree that the misconduct complained of must be serious misconduct. Whether it is of a sufficiently serious nature will depend on the factors stated by Sir Anthony Mason NPJ along with the seriousness of the consequences which may follow from an act or omission. An act or omission which may have as its consequence a death, viewed in terms of the need for maintenance of public standards to be marked and the public interest to be asserted, is likely to be more serious than one which would cause a trivial injury. This factor is likely to have less significance where, as in Shum Kwok Sher, the allegation is of corruption where the judgment upon the conduct may not vary directly in proportion to the amount of money involved.” (emphasis added) 29. In cases where corruption, dishonesty or other illegal practices are involved, it is not necessary to specifically consider the consequences of the misconduct in deciding whether it is serious enough as to constitute the offence of misconduct in public office. The misconduct speaks for itself: the seriousness of the consequences of such corrupt, dishonest or illegal practices will be obvious. 30. In other cases, where corruption, dishonesty or other illegal practices are not involved, the consequences of the misconduct may not be obvious. Nevertheless, this must be a factor which is also relevant when considering whether the misconduct is serious enough as to merit criminal sanction. I do not think the prosecution is disputing the relevance of this factor. Nor can this be disputed. This factor was not mentioned in the discussion of this offence in Shum Kwok Sher. That was a case of granting preferential treatment to a close relative in relation to some government contracts and the seriousness of such misconduct and its consequences could be readily seen. I also do not think the list of factors mentioned in that case was intended to be exhaustive. Further, in most cases, the consequences of the misconduct will usually have been considered when one is examining the nature and extent of the departure from those responsibilities. 31. The approach to be adopted in considering whether any misconduct was serious enough as to call for condemnation and punishment was further discussed in Chan Tak Ming v HKSAR (2010) 13 HKCFAR 745. Bokhary PJ said at para.27: “ … trivial misconduct will of course not support a charge of misconduct in public office. That said, the question is whether the offence is serious having regard to – as stated in that item (5th item of the reformulation) – the responsibilities of the office and the office holder, the importance of the public objects which they serve and the extent of the departure from those responsibilities. It is in that way – and not by saying that it must be sufficiently serious since it is not trivial – that one goes about deciding whether the necessary seriousness exists.” 32. One must consider all the circumstances of the case including the factors mentioned above. It would be wrong simply to conclude that if the misconduct in question is not trivial, then it must be serious enough to merit criminal sanction. It is a high threshold for the prosecution. As Pill LJ said in AG’s Reference (No.3 of 2003) in para.56: “It supports the view expressed in the criminal cases, from R v Borrow 3 B & Ald 432 to Shum Kwok Sher v HKSAR 5 HKCFAR 381 that there must be a serious departure from proper standards before the criminal offence is committed; and a departure not merely negligent but amounting to an affront to the standing of the public office held. The threshold is a high one requiring conduct so far below acceptable standards as to amount to an abuse of the public’s trust in the office holder. A mistake, even a serious one, will not suffice. The motive with which a public officer acts may be relevant to the decision whether the public’s trust is abused by the conduct.” 33. In the present case, the Magistrate held that the appellant’s conduct was serious enough as to amount to misconduct in public office. Her reasoning was as follows: “Making a booking for a family member per se is not improper. If the Appellant had made a booking for his family members and they attended the clinic in the manner that is required of other members of the public, there may not be any impropriety. However, if the Appellant uses his position to make a booking for a family member who does not attend the clinic and to obtain medicine which is only for use in “just in case” scenarios, the Appellant has shown favouritism towards his family members as it is clear that other members of the public are not entitled to such special services from the Appellant. By making a booking for his family members and registering their attendance, the appellant has denied the right of one member of the public to use the HA services on that particular day at that particular time. For the Appellant to claim that he thought he was acting properly, in such circumstances is, unbelievable and unreasonable.” (para.74) 34. It can readily be seen from the above passage that the Magistrate did not consider the making of appointments for family members in itself was anything improper; nor was it improper if the patients turned up at the Clinic. It was, according to her, the making of a booking for someone who would not attend and the obtaining of medicine which would not be immediately needed that was improper. This, she said, was because this would be showing favouritism to that person and would deprive one member of the public of the use of public medical services. 35. Such reasoning ignores the fact that there is no requirement that a patient must be diagnosed inside the Clinic. There is evidence from some of the doctor witnesses that while the normal practice is to diagnose a patient face to face, there can be exceptional circumstances when this is not done and that although this is not the best or most desirable practice, it is acceptable in some cases, for example, in the updating of a consultation summary or the changing of medication over the phone. 36. Secondly, as submitted by counsel for the appellant, it has not been proved that the public had been deprived of medical services in that the public was prevented from making a booking or that the public had been deprived of access to the medicines prescribed to the appellant’s family members. The Magistrate had also overlooked the fact that the appellant’s family members were entitled to public medical services. There was nothing to suggest that the diagnoses on the occasions in question were not genuine. The appellant had gone through the normal procedure and paid the requisite fee. It cannot be said that public resources had been wasted. The only favour which the appellant had shown to his family members was to dispense with their personal attendance at the Clinic. 37. The Magistrate’s conclusion that the appellant’s misconduct was serious was upheld by the Judge. He was satisfied that the Magistrate had considered all the relevant factors in coming to this conclusion. However, the Judge apparently went further than the Magistrate. He said: “Under normal circumstances, it may be beyond reproach that a doctor prescribes medicine for his family members after proper diagnosis even the consultation is not conducted in a clinic as judging from the present case’s evidence. However, the present case cannot be judged from this angle only, because the Appellant had obviously made a series of actions that had affected other public officers’ judgment, damaged the integrity of the system, wasted the public resources and shaken the confidence of the public in the system.” (para. 107) 38. It can be noted that the Judge, unlike the Magistrate, did not consider it was reproachable conduct for the appellant to prescribe medicine to his family members after proper diagnosis even though this was not done inside the Clinic. His main objection was the possible damage to the integrity of the system and the effect on public confidence in the system. 39. The integrity of the system and the public confidence in such a system are clearly relevant matters for consideration in deciding whether the conduct of a public officer is so serious as to amount to the offence of misconduct in public office. But these matters must be considered in context and together with other relevant factors. 40. The object of the system operated at the Clinic is to provide out-patient medical services to members of the public. The appellant was one of the doctors entrusted with the responsibility of running the Clinic fairly and efficiently. His family members were entitled to the medical services provided at the Clinic and had made use of such services and personally attended the Clinic before. On the occasions in question, he had followed the normal procedure for his family members in order to make use of the services at the Clinic. When what he did is analyzed in the proper context, the only departure from his responsibilities was that he had dispensed with the attendance of his family members at the Clinic. In this way, it can be correctly said that he has shown preferential treatment to his family members. The consequences of what he did might have the possible effect of blocking one or two patients from making an appointment during consultation hours and this might also give rise to the perception that the system had not been operated fairly and impartially in that family members of the staff are given preferential treatment. 41. But apart from that, the appellant had breached no rule set by the Hospital Authority: there is no express requirement that a doctor can only diagnose a patient inside the Clinic or that a doctor must state clearly in the records that the patient is or is not present at the Clinic. Another significant consideration is that non-attendance diagnosis is not against the Code of Professional Conduct. One can always say that the appellant could have made it clear in the records kept in the computer that his family members did not personally attend on the occasions in question. Alternatively, he could have made use of his discretionary power to deal with his family members’ cases after he had finished the usual quota for the day and that would not affect other patients’ interests. He may also be criticized for taking advantage of the lack of more effective control in the system. 42. However, I do not think the appellant’s conduct was, in the words of Pill LJ in AG’s Reference (No.3 of 2003), a departure which was “so far below acceptable standards as to amount to an abuse of the public’s trust in the office holder.” In my view, the appellant’s shortcomings cannot be regarded as an abuse of power which is the essence of the offence. In the circumstances of this case, I do not believe that the integrity of the system of out-patient service at the Clinic would be seriously tarnished by what the appellant had done. The prosecution has failed to cross the high threshold of proving that his misconduct was so serious as to call for criminal sanction. Conclusion 43. For these reasons, I do not think this is a proper case to charge the appellant with the offence of misconduct in public office. I would allow the appeal and quash his conviction. Mr Justice Ribeiro PJ: 44. I agree with the judgment of Mr Justice Chan, Acting CJ. Mr Justice Tang PJ: 45. For the reasons given by Mr Justice Chan, Acting CJ, I would also allow this appeal and quash the appellant’s conviction for misconduct in public office. Mr Justice Mortimer NPJ: 46. For the reasons given by Mr Justice Chan, Acting CJ, I also would allow this appeal and quash the appellant’s conviction for misconduct in public office. Sir Thomas Gault NPJ: 47. I agree with the judgment of Mr Justice Chan, Acting CJ. I too would allow the appeal and quash the conviction. Mr Justice Chan, Acting CJ: 48. The Court unanimously allows the appeal and quashes the conviction on misconduct in public office. Mr Peter Duncan SC, Mr Derek Chan and Ms Fiona Nam, instructed by Cheung & Yip, for the Appellant Mr Gerard McCoy SC, instructed by the Department of Justice and Mr Michael Wong, SPP of that Department, for the Respondent 
Mr Justice Fok PJ: A. Introduction 1. This appeal concerns the mental element of the offence under section 37(a) of the Crimes Ordinance[1] (set out below), which makes it an offence to make a false declaration to obtain registration for carrying on a vocation. The section requires that the defendant “knows” the declaration “to be false or fraudulent” and the appeal raises the questions: (a) what, as a matter of law, must a defendant know in order to be guilty of the offence; and (b) did the appellant in this case have the necessary guilty mind? As will be seen, the appeal arises in the context of an application for enrolment as a nurse and the false declaration concerns the existence of previous criminal convictions. 2. At the conclusion of the hearing of the appeal, the Court made the orders set out in Section G below and indicated that its reasons for doing so would be handed down in due course. These are the reasons of the Court. B. The background facts 3. The salient background facts taken from the admitted facts and the exhibits below are as follows: (1) On 10 January 2007, the appellant was convicted of two charges of obtaining property by deception, for which he was sentenced to seven weeks’ imprisonment on each (to be served concurrently). (2) Between September 2011 and August 2013, the appellant took a course for pupil enrolled nurses at the Open University of Hong Kong. After completing the course, the appellant filled out an application form[2] addressed to the Nursing Council of Hong Kong (“the Nursing Council”) seeking enrolment as an enrolled general nurse. (3) The application form was Exhibit P1 at trial and was accompanied by a Declaration Form addressed to the Secretary of the Nursing Council and signed by the appellant (“the Declaration Form”).[3] The Declaration Form included a number of declarations, including one stating “I declare that … (a) I have/have not been convicted of any offence punishable with imprisonment in Hong Kong or elsewhere.” An asterisk after the words “have/have not” invited the appellant to delete whichever was inappropriate and, on his form, the word “have” was crossed out. (4) On 11 November 2013, the Open University of Hong Kong submitted Exhibit P1, together with the Declaration Form, and other required documents to the Nursing Council for his application to be processed. (5) The Nursing Council duly processed the appellant’s application and, based on his application form and accompanying declarations, issued him with the enrolment certificate (Exhibit P8) and practising certificate (Exhibit P9) of an enrolled nurse which the appellant collected from the Open University of Hong Kong. (6) Thereafter, from 6 February 2014 to 1 December 2014, the appellant worked as an enrolled nurse at Chung Shak Hei (Cheung Chau) Home for the Aged. In applying for that employment, the appellant had submitted to the Home a copy of his practising certificate. 4. The appellant was charged with the offence of making a false declaration to obtain registration for the carrying on of a vocation, contrary to section 37(a) of the Crimes Ordinance (set out below). The particulars of the offence alleged that, on or about 16 October 2013, the appellant attempted to procure himself to be registered on the roll of enrolled nurses kept under the Nurses Registration Ordinance of persons qualified by law to practise nursing by wilfully making in writing a declaration which he knew to be false or fraudulent, namely a declaration that he had not been convicted of any offence punishable with imprisonment in Hong Kong or elsewhere. C. The falsity of the declaration 5. It is not in dispute that the relevant inquiry in the Declaration Form required him to declare whether he had any disclosable convictions of any offence punishable with imprisonment in Hong Kong or elsewhere. This is clearly correct as a matter of construction of the Declaration Form in the context of the laws of Hong Kong, which include the Rehabilitation of Offenders Ordinance.[4] The purpose of that ordinance, as reflected in its Long Title is “[t]o rehabilitate offenders who have not been reconvicted for 3 years, to prevent unauthorized disclosure of their previous convictions and for connected purposes.” 6. Thus, section 2 of the ROO excuses the disclosure of certain convictions by providing: “(1) Where – (a) an individual has been convicted in Hong Kong (before or after the commencement of this Ordinance) of an offence in respect of which he was not sentenced to imprisonment exceeding 3 months or to a fine exceeding $10,000; (b) he has not been convicted in Hong Kong on any earlier day of an offence; and (c) a period of 3 years has elapsed without that individual being again convicted in Hong Kong of an offence, then – (i) subject to section 3(3) and (4), no evidence shall be admissible in any proceedings which tends to show that that individual was so convicted in Hong Kong; (ii) any question asked of that individual or any other person relating to, or any obligation imposed on that individual or any other person to disclose, that individual’s previous convictions, offences, conduct or circumstances shall be treated as not referring to that conviction; and (iii) that conviction, or any failure to disclose it shall not be a lawful or proper ground for dismissing or excluding that individual from any office, profession, occupation or employment or for prejudicing him in any way in that office, profession, occupation or employment.” Relevantly to the present case, by section 2(1)(ii) of the ROO, a fiction is created in that where a person is asked about his previous convictions, a question about (and any obligation on him to disclose) any previous conviction is treated as not referring to a conviction falling within paragraphs (a), (b) and (c) of section 2(1). 7. However, there are exceptions to the confidentiality afforded by that section, which include those set out in section 4 of the ROO. For present purposes, it is sufficient to refer to section 4(1)(c), which provides: “(1) Section 2(1) and (1A) shall not apply to – … (c) proceedings relating to a person’s suitability to be granted, or to continue to hold, any licence, permit or dispensation, or to be registered, or continue to be registered, under any law; …”. 8. It is not in dispute in this appeal that section 4(1)(c) of the ROO applied in respect of the appellant’s application for enrolment as an enrolled nurse under the Nurses Registration Ordinance.[5] As such, he could not rely on section 2 of the ROO to excuse him from disclosing, in completing the Declaration Form, that he had indeed been convicted of an offence punishable with imprisonment in Hong Kong. By signing the Declaration Form stating that he had not been convicted of an imprisonable offence, the appellant made a false declaration. Whether he did so knowing the declaration was false lies at the heart of this appeal. D. The procedural history D.1 The magistracy trial 9. At trial in the magistracy,[6] the appellant pleaded not guilty and testified that: (1) He had been told, by his lawyers at the time of his convictions in 2007 and by staff of the Correctional Services Department when he served his term of imprisonment, that the ROO might assist him. (2) Before he filled in the Declaration Form, he read the relevant provisions of the ROO and sought the views of the Nursing Council, who suggested that he seek legal advice. (3) In October 2010, through his ex-wife, Ms Fan Hoi Wan, he had sought legal advice from the Free Legal Advice Scheme provided by the Duty Lawyer Service. Ms Fan approached a Scheme lawyer and asked the following question (as recorded on the Free Legal Advice Scheme application form): “I would like to enquire about the Rehabilitation of Offenders Ordinance. (That is), there is a person who had not had any previous criminal conviction, but was sentenced to imprisonment of 5 weeks (i.e. less than 3 months) at the Magistrates’ Courts for (the offence of fraud). Since then, (he) has not got involved in any (other) criminal offences. Having studied hard for 5 years, he managed to complete the (Nursing) and (Social Work) courses soon. In that case, when filling out the application form for registration, can he apply the (Rehabilitation of Offenders Ordinance) to state on the application form for registration that ‘(he) has never been convicted of any offences punishable by imprisonment in Hong Kong’? Assuming that he writes on the application form for registration that ‘(he) have [sic] never been convicted of any offences punishable by imprisonment in Hong Kong’, does he breach the law by doing so?” (4) The advice given to Ms Fan by the Scheme lawyer, as subsequently summarised in writing (adduced at trial as Exhibit D1), was that: “Since the application form for the social worker position required Ms Fan to declare if she has been convicted previously and it required the applicant to disclose such record even though he/she is protected under the Ordinance. I advised Ms Fan to disclose her record if such form had such declaration but if other forms do not have such provision, she would declare she has no previous conviction.” (5) He genuinely believed that, by reason of the provisions of the ROO, as confirmed by the legal advice given to Ms Fan, he was not required to disclose his previous convictions in the Declaration Form. 10. It will be necessary to address below some of the provisions of the Social Workers Registration Ordinance[7] (together with the application form to which the Scheme lawyer referred in his advice to Ms Fan) to put the legal advice provided to Ms Fan into context. It is not in dispute that the advice was incorrect as a matter of law. 11. The Deputy Magistrate held that section 4(1)(c) of the ROO was applicable in the present case so that the appellant was unable to rely on section 2 of the ROO. He also concluded that the legal advice given to Ms Fan contained in Exhibit D1 did not convey what the appellant said was the advice given to her and did not support the appellant’s explanation as to why he did not disclose his previous convictions. As will be seen, there is an issue between the parties as to whether the Deputy Magistrate (and also the Deputy Judge on appeal) fundamentally misconstrued the evidence as to what advice was given to Ms Fan. Be that as it may, as the Deputy Magistrate understood the advice, the appellant was told he had to disclose his previous convictions. 12. For this reason, the Deputy Magistrate disbelieved the appellant’s evidence that he had not disclosed those convictions because he believed the legal advice he had sought. Instead, he found that the appellant wilfully made a false declaration, knowing that he had previous convictions but wilfully indicating that he had never been convicted of any offences punishable with imprisonment in Hong Kong. Accordingly, he convicted the appellant of the offence charged and sentenced him to a fine of HK$3,000 and ordered him to perform 160 hours of community service. D.2 The intermediate appeal to the Court of First Instance 13. The appellant appealed against his conviction to the Court of First Instance.[8] The sole ground of appeal advanced was that the Deputy Magistrate was wrong to hold that section 4(1)(c) of the ROO (set out above) applied, so that the conviction was unsafe and unsatisfactory. The Deputy Judge concluded that section 4(1)(c) of the ROO was applicable in the present case, but for different reasons to those given by the Deputy Magistrate, so that the appellant was unable to rely on section 2 of the ROO to avoid disclosure of his previous convictions. 14. Having disposed of that only ground of appeal, the Deputy Judge dealt very briefly with the evidence at trial concerning the legal advice obtained by the appellant. In his judgment dismissing the appeal,[9] he concluded: “I considered the evidence in the present case by way of rehearing. I find that the prosecution by the evidence adduced in the trial has successfully proved beyond all reasonable doubts all the constituent elements of the offence. The magistrate is right in finding that the appellant shall not rely on the legal advice he obtained to believe that he is not required to disclose his past conviction record.”[10] D.3 Leave to appeal to the Court of Final Appeal 15. The Deputy Judge dismissed the appellant’s application to certify points of law for appeal to the Court of Final Appeal.[11] These related to the applicability of section 4(1)(c) of the ROO, a point which, as already noted, is no longer in dispute. 16. After the appellant filed an application to the Appeal Committee for leave to appeal, the Registrar of the Court of Final Appeal issued a summons under Rule 7(1) of the Court of Final Appeal Rules in respect of the application.[12] Following the filing of written submissions on behalf of the appellant pursuant to Rule 7(1), the Appeal Committee listed the application for an oral hearing and granted the appellant leave to appeal in respect of the following question of law: “In the light of the rule that ignorance of the law is generally no defence, what, on its true construction, is the mental element of the offence under section 37(a) of the Crimes Ordinance (Cap.200)?”[13] 17. Leave to appeal was also granted on the basis that it was reasonably arguable that, by reason of the courts below having materially misapprehended the evidence, substantial and grave injustice was done to the appellant.[14] E. Section 37(a) of the Crimes Ordinance 18. Section 37 of the Crimes Ordinance provides as follows: “37. False declarations, etc., to obtain registration, etc., for carrying on a vocation Any person who – (a) procures or attempts to procure himself to be registered on any register or roll kept under or in pursuance of any enactment for the time being in force of persons qualified by law to practise any vocation or calling; or (b) procures or attempts to procure a certificate of the resignation of any person on any such register or roll as aforesaid, by wilfully making or producing or causing to be made or produced either verbally or in writing any declaration, certificate or representation which he knows to be false or fraudulent, shall be guilty of an offence and shall be liable on conviction upon indictment to imprisonment for 12 months and to a fine.” 19. As a matter of legislative history, section 37 of the Crimes Ordinance was originally enacted as section 8 of the Perjury Ordinance.[15] The Perjury Ordinance was based on the English Perjury Act 1911[16] and section 8 of the Perjury Ordinance was relevantly the same as section 6 of the Perjury Act 1911. E.1 The elements of the offence 20. The actus reus of the offence is constituted by a defendant: (1) making or producing (or causing to be made or produced) a false declaration, certificate or representation; and (2) doing so for the purposes of procuring (or attempting to procure) his registration on a register or roll, kept under or in pursuance of any enactment, of persons qualified by law to practise any vocation or calling. It is not in dispute that these elements were satisfied in this case. 21. The mens rea of the offence is constituted by the defendant making the declaration (etc.): (3) wilfully; and (4) knowing it to be false or fraudulent. So far as wilfulness is concerned, it is not in dispute in this appeal that this element is satisfied on proof by the prosecution that the defendant made the declaration “deliberately and not inadvertently or by mistake”: see R v Millward (Neil) [1985] 1 QB 519 per Lord Lane CJ at 524G-525A. It is common ground the element of wilfulness was satisfied in this case. The critical question here is whether the declaration made by the appellant was one which he knew to be false. 22. This requires a determination, as a matter of statutory construction, of what is meant by the words “which he knows to be false or fraudulent” in section 37(a). Is it sufficient, as the prosecution contends,[17] that the appellant knew he had previous convictions but declared he had none and that his claimed mistaken belief that he was entitled not to disclose those convictions is irrelevant? Or, is it necessary, as the appellant contends, for the prosecution to prove that the appellant appreciated the false character of his declaration and, despite such appreciation and knowing its falsity, made the declaration.[18] 23. Before addressing the construction of section 37(a), because of the form of the question of law for which leave to appeal was granted, it is relevant to mention some general principles concerning ignorance and mistake of law. E.2 Ignorance of the law and mistake of law 24. There is no rule or principle that ignorance or mistake of law is not a defence to a criminal charge. Instead, there is only the self-evident truism that if, as a matter of construction, the requisite mens rea for a given offence does not include knowledge that it was wrongful, the prosecution does not have to prove that the accused knew it was wrongful. Whether such knowledge is required or not depends upon the construction of the language of the statute creating the offence. Many offences have no such requirement but some do. As Sir Garfield Barwick CJ put it in Iannella v French (1968) 119 CLR 84 at 97: “Mens rea may in some cases, depending … on the context and the subject matter, require that the defendant should know that the act is unlawful. That element of the offence itself cannot be eliminated in such a case by saying that ignorance of the law is no excuse. The defendant who is not shown in such a case to know that the act is unlawful needs no excuse. The offence has not been proved against him.” 25. Whether this is so will depend as a matter of construction on the mental element required by the statute. 26. In their printed Cases, both parties referred to decisions on the construction of various statutory offences, some of which were construed as imposing a mens rea requirement that included knowledge of the quality of the particular act or circumstances constituting the offence and some of which were construed as not requiring a particular mental state but merely knowledge of the facts necessary to establish guilt. 27. Thus, R v Smith (David) [1974] 1 QB 354,[19] R v Taaffe [1983] 1 WLR 627,[20] and Secretary of State for Trade and Industry v Hart [1982] 1 WLR 481[21] were referred to as examples of the former category of cases. Grant v Borg [1982] 2 All ER 257[22] and Attorney General’s Reference (No 1 of 1995) [1996] 4 All ER 21[23] were referred to as examples of the latter. It is unnecessary to discuss those cases further because each is simply a decision on the construction of the particular statutory provision in question. E.3 The construction of section 37(a) of the Crimes Ordinance 28. For the following reasons, we would hold that the mental element of the offence under section 37(a) requires the prosecution to prove that the defendant has an appreciation of the falsity of the declaration (etc.) he is making or producing and that, therefore, the defendant’s state of mind as to the falsity or otherwise of the declaration is relevant. If the defendant honestly believes that the declaration is not false, the mental element of the offence will not be established. 29. What the section requires is that the defendant “knows” the declaration is “false or fraudulent”. The nature of a particular statement may require an evaluation of other circumstances in order to determine its truthfulness or falsity. Unless the maker of such a statement appreciates that it is untrue, he cannot be said to know it is false or fraudulent.[24] Thus, in the present case, the declaration required of the appellant was whether he had any disclosable previous convictions (as opposed to whether he had any previous convictions at all). It is only by construing the ROO that the appellant could properly answer the inquiry as to whether he had any disclosable previous convictions. If section 2 of the ROO applied, but none of the exceptions in section 4 were relevant, his declaration would not have been untrue. 30. This construction of “declaration … which he knows to be false or fraudulent” is a natural construction of those words in the context of the section which is seeking to prohibit false statements made to procure registration on a register or roll for carrying on a vocation. 31. The above construction of section 37(a) is consistent with the English Court of Appeal’s construction of the words “knowingly and wilfully” in section 36 of the Crimes Ordinance: see R v Rajinder Kumar Sood [1998] 2 Cr. App. R. 355. Although Sood was cited by the respondent[25] as supporting its construction of section 37(a) in this appeal, that case is authority for the proposition that the words “knowingly and wilfully” in section 36 do not import as an additional requirement a further or ulterior intention on the part of a defendant. They nevertheless do require proof of an intention to do the act proscribed “with knowledge of the material circumstances which render it an offence”.[26] 32. Since, on a proper construction of section 37(a), an appreciation of the falsity of the declaration made is required, an honest and genuine belief by the defendant that the statement in question is in fact true, even if the result of a mistake of law, will preclude his being found to have the necessary mental element for the offence. In this way, a mistake of law may provide a defence to a charge under section 37(a). In the present case, such a mistake was relevant because of the application of the ROO. The effect of sections 2 and 4 of that ordinance (see paragraphs [6] and [7] above) was directly relevant to the fact of whether there was a disclosable conviction. F. Whether the courts below materially misapprehended the evidence? 33. The basis on which the Deputy Magistrate disbelieved the appellant’s evidence that he genuinely believed that he was excused from disclosing his previous convictions by reason of the ROO is his understanding of the content of the legal advice obtained by Ms Fan from the Free Legal Advice Scheme. [27] The Deputy Judge simply adopted the Deputy Magistrate’s reasoning in this regard. The appellant contends that substantial and grave injustice was done to him in that the Deputy Magistrate materially misapprehended the evidence below concerning that legal advice, and that the finding that his evidence was not to be believed cannot therefore be sustained. 34. The material parts of that legal advice and the Deputy Magistrate’s interpretation of it are described above. It will be recalled that Ms Fan had inquired about registration as a social worker and reference was made in the advice to the application form for registration as a social worker. A copy of the relevant registration form for social workers, with its accompanying notes, was admitted into evidence at trial as Exhibit MFI-1. 35. In this regard, it is material to note that the application for enrolment as a social worker under the Social Workers Registration Ordinance[28] differs from that for enrolment as an enrolled nurse. Section 37(5) of that Ordinance provides that: “(5) The Board shall exercise its power under this section in such a way as to require a person seeking to be registered as a registered social worker to make a statutory declaration as to – (a) whether he has been convicted of any offence, whether in Hong Kong or elsewhere; (b) if he has been so convicted, the nature of each such offence.” 36. The notes forming part of Exhibit MFI-1 provide, at paragraph 2 in respect of the statutory declaration to be completed, that: “According to Section 37(5) of the Social Workers Registration Ordinance, a person seeking to be registered as a registered social worker is required to make a statutory declaration as to (a) whether he has been convicted of any offence, whether in Hong Kong or elsewhere; (b) if he has been so convicted, the nature of each such offence. No exemption will be granted under the Rehabilitation of Offenders Ordinance (Cap.297). You are therefore required to make such a declaration in any circumstances.” [Emphasis added] 37. However, as already noted, it is common ground that the requirement on the appellant in completing the Declaration Form was to state whether he had any disclosable previous convictions. There is no equivalent of the provision in section 37(5) of the Social Workers Registration Ordinance in the Nurses Registration Ordinance. Instead, section 15 of the Nurses Registration Ordinance relevantly provides: “15. Enrolment (1) Any person who considers himself qualified to be enrolled in any part of the roll may apply in the manner prescribed to the secretary for enrolment. … (3) If, after due inquiry, the Council is satisfied that a person applying under subsection (1) has in Hong Kong or elsewhere – (a) been convicted of an offence punishable with imprisonment; or (b) been guilty of unprofessional conduct, the Council may, in its discretion, refuse to enter the name of that person upon the roll.” 38. The Declaration Form constitutes a form of “due inquiry” by the Council under section 15(3) of the Nurses Registration Ordinance as to whether an applicant has any previous convictions. However, there is no corresponding note in the Declaration Form for nurses to that in paragraph 2 of the notes forming part of Exhibit MFI-1 which excludes the ROO. 39. The advice given to Ms Fan was understood by the appellant as meaning that the exclusion of the ROO in Exhibit MFI-1 meant that any previous conviction had to be disclosed in that application form, regardless of the ROO, but if the form had no such express exclusion, he could properly state that he had no previous conviction. On the face of Exhibit D1, recording the advice given to Ms Fan, that would appear to be a correct construction of the legal advice she obtained on behalf of the appellant. 40. That this was his understanding of the legal advice is supported by his evidence in cross-examination where, in answer to the suggestion that Ms Fan had been advised “to disclose her record”, the appellant answered, “This is only the case for social worker.” He agreed that the inquiry had been about both social workers and nurses but disagreed with the suggestion that the advice did not mention nurses, presumably on the basis of his understanding that they fell into a category in which the form did not have a statement excluding the operation of the ROO. The appellant accepted in cross-examination that, taken literally, his declaration was false but that he understood the legal advice to mean that he did not have to disclose his previous convictions. 41. In paragraph 9 of the Statement of Findings, the Deputy Magistrate stated: “However, the content of exhibit D1 did not convey what the defendant said at all nor did it support his account. … I am also certain that he understood the content of D1. The legal advice at D1 indicated that if applicants are required to disclose their previous convictions on a form, they are then required to make disclosure. The defendant also made it clear that he understood that the legal advice at D1 was applicable to not only nurse applications but also social worker applications as his enquiry covered both professions.” 42. With respect, this was not correct. The Deputy Magistrate did not take any account of the differences in the statutory provisions referred to above or the respective forms for social workers and nurses and did not attempt to construe the second part of the legal advice dealing with the case where an applicant could legitimately declare that he did not have any previous conviction.[29] To make sense, that advice must have meant that an applicant might be able to declare that he had no previous conviction even though he actually did. In that advice, “such declaration” and “such provision” must have referred to the note excluding the operation of the ROO in Exhibit MFI-1. 43. Although, on the intermediate appeal, the appellant did not challenge the Deputy Magistrate’s refusal to accept his evidence,[30] it was incumbent on the Deputy Judge to consider whether the evidence was sufficient to convict the appellant of the offence. The Deputy Judge’s understandably brief conclusion (since this was not challenged on appeal) that the Deputy Magistrate “is right in finding that the appellant shall not rely on the legal advice he obtained to believe that he is not required to disclose his past conviction record” (see paragraph [14] above) is, with respect, an incorrect assessment of the evidence concerning the legal advice obtained and the belief which the appellant claimed he formed as a result of that advice. 44. In the circumstances, we consider that the appellant has established that the Deputy Magistrate and Deputy Judge both materially misapprehended the evidence concerning the legal advice obtained by him and that their respective conclusions that the appellant’s evidence (as to his belief that he was excused from the need to disclose his previous convictions in the Declaration Form) was to be disbelieved were fundamentally flawed. As a consequence, substantial and grave injustice was done to the appellant by reason of his conviction on this flawed basis. G. Conclusions and disposition 45. In the light of the conclusions that: (i) in answer to the question of law for which leave to appeal was granted, a defendant’s state of mind as to the falsity of the declaration was relevant and that the mental element of the offence would not be made out where a defendant held an honest and genuine belief that the declaration was not false (Section E.3 above); and that, (ii) substantial and grave injustice was done to the appellant in disbelieving his evidence that he held such an honest and genuine belief (Section F above), it follows that the appeal must be allowed and his conviction set aside. 46. Although there was no determination by the Deputy Magistrate or Deputy Judge as to whether, on a proper understanding of the legal advice, the appellant’s evidence as to his belief that the declaration he made was not false should be believed, it was not suggested by the respondent that a re-trial should be ordered. Once the evidence concerning the legal advice he obtained is properly understood, there is no proper basis for disbelieving the appellant. In the light of his honest and genuine belief that he was excused from disclosing his previous convictions by reason of the ROO, he could not be said to know that the Declaration Form contained a false or fraudulent statement and so he could not be guilty of the offence charged. Since, additionally, the appellant has already served the sentence imposed by the Deputy Magistrate, a re-trial would not be appropriate. 47. These are the Court’s reasons for our orders, made at the conclusion of the hearing of the appeal, to (1) allow the appellant’s appeal; (2) quash the appellant’s conviction below; and (3) order the respondent to pay the appellant’s costs of trial and of this appeal, to be taxed if not agreed. Mr Robert Pang SC and Mr Joseph Lee, instructed by Wat & Co., assigned by the Director of Legal Aid, and Mr Lawrence Pang, instructed by Wat & Co., on a pro bono basis, for the Appellant Mr Ned Lai SADPP and Mr Nicholas Wong SPP, of the Department of Justice, for the Respondent [1] (Cap.200). [2] The form follows the requirements of rule 4 of the Enrolled Nurses (Enrolment and Disciplinary Procedure) Regulations (Cap.164B) made under the Nurses Registration Ordinance (Cap.164). [3] This was Exhibit P4 at trial and its contents follow the requirements of section 15(3) of the Nurses Registration Ordinance. [4] (Cap.297) (“ROO”). [5] The Appellant’s Case at [19], read with [35]-[37]. [6] In TWCC 2202/2015, before Deputy Magistrate, Mr Jim Chun Ki. [7] (Cap.505). [8] In HCMA 113/2016, before Deputy High Court Judge Johnny Chan. [9] Judgment dated 13 March 2017 (“CFI Judgment”). [10] CFI Judgment at [46]. [11] HCMA 113/2016, Decision dated 16 May 2017. [12] (Cap.484A). [13] FAMC 38/2017, [2018] HKCFA 36, Determination dated 8 August 2018 at [1]. [14] Ibid. at [2]. [15] Ordinance No.21 of 1922. [16] See the speech of the Attorney General in Hong Kong Hansard, 21 September 1922, at p.84. [17] The Respondent’s Case at [44], [51]-[52] and [56]. [18] The Appellant’s Case at [24], [38] and [41]. [19] Concerning section 1(1) of the Criminal Damage Act 1971. [20] Affirmed on appeal, see R v Taaffe [1984] 1 AC 539 and concerning section 170(2) of the Customs and Excise Management Act 1979. [21] Concerning section 13 of the Companies Act 1976. [22] Concerning section 24(1)(b) of the Immigration Act 1971. [23] Concerning the offence of consenting to the acceptance of a deposit contrary to sections 3 and 96 of the Banking Act 1987. [24] It is unnecessary in this appeal to consider if the mental element of the offence can also be satisfied if the defendant is reckless as to whether the statement in question is true or false. [25] The Respondent’s Case at [38]. [26] [1998] 2 Cr. App. R. 355 per Potter LJ at 358D-F. [27] See paragraph [11] above and paragraph [41] below. [28] (Cap.505). [29] “I advised Ms Fan to disclose her record if such form had such declaration but if other forms do not have such provision, she would declare she has no previous conviction.” [30] CFI Judgment at [10]. 
Chief Justice Ma: 1. I agree with the judgments of Mr Justice Ribeiro PJ and Mr Justice Tang PJ. Mr Justice Ribeiro PJ : 2. I have had the benefit of reading in draft the judgment of Mr Justice Tang PJ and I am respectfully broadly in agreement with it. A. The background and the decisions below 3. Section 14 of the Buildings Ordinance[1] (“BO”) prohibits anyone from commencing or carrying out any building works without having first obtained approval by the Building Authority (“BA”) of the building plans and his consent for commencement of the building works. Accordingly, anyone who wishes to develop any site must first submit building plans for approval. 4. On 20 October 2010, the BA issued a circular to all authorized persons,[2] registered structural engineers and registered geotechnical engineers setting out its policy as follows: “The Building Authority has always taken the stance that the site of a proposed building for the purpose of the Buildings Ordinance ... can only include land which the applicant owns or which he has a realistic prospect of controlling. To facilitate the Building Authority to consider whether this requirement is satisfied, the applicant is required to submit a Form BA5 ... and particulars of ownership or realistic prospect of control of the land forming the site, together with the required documentary proof, when submitting new general building plans of any proposed new building on or after 21 October 2010.” 5. That circular followed earlier correspondence between the appellant (“REDA”) and the BA which stated its policy along the same lines and gave examples of the types of documents required: “Examples of documents or records which have been accepted as demonstrating a realistic prospect of controlling the land forming the site include agreements for sale and purchase, as well as authorization letters signed by registered owners authorizing an applicant to act on their behalf in pursuing the redevelopment of the subject building such as applying for approval of building plans for the proposed redevelopment. Moreover, an applicant may be considered as having a realistic prospect of controlling the land where he has successfully bid for a site at an auction (although he has yet to complete the payment and other land ownership registration procedures), or where he has a reasonable chance of success in a case involving land exchange with the Government.”[3] 6. Those communications took place against the background of Attorney General v Cheng Yick Chi,[4] a 1983 Privy Council decision on appeal from Hong Kong (to which I shall return) in which Lord Fraser of Tullybelton held that the land which forms a “site” for the purposes of the relevant legislation means land which a developer bona fide proposes to include in the development, being land which he owns or which he has a realistic prospect of controlling. 7. On 2 August 2013, applying Cheng Yick Chi,the Court of Appeal delivered judgments in two cases upholding the BA’s disapproval of the building plans submitted by two developers on the basis that they had failed to provide sufficient particulars of ownership or of a realistic prospect of control over parts of the proposed development sites. These were the Court of Appeal’s decisions Lai Siu Kin Rembert v Building Authority,[5] and Building Authority v Appeal Tribunal (Buildings)[6] respectively. 8. The parties in those proceedings did not seek leave to appeal to this Court. However, the present appeal flows from judicial review proceedings brought by the appellant for an order quashing the BA’s abovementioned policy as ultra vires and a declaration that the BA is not entitled, when considering building plans submitted, to require particulars and proof of ownership or realistic prospects of control, or to refuse his approval for failure to provide such particulars and proof. These proceedings therefore do not involve the submission or rejection of any actual building plans. 9. The application for judicial review was dismissed by Au J on the footing that he was bound by the abovementioned Court of Appeal decisions.[7] The present appeal is effectively a leap-frog appeal from his Lordship’s judgment to this Court since the Court of Appeal[8] held that it too was bound by those judgments and did not enter into any analysis of the issues. It granted leave to appeal to this Court on the basis of the following question of law: Does the Building Authority have an unqualified power in all cases, subject only to ordinary public law requirements (such as fairness or rationality), to: (1) reject building plans submitted for approval under s.14(1) Buildings Ordinance on the basis that a developer does not own or have a realistic prospect of controlling the site shown on the plans, or (2) require particulars or proof of ownership or realistic prospect of control of a site shown on building plans as pre-requisites for his approval of such plans under s.14(1) Buildings Ordinance? B. A question of statutory construction 10. The issue being one of vires,the question is whether power to refuse approval of building plans because the applicant has not provided sufficient particulars or proof of ownership or of a realistic prospect of control over the relevant site is conferred on the BA by the BO and the Regulations made thereunder. 11. This raises a question of statutory construction that has at its core, section 16(1) of the BO which specifies a series of grounds on which the BA may refuse to give his approval to building plans or refuse consent to the start of building works. By section 16(1)(i), he may refuse approval where: “... in his opinion, it is necessary for him to have further particulars of such plans or of the building works shown thereon ... to enable him fully to consider such plans.” 12. Lord Pannick QC[9] submits that this power is confined to particulars which are reasonably necessary to enable the BA to decide whether to disapprove the plans under one or more of the other limbs of s.16(1).[10] Counsel also accepted that the BA has implied power to do that which is incidental to, or consequential upon, the express functions which are conferred by the legislation.[11] C. BO section 16 construed in the light of its context and purpose 13. In my view, construing section 16 in the light of its context and purpose, the BA’s policy of seeking particulars or proof of ownership or realistic prospects of control of a site is reasonably necessary to enable him fully to consider whether submitted plans should be approved. 14. First, in my view, section 16(1)(d), taken in combination with section 16(1)(i), supplies a statutory basis for the BA’s abovementioned policy. This involves construing section 16(1)(d) in the context of the provisions of the BO and the Building (Planning) Regulations (“BPR”)[12] which limit the density of development on building sites. Secondly, the BA’s policy is justifiable on a broader construction of section 16 in the context of other BO provisions which make clear the purpose of the approval process. C.1 Section 16(1)(d) and the statutory maxima 15. Section 16(1)(d) provides: The Building Authority may refuse to give his approval of any plans of building works where ... (d) the carrying out of the building works shown thereon would contravene the provisions of this Ordinance or of any other enactment ...” 16. One of the main concerns of property developers as well as the BA relates to the maximum development potential of any particular site. Arguing against the BA’s policy and advocating greater commercial certainty for developers, REDA submitted: “The reason why unapproved plans give rise to uncertainty is because there are a large number of development parameters which cannot be known until the BA gives approval (for example, whether the site will be accepted as a class A, B or C site, whether bonus plot ratio will be granted, whether discretionary gross floor area will be granted, whether modifications or exemptions from the operation of the BO will be granted, and so on).”[13] 17. Such development parameters bear on permissible building densities and the setting of statutory maxima constitutes one of the most important functions[14] of the BO and its subsidiary legislation. As Lord Keith put it in Cinat Co Ltd v Attorney General,[15] referring to the BPR: “... the whole purpose of the regulations ... is to secure that in a particular locality the density of commercial and domestic buildings is no greater than accords with the public interest.” 18. It is therefore the BA’s duty to satisfy himself that all plans submitted do not propose building works which would contravene the BO and its subsidiary legislation, particularly the BPR, by exceeding the statutory maxima. 19. Building density is regulated by the combined effect of the BPR provisions on site classification, site coverage, plot ratio and building height. (a) “Site coverage” is defined relevantly to mean “the area of the site that is covered by the building that is erected thereon”[16] and “plot ratio” is defined as a value “obtained by dividing the gross floor area[17] of the building by the area of the site on which the building is erected”.[18] (b) “Permitted plot ratio” is defined to mean “the maximum plot ratio permitted under ... regulation 21”.[19] (c) Both “site coverage” and “permitted plot ratio” depend on how the site is classified and BPR reg 18A classifies sites according to how many streets of a stated width the site abuts on.[20] (d) BPR reg 19 provides that maximum site coverage and maximum plot ratio are to be ascertained, depending on the site classification, in accordance with regulations 20 and 21.[21] (e) BPR reg 20 lays down the maximum permitted site coverage by reference to percentages laid down in the First Schedule, related to the height of the building in question.[22] (f) Similarly, BPR reg 21 lays down the maximum permitted plot ratio by reference to the First Schedule.[23] (g) By way of illustration, under the First Schedule a domestic building of a height of over 36 m but not exceeding 43 m is permitted a maximum site coverage of 39% for a Class A site, 44% for a Class B site and 47% for a Class C site, with corresponding maximum plot ratios of 5.4, 6.1 and 6.5 for the respective site classifications. 20. The concept of “the site” is therefore plainly an essential element of this regulatory scheme. It is, however, nowhere defined in the legislation. Yet, it is necessary to know, as a matter of law, what constitutes a “site” to be able to classify the relevant land as a Class A, B or C site; to be able to determine what percentage of the land area ascertained to constitute the “site” would be covered by the proposed building; to be able to work out the plot ratio by reference to the gross floor area and the site area; and thus to determine whether the site coverage and plot ratio of the building shown on the plan would contravene the BO by exceeding the permitted maxima. 21. Faced with the absence of a statutory definition, Lord Fraser had to decide what ‘site’ means for the purposes of the BPR. His Lordship held that the land which forms a “site” must be ascertained as a question of fact in each case and that the concept includes “in addition to the land on which it is proposed to erect buildings, any land which the developer bona fide proposes to include in the development”. He held that a “site” could “only include land which he owns or which he has a realistic prospect of controlling”. 22. In my view, the approach adopted by Lord Fraser to the meaning of “site” is compelling and should be adopted for present purposes. This is because ownership and realistic prospects of control are often directly relevant to ascertaining the extent, position and nature of the site as an essential step in calculating the permitted parameters of the development. 23. This is illustrated by the Privy Council’s decision in Cheng Yick Chi[24]itself. In that case, the developers owned five adjoining houses which they wished to develop into a single new building. One of the main questions was whether a strip of land which had been dedicated to the public for pedestrian passage running in front of the five houses could be included as part of the “site”. If it could, the gross floor area of the proposed new building permitted by the BPR would be materially larger than if that strip of land were excluded from the calculations. 24. In addressing this issue, Lord Fraser elucidated the difference between a “lot” and a “site” as follows: “Mr Widdicombe for the respondents said that the site for purposes of development is normally a whole ‘lot’, using ‘lot’ to mean the leased area held directly or indirectly under a lease from the Crown. That may well be so. But (rightly), he did not contend that the ‘site’ was necessarily coextensive with the ‘lot’, because the word site must be flexible enough to apply to a case in which the person holding several adjacent lots under Crown leases proposes to develop them together as a single site. It must also apply where the owner develops only part of his lot, or develops it bit by bit at different times. For example, if a development scheme applies to a substantial area of perhaps an acre or more, to be developed by building a detached house on part of it, with the rest of the land being laid out as a garden and tennis court, the whole area would be the site. If some years later the tennis court is sold off in order to build a separate house on it, the site for that later development would be limited to the tennis court.[25]” 25. This passage demonstrates the close nexus between ownership or control of the land and ascertaining the nature and extent of the site. In the instances adumbrated by Lord Fraser, what constitutes the “site” is wholly dependent on decisions taken by owners or those in control of the land. Thus, proposals to amalgamate different lots to form one development site; to develop only part of a lot as a site; or to develop a lot in phases as a series of sites, are proposals that necessarily rest on decisions that can only be taken by the person who owns or who at least has realistic prospects of controlling the land. They are decisions which someone who has no realistic prospects of controlling the land could not possibly take. 26. The Privy Council accordingly concluded that “site” has to be given a meaning which recognizes that nexus, holding that a “site” for the purposes of the BPR, can only include land which the applicant for approval owns or has a realistic prospect of controlling. Lord Fraser’s approach was adopted in Wharf Properties Ltd v Eric Cumine Associates, Architects, Engineers and Surveyors,[26] where Lord Oliver of Aylmerton stated: “In general, the owner of land which he desires to develop is at liberty to designate as his site the whole of any part of the land which he owns, including land which already has existing buildings on it, although, of course, the floor area of the existing buildings will have to be taken into account in determining whether any new buildings planned will exceed the permitted plot ratio. This may seem almost self-evident but if authority is needed for the proposition it is contained in the decision of the Board in ... Attorney General v Cheng Yik Chi ... [citing the aforesaid passage from Lord Fraser]”. 27. Lord Fraser’s approach was also acknowledged in the Privy Council decisions in Hinge Well Co Ltd v Attorney General,[27] and Cinat Company Limited v The Attorney General of Hong Kong.[28] 28. It was also applied by the Court of Appeal in Building Authority v Appeal Tribunal (Buildings)[29] which provides a further illustration of how ownership of the land can be decisive in determining the extent or area of a site. The development proposal in that case was in respect of Nos 7-9 Ying Wa Terrace which was adjacent to Nos 10-12. The issue was whether the area in front of Nos 7-9, over which the owners of units in Nos 10-12 had a right of way, could be included as part of the site area of Nos 7-9 with consequences for calculating permitted site coverage and plot ratio.[30] The BA refused building plan approval pending clarification by the developers of “whether the developer of 7-9 Ying Wa Terrace has full control of the land at 10-12 Ying Wa Terrace”.[31] Applying Cheng Yick Chi,Fok JA (as he then was) held that “the BA was entitled to seek information from the Developers regarding ownership and control of Nos 10-12 for the purpose of determining whether or not the right of way should be excluded from the calculation of site area and plot ratio for Nos 7-9.”[32] As his Lordship explained:[33] “In order to ascertain the status of that right of way, it was necessary to ask for particulars of the Developers' control of Nos 10-12. If that site was in the common ownership of the Developers then the right of way would be extinguished and it could be ignored for the purposes of reg.23(2)(a)[34].” 29. REDA accepts that the BA has power to seek the relevant information in cases like Building Authority v Appeal Tribunal (Buildings)[35] “where such information is necessary in order to determine whether a specific, identified provision of the building legislation has been or will be complied with”, but argues that the BA’s policy impermissibly claims a general, free-standing power to reject general building plans for lack of proof of ownership or control “even where there is no other ground for disapproval in issue”.[36] That is too narrow an approach. Common experience shows that there are often cases where the connection between ownership or control and ascertainment of the statutory maxima is important and may be decisive. Possible grounds for disapproval because of contravention by exceeding such maxima may only emerge as issues as a result of the BA requiring particulars and proof in the first place, making the BA’s policy intra vires, rational and reasonable. C.2 Section 16(1) in its broader context and purpose 30. REDA’s challenge to the BA’s policy connotes the proposition that a person who does not own and does not have any realistic prospect of controlling the land comprised in the building plans is legally entitled to require the BA to undertake the approval process. That is a surprising suggestion since it would be natural to assume that a person asking for approval of specific plans for putting up a building on a particular site is able or has realistic chances of being able to carry out the proposed development because he owns or has realistic prospects of controlling the land in question. 31. Moreover, by virtue of BO section 15(1)[37] and regulation 30(3) of the Building (Administration) Regulations (“BAR”),[38] the BA would have no more than 60 days to reach his decision on approval, in default of which he would be deemed to have given his approval. If REDA’s proposition is correct, it would follow that anyone could impose that public duty on the BA even if he has no realistic prospect of controlling the site. It would also mean that there could be several separate applications for approval of plans regarding the same site where none of the applicants have a realistic prospect of gaining control thereof, even if, for instance, the owner had made it clear that he had no interest in selling the property or having anyone else develop it. 32. Resources naturally being limited, such a state of affairs would place heavy administrative burdens on the BA and cause delays in the approval process in relation to owners or persons who do have genuine and bona fide development projects. 33. Lord Pannick QC reminded the Court of Lord Millett’s often-cited warning in China Field Ltd v Appeal Tribunal (Buildings)(No 2),[39] against a consequentialist interpretation of a statute unsupported by the statutory language, where his Lordship said: “There can be no quarrel with the principle that statutory provisions should be given a purposive interpretation, but there has been a distressing development by the courts which allows them to distort or even ignore the plain meaning of the text and construe the statute in whatever manner achieves a result which they consider desirable. It cannot be said too often that this is not permissible. Purposive construction means only that statutory provisions are to be interpreted to give effect to the intention of the legislature, and that intention must be ascertained by a proper application of the interpretative process. This does not permit the Court to attribute to a statutory provision a meaning which the language of the statute, understood in the light of it context and the statutory purpose, is incapable of bearing: see HKSAR v. Lam Kwong Wai (2006) 9 HKCFAR 574 at §63.” 34. However, for the reasons discussed in Mr Justice Tang PJ’s judgment which I respectfully adopt, REDA’s argument, with its untoward consequences, can properly be rejected by construing section 16 purposively in the broader context of the other provisions of the BO dealing with the approval of building plans and consent to the commencement of building works. 35. As Tang PJ points out, the statutory scheme is not merely concerned with paper approvals of what may only be speculative and academic applications but with approvals as a necessary condition of carrying out building works. The whole point of obtaining approval is to be permitted to carry out building works. BO section 14 provides: “Save as otherwise provided, no person shall commence or carry out any building works or street works without having first obtained from the Building Authority – (a) his approval in writing of documents submitted to him in accordance with the regulations; and (b) his consent in writing for the commencement of the building works or street works shown in the approved plan.” 36. Section 14 is thus structured as a prohibition against carrying out such works without first having obtained building plan approval and without first obtaining consent to the commencement of the building works shown on the plan. The implicit premise is that the applicant owns or controls the site and so is able to carry out the building works depicted in the submitted building plans. 37. The provisions of section 15 which place pressure on the BA to reach a decision regarding approval or to give reasons for a refusal within 60 days, deeming approval given unless a refusal is notified, have been mentioned above. The injection of such urgency is only explicable on the basis that the BO recognizes that plan approval is the first step of the process and that the second step involving the commencement of building works should not be held up. The statutory intent therefore maintains its focus on an actual, and not a purely hypothetical, development project. The BO would hardly have injected such urgency into the scheme if it was intended that plan approvals could be sought by persons with no realistic prospects of controlling the site and so no realistic prospects of starting construction work thereon if approval is given. Section 16(3) makes it clear that building works should begin not more than two years after plan approval.[40] 38. Construing section 16 in the aforesaid context, I respectfully agree with Lord Fraser’s conclusion in Cheng Yick Chi that, for the purposes of the BO and its subsidiary legislation, for land to qualify as a site it must be “land which the developer bona fide proposes to include in the development” and thus, “can only include land which he owns or which he has a realistic prospect of controlling”. That approach was adopted by the Court of Appeal in Lai Siu Kin Rembert v Building Authority,[41] which was, in my view, correctly decided. 39. For the foregoing reasons, the BA’s policy of requiring the applicant to provide the requisite particulars and proof is not ultra vires. Mr Justice Tang PJ: Introduction 40. Section 14(1) of the Buildings Ordinance (“the Ordinance”)[42] provides that no person may commence or carry out any building work[43] without having first obtained from the Building Authority (“BA”): “(a) his approval in writing of documents submitted to him in accordance with the regulations; and (b) his consent in writing for the commencement of the building works … shown in the approved plan.” 41. Under Regulation 8 of the Building (Administration) Regulations (“BAR”), the documents which have to be submitted include: “(1)(b) plans showing – (iii) the width of every street adjacent to the site; and … (f) a block plan, showing the size and position of the building and of other buildings in the immediate vicinity, and neighbouring streets; (g) where the position of the site is not clearly identifiable from such block plan, a key plan showing the position of the site; (h) a diagrammatic plan, with full dimensions and calculations, showing – (i) the height, site coverage, plot ratio and disposition of the building in accordance with regulations 19, 20 and 21 of the Building (Planning) Regulations … ; 42. The grounds on which approval of plans or consent to the commencement of building works may be refused are set out in s 16.[44] Section 16(1) sets out the grounds upon which the plans could be refused approval and s 16(3) where the BA may refuse consent to commence work.[45] 43. In these proceedings, the applicant, the Real Estate Developers Association of Hong Kong (“REDA”), challenges the lawfulness of the BA’s disapproval under s 16 of building plans on the basis that the developer does not own or have a realistic prospect of controlling the site of the proposed development shown on those plans. 44. Lord Pannick QC who appeared for the applicant, together with Ms Yvonne Cheng SC and Mr Wilson Leung, submitted that the Ordinance does not expressly or by implication require an applicant to be the owner or have a realistic prospect of control of the site and that any requirement by the BA for proof of the same is ultra vires and unlawful. 45. In this court, Mr Benjamin Yu SC who appeared for the BA together with Mr Bernard Man SC relied on s 16(1)(a), (c), (d) and (i) in support of the BA’s policy to require proof that the developer owns or has reasonable prospect of control over the site the subject of the building plans and to reject them when such proof is unavailable. In brief, the BA’s case is that the concept of site is central to the regulation of the design of buildings and hence, of approval of building plans under the Ordinance and Regulations. 46. This is not the first time that the BA’s policy has been challenged, although it is the first time that it has come before this court. In Lai Siu Kin Rembert v Building Authority[46]the Court of Appeal upheld the BA’s decision to reject plans submitted by an applicant who owned 85% of the units on the proposed site. There, the court followed Attorney General v Cheng Yick Chi & Others,[47] by which it regarded itself bound, where Lord Fraser, in the advice of the Board said: “Their Lordships are of the opinion that the land which forms a ‘site’ for the purposes of the Regulations must be ascertained as a question of fact in the case of each development. It means, in addition to the land on which it is proposed to erect buildings, any land which the developer bona fide proposes to include in the development. It can only include land which he owns or which he has a realistic prospect of controlling.”[48] (Emphasis added) 47. Cheng Yick Chi is not binding on this court and Lord Pannick submitted that the words of Lord Fraser relied on by the Court of Appeal in the Case Stated Decision were not of general application and if they were meant to be of general application, they were wrong and we should take this opportunity to say so. 48. In these proceedings, the courts below regarded themselves bound by Cheng Yick Chi and on 10 September 2015, the Court of Appeal[49] granted leave to appeal to this court on the following question: “1. Does the Building Authority have an unqualified power in all cases, subject only to ordinary public law requirements (such as fairness or rationality), to: (1) reject building plans submitted for approval under s.14(1) Buildings Ordinance on the basis that a developer does not own or have a realistic prospect of controlling the site shown on the plans, or (2) require particulars or proof of ownership or realistic prospect of control of a site shown on building plans as pre-requisites for his approval of such plans under s.14 (1) Buildings Ordinance?” 49. The parties are agreed, and it is plainly right, that the resolution of the question turns on the proper construction of the Ordinance. That being the case, I shall first endeavour to come to a view on the proper construction before turning my attention to Cheng Yick Chi. Background 50. The immediate origin of these proceedings could be traced to the 2010-2011 Policy Address[50] on 13 October 2010 where it was announced that concessions regarding gross floor areas[51] in return, for, inter alia, green and amenity features would not apply to new building plans submitted on or after 1 April 2011. 51. Following the Policy Address, a BA circular letter dated 20 October 2010, to all authorised persons,[52] registered structural engineers, and registered geotechnical engineers, stated that the site of a proposed building for the purpose of the Ordinance could only include land which the applicant owned or which he had a realistic prospect of controlling and an applicant was required to submit a new Form BA 5 and particulars of ownership or realistic prospect of control of the land forming the site, together with the required documentary proof, when submitting new general building plans of any proposed new building on or after 21 October 2010, failing which BA might refuse to approve the plans under s 16(1) of the Ordinance.[53] 52. BA also explained in correspondence with REDA, dated 8 August 2011, that the “step up enforcement … in respect of ownership or realistic prospect of control of the … site” was made in the midst of “rising public concern”, that plans approved could not be implemented within a reasonable time so that by the time the plans were implemented, they would no longer “be conducive to a quality and sustainable built environment”.[54] 53. Litigation followed the stepped up enforcement of the ownership and control requirement. The plans which were the subject of the Case Stated Judgment were submitted during the window period between 13 October 2010 and 1 April 2011. The plans in the Dobrowen appeal had been submitted earlier but one of the reasons for disapproval was that the applicant had failed to demonstrate ownership or realistic prospect of control of the entire site. Both these proceedings terminated in the Court of Appeal. There was no appeal to this court. REDA had sought leave to join the Case Stated Proceedings which was refused. The present application was commenced by REDA so that this court should settle the point. It was said in the Case Stated Judgment that there were 116 appeals before the Tribunal which were in abeyance pending the resolution of the question.[55] Presumably they now await our decision. Our decision is likely to have important practical consequences, because if plans had been wrongly rejected which ought to have been approved earlier, they might be entitled to the concessions referred to in para 50 above. Also, according to Mr Lau Chi Keung[56] the earlier approval of building plans might reduce the risk of the developer being affected by changes in law including adverse changes in the zoning plans. Interpretation 54. Lord Pannick submitted that as “site” is not defined in the Ordinance or the Regulations, the starting point is to take the ordinary meaning of the term, viz a piece of land. It is simply a question of physical area and “site” is what the applicant chooses to put on the plan. But as the Chief Justice said in Fully Profit(Asia) Ltd v Secretary for Justice:[57] “… that in the area of statutory and constitutional interpretation, it is context that is key: context is the starting point (together with purpose) rather than looking at what may be the natural and ordinary meaning of words.” 55. I have no doubt that I should interpret “site” according to its context and purpose. Context and Purpose 56. The context is the approval of plans and consent to commencement of building works and the purpose of the Ordinance can be gathered from its long title, viz “To provide for the planning, design and construction of buildings …”. 57. BAR 8(1)(g) requires the site to be identified. Identification of the site on the plans is of critical importance because depending on its classification, location and size, the permissible building is determined. Three classes of sites are defined in the Building (Planning) Regulations (“BPR”). Classes B and C respectively are corner sites which abut on 2 or 3 specified[58] streets, none of which is less than 4.5 m wide. A Class A site is a site, not being a Class B or Class C site, which abuts on one or more of such specified streets. The maximum site coverage[59] and plot ratio[60] for Class A, B or C sites are regulated by BPR 20 and 21 respectively.[61] Site coverage and plot ratio in turn depend on the height of the proposed building.[62] For a site which is neither Class A, B or C, its height, and the maximum site coverage and plot ratio are determined by the BA.[63] 58. Under BPR 25 and the second schedule of the BPR, each domestic building must have an open space at the rear. The size of such open-space differs according to whether the site is a Class A, B or C site. Also, under BPR 2, the definition of site will affect whether a building can be classified as a detached or semi-detached building, which in turn will determine (a) whether it is necessary to build a service lane and (b) the height of windows. 59. It is no exaggeration to say that the site is the single most important determining factor in a plan. 60. If Lord Pannick is right and a site is simply any physical area delineated on the plan and it could be hypothetical, then as Lam J (as he then was), said at first instance in the Case Stated Appeal, a site could cover the whole of the Hong Kong Island or as Reyes J said in Ashley 121 Ltd v Appeal Tribunal (Buildings) [2011] 2 HKLRD 728 at para 83, the BA could be vexed by a flood of hypothetical proposals. These are far-fetched examples and Lord Pannick submitted that given the substantial expense involved in any submission of plans, it is unlikely that there would be a flood of frivolous applications following a decision in favour of REDA. Even so, these possibilities support the view that Lord Pannick’s submission is unlikely to be correct. 61. I have mentioned the opening words of its long title: “to provide for the planning, design and construction of buildings…”. Control over density of buildings is unquestionably one of its objects of the Ordinance.[64] BA’s functions and duties in these regards are no less important than their functions and duties in ensuring the safety of buildings. Lord Pannick submitted, however, that anyone who is prepared to incur the expenditure of an application should be permitted to do so even though there is no realistic prospect of his being able to build on the site. He sought to justify this submission by saying that this would enable a developer to know the potential of a made-up site. It is said that one of the benefits is that a developer would know how much he should pay to acquire the site. But this has to be balanced against the rising public concern referred to in BA’s letter of 8 August 2011.[65] Moreover, the approval process is transparent and there is an enquiry service provided by the BA for persons who desire clarification relating to the interpretation of or compliance with the Ordinance and its Regulations, including the development potential of a site. In any event, I do not believe these considerations assist in the interpretation. 62. Approval of building plans is dealt with in the context of a prohibition against building works without the necessary approval. Section 14 not only requires approval of plans, it also provides that building work could not be commenced without BA’s consent. The context is clearly of an approval of plans for intended building works. 63. Section 14 requires the submission of documents in accordance with the regulations for the approval of plans. The regulations, as discussed above, require the identification of the site on which building work was to be performed, and any approval would be site specific. Section 14 also provides for application for the consent to commence work at such site and in accordance with such approved plans. I do not believe it was ever envisaged that a site could be hypothetical. Under s 15, the plansare deemed to have been approved and consent to commencement of work are deemed to have been given unless the BA had notified his refusal to approve or consent setting out the grounds for such refusal within a prescribed number of days, namely, 60 days,[66] after their respective applications. Also, s 16(3) provides that the BA may refuse to give consent to the commencement of work if (d) a period exceeding two years has elapsed since (the plans were approved). Thus, the building plans were required to be approved or disapproved within a tight time table for a development which is expected to be commenced within 2 years. They show clearly that the plans to be submitted were expected to be plans for intended developments. 64. Lord Pannick relied on s 14(2)(a) which provides that neither the approval of any plans nor the consent to the commencement of any building works can be deemed to confer any title to land. This does not support the view that the legislature had in mind hypothetical applications. The purpose of s 14(2)(a) was to avoid any argument that any defect in title had been cured by such approvals. In other words, third party property rights should not be prejudiced by such approvals. 65. Lord Pannick submitted that s 16(1) does not expressly say that plans could be disapproved if the applicant was not the owner of the site or had no realistic prospect of control of the site. That is so. Nor does it expressly provide under s 16(3) that consent to commence work could be refused if the applicant was not the owner or had no realistic prospect of control of the site. In the case of consent to commence work, it is clearly unarguable that the BA would be obliged to give consent even if the applicant was neither the owner nor had any realistic prospect of control of the site. That would make any such consent meaningless. I believe there is no express provision under s 16(1) or s 16(3) because it was assumed that the plans would have been submitted by or on behalf of a person who bona fide[67] intended to develop the site. An applicant would not be so regarded unless he had the means and ability to do so. That is why requirement of proof of ownership or realistic prospect of control may properly be required before plans are approved. 66. Section 16(1) contains 17 sub–clauses and Mr Yu relied on four of them: “(1) the Building Authority may refuse to give his approval of any plans of building works where – (a) the plans are not such as are prescribed by regulations or are not such as he may have required under this section; … (c) he has not received application for their approval in the specified form or any such application does not contain the particulars required therein; (d) the carrying out of the building works shown thereon would contravene the provisions of this Ordinance or of any other enactment, or would contravene any approved or draft plan prepared under the Town Planning Ordinance (Cap 131); … (i) in his opinion, it is necessary for him to have further particulars of such plans or of the building works shown thereon or, where all the plans prescribed by regulations have not been submitted, to have one or more of the other plans prescribed by regulations, to enable him fully to consider such plans;” 67. Given my view that the site for which application for approval was made must be a site on which it was bona fide intended that the approved building would be built, the BA was entitled to require particulars of ownership or realistic prospect of control of the site under s 16(1)(i). It also follows that the other provisions were also engaged: under (1)(a), the plans could be refused approval because that did not relate to such a site; under (1)(c) because the application did not contain the particulars of ownership or realistic prospect of control required and under (1)(d) because only building works in relation to which plans had been properly approved could be built. 68. For the above reasons, as a matter of construction, I am of the view that the certified question should receive an affirmative answer. 69. I turn now to consider Cheng Yick Chi. Here I have the benefit of the Case Stated Judgment where Fok JA (now Fok PJ) dealt fully with the scope and application of Cheng Yick Chi. In Cheng Yick Chi, the issue was what was the meaning of a site for the purpose of the BPR and BO. Lord Fraser’s answer was quoted above at para 46 and I will not repeat. It was argued before Fok JA that that was not part of the ratio and therefore not binding on the Court of Appeal. That is not a matter which concerns us. But, with respect, I fully agree with Fok JA that it was part of the ratio of Cheng Yick Chi. In order to examine the contrary argument it is necessary to go briefly into the facts. There, the respondent owner wanted to include in a site, the subject of the application for the approval of plans, a strip of land which he owned but which he had previously dedicated to the public for passage and the question was whether it remained available for inclusion in the proposed development. The passage would not be built upon and would remain a passage in the proposed plans. It was in such context that their Lordships considered the meaning of “site” in the BPR, after noting at 19C that “Neither the Regulations nor the Buildings Ordinance contains a definition of site”, they concluded that a site “can only include land which he owns or which he has a realistic prospect of controlling.” This was the answer given to one of the declarations appealed against, namely, declaration (c), which concerned whether the proposed site formed “one site for a single building is a Class A site within the meaning of reg. 2 of (BPR)”. It was in that context, that Lord Fraser said: “… the land which forms a ‘site’ for the purposes of the Regulations must be ascertained as a question of fact in the case of each development … It can only include land which he owns or which he has a realistic prospect of controlling.” (19I)[68] 70. After dealing with declaration (c), Lord Fraser went on to deal with another question, declaration (a) which concerned the proper construction of Regulation 23(2)(b) . 71. Regulation 23(2)(b) has two limbs and reads : “In determining for the purposes of regs. 20, 21 or 22 the area of the site on which the building is erected – (a) no account shall be taken of any part of any street or service lane; and (b) there shall be included in any area dedicated to the public for the purposes of passage.” 72. The owner relied on BPR 23(2)(b) on the basis that the passage had been dedicated to public passage. Lord Fraser was of the view that BPR 23(2)(b) only applied to the areas which, in accordance with the relevant scheme for development, are proposed for dedication in the future, as if the words “to be” occurred between “area” and “dedicated” (p 21 I). That being the case, Lord Fraser said at 22D: “… in determining for the purposes of regs 20, 21 and 22 the area of the site, no account is to be taken of any part of any existing street, but account is to be taken of any area which, in the proposed development, is to be dedicated to the public for passage. Areas dedicated in the past are excluded.” 73. Lord Pannick is right that the answer given to declaration (a) would have disposed of the appeal in Cheng Yick Chi. But Lord Fraser’s dictum regarding the meaning of “site” for the purposes of the regulations was given in answer to one of the grounds of appeal. So, I agree with Fok JA that he was bound by Cheng Yick Chi. 74. Lord Pannick noted that Lord Fraser did not give any reason for his statement that a site could only include land which the developer owns or controls. I rather think that was because the conclusion was too obvious to require elaboration. 75. Nor does Cheng Yick Chi stand alone. I will mention only three decisions of the Privy Council in which Lord Fraser’s dictum was cited without reservation. I will deal with them chronologically. 76. Hinge Well Co Ltd v Attorney General [1988] 1 HKLR 32 [69] was concerned with the construction of Regulation 23(2)(a). (See pp 43-44) There, a scavenging lane divided two plots of land owned by the applicant and the question was whether the two plots formed one site for the purpose of BPR. Lord Pannick submitted that Hinge Well turned on Regulation 23(2)(a) and provided no support for Lord Fraser’s dictum. But at p 39 Lord Oliver said the respondent’s (BA) argument was based on Lord Fraser’s words “land which he owned or which he had realistic prospect of controlling”. In response, his Lordship said a landowner whose land is subject to a private easement could negotiate with those entitled to the easement, and whether he had any realistic prospect of controlling the land over which the easement is exercisable must be a question of fact in each case. This, I believe, was an affirmation of Lord Fraser’s dictum. 77. Wharf Properties Ltd & Another v Eric Cumine Associates, Architects, Engineers and Surveyors [1991] 2 HKLR 6, was concerned with professional negligence and the facts do not concern us. But at p 13, Lord Oliver of Aylmerton,[70] cited Lord Fraser’s dictum as authority that a developer was at liberty to designate as his site the whole of any part of land which he owns. 78. In Cinat Company Limited v The Attorney General of Hong Kong [1995] 1 HKLR 128,[71] the question was whether development could take place on a vacant piece of land once it had been included as part of a site for the purpose of calculating permitted site coverage and permissible plot ratio of another development. The only authority cited in the judgment was Cheng Yick Chi and the dictum at para 46 above was cited together with the words which followed, namely that the site must not include: “land [which] …, so to speak, used up in enabling some other existing building to comply with the regulations.” The additional words were decisive of the appeal in Cinat, because: “It was thus recognised that a development site might include some particular area of land owned or controlled by the developer which was not intended to be built on but which was necessary to enable the proposed building to comply with the regulations.” (My emphasis) 79. I also note that Lord Keith said at p 131 the whole purpose of the regulations, was to secure that in a particular locality the density of commercial and domestic buildings is no greater than accords with the public interest. 80. With respect, but for the fact that our decision is of great importance to the community and the real estate industry, I would have been content to say that I agree with Cheng Yick Chi without further elaboration. As it is I have endeavoured to explain why as a matter of construction I would have come to the same conclusion as Cheng Yick Chi. 81. For the above reasons, I would dismiss REDA’s appeal. Mr Justice Fok PJ: 82. In Lai Siu Kin Rembert v Building Authority[72] and Building Authority v Appeal Tribunal (Buildings)[73], one of the issues had been whether Lord Fraser of Tullybelton’s dictum in Attorney General v Cheng Yick Chi & Others[74] was part of the ratio decidendi of the Privy Council’s decision and therefore binding on the Court of Appeal hearing those cases (of which I was a member). The present appeal from the Court of Appeal (of which I was not a member) is not an appeal from those decisions but is instead a challenge to the correctness of Lord Fraser’s holding that, for the purposes of the Buildings Ordinance (Cap.123), a site can only include land which a person owns or which he has a realistic prospect of controlling. That was not an issue determined in the two earlier cases before the Court of Appeal in which I participated. I have had the benefit of reading in draft the judgments of both Ribeiro PJ and Tang PJ in which they each conclude that Lord Fraser was correct in so holding. I agree with their reasons for reaching that conclusion and that, accordingly, REDA’s appeal should therefore be dismissed. Lord Clarke of Stone-cum-Ebony NPJ: 83. I agree that REDA's appeal should be dismissed for the reasons given by both Mr Justice Tang PJ and Mr Justice Ribeiro PJ. Chief Justice Ma: 84. For the above reasons, the appeal is dismissed. We would also make an order nisi that the appellant pay to the respondent the costs of the appeal, such costs to be taxed if not agreed. Should any party wish for a different order for costs, written submissions should be served on the other party and lodged with the Registrar within 14 days of the handing down of this judgment, with liberty to the other party to serve and lodge with the Registrar written submissions in reply within 14 days thereafter. In the absence of any submissions seeking to vary within the time limited for their service, the costs order nisi will become absolute. Lord Pannick QC, Ms Yvonne Cheng SC and Mr Wilson Leung, instructed by Mayer Brown JSM, for the applicant (appellant) Mr Benjamin Yu SC and Mr Bernard Man SC, instructed by the Department of Justice, for the respondent (respondent) [1] Cap 123. [2] Defined in the BO as persons whose names are in a statutory register as architects, engineers or surveyors. [3] BA to REDA 8 August 2011. [4] [1983] 1 HKC 14 at 19. [5] [2013] 4 HKLRD 74, referred to below as the “Case Stated appeal”. [6] [2013] 4 HKLRD 52 referred to below as the “Dobrowen appeal” since the 1st Interested Party was Dobrowen Investment Limited. [7] HCAL 95/2011 (Reasons, 30 June 2014). [8] Lam VP, Chu and McWalters JJA, CACV 105/2015 (10 September 2015). [9] Appearing for REDA with Ms Yvonne Cheng SC and Mr Wilson Leung. [10] Appellant’s Case §38. [11] Appellant’s Case §31(3). [12] Cap 123. [13] Appellant's Case §4. [14] Along with promoting safety in building design and structure: Gammon v A-G of Hong Kong [1985] AC 1 at 14; and Mariner International Hotels v Atlas Ltd (2007) 10 HKCFAR 1 at §51. [15] [1995] 1 HKLR 128 at 131. [16] BPR reg 2. [17] “Gross floor area” being “the area contained within the external walls of the building measured at each floor level (including any floor below the level of the ground), together with the area of each balcony in the building, which shall be calculated from the overall dimensions of the balcony (including the thickness of the sides thereof), and the thickness of the external walls of the building.” (BPR reg 23(3)(a)). [18] BPR reg 21(3). [19] BPR reg 2. [20] BPR reg 18A: “(1) In this Part and the First and Second Schedules - ‘class A site’ means a site, not being a class B site or class C site, that abuts on one specified street not less than 4.5 m wide or on more than one such street; ‘class B site’ means, subject to paragraph (2), a corner site that abuts on 2 specified streets neither of which is less than 4.5 m wide; class C site’ means, subject to paragraph (2), a corner site that abuts on 3 specified streets none of which is less than 4.5 m wide.” Sub-regulations (2) and (3) contain elaborations. [21] BPR reg 19: “(1) The maximum site coverage permitted in respect of a building or buildings on a class A site, class B site or class C site shall be determined in accordance with regulation 20. (2) The maximum plot ratio permitted in respect of a building or buildings on a class A site, class B site or class C site shall be determined in accordance with regulation 21.” Sub-regulation (3) deals with special cases requiring ad hoc determination of the maxima. [22] BPR reg 20(1)(a): “Subject to regulation 22 and depending on the height of the building- the site coverage for a domestic building, or for the domestic part of a composite building, on a class A site shall not exceed that percentage of the area of the site specified in the second column of the First Schedule...” The rest of reg 20 deals with sites with other site classifications, non-domestic buildings and so forth. [23] BPR reg 21(1)(a): “Subject to regulation 22 and depending on the height of the building- the plot ratio for a domestic building on a class A site shall not exceed the plot ratio specified in the fifth column of the First Schedule ...” The other provisions of reg 21 deal with other site classifications, non-domestic buildings and so forth. [24] [1983] 1 HKC 14. [25] [1983] 1 HKC 14 at 19. [26] [1991] 2 HKLR 6 at 12-13. [27] [1988] 1 HKLR 32 at 39. [28] [1995] 1 HKLR 128 at 131. [29] [2013] 4 HKLRD 52. [30] At §14 and §36. [31] At §16. [32] At §39. [33] At §38. [34] Which excludes streets or service lanes from the calculation of permitted site coverage and plot ratio. [35] [2013] 4 HKLRD 52. [36] Appellant's Case §54. [37] BO s 15(1): “Where an application is made in the specified form for the Building Authority to approve plans or to consent to the commencement of building works or street works, he shall be deemed to have given his approval or consent, as the case may be, unless within the period prescribed by the regulations he has notified his refusal to give his approval or consent, as the case may be, in writing setting out the grounds for such refusal, and where one of such grounds is that further particulars and plans are required, he shall specify such plans and particulars.” [38] BAR reg 30(3): “For the purposes of section 15 of the Ordinance, the period after which the Building Authority shall be deemed to have given his approval of plans submitted to him shall be- (a) in relation to plans which are submitted for the first time to the Building Authority for approval, 60 days from the date on which the plans were submitted;...” [39] [2009] 12 HKCFAR 342 at §36. [40] BO section 16(3)(d): “The Building Authority may refuse to give his consent to the commencement of any building works or street works where … a period exceeding 2 years has elapsed since the approval of any of the prescribed plans in respect of the building works or street works.” [41] [2013] 4 HKLRD 74. [42] All references are to the Ordinance unless otherwise stated. [43] In these proceedings we are concerned with building plans for new buildings. [44] It is unnecessary to decide whether these grounds are exhaustive since the BA only relied on s 16. [45] Section 16(2) concerns street works and is irrelevant. [46] [2013] 4 HKLRD 74. A decision of the Court of Appeal, Cheung and Fok JJA and McWalters J. This was referred to the Court of Appeal by the Appeal Tribunal under s 53C(1) and has been referred to as the Case Stated Judgment. I will refer to it as such. This judgment was delivered on the same day as the judgment in Building Authority v Appeal Tribunal (Buildings)[2013] 4 HKLRD 52 (the Dobrowen judgment), which is intended to be read together with the Case Stated Decision. At p 56, para 3. [47] [1983] 1 HKC 14. Lords Fraser, Roskill, Brightman, Templeman and Sir John Megaw. [48] P 19. [49] Lam VP, Chu and McWalters JJA. [50] B140. [51] B143. [52] An authorized person is defined in s 2 and includes, amongst others, an architect. [53] The circular letter also mentioned a time limit for the validity period of modification/exemption under s 42 of the Ordinance, which has no direct relevance. [54] B140, BA’s letter to REDA, 9 December 2010. [55] Para 13 Case Stated Decision. [56] A fellow of the Hong Kong Institute of Surveyors and a fellow of the Royal Institution of Chartered Surveyors who has made an affirmation in support of REDA’s application. [57] (2013) 16 HKCFAR 351 at 361, para 15. [58] Specified street is defined in BPR 18A(3) and the details do not concern us. [59] Regulation 2 defines site coverage as: “… the area of the site that is covered by the building that is erected thereon and, when used in relation to a part of a composite building, means the area of the site on which the building is erected that is covered by that part of the building.” [60] Regulation 21(3) provides that “the plot ratio of a building shall be obtained by dividing the gross floor area of the building by the area of the site on which the building is erected.” [61] BPR 19(1) and (2). [62] First Schedule BPR. [63] BPR 19(3). [64] See para 79 below. [65] Para 52 above. [66] BAR 30(3). [67] It is to be noted that in the passage from Cheng Yick Chi referred to in para 46 above, there is a reference to “any land which the developer bona fide proposes to include in the development.” [68] Declaration (c). [69] Lords Keith, Roskill, Templeman, Ackner and Oliver of Aylmerton. [70] With Lords Keith of Kinkel, Brandon of Oakbrook, Goff of Chieveley and Jauncey of Tullichettle. [71] Lords Keith, Mustill, Woolf, Lloyd of Berwick and Nicholls of Birkenhead. [72] [2013] 4 HKLRD 74. [73] [2013] 4 HKLRD 52. [74] [1983] 1 HKC 14 at 19. 
Press Summary (English) Press Summary (Chinese) FACV No. 9 of 2012 IN THE COURT OF FINAL APPEAL OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION FINAL APPEAL NO. 9 OF 2012 (CIVIL) (ON APPEAL from CACV NO. 55 OF 2011) BETWEEN Chief Justice Ma: A INTRODUCTION 1. The only issue for determination in this appeal was essentially a factual one, which I shall presently identify. The appeal came to the Court under s 22(1)(a) of the Hong Kong Court of Final Appeal Ordinance[1], the “as of right” provision. After hearing submissions from counsel for the Appellant[2], we dismissed the appeal with costs, our Reasons to be handed down on a subsequent date. 2. In the underlying action, the Appellant (I shall refer to him also as the Plaintiff) claimed both the sum of $44.5 million (and interest thereon) as against the Respondent[3] (as 1st Defendant) and also damages as against an individual, Mr Chan Sheung Wai (as 2nd Defendant). After a 14 day trial, Saunders J[4] held in favour of the Plaintiff as against RHL but dismissed the claim against Mr Chan. RHL appealed to the Court of Appeal[5] and the appeal was allowed.[6] Mr Chan took no part in the appeals to the Court of Appeal and to this Court. 3. Before identifying the relevant issue for determination, I should first briefly set out the factual background. 4. The Plaintiff was the co-founder of a group of companies loosely known as the Rockapetta Group. The Group was in the business of the design, manufacture, sale and distribution of children’s toys. RHL was incorporated in Bermuda in 1994 with a view to a listing on the Hong Kong Stock Exchange; it was so listed and this remains the position of the Respondent.[7] RHL was the holding company in the Group, holding the shares in Group companies such as Rockapetta Industrial Company Limited (“RICL”)[8], Grand Extend Investments Limited (“GEIL”)[9] and other companies. Being the majority shareholder in RHL, the Plaintiff was its chairman and was also an executive director of the company until he finally disposed of his shareholding in early 2001. Save as to a number of aspects to which I shall return, it is not important for present purposes to go into the details of the sale of the Plaintiff’s shares in RHL. It is sufficient to refer only to the fact that the sale by the plaintiff was to Mr Chan through their respective corporate vehicles. Before the sale of the shares in 2001, RHL increased its share capital, and agreements were entered into for the placing and subscription of shares involving the Plaintiff through his corporate vehicle. 5. Between May 1995 and March 1998, the Plaintiff made several advances totalling $83 million to Group companies: $23 million was advanced in 1995, $34 million in 1996, $21 million in 1997 and $5 million in 1998. The Plaintiff was repaid $38.5 million, leaving a balance of $44.5 million and this was the sum claimed by the Plaintiff in the present proceedings. The issue at trial, and the issue that has remained throughout in the appeal to the Court of Appeal and in the present appeal, was simply: who was or were liable for the repayment of the loans? This question was, as observed at the outset, essentially a factual one. B THE PLAINTIFF’S CASE ON THE ISSUE AND ITS RESOLUTION IN THE COURTS BELOW 6. In the Re-Re-Re-Amended Statement of Claim (“the Statement of Claim”), the Plaintiff pleaded the loans in the following way:- “5. Between March 1995 and March 1998, the Plaintiff advanced to the [Rockapetta] Group a total sum of HK$83 million (as to HK$69 million advanced to or through RICL and as to HK$14 million to or through GEIL (collectively, “the Loans”)) for the purpose of repaying the bank facilities granted by various banks to companies within the Group and the general working capital requirements of the Group. At all material times, funds and other income of the Group were treated as a common pool of funds for the whole Group. Therefore, although each company of the Group maintained its own bank accounts, funds of the Group were transferred freely from one bank account of one company to another within the Group as and when they were required, or a company in the Group would make payments on behalf of other companies in the Group when required (as the case may be). 6. By two letters of agreement both dated 31 March 1999 from the Plaintiff to RICL and GEIL and countersigned and agreed by RICL and GEIL respectively, it was agreed, inter alia, that the Loan shall not be repayable by 1st April 2000 and interest shall accrue thereon at the prime lending rate for Hong Kong Dollars quoted by the Hongkong and Shanghai Banking Corporation Limited from time to time (“the prime lending rate”) from 1st April 1999 up to and including the date of full repayment of the Loans.” 7. The claim for the balance $44.5 million made against RHL was pleaded in the Statement of Claim under two alleged agreements:- (1) In March 2000, it was agreed between the Plaintiff and the directors of RHL that in consideration of the Plaintiff not making demands for the immediate repayment of the loans and waiving his entitlement to some interest, RHL would repay the loans together with interest, and for such purpose would undertake a series of fund raising exercises.[10] This agreement was referred to in the pleading as “the Initial Agreement”. (2) The second alleged agreement was pleaded to have been made on 19 June 2000 whereby in consideration of the Plaintiff again not demanding immediate repayment of the loans and interest thereon, RHL would repay the outstanding balance of the loans and interest thereon, doing so from the price of goods sold or to be sold by the Group.[11] This agreement, referred to as “the Subsequent Agreement”, was said to be supplemental to the Initial Agreement. 8. The breaches pleaded by the Plaintiff against RHL referred only to the Initial Agreement and the Subsequent Agreement[12] and no other. Certainly, no claim was made against RHL on the basis of any pre-existing liability on its part. As mentioned above, claims were also made against Mr Chan, the 2nd defendant in the proceedings, on the basis of fraudulent misrepresentation but these are of no relevance for present purposes. 9. In his judgment, the trial judge dismissed the claims made by the Plaintiff under the Initial Agreement and the Subsequent Agreement[13] and that ought to have been the end of the matter as no other claim was pleaded against RHL. However, for reasons which are not entirely clear, the learned judge found that liability against RHL was established on the basis of an oral agreement said to have been made between the Plaintiff on the one hand and one Leung Yuk Lun[14] on the other to the effect that while funds advanced by the Plaintiff could be used by subsidiary companies within the Group, nevertheless the holding company (that is, RHL) would be responsible for the repayment, the loans being treated to have been advanced to RHL itself.[15] The basis for this finding was certainly not pleaded by the Plaintiff nor were we referred to the evidential basis for this finding either. 10. It therefore came as no surprise that RHL appealed to the Court of Appeal. 11. In the judgment of Kwan JA, with which the other judges agreed, this finding of the trial judge was reversed. After noting that this factual basis was not pleaded, Kwan JA went on to say[16] that there was simply no evidence of such an agreement having been reached and it was recorded that leading counsel for the Plaintiff could not point to any such evidence either. 12. The Plaintiff however adopted a fallback position in the Court of Appeal. It was argued that the liability of RHL to the Plaintiff was a joint and several liability assumed by that company together with the other companies in the Rockapetta Group. As far as I can see, this was the first time in the history of the action that a case based on joint and several liability had been raised by the Plaintiff. The factual basis for this cause of action appeared to be merely that monies had been advanced by the Plaintiff to the whole Group without specification as to which entity should receive any particular tranche, it being left to the Group’s accountants to deposit monies into whichever company needed funds. According to the Plaintiff’s testimony, all the companies in the Group had access to the loans extended by him. One part of the Plaintiff’s evidence, which it was said indicated his intention, came during his cross examination[17]:- “A. Well, but to me all the companies has equal access to use my money. Likewise all the companies bear the responsibility of repaying me the loans because every company use my money. Therefore I did not bother with which accounts of whichcompanies was the money deposit. If I did not deposit the money into the company, the company itself would have been closed down or liquidated much earlier on.” 13. The Court of Appeal also rejected the Plaintiff’s case made on this basis. The following extract from the judgment of Kwan JA made this clear:- “48. I would agree with Mr Chan. Lending to nine companies with each of them made jointly and severally liable for repayment of the loan to the external creditor is unusual, to say the least. Although companies in a group may pool their resources and internal lending may take place, and a loan obtained from an external creditor may be on-lent by a company to another company within the group, this does not provide a legal basis that vis-à-vis the external creditor, all the companies in the group should assume liability for the external loan. Mr Ramanathan changed his position in his oral submission and sought to argue that in a lending to a group of companies, not all the companies would be made liable but only the holding company and whichever company in the group that had the use of the funds. That would depend on the loan agreement that was actually made, and this is not a case pleaded by the plaintiff.” 14. The judgment of Kwan JA analysed the evidence at some length to reject the submission that the loans of $83 million were made to the Group companies as a whole and that liability would somehow attach to individual companies within the Group, in particular RHL (whether on the basis of some joint or several liability, or on some other alleged basis). The appeal was accordingly allowed and the Plaintiff’s action against RHL was ordered to be dismissed. C RESOLUTION OF THE APPEAL BEFORE THIS COURT 15. As mentioned earlier, the appeal to this Court did not require leave, it being “as of right”.[18] And as also mentioned earlier, the issue raised was really one of fact. If leave to appeal had been required under s 22(1)(b) of the Hong Kong Court of Final Appeal Ordinance, there is no doubt that such leave would have been refused. I shall say something more about this at the end of my judgment. 16. In the appeal before us, the Plaintiff all but abandoned the basis on which the trial judge had found in his favour.[19] Instead, he relied on the case against the Respondent being founded exclusively on the joint and several liability that it was said the companies in the Rockapetta Group (including RHL) had assumed regarding the loans made by the Plaintiff. 17. Valiantly argued though the appeal was by leading counsel for the Plaintiff, the appeal was a hopeless one. This is for the reasons which are dealt with under the following headings. C.1 Pleadings 18. The Statement of Claim, even though it had undergone four amendments, did not plead a case against RHL based on a joint and several liability that had been agreed by all the companies in the Group. The only agreements that had been pleaded in relation to the loans were those made by RICL and GEIL (see para 6 of the pleading[20]) and, as far as RHL was concerned, the Initial Agreement and the Subsequent Agreement.[21] 19. When pressed, Mr Westbrook SC referred to para 5 of the Statement of Claim[22] as pleading a case based on joint and several liability. It did no such thing. In support of his submission here, counsel stated that only material facts, rather than law, must be pleaded. This is of course correct but an agreement to establish a case on the basis of joint and several liability must – like any other agreement – be properly pleaded: when and how the agreement was made, its material terms, the consideration therefor, breach, the sum due thereunder, and where appropriate, damages. It is right to point out that in other parts of the Statement of Claim, the Plaintiff did adequately plead a case based on agreements having been made (viz. the Initial Agreement and the Subsequent Agreement), and in that part of the pleading dealing with the Initial Agreement, there was specific mention of the term “joint and several liability”. However, nothing in the rest of the Statement of Claim was even remotely sufficient to raise the case the Plaintiff sought to argue before this Court. 20. Mr Westbrook urged upon us that para 5 of the Statement of Claim led inevitably[23] to the legal conclusion that the companies in the Group were each jointly and severally liable to the Plaintiff for the loans. This submission cannot be accepted and no authority was provided to support it. Quite apart from the necessity to make one’s case clear on the pleadings, the mere fact that loans are made for the benefit of a group of companies, with monies being freely transferable within the group and payments made to any company as and when required, did not without more make the liability of the companies a joint and several one, or even a joint one. Where, as seems to have occurred in the present case, it was left to the Group’s accountants to book the loans, the more likely conclusion was that the companies to which any particular amount of money was specifically directed, would be the entity legally liable to repay the loan as far as the Plaintiff (the lender) was concerned. And how a group of companies chose, as between themselves, to treat monies lent to the group would generally be of no concern to the lender in the absence of specific agreement. 21. It should by now really be quite unnecessary to issue yet another reminder on the rationale behind pleadings. The basic objective is fairly and precisely to inform the other party or parties in the litigation of the stance of the pleading party (in other words, that party’s case) so that proper preparation is made possible, and to ensure that time and effort are not expended unnecessarily on other issues:- Wing Hang Bank Limited v Crystal Jet International Limited[24]. It is the pleadings that will define the issues in a trial and dictate the course of proceedings both before and at trial. Where witnesses are involved, it will be the pleaded issues that define the scope of the evidence, and not the other way round. In other words, it will not be acceptable for unpleaded issues to be raised out of the evidence which is to be or has been adduced. As the Court of Appeal remarked in Wing Hang Bank Limited v Crystal Jet International Limited[25]:- “(2) In a trial, particularly where evidence is given by witnesses, it becomes extremely important that each side knows exactly what are the live issues. Where issues are sought to be introduced that have not been adequately or properly pleaded, amendments must be sought unless the consent of the other party or parties has been obtained. It will simply not do for unpleaded issues to be ‘slipped in’ when evidence is being given in the hope that the other side is not sufficiently alert to object.” 22. In the present case, much reliance was placed on the Plaintiff’s testimony given in cross examination to evidence the assertion made before us to the effect that the Rockapetta Group companies accepted a joint and several liability in respect of the Plaintiff’s loans.[26] It is simply not permissible for an issue to be raised in this way: one does not sift through the evidence adduced in a trial in the hope that something was said that can conceivably found a cause of action. Issues, I would reiterate, must be properly pleaded unless for some reason the pleadings have assumed a less significant role in the proceedings. I would perhaps also observe here that in any event the said passage from the Plaintiff’s testimony at best indicates only a subjective intention on his part. This is not admissible to establish the terms (or even the existence) of an agreement. 23. The purpose of pleadings, in clearly and unambiguously setting out the true extent and nature of a dispute not just for the benefit of the parties but also for the Court in managing and trying cases, remains important under our system of civil justice. The retention of the old rules as to pleading as well as the introduction of new provisions over four years ago under the Civil Justice Reform, reinforce this. 24. One of the new provisions introduced under the Reform was RHC o.18 r.12A which reads: “ A party may in any pleading make an allegation of fact which is inconsistent with another allegation of fact in the same pleading if– (a) the party has reasonable grounds for so doing; and (b) the allegations are made in the alternative.” If the Plaintiff had, as he ought to have done, sought leave to plead a case on joint and several liability, questions would inevitably have been raised as to whether the Plaintiff could in the first place properly plead such a cause of action, which would have been (at least at first blush) quite inconsistent with his pleaded case based on the Initial Agreement and the Subsequent Agreement. The cause of action based on these Agreements assumed that there was no pre-existing liability as far as RHL was concerned, hence the need to enter into agreements which made RHL liable for the repayment of the loans to the Plaintiff. In these circumstances, it would therefore have been quite difficult to see how a pre-existing liability jointly and severally assumed with others, fitted in. Before us, the Plaintiff did not provide any enlightenment on this. To me, this only served to highlight even more the insuperable weaknesses in the Plaintiff’s case made on this new basis. 25. Another new provision is the important RHC o 41A dealing with statements of truth. If, in the present case, that provision had applied to any amendment which the Plaintiff might have sought to make in order to plead a case of joint and several liability, he would have been required to verify the amendment by a statement of truth.[27] Given this requirement, the rule would have posed yet another difficulty for the Plaintiff alongside o 18 r12A. Both these new provisions emphasise the continuing important role of pleadings in any modern system of civil justice. 26. This appeal is the second case in which this Court has recently had occasion to comment on pleadings (or rather, the lack thereof). In Sinoearn International Limited v Hyundai-CCECC Joint Venture,[28] Ribeiro PJ referred to the “extraordinary” position of a party being permitted to run at trial an unpleaded case. I associate myself with the sentiments expressed in that judgment. As stated by Ribeiro PJ[29]:- “34. Pleadings are not mere formalities. They impose a necessary discipline and are fundamental to enabling every procedural facet of the adversarial system to operate fairly.” 27. The failure properly to raise a cause of action based on joint and several liability would have been enough to dispose of this appeal. For the sake of completeness, however, I should also mention that the case was all the more hopeless when one examined the evidence. C.2 The evidence 28. The documents at trial overwhelmingly pointed to the fact that the loans made by the Plaintiff were to RICL and GEIL, and it was these two companies which became primarily liable towards the Plaintiff in relation thereto. There was no document which indicated otherwise; certainly no document pointed towards joint and several liability. 29. In the judgment of Kwan JA, reference was made to the annual reports of RHL, RICL and GEIL (for the relevant years when the loans were made) which all stated that the loans were made to RICL and GEIL.[30] These were signed by the Plaintiff, as indeed were written resolutions[31] of the Board of Directors of both these two companies which stated expressly that loans were made by the Plaintiff to RICL and GEIL. There were also letters passing between the Plaintiff and the two companies from 1999 to 2001 which again evidenced this.[32] 30. Mr Westbrook SC contended that some documents supported the Plaintiff’s case on joint and severally liability. With respect, they did no such thing:- (1) First, three of the 20 cheques issued by or on behalf of the Plaintiff which made up the $83 million loans were drawn in favour of “Rockapetta” as the payee. It was said that this evidenced a payment to the Group. It is curious that the trial judge treated this piece of evidence as indicating that payments were made to RHL itself, but the Plaintiff does not seek to uphold this finding and was content merely to say that it indicated payments made to the Group as a whole. Even if this was the effect of these three cheques (which I doubt), it goes nowhere near supporting a case based on joint and several liability. (2) Secondly, reliance was also placed on the annual reports of RHL for the years 1999 to 2001 in which was stated that the interest payable on the loans was payable by the Group. Whatever else this statement may have indicated, it could not in my view evidence a joint and several liability assumed by each Group company towards the Plaintiff in relation to the loans. In any event, these reports had to be seen against the other annual reports of RHL, to which reference has already been made.[33] Such reports referred unequivocally to the fact that the loans were made to subsidiary companies. (3) Thirdly, the Plaintiff relied on certain repayments of the loans[34] having been made by RHL as pointing towards an acceptance by RHL that it was liable to the Plaintiff for repayment. However, when seen against the other instances of repayment (when cheques were issued to the Plaintiff by RICL and another subsidiary company), the evidence pointed more to the financial workings of a group of companies[35] than any support for the joint and several liability case run by the Plaintiff. It is also important in this context to note that in his witness statement for the trial,[36] he referred to the repayments of $38.5 million as being payments made pursuant to the Initial Agreement and the Subsequent Agreement (rather than pursuant to any pre-existing liability). It will be recalled that the case based on these two agreements was rejected by the trial judge. (4) Lastly, reliance was placed on a draft deed in March 2000 to be made between the Plaintiff, RHL, RICL and GEIL in which (in the Recitals) it was stated that loans of $83 million had been advanced to RHL through RICL and GEIL. The trial judge placed reliance on this draft deed in his judgment. The Court of Appeal was right in holding that not much could be inferred from an unexecuted deed. It was also inconsistent with the case now advanced by the Plaintiff inasmuch as the only liability referred to in the draft deed was that of RHL, and not of the Group companies as a whole on the basis of joint and several liability. 31. A general submission made by Mr Westbrook SC regarding the documents which were against the Plaintiff’s case was that they somehow only represented the way the Group’s accountants chose to treat the loans. The clear suggestion implicit in this submission was that these documents did not reflect reality. It was also submitted that insofar as the documents made reference to RICL and GEIL being liable to the Plaintiff for the repayment of the loans, this was correct since they were liable under (according to the Plaintiff) a joint and several liability together with other Group companies (including RHL). 32. These submissions might conceivably have been arguable if there was evidence supporting this factual scenario. However, no evidence was adduced by the Plaintiff to this effect. I have already made reference to the Plaintiff’s witness statement. In para 32 of that statement, the Plaintiff dealt with the background to the Initial Agreement and why, according to him, it was in RHL’s interest to enter into it:- “32. I verily believe that the Initial Agreement was and would be inthe interests of the Defendant for the following reasons:- (1) Without the Defendant's agreement to repay theLoans and interest thereon to me, I could immediately after 1st April 2000 demand RICL and GEIL respectively to forthwith make repayment in full of the Loansand interest thereon to me, and would no doubt insist upon such repayment in view of my intended disposal of my interests in the Defendant. (2) At that time, the Group did not have enough available funds to repay the Loans and interest thereon in full to me. According to the Consolidated Balance Sheet of the Group as at 31" December 1999 (containedin the Annual Report of the Defendant covering the period from 1st April 1999 to31st December 1999 [Item 2 of the Plaintiff’s List of Documents], the amount of "Bank balances and cash" was only HK$17,237,000.00. If RICL and GEILwere demanded tomake immediaterepayment in full of` the Loans and interestthereon,then, even if the available funds of the whole of the Group were utilised for such purpose, it was highly likely that they would default in such repayment or would at least have great difficulties in doing so. (3) Since the core business and assets of the Group were held by RICL, the default on the part ofRICL (and the possible winding up of RICL consequent thereon) would no doubthave grave financial impact on the whole of the Group, including the Defendant. The listing status of the Defendant might well be affected. Allthese would have disastrous effectonthe Defendant.” 33. Far from assuming that RHL was already liable to him for the repayment of the loans, on the basis whether of joint and several liability or any liability, the Plaintiff was stating that the liability lay with RICL and GEIL. 34. This stance was entirely consistent (as we have seen) with the documents and also consistent with his conduct. I need only refer to one aspect of the Plaintiff’s conduct. The present litigation was not the only set of proceedings instituted by the Plaintiff which touched on the question of the repayment of the loans. In 2006, he issued writs against his financial consultant[37] and his solicitors[38], as well as the purchasers of his shares in RHL,[39] all on the basis that loans had been advanced by him to Group companies (which, in context, could only mean RICL and GEIL). Further, this was the position he adopted when he was sued by RHL in 2002[40]. He also petitioned for the winding up of RICL[41] on the basis of the failure to repay the loans made by him to that company (not on the basis that the loan was to someone else and that RICL was jointly and severally liable). D CONCLUSION 35. For these reasons, the appeal was dismissed. E POSTSCRIPT 36. I have already mentioned that the Plaintiff’s appeal to this Court was as of right under s 22(1)(a) of the Hong Kong Court of Final Appeal Ordinance. This is yet another and all too common example of an unmeritorious appeal which has reached the Court of Final Appeal by this route. Permanent Judges of this Court have frequently remarked on s 22(1)(a) of the Ordinance as being anachronistic and, in a case such as the present, giving rise to unfairness to respondents in an appeal: see the remarks of Ribeiro PJ in China Field Limited v Appeal Tribunal (Buildings) (No. 1)[42] and Bokhary PJ in Wealth Duke Limited v Bank of China (Hong Kong) Limited.[43] It is also to be noted that in the latter case, Bokhary PJ observed that Hong Kong ought to be in line with other common law jurisdictions by abolishing appeals as of right; see also similar observations made by Ribeiro PJ in Chinachem Charitable Foundation Limited v Chan Chun Chuen.[44] 37. In Champion Concord Limited v Lau Koon Foo (No. 2),[45] I referred to the fact that the experience of this Court has been that the vast majority of appeals brought by this route have been “a drain on resources, waste time and hinder in a very tangible way the resolution of other, far more meritorious proceedings”. Indeed they are and the present case demonstrates this point. 38. As we have seen, the present appeal was devoid of merit, and if leave had been required, there was no prospect of such leave being granted. Yet, it was listed before the Court for a two day hearing. Part B of the Record (the documents bundle) occupied 600 pages. When the parties were first approached in relation to the bundles to be used in the appeal, the Plaintiff (the Appellant) submitted 15 lever arch files comprising over 3,700 pages of documents. Following a series of paper directions and one oral hearing before the Registrar of the Court, the Court eventually agreed to the two Part B files used in the appeal. The Respondent eventually filed another bundle of documents for use in the appeal (comprising over 150 pages). Both the Appellant and the Respondent lodged Cases and Supplemental Cases (consisting of 34 pages of submissions on the Appellant’s part and altogether 46 pages on the Respondent’s). 35 authorities were referred to in the parties’ Cases and produced to the Court. 39. Cases such as the present appeal are inherently wasteful of resources which the Courts have in limited supply. Following the Civil Justice Reform introduced in 2009, this kind of wastage ought no longer be tolerated by the public. It is unfair to the successful parties in a litigation, it is unfair on other litigants who have deserving cases before the Courts and it is ultimately unfair to the community. Mr Justice Chan PJ: 40. I agree with the judgment of the Chief Justice and the observations made by Mr Justice Mortimer NPJ. Mr Justice Ribeiro PJ: 41. I agree with the reasons for judgment of the Chief Justice. Mr Justice Mortimer NPJ: 42. I agree with the reasons given by the Chief Justice for the dismissal of this appeal with costs. 43. I would however add my voice to his comments concerning unmeritorious appeals brought under the ‘as of right’ provision of s.22 (1)(a) of the Hong Kong Court of Final Appeal Ordinance. 44. Whatever good historical reasons there may have been for this provision, it no longer has any validity or proper purpose. This has led other common law jurisdictions to abolish similar provisions because hopeless appeals brought as of right result in injustice and unfairness. 45. It is unjust to the successful respondent whose remedy is delayed and who may not recover all his costs; it is of no benefit to the losing appellant who will fail and then pay all the costs; it is unfair to waiting litigants who suffer unnecessary delay in having their disputes resolved; it wastes the time of the court and the judges; and it is an unnecessary additional burden on the public purse. Sir Thomas Gault NPJ: 46. I agree with the reasons for judgment given by the Chief Justice. Mr Simon Westbrook SC and Ms Rachel Lam instructed by Gall for the Appellant Mr Edward Chan SC and Mr C Y Li SC instructed by Tso Au Yim & Yeung for the Respondent [1] Cap 484. [2] Mr Simon Westbrook SC, with him Ms Rachel Lam. The Respondent, 21 Holdings Limited, was represented by Mr Edward Chan SC and Mr CY Li SC. [3] Formerly known as Rockapetta Holdings Limited. I shall use the abbreviation RHL interchangeably with the Respondent in this judgment. [4] In a judgment handed down on the 2 March 2011. [5] Tang VP, Kwan JA and Yam J. [6] The appeal was allowed on 9 December 2011 after a 2 day hearing; the Reasons for judgment were handed down on 20 December 2011. [7] The company is now known as 21 Holdings Limited. [8] Which was the distribution and sales arm of the Group. It was described by the Plaintiff as the “major operating arm of the Group”, until it was wound up in October 2002. [9] Which at one stage held some property used by the Group companies. [10] Para 8A of the Statement of Claim. [11] Para 13 of the Statement of Claim. [12] Paras 15 to 17 of the Statement of Claim. [13] The Plaintiff at no stage appealed these findings. [14] The Group’s Financial Officer. [15] Paras 28, 46 and 48 of the judgment dated 2 March 2011. [16] Para 58 of the judgment of the Court of Appeal dated 20 November 2011. [17] This passage was referred to in para 51 of the judgment of the Court of Appeal. [18] The Plaintiff’s claim was said to be for a liquidated sum, being in the nature of a debt owed to it by RHL. [19] See para 9 above. [20] Set out in para 6 above. [21] See para 7 above. [22] Set out in para 6 above. [23] The expression he used was “as night followed day”. [24] [2005] 2HKC 638, at 643G-H (para 6(1)). [25] At 643H-I (para 6(2)). [26] See the extract set out in para 12 above. [27] RHC o.20 r.13. [28] FACV 22 of 2012. [29] In para 34 of the judgment. [30] The annual reports of RHL for the years 1996-1998 made reference to loans amounting to $83 million being granted “to a subsidiary” or “subsidiaries of [RHL]”. [31] Dated 24 August 1999. [32] For example , in letters dated 31 March 1999 from the Plaintiff to RICL and GEIL, reference was made to loans of, respectively, $69 million and $14 million from the Plaintiff to the companies, and these letters further stated that the loans were unsecured and repayable by the companies on 1 April 2000. The companies were asked to confirm this loan arrangement, and this was done by signatures to this effect. [33] Para 29 above. [34] The total amount of the repayments was $38.5 million, leaving a balance of $44.5 million. [35] See para 20 above. [36] This was treated as part of the Plaintiff’s evidence in chief. [37] HCA No. 567 of 2006. [38] HCA No. 568 of 2006. [39] HCA Nos 569 and 570 of 2006. [40] HCA No. 3296 of 2002. [41] HCCW 743 of 2002. [42] (2009) 12 HKCFAR 68, at 77C-F (para 16). [43] (2011) 14 HKCFAR 863, at 865 (para 1). [44] (2011) 14 HKCFAR 798, at 835-6 (para 107). [45] (2011) 14 HKCFAR 837, at 845 (para 6). 
Chief Justice Ma: 1. I agree with the Reasons for Judgment of Mr Justice Fok PJ. Mr Justice Ribeiro PJ: 2. I agree with the Reasons for Judgment of Mr Justice Fok PJ. Mr Justice Tang PJ: 3. I agree with the Reasons for Judgment of Mr Justice Fok PJ. Mr Justice Fok PJ: 4. This appeal, for which leave to appeal was granted on the substantial and grave injustice ground, concerns the fairness of the appellant’s trial in the particular circumstances of this case. However, the appeal involves important norms concerning the need for the trial judge’s summing-up to the jury to be fair and for the judge’s comments in that summing-up to truly and accurately reflect the evidence. This is all the more important in a jurisdiction where evidence from one language is often translated into another language in which the trial is conducted. 5. At the conclusion of the appeal, the Court allowed the appeal, quashed the appellant’s conviction and ordered a re-trial, remanding the appellant in custody pending such re-trial. The Court indicated that it would provide its reasons for doing so in due course. These are my reasons for the orders made. The facts 6. The appellant was convicted after trial before Beeson J and a jury[1] of unlawfully trafficking in a dangerous drug, [2] namely 466.09 grammes of a solid containing 389.58 grammes of cocaine, and sentenced to 13 years’ imprisonment. 7. The facts of the case were quite straightforward. The appellant was stopped by police in Yau Tong walking up and down beside a car and was approached by the police. He was found in possession of a box containing cocaine and also in possession of the keys of the car in which a further quantity of cocaine was found. 8. The only substantial issue was the appellant’s knowledge of the contents of the box and car. The prosecution case was that it was to be inferred that the appellant knew of the presence of the dangerous drugs in both. The defence case, on the other hand, was that he did not know what the box contained or of the presence of the drugs in the car. 9. The appellant did not give evidence at trial but relied on the content of his video recorded interview (“VRI”) in Cantonese. That was critical to his defence as, in it, he sought to explain the reason why he was present at the scene. His explanation was that he had previously delivered his car to someone called “Hak Chai”, who was a car repairman, to provide a quote for its repair and he had arranged to meet Hak Chai in Yau Tong to pay him. When he arrived, Hak Chai said he had to go to the lavatory and so asked the appellant to wait by the car for him. He left the car key with the appellant in case he had to move it to avoid being ticketed. He also left a box saying that there was something valuable inside it and that someone would come and collect it. He asked the appellant to look after the box in his absence. The appellant agreed, since he still wanted Hak Chai to take the car away for repairs. Hak Chai had previously done this twice before and the appellant assumed the parcels contained spare motor parts. 10. The appellant’s defence therefore hinged on the credibility of his explanation in the VRI. The jury’s verdict, convicting the appellant, was unanimous and so it is apparent that they did not believe the appellant’s defence. The mistranslated portions of the VRI 11. In the course of her summing-up to the jury, the Judge commented on the appellant’s defence as contained in the VRI. Unfortunately, and without any fault on the part of the Judge or on the part of trial counsel (both of whom were non-Cantonese speakers), she proceeded to do so on the basis of three mistranslations in the English transcript of the VRI. 12. The mistranslated answers were the answers identified as counters 74, 310 and 314 in the VRI. As originally, and incorrectly, translated, they were as follows: (1) Counter 74: “On that – that day, I reached there, I – I intended to – I reached there on that day, I paid him money for repairing (the car) for me, well, he left – left me an ignition key, he said, (‘)Hey, you look after the car first, because if a moment later, the car is to be moved, if (you are going) to be ticketed, you can drive (it) away.(’) Well, so I received the ignition key, well, I stood outside, outside the car and waited. I had no idea that there were such things on the car – car.” (Emphasis added) (2) Counter 310: “Because in the past there – there wasn’t any problem all along, there wasn’t any problem all along. Well, anyway – anyway, that is, (we) knew – knew each other, like that, well, therefore (I) handed over to him. He – that is, before I saw him, it was okay, (he) had a sense of responsibility, like that, and this time, when he quoted a price for me, he immediately gave me a reply on the price within one (or) two days or so, well, slightly over two thousand bucks for fixing (it), in any event, he fixed (it) for me, well, I – I ….” (Emphasis added) (3) Counter 314: “Well, that was why I had the car fixed, well, I – I arranged with him to meet – meet in Yau Tong, at my mother’s, I just arrived there, well, he – he was already there, waited for me there.” (Emphasis added) 13. Instead, the correct certified translation of those answers was: (1) Counter 74: “On that, that day, I went down there. I, I intended to, on that day, I reached there. My (purpose) was to bring him the money for the repair work for me. Well, he put down, also put down and gave me the car key. He said, ‘Hey, you watch the car first because if a while later, the car is to be moved (or) if (you’re to be) ticketed, you can drive (it) away.’ Well, so I took over the car key. Well, I stood outside, outside the car to wait. I had no idea that there was such thing in the car, the car.” (Emphasis added) (2) Counter 310: “Because all along, all along there hadn’t been any problem, all along there hadn’t been any problem. Well, given that, given that, that is, (I) know (him), (we) know each other, well, (I) handed (it) over to him. He, that is, I found that previously he was okay, a (person) with a sense of responsibility, and this time, when he helped me get a quotation, he immediately gave me a reply on the price in just about one or two days. Well, some two thousand dollars for the repair. All in all, he helped me see to it. Well, I, I …” (Emphasis added) (3) Counter 314: “Well, that was why I (would) have the car fixed. Well, I, I arranged to meet him at, at Yau Tong, at my mother’s. I just arrived there. Well, he, he was already there, waiting for me.” (Emphasis added) 14. The significance of these mistranslations in the context of the Judge’s summing-up will be addressed below. For present purposes, it is important to point out that the difference in the translations lay in the fact that, as properly translated, the appellant’s version was that he had earlier handed his car over to Hak Chai to quote for its repair, that Hak Chai had brought his car back to Yau Tong on the night in question to collect the cost of its repair and that the appellant’s car was at that time still awaiting repair. As the Court of Appeal noted, the emphasis was on the fact that the appellant’s car was still awaiting repair.[3] With the benefit of the correct translations (and having watched the VRI during the hearing), the Court of Appeal was in no doubt that the Judge was misled by the mistranslations.[4] The intermediate appeal and leave to appeal to this Court 15. The appellant applied to the Court of Appeal for leave to appeal against his conviction. One of the appellant’s two grounds of appeal against conviction was that, because of these mistranslations, the Judge misunderstood the appellant’s VRI and made unfounded criticisms of his explanation in that interview.[5] 16. Although accepting that the Judge was misled by the mistranslations,[6] the Court of Appeal came to the view, for two reasons (which I will address below) that her analysis of the VRI would not so colour the minds of the jury that their verdict was unsafe.[7] 17. Leave to appeal to this Court was granted by the Appeal Committee on the ground that it was reasonably arguable that the trial judge’s comments in her summing-up to the jury relying on the mistranslated portions of the appellant’s VRI may have resulted in the appellant not receiving a fair trial.[8] The judge’s duty in a summing-up 18. A trial judge is under a duty, when addressing a jury, to summarise the issues and the evidence that relates to them clearly, accurately and fairly. The principle is succinctly stated in HKSAR v Lin Ping Keung (2005) 8 HKCFAR 52 per Chan PJ at [24]: “A trial judge is under a duty to summarize the issues and the evidence relating to these issues clearly, accurately and fairly to the jury. Although the jury are the sole judges of the facts, the trial judge is also entitled to express to the jury his or her views on the evidence, provided that the jury are told in clear and unambiguous terms that those are only the judge’s personal views, that the jury must form their own views on the evidence and that they can ignore the judge’s views if they do not agree with them. However, notwithstanding such direction, it is perhaps understandable that the jury might sometimes attach more weight to the views of the trial judge than those expressed by counsel. It is therefore important that a judge should be very careful in making comments on the evidence, particularly comments adverse to the accused.” 19. As Chan PJ also noted in that case (at [26]-[27]): “26. … what a judge says to the jury is very often the last word they hear before they retire and carries more weight than what defence counsel says in a closing address and what the judge says must truly and accurately reflect the evidence, must give a fair and balanced guidance to the jury … 27. If a trial judge chooses to make adverse comments on the defence, it is important that the facts on which the comments are based are correctly stated”. 20. Lord Hailsham of St Marylebone LC also drew attention to the requirement of accuracy and correctness in a judge’s direction to a jury in R v Lawrence [1982] AC 510 at p.519H, where he said that the direction: “… should also include a succinct but accurate summary of the issues of fact as to which a decision is required, a correct but concise summary of the evidence and arguments on both sides, and a correct statement of the inferences which the jury are entitled to draw from their particular conclusions about the primary facts.” 21. This principle in relation to the summing-up is a facet of the fundamental right of a defendant charged with a criminal offence to a fair trial. As Litton PJ held in Tang Siu Man v HKSAR (No.2): “Every defendant in a criminal trial in the High Court has a fundamental right to a fair trial, conducted in accordance with law. … When it comes to the summing-up, the fundamental rule is that this must be fair and balanced, so that the jury can properly weigh up the true issues.”[9] 22. Self-evidently, where evidence in one language is translated for the benefit of the court or parties, as often happens in this jurisdiction, it is important that the translation is true and accurate. Meaning, sometimes difficult in one language on its own, may become elusive in translation and, in mistranslation, unattainable. Whether a mistranslation leads to unfairness in a trial, however, will necessarily depend on the nature and context of the mistranslation and its importance to the issues in the particular case. The Court of Appeal’s reasons for dismissing the intermediate appeal 23. The Court of Appeal’s first reason for its conclusion that the mistranslated portions of the VRI did not result in any material unfairness was that much of what the Judge said was directed at the generally confused and muddled nature of the appellant’s answers in his VRI. The reasoning, set out in the CA Reasons for Judgment, was as follows: “16. First, we have watched the VRI during the hearing and although the applicant was orally fluent, what he said, when shown through a verbatim record in writing, is not the best example of clear communication. On the contrary, there was much disregard for proper sentence construction and not a few instances of going back and forth in time. At one point, when relating a past incident (counter 70), and whether or not he was conscious of it, the applicant even subverted his otherwise consistent position and pointed to himself and not Hak Chai as the one giving the excuse to leave the car. All of that, plus the direct and other-than-idiomatic rendition that pervades the original certified translation, does give a certain factually questionable quality to the VRI, which was not helped by the fact that some of the more important information contained in it was offered piecemeal and not all in one go. Looking at the matter in the round, we are satisfied that much of what the judge said was directed at that aspect of the VRI, albeit that some of it may have been generated by the said mistranslations. That it was so would have been obvious to the jury who, we have ascertained, had watched the VRI during the trial and had the assistance of the transcripts at the same time.” 24. The Court of Appeal’s second reason for dismissing the appellant’s application was as follows: “17. Second, upon our enquiry, Mr Tze [counsel for the appellant] has fairly accepted that, notwithstanding the anomaly at counter 70 (see previous paragraph), the bulk of what the applicant said in the VRI should be clear to the native Cantonese speaker. It should be clear that, according to the applicant, in the instant case as in the past, Hak Chai would, on occasions of the applicant’s car needing repair, exploit the opportunity, drive it to a certain location and, drumming up an excuse, leave the applicant to watch over it with something ‘valuable’ onboard, the suggestion being that the drugs belonged to Hak Chai and not him. This concession of Mr Tze is important because as this Count has discovered, almost immediately after the jury was retired (12:38 pm), they had asked to review the VRI on disc (12:45 pm). Their note to the judge reads: ‘The juror team would like to take a look of (sic) the video file no. KTRN 12017847 case no. HCCC 351/2012 again!’ This is without a doubt the best proof that, whatever view the judge may have expressed about the VRI, the jury had gone to the primary material to determine for themselves what to make of the applicant’s account.” 25. In this Court, the appellant contended that the Judge’s reliance on the mistranslations and her adverse comments based on them were such as to render his trial unfair. Counsel for the appellant, Mr James Tze, essentially maintained the same arguments advanced below in support of the appeal. For its part, counsel for the respondent, Mr Eddie Sean,[10] invited us to uphold the Court of Appeal’s reasons for rejecting the appellant’s ground of appeal and to dismiss this appeal. An analysis of the Court of Appeal’s reasoning 26. As regards the Court of Appeal’s first reason for dismissing the appellant’s application, there is some justification for its view that the VRI was “not the best example of clear communication”. It may be a fair comment that some of what the Judge said was directed at the ungrammatical and sometimes opaque nature of the appellant’s answers in his VRI but, for the following reasons, this is not a complete answer to the appellant’s complaint. 27. In her summing-up, at the outset, the Judge outlined the prosecution case and pointed out that the key issue was knowledge. After directing them on the burden and standard of proof, she quite properly directed them that they should acquit if what the appellant said in the VRI was or might be true. She then directed them on the elements of the offence and on the drawing of inferences. She reminded them of the oral evidence of the two prosecution witnesses and then proceeded to address the appellant’s VRI. 28. At that point in the summing-up, the Judge told the jury that the VRI consisted of incriminating parts and also excuses or explanations and she directed them to consider the whole of the statement when deciding where the truth lay. She then proceeded to go through the VRI and to point out areas that the jury might like to give their attention to. Before doing so, she said this: “I should point out that throughout the Video Interview the defendant gives slightly different answers at various stages to the same question. It is not entirely clear whether he was waiting to have the car assessed, in other words the quote was to be given, that he was to pay for the work to be done and then have it done, or whether it had been arranged before. At one point it appears that he left the car with ‘Hak Chai’ for two days before he came out to pick it up.”[11] 29. With the benefit of the correct translations, it is highly doubtful that the Judge would have made those remarks since the correct translations would not have suggested these apparent chronological difficulties in the appellant’s VRI answers. After all, as was implicit in the Court of Appeal’s acceptance that the Judge was misled by the mistranslations, the VRI was clear as to what the appellant’s account was.[12] 30. Further, in a later part of the summing-up, the Judge referred to the appellant’s answers at counters 310 and 312 and said of counter 312 that this was “another variation of the account of why he was at ‘Hak Chai’s’ place”.[13] That the latter answer in counter 312 was an inconsistent account proceeds on the basis of the mistranslated counter 310: counter 312 stated that the car had not yet been fixed, whereas the mistranslated counter 310 gave the impression that it had already been fixed. Had the correct translation of counter 310 been relied upon, there would have been no inconsistency and so no reason for the Judge to make the adverse comment she did. 31. Given the importance to the appellant that the jury should accept his explanation in the VRI, it was critical to his defence that the jury be directed on the basis of correct translations of his answers and that any comments on those answers by the Judge be accurate and fair. 32. Turning to the Court of Appeal’s second reason for dismissing the appellant’s application, the thrust of the reasoning is that the content of the VRI would have been clear to a native Cantonese speaker and that, since the jury asked to see the VRI again, this meant that the jury “had gone to the primary material to determine for themselves what to make of the applicant’s account”. That account was that, on this occasion as before, Hak Chai would use the request to effect repairs to the appellant’s car to create an opportunity to leave the appellant to watch over it with something valuable inside. The suggestion was therefore that the drugs belonged to Hak Chai and not the appellant, who was ignorant of their presence. That was the critical issue in the case. 33. However, the Court of Appeal’s second reason is problematic. First, it is speculative to assume that all the jurors were native Cantonese speakers and would not have relied on the translation of the VRI since the trial was conducted in English. 34. Secondly, the jury did not raise questions with the Judge regarding the mistranslated portions of the VRI. The reason for their request to review the VRI is not known and it is also speculative to conclude that they must have done so in order to check the accuracy of the Judge’s comments on the appellant’s answers. 35. Thirdly, and more importantly, it devalues the significance of the summing-up in the trial process. The jury is bound, as well as entitled, to take the Judge’s summing-up into account in its deliberations on their verdict. For the most part, juries pay close attention to a judge’s comments in summing-up and those comments, coming as they do from the judge, are likely to carry considerable weight with the jury, albeit that they are free to come to a different view on the facts: see, in this context, the reference to HKSAR v Lin Ping Keung (2005) 8 HKCFAR 52 per Chan PJ at [26] quoted above. To suggest that the effect of the mistranslations and the Judge’s adverse comments would be mitigated by the jury conducting its own analysis of the evidence is to attach too little weight to the summing-up as part of the trial process. As has been held, the repetition of formulaic directions to the effect that the jury are to make their own assessment of the facts and evidence is not sufficient to cure unfair comments on the part of a judge: see, in this context, the judgment of the Court of Appeal in HKSAR v Zhu Jinni [2012] 4 HKLRD 451 per Yeung VP at [37]-[38] and [41]. Whether substantial and grave injustice 36. It is well-established that, for an appeal to succeed on the substantial and grave injustice ground, it must be shown that there has been to the appellant’s disadvantage a departure from accepted norms which departure is so serious as to constitute a substantial and grave injustice: So Yiu Fung v HKSAR (1999) 2 HKCFAR 539. 37. In my view, in the context of this case where there was only one issue and where the contents of the VRI and the appellant’s credibility were critical to that issue, the Judge’s comments on the basis of the mistranslated portions of the VRI resulted in the summing-up being unbalanced and unfair to the appellant. Neither of the reasons given by the Court of Appeal for treating the comments based on the mistranslations as immaterial are, with respect, sufficient to overcome the unfairness to the appellant. In the circumstances, the Judge’s comments in her summing-up to the jury relying on the mistranslated portions of the appellant’s VRI did result in the appellant not receiving a fair trial and there was therefore to his disadvantage a departure from accepted norms amounting to substantial and grave injustice. Conclusion 38. These are my reasons for allowing the appeal and quashing the appellant’s conviction. Given the nature of the case and the reasons for quashing the appellant’s conviction, it was plainly appropriate to order a re-trial, as Mr Tze fairly accepted. Lord Walker of Gestingthorpe NPJ: 39. I agree with the Reasons for Judgment of Mr Justice Fok PJ. Mr James Tze, instructed by Au-Yeung, Chan & Ho, assigned by the Director of Legal Aid, for the Appellant Mr Eddie Sean SADPP and Mr Ivan Cheung SPP (Ag.), of the Department of Justice, for the Respondent [1] In HCCC 351/2012. [2] Contrary to section 4(1)(a) and (3) of the Dangerous Drugs Ordinance (Cap.134). [3] CACC 226/2013 (Lunn VP, Pang JA and M Poon J), Reasons for Judgment dated 14 March 2017 (“CA Reasons for Judgment”), at [8]. [4] CA Reasons for Judgment at [14]. [5] CA Reasons for Judgment at [5]. [6] CA Reasons for Judgment at [14]. [7] CA Reasons for Judgment at [15]. [8] FAMC 22/2017 (Ribeiro, Tang and Fok PJJ), Determination dated 6 February 2018. [9] (1997-98) 1 HKCFAR 107 at pp.120F & 121B. [10] Senior Assistant Director of Public Prosecutions, appearing with Mr Ivan Cheung, Senior Public Prosecutor (Acting). [11] HCCC 351/2012, Transcript of Summing-up, 20 June 2013, pp.13R-14B. [12] See para. [14] above. [13] For counter 312, see CA Reasons for Judgment at [11]. 
Chief Justice Ma : 1. For the reasons contained in the joint judgment of Ribeiro and Fok PJJ, I would also allow the present appeal. Given that Tang PJ is of a different view, I should make some brief observations to explain my concurrence. This judgment should be read after reading the joint judgment. I gratefully adopt the facts as stated in that judgment and also use the same abbreviations. 2. Insider dealing is an insidious activity that is detrimental to the reputation of any major financial centre. Hong Kong is no exception. Under Article 110 of the Basic Law, there is a requirement to safeguard, regulate and supervise Hong Kong’s business and financial markets. 3. The Securities and Futures Ordinance[1] (the SFO) is the principal statute fulfilling the functions stated in the Basic Law. Part XIII of the Ordinance deals with the Market Misconduct Tribunal (the MMT). It is within this part of the SFO where insider dealing is addressed. Division 4 of Part XIII deals specifically with insider dealing in the context of proceedings before the MMT. The present appeal is concerned with proceedings that took place before the MMT. Insider dealing is also a criminal offence; Part XIV of the SFO deals with insider dealing in the context of criminal proceedings. 4. As the joint judgment demonstrates by reference to s 270 of the Ordinance,[2] insider dealing essentially takes place when a person connected with a listed company, such as a shareholder or officer of the company, having what is now known as inside information[3] (this being price sensitive information about the company not known to the investing public – or the market – and which the connected person knows is price sensitive information), deals with the securities of that company. Insider dealing is a form of market misconduct.[4] The MMT has jurisdiction under the Ordinance, in proceedings instituted by the Securities and Futures Commission (the SFC), to determine whether any market misconduct has taken place and if so, to identify the person who has engaged in such market misconduct.[5] In the present case, the Respondents in the appeal (Charles and Marian), together with others, were the implicated persons in MMT proceedings involving a publicly listed company, Asia Telemedia Limited (ATML). Where any person has been identified as having engaged in market misconduct, the MMT has power to make consequential orders, such as an order of payment to the Government of an amount reflecting any profit gained or loss avoided as a result of the market misconduct. 5. As set out in the joint judgment, the composite elements in s 270 of the Ordinance were satisfied in the present case before the MMT. This was, subject to the provisions of s 271 of the Ordinance, enough to constitute a finding of market misconduct; nothing more needed to be shown. However, s 271 sets out a number of defences, among them the defence which was relevant in the present proceedings, s 271(3), the so‑called innocent purpose defence. That defence involves an implicated person (the burden is on that person) having to prove that the purpose(s) in his or her dealing with the relevant securities did not include the purpose of securing or increasing a profit or, correspondingly, avoiding or reducing a loss by using inside information. 6. It is of crucial importance to bear in mind that the relevant time at which to examine both the question of whether the composite elements of s 270 are satisfied and also whether the defence in s 271(3) is applicable, is the time of the dealing in the securities. In the present case, the relevant time were the times when Charles and Marian sold the ATML shares. 7. At those points in time, the MMT was satisfied that Charles and Marian were in possession of price sensitive information and which they knew was price sensitive information.[6] Both knew that if the relevant inside information had been out in the open (as it should have been), the price of ATML shares would be substantially reduced. In other words, they knew that the prices at which they sold the shares were artificially high. 8. The MMT however concluded that a defence under s 271(3) of the Ordinance was made out. The Tribunal was of the view that Charles and Marian’s sole motivating factor was to secure profits from the unexpected speculative boom in the price of ATML shares. Not only that, as far as the inside information was concerned, the Tribunal was of the view that both Charles and Marian thought that the indebtedness owed by ATML to Goodpine would be eventually sorted out behind the scenes and would never enter the public domain – the “behind closed doors” justification. The inside information was therefore treated by Charles and Marian as being irrelevant to their decision to sell the shares at the times they did. 9. Like Ribeiro and Fok PJJ, I am of the view that the Tribunal’s conclusion on the applicability of the s 271(3) defence was an erroneous conclusion. At the material time when the matter had to be examined (that is, when the shares were sold), Charles and Marian knew only too well that the prices at which they managed to sell the shares were artificially high as a result of the inside information not being publicly known. It was with this knowledge that they managed to earn the massive profits they did.[7] In other words, at these points in time, they were seeking to make massive gains (or in the language of s 271(3), securing a profit) knowing that they would be unable to do this if the inside information was made public. In these circumstances, it is impossible not to conclude that at least one of the purposes of their dealing in the shares included making gains by the use or utilization of the inside information they had. Charles and Marian could point to no purpose other than selling the shares at high prices. The fact also they thought that the Goodpine indebtedness would eventually be solved “behind closed doors” in the future was not to the point. In real time terms (that is, at the material points in time they sold the shares), the inside information was far from irrelevant : this information was decisive in terms of allowing them to make the profits they did and they fully appreciated this. In my view, it was this failure to take sufficiently into account the position of Charles and Marian in real time terms that constitutes the error in the conclusion reached by the MMT. Mr Justice Ribeiro PJ and Mr Justice Fok PJ: 10. Insider dealing is an “insidious mischief” which threatens the integrity of financial markets.[8] It is dealt with in Parts XIII and XIV of the Securities and Futures Ordinance,[9] the object of which is “to eliminate insider dealing and to reinforce the transparency of the markets, thereby enhancing and preserving Hong Kong’s position as an international financial centre”.[10] There are a number of defences available to an allegation of insider dealing. This appeal concerns the ambit of the defence provided in section 271(3) of the SFO (set out below). A. The proceedings below 11. By a notice issued by the Securities and Futures Commission (“SFC”),[11] a Market Misconduct Tribunal (“MMT”) was constituted[12] to determine whether any market misconduct had taken place in connection with dealings in the shares of a listed company called Asia TeleMedia Limited (“ATML”)[13] between 5 February and 6 June 2007, and if so, to determine the identities of any person who had engaged in such market misconduct and the amount of any profit gained or loss avoided in consequence. The MMT’s Report was published on 26 November 2015. 12. The notice named the 1st and 2nd respondents as persons suspected of having engaged in the market misconduct in question, the 1st respondent, Yiu Hoi Ying Charles (“Charles”), being ATML’s Director of Finance and an executive director; and the 2nd respondent, Wong Nam Marian (“Marian”) being its Company Secretary. Two other persons were also named as suspects but the MMT made no findings of market misconduct against them. They are Lu Ruifeng (“Lu”),[14] ATML’s Chairman, CEO, executive director and controlling shareholder; and Ho King Lin, Cecilia (“Cecilia”),[15] the Assistant Company Secretary. The SFC did not appeal their exoneration. 13. The MMT found that the elements of insider dealing[16] were established against Charles[17] and Marian,[18] holding that when selling their shares, they had information which they knew constituted inside information, that is, price sensitive information.[19] They were, however, acquitted of market misconduct on the basis that they had discharged the burden of bringing themselves within the defence provided by SFO, section 271(3). 14. The SFC’s appeal to the Court of Appeal[20] was dismissed. It rejected a proposed construction of section 271(3) and also rejected the submission that the MMT had made wrong findings of fact regarding the respondents’ beliefs and motivations in selling the shares. 15. The Appeal Committee[21] granted the SFC leave to appeal to this Court in respect of the following questions of law: (a) Whether, in the context of the statutory defence to insider dealing provided for in section 271(3) of the SFO, the meaning of “using” relevant information for the purpose of securing or increasing a profit or avoiding or reducing a loss is broad enough to encompass the “withholding” or “non-disclosure” of relevant information and the taking advantage of such withholding or non-disclosure for the purpose of securing or increasing a profit or avoiding or reducing a loss? (Question 1) (b) Whether the effect of the interpretation contended [for] by the Applicant (as identified in paragraph 40 of the Judgment, referred to as the “Applicant’s Interpretation”) would be to equate the “use” of such information with the mere “possession” of it so as to render the statutory defence in section 271(3) wholly or largely inoperative, otiose or illusory? (Question 2) (c) Whether the statutory defence provided in section 271(3) is intended by the legislature to be exceptional or of limited application? (Question 3) (d) Whether the adoption of the Appellant’s Interpretation means that the provisions of the SFO concerning insider dealing are being employed to “further the same objective” as that which governs disclosure of price sensitive/relevant information prescribed in Rule 13.09 and the related provisions of the Listing Rules (as now codified into Part XIVA of the SFO)? (Question 4) 16. Leave to appeal on the “or otherwise” ground was refused. 17. Shortly before the hearing, the Court invited submissions from the parties on the following question, namely: “Whether, on the findings of the Tribunal, upheld by the Court of Appeal, namely, the findings that the respondents’ sales of ATML shares while they possessed price sensitive information, although prima facie constituting [market misconduct], were solely motivated by the unexpectedly high prices achievable taken together with their belief that the relevant information would remain behind closed doors, it was correct as a matter of law to hold that the respondents were entitled to rely on the defence provided by section 271(3) of the SFO.”[22] In their oral submissions to the Court on this appeal, the parties concentrated on this core question since it was dispositive of the appeal. B. The course of events B.1 ATML’s condition 18. In 2002, when Lu acquired control of ATML,[23] the company was already in a parlous financial state. As acknowledged in an agreement between ATML and Madam Liu Lien Lien (“Madam Liu”) dated 29 July 2002, the company then owed Madam Liu sums totalling $83,388,308.00 which it promised to repay in instalments.[24] It was insolvent and no more than a listing shell. Lu planned to inject new businesses into ATML but, as it turned out, that did not occur. Defaults in repayment under the agreement led Madam Liu to serve five statutory demands on ATML between October 2002 and April 2006. However, on each occasion, she had been willing to negotiate and did not follow up the statutory demands by serving any winding-up petitions. 19. Meanwhile in 2005, ATML granted stock options to employees including the respondents, exercisable at $0.20 per share. This was described by the MMT as “at best aspirational” since the price at which ATML shares traded rarely reached $0.20, and fell to as low as $0.05 in November 2005.[25] 20. ATML’s auditors qualified its financial statements for the years ended 31 December 2004, 2005 and 2006 on the going concern basis so that the market was aware of its insolvent position. The evidence was that it was kept on “life support” as a possible vehicle for a back door listing in the hope of realising between $100 million and $300 million if someone acquired it for that purpose.[26] B.2 The assignment 21. The events giving rise to the present proceedings occurred in the first half of 2007. On 5 February 2007, Madam Liu’s solicitors, Messrs Woo, Kwan, Lee & Lo (“WKLL”) served on ATML notice of an assignment dated 1 February 2007 by Madam Liu of the balance of ATML’s debt in the sum of $58,083,992.00, plus accrued interest at the rate of 7% per annum, to a BVI company named Goodpine Limited (“Goodpine”) for a consideration of $25 million stated to have been paid (“the assignment”).[27] Acting also for Goodpine, WKLL demanded payment in full of the amount of the debt by 9 March 2007, threatening legal proceedings thereafter.[28] This development was not publicly announced. 22. The respondents testified that they believed that Goodpine was simply Madam Liu under a new guise.[29] However, the MMT described this as “no more than a surmise and not a compelling one”, pointing out that they expressed such belief “simply because the same lawyers represented Madam Liu and the corporation” and observing that where no conflict of interest is involved “it is common for one set of solicitors to represent all parties involved”.[30] 23. The MMT found that: “... at the relevant time ... nobody in Asia Telemedia knew who was the beneficial owner of Goodpine. At that time, therefore, if news of Goodpine had become known to the market, Goodpine would have been seen as a new actor on the stage. Its beneficial ownership would have been unknown and – critically – its intentions would have been unknown too.”[31] 24. Marian consulted ATML’s solicitors, Messrs Chiu & Partners (“C&P”) on the assignment. They advised that “it does not seem to us that you have a real defence to the demand from Goodpine” and sought instructions for putting forward a settlement proposal.[32] Two without prejudice offers made during May 2007 proposing payment of $8 million in final settlement received no response from Goodpine.[33] As the MMT found, “Marian Wong clearly appreciated the potential seriousness to the viability of the Company of this change of circumstance”.[34] B.3 A speculative surge and the statutory demand 25. At the date of the assignment to Goodpine, ATML shares closed at $0.20 on a trading volume of 1,540,000 shares.[35] About a fortnight later, there commenced a surge of speculative interest in ATML shares which sharply drove up their price and trading volumes. Thus, on 16 February 2007, they closed at $0.32. On the next trading day, they increased by a further 43%, closing at $0.46 on a turnover of 133,975,815 shares.[36] In March and April, the share price hovered at around $0.40.[37] 26. Goodpine re-entered the picture on 26 April 2007, when WKLL served a statutory demand on ATML seeking payment of a total sum of $70,270,491.00 and stating that ifwithin21days it failedtopaythefullamount,apetition forwinding-upwouldbe presented. Noting that ATML had been “formally warned” by C&P of “the dangers of a winding-up petition”, the MMT stated: “Despite the change of circumstances brought about by knowledge of the deed of assignment and the following statutory demand, no consideration was given to any suspension of trading in the shares of the Company pending the public being informed of the changed circumstances.”[38] 27. In fact the public were never informed of the assignment and statutory demand. The MMT made the following findings as to the respondents’ knowledge at that stage: “At that moment in time, therefore, what was known to the specified persons, but not to the market, was the following: (i)	that, having failed to secure any reduction of the debt due to her in several years and having received no positive response for a $10 million good faith payment as a prerequisite to further negotiations, there still being in excess of $70 million due, Madam Liu had now assigned the debt to a corporate third party; (ii)	that, on its face – the assignment being a formal deed – the third party had already paid a sum of $25 million to acquire all rights in the debt; (iii)	that, while it may have been surmised that the new creditor, Goodpine, was controlled by Madam Liu, the true identity of those behind the company was not known; nor was the purpose of the assignment known. However, after some five years of default and delay, the probabilities suggested something more than an exercise in passive internal administration. On any informed reckoning the assignment had to hold out the inevitability of some new, more aggressive set of moves to recover the debt; and (iv)	that this corporate third party, Goodpine, had now served a statutory demand seeking full payment within 21 days, failing which proceedings for the winding up of Asia Telemedia would be instituted.”[39] 28. On 7 May 2007, ATML granted further options for 37.5 million shares to its employees (including 5 million to Marian) exercisable at $0.40 per share. The speculative surge continued and gathered pace with the share price closing at $0.61 on 17 May and $0.69 on 22 May. It peaked at $0.97 on 29 May on a turnover exceeding 156 million shares.[40] B.4 The winding-up petition 29. On 6 June 2007, about six weeks after serving the statutory demand, Goodpine presented a winding-up petition. Trading in ATML shares was suspended. They had been trading at $0.83. When trading resumed on 18 October 2007, the share price fell by 62% to close at $0.315. 30. Although the winding-up proceedings progressed, no final winding-up order was made. As the MMT explained: “... before a final winding up order was made, a scheme of arrangement was agreed with a third party. In the result, the Company continues to operate today as a listed corporation, doing so under the name of Reorient Group Limited. Even if it was at the ‘eleventh hour’, it appears that the Company’s value as a ‘listed shell’ prevailed.”[41] B.5 The respondents’ trading in ATML shares 31. Marian exercised her share options to purchase 8 million shares at the price of $0.20 and 2 million shares at the price of $0.40. Between 28 February and 26 April (the date of the statutory demand), she sold 6.2 million shares at prices ranging from $0.37 to $0.494 per share. Between 27 April and 5 June (the day before the winding-up petition was presented), she sold an additional 3.8 million shares at prices ranging from $0.395 to $0.98. She made a net profit of $5.1 million from those sales.[42] 32. Charles exercised his share options to acquire 6 million shares at $0.20 per share and sold them between 28 and 31 May at prices ranging from $0.85 to $0.91, netting a profit of $5.303 million.[43] 33. It was on the basis of those share sales in the light of the information possessed by the respondents regarding the threatened winding-up proceedings that the insider dealing proceedings were brought. C. The insider dealing provisions in the SFO C.1 Section 245: Definition of market misconduct 34. Market misconduct is and was at all material times defined by SFO section 245 as meaning, among other things, “insider dealing”. C.2 Section 270: Definition of insider dealing 35. Insider dealing was at the material time,[44] relevantly defined by SFO section 270(1) as follows: “Insider dealing in relation to a listed corporation takes place – (a) when a person connected with the corporation and having information which he knows is relevant information in relation to the corporation – (i) deals in the listed securities of the corporation ...” 36. It involves five elements: (a) the corporation concerned must be publicly listed; (b) the person concerned must be “connected with the corporation”, usually called a “connected person”; (c) he or she must have information which constitutes “relevant information”, now referred to as “inside information”; (d) he or she must know that such information is inside information; and (e) he or she deals with the corporation’s listed securities with such knowledge. 37. If the five elements are established, it follows that the relevant activity is insider dealing. Furthermore, as the definition of “market misconduct” (see above) shows, such insider dealing will amount to culpable market misconduct giving rise to sanctions under the legislation, unless excused by any available statutory defence such as that provided under section 271(3) (discussed below). 38. The first two elements were not disputed.[45] Nor was it disputed that they had dealt with ATML’s securities. Issue was joined in the MMT as to whether elements of items (c) to (e) were established. 39. Before considering those disputed elements, it is important to note from the definition of insider dealing in section 270 that the activity of insider dealing takes place at the point in time when the connected person having the relevant information and knowledge deals in the relevant publicly listed securities. This temporal aspect applies to each of the five elements so that, to constitute insider dealing, all the elements must be shown to exist at that point in time. C.3 Section 245: Definition of relevant/inside information 40. “Relevant information” (now called “inside information”) was at the material time relevantly defined by section 245 as follows: “‘relevant information’, in relation to a corporation, means specific information about – (a) the corporation; … which is not generally known to the persons who are accustomed or would be likely to deal in the listed securities of the corporation but which if it were generally known to them be likely to materially affect the price of the listed securities; ...”. 41. There are accordingly four elements to this definition: (a) the information must be specific; (b) it must be about the corporation or its listed securities; (c) it must be information which is not generally known to those who deal or are likely to deal in those securities, who might for brevity be called “the market”; and (d) if the information were generally known to the market would be likely materially to affect the price of those securities, in other words, the information must be price sensitive. C.4 Section 271(3): The innocent purpose defence 42. The defence relied on by the respondents is provided by section 271(3) which materially states: “A person shall not be regarded as having engaged in market misconduct by reason of an insider dealing taking place through his dealing in ... listed securities ... if he establishes that the purpose for which he dealt in ... the listed securities ... was not, or, where there was more than one purpose, the purposes for which he dealt in ... the listed securities … did not include, the purpose of securing or increasing a profit ..., by using relevant information.” 43. The following aspects of the defence may be noted: (a) It is a defence which only comes into play where a prima facie case of market misconduct has been established. (b) The burden of establishing the defence is on the person seeking to rely on it, discharged on a balance of probabilities. (c) That person must establish that the purpose for which he or she dealt with the securities was not and (if there was more than one purpose) did not include, the proscribed purpose of securing or increasing a profit by using relevant information. (d) To discharge that burden, the specified person might often be expected to give direct evidence of his or her subjective purpose to show that he or she was acting for what might be called an “innocent purpose”. If such direct evidence is not given, that person must nevertheless be able to point to evidence which demonstrates that he or she acted for a purpose or purposes which entirely excluded the abovementioned proscribed purpose when dealing with the securities. 44. It is also to be noted that the purpose which the specified person relies upon in support of the defence is that which pertains at the time of his dealing in the relevant listed securities (“… if he establishes that the purpose for which he dealt in … the listed securities…”). This coincidence of purpose and dealing in relation to the innocent purpose defence under section 271(3) ties back to the temporal aspect applying to each of the elements constituting the definition of insider dealing in section 270(1) (see paragraph [39] above). C.5 Section 271(3): Using relevant information 45. It will be seen that the innocent purpose defence looks to the purpose for which the specified person dealt in the listed securities, with the focus of the inquiry being on whether or not it was for, or included, the proscribed purpose. That proscribed purpose, to set it out in full by reference to the statutory language, is “the purpose of securing or increasing a profit or avoiding or reducing a loss, whether for himself or another, by using relevant information” (italics added). 46. The central question in this case being whether, on the MMT’s findings, the respondents were able, as a matter of law, to rely on section 271(3), the meaning of those italicised words became the critical issue in this appeal. Their equivalent in the predecessor provisions to section 271(3) were “by the use of relevant information”,[46] but nothing turns on that legislative history (although it may be noted that, at no stage, was it considered desirable to adopt the very different language of the statutory defence to a charge of insider dealing in the UK laid down in the Criminal Justice Act 1993[47]). 47. It was the case of the SFC that “using relevant information” simply meant dealing in the listed securities when in possession of undisclosed price sensitive information which he knew, if disclosed, would be likely to affect the share price. For their part, the respondents contended that “using relevant information” must mean something other than mere possession or knowledge of the relevant information. 48. It is to be observed that the phrase “using relevant information” is not to be found in section 270(1) itself. However, in sub-paragraphs (c) and (d) of section 270(1), which deal with a person (commonly referred to as a “tippee”) to whom price sensitive information has been disclosed by a connected person, the phrase to “make use of the information” or its equivalent is used. So, in section 270(1), insider dealing takes place: “(c) when a person connected with the corporation and knowing that any information is relevant information in relation to the corporation, discloses the information, directly or indirectly, to another person, knowing or having reasonable cause to believe that the other person will make use of the information for the purpose of dealing, or of counselling or procuring another person to deal, in the listed securities of the corporation …”. (Emphasis added.) Insider dealing as described in section 270(1)(c) or (d) is also subject to section 271(3), so the making use of inside information may reasonably be understood to relate to the same activity in both the section constituting the market misconduct and in the section providing the defence to such misconduct. 49. Construing the provision purposively and in context, as one must, “using relevant information” in section 271(3) simply means making one’s decision to buy or sell the listed securities because of the quoted market price, knowing that price to be either artificially high or artificially low because the relevant information is not generally known to those accustomed or likely to deal in the securities. By doing so, one is employing the price sensitive information to one’s own advantage in order to steal a march on the rest of the market since, were that information generally known, it “would … be likely to materially affect the price of the listed securities” and therefore would have negated the insider dealer’s advantage. Since the statutory purpose of the prohibition on insider dealing is to prevent that particular mischief of taking advantage of price sensitive information to steal a march on the rest of the market, section 271(3) places a burden on an insider dealer to prove some other purpose in the dealing. 50. The above conclusion as to the meaning of the phrase “by using relevant information” does not make those words redundant, as contended by the respondents. Nor are those words merely conveying the fact of possession and knowledge of the price sensitive information. As we have sought to explain, it is the turning of the possession of that knowledge into action which constitutes the use of the relevant information. This construction is consistent with the concept of “consciously making use of relevant information”, which is how the Insider Dealing Tribunal construed the equivalent words in section 10(3) of the Securities (Insider Dealing) Ordinance in the International City case.[48] It is also consistent with the view of the Ontario Court of Appeal that the words “make use of” in section 113 of the Securities Act 1966 (Ontario) meant that the information “must be a ‘factor’ in the insider’s participation in the transaction”.[49] 51. This construction also addresses a question that at one stage had greater significance in the parties’ arguments on this appeal and in the Court of Appeal (see Section E of this judgment, below), namely whether “using relevant information” could be constituted merely by withholding or failing to disclose information to the wider market. That is not the meaning of those words since it is not the mere withholding or failure to disclose price sensitive information that brings about the legislative mischief, namely preventing the gaining of an unfair advantage over the rest of the market by the employment of price sensitive information to one’s financial advantage. Something other than mere withholding or non-disclosure is required and that is the exploitation of the price sensitive information for financial advantage in the way we have described. 52. After the conclusion of the oral arguments in this appeal, our attention was directed[50] to a decision of the European Court of Justice, namely Spector Photo Group NV and another v Commissie voor het Bank-, Financie- en Assurantiewezen.[51] There, the ECJ considered article 2(1) of European Council Directive 2003/6 which prohibits insider dealing. Although therefore not dealing with a defence to an allegation of insider dealing such as that contained in section 271(3) of the SFO, the ECJ’s judgment addresses the meaning of “using” inside information. The ECJ’s conclusion that, for a person to have used inside information within the meaning of article 2(1), the information must “have played a role in his decision-making”[52] is consistent with our conclusion (in paragraph [49] above) as to the meaning of “using relevant information” in section 271(3). D. The MMT’s findings 53. The MMT framed three questions to be answered: “(i) Whether the assignment by Madam Liu to Goodpine and Goodpine’s statutory demand, taken separately or together, constituted relevant information? (ii) Whether the three specified persons knew that the assignment and statutory demand, taken separately or together, was relevant information? (iii) Whether, if the information is found to be relevant information, that fact was in any way a motivating factor when each of the three specified persons dealt in the shares of Asia Telemedia?”[53] 54. Questions (i) and (ii) relate to the SFC establishing culpability for insider dealing on the part of the respondents. Question (iii) concerns the question whether the respondents have brought themselves within section 271(3). D.1 MMT’s findings as to prima facie culpability (a) Was it inside information? 55. As indicated above,[54] the information that was known to the respondents but not to the market when they were selling their shares included the fact (i) that Madam Liu (who, as experience had shown, was a relatively complacent creditor) had assigned the debt for a stated consideration of $25 million to Goodpine, an unknown quantity, which was demanding payment of the debt;[55] (ii) that their solicitors had advised that ATML had no defence to the claim;[56] (iii) that without prejudice offers by ATML had gone unanswered;[57] and (iv) that Goodpine had followed up its initial demand for repayment by serving a statutory demand stating that a winding-up petition would follow if full payment was not made.[58] 56. 56.	The MMT had no difficulty in concluding that such information was “specific”.[59] It also held that the assignment and statutory demand, taken together, clearly constituted inside information, being price sensitive. This was particularly so in the context of the speculative surge. Such information would have been regarded as posing “an existential threat to Asia Telemedia, a small cap share enjoying a speculation bubble, one that was not supported by any realistic fundamentals”.[60] The Tribunal observed that: “[with] a share of greater resilience, a ‘wait and see’ attitude may have prevailed. But Asia Telemedia shares, being traded largely, as a short-term speculative stock, had no such resilience. The obvious answer to avoid the general risk would surely have been to sell. Certainly sell if you are only holding the shares looking for a short term gain before selling anyway.”[61] 57. The MMT concluded: “The Tribunal is satisfied that Asia Telemedia, having received notice of the deed of assignment and the statutory demand – more especially in circumstances in which the new creditor was giving no indication of a desire to negotiate terms of repayment – was under an obligation to make an announcement to the public. That it did not do so was a palpable failing. The Tribunal is further satisfied that if such a notice had been published it would certainly have acted to put pressure on any further upward pressure and in all likelihood would have resulted in a material decrease in the share price. In summary, the Tribunal is satisfied that, taken together, the deed of assignment and the consequent statutory demand constituted price sensitive information.”[62] (b) Did the respondents know that the assignment and statutory demand constituted inside information? 58. The Tribunal found that the answer was clearly “Yes” in relation to Charles. He had sought to downplay his knowledge and role in ATML, suggesting that he was “a director in name only”.[63] But this was roundly rejected by the MMT which noted that it was never disputed that he knew of the assignment and its terms and that a statutory demand had been received.[64] It found that he “was at the very centre of this new debt issue” and “would have played a guiding role in what amounted to a supine attempt to deal with it”, adding: “... it follows that Charles Yiu would have had a firm grasp of the potential seriousness of the new environment brought about by the assignment and the consequent statutory demand and would also have understood only too well the vulnerability of the Company to the threat that was now posed.”[65] 59. The MMT pronounced itself: “... satisfied that Charles Yiu knew that, if, as it should have been, a public announcement was made by Asia Telemedia concerning the deed of assignment (involving a ‘purchase price’ of $25 million by a corporate third party) and the consequent statutory demand, such an announcement would in all likelihood not only have cancel[led] out the continuing rise in the value of the Company’s shares but, having regard to the company’s known frailty, would have brought about a material reduction in their price.”[66] It commented as follows: “The fact that Charles Yiu chose to turn a blind eye to the threat that was now posed does not detract from the fact that, in truth, he understood full well the nature of the threat and, in the course of the hearing before the Tribunal, contrived to put on a façade of ignorance.”[67] 60. Similarly, the MMT had “no hesitation in rejecting Marian Wong’s assertion that she never believed the information was price sensitive”, finding that she “did quickly come to understand the true nature of the threat that presented itself ...”.[68] 61. The Tribunal was “satisfied that, as with Charles Yiu, ... she knew that, if news of the deed of assignment and subsequent statutory demand fell into the public domain, it would be likely to have a material effect on the price of the Company’s shares ...”.[69] (c) Prima facie culpability established 62. It follows that the MMT found that, subject to their possible reliance on the section 271(3) defence, the respondents were culpable for market misconduct by reason of insider dealing.[70] The requirements of section 270 were satisfied: they were connected persons who had, and knew they had, price sensitive information concerning ATML which had not been disclosed to the investing public when they exercised their share options and sold their shares at the high prices generated by the speculative surge. They knew that if the inside information had been published, they would not have been able to achieve such profits since the share price would have fallen significantly. They knew, as the MMT put it, that ATML shares were “enjoying a speculation bubble”[71] and news of the assignment and statutory demand would have “puncture[d] the exuberance”.[72] D.2 The MMT’s findings regarding the section 271(3) defence 63. To be able to rely on section 271(3), the respondents had to prove that the purpose for which each of them had sold the ATML shares was not, and did not include, the proscribed purpose of securing or increasing a profit by using relevant information. They chose to give direct evidence of their subjective motivations. (a) Why they exercised their options and sold 64. It was clearly the opportunity to profit from the speculative surge in ATML’s share price that motivated the sales. The MMT recorded Marian as having stated: “Given that for years the Company’s employees had been forced to sit on the options, the materialization of the opportunity to exercise our share options and sell the shares [at a profit] got everyone excited and eager”.[73] 65. It found that her purpose in selling: “[put] simply ... was to seize upon an unexpected opportunity to make a profit when independent of any matter they knew of or could control, a profit presented itself; as the Tribunal has described it earlier – to pick up the gift of manna from the desert floor.”[74] 66. Charles was similarly motivated to take advantage of the surge in the share price. As the MMT found: “As to why he sold when he did, ... Charles Yiu said that when the share price rose close to four times from 20 cents to close to one dollar the temptation to sell was simply too great. As he put it: ‘I couldn’t even dream of that, you know. And that’s why it was at that point in time I started selling off my shares.’”[75] 67. 67.	The MMT summarised their evidence stating: “... the exercise of the share options was for all three[76] of them quite literally the chance of a lifetime and all three chose to exercise their options and sell their shares to seize that chance of a lifetime.”[77] 68. One may note at this point, that far from helping to establish a section 271(3) defence, their evidence merely served to confirm that theirs was “the purpose of securing or increasing a profit” – an unexpectedly large profit – taking advantage of the high prices produced by the speculative surge. It was in other words (subject to what is said below) evidence tending to confirm a proscribed purpose which would exclude the operation of section 271(3). 69. On a number of occasions, the MMT appears to have held that the section 271(3) was nevertheless available because the aforesaid purpose of seizing the opportunity to sell at the surge prices could be said to be their “sole motivation”: “... all three chose to exercise their options and sell their shares to seize that chance of a lifetime. That was, therefore, their sole motivation: to seize a sudden and unexpected speculative surge in the price of Asia Telemedia shares and, like others employed by Asia Telemedia in Hong Kong and the Mainland, to profit from the windfall.”[78] 70. And in relation to Charles, the Tribunal found: “That, it must be accepted on balance, was his sole motivating factor – to take his share of manna found on the desert floor, that is, to profit from an unexpected speculative boom in the share price - and, at the time he dealt, was unconnected with any desire to avoid a loss by reason of the price sensitive information in his possession.”[79] 71. As we shall presently see, the Tribunal did stop there. It may however be worth pointing out that there is no virtue in the respondents asserting simply that their sole purpose was to secure an unexpectedly high profit, given that such a purpose is part of the purpose prohibited by section 271(3). The question is whether it was also part of their purpose to use inside information to secure such profits. (b) The “behind closed doors” justification 72. The Tribunal had to find that while the respondents’ purpose was indeed to secure or increase profit, they had intended to achieve and in fact did achieve that objective without in any way “using inside information”. Given that the MMT had found that both Charles and Marian well understood the true nature of the existential threat ATML faced; that they knew this was price sensitive information; and that they were selling their shares at high prices which they knew would have been depressed if that information should become public, it is difficult to see how they could escape a finding that they were using that price sensitive information in deciding to sell at a time when that information was still unknown to the market and the high prices were achievable. 73. The sole basis upon which the MMT concluded that they had not “used” the price sensitive information was by accepting what might be called the “behind closed doors” justification. 74. Thus, in relation to Marian, the MMT found as follows: “In the context of all the evidence, the Tribunal is satisfied that, as with Charles Yiu, while she knew that, if news of the deed of assignment and subsequent statutory demand fell into the public domain, it would be likely to have a material effect on the price of the Company’s shares, she believed that – as it had been in the past with Madam Liu – the matter would somehow, however slow and muddled the process, be dealt with behind closed doors. She therefore turned a blind eye to the issue of whether the market should be informed although in truth she knew the answer.”[80] 75. In relation to Charles, the MMT recorded a submission made by the Presenting Officer which is echoed by the comment we have just made: “The Presenting Officer further submitted that, unless the Tribunal accepted that Charles Yiu did not appreciate he was in possession of price sensitive information, it would be wholly unrealistic to find that, when he sold his shares, he had no intention somehow to use that price sensitive information to his advantage.”[81] This was, however, rejected by the Tribunal on the basis of the “behind closed doors” justification: “... an insider may know that information in his possession is price sensitive (in the sense that, if passed into the public domain, it will likely have a materially adverse impact on the share price) but, for his own reasons, whether sound or suspect, believe that the information will not pass into the public domain; put another way, that whatever threatens the share price will be resolved behind closed doors.”[82] 76. At the heart of the MMT’s acceptance of the section 271(3) defence is the following passage in the Report: “... Yes, he [Charles] knew he was in possession of information which, if it became known to the market, would in all likelihood materially depress the share price. However, on balance, the Tribunal is satisfied that Charles Yiu believed that what was known to him would, by one means or another, be sorted out behind closed doors (as it had been in the past with Madam Liu) and would not therefore become a matter to influence the market. In the judgment of the Tribunal, on balance, that must explain why Charles Yiu paid no heed to the 21 day deadline imposed by the statutory demand. It was not because he lacked any appreciation of the events unfolding (although he tried his best to suggest such was the case) or because he was reckless. As he said, he sold because the share price, which clearly he was watching carefully, had gone so high that it was time to take his profit. In that sense, it was an undeniably sensible decision and, in the judgment of the Tribunal, not a decision that in any way indicates a conscious intent to misuse the price sensitive information in his possession. That, it must be accepted on balance, was his sole motivating factor – to take his share of manna found on the desert floor, that is, to profit from an unexpected speculative boom in the share price - and, at the time he dealt, was unconnected with any desire to avoid a loss by reason of the price sensitive information in his possession.”[83] 77. With great respect, it is our view that the Tribunal erred in law in holding that the “behind closed doors” justification was capable of assisting the respondents to establish the section 271(3) defence. 78. On its face, the respondents’ share dealings would appear plainly to have involved “using” the inside information. They sold taking advantage of their knowledge that the prices and profits they were securing were significantly greater than they ought to have been and which would no longer be achievable if the information were to find its way into the market. By doing so, they were “using relevant information” as that phrase is to be construed (see Section C.5 of this judgment, above). 79. Such culpable use cannot possibly be affected by their assertion that they subjectively believed that such information would not pass into the public domain but that the threat would somehow be sorted out behind closed doors. Note that this involves a belief in two separate facts, namely that (i) the negative news about the company would remain “behind closed doors” and, (ii) the negativity would disappear in the future because “whatever problems face the company will be successfully resolved”.[84] 80. “Relevant information” (or “inside information”) was defined by section 245 as information “which is not generally known to [potential investors] but which would if it were generally known to them be likely to materially affect the price of the listed securities”. Thus, the fact that the information does not in fact become known to the public in any particular case does not stop it from being inside information so long as it can be shown that if it were disclosed,it would affect the share price. Insider dealers may well hope that problems facing the company will be solved and that the inside information will never become publicly known so that the fact that they had dealt in the company’s securities taking advantage of such information will never be discovered. Or they may hope that their trades will be completed without knowledge being attributed to themselves before the information is made public. But whether or not the information actually enters the public domain does not affect its status as inside information. An insider may obviously still use that information while hoping or believing (rightly or wrongly) that it would not leak outside. It is a non sequitur to say that because a person believed it would not become public, that person did not use the price sensitive information. Relevant information is, by definition, always “behind closed doors”. 81. As regards the prospective successful resolution of a corporate problem, this ignores the fact that, when a person with price sensitive information deals in the securities for financial advantage, he is “using relevant information” at that very time and so his belief as to whatever might prospectively happen in the future to resolve the company’s dilemma is nothing to the point. Let us assume the dilemma is caused by the fact of a statutory demand having been served on an insolvent company. Because that statutory demand is not known to the market generally, the share price remains at $1. However, the negative news of service of the statutory demand would have depressed the price to 50 cents. The correct question to ask is, “If the market as a whole knew of the negative news at that time when the relevant dealing took place, what would the share price have been?” It cannot be right to answer that question, “Well, since I believed the statutory demand would not lead to a petition being presented, the price of $1 was justified and therefore such information was to me completely irrelevant.” At the time when the dealing took place (which is the critical time to judge whether an act constitutes insider dealing: see paragraph [39] above), the problem remained to be resolved and so a belief in a prospective resolution is simply irrelevant as a matter of law. 82. It would indeed be surprising if the law were to permit an insider to escape culpability by reliance on his or her own subjective belief, saying: “Yes, I traded in shares knowing that their price would be affected by undisclosed price sensitive information which I had, but that is okay because I believed that the market would never find out”. It seems that the MMT itself had some doubts about its acceptance of such an argument: “There is a moral dimension to this which some may find unappealing. How can it be that an Executive Director of a listed company can know of damaging matters that, if known to the market, would materially depress the share price and still be able to deal in the shares of the company without being found culpable of insider dealing simply because he believes that the damaging information will remain confidential and further believes, whether his belief is soundly based or not, that whatever problems face the company will be successfully resolved?”[85] 83. With respect, its answer to that question is less than convincing: “The answer perhaps, in so far as one may be required, is twofold. First, the mischief to be avoided is the use of confidential information to ‘steal a march’ on ordinary investors. If, however, the confidential information has not been used in any way in that manner then there has been no mischief. Second, there is no risk of the floodgates being opened because the circumstances will be rare when a person who deals in the shares of a listed company while in possession of price sensitive information will be able to demonstrate that his dealing was totally unconnected with any desire to avoid a loss or make a profit by reason of the price sensitive information.”[86] 84. The first answer, asserting that no use of the information is involved, begs the very question at issue, which is: Did the “behind closed doors” justification allow one to conclude that the respondents did not steal a march on ordinary investors? In our view, the answer is clearly “No”. Their belief that the information would not get out from behind those closed doors does not mean that they did not use it. Deciding to sell – to “seize that chance of a lifetime”[87] – with the benefit of price sensitive information, did involve stealing a march by using the relevant information at the material time, that is, at the point in time they dealt in the shares. 85. The second answer, namely, that it will rarely be the case that someone who deals in relevant shares while in possession of price sensitive information “will be able to demonstrate that his dealing was totally unconnected with any desire to avoid a loss or make a profit by reason of the price sensitive information”[88] is essentially the same as the first answer. Its premise is the non sequitur that a person who believes that the inside information will not leak out is not using it. Supposed rarity does not in any event justify the argument as a matter of principle. 86. We would emphasise that this analysis does not involve any challenge to the findings of the Tribunal. The respondents were found, no doubt correctly, to have genuinely believed “for ... reasons, whether sound or suspect”[89] that the threat would be resolved and the information would not get out from behind closed doors. The point is that such a belief is legally irrelevant and does not help them establish the section 271(3) defence: it does not demonstrate that they did not use the price sensitive information they possessed. (c) Conclusion as to the MMT’s approach to section 271(3) 87. It is accordingly our view that the Tribunal fell into error in holding that the respondents had succeeded in establishing the section 271(3) defence. A passage from the Written Case lodged on Marian’s behalf is revealing: “The prosecution must prove that the person has dealt in shares of a listed company while possessing information which he knew, if disclosed to the public, would likely to materially affect the share price. In most circumstances, the inference from these facts would already be that the person’s purpose was or included a purpose to make a profit or avoid a loss by using inside information.”[90] It goes on to submit that this usual inference is avoided in “truly exceptional circumstances”[91] in the present case. Those circumstances are essentially that the 2nd respondent’s purpose was to take advantage of the surge prices, supported by the “behind closed doors” justification.[92] 88. The Written Case correctly states that a finding that the respondents dealt in ATML shares while in possession of information which they knew was price sensitive, normally supports an inference that their purpose was or included the purpose of making a profit or avoiding a loss by using inside information. As Lord Nicholls of Birkenhead NPJ recognised in Insider Dealing Tribunal v Shek Mei Ling,[93] there would be insider dealing by an insider who “sold her shares before the price sensitive information had become fully available to the market” and by an insider “who buys shares improperly by misusing confidential information [and] seeks thereby to steal a march on the market”. Such insiders are required to abstain from dealing with the company’s securities until the market has had an opportunity to receive that information.[94] As Sir Anthony Mason NPJ noted in Koon Wing Yee v Insider Dealing Tribunal,[95] the statutory policy is to “eliminate insider dealing and to reinforce the transparency of the markets, thereby enhancing and preserving Hong Kong’s position as an international financial centre”. 89. A section 271(3) defence is not easy to establish. It only arises after prima facie culpability is established. It requires the specified person to prove on a balance of probabilities that dealings which prima facie constitute insider dealing were done without in any way involving the prohibited purpose. Generally, one would expect the insider positively to establish an innocent purpose. Thus, the defence might arise, for instance, where he or she dealt in the securities pursuant to a prior contractual obligation and had to sell whether it entailed realising a profit or a loss. Or the defence might arise where a person sells shares in compliance with an order of the Court made, for example, in matrimonial financial relief proceedings.[96] 90. But in the present case, there was nothing to stand in the way of drawing the obvious inference. The respondents could not point to any innocent purpose. Their testimony only served to emphasise that they were pursuing the purpose of securing profit, necessarily taking into account the price sensitive information in deciding to sell at such favourable prices. This constituted insider dealing and the “behind closed doors” justification did not enable them to rely on section 271(3). To permit such a justification to succeed would be to furnish insiders with a playing field that is not level and transparent. 91. This conclusion follows from our construction of section 271(3) (see Section C.5 above). So construed, a connected person in possession of price sensitive information cannot deal in listed securities unless he brings himself within the statutory defence by proving an innocent purpose in his dealing which does not include the purpose of making a profit or avoiding a loss. This may well put him at a disadvantage compared to others who may deal in the same securities (such as happened in the present case in respect of the other employees of ATML who exercised their options and sold shares when the price was surging) but that is the consequence of being a connected person who comes into possession of price sensitive information. A restriction on that connected person from dealing in listed securities may disadvantage him but that, it seems to us, is the reasonable price of achieving a fair and level playing field in the market. It is akin to restrictions placed on trustees and other fiduciaries from self-dealing or profiting from their positions. E. The Court of Appeal’s decision 92. The Court of Appeal upheld the MMT’s decision. Kwan JA, writing for the Court, focussed on the SFC’s two main grounds of appeal. 93. The first ground involved the allegation that the respondents had “used” the inside information by withholding its disclosure from the public in breach of their respective obligations to disclose, thus contributing to the maintenance of a falsely inflated share price which they knew was the consequence of their non-disclosure.[97] Her Ladyship summarised the argument thus: “The Tribunal has found that under the Listing Rules, ATML was under an obligation to disclose the Assignment and the statutory demand to the public, which in all likelihood would have resulted in a material decrease in the share price. The non-disclosure of the relevant information to the public falsely sustained an inflated share price of ATML and avoided puncturing the exuberance of the speculation bubble. By withholding or causing ATML to withhold disclosure, Charles and Marian had knowingly and directly contributed to the maintenance of the falsely inflated share price. In selling their shares, they knowingly and directly profited from the falsely inflated share price which they had contributed to maintaining through non-disclosure. Mr Wong submitted this plainly was ‘use’ of the relevant information for the purpose of making profit.”[98] 94. It may be noted that while in common with the analysis we have adopted, this argument highlights the advantage taken by the respondents of the inflated share price while possessing price sensitive information, it differs from that analysis by introducing the allegation that the respondents had themselves wrongfully withheld or caused ATML to withhold publication of the inside information, constituting “use” of the price sensitive information proscribed by section 271(3). 95. Thus, the SFC’s case (with italics supplied) was that: “Regardless of their prime motivation in selling their shares, or their belief that the news of the Assignment and the statutory demand would not get out to the market, Mr Wong [counsel for the SFC] submitted that the fact that they sold their shares for profit in a false market which they knew well was maintained by the non-disclosure of the relevant information procured or knowingly contributed to by them necessarily means that they were using the PSI [price sensitive information] for the purpose they aimed to achieve, namely to make profits for themselves.”[99] And that: “... ‘use’ of the PSI would include withholding it only where that person plays a part in withholding the information. ... this would cover the situation where a company is under an obligation to disclose the PSI under rule 13.09 of the Listing Rules, but has failed to do so and the senior executive would be regarded as having withheld or caused the company to withhold information by virtue of his corporate position and duties.”[100] 96. Kwan JA rejected that argument as involving a strained interpretation of the word “using” in section 271(3) by equating “withholding” with “use”; and because importing such a broad basis for excluding operation of the defence would result in its being rendered otiose.[101] Her Ladyship also rejected the argument on procedural fairness grounds: it had not been put to the respondents that they had knowingly and directly contributed to the maintenance of a falsely inflated share price by suppressing the information, and there was no finding to that effect.[102] With respect, we agree with those conclusions. 97. By its second ground of appeal: “The SFC sought to challenge two findings of fact made by the Tribunal, that Charles and Marian believed the settlement of the debt would be settled behind closed doors and that their sole motivating factor in exercising their share options was to profit from the unexpected speculative boom. It was contended that these two findings were plainly wrong.”[103] 98. It should be observed that the complaint concerning the finding of a belief that the information would remain behind closed doors differs from our criticism of the “behind closed doors” justification discussed above. We have accepted the correctness of the MMT’s findings as to the respondents’ subjective beliefs but have sought to demonstrate that such beliefs are legally irrelevant and do not assist in establishing the section 271(3) defence. In the Court of Appeal, the SFC’s argument was that, for various reasons, “they could not possibly have believed that the information would never come to light and that ATML’s vulnerability would never be realised.”[104] 99. Kwan JA held that, on well-established principles, there was no basis for the Court of Appeal to interfere with those findings since it was impossible to say that they were unsupported by evidence or plainly wrong.[105] Again, we respectfully agree. 100. The question of whether, on the evidence, the section 271(3) defence was in law made out by the respondents, and in particular whether the “behind closed doors” justification enabled them to rely on that defence notwithstanding the inculpatory findings made against them, does not appear to have been raised or addressed. However, the Court of Appeal appears implicitly to have agreed with the MMT’s approach to that defence. Thus the Court upheld the finding that the respondents believed that information about the assignment and statutory demand would stay behind closed doors, evidently accepting that such belief permitted the respondents’ exculpation pursuant to section 271(3). To the extent that the Court of Appeal so decided, in our respectful view, they upheld an erroneous conclusion reached by the MMT. F. Disposition of this appeal 101. As noted at paragraph [17] above, the question of whether, on the MMT’s findings, the respondents were able, as a matter of law, to rely on section 271(3) was the core question on this appeal. 102. On the basis of the foregoing analysis, it is our conclusion that the respondents were erroneously acquitted of market misconduct. Correctly applying section 271(3) to the facts found by it, the MMT ought to have held that the respondents had, as a matter of law, failed to make good that defence. The MMT found that their purpose in selling their shares was undoubtedly to secure profit when they possessed information which was price sensitive. It found that prima facie the respondents were involved in insider dealing. The MMT should not have accepted that the “behind closed doors” justification supplied a basis for exoneration. Allowing the appeal on this ground does not involve interfering with the MMT’s findings or raising questions on which the respondents might have wished to call additional evidence and so, although it was not the focus of the arguments in the MMT or the Court of Appeal, there is no procedural obstacle to this course, nor was any procedural objection raised. 103. Accordingly, it therefore follows that the appeal should, in our view, be allowed and the matter remitted to the MMT to deal with the question of sanctions. Such remitter will be on the basis that the respondents’ dealings in ATML shares subsequent to 26 April 2007 (the date of the statutory demand) constituted insider dealing which was not excused by the innocent purpose defence provided for in section 271(3). Mr Justice Tang PJ : 104. I regret I am unable to agree with Ribeiro and Fok PJJ. The essential facts have been stated in their joint judgment, which I gratefully adopt. I will state my reasons as briefly as I can. 105. Stripped to bare essentials for the purpose of this appeal, under s 270 of the Securities and Futures Ordinance, Cap 571, insider dealing takes place when a person connected with the corporation, in this case, Asia TeleMedia Limited (“ATML”), a listed company, who has relevant information which he knows is relevant information, deals in the securities of the corporation. The four elements required under s 270 are: (i) connection with the listed corporation, (ii) possession of relevant information; (iii) knowledge that the information is relevant information, and (iv) dealing in the securities of the corporation. Connection is straight-forward; Charles was an Executive Director as well as the Director of Finance of ATML. Marian, the Company Secretary. Relevant information was then defined in s 245(2)[106] as information which “is not generally known to those persons who are accustomed or would be likely to deal in the listed securities of the corporation but which would if it were generally known to them be likely to materially affect the price of the listed securities,” in other words, price sensitive information. The Tribunal regarded the assignment of the debt (notice of which was given on 5 February 2007) together with the statutory demand of 26 April 2007 as relevant information and held that Charles and Marian were aware that they were so. Importantly, the Tribunal was of the view “that the deed of assignment on its own would not have constituted price sensitive information.”[107] Hence, dealings prior to 26 April 2007 would not be caught by s 270. A winding-up petition was presented on 6 June 2007 pursuant to the statutory demand and, when it became publicly known, trading in ATML shares was suspended. The Tribunal held that had the relevant information been known, it would have substantially and adversely affected the share price. Dealing was widely defined in s 249 and included selling, purchasing, agreeing to sell or purchase. The Tribunal found that both Charles and Marian sold ATML shares whilst in knowing possession of relevant information after 26 April and before 6 June 2007. 106. On such findings, subject to the defence under s 271(3), Charles and Marian were insider dealers under s 270 and as such guilty of market misconduct.[108] The Tribunal, however, concluded that the defence under 107. The answer depends on the correct interpretation of s 271(3) which provided: “A person shall not be regarded as having engaged in market misconduct by reason of an insider dealing taking place through his dealing … if he establishes that the purpose for which he dealt … did not include, the purpose of securing or increasing a profit or avoiding or reducing a loss, … by using relevant information.”[109] 108. The s 271(3) defence only becomes relevant when it has been established, the burden being on the SFC in proceedings before the Tribunal, that insider dealing had taken place.[110] The words “by reason of an insider dealing taking place through his dealing” leave no doubt. 109. I turn to consider how insider dealing might take place. As I said, four elements are involved. For present purposes, I note, in particular, that trading whilst in possession of relevant information, which by definition, is not information in the public domain, the insider dealer would have had the benefit of the relevant information, namely, an artificially high price. That being so, I turn to the s 271(3) defence and ask, how might the insider dealer, who has benefited from the relevant information, satisfy the tribunal or the Court, that: “the purpose for which he dealt in … the listed securities … in question … did not include, the purpose of securing or increasing a profit or avoiding or reducing a loss, … by using relevant information.”[111] 110. In my opinion, the defence is not confined to the insider dealer establishing that the purpose(s) of his dealing did not include the purpose of securing a profit or avoiding a loss, for that would render the words “by using relevant information” otiose. 111. Nor is it limited to situations where the dealer was under compulsion to deal. This was the subject for decision in Henry Tai Hon Leung v Insider Dealing Tribunal.[112] There the Court of Appeal[113] was concerned with the Securities (Insider Dealing) Ordinance, Cap 395, in particular, s 10(3) of that Ordinance, a provision which is closely comparable to s 271(3), which provided: “A person who enters into a transaction which is an insider dealing shall not be held to be an insider dealer if he establishes that he entered into the transaction otherwise than with a view to the making of a profit or the avoiding of a loss (whether for himself or another) by the use of relevant information.” 112. There the tribunal had held that the insider dealer must show that he was compelled to sell and that, “without alternative resources, he had no choice but to sell at that time, regardless of whether or not he had come into possession of the relevant information.”[114] Rogers VP, with the concurrence of the other members of the Court, said: “28. Of course, if a person can establish that he had no choice but to sell securities he will, no doubt, be in a strong position to establish a defence under section 10(3) on the basis that there was not an intention to make a profit or avoid a loss. However, the subsection is clear. What has to be determined is whether there was any desire or intention to make a profit or avoid a loss by use of the relevant information. The section does not incorporate any test as to whether the person was compelled to or had no choice but to sell securities. In those circumstances it seems to me it would be wrong to interpret the Ordinance as if it incorporated this as part of the statutory defence.” 113. Later in the judgment, Rogers VP said it was for the insider dealers: “to establish that the sales … were not in any way influenced by their knowledge of the [relevant information].”[115] (my emphasis). 114. A similar approach was adopted in a criminal appeal,[116] where Henry Tai Hon Leung was cited.[117] Section 292(3) required the defendant to prove: “the purpose for which he dealt in … the listed securities … was not, or, … did not include, the purpose of securing … a profit … by using relevant information.” 115. The Magistrate had held that: “Once you are in possession, have the relevant information, know the relevant information, and know that the information would affect the share price, you may not deal in the shares.”[118] 116. Derek Pang J said that was wrong because that ignored the defence under s 292(3). He said: “43. … in order to invoke the defence under s.292(3) successfully, what the person having relevant information did must not be affected by the relevant information, even if it was in the least affected.” 117. In Canada, Green v Charterhouse Group Canada Ltd (1976) 68 DLR (3d) 592, a decision of the Ontario Court of Appeal, was concerned with s 113(1) of the Securities Act 1966 (Ontario), which provided a private right of action against insider dealer: “who, in connection with a transaction … , makes use of any specific confidential information for his own benefit or advantage that, if generally known, might reasonably be expected to affect materially the value of such securities, is liable to compensate …” 118. The argument on behalf of the alleged insider dealer which was accepted by the trial judge was that: “… it is not enough to have the information. To ‘make use of’ it, they submit, the information must be a ‘factor’ in the insider’s participation in the transaction which the insider carries out with the person alleged to be aggrieved, ‘either by inducing him to enter into it or by assisting him or otherwise influencing him in the manner in which he performs it.’”[119] 119. The judgment of the Ontario Court of Appeal[120] was delivered by Arnup JA who said the burden of proof was upon the insider “to show that in fact he did not make use of the information in the transaction, that is, that the information was not a factor in what he did.” And that: “In my view it is a question of fact in each case, and with respect to each individual in a case, whether the individual made use of specific confidential information.” (at 619) 120. Green was cited and adopted by the Insider Dealing Tribunal[121] in the International City Holdings Limited Enquiry.[122] At that time insider dealing was covered by Part XIIA of the Securities Ordinance, Cap 333 “(1) Insider dealing in relation to the securities of a corporation takes place and, pursuant to section 141C, may be culpable for the purposes of this Part – (a) when a dealing in securities is made, procured or occasioned by a person connected with that corporation who is in possession of relevant information concerning the securities;” 121. The defence under s 141C(3) provided: “A person who enters into a transaction which is an insider dealing within section 141B(1)(a) may be held not culpable for the purposes of this Part if his purpose is not, or is not primarily,[123] the making of a profit or the avoiding of a loss (whether for himself or another) by the use of relevant information.” 122. At para 2.9 of the report, the tribunal said: “… Another way of approaching the all-important question of the use of relevant information is to enquire and determine whether the evidence satisfies the Tribunal that the relevant information was a factor in the insider’s participation in the dealing transaction either by inducing him to enter into it or by assisting him or otherwise influencing him in the manner in which he performs the transaction. This was the approach of the trial judge and expressly approved by Arnup JA delivering the judgment of the Ontario Court of Appeal in Green v Charterhouse Group Canada Ltd …” 123. I would also note that the tribunal at para 2.10 agreed with the submission of the counsel, Mr Leslie Wright that: “… making use of relevant information in dealing in securities (as distinct from merely dealing in securities when in possession of relevant information) was the touchstone of culpability. We agree.”[124] 124. In England, in R v Cross [1990] BCC 237, the Court of Appeal[125] considered s 3(1) of the Company Securities (Insider Dealing) Act 1985, and held that the trial judge had seriously misdirected the jury because he had removed the option of the jury finding: “… that the defendant had price-sensitive information (in other words, the prosecution had proved their side of the case) but that the defendant had in turn proved that he had not used that information in order to make his profit or avoid his loss.” (at 248F) 125. Here, in the Court of Appeal, Kwan JA said with the concurrence of Lam VP and Cheung JA: “31. Pursuant to [section 271(3)], it is open to a person to establish on a balance of probabilities that, although at the time of dealing he was knowingly in possession of PSI, that was not a factor inducing him to deal. If he is able to establish that fact, then he is not to be identified as being an insider dealer. It is not sufficient to establish that the PSI was only a subsidiary motivating factor, it must be established it was not in any way a causative factor. As stated by Rogers VP in Henry Tai Hong Leung at para 28: ‘What has to be determined is whether there was any desire or intention to make a profit or avoid a loss by use of the relevant information.’” 126. These authorities support my view which is based on the language of s 271(3) read with s 270, that notwithstanding a finding that insider dealing has been established under s 270, if the insider can prove on a balance of probabilities that the inside information had not in any way, influenced, motivated or been a factor, in his dealing, he should “not be regarded as having engaged in market misconduct by reason of an insider dealing”. As Arnup JA said in Green: “… it is a question of fact in each case, and with respect to each individual in a case, whether the individual made use of specific confidential information.” (At 619) 127. Indeed, as R v Cross shows, it is a question for the jury in a criminal prosecution to decide. Crucially, the Tribunal regarded this as “a subjective issue, namely, whether, when he exercised his options and sold Asia Telemedia shares, Charles Yiu was motivated in any way by the fact that he knew he was at the time in possession of price sensitive information.”[126] 128. I agree it is a subjective issue and the fact that s 271(3) placed the burden of proof on the insider supports this view. 129. Nonethless, the burden of proof will not be easily discharged, as the Tribunal explained: “279 … there is no risk of the floodgates being opened because the circumstances will be rare when a person who deals in the shares of a listed company while in possession of price sensitive information will be able to demonstrate that his dealing was totally unconnected with any desire to avoid a loss or make a profit by reason of the price sensitive information” 130. A Market Misconduct Tribunal is typically chaired by an eminent judge or retired judge and “2 market practitioners”,[127] and can be expected to approach such defence with healthy scepticism. Here, the Tribunal held that it was Charles and Marian’s “sole motivation: to seize a sudden and unexpected speculative surge in the price of [ATML] shares and, like others employed by [ATML] in Hong Kong and the Mainland, to profit from the windfall.”[128] I now turn to consider whether the Tribunal was entitled so to find and whether, in so finding, it committed any error of law. 131. The evidence has been set out in some detail in the joint judgment. I will highlight those which I regard important to my judgment. 132. Between February and June 2007, shares in ATML were caught in what was described as a “frenzy” of “short-term speculation.” ATML was a speculators’ stock.[129] It is clear that the “value” of ATML resided in the fact that it was a listing shell and a possible vehicle for a backdoor listing. The Report highlighted the frenzy: “104. On 16 February 2007, after modest rises over the previous few days, the share price of Asia Telemedia rose by over 42%, closing the day at $0.320. 105. The next trading day, on a turnover of 133,975,815 shares, it rose by over 43%, closing at $0.460. 106. Thereafter the shares fell back slightly but for April 2007 they stayed above 40 cents, hitting a closing day high of $0.510 on 17 April 2007. 107. In May 2007, there was a further surge in the share price, rising during the course of the month to over 90 cents. On 29 May 2007, on a turnover in excess of 156 million shares, Asia Telemedia peaked at $0.970.” 133. Before the Tribunal, there were two other specified persons. I am not concerned with Lu Ruifeng. He was not identified as a person who had engaged in market misconduct because the Tribunal took the view that he was not given a reasonable opportunity of being heard.[130] Cecilia Ho was exonerated of insider dealing because the Tribunal was not satisfied that she was in knowing possession of price sensitive information.[131] Her circumstances may throw light on Marian’s case. 134. The shares which Charles and Marian sold came from share options exercised by them. There were two tranches of options, granted in 2005 and 2007 respectively. 135. The 2005 options were granted on 23 March 2005. Lu, Charles, Marian and Cecilia were each granted options exercisable until 2010 at $0.20 per share, in the following amounts: 1 million, 8 million, 8 million, and 3 million respectively.[132] It is clear that other employees were also granted options at the time. Paragraphs 284 to 287 of the Report summarized Marian’s evidence that the first person to exercise the option was a Ms Chan, followed by eight other employees on 21 February 2007 and then on the following day four more employees. She said, by the end of February 2007, all the employees in the Hong Kong office had exercised their options. As Mr Laurence Li, counsel for Marian submitted: “The surge [in ATML’s share price] caused considerable excitement among the employees. Between 16 and 22 February 2007, 13 employees exercised their stock options and sold the resultant shares.”[133] 136. On 7 May 2007, 37.5 million share options were granted with an exercise price of $0.40 per share. Marian was granted 5 million share options and Cecilia 1 million, exercisable immediately.[134] The evidence is not clear but it is likely that some or all of the 13 employees referred to above were also beneficiaries under the 2007 option. Both Marian and Cecilia sold shares under the 2007 option. In the case of Marian, 2 million shares out of 5 million, and Cecilia, 600,000 out of 1 million.[135] 137. So, the evidence was that 13 employees, who did not possess any price sensitive information, exercised their options and sold the shares. Of course, insofar as they did so before 26 April 2007, they would have sold prior to the existence of price sensitive information and no question of insider dealing could arise. It is not clear[136] whether any of the 13 employees sold after 26 April. But even if they had, since they were not in possession of any price sensitive information, they would not have come within s 270. We know that Cecilia sold both before (commencing on 26 February 2007) and after 26 April and she did so without being knowingly aware of any price sensitive information. She sold 300,000 shares on 11 May at $0.5017, 500,000 shares on 28th May at $0.8400 and 100,000 shares on 29th May at $0.9600. Presumably, she continued to sell because of the rising market price. 138. Marian had also started to sell before 26 April (from 28 February 2007), and had sold 6.2 million shares by 26 April 2007. Those sales fell outside s 270. But she also sold 3.8 million shares after 26 April.[137] I believe some may think that, like Cecilia, it was possible that Marian’s continued sale was not influenced in any way by the price sensitive information. Of course, the burden was on her to prove that this was the case.[138] 139. The Tribunal said: “292. It was the central assertion of Marian Wong’s evidence that all the Asia Telemedia employees, both in Hong Kong and the Mainland, exercised their options – including herself – for one very obvious reason, a reason that had nothing to do with their faith in the longer term viability of the Company or indeed their fear that it had no viable future. That reason was the desire – at last – to exercise their options at a time when, unexpectedly, the share price was surging. Put simply, it was to seize upon an unexpected opportunity to make a profit when independent of any matter they knew of or could control, a profit presented itself; as the Tribunal has described it earlier – to pick up the gift of manna from the desert floor.” 140. The Tribunal then concluded: “293. Even taking into account Marian Wong’s less than impressive evidence, her evasiveness in answering questions being very evident, nothing of substance arose during the course of the hearing to give the Tribunal reason to question her assertion that she had exercised her options for the single reason given above, a reason that was not in any way coloured by the price sensitive information in her possession. 294. As to how it could be that her possession of price sensitive information played no role, the Tribunal reiterates what has been said in respect of Charles Yiu. The Tribunal is satisfied that, while Marian Wong was in possession of information which she knew should properly form the basis for a public announcement by the Company, she nevertheless believed that somehow, in some way, any threat presented by Goodpine would be dealt with behind closed doors. On balance, therefore, the Tribunal is satisfied that Marian Wong demonstrated that her sales were motivated by the single reason amplified above.” 141. With respect, I see no legal or factual error. The motivation of the sale was a question of fact for the Tribunal, who had the benefit of the oral evidence of the witnesses. As Arnup JA in Green v Charterhouse Group at 620: “The Trial Judge, approaching this evidence with a healthy scepticism, might well have disbelieved it. In fact he chose to accept it.” 142. The Tribunal’s finding was not disturbed by the Court of Appeal. I do not believe I am entitled to interfere. Lest it be thought that this is a grudging acceptance of a finding which I am not comfortable with, for what it is worth, I am of the view that Marian was rightly exonerated by the Tribunal. 143. I will look closely at the Tribunal’s reasons exonerating Charles. But first, I wish to pause and consider whether there was any telltale sign of insider dealing in Charles’ case. It is common sense that price sensitive information has a limited lifespan, normally, between the inception of the information and its public disclosure, during which insider dealers would normally deal to make a profit or avoid a loss. Here, the statutory demand was issued on 26 April and the 21-day period expired on 17 May 2007. Thereafter, a winding-up petition based on the statutory notice could be presented, although, it was not presented until 6 June 2007.[139] A typical insider dealer could be expected to deal in the securities between 26 April and 17 May to profit from his private knowledge of the price sensitive information. 144. Charles sold a total of 6 million shares between 28 May and 31 May 2007 as follows: 145. Charles was not a beneficiary under the 2007 option, in other words, his shares came from the 2005 option and the exercise price was 20 cents. If Charles was prompted by the inside information to sell, one might expect him to start selling within the window of opportunity between 26 April and 17 May. He did not do so. Of course, this does not prove that he was not influenced. But his conduct is consistent with his case. 146. According to the Tribunal: “267. Central to Charles Yiu’s defence was that, if he had been in any way motivated by a desire to exercise his options and sell his shares before Goodpine instituted winding up proceedings, he would surely have done so before the 21 day deadline set out in Goodpine’s statutory demand had expired. Within that 21 day period, knowing that no public announcement was being made by Asia Telemedia as to Goodpine and the statutory demand, he would have been comparatively safe and would have had the ability to pick the best dates to sell. But after that 21 day deadline he would have appreciated that he was increasingly at risk of the winding up petition being issued and trading in Asia Telemedia shares being suspended. The statutory demand was received by Asia Telemedia on 26 April 2007, its 21 day deadline expiring on or about 17 May 2007. The only reasonable inference to be drawn therefore was that, if motivated in any way by the price sensitive information of the statutory demand following the deed of assignment, he would surely have exercised his options and sold his shares before 17 May 2007.” 147. The Tribunal went on to point out, not only did he only commence to sell on the 28 May, 10 days after the deadline, he chose not to sell them at once, but over four days between 28 and 31 May, when the heavy trading on 28 May alone could have easily accommodated his sales.[140] 148. The Tribunal then said: “271. As to why he sold when he did, as cited earlier in the report [at para 133], Charles Yiu said that when the share price rose close to four times from 20 cents to close to one dollar the temptation to sell was simply too great. As he put it: ‘I couldn’t even dream of that, you know. And that’s why it was at that point in time I started selling off my shares.’” 149. That the Tribunal had approached Charles Yiu’s evidence with healthy scepticism is clear from the following paragraph in the Report. “276. Objectively, there are a number of grounds for criticising the validity of such beliefs. But the Tribunal is given the task here of considering a subjective issue, namely, whether, when he exercised his options and sold Asia Telemedia shares, Charles Yiu was motivated in any way by the fact that he knew he was at the time in possession of price sensitive information. Yes, he knew he was in possession of information which, if it became known to the market, would in all likelihood materially depress the share price. However, on balance, the Tribunal is satisfied that Charles Yiu believed that what was known to him would, by one means or another, be sorted out behind closed doors (as it had been in the past with Madam Liu) and would not therefore become a matter to influence the market. In the judgment of the Tribunal, on balance, that must explain why Charles Yiu paid no heed to the 21 day deadline imposed by the statutory demand. It was not because he lacked any appreciation of the events unfolding (although he tried his best to suggest such was the case) or because he was reckless. As he said, he sold because the share price, which clearly he was watching carefully, had gone so high that it was time to take his profit. In that sense, it was an undeniably sensible decision and, in the judgment of the Tribunal, not a decision that in any way indicates a conscious intent to misuse the price sensitive information in his possession. That, it must be accepted on balance, was his sole motivating factor – to take his share of manna found on the desert floor, that is, to profit from an unexpected speculative boom in the share price - and, at the time he dealt, was unconnected with any desire to avoid a loss by reason of the price sensitive information in his possession.” 150. It is in this context that I will examine the paragraph which has troubled this Court:[141] “274. As the Tribunal has attempted to make clear, an insider may know that information in his possession is price sensitive (in the sense that, if passed into the public domain, it will likely have a materially adverse impact on the share price) but, for his own reasons, whether sound or suspect, believe that the information will not pass into the public domain; put another way, that whatever threatens the share price will be resolved behind closed doors.” 151. I accept relevant information is, by definition, always “behind closed doors”.[142] The “behind the closed doors” argument was never relied on as a defence under s 271(3), rather it was used to explain why Charles and Marian’s dealings were not in any way influenced by or motivated by the price sensitive information. I would not read too much into it. It is clear from the judgments of the Tribunal and the Court of Appeal that they understood the question before them was whether the respondents’ dealings had in any way been motivated or influenced by the price sensitive information. 152. The question remains whether the information in any way influenced the decision to deal, and not whether the insider dealer had had the benefit of the information. That, in my view, is an anterior question which is covered by s 270. Personally, I believe para 274 of the Report says no more than that, subjectively, the price sensitive information played no part whatsoever in Charles’ dealings. The Tribunal was saying no more than that, like the other employees of ATML, Charles sold because of the speculative bubble in the shares and the relevant information was not a factor. He was not influenced or bothered by it. In other words, like the 13 other employees, Cecilia, and indeed Marian, he sold regardless of any price sensitive information. The Tribunal compared this with picking up manna from the desert floor, a polite way of saying that they were taking part in a feeding frenzy. 153. I return to the construction of s 271(3). Mr Laurance Li has very helpfully in his written case traced the history of this provision. It was consolidated into the SFO from s 10(3) of the Securities (Insider Dealing) Ordinance (“SIDO”) Chapter 395 which provided: “A person who enters into a transaction which is an insider dealing shall not be held to be an insider dealer if he establishes that he entered into the transaction otherwise than with a view to the making of a profit or the avoiding of a loss (whether for himself or another) by the use of relevant information.”[143] 154. Section 10(3) of SIDO could in turn be traced to s 141C(3) of the Securities Ordinance, Cap 333 which provided: “A person who enters into a transaction which is an insider dealing … may be held not culpable for the purposes of this Part if his purpose is not, or is not primarily, the making of a profit or the avoiding of a loss (whether for himself or another) by the use of relevant information.”[144] 155. Section 141C(3) had a chequered history, and was derived[145] from s 14(1) of the UK Companies Bill 1973 and s 57(6) of the UK Companies Bill 1978, which eventually became s 68(8) of the UK Companies Act 1980 which provided: “The provisions of this section shall not prohibit an individual by reason of his having any information from – (a) doing any particular thing otherwise than with a view to the making of a profit or the avoidance of a loss (whether for himself or another person) by the use of that information.” 156. That later became s 3(1) of the Company Securities (Insider Dealing) Act 1985, which stated: “Sections 1 and 2 do not prohibit any individual by reason of his having any information from – (a) doing any particular thing otherwise than with a view to the making of a profit or the avoidance of a loss (whether for himself or another person) by the use of that information.”[146] 157. I agree with Mr Li it is clear that the object of s 271(3) is that innocent dealing should not be prohibited. That was the objective in England.[147] As Mr Li also submitted: “That was also the view of the Hong Kong Companies Law Revision Committee, which said in its second report dated 12 April 1973: ‘7.131 … In Hong Kong there are many companies with interlocking interests and directorships, and directors of such companies could find it very awkward if the provisions do not indicate with sufficient clarity the types of dealing, by them. We would therefore emphasize that the provisions should be restricted so as to apply to people acting with a guilty intention. The difficulties are also fully realized in Britain …’” 158. Sir Anthony Mason NPJ said: “46. That insider dealing amounts to very serious misconduct admits of no doubt. It is a species of dishonest misconduct.”[148] 159. I agree it is a species of dishonest misconduct. That being the case, it is only right that innocent dealing should not be covered. I would not regard any person who has satisfied the Market Misconduct Tribunal that he dealt without being in any way influenced by the inside information as dishonest. 160. In the course of the submission, I asked Mr Horace Wong SC, counsel for the Commission, whether the s 271(3) defence should avail a connected person who has entered into a voluntary but revocable scheme to purchase a fixed number of shares in the corporation monthly on every payday at the prevailing market price and who allowed the purchase to continue after he had become possessed of inside information, but not because he was in any way influenced by the information. Mr Wong answered in the affirmative. On my understanding of the defence, provided the person could satisfy the Tribunal that it was in fact the case, he should be exonerated. Just as, if he should be prosecuted, he should be acquitted. We are not concerned with criminal proceedings, but it may be helpful to note that Criminal Justice Act 1993, s 52(1) which concerned the criminal offence of insider dealing, it was a defence under s 53(1)(c) for a defendant if he shows “that he would have done what he did even if he had not had the information.” This is jury-friendly language which admits of no doubt and captures, in a few words, the essence of the innocent dealing defence.[149] I believe the defence under s 271(3) should be interpreted to provide a similar defence.[150] 161. As was pointed out in the joint judgment by Ribeiro and Fok PJJ, the phrase “using relevant information” is not found in s 270(1) itself. However, in subparagraphs (c) and (d) of s 270(1), which deal with disclosure of price sensitive information by a connected person to a tippee, the phrase to “make use of the information” or its equivalent is used. For example, in s 270(1)(c), insider dealing takes place: “(c) when a person connected with the corporation and knowing that any information is relevant information in relation to the corporation, discloses the information, directly or indirectly, to another person, knowing or having reasonable cause to believe that the other person will make use of the information for the purpose of dealing, or of counselling or procuring another person to deal, in the listed securities of the corporation…” 162. I believe these provisions support my view that more than simply dealing whilst in possession of the relevant information is required. Were it otherwise, they could simply have provided that insider dealing takes place if the tipper discloses the relevant information … knowing or having reasonable cause to believe that the tippee will deal in the listed securities of the corporation for the purpose of making a profit or avoiding a loss. 163. Nor am I concerned with questions of temporal aspects of dealings or profits or loss. Section 271(3) is concerned with exactly the same dealing(s) found to have taken place under s 270(1). They are not different dealings. Nor am I concerned with profit or avoidance of loss at different times. Section 270(1) does not require actual or prospective profit nor actual or prospective avoidance of loss. Nor does it matter whether any profit was actually made or loss avoided. It can be misleading to talk about actual or prospective profits or avoidance of loss under s 271(3). The trading had occurred and s 270(1) triggered. Actual profit made or loss avoided may be important for penalties if the s 271(3) defence is not made out but the fact that there was or might be such loss avoided or profits made is not decisive of the s 271(3) defence. The defence under s 271(3) is made out if it is proved that the purpose of the trade did not include the purpose of making a profit or avoiding a loss by using the relevant information. 164. Read together with s 271(3), I believe s 270(1) targets dealings where relevant information was used for the purpose of making a profit or avoiding a loss. But just as it would not matter under s 270(1) whether any profit was or would be made or loss avoided or would be avoided, it also would not matter under s 271(3). In other words, dishonest dealing involves the use of relevant information for the purpose of gain or avoidance of loss and not whether any gain had or would be made, or loss avoided or would be avoided. Spector Photo Group 165. Spector Photo is a decision of the Court of Justice of the European Union and concerns the interpretation of Eurpoean Parliament and Council Directive 2003/6/EC on insider dealing and market manipulation, Article 21 of which required Member States to prohibit specified categories of persons (primary insiders) who process inside information from “using” that information by dealing or trading or trying to deal in the instruments. 166. There, the court said: “… on a proper interpretation of Article 2(1) of Directive 2003/6, the fact that a person as referred to in the second sub-paragraph of that provision, in possession of inside information, acquires or disposes of, or tries to acquire or dispose of, for his own account or for the account of a third party, either directly or indirectly, the financial instruments to which that information relates implies that that person has ‘used that information’ within the meaning of that provision, but without prejudice to the rights of the defence and, in particular, to the right to be able to rebut that presumption. The question whether that person has infringed the prohibition on insider dealing must be analysed in the light of the purpose of that directive, which is to protect the integrity of the financial markets and to enhance investor confidence, which is based, in particular, on the assurance that investors will be placed on an equal footing and protected from the misuse of inside information.” 167. I do not believe Specter Photo assists in the interpretation of s 271(3). The language of the European Directive is not truly comparable. The decision arose out of a reference for a preliminary ruling on the proper interpretation of the article. It seems to me that the decision says no more than that an insider, possessed of inside information, who had traded in the relevant security is to be “taken” to have “used that information” “but without prejudice to the rights of the defence and, in particular, to the right to be able to rebut the presumption.” It is not entirely clear what those defences might be though some perhaps could be gathered from the Preamble to the Directives. Here, as noted, s 270 imposes a blanket prohibition against insider dealing, whereas s 271 provides defences against such blanket prohibition. 168. Also, unlike the UK,[151] the European Directive has no effect here. I would also note that, in legislative proceedings leading to the amendment in 2002 which brought the Hong Kong law to its current form, the legislative ad hoc group said that: “the existing United Kingdom definition[152] will shortly be replaced by the definition in the European Community Council directive co-ordinating regulations. We do not know the reasons motivating such a proposed change in the United Kingdom and we must not assume that what is good for the United Kingdom must necessarily be suitable for Hong Kong. If we are to adopt the European Community definition, we must be sure that it is more appropriate to the local circumstances of Hong Kong than the wording now proposed. For the time being, the ad hoc group is not yet so convinced.”[153] 169. For the above reasons, I respectfully agree with the Tribunal and the Court of Appeal and would dismiss the appeal. Lord Neuberger of Abbotsbury NPJ: 170. I agree with the reasoning and conclusion of Ribeiro and Fok PJJ. I also agree with the judgment of Ma CJ. Like him, I add a few words of my own simply to summarise why I respectfully differ from Tang PJ – and indeed from the Court of Appeal and the Market Misconduct Tribunal. 171. The respondents argue that the Tribunal was right to hold that they could rely on section 271(3), so that they should “not be regarded as having engaged in market misconduct by reason of an insider dealing taking place through [their respective] dealing[s] in” ATML shares (“the Shares”). This involves the respondents establishing (and the burden is on them) that “the purposes for which [they] dealt … in [the Shares] did not include … the purpose of securing or increasing a profit or avoiding or reducing a loss … by using relevant information”. 172. In effect, two arguments have been advanced on behalf of the respondents to support their case in this connection. First, that they assumed that the problem in connection with the Goodpine indebtedness, which constituted the inside information (“the information”) in this case, would be sorted out “behind closed doors” and would never be known to the market. Secondly, that their decision to sell the Shares was motivated solely by the desire to take advantage of what the Tribunal called “an unexpected speculative boom” in the market price of ATML shares. Section 271(3) is, I accept, so worded that it could be interpreted so as to accommodate each of those arguments, but in my opinion both arguments should be rejected. 173. So far as the respondents’ first argument is concerned, I consider that it proceeds on a misapprehension as to the nature of the “profit” or “loss” referred to at the end of section 271(3). It is not, as that first argument implicitly assumes, a possible profit or loss at some unspecified future time: it is a notional profit or loss as at the time of the insider dealing. Thus, in this case, the effect of the respondents’ sales of the Shares was to increase the profit they realised, because, as at the dates they sold the Shares, the respondents knew that the ATML share price was higher than it would have been if the inside information had been available to the market. Such an interpretation is consistent with the principal objection to insider dealing, namely that it distorts the market at the time it takes place. An orderly and fair market involves the buyer and the seller of shares enjoying equality of arms, and, more specifically, equality of access to information. 174. This interpretation also seems to me to be more consistent with the statutory scheme. Section 271(3) is only invoked by a person who has been held to fall within section 270, and, at least where section 270(1) applies (as in this case), that means that the person has dealt in shares knowing that he has information which “is not generally known … but would if generally known … be likely to materially affect the price” of the shares. That suggests that one is looking at the share price at the date of dealing. Furthermore, it would be rather odd to hold that a person has satisfied a requirement that he dealt at a time when he had information which he knew “would … be likely to materially affect the price” of those shares, and then to be able to hold that he believed that the information would never affect the price of those shares. 175. The respondents’ first argument also has the disadvantage of looking to the future, which is far more likely to involve the Tribunal having to consider the subjective expectations of the insider dealer, which is a difficult and potentially unsatisfactory exercise, and is unlikely to have been intended by the legislature. I discuss this point a little more fully in [177] and [178] below. 176. Turning to the respondents’ second argument, it appears to me to be unlikely that section 271(3) was intended to be available to any insider dealer who was able to persuade the Tribunal that he did not in fact rely on the information when purchasing or selling the shares concerned. Such an interpretation would involve the section 271(3) defence being based on the subjective intentions of the insider dealer which would inevitably have to be assessed on the basis of the evidence from the person concerned, and normally without the benefit of any objective and directly relevant independent evidence. 177. Inquiries as to what an insider dealer believed are inherently difficult, because he will be the only source of direct information, and he will have a vested interest in the outcome, and there will rarely be any independent evidence which is of much assistance on the issue. Further, if subjective intention is the basis of the section 271(3) defence, while no doubt the Tribunal would be sceptical in such a case, it is not hard to imagine an unscrupulous prospective insider dealer setting up a paper trail ahead of his trading with a view to supporting his contention that, for instance, he did not believe that the inside information would ever come out. Accordingly, an interpretation of section 271(3) which minimises the likelihood of such inquiries is to be preferred. 178. In this connection, I derive support from the decision of the Court of Justice of the European Union said in C-45/08 Spector Photo Group & Van Raemdonck [2010] Bus LR 1416, a case concerned with “the interpretation of Articles 2 and 14 of Directive 2003/6/EC of the European Parliament and of the Council of 28 January 2003 on insider dealing and market manipulation (market abuse)”, to quote from [1] of the judgment. At [36], the Court of Justice said that “entering into a market transaction is necessarily the result of a series of decisions forming part of a complex context which, in principle, makes it possible to exclude the possibility that the author of that transaction could have acted without being aware of his actions” and “where such a market transaction is entered into while the author of that transaction is in possession of inside information, that information must, in principle, be deemed to have played a role in his decision-making”. Accordingly, as the Court of Justice went on to say in [37], “the effectiveness of [the sanctions against insider dealing] would be weakened if made subject to a systematic analysis of the existence of a mental element”. 179. I also consider that section 271(3) would be remarkably lugubriously worded if it was simply intended to provide that an insider dealer should escape liability if he could establish that he would have dealt as he did even if he had not possessed the information, especially as an established statutory formula which provided for this was available to the draftsman – see section 53(1)(c) of the Criminal Justice Act 1993 (UK), quoted by Tang PJ in [160] above. 180. In other words, given that the purpose of provisions such as those with which this appeal is concerned is to ensure that the market is, and is seen to be, orderly, fair and undistorted, those with inside information relating to a company should not be able to deal in the company’s quoted shares, save in the exceptional circumstances, and, indeed, in circumstances which can, at least normally, be objectively verified. 181. The considerations discussed in the preceding paragraphs, and those discussed in the first two judgments above, satisfy me that section 271(3) is intended only to apply where the purpose of the insider dealer’s purchase or sale of the shares concerned can be shown to be unconnected with the market price of the shares. In other words, in the case of a normal transaction – i.e. one motivated (at least in part) by the quoted price of the shares – the insider dealer will not be able to invoke section 271(3) where he has been objectively advantaged as against the market by having the information. It would be inappropriate and unnecessary (and indeed impossible) to provide an exhaustive list of circumstances in which section 271(3) could avail an insider dealer, but they would include a sale or purchase pursuant to a specific contractual obligation or a court order, and a sale or purchase when, if the information had been publicly available, it would, respectively, have increased or decreased the quoted price. 182. I accept that this conclusion places people in the position of the respondents in this case at a disadvantage to other people, in that they may be unable to sell or buy shares in the company concerned when they would do so even if they had not got the information. From the insider’s perspective that is the price of being an insider; from the public perception, that is the price of ensuring a perceptibly fair and undistorted market. Chief Justice Ma : 183. By a majority of four to one, the appeal is allowed. It is also ordered that the matter should now be remitted to the MMT to deal with the question of the appropriate sanctions. The remitter will be on the basis that the 1st and 2nd Respondents (Charles and Marian) are found to be culpable of market misconduct by insider dealing and as stated in para. 103 above. The orders made by the Court of Appeal and the MMT are set aside. As to costs, we would make an order nisi that the 1st and 2nd Respondents pay the costs of the Appellant in this appeal, in the Court of Appeal and before the MMT, such costs to be taxed if not agreed. Should any party seek a different order as to costs, written submissions should be lodged with the Registrar (and served on the other parties) within 14 days of the handing down of this judgment, with liberty on the other parties to lodge and serve written submissions in reply within 14 days thereafter. If no written submissions are received seeking a different order as to costs before the expiry of the relevant period, the order nisi will become absolute. Mr Horace Wong SC and Mr Norman Nip, instructed by Securities and Futures Commission, for the Appellant Mr Russell Coleman SC and Mr Samuel Wong, instructed by Sit, Fung, Kwong & Shum, for the 1st Respondent Mr Laurence Li, instructed by Raymond Chan Solicitors, for the 2nd Respondent Market Misconduct Tribunal, the 3rd Respondent, in person (absent) [1] Cap. 571. [2] This provision is contained in Part XIII of the SFO. [3] In the version of the Ordinance applicable at the material time of this case, the term was known as “relevant information”; in the current form of the Ordinance, this is now known as “inside information”. [4] See s 245(1) of the SFO. [5] See s 252 of the SFO. [6] This information consisted of the assignment of the substantial debt owed by ATML to Goodpine and the service of the statutory demand by that company on ATML. [7] In Charles’ case, profits ranging between 425% and 450%; in Marian’s case between 100% and 500%. [8] Insider Dealing Tribunal v Shek Mei Ling (1999) 2 HKCFAR 205 per Lord Nicholls of Birkenhead NPJ at 207I. [9] (Cap.571) (“the SFO”), repealing and replacing the Securities (Insider Dealing) Ordinance (Cap.395). [10] Koon Wing Yee v Insider Dealing Tribunal (2008) 11 HKCFAR 170 per Sir Anthony Mason NPJ at [45]. [11] Pursuant to section 252(2) of the SFO, dated 16 January 2014. [12] Consisting of Mr Justice Hartmann NPJ (as he then was), Dr Chu Keung Wah and Mr Chan Sai Hung. [13] Subsequently renamed Reorient Group Limited. [14] The MMT found that because of acute illness, Lu was not given a reasonable opportunity of being heard and held that it was accordingly not permitted to identify him as a person who engaged in market misconduct (MMT at [53]). [15] The MMT accepted that Cecilia did not know that the relevant matters constituted price sensitive information and found that she had not engaged in market misconduct: MMT at [263]. [16] As laid down by SFO, section 270: MMT at [158]-[159]. [17] MMT at [224]-[225], [235], [240]-[243]. [18] MMT at [253]-[255]. [19] “Relevant information” is now referred to in the Ordinance as “inside information” and it is convenient to use the present designation. [20] Lam VP, Cheung JA and Kwan JA, [2017] 3 HKLRD 157 (26 April 2017), Kwan JA writing for the Court. [21] Ribeiro, Tang and Fok PJJ, [2018] HKCFA 7 (6 February 2018). [22] Letter from the Registrar to the parties dated 27 August 2018; the words in square brackets were originally “insider dealing” but it was common ground that they should, more appropriately, read “market misconduct”. [23] Then called Mansion House Group Ltd. [24] MMT at [58]. [25] MMT at [283]. [26] MMT at [74]-[76]. [27] MMT at [94]. [28] MMT at [100]. [29] MMT at [103]. This eventually proved to be the fact. [30] MMT at [203], footnote 23. Noted by the CA at [10]. [31] MMT at [203]. [32] MMT at [97]. [33] MMT at [128]-[129]. [34] MMT at [98]. [35] MMT at [95]. [36] MMT at [104]-[105]. [37] CA at [22]. [38] MMT at [127]. [39] MMT at [122]. [40] MMT at [107]. [41] MMT at [154]. [42] SFC Notice at [14]; MMT at [289]. [43] SFC Notice at [14]; MMT at [269]. [44] In the current version, section 270 substitutes “inside information” for “relevant information” but is otherwise identical. [45] It was not disputed that ATML was listed on the Hong Kong Stock Exchange and that the respondents were connected persons being a director and/or an employee of ATML: SFO section 247(1)(a). See: MMT at [162]-[163]. [46] Securities Ordinance (Cap.333) section 141C(3) and Securities (Insider Dealing) Ordinance (Cap.395) section 10(3). [47] Section 53(1)(c) of which provided a defence if the person charged could show “that he would have done what he did even if he had not had the information”. [48] Report of the Insider Dealing Tribunal in International City Holdings Limited, 27 March 1986, Vol.1 at [2.9]. [49] Green v Charterhouse Group Canada Ltd. (1976) 68 DLR (3d) 592 at p.619. [50] Due to the commendable research of the Court’s Judicial Assistants. [51] (Case C-45/08), 23 December 2009; [2010] Bus. L.R. 1416. [52] [2010] Bus. L. R. 1416, Judgment at [36]. [53] MMT at [196]. [54] Section B.3 of this judgment. [55] MMT at [146], [203] and [208]. [56] MMT at [97]. [57] MMT at [128]-[129]. [58] MMT at [121]. [59] MMT at [171]-[174]. [60] MMT at [218(ii)]. [61] MMT at [218(iii)]. [62] MMT at [219]. [63] MMT at [232]-[234]. [64] MMT at [224]. [65] MMT at [225]. [66] MMT at [241]. [67] MMT at [242]. [68] MMT at [254]. [69] MMT at [255]. [70] See paragraph [37] above. [71] MMT at [218(ii)]. [72] MMT at [219]. [73] MMT at [286]. [74] MMT at [292]. [75] MMT at [271]. [76] The Tribunal was referring also to Cecilia. [77] MMT at [266]. [78] MMT at [266]. [79] MMT at [276]. [80] MMT at [255]. [81] MMT at [273]. [82] MMT at [274]. [83] MMT at [276]. [84] MMT at [278]. [85] MMT at [278]. [86] MMT at [279]. [87] MMT at [266]. [88] MMT at [279]. [89] MMT at [274]. [90] R2’s Written Case at [8]. [91] Ibid at [9]. [92] Ibid at [43]-[52]. [93] (1999) 2 HKCFAR 205 at 209H. [94] Ibid. [95] (2008) 11 HKCFAR 170 at [45]. [96] It is not, however, necessary in the present appeal to determine whether it is only in cases of compulsion that the defence can be established: cf Henry Tai Hon Leung v Insider Dealing Tribunal, unrep, CACV 333-334/2004, 3 November 2005, at [27]-[28]. [97] CA at [36]-[39]. [98] CA at [40]. [99] CA at [42]. [100] CA at [50]. [101] CA at [46]-[51]. [102] CA at [52]-[60]. [103] CA at [61]. [104] CA at [62]. [105] CA at [64]-[68]. [106] The implicated dealings took place in 2007. All references (unless otherwise stated) are to provisions of the Securities and Futures Ordinance current at the time. [107] Para 198 of the MMT report. Before the Tribunal, the Commission’s case was that insider dealing had taken place on various dates between 1 February 2007 and 6 June 2007. However the expert witness provided by the Commission in his evidence said that in his opinion the deed of assignment plus the accompanying demand letter was not price sensitive, but the statutory demand combined with the deed of assignment was price sensitive. At closing, the Presenting Officer for the commission submitted that if the Tribunal accepted this view then only trading that took place after the receipt of the statutory demand could amount to insider dealing. [108] Section 245(1). It is also a criminal offence under s 291. [109] Now, it is called inside information but there is no material difference between the two expression. [110] For the time being, I concentrate on dealing by way of purchase or sale of shares. I will later briefly consider some of the other provisions in s 270. [111] There is a similar defence in the event of a criminal prosecution: Section 292(3). [112] Unrep, CACV 333/2004 and CACV 334/2004, 3 November 2005. [113] Rogers VP, Le Pichon and Tang JJA. [114] Court of Appeal, para 27. [115] Para 33. [116] Securities and Futures Commission v Lam King Hung [2010] 2 HKLRD 623, where the official translation of the judgment is reported. [117] Para 26. [118] Judgment of Derek Pang J, para 28. [119] Page 619. [120] Arnup, MacKinnon and Howland, JJ.A. [121] Clough J (as he then was) together with Mr Gordon M Macwhinnie, and Mrs Barbara M Wong. [122] IDT Report of Re International City, 27 March 1986. [123] One should note, though it is irrelevant to the present discussion, in the current appeal, the purpose must not include any such purpose. [124] Leading counsel for the Tribunal was Mr Henry Litton QC (as he then was), and one of his juniors was the future Mr Justice McMahon. It does not appear that counsel for the Tribunal took issue with Mr Wright’s submission. [125] McCowan LJ, Jupp and Potter JJ. [126] Report, para 276. [127] Para 11.24 Consultation Document on the Securities and Futures Bill, April 2000. [128] Report, para 266. [129] Report, paras 110 to 112. [130] Report, para 53. [131] Report, paras 262 & 263. [132] Report, para 82. [133] Para 27, 2nd respondents submissions. In the presenting officer’s closing submissions, when dealing with the defence under section 271 (3) he said at 43, “[S]o far as the other Specified Persons are concerned, it appears to be their case that they would have sold the shares in any event, regardless of the knowledge of the relevant information. They also point out that others, who did not possess the relevant information, were also selling ATML shares during the material period.” The material period according to the Commission was between Feb and June 2007. [134] Report, para 123. [135] Report, para 14. [136] They had no reason not to, since the price went hyperbolic in May. [137] 2.8 million shares out of the 2005 option and 1.2 million out of the 2007 option. [138] The Presenting Officer put it succinctly and submitted that it was her case that she would have sold the shares in any event regardless of the knowledge of the price sensitive information. Para 43, the Presenting Officer's closing submissions. [139] That was not in ATML’s control. [140] The shares were heavily traded. At para 105 of the Report, the Tribunal said the turnover on 21 February (the next trading day after 16 February 07), was 133, 975, 815 shares. The turnover fluctuated. But the turnover on 11 May, was about 150 million, 14 May about 200 million, 22 May about 100 million, 28 May over 150 million, 29 May, about 150 million, 30 May about 50 million. See volume-high-low-close chart of stock between 1 February 07 and 14 December 07. [141] Joint judgment, para 75. [142] Joint judgment, para 80. [143] Considered in Henry Tai Hon Leung, para 111 above. [144] Considered in International City, para 120 above. [145] Hong Kong Hansard, 5 Oct 1977 at para 14. [146] Considered in R v Cross. Para 124 above. [147] The Conduct of Company Directors. Cmnd 7037, November 1977. [148] Koon Wing Yee v Insider Dealing Tribunal (2008) 11 HKCFAR 170, with the concurrence of the other members of the court, in proceedings under SIDO. [149] Imagine the difficulty of directing a jury on the nicety of the meaning of “using”. [150] Naturally, I believe this should be a defence under s 292(3) in a criminal prosecution. [151] And I will not pause to consider how subsequent UK legislation was or might have been influenced by the European Directive. [152] I think this referred to the Financial Services and Markets Act 2000. I will not go into it. My point is simply that the legislature never intended to model our law on the European Directives. [153] Hong Kong Hansard, 25 July 1990, at pages 165-166. 
Mr Justice Chan PJ: 1. I agree with the judgment of Lord Hoffmann NPJ. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Lord Hoffmann NPJ. Mr Justice Bokhary NPJ: 3. I agree with the judgment of Lord Hoffmann NPJ. Mr Justice Hartmann NPJ: 4. I agree with the judgment of Lord Hoffmann NPJ. Lord Hoffmann NPJ: 5. This appeal arises out of two rather clumsily drafted consent orders which have resulted in a straightforward claim for money lent turning into a litigation saga which has lasted ten years. The answer to the point raised in this appeal is in my opinion tolerably clear and we did not find it necessary to call upon Mr Anthony Houghton SC on behalf of the respondent. But the argument of Mr Daniel Fung, SC for the appellants, suggested that there are certain aspects of the law on rectification which could usefully be clarified. 6. Kowloon Development Finance Limited (“KDFL”) is a licensed moneylender. Between 1997 and 2002 it advanced some $20 million to Pendex Industries Limited (“Pendex”) on the terms of facility letters which were from time to time revised and renewed. The loans were secured by the personal guarantees of the directors, Chan Wah Sun (“Mr Chan”) and Yu Kah Yee, who were husband and wife, and by a mortgage on a property at Shatin. Repayment was to be by monthly instalments in amounts which KDFL, after consultation with Pendex, would fix from year to year. 7. In 2003 Pendex experienced financial difficulties and fell behind with its payments. It tried to negotiate a rescheduling of the debt repayments but its initial proposals were not acceptable to KDFL, which issued an originating summons claiming payment of the outstanding capital and interest. Further negotiations resulted in an agreement as to discharge of the arrears and the amounts of the monthly sums which would be payable in 2004. The agreement was embodied in a Tomlin order made by consent by Master Lung on 8 January 2004. The relevant provisions are as follows: “1. All the Defendants do pay the Plaintiff the sums of money upon signing this summons as follows:- (i) HK$340,000.00 being part of the outstanding loan and interests accurred [sic]; (ii) HK$28,000.00 being the agreed legal costs and disbursements incurred by the Plaintiff … (iii) HK$ 9,600.00 being the outstanding insurance premium of the Property. 2. All the Defendants do pay the Plaintiff 12 monthly mortgage instalments commencing on 21st January 2004 by 24 post dated cheques and presented the same to the Plaintiff upon signing of this summons as follows:- [There followed a table of sums for which the cheques were to be payable. For the first six months the monthly payments were to be HK$170,000 and for the second six months $190,000, making a total of $2,160,000 during the year.] 3. All the Defendants do pay the Plaintiff interests on the outstanding loan in the sum of HK$21,591,264.12 at the rate of 10 per cent per annum from the date of arrears up to 18th December 2003 and at the rate of 8 percent per annum thereafter; … 6. This arrangement will be commenced from the date of this Summons till 21st December 2004. If all the Defendants do pay the plaintiff the total sum of HK$2,537,600.00 [being the sum total of the amounts payable by the end of 2004] during the aforesaid period, the Plaintiff will review the amount of repayment of Mortgage instalments annually thereafter;” 8. Paragraph 6 makes it clear that although the claim was for the full amount of HK$21,591,264.12 outstanding at the date of issue of the originating summons, and although Master Lung’s order provided that “all further proceedings herein be stayed except for the purpose of carrying this Order … into effect …”, the agreement in the Schedule was in the nature of a truce rather than a general treaty of peace. The truce would expire at the end of 2004. If the defendants complied with the terms, KDFL would revert to annual reviews of the instalments repayable. If they did not, clause 6 implied that hostilities could recommence. Whether this would take the form of an application to lift the stay or the commencement of a fresh action was not spelled out. During the negotiations, KDFL’s solicitor proposed the insertion of a clause which read: “7. If all the Defendants do not comply with any term of the aforesaid arrangements during any time of the aforesaid period, the Plaintiff shall be at liberty to restore this action and all the legal costs and disbursements arising therefrom shall be borne by the Defendant on an indemnity basis.” 9. The defendants’ solicitors objected to this clause. It is not clear what exactly they objected to. Perhaps it was the prospect that a single dishonoured cheque would be fatal, even if they could find the money. Perhaps it was the costs. At any rate, KDFL’s solicitor advised his client that it would do them no harm to remove it and it was deleted. The solicitor appears to have advised KDFL that they could always start a fresh action. I think that this was sound advice. Clause 6 clearly showed that compliance with the 2004 terms was not to discharge the post-2004 indebtedness. In the event, the defendants complied with the 2004 terms and no difficulties arose. 10. Further negotiations took place after October 2004 to agree the monthly amounts to be repaid in 2005. What the parties said to each other during these negotiations was a matter in dispute at the trial of this action and I shall return to it later. For the moment it is sufficient to say that they bore fruit in the shape of another Tomlin order dated 20 January 2005 in the same action which KDFL had commenced in 2003. This time the order was made by Master S Kwang. The operative part read as follows: “1. The Defendants do pay to the Plaintiff HK$9,600.00 being the Fire Insurance Premium of the property … 2. The Defendants do pay the Plaintiff HK$325,000.00 in 12 equal monthly instalments, by way of 12 post-dated cheques commencing from 21 January 2005; 3. Notwithstanding the order of Master Lung dated 8th January 2004, all further proceedings in this action be stayed, save that the Plaintiff be at liberty to apply for the purposes of carrying into effect of this Order; and 4. The Defendants do pay the Plaintiff HK$8,000.00 as the costs of this application.” 11. Despite the rather awkward language of clause 1, it appears to have been accepted that HK$325,000 was the sum payable each month and not the total for the year. The former was a considerable increase over the $190,000 payable in each of the last six months of 2004, but the latter would have been an improbably large reduction compared with the total of $2,160,000 which had been payable and paid in the previous year. I shall come back later to the reason why the instalments had been increased. The defendants provided post-dated cheques in the sum of $325,000 each but the February and later cheques were dishonoured. In September 2005, KDFL decided to recommence proceedings. 12. On 13 September 2005 KDFL issued a summons in the 2003 action, seeking an amendment of Master Kwang’s order by substituting something along the lines of the deleted para 7 from the draft of the 2004 order, or alternatively giving KDFL liberty to execute for the unpaid balance of the 2005 instalments, or alternatively setting aside the order for mistake or irregularity. None of these applications was very promising. It was hard to see how the Master could simply amend the consent order. Rectifying it or setting it aside would be a matter for a fresh action rather than a summons in the existing one. On 12 December 2005 the summons came before Master de Souza and he gave it short shrift. 13. On 31 December 2005 KDFL sent the defendants a letter before action, demanding repayment of the full capital and interest outstanding. On 12 January 2006 Pendex wrote back expressing surprise. They said that in their view, the effect of Master S Kwang’s order was that payment of the 2005 instalments discharged the whole debt and they asked for the mortgage to be released. 14. Pausing at that point, it seems to me that KDFL could simply have commenced an action for the whole outstanding debt. The court would then have had to decide whether, as a matter of construction, Master S Kwang’s order had the effect claimed by the defendants. I think it is very improbable that a court would have concluded that a reasonable observer, reading the order against the background known to the parties and in particular the scheme of repayment under the facility letters and the 2004 order, would have come to such a conclusion. The 2005 order says nothing about the discharge of any debt. It would, I think, have been construed as simply carrying forward the 2004 truce into 2005, substituting the new instalment amounts and, as in the case of the 2004 order, leaving the rest of the indebtedness to be paid off in accordance with subsequent annual reviews. 15. On 12 April 2006, KDFL commenced new proceedings claiming to have the order of Master S Kwang set aside on the ground of mistake and payment of the sum of $19,394,898.73, said to have been outstanding as at 20 March 2006, with interest thereafter. The statement of claim was afterwards amended and re-amended to include a claim for rectification of the order, substantially to make it clear that the remaining debt had not been discharged. 16. At the trial before Deputy High Court Judge Simon Mayo, which lasted 6 days, KDFL advanced four alternative cases. First, that as a matter of construction, the two Tomlin orders had not discharged the outstanding debt. Secondly, that one should imply a term into the orders to the effect that they did not discharge the outstanding debt. Thirdly, that the orders should be rectified to make it clear that they did not discharge the outstanding debt. Fourthly, that the 2005 order should be set aside on the ground of mistake and the plaintiff allowed to enforce the agreement as contemplated in the 2004 order. 17. I have already indicated that in my opinion the first argument should have succeeded. But the judge said (at para 175 of his written judgment) that “the interests of justice will be better served” if he left this point open and decided the rectification case instead. He did not enlarge upon his reason, but I think he probably meant that the rectification case gave him the opportunity to express views on the merits and the credibility of the witnesses which would not have been relevant if he decided the case simply as a question of construction. In the particular circumstances of the case, in which the judge had heard all the evidence on rectification and was in a position to make findings of fact, that was a reasonable decision. He also did not deal with the implied term argument, which was really the construction argument dressed up in different clothes. 18. The third argument was that the 2004 and 2005 orders ought to be rectified on the grounds of mutual, or alternatively unilateral mistake. As we have seen, the 2004 order did not require any rectification. Clause 6 was inconsistent with a discharge of the post-2004 indebtedness. It was the 2005 order upon which the defendants relied and which KFDL needed to rectify. 19. I will come in a moment to the judge’s findings of fact, but first I should say something about the legal requirements of mutual and unilateral mistake in an action for rectification. They sound like two varieties of mistake about the same thing, made in the one case by both parties and in the other by only one of them. But they are actually the expression of quite different principles. They deal with different kinds of mistakes. In the case of mutual or common mistake – the adjectives are in this context interchangeable – the mistake is about whether a written document correctly reflects what the parties had, on an objective assessment, agreed it should contain. As Denning LJ said in the well known case of Frederick E Rose (London) Ld v William H Pim Jnr & Co. Ld [1953] 2 QB 450, 461: “Rectification [for mutual mistake] is concerned with contracts and documents, not with intentions”. In Lovell & Christmas Ltd v Wall (1911) 104 LT 85, 88 Cozens-Hardy MR described rectification for common mistake as “a branch of the doctrine of specific performance”. By this he meant that if parties have agreed to execute a document in certain terms and by mistake it contains different terms, the court can specifically perform the prior agreement by rectifying the document. There was accordingly at one time a view that the remedy of rectification was available only if the prior agreement was itself actionable (like an agreement to grant a lease) and not, for example, an agreement subject to contract. But this was disavowed by the Court of Appeal in Joscelyne v Nissen and Another [1970] 2 QB 86. Nevertheless, it is true to say that the concept of rectification for common mistake involves carrying into effect what the parties appear to have actually agreed that the document should say. And in deciding what the parties have agreed, the common law adopts its usual objective stance, looking at what a reasonable observer would have understood the parties to mean and not concerning itself with their uncommunicated states of mind: Chartbrook Ltd and Another v Persimmon Homes Ltd and Another [2009] AC 1101. 20. Rectification for unilateral mistake, on the other hand, is very much concerned with the subjective states of mind of the parties. If the contract contains a provision which one party knows that the other party thinks is not there, or knows that the other party is mistaken about its meaning, the court may, as a matter of discretion, either refuse to allow him to enforce the contract as it would ordinarily be construed (Hartog v Colin and Shields [1939] 3 All ER 566) or go further and rectify the written agreement to give effect to what the mistaken party thought had been agreed (A Roberts & Co. Ltd and Another v Leicestershire County Council [1961] Ch 555). A civilian system of law would deal with such a case as a breach of the principle of good faith in contractual negotiations. To claim to enforce a contract in terms to which you know the other party never meant to agree is a breach of good faith. The common law has no such general doctrine of good faith in negotiation but a number of individual rules which provide remedies against specific forms of bad faith. Rectification for unilateral mistake is one of these: compare Bingham LJ in Interfoto Picture Library Ltd v Stiletto Visual Programmes Ltd [1989] QB 433. 21. The difference between the two grounds for rectification may be illustrated by the facts of Rose v Pim. The plaintiff was a London merchant who placed a written order for Moroccan horsebeans in the belief that his Egyptian buyer would accept them under the description “feveroles”. He had discussed this with the seller, another London merchant who was of the same opinion. But the parties were mistaken. In Egypt, horsebeans and feveroles are different. The Court of Appeal refused to rectify the order by substituting “feveroles” for “horsebeans” because the document did not differ from what, to all outward appearances, the parties had agreed. They had agreed on a sale of horsebeans and the order document said “horsebeans”. On the other hand, if the seller knew that the buyer mistakenly thought that it was a term of the contract that horsebeans could be sold as feveroles, a court might have thought he had contracted in bad faith and that the order should be rectified on the ground of unilateral mistake. 22. Some commentators have expressed surprise that a party might find that, as a result of rectification on grounds of mutual mistake, he is bound by a contract which is not only different from the terms of the final document but is one which, subjectively, he never intended to agree to. That is what happened in the Chartbrook case. But Chartbrook was by no means the first time that this had happened: see, for example, George Cohen Sons & Co. Ltd v Docks and Inland Waterways Executive (1950) 84 Ll L Rep 97. Objective interpretation of contractual agreements is a fundamental principle of the common law. In Daventry District Council v Daventry & District Housing Limited [2012] 1 WLR 1333, Toulson LJ (as he then was) expressed some sympathy with these academic comments on Chartbrook. However, he also quoted the well known passage from the judgment of Blackburn Jin Smith v Hughes (1871) LR 6 QB 597, 607, which is the classic statement of the principle of objective interpretation: “If, whatever a man’s real intention may be, he so conducts himself that a reasonable man would believe that he was assenting to the terms proposed by the other party, and that other party upon that belief enters into the contract with him, the man thus conducting himself would be equally bound as if he had intended to agree to the other party’s terms.” 23. Thus in cases in which there is no intention to embody the agreement in some formal document, a party may well find himself bound by terms which, subjectively, he did not intend to agree to. Why should it be different because the parties have agreed to record those terms in a written instrument? The function of the court is to make the document accord with what the parties objectively agreed. It is not necessary for this purpose to show that in resisting rectification the other party is acting in bad faith. He may have been entirely in good faith in thinking that the written document reflects what was agreed, but that makes no difference. Importing notions of good faith into rectification for mutual mistake does not recognise that the important difference between mutual and unilateral mistake lies in what the mistake must be about. In mutual mistake, the mistake is about whether the document correctly reflects the terms previously agreed. In unilateral mistake, it is about the mistaken belief of one of the parties, known to the other, about what the contract said or meant. 24. Before leaving this discussion of general principles, I should emphasise that in claims for rectification of contracts for mutual mistake, it is necessary for the court to be confident that the formal document does not reflect what was previously agreed. As Denning LJ said in Rose v Pim, 461, “if you can predicate with certainty what their contract was, and that it is, by common mistake, wrongly expressed in the document, then you rectify the document; but nothing less will suffice”. It is common for commercial agreements to be preceded by heads of agreement, term sheets or the like, followed by further negotiations to arrive at a final expression of the contractual obligations of the parties. As Hobhouse LJ explained in Britoil plc v Hunt Overseas Oil Inc & Ors [1994] CLC 561, you do not construe the earlier heads of agreement as if they were a contract and simply compare them with the final document. If there is room for ambiguity in the heads of agreement or if they might have been varied in the course of subsequent negotiations, a claim for rectification must fail. The heads of agreement are only part of the material upon which the court must decide whether it can “predicate with certainty” what an objective observer would have thought that the parties had agreed and continued to agree to record in the final document. 25. I come now to the evidence and the judge’s findings of fact. The witnesses for KDFL were Monica Wai (“Ms Wai”) and Cindy Yu (“Ms Yu”). Ms Yu reported to Ms Wai, whom she described in evidence as “the decision maker”, but the actual negotiations which led to the two consent orders were conducted by Ms Yu for KDFL and Mr Chan for Pendex. Ms Yu reported these to Ms Wai, who gave KDFL’s consent to the Tomlin orders, but she had no independent knowledge of the negotiations. The judge accepted Ms Yu as an honest and careful witness. Her evidence was supported by notes she had made at the time. The object of the discussions was to agree upon an amount for instalment payments which would enable the entire advances to be repaid within a time acceptable to KDFL. Thus the discussions which preceded the 2005 order began with Ms Yu telephoning Mr Chan and saying that the monthly payments would have to be substantially increased if the advances were to be repaid within a reasonable time. Mr Chan offered $210,000 a month for the first 6 months of 2005 and $230,000 for the second. After obtaining a valuation of the mortgaged property, Ms Yu calculated that if repayments were made at the rate offered by Mr Chan, it would take nearly 8 years for the debt to be reduced even to the value of the mortgaged property. She reported this to Ms Wai, who said it was unacceptable and that the instalments should be in an amount sufficient to repay the debt within 5 years. That worked out at $325,000 a month. 26. Ms Yu reported these calculations to Mr Chan, who said he had no option but to agree. Ms Yu’s evidence was that there was never any question of the entire debt being discharged by the payments in 2005. This would have been quite inconsistent with the basis upon which the amount of the instalments had been calculated. Mr Chan, she said, knew this perfectly well. 27. Ms Wai also gave evidence to confirm those parts of Ms Yu’s evidence which were within her own knowledge. 28. Mr Chan gave evidence for Pendex. He said it had been agreed that if he made the payments for 2004 and 2005, no further payments would be required and the action would be discontinued. The judge did not regard Mr Chan as an honest witness and rejected his evidence. He found that Mr Chan at all times knew that KDFL had no intention of releasing the rest of the indebtedness. 29. The judge started his discussion by saying “it seems likely that it was a common mistake”. In my opinion there was ample evidence for this conclusion. First, on the evidence of Ms Yu, which the judge accepted, an objective assessment of what the parties had agreed did not include any term that the post-2005 debt was to be released. Secondly, the parties appeared to have agreed that the result of their agreement should be embodied in the Tomlin order. Therefore, so far as the Tomlin order could be construed as implying such a release, it did so by mistake. 30. The judge went on to consider unilateral mistake. Having regard to the opinion he had formed of Mr Chan, that was how he preferred to decide the case. He said Mr Chan always knew that it was never the intention of KDFL to release its debt and that it was misconduct (“sharp practice”) on his part to use the form of the Tomlin order (assuming it admitted of such a construction) to support a claim that they had done so. Again, I think there was ample evidence for this conclusion. 31. The Court of Appeal (Tang VP, Yuen JA and Lam J) upheld the judge’s decision on the ground of common mistake. They had some doubts about the finding of unilateral mistake. I entirely agree with the reasoning and conclusions of the Court of Appeal on the question of common mistake. But in my opinion, however, the evidence before the judge was also sufficient to justify rectification on the ground of unilateral mistake. 32. Mr Daniel Fung’s main submission before this Court was that common and unilateral mistake were mutually inconsistent. KDFL was obliged to elect which one it wanted to rely upon. In making this submission, I think that Mr Fung was inclined to assume that the difference was simply that in one case the mistaken belief was shared by both parties and in the other it was held by only one. Obviously, if you assert that the mistake was common to both parties, it is inconsistent to say that it was only on the part of one of them. It would not be unreasonable to say that you should make up your mind whether your case is the one or the other. But the reason why I have taken some time to explain the difference between common and unilateral mistake is because this is an oversimplified statement of the difference between them. The difference does not lie only in whether one or both parties held a mistaken belief but also in what the mistake must be about. Accordingly, there is no inconsistency in asserting that the parties had made an agreement including a term which the document has mistakenly left out, but if no such objective agreement can be proved, the one party knew both that the term had been omitted and that the other party thought it had been included. This is what happened in George Wimpey UK Limited v Vic Construction Limited [2005] BLR 135, in which a claim for rectification for common mistake was abandoned at the opening of the case but the claimant tried instead (unsuccessfully, in the end) to prove that he had made a mistake of which the defendant knew. 33. Mr Fung said that in that case the claimant made his election in opening to rely on unilateral mistake. But I see no reason why a party should not run both cases and submit to the judge that if he finds that the facts do not justify rectification for common mistake, he should consider whether they support a case of unilateral mistake. 34. I would therefore dismiss the appeal. Mr Justice Chan PJ: 35. The Court unanimously dismisses the appeal with costs. Mr Daniel R Fung SC and Mr Frances Lok, instructed by Robertsons, for the appellants Mr Anthony Houghton SC and Mr CY Li SC, instructed by Vincent TK Cheung, Yap & Co., for the respondent 
Chief Justice Ma: 1. I agree with the judgment of Lord Hoffmann NPJ. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Lord Hoffmann NPJ. Mr Justice Tang PJ: 3. I agree with the judgment of Lord Hoffmann NPJ. Mr Justice Fok PJ: 4. I agree with the judgment of Lord Hoffmann NPJ. Lord Hoffmann NPJ: Introduction 5. In 2005 Innomaxx Biotechnology Group Limited (“Innomaxx”) was a Bermuda company listed on the Hong Kong Stock Exchange which carried on, through a subsidiary, the business of blood cord banking. It stored blood from the umbilical cords of newly born infants as a source of stem cells for potential future therapeutic use. Its Chief Executive and one of its directors and principal shareholders was the first defendant Luk Kin Peter Joseph (“Mr Luk”). 6. In 2006, after a reverse takeover organized by the South African finance company Investec Plc, Innomaxx morphed into a mining company. It acquired a titanium mine in Shanxi Province and Investec became its principal shareholder. It changed its name to China Mining Resources Group Limited (“China Mining”). On 7 July 2007, Mr Luk resigned from the board and as Chief Executive. His functions were taken by over by Richard Leung (“Mr Leung”), the Chief Financial Officer and a nominee of Investec. But Mr Luk continued to be a director of the subsidiary carrying on the blood cord business. 7. After the takeover, China Mining bought another mine in north-east China. Investec and Mr Leung decided that its future lay in mining rather than biotechnology. They were also concerned about the potential liabilities of the blood cord banking business. It had taken payments in advance in return for undertaking to store the customer’s blood for many years against possible future need. If it became unable to provide the necessary facilities, the customer might ask for his money back. China Mining therefore wanted to sell the business. 8. In July 2007, after Mr Luk’s resignation from the board of China Mining, Mr Leung proposed to him that he should buy the blood cord banking business. However, as China Mining was a listed company, it had to comply with the Stock Exchange Listing Rules. Mr Luk was a person “connected” with China Mining for the purposes of the Rules and it would be therefore necessary for China Mining to disclose this fact to the Stock Exchange, secure the advice of independent financial advisers and obtain the approval of the independent shareholders in general meeting. 9. Perhaps for this reason, the sale to Mr Luk did not proceed. In the following year, the proposal to sell the blood cord banking business was revived. This time, Mr Leung suggested that Mr Luk find a third party buyer. He said he would look out for someone. In late October 2008, he informed Mr Leung that a company controlled by his wife’s aunt, Ho Pui Fan (“Miss Ho”) was interested. A price of HK$15 million was agreed. 10. There was then a discussion of how the sale would be structured. The subsidiary carrying on the blood cord banking business was Cell Therapy Technologies Centre Limited (“Cell Therapy”). It was not however an immediate subsidiary of China Mining. There was an intermediate holding company called Biogrowth Assets Limited (“Biogrowth”). Biogrowth had no function except to hold the shares in Cell Therapy and at first it was proposed that China Mining would sell the shares in Biogrowth. But then it was decided that Biogrowth would remain in the China Mining group and sell Miss Ho’s company the shares in Cell Therapy. From a practical point of view, it made no difference. What was being sold was the blood cord banking business. 11. The board of Biogrowth, as seller, therefore had to authorize the sale of its shares in Cell Therapy to Miss Ho’s company, United Easy Investments Limited (“United Easy”). Furthermore it was necessary that the resolution declare any interest of its directors in the transaction. If there was none, China Mining would be able to notify the Stock Exchange that United Easy and its ultimate beneficial owners were independent of both the company and its connected persons. 12. The directors of Biogrowth were Mr Luk and the second defendant Yu Oi Kee (“Miss Yu”), an accountant who had been Financial Controller and Company Secretary of Innomaxx before the mining takeover. The group’s solicitors drafted the documents necessary for the sale and sent Mr Luk and Miss Yu a draft board minute authorizing the sale and, under the heading “Declaration of Interests”, stating “It was NOTED that none of the Directors was interested in the transactions herein contemplated”. The draft had a note saying “Please confirm”. Miss Yu duly confirmed. Before completion, Mr Luk and Miss Yu signed the Biogrowth board minutes (“Board Minutes”). The Charges 13. After an investigation by the Independent Commission Against Corruption, it was alleged against Mr Luk and Miss Yu that the declaration in the Board Minutes was to their knowledge false. Miss Ho was a nominee for Mr Luk, who was the true beneficial owner of United Easy. Further, it was alleged that Mr Luk had bribed Miss Yu to co-operate in publishing the false statement by offering her 1.5 million of his shares in China Mining. Three charges were laid against them. The first, against both, was that they conspired to commit an offence under section 9(3) of the Prevention of Bribery Ordinance Cap 201 (“the Ordinance”): “Any agent who, with intent to deceive his principal, uses any receipt, account or other document – (a) in respect of which the principal is interested; and (b) which contains any statement which is false or erroneous or defective in any material particular; and (c) which to his knowledge is intended to mislead the principal, shall be guilty of an offence.” 14. The document alleged to have been used was the Biogrowth Board Minutes. 15. The second charge was against Mr Luk in relation to the alleged bribe, under section 9(2)(a) of the Ordinance: “Any person who, without lawful authority or reasonable cause, offers any advantage to any agent as an inducement to or reward for or otherwise on account of the agent’s – (a) doing or forbearing to do, or having done or forborne to do, any act in relation to his principal’s affairs or business… shall be guilty of an offence.” 16. The third charge was against Miss Yu for accepting the bribe, contrary to section 9(1)(a) of the Ordinance: “Any agent who, without lawful authority or reasonable excuse, solicits or accepts any advantage as an inducement to or reward for or otherwise on account of his – (a) doing or forbearing to do, or having done or forborne to do, any act in relation to his principal’s affairs or business… shall be guilty of an offence.” 17. Section 2(1) of the Ordinance defines “agent” to include “any person …acting for another”. The Trial 18. At the trial in the District Court before His Honour Judge Browne, the issues on the first charge appear to have been (a) whether Mr Luk was in fact the beneficial owner of United Easy and whether Miss Yu knew this to be the case; (b) whether they used the Board Minutes to deceive and (c) whether the Board Minutes was an “other document” within the meaning of section 9(3) of the Ordinance. The judge answered all these questions in the affirmative and convicted. 19. On the bribery charges there was no dispute that on 3 November 2008, Mr Luk had given Miss Yu a certificate for 1.5 million shares in China Mining. The issue was whether it was given as a reward for past lawful services or as an inducement to secure her participation in the conspiracy alleged in the first charge. The judge found that the latter was the case and convicted both defendants. 20. The particulars of the first charge stated that the defendants were agents of Biogrowth, a wholly owned subsidiary of China Mining and that they acted with intent to deceive “Biogrowth and/or [China Mining]”. No one seems to have thought that the distinction between the two companies made any difference and of course from a practical point of view that was right, because Biogrowth, although in law a separate legal person, was in practice merely a vehicle through which China Mining held the blood cord banking business. The judge recorded in his reasons for judgment that the question of agency was not in dispute. He said that they were agents “of Biogrowth and [China Mining]”. Court of Appeal 21. In the Court of Appeal however, the separate legal personalities of China Mining and Biogrowth took centre stage. The defendants submitted that they could not be agents of China Mining because they were not its directors or employees. They owed it no legal or fiduciary duties. They were directors of Biogrowth and owed it fiduciary duties, but they could not have intended to deceive Biogrowth because they were themselves the only persons through whom Biogrowth could be deceived. As they could not deceive themselves, the charge had to fail. Likewise on the bribery charges, section 9(4) of the Ordinance provides a defence that the advantage in question was accepted or offered “with the permission of [the recipient’s] principal” and the defendants, as directors of Biogrowth, must be taken to have given themselves permission. As they were not directors of China Mining, it was not their principal and they did not need its permission. 22. The Court of Appeal (Lunn VP, Macrae and McWalters JJA) accepted the submission that the defendants, as the only directors of Biogrowth, could not have intended to deceive it: paragraph 115. But they held that the defendants were also, and more relevantly, agents of China Mining. The definition section of the Ordinance provided that the term “agent” included “any person…acting for another”. In finding a purchaser for Cell Therapy, they were acting for China Mining. As Lunn VP put it (at paragraph 99 of the Court of Appeal judgment): “Obviously, in providing Richard Yeung with [United Easy] as the prospective purchaser of [Cell Therapy], in response to his request for a purchaser to be identified, [Mr Luk] took the first step, of a number of steps that he and then [Miss Yu] took, that evidence the fact that they were acting for [China Mining]. No doubt, there was an element of overlap in the capacity in which they did so. Clearly the fact that the appellants were directors of [Biogrowth] did not prevent them acting for [China Mining] in processing the sale of [Cell Therapy].” 23. Likewise, in relation to the bribery charges, the Court of Appeal held that Miss Yu was given the shares for participating in a conspiracy against China Mining as her principal and therefore any deemed permission she may have from Biogrowth was of no avail. The Court did not decide whether such permission should be deemed to have been given. 24. Finally, there was the question of whether the Board Minutes was a “document” for the purposes of section 9(3). The judge had held that it was. In the Court of Appeal, the appellants contended that “other document” should be confined to documents eiusdem generis as receipts and accounts, the genus being “financial documents”. Lunn VP, after referring to some authorities, held that the genus was wider. It included “business records within the relationship of principal and agent”. The minute, he said, fell within this category. Macrae JA agreed. McWalters JA thought there was no restriction to a genus and that “any…other document” meant any other document. The Certified Questions 25. The Court of Final Appeal gave leave to appeal on the following questions of great and general importance: (1) What is the meaning of “agent” for the purposes of s.9 of the Prevention of Bribery Ordinance, Cap 201 (“the POBO”) and specifically whether a person who is under no legal, contractual or fiduciary obligation to act in relation to the affairs or business of another is that person’s “agent” for the purposes of s.9? (2) Whether the words “or other document” in s.9(3) of the POBO are to be construed ejusdem generis with the words which precede them and specifically whether board minutes are documents which come under this definition? (3) Whether or not the common law principles as espoused in R v McDonnell [1966] 1 QB 233 and Attorney-General’s Reference (No. 2 of 1982) [1984] QB 624, in the context of conspiracy to defraud or theft, namely, the principle that the mind and will of the sole director(s) are treated in law as the mind and will of the limited company, applies in the context of the statutory anti-corruption regime under the POBO and, in particular, to the offence of conspiracy for agents to use a document with intent to deceive their principal, contrary to sections 9(3) and 12(1) of the POBO and sections 159A and 159C of the Crimes Ordinance, Cap. 200. Charge 1: Conspiracy 26. I shall first deal with the conspiracy charge. For the purpose of the conspiracy charge, the first question is whether the judge and the Court of Appeal were right to hold that the appellants were agents of China Mining, or at any rate that Mr Luk was and that Miss Yu knowingly participated with him in using a document to deceive it. This turns upon the definition of “agent”, which includes “any person…acting for another”. 27. Counsel for Mr Luk submitted in his printed case that these last words could not be given their conventional meaning. A person cannot, he said, be an agent for the purposes of the Ordinance unless there is a pre-existing contractual or fiduciary relationship which obliges him to act. “[M]erely because a person acts in accordance with another’s request does not render him an agent of that other unless he is bound to act upon [it]”. 28. There is however a consistent line of authority in Hong Kong which makes it clear that no pre-existing duty is required. Acceptance of a request to act may itself create a duty to do so honestly and in good faith. 29. In R v Chong Chui Ha [1997] 4 HKC 518 a firm of estate agents received an offer of the asking price for a flat which they were marketing. Later someone else made them a higher offer. An employee of the estate agents told the representative of the first buyer that if he paid her $200,000, she would not pass on the higher offer to the owner. Charged with soliciting a bribe contrary to section 9(1), the defendant submitted, as in this case, that she was not an agent for the purposes of the Ordinance: “it is not sufficient for a person merely to act for another person”. It is true that she had no power to contract on the owner’s behalf. Nor was she obliged to do anything. If she had not lifted a finger to find a buyer, she would not have been in breach of contract. Nor was the owner going to pay for her services. She was hoping to share the commission of the buyer’s agent. But Keith J held, rightly if I may respectfully say so, that by accepting instructions she came under a duty, so far as she acted for the owner at all, to do so honestly and in his interest. That made her his agent for the purposes of the Ordinance. 30. It is not even necessary that there should have been a request to act. A person who is in a position to act on behalf of another and voluntarily does so may also thereby assume fiduciary duties. In Hung John Terence v HKSAR (unreported, FAMC 85/2010) a voting member of the Jockey Club solicited and obtained a bribe for proposing someone for full membership. He was convicted under section 9(1)(a) of the Ordinance and his appeal was dismissed by the Court of Appeal. He petitioned for leave to appeal to the Court of Final Appeal on the ground that he had not acted as agent of the Jockey Club or anyone else. Bokhary PJ said (at paragraph 11): “In our view, it is plain beyond reasonable argument to the contrary that a voting member is acting for the Club when he proposes someone as a suitable candidate for full membership. It follows…that the voting member is an agent, that the Club is his principal and that he was doing something in relation to its affairs.” 31. Mr Hung was not under any contractual or fiduciary duty to propose members of the Club. He had not even been asked to do so. But, having chosen to recruit a member on behalf of the Club, he came under a duty to do so in good faith and not to exploit his position to obtain a bribe. 32. Mr McCoy SC, on behalf of Miss Yu, commended to the Court the observations of Finn J in his work on Fiduciary Obligations (1977) and his judgment in Grimaldi v Chameleon Mining NL and another (No 2) (2012) 287 ALR 22. The learned judge points out in his book that a fiduciary is a person who has assumed a relationship with another which attracts a rule requiring him to act in accordance with what are usually described as fiduciary duties: not to take bribes or secret commissions, not to give false information and so on. In the Grimaldi case he said (at paragraph 177): “[A] person will be in a fiduciary relationship with another when and in so far as that person has undertaken to perform such a function for, or has assumed such a responsibility to, another as would thereby reasonably entitle that other to expect that he or she will act in that other’s interest to the exclusion of his or her own or a third party’s interest...” 33. I have no difficulty in accepting that this describes at least one of the ways in which a person may become an agent for the purposes of section 9. And I think it clearly applied in this case. Having agreed with Mr Leung, acting for China Mining, that he would find a buyer for its unwanted blood cord banking business, Mr Luk created a reasonable expectation that he would act in the interest of China Mining and to the exclusion of his own interest. More specifically, he assumed a duty to act in good faith and not deceive China Mining into making a false statement to the Stock Exchange. Miss Yu was aware that this was the basis upon which he was putting forward United Easy as a buyer and participated in his deception. That is sufficient for liability under Charge 1. 34. This shows some ambiguity in the first certified question. If it asks whether someone who acts on behalf of another must have had some pre-existing legal, contractual or fiduciary obligation to act, the answer is no. But if one asks whether such an obligation must exist, perhaps merely as a result of having acted, at the time when he uses the document in question, the answer is yes. 35. This conclusion makes it strictly unnecessary to decide whether the defendants could also have deceived Biogrowth. The Court of Appeal said they could not because the concept of deception necessarily refers to a state of mind. It must have the effect of making someone think that something is the case when in fact it is not. It would therefore be necessary to find someone whose deceived state of mind counted as the mind of Biogrowth. Since, in the view of the Court of Appeal, the only candidates for having their states of mind attributed to Biogrowth were the two defendants, who knew the true state of affairs, there was no way in which the company could be said to be deceived. 36. I do not think this is right. The offence under section 9(3) is not deceiving but using a document with intent to deceive the principal, i.e. Biogrowth. There is no reason why the deceiver needs to have in mind the particular person whose state of mind will count as that of Biogrowth. In the present case, the Board Minutes which contained the false statement about Mr Luk having no interest in the transaction also authorized Mr Chen Shou Wu, an officer of China Mining, to execute the sale agreement on behalf of Biogrowth. That was an act of Biogrowth and in agreeing on its behalf to sell to United Easy, Mr Chen was deceived into thinking it was an independent third party. Why should his deceived state of mind not be attributed to Biogrowth? If the sole director of a subsidiary company forges an invoice to secure payment from the cashier, why should the deception of the cashier who pays on behalf of the company not count as deception of the company? Furthermore, it is unclear from the timetable of events on the day the transaction took place whether Mr Luk and Miss Yu (whose resignations are recorded in the minutes of a meeting recorded as having taken place at 5 p.m.) were still directors when the Board Minutes of the earlier meeting was sent on behalf of Biogrowth to the Stock Exchange. Nor should it matter. 37. Finally, in relation to the conspiracy charge, there is the question of whether the Board Minutes was a “document” for the purpose of section 9(3). Like McWalters JA, I do not see why “any…other document” should not mean any other document. There is no need to create a genus on the basis of the references to receipts and accounts. The genus is created in paragraphs (a), (b) and (c) of section 9(3). It cannot be any document but must be a document in respect of which the principal is interested and which contains a false or erroneous statement intended to mislead. My answer to the second certified question would therefore be that no eiusdem generis construction is required and that “document” includes a board minutes. Charges 2 and 3 38. On the bribery charges, Mr McCoy SC submitted that the prosecution could not make up its mind who the principal was alleged to be. Sometimes it appeared to be China Mining, sometimes Biogrowth, sometimes both, sometimes one and/or the other. At the trial, as I have said, neither side regarded this as being a matter of any significance. On the facts of this case, I do not think it was. The only difference was that if the defendants were agents of China Mining, they plainly had no defence of permission under section 9(4). Only in relation to Biogrowth could they argue that they must be deemed to have given themselves permission. 39. Like the Court of Appeal, and for the reasons already given, I think that Mr Luk was as agent of China Mining. However, even if the agency was confined to Biogrowth, I do not think that his and Miss Yu’s knowledge of the bribe can be attributed to the company, let alone treated as a grant of permission to give and receive it. 40. The third certified question asks whether “the common law principles” as expressed in two old cases about conspiracy to defraud and theft were applicable to statutory offences under the Ordinance. That suggests that there are uniform common law principles by which one will attribute acts, knowledge, states of mind etc to a company. 41. In my opinion it cannot be too strongly emphasized that there are no such “common law principles”. The authorities since Meridian Global Funds Management Asia Ltd v Securities Commission [1995] 2 AC 500 and in particular the more recent cases of Moulin Global Eyecare Trading Ltd v Commissioner of Inland Revenue (2014) 17 HKCFAR 218 and Bilta (UK) Ltd (in liquidation) and others v Nazir and Ors (No 2) [2016] AC 1 make it clear that in every case the criteria for attribution must be such as will give effect to the purpose and policy of the relevant substantive rule, whether that rule is contained in a statute or the common law. The fact that the knowledge or state of mind of, say, a director, must be attributed to a company for the purpose of one rule (for example, imposing liability on the company to a third party) does not mean that his knowledge or state of mind must be attributed to the company for the purpose of a different rule (for example, imposing liability on the director for defrauding the company). In the case of section 9(1) and (2) of the Ordinance, it would be absurd to hold that the knowledge by the directors of their own breach of duty to the company by giving and taking a bribe was to be attributed to the company. That would defeat the purpose and policy of the rule. 42. Accordingly I do not think that the defendants were prejudiced by the indeterminacy about which of the two companies was their principal for the purposes of charges 2 and 3. I would dismiss the appeals. Chief Justice Ma: 43. For the above reasons, the appeals are dismissed. (Joseph Fok)				Permanent Judge (Lord Hoffmann)				Non-Permanent Judge Mr Edwin Choy and Mr Joe Chan, instructed by V. Hau & Chow, for the 1st Defendant (Appellant in FACC 6/2016) Mr Gerard McCoy SC, Mr Albert N.B. Wong and Mr Kim J. McCoy, instructed by Leung & Lau, for the 2nd Defendant (Appellant in FACC 7/2016) Mr William Tam SC, DDPP, and Ms Samantha Chiu, SPP, of the Department of Justice, for the HKSAR (Appellant in FACC 8/2016) 
Introduction 1. By way of judicial review, the Applicant[1] seeks to challenge the decision of the Principal Magistrate Mr Peter Law (“Magistrate”) made on 25 April 2022 (“Decision”) refusing to lift reporting restrictions for committal proceedings upon her application made under s87A(2) of the Magistrates Ordinance (Cap 227) (“MO”). When this court granted leave to the Applicant for her proceeds with this judicial review, direction was also given that the Secretary for Justice (“SJ”) be added as the Respondent[2]. The issues 2. The issues in this review are within a narrow compass, namely: (i) whether the first “shall” in s87A(2), MO bears its plain and literally meaning so that it imposes a mandatory duty on the Magistrate to lift the reporting restrictions upon an application by an accused pursuant to that subsection; (ii) if the answer to (i) is in the negative so that the Magistrate has a discretion, whether he had taken into irrelevant considerations or had failed to take into account relevant considerations in coming to the Decision. 3. On behalf of the Applicant, Mr Dykes (with him, Mr Wong) submits, in a nutshell, that s87A means what it says, to the effect that reporting restrictions must be lifted if an application is made requiring that the default position under section 87A(1) should no longer apply. Therefore, the magistrate erred in supposing that a discretion existed. It is further submitted that even if such a discretion did exist, the Magistrate’s reasoning was totally in opposition to the Principles of Open Justice that govern the exercise of judicial power in the context of restricting access to, or reporting of, court proceedings. 4. On the other hand, Mr Ma (with him, Mr Tang) submits, in a nutshell, that purposively and properly construed and in particular having regard to the importance of ensuring a fair trial, s87A(2) carries a meaning which differs from its plain and natural meaning in that it confers a discretion on examining magistrates presiding over committal proceedings whether or not to lift the general reporting restrictions imposed by s87A(1). Furthermore, it is submitted that the Magistrate had properly exercised his discretion in the matter. Basic Facts 5. The facts, which are not in dispute, are as follows. The Applicant, together with the three Interested Parties are prosecuted for one charge of “incitement to subversion”[3]. The prosecution intends to seek a committal of the defendants to the Court of First Instance for trial and for that purpose the committal bundles were served on the defence in February 2022. 6. On 14 February 2022, which was the first appointed Return Day, the Applicant made an application to the Magistrate asking that the reporting restrictions under s87A(1), MO (which is specific to committal proceedings) be lifted. Subsequent court appearances were fixed and re-fixed for argument on several issues. Eventually, on 25 April 2022 the aforesaid application was argued[4] , immediately after which the Magistrate made the Decision and gave oral reasons for it. 7. On 30 May 2022, the Applicant filed her Form 86[5] applying for leave for judicial review. On 31 May 2022, the Applicant elected for a preliminary inquiry (“PI”) the hearing of which is to commence on 2 September 2022. So far, none of the defendants have been committed to the High Court. 8. As aforesaid, leave for judicial review was granted by this court on 2 June 2022 and at the same time directions were given for the service of the papers on the Secretary of Justice and all interested parties and also for the filing of affirmations and skeleton arguments (if so advised). It is noted, however, that none of the interested parties seeks to file anything or to be heard in the present judicial review[6]. I take it, therefore, that the Interested Parties all take a neutral stand in the matter. The Magistrate’s reasons for the Decision 9. The Magistrate’s reasons, given orally in Cantonese, can be found in the transcript of his ruling[7]. He made the following points: (a) the concept of fair trial, which is provided for in the Basic Law (BL), the Hong Kong Bill of Rights Ordinance[8] (BOR) and the NSL, is the cornerstone of our legal system[9]; (b) s87A, MO, as an exception to the general principle which allows reporting of court proceedings, was enacted to realise the spirit of fair trial[10]; (c) the court has an implied power of discretion when considering application made pursuant to s87A(2)[11]; (d) the criminal case has several unique features: (i) it is a sensational case that has attracted much attention, both locally and overseas; (ii) the case had been widely reported by the media; (iii) a great number of members of public attended in court, some of them shouted in court, and that presented problems in maintaining court order and discipline; (iv) members of the society have expressed different opinions about the case, some of them are sharp and aggressive[12]; (e) if the reporting restriction is lifted, it would have to be lifted for all proceedings in order to avoid speculations[13]. That would lead to wide and sharp discussions – even attacks, before trial. Secondly, some of the people who attended court had shown reckless disregard for order. The above will certainly bring mental pressure to witnesses who will need to appear in court in the future and they may even be daunted and deterred, thus leading to a serious undermining of a fair trial[14]; and (f) on the other hand, since the case has already been widely reported, if anyone wants or needs to assist the defence in any way, eg, by giving evidence, they must have been heard about the case already and they would approach the defence if they wish. Therefore, there would be no prejudice to the defence in this regard. He did not see any other prejudice to the defence in case the application was refused and the defence had not to mention any[15]. Grounds of challenge 10. In the Form 86, the Applicant raises the following grounds of challenge: • Ground 1: Error of law on s87A / Ultra Vires It is submitted that in reaching the Decision, the Principal Magistrate mistakenly relied on a discretion he does not have under s87A, MO. • Ground 2: Error of law in the exercise of his discretion It is submitted that even if the court does have a discretion, the Magistrate erred in subverting or overlooking the principles of open justice. It is also submitted that in effectively requiring the defendant to give reasons to justify wishing for unrestricted reporting, the Magistrate subverted the fundamental principle of open justice, turning the right to a public trial into a privilege. • Ground 3: Taking into account irrelevant considerations and ignoring relevant considerations It is submitted that the Magistrate gave no consideration at all to the risk to the impartiality of potential jurors, which is the raison d'être of s87A, MO. Instead, he relied much on the conduct of the public and pressure on (prosecution) witnesses, which are entirely irrelevant. Moreover, the risk of disruption to court proceedings, which was neither raised by the prosecution nor the Applicant, was not substantiated by any evidence nor argued before the Magistrate. The risk of intimidating potential witnesses is neither real, nor linked to whether reporting restrictions are lifted or not. The Magistrate is also wrong to assume that there will be no prejudice to the defence just because the case was previously reported. Most detrimentally, the Magistrate gave no consideration at all of the importance of maintaining public confidence in the administration of justice, both locally and internationally. • Ground 4: Wednesbury Unreasonableness It is submitted that given the context of the case, the importance of open justice and the utter lack of any real risk to a fair trial that could be caused by reporting, the Decision is Wednesbury unreasonable even if the Principal Magistrate does have a discretion under s87A, MO. Committal proceedings 11. As s87A, MO is specific to committal proceedings, in order to provide the context for consideration, it would be necessary for me to give a brief outline of committal proceedings in general. 12. First of all, criminal offences in Hong Kong are divided into summary offences and indictable offences: ‑ (1) Summary offences are criminal offences other than treason, or where the words “upon indictment” or “an indictment” appear in a relevant statute, or where an offence has been transferred to the District Court for trial under Part IV of the Magistrates Ordinance (see s14A(1) of the Criminal Procedures Ordinance, Cap 221 (“CPO”)). (2) An indictable offence is an offence other than a summary one. It means a crime or offence for which a magistrate is authorized or empowered or required to commit an accused for trial before the District Court or the Court of First Instance (see s2, MO). 13. However, many indictable offences are also triable summarily in Magistrates’ Court. Where the offence is treason, or where the words “upon indictment” or “on indictment” appear and it is not further stated that the offence can be tried or is punishable either summarily or on indictment, then the offence can only be tried on indictment: s14A(2), CPO. As for an offence which is triable either way, the choice of venue would be a prosecutorial decision solely for the Secretary for Justice: BL 63; see also Chiang Lily v Secretary for Justice[16]; and Tong Ying Kit v Secretary for Justice[17]. 14. MO contains detailed provisions regarding trial by a magistrate of indictable offences and the transfer or committal, as the case may be, from the Magistrates Court of indictable offences to the District Court or the Court of First Instance. 15. By NSL 41(3)[18], the offence with which the Applicant is charged is indictable. Accordingly, her criminal case was first dealt with in the Magistrates’ Court in committal proceedings in accordance with the procedures laid down in Part III of the Magistrates Ordinance: Attorney General v O’Donnell[19]. As pointed out by Mr Dykes in his skeleton argument, except in cases where the prosecution prefers a voluntary bill, committal proceedings are a part of the process of trial on indictment. It should also be noted that committal proceedings may include a PI which is defined in s71A, MO somewhat circularly as ‘proceedings pursuant to an election by the accused to have a charge against him heard at a preliminary inquiry’. 16. At common law, a magistrate upon a preliminary inquiry into a charge of an indictable offence has a duty to determine whether there is a case against the accused which ought to be sent for trial before a judge and a jury: R v Carden[20]. As said the then Attorney General at the Second Reading of the “Criminal Procedure (Preliminary Proceedings on an Indictable Offence) Bill 1983[21]: “Traditionally, it has been regarded as an important safeguard against unfair prosecution upon flimsy or incomplete evidence for all serious criminal cases to be the subject of a preliminary investigation before the magistrates. The oral and documentary evidence said to support the charges had to be presented to the Magistrate in the presence of the accused person who could test it by cross­ examination and answer it by argument or evidence of his own. Only if a Magistrate was satisfied that there was a prima facie case to answer would the accused person be committed to stand trial in a superior court.” This jurisdiction, however, has been modified extensively by the reform introduced through the aforesaid Bill[22], the purpose of which, according to the Attorney General, is to[23]: “contribute to the saving of time and unnecessary expense in the administration of criminal justice.” 17. After the legislative reform in 1983, when an accused appears or is brought before a magistrate upon a charge of an indictable offence the magistrate “may, and shall if the prosecutor so requires”, appoint a day (return day) for the continuation of the committal proceedings: s80A(1). Since both “may” and “shall” appear together, there can be little doubt that the latter implies a mandatory duty on the part of the presiding magistrate to appoint a return day if and when the prosecution so requires. 18. By s80A(4)(c), upon first appointing the return day, the presiding magistrate “shall” inform the accused, among other things, that on the return day, he will have the right to require a PI and, if he does so require, he may, at the PI, call witnesses to give evidence on his behalf. 19. Not less than 7 clear days before the return day, or such shorter period as the accused may consent to, the prosecutor “shall serve” on the accused the documents listed in s80B(1)(a)-(d) (commonly known as the “committal bundle”). Again, this requirement for the prosecution to serve the committal bundle on the defence in a timely manner appears to be mandatory, as the failure of which would result in an adjournment in the absence of the accused’s consent: s80B(6). 20. On the Return Day, the prosecutor “shall”, if the requirements of s80B(1) are satisfied, hand into court the original of the committal bundle. The presiding magistrate “shall” then inform the accused that the prosecutor seeks his committal; and unless the accused elects to have the charge(s) heard at a PI he will be committed without such an inquiry. The accused “shall”, thereupon, be asked whether or not he elects a PI; and the accused may make such an election. 21. By s80C(2), if the accused elects, or is deemed to have elected[24], a PI, then the magistrate “shall”— (a) inform the accused that at the inquiry a witness for the prosecution whose statement was served on him under section 80B(1) will not be called to give evidence and his written statement will be received in evidence unless the accused, either then or not less than 7 days before the inquiry, notifies the magistrate and prosecutor of his desire to have the witness called; and (b) ascertain, without prejudice to any right of the accused to give notice thereafter, what prosecution witnesses the accused wishes to have called at the hearing and whether or not the accused wishes to call any witnesses of his own, and thereafter the PI shall be conducted in accordance with the provisions of ss81[25], 81A[26], 82[27], 83[28], 84 [29]and 85[30]. Dates for hearing of the PI would be fixed. 22. At this juncture, it is pertinent to note that when there is a PI, s80 becomes relevant. The provision says[31], “80. Place where examination taken not an open court The room or building in which a magistrate conducts a preliminary inquiry shall not be deemed an open court for that purpose; and it shall lawful for the magistrate, in his discretion, to order that no person shall have access to or be or remain in such room or building without the consent or permission of the magistrate, if it appears to him that the ends of justice will be best answered by so doing. ” (Emphasis supplied) 23. This section was considered in R v B Suttill Esq, Magistrate, ex parte Asiaweek Ltd & Ors[32]. Kemspter JA prefaced his judgment by referring to the famous dictum of Lord Halsbury in Scott v Scott[33]: “…It may often be necessary, in order to attain its primary object, that the Court should exclude the public … As the paramount object must always be to do justice, the general rule as to publicity, after all only the means to an end, must accordingly yield. But the burden lies on those seeking to displace its application in the particular case to make out that the ordinary rule must as of necessity be superseded by this paramount consideration. …” (Emphasis supplied) In the latter part of his judgment, Kempster JA said that s80 was calculated to demonstrate that the power of the magistrate was to confine the number of persons attending a PI to those who could reasonably be accommodated in his court; to exclude the unruly; on occasion to protect the anonymity of prosecution witnesses and to sit in camera when the security of the territory might otherwise be impaired was not open to question. Two considerations of public policy fell to be considered by the examining magistrate, namely the provision of a jury free of bias or prejudice and the interest of the public in the manner and quality of the administration of justice. In my view, based on the judgment of Kempster JA, an examining magistrate exercising his discretion under s80 would be required to conduct a balancing exercise, in which the aforesaid “principle of necessity” expounded by Lord Halsbury in Scott v Scott should not be ignored or overlooked. 24. Return to committal proceedings in general, in case the accused does not elect, and is not deemed to have elected to have a PI, then by s80C(3), the magistrate “shall” inform him that— (a) he is not obliged to say anything in respect of the charge but that he may plead guilty to the charge and that such a plea will result in his being committed for sentence on that charge; (b) unless he pleads guilty to charge, he will forthwith be committed for trial on that charge without an inquiry; and (c) after committal for trial he has the right to apply once only to a judge for discharge on the grounds that there is no prima facie case against him. 25. After complying with the above, by s80C(4), the magistrate “shall” then— (a) except where the accused is a juvenile or the offence is punishable by death, ascertain if the accused desires to plead guilty to the charge and, if he does so desire, he or she would be committed to the Court of the First Instance for sentence pursuant to s81B; and (b) if the accused is a juvenile or if the offence is punishable by death or if the accused does not then plead guilty, then he or she would be committal to the Court of First Instance for trial pursuant to s81B(2) (commonly referred to as a “straight committal” in contrast to a committal after a PI). 26. If the accused elects to plead guilty in the committal proceedings, then by s81B(2), the magistrate, before he accepts the plea “shall”— (a) require the prosecutor to outline to the accused, to the satisfaction of the magistrate, the alleged facts upon which the charge is based; (b) explain to the accused the offence with which he is charged and the ingredients which at law constitute such offence; and (c) record any statement made by the accused in pleading guilty to the charge and that the offence and the ingredients which at law constitute the offence were explained to the accused, and, if the magistrate is satisfied that the plea is made voluntarily and with an understanding of the nature of the charge, he “shall” commit the accused for sentence. It is pertinent to note that s86(1)(c) requires the magistrate to sign a certificate saying that he has observed the requirements of s81B(2) and it also provides that the certificate “shall” be transmitted to the Registrar as soon as may be after the committal. This shows that the aforesaid requirements stipulated in s81B(2) are mandatory. 27. In case the accused is committed for trial, whether after a straight committal or a PI, s85A requires that the magistrate “shall”: (a) issue his warrant committing the accused to prison until delivery by due course of law, provided that he may admit the accused to bail; (b) inform the accused, unless he is already in receipt of legal aid, of his right to apply for legal aid; (c) give the accused the alibi warning[34]; and (d) in case of a straight committal, inform him that he has the right to apply to a judge for discharge on the grounds there is no prima facie case against him[35]. By s86(1)(a)-(b), (d)-(e), there “shall” be transmitted to the Registrar as soon as may be after the committal of the accused, the written charge, the original committal bundle or depositions (as the case may be) and a certificate signed by the magistrate stating that he has observed the requirements of s85A. Again, this requirement of a certificate shows that the things that the magistrate is required to do under s85A are mandatory. 28. From all the above, it is clear that a major (if not dominant) purpose of the various procedural safeguards described above is to ensure that the accused is informed of his rights and what is going to happen. The requirement that the accused shall have the committal bundle served on him at least 7 days before the actual committal (in the absence of consent to the otherwise) is to enable him to obtain timely legal advice and to make an informed decision whether to elect a PI or to consent to a straight committal. As the Attorney General explained[36]: “It is recognized that at that early stage of electing whether to bow to the inevitability of a trial or whether to require the prosecution to present their case to the Magistrates in the traditional way, the accused person needs legal advice. For the first time therefore in the Magistrates' Court, Legal Aid is to be made available to accused persons at this early stage of the criminal process so that they may be advised in each case whether or not the dossier of evidence tendered by the prosecution discloses a prima facie case and whether or not anything can be gained by testing it before the magistrate. If past experience is a guide, it is anticipated that in the vast majority of cases the accused person will be advised to assent to the new procedure. ” 29. From the court’s experience, ever since the 1983 reform the vast majority of criminal cases committed to the Court of First Instance for trial are by way of straight committal which does not require any live witness (whether for prosecution or defence) to give evidence in committal proceedings. For committal proceedings without a PI, the role of committing magistrates becomes largely administrative in that they would not be required to consider the evidence of the charge. In R v Horsham Justices, Ex parte Farquharson[37], a similar observation was made by Ackner LJ as regards the position in UK. Section 87A, MO 30. Insofar as it is relevant to the present judicial review, s87A, MO provides as follows: “87A. Restrictions on reports of committal proceedings (1) No person shall publish in Hong Kong a written report, or broadcast in Hong Kong a report, of any committal proceedings in Hong Kong containing any matter other than that permitted by subsection (7). (2) Notwithstanding subsection (1), a magistrate shall, on an application for the purpose made with reference to any committal proceedings by the accused or one of the accused, as the case may be, order that subsection (1) shall not apply to reports of those proceedings, and any such order shall be entered in the Magistrate’s Case Register. (3) If the accused is not represented at any preliminary inquiry by counsel or by a solicitor, the magistrate shall, immediately before taking depositions of witnesses, explain to the accused the restrictions on reports of committal proceedings imposed by subsection (1) and inform him of his right to apply to the court for an order removing those restrictions. (4) Where a magistrate has made an order under subsection (2) removing the restrictions on reports of committal proceedings and has adjourned those proceedings to another day, he shall, at the beginning of the adjourned hearing of the proceedings, state that the order has been made. (5) Notwithstanding subsection (1) a report of committal proceedings containing matter other than that permitted by subsection (7) may be published or broadcast— (a) where the magistrate determines not to commit the accused for trial, after he has so determined; (b) where the magistrate commits the accused or any of the accused for trial, after the conclusion of his trial or, as the case may be, the trial of the last to be tried. (6) Notwithstanding subsection (1), where at any time during committal proceedings the magistrate assumes power to deal with the offence summarily under section 91 or 92, a report of so much of the committal proceedings containing any such matter as takes place before the magistrate assumes such power may be published or broadcast, after the magistrate has assumed power, as part of a report of the summary trial. (7) A report of committal proceedings published or broadcast without any order under subsection (2) and before the time specified in subsections (5) and (6) may contain— (a) the identity of the court and the name of the magistrate; (b) the names, addresses, occupations and ages of the parties and witnesses; (c) the offence, or a summary thereof, with which the accused is charged; (d) the names of counsel and solicitors engaged in the proceedings; (e) any decision of the magistrate to commit the accused for trial, and any decision of the magistrate on the disposal of the case of any defendants not committed; (f) where the magistrate commits the accused for trial, the charge, or a summary thereof, on which he is committed and the court to which he is committed; (g) where the committal proceedings are adjourned, the date and places to which they are adjourned; … (i) whether legal aid was granted to the accused. … (10) Subsection (1) shall be in addition to, and not in derogation from, the provisions of any other Ordinance with respect to the publication of reports and proceedings of magistrates’ and other courts.” (Emphasis supplied) 31. It should be noted that, by s86(1)(f), where the magistrate has made an order under s87A(2) (“the lifting order”), a statement to that effect “shall” be transmitted to the Registrar as soon as may be after the committal of the accused. This requirement, in my view, highlights the importance of having a proper record of the lifting order if one is made. 32. Before turning to the case law, some initial observations can be made at this juncture: (1) the reporting restrictions imposed by s87A(1) apply only to “committal proceedings”; (2) by s87A(10), the restriction imposed by s87A(1) is in addition to, and not in derogation from, other similar statutory restrictions. One such example would be s9P of the Criminal Procedure Ordinance, Cap 221, which restricts reporting of bail proceedings[38]; (3) both “shall” and “may” appears several time in various places throughout the section. Putting aside for the moment the dispute about the meaning of the first “shall” in s87A(2) (which is the subject matter of this judicial review), there can be little doubt that on every other occasion when “shall” is used in the section in contrast to “may”, it is used in a mandatory rather than permissive sense: s86(1)(f), MO. This respective usage of “shall” and “may”, as we have seen, is consistent with their usage in other sections in MO about committing proceedings; and (4) Since committal proceedings include a PI, if the accused elects one, then both s87A and s80 would be relevant. 33. The provisions of s87A, which were modelled on s3 of the Criminal Justice Act 1967 of the UK (“the 1967 Act”), were first introduced by the Magistrates (Amendment) Bill 1971[39]. As to its purpose, the then Attorney General, moving the second reading of the Bill on 2 December 1970, had the following to say, “The object, Sir, of these restrictions is to prevent potential jurors from reading details of the prosecution case, as it comes out in the committal proceedings, and from forming a prejudice against he accused. However, there may be occasions on which publicity is of some benefit to the accused, for instance if he cannot trace a witness who might see a newspaper report and come forward to give evidence as a result. Therefore, the accused is given this right to apply for an order by the magistrate removing the restrictions, though the prosecution does not have any similar right.” (Emphasis supplied) Thus, the main purpose of the section is to protect an accused from pre-trial adverse publicity which may affect the future jury. For this reason, the accused, not the prosecution, is given the right to waive this protection. Applicant’s case authorities 34. Mr Dykes starts with the common law principle, which he described as “unquestioned”, that justice must not only be done, but seen to be done: Scott v Scott, ante. It is now recognized that, subject to some well-recognised exceptions, the principles of open justice require that the contents of a public hearing can be fully reported: AG v Leveller Magazine Ltd [40]; R (Mohamed) v Foreign Secretary (No. 2)[41]; Al Rawi v Security Service[42]; and ATV Ltd v Communications Authority [43]. Against this backdrop, Mr Dykes continues and relies on the following English case authorities on the 1967 Act which, he submits, support the proposition that the lifting of reporting restrictions is mandatory at the instance of the accused. 35. In R v Russell, ex parte Beaverbrook Newspapers Ltd[44], 5 defendants were charged with a total of 40 inter-connected charges which formed the subject of one set of committal proceedings. One defendant, X, who was concerned in only six of the charges, applied for an order under s3(2) of the 1967 Act[45]. The magistrate made the order but limited its effect to those parts of the proceedings relating to X, on the basis that an unlimited order would result in publicity that might prejudice other defendants in respect of other, different charges in the proceedings. The proprietors of a newspaper applied for an order of certiorari to quash the magistrate’s limited order, and for an order of mandamus directing him to make an order in the terms of s3(2). On the question whether the newspaper proprietors had a standing to make the application, Lord Parker CJ[46] said, among other things, that where charges were so inter-connected as to be properly the subject of one set of committal proceedings, an order made under s3(2) of the 1967 applied to the totality of the proceedings; and that the magistrate had no jurisdiction to make the limited order[47]. 36. In R v Bow Street Magistrate, Ex parte Kray (Reginald)[48], before preliminary proceedings started, the accused applied for and was granted a lifting order under s3(2) of the 1967 Act in general terms. Subsequently, his charges were divided into eight different groups and there were preliminary proceedings in respect of each group before different magistrates. When the final group of charges came before the original magistrate who made the lifting order, an issue arose as to whether the proceedings would be subject to reporting restrictions. The accused asked the magistrate to retain the restriction under s. 3(1), but the magistrate ruled against him. The accused then applied for an order of certiorari to quash the magistrate’s order to lift the reporting restrictions and an order of mandamus to require the magistrate to order that the reporting restrictions had not been lifted. In dismissing the application, the Divisional Court held that: (1) an application for an order under s3(2), and the grant of such an order, may be made before the preliminary inquiry has begun; and (2) when an order is made under section 3 (2), the proceedings to which it relates, that is to say, the particular committal proceedings to which it relates, must be ascertained in the light of the circumstances prevailing at the time when the order was made. In passing, Lord Widgery LJ[49] commented on the effect of s3 of the 1967 Act and said, “So far then the position is clear; there is a general embargo on publicity in regard to committal proceedings, unless an application is made by a defendant for the lifting of the embargo, whereupon the court must make an order to that effect. … ” (Emphasis supplied) 37. In R v Blackpool Justices ex p Beaverbrook Newspapers Ltd [50], which was a multiple-defendant situation, three of the defendants applied for and were granted a lifting order. However, there were other defendants, some of them had been remanded prior to and some of them after the lifting order was made. When all the defendants, there were 10 of them, were before the justices for the purpose of taking depositions, the justices ordered that the proceedings were to be subject to reporting restrictions. As a consequence, newspaper proprietors applied for an order of mandamus for lifting the reporting restrictions or an order of certiorari to quash the restriction order. The defendants opposed the motion, contending among other things that the lifting order was ineffectual and could be withdrawn by the justices because it was applied for and made before the identity of all the defendants was known and other defendants had been brought into the proceeding after it was made. Lord Widgery CJ, giving the judgement of the court, referred to R v Russell ex p Beaverbrook Newspaper Ltd, ante; and R v Bow Street Magistrate, Ex parte Kray (Reginald), ante and said[51], “There is some authority to assist in the interpretation of this section, although it is still a relatively new one. …it suffices to say that the effect of that decision is that, where there are a number of defendants concerned in committal proceedings affecting them all, if one such defendant asks for an order under section 3 (2), that order must be made and will apply to all the defendants concerned in those proceedings. …The stark choice of whether in such an instance reporting shall be permissible in respect of all, or in respect of none, is taken by the Act of 1967, and the effect of the Act is that, if one defendant applies for an order and gets it, as he must on application, all other defendants involved in those committal proceedings are also subject to the full glare of publicity which used to apply to all defendants before 1967. … For present purposes the importance of that decision is that it lays down that an application for an order under section 3 (2), and the grant of such an order, may be made before the committal proceedings proper have begun. I use the phrase ‘committal proceedings proper’ because, by virtue of section 35 of the Act of 1967, when an accused person is brought before the magistrates' court charged with an indictable offence, the justices are to be treated as sitting as examining justices as soon as the defendant appears or is brought before the court. …” (Emphases supplied) 38. In R v Horsham Justices, ex parte Farquharson, ante, an issue arose as to whether the examining justices would have any power to re-impose the reporting restrictions after a lifting order had been made pursuant to s3 of the 1967 Act. The facts were as follows. Before the start of the preliminary inquiry, the defendants applied for and were granted an order lifting the reporting restrictions. Whilst the preliminary inquiry was taking place, s4 of the Contempt of Court Act 1981 became operative, then the justices made another order prohibiting reporting of any part of the proceedings until the commencement of any trial herein. A journalist, his union and a local paper applied for orders of certiorari to quash the second order. The Divisional Court quashed the second order of the justices but remitted the matter to the justices for them to consider whether an order should be made postponing publication of all or only part of the committal proceedings. Regarding the effect of s3 of the 1967 Act, MO, Forbes J , with whom Glidewell J agreed[52], said, “ Weighing in the balance the importance of public dissemination of information with the importance of ensuring a fair trial, Parliament decided in section 3 of the Act of 1967 that the scales moved against public dissemination and hence that section provided for a restriction on reporting of committal proceedings unless a defendant applied for that restriction to be lifted. As I read the section, once any defendant applies the justices have no discretion to refuse and, what is more, no reasons have to be given by the defendant who so applies. It would be wholly wrong to suggest – and I think there was a suggestion of it during the course of the argument – that the justices had any duty to weigh the interests of public reporting against the interests of the defence or the interests of justice. Any such balancing of interests had already been done by Parliament in enacting the provision. The cases, and I need not rehearse them, show two things. First, the interest which any co-accused might have in maintaining restriction on reporting was not a matter which justices could take into account. Indeed, as I have said, they appear to have no discretion to exercise in the matter at all. Secondly, once the application and the inevitable order had been made, there was no possibility that that order could be recalled. Even cogent and respectable reasons for a change of heart by a defendant could not result in the revocation of the order. The justices simply have no power to revoke. ” [53] “ What section 3 of the Criminal Justice Act 1967 , did was to add a new statutory offence; to make it unlawful to publish any report at all of the proceedings (other, of course, than those matters set out in subsection (4) ) irrespective of whether a report was likely to prejudice a fair trial. A successful application under section 3(2) did no more than remove this additional and very wide prohibition on reporting and thus left the position exactly as it had been before the Criminal Justice Act 1967; namely, that reports of what I shall call "prejudicial material" might still attract proceedings for contempt. Thus, although a decision to lift restrictions was irrevocable this was because the court simply had no power to make an order banning publication. It did not affect the law that publication of prejudicial material might still be contempt.”[54] (Emphasis supplied) 39. On appeal, one of the questions which the English Court of Appeal was asked to address was whether the publication of committal proceedings was governed by, and only by, the provisions of s3 of the 1967 Act. In upholding the judgment of the Divisional Court, Lord Denning said[55], “… Under the statute any one of the accused can ask for restrictions to be lifted: see section 8 (2) of the Magistrates' Courts Act 1980[56]. If he asks, the magistrates have no option. They have to make an order permitting all the committal proceedings to be reported as against all the accused: see Reg. v. Russell, Ex parte Beaverbrook g Newspapers Ltd. [1969] 1 Q. B. 342” (Emphasis supplied) Shaw LJ explained the rationale behind s3 of the 1967 Act as follows[57]: “ Before the Act of 1967, committal proceedings were freely reported subject only to the general principle that reports had to be fair and accurate and so on. … By section 3 it prohibited, subject to the exception of some innocuous or essential details, all publication of reports of committal proceedings. This prohibition was solely in the interests of the justice to the defendant… Parliament recognised, however, that in exceptional situations a defendant (who might be one of a number of accused) might have an interest in the dissemination of news of his predicament…It enacted that if a defendant (possibly one of a number) asked that the prohibition on publication should be lifted, then, ipso facto, whatever other others charged might say or the justices might think, the statutory ban on publication went…. ” (Emphases supplied) Echoing the above, Ackner LJ said, “ Section 3 (1) (section 8 (1) of the Magistrates' Courts Act 1980) provided the general restriction on reports of committal proceedings. Section 3 (2) (section 8 (2) of the Act of 1980) provided for the lifting of these restrictions on the application by the defendant or any one of the defendants. Section 3 (4) (section 8 (4) of the Act of 1980) specified those matters that may be contained in a report of committal proceedings without any order having been made under section 3 (2) to lift the restrictions. They included the identity of the court and the names of the examining justices, the names, addresses and occupation of the parties and witnesses, etc. It is clearly established that once any defendant applied the Act of 1967 gave the justices no discretion to refuse and no reasons had to be given by the defendant who so applies. The interest which any co-accused might have in maintaining restriction on reporting was not a matter which justices could take into account. Further, once the application and the inevitable order had been made, there was no possibility of recalling that order. …”[58] “… The Act of 1967 was clearly designed to provide protection to an accused against the wide publication previously permitted of committal proceedings. …”[59] (Emphases supplied) Respondent’s response to case authorities 40. In a chivalrous attempt, Mr Ma seeks to distinguish the above English case authorities by submitting that the question of the proper construction of s3 of the 1967 Act as to whether it could and should be construed as conferring a power or discretion upon the magistrates, was not really raised or argued in any of those cases. If I understand it correctly, what is being implied by Mr Ma is that in the above-quoted judicial pronouncements on the effect of s3 of the 1967 are strictly speaking obiter. 41. With respect, I am unable to accept this submission of Mr Ma. In my view the mandatory nature of s3(2) of the 1967 Act (and hence of our s87A(2)) forms an integral and essential part of the reasoning of the courts in R v Blackpool Justices ex p Beaverbrook Newspapers Ltd andR v Horsham Justices, ex parte Farquharson which cannot be dismissed as mere obiter. Furthermore, as submitted by Mr Dykes, English judges have spoken in one voice on the interpretation of the 1967 Act. To say the least, this unanimous view of those eminent judges merits a serious consideration. 42. Then, Mr Ma submits that the Courts in those English cases had apparently proceeded upon “a literal, plain and ordinary meaning” approach, in giving the word “shall” a mandatory meaning and effect. This, so the argument goes, is not the modern approach on statutory interpretation which is to start with the context and purpose of the legislation in question. 43. The above submission contains an acknowledgement, rightly in my view, that the Applicant’s interpretation of s87A(2) accords with the plain and ordinary meaning of the section. However, contrary to Mr Ma’s contention, the English courts have not lost sight of the context and purpose of the legislation under consideration. This is amply demonstrated by the judgments in Farquharson. In the judgment of the Divisional Court, Forbes J looked into the policy reasons behind s3 of the 1967. Immediately preceded the passage of his judgment which is quoted above, his lordship said[60], “ The argument on the first ground, that there was no jurisdiction to make an order under section 4 (2) in the circumstances of this case, provoked an investigation into the reasons behind the passing of section 3 of the Criminal Justice Act 1967 . We were referred to passages from speeches in the House of Lords in the Attorney-General v. Leveller Magazine Ltd. [1979] A. C. 440 , notably that of Lord Diplock, at pp. 449-450: ‘As a general rule the English system of administering justice does require that it be done in public: Scott v. Scott[1913] A. C. 417 . If the way that courts behave cannot be hidden from the public ear and eye this provides a safeguard against judicial arbitrariness or idiosyncrasy and maintains the public confidence in the administration of justice. The application of this principle of open justice has two aspects: as respects proceedings in the court itself it requires that they should be held in open court to which the press and public are admitted and that, in criminal cases at any rate, all evidence communicated to the court is communicated publicly. As respects the publication to a wider public of fair and accurate reports of proceedings that have taken place in court the principle requires that nothing should be done to discourage this. However, since the purpose of the general rule is to serve the ends of justice it may be necessary to depart from it where the nature or circumstances of the particular proceeding are such that the application of the general rule in its entirety would frustrate or render impracticable the administration of justice or would damage some other public interest for whose protection Parliament has made some statutory derogation from the rule.’ Of course, the public nature of the administration of justice and the right of the wider public to be informed by the press of what is taking place are matters of the greatest importance. But, as their Lordships point out in the Leveller case, sometimes in particular cases these matters must be subordinated to the interests of justice. A typical case in my view is afforded by the provisions of section 3 of the Act of 1967. The wide public dissemination of the evidence given in committal proceedings may work to the prejudice of the defendants at the subsequent trial because the memory of what they have read may make it more difficult for a jury to try the case solely on the evidence before them. Most juries are robust enough to surmount this difficulty but the possibility of prejudice to the defence is there.” Clearly, Forbes J was fully alive that the principles of open justice must be subordinated to the interest of justice. 44. In the Court of Appeal, Lord Denning also referred to the policy behind the UK legislations concerning report of court proceeding[61]: “ This interpretation is, in my mind, necessary so as to ensure two of our most fundamental principles. One is open justice. The other is freedom of the press. It is of the first importance that justice should be done openly in public: that anyone who wishes should be entitled to come into court and hear and see that takes place; and that any newspaper should be entitled to publish a fair and accurate report of the proceedings — without fear of a libel action or proceedings for contempt of court. Even though the report may be most damaging to the reputation of individuals, even though it may expose wrongdoing in high places, even though it may be embarrassing to the most powerful in the land, even though it may be political dynamite, nevertheless it can be published freely — so long as it is part of a fair and accurate report. The only case in which it will be punishable as a contempt of court is when the court makes an order postponing publication in the legitimate exercise of its power in that behalf. … I cannot think that Parliament in section 4 (2) ever intended to cut down or abridge the freedom of the press as hitherto established by law. All it does is to make clear to editors what is permissible and what not. In considering whether to make an order under section 4 (2), the sole consideration is the risk of prejudice to the administration of justice. Whoever has to consider it should remember that at a trial judges are not influenced by what they may have read in the newspapers. Nor are the ordinary folk who sit on juries. They are good, sensible people. They go by the evidence that is adduced before them and not by what they may have read in the newspapers. The risk of their being influenced is so slight that it can usually be disregarded as insubstantial—and therefore not the subject of an order under section 4 (2).” (Emphasis supplied) Whether or not one agrees with Lord Denning’s assessment as to the likelihood or otherwise of jury being influenced, his lordship clearly had considered the risk of prejudice to the administration of justice as a result of pre-trial reporting. 45. Shaw J explained that s3 of the 1967 Act was a “simple expedient” to deal with the prejudice which might result from pre-trial “propagation of the evidence”[62]: “ Before the Act of 1967, committal proceedings were freely reported subject only to the general principle that reports had to be fair and accurate and so on. After a notorious case which attracted much public attention it became more and more apparent that the propagation of the evidence which was almost invariably only the prosecution side of a grave accusation might prejudice the subsequent trial of that accusation by a jury. The media disseminate such matters widely and impressively. Accordingly the Act of 1967 adopted a simple expedient.” 46. As already noted above, Ackner J identified that the 1967 Act was clearly designed to provide protection to an accused against pre-trial adverse publicity. Respondent’s submission on local context and legal framework 47. Then, Mr Ma summits that our local legal context, which includes the existing state of our laws, the BL, the BOR as well as the NSL, enshrines the importance of the interests of justice and fair trial at the constitutional level and in safeguarding national security. Under such context, so the argument goes, the Court should generally lean in favour of the interpretation which affords greater protection to fundamental rights. This constitutional, fundamental rights and safeguarding national security context, it is submitted, was not present in those English cases, which were all from the pre-Human Rights Act 1998 era. 48. As regards the above submission, I fully accept that local context and legal framework are important in construing legislation. The rules of statutory interpretation are well-established and are recently encapsulated by Cheung CJ who gives the recent judgment of the Court of Final Appeal in HKSAR v Chan Chun Kit[63]. “10. The rules of statutory construction are well established. Words are construed in their context and purpose. They are given their natural and ordinary meaning with context and purpose to be considered alongside the expressed wording from the start, and not merely at some later stage when an ambiguity is thought to arise. A purposive and contextual interpretation does not mean that one can disregard the actual words used in a statute. Rather, the court is to ascertain the intention of the legislature as expressed in the language of the statute. As has been repeatedly pointed out, one cannot give a provision a meaning which the language of the statute, understood in the light of its context and purpose, cannot bear[64]. 11. Context here is to be taken in its widest sense and includes other statutory provisions and the general law. The purpose of a statutory provision may be evident from the provision itself, the recommendation of a report such as that published by the Law Reform Commission, the explanatory memorandum to the relevant bill or a statement by the responsible official of the government in relation to that bill in the Legislative Council[65]. It may also be relevant in any given case to look at the history of the provision concerned[66]. ” 49. I also fully accept that open justice is a means to fair trial and the latter is the ultimate aim itself: ATV Ltd v Communications Authority, ante. What I am unable to agree with Mr Ma is that the local context and legal framework in Hong Kong call for an interpretation of our s87A which is different from its plain meaning as supported by the English case authorities. 50. First, I do not understand that Mr Ma is going so far as to say that the right to fair trial (meaning fairness to both the prosecution and the defence) which is the main plank of his submission formed no part of the Common Law in Hong Kong before the BOR came into operation in 1991: see R v Sang[67]; and R (Begum) v Special Immigration Appeals Commission[68]. Moreover, it has always been the case, before or after the handover in 1997, that the prosecution has a legitimate interest in maintaining the fairness of the trial: Tong Ying Kit v Secretary for Justice[69]. 51. Secondly, the Common Law principles of open justice and closely connected to it, press freedom, also eminently feature in the BL, the BOR and the NSL: BL 27 Hong Kong residents shall have freedom of speech, of the press and of publication; freedom of association, of assembly, of procession and of demonstration; and the right and freedom to form and join trade unions, and to strike. BL 87 In criminal or civil proceedings in the Hong Kong Special Administrative Region, the principles previously applied in Hong Kong and the rights previously enjoyed by parties to proceedings shall be maintained. Anyone who is lawfully arrested shall have the right to a fair trial by the judicial organs without delay and shall be presumed innocent until convicted by the judicial organs. (Emphasis supplied) BOR 10[70] Equality before courts and right to fair and public hearing All persons shall be equal before the courts and tribunals. In the determination of any criminal charge against him, or of his rights and obligations in a suit at law, everyone shall be entitled to a fair and public hearing by a competent, independent and impartial tribunal established by law. The press and the public may be excluded from all or part of a trial for reasons of morals, public order (ordre public) or national security in a democratic society, or when the interest of the private lives of the parties so requires, or to the extent strictly necessary in the opinion of the court in special circumstances where publicity would prejudice the interests of justice; but any judgment rendered in a criminal case or in a suit at law shall be made public except where the interest of juvenile persons otherwise requires or the proceedings concern matrimonial disputes or the guardianship of children. (Emphasis supplied) NSL 4 Human rights shall be respected and protected in safeguarding national security in the Hong Kong Special Administrative Region. The rights and freedoms, including the freedoms of speech, of the press, of publication, of association, of assembly, of procession and of demonstration, which the residents of the Region enjoy under the Basic Law of the Hong Kong Special Administrative Region and the provisions of the International Covenant on Civil and Political Rights and the International Covenant on Economic, Social and Cultural Rights as applied to Hong Kong, shall be protected in accordance with the law. (Emphasis supplied) NSL 41(4) The trial shall be conducted in an open court. When circumstances arise such as the trial involving State secrets or pubic order, all or part of the trial shall be closed to the media and the public but the judgement shall be delivered in an open court. (Emphasis supplied) As such, the right to fair trial is entrenched rather than altered by the BL; and the common law principles of open justice (subject to certain well-defined exceptions) remain the norm after the handover and the NSL does not change that. In the latter regard, it is pertinent to note that the prosecution has not asked for a closed court order under NSL 41(4), nor does NSL 41(4) form any part of the Magistrate’s reasoning in arriving at the Decision. 52. Thirdly, as can be seen from the 1979 Hansard quoted above, the major object of the defaulted reporting restrictions provided for in s87A(1), MO is to prevent “prejudice against the accused”. This is not different from its English equivalent and hence the English case law is directly relevant. More importantly, it must be taken that the Legislature was aware of the existing English case authorities on s3 of the 1967 Act when it enacted s87A in 1979. As noted above, both “shall” and “may” are used in s87A, so that there can be no doubt that the Legislature appreciated at the time that the two words bear different meanings. Therefore, it is clearly evident that the Legislature, by enacting s87A in its present form, intended that the section should operate with a similar effect as its English equivalent; otherwise, it would not have been difficult for the Legislature to replace the “shall” under consideration with “may”. 53. Fifthly, as aforesaid, there is a consistent usage of “shall” in all other provisions in the MO relating to committal proceedings. This, in my view, strongly suggests that the first “shall” in s87A(2) should bear the same meaning as it appears elsewhere in the other places of the MO dealing with committals. 54. Sixthly, I have no difficulties with the proposition that the court can suitably adjust its own procedure taking into account the interests of witnesses, without compromising the principles of open justice: HKSAR v Shamsul Hoque[71]. However, the respondent’s concern that prosecution witnesses might be deterred from giving evidence in the coming trial in the absence of reporting restrictions is speculative. As noted above, except when there is a PI, return days are largely administrative in nature which do not involve examination of evidence in support of the charge. On the other hand, the hearing of a PI is not in open court and the public can be lawfully excluded if “the ends of justice will be best answered by so doing”: s80, MO. Apart from that, there are also other measures at the court’s disposal for protecting witnesses, which I will come to in due course. All in all, I am simply unable to see how reporting restrictions on its own can serve as an effective means to protect prosecution witnesses from interference or undue pressure. In my judgement, the Respondent’s contention that the lifting of reporting restrictions would lead to the frustration of the ultimate aim of doing justice is exaggerated and untenable: cf Asia Television Ltd v Communications Authority, ante. Respondent’s submission on facultative meaning 55. Mr Ma submits that, as a matter of common law, depending on the context and purpose, the word “shall” is capable of bearing a facultative meaning, and confers a power or discretion. I have no difficulties with that proposition. However, as shown by the case authorities relied upon by Mr Ma, the application of this common law principle is context-specific. Therefore, it begs the question as to whether the same is applicable to the present case. 56. In re Davis[72] was about admission of barristers in Australia pursuant to s10 of the Legal Practitioners Act 1898-1936 which provided that: “Every candidate whom the Board shall approve as a fit and proper person to be made a barrister shall be admitted as a barrister by the Court on any day appointed for that purpose. ” Although the High Court of Australia held[73] that the Supreme Court was not bound by the provision to admit to the Bar a candidate who was approved by the Board, the context in that case was entirely different from the present one. Moreover, the following passages from the judgment of Dixon J[74] (with whom Williams J agreed[75]) are, in my view, instructive: “This contention, in my opinion, gives too literal a construction of s.10. It would be absurd to lay hold of the word “shall” and interpret the provision as intending to place upon the Court the imperative duty of admitting to the Bar, without regard to any other condition, a person who showed that the Board approved him as a fit and proper person. … … But I am clearly of opinion that to treat s. 10 as it now stands as imposing an imperative duty upon the Court without regard to any other condition to admit a candidate to the Bar once he shows the Board’s approval, produces such absurd and inconvenient consequences and is so improbable an intention that some other construction should be adopted, if one is possible. I think that another meaning is open and that the real effect of the section is to provide no more than that, if the candidate has obtained approval of the Board, then his admission shall be in open Court and upon some day appointed for the purpose. Its purpose is not to entitle him to admission independently of every other consideration. When the legislation wishes to give a right to admission it uses the word “entitles,” as for example in ss. 11 and 12. ” I am not suggesting that “absurdity” is the only criteria. However, the point is that the natural and ordinary meaning of the first “shall” in s87A(2), MO does not lead to any absurdity. To the contrary, a mandatory “shall” matches the legislative intent and is also consistent with the usage of that word in the other parts of MO relating to committal proceedings. 57. In Karus v Scottish Legal Complaints Commission[76], one of the issues was whether K would still be subject to disciplinary proceedings when he had already requested that his name be removed from the roll of solicitors, although that had not yet been done. The relevant Scottish provision said, “An enrolled solicitor who wishes his name to be removed from the roll of solicitors may make an application to the Council in that behalf, and the Council shall…on the solicitor satisfying the Council that he has made adequate arrangements with respect to the business he has then in hand, remove the name of that solicitor. ” (Emphasis supplied) As to this, the Court of Session said[77], “…The word ‘shall’ has frequently been construed as permissive rather than mandatory; for example, Stroud, Judicial Dictionary of Words and Phrases … gives 17 examples of cases where the word has been so construed, usually with reference to the underlying policy of the Act in question. In the present case, we are of opinion that the policy underlying the existence of the roll of solicitors demands that the second respondent should have an element of discretion as to whether a request to remove a solicitor's name from the roll should be granted, at least immediately. Provided that good cause exists for doing so, we consider that the second respondent should be entitled to refuse a request for removal. The likelihood of significant disciplinary proceedings would provide a sufficient reason. …” (Emphasis supplied) That case is not authority for the general proposition that “shall” should always be construed as permissive rather than mandatory so as to achieve a perceived desirable result. Rather, the case shows that a strong and clear policy reason would be required before that could be done. In the present case, however, there is no evidence that the policy behind s87A is for the protection of (prosecution) witnesses. Rather, there is every evidence that the policy behind that section is to protect the accused, so that it is up to the accused to decide whether he or she would waive that protection. Conclusion on ultra vires 58. To conclude, I can see no cogent or convincing reasons for the Respondent’s proposition that the provisions in s87A, MO should bear a meaning other than their plain and ordinary meaning. To the contrary, a purposive and contextual interpretation and a consideration of the case law point in unison to the conclusion that s87A(2) means what it says in that the magistrate is at the instance of the accused under a mandatory duty to lift the reporting restrictions imposed by s87A(1). It follows that the Decision is ultra vires in that the Magistrate purported to exercise a discretion which did not exist. 59. In so ruling, I have not neglected the point made by Mr Ma that the operation of s87A(2), MO might result in unfairness in a multi-defendant situation where not all of them agree to the lifting of reporting restrictions. The effect of the above-quoted English case authorities is that the magistrate would have no discretion even when that happens. I note, however, that in the UK, provisions were inserted to ameliorate that contingency by conferring a discretion on the examining justices[78]. This, however, has not happened in Hong Kong and this may be an area of reform which the Legislature would like to consider. 60. That said, I appreciate that under the constitutional and legal framework of Hong Kong, an issue may arise as to whether s87A(2) is so wide as to violate the fair trial right of the co-accused. However, there are two reasons why this court should restrain from expressing any conclusive views on this issue. First, the Applicant’s application had not been objected to by the Interested Parties, who have all taken a neutral stand and do not seek to be heard in the present judicial review. Therefore, the present case is not a suitable vehicle for considering the issue. 61. Secondly, for the reasons which I have already given, the “discretion” which the Respondent seeks to read into s87A(2) cannot be supported by common law rules of statutory interpretation. However, whether or not a “remedial interpretation” could and should be adopted (as in cases like HKSAR v Lam Kwong Wai & Anor[79] ) has not been argued before me by any of the parties. This, too, should be left for future consideration in a suitable case. Other grounds 62. In view of my ruling on the ultra vires ground, it would not be necessary for me to deal with the other grounds relied upon by the Applicant. However, out of deference to counsel, perhaps I shall express my view on the alternative case of the Applicant. 63. Even assuming that the Magistrate did have a discretion under s87A(2), MO, so that the protection and interests of prosecution witnesses could be a legitimate consideration, I am inclined to the view that a magistrate presiding over committal proceedings should not refuse to acceded to the application of an accused under s87A(2) unless such refusal is “strictly necessary” in the interests of justice: BOR 10. In ATV Ltd v Communications Authority, Cheung CJHC (as he then was) said[80], “27. The case law has very often expressed this in terms of a requirement of “necessity”, that is, where justice would be frustrated if open administration of justice in a particular case is not restricted, then, to the extent necessary to prevent that from happening, there may be restriction on doing justice openly. 28. This requirement of ‘necessity’ is founded on the common law, and has also found expression in article 10 of the Hong Kong Bill of Rights and, in the case of the Court of Final Appeal, section 47(3) of the Hong Kong Court of Final Appeal Ordinance (Cap 484). Article 10 of the Hong Kong Bill of Rights relevantly provides that the press and public may be excluded from a hearing ‘to the extent strictly necessary in the opinion of the court in special circumstances where publicity would prejudice the interests of justice.’ ” 64. It is arguable that the exercise of the discretion (assuming there is one) under section 87A(2) would involve the application of a proportionality test; Leung Kwok Hung & Ors v HKSAR[81]; and Hysan Development Co Ltd v Town Planning Board [82]. I need to emphasize, however, that this point has not been argued by counsel before me at any length. The point is that if the Magistrate was required to consider “proportionality” in the exercise of his discretion, then he did not seem to have done that in his oral reasons. 65. Putting aside whether or not the Magistrate as a matter of law was required to consider proportionality, what make reporting restrictions “strictly necessary” would be fact and case sensitive. Nevertheless, in his oral reasons he did not seem to have considered the following matters which are, in my view, relevant: (1) Whether, and if so to what extent, reporting restrictions were effective in achieving the perceived aim of protecting prosecution witnesses; (2) if the worry was about unruly behaviour of some members of the public present in court, whether that concern could be alleviated by excluding them pursuant to s122, CPO and (in case of a PI) s80, MO and/or in an extreme case by ordering a hearing in camera pursuant to s123, CPO; (3) if the concern was about the revelation of identities of prosecution witnesses, then whether an anonymity order could help; (4) if the concern was about the risk to the impartiality of potential jurors, then there was an option open to the SJ of a trial by a panel of three judges instead of a jury trial: NSL 46; see also Tong Ying Kit v Secretary for Justice, ante; (5) whether there are civilian witnesses and if so, what the nature of their evidence is. If the prosecution witnesses are all police officers, then whether it can be expected that they are less likely to yield to improper pressure or be deterred from giving evidence; and (6) whether counsel could so tailor their speeches in court as to avoid revealing the identity of the prosecution witnesses and the contents of their evidence. As such, I am inclined to the view that even if there were such a discretion as contended by the Respondent, the Decision would still be flawed in that it entails a failure to take into account relevant considerations and as a result of which it has not been shown that the reporting restrictions were “strictly necessary” in the circumstances. However, to be fair to the Magistrate, he did not have the benefit of the assistance that this court has when he came to the Decision “on the spot”. Orders 66. Based on the above, I grant: (1) an order of certiorari to quash the Decision; (2) an order of mandamus that the Magistrate shall make an order to lift the reporting restrictions in accordance with s87A(2), MO when the Applicant is next appeared before him; For avoidance of doubt, the aforesaid orders do not affect the restrictions on reporting bail proceedings imposed by provisions other than s87A, MO. 67. In view of the result of this judicial review, I make an order nisi that the Applicant shall have the costs of these proceedings, with certificate of two counsel, to be taxed if not agreed. Mr Philip J Dykes, SC, leading Albert N B Wong, instructed by Messrs Kenneth Lam, Solicitors, for the Applicant Mr Johnny Ma, SC, leading Alexander Tang, instructed by the Department of Justice, for the Respondent The 1st, 2nd, 3rd Interested Party, did not appear [1] D4 in WKCC 3632/2021. The Interested Parties are the other defendants in the same criminal case. [2] See the Order by this court made on 7 June 2022. [3] Contrary to Articles 22 and 23 of the Law of the People’s Republic of China on Safeguarding National Security in the Hong Kong Special Administrative Region in Schedule to the Promulgation of National Law 2020 (LN 136 of 2020) (“NSL”) [4] There were no similar applications from the other defendants. [5] Dated 27 May 2022 HB/A/1. [6] Although Mr Henry Hung of the Official Receiver’s Office was holding a “watching brief” for the 1st Interested Party and Ms Catherine Wong and Mr Jay Koon (on instruction of Ho Tse Wai & Partners) were in court holding a “watching brief” for the 2nd and the 3rd Interested Parties. [7] Hearing Bundle A, Item 4 (HB/A/4). Agreed English translation is HB/A/4A. [8] Cap 383 [9] HB/A/4A/49 [10] Ibid [11] HB/A/4A/50 [12] HB/A/4A/50-51 [13] The Magistrate appears to be saying that an order for lifting of the reporting restrictions is not “once and for all”, so that there could be more than one order, each covering a particular attendance of the committal proceedings. However, as will be seen in due course, this proposition is not supported by the case authorities. [14] HB/A/4A/51 [15] HB/A/4A/52 [16] (2010) 13 HKCFAR 208 [17] [2021] 3 HKLRD 350 [18] “Cases concerning offence endangering national security within the jurisdiction of the Hong Kong Special Administration Region shall be tried on indictment.” See also HKSAR v Ng Hau Yi Sidney (2021) 24 HKCFAR 417, [27] as to the meaning of the phrase “offence endangering national security”. [19] [1985] 2 HKC 283, at 289B-H [20] (1879) 5 QBD 1 [21] Hong Kong Hansard, Session 1982/83, 1161 (27 July 1983) [22] Ord 48 of 1983, which came into operation on 11 January 1984: LN 415/83. [23] Hong Kong Hansard, ante. [24] By s80C(5): “An election by the accused to have a charge against him heard shall be deemed, if there is more than one charge, to relate to all charges. ” [25] Taking of evidence at the PI. [26] Admission of statements and exhibits in committal proceedings. [27] Taking of depositions, and caution to and statement of the accused on proceedings before the examining magistrate. [28] Evidence of the accused and defence witnesses before the examining magistrate. [29] Order by committing magistrate for attendance at trial of witnesses. [30] Discharge or committal of the accused. [31] The section was amended by s3(d), Ord 48 of 1983 by (i) deleting “hears an indictable offence” and substituting “conducts a preliminary inquiry” and (ii) by deleting “hearing the case”. [32] [1984] HKC 474 [33] [1913] AC 417, 437-438 [34] Where the magistrate has given the warning as required, his clerk “shall” give the accused written notice of the provisions of section 65D, the Criminal Procedure Ordinance, Cap 221 (“CPO”). [35] See s16 of CPO. [36] Hong Kong Hansard, Session 1982/83, 1162 (27 July 1983) [37] [1982] QB 762, at 799B-C. [38] 9P. Restriction on reports of bail proceedings (1) Unless it appears to the court that the interests of public justice otherwise require, no person shall publish in Hong Kong a written report, or broadcast in Hong Kong a report, of any bail proceedings containing any matter other than that permitted under subsection (2). (2) A report of bail proceedings may contain— (a) the name of the person being the subject of those proceedings; (b) the offence with which the person being the subject of those proceedings is charged; (c) the identity of the court and the name of the magistrate, District Judge or judge, as the case may be; (d) the names of counsel and solicitors, if any, engaged in the bail proceedings; (e) the result of the bail proceedings and where the person being the subject of those proceedings is admitted to bail subject to any condition under section 9D(2), the details of any such condition; (f) where the bail proceedings are adjourned, the date and place to which they are adjourned. [39] Which later became Ord 6 of 1971. [40] [1979] AC 440, 449H-450B. [41] [2011] QB 218, [134]. [42] [2012] 1 AC 531. [43] [2013] 2 HKLRD 354, [17]-[36]. [44] [1969] 1 QB 342 [45] Which is virtually identical to our s87A(2). [46] With whom Melford and Bridge JJ agreed. [47] Ibid, 347E-348A [48] [1969] 1 QB 473 [49] Whose judgment James and Bridge JJ agreed. [50] [1972] 1 WLR 95 [51] Ante, 98E-H. [52] Ibid, 775E-F [53] Ibid, 767G-768B, the passage in quote represented the position in UK before the Criminal Justice (Amendment) Act 1981. [54] Ibid, at 770G. [55] Ibid, at 786D-E. [56] Magistrates’ Court Act 1980, s8, which is a “successor” of s3 of the 1967 Act, provides: “(1) Except as provided by subsections (2), (3) and (8) below, it shall not be lawful to publish in Great Britain a written report, … in Great Britain a report, of any committal proceedings in England and Wales containing any matter other than that permitted by subsection (4) below. (2) Subject to subsection (2A) below a magistrates’ court shall, on an application for the purpose made with reference to any committal proceedings by the accused or one of the accused, as the case may be, order that subsection (1) above shall not apply to reports of those proceedings.” [57] Ibid, at 796A-E. [58] Ibid, at 802E-G. [59] Ibid, at 803H. [60] Ante, at 766H-767F [61] Ibid, at 793H-794H. [62] Ibid, 796A-B. [63] FACC 1/2022; [2022] HKCFA 15 (dated 15 July 2022) [64] Chan Ka Lam v Country and Marine Parks Authority (2020) 23 HKCFAR 414, [26]-[27]. [65] HKSAR v Cheung Kwun Yin (2009) 12 HKCFAR 568, [13]-[14]. [66] Town Planning Board v Town Planning Appeal Board (2017) 20 HKCFAR 196, [29] (2). [67] [1980] AC 402, at 437 [68] [2021] AC 765, [87] & [90]. [69] [2021] 3 HKLRD 350, [42] (per Poon CJHC) [70] Which is identical to ICCPR Art.14.1 which is entrenched by BL 39. [71] [2014] 6 HKC 395 [72] (1947) 75 CLR 409 [73] By a majority (Strake, Dixon, McTiernan and Williams JJ, Latham CJ dissenting) [74] Ibid, at 424, 425 [75] Ibid, at 428 [76] [2014] CSIH 59 [77] Ibid, [19]. [78] Magistrates’ Courts Act 1980, s8C. [79] (2006) 9 HKCFAR 574 [80] Ante, at [27]-[28]. [81] (2005) 8 HKCFAR 229 [82] (2016) 19 HKCFAR 372 
Mr Justice Fok PJ: 1. The appellant was convicted of one charge of dealing with property known or reasonably believed to represent proceeds of an indictable offence contrary to section 25(1) of the Organised and Serious Crimes Ordinance.[1] Her application for leave to appeal to this Court against that conviction was heard on 22 March 2016, prior to the Court hearing the appeals in Yeung.[2] In the light of those pending appeals, the Appeal Committee granted leave to appeal on the substantial and grave injustice ground on the basis it was reasonably arguable that the charge on which the appellant was convicted was duplicitous.[3] 2. After hearing counsel for the appellant, we dismissed the appeal indicating that our reasons for doing so would be handed down in due course. These are the reasons of the Court for dismissing the appeal. A. The facts 3. The appellant co-habited with her partner, surnamed Chu, who was the 1st defendant (D1) at trial below, in premises at Euston Court, 6 Park Road, Hong Kong (“the Park Road flat”). On 31 January 2011, D1 was seen by police in suspicious circumstances outside commercial premises in Yuen Long. He was arrested and found to have in his possession equipment fit for stealing. 4. A subsequent search led to the discovery, in the Park Road flat and in two bank safe deposit boxes under the appellant’s sole control, of a large quantity of jewellery, coins, foreign currency notes and other property believed to have been stolen in various burglaries, namely: (1) 290 pieces of jewellery; (2) 24 watches; (3) 166 coins; (4) 38 gold nuggets; (5) 5 foreign exchange certificates; (6) Cash of HK$399,737.20, RMB¥36,217.06, US$9,998, NTW$49,980, €13,515, $58,600 Argentine Peso, S/97,100 Peruvian Nuevo Sol and 10,000 Romanian Leu. The total value of this property was estimated to be HK$3,962,358.20. 5. The appellant was also found to have bank accounts with five banks in which, as at 1 January 2006, there was a balance of about HK$4.5 million and into which, between 1 January 2006 and 26 October 2011, a total of over HK$80 million and RMB¥2.6 million had been deposited. Particulars of the five banks where the accounts were held, the total number of deposits and total amounts of deposits are set out in the Annex to this judgment. B. The charges 6. D1 and the appellant were charged in the same indictment with a total of seven charges. D1 was charged alone with going equipped for stealing (Charge 1), burglary (Charges 2 and 3) and resisting a police officer in the execution of his duty (Charge 4), whilst the appellant was charged alone with handling stolen goods (Charge 5). 7. Both D1 and the appellant were also charged together with handling stolen goods (Charge 6) and with dealing with property known or reasonably believed to represent the proceeds of an indictable offence[4] (Charge 7). 8. At the close of the prosecution case at trial,[5] a submission of no case to answer in respect of Charges 6 and 7 was made by the defence. The Judge gave leave to the prosecution to amend Charge 7 to add some of the property from Charge 6 to the particulars under Charge 7 and then found that there was no case for D1 and the appellant to answer in respect of Charge 6 but that there was a case for both to answer on amended Charge 7. 9. So amended, the Particulars of Offence under Charge 7 read as follows: “CHU Shue-lung and LI Shuk-woon, between the 1st day January [sic], 2006 and the 26th day of October, 2011, both dates inclusive, in Hong Kong, knowing or having reasonable grounds to believe that property, namely: (i) deposited money in the sum of $8,730,235.76 Hong Kong currency and ¥2,212,789.71 Renminbi in account of Bank of China (Hong Kong) Limited held in the name of LI Shuk-woon; (ii) deposited money in the sum of $15,216,450.17 Hong Kong currency, €1,750 Euro, £1,005 Pound Sterling, $800 Canadian currency, $20,582.09 United States currency and ¥435,818.65 Renminbi in account of Hang Seng Bank Limited held in the name of LI Shuk-woon; (iii) deposited money in the sum of $36,966,989.83 Hong Kong currency in account of Citic Bank International Limited held in the name of LI Shuk-woon; (iv) deposited money in the sum of $6,118,551.61 Hong Kong currency in account of Bank of East Asia Limited held in the name of LI Shuk-woon; (v) deposited money in the sum of $12,452,452.44 Hong Kong currency in account of Public Bank (Hong Kong) Limited held in the name of LI Shuk-woon; and (vi) concealed property, namely 290 pieces of jewelleries, 24 watches, 166 coins, 38 gold nuggets, one bank [sic] of China Foreign Exchange Certificate, cash of $399,737.2 Hong Kong currency, cash of ¥36,217.06 Renminbi, cash of $9,998 United States currency, cash of $49,980 New Taiwan currency, cash of €13,515 Euro, cash of $58,600 Argentine Peso, cash of S/.97,100 Peruvian Nuevo Sol and cash of 10,000 Romanian Leu, 3 gold nuggets and 10 gold coins.” C. The proceedings below 10. After trial, D1 was convicted by the Judge of Charges 1 to 4 and D1 and the appellant were both convicted of amended Charge 7.[6] The appellant was acquitted of Charge 5 by the Judge on the footing that her awareness of the status of the 13 items of stolen property particularised in the charge as stolen goods had not been proven. D1 was sentenced to a total of 7 years’ imprisonment and the appellant was sentenced to 6 years’ imprisonment. 11. As to amended Charge 7, D1’s defence was that the property in the Park Road flat and the two safe deposit boxes belonged to him and the appellant. The appellant’s defence was that the money in the bank accounts was her earnings obtained legally and some of the property in the flat and safe deposit boxes belonged to her or were gifts from relatives or friends, whilst the rest was obtained by D1 by legal means before being passed to her. 12. The Judge evaluated the evidence in relation to the deposits in the five bank accounts and considered the items of property particularised in amended Charge 7 by reference to three groups: Group A, being those the appellant claimed were gifts from relatives and friends (other than D1); Group B, being those she claimed to have purchased herself; and Group C, being alleged to be gifts from D1 and property kept on D1’s behalf. 13. The Judge’s conclusion on amended Charge 7 was as follows: “90. The prosecution has proved beyond reasonable doubt that D1 had, over the period of 70 months in question, dealt with the deposits in the said Hang Seng and Ka Wah accounts jointly with D2, and the majority of those deposits represented, directly or indirectly representing [sic] the proceeds of an indictable offence. The prosecution has already proved beyond reasonable doubt that D2 had, over the period of 70 months in question, dealt with the accounts invovled [sic], (including dealing with the said Hang Seng and Ka Wah accounts jointly with D1), and the majority of those deposits were property directly or indirectly representing the proceeds of an indictable offence. 91. The prosecution has proved beyond reasonable doubt that D1 had, over the period of 70 months in question, dealt with the exhibits of the above Group C, and those exhibits were property directly or indirectly representing the proceeds of an indictable offence. The prosecution has already proved beyond reasonable doubt that D2 had, over the period of 70 months in question, dealt with the exhibits of the above Groups A, B and C, and those exhibits are property directly or indirectly representing the proceeds of an indictable offence.” (Italics in original) 14. On appeal,[7] both D1 and the appellant initially sought leave to appeal against the convictions and sentence but D1 later abandoned his applications. On 4 February 2015, the Court of Appeal dismissed the appellant’s applications for leave to appeal against conviction and sentence. 15. Neither at trial, nor before the Court of Appeal, was any argument raised on behalf of the appellant that amended Charge 7 was duplicitous or that there was any prejudice to her in the conduct of her defence to that charge by reason of duplicity. D. Leave to appeal to this Court 16. The appellant applied for leave to appeal to this Court on the basis of substantial and grave injustice, relying on various grounds. As noted above, in the light of the appeals in Yeung, the Appeal Committee granted leave to appeal on the basis it was reasonably arguable that amended Charge 7 was duplicitous. The Appeal Committee refused leave to appeal on the other matters raised by the appellant.[8] The sole issue in the present appeal, therefore, is concerned with the issue of the alleged duplicity of amended Charge 7. E. Duplicity and money laundering 17. In Yeung, the Court considered the issue of duplicity in relation to a charge of money laundering under section 25 of OSCO. The Court held that it was not the case that criminal behaviour can never be treated as one offence for the purpose of rules 2 and 3 of the Indictment Rules[9] if it was capable of being broken down into a number of offences. However, conversely, the prosecution does not have an unconstrained discretion such that an offence is constituted by whatever conduct the prosecution decides to make the subject of one charge.[10] 18. The Court cited (at [135]) the following passage from the speech of Lord Diplock in DPP v Merriman[11] as stating the general principle: “The rule against duplicity, viz. that only one offence should be charged in any count of an indictment, which is now incorporated in rule 4 (1) of Schedule 1 to the Indictments Act 1915, has always been applied in a practical, rather than in a strictly analytical, way for the purpose of determining what constituted one offence. Where a number of acts of a similar nature committed by one or more defendants were connected with one another, in the time and place of their commission or by their common purpose, in such a way that they could fairly be regarded as forming part of the same transaction or criminal enterprise, it was the practice, as early as the eighteenth century, to charge them in a single count of an indictment.” 19. The crucial question is how to determine whether acts are so connected that they can fairly be regarded as forming part of the same transaction or criminal enterprise. The answer, given by the Court in Yeung at [138], is that: “A judgment as to what can fairly be regarded as the scope of a criminal enterprise for the purpose of framing an indictment is likely to be affected by the nature of the crime, the circumstances of its commission, potential grounds of defence or exculpation, and the considerations of practice and procedure that will follow from adoption of one course or another.” 20. In relation to charges of money laundering, the Court held that section 25 of OSCO does not create an offence of carrying on the business of money laundering, which would be a continuing offence giving rise to no question of duplicity. It held that most of the forms of “dealing” set out in the definition in section 2 of OSCO involve conduct that would normally consist of a single act but that some forms of dealing, in particular concealing, could cover a continuing process and the individual acts could be undertaken as part of that process of concealment.[12] 21. The principle in relation to a charge of money laundering was stated by the Court at [159] as follows: “Where a number of acts of money laundering are connected in such a way that they can be regarded as forming part of the same transaction or criminal enterprise then it will be legitimate to charge them in a single count unless there is a risk of injustice to the defendant. Such injustice might lie in uncertainty or inadequate notice as to the case the defendant has to meet, confusion or prejudice resulting from different defences to different aspects of the prosecution case, problems of admissibility of evidence, or uncertainty as to the scope of an ultimate verdict. If these or other considerations that may be raised in a particular case are capable of being met, without unfairness, by appropriate measures of trial management then a court may well conclude that the rule does not prevent aggregation for the purpose of framing a charge or charges.” 22. The aspect of concealment, in a charge of money laundering, is relevant to the duplicity argument because it may provide part of the connection which makes the individual acts of dealing of a similar nature which could fairly be regarded as forming part of the same transaction or criminal enterprise.[13] 23. Even if a charge is duplicitous, it does not follow that the indictment or the relevant count is a nullity. The duplicity can be cured by amendment at trial, either to a separate count for each offence or a group of charges depending on the defences raised. Where the complaint of non-compliance with the rule against duplicity involves one of the forms of unfairness that the rule aims to prevent, then the trial judge may need to consider whether that problem can be addressed by available powers of trial management.[14] 24. At [147], the Court held: “When the point is raised for the first time on appeal, the appellate court will consider whether the objection based on duplicity is well-founded and whether the form of the indictment resulted in the risk of injustice. The appellate court will have the benefit of hindsight. It will know the nature of the defence case at trial, whereas a trial judge may have had to consider potential unfairness on the basis of the charges and particulars, and the prosecution opening. It will have the trial judge’s reasons for verdict. If no complaint about duplicity was made at trial that may have a bearing on a claim of unfairness.” (Footnote omitted) F. The contentions raised on this appeal 25. As noted above, the appellant’s complaint of duplicity was not raised at trial or in the Court of Appeal. It was raised for the first time in this Court. Mr Andrew Bruce SC,[15] on behalf of the appellant, realistically accepted that this Court’s decision in Yeung limited the scope of argument that might be advanced in this appeal. However, he contended that the circumstances of the present case did not fall within the concept of “an enterprise” as discussed in Yeung because the offence pleaded in amended Charge 7 took place over a period of 70 months and, unlike Yeung where there was only one form of property, namely money, here the subject properties particularised in the charge included deposits in 5 bank accounts, hundreds of pieces of concealed property and cash in different currencies. 26. Mr Bruce submitted that the prosecution’s case against the appellant concerning the deposits in the bank accounts did not involve proof of any predicate offence but, instead, proceeded on an invitation to the Judge to consider the background of the appellant and her relationship with D1 to draw the adverse inference against her regarding her mens rea for the offence. Only a portion of the property particularised in the charge was identified by prosecution witnesses as having been stolen and the prosecution’s case against the appellant in respect of the remaining property was simply based on her possession of them without any examination of the circumstances in which they came into her possession. 27. It was therefore submitted that, had amended Charge 7 been appropriately amended to avoid offending the rule against duplicity, the Judge would have had to consider the state of mind of the appellant at different times and in different circumstances according to the approach laid down in HKSAR v Pang Hung Fai,[16] whereas the Judge in fact adopted a global approach in assessing the evidence before him. This, it was ultimately submitted, resulted in the appellant not receiving a fair trial such that substantial and grave injustice was done to her. G. Were the acts charged part of the same transaction or criminal enterprise? 28. The proper application of the rule against duplicity in the context of a charge under section 25 of OSCO is now that laid down in Yeung as summarised in Section E above. As there noted, the rule does not give the prosecution an unconstrained discretion to amalgamate any conduct into one single charge. There must be a sufficient connection between the acts charged together such that they may fairly be regarded as forming part of the same transaction or criminal enterprise. 29. In the present case, we were satisfied that there was a sufficient connection between the separate acts of dealing with the funds deposited in the appellant’s bank accounts, as well as the jewellery and other property discovered in the Park Road flat and her safe deposit boxes. We were also satisfied that the appellant had dealt with the property particularised in amended Charge 7 as part of a continuing activity, the common purpose of which was to conceal the funds or property from an illicit source in her control or custody, whether in her bank accounts, safe deposit boxes or home. 30. For the purposes of the present appeal, we would accept the contention advanced in the respondent’s written submissions that the conduct particularised in amended Charge 7 was such as fairly to be regarded as part of the same criminal enterprise. This conclusion is justified in the present case due to the particular circumstances of the appellant’s co-habitation with D1 in the Park Road flat and her control of the funds in the bank accounts and her control and custody of the property in her safe deposit boxes and the Park Road flat, all of which funds and property, it was alleged by the prosecution, were dealt with by the appellant for the purpose of concealing their illegitimate sources for the benefit of her and D1. The illegitimacy of those sources was clearly suggested by the lack of any documentary evidence to support either the appellant or D1 having the amount of property particularised in the charge and from her and D1’s earnings and occupations not being commensurate with the amounts of money and value of property involved. 31. Of particular relevance, in the present appeal, is the lack of any prejudice to the appellant in the conduct of her defence at trial, or her appeal to the Court of Appeal. There was no objection to the form of amended Charge 7 at trial and the appellant’s case was that all of the funds and property particularised in that charge had a legitimate provenance to which she testified at the trial. The trial took place before a judge alone and, in his Reasons for Verdict, he gave detailed reasons for his findings in relation to this charge by reference to the appellant’s different explanations. 32. Applying the approach in Yeung,[17] we therefore concluded that amended Charge 7 was unobjectionable and did not offend against the rule against duplicity. Having said that, we would make the observation that, given that the property the subject of the charge consisted of distinct types, namely money and other personal property, and was held in different places, had the question of how to charge the OSCO section 25 offence arisen before trial, a better prosecutorial approach might have been to prefer separate charges for (i) the funds in each of the bank accounts, (ii) the personal property found in the Park Road flat and (iii) the personal property found in each of the safe deposit boxes. This would have catered for the possibility of a different defence in respect of each of these discrete forms of property dealt with by the appellant. As it transpired, however, the same defence was advanced in respect of all the property charged, so this better approach would not have been more advantageous in the present case. Nevertheless, depending on the facts in other cases, an omnibus charge of the type preferred here can potentially give rise to a risk of unfairness due to the inherently duplicitous nature of such a charge. H. Conclusion 33. For these reasons, the appeal was dismissed. Mr Andrew Bruce SC and Mr Tony C.Y. Li, instructed by Charles Yeung Clement Lam Liu & Yip, for the Appellant Mr William Tam SC, DDPP and Ms Irene Fan SPP, of the Department of Justice, for the Respondent Annex Bank Total no. of deposits Total amount of deposits Bank of China (Hong Kong) Limited 428 HK$8,730,235.76RMB¥2,212,789.71 Hang Seng Bank Limited 686 HK$15,216,450.17€1,750 £1,005 CDN$800 US$20,582.09 RMB¥435,818.65 Citic Bank International Limited 161 HK$36,966,989.83 Bank of East Asia Limited 346 HK$6,118,551.61 Public Bank (Hong Kong) Limited 19 HK$12,452,452.44 [1] (Cap.455) (“OSCO”). [2] In this Judgment, I shall adopt this reference for the two appeals in HKSAR v Yeung Ka Sing Carson and HKSAR v Salim Majed, FACC 5 & 6/2015 and FACC 1/2015, which were heard together and in respect of which judgment was delivered on 11 July 2016. [3] FAMC 46/2015, Ribeiro, Tang and Fok PJJ, Determination dated 22 March 2016 at [1]. [4] Contrary to s.25(1) and (3) of OSCO. [5] In DCCC 249/2012, before HH Judge E. Yip. [6] DCCC 249/2012, Reasons for Verdict, 13 March 2013. [7] In CACC 161/2013, before Yeung VP, J. Poon and D. Pang JJ (as they then were), Reasons for Judgment dated 13 March 2015. [8] Determination at [2]. [9] Made in 1976 by the Criminal Procedure Rules Committee pursuant to powers conferred by the Criminal Procedure Ordinance (Cap.221). [10] Yeung at [134]. [11] [1973] AC 584 at 607. [12] Yeung at [155]. [13] Yeung at [158]. [14] Yeung at [146]. [15] Appearing with Mr Tony C.Y. Li. [16] (2014) 17 HKCFAR 778. [17] Recently also applied in HKSAR v Yang Sigai, FACC 8/2015, Judgment dated 23 September 2016, at [17]. 
Chief Justice Ma: 1. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Ribeiro PJ: 2. This appeal concerns the proper construction of a will and in particular, the question whether certain parcels of shares were intended to pass under the will or whether they fall to be dealt with on a partial intestacy. It requires consideration of whether the will’s true interpretation can be arrived at by applying ordinary principles of construction or whether extrinsic evidence should be admitted pursuant to section 23B of the Wills Ordinance[1] to assist in its interpretation. A. The Will and the persons and principal assets involved 3. The will in question is that of Mr Tan Kiam Toen (“the testator”) dated 6 February 2008 (“the Will”), made in the form of a joint will with his wife (and now his widow) Madam Ng Giok Oh (“Madam Ng”). Such a joint will operates as the separate will of each of the persons making his or her will in the single instrument.[2] Thus, the Will contains clauses which are equally applicable to the testator and to Madam Ng, using language expressed in the plural form. But each such clause must be read and understood for present purposes as a provision in the testator’s will operating separately. The Will is set out in full in its English translation (having been written in Chinese) in Annex A to this judgment. A codicil dated 8 May 2008 (“the Codicil”) is set out in translation in Annex B. 4. The testator was born in 1919 and died on 15 November 2008, domiciled in Hong Kong. He is survived by Madam Ng (who was born in 1922) and by three sons and two daughters, as well as certain grandchildren. 5. The testator’s five children (now aged in their fifties and sixties) are the parties to this litigation. The sons, Mr Tan Cheng Gay, Mr Tan Yok Koon and Mr Tan Chin Hoon, and the daughter, Ms Tan Choo Pin brought these proceedings as plaintiffs and are now the 1st, 2nd, 4th and 3rd appellants respectively. His other daughter, Ms Tan Choo Suan, the sole executrix and trustee of the Will, was sued as the 1st defendant and is now the 1st respondent. The Secretary for Justice is named as 2nd respondent because of the interest that charities have in the outcome. However, as he has played no substantive role in the litigation, it will be convenient to refer to the 1st respondent simply as “the respondent”. Madam Ng is in the respondent’s camp. 6. The testator had very modest origins and went to work in Indonesia as a labourer at the age of 17. However, he prospered greatly and, in 1961, he founded Afro-Asia Shipping Company (Private) Ltd (“AAS”) a Singapore company which traded in tin, tea, coffee, rice, rubber and cement. The evidence is that both at the time when he made the Will and at the time of his death, the registered owners of the issued shares in AAS were (as to 52.22% comprising 2,779,110 shares) Madam Ng and (as to 47.78% comprising 2,542,590 shares) the respondent. I shall refer to the AAS shares in the respondent’s registered ownership as “the AAS shares”. 7. The principal assets of AAS consist, and have throughout consisted, of a parcel of 15,674,500 shares constituting a 13.37% interest in a Singapore listed company called EnGro Corporation Limited (“EnGro”) and a commercial building known as the Afro-Asia Building in Robinson Road, Singapore. 8. The testator also caused to be formed a Hong Kong company called Afro-Asia International Enterprises Pte Limited (“AAIE”) to serve as a vehicle for holding additional shares in EnGro. A third party owns 30% of AAIE but the remaining 70% of its shares are registered in the names of the two daughters, namely, the respondent and the 3rd appellant, each holding a parcel of 1,750,000 AAIE shares together constituting a 35% interest in AAIE (the daughters’ combined interest referred to as “the AAIE shares”). 9. The Court was informed of the existence (but few of the details) of proceedings before the Singapore courts regarding the beneficial ownership of the AAS and AAIE shares. However, for the purposes of the present proceedings, everyone has proceeded on the footing that the AAS shares held by the respondent, and the AAIE shares held by the respondent and the 3rd appellant are acknowledged by them to be held as trustees for the testator and his estate. 10. Financed by funds lent by AAS to the respondent[3] and on-lent by the respondent to AAIE, AAIE acquired further EnGro shares. 70% of the shares acquired, amounting to 31,124,100 shares, reflect the estate’s 70% interest in AAIE and constitute the principal asset held by the estate through AAIE. The evidence is that the closing price of EnGro shares on 14 November 2008, the day before the testator died, was S$0.66 per share. 11. Additionally, the testator’s estate consists of a very valuable collection of artworks and antiques owned by the testator and kept at his residence in a flat which he owned with Madam Ng in Bowen Road, Hong Kong (“the Bowen Road property”). Further artworks and antiques were also owned by the testator and kept by him at two addresses in Singapore. I shall refer to all these artworks together as “the art collection”. 12. Minor assets of the estate include about HK$2 million in balances in bank accounts held either solely by the testator or jointly by him and Madam Ng, plus one share in his name in a British Virgin Islands company called Balmain Industries Ltd (“Balmain”), the only other shareholders being Madam Ng and the respondent, each also holding one share. Balmain’s only asset is US$4.5 million cash in a bank account. B. The events leading up to and following upon the testator’s death B.1 The 2006 Will and 2006 Undertaking 13. On 21 April 2006, the testator executed a will (“the 2006 Will”) drafted by a Singapore lawyer.[4] It appointed his two daughters as executrices and trustees of the will and did not make any gift to charity but left “all [his] properties moveable and immovable, whatsoever and wheresoever situate ...” to Madam Ng (50%) and his five children (10% each). It stated in its Clause 6: “For the avoidance of doubt, I declare that I have no assets which are held by my daughters or by any other person in trust for me and that all the assets previously belong [sic] to me or financed by me which are currently held by my daughters are not held in trust for me and I further declare that the shares in [AAS ] which are registered under my name belong absolutely to my daughter [the respondent] and I had executed a Trust Deed in favour of [her] accordingly.”[5] 14. Clause 6 therefore excluded the AAS and AAIE shares from the testator’s estate. However, some seven months later, on the testator’s instruction,[6] the two daughters wrote jointly to their parents in a letter dated 21 December 2006, giving what has been referred to as “the 2006 Undertaking”, stating: “.... even though we are the beneficial owners of 3,500,000 ordinary shares in .... [AAIE] ... which in turn owns 29,641,700 shares in [EnGro] ... we have promised you that when father is not around, we will give some AAIE shares to our mother and our brothers. If AAIE is liquidated, we will still give the shares in EnGro which we received from AAIE to our mother and brothers.” There followed a distribution list whereby the 1st appellant was to get 700,000 shares in AAIE and everyone else (Madam Ng and the other siblings) was to get 560,000 shares.[7] B.2 The 2007 Will 15. Less than a year later, the 2006 Will was revoked when the testator executed his will dated 30 March 2007 (“the 2007 Will”) prepared by a Hong Kong solicitor, Mr Bruce Fu Chi Kai, from the firm of Hastings & Co. The 2006 Will appointed the respondent his sole executrix and trustee of the will and made a gift to Madam Ng (in translation) in the following terms: “.... I give as a gift all the properties under my name in Hong Kong and any other places (including movable properties and immovable properties) (hereinafter called ‘my remaining properties’) to my wife, Ng Giok Oh...” The italicised words[8] may be compared with the words in the Will, particularly the words “under my name” (“名下”), which are at the heart of the present appeal, as discussed below. 16. After proceeding to make gifts of all his “remaining properties” as defined to charity in the event of Madam Ng pre-deceasing him, the testator declared (as he does similarly in Clause 11 of the Will) that, his children having been adequately taken care of, he did “not intend to give any portion of [his] remaining properties to them as a gift”. B.3 The 2008 statutory declaration 17. Then on 31 January 2008, the testator made a statutory declaration upon which witnesses were examined in the Singapore court, but a copy of which has not been produced in these Hong Kong proceedings. Extracts from the transcript of evidence in the Singapore court referring to the statutory declaration have been relied on by the appellants to show that less than a week before making the Will, the testator was making it clear that he wished to distribute the AAIE shares and the underlying EnGro shares to members of the family and did not envisage such shares going to charity. B.4 The Will 18. As previously noted, on 6 February 2008, the testator executed the Will which is the subject matter of this appeal, revoking all prior wills. Its provisions will require close scrutiny but, in outline, the Will provides (in Clause 2) that it governs and deals with “all the properties under [the testator’s] name (wheresoever situate worldwide)”; and (in Clause 5) that Madam Ng, who survives the testator, is to “enjoy during her lifetime the income derived from the estate of” the testator; and that upon her death, the entire residue is to pass to specified charities, the testator declaring (in Clause 11) that he does not intend to give any portion of his residuary estate to his children, stating that they have received adequate and appropriate care during their parents’ lifetime. B.5 The Codicil 19. Three months after the Will was executed, the testator executed[9] the Codicil dated 8 May 2008. It deals with a property in Australia and the Bowen Road Property, both of which were owned by the testator and Madam Ng as joint tenants, declaring that the respondent had been made the third joint tenant (and therefore enjoyed the right of survivorship) in respect of the Australian property and that, on the death of both the testator and Madam Ng, the Bowen Road Property was to pass to the respondent. B.6 The Parents’ Letter 20. By a letter dated 16 August 2008 addressed to their children, the testator and Madam Ng noted that the children were aware that their parents “would donate most of the properties to charitable and Buddhist organisations in Singapore, Hong Kong and Mainland China, and would leave the Engro shares held through AAIE to you people.” Proposals were put forward as to how those AAIE shares should be freed from existing debt and distributed to the children, the parents stressing however, that they were to keep for themselves the AAS shares. They emphasised that they wished to avoid disputes among their children and would be instructing a lawyer to draft a Deed of Family Arrangement which the children were required to sign before the proposed distribution of AAIE shares as would be set out in a schedule was to be implemented. B.7 The sons’ response 21. The sons did not agree with their parents’ proposal. Instead, by letter dated 12 October 2008, they sought to persuade them to include the EnGro shares held through AAS to be distributed to the family, offering an undertaking as to how the dividends should be dealt with. 22. Some further discussions involving Hastings & Co ensued on 28 October 2008, but the proposed inter vivos distribution of the AAIE shares was not achieved when the testator died on 15 November 2008. B.8 Madam Ng’s attempts to implement the testator’s wishes 23. Between November 2008 and August 2009, Madam Ng, assisted by Hastings & Co and counsel, sought to find a means of implementing her testator husband’s wishes by providing the children with a parcel of EnGro shares from her own resources. This progressed to the stage of a Deed of Family Arrangement being prepared and sent to the appellants for execution. But the Deed was never executed and instead, proceedings were instituted by the appellants in Singapore and also in Hong Kong. C. The present proceedings C.1 The grant of probate 24. The respondent, as executrix and trustee of the testator’s will obtained a grant of probate from the Hong Kong Court on 6 May 2009. The annexed Schedule of Assets and Liabilities of the Testator[10] listed as the testator’s assets in Hong Kong (i) cash at bank totalling HK$1,981,200.06; (ii) the AAIE shares (listed as two parcels of 1,750,000 shares each held by the respondent and the 3rd appellant respectively in trust for the testator); and (iii) “various pieces of antiques, paintings and works of art”. 25. The grant was re-sealed on the respondent’s application by the Singapore Court on 15 November 2010, annexing a statement that the testator died possessed of property in Singapore comprising (i) the AAS shares (stated to be 2,542,590 shares in AAS held by the respondent as nominee for the testator) and (ii) various pieces of art works, antiques, porcelain vases, furniture, etc located at two Singapore addresses. C.2 The originating summons 26. On 4 February 2013, the appellants issued proceedings which (as amended[11]) sought the Court’s determination as to: “... whether upon the true construction of the [Will and Codicil, those documents] shall govern only the administration and disposition of assets and/or properties of the Testator which are held under or in the name of the Testator or held by the Testator personally wherever situated and shall not govern the administration and disposition of assets and/or properties of the Testator which are held in the name of or by others but beneficially and on trust for the Testator.” 27. If the question were affirmatively answered, consequential orders were sought for rectification of the Grant of Probate and the attached Schedule of Assets to exclude the AAIE shares from the testator’s estate. It may be noted in passing that no consequential relief is sought regarding the AAS shares held by the respondent as trustee for the testator and his estate. 28. The dispute therefore pivots around the wording of Clause 2 of the Will which states that the Will “shall govern and deal with all the properties under [the testator’s name] (wheresoever situate worldwide).” Does this exclude properties, such as the AAIE shares whose registered owners are the respondent and the 3rd appellant, even though it is acknowledged that they are held for the testator beneficially? C.3 The decisions of the Courts below 29. At first instance,[12] Deputy High Court Judge Chu (as B Chu J then was) concluded that such property is not excluded. Looking at the Will as a whole, her Ladyship held that the relevant words: “ ‘名下所有(不論在世界任何地方)的財産’ can be all embracing and are entirely apt to cover all the assets of the Father and the Mother and to include assets beneficially belonging to them or either of them but held on trust on his/her behalf by another, and not only restricted to assets held or in the personal name of the Father or the Mother.”[13] 30. The Judge held that since the meaning of the relevant provisions was clear, it was unnecessary to consider section 23B of the Wills Ordinance which only operates in case of ambiguity. She added however, that particularly in the light of the evidence given in the Singapore proceedings by Mr Fu, the solicitor who had prepared the Will, and by Madam Ng on affirmation, even if her Ladyship had taken the extrinsic evidence into account, the outcome would have been the same.[14] 31. The Court of Appeal[15] upheld the Judge. It held that “名下” is an ordinary term without any technical meaning and admits of no ambiguity. In its dictionary meaning, when used in connection with assets “it means assets ‘belonging’ (屬) to the person concerned. It is not limited to those assets which are held under or in the name of that person alone.”[16] 32. Construing Clause 2 in the light of the other Clauses in the Will, the Court considered the meaning “crystal clear”: ‘It means ‘all the properties belonging to us’. It embraces all assets beneficially belonging to the Testator and Madam Ng.”[17] D. The principles applicable to construction of the Will D.1 The common law principles 33. We have the benefit of the authoritative exposition of the principles applicable to the construction of wills provided by Lord Walker of Gestingthorpe NPJ in the recent decision of this Court in Chinachem Charitable Foundation Ltd v The Secretary for Justice,[18] a decision drawing substantially upon the judgment of Lord Neuberger of Abbotsbury in Marley v Rawlings.[19] 34. As Lord Walker NPJ points out, wills have as two important features (i) the fact that they are unilaterally created documents and (ii) that they have an ambulatory character, that is, that they are intended to take effect from the time of the testator’s or testatrix’s death, which may be some time after the will was executed.[20] These features must be borne in mind when construing a will. 35. Against that background, it is now established that, subject to statutory regulation, one adopts the same approach to interpreting wills as one does to interpreting contracts and other legal documents: “... the aim is to identify the intention of the party or parties to the document by interpreting the words used in their documentary, factual and commercial context”.[21] 36. In other words, subject to statute: “...the court is concerned to find the intention of the party or parties, and it does this by identifying the meaning of the relevant words, (a) in the light of (i) the natural and ordinary meaning of those words, (ii) the overall purpose of the document, (iii) any other provisions of the document, (iv) the facts known or assumed by the parties at the time that the document was executed, and (v) common sense, but (b) ignoring subjective evidence of any party's intentions.”[22] 37. This approach is consistent with the well-known “arm-chair” principle, save that it must be kept in mind that that principle looks to context as at the time the will is made whereas the will speaks from the time of death regarding any general description of property.[23] 38. Lord Walker NPJ’s summary of the “two all-important principles of construction” are of particular present relevance, namely “that the words must be read and understood in their context, and that the will must be read as a whole”.[24] This requires the Court to adopt an “iterative” process, checking each of the rival meanings against the other provisions of the document and investigating its practical consequences,[25] and: “It may require the court to go forwards and backwards painstakingly between the various words and phrases, occurring in different parts of the document, which give rise to the problem.”[26] D.2 Section 23B of the Wills Ordinance 39. Section 23B[27] provides: (1) This section applied [applies[28]] to a will- (a) in so far as any part of it is meaningless; (b) in so far as the language used in any part of it is ambiguous on the face of it; (c) in so far as evidence, other than evidence of the testator's intention, shows that the language used in any part of it is ambiguous in the light of surrounding circumstances. (2) In so far as this section applies to a will extrinsic evidence, including evidence of the testator's intention, may be admitted to assist in its interpretation. 40. As we have seen, Deputy High Court Judge Chu did not consider it necessary to resort to section 23B but held that even if extrinsic evidence was taken into account, it would not have altered her decision. The Court of Appeal thought it unnecessary to consider any extrinsic evidence. 41. In this appeal, the respondent submits that the Court should adopt a position similar to that adopted below, whereas the appellants argue that extrinsic evidence is both admissible in the light of ambiguity in Clause 2 of the Will and highly supportive of the narrow construction of the words “名下” (excluding assets held on trust) urged by them. I shall first endeavour to construe the Will adopting the common law principles identified above before considering the possible applicability and relevance and section 23B. E. Construction of the Will 42. Clause 2 states: “We hereby declare that this Will shall govern and deal with all the properties under our names (wheresoever situate worldwide).” In the original Chinese, it states: “我們謹此聲明這份囑書將管轄及處理我們名下所有(不論在世界任何地方)的財産”. 43. The central question is whether, by using the words “under our names” or “名下” (transposing those words to “under my name” since we are viewing the testator’s will separately) has the effect, as the appellants contend, of confining the property intended to pass under the will to property which is in some way registered or officially recorded to be property held by the testator referred to by name, and thus excluding property held in the names of others, albeit held by them in trust for the testator (“the narrower construction”). The respondent submits that this is an unwarranted construction which is inconsistent with the ordinary meaning of the words used and inconsistent with the purpose and the other provisions of the Will in the light of the assets constituting the estate. The respondent submits that the words were intended merely to mean “belonging to the person concerned (ie, the testator)”, encompassing property held on trust for him or her (“the wider construction”). E.1 “名下” as a matter of ordinary language 44. Adopting an iterative approach, the rival meanings given by the parties to the words “名下” may be considered, before checking each of those meanings against the other provisions of the Will and examining the practical consequences. 45. The four judges who decided this case at first instance and in the Court of Appeal are native Chinese speakers and, in my view, great weight must be given to their unanimous conclusion that the words used in Clause 2 are capable of bearing a meaning embracing the whole of the testator’s property and do not confine the assets disposed of to property expressly recorded in some way to be the property of the testator by name. 46. Thus, Deputy High Court Judge Chu stated: “Having considered the whole of the Joint Will, I am of the view that Clause 2 should not be construed in such a narrow sense. I accept Mr Chow’s submission that the words “名下所有(不論在世界任何地方)的財産” can be all embracing and are entirely apt to cover all the assets of the Father and the Mother and to include assets beneficially belonging to them or either of them but held on trust on his/her behalf by another, and not only restricted to assets held or in the personal name of the Father or the Mother.”[29] 47. And Poon J, writing for the Court of Appeal, stated: “38. Mr Ho, SC, for the plaintiffs, submitted that in its natural, plain and ordinary meaning, ‘名下’ means ‘under/in the name of” somebody. It follows that only assets under/in the name of the Testator were covered by the Joint Will. It does not include those held under/in the name of some ... others on trust for him. Mr Chow, SC, for the 1st defendant submitted that the words ‘名下所有…的財產’ are entirely apt to cover or include assets beneficially belonging to a person but held on trust for his benefit by another person. With respect, we agree with Mr Chow. 39. As rightly accepted by the parties, ‘名下’ is an ordinary term without any technical meaning. It must be understood in its plain and ordinary usage in Chinese. ... The dictionary meaning of ‘名下’ in its plain and ordinary Chinese usage is beyond doubt and admits no ambiguity. When it is used in connection with assets, it means assets ‘belonging’ (屬) to the person concerned. It is not limited to those assets which are held under or in the name of that person alone. 40. ‘名下’ is used in Clause 2 in the phrase ‘我們名下所有…財產’ to describe the assets to be covered by the Joint Will. Read with the plain and ordinary meaning of ‘名下’ in mind, the testamentary intention as expressed in that phrase is crystal clear. It means ‘all the properties belonging to us’. It embraces all assets beneficially belonging to the Testator and Madam Ng.” 48. It is significant that the appellants accept that the wider construction (“belonging to”) is an available meaning of “名下”. This is put as follows in their printed case: “The term ‘名下’ (‘under the name’), if the Court should come to the view that it is not clear enough, is capable of bearing more than one ordinary meaning, (1) one meaning, i.e. ‘屬於某人’ (‘belonging to some person’), would be apt to include properties beneficially owned by a person irrespective of whether those properties are held under the name of that person or not, so that the phrase ‘我們名下所有的財產’ (‘all the properties under our names’) can be construed as meaning ‘all the properties belonging to us’; (2) another meaning, i.e. ‘某人名義之下’ (‘under the name of some person’), would be apt to include only those properties beneficially owned by and held under the name of a person, but not properties held under the name of somebody else, so that the phrase ‘我們名下所有的財產’ (‘all the properties under our names’) can be construed as meaning ‘all the properties held under our names and owned by us’.” E.2 What assets would be left to pass under the Will on the narrower construction? 49. As noted in Section C.1 of this judgment, in the Schedules to the Grant of Probate issued in Hong Kong and re-sealed in Singapore, the respondent listed as the testator’s assets, apart from relatively small bank balances, the AAIE shares held by the respondent and the 3rd appellant in trust for the testator; the AAS shares held by the respondent as nominee for the testator; and the art collection. 50. If the appellants’ narrower construction were to be accepted, it would mean that the AAIE and AAS shares would plainly be excluded from passing under the Will: they were not held under the testator’s names but under the names of the respondent and the 3rd appellant on trust for the testator. It would, in my view, also mean that the art collection would be excluded. While one would readily accept that each item of the art collection might be said to “belong to” the testator (where for instance, evidence of his having purchased or otherwise having acquired the object existed), it is hard to see how, as a matter of language, it would be appropriate to say that the object in question was held “under his name” unless those words were given the wider construction. 51. It seems likely that those responsible for drafting the originating summons were conscious of this difficulty since, as noted in Section C.2 above, the question formulated for the determination of the Court is: “... whether upon the true construction of the [Will and Codicil], those documents] shall govern only the administration and disposition of assets and/or properties of the Testator which are held under or in the name of the Testator or held by the Testator personally wherever situated...” 52. The words I have italicised implicitly acknowledge the difficulty of fitting the ownership of chattels such as items in the art collection within the phrase “held under ... the name of the Testator” if given the narrower construction. Thus the need was evidently felt to add “or held by the Testator personally” as a way of encompassing the art collection. But if that extension is accepted as coming within the meaning of the words “held under the name of the testator”, it is hard to see why other property “belonging personally to” but not “held under the name of” the testator should be excluded from the Will. Could shares to which the testator was beneficially entitled not be said to “belong personally to him”? 53. In the argument before the Court, the “held personally” extension referred to in the originating summons was not advanced. Instead, the construction urged by the appellants was the strict, narrower construction, put in their printed case as follows: “The term ‘名下’ (‘under the name’) should be taken as being inserted specifically on purpose, so that the phrase ‘我們名下所有 … 的財產’ (‘all the properties under our names’) should be given its plain and ordinary meaning to embrace properties owned by and held under the names of the Testator and/or Madam Ng, but not properties which were/are beneficially owned by them or either of them but held under the name of others on trust for them or either of them.”[30] 54. But if such a narrow construction is adopted, little would be left from among the testator’s known assets to pass under the will. The only property that Mr Patrick Fung SC[31] could point to as held under the testator’s name is the single share in Balmain, a company whose only asset is US$4.5 million cash in a bank account, there being two other shareholders, namely Madam Ng and the respondent, and there being nothing known of any liabilities. 55. The narrower construction urged by the appellants would therefore cause the Will to attach only to an insignificant part of the testator’s known assets. This calls into question the validity of such an interpretation. It is implausible to suggest that the testator went to the trouble of making a will to provide for how his property should be dealt with after his death but chose to exclude his main assets from that testamentary exercise. Why should he have taken the trouble of making a will only to leave his most valuable assets to be dealt with on a partial intestacy? 56. This is reflected in Williams[32] where the learned authors state: “A testator may well intend to die partly intestate; and, when he makes a will, he is testate only so far as he has expressed himself in his will. Where, however, the construction of the will is doubtful, the court acts on the presumption that the testator did not intend to die either totally or partly intestate, provided that on a fair and reasonable construction there is no ground for a contrary conclusion.” 57. In support, Re Harrison, Turner v Hellard[33] is cited, where Lord Esher MR made the point in colourful terms: “There is one rule of construction, which to my mind is a golden rule, viz, that when a testator has executed a will in solemn form you must assume that he did not intend to make it a solemn farce – that he did not intend to die intestate when he has gone through the form of making a will. You ought, if possible, to read the will so as to lead to a testatcy, not any intestacy. This is a golden rule. I do not deny that this will may be read in two ways, or that it requires that a blank should be filled up. But it may be read in such a way as not to amount to a solemn farce.” E.3 The Australian and Bowen Road properties 58. Might it be suggested that the Australian and Bowen Road properties were the assets held “under his name” contemplated to pass under the Will given the narrower construction? At the time of making the Will, the testator had not made provision for those properties and it was only three months later that they received specific mention in the Codicil. 59. In my view, the Australian property plainly cannot be regarded as an asset intended to pass under the Will as property “held under the name of the testator”. It was originally held by the testator and Madam Ng as joint tenants and, as the Codicil makes clear, on 24 May 2003 they arranged for the respondent to become the third joint tenant. Her parents acknowledged the legal consequences in the Codicil as follows: “We are aware and also understand that as one of the joint tenants of the property, after both of us have passed away, according to the legal principle of right of survivorship, Choo Suan will automatically become the sole owner of the Australia Sydney Property.” 60. It was therefore intended, by arrangements made almost five years before the Will was executed, that the Australian property should pass from the testator and Madam Ng to the respondent, not by operation of any will, but by virtue of her survivorship right as joint tenant. The Australian property is thus not an asset passing under or relevant to the construction of the Will. 61. Similar considerations apply to the Bowen Road property. Although no arrangements favouring the respondent had been made prior to execution of the Codicil, the Bowen Road property was, at the time when the Will was executed, held by the testator and Madam Ng as joint tenants. In other words, viewed from the perspective of the testator’s separate Will, upon his predeceasing Madam Ng, she would become, by virtue of the joint tenancy, automatically entitled to ownership of the property. It was not an asset which would pass in accordance with the Will whereby Madam Ng would merely enjoy a life interest in income with the residue passing to the specified charities. It would pass to her absolutely as the surviving joint tenant. Conversely, if Madam Ng had predeceased the testator, the Bowen Road property would automatically have passed to the testator by virtue of his survivorship right and not pursuant to Madam Ng’s will. 62. The effect of a joint tenancy in such a context is noted in Williams as follows: “Before speaking of joint wills it is perhaps necessary to say that so far as English law is concerned, they have nothing whatsoever to do with joint property, ie property held by beneficial joint tenants. Such property (provided always that the joint tenancy is subsisting at death and has not been severed) cannot be subject to any will but passes automatically to the survivor or surviving joint tenants.”[34] 63. By the Codicil, the respondent’s parents stipulated that the survivor of the two of them thereby bequeathed the Bowen Road property to the respondent. Madam Ng survived the testator and automatically became owner of the Bowen Road property. No interest in that property would vest in the respondent unless and until Madam Ng predeceases her and the property passes to her in accordance with the Codicil. It is thus plain that the gift of the Bowen Road property was not a gift of property “under the name of the testator” intended to pass under the Will. E.4 Clause 5 and income-earning assets 64. Since, on the appellants’ narrower construction, the AAS and AAIE shares and their underlying EnGro shares, as well as the Australian and Bowen Road properties are not property “under the name of the testator” passing under the Will, it is difficult to see how, on that construction, Clause 5, which makes a gift of income for life, could have any meaningful effect. 65. Clause 5 states: “We mutually undertake to each other that in the event one of us predeceases the other, subject to the payment of all the necessary expenses ..., the survivor of us can enjoy during his/her lifetime the income derived from the estate of the first-deceased of us.” 66. The EnGro shares held through AAS and AAIE are the assets of the testator’s estate (being “the estate of the first-deceased of us”) that represent the natural source of income envisaged under Clause 5. Eliminating them from the Will on the appellants’ narrower construction makes it hard to see any other income-producing source. The Australian and Bowen Road properties have also been eliminated for the reasons just discussed and the art collection, even if not eliminated on the basis mentioned above, does not produce income. It follows that the central gift to Madam Ng under the Will – the gift of the income of the testator’s estate for life – could have little effect if the appellants’ narrower construction is adopted since it excludes the only obvious income-earning assets from the operation of the Will. Whatever income the relatively small cash balances might be able to produce cannot provide the rationale for Clause 5. 67. Thus, considering the Will as a whole and checking the rival meanings against Clause 5 of the Will and against the implausible practical consequences of the narrower construction discussed above, the wider construction strongly commends itself. For the foregoing reasons, I conclude that on its true construction, unaided by any extrinsic evidence but taking into account the assets which the testator was seeking to dispose of, the operation of the Will is not confined to properties held in some formal or officially recorded sense “under the name of the testator” but applies to all properties belonging to him, including properties held on trust for him and to which he was beneficially entitled. So construed, I do not consider there to be any ambiguity in the words used since the alternative meanings put forward appear to me to be less than reasonably arguable in the context of the Will read as a whole. I therefore respectfully agree with the decisions arrived at by the trial Judge and the Court of Appeal and would dismiss this appeal. 68. It is strictly unnecessary and indeed, given my finding of an absence of ambiguity, not strictly permissible to proceed to examine the effect of any potentially relevant extrinsic evidence for section 23B purposes. However, since it has been argued at some length, it may be desirable for me to touch briefly on such evidence to explain why, in my view, it would not have led to a different outcome. F. Extrinsic evidence 69. The events outlined in Section B above undoubtedly show that both before and after he executed the Will, the testator wished to make an inter vivos gift of the AAIE shares and therefore of the Engro shares held by AAIE to Madam Ng and the children. He was, however, anxious that there should not be disputes or litigation with regard to that gift and so insisted on a Deed of Family Arrangement being signed to settle all the details before the proposed gift could take effect. The gift failed because the sons did not agree to the arrangement but sought to persuade the testator to add the AAS shares to the gift, and the testator died before the inter vivos gift could be accomplished. The evidence also shows that after his death, Madam Ng sought to implement the testator’s intention to make an inter vivos gift of EnGro shares from her own resources but that this was not achieved and the litigation ensued. 70. Assuming that the extrinsic evidence (referred to in Section B above) were admissible, the question is whether it throws any light on the testator’s testamentary intentions. It is one thing to say that the evidence clearly establishes that the testator wanted at some stage to make an inter vivos gift to the family of the relevant shares (which he obviously regarded as his own to dispose of as he pleased). But it is quite another to say that such evidence helps us to ascertain the scope of the property he intended to dispose of by his Will in the event that the inter vivos gift was not accomplished. Failure of the intended inter vivos gift does not necessarily mean that the property in question falls outside the scope of the Will so as to pass on a partial intestacy. Whether that was the intention requires the Will to be construed as a whole, as has been done in the foregoing parts of this judgment. 71. Mr Patrick Fung SC sought to rely on the extrinsic evidence for the negative proposition that the testator did not intend the AAIE shares to go to charity but to the family instead. That proposition is true as far as it goes, but with respect, it does not go far enough. The testator’s intention was that the AAIE shares should go to the family only upon the terms of the proposed inter vivos gift and not otherwise. Indeed, he made it clear in Clause 11 of the Will that his intention was that the children should not take any portion of his assets passing under the Will. Extrinsic evidence establishing an intention to make an inter vivos gift (which does not eventuate) does not, at any rate without more, throw light on the question of what, on its true construction, the Will means when it refers to property “under the testator’s name”. 72. I have held that, properly construed, the Will applies to the AAIE shares and other property belonging to the testator. Failure of the proposed inter vivos gift merely means that those shares, not having been removed from the estate by the making of such gift, remain to be dealt with in accordance with the Will. This is unaffected by any of the extrinsic evidence sought to be relied on. 73. Accordingly, for the reasons developed above, I have concluded that the AAIE shares were intended to be covered by the Will on its true construction and that the failure of the intended inter vivos gift means that those shares remain to be dealt with under the Will and not under a partial intestacy. G. Conclusion 74. I would therefore dismiss this appeal and make an order nisi requiring the appellants to pay to the respondent the costs of and occasioned by this appeal. I would not make any order in relation to the Secretary for Justice who has not taken an active part. I would grant the parties liberty to make written submissions as to costs to be lodged within 14 days from the date of this judgment, in default of which I would direct this order as to costs to stand as an order absolute without further direction. Mr Justice Tang PJ: 75. Mr Tan Kiam Toeh (“the testator”) was a self-made man who died on 15 November 2008 aged about 90. He was survived by his wife, Madam Ng Giok Oh (“Madam Ng”), and their three sons, Cheng Gay, Yok Koon and Chin Hoon, (1st, 2nd and 4th appellants respectively), two daughters, Choo Pin, (3rd appellant) and Choo Suan (“1st respondent”). 76. On 6 February 2008, the testator and his wife (who was born in 1922) made a joint will in the Chinese language which was prepared by Messrs Hastings (“Hastings”). This litigation concerns the construction of Clause 2 which reads “我們謹此聲明這份平安囑書將管轄及處理我們名下所有（不論在世界任何地方）的財產。” and as translated “We hereby declare that this Will shall govern and deal with all the properties under our names (wheresoever situate worldwide)”, in particular the expression “名下” “under our names”. Under the joint will, subject to a life interest in the income in favour of the survivor, the residuary estate would go to charities, for that reason, the Secretary for Justice has also been made a respondent.[35] 77. The appellants’ primary contention was that since the will was drafted by solicitors practising common law in Hong Kong, the plain and ordinary meaning of these words covered only “properties owned by and held under the names of ‘the deceased or Madam Ng’, but not properties which were/are beneficially owned by them or either of them but held under the name of others on trust for them or either of them.” 78. Alternatively, the appellants relied on s 23B of the Wills Ordinance, in particular, that because the expression is “ambiguous on the face of it”, “extrinsic evidence, including evidence of the testator’s intention, may be admitted to assist in its interpretation.” Under s 23B, such evidence is also admissible “in so far as evidence, other than evidence of the testator’s intention, shows that the language used in any part of it is ambiguous in the light of surrounding circumstances.” Mr Fung submitted that the extrinsic evidence supported the above construction. 79. The 1st respondent’s case was that the expression “properties under our names” was not ambiguous and it covered all properties belonging to the testator.[36] 80. At first instance, Deputy Judge Bebe Chu (as she then was) concluded that as a matter of construction, it was clear that the testator had intended the will to cover all his assets whether they were held in his name or not.[37] She further concluded that had she been of the view that there was ambiguity, after taking into account extrinsic evidence on the testator’s subjective intention, she would have reached the same conclusion. 81. The Court of Appeal[38] regarded the will as clear and unambiguous and did not find it necessary to consider any extrinsic evidence. 82. On 4 February 2015, the appeal committee[39] granted leave to appeal on the basis “that the true construction of the will poses particular difficulties and raises issues as to the proper scope and application of section 23B of the Wills Ordinance.” 83. I have had the advantage of reading the judgment of Ribeiro PJ in draft and like his lordship I have come to the conclusion that on the proper construction of the will, the expression “all properties under our names” covered all properties belonging to the testator, whether or not they were held under his name. I also note that in the courts below, four bilingual judges concluded that the expression was clear and unambiguous. With respect, although I am also bilingual, when I read the papers for the first time when considering whether the application for leave to appeal should be dismissed without a hearing[40], I was troubled by the fact that, given the highly unusual circumstances of this case, a solicitor in Hong Kong had seen fit to use the expression “all the properties under our names” instead of the more common and direct expression “all our properties”. Mr Wong Yan Lung SC for the 1st respondent, in the course of his submission, said that the Chinese expression “名下” “under my name” was recorded to have been first used in the 11th century by a famous poet to denote “properties belonging to me”, a usage which predated the first development of the Law of Uses in the 12th century in England. It is pertinent to mention that traditional China never developed Uses or Trusts so that one would have little, if any, reason at all to distinguish between “my properties” and “properties under my name”. As Ribeiro PJ pointed out at para 48, Mr Patrick Fung, accepted, and rightly so in my view, one of the meanings of “名下” “under my name” is “belonging to me.” I accept that, in a homemade will, a Chinese testator, who was not aware of the niceties of ownership under the common law or the rules of equity, might use such an expression “我名下所有…的財產” “all properties under my name” to denote all his properties, although I would expect a modern testator to use the simpler, more direct, and more modern expression “我所有的財產” “all my properties” instead. Be that as it may, I would not expect a bilingual lawyer, trained in the common law, to use the expression “all the properties under our names” when as here there were substantial assets which were not held under the testator’s name, and as Ribeiro PJ’s judgment will have shown, at one time, the testator had stated unequivocally that he was not the beneficial owner of such assets. In the circumstances of this case, I am not at all surprised that the expression has given rise to these proceedings. If the simple and modern Chinese expression “all our properties” had been used in the 2008 will, I do not believe any construction issue would have arisen at all. Certainly not one which would reach this court. 84. This is why I feel I should examine the extrinsic evidence relied on by Mr Fung and explain in my own words why notwithstanding such evidence and my own misgiving over the use of the expression, I have come to the conclusion that the expression “properties under my[41] name” properly understood, covered all properties belonging to the testator whether or not they were held under his name. 85. Ribeiro PJ has set out in some detail the various assets which belonged to the testator. Amongst such assets were a controlling interest in EnGro Corporation Limited (“EnGro”), a Singaporean listed company. Such controlling interests are held by two companies, namely, Afro-Asia International Enterprises PTE Limited (“AAIE”) a Hong Kong company and Afro-Asia Shipping Company (“AAS”), a Singapore company. 86. The testator did not hold any shares in AAIE or AAS in his name. Of the 3,500,000 shares in AAIE said to be beneficially owned by him, 1,750,000 were registered in the name of each of his daughters. AAIE in turn held shares in EnGro[42], and on such basis, in these proceedings, the testator was regarded as the owners of 70% of these shares in EnGro, namely, 31,124,100 shares[43]. As for AAS, similarly, the testator was said to be the beneficial owner of 2,542,590 shares with the remaining 2,779,110 shares said to be beneficially owned by Madam Ng. The 2,542,590 shares were registered in the name of the 1st respondent. AAS in turn owned 15,674,500 shares in EnGro, constituting about 13.37% of the issued shares on EnGro. Apart from other assets, AAS also owned the Afro-Asia building, at 63, Robinson Road, Singapore, a very valuable property.[44] 87. The beneficial ownership of these shares[45] is under litigation in Singapore, where, I believe, the appellants claimed that they were the beneficiaries of some of these shares and the 1st respondent claimed that these shares were held in trust for the father. 88. We assume for the purpose of the present proceedings that these shares were beneficially owned by the testator. It follows that if the appellants are right and these shares had not been disposed of by the 2008 will, they would devolve as on intestacy. 89. I turn to consider the extrinsic evidence relied on by Mr Fung. Since, in my view, they do affect the outcome, I shall be as brief as I can, and I will not pause to examine their admissibility or relevance, I shall take them all into consideration in the construction of the 2008 will. 90. The testator and Madam Ng had made two earlier wills. On 21 April 2006, the testator executed a will (“2006 will”) in English drafted by a Singapore lawyer. The testator appointed his two daughters as executrices and trustees of the will and did not make any gift to charity but left “all (his properties) movable and immovable, whatsoever and wheresoever situate (including any property over which I may have a general power of appointment or disposition by will)” to Madam Ng as to 50% thereof and his five children 10% each. Clause 6 is highly unusual and provided: “For the avoidance of doubt, I declare that I have no assets which are held by my daughters or by any other person in trust for me and that all the assets previously belonged to me or financed by me which are currently held by my daughters are not held in trust for me.” 91. On the same day, Madam Ng executed a will which was the mirror image of the 2006 will. However, Clause 6 in Madam Ng’s will went on to say that shares in AAS which were registered “under my name belonged absolutely to my daughter, (the 1st respondent) and I had executed a trust deed in favour of (the 1st respondent) accordingly”. 92. On the same day, the testator and Madam Ng made a joint statement (“the parents’ statement”), witnessed by the Singapore lawyer who had witnessed their wills, which read: “We, (the deceased) and(Madam Ng), husband-and-wife, wish to state that we have no assets which are held by our daughters or by any other person in trust for either or both of us. We wish to further state that all the assets previously belonged to any or both of us or financed by any or both of us which are currently held by our daughters are beneficially owned by our daughters and are not held in trust for anyone or both of us. I, (Madam Ng), wish also to put on record that the shares in (AAS) which are registered under my name belonged absolutely to my daughter, (1st respondent) and I had executed a trust deed in favour of my daughter, (1st respondent) accordingly. We make this statement in order to avoid any doubt that our other family members may have as regards the assets currently owned or held by our daughters.” 93. Leaving aside consideration of the legal effect, if any, of Clause 6 of the 2006 will or the parents’ statement, it seems clear that the shares in AAIE or in AAS, now said to be owned beneficially by the testator were not then regarded by the testator as part of his estate nor were they meant to devolve under the 2006 will.[46] However, if the testator was, notwithstanding Clause 6 and the joint statement, in fact the beneficial owner of these shares, the language, (“all his properties”)[47] of the 2006 will was wide enough to cover them such that they would devolve as part of his estate. 94. On 21 December 2006, the two daughters wrote a joint letter (“the joint undertaking”) to their parents which was signed and sealed by them in the presence of the same Singapore solicitor who had witnessed the 2006 wills. In the joint undertaking the daughters stated that although they were the beneficial owners of 3,500,000 ordinary shares in AAIE, they undertook that “when father is not around”, to give some[48] of the AAIE shares[49] to their mother and brothers in the manner stated in the letter. The joint undertaking was said to have been made at the request of the testator. If it was meant to have legal effect, it might be enforceable.[50] There was no mention of the 2,542,590 shares in AAS held in the name of the 1st respondent. 95. On 30 March 2007, the testator made a new will in Hong Kong in the Chinese language[51], prepared by Hastings, a long established firm in Hong Kong. The 1st respondent was appointed the sole executrix and trustee. Under the 2007 will, subject to payment of debts etc, the testator: “(give) as a gift all the properties under my name in Hong Kong and any other places (including movable properties and immovable properties)(hereinafter called “my remaining properties”) to my wife, … for her to inherit alone, and to fully possess and enjoy.” However, should Madam Ng not survive the testator by at least 30 days, all his remaining properties would be devoted to charitable uses. Clause (VI) of the 2007 will provided: “I hereby declare, all my children and descendants have received adequate and appropriate care during my lifetime; I hope that they can be self-reliant, and do good to the society and mankind. Therefore, I do not intend to give any portion of my remaining properties to them as a gift. I also hope that all my children and descendants can get along harmoniously, and (they) shall not rashly commence any lawsuits or dispute over my remaining properties.” Madam Ng also made a mirror image will on the same day. 96. The 2007 will, as one would expect, expressly revoked the 2006 will. However, we were left in the dark about the testator’s attitude over Clause 6 of the 2006 will, or the parents’ statement of 21 April 2006. It is not clear whether the AAS or AAIE shares held in the names of one or more of his daughters were then regarded or understood by the testator to be part of the “remaining property” disposed of under the 2007 will. The use of the expression “all the properties under my name” in this context, is not illuminating. We do not know why Clause 6 of the 2006 will was not repeated. The testator sitting in his arm-chair would have been aware that he had said in Clause 6 of the 2006 will and in the parents’ statement that these shares were not beneficially owned by him. He would also know that in the joint undertaking made at his request, his daughters stated that they were the beneficial owners of the 3,500,000 shares in AAIE. Unfortunately, the solicitors’ notes in connection to the 2007 will have not been produced. I believe one may fairly proceed on the basis that the solicitors must have been aware of the content of the 2006 will and the existence of the parents’ statement and the joint undertaking. It is a mystery why no illumination was provided in the 2007 will or at all. In such context, the simple language chosen, viz. “all the properties under my name” is puzzling. Madam Ng’s 2007 will also used the expression “all the properties under my name”, given that she had admitted that she held the AAS shares on trust for the 1st respondent, if that was true, no beneficial interest in the AAS shares could pass under her will. But since we are not concerned with her will, nor were any submissions made in regard to her will, I will say no more. 97. The next will was made on 6 February 2008 and this is the will in suit. 98. This is the joint will[52] of the testator and Madam Ng but we are only concerned with the testator’s part of the will. It is in Chinese and drafted by Hastings with the 1st respondent as the sole executrix and trustee. It revoked all previous wills. Clause (II) stated “We hereby declare that this Will shall govern and deal with all the properties under our names (wheresoever situate worldwide).” In this respect, there is no difference in substance from the 2007 will. However, it differed importantly from the 2007 will in that instead of leaving the residuary estate to the survivor absolutely, the survivor was given a life interest in the income (Clause V) with a gift over to charities. Some details were provided, for example, HK$100,000,000 to different charitable uses in Hong Kong, and a similar amount to charitable uses in their native village in the Mainland. There were also other charitable uses, in respect of which no specific amount was stated but in relation to which substantial sums were likely to be required, for example, a donation to Singapore Buddhist Lodge for study and research, scholarships for two universities in Singapore as well as ten top universities in the world. 99. There is a Clause (XI) which is similar in effect to the Clause (VI) of the 2007 will. 100. In the construction of the 2008 will, I have not found it necessary to resort to either Clause (V) which gave the survivor a life interest in the income or Clause (XI). Clause (V) is a clause which one would expect to find in a will which provided for a life interest. On the available evidence it does not appear to me that any such income was necessary for the livelihood of the testator or Madam Ng. Clause (XI) is common in a will where no provision is made for persons for whom one would normally provide. My view on the proper construction of the 2008 will does not depend on such standard clauses. 101. However, the generous provisions made for charities in the 2008 will are important because they show that, rightly or wrongly, the testators thought their estates were substantial. However, if Mr Fung is correct, as Ribeiro PJ’s judgment explains[53], the assets in the estates would have been quite inadequate.[54] 102. A document much relied on by Mr Fung which postdated the 2008 will is the letter dated 16 August 2008 addressed to their children by the testator and Madam Ng. Ribeiro PJ dealt with them in paras 20 and 21 of his judgment. I do not believe this letter helps Mr Fung at all. It is true that the letter showed that the testator had intended to give the AAIE shares to Madam Ng and the children.[55] The letter showed that, rightly or wrongly, the testator regarded the shares as being at his disposal. Whether that was because he was their beneficial owners or because of his moral authority over his children is not for me to decide. Nor was Mr Fung content to stop at the AAIE shares. He argued that the testator had intended to make a gift of the AAS shares or at least AAS’s holding in EnGro to the appellants or some of them. I must confess that here, Mr Fung has an impossible task. Far from supporting Mr Fung’s contention, the letter clearly contradicts it. The parents were clear that their other assets, i.e. other than the AAIE shares should go to charities. Nor does the letter assist in the construction of the expression “under our names”. Mr Fung suggested that when the testator used the expression “under our names”, he intended to exclude from his will properties owned by him which were not held in his name. He submitted that the testator had intended to deal with such properties during his lifetime, but since unfortunately, he had failed to dispose of them inter vivos, there was an intestacy in relation to them. I can readily accept that as the letter of 16 August 2008 made clear the testator and Madam Ng intended to deal with the AAIE shares and the AAS shares in a deed of family arrangement to be drafted by solicitors “to avoid family members having any arguments or disputes amongst themselves in future”, I cannot accept that the testator intended by the use of the expression “all the properties under our names” to indicate that failing disposition inter vivos they should devolve as on intestacy. That flies in the face of his intention to benefit the various charities so clearly stated in the 2008 will as well as in the letter. If it was the testator’s conscious decision to exclude properties beneficially owned by him but not held in his name from his will, I do not believe he would have chosen to communicate his intention so obliquely[56]. Indeed, at the making of the 2008 will, either the deceased regarded the shares as beneficially owned by him or he did not. If he did and he intended to exclude them, I believe he would have said so clearly in the will.[57] If he did not regard them as beneficially his, then the expression “all our properties” could not have been chosen with the exclusion of those assets in mind. Ribeiro PJ in his judgment has highlighted other difficulties with Mr Fung’s construction, for example, the possible exclusion of valuable properties such as paintings and antiques. I regret to say, since no inter vivos disposition was made, the AAIE shares, if they were beneficially owned by the testator, would devolve under the will. 103. Since, as Mr Fung accepted, as he must, that one of the meanings of “我們名下…的財產” “properties under our names” is “properties belonging to us”, I am of the view that properly construed in the light of all the circumstances highlighted above, the testator intended by the expression “all the properties under our names” to cover all properties belonging to him. 104. For the above reasons, I agree that the appeal must be dismissed. It is a matter of regret that notwithstanding Madam Ng’s admirable effort[58] to give effect to the wishes of the testator[59], the parties were unable to settle their dispute amicably. I can only say that even now, with the co-operation of Madam Ng, a settlement may still be achieved. Mr Justice Fok PJ: 105. I agree with the judgment of Mr Justice Ribeiro PJ. Lord Neuberger of Abbotsbury NPJ: 106. I agree with the judgment of Mr Justice Ribeiro PJ. Chief Justice Ma: 107. The Court unanimously dismisses the appeal and makes the orders as to costs set out in the final paragraph of the judgment of Mr Justice Ribeiro PJ. (Joseph Fok)				Permanent Judge (Lord Neuberger of Abbotsbury)				Non-Permanent Judge Mr Patrick Fung SC and Mr Johnny Ma, instructed by Fairbairn Catley Low & Kong, for the 1st to 4th Plaintiffs (Appellants) Mr Wong Yan Lung SC and Mr Derek Hu, instructed by ONC Lawyers, for the 1st Defendant (1st Respondent) Ms Agnes CM Chan SGC, of the Department of Justice, on watching brief for the 2nd Defendant (2nd Respondent) Annex A Joint Will of	TAN Kiam Toen	NG Giok Oh I, TAN Kiam Ton, holder of Hong Kong Permanent Resident Identity Card Number [XXX], and I, NG Giok Oh, holder of Hong Kong Identity Card Number [XXX], currently residing at [XXX], Bowen Road, Hong Kong, now make this our Will and the contents are as follows : (I) We hereby revoke all our previous wills and testamentary dispositions heretofore made by us and declare that this to be our last Will. (II) We hereby declare that this Will shall govern and deal with all the properties under our names (wheresoever situate worldwide). (III) We are both permanent residents of Hong Kong Special Administrative Region, and Hong Kong is our place of domicile. We hereby declare that the interpretation and execution of this Will shall be in accordance with the laws of Hong Kong. (IV) We hereby appoint our daughter, TAN Choo Suan, holder of Hong Kong Identity Card Number [XXX], to be the sole Executrix and Trustee of this our Will (hereinafter referred to as “our Trustee”). (V) We mutually undertake to each other that in the event one of us predeceases the other, subject to the payment of all the necessary expenses (such as payment of debts, payment of funeral expenses and payment of the requisite testamentary expenses, and also payment of government taxes, et cetera), the survivor of us can enjoy during his/her lifetime the income derived from the estate of the first-deceased of us. (VI) When the survivor of us has also passed away, we handover all our properties to our Trustee for her to make the judgment/s based on the actual circumstances and needs, and to exercise her power to sell, resell, lease, mortgage, convert into cash our properties or any portion thereof. Our Trustee shall also have the power to revise, amend, (or) cancel the aforesaid act, or retain the said properties and maintain the status quo. However, she shall not be liable for any loss that may arise from the sale or from a delay in the sale of our properties. And subject to the payment of all the necessary expenses (such as payment of debts, payment of funeral expenses and payment of the requisite testamentary expenses, and also payment of government taxes, et cetera), our Trustee shall hold our residuary estate on trust and shall make the following arrangements. (VII) Our Trustee is required to allocate a sum of Hong Kong dollars one hundred million (HK$100,000,000) from our residuary estate and donate it to organizations, bodies, or units in Hong Kong to be used for charity, medical, education, elderly care, and help for the young and the poor. As for the decision on the specific beneficiary organizations, bodies or units, the amount to donate, and the form of donation, it shall be for our Trustee to exercise her absolute discretion. The receipts issued by the persons in charge of the beneficiary organizations, bodies or units, (such as the chairman, director, administrative secretary, manager or other authorized personnel) shall be valid, and can discharge our Trustee of her responsibilities. (VIII) Our Trustee is required to allocate a sum of Hong Kong dollars one hundred million (HK$100,000,000) from our residuary estate and donate it to the organizations, bodies, or units in our home town – Shanting Village, Maxiang Town, Xiang’an District, Fujian Province, China – and in Xiamen to be used for charity, medical, education, elderly care, and help for the young and the poor. As for the decision on the specific beneficiary organization, bodies or units, the amount to donate, and the form of donation, it shall be for our Trustee to exercise her absolute discretion. The receipts issued by the persons in charge of the beneficiary organisations, bodies or units, (such as the chairman, director, administrative secretary, manager or other authorized personnel) shall be valid, and can discharge our Trustee of her responsibilities. (IX) Our Trustee is required to use our residuary estate for the following purposes : 1. Donate a sum of money to Singapore Buddhist Lodge to be used for setting up a school for Buddhist studies. As for the decision on the amount to donate and the form of donation, it shall be for our Trustee to exercise her absolute discretion. 2. Donate a sum of money to National University of Singapore. As for the decision on the amount to donate and the form of donation, it shall be for our Trustee to exercise her absolute discretion. 3. Set up a permanent scholarship, to encourage and reward Chinese students who could gain admission into National University of Singapore, Nanyang Technological University of Singapore, and the top ten institutions of higher learning in the world to pursue a bachelor’s degree course. As for the decision on the list of students to be awarded the scholarship, the amount and number of scholarships to be awarded, and which universities are “the top ten institutions of higher learning in the world”, our Trustee may exercise her absolute discretion. 4. Donate monies to organizations, bodies or units in need, to be used for the care of the aged, for medical and other social charitable services in the Republic of Singapore. As for the decision on the specific beneficiary organizations, bodies or units, the amount to donate, and the form of donation, it shall be for our Trustee to exercise her absolute discretion. The receipts issued by the persons in charge of the beneficiary organizations, bodies or units, (such as the chairman, director, administrative secretary, manager or other authorized personnel) shall be valid, and can discharge our Trustee of her responsibilities. We hope that the donations for the foregoing Item No.(IX) could be executed and operated through a charitable trust fund set up in our names in the Republic of Singapore. (X) We hereby direct that during the lifetime of our Trustee, based on the actual circumstances and needs, she can decided whether it is necessary to appoint other person/s or organization/s to replace her as executor of this Will or to be the joint executor of this Will with her. In the event the executor of this Will is a professional or is an organization, the said executor can enjoy the usual remunerations a professional/organization is entitled to in administering our estate and in conducting the businesses in connection with the trust (including some acts that could also be carried out by non-professional persons or organizations). (XI) We hereby declare, all our children and descendents have received adequate and appropriate care during our lifetime. We hope that they can be self-reliant, and do good for society and mankind. As such, we do not intend to give any portion of our residuary estate to them. We also hope that all our children and descendents can get along harmoniously, and (they) must not commence any lawsuits or disputes indiscriminately over our residuary estate. We, TAN Kiam Toen and NG Giok Oh, signed this Will as our last Will on 6 February 2008 in the presence of the undersigned witnesses; the witnesses have at the request of the testators signed as witnesses at the same time, and when signing as witnesses, the testators and the witnesses were all present at the same time, attesting for each other. Signed (Testator) Signed (Witness) Annex B CODICIL I, TAN Kiam Toen, holder of Hong Kong Permanent Resident Identity Card Number [XXX], and I, NG Giok Oh, holder of Hong Kong Identity Card Number [XXX], currently residing at [XXX], Bowen Road, Hong Kong, hereby declare this to be a First Codicil to our Joint Will made on 6 February 2008. We hereby make the following additions to our Joint Will made on 6 February 2008 : - (1) In Australia we own a real estate property located at [XXX], Castle Cove, NSW 2069, Australia, as joint tenants (hereinafter referred to as the “Australia Sydney Property”). (2) On 24 May 2003, we arranged for our eldest daughter, TAN Choo Suan, (Hong Kong Identity Card Number [XXX]) to become the third joint tenant for our Australia Sydney Property, to own the said Australia Sydney Property jointly with us. (3) In Hong Kong we also own a real estate property located at [XXX] Bowen Road, Hong Kong, as joint tenants (hereinafter referred to as the “Hong Kong Bowen Mansion Property”). (4) As all along Choo Suan is not married and (she) has no children, during our old age she has also resigned from an ideal job in America and returned to Hong Kong, and she has all along been keeping us company and taking care of us, as such, our wish is that after both of us have passed away, Choo Suan will solely inherit and wholly own and enjoy the use of our Australia Sydney Property and Hong Kong Bowen Mansion Property. (5) We are award and also understand that as one of the joint tenants of the property, after both of us have passed away, according to the legal principle of right of survivorship, Choo Suan will automatically become the sole owner of the Australia Sydney Property. (6) As for the Hong Kong Bowen Mansion Property, we now direct that after both of us have passed away, Choo Suan will solely inherit and wholly own and enjoy the use of (the property). (7) In the event during our lifetime (or during the lifetime of the survivor of us), the Hong Kong Bowen Mansion Property is sold and another new property is purchased to be used as (our) residence, then the directions in paragraph (6) shall apply to the newly purchased property. In other words, the Principle of Ademption shall not apply to the Hong Kong Bowen Mansion Property. (8) Other than the foregoing additions/amendments, we confirm that the other contents of the “Joint Will” we made on 6 February 2008 are correct, and are still valid. We, TAN Kiam Toen and NG Giok Oh, signed this instrument on 8 May 2008 to be a First Codicil to our Joint Will made on 6 February 2008 in the presence of the undersigned witnesses; the witnesses, at the request of the testators, have signed as witnesses at the same time, and when signing as witnesses, the testators and the witnesses were all present at the same time, attesting for each other. Signed (Testator) Signed (Witness) [1] Cap 30. [2] See Williams on Wills (Lexis Nexis, 10th Ed), para 2.2. [3] The details of this transaction may be subject to dispute in the Singapore proceedings, but are accepted as stated above for the purposes of the Hong Kong proceedings. [4] On the same day, Madam Ng executed a will which was the mirror image of the 2006 Will. [5] This was also stated in a signed and witnessed statement of the same date. [6] DHCJ Chu (see below) at §8(11). [7] If AAIE were wound up, the EnGro shares were to be distributed along the same lines. [8] In Chinese: “名下所有在香港以及其它任何地方的財産(包括動産及不動産)”. [9] Again jointly with Madam Ng. [10] As amended by additions made on 9 June 2011 and 18 March 2013. [11] On 11 July 2013. [12] HCMP 246/2013 (6 September 2013). [13] Judgment §40 and §42. [14] Judgment §§48-50. [15] CACV 200/2013 (23 May 2014) Hon Lam VP, Kwan JA and Poon J, Poon J giving the Judgment of the Court. [16] Court of Appeal §39 [17] Court of Appeal §40. [18] (2015) 18 HKCFAR 169. [19] [2015] AC 129. [20] Chinachem Charitable Foundation at §27. [21] Chinachem Charitable Foundation at §29, citing Marley at §20. [22] Chinachem Charitable Foundation at §29 citing Marley at §19. [23] As Lord Walker NPJ points out (Chinachem Charitable Foundation at §30), section 19 of the Wills Ordinance provides that: “Every will shall, with reference to the property comprised in it, be construed to speak and take effect as if it had been executed immediately before the death of the testator, unless a contrary intention appears from the will”, the same applying equally to any general description of a class of beneficiaries. [24] Chinachem Charitable Foundation at §31. [25] Chinachem Charitable Foundation at §31 citing Lord Mance in Re Sigma Finance Corporation [2010] 1 All ER 571 at §12. [26] Chinachem Charitable Foundation at §31. [27] Which is based on and identical to section 21 of the English Administration of Justice Act 1982. [28] “Applied” appears to be a typographical error. [29] Judgment §40. [30] Appellants’ printed case, §10. See also §36. [31] Appearing with Mr Johnny Ma for the appellants. [32] Williams on Wills (Lexis Nexis, 10th Ed), para 51.1. [33] (1885) 30 Ch D 390 at 393. [34] Williams on Wills (Lexis Nexis, 10th Ed), para 2.1. [35] The Secretary for Justice, however, has not taken any active part in the proceedings. [36] We are not concerned with Madam Ng, who, happily, is still with us. [37] Para 42. [38] Lam VP, Kwan JA and Poon J. [39] Ribeiro, Tang and Fok PJJ. [40] Pursuant to Rule 7 HKCFA rules Cap 484. [41] We are not concerned with Madam Ng. [42] A 70% interest in EnGro. [43] The closing price of these shares was S$0.66 on 14 Nov 2008, the day before the testator died. [44] This was given an open market value of $65,000,000 as at 26 May 2009. [45] Namely, 3,500,000 shares in AAIE and 2,542,590 shares in AAS. [46] The same might be said of Madam Ng’s holding of AAS shares in regard to her 2006 will. [47] See para 90 above. [48] Some only in the sense that they would retain some of the shares for themselves. [49] In the event of the liquidation of AAIE, shares in EnGro. [50] Beswick v Beswick[1968] AC58. [51] I shall use the English translation of the 2007 will and the 2008 will unless it is necessary to consider the Chinese original. [52] There is a codicil dated 8 May 2008, see para 19, Ribeiro PJ’s judgment above. [53] In sections E.2 and E.3. [54] I have not ignored Madam Ng’s shares in AAS but it is reasonably clear that the testator’s estate was the more substantial. [55] But they would share them differently from what was stated in the joint undertaking. [56] And if he had communicated such intention to his solicitors (as one would expect), I believe such intention would have been clearly spelt out in the will. [57] One example is Clause 6 of the 2006 will. Though if the testator was wrong in thinking that they were not beneficially his, they would nevertheless pass under his will if the expression “under our names” properly construed meant “belonging to us”. [58] Apparently with the agreement of the 1st respondent. [59] Para 23, Ribeiro PJ’s judgment above. 
Chief Justice Ma: A INTRODUCTION A.1 The issues in the appeal 1. Employees under contracts of employment sometimes unfortunately find themselves made redundant.[1] The law steps in to soften such a situation in a number of ways. Part VA of the EO deals with the obligation on the part of an employer to make what are known as severance payments to the employee who has been made redundant. Severance payments are payments reflecting the length of service of an employee subject to maximum payments stipulated in the EO.[2] In the calculation of the severance payment due to a redundant employee, deductions are, however, to be made to what otherwise would be the severance payment where the employee benefits from gratuities under the relevant contract of employment based on length of service, or from an occupational retirement scheme (“OPS”), or certain benefits under a mandatory provident fund (“MPF”) scheme.[3] I shall refer to these benefits under s 31I of the EO as “the s 31I benefits”. As we shall see, the proper treatment of these benefits is of considerable importance in this appeal. 2. Where severance payments are due to an employee but the employer fails to pay the employee, either wholly or in part, owing to insolvency, limited protection is given to the employee in the form of ex gratia payments made under the Protection of Wages on Insolvency Ordinance,[4] the statute with which this appeal is concerned. A fund has been set up under the PWIO[5] (“the Fund”) out of which the ex gratia payments are made. There are limits spelt out in the Ordinance as to the amount of ex gratia payments referable to severance payment that may be made. The relevant provision under consideration in this appeal is s 16(2)(f)(i)[6] which states that the amount of ex gratia payment referable to severance payment is capped at and therefore cannot exceed the aggregate of $50,000 plus half of the employee’s entitlement to severance payment in excess of $50,000. The Fund is administered by the Protection of Wages on Insolvency Fund Board (the 1st Respondent in the appeal).[7] The person responsible for determining and paying the amount of ex gratia payment under the PWIO is the Commissioner for Labour, to whom any application for ex gratia payment under the Ordinance is made.[8] 3. This appeal,[9] brought in judicial review proceedings instituted by the appellant (I shall refer to him from now on as the Applicant) against both Respondents, involves the determination of the approach to be adopted in the calculation of ex gratia payments payable under the PWIO as relate to severance payments. As will be seen, the general question for determination is how the cap in s 16(2)(f)(i) is calculated. The specific question within this is how and at what stage are the s 31I benefits to be taken into account in the computation of any ex gratia payment to be made under the PWIO. This is what divides the parties :- (1) The Applicant contends that the relevant figure representing the severance payment to which an employee is entitled for the purposes of the application of the cap in s 16(2)(f)(i) of the PWIO is a net figure, meaning the actual amount of severance payment that is due to an employee. Where s 31I benefits exist, these should be taken into account beforehand in arriving at the net figure to which the statutory cap should be applied. Once the statutory cap is applied in this way, the amount of ex gratia payment can then be ascertained. (2) The Respondent argues on the other hand that the statutory cap in s 16(2)(f)(i) is applied to the gross amount of severance payment payable before any deductions are made; in other words, one does not take the amount of severance payment that is actually owing to an employee but only the original, gross amount of severance payment that was owing. Where s 31I benefits exist, these are deducted only after the capped figure is calculated in order to arrive at the extent of the ex gratia payment payable. A.2 The facts 4. The essential facts in the proceedings were not in dispute. The Applicant was formerly employed as a driver by Fu Ming Transport Co Ltd from 6 November 1999 to 7 October 2011. On 7 October 2011, Fu Ming went into voluntary liquidation and dismissed its employees (including the Applicant) with immediate effect. Like other employees, the Applicant was owed arrears of wages, wages in lieu of notice and severance payment. He applied to the Respondent for ex gratia payments in relation to all these items.[10] Ex gratia payments were duly made in relation to arrears of wages and wages in lieu of notice, and there is no dispute about these. As regards severance payment, the Respondent was of the view that no ex gratia payment should be made in relation to this item. His decision was upheld by the Board.[11] The present proceedings were commenced to challenge this decision and to have it quashed. 5. As regards severance payment,[12] the Applicant was entitled to a gross amount of $131,696.54. His s 31I benefits amounted to $106,319.04. The respective stances of the Applicant and the Respondent as set out above,[13] reflect the practical implications of the questions to be determined in this appeal :- (1) The Applicant’s position is simple. The net amount of severance payment to which he is entitled – in other words, the amount he is actually owed – amounts to $25,377.50.[14] This figure being less than the statutory cap contained in s 16(2)(f)(i) of the PWIO,[15] this is, according to the Applicant, the sum payable as ex gratia payment to him. (2) The Respondent’s method of calculation results in a quite different conclusion. The Respondent takes the gross figure of severance payment owed, that is $131,696.54 and this, he says, is the relevant figure to which the statutory cap is applied. This results in a capped figure of $90,848.27.[16] The s 31I benefits of $106,319.04 are then deducted from this to arrive at the ex gratia payment payable to the Applicant. In the present case, the Respondent contends, this results in a negative figure[17] and therefore no ex gratia payment is payable. 6. Before leaving the facts, I would just add the point that the methodology adopted by the Respondent has been used by the Commissioner “for decades”.[18] This approach is reflected in the standard forms used by the Respondent. In the Notice of Payment,[19] issued by the Respondent to the Applicant in the present case, the Respondent’s approach is clearly set out in the part headed “Points to Note”. A.3 The decisions in the courts below 7. Both the lower courts agreed with the Respondent’s position and the Applicant’s application for judicial review was dismissed. It will presently be necessary to analyze in greater detail their reasoning in Section C below. At this point, however, it is convenient to dispose of one argument raised by the Respondent in the lower courts, a point that is no longer pursued in this Court. It was at one stage contended by the Respondent that the Commissioner had an open discretion to do as he thought fit as regards the making of ex gratia payments, both as to whether to make any payment at all and if so, as to what amount, subject only to the statutory cap.[20] This argument was rejected by the Court of Appeal.[21] With respect, Cheung CJHC was correct to reject this submission. Before us, counsel for the Respondent[22] submitted that the discretion vested in the Respondent was exercisable in cases where, for example, there may be collusion, but it was accepted that the Respondent was bound to make ex gratia payments in accordance with what the PWIO required him to do. The only dispute, therefore, was what the PWIO, properly construed, prescribed as far as the calculation of ex gratia payments were concerned. A.4 The certified Question 8. On 3 June 2015, the Court of Appeal granted leave[23] to appeal to this Court on the following question of great general or public importance :- “The true interpretation of section 16(1) and other related provisions in the Protection of Wages on Insolvency Ordinance (Cap 380) regarding how the amount of an ex gratia payment should be fixed in circumstances such as the present – that is, where it is known to the Commissioner for Labour at the time of application for an ex gratia payment under the Ordinance that the employee/applicant is entitled to, or, as the case may be, has been paid one or more gratuities and/or benefits as mentioned in section 3II of the Employment Ordinance (Cap 57), or in the other similar or related circumstances.” B THE STATUTORY SCHEME REGARDING EX GRATIA PAYMENTS UNDER THE PWIO 9. The making of ex gratia payments in the context with which this appeal is concerned is statutory and it is therefore necessary to construe relevant Ordinances by applying well‑known principles of statutory construction. As will appear under this Section and the next, for my part, I agree with the Applicant’s position in this appeal and in turn, with respect, disagree with the approach of the lower courts. In my view, the main error of the courts below has been to misapply relevant principles of statutory construction. 10. The present case is about severance payments. As we have seen,[24] severance payments are payable in the event of redundancy and the relevant provisions here are contained in Part VA of the EO :- (1) The liability to make severance payments is contained in s 31B(1) of the EO. Redundancy is defined in s 31B(2). (2) The amount of severance payment to which an employee is entitled is set out in s 31G and it is subject to the monetary limits set out therein. Calculation of severance payment under s 31G looks essentially to salary and the number of years service of the relevant employee. However, any deductions which should be made to the severance payment due to an employee to arrive at a net figure, are contained in other provisions in Part VA. It is to be noted that s 31G expressly states that its provisions are “Subject to this Part [of the EO].” (3) Deductions to be made to severance payments are set out in s 31I, also contained in Part VA of the Ordinance. Section 31I identifies what I have earlier referred to as the s 31I benefits[25] :- “31I. Severance payment to be reduced by amount of gratuities and benefits in certain cases If an employee becomes entitled to payment of a severance payment under this Part and- (a) because of the operation of the employee’s contract of employment, one or more gratuities based on length of service or one or more relevant occupational retirement scheme benefits have been paid to the employee; or (b) a relevant mandatory provident fund scheme benefit is being held in a mandatory provident fund scheme in respect of the employee, or has been paid to or in respect of the employee, the severance payment is to be reduced by the total amount of all of the gratuities and benefits to or in respect of the employee to the extent that they relate to the employee’s years of service for which the severance payment is payable.” The deductions to be made as set out in this provision are expressly stated to be “under this Part”. The operation of s 31I is therefore restricted to the purposes of Part VA of the EO. (4) The link between severance payments and s 31I benefits is further developed in s 31IA which allows an employer, where severance payment under the EO has been made to an employee, to make deductions (to the extent of the severance payment made) from any applicable gratuity based on length of service under the relevant contract of employment, or from an ORS, or from an applicable benefit under an MPF scheme. Further provisions in this regard are identified in s 31IA(3), namely s 70A of the Occupational Retirement Schemes Ordinance[26] and s 12A of the Mandatory Provident Fund Schemes Ordinance.[27] As in the case of s 31G and s 31I, the provisions contained in s 31IA are also expressly stated to apply “under this Part”. 11. Where there is a failure to make severance payments owing to the employer’s insolvency, employees may apply to receive ex gratia payments. The relevant provisions in this context are contained in the PWIO :- (1) Part 5 of the PWIO deals with payments to be made out of the Fund. (2) The entitlement to apply for an ex gratia payment is contained in s 15 :- “15. Entitlement to apply for ex gratia payment (1) Subject to this Part, an applicant to whom- (a) wages are due and unpaid; (b) wages in lieu of notice are due and unpaid; (c) the liability to be paid a severance payment has arisen and the severance payment is unpaid, whether or not the severance payment is then due; (d) pay for untaken statutory holidays is due and unpaid; or (e) pay for untaken annual leave is due and unpaid, may apply for an ex gratia payment from the Fund in respect of the wages, wages in lieu of notice, the severance payment, the pay for untaken statutory holidays or the pay for untaken annual leave, as the case may be, or all or any of them.” We are of course concerned only with the aspect of severance payments in the present case. (3) Accordingly, applications can be made where the liability to make a severance payment has arisen and the severance payment is unpaid. This immediately raises the question what is meant by a severance payment in this context of the PWIO. Severance payments are dealt with generally in the EO.[28] However, as far as the PWIO is concerned, it is clear that what must be referred to is the actual amount of severance payment that is due to an employee. The definition section of the PWIO[29] defines severance payment as follows :- “severance payment (遣散費) means a severance payment in respect of which an applicant would, on the winding up of a company, be entitled to priority under section 265(1)(ca) of the Companies (Winding Up and Miscellaneous Provisions) Ordinance (Cap 32) or would, on a bankruptcy, be entitled to priority under section 38(1)(ca) of the Bankruptcy Ordinance (Cap 6), save that the amount specified in section 265(1)(ca) of the Companies (Winding Up and Miscellaneous Provisions) Ordinance (Cap 32) or section 38(1)(ca) of the Bankruptcy Ordinance (Cap 6) (as the case may be) as the maximum amount in respect of which an applicant would be entitled to priority, or any amount substituted therefor under section 265(1B) of the Companies (Winding Up and Miscellaneous Provisions) Ordinance (Cap 32) or section 38(2B) of the Bankruptcy Ordinance (Cap 6) (as the case may be), shall not apply;” (4) From this definition, it is clear that the relevant severance payment (for the purposes of the PWIO) is the actual severance payment to which an employee would be entitled; quite simply it is the net entitlement to severance payment as to which an employee would be claiming in a winding up (in the case of a company) or in a bankruptcy (in the case of an individual employer), and in respect of which the employee would be able to claim priority. It is also to be noted that those provisions of the Companies (Winding Up and Miscellaneous Provisions) Ordinance Cap 32[30] and the Bankruptcy Ordinance Cap 6[31] expressly referred to in the definition – these deal with priority of certain claims in a winding up or bankruptcy – both use the term “any severance payment payable”. Priority is given to actual amounts owing, not theoretical ones. (5) Where an employee makes an application under s 15 for an ex gratia payment in relation to unpaid severance payment, it is for the Respondent to consider the application. Where it appears to the Commissioner for Labour that an employer has failed to pay any or any part of a severance payment, that employer being insolvent,[32] an ex gratia payment may be made to the Applicant out of the Fund. Section 16(1) states :- “(1) Subject to subsections (1B) and (2) where it appears to the Commissioner that an employer has failed to pay any wages, wages in lieu of notice, severance payment, pay for untaken statutory holidays or pay for untaken annual leave, as the case may be, or all or any of them to an applicant and that – (a) in the case of an employer who is not a company – (i) a bankruptcy petition has been presented against him; or (ii) he would, but for the existence of section 6(2)(a) of the Bankruptcy Ordinance (Cap. 6), be liable to have a bankruptcy petition presented against him; or (b) in the case of an employer who is a company, a winding-up petition has been presented against that employer, he may make an ex gratia payment to the applicant out of the Fund of the amount of the wages, wages in lieu of notice, severance payment, pay for untaken statutory holidays or pay for untaken annual leave, as the case may be, or all or any of them.” (6) Under s 16(1B), where a severance payment is “not due” at the time an application is made, the Respondent may defer consideration of it until that severance payment becomes due. (7) It is in this context in which one then comes to consider s 16(2)(f)(i), very much the critical provision to be considered in this appeal. Section 16(2) is put in negative terms, that the Respondent is not to make any payment in the circumstances set out. Section 16(2)(f)(i) essentially sets out the financial limits of the ex gratia payment payable in respect of severance payment. That provision states :- “(2) The Commissioner shall not make any payment under subsection (1) …. (f) in respect of a severance payment – (i) of an amount exceeding the aggregate of $50,000 and half of that part of the applicant’s entitlement to severance payment in excess of $50,000 ….” (8) Given the statutory definition of severance payment in s 2 and in the context set out in sub-paras (5) to (7) above, it is clear that the financial limit set out in s 16(2)(f)(i) is to be applied to the actual amount of severance payment owing to an employee. And the actual amount of severance payment due to an employee – in the words of the provision, the “applicant’s entitlement” – is the net amount due to that employee. (9) Where, as in the present case, s 31I benefits are involved, the actual or net amount of severance payment due to an employee is the original entitlement to severance payment calculated in accordance with s 31G of the EO less the s 31I benefits. This net amount will accordingly be the figure in respect of which the financial limits in s 16(2)(f)(i) are to be applied. 12. In my view, the system of ex gratia payments regarding severance payments is to be operated in the way set out in the previous paragraph. The Applicant’s position in this appeal accords with this analysis.[33] I have somewhat painstakingly gone through the relevant provisions in view of the contrary way both courts below had approached the exercise and of the methodology employed by the Respondent up to now.[34] 13. I must now deal with the reasoning of the lower courts. This reasoning was supported by the Respondent in this Court. C THE REASONING OF THE LOWER COURTS 14. Louis Chan J dismissed the application for judicial review and this was upheld by the Court of Appeal. The crucial effect of the lower courts’ decision was in relation to the application of the financial limits set out in s 16(2)(f)(i) of the PWIO. They took as the relevant figure as to which the limits in that provision were to be applied, the original or gross figure as to an employee’s entitlement to severance payment, namely, the figure calculated in accordance with the formula set out in s 31G of the EO. Thus, in the present case, the Respondent takes as the applicable figure the sum of $131,696.54,[35] and it was to this figure that the financial limits in s 16(2)(f)(i) had to be applied. 15. The Judge at First Instance was of this view. The Court of Appeal agreed with the Judge but its conclusion here and its analysis of the application of s 16(2)(f)(i) are with respect in error :- (1) The root of the error was perhaps what the Court of Appeal perceived to have been common ground between the parties. In paragraph 41 of its judgment, it is stated :- “41. As explained, although the parties were in disagreement before the judge (see paras 4 and 5 of the judgment), they are agreed before this court regarding how the cap is to be calculated. Both sides now agree that the Commissioner (and the judge below) are correct in using the section 31G formula (disregarding the reduction in section 31I) to calculate the cap.” It is far from clear how this concession came about because it would appear from both the judgment of the Court of Appeal and at First Instance that the Applicant was contending for quite the opposite (and this position was maintained by counsel in the appeal before us). Neither Mr Shieh SC nor Mr Ho SC was quite able to explain the so‑called concession either. Be that as it may, it perhaps suffices to say that even if a concession had been made, this Court is not bound by it. (2) No real analysis of the relevant provisions of the PWIO (referred to in para 11 above) was undertaken. Instead, there seemed almost to be an assumption that s 16(2)(f)(i) applied only in respect of the gross amount of severance payment (as I have used this term) due to an employee. (3) Reliance was placed on s 16(2B)(a)(A) to support the Court of Appeal’s conclusion.[36] This provides :- “(2B) (a) Where it appears to the Commissioner that- (i) an applicant’s wages have been reduced during the period of 12 months immediately before he is dismissed or laid off; and (ii) before the wage reduction took effect, the employer of the applicant had given an undertaking to the applicant to the effect that if the applicant was dismissed or laid off after the wage reduction, the severance payment payable to him would be calculated in a manner more favourable to him than that provided for in section 31G of the Employment Ordinance (Cap 57), then, for the purposes of subsection (2)(f)(i), the applicant’s entitlement to severance payment may, if it is more favourable to the applicant, be calculated- (A) subject to paragraph (c), in accordance with section 31G of the Employment Ordinance (Cap 57); or (B) in the manner specified in the undertaking, whichever results in a lesser amount.” It was said that this provision, by its specific reference to s 31G of the EO in describing an applicant’s “entitlement to severance payment”, indicated that s 16(2)(f)(i) was to be applied to the gross amount of severance payment. This is misplaced :- (a) It ignores the wording and context of s 16(2)(f)(i) as explained earlier. (b) The application of s 16(2B)(a)(A) is, as can be seen from its wording, restricted to a rather peculiar factual situation. It goes no further than provide a means of calculation in the factual situation set out when a comparison has to be made between the effect of a statutory provision (s 31G) and that of an employer’s undertaking. (c) As pointed out earlier,[37] s 31G in any event is itself expressly stated to be “Subject to this Part [i.e. Part VA of the EO]”. The deductions which should be made to severance payments under s 31I are under this Part of the EO. (4) It also appeared to be reasoned[38] that as s 31G stated the amount of severance payment to which an employee was “entitled”, this was therefore the same entitlement to severance payment referred to in s 16(2)(f)(i). This was again in error for the same reasons set out in sub‑paras (3)(a) and (c) above. 16. With its erroneous starting point, the Court of Appeal[39] then approached the question of how deductions were to be made in respect of the s 31I benefits in order to arrive at the appropriate amount of ex gratia payment payable under the PWIO. The Court of Appeal concluded that it was only after applying the s 16(2)(f)(i) limits to the gross entitlement to severance payment, would the s 31I benefits be deducted in order to arrive at the ex gratia payment payable. The Court of Appeal said this[40] :- “49. In other words, where the section 31G figure does not exceed the cap, double‑payment is taken care of by simply reducing the ex gratia payment entitlement with the section 31I figure to arrive at the ex gratia payment amount. However, when the section 31G figure exceeds the cap, the cap replaces the 31G figure as the starting point, and the concern about double‑payment is addressed by reducing the section 31I figure from the cap to arrive at the ex gratia payment amount.” Accordingly, on the facts of the present case, no ex gratia payment was payable; the Respondent’s position as I have earlier identified.[41] 17. In support of this method of deducting the s 31I benefits, reliance was placed on what the Court of Appeal perceived to be the legislature’s intention behind the deductions to be made on account of the s 31I benefits; it reasoned as follows :- (1) Since the purpose of s 31I is to reduce the amount of severance payment payable to an employee, so the same intention is to be applied in relation to ex gratia payments in respect of severance payments under the PWIO. (2) Accordingly, since the figure arrived at after an application of s 16(2)(f)(i) of the PWIO is, in a sense, to be treated as severance payment, so the deductions to be made under s 31I should also be made to this figure. Should the s 31I benefits not be so deducted, this would result in what was termed a “double‑payment”, meaning that no account would then be taken of deductions which had to be taken into account under that provision. The Court of Appeal said this[42] :- “47. That the legislation has chosen to cap the section 31G amount is of significance in the present discussion once it is remembered that section 31G only gives one the gross severance payment payable. It comes before section 31I. One only arrives at the net severance payment payable under Cap 57 by reducing the section 31G amount with the section 31I amount, if any. In my view, the legislature obviously treats the payments or entitlement included in section 31I as equivalent or similar in nature to severance payment, so that if they are not reduced from the amount calculated under section 31G, there will be double‑payment of what the legislature intends the employee to receive upon termination of employment by way of severance payment. Whether one calls the section 31I amount a “part‑payment” of the severance payment calculated under section 31G is a matter of language. The important thing is the legislature’s true intention in relation to severance payment and prevention of double‑payment. 48. The importance of all this is that in construing the provisions in Cap 380 in accordance with the legislature’s intention under the scheme to provide quick relief money to an employee who has lost his job suddenly and who has not been paid his severance payment due to him, one should also proceed on the assumption that no double‑payment of the type just described is intended. The only difference is that where the section 31G figure exceeds the cap, the cap, instead of the section 31G figure, is used as the starting figure for working out the ex gratia payment amount.” It was therefore this provision (s 31I of the EO), or rather the legislative intention behind it, that required the Respondent to make the deductions of the s 31I benefits only after the figure (capped by s 16(2)(f)(i)) had been ascertained. 18. Reliance was also placed on the subrogation provisions contained in s 24(2B) of the PWIO to support the Court of Appeal’s analysis.[43] The relevant parts of s 24 state :- “(1) Where a payment is made under section 16 to an applicant in respect of wages due to him, or in respect of any liability to pay him a severance payment, whether or not the severance payment is due when that payment under section 16 is made, all his rights and remedies with respect to those wages or severance payment, as the case may be, existing immediately before that payment under section 16 shall, to the extent of the amount of that payment under section 16, be transferred to and vest in the Board for the benefit of the Fund and the Board may take such steps as it considers necessary to enforce those rights and remedies. (2) The rights and remedies of an applicant in respect of a severance payment transferred to and vested in the Board pursuant to subsection (1) shall include the rights and remedies of the applicant in respect of so much of that severance payment as the applicant would, on the winding up of a company, be entitled to priority under section 265(1)(ca) of the Companies (Winding Up and Miscellaneous Provisions) Ordinance (Cap 32) or would, on a bankruptcy, be entitled to priority under section 38(1)(ca) of the Bankruptcy Ordinance (Cap 6). ……. (2B) If- (a) an applicant is entitled to a payment under an occupational retirement scheme or is a person for whom accrued benefits in a mandatory provident fund scheme are held; and (b) an ex gratia payment in the form of a severance payment is made to the applicant under section 16, the applicant’s rights and remedies are, to the extent of the amount of the ex gratia payment, transferred to, and vested in, the Board for the benefit of the Fund. The Board may take such steps as it considers necessary to enforce those rights and remedies.” 19. As I understand the reasoning, it was said that s 24(2B) gave support in that this provision, by referring specifically to the s 31I benefits, clearly envisaged the Respondent being obliged to make deductions on account of these benefits. Mr Ho SC submitted that s 24(2B) gave the statutory basis for the Respondent to make the deductions of s 31I benefits in the way the Court of Appeal held to represent the correct method. 20. In my view, this approach to the way ex gratia payments are to be computed under the PWIO is wrong and the reasoning employed to arrive at this result is similarly in error :- (1) As stated above, the starting point of the Court of Appeal’s analysis in applying the s 16(2)(f)(i) limits to the gross amount of severance payment payable, is wrong.[44] Nothing in s 16(2)(f)(i), for the reasons already discussed, requires the application of the limits stated therein to a theoretical gross sum, as opposed to an actual, net one. This is enough by itself to undermine and dispose of the approach adopted by the Court of Appeal. (2) Further, there is simply no provision in the PWIO permitting, much less requiring, the deduction of s 31I benefits only after applying the s 16(2)(f)(i) limits to the gross sum. Section 31I, which the Court of Appeal appeared to utilize for this purpose, only applies to make deductions from severance payments for the purposes and in the context of Part VA of the EO. Nothing in the PWIO enables s 31I to be transferred to be used for the purposes of making deductions of s 31I benefits from the relevant sum arrived at after applying the limits contained in s 16(2)(f)(i). The Court of Appeal fell into error by reading into the PWIO words which were not there and also reading into the PWIO a legislative intention which is not manifested in the Ordinance; on the contrary, the legislative intention manifested by the actual language of the relevant statutory provisions in that Ordinance point to an opposite result. I accept that there ought not to be any double payment in the sense that s 31I benefits should be taken into account, but in the context of ex gratia payments under the PWIO, this is done by the calculation of the net sum of severance payment payable for the purposes of applying s 16(2)(f)(i), as we have seen earlier. (3) Nor does s 24(2B) of the PWIO provide any support. In alluding to this provision, the Court of Appeal was of the view that it only applied in circumstances where at the time an application for an ex gratia payment was made under the PWIO, neither the applicant employee nor the Respondent was aware of the existence of s 31I benefits.[45] This is a somewhat odd analysis because nothing in s 24(2B) suggests such a restrictive application. This was in all likelihood a somewhat strained construction to justify the view reached that the s 31I benefits should only be deducted after a gross sum severance payment had been capped under s 16(2)(f)(i). The examples employed in the judgment suggest this.[46] Be that as it may, whatever the reason for this approach, it is wrong. Section 24(2B) is, in the context in which it is found, a provision dealing with subrogation. Like the earlier provisions contained in ss 24(1) and (2), s 24(2B) deals with a situation where the Respondent steps into the shoes of the employee. It applies where, for any reason, an ex gratia payment has included an element of s 31I benefits.[47] Seen in the context of subrogation, this provides no basis (as the Respondent has argued[48]) to enable deductions of s 31I benefits to be made only after applying the s 16(2)(f)(i) limits to the gross severance payment. 21. The factor that appeared to influence the Court of Appeal most in arriving at the conclusion it did was that it saw the making of ex gratia payments under the PWIO as somehow requiring equal treatment for all employees, whether or not the employees were beneficiaries of s 31I benefits. The examples given by the Court of Appeal in its judgment all made this assumption. In other words, the Court of Appeal was anxious to reach a result that enabled all employees to be treated in the same way. If this could not be achieved, it was said that “injustice” would result.[49] The Respondent contended for its part that this would also be “absurd”.[50] 22. The only conceivable legal basis for this line of thinking was the operation of s 31I of the EO which, as we have seen, provides for the deduction of s 31I benefits for what otherwise would be the severance payments payable. However, in my view, this is a long way short of providing a tenable justification for the view arrived at by the Court of Appeal and as submitted by the Respondent. It seeks to transfer a legislative intention in one context (dealing with severance payments in the EO) into another context (the making of ex gratia payments under the PWIO) and there is no provision in the PWIO similar to s 31I expressly permitting such a reduction. Nothing in the PWIO remotely suggests that such an approach is permissible. There is certainly no hint in the PWIO that there should be parity between all types of employees. The intention and purpose of this piece of legislation was simply to provide for the making of ex gratia payments where there has been a failure on the part of insolvent employers to pay what is due to employees by way of, amongst other items, severance payments. More important, the actual words of the Ordinance[51] point to an opposite result. On analysis, what the Court of Appeal essentially sought to do was to use what it saw as the purpose of the PWIO to construe the effect of the statute. This Court has on numerous occasions[52] stated that statutory construction requires a purposive approach to be adopted; in other words, the words of a statute must be construed in the light of their purpose. Context of course also plays an important part. However, it must always be borne in mind that context and purpose are not to be seen in isolation. Just as it would be wrong to construe words in a statute without regard to context and purpose, it is equally impermissible to ignore the actual words used in a statute in order to construe its effect. In China Field Ltd v Appeal Tribunal (Buildings)(No 2),[53] Lord Millett NPJ warned of the impermissibility of adopting an approach which would “distort or even ignore the plain meaning of the text and construe the statute in whatever manner achieves a result which they [the courts] consider desirable.” While the plain or natural meaning of the relevant text may not always be clearcut (hence the obvious need to bear in mind as a starting point context and purpose[54]), the actual words used cannot be ignored. The Court is after all an “interpreter not a legislator.”[55] As Fok PJ said in HKSAR v Fugro Geotechnical Services Ltd[56] :- “22. When it is said that context is the starting point, together with purpose, in statutory interpretation, that is not to say that one puts the words being construed to one side. On the contrary, since contextual and purposive construction is a tool or aid to assist a court in arriving at an interpretation that gives effect to the legislative intention, one must always have regard to the particular words used by the legislature in expressing its will. A court cannot attribute to a statutory provision a meaning which the language of the statute, understood in the light of its context and the statutory purpose, is incapable of bearing. For that reason, one must necessarily look to the statutory language to see what meaning or meanings it is capable of bearing.” I fear that the Court of Appeal, in ignoring the actual wording of the relevant provisions of the PWIO adopted a wrong approach in construing the Ordinance. Quite simply, if parity between employees was intended under the PWIO, the statute would and should have said so. D CONCLUSION 23. In my opinion, for the above reasons, this appeal should be allowed. If the correct approach had been adopted, on the facts of the present case, the sum against which the limits in s 16(2)(f)(i) of the PWIO were to be seen, was the net amount of severance payments due to the Applicant, namely $25,377.50.[57] This being less than the s 16(2)(f)(i) limits, the Applicant is entitled to an ex gratia payment in this amount under the PWIO. 24. I would accordingly make the following orders :- (1) The Applicant’s appeal is allowed. (2) As to the Applicant’s application for judicial review :- (a) The decision of the 1st Respondent dated 23 November 2012 is hereby quashed. (b) There be a declaration that the Applicant is entitled to be paid as an ex gratia payment under the PWIO the sum of $25,377.50 on account of severance payments. 25. As to costs, I would make an order nisi that both Respondents pay to the Applicant the costs of this appeal, and the costs in the courts below, such costs to be taxed if not agreed. The Applicant’s own costs should also be taxed in accordance with the Legal Aid Regulations, Cap 91. Should any party wish to have a different order for costs, written submissions should be served on the other parties and lodged with the Court within fourteen days of the handing down of this judgment, with liberty on the other parties to lodge written submission in reply within fourteen days thereafter. In the absence of such written submissions, seeking to vary the order nisi, such order will stand absolute upon the expiry of the time limited for any submissions to vary. Mr Justice Ribeiro PJ: 26. I agree with the judgment of the Chief Justice. Mr Justice Tang PJ: 27. I also agree with the judgment of the Chief Justice. Mr Justice Fok PJ: 28. I also agree with the judgment of the Chief Justice. Lord Clarke of Stone-cum-Ebony NPJ: 29. I also agree with the judgment of the Chief Justice. Chief Justice Ma: 30. The appeal is accordingly allowed. The Court makes the orders as set out in para 24 above and also makes a costs order nisi as set out in para 25 above. Mr Paul Shieh SC and Mr Ng Man Sang Alan, instructed by Kwok, Ng & Chan, assigned by the DLA, for the Appellant The 1st Respondent was not represented and did not appear Mr Ambrose Ho SC and Mr Jonathan Chang, instructed by the Department of Justice, for the 2nd Respondent [1] For the meaning of redundancy, see s 31B(2) of the Employment Ordinance Cap 57 (“the EO”). [2] See s 31G(1) of and Table A in the Seventh Schedule to the EO. [3] Section 31I of the EO. [4] Cap 380 (“the PWIO”). [5] Known as the Protection of Wages on Insolvency Fund : s 6 of the PWIO. [6] This provision is set out in full in para 11(7) below. [7] See ss 3 and 4(1)(a) of the PWIO. The Board did not appear in the courts below and did not appear before this Court although it is presumably to be bound by the decision of this Court, including any orders as to costs. [8] See ss 15 and 16 of the PWIO. Given that the Board did not appear either in the courts below or before us, the Commissioner was effectively the only Respondent in the present appeal. I will simply refer to the 2nd Respondent as the Respondent. [9] From the Court of Appeal (Cheung CJHC, Peter Cheung JA and Poon J (now Poon JA)) dismissing the appeal from the decision of Louis Chan J. The judgment of the Chief Judge in the Court of Appeal was agreed by the other two members of the court. [10] Under the PWIO, ex gratia payments are also available in respect of arrears of wages and wages in lieu of notice (see ss 15(1)(a) and (b)), but we are concerned only with severance payments (s 15(1)(c)). [11] The relevant decision of the 1st Respondent is stated in the Notice of Application for Judicial Review to be dated 23 November 2012. [12] Calculated under s 31G of the EO. [13] Paras 3(1) and (2) above. [14] $131,696.54 - $106,319.04. This applies s 31I of the EO : see para 1 above. [15] Referred to in para 2 above. [16] $50,000 + (50% of $81,696.54 (this being the excess of the severance payment due to the Applicant beyond $50,000)). [17] $90,848.27 - $106,319.04. [18] This was the evidence of the Assistant Commissioner for Labour in an affirmation filed in the present proceedings. [19] Dated 30 July 2012. The form is a standard form (Form 8 (Revision 2006)). [20] See paras 5, 9, 10 and 39 of the judgment of the Court of Appeal. [21] See paras 51 and 55 of the judgment of the Court of Appeal. [22] Mr Ambrose Ho, SC and Mr Jonathan Chang represented the Respondent. The Applicant was represented by Mr Paul Shieh, SC and Mr Alan Ng. [23] Pursuant to s 23(1)(a) of the Hong Kong Court of Final Appeal Ordinance Cap 484. [24] Para 1 above. [25] See para 1 above. [26] Cap 426. [27] Cap 485. [28] See para 10 above. [29] Section 2. [30] Section 265(1) states :- “265. Preferential payments (1) In a winding up there shall be paid in priority to all other debts - … (ca) any severance payment payable to an employee under the Employment Ordinance (Cap 57), not exceeding in respect of each employee $6,000;”. [31] Section 38(1) states :- “38. Priority of debts (1) In the distribution of the property of a bankrupt there shall be paid in priority to all other debts - … (ca) any severance payment payable to an employee under the Employment Ordinance (Cap 57), not exceeding in respect of each employee $6,000;”. [32] The situations set out in s 16(1)(a) and (b) of the PWIO. [33] See paras 3(1) and 5(1) above. [34] A methodology that has been used “for decades” : see para 6 above. [35] See para 10(2) above. [36] Para 42 of the judgment of the Court of Appeal. [37] Para 10(2) above. [38] Paras 43 to 50 of the judgment of the Court of Appeal. [39] As well as the Judge. [40] At para 49. [41] See para 5(2) above. [42] At paras 47-48 of the judgment of the Court of Appeal. [43] At paras 53-61 of the judgment of the Court of Appeal. [44] See para 15 above. [45] Para 54 of the judgment of the Court of Appeal. [46] See paras 56-60 of the judgment of the Court of Appeal. [47] This may, for example, happen where the extent of such benefits, or even their existence, was unknown to an applicant employee or the Respondent, or even where the existence may be known but such benefits had not yet accrued; in such a situation, the Respondent may want to make an early payment. Afterall, s 31I benefits are only deductable in the specific circumstances set out in that provision, and these circumstances may or may not actually have materialized at the time of the application for an ex gratia payment. [48] See para 19 above. [49] See para 32 of the judgment of the Court of Appeal. [50] See para 31 of the judgment of the Court of Appeal. [51] See the analysis in Section B above. [52] See, among other cases, HKSAR v Lam Kwong Wai & Another (2006) 9 HKCFAR 574, at para 63 (per Sir Anthony Mason NPJ). [53] (2009) 12 HKCFAR 342, at para 36; see also T v Commissioner of Police (2014) 17 HKCFAR 593, at paras 194-5 and 222 (per Fok PJ) and at para 278 (per Lord Neuberger of Abbotsbury PJ). [54] See Fully Profit (Asia) Ltd. v Secretary for Justice (2013) 16 HKCFAR 351, at para 15. [55] See T v Commissioner of Police at para 278. [56] (2014) 17 HKCFAR 755. [57] See para 5(1) above. 
Chief Justice Cheung: 1. I agree with the joint judgment of Mr Justice Ribeiro PJ and Mr Justice Gummow NPJ. Mr Justice Ribeiro PJ and Mr Justice Gummow NPJ: 2. There has reached this Court litigation commenced by the Appellant (“P”) in 2012. It concerns family disputes respecting two parcels of land located in the Sha Kok Mei Village (“the Village”), Sai Kung, in the New Territories. The Cheung family are indigenous inhabitants of the Village. The land law in the New Territories in 1898 and its subsequent development provides the background to the issues in this litigation and has been traced in Winfat Enterprises (HK) Co Ltd v A-G [1983] HKLR 211, and on appeal [1985] 1 AC 733 at 744. 3. The first dispute relates to the parcel (“the Disputed Land”) being Lot 1101 in Demarcation District 221. The principal issues in this Court concern equitable doctrine and remedies, in particular the existence and incidents of proprietary estoppel in favour of the Third Respondent (“D3”). 4. The second dispute concerns Lot 774 on which a house is erected (“House 774”). House 774 is about two minutes walking distance from the Disputed Land. The First and Second Respondents (“D1” and “D2”) together hold a 1/3 undivided share in the land. P was born in 1960 (adopted in 1961). She grew up in House 774 and lives there with her daughter and granddaughter on the ground floor. She derives rental income from the other two floors. The trial judge ordered that P pay to D1 and D2 1/3 of that rental income from 1997 (the year of the death of P’s father) and 1/3 of the mesne profits from 1997 in respect of the ground floor. Question 1 5. It is convenient to deal first with the Disputed Land and to begin with the family history which lies behind the litigation. 6. Cheung Tak Ming died in 1934. He was owner of the Disputed Land, along with other land in the Village. His three sons, Wan, Kau and Fuk, became registered co-owners of the Disputed Land. The eldest, Wan, died in 1999. D1 and D2 are respectively his son and daughter. They together are registered owners as to 1/3 of the Disputed Land, in succession to their father. D1 is the father of D3. D3 was born in 1971 and is the sole great grandson of Cheung Tak Ming. This was a matter of much significance from the traditional viewpoint of Wan, Kau and Fuk. 7. The second son, Kau, died in 1997. P is his adopted daughter, executrix of his will under a grant made in 1998 and beneficiary of his estate with her daughters. 8. The third son, Fuk, had a physical disability, never married and suicided on a date between September 1991 and May 1992. He died intestate and P is administratrix of his estate under a grant made as late as 2006. By operation of s 4(8) of the Intestates’ Estates Ordinance, Cap 73, Fuk’s assets passed to his brothers Wan and Kau who survived him. 9. In this litigation P sought, among other relief, an order for sale of the Disputed Land under the Partition Ordinance, Cap 352. After a 16 day trial Wilson Chan J dismissed her claims. He found P not to be a credible or reliable witness, whereas D3 was a credible witness whose evidence was consistent with the objective circumstances. 10. On the counterclaim by D3, his Lordship declared that D3 is the sole beneficial owner of the Disputed Land and P is a constructive trustee holding 2/3 interest therein for D3. 11. The inconclusive outcome of the appeal by P to the Court of Appeal, [2020] HKCA 148, is described in the Reasons for the Determination of the Appeal Committee (Ribeiro, Fok, Cheung PJJ) given on 18 December 2020 ([2020] HKCFA 41). The Court of Appeal (Lam VP, Cheung and Au JJA) set aside the judgment of the trial judge and remitted certain issues for determination in accordance with its legal analysis, but not on a retrial. Leave was granted by the Appeal Committee on several questions of law. Question 1 concerns proprietary estoppel and the position of D3 with respect to the Disputed Land. 12. Before turning to the text of that question it is appropriate to consider the findings at trial respecting what was identified as the “Common Understanding” and the reliance of D3 upon it. 13. The Disputed Land is situated in front of Lot 1099 where the family of D3 lived since 1974 and he grew up. The evidence of D3, accepted at trial was that since the 1970’s there had been a “Common Understanding” among the elders of the family, in particular Wan, Kau and Fuk, that D3 could use and own the Disputed Land and build a house there when he became an adult at the age of 18. The Common Understanding did not include an assurance that by way of testamentary disposition the three brothers would bequeath the Disputed Land to D3. 14. D3 is a professional construction contractor and builder. Knowing about the Common Understanding, when D3 was a teenager in the 1980’s he had started carrying out works to build stone walls to surround the Disputed Land. In 1991/1992 D3 applied substantial physical effort to heighten the stone walls and install underground electricity cables connected to the lights at the entrance gate. He also installed a new gate to prevent strangers entering the Disputed Land and maintained the trees, plants and storage house there. Significantly, in March 1997, Wan and Kau, the surviving brothers, signed a consent statement to facilitate the intended application by D3 to the District Lands Office of Sai Kung for permission to build a house on the Disputed Land. In this Court, D3 submits that by 1999 when the last of the three brothers died he had already acted upon the Common Understanding and sustained significant detriment. 15. In about 2002, D3 erected a one storey building, referred to in the pleadings as “Structure B” and in 2003 a second building (“Structure A”). In September 2003, with his wife and infant son, D3 commenced living on these premises and in the following years D3 made further improvements to the Disputed Land. He estimated that from the early 1990s he spent at least HK$2 million on construction and improvement works to the Disputed Land and did so in reliance on the Common Understanding. 16. The trial judge accepted that the case for proprietary estoppel had been established by D3. 17. To displace that outcome, P submits that (i) the activities of D3 from 1992 to 1999 do not constitute the requisite “substantial” detrimental reliance (ii) no “equity” had arisen in favour of D3 at the time of the death of Wan, the last of the three brothers, in 1999 and (iii) the activity of D3 after 1999 cannot supply the necessary detrimental reliance because this must occur before the death of the last of the authors of the Common Understanding. 18. D3 responds that, contrary to (i), the activities of D3 from 1992 to the death of Wan in 1999 did constitute sufficient detrimental reliance, and this Court should accept that submission and uphold D3’s case on proprietary estoppel giving rise against P to a constructive trust over the Disputed Land as the measure of relief. 19. Further, and in the alternative, D3 submits that the determinative question is whether D3 had suffered sufficient detriment when P, as personal representative of the estates of Kau and Fuk, sought in 2012 to resile from the Common Intention. 20. Before considering those submissions, several observations of a general nature are appropriate. The first concerns the approach of equity in such property disputes as those concerning undue influence and, here, proprietary estoppel. In Muschinski v Dodds (1985) 160 CLR 583 at 616 Deane J emphasised that, while “general notions of fairness and justice” inform “the content and application of equity”, relief by way of constructive trust should not reflect “the formless void of individual moral opinion”. Nevertheless,in Jenyns v Public Curator (Q) (1953) 90 CLR 113 at 119, Dixon CJ, McTiernan, Kitto JJ emphasised the “comprehensive view” taken by a court of equity of connected circumstances which influence determination upon “the real justice of the case”. Recently, Lady Arden has emphasised that equity operates by “principles” rather than “rules”, to produce a “fair result” (Manchester Ship Canal Co Ltd v Vauxhall Motors Ltd [2020] AC 1161 at [63]-[64]). 21. In the current, 21st, edition of Hanbury and Martin “Modern Equity” (Glister and Lee eds.) §30-022, it is said that the doctrine of proprietary estoppel has developed rapidly in the past half century. That may be so, as illustrated by the numerous cases decided in Hong Kong, the United Kingdom and Australia to which the parties to this appeal referred the Court. 22. But the foundation decision of Dillwyn v Llewelyn[1]was decided in 1862. There, a father, having made an incomplete gift of land to his son for the express purpose of the son building a house on the land, thereafter assented to and approved of the son proceeding with the building; the father’s conduct after making the incomplete gift, with the expenditure by the son thereby induced, supported the Court of Chancery obliging the father’s executors to convey the land to the son. 23. Put shortly, the requirements of equity to recognise and give relief based on proprietary estoppel are (a) a representation or assurance made to the claimant (b) reliance thereon by the claimant and (c) detriment to the claimant in the consequence of that reliance.[2] 24. More specifically, in Gillett v Holt [2001] Ch 210 at 225, 232, Robert Walker LJ (as Lord Walker then was) emphasised that (i) “the doctrine of proprietary estoppel cannot be treated as subdivided into three or four watertight compartments”, (ii) “the quality of the relevant assurances may influence the issue of reliance”, (iii) “reliance and detriment are often intertwined”, (iv) “the fundamental principle that equity is concerned to prevent unconscionable conduct permeates all elements of the [proprietary estoppel] doctrine” and (v) detriment is “not a narrow or technical concept”; it need not consist of expenditure of money or other quantifiable financial detriment and may consist of manual labour. His Lordship added “In the end the Court must look at the matter in the round”; there is “a broad inquiry as to whether repudiation of an assurance is or is not unconscionable in all the circumstances.” 25. With respect to (v) in Clarke v Meadus [2010] EWHC 3117 (Ch) at [86] Warren J emphasised that “where promises are made over a period of years, it is necessary to stand back and look at the claim in the round.” 26. With respect to (iii), and to later statements by Robert Walker LJ in Jennings v Rice [2002] EWCA Civ 159 at [50], Lady Arden (then Arden LJ) in Suggitt v Sugitt [2012] EWCA Civ 1140 at [44] observed that there need not be “a relationship of proportionality between the level of detriment and the relief awarded”; the question on an appeal is whether the relief granted was “out of all proportion to the detriment which the claimant has suffered?” 27. It may be confusing to use the term “cause of action” in this context. Rather there are two steps involved. First, once the elements of the proprietary estoppel are established an equity arises; secondly the court then must decide the most appropriate form of relief, and may have regard to the circumstances as they then exist.[3] 28. The facts of the present case give rise to another temporal question. When are the representations spent so that subsequent reliance upon them of itself cannot complete the accrual of an equity in the claimant? This issue is presented by Question 1. 29. Question 1 on which the Appeal Committee granted leave asks, with respect to the claim by D3 based on proprietary estoppel and the building of the two structures in 2002 and 2003, “Whether there can be proprietary estoppel arising from an oral promise relating to [an] interest in land binding on the estate of the promisor in the absence of any or any reasonable detrimental reliance by the promisee prior to the death of the promisor; and in such case whether detrimental reliance of the promisee only after the death of the promisor can give rise to proprietary estoppel against the estate of the deceased promisor overriding the interest of the successors or beneficiaries to the estate?” 30. In Webster v Ashcroft [2012] 1 WLR 1309 at [27], it was said “Normally, the promisee would not bring proceedings unless and until the [promisor’s] personal representative refused to give effect to the equity. In Gillett v Holt [2001] Ch 210, the promisor made inconsistent dispositions in his own lifetime, and earlier proceedings were justified.” In both instances the right in equity to relief against the promisor had accrued. Question 1 postulates occurrence of the necessary reliance by the promisee being present only after the death of the promisor. 31. P correctly submits that where there is not the necessary reliance by the promisee prior to the death of the promisor, the promise or assurance by the promisor must be taken to have lapsed. The executor or administrator of the promisor otherwise, if it were sufficient that the detriment occurred after the death of the promisor, would be giving precedence to an interest which only accrued after the interests of the testamentary beneficiaries or of the next of kin (under the Intestates’ Estates Ordinance, Cap 73) had taken effect. 32. Two further points should be made. The first is that where the requirements for the equity of the promisee to accrue have been satisfied before the death of the promisor, the state of affairs subsequent to the “cut-off” point may be relevant at the stage where the Court is considering the appropriate remedy to be awarded. 33. The second point concerns the situation where, as in the present case, the promise or assurance was by co-owners of the property in question. Reference has been made above in [14] to the significant steps by Wan and Kau in March 1997 supporting the intended application by D3 to build a house on the Disputed Land. This was not after the “cut-off” point because it postdated the death of Fuk. Where the subject land is in co-ownership the “cut-off” date is that of the last survivor. As Cheung JA put it in the Court of Appeal at 6.35, this is the last chance to go back on the promise or representation. 34. Question 1 should be answered accordingly. 35. That, however, opens up a further issue which was debated on the appeal. It is whether there had been sufficient activity by D3 with respect to the Disputed Land before the death of Wan in 1999. 36. It is here that the principles dealt with above in [23]–[27] come into play. 37. The trial judge emphasised that the litigation concerned a traditional Chinese family living in the New Territories. The steps taken by D3 described in [14] were those of a young man preparing for an adult life in occupation of the improved Disputed Land, he doing so with the support and encouragement of his elders. Those steps were in the setting of this case sufficient reliance by D3 upon the Common Understanding. 38. The question is whether the remedy for the equity should be the constructive trust by P of her 2/3 interests in the Disputed Land for D3, or some lesser remedy such as a sum of money. The former remedy is appropriate. It would “enable the claimant to have the benefit of the equitable right which he is held to have” (Luo Xing Juan v Estate of Hui Shui See (2009) 12 HKCFAR 1 at [68]. 39. As noted at [32], at this stage the Court may have regard to the state of affairs as it developed after the death of Wan in 1999. This is not a case such as Giumelli v Giumelli (1999) 196 CLR 101, where other parties have made contributions to the land in question so that relief by way of constructive trust would inflict an injustice. The trial judge found that after 2003 and the birth of D3’s son and when P had become friendly with D3, she would occasionally visit his house on the Disputed Land and stay for dinner with his family. The infant son called P “aunt”. P did not assert the titles of Kau or Fuk until relations with the defendants soured and she embarked on this litigation, joining D3 as a party only in 2013. D3 believes this change in attitude by P occurred after she had become interested in the sale of the Disputed Land to a developer. For his part, it appears that D3 did not invest in any other properties and his emotional and sentimental attachment to the Disputed Land, his family home for many years, would make his dispossession unconscionable in the sense used by Lord Walker in Gillett v Holt, discussed above [24].[4] 40. The trial judge held that the facts gave rise in favour of D3 both to a proprietary estoppel and a common intention constructive trust (at [89]). In Stack v Dowden [2007] 2 AC 432 at [37] Lord Walker warned against the assimilation of the common intention constructive trust and proprietary estoppel. The death of the legal owner also supplies the “cut-off” date for any common intention constructive trust, as Cheung JA held in the Court of Appeal at 6.38. But in other respects the common intention as constructive trust and proprietary estoppel are distinct.[5] 41. The matter need not be further pursued here. The arguments in this Court focused upon proprietary estoppel and D3 succeeds on that ground. There remains the second dispute and the issue presented by Question 2. Question 2 42. Question 2 is in these terms: “Whether a co-owner in sole occupation of land, in cases other than partition or ouster and in the absence of agreement, should be ordered to account to the other co-owners for occupation rent. Alternatively, whether the proposition of law in Re Pavlou (A Bankrupt) [1993] 1 WLR 1046 at 1050D that ‘a court of equity will order an inquiry and payment of occupation rent even if there is no ouster when it is necessary to do so to do equity between the parties’ should be confined to partition or analogous proceedings?” 43. It arises in relation to the counterclaim mounted by D1 and D2 against P for an account and payment of 1/3 of the rental income from the first and second floors of House 774 and of 1/3 of mesne profits claimed in respect of the ground floor in which P resided. The trial judge made the orders sought.[6] The Court of Appeal held however that the issue regarding the expenses of rebuilding House 774 should be remitted to the trial judge to be taken into account in applying a specified approach to the law (referred to below) but that otherwise, his Lordship’s findings should be left undisturbed.[7] 44. The following facts are not in dispute, namely that: (a) As noted above at §4, House 774 is erected on Lot No 774 in DD 221. We will refer to the property as a whole as “House 774”. (b) P had grown up in House 774 which was occupied by her grandparents and parents; (c) the building had collapsed in 1984 after a typhoon and was rebuilt in 1985, initially as a house with two floors, using funds provided by P’s family, after which she took up occupation in it; (d) in 1997, P let out the first floor and received rent from the letting which she retained; (e) in 2008, P sold her interests in other plots of land and used the proceeds to convert House 774 into a three-storey building; and (f) in 2009, she let out the second floor and received the rent. 45. It is common ground that D1 and D2 together hold an undivided 1/3 interest in House 774 as tenants in common with P who, as personal representative succeeding to the estates of Fuk and Kau, holds an undivided 2/3 interest therein. The claim and decision at first instance 46. It is evident from the pleadings that the case as framed at first instance suffered from certain misconceptions held on either side. 47. Although D1 and D2 admit that they are co-owners of the property as tenants in common with P, their counterclaim is posited simply on the basis that “[as] the 1/3 co-owner” of House 774 they are entitled to a 1/3 share of the rent P received and “1/3 of the mesne profits in respect of the ground floor” where P lived, reflecting their “proprietary interests and rights”.[8] 48. So pleaded, it was not a viable claim since it is well-settled that a co-owner who is not in occupation has no claim for rent against the occupying co-owner simply by virtue of the latter’s sole occupation of the co-owned property.[9] This is because unity of possession among the co-owners is central to a tenancy in common. As Meagher JA explained in a valuable summary of the rights of co-owners inter se in Forgeard v Shanahan:[10] “1. Since both joint tenants and tenants in common have joint possession of the land in which they have the estate, it was a settled rule of law that the possession of any one of them was the possession of the other of them, so as (for example) to prevent the statutes of limitation from affecting them; nor did the bare receipt of all the rents and profits by one operate as an ouster of the other...[11] 2. It follows that, where one co-owner is in occupation and the other not, but there has been no actual ouster or exclusion by the former of the latter, the law treats the latter simply as someone who has chosen not to exercise his legal right to occupy the land.” 49. As Lord Cottenham LC explained, the effect of imposing an obligation to pay rent in such circumstances “would be, that one tenant in common, by keeping out of the actual occupation of the premises, might convert the other into his bailiff; in other words, prevent the other from occupying them, except upon the terms of paying him rent.”[12] That was in principle unacceptable. 50. As indicated in the passage from Meagher JA cited above, a co-owner who has been ousted, that is, excluded from the co-owned property, does have an action in trespass and can recover mesne profits or an occupation rent from the co-owner in occupation.[13] As Lord Hatherley LC stated: “Now, as regards the question of trespass, it appears to be perfectly settled ... that unless there be an actual ouster of one tenant in common by another, trespass will not lie by the one against the other so far as the land is concerned.”[14] 51. Other bases exist for founding a claim by a tenant in common for occupation rent or for an account of rent against a fellow tenant in common in sole occupation. These are considered later in this judgment and include cases where partition or analogous proceedings are brought with a view to ending the co-ownership; where there is an agreement giving rise to a duty to account and where a statutory action may be available. But for immediate purposes, ouster is the relevant basis for consideration. 52. This is so because of the nature of the defence mounted by P against the counterclaim. As noted by the Judge,[15] it was a purported defence based on adverse possession. P was at pains to argue that the counterclaim, viewed as an action for trespass, was barred by her adverse possession of House 774. This was misconceived since, as tenants in common, they had joint possession of the property and no question of adverse possession could arise. Indeed, such joint possession was pleaded by P.[16] 53. In furtherance of her purported defence of adverse possession P’s pleadings include the following averments: “At all material times, all of the Defendants in this action were not allowed to enter the house on said Lot 774 unless they were invited by the Plaintiff or her father, namely Kau.” (Defence to Counterclaim §27(f)) “From the aforesaid [...] time in 1997, no one can enter House 774 without the Plaintiff’s permission.” (Ibid at§28(d)) “At all material times since about 1997, the Plaintiff had the [sic] every intention to possess House 774 adverse to the interests of other registered owners of the same.” (Ibid at §28(e)) “By reason of the exclusive possession, occupation and control of House 774 by the Plaintiff with the requisite intention to exclude all others at large, the rights, interest and title of the Ds in respect of [House 774], have been extinguished and the Ds are barred from recovering the same, under the Limitation Ordinance (Chapter 347).” (Ibid at §28(f)) 54. P’s witness statements supported her pleadings: “Based on my memory, all Defendants were not allowed to enter the Replacement House unless they were invited by My Father or me. We had always treated the Replacement House as our solely owned property and enjoyed exclusive possession, occupation and control of the same.”[17] “After My Father passed away in 1997, I had always enjoyed exclusive possession, occupation and control of the Replacement House. No one has ever made any queries of my right to do so. Since 1997, no one can enter the Replacement House without my permission. I have always had and continue to maintain absolute exclusive use of the Replacement House and land on which it sits on [sic].”[18] 55. D1 and D2’s evidence also lent support to their having been excluded. They gave evidence of an incident in 1997 when they approached P asking for a share of the rent she had started collecting from letting out the first floor of House 774. P reacted with great hostility and threw a hard object through the window of D1 and D2’s house, leading to P being prosecuted (and apparently convicted[19]). 56. One would have thought that those pleadings and the evidence filed provided a basis, self-inflicted by P, for supporting D1 and D2’s counterclaim, enabling them to contend that they had been ousted by P from House 774 and therefore that notwithstanding co-ownership, they had a good claim against P for trespass and mesne profits.[20] 57. However, the point has not been taken at any stage. Instead, D1 and D2 were intent on showing that there was no ouster, believing that this advanced their cause against P rather than barring its success. Although his Lordship had held that adverse possession was a bad defence in law, he proceeded to make findings aimed apparently at negating adverse possession by confirming that there had been no ouster, believing that such findings supported D1 and D2’s counterclaim. Thus, he held: (a) “The plaintiff cannot establish ouster by simply showing that she has pocketed all the rental income of House 774.”[21] (b) “... in the present case, the 1st and 2nd defendants have done more than enough to negative ouster.”[22] The Court of Appeal 58. In the Court of Appeal, the “no ouster” finding was not challenged and the Court proceeded on the basis that there had been no ouster of D1 and D2 from the property.[23] Ms Audrey Eu SC[24] (who did not appear at the trial) seized on that finding and argued: “... that the Judge erred in ordering an account and enquiry as to one-third of the rental income arising from the 1st and 2nd floors of House 774 and one-third of the mesne profits in respect of the ground floor of House 774 as from 1997 onwards and ordering the plaintiff to pay them, in that it is well settled that co-owners are not agents or fiduciaries for one another and a co-owner out of occupation has remedies in law in two situations: (1) damages for occupation rent for ouster, or (b) an account if the other co-owner has rendered himself liable as agent or bailiff (Chen Yu Tsui v Tong Kui Kwong [2006] 1 HKC 23 at paragraphs 87-105, cited with approval by the Court of Appeal in Tang Tak Sum v Tang Kai Fong [2015] 1 HKLRD 286). None of the two situations arise on the facts of this case and the Judge expressly found in paragraphs 112 to 113 of the judgment that there was no ouster.”[25] 59. In response, Mr William Wong SC,[26] submitted: “... that it is now recognised that a co-owner may claim occupation rent from another co-owner where it is ‘fair’, ‘just’ or ‘in order to do equity between the parties’: Gray and Gray, Elements of Land Law (5th ed) at paragraph 7.4.45.”[27] 60. He cited in particular In re Pavlou (A Bankrupt),[28] in which Millett J (as Lord Millett then was) stated: “[A] court of equity will order an inquiry and payment of occupation rent, not only in the case where the co-owner in occupation has ousted the other, but in any other case in which it is necessary in order to do equity between the parties that an occupation rent should be paid.” 61. Mr Wong SC also cited Re Byford (deceased),[29] where Collins J (as Lord Collins of Mapesbury then was), stated: “[W]hat the court is endeavouring to do is broad justice or equity as between co-owners.” Counsel also referred to French v Barcham,[30] for the proposition that “whether occupational rent would be charged on the facts of each case is ‘ultimately a matter of discretion’”. 62. Thus, Ms Eu SC was advancing the traditionally understood position (as indicated above) that one tenant in common could not, in the absence of ouster or another recognised ground, bring a claim for rent against another tenant in common simply based on the latter’s sole occupation of the property in question. 63. While their Lordships did not do so explicitly, it would appear that they accepted that absent ouster, the counterclaim of D1 and D2 could not be sustained on traditional grounds. However, they considered that the counterclaim could be viable if what was referred to as “the modern approach” were adopted. Thus, in the majority, Lam VP (Au JA agreeing) held that: “As regards the order for account on the counterclaim, as the modern approach has not been addressed by the judge and he found that there was no ouster, the claim should be remitted to the judge.”[31] By the “modern approach”, his Lordship had in mind the passages cited above from In re Pavlou and Re Byford: “In respect of the finding that there was no ouster, it should not be a bar to a claim for equitable account in light of the authorities cited ... below. ... That should not bar this Court from re-examining the law in light of the modern position as reflected in the English authorities of Re Pavlou (A Bankrupt) [1993] 1 WLR 1046 and Re Byford (deceased) [2004] 1 P & CR 12 159, which has been followed by Ng J in Leung Siu Man v Leung Yiu Tai [2019] HKCFI 1971.”[32] 64. His Lordship concluded: “I see no reason why the modern approach should not be followed in Hong Kong. In other words, a court of equity will order an inquiry and payment of occupation rent even if there is no ouster when it is necessary to do so to do equity between the parties.”[33] 65. The majority however, held that to enable this “modern approach” to be considered it was necessary for there to be a remitter to the trial judge to enable findings to be made as to the expenses of rebuilding House 774 to be taken into account in doing equity between the parties.[34] 66. Cheung JA held that the Judge had failed to articulate reasons why he had ordered an account and considered that the whole matter should be remitted, with a wide range of inquiries to be pursued.[35] On the scope of the remitter, he was in the minority. 67. The Appeal Committee granted leave to appeal on Question 2 which calls for examination of the Court of Appeal’s holding that a co-owner in sole occupation of land, in cases other than partition or ouster and in the absence of agreement, may be ordered to account to the other co-owners for occupation rent and rental receipts adopting a broad, discretionary “modern approach” regarded as an independent basis for such a claim. The established principles Ouster and agreement as basis for action 68. Initially, a co-owner out of occupation had very limited remedies against his co-owner in occupation in relation to the land. As Meagher JA put it: “... a co-owner not in occupation was normally virtually without remedy. He could not sue in trespass unless there was an ouster. ... A co-owner out of occupation could not even recover his share of rents and profits if the co-owner in occupation appropriated them to himself; no action of account lay either at law or in equity.”[36] “Apart from statute, a co-owner out of occupation had remedies at law in two situations, and no more. If he had been ousted, he could bring ejectment and mesne profits ... If, on the other hand, his co-owner were in occupation by agreement that co-owner became an agent or bailiff and rendered himself liable in a common law action of account. In either case (ie of ouster or occupation by agreement) he would be liable for rents actually received and possibly also for an occupation fee.”[37] His Honour added: “Apart from Statute, in equity the plight of a co-owner not in occupation was little better. There did not seem to be any action which would render a co-owner in occupation liable to refund any rents received, much less liable for an occupation fee.”[38] Statutory action 69. The position was considered unsatisfactory and in England, led to the enactment in 1705 of the Statute of Anne[39] which created a statutory right of action against a tenant in common for rents actually received and which constitute “more than ... his just share or proportion”. While this statutory claim was once available in Hong Kong, it ceased to have any application as from 7 January 1966.[40] It therefore needs no further discussion here save to note that no such statutory action presently exists in this jurisdiction. Constructive ouster 70. There have, however, been incremental improvements to the position of co-owners out of occupation. One development involves widening the notion of “ouster” to include constructive exclusion of one co-owner especially in cases involving breakdown of a matrimonial or co-habiting relationship, and especially where domestic violence or the threat thereof is involved.[41] Equity’s intervention in partition and analogous proceedings 71. Another development, of particular importance in the present case, occurred in the context of equitable accounting. While equity does not provide a basis for an independent, proactive claim for an account of rent against a co-owner in occupation, it does intervene where a co-owner institutes a partition or analogous action and may impose a requirement for adjustments to be made in the process of equitable accounting in order to do equity between the parties. 72. An illustration is provided by Leigh v Dickeson,[42] where one tenant in common brought an action to recover rent from another and faced a counterclaim for the cost of repairs. The Court held that the claim failed but Cotton LJ explained the position which would have arisen if it had been a partition action: “... no remedy exists for money expended in repairs by one tenant in common, so long as the property is enjoyed in common; but in a suit for a partition it is usual to have an inquiry as to those expenses of which nothing could be recovered so long as the parties enjoyed their property in common; when it is desired to put an end to that state of things, it is then necessary to consider what has been expended in improvements or repairs: the property held in common has been increased in value by the improvements and repairs; and whether the property is divided or sold by the decree of the Court, one party cannot take the increase in value, without making an allowance for what has been expended in order to obtain that increased value; in fact, the execution of the repairs and improvements is adopted and sanctioned by accepting the increased value. There is, therefore, a mode by which money expended by one tenant in common for repairs can be recovered, but the procedure is confined to suits for partition. Tenancy in common is an inconvenient kind of tenure; but if tenants in common disagree, there is always a remedy by a suit for a partition, and in this case it is the only remedy.” 73. In Brickwood v Young,[43] Griffith CJ elaborated upon Leigh v Dickeson as follows: “The principle appears to be that the making of permanent improvements by one tenant in common in sole occupation gives rise to an equity attaching to the land, analogous to an equitable charge created by the owners for the time being, but enforceable only in the event of partition or a distribution of the value of the land amongst the tenants in common. There can be no reason why such a charge should not run with the land in favour of purchasers from the person originally entitled to it.” 74. And in Squire v Rogers,[44] Deane J further explained: “As a general rule, capital expenditure upon permanent improvements to land by one joint owner without the authority of his co-owner creates a passive equity which attaches to the land. The joint owner making the improvements is not entitled to bring proceedings for contribution against his co-owner. In circumstances where his co-owner (or a successor in title of his co-owner other than a purchaser for value without notice) would otherwise unfairly benefit under an order in equity (including partition or sale of the property), he is entitled to an allowance for his expenditure on such improvements to the extent to which they result in the present enhancement of the value (or the price on sale) of the land ...” 75. Of particular relevance to Question 2 are cases where, as part of the equitable accounting exercise in a partition action or analogous proceedings, an occupying owner who claims an allowance for his expenditure may be permitted such an allowance only if he consents to be charged an occupation rent. These represent cases where an occupation rent is claimable by the owner out of occupation without proof of ouster. 76. Thus, in Teasdale v Sanderson,[45] a suit was instituted by one of several tenants in common against the others, for partition of the estate and for an account of the rents and profits received by the defendants, who had been in exclusive possession for about fourteen years. One of the defendants was in personal occupation of part of the premises, and sought to claim credit for the sum of £60 spent on necessary repairs and improvements of the property. Sir John Romilly MR held “that these accounts must be reciprocal, and, unless the Defendant is charged with an occupation rent, he is not entitled to any account of substantial repairs and lasting improvements on any part of the property”. 77. In McCormick v McCormick,[46] Salmond J also referred to such reciprocity, pointing out that: “ ... there is no general right even in a partition suit to charge an occupying owner with an occupation rent. I think that the obligations of co-owners to account to each other are the same in equity as at law, and are the same in a partition suit as in other proceedings, save only that in a partition suit, if an occupying owner claims an allowance for his expenditure, he can obtain it only if he consents to be charged with an occupation rent.” 78. And in Forgeard v Shanahan, Meagher JA explained the position in Equity as follows: “As far as equity is concerned, an occupation fee will be exacted in at least two circumstances: first, in a partition suit (or related litigation): if there has been an exclusion, the tenant in occupation will be charged with an occupation fee: this is an example of equity following the law; and secondly, if the owner in occupation claims an allowance in respect of improvements effected by him, equity will permit such an allowance only on terms that he is accountable for an occupation fee – this is an example of he who comes to equity having to do equity.”[47] 79. Cases in this category should be borne in mind when examining the “modern approach” favoured by the Court of Appeal. Is there an independent “modern approach”? 80. In D1 and D2’s Written Case,[48] it is contended that: “The principles in Re Pavlou clearly establish a flexible approach in equity to order occupation rent without confinement to ouster, partition or anything in analogy.” “As a matter of law, it is now recognised that a co-owner may claim occupation rent from another co-owner where it is ‘fair’, ‘just’ or ‘in order to do equity between the parties’: Gray and Gray, Elements of Land Law (5th ed) at §7.4.45, without restrictions to ‘partition’, ‘ouster’, ‘agreement’.” 81. In support, the following citations and propositions are put forward: “[A] court of equity will order an inquiry and payment of occupation rent, not only in the case where the co-owner in occupation has ousted the other, but in any other case in which it is necessary in order to do equity between the parties that an occupation rent should be paid”: In re Pavlou (A Bankrupt) [1993] 1 WLR 1046, per Millett J (as he then was), at 1050D. In determining the question of occupation rent, “[w]hat the court is endeavouring to do is [broad] justice or equity as between co-owners”, and the Court would take into account of the use or benefit derived by the co-owner in occupation: Re Byford (deceased) [2004] 1 P & CR 12 159, per Collins J (as he then was), at 168. Whether occupational rent would be charged on the facts of each case is “ultimately a matter of discretion”: French v. Barcham and another [2009] 1 WLR 1124, per Blackburne J at §43. 82. We will say at once that we do not accept that the authorities justify the proposition that a “modern approach” has been introduced so that under the guise of “flexible equitable relief”, the principles incrementally developed over the centuries, each subject to principled rationales for dealing with issues between co-owners encountered in practice, have been overridden and replaced by an amorphous broad discretion. 83. As always, context is all important. The line of cases of which In re Pavlou forms a part are all illustrations of the courts doing justice in making adjustments and imposing set-offs in the process of equitable accounting. It is in that context that the dicta cited by D1 and D2 must be understood. They do not signal any departure from the principles explained above and in particular, do not eliminate the threshold requirements for equity’s intervention. 84. We may begin by reference to Bernard v Josephs,[49] where the principles on which the court acts in making adjustments to the shares in which tenants in common are entitled to the proceeds of the sale of property are elucidated. Lord Denning MR explained: “After ascertaining the shares, the next problem arises when it is to be turned into money. Usually one of the parties stays in the house, paying the mortgage instalments and the rates and other outgoings. The house also increases in value greatly owing to inflation. None of that alters the shares of the parties in the house itself. But it does mean that when the house is sold — or the one buys the other out — there have to be many adjustments made. The value of the house itself is taken at the value at the time of sale or buying out. There must be deducted from it all the money needed to redeem the mortgage. Then the one in possession must be given credit for paying the other’s share of the mortgage instalments and be debited with an occupation rent for using the other’s share of the house. Other adjustments may be needed for other outgoings. Then the net amount must be divided according to the shares.”[50] And Griffiths LJ stated: “When the proceeds of sale are realised there will have to be equitable accounting between the parties before the money is distributed. If the woman has left, she is entitled to receive an occupation rent, but if the man has kept up all the mortgage payments, he is entitled to credit for her share of the payments; if he has spent money on recent redecoration which results in a much better sale price, he should have credit for that, not as an altered share, but by repayment of the whole or a part of the money he has spent. These are but examples of the way in which the balance is to be struck.”[51] In such cases, occupation rent may be an element in the adjustment where entitlement does not depend on ouster or agreement. 85. That decision was applied in In re Gorman (A Bankrupt),[52] an authority later cited by Millett J in In re Pavlou. In re Gorman involved a husband who had become bankrupt after being divorced from his wife. His trustee in bankruptcy sought an order for sale of the former matrimonial home which was in the former couple’s joint names. The issue, as Vinelott J noted, was “whether the trustee is entitled to an immediate order for sale and whether, when sold, there should be any equitable accounting between Mrs. Gorman and the trustee to reflect the fact that the whole of the mortgage instalments have been paid by her.”[53] In particular, the question was “whether Mrs. Gorman is entitled to be credited with one half the mortgage instalments she has paid, and whether she is liable to be debited with one half of an occupation rent.”[54] Applying Bernard v Josephs, Vinelott J ordered: “... an inquiry into the amount of the proper occupation rent, between the making of the receiving order and the sale of the property, and an account of all the mortgage instalments paid by Mrs. Gorman before and after the receiving order was made. One half of the occupation rent will have to be debited and one half of the mortgage instalments credited to Mrs. Gorman's share.”[55] 86. The situation in In re Pavlou[56]was similar. A wife petitioned for divorce against her husband for desertion (although she was apparently willing and anxious to have him back). He was made bankrupt and the trustee in bankruptcy applied for an order for sale of the property jointly owned by the couple as tenants in common in equal shares.[57] It was agreed that there would be an order for sale and that “there must be an equitable accounting between the parties.”[58] Millett J was throughout concerned with that equitable accounting process. His Lordship took the guiding principle from Leigh v Dickeson,[59] explaining that: “On a partition suit or an order for sale adjustments could be made between the co-owners, the guiding principle being that neither party could take the benefit of an increase in the value of the property without making an allowance for what had been expended by the other in order to obtain it.”[60] 87. It was as part of the accounting exercise that Millett J posed these questions: “More difficult questions arise from the wife’s continued occupation of the property. Is she chargeable with an occupation rent in respect of any period prior to the date of the bankruptcy order? And is she entitled to credit for the interest element in the mortgage repayments which she has made since the husband left her?”[61] 88. He proceeded to review cases where ouster had been the basis for claiming occupation rent and where domestic violence cases were treated as constructively involving ouster,[62] and then made the statement relied on by the Court of Appeal and the respondents for the existence of a “modern approach”, namely: “[A] court of equity will order an inquiry and payment of occupation rent, not only in the case where the co-owner in occupation has ousted the other, but in any other case in which it is necessary in order to do equity between the parties that an occupation rent should be paid.” 89. Millett J was thus concerned with possibly ordering an inquiry and payment of occupation rent as part of an equitable accounting exercise where an order for sale had been made. It is entirely consistent with previous authority reviewed above to hold that in that context, occupation rent may be ordered not only where there has been ouster but where it is necessary to do equity between the parties in the accounting process. His Lordship ended his judgment by stating that the wife was prima facie entitled to credit for the interest element in the mortgage payments made since her husband left the property. 90. The circumstances in Re Byford (deceased),[63] were similar except in one respect, namely, that a husband who had been declared bankrupt continued to live with his wife in their house. The trustee in bankruptcy did not seek to realise the husband’s interest in it until after the husband died, when he claimed a half interest and an order for sale if the wife did not agree to pay an appropriate amount for the trustee’s interest. So this was a case where there had been no actual ouster of the husband from the property during his lifetime. Meanwhile, the wife had continued to pay mortgage interest and make repairs, etc. She did not dispute the trustee’s half interest but claimed credit for her outgoings and the trustee claimed to be entitled to set off occupation rent. 91. Collins J cited In re Gorman and In re Pavlou, and asked whether the same result as in those cases should follow, noting the husband’s continued residence for nearly 10 years after the bankruptcy. His Lordship reviewed the authorities, including most of the cases cited in this judgment, stating: “In the typical case an occupation rent has been charged where the party in occupation has actually or constructively excluded the other party from occupation.”[64] 92. However, as Collins J pointed out, the issue before him was one of set-off in equitable accounting: “The principal issue debated in this case is whether an occupation rent will be set off against mortgage interest payments made by the wife when there has been no marital breakdown and the bankrupt husband has remained in the matrimonial home.”[65] 93. His Lordship referred to Leigh v Dickeson,[66] and In re Pavlou for the “guiding principle”: “The principles of equitable accounting apply equally to beneficial tenancies in common and beneficial joint tenancies. The guiding principle is that neither party can take the benefit of an increase in the value of the property without making an allowance for what had been expended by the other in order to obtain it.”[67] 94. This was the context in which his Lordship made the statement relied on by the Court of Appeal and the respondents: “What the court is endeavouring to do is broad justice or equity as between co-owners. As Millett J. said in Re Pavlou, the fact that there has not been an ouster or forcible exclusion is not conclusive. The trustee cannot reside in the property nor can he derive any financial enjoyment from the property while the bankrupt’s spouse resides in it, and the bankrupt spouse’s creditors can derive no benefit from it until he exercises his remedies.”[68] 95. Plainly, neither of the decisions of Lord Millett and Lord Collins is authority for any sweeping new “modern approach”. Rather, they illustrate the need to consider a judgment as a whole when extracting a proposition of law for which it is authority. 96. The respondents also rely on the decision of Blackburne J in French v Barcham,[69] for saying that “Whether occupational rent would be charged on the facts of each case is ‘ultimately a matter of discretion’.” The circumstances were like those in In re Byford. A husband and a wife occupied a property owned by them as tenants in common. He was made bankrupt and the couple remained in occupation. It was only 12 years after the bankruptcy order that the trustee in bankruptcy applied to the court for the sale of the property, seeking to set off against the wife’s 50% share a notional occupation rent to reflect their continued occupation. This ought therefore to have been approached as another case of equitable accounting requiring a decision as to whether it was equitable to charge her occupation rent. This was indeed how the District Judge had dealt with the case. 97. However, after dealing with statutory provisions that do not concern us,[70] and despite his citation of the line of cases referred to above, Blackburne J focussed narrowly on the question of ouster rather than on the process of equitable accounting. He held that “[without] actual ouster of the one co-owner by the other any liability to pay an occupation rent simply did not arise”.[71] This required him to deal with the fact that there had been no actual ouster of the bankrupt husband for some 12 years. The question to be determined, as he saw it, was “what would amount to ouster (or exclusion) so as to give to the tenant in common who was out of occupation a claim for compensation against the tenant in common who remained in occupation.”[72] 98. His Lordship proceeded to hold that the concept of “ouster” should be greatly extended, adopting a notion of “unreasonableness” in formulating the following test: “The essential point, in my view, is that when on inquiry it would be unreasonable, looking at the matter practically, to expect the co-owner who is not in occupation to exercise his right as a co-owner to take occupation of the property, for example because of the nature of the property or the identity and relationship to each other of the co-owners, it would normally be fair or equitable to charge the occupying co-owner an occupation rent.”[73] By way of elaboration: “... if there is some reason why that co-owner is not in occupation and it would be unreasonable in the circumstances for him to take up occupation fairness requires the occupying co-owner to compensate the other for the fact that the one has enjoyment of the property while the other does not.”[74] 99. The application of this approach to the trustee in bankruptcy’s claim led to his Lordship treating the trustee as the co-owner out of occupation (while the husband continued to live in the house) and concluding as follows: “When a trustee in bankruptcy has been appointed of the estate of a co-owner so that that co-owner’s interest vests in the trustee, but the other co-owner remains in occupation of the property, application of the principle will ordinarily, if not invariably, result in the occupying co-owner having to account to the trustee of the beneficial interest to which the bankrupt co-owner was formally entitled for an occupation rent. This is because it is not reasonable to expect – even if it were otherwise practicable for him to do so – the trustee in bankruptcy to exercise the right of occupation attaching to the interest in the property that vested in him on his appointment as trustee of the bankrupt co-owner.”[75] 100. With respect, the failure to appreciate that the relevant line of cases deals with equitable accounting upon partition or an order for sale, permitting occupation rent to be charged often by way of set-off and without a requirement for ouster, appears to have led the analysis in French v Barcham astray. Faced with an absence of any actual ouster over nearly 12 years, Blackburne J resorted to a notional ouster on the footing that it would have been unreasonable to expect the trustee to exercise the right of occupation. He appears to have regarded such occupation to involve a personal, physical taking up of residence by the trustee (rather than say, occupation for the purpose effecting sale or of renting out the property) and therefore reached the conclusion that the trustee would almost always be entitled to charge the occupying co-owner with occupation rent since the trustee could not reasonably be expected to take up residence. That destroys the flexible equitable approach adopted in the cases by introducing a test based on a highly artificial scenario. 101. In Davis v Jackson,[76] Snowden J declined to follow French v Barcham. This was yet another case involving an application by a trustee in bankruptcy for sale of a house held in the joint names of a formerly married couple. The background facts were unusual in that: “... the parties have at all relevant times been estranged and lived apart when the house was originally acquired by the wife in her sole name, and the husband was never intended to live in the house and has not done so at any time. The property was only transferred into joint names when the house was re-mortgaged, but at no time did the husband pay any of the instalments of either mortgage or contribute to any of the outgoings.”[77] 102. His Lordship approached the case as one concerning equitable accounting.[78] He regarded Blackburne J’s “unreasonableness” test as inconsistent with earlier authorities, commenting that, contrary to the “default position” where no occupation rent is payable, under the proposed test “it would invariably be unreasonable for a trustee in bankruptcy to seek to take up occupation, [so that] Blackburne J’s approach would have the result, as a virtually immutable rule, that an occupation rent should be payable.”[79] As such, it would exclude “the possibility of the court having any regard to the position that existed prior to the bankruptcy, or to the conduct or circumstances of the non-bankrupt party”[80] in deciding whether to order payment of occupation rent. 103. As indicated in cases like Pavlou and Byford, the court adopts a broad equitable approach in the exercise of equitable accounting, requiring the person seeking equity to do equity in the circumstances of each individual case. Those authorities do not discard the established principles which determine when a viable claim for an account may arise. On the unusual facts in Davis v Jackson, Snowden J refused to charge Mrs Jackson occupation rent, holding that: “[If she] were to be charged occupation rent for the period since the vesting in the Trustee of Mr Jackson’s interest in the Property, so as to offset the substantial mortgage payments which she has continued to make after that time, the result will, to my mind, represent an unjust windfall for Mr Jackson’s creditors”.[81] Conclusion on Question 2 104. We conclude that the authorities considered above do not establish any new, free-standing “modern approach” such as that urged by the respondents and favoured by the Court of Appeal. Claims by one co-owner against a co-owner in occupation for payment of occupation rent or for an account of rent can only arise in accordance with the principles laid down in the established authorities. Unity of possession precludes such claims otherwise than in cases of ouster (including “constructive exclusion” as in domestic violence cases); or where an operative agreement renders the co-owner in occupation an agent or bailiff so as to come under a duty to account to the other. Where partition or analogous proceedings have been instituted, apart from cases of ouster, equity may recognise a defensive equity in favour of one of the co-owners regarding expenditure appropriately incurred and may, in the process of equitable accounting, require the other, viewed as a seeker of equity required to do equity, to be debited with an occupation rent to set off the expenditure incurred, thus reciprocally balancing the parties’ interests in the distribution of the realised proceeds of the co-owned property. 105. The present case has proceeded throughout (including in this Court) on the basis that there has been no ouster. None of the other established bases for claiming occupation rent or an account of rent have been alleged, let alone made out. Accordingly, the action against P for “mesne profits”, occupation rent and an account of rent must fail. There is thus no basis for ordering any remitter to ascertain P’s building expenses since there is no viable claim for rent against which such expenses might be set off. 106. We would therefore allow P’s appeal and dismiss the counterclaim of D1 and D2, setting aside the orders made by the Courts below in relation to such counterclaim. Disposal of the appeal 107. We would accordingly make the following orders, namely, that: The appeal be allowed and the order of the Court of Appeal dated 5 March 2020 be set aside and in place thereof: (1) it be declared that the Plaintiff is estopped from denying that the Third Defendant is beneficially entitled to Lot 1101 in DD221; (2) the Plaintiff cause the undivided 2/3 share she holds in Lot 1101 DD221 be transferred to the Third Defendant; (3) the parties seek to agree the terms to implement order (2) and in default of such agreement the parties within 14 days of these orders lodge their proposed order, for determination by this Court; (4) the claims by the Plaintiff against the First and Second Defendants, and against the Third Defendant be dismissed; (5) the claim by the First and Second Defendants against the Plaintiff be dismissed; and, (6) within 14 days of these orders the parties lodge their submissions as to the costs of the proceedings in the Court of First Instance, the Court of Appeal and this Court to be determined by the Court on the papers. Mr Justice Fok PJ: 108. I agree with the joint judgment of Mr Justice Ribeiro PJ and Mr Justice Gummow NPJ. Mr Justice Bokhary NPJ: 109. I agree with the joint judgment of Mr Justice Ribeiro PJ and Mr Justice Gummow NPJ. All that I add is this. On its own, each side has achieved a measure - although not necessarily an equal measure - of success in this appeal. Together they share the success of a final disposal here and now of this unhappy case without a remitter. For reasons which I can understand, the learned judges of the Court of Appeal felt unable to avoid remitting certain matters to the trial court. Our avoidance of a remitter was, as it seems to me, materially assisted by the able and responsible way in which Ms Audrey Eu SC for the Plaintiff and Mr William Wong SC for the Defendants presented their respective cases. I am grateful to them for that. So should the parties be. Finality is an important element of practical justice. Chief Justice Cheung: 110. The appeal is unanimously allowed and the order of the Court of Appeal dated 5 March 2020 is set aside. In place thereof we make the orders set out in paragraph 107 above. Ms Audrey Eu, SC, Mr Anson Wong Yu Yat and Mr Jason Kung, instructed by Chan & Chan, for the Plaintiff (Appellant) Mr William Wong, SC, Mr Alan Kwong and Ms Stephanie Wong, instructed by Tung, Ng, Tse & Lam, for the 1st to 3rd Defendants (Respondents) [1] (1862) 4 De G.F. & J517 [45 ER 1285]. [2] Thorner v Major [2009] 1 WLR 776 at [29] per Lord Walker of Gestingthorpe. [3] Clarke v Meadus [2010] EWHC 3117 (Ch) at [75], Gillett v Holt [2001] Ch 210 at 235; Giumelli v Giumelli (1999) 196 CLR 101 at [49]-[50]; Chan Gordon v Lee Wai Hing [2011] 2 HKLRD 506 at [52]-[55]. [4] See also Repatriation Commission v Tsourounakis [2004] FCAFC 332 at [56], per Spender, Kiefel, Emmett JJ. [5] See the papers by Sir Terence Etherton “Constructive trusts and proprietary estoppel: the search for clarity and principle” (2009) 2 Conveyancer and Property Lawyer 104. Paul Matthews “The Words Which Are Not There: A Partial History of the Constructive Trust” in Charles Mitchell (ed) “Constructive and Resulting Trusts”, Hart Publishing 2010. [6] Mr Justice Wilson Chan, HCA 1652/2012 (10 April 2017). [7] Lam VP, Cheung and Au JJA [2020] HKCA 148 at §§1.41 and 1.44. The scope of the remitter was decided by a majority (Lam VP and Au JA). [8] Re-Re-Amended Defence of D1 and D2 and Counterclaim (“the Counterclaim”), §§24-29. [9] M’Mahon v Burchell (1846) 2 Ph 127 at 134-135, 41 ER 889; Henderson v Eason (1851) 17 QB 701 at 720, 117 ER 1451; Griffies v Griffies (1863) 8 LT 758 at 758; McCormick v McCormick [1921] NZLR 384 at 385; Rees v Rees [1931] SASR 78 at 81; and Jones (AE) v Jones (FW) [1977] 1 WLR 438 at 441-442. [10] (1994) 35 NSWLR 206 at 221. [11] Citing Ford v Grey (1703) 1 Salk 285; 91 ER 253 ; 6 Mod 44 ; 87 ER 807. [12] M’Mahon v Burchell (1846) 2 Ph 127 at 134-135, 41 ER 889. Followed in Kennedy v De Trafford [1897] AC 180 at 190. [13] Cresswell v Hedges (1862) 1 H & C 421, 158 ER 950; McCormick v McCormick [1921] NZLR 384 at 385; Rees v Rees [1931] SASR 78 at 80; Luke v Luke (1936) 36 SR (NSW) 310 at 314; Dennis v McDonald [1982] Fam 63 at 70-71; Forgeard v Shanahan (1994) 35 NSWLR 206 at 221. [14] Jacobs v Seward (1872) LR 5 HL 464 at 472. [15] Judgment §110. [16] Re-Amended Statement of Claim, §§1, 3-7 and §8: “In the premises, [P, D1 and D2] as tenants-in-common of the Land [including House 774], are all entitled to the possession and occupation of the same ...”. [17] Supplemental Witness Statement of the Plaintiff at §7. [18] Ibid at §8. [19] P states that she does not remember the sentence imposed: Second Supplemental Witness Statement of the Plaintiff, at §43. [20] And possibly, by re-casting the counterclaim as one for partition, a claim for an account of rent received. [21] Judgment §112. [22] Ibid at §113. [23] Court of Appeal, per Lam VP at §§1.40, 1.42 and 1.43. [24] Appearing for P with Mr Anson Wong Yu Yat. [25] Court of Appeal, per Cheung JA at §7.1. [26] Appearing for the respondents with Mr Alan Kwong and Ms Stephanie Wong. [27] Court of Appeal, per Cheung JA at §7.2. [28] [1993] 1 WLR 1046 at 1050. [29] [2004] 1 P & CR 12 159 at 168; referred to by Cheung JA at §7.3. [30] [2009] 1 WLR 1124 per Blackburne J at §43. [31] Court of Appeal §1.40. [32] Court of Appeal per Lam VP at §1.42. [33] Ibid at §1.43. [34] Ibid at §§1.41 and 1.44. [35] Court of Appeal per Cheung JA at §§7.5 and 7.6. [36] Forgeard v Shanahan (1994) 35 NSWLR 206 at 221-222 (citations omitted). [37] Ibid. [38] Ibid. Meagher JA qualified this by reference to “a solitary and curious decision in 1685 which might suggest the contrary. That case is Strelly v Winson (1685) 1 Vern 297; 23 ER 480, an Admiralty case which seems to have wandered into the Chancery Courts. In the course of the judgment the Lord Keeper said: ‘and so where one tenant in common receives all the profits, he shall account in this court as bailiff to the other two for two-thirds’, a proposition which has never been relied on, or even noticed, in any case decided since that date.” We note that this was subsequently viewed favourably in Ryan v Dries [2002] NSWCA 3. However, Strelly v Winson is described as “ancient but frequently forgotten authority” and as a case referring to a “shadowy equitable jurisdiction” in J D Heydon, M J Leeming and P G Turner, “Meagher, Gummow and Lehane’s Equity Doctrines & Remedies, 5th Ed, §26-130. It was not relied on in the present case and is very much an outlier of very uncertain authority. It hardly qualifies as involving “a modern approach”. In Chen Yu Tsui v Tong Kui Kwong [2006] 1 HKC 23 at §§103-104, Deputy High Court Judge Barnabas Fung declined to follow Strelly v Winson. [39] 4 & 5 Anne c 3, s 27: “And from and after the said first day of Trinity term shall and may be brought... by one joynt tenant and tenant in common his executors and administrators against the other as bailiff for receiving more than comes to his just share or proportion and against the executor and administrator of such joynt tenant or tenant in common.” [40] The relevant provision was initially applied to Hong Kong by the combined effect of the Supreme Court Ordinance (1844), section 3, Ordinance No 6 of 1845 and Ordinance No 2 of 1846, section 3 which made applicable to Hong Kong the laws of England as they existed on 5 April 1843 (except those inapplicable to local circumstances). The Statute of Anne (1705) was thus included. However, that Statute ceased to have effect upon enactment of the Application of English Law Ordinance (Cap 88) because it was not expressly included among the applicable laws (as required by section 4 of Cap 88). Cap 88 did not in any event survive the transition to the HKSAR in consequence of the Decision of the Standing Committee of the National People’s Congress on Treatment of the Laws Previously in Force in Hong Kong in accordance with Article 160 of the Basic Law which stated in paragraph 2 and Annex 1 that Cap 88 is not adopted. See Chen Yu Tsui v Tong Kui Kwong [2006] 1 HKC 23 at §88 on the Statute of Anne and Suen Toi Lee v Yau Yee Ping (2001) 4 HKCFAR 474 at §33 and Liu Ying Lan v Liu Tung Yiu [2003] 3 HKLRD 249 at §61 on the fate of Cap 88. [41] Dennis v McDonald [1982] Fam 63 at 71; In re Pavlou (A Bankrupt) [1993] 1 WLR 1046 at 1048-1049; Re Byford (deceased) [2004] 1 P & CR 12 at §30. [42] (1884) 15 QBD 60 at 67. [43] (1905) 2 CLR 387 at 396. [44] (1979) 27 ALR 330 at 346 (citations omitted). See also Brickwood v Young (1905) 2 CLR 387 at 395. [45] (1864) 33 Beav 534 at 534; 55 ER 476. [46] [1921] NZLR 384 at 387-388. [47] (1994) 35 NSWLR 206 at 223 (citations omitted). [48] At §§116-117. [49] [1982] Ch 391. [50] Ibid at 400. [51] Ibid at 405. [52] [1990] 1 WLR 616. [53] Ibid at 624. [54] Ibid at 625. [55] Ibid at 627. [56] [1993] 1 WLR 1046. [57] It had been acquired by them as joint tenants but the effect of the bankruptcy order against the husband was to sever the joint tenancy and cause the property to be owned by them as tenants in common in equal shares. (Ibid at 1048) [58] Ibid at 1048. [59] (1884) 15 QBD 60. [60] [1993] 1 WLR 1046 at 1048. [61] Ibid at 1049. [62] Hill v Hickin [1897] 2 Ch 579 and Dennis v McDonald [1982] Fam 63. [63] [2004] 1 P & CR 12 159. [64] Ibid at §30. [65] Ibid at §31. [66] (1884) 15 QBD 60. [67] [2004] 1 P & CR 12 159 at §22. [68] Ibid at §40. [69] [2009] 1 WLR 1124. [70] Trusts of Land and Appointment of Trustees Act 1996, sections 12-15. [71] [2009] 1 WLR 1124 at §23. [72] Ibid at §24. [73] Ibid at §34. [74] Ibid. [75] Ibid at§35. [76] [2017] 1 WLR 4005. [77] Ibid at §1. [78] Ibid at §30: “The principles of equitable accounting were developed by the courts of equity to enable an adjustment to be made to the division of the net proceeds of sale of a co-owned property in appropriate cases.” [79] Ibid at §63. [80] Ibid at §64. [81] Ibid at §77. 
Chief Justice Ma: 1. For the reasons contained in the judgment of Mr Justice Fok PJ, I agree that this appeal should be dismissed. The policy behind the Motor Vehicles Insurance (Third Party Risks) Ordinance, Cap.272 is, as Mr Justice Fok has pointed out, to protect the public from the use of motor vehicles intended or adapted for use on roads to which the public has access. Whether or not a vehicle will come under the requirements of the Ordinance is a mixed question of law and fact. In the present case, the relevant vehicle (a light utility vehicle) was subject to the requirements of compulsory third party insurance under the Ordinance. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Tang PJ: 3. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Fok PJ: A. Introduction 4. The appellant was convicted of one charge of using a motor vehicle without third party insurance contrary to sections 4(1) and 4(2) of the Motor Vehicles Insurance (Third Party Risks) Ordinance, Cap.272 (the MVI(TPR)O). The two contested elements of the offence were the status of the place where the use of the vehicle occurred and the status of the vehicle being used and this appeal accordingly raises questions concerning the definitions of “road” and “motor vehicle” in the MVI(TPR)O. A.1 The facts 5. Kau Sai Chau is an island located off Sai Kung, on which there are three 18-hole public golf courses, namely the North, South and East Courses. The golf courses are operated by the appellant, having been built with funds donated by the Hong Kong Jockey Club. On the island there is a club house, a driving range and the three golf courses in question. The evidence at trial was that any member of the public may visit the club house and use the driving range. Subject to meeting certain qualifying handicap requirements, any member of the public can play on the golf courses. Being an island, access is by ferry from Sai Kung and shuttle buses are available to take visitors from the ferry pier to the club house for access to the golfing facilities. 6. On 18 May 2009, at about 1pm, one of the appellant’s employees was instructed to drive four workers in a vehicle owned by the appellant to the 4th hole of the Kau Sai Chau East Course. The workers were engaged by a contractor of the appellant to carry out electrical engineering work at the golf courses. One of the workers sat in the front passenger seat of the vehicle next to the driver and the other three sat behind them on a toolbox in the cargo bed of the vehicle. At about 1.15pm, whilst travelling downhill on an unnamed road near the 4th hole of the East Course, the vehicle overturned and the driver and its passengers were thrown from it. As a result, all of the vehicle’s occupants were injured and, tragically, one of the passengers, who was in the rear of the vehicle, died from his injuries. 7. Photographs tendered in evidence showed the unnamed road to be part of what the Judge below described as “a very well-constructed road system with clear road markings and signs”, part of which was a “dual carriageway of substantial width”.[1] The evidence at trial established that golf carts could be driven, and golfers could walk, on the unnamed road. The users of the road might be golfers playing the 4th hole of the East Course or golfers using the unnamed road for access to the North or South Courses. There was no physical obstruction preventing access to the unnamed road and there was no sign declaring the road to be private or restricted. The state of the evidence was that any person who went to the island, whether golfers playing on one of the courses or users of the driving range, could go to the unnamed road without any hindrance. 8. The vehicle, described below as a light utility vehicle, was purchased new in about 2005 and used exclusively on the golf courses at Kau Sai Chau. It was a left-hand drive four-wheeled vehicle with a petrol engine and its maximum speed was 24 km/h. It was 2.8 m long and 1.25 m wide and had two seats in the front for the driver and a passenger and a cargo bed at the rear which tilted manually. The owner’s manual for the vehicle described it as a “light duty utility vehicle” which was “designed and manufactured for off-road use” and “for use on improved roads (but not on public highways)”. It had a flat plastic roof mounted on a metal frame over the driver’s and front passenger’s seats and a windscreen but it did not have doors. Nor did it have rear view mirrors, seatbelts, left/right indicator lights at the front or rear, a speedometer or plates for the vehicle’s registration number or licence. On the other hand, it did have brake lights, rear lights, an engine hour meter and a low oil pressure light as additional equipment. A.2 The charge 9. The appellant faced a sole charge of using a motor vehicle without third party insurance contrary to sections 4(1) and 4(2) of the MVI(TPR)O. 10. Section 4(1) of the MVI(TPR)O provides: “Subject to the provisions of this Ordinance it shall not be lawful for any person to use, or to cause or permit any other person to use, a motor vehicle on a road unless there is in force in relation to the user of the vehicle by that person or that other person, as the case may be, such a policy of insurance or such a security in respect of third party risks as complies with the requirements of this Ordinance.” And section 4(2) of the MVI(TPR)O specifies the penalties to which a person convicted under section 4(1) is liable. 11. As will be apparent, there are four constituent elements of the offence, namely: (i) the use, (ii) of a motor vehicle, (iii) on a road, and (iv) without a complying policy of insurance covering third party risks. A.3 The proceedings below 12. The case was tried in the Kowloon City Magistrates’ Court in September 2011. The appellant did not dispute the use of the vehicle (element (i)) or that there was no third party risks insurance policy in effect in respect of that user (element (iv)) and the Magistrate[2] was satisfied these two elements of the offence were established. 13. The Magistrate was also satisfied that the other two elements of the offence, which the appellant did contest, were also established beyond reasonable doubt. 14. He found, applying the definition of “public” in section 3 of the Interpretation and General Clauses Ordinance, Cap.1 (the IGCO),[3] that a class of the public, namely golfers playing the East Course, could have access to the section of the road where the incident occurred and so held that the unnamed road near the 4th hole of the East Course was a “road” within section 2 of the MVI(TPR)O (see below) (element (iii)). Alternatively, he held that it was a “private road”[4] so that this element of the offence was made out by reason of section 3 of the MVI(TPR)O (see below). 15. Applying the test laid down in Burns v Currell [1963] QB 433, the Magistrate also held that the light utility vehicle was a “motor vehicle” within section 2 of the MVI(TPR)O (see below) (element (ii)). 16. The consequence of the Magistrate’s conclusion in the alternative regarding the road (element (iii)) is that, by reason of the provisions of section 119 of the RTO[5] and assuming the light utility vehicle was correctly held to be a motor vehicle, it rendered moot the question of whether the road was in fact public or private. 17. Upon conviction, the appellant was fined $6,000. 18. The appellant’s appeal against its conviction for the offence was heard in the Court of First Instance on 30 August 2012. By his judgment dated 8 October 2012, the Judge[6] dismissed the appeal affirming the Magistrate’s conclusions on the two contested elements of the offence. 19. As to the status of the unnamed road, the Judge held that the public had access to the unnamed road and thus affirmed the Magistrate’s conclusion that it was a “road” within section 2 of the MVI(TPR)O. The Judge held that the Magistrate had been correct in adopting section 3 of the IGCO in construing the meaning of “public” in the definition of “road” in section 2 of the MVI(TPR)O. He found that the golfers playing on the Kau Sai Chau courses were a “class of the public” within the IGCO and that they were using or had access to the unnamed road. Alternatively, he found that the users of the driving range were members of the general public and that they were using or had access to the unnamed road. In any event, the Judge agreed with the Magistrate that, even if the road was not a public road, it was a private road within section 2 of the RTO and therefore this element of the offence was made out by reason of section 3 of the MVI(TPR)O. 20. The Judge held that the applicable test for whether a vehicle is “intended” for use on roads was that in Burns v Currell and that the Magistrate had correctly applied this test in concluding that the light utility vehicle was a motor vehicle within section 2 of the MVI(TPR)O. A.4 The certified questions 21. On 4 December 2012, on the appellant’s application, the Judge certified four questions as points of law of great and general importance and, on 8 April 2013, the Appeal Committee granted leave to appeal on those four questions, namely: (1) “Whether the inclusion of ‘public’, under section 3 of the IGCO, in the word ‘road’, under section 2 of the MVI(TPR)O, should be displaced by any contrary intention appearing either from the IGCO or from the context of MVI(TPR)O?” (2) “What constitutes a special class of the public under common law or a ‘class of the public’ in section 3 IGCO?” (3) “Whether the application of MVI(TPR)O to ‘private roads’ in section 3 of MVI(TPR)O expanded the Burns test of whether a reasonable person looking at the vehicle would say that one of its users was use on a ‘road’ as defined in section 2 MVI(TPR)O?” (4) “Whether the Legislature intended a vehicle ‘constructed or intended for use for the carriage of passengers on golf courses’ (cf. Interpretation of ‘golf cart’ in section 2 RTO) to be a ‘motor vehicle’ in section 2 MVI(TPR)O?” 22. The first two questions relate to the status of the unnamed road where the accident occurred and the latter two questions relate to the question of whether the light utility vehicle was a motor vehicle within the statutory provision creating the offence. B. The road element of the offence (element (iii)) B.1 The provisions of the MVI(TPR)O apply even if the unnamed road is a private road 23. The relevant offence is committed by user of an uninsured motor vehicle on a “road” as defined in section 2 of the MVI(TPR)O. Section 2 defines “road (道路)” as meaning: “any highway and any other road to which the public have access and includes any road upon which a member of the public is allowed to drive a vehicle only if in possession of a permit issued by the Commissioner of Police or the Commissioner for Transport.” 24. The word “road” in the phrase “any other road to which the public have access” is not separately defined in the MVI(TPR)O. In the IGCO, “road” is defined as having the same meaning as “street” and “street” is there defined as meaning: “(a) any highway, street, road, bridge, thorough-fare, parade, square, court, alley, lane, bridle-way, footway, passage or tunnel; and (b) any open place, whether or not frequented by the public or to which the public have or are permitted to have access”. 25. Subject to the question of whether the public had access to it, to which it will be necessary to return for the reasons explained in Section B.2 below, the unnamed road in the present case was a “road” since it clearly fell within paragraph (a) of the IGCO definition of “street”. Nor can there be any doubt that the IGCO definition of “road” should apply here since no contrary intention appears either from the IGCO or the context of the MVI(TPR)O. [7] But, in any event, even applying the ordinary dictionary meaning of the word “road”[8], the unnamed road was plainly a road. It was a thoroughfare designed for or regularly used by vehicles to get from one place to another and this characteristic is sufficient to constitute it a road.[9] 26. Be that as it may, the MVI(TPR)O was amended in 1988[10] to reflect a fundamental revision to the RTO (by the same amending ordinance) by the introduction of a definition of “private road” and the exclusion of a “private road” from the definition of “road” in the RTO and the express application of most (but not all) of the provisions of the RTO to private roads. 27. The amendment of the MVI(TPR)O in 1988 made it clear that the offence of using a motor vehicle without third party insurance could be committed by such user on a “private road”. Section 3(1) of the MVI(TPR)O provides: “For the avoidance of doubt, it is hereby declared that the provisions of this Ordinance shall apply to private roads, other than private roads in areas wholly or mainly used for the carrying on of construction work or industry, as they apply to roads and, for that purpose, the provisions of any other Ordinance which relate, in any way, to the provisions of this Ordinance shall apply accordingly.” 28. Section 3(2) of the MVI(TPR)O provides that the term “private road (私家路)” means: “a private road within the meaning of the Road Traffic Ordinance (Cap. 374).” And in section 2 of the RTO, “private road (私家路)” is given the following very wide meaning: “every thoroughfare, street, lane, alley, court, square, car park, passage, path, way and place to which access by the public may be restricted under common law, irrespective of whether such access is so restricted, but does not include- (a) (Repealed 23 of 2002 s. 91) (b) any part of the carriageway of the North-west Railway designated by the Commissioner for the purposes of the definition of road (道路) in this section by notice in the Gazette; (Added 80 of 1988 s. 3)”. 29. There can be no doubt that, if it is not a road to which the public have access, the unnamed road adjacent to the 4th hole of the Kau Sai Chau East Course, where the use of the light utility vehicle at the time of the offence occurred, is nonetheless still a “private road” within section 3(2) of the MVI(TPR)O. 30. That being the case, the issue of whether the unnamed road was a “road” within the definition in section 2 of the MVI(TPR)O is academic so far as this particular element of the offence is concerned because, even if the unnamed road were a “private road” within section 3(2) of the MVI(TPR)O, user of a motor vehicle there would satisfy this element of the offence. B.2 Was the unnamed road “a road to which the public have access”? 31. Although the relevant offence may now be committed by user of a motor vehicle on a private road, it remains relevant to consider whether the unnamed road was a road to which the public had access because, as will be seen, its status as such a road, or not, is relevant to the question of whether the light utility vehicle was within the definition of “motor vehicle” in the MVI(TPR)O. 32. The statutory definition stipulates that a “road” is either “any highway” or “any other road to which the public have access”. The reference to a highway is to a road which forms part of the public road system of Hong Kong, which for convenience may be referred to as a public highway. The unnamed road in the present case is not a public highway, and it was not the prosecution’s case at trial that it is, so one must ask if the public had access to the unnamed road to ascertain if it fell within the relevant definition. 33. The question of whether a place is one to which the public have access for the purposes of road traffic and ancillary legislation is largely a matter of fact and degree but it is a question of law in any case whether the facts are sufficient to support a particular conclusion. B.3 The evidence of user of the unnamed road 34. The evidence before the Magistrate relevant to the user of the unnamed road was as follows: (1) PW1, one of the four workers on board the light utility vehicle, who had worked at the East Course many times, said that pedestrians were not allowed to walk along the unnamed road when someone was playing golf there, but vehicles for playing golf were allowed to use the road and golfers were allowed to walk along the road. (2) PW3, the workshop manager of the appellant’s greens maintenance department said that he had seen vehicles other than the light utility vehicle using the unnamed road. He said that golfers playing the 4th hole of the East Course would use their golf carts along the unnamed road and that the road led to the North and South Courses. There was no sign indicating the unnamed road was a private road. (3) PW5, the appellant’s human resources manageress, gave evidence that members of the public could on advance booking and payment of a fee use the driving range. The appellant did not have a membership system and members of the public who satisfied the relevant handicap qualifications could play on the various courses. (4) PW6, an officer of the police traffic accident investigation team, gave evidence that it was not necessary to pass through any gate when going from the club house to the East Course and no personnel checking people entering the golf course, nor was there a “Private Road” sign on the unnamed road. 35. On the above evidence, the Judge held that: “24. … golf carts were allowed to be used on the unnamed road and golfers were allowed to walk there. There was no physical obstruction preventing access to the unnamed road and there was no sign declaring the road to be a private road or restricted road. 25. PW3 said that it was the golfers playing at hole number 4 of the East Course who used their golf carts there, but the evidence of PW1 was simply that golf carts could use and golfers could walk on the unnamed road. PW3 further said that the unnamed road could lead to South Course and North Course. Although he added that they were far away it showed that golfers at these 2 courses could also use the unnamed road. 26. According to the evidence of PW6 there was no gate separating East Course and the Club House. PW5 said that any member of the public on advance booking and payment of fee could use the driving range on the island. These people could always, if they chose to do so, go to the East Course and use the unnamed road. The state of the evidence is such that any person who had gone up to the island, be they golfers or people using the driving range, could go to the unnamed road without any hindrance.” B.4 The Judge’s approach 36. The Judge approached the status of the unnamed road by reference to the meaning of public in section 3 of the IGCO, which he held applied to define the word “public” in section 2 of the MVI(TPR)O. He held that: “27. … The golfers allowed to go to play at the 3 golf courses on the island were no doubt a special class of the public as distinct from the general public.” (Emphasis added) 37. He then addressed the question of whether those golfers were “a class of the public” within the definition in the IGCO and how such a class of the public was to be determined. He approached that question by reference to the size of the class, the nature of the qualification required to become a member of the class, the difficulty involved for a member of the public to become qualified and who had control over how the qualification could be obtained and the number of people so qualified.[11] He concluded that golfers playing on the Kau Sai Chau courses were a sufficiently large class of persons to constitute a class of the public within the IGCO definition.[12] 38. The Judge considered the users of the driving range and said they were of a different category. Although he expressed his conclusion somewhat equivocally, the Judge appeared to conclude that they were members of the general public.[13] B.5 Golfers on the golf course remained members of the general public 39. In my view, the Judge was, with respect, wrong to conclude that the golfers playing on the three courses at Kau Sai Chau were not members of the public in general. 40. Kau Sai Chau is a public golf facility and not a private club. Membership is not required in order to enter and use the facility and members of the public do so for the purpose of playing golf, practising or enjoying the surroundings of the golf courses. The fact that having a particular golf handicap may be required to play on any of the three golf courses does not mean that they enter for any purposes other than their own purposes and in their capacity as members of the public. They are not there to exercise their rights or privileges as a member of a private club. 41. The appellant contended that the imposition of handicap requirements for access to the golf courses constituted a screening process that distinguishes qualifying golfers from other members of the public and relied on Harrison v Hill (1932) J.C. 13 to support the distinction between the general public and golfers having a qualifying handicap as a special class of the public. 42. In Harrison v Hill, the Scottish High Court of Justiciary addressed the construction of “public” in section 121(1) of the Road Traffic Act 1930 (the RTA 1930) which provided that: “‘Road’ means any highway and any other road to which the public has access, and includes bridges over which a road passes.” An issue arose as to the status of an ordinary farm road which formed the access to a farm from a public highway. It was found that members of the public not having business at the farm frequently walked upon it. On the question of whether the road was one to which the public had access, the Lord Justice-General (Lord Clyde) said (at p.16): “I think that, when the statute speaks of ‘the public’ in this connexion, what is meant is the public generally and not the special class of members of the public who have occasion for business or social purposes to go to the farmhouse or to any part of the farm itself; were it otherwise, the definition might just as well have included all private roads as well as all public highways.” (Emphasis added) 43. It was said that this approach was “echoed” in A-G v Ng Kwan [1987] 1 HKC 183, a case concerned with whether a road leading to a refuse tip was a road within the meaning of section 2 of the RTO. The general public could not use the road in question unless authorised but there was no restriction on any member of the public coming onto the road during the time when it was opened, provided they were doing so in order to dump refuse. The Court of Appeal held that members of the public who entered on a road leading to a refuse tip were using the road as general members of the public and not as a special class of persons and said (at p.190D-F): “The question is then whether or not the public had access to it. We are satisfied that any member of the public, if he so choose [sic], could enter into the road for the purpose of disposal of refuse. The trial judge held that the general public could not use the road unless authorised, but he also held that there was no restriction on entry for vehicles bringing in refuse. We are satisfied that this does not mean that only a special class of persons had access to the road within the meaning of that expression as used in R v Beaumont. Here the class of persons who could use the road were such members of the public who elected at any time to take refuse to the tip. Any member of the public could at any time, then bring himself within that class.” 44. I am unable to accept that the screening process constituted by the requirement of having a qualifying handicap was such as to change the character of those particular qualifying golfers as members of the public generally. The screening process, which only applies to the golf courses themselves and not to the driving range or club house, does not endow them with any special characteristic that deprives them of their identity as members of the general public and instead turns them into a special class of the public. To the point of the boundary of the golf facility, in practical terms the jetty on the island, these persons are plainly members of the general public and, when they disembark on the island to play golf, use the driving range and walk the courses, they do not lose that characteristic as members of the general public. 45. The evidence supports the findings of the Magistrate and Judge that golfers playing the golf courses used the unnamed road. Although slightly equivocal as regards those playing the North and South Courses, there is no doubt on the evidence that golfers playing the East Course actually used that road. The evidence regarding the users of the driving range is less clear and, whilst it demonstrated that they could use the unnamed road, I have some doubt that it amounted to evidence of actual user of that road on their part. Be that as it may, the evidence amply supported the finding that golfers used and had access to the unnamed road. For the reasons I have stated, those golfers did so as members of the general public and therefore the unnamed road was a road to which the public had access. 46. It is not necessary to demonstrate that any and all members of the public have satisfied the relevant screening restrictions imposed and I do not regard Harrison v Hill or A-G v Ng Kwan as suggesting this necessity. It is sufficient that any member of the public can acquire a qualifying golf handicap. Neither its acquisition nor its lack in any way diminishes the golfer’s or non-golfer’s status as a member of the public. In the present case, this particular conclusion is supported by the wording of section 2 of the MVI(TPR)O which provides that a “road … includes any road upon which a member of the public is allowed to drive a vehicle only if in possession of a permit issued by the Commissioner of Police or the Commissioner for Transport.” This demonstrates that the mere need to have a particular skill, namely an ability to drive sufficiently competently to pass a driving test, which clearly not everyone has, is not critical to remaining a member of the public. B.6 Section 3 of the IGCO 47. That the unnamed road was one to which the public had access is all the more clear when one takes into account, as the courts below did, the fact that, by reason of the definition of “public” in section 3 of the IGCO, the public includes a class of the public. On the hypothesis they are not members of the general public, there can be no doubt that golfers on the Kau Sai Chau public golf courses having the required qualifying golf handicap are members of a class of the public. 48. In the face of this difficulty, it was the appellant’s case that the IGCO definition of “public” was displaced by a contrary intention and that, instead, Lord Clyde’s construction of “public” in section 121(1) of the RTA 1930 in Harrison v Hill should be adopted so that access merely by a class of the public was insufficient to constitute the road one to which the public had access. This contention was supported by extensive argument in the appellant’s printed case. 49. The appellant’s first argument was that the legislative history of the MVI(TPR)O is such that it should be construed consistently with equivalent provisions in the RTA 1930. The argument proceeded as follows: The MVI(TPR)O came into operation on 9 November 1951 before the IGCO, which came into operation as from 31 December 1966. Although the IGCO applies retrospectively before its commencement, Hong Kong courts prior to its enactment must have applied English authorities. In 1988, the Legislature amended the MVI(TPR)O to apply to private roads. If the word “public” in section 2 of the MVI(TPR)O was understood to comprise a class of the public, then, as Lord Clyde observed in Harrison v Hill, a “road” in section 2 would include all private roads as well as public highways. However, the Legislature chose to enact section 3 to extend the MVI(TPR)O to private roads. Thus, the meaning of a road to which the public has access in section 2 remained the same as Lord Clyde’s definition of road in section 121(1) of the RTA 1930. 50. The appellant’s second argument was that the context of section 2 of the MVI(TPR)O indicates a contrary intention. Here, the appellant’s submission was that the Judge attributed too much weight to the omission of the Legislature to define “public” in the MVI(TPR)O and that a contrary intention to the IGCO could also appear in an ordinance where the Legislature omitted to define a particular word. Thus, it was submitted that a contrary intention on the part of the Legislature could be inferred from the definition of “road” in section 2 of the MVI(TPR)O itself because, in contrast to the detailed definition of “road” in the RTO, there is a lack of detail in the MVI(TPR)O definition. It was submitted that the Legislature’s decision not to qualify the word “public” in section 2 of the MVI(TPR)O was similarly deliberate. Thus, it was submitted, a contrary intention to the application of section 3 of the IGCO was apparent and the Harrison v Hill definition of “public” should be followed. 51. The appellant’s third argument as to a contrary intention to the application of section 3 of the IGCO was based on the decision in A-G v Ng Kwan. It was submitted that the Judge misinterpreted A-G v Ng Kwan. The Judge correctly noted that Court of Appeal there did not decide whether the word “public” in the definition of “road” in section 2 of the RTO had the same definition as in section 3 of the IGCO. However, the manner in which the Court of Appeal determined whether the public had access “echoed”, it was submitted, the Harrison interpretation. 52. I do not accept those arguments and do not agree that there is an apparent intention not to apply the definition of “public” in section 3 of the IGCO to the meaning of the word “public” in section 2 of the MVI(TPR)O. 53. The absence of the IGCO section 3 definition of “public” prior to 1966 does not, in my view, mean that the Hong Kong courts would inevitably have reached the same conclusion on the meaning of the word “public” in section 2 of the MVI(TPR)O as was reached in Harrison v Hill. In that case, the issue was whether a road came within the definition of that term in section 121 of the RTA 1930 and arose in the context of a charge of driving whilst disqualified. The context of the MVI(TPR)O is different. The long title to the MVI(TPR)O states that its purpose is: “To make provision for the protection of third parties against risks arising out of the use of motor vehicles”. The primary focus of the MVI(TPR)O is therefore the protection of the public from the use of vehicles rather than the regulation of the places where such vehicles are used. 54. The contrast between the detail of the definition of “road” in section 2 of the RTO and the relatively more general definition in section 2 of the MVI(TPR)O does not take matters any further. The RTO and the MVI(TPR)O are different ordinances with different objects and purposes. The former is an ordinance “To provide for the regulation of road traffic and the use of vehicles and roads (including private roads) and for other purposes connected therewith”, whereas the purpose of the latter is that set out above. There is no reason to think that the more general definition of “road” in section 2 of the MVI(TPR)O was the result of a deliberate legislative decision to adopt the definition of “road” from the RTA 1930 or the specific ruling in Harrison v Hill. 55. Nor, in my view, does the addition in 1988 of section 3 of the MVI(TPR)O to extend the provisions of that ordinance to private roads assist the appellant. In the first place, the provision was inserted “[f]or the avoidance of doubt”, thereby indicating that the intention had always been to include such roads. Further, since, as I have held, one cannot discern a deliberate decision to adopt the Harrison v Hill definition of “road”, the argument that the Legislature should be taken to have intended to retain it when section 3 was added is not sustainable. 56. I do not consider the argument based on the omission to define “public” in the MVI(TPR)O to be sound. Whilst I accept that a contrary intention to the IGCO could also appear in an ordinance where the Legislature omitted to define a particular word, this would likely be in the nature of an exception rather than the rule. It is true that, by way of example, the appellant was able to point to the Film Censorship Ordinance, Cap.392 (the FCO). The FCO section 9(2) refers to the public without defining it, but it is clear from a reading of section 9(2) that section 3 of the IGCO clearly could not apply without producing an absurd or anomalous meaning. In general, where an ordinance does not define a word, for which a definition is provided in the IGCO, the prima facie assumption will be that the latter general definition is intended to apply to the use of that same word in the ordinance in question unless, as in the case of the FCO but not here, a contrary intention applies by reason of absurdity or anomaly. 57. Finally, in relation to A-G v Ng Kwan, a first point to note is that it is not authority for the proposition that a contrary intention to the application of the definition of “public” in section 3 of the IGCO appears from the RTO. Notwithstanding the headnote of the reported decision, whilst the ratio of A-G v Ng Kwan was that the definition of “road” in section 3 of the IGCO was displaced by a contrary intention in section 2 of the RTO, the court did not hold that the definition of “public” in section 3 of the IGCO was similarly displaced by the RTO. 58. Instead, it is clear from the judgment of the court[14] delivered by Penlington J that the decision proceeded on the basis that the court was satisfied that the road was a road within the meaning of section 2 of the RTO because the public had access to it (see pp.189H-190D). After concluding that members of the public who entered the road for the purposes of disposing refuse were doing so as members of the public generally in the passage set out in paragraph 43 above, the Court continued (at p.190F): “We are therefore satisfied that even without calling in aid the definition of ‘public’ in the Interpretation Ordinance, the public did have access to this road and that it was therefore a road within the meaning of the Road Traffic Ordinance. When one also considers the definition of public in the Interpretation Ordinance, it is then clear beyond any doubt.” The Court would clearly not have said this if it were holding that the IGCO definition of “public” was displaced. 59. A second point to note is that A-G v Ng Kwan was not concerned with the MVI(TPR)O at all but with the definition of “road” in the RTO. The court held that the word “include” in the definition of road in the RTO meant “means and includes” because of the “extreme detail of the definition in the [RTO]” (p.190B-D). Because of this a contrary intention to the definition of road in the IGCO was evinced. No such detail exists in the definition of “road” in the MVI(TPR)O and there is no definition of “public” in that ordinance. 60. I do not therefore agree that the way the Court of Appeal construed “public” in A-G v Ng Kwan, even if it may be correct to characterise that construction as having “echoed” the interpretation in Harrison v Hill, means that the definition of “public” in section 3 of the IGCO is displaced for the purposes of the MVI(TPR)O. 61. For the above reasons, the unnamed road was clearly one to which the public had access. C. The motor vehicle element of the offence (element (ii)) 62. The other contested element of the offence was whether the light utility vehicle was a “motor vehicle” as defined in the MVI(TPR)O. 63. Section 2 of the MVI(TPR)O, provides that “motor vehicle (汽車)” means: “a vehicle intended or adapted for use on roads, and propelled by any form of mechanical power, and includes a motor bicycle with or without a side-car or trailer, a motor tricycle and a bicycle or tricycle assisted by a motor, and a village vehicle within the meaning of section 2 of the Road Traffic Ordinance (Cap.374), but not a vehicle drawn by any other motor vehicle, nor any conveyance for use solely on railways or tramways; (Amended 31 of 1986 s. 8) [cf. 1930 c. 43 s. 1 U.K.]” 64. The term “village vehicle (鄉村車輛)” is in turn defined in section 2 of the RTO as meaning: “(a) a driver operated village vehicle; (b) a pedestrian controlled village vehicle; or (c) a golf cart; (Replaced 80 of 1988 s. 3)”. 65. The latter term “golf cart (高爾夫球車)” is defined in the same section of the RTO as meaning: “a motor vehicle having an overall length not exceeding 3.2 m and an overall width not exceeding 1.2 m constructed or intended for use for the carriage of passengers on golf courses, in addition to the carriage of a driver; (Added 80 of 1988 s. 3)”. 66. Although the light utility vehicle in the present case bears some resemblance to a golf cart, having a similar chassis and body structure, particularly at the front of the vehicle, it is not a “golf cart” as defined in the RTO since its width of 1.25 m exceeded the statutory maximum of 1.2 m. Furthermore, it was not “constructed or intended for use for the carriage of passengers on golf courses, in addition to the carriage of a driver”. The vehicle was called, in the Owner’s Manual & Service Guide, a “Workhorse” and it was described there as “a light duty utility vehicle”. Instead of a platform at the rear on which to store and secure golf clubs, typically found on a golf cart, this vehicle had a flat cargo bed and was clearly designed to be a work vehicle able to transport cargo. Even disregarding the statutory dimensions, by reason of its design and obvious intended purpose as a work vehicle, the light utility vehicle could not properly be described as a golf cart and it was not suggested in argument that it was a golf cart. 67. Nor did either party suggest it was a village vehicle, that type of vehicle being a motorised cart “used to transport agricultural produce, building materials and other goods in rural and other areas which are inaccessible or closed to ordinary vehicles”.[15] 68. The material question is therefore whether the light utility vehicle was “a vehicle intended or adapted for use on roads”. C.1 The test to determine if a vehicle is a motor vehicle 69. The correct test to be applied is the objective test propounded by Lord Parker CJ in Burns v Currell (at p.440),[16] namely whether a reasonable person looking at the vehicle would say that one of its users would be a road user. This does not mean what the user or manufacturer intended and also does not mean user by someone “losing his senses” or isolated user or user in an emergency. The tribunal of fact will have to ask whether it has been proved beyond reasonable doubt that any reasonable person looking at the vehicle would say that one of its users would be a use on the road. 70. The test was explained by Glidewell LJ in Chief Constable of Avon and Somerset v F (1987) 84 Cr. App. R. 345[17] at pp.350-351 as being: “… what would be the view of the reasonable man as to the general user of this particular vehicle; not what was the particular user to which this particular defendant put it, either at the time in question or, indeed, generally. In other words, if a reasonable man were to say: ‘Yes, this vehicle might well be used on the road,’ then, applying the test, the vehicle is intended or adapted for such use.” 71. In DPP v Saddington [2001] R.T.R. 15,[18] Pill LJ observed (in 2000) that the test had been applied “for approaching 40 years” and should not readily be departed from. 72. In Hong Kong, the test has been consistently applied in order to determine whether a vehicle is within the definition of “motor vehicle” in section 2 of the MVI(TPR)O[19] and section 2 of the RTO.[20] 73. In applying the Burns test there will be cases in which it will not be necessary to have any evidence other than photographs and a description of the vehicle. As Lord Parker observed (at p.440): “… in the ordinary case it seems to me that there will be little difficulty in saying whether a particular vehicle is a motor vehicle or not.” Thus, to take an obvious example, a standard motor car would readily be held to satisfy the Burns test on the basis that the tribunal of fact would conclude that any reasonable person looking at it would say that one of its users would be a use on the road since it was clearly intended for such a use as disclosed by its design, condition, equipment, functions and attributes. 74. There will, however, be other cases in which the answer to the question posed is not so obvious. In those cases, it may be necessary, if the tribunal of fact is to be satisfied that it has been proved to the requisite criminal standard of proof that any reasonable person looking at the vehicle would say that one of its users would be a use on the road, for there to be other evidence beyond just photographs and a description of the vehicle. An obvious example of such a vehicle which would not, without more, be so regarded, is a motorised lawnmower. Another example might be a recreational vehicle patently designed to be used off-road, such as a dune buggy or a Go-Kart. 75. One such type of evidence would be evidence that the vehicle was registered and licensed for use on the road.[21] Other types of evidence would be evidence as to the vehicle’s capability to get through traffic and its speed[22] or evidence that the vehicle was in good condition and roadworthy.[23] 76. A further type of relevant evidence would be evidence of user. In Percy v Smith [1986] R.T.R. 252, the Divisional Court held, at p.258H, that the justices were entitled to take into account the actual use made of the vehicle by the defendant. The evidence of use on a road might demonstrate an intention on the part of the user that it be so used and could be evidence supporting an inference that the user would continue so to use it.[24] On the other hand, evidence of the user of the vehicle might demonstrate that it was not intended to be used on the road.[25] 77. Such evidence of user might also include evidence concerning the number of vehicles of similar design in circulation[26] or the user of the same type of vehicle other than the particular vehicle the subject of the charge.[27] Thus, in Burns v Currell, Lord Parker noted (at p.440) that “[t]here was no evidence that other people used these vehicles on the road”. C.2 The evidence of user in the present case 78. The evidence before the Magistrate relevant to the user of the light utility vehicle was as follows: (1) PW1 said that he had ridden on this type of vehicle about 5 times when he had previously gone to the golf course for work. (2) PW2, one of PW1’s co-workers, said he had ridden on the light utility vehicle in question previously when working on the golf course and about twice before on the day of the accident. (3) PW3 said that both utility vehicles and golf carts would go past the section of the road where the accident occurred. This meant that players on the 4th hole of the East Course would go past this section of the road which could also lead to the North and South Courses. His evidence was that the light utility vehicle was only used within the golf courses and would not be used outside the golf courses. The appellant did not require drivers of the light utility vehicle to have a driver’s licence. (4) PW4, the supervisor of the appellant’s property maintenance department, said the light utility vehicle had been used since around 2004[28] by the maintenance department to take workers out onto the golf courses for work and that he had personally used the light utility vehicle to do so. 79. In the light of that evidence, the Judge held as follows: “72. According to the evidence of PW1 he had travelled on the light utility vehicle or a similar type of vehicle to go to work on the island for roughly 5 times or more before the accident and PW2 said that he had also done the same. PW2 even said that on the day of the accident he had travelled on it for 2 times. PW3 said he had seen light utility vehicles as well as golf carts travelling along the unnamed road. All these show that the light utility vehicle was for general use on the roads on the island including the unnamed road. On the day of the accident the use of the light utility vehicle on the unnamed [road] was not an isolated incident and general and habitual use of the vehicle on the unnamed [road] was contemplated.” C.3 Did the courts below apply the correct test? 80. The Magistrate and Judge both expressly applied the Burns test in arriving at the same finding that the light utility vehicle was a “motor vehicle” within the statutory definition. 81. The appellant’s printed case posited a misdirection in law on the part of the Judge, namely that he allowed the fact that the offence could be committed on a “private road” to infect his objective assessment of whether the light utility vehicle was intended for use on roads. The appellant contended that this was an unconventional and therefore erroneous application of the Burns test. 82. The definition of “motor vehicle” in section 2 of the MVI(TPR)O refers to a “vehicle intended or adapted for use on roads …”. The term “road” is itself defined in section 2 to mean “any highway and any other road to which the public have access …”. Although section 3 declares the provisions of the MVI(TPR)O to apply to “private roads” (as defined in the RTO), there is nothing in that section which suggests that the definition of “motor vehicle” or “road” in section 2 – or, more particularly, the meaning of the word “roads” in the definition of “motor vehicle” there – is affected by that declaratory statement. Further, the linkage between road user and the relevant statutory definition of “road” is supported by Burns v Currell at p.441 and A-G v Ho Kam Mun at p.614F.[29] 83. The appellant is therefore correct, in my view, in contending that the Burns test must be posed by reference to road user on a highway or any other road to which the public have access. 84. I do not accept, however, the appellant’s further contention that the Judge failed to apply the conventionalBurns test and that the Judge erroneously applied some expanded version of that test by reference to the reasonable person’s view that the light utility vehicle would be used on a private road. 85. The relevant part of the Judge’s judgment reads as follows: “73. It has already been found by the court that the unnamed road was a road for the purpose of MVI(TPR)O. Applying the test in Burns v Currell any reasonable man looking at the light utility vehicle would say that one of its uses would be use on the roads. The magistrate correctly applied this test and his conclusion that the light utility vehicle was a motor vehicle within the meaning of MVI(TPR)O cannot be faulted. 74. The court has to ask one question – what if the finding of the unnamed being a road under MVI(TPR)O by the magistrate and this court was wrong and the road was a private road because golfers allowed to use the unnamed road were a special class of people not constituting ‘any class of the public’? Could it then be argued that general use of the light utility vehicle on a ‘private road’ was different from such use on a ‘road’ and therefore the light utility vehicle was not intended or adapted for use on ‘roads’? 75. There is clear evidence that golfers had access to the unnamed road. Even if they were wrongly held by the magistrate and this court to be ‘a class of the public’, they were still people using the unnamed road to which MVI(TPR)O applied. As can be seen from the photographs of the scene the unnamed road was part of a very well-constructed road system with clear road markings and signs and some part of it was of dual carriageway of substantial width. Taking all these into consideration and applying the test in Burns v Currell any reasonable man looking at the light utility vehicle would say that one of its uses would be use on the roads. The light utility vehicle would still be a motor vehicle for the purpose of MVI(TPR)O.” 86. The Judge expressly proceeded to apply the Burns test in paragraph 73 of his judgment on the basis of his finding that the unnamed road was a road to which the public had access and not by reference to the fact that the offence could be committed by user of the vehicle on a private road by virtue of section 3 of the MVI(TPR)O. 87. That being the case, his application of the Burns test to the light utility vehicle taking into account its user on a road to which the public had access was an orthodox and proper approach. Applying that test, the Judge came to the same conclusion as the Magistrate and therefore confirmed the latter’s finding that the light utility vehicle was a “motor vehicle” within section 2 of the MVI(TPR)O. 88. The Judge’s remarks in paragraphs 74 and 75 of his judgment, on the footing that the road was one to which the MVI(TPR)O applied because it was a “private road”, were obiter and, for the reasons set out above, should be disregarded. C.4 Did the Judge reach the correct conclusion in applying the Burns test? 89. In applying the Burns test, each case will turn on its own facts.[30] 90. I have described the light utility vehicle in Section A.1 above and summarised the evidence of its user in Section C.2 above. 91. The fact that the absence of doors, rear view mirrors, seatbelts, indicator lights, a speedometer and licence plates meant that the light utility vehicle might not have been roadworthy in the sense of being able to be used safely on roads,[31] or for the purposes of the legislation relating to motor vehicles, [32] is not decisive. Nor is the stated intention in the owner’s manual that the vehicle was intended for off-road use and not for operation on public streets.[33] It is not necessary, to be a motor vehicle, that it be capable of being driven in traffic in the way that another vehicle might, and it may be one even if its being driven might require the adoption of an ungainly posture.[34] It does not have to be capable of being driven at a particularly fast speed.[35] 92. Although use on a road will not be determinative,[36] there was evidence of user of the light utility vehicle itself and other similar vehicles on the unnamed road and other roads on Kau Sai Chau. I have set out paragraph 72 of the Judge’s judgment above. It is not clear that it was established by evidence that there are roads on the island other than roads within the golf course and related to that facility, nor was there evidence of actual use of them by members of the public. However, for the reasons explained above, the unnamed road where the accident took place was clearly a road to which the public had access. In my view, the evidence before the Judge justified his holding that the general and habitual use of the light utility vehicle on the unnamed road, a road to which the public had access, was intended. The evidence was that the light utility vehicle was regularly used to transport people and equipment along a road and the user was clearly not an isolated or emergency user, nor could it be said to be a user by someone losing his senses. 93. Having regard to the evidence concerning the light utility vehicle and in particular of its user on a road to which the public had access, I have no doubt that in the present case the Magistrate and Judge were correct to find, applying the Burns test, that a reasonable man looking at the vehicle with full knowledge of its characteristics would say that one of its users was use on a “road” as defined in section 2 of the MVI(TPR)O. D. Not necessary to address the certified questions 94. It follows from the discussion above that the unnamed road was a road to which the provisions of section 4 of the MVI(TPR)O applied and the light utility vehicle was a motor vehicle. Therefore, the two contested elements of the offence are determined against the appellant without having to resolve the issues which gave rise to the four certified questions and it is therefore unnecessary to give specific answers to those questions. 95. However, since the issue giving rise to certified question 1 has been addressed above, the answer to that question is “No”. Similarly, the issue giving rise to certified question 3 has been addressed above and the answer is also “No”. 96. Certified questions 2 and 4 do not arise for decision. However, I would add, in respect of certified question 2, which arose from the Judge’s approach as to how to determine what constitutes a “class of the public” under section 3 of the IGCO, that, whilst the Judge was certainly correct to say that this will depend on the circumstances of individual cases, I would not endorse his approach of looking to factors such as the size of the class, the nature and difficulty of the qualification required to become a member of it and the degree of control over how the qualification is obtained.[37] But it is not necessary to this appeal to resolve certified question 2 because it is clear that, since golfers having the required qualifying golf handicap are members of the general public, they must in any event form a class of the public. E. Conclusion 97. For the reasons set out above, the appellant was properly convicted of the offence under section 4 of the MVI(TPR)O and the appeal against conviction must be dismissed. The light utility vehicle was a motor vehicle which was being used on a road to which the provisions of the MVI(TPR)O applied and so there should have been in force in relation to that user an effective policy of third party insurance. Lord Phillips of Worth Matravers NPJ: 98. I agree with the judgment of Mr Justice Fok PJ, and just wish to add some brief comments about the consequences of this judgment. I shall adopt the same shorthand references to the relevant legislation. 99. The MVI(TPR)O was passed in 1951. Roads in Hong Kong were very different in 1951 from roads in Hong Kong today. And the types of motorized vehicles found in Hong Kong in 1951 were very different and much less varied in nature than the motorized vehicles that exist today. 100. The origins of the RTO can be traced back to 1947, but the RTO has been amended and revised on a number of occasions since then. The current RTO, for the first time, draws a clear demarcation between “roads” and “private roads”. Many of the provisions of the RTO apply to both, but some do not – see sections 117 and 118. Significantly, if a vehicle is only driven on private roads it does not have to be licensed and, in consequence, does not have to comply with construction regulations that have to be satisfied if a vehicle is to be registered and licensed. 101. Compulsory third party insurance in relation to motor vehicles reflects the enhanced risk of damage or injury to third parties inherent in the use of such vehicles. That risk is readily apparent in relation to vehicles designed to be used on “roads” as defined in the RTO. 102. There are, however, types of vehicle designed to be used only on “private roads” that do not pose the same degree of risk to third parties, such as battery driven vehicles capable only of low speeds. I question whether, as a matter of public policy, the risk posed by such vehicles justifies requiring those who own and operate them to take out motor insurance against third party risks in respect of them. Indeed, it is open to question whether the risk posed by the vehicle the subject of these proceedings justified imposing a requirement to take out such insurance. 103. The MVI (TPR)O has not been the subject of significant amendment since 1951. The object of this short judgment is to suggest that there is a case for reviewing the ambit of its operation. Mr Michael Ozorio SC and Mr Eric Tsoi, instructed by Baker & McKenzie, for the Appellant Ms Vinci Lam Ag SADPP, of the Department of Justice, for the Respondent [1] Judgment §75 – the reference to “dual carriageway” is a misnomer since, although the road had two lanes, the lanes were not separated by a central reservation and so the road was therefore strictly a single carriageway, but nothing turns on this misdescription. [2] Mr Wong Kwok-Fai, Raymond [3] IGCO, section 3 provides that “public (公眾、公眾人士) includes any class of the public.” [4] Within section 2 of the Road Traffic Ordinance, Cap.374 (the RTO) (see below) [5] RTO, section 119 provides: “Where any provision of this or any other Ordinance creates an offence an element of which is that the offence occurs on or in relation to a place which is a road or private road, then, without prejudice to the proof of any other elements of that offence, it shall be sufficient in any proceedings alleging the commission of that offence for the prosecution to show that the place where that offence was committed was either a road or private road, without showing that the place was one or the other, and a charge alleging the commission of that offence may be framed accordingly.” [6] Deputy High Court Judge Joseph Yau [7] IGCO, section 2(1) [8] See Shorter Oxford English Dictionary (6th Ed.), Vol.2, p.2595, which includes the following definition: “A path or way between different places, usu. one wide enough for vehicles as well as pedestrians and with a specially prepared surface. Also, the part of such a way intended for vehicles, the roadway”. [9] See also, in this context, the useful discussion in the speech of Lord Clyde in Cutter v Eagle Star Insurance Co Ltd [1998] 4 All ER 417 at pp.422h-423e. [10] By section 15 of the Road Traffic (Amendment) (No.2) Ordinance 1988 (Ord. No.80 of 1988) [11] Judgment §§30-35 [12] Judgment §36 [13] Judgment §§37, 39 & 41 [14] Cons VP, Kempster JA and Penlington J [15] Legislative Council Brief (File Ref.: TRAN 3/10/03 IV) dated 7 May 1986 at §2 [16] A case concerning a Go-Kart (held not to be a motor vehicle) [17] Concerning a motorcycle adapted for scrambling (held not to be a motor vehicle) [18] Concerning a “Go-ped” which was a motorised two-wheeled scooter consisting of a small foot platform on which the rider stood, powered by a 22.5cc engine and with a maximum speed of 20 mph (held to be a motor vehicle) [19] A-G v Ho Kam Mun [1996] 2 HKC 609 (concerning a forklift truck, held not to be a motor vehicle) [20] R v Lai Tung-sang [1991] 1 HKLR 100 and HKSAR v Tse Ping-kwok, unrep., HCMA 915/2006 (15 January 2007) (the former concerning a truck crane and the latter a forklift truck, both held to be motor vehicles) [21] A-G v Ho Kam Mun at p.614H; Percy v Smith [1986] R.T.R. 252 at pp.255H-J, 256B-C & 256J (concerning a forklift truck, held to be a motor vehicle); and Siciliano v Acme Knitters and Dyers Pty Ltd [1994] 1 V.R. 632 at p.636 (also concerning a forklift truck, held not to be a motor vehicle) [22] DPP v Saddington at §18; DPP v King [2008] EWHC 447 at §15 (concerning a “City Mantis” brand electric scooter, capable of up to 10 mph, and described in argument as an executive toy – held to be a motor vehicle) [23] DPP v Saddington at §15 [24] Clark v HM Advocate (2004) J.C. 66 at §19 (concerning an “off road” trail bike, held to be a motor vehicle) [25] Newton v Incorporated Nominal Defendant [1970] V.R. 257 at p.262 (concerning a diesel excavator mounted on caterpillar tracks (i) which could only travel at 4 mph, (ii) the tracks of which would tear up bitumen if it travelled along a bitumen surface, and (iii) parts of which, after travelling about 200 yards under its own power, had to be greased in an operation which took half to three-quarters of an hour – held (unsurprisingly) not to be a motor vehicle) [26] DPP v Saddington at §18; DPP v King at §14 [27] Maclean v McCabe (1964) S.L.T. (Sh. Ct.) 39 at p.40 (concerning a diesel dumper, held not to be a motor vehicle) [28] PW3’s recollection was that it was purchased in around 2005 [29] See also, Grant v McHale (2006) J.C. 81 at §9 [30] Clark v HM Advocate at §19 [31] DPP v Saddington at §15 [32] Clark v HM Advocate at §19 [33] Burns v Currell at p.440; DPP v Saddington at §20; and Coates v CPS [2011] EWHC 2032 at §71 – a case concerning a Segway (held to be a motor vehicle) [34] Grant v McHale (2006) J.C. 81 at §9 (concerning a “mini-moto”, which was a small scale motorcycle the dimensions of which gave rise “to an impression that it was no more than a toy” (see §6) – held to be a motor vehicle) [35] DPP v King at §15 [36] DPP v King at §16 [37] Judgment §§30-35 
Mr Justice Fok PJ: 1. This appeal arises out of the prosecution of the respondent cyclist for the offence of failing to comply with a requirement indicated by a traffic sign. Her conviction by the magistrate[1] was quashed on intermediate appeal to the Court of First Instance.[2] The Appeal Committee[3] granted the prosecution leave to appeal to this court on the basis that the following points of law of great and general importance were involved in the decision of the Court of First Instance, namely: “In the context of the regulatory regime under the Road Traffic Ordinance Cap. 374 and its regulations governing road traffic offences and, in particular, in relation to the offence of failing to comply with a requirement indicated by a traffic sign contrary to Regulations 50(2) and 61(2) of the Road Traffic (Traffic Control) Regulations Cap. 374G (“the Regulations”):- (1) How does the principle of legal certainty operate and what is the extent of its requirements? (2) Whether or not misunderstanding or ignorance of the legal requirement as indicated by a traffic sign - when such requirement is expressly set out in the Regulations and the Road Users’ Code - can give rise to a reasonable excuse under Regulation 61(2) of the Regulations?” 2. As will be seen, it was not sought to be argued that the offence infringes the principle of legal certainty. Rather the focus of the appeal was instead on the nature and scope of the offence of failing to comply with the requirements of a traffic sign and the limits of the statutory defence of reasonable excuse. Specifically, does the meaning of a traffic sign depend on the cyclist’s subjective interpretation and, if not, does her mistaken belief in its meaning constitute a reasonable excuse for failure to comply with the requirement indicated by it? 3. At the conclusion of the hearing, the court allowed the appeal and made the orders set out in Section G below, indicating that our reasons for doing so would be handed down in due course. These are our reasons. A. Background A.1 The charge, location of the alleged offence and the applicable traffic signs 4. The charge was that, on 13 May 2013 at 10.12 am at Yu Tung Road Cycling Track near lamp post AC 1509, the respondent being a person riding a bicycle on a road did, without reasonable excuse, fail to comply with the requirement indicated by a traffic sign of the type shown in Figure No. 155 in Schedule 1 of the Regulations. Figure No. 155 (addressed in greater detail below) is a cycling restriction sign and indicates that cycling is prohibited beyond the sign and that cyclists must dismount and push their bicycle if they wish to proceed beyond the point of the sign. 5. Yu Tung Road in Lantau runs alongside the rear of the Lantau District Police Headquarters and Lantau North Police Station. Between the road and the police station there is a cycleway. The cycleway is intersected by a driveway which leads to the rear of the police station. At the time of the alleged offence, facing a cyclist in each direction on the cycleway, just before the point where the cycleway intersects the driveway, there is a road sign of the Figure No. 155 type and a road marking on the cycleway of the Figure No. 619 type with the word “End” in English and the words “終止” in Chinese. On the other side of the driveway, where the cycleway resumes, there is a road sign of the Figure No. 156 type and another road marking of the Figure No. 619 type.[4] 6. The forms, dimensions, requirements and information of the types of traffic signs (Figure Nos. 155 and 156) and road markings (Figure No. 619) referred to above are depicted in Annex I to this judgment and will be addressed further below. The actual traffic signs in the form of Figure Nos. 155 and 156 are shown in Annex II to this judgment. 7. The alleged offence took place beyond the Figure No. 155 sign which is next to lamp post AC 1509 which is situated alongside the cycleway just before the intersecting driveway. The respondent was stopped by a police sergeant (PW1) when she failed to dismount from her bicycle beyond that sign and continued to ride her bicycle into the driveway and before she had reached the continuation of the cycleway and the Figure No. 156 sign. A.2 The respondent’s conviction by the magistrate 8. The magistrate accepted PW1’s evidence that the traffic sign in question was not covered by leaves or otherwise obscured from sight at the time. The respondent admitted that she failed to dismount from her bicycle when she reached the traffic sign. She accepted in cross-examination that the traffic sign was visible and that she was aware that it was always there. It was her defence that she was honestly and reasonably confused by the sign and thought that it meant cycling was permitted. 9. The magistrate did not accept that the respondent had a reasonable excuse for failing to comply with the requirement to dismount from her bicycle. He was not satisfied that a reasonable person would be honestly and reasonably confused by the traffic sign.[5] He did not find it reasonable for the respondent to make an assumption from the surrounding circumstances of the location as to what the sign meant, rather than by ascertaining its clear meaning from the Road Users’ Code (“the Code”).[6] The magistrate went further and, in effect, disbelieved her defence that she was honestly and reasonably confused by the traffic sign, finding instead that she “willfully neglected” the sign.[7] 10. Accordingly, the magistrate was not satisfied that the evidence adduced by the respondent could constitute evidence capable of raising a reasonable doubt that she acted in the honest and reasonable belief that the circumstances of her conduct were such that, if true, liability would not attach.[8] He therefore found that the prosecution had proved beyond reasonable doubt that the respondent, without reasonable excuse, failed to comply with the requirement indicated by the traffic sign in question and so convicted the respondent.[9] By way of sentence, the magistrate imposed a fine of $500.[10] A.3 The basis on which the conviction was quashed by the judge 11. The judge approached the case differently to the magistrate. He began by identifying the two principal issues in the appeal before him as being: (1) whether the traffic sign in question (i.e. the Figure No. 155 sign) was ambiguous, unclear or confusing; and (2) whether a road user, including the respondent, was obliged in law to know the meaning of traffic signs.[11] 12. Before dealing with the first issue, in his review of the evidence at trial, the judge referred to a defence application, which the magistrate refused, to call an expert witness (Mr Steven Coward) on the issue of whether the design of the sign was ambiguous, unclear or confusing.[12] He also referred to evidence from which he concluded that it was acknowledged by both PW1 and the respondent that the sign was not generally understood by members of public.[13] Whilst upholding the magistrate’s decision to refuse to admit evidence that other road users were confused by the Figure No. 155 sign, the judge considered that part of the expert evidence of Mr Coward on the regulation of cyclists in Hong Kong and abroad would have been of assistance to the court.[14] 13. The judge then considered a complaint by the respondent that certain material,[15] in which it was noted that some traffic signs alongside cycle tracks might be quite confusing and might convey misleading information to cyclists and that traffic signs of the Figure Nos. 155 and 156 types were commonly misinterpreted as meaning the opposite of what they are intended to convey, was not disclosed by the prosecution. The judge held that the failure to disclose the Atkins China Report was a material irregularity since it was relevant material that might well have provided a train of inquiry to admissible evidence. He also held that, whilst not published at the time of trial, it might have been that there was information in the Audit Commission Report available in relation to its findings about the sign that could and should have been disclosed. He concluded that there had been material non-disclosure and that on this ground alone the conviction should be quashed.[16] 14. In light of the non-disclosed materials, the judge considered that Mr Coward’s opinions on cycling regulation in Hong Kong and abroad to be of assistance to the court and admitted his expert opinion on those specific matters into evidence on the appeal.[17] 15. The judge then returned to address the first principal issue he had identified, namely whether traffic signs of the Figure Nos. 155 and 156 types were confusing. He proceeded to consider what he thought the signs indicated to him and considered that road users might be in a state of confusion over the signs. He held, applying MacLeod v Hamilton[18] and R (Oxfordshire County Council) v Bus Lane Adjudicator,[19] that traffic signs should convey a clear and simple message in order to provide adequate information or notice to a road user to enable him appropriately to regulate his conduct. [20] He therefore concluded: “55. I find Figures 155 and 156 to be ambiguous, unclear and confusing. The depictions in the signs are confusing and would benefit by being accompanied with clear language indicating that a cyclist should dismount or resume cycling as the case may be. This has been the practice in other jurisdictions to which Mr Coward has provided various examples.” 16. Having so held, the judge then examined[21] whether the Figure No. 155 sign was sufficiently clear and precise to satisfy the principle of legal certainty laid down in Sunday Times v United Kingdom,[22] applied by this court in Shum Kwok Sher v HKSAR[23] and Winnie Lo v HKSAR.[24] Although he noted he had not been addressed on the principle of legal certainty by the parties, he held that: “58. The principle of legal certainty applies with equal force with a sign that results in an offence being committed if disobeyed. If the sign which is solely graphical does not with sufficient precision convey to the road user what it is that he can or cannot do, then a fundamental requirement of legal certainty is not satisfied. … 62. Whilst the present case is concerned with whether the sign itself is ambiguous or confusing, which I have held it is, then the [respondent] avoids conviction of any contravention for want of legal certainty of the sign.” 17. Finally, the judge addressed the question of whether the respondent had a reasonable excuse for disobeying the sign. He found that it was not reasonable to expect the respondent to find out the meaning of the sign either by being familiar with or accessing the Code on the matter.[25] He thus concluded: “72. The key issue in this case was that the sign was ambiguous and confusing and this was also confirmed by reports of the Transport Department. This would not be corrected by what is stated in the Code for the sign remains ambiguous and confusing. It therefore in my view offended the principle of legal certainty and also gave the basis for the [respondent] to have a reasonable excuse for not obeying it.” 18. Accordingly, he quashed the respondent’s conviction and set aside her sentence. 19. In his decision refusing to certify a point of law of great and general importance, the judge expanded on the application of the principle of legal certainty to the sign in question.[26] He said: “10. The primary consideration is what the sign says. If there is no transference of a clear or understood meaning then the sign may offend the principles of legal certainty. See R (Oxfordshire County Council) v Bus Lane Adjudicator [2010] EWHC 894 (Admin) at paras 62 and 65. This is a contextual matter. A regulatory sign to a certain extent is different from a regulatory statutory provision. A traffic sign is immediate and a person is required at the time to comply with what is depicted visually by the sign at the place where it is erected. It is not an answer to say that an explanation of the sign is provided elsewhere because that does not cure the inherent defectiveness of the sign itself. Otherwise, a traffic sign could depict any form of graphics, even one unrelated to the subject matter it seeks to indicate which would still be valid, according to the respondent’s submission, as long as there was an accessible explanation of it elsewhere.” B. The legislative regime B.1 The context of the statutory offence 20. The statutory provisions relevant to this appeal are addressed in Section B.2 below. It is important, as a starting point, to bear in mind that the context of those provisions is the regulation of road traffic. The Regulations are made under the Road Traffic Ordinance,[27] the long title of which provides that it is “[t]o provide for the regulation of road traffic and the use of vehicles and roads (including private roads) and for other purposes connected therewith.” It is self-evident that the statutory context and purpose of the Ordinance and the Regulations made thereunder is the effective regulation of the use of vehicles and roads for the avoidance or minimisation of the risk of potentially fatal accidents and of injuries and damage to persons and property. It is stating the obvious that one of the principal purposes of traffic signs and road markings is to make use of the road as safe as reasonably practicable. There is an important public interest in this statutory context and purpose.[28] B.2 The relevant statutory provisions 21. Section 11 of the Ordinance provides: “The Secretary [of Transport and Housing] may make regulations to provide for – (a) the classification, design, colour, erection, placing, operation, maintenance, alteration and removal of permanent, temporary and variable traffic signs and road markings; … (b) prohibiting, controlling and restricting vehicular and pedestrian movements by or in relation to traffic signs or road markings; …”. 22. Under section 15: “Any regulation made under section 6, 7, 8, 9, 10, 11, 12, 12A, 121 or 131 may provide that a contravention thereof shall be an offence, and may provide penalties for such offence not exceeding a fine of $15,000 and imprisonment for 9 months.” 23. The Regulations are the regulations made by the Secretary pursuant to his power under section 11 of the Ordinance. A “prescribed traffic sign” is defined in regulation 2 as meaning “a traffic sign of the size, colour and type prescribed in Schedule 1, 3 or 4” and under regulation 3(1): “The Commissioner [for Transport] may cause or, by permit in writing (which permit shall be subject to such conditions as the Commissioner thinks fit and are specified therein), permit to be erected or placed on or near any road a traffic sign, and in the case of a prescribed traffic sign the meaning of that sign shall be in accordance with its content and the note relating to the Figure of that sign in Schedule 1, 3 or 4.” 24. Regulation 50(2) provides: “A person riding a bicycle or tricycle on a road shall comply with the requirement indicated by a traffic sign of the type shown in any of the Figure Nos. 126, 127, 137, 138, 154, 155, 156 or 159 in Schedule 1.” 25. Regulation 61(2) creates an offence in the following terms: “Any person who without reasonable excuse contravenes any of the provisions of, or any requirement under, regulation 10(1), 10A(1), 12(1), 14(5), (6) or (7), 20(4) or (5), 21(4), 23(2), 33(6), 34(2), 36, 39, 41A, 42, 43, 44, 45, 46, 50 or 51 commits an offence and is liable to a fine of $2,000.” 26. Schedule 1 of the Regulations sets out express provisions concerning traffic signs and road markings and includes the dimensions, form and contents of those signs and markings together with a description in words of the relevant regulatory or informatory nature of those signs and markings. Those for Figure Nos. 155, 156 and 619 are reproduced in Annex I to this judgment and the actual signs in the form of Figure Nos. 155 and 156 are shown in Annex II. 27. Section 109 of the Ordinance addresses the Code and provides, among other things, as follows: “(1) The Secretary may cause to be prepared a code (in this section referred to as the road users’ code) comprising such directions as appear to him to be proper for the guidance of persons using roads and may from time to time revise the code by revoking, varying, amending or adding to the directions therein in such manner as he thinks fit. … (5) A failure on the part of any person to observe any provisions of the road users’ code … shall not of itself render that person liable to criminal proceedings of any kind but any such failure may in any proceedings whether civil or criminal and including proceedings for an offence under this Ordinance be relied upon by any party to the proceedings as tending to establish or negative any liability which is in question in those proceedings.” B.3 The act constituting the relevant offence 28. The offence under regulation 61(2) is the contravention, without reasonable excuse, of the provisions of various regulations, including regulation 50(2). Regulation 50(2) imposes on cyclists a duty to “comply with the requirement indicated by a traffic sign of the type shown in any of the Figure Nos. 126, 127, 137, 138, 154, 155, 156 or 159 in Schedule 1”. A traffic sign of those types is a prescribed traffic sign and, by regulation 3(1), “the meaning of that sign shall be in accordance with its content and the note relating to the Figure of that sign in Schedule 1”. The actus reus of the offence is therefore the cyclist’s act of failing to comply with the requirement indicated by the relevant traffic sign and in the case of Figure No. 155 that is the act of failing to dismount when proceeding beyond the point of the sign. 29. It is important to note that the focus of the statutory offence is on compliance with the requirement indicated by the traffic sign, being a prescribed traffic sign, which requirement is “in accordance with its content and the note relating to the Figure of that sign in [the relevant] Schedule”. In the case of a traffic sign in the form of Figure No. 155, that requirement is expressly set forth in Schedule 1 of the Regulations and it is in clear and unambiguous terms: if a cyclist wishes to proceed beyond the point of the sign, he must dismount and push his cycle on foot. 30. The offence is not a failure to comply with a traffic sign but with the requirement that the traffic sign indicates. The interpretation of the sign is not dependent on the cyclist’s subjective appreciation or non-appreciation of the meaning conveyed by it. Instead, the interpretation of the sign is approached taking in the meaning assigned to it by the Schedule. It is therefore no part of the offence that the sign alone must be interpreted, either by the cyclist or by a court, in order to determine what it is that the cyclist should or should not do. For the reasons addressed below (in section D.2), the respondent’s contention that interpretation of the meaning of the sign on its own and without reference to the Schedule is rejected. B.4 The mental element of the offence 31. Failure to comply with the requirements of a relevant traffic sign does not inevitably lead to conviction for the offence under regulations 50(2) and 61(2). The offence is only constituted by the relevant act committed “without reasonable excuse”. This is a statutory defence expressly provided for within regulation 61(2). 32. It was not suggested in argument by either the prosecution or the respondent that the presumption of mens rea persists so that the prosecution must prove knowledge, intention and recklessness in relation to the constituent elements of the offence, specifically the cyclist’s failure to comply with the requirement of the relevant traffic sign. There is no need to prove knowledge of requirement of the traffic sign and intention to disobey that requirement. As a matter of statutory construction, it is plain that the offence, being a regulatory offence, is one in respect of which the presumption of mens rea has been displaced and is intended to be subject only to the statutory defence expressly provided for. 33. In the parties’ respective printed cases for the appeal, there was a difference as to whether the appropriate mental requirement of the offence is that falling within the second, third or fourth alternatives set out in this court’s judgments in Hin Lin Yee v HKSAR[29] and Kulemesin v HKSAR.[30] As will be recalled:- (1) The second alternative is where there is evidence capable of raising a reasonable doubt that the accused acted or omitted to act in the honest and reasonable belief that the circumstances or likely consequences of his conduct were such that, if true, liability would not attach, in which event the accused must be acquitted unless the prosecution proves beyond reasonable doubt the absence of such exculpatory belief or that there were no reasonable grounds for such belief.[31] (2) The third alternative is where the accused has a good defence if he can prove on the balance of probabilities that he acted or omitted to act in the honest and reasonable belief that the circumstances or likely consequences of his conduct were such, if true, he would not be guilty of the offence. (3) The difference between these two alternatives is that, in the second, the prosecution bears the burden (beyond reasonable doubt) of proving the absence of the exculpatory belief whereas, in the third, there is a reverse onus on the accused (the burden being on a balance of probabilities) to prove the exculpatory belief. (4) The fourth alternative is where the accused is confined to relying on the statutory defence expressly provided for, the existence of such defence being inconsistent with the second and third alternatives. 34. The statutory defence of reasonable excuse is potentially wider than the halfway house defence of honest and reasonable belief since (as discussed below) a person may have a reasonable excuse for failing to comply with the requirement of a traffic sign, even though he deliberately intends that non-compliance. For that reason, the appropriate alternative mental requirement of the offence under regulations 50(2) and 61(2) is the fourth alternative. The onus of establishing a reasonable excuse rests on the accused[32] (the burden being on a balance of probabilities). 35. In any event, insofar as the appropriate alternative mental requirement of the offence were one of the second or third of the restated Hin Lin Yee alternatives, the appropriate alternative for this offence would be the third alternative, under which the onus of proving the defence of honest and reasonable belief is on the accused. This is because the offence is not a serious offence carrying severe penalties[33] and a prosecutorial burden to disprove honest and reasonable belief would be inconsistent with the statutory defence of reasonable excuse under which the burden is clearly on the accused. B.5 The defence of reasonable excuse 36. The expression “without reasonable excuse” occurs in various statutory contexts. A consideration of the defence involves looking to three matters. First, self-evidently, the matters said to constitute reasonable excuse must be identified. Secondly, the court will then examine whether the excuse is genuine, since the reason asserted for departing from a relevant prescription must be the real reason for doing so. Thirdly, the court must make an assessment of whether that excuse is reasonable, which the court will do on an objective standard depending on the particular facts of the case.[34] 37. In determining whether an excuse is reasonable or not, it will be relevant to have regard to the context in which the defence of reasonable excuse arises, since that context may suggest either a narrow or wide range of circumstances that might constitute a reasonable excuse. For example, the range of circumstances in which there is a reasonable excuse for failing to provide a sample of blood or urine in the context of the laws against driving under the influence of drink has been held to be narrow, since the circumstances giving rise to the offence are always essentially similar so that what might be a reasonable excuse for committing it can be envisaged.[35] In other contexts, the defence may be construed more widely and the question of whether or not an excuse is reasonable will be determined in the light of the particular facts and circumstances of the individual case.[36] 38. In the present context, the offence of failing to comply with the requirements of a traffic sign can be committed in a large variety of circumstances and for many different reasons. There is therefore no reason, as a matter of context, to hold that only a narrow range of circumstances can constitute a reasonable excuse under regulation 61(2). Instead, for the purposes of that regulation, the concept should be regarded as potentially broad. Thus, a reasonable excuse for failure to comply with the requirement of the Figure No. 155 sign might include where the failure is inadvertent because the sign has become obscured from view by surrounding foliage. It might also include where the failure is deliberate but has occurred because the cyclist does so to avoid or prevent an accident or to escape an assailant. These examples are, however, only illustrative and it is not intended to identify an exhaustive list of the circumstances that could constitute a reasonable excuse for failure to comply. 39. A cyclist’s belief as to the meaning of a prescribed traffic sign may be relevant to the determination of the defence of reasonable excuse. It will certainly be relevant to the genuineness of an asserted excuse that the cyclist thought that he was in fact complying with the requirement of the traffic sign. It may also be relevant to the question of whether the asserted excuse was reasonable, although the court may accept the cyclist believed he was complying with the requirement of the traffic sign but nevertheless hold that, objectively, this was unreasonable in all the circumstances.[37] 40. The reasons for concluding that the excuse relied on by the respondent in the present case was not reasonable are discussed in Section D.3 below. C. The principle of legal certainty 41. The principle of legal certainty requires that “[a] criminal offence must be so clearly defined in law that it is accessible and formulated with sufficient precision to enable the citizen to foresee, if need be with appropriate advice, whether his course of conduct is lawful or unlawful”.[38] The principle is derived from the majority judgment in Sunday Times v United Kingdom[39] and has been considered by this court in a number of cases.[40] 42. Once it is concluded (see Section B.3 above) that the correct meaning of a prescribed traffic sign is that set out in the relevant Schedule of the Regulations, it is immediately apparent from the statement of the principle in the preceding paragraph that there can be no basis for suggesting that the offence in the present case offends the principle. It cannot sensibly be contended that, so construed, the law in question is otherwise than adequately accessible and sufficiently precise. Reference to the Regulations, publicly available as part of the Laws of Hong Kong published in various forms (including print and online versions), provides a clear and unambiguous explanation of the requirement of the prescribed traffic signs, including that of the Figure No. 155 type. So too does reference to the Code.[41] The statutory defence of reasonable excuse is well-recognised in the law and, notwithstanding that it must be considered contextually and in the light of all relevant circumstances, clearly sufficiently precise to satisfy the principle of legal certainty. 43. Contrary to the suggestion in the judge’s decision refusing a certificate (see paragraph 19 above), the judge was, with respect, wrong to suggest that the meaning of the traffic sign (rather than the meaning of the law) must be immediately clear if the principle of legal certainty is not to be infringed. In any event, the principle of legal certainty does not mean that the content of a legal rule must be immediately apparent, since this would be inconsistent with the concept of adequate accessibility, which recognises that appropriate advice may be required for an individual to be informed of the content of the law in question. 44. In the circumstances, it is unsurprising that it was common ground between the parties that the application of the principle of legal certainty was not the relevant issue in this case. Instead, the real issue related to the nature of the offence in question and the ambit of what might or might not constitute a reasonable excuse for non-compliance with the traffic sign in question. D. The contentions on appeal D.1 The correct starting point 45. The correct starting point for a consideration of the offence in question is the legislative scheme, which is addressed in Section B above. As stated in Section B.3, the interpretation of the sign on its own and without reference to its meaning indicated in the Schedule is not a matter for the cyclist or the court. However, in the Court of First Instance, the judge identified the first issue before him as being the question of whether the sign was ambiguous, unclear or confusing. Mr Nigel Kat SC[42] submitted that the judge was correct to have done so and, on that basis, he sought to rely on the judge’s finding that the Figure No. 155 sign is ambiguous, unclear and confusing. D.2 Not necessary to consider the meaning of the sign separately to its meaning as set out in Schedule 1 of the Regulations 46. Mr Kat submitted that the judge’s approach was correct as a matter of construction of regulation 3(1) of the Regulations, which is set out above. He submitted that the requirement of that regulation was that the meaning of a traffic sign shall be in accordance with (i) its content and, separately, (ii) the note relating to it in Schedule 1. Thus, Mr Kat submitted that it was a matter for the court to determine the meaning of the traffic sign by reference to its content. If that content is not clear and unambiguous on its own, it is not open to the court to convict a cyclist for failing to comply with it. 47. We reject that construction of regulation 3(1). Mr Kat’s submission involves reading the words “in Schedule 1, 3 or 4” in the phrase “the meaning of that sign shall be in accordance with its content and the note relating to the Figure of that sign in Schedule 1, 3 or 4” as being solely referable to “the note relating to the Figure of that sign” and not qualifying “its content”. That is an artificial and strained construction of regulation 3(1). As a matter of structure, regulation 3(1) does two things: first, it gives power to the Commissioner to erect and place traffic signs; secondly, it provides that “in the case of a prescribed traffic sign” its meaning will be in accordance with “its content and the note relating to the Figure of that sign in Schedule 1, 3 or 4”. As a matter of language, it is more natural to read “in Schedule 1, 3 or 4” as qualifying both “its content” and “the note relating to the Figure of that sign”, and not merely the latter phrase. 48. That construction is all the more apparent when one considers regulation 3(1) in context. Schedule 1 sets out the form and content of various prescribed traffic signs, including their precise dimensions and colours, as well as a description in words of the requirement or restriction thereby indicated. There are a number of signs specified in Schedule 1, the graphical content of which on their own do not readily indicate their meaning but, when read together with the accompanying note, their meanings are abundantly clear. For example, Figure No.111 shows the outline of a person in front of the front of a vehicle. The graphical content of this sign (which is provided for in Schedule 1) is not immediately apparent. It is only with the benefit of the description in the Schedule, which indicates that the sign denotes the commencement of a “pedestrian priority zone” in which vehicles are to give way to pedestrians, that the meaning becomes clear. Another example is Figure No. 140, which is simply a red cross in a red border with a blue background. Without reference to the note in Schedule 1, it is not apparent that the sign indicates the requirement “no stopping”. 49. Mr Kat sought to support his submission of regulation 3(1) by reference to section 110 of the Ordinance, which provides: “A traffic sign or road marking which is placed on or in a road and is similar to a prescribed traffic sign or prescribed road marking, as the case may be, shall be deemed to be a prescribed traffic sign or prescribed road marking and to have been lawfully so placed unless the contrary is proved, and the fact that a traffic sign or road marking differs slightly in positioning, size, colour or type from a prescribed traffic sign or prescribed road marking shall not prevent such traffic sign or road marking from being a prescribed traffic sign or prescribed road marking, as the case may be, so long as it is visible and the general appearance and meaning of the sign or marking is not thereby materially impaired.” The argument was that this section requires the court to make a value judgment as to the meaning of a traffic sign since a sign will cease to be a traffic sign if its general appearance and meaning is materially impaired by a variation in its form. Thus, the importance of the meaning of a sign itself is emphasised in the legislation. 50. However, this argument misconstrues the purpose and effect of section 110. That section relates to variations in traffic signs and road markings and its purpose is to deem a sign which is “similar to” a prescribed sign to be a prescribed sign and therefore to have the meaning which is given to it in the Regulations. One need only have regard to the meaning of the sign conveyed by its content where a variation to a prescribed sign has been used. In those circumstances, one must ask whether the variation materially impairs its meaning so that it should not be permitted to be deemed to be a prescribed sign with its prescribed meaning. That inquiry is not engaged in the present case. But, more importantly, reliance on this section does not afford any support for Mr Kat’s construction of regulation 3(1). 51. Mr Kat also sought to support his construction of regulation 3(1) by reference to the speech of the Secretary for Transport when moving the Bill which became the RTO in the Legislative Council. He submitted that the effect of the Secretary’s speech was that, in order to ensure that Hong Kong’s traffic signs would give road users adequate notice of what is required of them, the signs would conform to simple, established and internationally recognised norms and that the system of signage implemented under the Regulations would reflect UK and international practice (subject to necessary variations such as for rickshaws).[43] 52. It is reading too much into that statement promoting the RTO to treat this as requiring the artificial and strained construction of regulation 3(1) advocated on behalf of the respondent. The proper construction of regulation 3(1) set out above is clear as a matter of construction of the statutory language in the light of its context and purpose (as to which see paragraph 20 above) and nothing said by the Secretary for Transport suggests that another construction is to be preferred. D.3 Respondent failed to comply “without reasonable excuse” 53. It was contended on behalf of the respondent that her failure to comply with the requirement of the Figure No. 155 sign in question was with reasonable excuse because she was confused by the sign. Her case was therefore that she honestly believed that the sign meant that she could ride her bicycle at the point where she was intercepted by the police sergeant. Although the magistrate’s finding that the respondent “willfully neglected” the sign might suggest that he did not accept that the respondent honestly so believed, it is questionable whether there was a proper basis for him to reject her evidence in this regard. 54. Taking the respondent’s excuse, therefore, to be genuine and her belief to be honestly held, the reasonableness of this excuse falls to be assessed in the context of the offence and in the light of the correct answer to the question of whether the meaning of the sign is to be determined by reference to its content alone or also by reference to the note in the Schedule explaining its requirement. Given that, as already discussed above (in Sections B.3 and D.2), the prescribed meaning of the Figure No. 155 sign is that set out in Schedule 1 and not as determined either by the cyclist or by the court by reference to the sign alone, there is no room, objectively, to conclude that the respondent’s honestly held but mistaken belief as to the meaning of the sign was reasonable. 55. The judge’s conclusion to the contrary is tainted by his erroneous approach of seeking to interpret the meaning of the traffic sign in question in isolation and without reference to its clear and unambiguous meaning set out in Schedule 1. This was not a case in which the respondent claimed that she was not aware of the existence of the sign (for example because it was covered by leaves) or that she had good reason to deliberately fail to comply with it (for example because she was fleeing from an assailant). She simply misunderstood it. 56. Her misunderstanding was not, however, reasonable. As already noted, there are a number of traffic signs, the meanings of which are not immediately apparent merely from looking at them. Different people may have different views as to the meaning of those signs. It would be a recipe for traffic chaos if an honest belief that those signs had different meanings to those set out in the relevant Schedule of the Regulations could constitute a reasonable excuse for not complying with their requirements. In any event, the meaning of a prescribed traffic sign being that set out in the relevant Schedule of the Regulations and reproduced in the Code, it is a simple matter for any cyclist to ascertain the meaning of any applicable traffic sign. All users of the road have a responsibility to familiarise themselves with the meaning of traffic signs and road markings, since this is important not only in terms of compliance with the Ordinance and Regulations, but also from the perspective of road safety. This responsibility is also reflected in the treatment of the Code in section 109(5) of the Ordinance. It is, in these circumstances, unreasonable for a cyclist to rely on a mistaken belief as to the meaning of a particular sign as a reasonable excuse. This is simply an application of the general principle of law that ignorance of the law is not a defence, which has been applied in the context of the defence of reasonable excuse.[44] 57. In an argument that ranged over circumstances that included some that were not present in this case, Mr Kat submitted, to the contrary, that the respondent’s mistaken belief was reasonable. He relied on the fact that other cyclists had not complied with the Figure No. 155 sign, that the judge had found the sign to be “ambiguous, unclear and confusing”, that the driveway intersecting the cycleway where the sign was placed was disused so that the sign was redundant, and that there was no evidence that the respondent posed any risk to any other road user when she cycled beyond the point of the Figure No. 155 sign. He gave, as an example, a motorist confronted by a traffic light at 3 am which was stuck on red who carefully proceeds after checking that there is no traffic coming against the light. 58. These matters do not assist the respondent. The fact that other cyclists may have failed to comply with the requirements of the Figure No. 155 sign at the particular location in question may reflect a deliberate choice on their part to disobey the traffic sign. Even if it shows that other cyclists were confused, this on its own cannot make the mistaken understanding of the sign reasonable. The judge’s finding that the sign was “ambiguous, unclear and confusing” is, as already noted, tainted by his erroneous analysis of the statutory scheme and, as will be discussed below, somewhat surprising. It was not the respondent’s evidence at trial that she failed to comply with the traffic sign because she thought it was redundant or that her riding would not pose any risk to any other road user. In any event, it is not for the court to determine the reasonableness of the placement of the traffic sign, which might be a matter for a judicial review (if proper grounds existed for such a challenge, which was not suggested to be the case here), and the absence of risk to other road users by the commission of the offence is a matter in mitigation of penalty rather than a reasonable excuse for non-compliance. The traffic light stuck on red at 3am is simply an example of a malfunctioning light, which simply provides another example of a type of circumstance which might constitute a reasonable excuse for non-compliance. 59. In support of reasonable excuse on the part of the respondent, Mr Kat relied on the case of R v Murray-Jones Paul.[45] In that case, the defendant’s conviction for failing to comply with the requirement of a road marking (indicating a bus lane) was quashed on the basis that the relevant road marking was not noticeable because of the conditions prevailing and, furthermore, was inconsistent with road signs placed at the location of his alleged offence which contradicted the road marking.[46] Although there is no doubt each of these excuses for non-compliance could properly amount to reasonable excuse,[47] neither of these contentions was raised by the respondent in the present case so the decision does not assist the respondent. 60. Mr Kat also submitted that reference to the Code was not reasonable since a perusal of the section on cycleways (at p.34) indicates that a cyclist will usually have to join or return to the normal road at the end of a cycleway and that a cycle route may cross a road and the cyclist may ride across the road. The respondent did not suggest these statements in the Code led her to be confused by the traffic sign. But, in any event, such confusion would not have been reasonable since those statements were not definitive and the requirements of the Figure No. 155 sign are clearly indicated elsewhere in the Code (at pp.30 and 101) which must be read as a whole. Mr Kat’s submission that cyclists might include those who were illiterate or were of children of young age, who would not therefore read the Code, is simply irrelevant on the facts of this appeal. 61. Mr Kat relied on an appendix in the evidence of Mr Coward setting out a comparison of various cycling signs in different jurisdictions around the world. He pointed to the fact that in other jurisdictions, where a sign indicating that cyclists must dismount is depicted by an ideogram (i.e. graphic symbol) it is usually accompanied by words such as “cyclists dismount”. That may be so, and the conclusion from this is that it may be possible to make a clearer traffic sign, but it does not follow from this that it is reasonable for a cyclist to ignore the prescribed meaning of the Figure No. 155 sign set out in the Schedule. 62. In sum, the meaning of the Figure No. 155 traffic sign is not for the cyclist to assume and it is not a matter of interpretation by the cyclist of the sign itself divorced from its meaning stipulated in Schedule 1. A belief, albeit honestly held, that the sign indicates that cycling is permitted, is not a reasonable excuse given the readily available means of ascertaining its meaning from the Schedule and the Code. E. An analysis of the judge’s reasoning 63. The basis on which the conviction was set aside by the judge is set out in Section A.3 above. With respect, he was in error in identifying the first issue in the appeal before him as being whether the traffic sign in question was ambiguous, unclear or confusing. For the reasons set out above, that was not the correct starting point and it was not a matter for the court to interpret the meaning of the sign rather than to look to its prescribed meaning in Schedule 1. 64. Although the judge explained his reasons for finding that the sign in question was “ambiguous, unclear and confusing” and that finding is not the subject of this appeal, it is a finding that is tainted by the incorrect approach he adopted. In any event, it is a finding of fact which, with respect, is somewhat surprising. It is difficult to see how an ideogram of a person with one hand on the handlebar of a bicycle and the other hand on its seat with the lower limbs in the attitude of a person walking (and not showing the person’s feet on the pedals of the bicycle) could be said to introduce ambiguity or confusion. But for present purposes it is sufficient to hold that it was not a matter for the judge to determine the meaning of the sign. 65. For that reason, the judge’s reliance on MacLeod v Hamilton and R (Oxfordshire County Council) v Bus Lane Adjudicator were, with respect, misplaced. Those cases involved statutory schemes each of which required the court to consider whether the traffic signs in place gave “adequate information as to the effect of the [traffic regulation] order” to road users. That imposed a duty on the part of the relevant authority to publicise the traffic regulation order by placing signs that provided adequate information as to the effect of the order. Although there are provisions in the Ordinance that similarly involve a consideration of the adequacy of signage,[48] that is not the legislative scheme here in relation to regulations 50(2) and 61(2). 66. Whilst the non-disclosed materials might have been relevant to the genuineness of the respondent’s excuse that she was confused by the sign, their actual relevance would have depended on whether the genuineness of that excuse was put in issue by the prosecution. Since it is not clear that this was conceded by the prosecution, those materials might have been relevant to that extent. However, they were not relevant to the reasonableness of the asserted excuse since, at most, the non-disclosed materials merely showed that other cyclists entertained the same mistaken belief about the meaning of the sign. The expert evidence concerning the regulation of cyclists in Hong Kong and abroad might be relevant to the exercise of determining whether other signs might be used to indicate a requirement to dismount but that was not the issue before the judge. 67. As stated above, the issue of the meaning of the traffic sign not being a matter for the judge, the principle of legal certainty simply did not arise as an issue and the judge erred in holding that it was engaged in the context of the present case. Neither party had raised this issue before him and, with respect, the judge erred in holding that the principle of legal certainty was not satisfied in relation to the Figure No. 155 sign. 68. Rather than the meaning of the sign in question, the critical question in this case was whether the respondent had demonstrated a reasonable excuse for failing to comply with the requirements of that sign. Her only asserted excuse was an honest but mistaken belief as to its meaning. For the reasons set out above, this was not a reasonable excuse and the judge was, with respect, wrong to reach a contrary conclusion. F. Answers to the certified questions 69. Insofar as necessary, in view of what is set out above, we would answer the certified questions as follows: (1) As to question (1), see Section C above. The principle of legal certainty is clearly satisfied in relation to this offence (and other similar offences under the Regulations) because the meanings of the prescribed traffic signs are “adequately accessible and formulated with sufficient precision to enable the citizen to foresee, to a degree that is reasonable in the circumstances, the consequence which a given action may entail so that he may regulate his conduct”. (2) As to question (2), see Section D.3 above. A mistaken albeit honest belief as to the meaning of a traffic sign will not, where its meaning is expressly stated in the Regulations and the Schedules thereto and also consistently stated in the Code, constitute a reasonable excuse for failure to comply with the requirement indicated by it. G. Conclusion and orders 70. For these reasons, we allowed the appeal. 71. In the light of this outcome and in view of the minor nature of the offence and the penalty imposed on her, Mr William Tam SC, Deputy Director of Public Prosecutions,[49] sensibly indicated that he did not press for the restoration of the conviction and so we did not order that it be restored. 72. As to costs, the respondent being legally aided, we made no order as to the costs in this court and in the Court of First Instance. 73. As to the costs of the trial, we ordered that any submissions as to why the order in favour of the respondent made by the judge should not be set aside be made in writing. We now direct the filing of submissions on behalf of the respondent in respect of that matter within 14 days of the date of this judgment, with any reply thereto by the appellant within 14 days thereafter. Mr William Tam SC, DDPP and Mr Ivan Cheung PP, of the Department of 	Justice, for the Appellant Mr Nigel Kat SC and Mr Azan Marwah, instructed by Vidler & Co., assigned 	by the Director of Legal Aid, for the Respondent Annex I AnnexII Actual sign of Figure No. 155 type: Actual sign of Figure No. 156 type: [1] In TWS 12847/2013, before Mr Andrew T.C. Mok, Deputy Special Magistrate (Statement of Findings dated 5 May 2014 (“SOF”)). [2] In HCMA 280/2014, before Zervos J (Judgment dated 20 April 2015 (“CFI Judgment”). [3] Ma CJ, Ribeiro & Tang PJJ (Determination dated 18 August 2015). [4] In the respondent’s case, there was reference to previous inconsistent signage at the location in question, namely a Figure No. 155 sign facing cyclists on the other side of the driveway from the end of the cycleway adjacent to lamp post AC 1509. It was inconsistent in that it suggested a second requirement to dismount just before the resumption of the cycleway. However, there was no evidence at the trial that the respondent believed that this inconsistent signage was in place at the time of the alleged offence. In fact, her evidence was that she paid attention to the sign on the other side of the driveway which was the (consistent) Figure No. 156 sign, so the previous inconsistent signage was irrelevant to the appeal. [5] SOF at [28]. [6] SOF at [32]. [7] SOF at [36]. [8] Ibid. at [39]; this being a reference to the alternative mens rea regime applicable where a statutory offence is silent or ambiguous as to its mental requirements (discussed further below). [9] Ibid. at [40]. [10] Ibid. at [43]. [11] CFI Judgment at [4]. [12] Ibid. at [19]. [13] Ibid. at [23]. [14] Ibid. at [23] and [24]. [15] Namely a Final Report by Atkins China Ltd dated March 2013 and an Audit Commission Report dated 30 October 2014. [16] [39]-[40]. [17] Ibid. at [41], [47] and [48]. [18] [1965] SLT 305 per Lord Clyde at 308. [19] [2010] EWHC 894 (Admin) per Beatson J at [65]. [20] CFI Judgment at [50]-[54]. [21] Ibid. at [56]-[57]. [22] (1979-1980) 2 EHRR 245 at [49]. [23] (2002) 5 HKCFAR 381 per Sir Anthony Mason NPJ at [63]. [24] (2012) 15 HKCFAR 16 per Bokhary PJ at [18]-[21] and per Ribeiro PJ at [71]-[77]. [25] Ibid. at [63] and [69]. [26] HCMA 280/2014 (Decision dated 10 June 2015). [27] (Cap.374) (“the Ordinance”). [28] Secretary for Justice v Latker [2009] 2 HKC 100 per Ma CJHC (as Ma CJ then was) at [41]. [29] (2010) 13 HKCFAR 142 (“Hin Lin Yee”). [30] (2013) 16 HKCFAR 195 (“Kulemesin”). [31] Hin Lin Yee at [96] and Kulemesin at [41] and [83]. [32] Criminal Procedure Ordinance (Cap.221), s.94A. [33] See the discussion in Kulemesin at [87]-[89]. [34] HKSAR v Adams Secuforce (International) Ltd [2008] 1 HKLRD 207 at [10]; Securities and Futures Commission v Ernst & Young [2015] 5 HKLRD 293 at [102]-[103]. [35] R v Lennard [1973] 1 WLR 483 at 487G-488B; applied in HKSAR v Poon Kin Bong [2006] 3 HKC 421 at [31], [33]; and discussed in R v G [2010] 1 AC 43 at [81]. [36] R v G (supra) at [81]. [37] See, in a very different context, the case of R v Unah [2012] 1 WLR 545 at [5]-[6]. [38] Mo Yuk Ping v HKSAR (2007) 10 HKCFAR 386 at [61]. [39] (1979-1980) 2 EHRR 245. [40] Shum Kwok Sher v HKSAR (supra) at [63] (offence of misconduct in public office held not to offend principle); Lau Wai Wo v HKSAR (2003) 6 HKCFAR 624 at [36] (application of principle to bind-over orders); Leung Kwok Hung & Others v HKSAR (2005) 8 HKCFAR 229 at [27] (principle mandated by the expression “prescribed by law”); Mo Yuk Ping v HKSAR (supra) (offence of conspiracy to defraud held not to offend principle); Winnie Lo v HKSAR (supra) at [19]-[21] and [71]-[77] (offences of maintenance and champerty held not to offend principle). [41] See Chapter 4 at pp.26 and 30, and Chapter 8 at p.101 of the Code (May 2000 Edition), also available online at http://www.td.gov.hk/en/road_safety/road_users_code/index.html. [42] Appearing, with Mr Azan Marwah, on behalf of the respondent. [43] Respondent’s Case at [33], referring to Hong Kong Hansard (28 July 1982) at p.1109. [44] See: Johnson v Youden [1950] 1 KB 544 at 546; and R v Jones [1995] QB 235 at 242. [45] HCMA 1189/1988, unrep., Hooper J, Judgment dated 10 January 1989. [46] Ibid. at [27], [36], [41]-[43]. [47] See also, e.g. as to inconsistency between a road marking and a traffic sign, HKSAR v Lai Tai Kwong, HCMA 917/1986, unrep., Hopkinson J, Judgment dated 2 October 1986. [48] See sections 40(3) and 40(6) of the Ordinance concerning variations of speed limits and the requirement for speed limit signage to “give adequate guidance or notice to drivers” as to the applicable speed limit. [49] Appearing with Mr Ivan Cheung, PP, on behalf of the appellant. 
Chief Justice Cheung: 1. Three times between 2010 and 2018, the appellant was tried and convicted after trial of murdering his cohabitee Madam Yeung Sau Yu on 12 September 2009. On each occasion, the conviction was overturned on appeal – twice by the Court of Appeal[1] and once by this Court[2] (reversing the Court of Appeal’s dismissal of the appellant’s application for leave to appeal[3]) – and a retrial was ordered. This appeal arose out of the last conviction of the appellant for murder on 26 April 2018. The conviction was unanimously quashed by the Court of Appeal on 20 November 2020[4]. By a majority decision (Macrae VP and Zervos JA, McWalters JA dissenting), the Court of Appeal ordered a third retrial. On 29 March 2021, the Appeal Committee gave leave to appeal against the order for retrial[5], and thus this appeal. The facts 2. There is no dispute that the appellant killed the deceased on 12 September 2009 in the flat where they cohabited and that he had done so by chopping her with a knife approximately 213 times. The appellant denied the charge of murder but was prepared to plead guilty to manslaughter by reason of provocation. At his third trial, a further defence of diminished responsibility, supported by newly obtained psychiatric evidence, was also run. 3. There was no eye witness to the killing. Only the appellant and the deceased were present inside the flat at the time. The defence case on provocation was essentially based on two video records of interview made by the police shortly after the killing, the voluntariness of which was admitted, as well as the appellant’s own evidence at trial. Essentially, his case on provocation was that the deceased was having an affair with another man (who gave evidence as prosecution witness). Although the appellant was not sure about it at the time, he had a strong suspicion about the deceased’s infidelity. Used condoms were on several occasions found by the appellant in the rubbish bin at the rear staircase outside the flat, and on the last time that happened, that is, in the afternoon of 11 September 2009, the appellant confronted the deceased with a used condom which he found in the rubbish bin, which he said was still warm. That led to heated quarrels between the two that same evening and the following morning during which, according to the appellant, the deceased taunted him about becoming a garbage collector, his poor performance as a sexual partner, and that he was not his daughter’s father and she had been fathered by a man who had an affair with his ex‑wife. According to the deceased’s ex‑husband, the deceased had called him on his mobile at 5:40 am and told him that the appellant had gone crazy, taken off all his clothing and attempted to jump off the building, and she was scared. According to the appellant, after that telephone call, the deceased twice thrust the used condom into his face near his mouth, asking him to eat it so that he would stop asking her about her new boyfriend. The appellant claimed that he lost control and his mind went blank. He fetched a knife from the kitchen, attacked the deceased frenziedly and killed her. That happened between 6 and 7 am. The first trial and appeal 4. The appellant’s first trial took place before Beeson J sitting with a jury in October 2010. On 21 October 2010, he was convicted of murder by the unanimous verdict of the jury. On 30 October 2013, the Court of Appeal allowed his appeal and quashed the conviction because impermissible hearsay evidence was adduced at trial, with no objection from defence counsel who somehow considered it advantageous to his case. The Court took the view that given the degree to which the inadmissible evidence went to undermine the defence of provocation, it could hardly be said that the conviction for murder was safe or satisfactory. The Court of Appeal ordered a retrial. The second trial and appeals 5. The retrial took place before V Bokhary J and a jury in March 2014, and again the appellant was convicted of murder by the unanimous verdict of the jury on 31 March 2014. The sole issue at trial was, as before, whether he had been provoked by the deceased. He again pleaded guilty to manslaughter on the basis of provocation which was not accepted by the prosecution. 6. On 30 June 2015, the Court of Appeal, by a majority (Lunn VP and Macrae JA, McWalters JA dissenting on the gravity of the provocation ground), refused leave to appeal against the conviction. The appellant’s further appeal to this Court was, however, successful. In its judgment dated 7 February 2017, this Court allowed the appeal and quashed the conviction upon a construction of the phrase “do as he did” in section 4 of the Homicide Ordinance (Cap 339)[6], and held that it had not been dealt with accurately and adequately by the trial judge in her directions to the jury. This Court ordered a second retrial. The third trial 7. The appellant’s third trial before Anthea Pang J and a jury commenced on 8 August 2017. It was aborted shortly afterwards because of adverse media reporting of the case. Following a series of pre‑trial reviews, in which for the first time the defence indicated its intention to rely on psychiatric evidence to run a new defence of diminished responsibility, the third trial commenced on 10 April 2018 before Maggie Poon J (as she then was) sitting with a jury. The defence of diminished responsibility, run alongside the original defence of provocation, was supported by a psychiatric report prepared by Dr Wong Chung Kwong who had seen the appellant on 30 June 2017. The prosecution called Dr Oliver Chan, who had seen the appellant on 2 August 2017 and prepared a psychiatric report dated 22 April 2018, in rebuttal. According to Dr Wong, the appellant was suffering from an abnormality of the mind arising from the combined effects of two psychiatric disorders, namely a Major Depressive Disorder (“MDD”) of mild to moderate severity, and Acute Dissociative Reactions to Stressful Events (“ADR”) which was a brief, acute and serious disorder, coupled with four underlying factors, namely, hypoglycaemia, sleep deprivation, alcohol and emotional memory of his ex‑wife’s adultery. Dr Wong took the view that at the time of the killing, the appellant’s abnormality of mind due to these two psychiatric disorders plus the four underlying factors severely compromised his mental capacity to think and judge rationally, and it impaired his ability to control his emotion, impulse and behaviour. Dr Chan disagreed with Dr Wong’s finding that the appellant had been suffering from MDD at the time of the killing. Instead he found that he had been suffering from an adjustment disorder. Dr Chan accepted that the appellant’s discovery of the deceased’s infidelity had been a major life stressor in the two months preceding the killing. He took the view that an adjustment disorder might be considered as an abnormality of mind, but considered that the appellant’s condition did not substantially impair his mental responsibility for his act, and could not provide a causal explanation for the killing. However, Dr Chan agreed that the appellant might have been on a highly emotional roller coaster ride just prior to the killing, and that he went into a dissociative state during the attack. 8. Before counsel’s closing addresses and the judge’s summing up, the judge prepared draft written directions to the jury, setting out the law and issues relating to murder, provocation and diminished responsibility, and invited counsel to comment on them, which counsel did. In short, the draft directions referred to the psychiatric evidence in the context of diminished responsibility but not provocation. From the comments made by the appellant’s leading counsel, Mr Andrew Bruce SC, to the judge on the draft directions, it is clear that on the issue of provocation, the only concern leading counsel had was in respect of factual provoking conduct, and the expert evidence on the appellant’s mental health was seen as relevant to the special characteristics of the appellant in considering the “do as he did” element of provocation only. No mention was made by Mr Bruce of the relevance of the psychiatric evidence to the question of whether the appellant had lost his self‑control. Mr Philip Chau, counsel for the prosecution, simply expressed his “complete agreement” with Mr Bruce’s comments. In those circumstances, the judge finalised her written directions to the jury on the issues. 9. In his closing address to the jury, Mr Bruce relied on the psychiatric evidence of Dr Wong in support of his client’s defence of diminished responsibility. Consistent with his comments on the judge’s draft written directions, when he came to the defence of provocation, the psychiatric evidence was only relied on in relation to the objective question of whether an ordinary sober person would have been provoked given the appellant’s experiences, background, features and attributes relevant to the provocation. It was not relied on or referred to in relation to the prior, factual question of whether the appellant had or might have lost his self‑control. 10. The judge’s summing up, in which she directed the jury along the lines of her written directions on the issues, also referred to the psychiatric evidence in the context of the objective question of “do as he did” only, but not the factual question of loss of self‑control. 11. The jury unanimously returned a guilty verdict of murder and the appellant was sentenced to life imprisonment accordingly. The appeal below 12. On appeal, Mr Bruce, for the appellant, successfully ran the argument that the judge misdirected the jury by failing to draw their attention to the psychiatric evidence and its relevance in relation to the issue of whether the appellant lost or might have lost his self‑control[7]. The Court of Appeal unanimously agreed that the psychiatric evidence was “highly relevant”[8] to the factual issue of loss of self‑control, and concluded that the judge’s “clear failure”[9] to direct the jury on the relevance of the psychiatric evidence to this issue resulted in the defence of provocation not being fully and fairly placed before the jury. The failure “went to the heart of the subjective aspect of the defence”[10], namely whether the appellant lost his self‑control. The appeal was allowed and the conviction quashed, for the third time, accordingly[11]. 13. What divided the Court in its judgment dated 20 November 2020 was the question of ordering a third retrial, that is, a fourth trial for murder. Zervos JA, who gave the leading judgment in favour of a third retrial, summarised his reasons for so ordering in paragraph 257 of the judgment: “ In summary, I am persuaded that the applicant should be retried for a fourth time and my decision includes the following considerations: (1) The number of trials and appeals. The applicant has undergone three trials and various appeals and this is a significant factor against ordering a retrial, but this has been outweighed by other factors that have led to the decision that the applicant should be tried for a fourth time. I do not consider that a fourth trial would be oppressive in the circumstances. (2) The lapse of time. The delay in putting the applicant on trial for a fourth time is also a significant factor against ordering a retrial, but bearing in mind the history of the proceedings and nature of the case I do not consider that this will cause any material prejudice to the applicant. (3) The need for finality. This is an important consideration that must also be considered together with the seriousness of the offence. It is desirable that the guilt or innocence of the applicant for the alleged offence of murder be finally determined by a jury. This consideration not only takes into account the interests of the applicant, but also the interests of the victim’s family and the interests of the public. (4) The seriousness of the offence and the strength of the case. This case involves an alleged offence of murder. There is no question that the applicant killed the deceased. The issue at trial is whether the applicant killed the deceased with murderous intent or whether he was provoked or suffering from diminished responsibility at the time of the killing. (5) The length and complexity of a fourth trial. This is a fairly straightforward case where the issues are well defined. (6) The nature of the case and the matters in issue. The questions to be addressed in evaluating the defences are preeminently ones for a jury and it is clearly preferable for a jury to determine if the applicant’s conviction for killing the deceased should be for murder or the lesser crime of manslaughter. (7) The alternative conviction of manslaughter. A serious crime of murder is alleged against the applicant and it is in the interests of the public that persons charged with such serious crime have the charge determined by a court of law, and if found guilty, are appropriately punished. (8) The length of time served. The applicant if convicted of murder would be sentenced to life imprisonment. (9) Whether the applicant can receive a fair trial. It is said that there will be difficulties in the applicant receiving a fair trial because of (i) his previous statements and evidence; and (ii) media publicity of the case. If an issue should arise in relation to these matters they can be appropriately dealt with in the trial process. (10) The history of the proceedings and the applicant’s responsibility for the mistrials and delay. It would seem that the applicant is partly responsible for the delay in the proceedings because of the way he has run the defence case which at the third trial changed to include the defence of diminished responsibility. There is also the factor that the applicant’s conviction has been quashed because of a misdirection by the judge for which leading counsel was partly responsible. (11) The applicant’s mental condition. It does not appear that the applicant’s mental condition would preclude ordering a retrial either on that basis or together with other factors, namely undergoing a fourth trial and the lapse of time since the offence took place.” 14. Earlier in his judgment, Zervos JA had said that this was a case where “the circumstances which brought about the possibility of a retrial were, in no small measure, due to the conduct of the [appellant] or those acting for him”[12]. The judge was critical of Mr Bruce’s failure to point out to the trial judge the relevance of the psychiatric evidence to the factual question of whether the appellant had or might have lost his self‑control under the provocation defence[13]. He referred to it as a possible “tactical decision” on the part of leading counsel[14]. In paragraph 212 of the judgment, Zervos JA commented: “ It is clearly apparent from the highlighted comments of Mr Bruce in his closing address that he made no mention of the psychiatric evidence under the first limb of provocation, and only specifically discussed the psychiatric evidence under the second limb. This approach by the defence gives the impression that a tactical decision may have been made to lay emphasis on the alleged provocative conduct of the deceased, ensuring the impact of it was not lessened by reference to the psychiatric evidence, in order to establish that the applicant was or might have been provoked.” 15. In his short concurring judgment, Macrae VP gave five reasons for favouring another retrial: first, this was a case of murder, the most serious offence in the criminal calendar. It was important from everyone’s point of view, not least the family of the deceased, that there be a proper adjudication, according to law, of the guilt of the accused in respect of such a serious offence[15]. Secondly, there was no issue as to whether the appellant killed the deceased. The issues arising from the appellant’s two defences would best be resolved by a jury which hears the evidence rather than by an appellate court which reads it[16]. Thirdly, given the nature of the appellant’s defences, the issue would necessarily turn on what was in the mind of the appellant at the time he killed the deceased. The appellant’s own recollection was likely to be vivid and graphic, and thus the lapse of time and its effect on the appellant’s memory was more apparent than real as a problem. Appropriate directions could also be given at the retrial to make allowances for the fact that with the passage of time memories fade[17]. Fourthly, the third trial was the first trial at which the appellant had advanced the issue of diminished responsibility[18]. Fifthly, the judge’s error was contributed to by the submissions of the appellant’s leading counsel himself[19]. 16. Disagreeing with the majority, McWalters JA took the view that enough was enough. In paragraph 143 of the Court of Appeal’s judgment, he summarised the factors which persuaded him to come to that conclusion: “ I am persuaded into adopting this course by the following factors: (i) on the assumption that the applicant was awarded a discount of one third for his offer to plead guilty to manslaughter[20] then the applicant has, in effect, already served a fixed term sentence of approximately 23 years’ imprisonment[21]; (ii) the applicant has been forced to undergo the ordeal of three trials and multiple appeals over the course of 11 years and as a consequence a fourth trial to determine whether he is guilty of murder, as opposed to manslaughter, would be oppressive to him; (iii) there would be considerable difficulties in ensuring that a fourth trial would be a fair trial; (iv) a substituted verdict means that the applicant does not escape justice and does not go unpunished for his crime; (v) a substituted verdict will bring finality to the proceedings arising from this crime and this will provide additional benefits of: (a) a saving to the public purse; and (b) closure for the applicant and his family; and (vi) on a properly conducted retrial a verdict of manslaughter cannot be said to be an unlikely verdict as there was independent evidence of provocative conduct by the deceased and the murder was clearly committed in a frenzied state consistent with a loss of self‑control. I note that in R v Bell the English Court of Appeal said of a third trial for murder: ‘The broad public interest in the administration of criminal justice leads us to the clear view that a second re‑trial should be confined to the very small number of cases in which the jury is being invited to address a crime of extreme gravity which has undoubtedly occurred (as here) and in which the evidence that the defendant committed the crime (again, as here) on any fair minded objective judgment remains very powerful.’[22] I am not persuaded that it can be said that on a fair minded objective judgment, the evidence ‘remains very powerful’ for the return of a verdict of murder, as opposed to a verdict of manslaughter.”[23] 17. As for the suggested “tactical decision” of Mr Bruce which contributed to the trial judge’s error in her directions to the jury, McWalters JA had this to say: “ In respect of the conduct of Mr Bruce, Zervos JA is particularly concerned that it may reflect a deliberate tactical choice by him in the conduct of the defence case. I am not persuaded that the transcript supports such a conclusion. I cannot see in this case, in contrast with what occurred in the first trial, any positive action by Mr Bruce or Mr Chau to persuade the Judge to pursue a particular course and nothing was said or done by either counsel to mislead the Judge. I cannot see any basis for suggesting that Mr Bruce intentionally stood by and allowed the Judge to fall into error. Nor can I see how any tactical benefit would accrue to the applicant by not reminding the jury of the relevance of the psychiatric evidence to their decision on the issue of loss of control. Indeed, quite the contrary. It was positively adverse to the applicant and that is why we are allowing the appeal. I cannot, therefore, agree that there is a basis for concluding that Mr Bruce deliberately pursued a tactical course in his conduct of the defence case which contributed to the error of the judge which lead to this appeal being allowed.”[24] Leave to appeal 18. On 29 March 2021, the Appeal Committee (Ribeiro PJ, Fok PJ and Tang NPJ) granted leave to appeal on the substantial and grave injustice basis on the footing that it was reasonably arguable that in ordering a third retrial for murder, the exercise of discretion by the majority of the Court of Appeal miscarried. The arguments before us 19. Before us, Mr Robert Pang SC (Ms Denise Souza with him), for the appellant, relied on three grounds[25]. First, the majority of the Court of Appeal wrongly took into account the conduct of the appellant’s legal representatives in his previous trials, particularly the third trial, as factors which were relevant and justified a further retrial. In particular, counsel argued that although Mr Bruce did comment on the judge’s draft written directions and failed to point out the relevance of the psychiatric evidence to the factual question of losing self‑control under the defence of provocation, that was a mistake shared by prosecuting counsel as well as the judge herself. It was not the result of any “tactical decision” on the part of Mr Bruce and it was an irrelevant consideration. No tactical advantage was ever identified by the majority. As for the raising of the defence of diminished responsibility and the reliance on psychiatric evidence only for the first time at the third trial, Mr Pang submitted that it was a viable defence and the appellant was entitled to raise it. The new defence did not cause any delay or adjournment. There was no suggestion that the psychiatric evidence was unnecessary or was time wasting. 20. Secondly, counsel argued that the majority did not sufficiently consider whether the prosecution had a strong case for murder. The fact that the killing was admitted did not mean that the prosecution had a strong case. The issue was whether the prosecution had a strong case that the appellant was not legally provoked into killing the deceased. In light of the evidence of the appellant, the circumstances and the psychiatric evidence, counsel submitted that the strength of the prosecution’s case was not as strong as the majority thought it to be. 21. Thirdly, counsel submitted that the majority failed to sufficiently consider whether the delay or lapse of time in proceedings rendered an order for retrial oppressive and unjust, or a breach of the appellant’s right to be tried without undue delay under article 11(2)(c) of the Hong Kong Bill of Rights[26]. Counsel submitted that the majority failed to realise that the right to be tried without undue delay is a separate or independent right from the right to a fair trial[27] and the order for a third retrial breached the appellant’s right. 22. Counsel asked the Court to substitute the conviction for murder with a conviction for manslaughter on the basis of provocation, and sentence the appellant accordingly. 23. For the respondent, Mr William Tam SC (Mr Ira Lui and Mr Andy Lo with him) pointed out that a fourth trial for murder is not unprecedented[28]. The law has given the discretion to order a retrial to the Court of Appeal, rather than to this Court. This Court should, therefore, not lightly disturb the exercise of discretion by the Court of Appeal. The Court of Appeal was in disagreement only on the weight to be attached to certain factors and ultimately favoured a retrial by a majority. The majority and the minority each had the benefit of considering and debating the detailed reasons of the other side. There was no cogent reason for suggesting that the majority had erred in principle or had taken into account irrelevant factors. Whilst the issue might be finely balanced, there was no reason to interfere with the exercise of discretion by the majority, even if this Court might “lean in favour of an exercise of it the other way”[29]. 24. As for the conduct of the appellant’s counsel at the third trial, Mr Tam emphasised that the jurisdiction to order a retrial is put in the widest possible terms. As a matter of principle, there is no justification for excluding from the Court of Appeal’s consideration the conduct of an accused’s legal representatives. 25. Counsel also submitted that the majority’s focus, when referring to the appellant’s counsel’s conduct at the third trial, was on the cause and consequence of the technical blunder by the trial judge. The majority’s emphasis was on the nature of the error and not on an attribution of fault. In any event, the majority only considered that counsel’s contributing fault was a material matter to be taken into account. It did not treat it as a determinative one[30]. 26. Counsel accepted that the majority of the Court of Appeal did not elaborate on why the raising of diminished responsibility as a partial defence at the third trial only was relevant to the question of retrial. He suggested that if the defence had been raised earlier, the relevance of the psychiatric evidence could have been sorted out either at trial or on appeal on an earlier occasion. 27. As for the strength of the prosecution case, Mr Tam pointed out that the appellant’s defence turned entirely on his own evidence and the psychiatric evidence. He submitted that the appellant’s testimony was self‑contradictory, unreasonable and inconsistent with the account in his video interviews. In relation to expert evidence, Dr Oliver Chan’s evidence in rebuttal was obviously preferred by the jury. Counsel also pointed out that the appellant had been disbelieved unanimously in all his previous trials. The defence of diminished responsibility was plainly rejected by the jury. Whether the appellant was or might have been provoked into killing the deceased was eminently a task for the jury. 28. As regards the question of delay and fair trial, Mr Tam pointed out that a fair trial is a trial fair to all parties, and not only to the accused. He emphasised that murder is one of the most serious crimes and the public interest in seeing that the murder charge against the appellant for the brutal killing of the deceased was resolved by a jury at trial was very strong. It was submitted that the issues involved in the retrial would be within a very narrow compass and the retrial was unlikely to be lengthy. There was no suggestion of the appellant having difficulties in recalling the events due to the lapse of time, and indeed he had no difficulties in giving a very detailed account to the psychiatrists when they interviewed him over eight years after the killing. In any event, any difficulties would be mitigated by reference to the appellant’s police interviews and transcripts of his evidence in the earlier trials. Moreover, the trial judge could, if required, give the jury appropriate directions. A lapse of time of some twelve years would not in itself forbid a further retrial. 29. Mr Tam also argued that a lapse of time amounting to a breach of the right to a fair trial within a reasonable time does not per se render an otherwise fair trial and a resulting conviction unlawful. A breach of the reasonable time requirement would only entitle the convicted person to remedies such as a reduction in penalty and payment of compensation[31]. The legal principles 30. The power to order a retrial is found in section 83E(1) of the Criminal Procedure Ordinance (Cap 221). It provides: “Where the Court of Appeal allows an appeal against conviction and it appears to the Court of Appeal that the interests of justice so require, it may order the appellant to be retried.” 31. The principles governing the exercise of this statutory discretion to order a retrial are well established. They were revisited by this Court as recently as last year in HKSAR v Zhou Limei (No 2)[32], and it is quite unnecessary to go into the earlier authorities again. In paragraph 9 of this Court’s judgment, Chief Justice Ma, giving the leading judgment of the Court, reiterated a number of relevant principles from the decided cases on retrial: “ Section 83E of the Criminal Procedure Ordinance states that the Court of Appeal, if it decides to allow an appeal, may order a retrial if ‘the interests of justice’ so require. The jurisdiction to order a retrial is thus put in the widest possible terms. In HKSAR v Tam Ho Nam (No 2),[33]Fok PJ restated the relevant principles regarding retrials by reference to Au Pui Kuen v Attorney General of Hong Kong,[34]Ting James Henry v HKSAR[35]and Kissel v HKSAR.[36] The following principles, distilled from the above and other cases, are relevant in the present case: (1) Whether or not a retrial should be ordered is a matter of discretion. This discretion is usually exercised, as it should be, by the Court of Appeal, relying on their ‘collective sense of justice and common sense.’[37] And, as was put by Lord Bingham of Cornhill, there must be ‘an informed and dispassionate assessment of how the interests of justice in the widest sense are best served’; it is important to maintain ‘confidence in the efficacy of the criminal justice system.’[38] (2) The discretion whether or not to order a retrial depends entirely on what justice requires (this being the ‘critical question’).[39] (3) The interests of justice of course include a consideration of an accused’s interests and circumstances. The criminal justice system is there to bring matters to a conclusion without undue delay and without oppression; these are ‘accepted norms’.[40] It should be acknowledged that any criminal trial is to some degree an ordeal for the accused.[41] The interests of justice also include the interest of the public in seeing those who are guilty of serious crimes brought to justice and not escape merely because of a technical error in the conduct of a trial or in the summing up to a jury.[42] In Au Pui Kuen v Attorney General of Hong Kong,[43]Lord Diplock referred to the following passage from the judgment of Gould Ag CJ in Ng Yuk Kin v The Crown:[44]that there may be cases where it ‘is in the interest of the public, the complainant, and the appellant himself that the question of guilt or otherwise be determined finally by the verdict of a jury, and not left as something which must remain undecided by reason of a defect in legal machinery’. In assessing the public interest, a court must take into account the views of the prosecution which is best qualified (and I would add has the duty) to present the views of the public,[45]although it must ultimately be for the court to determine what is in the public interest. The strength of the prosecution case is also a relevant consideration. (4) The interests of justice require all relevant factors, both for and against a retrial, to be taken into account. Such factors will not only vary from case to case, but their relative importance and weight will also be different in any given case.[46] (5) The above said, one factor that must be given significant weight is the fact that the accused has already undergone a trial, in particular where the trial is long and complex.[47] This is all the more so when there is involved a second retrial, which means of course the possibility of a third trial for the same offence. In Mok Kin Kau v HKSAR,[48]the ordering of a second retrial after two concluded trials and appeals, and the serving of the whole sentence, was said to be an ‘unusual course’ and in such a situation, in the absence of a special or compelling reason, this was a ‘departure from accepted norms’ sufficient to constitute a substantial and grave injustice. Although it is not unprecedented for a second retrial to be ordered, the cases accept that this is an ‘unusual’ course to take,[49]even where the accused has not served the whole of his or her sentence as was the case in Mok Kin Kau v HKSAR. Given that the ordering of a second retrial is an unusual course, a court would have to be persuaded by cogent and compelling reasons to make such an order. This is consistent with the interests of justice, but of course all relevant factors must be carefully weighed in this exercise of discretion. (6) Another factor that should also be taken into account is the time that an accused has spent in custody and in relation to this facet, the time that an accused has been in custody will have to be seen against the likely sentence that he or she might receive on a retrial.[50]”[51] 32. The printed cases of the parties referred to a number of cases in which the courts considered how the discretion to order a retrial should be exercised[52]. Of course, each case turns on its own facts. The reference to how the discretion to order a retrial was exercised in a previous case, even of broadly similar facts, is bound to be of limited assistance. 33. As already mentioned, the primary responsibility for exercising the discretion to order a retrial lies with the Court of Appeal, after quashing a conviction below. As explained in Zhou Limei (No 2)[53], this Court would only intervene in limited circumstances: “ In common with the review of discretion in other areas, this Court will not disturb the exercise of discretion by the Court of Appeal in the determination whether or not to order a retrial unless there has been a serious error of law or principle or approach, such as where the court below has failed to take into account a relevant consideration or has taken into account an irrelevant one. Only in such a situation would the Court of Final Appeal be justified in exercising the discretion afresh. Furthermore, it is also important to emphasise that the relative weight which the Court of Appeal ‘ascribes to each relevant factor’ is a matter within its discretion and it is not open to the Court of Final Appeal to seek to ascribe a different weight.[54]”[55] The majority’s exercise of discretion 34. Turning to the facts of the present case, a major difference between the majority and the minority on the question of retrial was whether Mr Bruce (and, vicariously, the appellant himself) contributed to[56] or was “partly responsible”[57] for the error made by the judge in her directions on the issue of loss of self‑control insofar as the relevance of the psychiatric evidence was concerned. Although couched in cautious terms, it is plain from a fair reading of the judgments of Macrae VP and Zervos JA that they considered defence counsel was deliberate in not drawing to the judge’s attention the relevance of the psychiatric evidence to the question of loss of self‑control. Instead, he simply stood by and allowed her to fall into error, for the purpose of achieving a tactical benefit. That certainly was how McWalters JA understood their position[58]. He gave his reasons for disagreeing with the view of the majority on the role played by defence counsel in the judge’s mistake. I find his reasons persuasive. 35. No doubt, the judge was in error when she overlooked entirely the relevance of the psychiatric evidence to the question of provocation in her initial draft directions which she showed to counsel for comments. It is also true that Mr Bruce only referred to the psychiatric evidence in relation to the objective question of “do as he did” under the defence of provocation. He made no mention of the relevance of the expert evidence to the question of loss of self‑control. The same omission was repeated in his closing address. To that extent, Mr Bruce, like prosecuting counsel, may properly be regarded as having played a part in the judge’s mistake in overlooking the relevance of the expert evidence to the question of loss of self‑control. 36. However, the majority obviously went much further. They thought that defence counsel must have also been aware of the relevance of the expert evidence to the question of loss of self‑control but by a deliberate, tactical decision he chose to keep quiet about it and let the judge err. According to Zervos JA, this was to achieve a tactical advantage which he explained in paragraph 212 of his judgment. If the learned judge were right on his view, this would have been a very serious matter, contributing substantially to the trial judge’s omission in her directions and thus the eventual quashing of the jury’s verdict, and it would have been right to take it critically against the appellant on the question of retrial. And indeed quite apparently, that was Zervos JA’s approach although he was careful to say that this was not by itself determinative of the question of retrial[59]. 37. With respect, I am unable to subscribe to Zervos JA’s view. First, it was the trial judge who first overlooked the relevance of the expert evidence to the question of provocation altogether in her initial draft directions. There is no suggestion that Mr Bruce played any part in it. According to the transcript, after reading the initial draft directions, Mr Bruce recognised the relevance of the expert evidence to the objective question of “do as he did” and made his comments to the judge accordingly. Significantly, after he had done so, neither the judge nor prosecuting counsel who agreed with Mr Bruce on the relevance of the evidence to the objective question realised that the expert evidence was also material to the earlier question of loss of self‑control. In terms of failing to appreciate the relevance of the expert evidence to that question, all three of them – the judge, prosecuting counsel and defence counsel – were on an equal footing. Judging from the transcript[60], there was really nothing to differentiate them from one another in this regard, and there was certainly nothing to justify singling Mr Bruce out by suggesting that his failure to point out the relevance of the expert evidence to the question of loss of self‑control was a deliberate, tactical one, whereas the same failure of the judge and prosecuting counsel was a mere error on their part. It is true that at the hearing before the Court of Appeal, when asked about it, Mr Bruce could not give a satisfactory answer for his not spotting the relevance of the expert evidence to the question of loss of self‑control at the time of trial[61]. But there could be many reasons for this, and in the absence of convincing evidence, I am not prepared to draw any adverse inference against leading counsel. 38. This brings me to the so‑called “tactical benefit” that Mr Bruce was supposed to be seeking to obtain for his client. Zervos JA sought to articulate it in paragraph 212 of his judgment. However, with respect, I share McWalters JA’s inability to comprehend the supposed tactical advantage that defence counsel was seeking to achieve[62]. The psychiatric evidence was supportive of the appellant’s case on provocation not only on the objective question of “do as he did”, but also on the factual question of loss of self‑control. It is difficult to see how any tactical benefit would be achieved for the appellant by not pointing that out to the judge and to the jury. Any suggestion that defence counsel was seeking to lay the foundation for a ground of appeal would not make any sense at all unless the conviction of the appellant for murder by the jury was a foregone conclusion. However, I do not see any material to suggest that. 39. In any event, whether the fault of defence counsel should be attributed to the defendant must depend on the facts of each case. Even if Mr Bruce were really guilty of deliberately withholding the relevance of the psychiatric evidence to the question of loss of self‑control from the judge (and the jury), the extent to which the appellant should be held vicariously responsible for it would still be highly debatable, particularly when there was nothing to suggest that the appellant was a party to whatever tactical decision that leading counsel might have made on his behalf. 40. Both Macrae VP and Zervos JA thought the fact that diminished responsibility was only raised for the first time at the third trial was a matter that should be counted against the appellant[63]. With respect, I disagree. There is no suggestion that diminished responsibility was not an arguable defence in the present case. After all, it was supported by expert evidence. It was available to the appellant, and he was entitled to run it. The running of that defence at the third trial did not cause any significant delay in the proceedings. The fact that the running of that defence entailed the introduction of the psychiatric evidence at the trial which, as it happened, played a crucial part in the error committed by the judge in her summing up was neither here nor there. Zervos JA thought that the raising of this additional defence at the third trial “was partly responsible for complicating the defence of provocation”[64]. I cannot see this as a valid criticism against the appellant on the question of retrial, nor do I see this as a reason for visiting the judge’s error upon the appellant for running an additional defence at the third trial. 41. Mr Tam’s suggestion that had the partial defence been first raised at the first or second trial, the relevance of the psychiatric evidence to the question of loss of self‑control could have been sorted out earlier is speculative at best. In any event, it is difficult to see how and why the appellant should be held responsible for its late raising by his legal representatives given the highly technical nature of the defence. 42. For these reasons, I take the view that the majority of the Court of Appeal has misapprehended what happened at the trial and taken into account irrelevant considerations when exercising the discretion to order a third retrial. The exercise of discretion miscarried, and it falls to this Court to re‑exercise it. This is one of those rare cases where it is justified for this Court to interfere with the exercise of discretion on retrial by the Court of Appeal. Re‑exercising the discretion 43. Section 83E(1) of the Criminal Procedure Ordinance provides that the Court of Appeal may order a retrial “if it appears to the Court of Appeal that the interests of justice so require”. Exercising afresh the discretion conferred on the Court of Appeal, I cannot be satisfied that the interests of justice require a further retrial. Indeed I am satisfied that they do not. 44. The relevant considerations that should be taken into account have been discussed in quite some detail by McWalters JA in his dissenting judgment[65], and they have been summarised in paragraph 143 of his judgment which I have already quoted. Leaving aside the supposed fault and responsibility of defence counsel at trial and the fact that diminished responsibility was only relied on for the first time at the third trial, the other considerations mentioned by Zervos JA in his judgment[66] are also relevant ones and I bear them in mind as well. 45. Amongst these relevant considerations, I would emphasise the following matters which, when considered together, decidedly tilt the balance against ordering a third retrial. First, if such an order were made, this would be the fourth full trial for murder that the appellant would be forced to face. At the retrial, he would have to undergo the substantial forensic challenge of recalling the parties’ relationship and the details of the killing at trial, and be subjected to cross‑examination on them. He would also have to face the stress and anxiety over the uncertainty of the outcome of the retrial. These would only be some of the matters that the appellant would have to endure in a fourth trial. As has been observed[67], any criminal trial is “to some degree an ordeal for the accused” and “it goes without saying that no judge exercising his discretion judicially would require a person who had undergone this ordeal once to endure it for a second time unless the interests of justice required it”. McWalters JA was obviously correct when he pointed out that “[t]he more often the defendant has to face trial, the greater the ordeal will be for him, and the less fair the criminal justice process will be perceived to be”[68]. In the present case, the appellant has already undergone three full trials, three appeals in the Court of Appeal, as well as two appeals in this Court over the last eleven years, which is, on any view, a very lengthy period of time. There must come a point at which the question – what do the interests of justice require? – is answered by saying “enough is enough”, and to order another retrial would simply be oppressive to the defendant. 46. Secondly, the appellant has already been remanded in custody for almost twelve years. That is equivalent to a starting point of almost 27 years’ imprisonment assuming a conviction for manslaughter, given the conventional one‑third discount for his early plea of guilty, as well as another one‑third remission by the Commissioner of Correctional Services for good behaviour (as to which there is no dispute)[69]. It is true that if the appellant were to undergo a fourth trial and be found guilty of murder, he would be sentenced to imprisonment for life, and this would mean, very generally speaking, anything between 20 to 30 years’ imprisonment before he might be paroled. Yet it has to be remembered that if upon retrial, he were only found guilty of manslaughter, a starting point of almost 27 years’ imprisonment would be way beyond the higher end of the usual range of sentences for manslaughter, even after taking into account the horrendous manner in which the killing was committed[70]. 47. Thirdly, I fully bear in mind that the appellant is charged with murder, one of the most serious crimes in our statute books, and this is a particularly brutal case. The public interest in seeing that the murder charge is resolved by a jury after trial is extremely great, not to mention the interest of the deceased’s family in seeing that justice is done. I also bear in mind that the issues involved in the case are relatively straightforward and narrow, and the trial is not likely to be lengthy. I give due weight to the view of the prosecution as representing the public interest that although this case is testing the limits of the discretion to order a retrial, the public interest inclines in favour of a fourth trial. However, after the lapse of almost twelve years, the memory of the appellant as well as that of the witnesses of the relevant events must have been affected to some extent, and there is a difference between giving evidence on the basis of one’s recollection and impression at the time and reconstructing one’s memory and perception from witness statements and transcripts. The quality of the evidence as well as the demeanour of the witnesses are bound to be affected. Given that the issues in a fourth trial would turn very much on the performance of the appellant in the witness box, I would be slow to say that any prejudice that he might suffer at the trial could be satisfactorily remedied by appropriate directions to the jury. On the other hand, unlike those cases where the choice faced by the court was either to order a retrial or to let the accused go free without any conviction, not ordering another retrial here would not mean the appellant could leave as an innocent man. He would still be convicted and sentenced for manslaughter. 48. This brings me to the fourth matter, namely, the strength of the prosecution case for murder. The issue here is a narrow one, as the killing of the deceased by the appellant has been admitted from day one. The contest is between murder and manslaughter (by reason of provocation or diminished responsibility). It is true that the appellant has been convicted for murder three times by three juries. However, the trial process on each occasion was flawed. On the facts and evidence before us, I would say that the defence of provocation, which is supported by the appellant’s story (which to some extent is corroborated by the deceased’s former husband’s evidence), the circumstances under which the killing took place and the psychiatric evidence, is at least reasonably arguable and a murder verdict is certainly not a foregone conclusion. 49. Fifthly, the quashing of the convictions in the first and third trials was in part due to mistakes contributed to by defence counsel then representing the appellant. However, for the mistake in the first trial, as McWalters JA explained[71], the primary responsibility lay with the prosecutor. As for the mistake in the third trial, for the reasons given above, it was not the result of any deliberate, tactical decision on the part of defence counsel, still less that of the appellant. In my view, they carry little if any weight when considering how the discretion should be exercised. 50. Looking at the matter in the round, I have come to the conclusion that the discretion should be exercised against ordering a retrial. Regrettably, the reality in the present case is that the criminal justice system, despite three attempts, has failed not only the deceased’s family and the public, but also the appellant. I think we have reached a point where enough is enough, and the interests of justice are no longer served by continuing with the prosecution of the appellant for murder. Rather, a conviction for manslaughter on the basis of provocation should be entered on the appellant’s plea of guilty and the appellant sentenced accordingly, thereby bringing finality and closure to this tragic case. Sentence 51. Generally speaking, it is not the function of this Court to pass sentences in criminal cases. That is the function of the trial court and, on appeal, the Court of Appeal. However, on the wholly exceptional circumstances of this case, it is right for this Court to sentence the appellant for manslaughter. As mentioned, the appellant has been in custody for almost twelve years which will be counted as service of his sentence. Mr Tam, Deputy Director of Public Prosecutions, very fairly accepted that bearing in mind the one‑third discount for the early plea of guilty and the one‑third remission for good behaviour by the Commissioner of Correctional Services, the time that the appellant had spent in custody would have exceeded whatever sentence that this Court might legitimately pass in the present case on the basis of a manslaughter conviction. Mr Tam agreed that amongst the authorities for sentencing placed before the Court[72], the highest sentence imposed was 16 years[73], which is of course far less than the starting point of almost 27 years that the time the appellant has already spent in custody would represent. 52. In the circumstances, I would quash the order for a third retrial, enter a conviction for manslaughter on the basis of provocation and sentence the appellant to such sentence that would allow for his immediate release from custody. Mr Justice Ribeiro PJ: 53. I agree with the judgment of the Chief Justice. Mr Justice Fok PJ: 54. I agree with the judgment of the Chief Justice. Mr Justice Tang NPJ: 55. I also agree. Mr Justice French NPJ: 56. I agree with the orders proposed by the Chief Justice and his reasons for them. Chief Justice Cheung: 57. Accordingly, the court unanimously allows the appeal and makes the orders set out in paragraph 52 above. Mr Robert Pang SC and Ms Denise Souza, instructed by Tse Yuen Ting Wong, assigned by the Director of Legal Aid, for the appellant Mr William Tam SC, DDPP, Mr Ira Lui, ADPP and Mr Andy Lo, SPP, of the Department of Justice, for the respondent [1] [2014] 4 HKC 145 and [2021] 1 HKLRD 26 (“CA judgment”). [2] (2017) 20 HKCFAR 1. [3] [2015] 5 HKLRD 743. [4] [2021] 1 HKLRD 26. [5] [2021] HKCFA 12. [6] Section 4 reads: “Where on a charge of murder there is evidence on which the jury can find that the person charged was provoked (whether by things done or by things said or by both together) to lose his self‑control, the question whether the provocation was enough to make a reasonable man do as he did shall be left to be determined by the jury; and in determining that question the jury shall take into account everything both done and said according to the effect which, in their opinion, it would have on a reasonable man.” [7] CA judgment, [62]. [8] Ibid, [84]. [9] Ibid, [84]. [10] Ibid, [90]. [11] Ibid, [259]. [12] Ibid, [226]. [13] Ibid, [205]-[214], [226], [251]. [14] Ibid, [146], [212]. [15] Ibid, [3]. [16] Ibid, [4]. [17] Ibid, [5]. [18] Ibid, [6]. [19] Ibid, [7]. [20] In my view, as a matter of legal principle, the applicant should be accorded a full one‑third discount on the sentence as he has all times offered to plead guilty to manslaughter and in none of his trials has he ever sought a complete acquittal. On arraignment, he pleaded guilty to manslaughter. [21] 23 years discounted by one third is 15 years 4 months which when discounted again by one third to allow for remission by the Commissioner for Correctional Services is 10 years 2 months. [22] [2010] 1 Cr App R 27, 418, [46]. [23] Original footnotes included. [24] CA judgment, [124]. [25] At the oral hearing, counsel’s submissions focused on the first ground only. [26] The Hong Kong Bill of Rights Ordinance (Cap 383), section 8. [27] Hong Kong Bill of Rights, article 10. [28] R v Davis (Iain Lawrence) [2011] EWCA Crim 2156; Andrew John Hawkins (No 3) (1994) 76 A Crim R 47; Shane Michael Martin (No 4) (1999) 105 A Crim R 390. [29] HKSAR v Chu Chi Wah FAMC 58/2010, 2 November 2010, [3]. [30] CA judgment, [252]. [31] Attorney General’s Reference (No 2 of 2001) [2004] 2 AC 72; Henworth v The United Kingdom (2005) 40 EHRR 33; Vlad v Romania (2014) 59 EHRR 13. [32] (2020) 23 HKCFAR 169. [33] (2017) 20 HKCFAR 414. [34] [1980] AC 351. [35] (2007) 10 HKCFAR 632. [36] (2010) 13 HKCFAR 27. [37] Au Pui Kuen v Attorney General of Hong Kong [1980] AC 351, 357D–E. [38] B (a Child) v R [2001] UKPC 19, [39]. [39] HKSAR v Tam Ho Nam (No 2) (2017) 20 HKCFAR 414, [21]. [40] Mok Kin Kau v HKSAR (2008) 11 HKCFAR 1, [10]. [41] Au Pui Kuen v Attorney General of Hong Kong, 356H. [42] Au Pui Kuen v Attorney General of Hong Kong, 357C–D. [43] At 359D–E. See also Reid (Dennis) v R [1980] AC 343, 350G–H. [44] (1955) 39 HKLR 49, 60. [45] Ting James Henry v HKSAR, [51]. [46] Au Pui Kuen v Attorney General of Hong Kong, 357D–E. [47] Ting James Henry v HKSAR, [50]. [48] At [12] and [14]. [49] See HKSAR v Tam Ho Nam (No 2), [24] referring to R v Chau Mei Ling [1981] HKC 542, 545B–C; Mok Kin Kau v HKSAR, [7]; HKSAR v Li Yanhong (No 2) [2016] 1 HKLRD 946, [14]. [50] See, for example, Ting James Henry v HKSAR, [52]. [51] Original footnotes included (with some omissions). [52] Ting Henry James v HKSAR; HKSAR v Tam Ho Nam (No 2); HKSAR v Zhou Limei (No 2); Au Pui Kuen v Attorney General; Mok Kin Kau v HKSAR; Reid v The Queen; Charles and Ors. v The State [2000] 1 WLR 384 (PC); Henworth v United Kingdom; Vlad v Romania; R v Lazarus [2017] NSWCCA 279; HKSAR v Lee Ming Tee HCCC 191/1999, 8 June 2004; HKSAR v Li Yanhong (No 2); Nguyen Anh Nga v HKSAR (2017) 20 HKCFAR 149; HKSAR v Bian Zhenju [2015] 2 HKLRD 1089; R v Bell; R v Davis (Iain Lawrence); DPP (Nauru) v Fowler (1984) 154 CLR 627; Andrew John Hawkins (No 3); Shane Michael Martin (No 4); HKSAR v Ha But-yee FAMC 41/2016, 11 November 2016; HKSAR v Chu Chi-wah; HKSAR v Liang Yaoqiang; HKSAR v Lee Wai-man [2020] 3 HKLRD 310; B (a child) v R and HKSAR v Nayab Amin [2020] 2 HKLRD 1051. [53] [10]. [54] See, in the context of sentencing, Secretary for Justice v Wong Chi Fung (2018) 21 HKCFAR 35, [62]. [55] Original footnote included. [56] CA judgment, [7] per Macrae VP. [57] Ibid, [257](10), per Zervos JA. [58] Ibid, [124]. [59] Ibid, [252]. [60] Extracted by Zervos JA in [208] of his judgment. [61] CA judgment, [146]. [62] Ibid, [124]. [63] Ibid, [6], [251], [257](10). [64] Ibid, [251]. [65] Ibid, [129]-[142]. [66] Ibid, [221]-[257]. [67] Au Pui Kuen v Attorney General of Hong Kong, 356G–H, per Lord Diplock. [68] CA judgment, [133]. [69] A starting point of 27 years discounted by one‑third for an early guilty plea would give a sentence of 18 years. A one‑third remission for good behaviour would reduce the actual period of imprisonment to 12 years. [70] See below. [71] CA judgment, [119]. [72] R v Yu Wing-sze CACC 372/1995, 15 November 1995; HKSAR v Tam Shu-kin CACC 444/2004, 13 May 2005; HKSAR v Yiu Kam‑tin CACC 510/2004, 1 June 2005; HKSAR v Lau Bo‑ki CACC 412/2005, 9 November 2007; HKSAR v Lim Khi‑chong CACC 159/2007, 9 November 2007; HKSAR v Liu Chui-fa HCCC 8/2012, 13 August 2012; HKSAR v Yeung Ki HCCC 242/2013, 14 March 2014; HKSAR v Chen Peng HCCC 115/2016, 20 June 2018; HKSAR v Francisco Reynaldo [2000] 3 HKLRD 688; HKSAR v Wong Kam‑shing Jackie [2010] 4 HKC 580; HKSAR v Yau Kit‑keung [2010] 6 HKC 473. [73] HKSAR v Lau Bo‑ki, [11], a case of total denial of responsibility for the killing. 
1. This is an assessment of damages for personal injuries suffered in a road traffic accident that occurred on 27 June 2014. At the time of the accident, the plaintiff, who was born on 1 October 1959, was about 54.5 years of age. Liability was admitted on 25 November 2015, before proceedings were commenced on 29 May 2017. Interlocutory judgment for damages to be assessed was entered on 8 June 2017 against the 1st and 2nd defendants. At the time of the accident, the 1st defendant was the driver of the 2nd defendant’s bus bearing registration number HS539. 2. The plaintiff was legally aided. However, legal aid was discharged on 22 May 2020 and the plaintiff appeared at trial in person. The defendants were represented by Mr Simon Wong of counsel. 3. The trial was set down in the fixture list for four days commencing on 18 June 2020. Prior to the commencement of trial, I had received opening statements from the plaintiff and the defendants. On the morning of the trial on 18 June 2020, I received into evidence the assessment bundles comprising of the following sections: A. Pleadings and orders [A/1-92]; B. Witness statements [B/93-144F]; C. Medical reports [C/145-201]; D. Experts reports [D/202-336]; E. Other documents [E/337-492B] – this section was subsequently enlarged to include further documents attached to the plaintiff’s closing submissions (including their English translations) at E/351A-351W, 248A, and 359A. These documents, which had been disclosed previously in the plaintiff’s list of documents filed on 27 November 2017, ought to have been included in the assessment bundles at the outset; and F. Medical records - the section was subsequently reduced to a handful of documents after the medical records, which were not relevant to the issues before me, were removed [F/543-550, 573, 901-3, 910-911]. Pursuant to directions made previously, the experts’ reports were adduced into evidence without the need to call the makers thereof. 4. The plaintiff elected to give evidence in Putonghua. He confirmed the truth of his 2 witness statements made on 13 February 2018 [B/93-128] and 12 October 2018 [B/129-139] and he relied on them as his evidence in chief. He also confirmed the truth of the factual statements that he had made in his opening statement. He then gave further evidence complaining that what the doctors said to him orally was very different from the written medical reports; that lot of facts stated orally by the doctors had not been reduced into writing; and that they did not want to bear the responsibility of stating how serious his injuries were (Transcript/Day1/13A-17G). 5. At this point I was about to adjourn for the morning break when the plaintiff complained to me about discomfort in his head and that he was not able to go on. I suggested to the plaintiff that he should take a rest during the morning break. When the hearing resumed, the plaintiff complained that he was feeling worse and that he had forgotten to bring his medicine with him. As he was residing in Kowloon City, I suggested to him that he should return home to take a rest and to take his medicine, and to return to court at 2:30 pm. I then adjourned the trial to 2:30 pm. Prior to the resumption of the proceedings at 2:30 pm, my clerk was informed that the plaintiff had been taken by ambulance to Queen Elizabeth Hospital (“QEH”) for medical treatment. At about 2:40 pm, I adjourned the trial to Monday 22 June 2020 at 10:00 am, rather than Friday 19 June 2020, as I was concerned that costs might be wasted if the plaintiff was admitted to hospital and was unable to come to court the next day. 6. On 19 June 2020, my clerk received a telephone call from Madam Zhang Qiongyun, the wife of the plaintiff, and was informed by her that the plaintiff had been admitted to QEH on 18 June 2020 and, further, that she was told by a doctor from QEH that the plaintiff was unlikely to be discharged on 22 June 2020. My clerk communicated this information to the defendants’ solicitors who suggested that the trial should be adjourned to 23 June 2020. I acceded to this suggestion and adjourned the trial to Tuesday 23 June 2020 at 10 am. 7. On 22 June 2020, Madam Zhang Qiongyun telephoned my clerk again and informed her that she was told by a Dr Lau of the Department of Psychiatry of QEH that the plaintiff needed to be hospitalized for a longer period of time and she did not know when he would be discharged. I then adjourned the trial to 3 – 5 August 2020 (3 days reserved). The parties were notified accordingly. 8. At the same time that I adjourned the trial to 3 – 5 August 2020, I also directed that: “(1) Leave granted to the Defendants to adduce the Investigation Report prepared by Verity Consulting Limited dated 12 May 2020 together with a DVD and 4 pages of photos extracted from the surveillance video taken on 7 May 2020 at the resumed trial; (2) Leave to the Plaintiff to file a further supplemental witness statement to respond to the Investigation Report prepared by Verity Consulting Limited dated 12 May 2020 together with a DVD and 4 pages of photos extracted from the surveillance video taken on 7 May 2020 on or before 24 July 2020.” On 18 June 2020, the defendants had made an application for leave to adduce this surveillance evidence which application the defendants withdrew upon my indication that I was unlikely to grant it. The last-minute admission of this evidence would have resulted in an adjournment of the trial as the plaintiff had to be given the opportunity to respond to the new evidence. However, I granted the application on 22 June 2020 as the trial was being adjourned for over a month which gave the plaintiff sufficient opportunity to respond to this factual evidence that was probative and relevant. 9. The trial resumed on the morning of 3 August 2020. The plaintiff was reaffirmed in Putonghua and he confirmed the truth of his additional statement of 23 July 2020 which he had sent to court prior to the resumption of the trial (Transcript/Day2/22P-23K). The plaintiff was then cross examined by counsel for the defendants who put the defendants’ case to him and put questions to him relating to certain photographs taken at the scene of the accident and to certain previous statements he had made (Transcript/Day2/23M-32D). 10. At about 11 am on the morning of 3 August 2020, I found the plaintiff guilty of a contempt in the face of the court. He had continued to use his mobile phone in court to dial someone despite my warning to him not to use his mobile phone in the court room. I ordered that he be sent down to the cells until 4 pm (Transcript/Day2/32E-33Q). However, he claimed to be unwell and an ambulance was called for him. He was conveyed to the Accident and Emergency Department (“AED”) of Queen Mary Hospital (“QMH”), accompanied by 2 bailiffs of the High Court. He was admitted to QMH on the afternoon of 3 August 2020. At about 4 pm, I released the plaintiff from the custody of the bailiffs and I released the bailiffs from their custodial duty (Transcript/Day2/33S). 11. As his contempt was unrelated to the issues at trial, I adjourned the trial to myself to resume at 10 am on Monday, 10 August 2020. I also indicated that I might adjourn it to a later date if the plaintiff was still in hospital on that day (Transcript/Day2/34A, J). However, the plaintiff was discharged from QMH on 5 August 2020. 12. The trial resumed and was quickly concluded on 10 August 2020 when counsel for the defendants informed the court that he had no further questions to ask the plaintiff. He had already put the defendants’ case to the plaintiff on the second morning of the trial. The plaintiff then completed his evidence. The plaintiff’s wife, Madam Zhang Qiongyun, was unable to come to court to give evidence. However, the defendants’ counsel agreed that her witness statement could be received into evidence as a hearsay statement [B/140-144F]. That completed the plaintiff’s case. Counsel for the defendants then informed me that the defendants were not calling any evidence and would not rely on the surveillance evidence. The defendants would only rely on the documentary evidence in the assessment bundles and the experts’ reports, in particular. I then gave directions for closing submissions to be lodged (Transcript/Day3/1A-6A). Written closing submissions were received from the defendant on 12 August 2020 and from the plaintiff on 18 August 2020. Overview of the Parties’ Cases 13. The defendants’ closing submissions contains a succinct overview of the parties’ cases which I set out below: “2. P’s case [A/75/(r)] is that as a result of the Accident he suffered from multiple serious physical injuries which included those to his face, neck, back, right upper limb, and right lower limb. Significantly, he claimed that he had weakness in right upper limb and total paraplegia of right lower limb and is now wheelchair bound. He also claimed to have suffered from hearing loss on both ears, impairment to both eyes, urinary urgency/incontinence and had dizziness which caused him fall frequently. Mentally, he had persistent insomnia, poor mood, nightmare, auditory hallucinations, flashbacks and vivid recollection of the accident, and had psychotic depression and post-traumatic stress response. He said he is totally dependent on his wife now. P claims damages for PSLA, pre-trial and post-trial loss of earnings and loss of earning capacity. He also claims special damages and damages for future expenses for medical treatments and purchasing aids and equipment. Further, he claims costs of care and attention on the basis that his wife had to quit her jobs in order to look after him. 3. Ds’ case is that P suffered from no more than contusion to his face and mild soft tissue injury to his neck, and those injuries resolved within one month, and that P grossly exaggerated his symptoms and is a malingerer. 4. The parties’ respective quantum figures are summarized as follows: The Documentary Evidence 14. At the time of the accident on 27 June 2014, the 1st defendant was the driver of the 2nd defendant’s bus bearing registration number HS539. He was convicted of careless driving on 27 January 2015 in respect of this occurrence and fined $900. 15. According to the statement made to the police by one, Chan Wai Lun, on 13 August 2014 [E/342-343D], the accident occurred at about 2:45 pm on 27 June 2014. The place of the accident was the Kowloon bound lane of Tsuen Wan Road, outside Wing Kin Industrial Building. Four vehicles were involved in the accident, namely, a medium goods vehicle bearing registration number RJ5615, which he had been driving, together with trailer bearing number 46888T which sustained damage to its rear; a private car bearing registration number RU2729 which sustained damage to its front and rear; a private car bearing registration number HJ3269 which sustained damage to its front and rear; and the 2nd defendant’s bus bearing registration number HS539. At the time the weather was fine, the road surface was dry and the traffic was smooth. Tsuen Wan Road at the location of the accident was a three lane one-way carriage way: the left lane merged with the branch lane and the second and third lanes were separated by double white lines. The speed limit was 70 kmh. At the time, he was driving at about 40 kmh on the second lane. When he approached the place of the accident, the vehicle in front suddenly reduced speed to almost stopping and appeared to change lane. He wanted to change lane in order to avoid it but there were vehicles on the other lane. He slowed down to several kilometres per hour and then stopped for 2 to 3 seconds. During this time the rear of his vehicle was bumped by the second vehicle. He was not injured. When he alighted from his vehicle he saw that the rear of his trailer was damaged whilst the three other vehicles were stopped behind his vehicle on the same lane. All the vehicles were damaged and the drivers of the second and third vehicles were injured. The police later arrived to deal with the matter. 16. The driver of the second vehicle, one, Chan Ka Lun, also made a statement to the police on 15 August 2014 about this occurrence [E/346-347D]. His description of the time and location of the accident, and the weather and the traffic at that time was the same as Chan Wai Lun’s description. At the time he was driving on the second lane at about 60 kmh. When approaching the place of the accident he saw that the vehicle in front stopped with its hazard warning lights on. He switched on his hazard warning lights and reduced the speed to several kilometres per hour following the vehicle in front. He then stopped for 1 to 2 seconds but was bumped by the vehicle travelling behind him which caused his vehicle to move forward and hit the rear of the trailer. He suffered injury to his back. When he got off from his vehicle to check, he saw that the front and rear of his vehicle was damaged, the third vehicle was stopped behind his vehicle, and the bus was stopped behind the third vehicle. The driver of the third vehicle was also injured. The police later arrived at the scene. 17. The plaintiff also made a statement to the police on 13 August 2014 [E/344–345E]. His description of the time and location of the accident, and the weather and the traffic at that time was the same as Chan Wai Lun’s and Chan Ka Lun’s description. There was traffic congestion on the first and second lanes and the traffic was smooth on the third lane. He was driving on the second lane at about 30 kmh. When approaching the place of the accident he saw that the first and second vehicles in front had stopped. He also stopped for 2 to 3 seconds. He saw from the rear view mirror that the 2nd defendant’s bus had not reduced its speed. The bus hit the rear of his car and caused it to move forward and bump the rear of the second vehicle. He then stated: “After the accident, I suffered from injuries in my right eye, neck and waist. When I got off the car and checked the damage to the car, I saw that the front and rear of my car were damaged…”. The first and second vehicles, his vehicle and the fourth vehicle, the bus, were all stopped on the same lane. All the vehicles were damaged and he and the driver of the second vehicle were injured. The police arrived to deal with the matter. He was sent to Princess Margaret Hospital (“PMH”) for treatment by an ambulance. He was hospitalised for treatment for 4 days. His car had fully stopped before the collision and had not struck the second vehicle. The distance between his car in the second vehicle was about 2 metres. There was only one collision throughout the whole process. He heard a bang sound at the back of his car and felt the back of his car being hit. His car was caused to rush forward and hit the back of the second vehicle. He saw the bus bump into his car without reducing its speed. As the airbag of his vehicle bounced out, he was unable to see the crashing of the first and second vehicles. The accident was caused by the bus driver who failed to stop. 18. The police took a cautioned statement from the 1st defendant on 26 August 2014 [E/351A–351E] in which it was recorded that the 1st defendant did not want to say anything and did not want to answer any questions. 19. Police constable, PC 6857, Lee Chi Hang, made police statement on 15 July 2014 [E/351F–351M] in which he stated that he had reached the scene of the accident at 2:55 pm on 27 June 2014. He noted that the sequence of the vehicles involved in the accident was RJ5615 (V1), RU2729 (V2), HJ3269 (V3) and HS539 (V4). The driver of V3, the plaintiff, had a red and swollen right eye. When he asked the plaintiff at the scene about the cause of the accident, the plaintiff said “I was also able to stop in time. The distance was one car space. Then I saw the bus ramming into my car.” 20. The brief facts of the case dated 27 January 2015 in respect of the charge of careless driving against the 1st defendant contained a summary of the statements made to the police by Chan Wai Lun, Chan Ka Lun, the plaintiff and PC Lee Chi Hang. The location of the accident on Kowloon bound Tsuen Wan Road outside Wing Kin Industrial Building was a three-lane one-way road which merged with a slip road next to the first left lane. There were double white lines dividing the second left lane and third left lane. The speed limit was 70 km. At the time of the accident, it was not raining and the road surface was dry and intact. The traffic flow was smooth. 21. The photographs taken by the police at the scene of the accident showed the damage to the vehicles concerned [E/348-351]. The top photograph on E/348 showed the damage sustained by the rear of the second vehicle, a private car: the back and the rear fender and the rear license plate of the vehicle were significantly dented. The bottom photograph on E/348 showed the damage sustained by the front of the plaintiff’s vehicle, also a private car: the front license plate and the front offside fender and front offside bonnet sustained damage. The top photograph on E/349 showed the second and third vehicles and the bus stopped in the second lane of this straight section of road. The bottom photograph on E/349 showed that the damage sustained by the rear of the plaintiff’s vehicle which was also dented, but not as significantly as the damage sustained by the second vehicle. 22. Most significantly, the top photograph on E/350 showed the plaintiff on a stretcher with his eyes open. This photograph was taken at the scene of the accident before the plaintiff was conveyed by ambulance to hospital. Injuries and Treatment 23. A clear and concise summary of the medical records detailing the injuries and treatment of the plaintiff after the accident appears in the joint neurological report of Drs Brian Choa and Edmund Woo dated 19 October 2017 which I set out as follows [D/231-240]: “2.1. The accident of 27th June 2014 and its aftermath On 27th June 2014, Yeung was injured in a traffic accident in Tsuen Wan. He was the driver of a stationary 7-seater van (HJ 3269) which was collided into at its rear by a bus. His van was pushed forwards and collided with another motor vehicle (a mini-cooper) in front. The airbag of his van was released and he hit his right face against the airbag. In his own statement of 13th August 2014 to the police, he stated that he could get out of his van to inspect the damages (我落車查看…). Photographs taken at the accident site shows that the front and the back of his van were somewhat dented; it by no means concertina’d. He was taken to the Princess Margaret Hospital (PMH) and admitted under the surgeons. It was noted under the section of past medical history that he had sciatica with follow-up at the family medicine clinic (he also had a past history of haemorrhoids, sebaceous cyst at the right groin and gastritis). He reported being hit on the face by the airbag. The notes specified there was no other injury. He had no loss of consciousness, nausea or vomiting. He had mild dizziness but he could walk after the accident. Examination showed he had a full Glasgow Coma Scale score of 15/15 and normal power in all 4 limbs. The right face was erythematous and tender. The abdomen was soft and the anal tone normal. X-rays of the facial bones showed air-fluid levels in the right maxillary sinus but no fracture. X-rays of the cervical spine showed normal alignment. CT (computed tomographic) brain scan was normal. He complained of occipital pain, neck pain and left hip pain while in hospital; there was no cervical spine tenderness. It was further specified that he had no limb weakness or numbness and he was ambulatory. He was discharged on 30th June 2014. 2.2 Follow-up with the surgeons of PMH He attended follow-up visits at the surgery clinic of PMH. The background history of sciatica with follow-up at the family medicine clinic was again noted. When first seen on 17th July 2014 (3 weeks after the traffic accident), he reported occasional dizziness and right peri-orbital pain with subjective decrease in visual acuity (he could not read small prints). He also complained of right lower limb numbness. Examination showed power in the right lower limb at grade 4/5 (it was stated that such weakness was the same as before the traffic accident). He had mild tenderness over the right zygoma but there was no cervical spine tenderness. Report X-rays of the face showed no air-fluid level or fracture. X-rays of the lumbosacral spine showed mild anterior osteophytes but the disc spaces were not narrowed. When seen again on 28th August 2014, he complained of low back pain shooting down to the right lower limb which was numb and he needed crutches to walk since the accident. He also complained of blurring of vision with increased lacrimation in the right eye. Examination showed numbness in the right lower limb and its power was grade 2-3/5. When seen on 20th November 2014, power in the right lower limb was grade 4/5. He had no sensation in the S1 (first sacral) region. Ophthalmological and orthopaedic assessments were arranged. 2.3 Physiotherapy at the Kwong Wah Hospital (KWH) He was seen by the physiotherapists at KWH from 20th October 2014. After 6 months of treatment, he reported an overall 30 to 40% improvement. He still had back pain and right leg numbness. When assessed on 13th January 2015, power was grade 4+/5 in the right lower limb and grade 5-/5 in the left lower limb. He could walk independently with 1 elbow crutch. He was discharged after the last session on 25th March 2015. 2.4 Orthopaedic assessment at KWH He attended the orthopaedic clinic of KWH on 11th December 2014. He complained of low back pain and weakness/numbness in the right lower limb. Examination showed tenderness over the lumbar spine. He demonstrated numbness and weakness from the L2 (second lumbar) to the S1 segments with power at grade 3-4/5 on the right. X-rays of the lumbosacral spine showed loss of lumbar lordosis. MRI (magnetic resonance imaging) scan of the lumbosacral spine on 19th March 2015 showed no spinal cord or nerve root compression. When seen on 2nd April 2015, power in the right lower limb had improved to grade 5-/5. When seen on 14th May 2015, He reported another fall in late April 2015 for which he was admitted to the Queen Elizabeth Hospital (QEH). Examination showed power at grade 0/5 in the right lower limb and grade 4/5 in the right upper limb. MRI scans of the brain, the cervical spine and the lumbar spine performed subsequently were all normal. When last seen on 11th November 2016, he could walk indoors with a frame and a knee-ankle-foot-orthosis. 2.5 Occupational therapy at KWH He was assessed by the occupational therapists at KWH on 4th February 2015 and 27th March 2015. He complained of right shoulder pain when elevating the right arm, pain over the thoracolumbar spine and numbness/weakness in the right leg. Examination showed his tolerance in sitting was 30 minutes while tolerance was 20 minutes each for walking and standing with one elbow crutch. His gait was inconsistent. Hand grip was 18 kg-force on the left and 10 kg-force on the right. It was opined that his working capacity did not meet his job demand (as a chauffeur). 2.6 The fall of 29th April 2015 and its aftermath He was seen at the Accident & Emergency Department of QEH at 2:21 am on 29th April 2015. He fell at home at 1:35 am. He landed on his right buttock. He complained of back pain and numbness in the right leg. Examination showed tenderness in the low back. His right leg was numb to touch. X-rays of the lumbosacral spine showed no fracture. He was admitted under the orthopaedic surgeons. A more detailed history taken by the orthopaedic surgeons pointed to previous sciatica since 2013; he had right-sided low back pain radiating to the right lower limb. It was also stated he had a past history of LUTS (lower urinary tract symptoms). As for the fall, he slipped in the toilet while urinating holding onto crutches. Examinations showed no movement of the right lower limb when instructed to do so but active movements including flexion of the right hip and extension of the right knee were noted at other moments. There was inconsistent sensory loss. MRI scan of the lumbosacral spine showed mild disc bulges at several levels but without any nerve tissue compression. He was given analgesic medications and physiotherapy. He was evaluated by the neurologists. He complained of pain over the whole spine, numbness in the ulnar aspect of the right upper limb, paraesthesiae in the right thigh and no sensation below the right knee. Examination showed normal muscle tone and no muscle wasting. Power was pain-limited at grade 4/5 in the right upper limb but grade 0/5 in the right lower limb. The deep tendon reflexes were normal and symmetrical. Sensation was absent in the right L1 (first lumbar) to L5 (fifth lumbar) dermatomes. He was also seen by the ENT (Ear, Nose & Throat) surgeons while in hospital. He complained of right hearing loss since the traffic accident of 27th June 2014. Pure-tone audiogram on 20th May 2015 showed bilateral sensorineural hearing loss, moderate on the left and severe on the right. When he was discharged on 26th May 2015, he walked with a frame. Subsequent nerve conduction study on 5th June 2015 at QEH showed prolonged distal latency and reduced compound muscle action potential in the left peroneal nerve; conduction in the nerves of the right lower limb was normal. It was opined there was no electrophysiological evidence to account for the right lower limb symptoms. 2.7 Neurological assessment at KWH He was admitted to KWH under the orthopaedic surgeons on 11th June 2015 and 27th July 2015 for consultation with the neurologists in respect of the diagnosis of a possible whiplash injury. His accident of June 2014 and his fall in April 2015 were noted. Examination showed weakness in the right upper and lower limbs but there was no definite upper motor neurone or lower motor neurone sign. Power was grade 0/5 in the entire right lower limb but the deep tendon reflexes were normal and the plantar response was flexor. Review of the MRI scan of the lumbosacral spine taken earlier in March 2015 showed only minimal disc bulge at the L4/5 (fourth/fifth lumbar) and L5/S1 levels with only minimal thecal sac indentation. MRI scan of the brain and the brainstem on 14th July 2015 showed only a few tiny foci of nonspecific signal changes in the cerebral hemispheres. MRI scan of the cervical spine on 2nd October 2015 showed only mild cervical spondylosis with a disc bulge at the C6/7 (sixth/seventh cervical) level, worse on the left, and possible encroachment on the left C7 nerve root. Somatosensory evoked potentials on 3rd December 2015 showed normal responses in the C5/6/7/8 (fifth/sixth/seventh/eighth cervical) dematomes, in both median nerves and in both posterior tibial nerves. Nerve conduction study on 7th June 2016 showed normal conduction in all the nerves of the upper and lower extremities. No conclusive diagnosis could be offered. 2.8 Psychiatric assessment at the Kwai Chung Hospital (KCH) He was seen at the common mental disorder clinic of KCH from 13th August 2015. He complained of insomnia for 7 to 8 months and headache for more than a year. His mood was low and he was irritable. He had fear of being hit from the back. He could commute on the bus and the MTR. He had only fleeting suicidal ideas. He was not violent. He had vague auditory hallucination but no visual hallucination. The diagnoses of post-traumatic stress disorder and psychotic depression were made. He was given antidepressant (sertraline) medication. At the next visits on 10th September 2015 and 8th October 2015, he was calm with a low mood. His speech was relevant and coherent. He was neither suicidal nor violent but he became wheelchair-bound. He was admitted to KCH on 5th November 2015 because of suicidal ideas. When reviewed on 27th October 2016, his mood was neutral. His sleep was improved. He had no active auditory hallucination. When last seen on 12th January 2017, his mood was irritable with poor sleep. He was still bothered by auditory and visual hallucinations. Examination showed he was calm with relevant and coherent speech. His mood was on the low side. He was not suicidal. His insight was partial. He was given antidepressant (mirtazapine, venlafaxine), antipsychotic (quetiapine) and anxiolytic (clonazepam, propranolol) medications. 2.9 ENT assessment at QEH He attended follow-up visits at the ENT clinic of QEH on 4th November 2015. Examination showed both eardrums were normal. Pure-tone audiogram on 4th November 2015 showed mild left sensorineural hearing loss and severe right sensorineural hearing loss. His pure-tone audiogram repeated on 19th June 2016 showed moderate left sensorineural hearing loss and profound right hearing loss but the test had fair to poor reliability as he did not co-operate and intended to exaggerate his hearing loss. Examination of his eardrums remained normal. Clinical voice test showed no response on the left ear while the hearing threshold was 30 to 60 dB on the right ear. Pure tone audiogram was repeated on 22nd June 2016 and it showed mild left sensorineural hearing loss; right hearing loss was exaggerated and Stenger test (a test for differentiating feigned from true unilateral hearing loss) was positive on the right. Cortical evoked response audiometry on the same day showed a hearing threshold at 40 dB from the 500 Hz to the 2,000 Hz range on both sides. He was discharged from further follow-up after the last visit on 6th July 2016. 2.10 Rehabilitation at the Kowloon Hospital (KH) He attended physiotherapy from 9th December 2015 and occupational therapy from 5th January 2016 at the Kowloon Hospital. In April 2016, he was fitted with a right knee-ankle-foot-orthosis. 2.11 Urological assessment at QEH He attended the urology clinic of QEH from 28th April 2016. He complained of deteriorating urinary symptoms since the accident of 27th June 2014. He had urgency and incontinence. He needed 2 to 3 diapers a day. Examination showed power in his right lower limb at grade 0/5 while power in the right upper limb was grade 3-4/5. He was wheelchair-bound. His anal tone was normal and he had a mildly enlarged prostate. His renal function on 28th April 2016 was normal. Urodynamic study on 16th May 2016 (he signed the consent form himself with the same signature as he did on the police statement of 13th August 2014) showed detrusor external sphincter dyssynergia with insignificant post-void residual urine. When seen on 21st July 2016, he complained of erectile dysfunction since the traffic accident; he was prescribed erectogenic (sildenafil) medication. Ultrasound scan of the urinary system on 25th July 2016 showed no abnormality. When reviewed on 15th September 2016, he was given an alpha-blocking agent as symptomatic treatment for his urinary symptoms. 2.12 Assessment by the clinical psychologist at the East Kowloon Psychiatric Centre He was seen by the clinical psychologist at the East Kowloon Psychiatric Centre on 31st October 2016. His accident of 27th June 2014 was noted. He came in a wheelchair. He complained of headache, poor sleep and frequent nightmares. He presented with a low mood. He wept when talking about his loss of occupational functioning. He heard high-pitch machinery sound like that of a brake. He avoided going out for fear of further traffic accidents. His speech was relevant and coherent. He was not suicidal. The diagnoses of post-traumatic stress disorder and psychotic depression were made. He was given psychotherapy. 2.13 Ophthalmological assessment at the Hong Kong Eye Hospital He attended the ophthalmology clinic of the Hong Kong Eye Hospital. When last seen on 22nd November 2016, his best corrected visual acuity was 0.6 on the right and 0.7 on the left (improved compared to July 2015). Optical coherence tomography, multifocal electroretinogram and visual field testing were all normal. The diagnosis of mild refractive error was made.” The Expert Evidence 24. The plaintiff was jointly examined by medical experts from four different areas of speciality: The Orthopaedic Expert Evidence 25. The joint orthopaedic report of Dr Lam Yan Kit and Dr David Cheng was dated 29 June 2017. There was no disagreement between the two experts. They expressed the following opinions [D/224-226]: “3. … From review of the medical documents, we opine that the diagnosis was a contusion of the face and possibly a mild neck soft tissue injury, without any complications. … We further opine that the accident should be the sole cause of the injury. 4. For this injury, Yeung was treated conservatively, and appropriate treatment. Subsequent Development 5. However, Yeung’s condition gradually deteriorated…. He ended up in this present condition, paraplegia of the right lower limb, and paraparesis (partial paralysis) of the right upper limb. Both motor and sensory functions are affected. 6. The treating orthopaedic surgeons, both at KWH and QEH did their very best to try to find out the cause of such a clinical presentation. Exhaustive investigations … were done. The opinion of neurologists was also sought. All investigations drew a blank. … 7. We opined that the treatment given to Yeung is appropriate and complete. In our joint examination today, we also found Yeung in a state of right lower limb paralysis, and [paraparesis] of the right upper limb. These findings are genuine. … We, like our colleagues in the HA hospital, are also at a loss of the cause of such a presentation. There is simply not a single orthopaedic condition that fits. Relation with the captioned accident of 27.6.2014 8. We consider: (i) The injury was described as a facial contusion and possibly a minor neck injury without complications. (ii) The neck symptom resolved quickly. (iii) Symptoms of right lower limb weakness and paraesthesia surfaced weeks after the accident, unlikely to be due to any acute involvement. (iv) From our experiences, it is highly unlikely that an injured person not to inform his treating doctors of any complaints, even if not on the first encounter, but during subsequent stay in hospital. (v) There was another episode of back injury some 10 months (April 2015), causing rapid deterioration, and also new symptoms of neck and right upper limb paraparesis. (vi) We note the neurologists had put forward a possible diagnosis of whiplash injury to neck. With due respect, we opine that this does not hold water. Yeung’s clinical presentation suggests severe neurological involvement of a broad degree, either in the spine at different levels, or the brain. Whiplash injury referred to soft tissue injury of the neck. Furthermore, symptoms appear much later.” 9. All considered, we have to say we do not know the cause. However, we must confess that we could not relate Yeung’s present state to any orthopaedic injury with such a presentation. We are unable to say that the captioned accident and/or the subsequent back injury is the sole cause.” [My emphasis] 26. Although the orthopaedic experts considered that the plaintiff’s symptoms and condition were genuine, they were unable to say what could have caused such a condition which did not fit any orthopaedic injury that they were aware of. The Neurological Expert Evidence 27. The neurological experts’ joint report of Dr Brian Choa and Dr Edmund Woo was dated 19 October 2017. They were agreed in almost all aspects, their only disagreement of note being Dr Choa’s opinion that a three-month sick leave was reasonable which differed from Dr Woo’s opinion that one month was sufficient [D/263]. I have already set out above, their review of the medical records. They also reviewed the X-rays and the various scans that were taken. The positive air fluid level in the right maxillary sinus shown on the X-ray of 27 June 2014 was resolved, as shown by the X-ray of 17 July 2014. All other X-rays and scans were normal save for mild degenerative changes of the thoracic and lumbar spines and a mild disc bulge at C6/7 and L4/5 levels without any spinal cord or nerve root compression [D/251-253]. 28. In the course of their examination, they observed: (1)“Yeung was evasive in his answers, mendacious in his account of events and prone to claim that all his symptoms were getting worse”. [D/247/paragraph 6.2] (2)“He used the right pincer grasp to hold onto a pencil to write his Chinese name and to copy a geometric figure in a feeble and clumsy manner; he held the pencil steadily in the right pincer grasp and he moved the piece of paper with his left hand so as to produce the characters and the drawings. When not being formally tested, Yeung was seen to use his right arm to raise himself from his wheelchair by forcefully extending at the right elbow; he also used his right hand in an agile fashion to assist in unlocking the knee-ankle-foot orthosis.” [D/250/paragraph 6.4.3] (3)“When transferring onto the reclining chair, there was a brief interval of at least 1 to 2 seconds when the plaintiff could actually stand stably with the knee-ankle-foot orthosis detached.” [D/251/paragraph 6.4.4] (4)“Head turning to one side is carried out by the contralateral sternomastoid muscle, i.e. the left sternomastoid muscle turns the head to the right while the right sternomastoid muscle turns the head to the left. The layman is not aware of this paradoxical crossed innervation and the malingerer will demonstrate head turning to the wrong side. Yeung demonstrated poor head turning to the right; this indicates weakness in the left, not the right, sternomastoid muscle.” [D/257-258]. 29. As regards the traffic accident itself, Dr Choa’s opinion was that the plaintiff’s current account that he had lost his memory for 30 to 60 minutes was fabricated, given his statement to the police to the contrary and the negative investigations, including more than one head scan and a full spinal MRI scan, which supported his opinion. Dr Woo was of the opinion that he did not have any head injury, brain injury or concussion and that any soft tissue injury to the neck should have resolved within a month [D/253-254]. 30. As regards the fall of 29 April 2015, Dr Choa was of the opinion that there was no mechanism whereby a fall onto his buttocks could lead to a paralysis of the limb in the absence of evidence of extensive nerve damage. Dr Woo was of the opinion that he sustained, at most, soft tissue injury to the low-back in the fall which should have resolved within a month [D/254-255]. 31. Dr Woo summarised the plaintiff’s disability is as follows. He complained of residual headache, impaired memory, irritability, neck pain and weakness in his right extremities. He claimed that the neurological deficits worsened progressively since the traffic accident, during the months between the accident in June 2014 and the fall in April 2015, and that the deficits in the right lower extremity became profound as a further result of the fall [D/257]. 32. Dr Choa’s opinion on the plaintiff’s residual disability was: (1)only severe spinal cord compression or a large brain lesion could account for his clinical picture and these had been excluded; (2)the series of negative results since 2014 have shown unequivocally that there was no organic basis for his alleged visual impairment, hearing loss and paralysis of right-sided limbs; (3)an MMSE (mental examination) score of 16/30 would identify a patient is being markedly demented which the plaintiff clearly was not; (4)although the difference of 3 cm in the girth of his two thighs was just outside the normal limit of 2.5 cm this could be the residuum of his long-standing sciatica or the result of disuse or a combination of both - his normal knee and ankle tendon reflexes were inconsistent with significant damage to his femoral or sciatic nerves; (5)the plaintiff’s evasive behaviour, and his observed ability to use his allegedly paralysed limb when not normally tested strongly suggest that his disability is mostly if not entirely feigned [D/255-256]. 33. Dr Woo’s opinion on the plaintiff’s residual disability was: (1)the plaintiff’s neurological examination demonstrated many unusual features; (2)the magnitude of cognitive impairment demonstrated by the plaintiff was observed only after a severe head injury in which there was significant, diffuse brain parenchymal damage (damage to the functional tissue of the brain) and in whom primitive reflexes and/or paratonia (inability to relax muscles) were evident on examination to corroborate frontal lobe dysfunction. In the plaintiff’s case, he had no head injury, there was no objective sign of frontal lobe dysfunction on examination, and there was no radiological evidence of any parenchymal brain damage; (3)the layman is not aware of the paradox that the left sternomastoid muscle turns the head to the right while the right sternomastoid muscle turns the head to the left and the malingerer will demonstrate head turning to the wrong side - the plaintiff demonstrated poor head turning to the right which indicated weakness in the left, not the right, sternomastoid muscle; (4)motor deficits in an extremity associated with a brain or cervical spine injury is associated with hypertonia (increase in muscle tone) and hyper-reflexia (hyperactive deep tendon reflexes). In the plaintiff’s case there was no associated hypertonia or hyper-reflexia; (5)if the plaintiff’s diffuse weakness and sensory loss in the right upper extremity were genuine, they would indicate an extensive lesion spanning from C5 down to T1 on the right. However, the objective signs of such weakness and sensory loss, namely, muscle wasting and absent deep tendon reflexes, were absent in the plaintiff’s case - he had no muscle wasting and all deep tendon reflexes were preserved. Further, MRI scans showed no compression of exiting nerve roots and no evidence of any nerve root lesion; (6)the plaintiff demonstrated profound, total sensorimotor loss in the right lower extremity. If genuine they would indicate a massive lesion spanning from L2 down to S1 nerve roots, all and only on the right; yet there was only mild muscle wasting in the right thigh and all deep tendon reflexes were preserved. Again MRI lumbar scans showed no compression of exiting nerve roots and other studies showed no evidence of nerve root lesion; (7)notwithstanding his claim, the plaintiff demonstrated for at least a second or two that he could bear weight on the right lower limb without the orthosis; (8)the distribution of right-sided sensory loss was non-physiological; (9)there were multiple inconsistencies between the plaintiff’s version of events and the recorded information; (a) he claimed to have been rendered unconscious upon the impact and not regain awareness until 2 hours later but, in his police statement, he said he got out of his vehicle and inspected the damage at the scene; (b)he claimed he was confined to bed and needed two canes to walk when discharged from PMH but the hospital notes clearly stated that he was ambulatory with no limb weakness; (c) he claimed that his right hand was too weak to write and he demonstrated marked clumsiness in writing at their examination which was incompatible with the signatures in his police statement in August 2014 and the consent form in May 2016 which were clearly made in an agile fashion; and (d)he claimed to have urgency of micturition and he had to micturite every hour but he sat through the almost 2-hour interview without the need to go to the toilet. Dr Woo concluded that the plaintiff was a malingerer. The deficits he demonstrated were completely nonphysiological. There was marked discrepancies between his version and the medical records. All these indicated that his deficits were feigned [D/257-261]. The Psychiatric Expert Evidence 34. The joint psychiatric report of Dr Benjamin Lai and Dr Gabriel Hung was dated 23 November 2017. Dr Hung offered the following opinions [D/280-286]: “16.3 Before proceeding to consider the possible psychiatric diagnosis, I will address the issue of Mr Yeung’s reliability. Most psychiatric diagnoses are based on reported and subjected symptoms. In a treatment situation the psychiatrist does not doubt the veracity of the patient’s symptoms. This is posited on the shared goal of the patient’s desire to get well and the psychiatrist’s commitment to help the patient get well. Whereas in litigation the plaintiff’s goal is to obtain compensation and the responsibility of the expert psychiatrist is to assist the court. In this context the expert psychiatrist has to look at all the information made available in a critical manner and to form a view on the reliability of the plaintiff. The core of reliability is consistency. 16.4 The most important issue in this case is the reliability of the physical disability. The orthopaedic experts could not find a reason for the physical disability. The experts in neurology agreed that his physical disability as feigned and there is no physical explanation for his symptoms. Dr Woo went [on] to state that Mr Yeung is a malingerer. 16.5 In the joint examination, there was also an inconsistency in the sequence of events. The medical records indicated that there were two injuries (the index Accident on 27 June 2014 and the fall at home on 29 April 2015). After the Accident, the medical records indicate that he was able to walk with a stick and his right leg was not immediately paralysed. However, he told us at the joint examination that his right leg was immediately paralysed after the Accident. The medical records indicated that it was after the fall at home on 29 April 2015 that he lost all power and movement in the right leg. 16.6 It was also very strange that he was unable to answer simple questions about his background (e.g. if his parents were alive, how many siblings he has, whether he had a second marriage, and his answered that he had no children when he had a son according to the medical records). This most likely indicate lack of effort or unwillingness to answer and not due to true loss of memory, as he was clearly able to answer other questions without any difficulty. … 16.8 … I will proceed to discuss the diagnosis on the assumption that he was truthful in his meetings with the treating psychiatrists and also at the joint examination. … 16.20 the basis for his psychiatric symptoms is the severe physical disability following the Accident and the subsequent fall at home. The orthopaedic experts could not find a reason for his severe physical symptoms. However, both neurological experts have opined that his symptoms were feigned. Accordingly, if his physical disability is feigned, the basis for his psychiatric symptoms also fails. 16.21 Therefore, in this case, I have doubts regarding the reliability of the diagnosis due to the presence of either malingering or an abnormal sick role. It is not advisable in this case to take his self-reported psychiatric symptoms in full. … 16.26 Therefore, despite Mr Yeung volunteering many symptoms of Major Depressive Disorder, it is unlikely that he was actually suffering from such severe symptoms. … 16.39 His main impairment at the moment is his severe physical impairment. His psychiatric symptoms are very mild compared to his physical condition.” [My emphasis] 35. Dr Lai offered the following opinions [D286-290]: “17.8 In the assessment of his current psychiatric condition, the following have been considered: … (h) … if according to the opinions of the neurology experts that from the neurological perspective there should not be any restriction on his activities of daily living and there is no loss of earning capacity, his current physical complaints should not be as severe as he has complained of, and thus his current psychiatric symptoms would likely be less severe as he has tried to portray. … (j) In the current joint psychiatric examination he said that he could not recall where he was born, where his parents were currently, whether they were currently alive, how many siblings he had, the age he came to Hong Kong, the year he left school, when he got married, his wife’s age, whether his current marriage is his second marriage and his home address. On the basis that he did not have severe degree of head injury at the time of the Accident and clinically he appears to have a mild to moderate degree of depressed mood, it is not expected that he would be affected in his long term memory to the degree of not being able to recall background information and affected in his recent memory to the degree of not being able to tell his home address. The inconsistency raises a caution that he has not tried his effort in answering the doctors’ questions in the current joint psychiatric examination or he has tried to impress the doctors of the seriousness of his psychiatric condition. 17.9 Based on the information from the available medical documents, current examination findings and the above discussion, there is inconsistency observed in the current joint psychiatric examination. If he has permanent physical disabilities as opined by the orthopaedic experts it is compatible that he is currently suffering from mild to moderate degree of psychiatric symptoms. However, if according to the neurology experts he has no restriction in his activities of daily living and there is no loss of earning capacity, it is compatible that he would be suffering from only mild degree of psychiatric symptoms.” The Ophthalmological Expert Evidence 36. The joint ophthalmological report of Drs Tsui Chung Wah and Dr Ng Wing Ho was dated 18 January 2018 [D/317-336]. During their joint examination, the experts were unable to obtain a reliable visual acuity test result: “Unaided visual acuity Right eye: 20/200; improved to 20/60+ with malingering test Left eye: 20/60; improved to 20/30 with malingering test Best-corrected visual acuity Right eye +2.00 = 20/50 (i.e. 2.00 dioptres of longsightedness) but fluctuated from 20/100 to 20/50 with malingering test” [D/324] The experts were in total agreement in their joint report [D/327-329]. They noted that the plaintiff complained of progressive blurring of vision in his right eye since the accident. His visual acuity was found to be impaired initially but was later confirmed to be 6/9 bilaterally by malingering test, which was satisfactory. Various ophthalmic investigations were conducted and the results were all normal. The only positive clinical finding observed was the mild retinal pigment epithelial changes at the macula of the right retinal fundus on 10 June 2015. All subsequent investigations conducted were essentially normal. During the joint examination on 15 December 2017, a Spatial Domain Optical Coherence Tomography (“SD-OCT”) examination was conducted that revealed at the fovea of the right eye a small retinal pigment epithelial detachment (“PED”) which was not present in the previous SD-OCT examination report of 31 August 2015. PED is a primary retinal disease not caused by ocular trauma. In the plaintiff’s case, his right eye PED could be idiopathic (a disease or condition which arises spontaneously or for which the cause is unknown) because it was unilateral and there was absence of clinical signs of other retinal diseases. As a result, his right eye vision was further blurred with distorted images. The newly diagnosed PED in the right eye and its subsequent visual impairment are not related to the accident on 27 June 2014 and his current ocular complaints are not attributable to the accident. Judging from the available normal clinical findings obtained in the left eye, it is highly probable that the plaintiff should be able to achieve a normal best corrected visual acuity of 6/6 in his left eye. The best corrected visual acuity of his right eye was 20/50 (i.e. 6/15) with metamorphopsia (a visual defect that causes linear objects to look curvy or rounded) which was due to the newly diagnosed PED and was not related to the accident. 37. The experts concluded that despite the plaintiff’s complaint of progressive blurring of vision in his right eye since the accident, repeated ophthalmic examinations and investigations conducted failed to delineate any pathology in his eyes. During the joint examination on 15 December 2017, he was found to be suffering from right eye PED which was newly diagnosed and not related to the accident. The Plaintiff’s Evidence 38. In his first witness statement dated 13 February 2018 [B/93-128], the plaintiff gave an account of the accident which was similar to the account contained in his statement to the police. He then stated: “6. …The airbag of the steering wheel of the private car immediately popped out and hit the left of my face. Everything turned white before my eyes and I passed out afterwards. 7. Soon afterwards, passers-by pried open the driver’s door and woke me up and then held me to leave the compartment. At that time, I already felt pain over my right eye, right face, neck and waist/back and numbness of right trunk. …” He was sent to the AED of PMH and admitted to the Department of Surgery. He was discharged on 30 June 2014 and had to walk on a pair of crutches. In paragraphs 9 to 13 of his witness statement, he described the various treatments he received for his right eye, right ear and weakness of spine. In paragraph 16 of his witness statement, he described the fall he had at night on 28 April 2015. He had severe numbness and weakness on his right leg and he was admitted to the Department of Orthopaedics and Traumatology of QEH. He required a walking frame when he was discharged on 26 May 2015 and received a recommendation to purchase a wheelchair. In paragraphs 17 to 27 of his witness statement, he detailed the various treatments he received at various hospitals including treatment for incontinence and psychiatric treatment. In paragraph 28 of his witness statement, he listed his disabilities as follows: (a) paralysis of right leg, loss of sensation and mobility; (b) persistent pain on back; (c) persistent severe pain on the back of the neck; frequent headache and dizziness; (d) weakness of right hand, numbness of thumb and index finger; loss of sensation of middle finger, ring finger and little finger; (e) impaired vision and persistent discomfort of right eye; pain in the eye and watery eyes; vision getting poorer; (f) impaired hearing of right ear which was getting poorer - advised to use hearing aid; (g) urinary incontinence, particularly at night; diaper was needed; (h) severe headache at night preventing him from falling asleep; (i) feeling despondent, slow in response, reduced concentration, occasional auditory hallucination, frequent dizziness and once had suicidal idea; (j) wheelchair-bound; and (k) inability to work. Except for toileting and eating, he required his wife’s assistance in all daily activities. He could no longer swim, fish and play badminton nor engage in any sport activities. He could not engage in sexual intercourse after the accident. In paragraph 33 of his witness statement, he described his mental state. In paragraphs 34 to 43 of his witness statement, he supported his various claims for loss and damage. 39. In his supplemental witness statement dated 12 October 2018, the plaintiff provided further information in support of his claims for aids and equipment and the cost of renovations and alterations [B/129-139]. 40. In paragraph 3 of his opening statement dated 9 June 2020, the plaintiff stated that the 1st defendant’s bus rammed into the rear of his private car at a speed of over 80 kmh. The tremendous impact just pushed his private car into the rear of the vehicle in front and his private car was turned instantly into a wreck. The bodywork was crushed at both ends and left almost 2 feet shorter. The engine was knocked down and machine oil was all over the road surface. As the airbag exploded in the car instantaneously, his view was full of white smoke. His right eye was struck by the plastic cover of the airbag and he fell unconscious as a result. At the same time, the driver’s seat was severely damaged and deformed. The offside car door was contorted and his right upper arm was in turn crushed. He sustained multiple bodily injuries, with those in his right eye, cervical spine, the back of his head and his lower back being more serious. He was then whisked off to PMH for emergency treatment. 41. He also stated in paragraph 12 of his opening statement that on the day when he went to give his statement to the police, he was very weak physically and his head was painful and dizzy as a result of the severe injuries he suffered. He finished giving his statement with difficulty. At that time, he did not read in detail the statement taken by the police. The contents of the statement taken by the police were prepared by them in advance. The statement taking was conducted by way of question-and-answer with the police asking the questions and him answering, giving the information he required. He did not have the chance nor any time to explain in detail what had happened. The format, wordings and contents of his statement and the statements of another two witnesses were basically the same. The contents of the statement did not accurately reflect what actually happened at the time of his accident. 42. In paragraph 3 of the plaintiff’s closing submissions dated 18 August 2020, the plaintiff sought to introduce additional evidence of receiving a threatening letter from a boy at about 8:30 am on 3 August 2020. The attachments to the plaintiff’s closing submissions included a copy of this threatening letter, a translation in English prepared by the plaintiff of the contents of the threatening letter, and a letter in English from the plaintiff addressed to the court dated 10 August 2020 the contents of which were similar to what was stated in paragraph 3 of the plaintiff’s closing statement. The full English translation of the letter is as follows: “Today in court you must cooperate with us, (and) admit that your disability is faked by you yourself, otherwise we will take action against you, you and your family will suffer more serious disaster.” The contents of the letter suggest that the threat to the plaintiff emanated from the defendants or from one of them. On 3 September, 2020, I directed the defendants to respond to the court in writing within 7 days regarding this new evidence and, in particular, whether it should be admitted after the parties had already closed their cases and, if admitted, what evidence, if any, the defendants would seek to adduce in response or rebuttal thereof. 43. On 10 September 2020, the defendants wrote to the court and submitted that this new evidence should not be admitted. It was irrelevant. It was obvious that the plaintiff had not been threatened to make any admissions. He vigorously denied the case put to him. Although the defendants doubted the veracity of the plaintiff’s evidence about the incident and the letter, the matter should best be left to be investigated by the police to whom the plaintiff had made a report. Although they did not seek to adduce evidence in rebuttal, they asserted that neither they nor their agents were in any way involved in the alleged incident. 44. In the exercise of my discretion, I admit into evidence the matters set out in paragraph 3 of the plaintiff’s closing submissions and the attachments to it, which I have set out in paragraph 42 above. As I am able to assess the veracity of the plaintiff’s complaint set out in paragraph 3 of his closing submissions without the need to hear or receive further evidence on this matter, I would not have granted leave to the defendants to introduce rebuttal evidence, had they sought to do. 45. In paragraph 8 of the plaintiff’s closing statement, he stated that the second defendant’s bus involved in the accident travelled at the speed of about 90 kmh and that it collided into the boot of his car without slowing down. 46. In her witness statement dated 18 May 2020 [B/140-144F], which was received into evidence as a hearsay statement, Madam Zhang Qiongyun stated that she had resigned from her 2 jobs as a manager in a restaurant and as a part-time salesperson in an electronics factory in Mainland China in order to take care of the plaintiff, her husband, in Hong Kong. 47. In his evidence in chief, in addition to relying on his two witness statements, the plaintiff gave further evidence complaining that what the doctors said to him orally was very different from the written medical reports; that lot of facts stated orally by the doctors had not been reduced into writing; and that they did not want to bear the responsibility of stating how serious his injuries were (Transcript/Day1/13A-17G). 48. When the trial resumed on the morning of 3 August 2020, the plaintiff confirmed the truth of his additional statement of 23 July 2020 which he had sent to court prior to the resumption of the trial (Transcript/Day2/22P-23K). The statement contained the additional evidence, in paragraph 4, of the plaintiff’s plan to consult an orthopaedic specialist to amputate his right leg so that the prosthesis could be fitted to him to enable him to stand and walk. During his cross-examination, the plaintiff stated that he had lost consciousness after the accident but he could not recall when he regained consciousness, whether at the scene or in hospital. When shown the photograph taken at the scene of him lying on the stretcher with his eyes open [E/350], he answered: “INTERPRETER: “The mere fact of my eye opening was not equivalent to my state of being sober. In the hospital, I have seen a lot of patients with their eyes wide open but they all lost their consciousness.” COURT: So your evidence is that although this is a photograph of you with your eyes open, you were unconscious. A.; That’s correct.” The plaintiff was also referred to his statement to the police and asked whether he knew at the time he signed it that he had to provide true and accurate statements [E/344, 345A]. He answered that he was not sober and had difficulty walking. His wife and elder sister accompanied him but his mind was not sober. The police officer told him to sign and he just did so. Whatever is told to him by the police, he will just do it. In addition, his eyes could not see well. He disputed the correctness of his police statement in which he had stated that he “got off the car and checked the damage to the car, I saw that the front and rear of my car were damaged” (E/344, 345, 345C). When referred to the statements that he had made to the experts during their examination of him [D/320, 241], he said that it was so long ago and that he could not remember what he had said. (Transcript/Day2/24M-32D). 49. The plaintiff did not give any further evidence in re-examination (Transcript/Day3/2D-3D). My Findings 50. I refer to the documentary evidence of the accident which I have set out in paragraphs 14 to 22 above. I find that this was an accident which occurred as described by the plaintiff and the 2 drivers, Chan Wai Lun and Chan Ka Lun to the police and as set out in the prosecution’s brief facts of the case against the 1st defendant on the charge of careless driving. The plaintiff submitted that the contents of the statement did not accurately reflect what actually happened at the time of the accident. The format, wordings and contents of the statements were basically the same and the contents of the statements were prepared by the police in advance. The statement taking was conducted by way of question-and-answer. I am not impressed by these submissions. It is quite natural for a police officer taking a statement about a traffic accident to assist a witness in giving his statement by asking him to first describe the date and time of the accident; the location of the accident and the speed limit at that location; the weather and the road and traffic condition at the relevant time; and then to describe how the accident occurred; the injuries suffered and the damage sustained by the vehicles involved; and what happened at the scene after the accident including the arrival of the police and the ambulancemen. In this case, each of the three witnesses, including the plaintiff, gave an account of how his own vehicle was involved in the accident. Each of them was able to stop his vehicle from colliding into the vehicle in front of him. The similarity of their accounts does not detract from the veracity of their accounts. 51. I find that, by reason of the fact that the vehicle in front of Chan Wai Lun’s vehicle suddenly reduced speed and appeared to be stopping, Chan Wai Lun, Chan Ka Lun and the plaintiff all slowed down and stopped their vehicles without colliding into one and other. However, the 2nd defendant, as a result of his carelessness, did not stop his bus in time which collided into the rear of the plaintiff’s vehicle, causing it to move forward and collide into the rear of Chan Ka Lun’s vehicle which, in turn, moved forward and collided in the rear of Chan Wai Lun’s trailer. The bottom photograph on E/348 showed the damage sustained by the front of the plaintiff’s vehicle: the front license plate and the front offside fender and front offside bonnet sustained damage. The bottom photograph on E/349 showed that the damage sustained by the rear of the plaintiff’s vehicle which was also dented. More significantly, the photographs showed, and I find, that the body of the plaintiff’s vehicle did not sustain any damage and, in particular, that the driver’s door and compartment was not damaged as a result of the collision and there was no engine oil leaking out from the vehicle onto the road surface. 52. I reject entirely the plaintiff’s evidence that the 1st defendant’s bus rammed into the rear of his private car at a speed of over 80 kmh (which, in his closing submissions, he said was 90 kmh), that the tremendous impact pushed his private car into the rear of the vehicle in front and turned his private car instantly into a wreck, that the bodywork was crushed at both ends and left almost 2 feet shorter, that the engine was knocked down and machine oil was all over the road surface, that the driver’s seat was severely damaged and deformed, and that the offside car door was contorted. I find that the 2nd defendant’s bus was not travelling faster that the plaintiff’s own speed of 30 kmh prior to the collision and that, because of inattention, he did not react quickly enough to prevent the collision. I also infer from the dent suffered by the rear of the plaintiff’s vehicle, which was not significant, that the 1st defendant did slow down the bus from its initial speed of 30 kmh, but not quickly enough to prevent the collision that occurred. 53. I refer to the medical records and reports in C/145-201 and to the summary of these medical records and reports appearing in the joint neurological report of Drs Brian Choa and Edmund Woo dated 19 October 2017 which I have set out in paragraph 23 above. I find that the contents of these medical records and reports are factually correct and that the said summary is an accurate summary of them. I do not accept the plaintiff’s complaints that what the doctors said to him orally was very different from their written medical reports; that lots of facts stated orally by the doctors had not been reduced into writing; and they did not want to bear the responsibility of stating how serious his injuries were. The medical records and reports I have seen in this case are in the standard form and contain concise information relevant to the condition of the plaintiff and the injuries suffered by him. I can discern no reason, and certainly there is no evidence, to support the plaintiff’s contention that the medical doctors and other medical personnel who examined and treated the plaintiff would deliberately conceal the true extent of his injuries; and that they did not want to bear the responsibility of stating how serious his injuries were. The plaintiff’s evidence that they did so is pure fabrication. 54. I have noted, in particular, the following medical records and reports which I found to be very relevant to the issues before me: (a) The extent of his injuries was clearly recorded in the notes of the AED of PMH which stated [F/545]: “Triage Assessment Condition on arrival: eye injury by airbag. LOCo (meaning no loss of consciousness). c/o [complain of] Rt [right] eye pain Clinical findings - Sprain neck - R face & eye hit over air bag - Blurred vision+ pain+ - No other injury - Limb weaknesso numbnesso (meaning no limb weakness nor numbness)” (b) He was admitted to the Department of Surgery on the same day. The medical report from the Department of Surgery [C/151] stated: “His right facial region was hit by airbag, and there was no other associated injury, On admission, he was alert with stable haemodynamics, 4 limbs power full, there was erythema over the right face with mild tenderness, abdomen was soft and non tender. XR facial bone showed some fluid level in right maxillary and frontal sinus, no fracture was seen. CT brain showed no intracranial haemorrhage or skull bone fracture. X-ray C spine was unremarkable.” (c) It was recorded that his pain improved with analgesics and he was discharged on 30 June 2014 [F/544]. The discharge note also recorded that he was “able to walk” and “4 limbs power full” [F/543]. However, it was the plaintiff’s evidence that upon discharge, he had to walk on a pair of crutches [B/114/paragraph 8]. (d) The plaintiff was then followed up 3 times at the outpatient department. It was recorded that during these follow-ups that he complained of “subjective blurring of right eye vision and lacrimation” and also “worsening of [right lower limb] numbness” [C/151]. (e) On 12 September 2014, the plaintiff attended the Eye Clinic of Caritas Medical Centre. He complained of right eye grittiness and tearing since the accident. The doctor noted that the condition was unlikely to be related to the accident [C/146]. (f) He started physiotherapy treatment in the Department of Physiotherapy of Kwong Wah Hospital on 20 October 2014. He attended a total number of 31 treatment sessions and reported 40% overall improvement of his right leg pain and numbness. In the final assessment on 25 March 2015, he reported that his right leg numbness was slightly improved during walking. Muscle strength of bilateral lower limbs was about normal. Straight leg raising test of bilateral legs were still limited with pain. He was discharged [C/153-154]. (g) He also started a course of occupational therapy treatments in Kwong Wah Hospital on 4 February 2015 [C/155-156]. He complained of persistent pain over his lumbar spine, bilateral iliac crests and gluteal regions, and numbness over the lateral and dorsal aspects of his right leg. Objective assessment did not reveal significant limitation on active movement of his neck and bilateral arms. He also complained of back pain and demonstrated difficulty in squatting and stooping. He walked with one elbow crutch. He continued with the treatment until the last work assessment done on 27 March 2015 when, based on what he reported, he was assessed to have failed to manage the job demands. (h) On 28 April 2015, the plaintiff claimed that he had a fall at home which resulted in a complete paralysis of his right leg. He said on the night of 28 April 2015, he suddenly felt severe pain on the back of his neck and passed out falling down with his buttock. He was sent by ambulance to the AED of QEH [B/116/paragraph 16]. However, other versions of this event are recorded in various other sources: (1) The ambulance record [F/903] recorded that “patient stated that accidentally slip fell down on floor in toilet about 0135 hrs”. No loss of consciousness was mentioned. (2) The note from the AED of QEH [F/902] and the medical report from the Department of Orthopaedics and Traumatology of QEH [C/163] also recorded that the plaintiff slipped and fell. (3) As recorded in their joint report, he told the neurologists that, when urinating, he could not balance properly and fell. He had no loss of consciousness. [D/242-243] [My emphasis] (i) He was admitted to the Department of Orthopaedics and Traumatology of QEH. Their medical record is telling [C/163]: “During physical examination, there was no movement of the right lower limb when the above-named was instructed to move, while active movement was observed at other moments. There was inconsistent sensory loss. X-rays of lumbosacral spine were unremarkable”. (j) On 5 June 2015, a nerve conduction test was performed and the result was unremarkable. It was recorded that “there was no electrophysiological evidence to account for the right lower limb symptoms” [C/163]. (l) On 9 July 2015, he visited Hong Kong Eye Hospital [C/157]. After physical examination, the doctor noted that his “eyes are essentially normal”. 55. As can be seen from the following table prepared by counsel for the defendants, the plaintiff has made very many inconsistent statements which have been recorded in the various statements, medical records and reports identified in the table which I have amended slightly and which I set out below: 56. Counsel for the defendants submitted that the plaintiff has given numerous and unexplainable inconsistent statements concerning his injuries, condition and disabilities; that his accounts of events were often contradicted by other documents; and that there was no credible evidence that could explain away these inconsistencies. 57. It was further submitted that, contrary to what the plaintiff claimed about his right upper limb disability, the right-handed plaintiff was able to write the 18-page opening statement on 9 June 2020; and the 6-page statement and a declaration on 23 July 2020 (Transcript/Day2/24A-J). This was remarkably different from what he demonstrated in the neurological joint examination where he demonstrated grave difficulty in even writing a signature properly. 58. I disbelieve the plaintiff’s evidence that his right eye was struck by the plastic cover of the airbag and he fell unconscious as a result; and that he sustained multiple serious injuries in his right eye, cervical spine, the back of his head and his lower back. The plaintiff is a blatant and unmitigated liar. He had no qualms asserting the bare faced lie that, although the photograph at the scene of the accident showed him with his eyes open, he was, in fact, unconscious; which lie he sought to support with another preposterous lie that, in hospital, he had seen a lot of unconscious patients with their eyes wide open. I also disbelieve his evidence that, at the time he made his police statement, he was physically very weak, his head was painful and dizzy, his mind was not sober and he had difficulty walking. I find that his statement to the police made on 13 August 2014 in which he stated that he had alighted from his car after the accident and checked and saw that that the front and rear of his car was damaged, was truthful. PC Lee Chi Hang was able to speak to him at the scene of the accident. The plaintiff never lost consciousness at the scene of the accident. 59. Equally unbelievable is his additional evidence that he planned to consult an orthopaedic specialist to amputate his right leg so that a prosthesis could be fitted to him to enable him to stand and walk. I refused his application at the beginning of the trial to conduct a “simple and yet brutal” nerve test by plunging an 8 inch steel needle into his right leg (paragraph 24 of his opening statement). The nerve conduction test performed on 5 June 2015 was unremarkable. These exaggerated theatrics only serve to prove the complete unreliability of his evidence. 60. Another instance of these exaggerated theatrics occurred during the joint neurological examination when he wrote his Chinese name and copied a geometric figure in a feeble and clumsy manner: holding the pencil steadily in the right pincer grasp and moving the piece of paper with his left hand so as to produce the characters and the drawings. Yet he was able to produce, in his own handwriting, an 18-page opening statement, a 6-page statement and a declaration, and a 15-page closing submissions. The hand written characters on these documents were written very clearly. 61. These exaggerated theatrics have continued with his new evidence of receiving a threatening letter to admit that his disability was feigned. It is ludicrous to suggest that the defendants would resort to such criminal behaviour to win this case when they were already armed with a wealth of clear and compelling evidence to prove that the plaintiff is a malingerer. I find this letter, and his evidence about it, to be pure fabrication by the plaintiff. This unsophisticated ruse only serves to confirm my finding that he is a calumnious and mendacious liar. 62. I totally reject the plaintiff’s submission that the defendants used their huge financial means to compile a large amount of false medical expert reports (paragraphs 12, 13 and 15 of plaintiff’s closing submissions and paragraph 3 of his “core” submissions). These experts were appointed by the defendants’ solicitors and by the plaintiff’s solicitors at the time when he was legally aided. These reports were joint reports that were produced by the experts after the plaintiff had been jointly examined by them. 63. I refer to the opinions of the neurologists set out in paragraphs 27 to 33 above. The plaintiff has doubted their expertise and professional ethics and has asserted that there was no basis to support the opinions of these “so-called medical specialists”, who were “colluding” with each other and with Drs Lai and Hung “to maximise their personal gains”. He did not explain the basis of his doubt or why he asserted that there was no basis for their opinions (paragraphs 11, 13, 15, 16 and 20 of plaintiff’s opening statement). I totally reject the plaintiff’s unsubstantiated allegation that these 2 experts were colluding with each other and with Drs Lai and Hung. Dr Choa and Dr Woo have given expert evidence in our courts for very many years. Their testimony has been accepted by the courts on a multitude of occasions. Both experts are well recognised and trusted by the courts. I accept the opinion of Dr Choa, for the reasons that he has given, that the plaintiff’s disability was mostly, if not entirely, feigned; and the opinion of Dr Woo, for the reasons that he has given, that the plaintiff was a malingerer and that all his deficits were feigned. I accept and rely on their expert opinions to support my findings in this case. I set out below my findings on the injuries and deficits he actually suffered as a result of the accident. I also deal below with the slight difference of opinion between these 2 experts on the amount of reasonable sick leave the plaintiff was entitled to. 64. I refer to the opinions of the psychiatrists set out in paragraphs 34 to 35 above. Again, I totally reject the plaintiff’s unsubstantiated allegation that these 2 experts were colluding with each other and with Drs Choa and Woo. Dr Lai and Dr Hung are well recognised psychiatric experts whose expert evidence has been accepted by our courts. I accept the opinion of Dr Hung that if his physical disabilities are feigned, the basis for his psychiatric symptoms also fails. Dr Lai was also of the opinion that he would be suffering from only mild degree of psychiatric symptoms if he had no restriction in his activities of daily living and there was no loss of earning capacity. However, Dr Lai has not explained, why, if the plaintiff has feigned his physical disabilities, he would suffer from even a mild degree of psychiatric symptoms. I prefer the opinion of Dr Hung that, if his physical disabilities were feigned, the basis for his psychiatric symptoms would not exist. I accept and rely on the expert opinion of Dr Hung to support my findings in this case. 65. I refer to the opinions of the ophthalmologists set out in paragraphs 36 to 37 above. I accept the joint opinion of Drs Tsui and Ng that that, despite the plaintiff’s complaint of progressive blurring of vision in his right eye since the accident, repeated ophthalmic examinations and investigations conducted failed to delineate any pathology in his eyes; and that, during the joint examination on 15 December 2017, he was found to be suffering from right eye PED which was newly diagnosed and not related to the accident. I accept and rely on their expert opinions to support my findings in this case. 66. I refer to the opinions of the orthopaedic experts set out in paragraphs 25 to 26 above. I accept the joint opinion of Drs Lam and Cheng that the diagnosis was a contusion of the face and, possibly, a mild neck soft tissue injury, without any complications; that the neck symptom resolved quickly; that the symptoms of right lower limb weakness and paraesthesia that surfaced weeks after the accident was unlikely to be due to any acute involvement with the accident; that they could not relate the subsequent deterioration to any orthopaedic injury; and that they did not know the cause of the subsequent deterioration. I accept and rely on their expert opinions, set out above, to support my findings in this case. 67. The plaintiff praised and relied on the orthopaedic experts as men of integrity and professionalism (paragraph 10 of plaintiff’s opening statement, second paragraph 11 of plaintiff’s closing submissions and paragraph 4 of plaintiff’s “core” submissions). However, I do not rely on their joint opinion that the plaintiff’s right lower limb paralysis and paraparesis of the right upper limb was genuine. Whether or not a plaintiff’s complaints are genuine is a question of fact for the court to determine, with or without the assistance of expert evidence. 68. During their physical examination [D/210-211], the experts found that he could stand up with the brace on but was unable to walk. They performed a straight leg raising test – both legs could be raised to 80 degrees out of a maximum of 90 degrees. The Babinski reflex was normal. There was little appreciable difference of the calf circumference (37.5 cm on the right and 38 cm on the left) and no difference of the arm circumference (31 cm on both sides) and the forearm circumference (27 cm on both sides). There is a typographical error in the report which stated that thigh circumference was 44 cm on the right and 4.75 cm on the left. 69. The reading was probably 47 cm on the left, a difference of 3 cm, which is what Dr Choa found [D/250]. Dr Choa noted that a 3 cm difference was just outside the normal limit of 2.5 cm between sides. He stated that [D/256]: “This could be the residuum of his long standing sciatica or the result of disuse, or a combination of both. This observation is not consistent with significant nerve damage to his femoral or sciatic nerves as his knee and ankle tendon reflexes are normal.” Dr Choa also noted that when not formally tested, the plaintiff was seen to use his right arm to raise himself from his wheelchair by forcefully extending at the right elbow; he also used his right hand in an agile fashion in unlocking the knee-ankle-foot orthosis [D/250]. Dr Choa also noted that when transferring onto the reclining chair, there was a brief interval of at least 1 to 2 seconds when he could actually stand stably with the knee-ankle-foot orthosis detached [D/251]. 70. The orthopaedic experts arrived at their conclusion that the plaintiff’s right lower limb paralysis and paraparesis of the right upper limb was genuine without expressly addressing the possibility that the plaintiff might have been malingering. They expressly stated that there was not a single orthopaedic condition that fitted his condition and that they could not relate his present state to any orthopaedic injury with such a presentation. Yet they did not expressly consider and rule out the possibility that he was malingering. Given the other evidence that has been adduced before me, both factual, which I have found to be truthful, and expert opinion evidence that I have preferred, I do not place any reliance on the orthopaedic experts’ opinion that the plaintiff’s right lower limb paralysis and paraparesis of the right upper limb was genuine. 71. The plaintiff has told me a multitude of lies. He is a consummate, flagrant and egregious liar and I disbelieve his evidence. I find that he is a malingerer. I find that he suffered a contusion of the face and a mild neck soft tissue injury without any complications in the accident on 27 June 2014. I entirely reject his evidence and his claims that he suffered a multitude of disabilities, including his claims that he suffered frequent headaches and dizziness and severe headaches at night; severe pain on the back of his neck and persistent pain on his back; weakness of right hand, numbness and loss of sensation of his thumb and fingers; impaired hearing in his right ear; urinary incontinence; mentally slow in response and reduce concentration; depression, auditory hallucinations and suicidal ideas; being wheelchair-bound because of paralysis and loss of sensation of right leg. 72. I find that sometime after 31 August 2015, as discovered by an examination on 15 December 2017, the plaintiff suffered from a small PED which was not present in the previous examination conducted on or about 31 August 2015. This was a primary retinal disease not caused by ocular trauma. As a result, his right eye vision was blurred with distorted images. The newly diagnosed PED in the right eye and its subsequent visual impairment were not related to the accident on 27 June 2014 and his current ocular complaints of blurred and distorted vision in his right eye are not attributable to the accident. I reject the plaintiff’s evidence that he suffered from visual problems as a result of the accident on 27 June 2014. Assessment of Quantum 73. I have referred to the authorities cited to me on the issue of damages for pain suffering and loss of amenities (“PSLA”): Yau Po Shan v The Express Lift Company Limited & Anor DCPI 2370/2014, 8 November 2019, [2019] HKDC 1495, DDJ S.H. Lee; Wong Ka Lai v Lau Wai Lam & Anor DCPI 1218/2016, 17 April 2019, [2019] HKDC 445, DDJ K.C. Hui; Chung Yin Ting v Chan Miranda (HCPI 1008/2015, 15 February 2019, [2019] HKCFI 270, Master Leong; Wong Eleven v China Way Transportation Ltd & Anor (DCPI 1697/2015, 21 August 2018, [2018] HKDC 1016, DDH Simon Ho; and Lee Chit Ming v Man Siu Hung HCPI 1242/2014, 15 March 2017, Master Chow. I assess damages for PSLA in the sum of $100,000 for the minor injuries, a contusion of the face and a mild neck soft tissue injury without any complications, suffered by the plaintiff in the accident on 27 June 2014. 74. The plaintiff was employed as a chauffeur at the time of the accident by Artwell Tapioca Limited. According to the plaintiff’s employment contract dated 11 March 2014, his monthly salary was $9,000 and he was given an additional $7,000 as parking and meal allowance [E/375-376B]. He was also entitled to double pay at Chinese New Year. I find that his income at the time of the accident was $16,750 ($9000 + $7,000 + ($9,000/12)). Although the defendants were prepared to accept three months as an appropriate sick leave period, I am not prepared to do so. I prefer the opinion of Dr Edmund Woo that it was reasonable for the plaintiff to have one month’s sick leave, which is supported by the fact that the plaintiff resumed work with his employer on 1 August 2014 [E/380]. He resigned on 29 August 2014. According to his letter of resignation, he resigned for personal reasons [E/397–397A]. I would have assessed damages for loss of earnings in the sum of $16,750. However, I note from his bank passbook records that the plaintiff was paid his usual salary of $8,550 a month ($9,000 – $450 being his 5% MPF contribution) for the months of June, July and August 2014 [E/416-417]. I also infer, from the payments of salary by the plaintiff’s employer and the deduction of the plaintiff’s own contribution towards MPF for these months, that the plaintiff’s employer would have paid its contribution towards the plaintiff’s MPF. I find that the plaintiff did not suffer any loss of MPF contributions. However, in the absence of evidence of the payment of his monthly allowance of $7,000 for the month of July 2014, I assess damages for loss of earnings in that amount. 75. I assess special damages in the sum of $5,000 being a sum that the defendants are prepared to agree. 76. I reject all other claims for damages advanced by the plaintiff. 77. The plaintiff received the amount of $652,808.33 as employees’ compensation from Artwell Tapioca Limited on or before mid- March 2019 [E/361-363] which far exceeds my assessment of damages in this case as shown on the following table. Accordingly, I do not award any damages nor any interest on damages to the plaintiff. Dismissal of Action and Costs 78. I dismiss the plaintiff’s action. I make a cost order nisi that the plaintiff pays the costs of the action to the 1st and 2nd defendants on an indemnity basis, to be taxed if not agreed. Such an order will become absolute unless a written application is made within 14 days by any party to vary it. I direct that such written application may be made by letter addressed to my clerk. Such letter should state the variation of my order that is sought and the reasons for seeking such variation. Upon application being made, I will dispose of the same on paper. I may or may not give directions to serve further written submissions before disposing of the same on paper. 79. As the cost order nisi will impact upon the Director of Legal Aid, I grant him leave to apply, by letter, to vary the order within 14 days and/or to apply for wasted costs orders, in respect of which I had made certain observations in paragraph 26 of my recent decision in Lai Sin Yan Elsie v. Tata Communications (Hong Kong) Ltd. HCPI 1092/2015, 14 August 2020. If application is made for wasted costs orders, I will give appropriate directions to enable affected parties to be heard and to deal with the same. 80. I also direct the solicitors for the defendants to send a copy of this judgment to Artwell Tapioca Limited or to their solicitors. The plaintiff appeared in person Mr Simon Wong, instructed by Deacons, for the 1st and 2nd defendants 
Chief Justice Ma: 1. I agree with the judgment of Lord Millett NPJ. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Lord Millett NPJ. Mr Justice Fok PJ: 3. I agree with the judgment of Lord Millett NPJ. Mr Justice Chan NPJ: 4. I agree with the judgment of Lord Millett NPJ. Lord Millett NPJ: 5. This is an appeal by the 1st and 3rd Defendants (“D1” and “D3”) from the unanimous judgment dated 6th November 2015 of the Court of Appeal (Cheung, Kwan and Barma JJA) dismissing their appeal from the judgment of Chow J dated 7th January 2015. The trial judge adjourned the case against the 2nd Defendant who suffers from senile dementia, and she has played no part in the proceedings. 6. D1 and D2 are the administrators of the estate of the late Tang Pui King (“the deceased”) who died intestate on the 4th July 1978. D1 is the son of the deceased, and D2 is his tin fong wife. The next of kin consisted of the five sons of the deceased, one of whom has since died. The Respondent, who was the Plaintiff in the action, is another son. The estate was a large one, and between 1985 and 2012 each of the sons received over HK$86 million by way of cash distribution from the estate. Despite the passage of nearly 40 years, it seems that the estate has not yet been fully administered. The facts 7. The facts, which are not in dispute, are set out at length in the judgments below and it is not necessary to repeat them. They can be shortly stated as follows. 8. In 2003 D1 bought a property in Yuen Long (“the Property”) for $27.3 million. On completion of the purchase he executed a declaration of trust stating that he held the Property in trust for his corporate vehicle D3 and that it had provided the whole of the consideration for the purchase. He later assigned the Property to D3. 9. It is common ground that D1 used $11.48 million of the estate’s money, representing 40.4% of the total purchase price, in payment of the amount due on completion, and that he never informed the Plaintiff that he had done so or sought his consent or that of his surviving brothers, though he says that he informed the executors of his deceased brother that he had borrowed the money from the estate. 10. In October 2003, some seven months after completion, D1 repaid the amount which he had taken from the estate with interest at approximately 3% per annum, a rate which compared favourably with the rates currently obtainable from the banks, though less than D1 would have had to pay to borrow from a bank. The trial judge accepted D1’s evidence that he had always intended to repay the money, which he regarded as a bridging loan, and that he could have raised the money from other sources but had chosen to resort to the estate’s money “as a matter of convenience”. 11. The Plaintiff became aware of D1’s use of the estate’s money in 2005, and commenced the present proceedings in 2009 to recover the profit which he had made by his breach of fiduciary duty. D1 continued to insist that he had discharged his liability to the estate in full by repaying the loan. By the time the case came to trial, however, the value of the Property had increased substantially, and the Plaintiff claimed to be entitled to his share of the increased value of the Property. The judgments below 12. The trial judge held that the transaction should be regarded as a loan and that this precluded the Plaintiff from tracing the money. This, he said, was because tracing requires the claimant to show that he had title to the original asset, whereas in the case of a loan legal title passes from the lender (the estate) to the borrower (D1)[1]. However, he also held that the loan was a misuse of the estate’s funds and that the Plaintiff was entitled to recover his share of the profit made by D1 in breach of his fiduciary duty. He considered that it would be inappropriate to grant a proprietary remedy[2], largely because of the existence of other beneficiaries who were not parties to the action and whose attitude to D1’s use of the estate’s money was not known. He held that D1 was liable to account to the estate for the profits which he derived from the acquisition and holding of the Property, and directed all necessary inquiries to enable the Plaintiff to recover his share of the profits, including the increase in the Property’s value, giving credit for the repayment in 2003 and other payments made by D1 in respect of the Property. 13. The Court of Appeal disagreed with the judge’s conclusion that the transaction was a loan, but confirmed his finding that it was a misapplication of money belonging to the estate. D1 had not appealed against the judge’s refusal to grant proprietary relief, and the Court of Appeal confirmed the accounts and enquiries directed by the judge, ordered D1 to pay the Plaintiff one fifth of the amount found due on taking the account, and dismissed the appeal. The nature of the claim 14. With all due respect to the courts below, they have made heavy weather out of a very simple case. Induced to do so by counsel, they have treated it as a claim to secret profits. This expression is normally used to describe the principal’s claim when a fiduciary, in breach of his fiduciary duty, exploits the fiduciary relationship to divert a business opportunity from his principal or to obtain a benefit for himself from a third party. The paradigm case is that of the agent who receives a bribe from a party with whom he is negotiating on behalf of his principal, giving rise to the receipt of an unauthorised profit and a conflict of interest and duty. It is precisely because the principal had no previous interest in the money or property obtained by the fiduciary in such a case that his right to bring a proprietary claim, though well established by authority, has until recently been questioned. 15. But the present case is concerned with a fiduciary who made a profit by applying his principal’s money for his own benefit. D1 did not take advantage of the fiduciary relationship or put himself in a position where his interest conflicted with his duty. He simply helped himself to money belonging to the estate and applied it for his own benefit. It is a straightforward case where a trustee or person in an analogous position has committed a breach of trust. As will appear, not only is the factual context different, so is the underlying policy which drives equity’s response. 16. The confusion of the two distinct kinds of claim led to a citation of much irrelevant authority[3], and a prolonged argument about causation. True cases of secret profits, particularly those concerned with the diversion of a business opportunity, may involve very difficult questions of causation. But in the present case questions of causation simply do not arise. The right to benefit from an increase in the value of a property (or to suffer from a reduction in its value) is not “caused” by the ownership of the property; it is an incident of ownership. Where property is acquired with the help of a loan, any increase in its value is attributable to the ownership of the property, not to the existence of the loan. D1 contended for the “but for” test of causation; but if that were applicable the relevant question would not be whether the property would have been acquired but for the loan, in many cases a hopelessly speculative enquiry, but whether the purchaser would have profited from the increase in value had he not acquired the property, to which the answer is obviously “no”. 17. Nor is the case concerned with difficult questions of tracing. A purchaser who buys and pays for property does not trace his money into the property. He obtains title by contract or conveyance. In the present case there is a modicum of tracing, for the Plaintiff must show that the Property was acquired at least in part with the estate’s money, and D1 paid the money into his own bank account before withdrawing money from the account to buy it. But that he used the estate’s money to finance part of the cost of the Property is not disputed. Taking the account 18. The facts are not in dispute and there are only two relevant transactions to consider. The dispute concerns their characterisation and financial consequences. These questions are easily solved if seen through the prism of an account and if two basic principles are observed. First, the characterisation of a transaction is a question of law; and secondly, where a trustee or a person in an analogous position has committed a breach of trust by misapplying trust money, the beneficiaries have the right to elect whether to reject or affirm the transaction. The first transaction 19. D1 contends that the first transaction, by which he took money belonging to the estate and applied it towards the purchase of a property for himself, was a (voidable) loan. He relies on Rowley Holmes & Co v Barber[4] to show that a person who has two capacities may make a loan from himself in one capacity to himself in another. But the defect in D1’s use of the money is not want of capacity but want of authority. It was an unauthorised disbursement of money belonging to the estate which D1 applied for his own purposes. He has labelled the transaction a loan; but it is for the court to characterise the transaction, not the parties, still less one of them. The fact that D1 intended to repay the money and did so within a very short time does not overcome his want of authority or convert his misappropriation of the money into a loan. 20. When the Plaintiff received the accounts and discovered the disbursement, he had the right to elect whether to reject or affirm it[5]. Had he elected to reject it, he would have asked for the disbursement to be disallowed and would have disclaimed any interest in the Property by treating it as bought with D1’s own money. This would have produced a deficit in the estate account which D1 would have been obliged to make good had he not already done so. 21. As the Property had substantially increased in value, however, the Plaintiff has naturally elected to affirm the transaction, ie to treat it as an authorised investment of the estate’s money in or towards the purchase of the Property for the benefit of the estate. This has prevented there from being any deficit in the estate account, for the amount of the debit (the disbursement) is matched by the credit (the value of the Property which it was used to acquire). The second transaction 22. Seven months later D1 repaid the estate the amount he had misappropriated with interest. By insisting that this discharged his liability to reimburse the estate for the money he had taken, he is effectively claiming the right to choose whether to affirm the use of the money to purchase the Property or reject it, a right which belongs to the beneficiaries not to the trustee. Given the Plaintiff’s decision to affirm the purchase, there was no deficit in the estate account which D1 was obliged to make good, but the estate had acquired an interest in the Property. D1’s payment must, therefore, be treated, not as a repayment of a loan or money wrongly taken from the estate, but as an attempt to buy out the estate’s interest, an attempt which the Plaintiff has elected to reject. Principle 23. The principle that the beneficiaries can elect to treat property purchased by an unauthorised but profitable application of trust money as part of the trust fund has been established for at least 200 years. In Scott v Scott[6] the High Court of Australia said[7]that there was “of course, abundant authority for the proposition that if trust moneys have been exclusively used in the purchase of property the beneficiary may elect to take the property itself” and cited the statement of Sir John Stuart in Mathias v. Mathias[8] where he said: “Lord Eldon and Lord Redesdale, in the case of Phayre v. Peree[9], in the House of Lords, laid it down as clear law that the trustees can never deal with the trust fund for their own benefit. Lord Redesdale said that the father, who was only tenant for life, could not take the purchase for his own benefit solely, and that his purchase of leasehold property, although unauthorised by the trust, being a beneficial purchase, the benefit must belong to the trust fund”. 24. The facts in Scott v Scott correspond closely to those of the present case. A trustee applied trust money together with his own in the purchase of a property for himself. Shortly before his death he repaid the trust money used by him in the purchase, having previously executed a declaration of trust by which he declared that he held the property in trust for the beneficiaries but limited their interest to the amount of the trust money which he had used to purchase the property, in effect creating an equitable lien for the repayment[10]. He contended that he had discharged his liability to the beneficiaries. By the time of the proceedings, however, the property had substantially increased in value, and the beneficiaries claimed a share in the increase. The issue was thus the same as in the present case. 25. The High Court of Australia[11] upheld the plaintiff’s claim. In a judgment of the court they said[12] “We may, for instance, take the case of a trustee who, in breach of trust, purchases shares for £2,000 by the use of £1,000 of trust moneys together with £1,000 of his own. There is no doubt that the beneficiaries might elect, either, to take one-half of the shares or, alternatively, to claim a lien on the shares for £1,000. But they may not know of the purchase and do neither. Then suppose that prior to any election[13] by the beneficiaries the trustee sells the shares for £3,000 and retains the proceeds in his hands. Is it to be thought that the right of the beneficiaries at this stage will be limited to a claim on those moneys for the specific sum of £1,000?[14] ……. But there can be no doubt that they would be entitled not only to have the sum originally misapplied made good but also to[15] obtain one-half of the resultant profit. We think the same conclusion must inevitably follow even if the property purchased with the mixed fund is property which is not “specifically severable” and that the argument to the contrary must be rejected.” Policy 26. A claim by a principal to recover secret profits made by a fiduciary by exploiting the fiduciary relationship for personal gain and a claim by a beneficiary against a trustee or person in an analogous position for breach of trust give rise to similar remedies, both personal and proprietary, but they raise different factual issues and their underlying policy is different. A claim to secret profits may involve difficult questions of causation as to the extent of the fiduciary relationship or of the business opportunity in question. The policy behind the claim is to enforce the trust which the principal places in the undivided loyalty of his fiduciary by preventing the fiduciary from deriving a personal benefit from the relationship in the absence of his principal’s informed consent. Equity’s response is to require the fiduciary to disgorge the benefit. 27. The policy behind a claim by a beneficiary for a breach of trust of the present kind is to deter the trustee from using the trust fund as his personal bank account, borrowing from it for his own private purposes and merely repaying the amount he has borrowed. Such conduct puts the trust fund at risk without hope of gain. Equity’s response is to insist that any profit is for the beneficiaries and any loss for the trustee. Conclusion 28. In the absence of any argument about the form of the order made below, the appeal should simply be dismissed. As to costs, I would make an order nisi that the appellants pay the costs of the respondent, such costs to be taxed if not agreed. Should any party or parties wish to have a different order as to costs, written submissions should be served on the other party or parties and lodged with the Registrar of the Court within 14 days of the handing down of this judgment, with liberty on the other party or parties to serve and lodge written submissions in reply within 14 days thereafter. In the absence of such written submissions, the order nisi will stand absolute at the expiry of the time limited for such submissions. Chief Justice Ma: 29. For the above reasons, the appeal is dismissed and an order nisi as to costs is made as set out in para 28 above. Mr Robert Ham QC, Mr Denis Chang SC and Ms Candy Chan, instructed by Wong, Hui & Co., for the 1st & 3rd Defendants (Appellants) Mr Brian Green QC and Mr Benjamin Chain, instructed by Pansy Leung Tang & Chua, for the Plaintiff (Respondent) [1] This would preclude following, not tracing, and then only because the borrower would normally be a bona fide purchaser for value without notice. But there is no need to follow the property where the lender and the borrower are the same person, and in any case D1 can hardly claim to be a bona fide purchaser without notice. [2] He refused “to impose a constructive trust”. Strictly speaking, this would not have been necessary. If he had decided to grant proprietary relief, he should simply have declared that an appropriate share of the Property formed part of the estate and was held on the original trusts. [3] Such as Regal (Hastings) Ltd. v Gulliver [1967] 2 AC 134; and Kao Lee & Yip v Koo Hoi Yan [2003] 3 HKLRD 296 [4] [1977] 1 WLR 371. [5] See Libertarian Investments Ltd. v Hall(2013) 16 HKCFAR 681 pp732-3. [6] [1963] 109 CLR 649 [7] At p. 660. [8] (1858) 3 Sm & Giff 552 at p. 563-4 [9] (1815) 3 Dow 116. [10] There was no need to provide for interest, as the trustee was life tenant. [11] McTiernan, Taylor and Owen JJ. [12] At p 662 [13] It had been argued that the trustee had remedied the breach before the beneficiaries made their election, and that this was sufficient to exonerate the trustee. The same argument was raised by D1 in the present case. [14] A rhetorical question which plainly expects the answer “no”. [15] Ie. to (£1,000 + £500) = £1,500, or more simply one half of the value of the property realised by the sale as if the purchase were a proper investment of the trust money. 
Press Summary (English) Press Summary (Chinese) DCCC 980/2015 IN THE DISTRICT COURT OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION CRIMINAL CASE NO. 980 OF 2015 REASONS FOR VERDICT 1. The defendants plead not guilty to one joint charge of causing grievous bodily harm with intent, contrary to section 17(a) of the Offences against the Person Ordinance[1] (charge 1). D5 also pleads not guilty to one charge of common assault, contrary to Common Law and punishable under section 40 of the Offences against the Person Ordinance (charge 2). Introduction 2. On the night of 14/15 October 2014 the police carried out Operation Solarpeak to clear the protestors of the Occupy Central movement. Many police officers both uniform and plainclothes officers were involved in the operation. 3. Crime Group A consisted of Quick Response Teams (“QRT”) and video teams. The defendants were all on duty that night in the area of Lung Wo Road. D1 led QRT A2-2, which included D2, D3, D5, D6 and D7. D4 was initially deployed as a member of a video team but later redeployed by D2. 4. At about 2:45 a.m. on 15 October the clearance operation commenced. The police moved eastbound along Lung Wo Road. After clearing the barricades near to the underpass on Lung Wo Road the police passed through the underpass. On reaching the end of the underpass Tsang Kin Chiu (“Tsang”) was seen on the planter above Lung Wo Road pouring liquid on the police. This was captured by the police video teams. Prosecution case 5. The prosecution case is that Tsang was pulled down from the planter to the pavement and subdued by several uniform police officers. After successfully handcuffing Tsang the uniform police officers handed Tsang over to D1-D6. D1-D6 escorted Tsang away in the direction of Lung Wo Road westbound. On the way Tsang was picked up and carried face down. 6. Protestors were to be taken to the escort coaches and cars on Lung Wo Road for transport to the Central Police Station. Tsang was not carried direct to where the coaches and cars were parked and instead was taken to the north side of the Lung Wui Road Government Building Pump Station East Substation (“the substation”) which was located on Lung Wo Road[2]. 7. On reaching the substation D1-D6 were joined by D7, who helped carry Tsang to the north side of the substation. On reaching the north side of the substation Tsang was dumped on the ground and assaulted by the defendants. 8. Tsang was then frogmarched to Lung Wo Road where he boarded a car. D5 and D6 sat on either side of Tsang and accompanied him to the Central Police Station. At the police station Tsang was taken to room 7 where he stayed until he was escorted by coach to the Police College in Wong Chuk Hang again accompanied by D5 and D6. While in the police station D5, in the presence of D6, slapped Tsang on the face. Evidence 9. Part of what happened that night was captured on video by TVB, Apple Daily, ATV and Now TV and the police video teams. Photographs from Apple Daily and Oriental Daily also showed Tsang being escorted and carried face down. Apart from the police video which shows the clearance operation, including Tsang pouring liquid on the police, the defence objected to the admissibility of the video and photograph evidence. 10. CCTV of the Central Police Station recorded two police officers taking Tsang into and out of the police station. The defence objected to the admissibility of the CCTV recordings. 11. A voir dire was held to determine the admissibility of the video evidence, photographs and CCTV recordings. I ruled all the video evidence, photographs and CCTV recordings admissible in evidence. 12. The prosecution and defence agreed that the evidence given on the voir dire may be taken into consideration on the general issue[3]. Fifteen witnesses were called (six were recalled) and the evidence of a further twenty one witnesses was read pursuant to section 65B of the Criminal Procedure Ordinance[4]. Facts were also admitted pursuant to section 65C of the Criminal Procedure Ordinance. Site visit 13. During the voir dire a joint application was made for the court to go on a site visit to view the various locations referred to in the evidence, including where Tsang was apprehended; the route Tsang was taken and the substation. No agreement could however be reached as to how the site visit was to be conducted. 14. The court therefore suggested for consideration of the parties that a video be taken of the scene showing all the various locations the parties wanted the court to see. After due consideration the parties agreed for a video to be made and withdrew their application for a site visit. 15. SPC 34185 (PW26) made the video, exhibit P20 (c). The witness statement of SPC 34185 was read, exhibit P20 (a). SPC 34185 also gave evidence during which the video was played in court. Screen captures 16. Annexed to my verdict are a series of screen captures taken from the video footage to assist in identifying which part of the video footage to view. Issues 17. The main issues on charge 1 were whether Tsang was the person seen assaulted on the video footage; whether the defendants were the assailants; whether the defendants were part of a joint enterprise and whether Tsang suffered grievous bodily harm. The main issues on charge 2 were whether Tsang was slapped on his face inside the Central Police Station and, if so, whether D5 was the police officer who assaulted him. 18. The defendants elected not to give evidence on the voir dire or the general issue. No witnesses were called on their behalf. No adverse inference is drawn against the defendants for remaining silent. That is their right. This proves nothing one way or the other. This does not establish their guilt. On the other hand, this means there is no evidence from the defence to undermine, contradict or explain the evidence led by the prosecution. 19. In reaching my verdict I remind myself of the burden and standard of proof and that the burden is on the prosecution throughout. The defendants have to prove nothing. I direct myself that I must be sure of the guilt of each defendant before I can convict, the case of each defendant to be considered separately and each charge to be considered separately. 20. I remind myself that when drawing inferences from the evidence the inference must be the only reasonable inference to draw from the proved facts. If from the facts proved there is a reasonable inference to draw against the defendants as well as one in their favour the adverse inference cannot be drawn. 21. Admitted in evidence is that the defendants all have a clear record.[5] I direct myself as to good character in accordance with the decision in HKSAR v Tang Siu Man[6]. 22. I have carefully considered all the evidence and the submissions of all counsel both oral and written. Admissibility of video footage, photographs and CCTV recordings 23. I admitted in evidence the video footage from TVB (exhibit P1); Apple Daily (exhibit P2); ATV (exhibit P3); and Now TV (exhibit P4); photographs from Apple Daily (exhibits P12 and P18 (c)) and Oriental Daily (exhibits P13 and P18 (d)) and the CCTV recordings (exhibit P15). 24. The defence objected to the admissibility of the video footage, photographs and CCTV recordings as particularised in the grounds of objection submitted to court[7]. In summary the grounds of objection were that there was no or no sufficient evidence as to the origin of the videos and photographs; that the videos and photographs had not been tampered with; the provenance and history of the photographs; and the authenticity and accuracy of the CCTV recordings. The defence also invited the court to exercise its discretion to exclude the evidence. 25. Notwithstanding witnesses would need to be recalled all counsel agreed that the court should first determine the admissibility of the video footage, photographs and CCTV recordings by way of voir dire and not adopt the alternative procedure. 26. The prosecution called nineteen of twenty six witnesses from the list of witnesses submitted to court[8]. The evidence of the other witnesses was admitted pursuant to sections 65B and 65C of the Criminal Procedure Ordinance and section 22A of the Evidence Ordinance[9]. Three of these witnesses were tendered for cross-examination. The defence elected to call no evidence. 27. The evidence was heard over ten days. Submissions lasted four days. Having carefully considered all the evidence and the submissions of counsel I found all the video footage, photographs and CCTV recordings relevant and prima facie authentic. 28. I considered each video footage, each photograph and the CCTV recordings separately. I found no grounds to exercise my discretion to exclude from evidence any of the video footage, photographs or CCTV recordings. Authenticity 29. In R v Murphy & Another, the Court of Appeal in Northern Ireland held that: “… in the case of video recordings, the issue for the judge is, is it relevant? If it is, is it prima facie authentic? If it is, then it is admissible and it is left then to the jury …. to decide whether its authenticity is beyond doubt and if its contents prove or add to the proof of guilt beyond reasonable doubt” [10]. 30. In HKSAR v Lee Chi Fai & others the Court of Appeal held that the approach in R v Murphy was wholly correct [11]. Relevance Video footage and photographs 31. The prosecution case is that the video footage, whilst not continuous, shows in sequence (1) Tsang pouring liquid on the police; (2) Tsang being apprehended and subdued; (3) Tsang being escorted by D1-D6; (4) Tsang being carried face down by his arms and legs; (5) Tsang being carried to the substation where D1-D6 are joined by D7; and (6) Tsang being dumped on the ground and assaulted by the defendants; and the photographs show Tsang being escorted and carried face down by his arms and legs. 32. The pouring liquid is shown in the Now TV footage [12] and the TVB footage[13]. This is also shown in the police videos (exhibits P5, 6, 7 & 8)[14], admitted into evidence pursuant to section 65C of the Criminal Procedure Ordinance[15]. The ATV footage shows Tsang being subdued and handcuffed[16]. Tsang being escorted and assaulted is shown in all the video footage whereas only the TVB footage shows Tsang being carried. 33. At the commencement of the voir dire Mr Leung SC submitted by way of aide-memoire an outline of the evidence in relation to the video footage and a table listing the video footage to be played in court[17]. I relied only on the evidence adduced in court and what I observed from viewing the video footage and not the descriptions in the aide-memoires. 34. The defence submitted that the video footage and the photographs were not relevant, for example Mr Lok SC submitted that the video footage showing Tsang spilling liquid, being subdued and escorted had no bearing at all regarding the criminal allegation against the defendants and were irrelevant to the charge[18]. A similar submission was made by Mr Choy[19]. 35. In my view the video footage of Tsang pouring liquid on the police and being apprehended are relevant in seeking to explain why Tsang was handed over and escorted by other police officers. 36. Similarly Mr Lok SC submitted that the photographs showing Tsang being escorted and carried were not relevant to the charge against the defendants[20]. In my view the photographs are relevant in seeking to showwho escorted and carried Tsang to the substation. 37. Mr Cheng SC submitted that on the face of the evidence adduced the court could not conclude that the video footage captured Tsang being assaulted and therefore the prosecution had failed to adduce sufficient and satisfactory evidence to establish the relevance of the video footage and photographs[21]. This submission largely rested on an attack on the credibility of Tsang. I address the credibility of Tsang later in my verdict suffice to say that I reject the submission of Mr Cheng SC. 38. I am satisfied that the video footage and the photographs are relevant to show the events leading up to and including the assault subject of charge 1. The video footage and the photographs are relevant to the issues of whether an offence was committed and who committed it. As was said in R v Murphy the video was clearly relevant to the facts because the prosecution alleged that what it showed pointed to the guilty participation of the appellants[22]. CCTV recordings 39. The prosecution case is the CCTV recordings capture Tsang being taken by D5 and D6 to room 7 of the Central Police Station. Mr Chung and Ms Lam made no submissions on relevance. In his written submission Mr Chung states D5 does not take issue on the relevance of the CCTV recordings[23]. 40. The CCTV recordings, showing Tsang being escorted by two police officers inside the Central Police Station, are clearly relevant to the allegation made by Tsang that inside room 7 of the police station one of the two police officers, in the presence of the other police officer, slapped him. Prima facie authentic 41. The prosecution and defence disagreed as to the test to be applied in determining whether the video footage, photographs and CCTV recordings were prima facie authentic. Mr Leung SC submitted that the evidence of each witness must be taken at its highest in determining whether the video footage, photographs and CCTV recordings were prima facie authentic[24]. The defence on the other hand submitted that the court must be satisfied on a balance of probabilities that the video footage, photographs and CCTV recordings were prima facie authentic[25]. 42. Reliance is placed by the defence on R v Robson & Harris[26] where the trial judge heard objections to the admissibility of tape recordings before the case was opened to the jury. One of the issues before the trial judge was whether the tape recordings were shown to be original. The trial judge said that all he was required to do was satisfy himself that a prima facie case of originality had been made out on the evidence. If the evidence appeared to remain intact after cross-examination it was not incumbent on the judge to hear and weigh other evidence which might controvert the prima facie case as this would trespass on the ultimate function of the jury[27]. 43. The trial judge said that the real gravamen of the objection was an attack on the authenticity of the tape recordings and that he should decide the question of whether the tape recordings were shown to be authentic on the balance of probabilities[28]. As Mr Leung pointed out that was because the trial judge accepted the proposition of counsel on both sides that the standard of proof in this regard was the balance of probabilities[29]. 44. The decision in R v Robson & Harris is cited in both R v Murphy and HKSAR v Lee Chi Fai. In “Social Networking Material as Criminal Evidence” by Micheal O’Floinn & David Ormerod, the authors state there is considerable ambiguity as to what a prima facie admissibility test is and point out no mention is made in R v Murphy of the balance of probabilities[30]. 45. Reference is also made to R v Robson & Harris in Cross & Tapper on Evidence where the authors state[31]: “Where the judge has merely to be satisfied that there is prima facie evidence – for example, that a confession was made, that a previous consistent statement amounted to a complaint, or that a tape recording was the original – he need hear evidence only from the party tendering the confession, previous consistent statement, or tape recording. In R v Robson, the Court of Appeal regarded this as equivalent to a requirement that such evidence need reach only the standard of the balance of probabilities for fear of usurping the function of the jury. It is submitted that the better view is that the requirement is no different from that applying to the satisfaction of an evidential burden, namely whether the evidence, if believed by the jury, would be sufficient to prove the matter asserted to the standard required to satisfy the legal burden, namely beyond reasonable doubt, when borne by the prosecution in a criminal case. This involves no usurpation of the function of the jury since the jury is free not to believe the evidence, and may well not do so after taking into account its contravention by the other side. Such a view is in complete harmony with the ordinary rules on the discharge of an evidential burden, the determination of whether or not there is a case to answer, and the proper distribution of functions between judge and jury”. 46. The authors of “Social Networking Material as Criminal Evidence” agree with Cross and Tapper on Evidence and respectfully submit that the trial judge in R v Robson & Harris was wrong to treat authentication as requiring proof on a balance of probabilities[32]. 47. I am satisfied the test is as articulated in Cross & Tapper on Evidence, namely that the requirement is whether the evidence, if believed by the jury, would be sufficient to prove the matter beyond reasonable doubt[33]. Section 22A of the Evidence Ordinance 48. Before considering proof of authenticity I will first address the submission of Mr Lam, and adopted by other counsel, that the provisions of section 22A of the Evidence Ordinance applied[34]. 49. Mr Lam submitted that the video footage having been inputted into computers by the cameramen; presumably edited and processed by computers for broadcasting and publication; and uploaded to and downloaded from open source internet websites, presumably with the use of computers, the provisions of section 22A applied. The prosecution having failed to satisfy the requirements of section 22A Mr Lam submitted the video footage and photographs should be ruled inadmissible[35]. 50. Mr Leung SC submitted that the provisions of section 22A did not apply the video footage, photographs and CCTV recordings being real evidence and not tendered for any hearsay purpose. 51. Real evidence is defined in Cross & Tapper on Evidenceas an independent species of evidence as their production calls upon the court to reach conclusions on the basis of its own perception, and not on that of a witness[36]. Section 22A provides an exception to the hearsay rule[37]. 52. I agree with Mr Leung that the video footage, photographs and CCTV recordings are real evidence. The representations are made by the video footage, photographs and CCTV recordings and not by a witness. The video footage, photographs and CCTV recordings are not therefore tendered for any hearsay purpose. I am satisfied section 22A is not engaged in this case. Further, I am of the view section 22A does not apply where direct oral evidence of the fact is called, as in this case[38]. Proof of authenticity 53. All the video footage and photographs being copies downloaded from the internet the defence submitted that full particulars of the history and provenance of the copies must be shown[39]. Reliance in this regard was placed on the passage in R v Murphy where the court said that[40]: “If the original tape is not available, then the 'provenance and history' of the copy will be a necessary requirement to prove authenticity”. 54. Earlier in the same passage the court said that[41]: “Authenticity, in our view, like most facts may be proved circumstantially. In the case of a video film, the direct way is to call the cameraman who took it and the court will normally expect him to be called. But if he is not available, he need not be called; other evidence will suffice if it is logically probative that the video was authentic. That evidence may be adduced in other ways and from other sources”. 55. What is required is evidence that the video footage and photographs accurately record the events of that night. Circumstantial evidence 56. Circumstantial evidence may include one or more witnesses confirming the accuracy of the events shown in the video footage, photographs and CCTV recordings[42]. In addition to witnesses from TVB and ATV the prosecution relied on the evidence of the police officers who apprehended Tsang[43] and the evidence of Tsang to say the video footage and photographs were consistent with their recollection of the events. 57. The Northern Ireland case of R v Quinn[44], which concerned video evidence downloaded from YouTube, is an illustration of how authenticity may be proved circumstantially. The trial judge held that the provenance of the film had to be established by the prosecution by evidence in a way that permitted the defence to investigate and examine the provenance and reliability of the material. 58. The Court of Appeal held that in ruling the video evidence inadmissible the trial judge appeared to have given no consideration to the inferences which properly could be drawn from statements made in the course of the police interviewing the defendant. During the interviews the solicitor for the defendant intervened indicating his client was present and that identity was not in issue. The Court of Appeal said that it would be for the trial judge to determine whether the defendant, by remaining silent in the face of the comments made by his solicitor, was admitting he was present and that the video clips demonstrated his activities. 59. In allowing the appeal the Court of Appeal was satisfied that the prosecution had established a prima facie case for the admission of the video clips. Having noted that said in R v Murphy that authenticity like most facts may be proved circumstantially the court said that authenticity can also be established by admissions[45]. 60. In my view, whilst the concerns expressed by the trial judge in R v Quinn are valid concerns in any case where footage is obtained from YouTube or other open source, I am satisfied, reading the passage as a whole, the court in R v Murphy when saying “the 'provenance and history' of the copy will be a necessary requirement to prove authenticity”, was not excluding a copy being proved circumstantially[46]. Comparisons 61. In R v Murphy the court stated that[47]: “The film may be proved authentic by comparing it with films taken by others of the same event, taken at the same time or even at a different time”. 62. In determining whether the video footage, the photographs and the CCTV recordings were prima facie authentic the prosecution invited the court to make comparisons between the video footage; to compare the photographs with the video footage and to compare the CCTV recordings with the video footage and the photographs[48]. 63. The defence submitted to make such a comparison the authenticity of the footage compared with must be beyond doubt, placing reliance on the next sentence in R v Murphy: “Or, as in the instant case of the heli-tele film, by comparing it with a film of the same events that is authentic beyond doubt”. 64. Mr Cheng SC submitted that unlike in R v Murphy the court has no comparative standard from which to conduct such a comparison[49]. In support Mr Cheng SC cited the decision of the trial judge in R v Quinn[50]that it would be quite inappropriate to take two unauthenticated pieces of material and use either of them to authenticate the other. As noted earlier the prosecution appealed the ruling of the trial judge that the video clips were inadmissible. Mr Cheng SC submitted that in allowing the appeal the reasoning of the trial judge was not criticised[51]. 65. I agree with the submission of Mr Leung SC that the court in R v Murphy, when saying “Or, as in the instant case of the heli-tele film comparing it with a film of the same events that is authentic beyond doubt” (my emphasis), were just giving an example taken from the circumstances of the case before them. The court was not, in my view, saying for every comparison the authenticity of the footage compared with must be beyond doubt. 66. Mr Leung did not however simply ask that I compare unauthenticated footage and use either of them to authenticate the other. Mr Leung submitted that if the court was to find the TVB footage prima facie authentic then that footage could be compared with the other footage and photographs. Evidence 67. Before addressing the individual video footage, photographs and CCTV recordings, I will first summarise the evidence given on the voir dire by the police officers who apprehended and subdued Tsang[52] and the evidence of Tsang[53]. Police evidence 68. At around 3:20 a.m. on 15 October 2014 Sgt 47574 (PW1) saw a man wearing a black T-shirt standing in a flower bed[54] pouring liquid from a bottle down onto Lung Wo Road. Sgt 47574 went up to the male and grabbed hold of him. Sgt 47574 said this can be seen in police video footage (exhibit P8)[55] and the Now TV footage (exhibit P4)[56], which were played to Sgt 47574 in court. 69. Neither Sgt 47574 nor the other police officers were asked to identify this male. Admitted in evidence is that the police video footage shows Tsang Kin Chiu throwing liquid from a plastic bottle over police officers[57]. 70. Tsang struggled at which time Sgt 34200 (PW2) and Senior Inspector Lau (“SIP Lau”) (PW32)[58] assisted Sgt 47574 to subdue Tsang and take him to the pavement. Sgt 34200 warned Tsang not to move or he would use OC foam (pepper spray). Tsang still continued to struggle therefore Sgt 34200 removed Tsang’s goggles and mask and sprayed OC foam onto his face. Tsang struggled even more intensely whereby he was able to separate himself from Sgt 34200. 71. On seeing this SSgt 52820 (PW4) went forward to assist Sgt 34200 by putting his arm around Tsang. Tsang continued to resist by swinging his body from side to side causing both of them to fall on the ground and bang their heads against the wall. Senior Inspector Fu (“SIP Fu”) (PW3) also went up to assist. Tsang continued to struggle vigorously and tried to kick SIP Fu. 72. Senior Inspector Wat (“SIP Wat”) (PW5) on hearing SIP Fu warn Tsang not to kick, turned round and saw SIP Fu and Sgt 34200together with two other officers struggling with Tsang on the ground. SIP Wat went up to assist. At the second attempt Sgt 34200, SSgt 52820, SIP Wat and SIP Fu succeeded in handcuffing Tsang at his back with a pair of plastic handcuffs carried by SIP Wat. The first 5 seconds of the video footage from ATV, (exhibit P3 (2)), played in court to SIP Wat, shows SIP Wat holding the larger pair of handcuffs. 73. SSgt 52820 and SIP Fu helped Tsang stand up. When Tsang said he had been sprayed with pepper spray SIP Fu instructed SSgt 52820 to rinse Tsang’s face with water from his water bottle. This is seen in the first 18 seconds of the video footage from ATV (exhibit P3 (2)), which was played in court to Sgt 34200, SSgt 52820 and SIP Fu. The rinsing of the face is also seen between 01:33-01:41 which was played to SIP Fu and SSgt 52820. 74. Tsang was then handed over to other police officers, described by SSgt 52820 and SIP Fu as crime officers. SIP Fu said Tsang was handed over to three or four crime officers[59]. 75. Sgt 34200; SIP Fu; SSgt 52820 and SIP Wat all confirmed that the video footage played to them in court was consistent with their recollection of the events of that night. 76. No video footage was played to SIP Lau, who was only called to give evidence on the general issue. SIP Lau said on 29 September 2015 he had viewed footage of part of the incident. The court was not however told what footage SIP Lau had viewed. In cross-examination SIP Lau agreed that the majority of what happened was not captured on the video footage he had viewed and that he did not appear on the footage[60]. Tsang Kin Chiu 77. The following is a summary of the evidence given by Tsang on the voir dire relating to the admissibility of the video footage and the photographs. Tsang’s evidence relating to the admissibility of the CCTV recordings is summarised later when considering the issue of admissibility of the CCTV recordings. Subdued 78. Tsang testified that at around 3:20 a.m. on 15 October 2014 he was demonstrating in the flower bed on the side of Lung Wo Road near to the LegCo Building. Tsang was wearing a black T-shirt with words on the front and stars on the back[61]. 79. While demonstrating several men pushed Tsang to the ground, assaulted him and handcuffed him with zip wire. Tsang did not know who did this but guessed they were police because only police use force in demonstrations[62]. At the same time Tsang was sprayed with pepper spray. Tsang was then dragged up and after a couple of steps someone washed his face with water. Escorted 80. Tsang was then handed over to several other men. One held his left arm and one his right arm while dragging him along for about 15-20 meters in the direction of Central. Another one twisted his left thumb and wrist. Two or three more men were standing next to them, one responsible for driving away the reporters and one for leading the way. 81. The first 17 secondsof theATV footage (exhibit P3 (2)) were played to Tsang. Tsang identified himself in the footage as the person in the black T-shirt with stars on the back. Tsang said thefootageshowed himbeing subdued; handcuffed with zip wire; being taken up and his face washed. The footage from 1:33-1:54 was also played. Tsang said the footage from 1:33-1:40 showed the second group and the reporters being driven away; and the footage from 1:40-1:54 showed the second group pushing and dragging him. Tsang said the ATV footage was consistent with his recollection of the events that night and that the events shown were in sequence. 82. The TVB footage (exhibit P1 (b)) from 5-13 seconds was also played to Tsang. Tsang identified himself in the footage as the person in the black T-shirt being escorted by the second group. Tsang said the TVB footage was consistent with his recollection of the events that night. 83. The Now TV footage (exhibit P4) from 40-44 seconds was also played to Tsang. Tsang said this also showed him being escorted by the second group and was consistent with his recollection of the events that night. Tsang identified himself in the photograph (exhibit P12) as the person in the middle in the black T-shirt. Tsang said this was after he was handed over to the second group. Carried 84. Tsang heard the group discussing where to take him. When asked to walk faster Tsang did not comply. After a short distance Tsang was lifted up by his arms and legs and carried face down to a place next to a substation. Tsang said this was the dark corner referred to by everyone and the media. Tsang did not know if the people who lifted his legs were from the same group but believed they were. 85. The first 9 seconds of the TVB footage (exhibit P1 (c)) was played to Tsang. Tsang said the footage showed him being carried to the dark corner and was consistent with his recollection of the events that night. Tsang identified himself in the photograph (exhibit P13) as the person in the black T-shirt, with stars on the back, being carried face down to the dark corner. Assault at the substation 86. After being carried a distance of about 500 meters Tsang was dumped on the ground face down. Tsang estimated this was approximately 1-2 minutes after being lifted up. Tsang was kicked and punched by more than one person; sensed he was also beaten with a hard object; that someone stepped on his face and someone hit his face. Tsang curled up his body and faced the wall. During the assault Tsang also heard someone say “demonstration”. 87. Tsang did not at the time know how long the assault lasted. Later, having viewed TV news footage, Tsang confirmed that the assault lasted about four minutes, which was consistent with his memory. 88. Tsang said that from the time his face was washed to the time he arrived at the substation no one left the group. Tsang did not know if during this time anyone else joined the group but later saw from the news footage that someone else did join. 89. The TVB footage (exhibit P1 (c)) from 10 seconds to the end and (exhibit P1 (h)) from 00:14 – 1:07 were played to Tsang. Tsang said the footage showed him being carried to the substation; thrown to the ground and assaulted. Tsang said the TVB footage was consistent with his recollection of the events that night. The Apple Daily footage (exhibit P2 (d)) and the Now TV footage from 18-29 seconds were played to Tsang. Tsang said the footage also showed the assault and was consistent with his recollection of the events that night. Video footage TVB (exhibit P1) 90. The TVB footage is contained on one disc (exhibit P1) and consists of eight videos (a)-(h). The footage was downloaded by the police between October 2014 and September 2015 from YouTube and news.tvb.com. On 17 December 2015 Sgt 46437 duplicated the footage which was then marked WSL1 (a)-(h)[63]. 91. WSL1 (a)-(h) was the exhibit number given to the footage in the affirmation of WSI Wong Sau Ling in High Court proceedings for a production order against TVB[64]. On 23 May 2016 Sgt 58223 (PW19) burnt a disc containing the footage WSL1 (a)-(h). The disc was produced to court and marked exhibit P1 with the eight videos referred to as (a)-(h). 92. In cross-examination Sgt 54850 (PW13), who downloaded some of the footage from YouTube said he was unable to trace who uploaded the footage to YouTube and had no knowledge what was done to the footage before being uploaded[65]. 93. Three witnesses from TVB gave evidence: Sum Ka Hung (PW9), a senior technician in the news ingestion department; Lam Ka Yu (PW10), a senior librarian and David Wong (PW8), the news production manager. 94. In summary Mr Sum explained that the news footage filmed by the cameramen would be transmitted real time by fibre optical or TVU and automatically recorded (ingested) onto the server exclusively used by the news department[66]. During the process of automatic recording Mr Sum, with the aid of a signal meter, would ensure the signal was normal and not interfered with. Mr Sum would find out the location of the transmission and then type in a file name, called a Slug. 95. Mr Sum was on duty between 22:30 on 14 October 2014 and 08:30 the next morning, checking and monitoring the signals transmitted from the cameramen. Between 03:20 and 03:43 images from two cameras at Tamar Park (camera A & camera B) were transmitted real time by TVU and automatically ingested and saved in the server. In cross-examination Mr Sum said the times were roughly when the incident happened[67]. David Wong also confirmed live-footage was automatically ingested onto the server real time[68]. 96. Mr Sum found the signals from both cameras to be satisfactory.Mr Sum assigned the following Slug names to the transmissions saved on the server fromcamera A and camera B: lungwo.live (camera A) and lungwo.topshot.live (camera B). These were the only images received from camera A and camera B between 03:20 and 03:43. The whole system including the server was operating properly during the time Mr Sum was on duty. 97. Mr Sum said that only authorised persons can access the server, which was password protected. In cross-examination Mr Sum said he did not know how many persons had access to the server or whether there was any other measures taken to prevent access to the server[69]. Mr Sum agreed he told the police apart from password protection there was no other measure to prevent unauthorised interference with the server. David Wong said once the footage was ingested and stored onto the server all colleagues from the news department could access the footage with their own log-in and password. 98. Mr Sum and his colleagues were unable to make any alteration to the file stored in the server. David Wong also understood it was not possible to alter, amend or edit news ingested into the server. 99. After transmission was automatically ingested and saved on the server, staff from the library would burn the footage to blu-ray discs. This was the responsibility of Mr Lam. When transferring the footage to blu-ray discs the re-writeable function of the disc would be disabled and Mr Lam would not alter or edit any of the footage. 100. In the afternoon of 15 October 2014 Mr Lam reported for duty and retrieved the footage: lungwo.live and lungwo.topshot.live, which he burnt onto blu-ray discs labelled Local VC 13451 (camera A) & Local VC 13452 (camera B) respectively. I will refer to the blu-ray discs as A & B. In cross-examination Mr Lam said he had no impression what time the raw footage was loaded onto the server because the footage was already there when he went to work[70]. 101. When transferring the footage to the blu-ray discs Mr Lam did not alter or edit the footage. After burning the discs Mr Lam compared the content with the footage saved on the server and found the content to be identical. 102. In cross-examination[71] Mr Lam said that he believed the content on the server was not altered by anyone because he did not have the right to alter those things on the server. As far as Mr Lam knew a password was required when alteration was to be made to the footage on the server. Mr Lam did not have this password and explained that the password used to access the server by members of the news department was “not enough” to alter anything. Mr Lam did not know what was required to alter anything. 103. The blu-ray discs were then stored in the library. Colleagues from the news department had access to the library. On 20 January 2016 David Wong requested the blu-ray discs to compare with the footage WSL1 (a)-(h)[72]. Mr Lam retrieved the discs from the library and handed them over to David Wong. In cross-examination Mr Lam confirmed David Wong returned the discs, which are now kept in the TVB library. 104. P1 (a) was played in court. David Wong identified the TVB News App logo, which is automatically added when the footage was uploaded to the TVB News App. Having compared the footage with blu-ray disc B, David Wong concluded that apart from certain blocked shots which were missing, the video footage on (a) had not been edited. David Wong also compared the sound recording heard on the footage and found the sound had also not been edited. 105. David Wong explained that exhibit WSL1 (a) was split into first shot and second shot. The first shot comprised of the first 10 seconds, which showed a man being carried by a group. In cross-examination[73]David Wong said he did not speak to the cameraman. When asked if he could say where the man was first picked up David Wong explained that prior to the footage there was another 4 seconds on the blu-ray disc. This four second shot however was very shaky and therefore David Wong could not see clearly what was depicted or where the man was first picked up. 106. The second shot is the remainder of the footage showing the assault at the substation. 107. When David Wong compared the blu-ray discs with the footage WSL1 (a) he saw that the first shot and second shot came from the same camera, camera B. After the first shot David Wong saw that the group turned into an area where they were blocked by a van[74] (“the blocked shot”). The second shot is when the group appeared again on camera. David Wong said the first shot, the blocked shot and the second shot could all be seen on blu-ray disc B as one continuous shot without editing. 108. P1 (c), (e) and (f) were all played in court. David Wong did not compare these with the blu-ray discs because from what he saw they were all exactly the same as (a). David Wong was unable to tell when (a), (c), (e) and (f) was made available to the public. Based on what he was told by a colleague David Wong guessed the time was around 6 a.m. via the TVB News App. 109. The first 1 minute and 8 seconds of P1 (b) was played in court. David Wong referred to the footage from 5-13 seconds as the “relevant shot”, showing the male being escorted. This was filmed by camera A. David Wong compared the footage with blu-ray disc A and concluded this was a continuous single shot without editing. In cross-examination David Wong said that he was able to say this was a single shot without editing because he had viewed the original blu-ray disc[75]. 110. David Wong also compared the sound recording heard on P1 (b) with blu-ray disc A and found they were the same. 111. The footage P1 (b) between 13 and 21 seconds was filmed by camera B, the same camera as (a), (c), (e) and (f) and shows part of the second shot. At the beginning of this part there is an extra four seconds which are not seen on (a), (c), (e) and (f)[76]. David Wong confirmed these four seconds were contained on blu-ray disc B. The extra four seconds can also be seen on P1 (d), (g) & (h). 112. The news footage seen in P1 (b) was broadcast by the TVB interactive news channel between 06:04 and 06:05 on 15 October[77]. The time and date as well as the TVB logo can be seen on the footage. The time and date was real time, automatically input when the footage was broadcast. The voice over was added by the news reporter after filming. 113. In cross-examination David Wong said apart from the images shot by camera A and camera B WSL1 (b) contained many other images from other cameras, which were not burnt onto blu-ray discs A & B[78]. David Wong explained there were a lot of cameras at the scene but the “relevant shot”, first shot and second shot came from camera A and camera B. David Wong mainly checked the “relevant shot”, first shot and second shot and did not make a full comparison of all the images. 114. P1 (d) was played in court[79]. David Wong compared (d) with (a) and (b) and concluded the main shots captured were from cameras A & B and were the same as (a) and (b). P1 (d) shows the relevant shot, the first shot and part of the second shot[80]. 115. P1 (g) was played in court[81]. David Wong compared (g) with (d) and concluded the beginning of the footage was the same except that the part showing a police spokesman was cut out from (g). A careful comparison also shows that, after the assault at the substation and just prior to the police spokesman, other parts of the footage are also not included in (g). The broadcaster seen at the beginning and the news reporter seen at the end of the footage are however the same. 116. David Wong said that (d) should have been broadcast at 07:02 on Good Morning Hong Kong and on the interactive I-news channel[82] but did not know when (d) was made available to the public on the internet. The time of the broadcast is not however seen on the footage. The news footage seen in P1 (g) was broadcast between 09:41 and 09:45 on 15 October[83], the time and date as well as the TVB logo being shown on the footage. 117. P1 (h) was played in court[84]. David Wong compared (h) with (b) and concluded they were the same except that the brightness of (h) was enhanced. The time of the broadcast between 06:04 and 06:05 can also be seen on the footage. 118. I was satisfied, if believed, that the evidence of Mr Sum; Mr Lam and David Wong was sufficient to prove the authenticity of the TVB footage beyond reasonable doubt. I therefore ruled the TVB footage prima facie authentic and admissible in evidence. 119. In addition, as summarised earlier the TVB footage P1(b) and P1(c), showing the escort and the assault at the substation was played to Tsang, who confirmed the footage was consistent with his recollection of the events that night. Tsang also identified himself as the person being carried in the TVB footage P1 (h) [85]. I was satisfied the evidence of Tsang, if believed, was also sufficient to prove the authenticity of the TVB footage beyond reasonable doubt. Discussion 120. The main criticism levelled at all the video evidence, and not just the TVB footage, is that there was no continuous recording, the footage having clearly been edited[86]. The defence submitted the court cannot be satisfied the first shot, the blocked shot and the second shot is one continuous shot, for example Mr Lok SC, submitted that by reason of there being no continuous recording and the difference in quality between the first shot and the second shot, the second shot could have depicted the activities of a different group[87]. 121. A similar submission was made by Mr Cheng SC[88] and Mr Lam[89]. Mr Lo submitted that it is unknown whether the first shot and the second shot are in sequence or on different occasions and that in view of the length of the blocked shot the footage cannot be relied upon as an accurate depiction of the events shown therein[90]. Mr Choy submitted that given the abundant uncertainties arising from the blocked shot minimal weight should be attached to the footage[91]. “The blocked shot” 122. David Wong’s evidence that after the first shot he saw the group turned into an area where they were blocked by a van (“the blocked shot”) has been the subject of much criticism. 123. When asked in cross-examination to estimate the time between the first shot and the second shot Tsang said, “I think more than 5 seconds, but it’s within 10 odd to 20 seconds.”[92] 124. David Wong also gave evidence about the time between the first shot and the second shot. In examination-in-chief David Wong said the group were blocked by a vehicle for about 20-30 seconds. In cross-examination[93] David Wong was referred to his affirmation[94] filed in the High Court proceedings in which the time was given and confirmed the gap between the two shots was in fact 49 seconds; he did not know what happened during those 49 seconds and he did not speak to the cameraman. 125. Mr Cheng SC submitted that David Wong’s evidence as to what was captured in the blocked shot was inadmissible hearsay and that it was not possible for anyone to determine what was in fact captured during the blocked shot[95]. I have no hesitation in rejecting this submission. The evidence of the blocked shot was adduced to explain why the footage was not continuous. 126. At no time in evidence did Tsang say he was taken behind a vehicle. Mr Lok SC submitted that no mention having been made by Tsang that he had been carried anywhere near any vehicle raises the question whether he was carried behind a van or this was an incident not involving Tsang[96]. Mr Lo submitted that Tsang’s evidence being inconsistent with David Wong’s evidence that the group turned behind a vehicle, it would be unsafe to rely on David Wong’s evidence[97]. 127. In cross-examination on the voir dire Tsang agreed he told the police he was not sure about the route he was carried. When put by Mr Cheng SC he had no personal memory of the route Tsang disagreed and said “It’s just a straight line”. In cross-examination on the general issue Tsang disagreed with Mr Cheng SC that the route involved a sharp right turn before entering the substation saying that the route was slightly sloping upwards, but was more or less a straight line. 128. Mr Cheng SC submitted that Tsang’s evidence the route was a straight line was irreconcilable with the video evidence which showed Tsang was not carried in a straight line. This Mr Cheng SC submitted cast serious doubt on the overall reliability and credibility of Tsang’s evidence[98]. Mr Lam in making a similar submission pointed out that the route taken by SPC 34185 (PW26), when making the video of the scene (exhibit P20), was different to that of the group seen walking alongside the substation[99]. Mr Lam submitted that the route taken by SPC 34185 was “the straight line route”. 129. I have no hesitation in rejecting these submissions. In my view it is not at all surprising that Tsang did not mention he was taken behind a vehicle. Tsang was carried face down and would not therefore have been able to see clearly the route he was taken or what he was taken past. I do not agree the route taken by SPC 34185 was a “straight line route”. The route SPC 34185 took was by turning right across the lawn as opposed to continuing on the path and turning right at the substation. 130. The fact Tsang did not say he was taken behind a vehicle and described the route he was carried as a straight line does not cause me to doubt the evidence of David Wong that after the first shot he saw the group turned into an area where they were blocked by a van. In addition, in cross-examination Mr Lok SC played to Tsang the first eight seconds of the TVB footage exhibit P1(c), which shows Tsang being carried (the first shot). After asking Tsang whether he was facing up or down Mr Lok SC asked, “And then it seemed you and the team had turned into a corner.” Tsang agreed. Impossibility for a van to enter the path 131. Mr Lok SC submitted that it was impossible for a van to enter the path where David Wong said the group were blocked by a vehicle[100]. Mr Lok SC, relying on the video of the location (exhibit P20), submitted there was a concrete barrier together with fixed and continuous railing thereby preventing vehicles entering the path from Lung Wo Road[101]. 132. I have no hesitation in rejecting this submission. Police video, exhibit P8 (file 00000), clearly shows that the railings are not continuous. Railings run parallel with the substation. On approaching the underpass and before reaching the railings there is a recess or lay-by where cars can pull in and park or drive alongside the substation. At the end of the railings there is clearly sufficient space for a vehicle to access the path[102]. 133. Similar submissions were made by Ms Lam and Mr Lo[103]. In support Ms Lam played the police video exhibit P5 (file 00000) to show the road and path were blocked by barricades. Screen shots for 11:39; 12:20 and 12:45 were submitted by Ms Lam[104]. As pointed out during the submission of Ms Lam this was before Tsang was seen pouring liquid down onto Lung Wo Road. That is shown in exhibit P5 (file 00001) at 13:48, just over 20 minutes after the screen captures submitted by Ms Lam. 134. The police video, exhibit P5, (file 00000) at 12:55 shows that the mills barriers were removed. After about five minutes the police begin to clear the path by moving the demonstrators back up the path (see exhibit P5, file 00000 from 18:00). The videos do not show the plastic barricades being removed. One orange plastic barricade is seen with one or more white barricades in police video, exhibit P7 (file M2U00038) at 04:45, which is about two minutes after the police started moving the demonstrators as seen in exhibit P5. 135. About ten minutes later the path is cleared and the police walk down the path to get ready to go through the underpass (see exhibit P5, file 00001 between 03:15 – 04:30). This can also be seen on police videos, exhibit P7 (file M2U00038 from 04:40) and exhibit P8 (file 00000 from 11:00). This is just less than ten minutes before Tsang was seen pouring liquid down onto Lung Wo Road, which is also shown in exhibit P7 (file M2U00038 at 23:00) and exhibit P8 (file 00001 at 06:13). 136. The cameras do not however show whether by this time the plastic barricades have been removed. Notwithstanding I am nevertheless satisfied, having viewed the police videos, that by removing the mills barriers there was more than sufficient space for a vehicle to access the path. 137. Ms Lam relying on the evidence of Superintendent Ng (“SP Ng”) (PW27) submitted that the evidence of David Wong should be rejected[105]. I have no hesitation in rejecting this submission. In cross-examination Ms Lam showed SP Ng photograph 11, exhibit P24. After SP Ng agreed the pavement was cleared that night Ms Lam asked SP Ng if he agreed that no car was allowed to drive on the red pavement. SP Ng replied, “All along no vehicles are supposed to be driven up to this pavement.” Ms Lam then showed photograph 12 and asked the same question to which SP Ng replied “Right.” 138. Photograph 12 was a close up of photograph 11. I am satisfied all that SP Ng was saying is no vehicles were supposed to be driven on the red pavement. This evidence does not cause me to doubt the evidence of David Wong that after the first shot he saw the group turned into an area where they were blocked by a van. 139. Mr Lok SC and Ms Lam further submitted that even if the path was cleared, with so many people around, it was unlikely a van could have reached the path David Wong said was the location of the blocked shot. Accepting there were many people around that early morning I am satisfied, having viewed the police videos, there was nothing to prevent a van from driving onto the path. In addition, as seen from the police videos, the police were by this time at the other side of the underpass where no doubt most of the demonstrators will have been. This submission does not cause me to doubt the evidence of David Wong that after the first shot he saw the group turned into an area where they were blocked by a van. Sequence 140. In cross-examination David Wong said he did not ask the cameramen whether what was seen on (a)-(g) was in natural sequence. Asked by Mr Cheng SC whether he could say the events were in natural sequence David Wong replied that for parts of (b), (d) & (g) he could not say as these were news stories with voiceovers but for the relevant shot, first shot and second shot these were in sequence according to the times on the blu-ray disc. 141. Mr Cheng SC submitted that there was insufficient evidence to establish whether the events shown in the relevant shot occurred before or after the events shown in the first shot and the second shot[106]. Relying on the Canadian case of R v Penney[107]Mr Cheng SC submitted that the first and second shot being separated by a lengthy gap cannot be relied upon as an accurate depiction[108] and that in absence of the true sequence of events no reliance can be placed on the video footage[109]. 142. Similarly Ms Lam submitted that all the video footage had been edited whereby the footage was seriously distorted giving a false pictorial presentation of the event[110]. Mr Lam submitted the possibility the TVB footage captured three totally different and unrelated events could not be ruled out[111]and that the editing of the footage may possibly have distorted the nature and sequence of events[112]. 143. I have no hesitation in rejecting these submissions. In Kajala v Noble, the Court of Appeal said where no attack is made on the integrity of the film in the sense that it had, by cutting, distorted the activities of the appellant it was irrelevant that the film might not have shown the events in their true sequence, so long as the film accurately showed the activities of the defendant[113]. 144. I am satisfied on the evidence of Sum Ka Hung, Lam Ka Yu, David Wong and Tsang that the video footage, whilst not continuous, shows the events of that night. I am satisfied the images have not been altered or changed. I am satisfied that the true sequence of events is that the relevant shot is followed by the first shot and then the second shot. The fact that there is a blocked shot in-between the first shot and the second shot does not cause me to doubt this is the true sequence of events. 145. The case of R v Penney is clearly distinguishable. The offence charged was that, “No person shall attempt to kill a marine mammal except in a manner that is designed to kill it quickly.” In the absence of a continuous video of the killing of the seal the court said that it was impossible to determine whether the manner used was designed to kill the seal quickly[114]. On the evidence I am satisfied that the video footage, albeit not continuous, accurately depicts the events of that night and that a continuous video is not necessary for proof of the charge as in R v Penney. Editing 146. In cross-examination David Wong was referred to §§25 & 26 of his affirmation filed in the High Court proceedings which stated there had been some editing to the relevant parts and the second shot with the uneventful parts removed[115]. David Wong explained that the uneventful parts referred to in §25 were the blocked shots. 147. With regard to §26 David Wong explained that due to the duration of news stories certain shots were cut out of P1(d) & P1(g) but having made a comparison of the footage David Wong concluded that the relevant shot in camera A was not edited[116]. For camera B David Wong explained there were several edit points to condense the story. The parts edited out were retained by TVB. Although TVB opposed the production of the blu-ray discs to safeguard journalistic independence and safeguard frontline staff, David Wong said that the parts edited out of (d) & (g) were in fact contained in (a), (c), (e) & (f)[117]. 148. I accept the evidence of David Wong. The fact some of the footage had been edited in the way David Wong described does not cause me to doubt the findings I have made that the video footage, whilst not continuous, shows the events of that night; that the images depicted have not been altered or changed; and that the true sequence of events is that the relevant shot is followed by the first shot and then the second shot. Enhancement (exhibit P1 (h)) 149. In cross-examination David Wong said that it appeared exhibit P1 (h) had been edited by adding digital enhancement of the brightness of the image[118]. This was not done by TVB and David Wong did not know how the brightness had been adjusted. Mr Cheng SC submitted exhibit P1 (h) had been tampered with and should therefore be ruled inadmissible[119]. 150. I have no hesitation in rejecting the submission the video footage had been tampered with, in the sense that the events shown have been altered or changed. Although David Wong did not compare exhibit P1 (h) with the blu-ray disc he did make comparison with exhibit P1 (b) and concluded that apart from the brightness they were the same. 151. I have also compared P1 (h) with P1 (b), not to establish the authenticity of P1 (h) but to see whether the events depicted are the same. Like Mr Wong I am satisfied the images depicted are the same and have not been altered or changed. P1 (h) is the same news broadcast as that shown in P1 (b) at 06:04. I am satisfied that the only enhancement is the brightness. The fact the court has not been told who or how the brightness of the image was enhanced does not cause me to doubt the authenticity of P1 (h). Real time 152. Mr Cheng SC submitted that from the time the footage was ingested onto the server to the time the footage was burnt onto blu-ray discs eleven hours passed during which time the footage could have been tampered with. The prosecution having failed to rule out the possibility the footage had been tampered with during this time Mr Cheng SC submitted the blu-ray discs would simply be the “fruit of a poisoned tree” and cannot be relied on as a means to authenticate the footage[120]. 153. In cross-examination when asked if there was any evidence showing the precise time the relevant shot was filmed by camera A, David Wong explained that the images recorded by camera A were broadcast live on the interactive channel[121]. As far as David Wong could remember from viewing the live footage the time should be around 03:30[122]. In re-examination David Wong confirmed the relevant shot was broadcast live. 154. Ms Lam submitted in court that David Wong changed his evidence with regard to when the relevant shot was made available to the public[123]. I have no hesitation in rejecting this submission. When asked by Mr Leung SC what time the footage on P1 (b) between 00:05 and 00:13, showing the relevant shot, was made available to the public David Wong, after clarifying the question was referring to P1 (b), replied 06:04. David Wong was saying the broadcast exhibit P1(b) was at 06:04 and not that the relevant shot was first made available at 06:04. I am satisfied David Wong did not change his evidence. 155. When asked in cross-examination where he obtained the information the footage from camera A was taken between 03:20 and 03:43 David Wong explained that he obtained the information from the blu-ray disc[124]. The time was real time and not added to the blu-ray disc. 156. Later in cross-examination when asked about the blocked shot (taken by camera B) David Wong confirmed that in his affirmation he said the first shot was taken between 03:32:08 and 03:32:19 and the second shot between 03:33:08 and 03:37:48. David Wong said he obtained these times from the blu-ray disc[125]. 157. In re-examination David Wong was asked to explain how the time was recorded in the blu-ray disc. David Wong explained that when the news footage was transmitted live and simultaneously recorded onto the blu-ray disc the real time mode was used whereas when the blu-ray disc was burnt later this was not in real time. 158. David Wong said blu-ray disc B was recorded real time and blu-ray disc A was recorded later and therefore not in real time. Although David Wong was able to give the time the first shot and the second shot were filmed his evidence blu-ray disc B was recorded simultaneously differs from the evidence of Mr Lam that he burnt both discs later that afternoon. Mr Lam being the one who burnt the discs I accept his evidence. I therefore do not rely on the times given by David Wong as being the actual time when the first shot and the second shot were taken. 159. I accept the evidence of Mr Sum that the footage from camera A and camera B was automatically ingested onto the server roughly between 03:20 and 03:43 and the evidence of David Wong that the relevant shot was broadcast live. 160. The relevant shot was filmed by camera A and is shown in P1 (b), (d), (g) & (h). The first shot and second shot were filmed by camera B. The first shot is shown in P1 (a), (c), (d), (e), (f) & (g). The second shot is shown in all the TVB footage[126]. 161. David Wong said P1 (b) was broadcast at 06:04 and P1 (d) at 07:02. The footage P1 (g) shows the broadcast was at 09:41. P1 (h) is the same as P1 (b) and shows the time of broadcast at 06:04. David Wong could only guess the time P1 (a), (c), (e) & (f) were made available to the public[127]. 162. I have no hesitation in rejecting the submission the footage was tampered with. Camera A was broadcast live and within a relatively short time, of less than two and a half hours, the footage from camera B had been broadcast. Although members of the news department had access to the server I have no hesitation in rejecting the suggestion that in the intervening period from filming to broadcast and from broadcast to the burning of the blu-ray discs someone had accessed the server and tampered with the footage whereby the images and appearance of the people depicted had been altered or changed. 163. In addition, I accept the evidence of Mr Sum that he and his colleagues were unable to make any alterations to the files stored in the server; the evidence of David Wong that it was not possible to alter, amend or edit the news ingested into the server; and the evidence of Mr Lam that he did not alter or edit the footage and after burning the discs compared the content with the footage saved on the server and found the content to be identical. Conclusion 164. I accept the evidence of Mr Sum; Mr Lam; and David Wong. Having carefully considered the evidence and viewed the footage I am satisfied there are no signs of tampering or discontinuity which cause me to doubt the authenticity, accuracy and integrity of the TVB footage. I find the prosecution have proved the authenticity of the TVB video footage beyond reasonable doubt. 165. In addition, I accept the evidence of Tsang that the TVB footage played to him in court was consistent with his recollection of the events that night. On this evidence I find the prosecution have proved the authenticity of the TVB footage beyond reasonable doubt. I discuss later the credibility of Tsang. Discretion to exclude 166. In ruling the TVB footage was prima facie authentic and finding the footage authentic I carefully considered the submissions that I should exclude the evidence, including that the defence did not have access to the blu-ray discs referred to by David Wong and from which David Wong made his comparisons; the absence of the cameramen; and the intelligibility and/or quality of the footage[128]. 167. I found no grounds on which to exclude any of the TVB footage on the grounds of unfairness. None of these matters cause me to doubt the findings I have made. Apple Daily (exhibit P2) 168. The Apple Daily footage is contained on one disc (exhibit P2) and consists of five videos (a)-(e). The footage was downloaded by the police from hk.apple.nextmedia.com; hk.dv.nextmedia.com and YouTube. On 17 December 2015 Sgt 46437 duplicated the footage which was then marked WSL2 (a)-(e) [129]. On 23 May 2016 Sgt 58223 burnt a disc containing the footage WSL2 (a)-(e). The disc was admitted in evidence and marked exhibit P2 with the five videos referred to as (a)-(e)[130]. 169. All the five videos show part of the assault at the substation. None however capture the beginning of the assault. The longest is (d), which shows 3 minutes 59 seconds of continuous footage. The other four videos show various segments of the assault seen in (d). 170. Videos (a) & (e) are also continuous footage however the footage consists of four separate segments of the assault. The four segments are the same in both videos. Video (b) shows three short segments of the assault and video (c) shows the same segment twice, the second time being a little longer. 171. Nobody from Apple Daily was called to give evidence. As summarised earlier the Apple Daily footage (d) was played to Tsang, who confirmed the footage was consistent with his recollection of the events that night. 172. I was satisfied, if believed, that the evidence of Tsang was sufficient to prove the authenticity of the Apple Daily footage beyond reasonable doubt. I therefore ruled the Apple Daily footage of the assault prima facie authentic and admissible in evidence. 173. In addition Mr Leung SC submitted that the content of the TVB footage P1 (a) from 42 seconds was identical with the Apple Daily footage P2 (d), as particularised in the table annexed to the written submission of the prosecution[131]. In court Mr Leung SC submitted that video (d) was filmed by a different camera and was not simply a copy of the TVB footage P1 (a). Mr Leung SC played to the court part of P1 (a) at 1 minute and 22 seconds, which Mr Leung SC submitted showed a slight tilting of the camera, which is not seen on the Apple Daily footage. 174. I was satisfied, if believed, that the Apple Daily footage of the assault at the substation was identical to the TVB footage and was filmed by a different camera, this was also sufficient to prove the authenticity of the Apple Daily footage beyond reasonable doubt. 175. The Apple Daily footage P2(c) between 21 and 27 seconds shows what I will refer to as a moving photograph. In oral submissions Mr Leung SC submitted the clothing and appearance of the persons shownare the same as seenin the TVB footage. 176. I was satisfied, if believed, that the clothing and appearance of the persons were the same as in the TVB footage or the still photographs, this was sufficient evidence to prove the authenticity of the moving photograph beyond reasonable doubt. I therefore ruled the moving photograph prima facie authentic and admissible in evidence. Discussion 177. I accept the evidence of Tsang that the Apple Daily footage played to him in court was consistent with his recollection of the events that night. I discuss later the credibility of Tsang. 178. Having carefully considered the evidence and viewed the footage I am satisfied there are no signs of tampering or discontinuity which cause me to doubt the authenticity, accuracy and integrity of the Apple Daily footage. I find the prosecution have proved the authenticity of the Apple Daily footage of the assault beyond reasonable doubt. 179. In addition, having carefully viewed the TVB footage and the Apple Daily footage I agree with the submission of Mr Leung SC that the content of the TVB footage P1 (a) from 42 seconds is identical to the Apple Daily footage P2 (d). Although the Apple Daily footage appears to be a wider angle than the TVB footage, I am not satisfied the “tilting” referred to by Mr Leung necessarily proves that the Apple Daily footage was filmed by a different camera. 180. I am nevertheless satisfied that the TVB footage and the Apple Daily footage were filmed by two different cameras. The beginning of the Apple Daily footage P2 (a) shows the camera panning across Lung Wo Road to film what was happening at the substation[132]. It is at this point the footage in P2 (d) begins. This explains why the Apple Daily footage, unlike the TVB footage, does not capture the group arriving at the substation and the beginning of the assault. 181. On this evidence I find the prosecution have proved the authenticity of the Apple Daily footage beyond reasonable doubt. 182. I am satisfied in determining whether the moving photograph in P2 (c) is authentic I can compare this with the TVB footage and the still photographs to see if the images are consistent. Notwithstanding Tsang is not seen in the moving photograph, I am satisfied by looking at the clothing and appearance of the persons, the moving photograph shows the time when Tsang was being carried face down. 183. This is the same time as in the photograph exhibit P13 and shortly after the police officers escorted Tsang as seen in the TVB footage P1 (b), (d) & (h) between 5-15 seconds and (g) between 02:47-02:55 and the photograph exhibit P12. Having carefully considered the evidence and viewed the footage and photographs, I find the prosecution have proved the authenticity of the moving photograph beyond reasonable doubt. Discretion to exclude 184. In ruling the Apple Daily footage was prima facie authentic and finding the footage authentic I carefully considered the defence submissions that I should exclude the evidence, including nobody was called from Apple Daily to confirm when the footage was filmed; the absence of the cameramen; how the footage was edited for broadcasting; and the quality of the footage[133]. 185. I found no grounds on which to exclude any of the Apple Daily footage on the grounds of unfairness. None of these matters cause me to doubt the findings I have made. ATV (exhibit P3) 186. The ATV footage is contained on one DVD, exhibit P3 and consists of a broadcast version, exhibit P3 (1) and a library copy, exhibit P3 (2)[134]. P3 (1) was downloaded by the police from the website of ATV[135] and shows the beginning of a news broadcast at 18:00 hours on 15 October 2014. P3 (2) was obtained from ATV[136]. 187. The news broadcast and the library copy show Tsang being subdued, escorted and part of the assault at the substation. Mr Leung SC submitted that the content of the broadcast copy from 41 seconds onwards was identical to the library copy[137]. 188. One witness Chim Yat Kin (PW7), a senior video editor in the video editing department of the news department of ATV, gave evidence. In summary[138] Mr Chim testified that one of his duties was converting video captured by the cameramen onto tape. Mr Chim explained that when the camera was filming, the footage was transmitted real time to the ATV studio by a link called “LiveU” and recorded onto what Mr Chim called the “raw tape”, which would be kept in the library. 189. After the filming was finished the cameraman would return to ATV and remove the storage card (called a P2 card) from the camera. The storage card was required for use every day therefore Mr Chim or one of his colleagues would copy the content of the storage card onto a tape, called the master tape. Depending on the quality and the content, the raw tape may also be used as a master tape. 190. Mr Chim explained that the cameraman could only record and not edit therefore the content of the master tape was the same as the footage filmed by the cameraman. The master tape was then placed in the library and the storage card returned to the cameraman for further use. 191. Mr Chim also explained how news was edited for broadcasting. After receiving the narration and the voice over from the reporter Mr Chim would listen to the narration and locate the relevant parts in the master tape. The relevant parts would be copied for broadcasting. The footage would be edited for example by inserting photographs of persons mentioned in the narration or inserting the date. After completing the whole story, the tape would be kept ready for broadcasting. In cross-examination Mr Chim agreed the length of the broadcast copy would be shorter than the master tape[139]. 192. Where photographs edited into the footage were from outside sources such as other news agencies or from members of the public, an acknowledgement, called a “courtesy”, would be given at the time of broadcasting. When the news was being broadcast the director would key in the relevant courtesy which would then be seen by those watching the broadcast. 193. Mr Chim demonstrated two examples from exhibit P3(1) where acknowledgement was given. First at 01:03 acknowledging footage from Apple Daily showing the assault at the substation and second at 01:20 acknowledging a photograph of Tsang from the Occupy Central movement. In cross-examination Mr Chim said that the footage from Apple Daily was normally obtained direct from Apple Daily and not from Apple Daily’s website[140]. Mr Chim did not however know how the photograph from the Occupy Central movement was obtained. 194. After the news was broadcast colleagues from the library would take the tape together with the other news stories broadcast and copy all the stories onto one disc called the “clean version”. As the courtesy was broadcast real time, these would not be seen on the clean version. The clean version was kept in the library for future use. 195. Only colleagues from the news department could have access to the library copies. When a copy was required a request would be made to the library and the librarian would locate the copy. In March 2015 Mr Chim was asked by his boss to assist the police, who requested copies of footage from various dates. On 31 March Mr Chim requested the copies from the library which he converted onto a USB to give to the police[141]. In answer to the court Mr Chim said the copies that he made for the police were from the clean version. 196. In cross-examination Mr Chim said he did not know who the cameramen were who filmed the footage provided to the police; did not know who edited the footage; that no records were kept by ATV showing who edited the footage; and did not know who was responsible for the post-production copy[142]. 197. I was satisfied, if believed, that the evidence of Mr Chim was sufficient to prove the authenticity of the ATV footage beyond reasonable doubt. I therefore ruled the ATV footage prima facie authentic and admissible in evidence. 198. In addition, as summarised earlier the ATV footage P3 (2) showing Tsang being subdued and escorted was played to Tsang, who confirmed the footage was consistent with his recollection of the events that night. I was satisfied this evidence, if believed, was also sufficient to prove the authenticity of this part of the ATV footage beyond reasonable doubt. 199. The footage was also played to Sgt 34200 (PW2); SIP Fu (PW3); SSgt 52820 (PW4) and SIP Wat (PW5), who all confirmed that the footage was consistent with their recollection of the events of that night[143]. I was satisfied this evidence, if believed, was also sufficient to prove the authenticity of this part of the ATV footage beyond reasonable doubt. Discussion 200. As noted earlier one of the main criticisms of all the video footage was that there was no continuous recording[144]. In cross-examination Mr Chim agreed that the length of raw footage could be cut; the sequence of events may be edited; that depending on the narration it was possible a subsequent event may become an earlier event; and that he was unable to say whether the footage supplied to the police was the true sequence of events[145]. 201. Specifically, complaints were made that the prosecution failed to prove the provenance of the assault footage, the footage having originated from Apple Daily and that the footage of the apprehension is prejudicial to the defendants. Assault footage originating from Apple Daily 202. Mr Choy submitted that the prosecution failed to adduce any evidence on the provenance and/or authenticity of the ATV footage, in particular the footage which originated from Apple Daily showing the assault at the substation[146]. A similar submission was made by Mr Lo[147]. 203. I have no hesitation in rejecting this submission. Mr Chim clearly explained the making of the broadcast copy and that the footage from Apple Daily was normally obtained direct from Apple Daily. For the reasons already given I am satisfied the prosecution have proved the authenticity of the Apple Daily footage beyond reasonable doubt. Prejudice 204. Mr Cheng SC submitted that the ATV footage of the apprehension of Tsang was extremely prejudicial to the defendants as this created an impression that minimal violence was actually used in subduing him[148]. A similar submission was made by Ms Lam[149]. 205. I have no hesitation in rejecting this submission. The footage, in my view, does not give the impression contended by Mr Cheng SC. The footage only shows what happened after Tsang had fallen to the ground and was about to be handcuffed with zip wire and not how he fell to the ground. 206. Six police officers have given evidence of how they subdued Tsang and Tsang has given evidence as to how he says he was subdued. Whether or not in the course of subduing Tsang the police used violence or Tsang himself used violence, is to be determined after consideration of all the evidence and not just what is seen in the footage. I found nothing prejudicial in the admission of this footage. 207. I was satisfied that I could compare the footage in P3 (1) with P3 (2) to see whether the footage downloaded by the police from the ATV website in November 2014 was the same as that kept by ATV. I am satisfied the footage broadcast in P3 (1), from the point where the two presenters had finished their introduction is the same footage seen in P3 (2), save that the courtesy and other captions are not seen. 208. No evidence was called as to when exhibit P3 (1) was uploaded by ATV onto their website. According to the footage the broadcast was the same day, 15 October at 18:00 hours. Further during the broadcast the female news reader refers to the announcement by the Chief Executive that the question and answer session at the Legislative Council the following day will be cancelled[150]. That announcement was made at 3:30 p.m. on 15 October 2014[151]. 209. I am satisfied the broadcast was the same day, 15 October 2014. Mr Leung SC referred the court to the passage in R v Murphy where the court referred to the trial judge’s observations that it seemed highly unlikely that a film clip shown the same day on the BBC news was a fabrication or had been altered or tampered with so that it ceased to be an accurate record of what the cameraman had seen[152]. Conclusion 210. I accept the evidence of Mr Chim. Although Mr Chim was not responsible for making the library copy supplied to the police, taking into account the evidence of Mr Chim, including that the cameraman could only record and not edit; the content of the master tape was therefore the same as the footage filmed by the cameraman; that P3 (1) was broadcast the same day; the footage broadcast was identical to the footage kept in the library of ATV, I find the prosecution have proved the authenticity of the ATV footage beyond reasonable doubt. 211. I am satisfied there was no tampering of the footage. In cross-examination when asked about electronic editing and whether the director and post-production team use different equipment to create images for broadcasting, Mr Chim replied “They would not produce a different footage for broadcasting, they would process it with the equipment and then broadcast it” [153]. I reject the suggestion that the footage is selective and has been tampered with whereby it ceases to be an accurate record of what the cameraman had seen. Having carefully considered the evidence and viewed the footage I am satisfied there are no signs of tampering or discontinuity which cause me to doubt the authenticity, accuracy and integrity of the ATV footage. 212. In addition, I accept the evidence of Tsang and the police officers that the ATV footage played to them in court was consistent with their recollection of the events that night. On this evidence I find the prosecution have proved the authenticity of the ATV footage beyond reasonable doubt. I discuss later the credibility of Tsang and the police officers. Discretion to exclude 213. In ruling the ATV footage was prima facie authentic and finding the footage authentic I carefully considered the submissions that I should exclude the evidence, including that Mr Chim had no personal dealings with the video footage; did not know the cameraman who filmed the footage; did not know who edited the footage for broadcasting; nobody was called from ATV to confirm or verify when the footage was uploaded to the ATV website; and the quality of the footage[154]. 214. I found no grounds on which to exclude the ATV footage on the grounds of unfairness. None of these matters cause me to doubt the findings I have made. Now TV (exhibit P4) 215. The Now TV footage consists of one disc showing Tsang being escorted(between 40-44 seconds) and part of the assault at the substation(between 18-29 seconds). The Now TV footage was downloaded by the police from the website of Now TV[155]. 216. Nobody from Now TV was called to give evidence. The footage was shown to Tsang, who confirmed the footage was consistent with his recollection of the events that night. 217. In addition, the footage between 31-40 seconds shows Tsang pouring liquid on the police. Admitted in evidence is the police video, exhibit P8 which, between 06:13 and 06:57, also shows Tsang pouring liquid on the police[156]. 218. Both the police footage and the Now TV footage were played to Sgt 47574 (PW1). Sgt 47574 confirmed the police footage showed him getting hold of the man pouring liquid, part of which he said could also be seen on the Now TV footage[157]. The police footage was also played in cross-examination to Tsang[158], who said he was the one splashing liquid on the police. 219. I was satisfied, if believed, that the evidence of Tsang and Sgt 47574 together with the police footage, exhibit P8, was sufficient evidence to prove authenticity of the Now TV footage beyond reasonable doubt. I therefore ruled the Now TV footage prima facie authentic and admissible in evidence. 220. I accept the evidence of Sgt 47574, whose evidence was not challenged[159]. The evidence of the Sgt was supported by the police video, exhibit P8. I accept Tsang’s evidence that the Now TV footage was consistent with his recollection of the events that night. 221. Having carefully considered the evidence and viewed the footage I am satisfied there are no signs of tampering or discontinuity which cause me to doubt the authenticity, accuracy and integrity of the Now TV footage. I find the prosecution have proved the authenticity of the Now TV footage beyond reasonable doubt. Discretion to exclude 222. In ruling the Now TV footage was prima facie authentic and finding the footage authentic, I carefully considered the submissions that I should exclude the evidence, including the absence of the cameramen; nobody was called from Now TV to confirm when the footage was filmed; and the quality of the footage[160]. 223. I found no grounds on which to exclude the Now TV footage on the grounds of unfairness. None of these matters cause me to doubt the findings I have made. Photographs Apple Daily (exhibit P12); Oriental Daily (exhibit P13) 224. Both the photographs were downloaded on 24 March 2016 by Sgt 54162 (PW23) from the websites of Next Media and Oriental Daily[161]. 225. Nobody from Apple Daily or Oriental Daily was called to give evidence about the photographs. Both photographs were shown to Tsang, who described P12 as showing the time when he was still walking after he had been handed over to the second group of police officers and P13 as showing him being carried face down after he had been lifted off the ground by his arms and legs. 226. I was satisfied, if believed, that the evidence of Tsang was sufficient evidence to prove the authenticity of the two photographs beyond reasonable doubt. I therefore ruled the two photographs prima facie authentic and admissible in evidence. 227. In addition Mr Leung SC submitted that the photographs matched what was captured in the TVB footage[162]. 228. In ruling the photographs prima facie authentic I did not rely on the evidence of Sgt 58223 (PW19) that on 4 June 2016 he also located the same two photographs (exhibit P18) from the websites of Apple Daily and Oriental Daily, in the news and past news sections respectively of the two websites. 229. I accept the evidence of Tsang that he was the person seen in both photographs. I am satisfied so I am sure that the photograph exhibit P12 was taken during the escorting of Tsang as seen in the TVB footage P1 (b), (d) & (h) between 5-15 seconds and (g) between 02:47-02:55[163]. 230. Mr Lam submitted that the facial features not being shown on photograph exhibit P13 there was no cogent evidence to suggest the person carried was Tsang and that Tsang’s identification by clothing was unsafe[164]. The fact Tsang was carried face down and therefore his facial features are not seen in the photograph do not cause me to doubt Tsang’s evidence that he was the person in the photograph being carried face down. Tsang identified himself by reference not only to his black T-shirt with stars on the back but also because he was carried in this way face down. 231. In addition, I agree with Mr Leung SC that the number and appearance of persons shown in the photograph exhibit P13 together with the colour and pattern and type of clothing matches that seen in the news footage, in particular the TVB footage P1 (b), (d) & (h) between 5-15 seconds and (g) between 02:47-02:55 showing Tsang being escorted just prior to being lifted up and carried face down. 232. I am satisfied there are no signs of tampering which cause me to doubt the authenticity, accuracy and integrity of either photograph. I find the prosecution have proved the authenticity of the two photographs beyond reasonable doubt. Discretion to exclude 233. In ruling the photographs prima facie authentic and finding the photographs authentic, I carefully considered the submissions that I should exclude the photographs, including that nobody was called from Apple Daily or Oriental Daily to testify when, where and by whom the photographs were taken; when the photographs were uploaded onto the websites; and whether the photographs had been edited[165]. 234. I found no grounds on which to exclude either of the photographs on the grounds of unfairness. None of these matters cause me to doubt the findings I have made. CCTV recordings (exhibit P15 (a) & (b)) 235. As noted earlier the grounds of objection to the admissibility of the CCTV recordings were that there was no or no sufficient evidence to establish the authenticity of the CCTV recordings and the accuracy of the recordings[166]. 236. The prosecution intended only to call SPC 16408 (PW21) who burnt the CCTV footage and Tsang to identify himself on the CCTV footage. In light of the grounds of objection two additional witnesses were called DSgt 47147 (PW22) and DSgt 45918 (PW24). A further witness DSgt 50037 (PW25) was tendered for cross-examination[167]. All three officers were from the Complaints Against Police Office (“CAPO”)[168]. Evidence Tsang Kin Chiu 237. The following is a summary of the evidence given by Tsang on the voir dire relating to the admissibility of the CCTV recordings[169]. Tsang testified that after he was assaulted the group took him behind the substation to Lung Wo Road where he boarded a private car. Two of the group that assaulted Tsang sat on either side of him in the back of the car. The group together with the driver discussed whether it was necessary to wait for others. After waiting several minutes the car was driven to the Central Police Station[170]. 238. Tsang was taken into the police station and escorted to interview room 7 by the same two police officers who had sat on either side of him in the car. Tsang guessed he was inside room 7 for about half an hour. Tsang was then taken by the same two police officers on board a coach and escorted to the Police Training School in Wong Chuk Hang[171]. 239. The CCTV recordings, exhibit P15, file 034000, cameras 12 & 14 and file 045000, cameras 1, 2 & 14 were played to Tsang[172]. Tsang identified himself as the person in the black T-shirt, with stars on the back. Tsang said file 034000, cameras 12 & 14 showed him being escorted by the two police officers to interview room 7 and file 045000, cameras 1, 2 & 14 showed him being escorted inside the police station by the same two police officers. In answer to the court Tsang said he did not know at the time where he was walking to in file 045000 because he had never been to the police station before. Tsang said the footage was consistent with his recollection of the events that night. Police evidence 240. In summary SPC 16408 (PW21) testified that in 2014 he was attached to the Administrative Support Section (“ASS”) at the Central Police Station[173]. His duties involved operating the CCTV system of the police station, which system had been installed by Chubb. The CCTV system was divided into four areas, each with their own recording machine and which operated separately. Area 1 (DVR 1) covered the vehicular exit/entrance of the police station. Area 4 (DVR 4) covered the report room, which as far as SPC 16408 remembered consisted of 16 cameras. 241. The CCTV system was for both surveillance and video recording. The video recording was done automatically. The footage was saved in the hard disc for around 30 days after which the footage would be automatically overwritten. The real time would be displayed on the recording. 242. All members of the ASS could access the hard disc with the prior approval of the Chief Inspector of ASS. After approval was given the members needed to key in a password to gain access. In cross-examination SPC 16408 said there were altogether six members in the team, two of whom could access the CCTV system with a password[174]. In re-examination SPC 16408 clarified that in addition to him there were two others who could access the system with a password and said that with the password the member was not able to alter or edit the content of the hard disc. 243. On 27 October 2014 SPC 16408 received instruction from his superior to assist officers from CAPO to burn discs of Area 1 and Area 4 for the period 3:35 - 4:55 a.m. on 15 October 2014. SPC 16408 had received one day’s training around January 2012, mainly on how to burn discs and including an explanation of the functions of the system. 244. When SPC 16408 checked the machines he found a problem with the operating system of DVR 1. No images had been recorded. DVR 4 was normal therefore SPC 16408 burnt the images onto a disc. This took about 20 minutes to complete. DVR 4 was in working order during the burning of the disc. In cross-examination SPC 16408 said he did not view the CCTV footage before burning the disc[175]. 245. SPC 16408 explained how he burnt the disc. The first step was to access the function for burning discs. A window would then automatically pop up into which SPC 16408 keyed in the date, the start time and the end time of the footage he required. The machine would then automatically search for the footage. After searching a window would pop up asking a disc to be inserted. The machine would then automatically save the footage onto the disc. When the copying was completed the tray would be automatically ejected and the disc removed. 246. After burning the disc SPC 16408, together with DSgt 47147 (PW22), took the disc to his superior Sgt 14827[176], who asked DSgt 47147 to sign acknowledging receipt. DSgt 47147 produced the receipt (exhibit P19). 247. At the beginning of cross-examination when Mr Chung asked how many times he did the disc burning exercise on 27 October 2014, SPC 16408 replied twice. SPC 16408 went on to explain that at the request of an officer from CAPO he had burnt two discs, one to be produced as an exhibit and one for investigation. In re-examination SPC 16408 explained that after burning one disc he burnt the second disc. 248. SPC 16408 opened one of the discs to show DSgt 47147 how to view the disc. The content of the disc was not however viewed at that time. SPC 16408 told DSgt 47147 the two discs were the same therefore the second disc was not opened. In cross-examination DSgt 47147 said this was the normal practice otherwise he would not know how to view the disc[177]. 249. DSgt 47147 then returned to his office and handed over the discs to DSgt 45918 telling DSgt 45918 how to view the discs. DSgt 45918 assigned one disc as the exhibit (exhibit P15 (a)) and attached a yellow exhibit label (exhibit P15 (c)). The second disc (exhibit P15 (b)) was kept as the working copy. 250. I was satisfied, if believed, that the evidence of SPC 16408, DSgt 47147 and DSgt 45918, was sufficient evidence to prove the authenticity of the CCTV recordings beyond reasonable doubt. I therefore ruled the CCTV recordings prima facie authentic and admissible in evidence. In addition, I was satisfied the evidence of Tsang, if believed, was also sufficient evidence to prove the authenticity of the CCTV recordings beyond reasonable doubt. Discussion 251. In the first part of his submission, what is called “the chain issue”, Mr Chung submitted the court cannot be satisfied that the disc, exhibit P15 (a) was burnt by SPC 16408 and that DSgt 45918 was given two discs[178]. Ms Lam also submitted the prosecution could not prove the chain of the two discs[179]. Identification of P15 (a) 252. When the CCTV footage was shown to Tsang no disc was produced to court as an exhibit. The two files shown to Tsang were referred to as PP15 (a) and (b)[180]. 253. In examination-in-chief SPC 16408 was shown a disc which he identified, as being the copy he burnt, from the writing on the disc. SPC 16408 wrote on the disc after burning the disc. The disc was marked as PP15. The yellow exhibit label (P15(c)) attached to the disc was not marked at this time. 254. Towards the end of cross-examination Mr Chung showed a second disc to SPC 16408. SPC 16408 again identified the disc from the writing on the disc, which he also wrote after burning the disc. At this stage the reference to the files shown to Tsang as PP15 (a) and (b) was no longer used. The disc shown to SPC 16408 in examination-in-chief was marked PP15 (a) and the disc shown in cross-examination marked PP15 (b). The yellow exhibit label attached to exhibit PP15 (a) was marked as exhibit PP15 (c) during the evidence of DSgt 45918 (PW24). 255. In cross-examination SPC 16408 said that apart from the words “Recorded by SPC 16408” written on the disc, exhibit PP15 (b), everything else on the two discs was written by him. SPC 16408 disagreed that the difference in what was written on the two discs was because only one disc was burnt[181]. In re-examination SPC 16408 said the words “Recorded by SPC 16408” were not written in his presence. 256. The evidence of DSgt 47147 differed from the evidence of SPC 16408[182]. DSgt 47147 testified the writing on the discs in red (RR CDIV/RR CDIST) was written by him after he returned to his office. In cross-examination DSgt 47147 disagreed that the difference in the red writing between the two discs was because they were not written at the same time and that he only had one disc in his possession[183]. 257. DSgt 47147 said when he wrote the words in red the words in black “Recorded by SPC 16408” had already been written on the disc, exhibit PP15 (b). In cross-examination DSgt 47147 said that after the discs were handed to him by SPC 16408 they never left his possession until after he had written the words in red. 258. Ms Lam submitted that someone else must have written “Recorded by SPC 16408” before the discs were passed to DSgt 47147 and that there was no evidence who wrote those words[184]. 259. The discs were burnt in October 2014. I do not find surprising that nearly 20 months later there is discrepancy in the evidence as to the writing on the discs. The identification by SPC 16408 clearly comes from the date on the discs (2014-10-15) and the period that was copied (03.35-04.55), which were in the handwriting of SPC 16408 and were the date and time SPC 16408 was requested to copy. The discrepancy in the evidence as to the red handwriting and when the words “Recorded by SPC 16408” were written does not cause me to doubt the evidence of SPC 16408 that he burnt two discs, exhibits P15 (a) and (b). 260. In addition, as was acknowledged by Mr Chung in court and in his written submission, the video content of the two discs are the same[185]. This lends support to the evidence of SPC 16408 that he burnt two discs. 261. Mr Leung SC submitted support for the evidence of SPC 16408 that he burnt two discs could also be found from looking at the data on the content page of the two discs, which showed that the discs were recorded one after the other[186]. I am satisfied Mr Chung is correct that before placing any reliance on this data the court requires expert evidence as to the meaning of the data[187]. 262. I do however agree with the submission of Mr Leung SC that the difference in the content pages of the two discs show that one disc is not a duplicate of the other. This submission was made in response to the suggestion put by Mr Chung to DSgt 45918 that he had made a copy of the disc given to him[188]. The data in this context is used not to prove the truth of the data but to show the data on the content page of each disc is different and therefore not a duplicate of each other. 263. Mr Chung further submitted that the evidence of SPC 16408 that he downloaded the footage in “1 file” and that the burning time would not be recorded in the disc created further doubt as to the authenticity of the two discs[189]. I have no hesitation in rejecting this submission. 264. The reference to “1 file” was when Mr Leung SC asked SPC 16408 if he copied the footage in one file or more than one file. The answer copied on one file must be viewed in context that the CCTV footage was automatically recorded on one disc for the whole period input by SPC 16408. 265. SPC 16408 did not say affirmatively that the burning time was not recorded on the discs. What he said was that as far as he remembers this was not recorded. This evidence does not cause me to doubt that SPC 16408 burnt two discs. 266. I have no hesitation in rejecting the submission that SPC 16408 did not burn the discs, exhibits P15 (a) and (b). The number of discs handled by DSgt 45918 (PW24) 267. Mr Chung submitted that all records show only one disc was handed over to DSgt 47147[190]. Acknowledgement of Receipt of a Copy of DVD (exhibit P19). 268. Mr Chung submitted that notwithstanding the nature of the document is said to be 2 DVD, the main content of the receipt refers to a single copy[191]. In cross-examination DSgt 47147 agreed that the main part of the receipt, “I DSgt 47147 acknowledge the receipt of a copy of the above-mentioned DVD” referred to a single copy. Ms Lam in her submission also refers to this part of the receipt[192]. Similarly, DSgt 47147 agreed that “This copy was issued to me”, also referred to a single copy. 269. DSgt 47147 however disagreed with Mr Chung that he only received one disc. In understanding DSgt 47147’s apparent agreement that the receipt only refers to a single copy it is to be noted that Mr Chung never drew to the attention of DSgt 47147 “2 DVD” under nature of document. 270. I do not agree with the interpretation sought by Mr Chung and Ms Lam that the receipt is for a single copy only. In examination-in-chief DSgt 47147 said the receipt was for acknowledging receipt of the two discs. Although the main content is written as a copy, this is “receipt of a copy of the above-mentioned DVD”. The above-mentioned DVD is 2 DVD. Computer certificate 271. Before calling SPC 16408 to give evidence the prosecution sought to admit in evidence pursuant to Section 22A of the Evidence Ordinance[193], a computer certificate relating to the making of a copy on 27 October 2014. The certificate was dated 4 June 2016 and made after the defence indicated on the first day of trial that the admissibility of the CCTV recording was to be challenged. There being no certified translation of the certificate I ruled the certificate inadmissible[194]. 272. Mr Chung submitted that the content of the certificate clearly indicated SPC 16408 had burnt only one disc on 27 October 2014, which disc was handed over to DSgt 47147[195]. Ms Lam also referred to the certificate mentioning only one disc[196]. 273. The certificate referred only to the disc intended to be produced in court by the prosecution (i.e. exhibit P15 (a)). As summarised earlier in examination-in-chief SPC 16408 referred only to one disc and when asked at the beginning of cross-examination how many times he did the disc burning exercise SPC 16408 said he had burnt two copies. 274. Towards the end of cross-examination SPC 16408 agreed that paragraph 6 of the certificate referred to the making of one copy, which was handed over to DSgt 47147. SPC 16408 explained that when he read the certificate he thought he had only recorded one disc. SPC 16408 disagreed with the suggestion put by Mr Chung that he had in fact burnt only one disc. 275. This was not the first time that SPC 16408 had said he burnt two discs. In cross-examination Mr Chung asked SPC 16408 whether he was given anything before signing the certificate to assist him remembering what happened on 27 October 2014. SPC 16408 told the court that he was given his witness statement made on 27 October 2014. Mr Chung then took SPC 16408 through his witness statement including that he said he burnt a total of two discs. SPC 16408 confirmed the content of his witness statement was true and accurate. 276. I accept the explanation given by SPC 16408 that on reading the certificate he thought he had only recorded one disc. The fact the certificate referred to only one disc did not cause me to doubt the evidence of SPC 16408 that he had burnt two copies, both of which were given to DSgt 47147. 277. Ms Lam submitted the first time SPC 16408 mentioned he gave a disc to DSgt 47147 was in the certificate, whereas in his witness statement he said he gave the discs to his superior, with no mention being made of any discs handed to DSgt 47147[197]. In cross-examination when asked by Mr Chung why there was no mention in his witness statement of dealing with DSgt 47147, SPC 16408 replied that he took DSgt 47147 to his office and passed the discs to his superior, who asked DSgt 47147 to sign the acknowledgement. This reply was consistent with the evidence given by SPC 16408 in examination-in-chief[198]. 278. I accept the evidence of SPC 16408 that he passed the discs to his superior Sgt 14827, who asked DSgt 47147 to sign the acknowledgement. Again I do not find surprising that nearly 20 months later there is a discrepancy in the evidence as to whether SPC 16408 personally passed the discs to DSgt 47147 or gave them to his superior who, in his presence, handed them to DSgt 47147. This discrepancy in the evidence does not cause me to doubt SPC 16408’s evidence that he burnt two discs which were given to DSgt 47147. Nor does the fact SPC 16408 made no mention in his witness statement of dealing with DSgt 47147 cause me to doubt SPC 16408’s evidence that he burnt two discs which were given to DSgt 47147. Yellow exhibit label, exhibit P15 (c) 279. The reverse of the exhibit label clearly refers to one piece of DVD. In examination-in-chief DSgt 45918 said DSgt 47147 decided which disc was to be the exhibit whereas in cross-examination DSgt 45918 said he decided which disc was to be the exhibit. When the court clarified which answer was correct DSgt 45918 said he decided which disc was to be the exhibit. 280. Ms Lam submitted that this discrepancy could not be reconciled and cast doubt on the credibility and reliability of the evidence of DSgt 45918[199]. I have no hesitation in rejecting this submission. Whoever decided which disc was to be the exhibit and which was to be the working copy, is in my view not material. This discrepancy in the evidence does not cause me to doubt that DSgt 47147 was given both discs which he handed to DSgt 45918. 281. The fact the exhibit label refers to only one disc is explained by the fact the disc was the exhibit to be produced in court as opposed to the disc kept as a working copy for which no exhibit label was attached. 282. In his submission Mr Chung questioned why the exhibit label stated the disc was seized on 29 October 2014 if in fact DSgt 47147 handed the disc to DSgt 45918 on 27 October 2014[200]. 283. In cross-examination DSgt 45918 said that he forgot to make a record that he received two discs from DSgt 47147 on 27 October 2014; did not make a record of which disc was assigned as the exhibit and which disc was the working copy; and did not record this information in the investigation report (Pol. 155), which was written by him. 284. DSgt 45918 was not the only one who forgot to make a record of how many discs he received. In cross-examination DSgt 47147 agreed that there was no record of what he had done at the Central Police Station on 27 October 2014 in either the investigation report (Pol. 155) or his witness statement dated 4 February 2015; and that when he did mention this in his witness statement dated 6 June 2016 he did not specify how many discs he received from SPC 16408[201]. 285. I agree with the criticisms made of DSgt 47147 and DSgt 45918 for their failure to keep proper records. This failure however does not cause me to doubt the evidence that DSgt 47147 was given both discs which he handed to DSgt 45918. I have no hesitation in rejecting the submission DSgt 45918 was given only one disc. 286. In making his submission on “the chain issue” Mr Chung also referred to what he described as the confusion over the evidence relating to the sketches (exhibit P16) drawn by DSgt 45918[202]. I will address this next when considering the second part of Mr Chung’s submission, “the content issue”. The content issue 287. The second part of Mr Chung’s submission concerned whether the footage had been tampered with after recording whereby the footage might not contain what actually transpired[203]. This submission was largely based on the evidence of DSgt 45918, in particular the sketches he drew of the Central Police Station showing first 12 cameras, then 14 cameras and finally 16 cameras, whereas the CCTV recordings show 16 cameras[204]. A similar submission was made by Ms Lam that the reliability or accuracy of the CCTV footage was in doubt and that the integrity of the exhibits had not been proved[205]. 288. In summary on 28 October 2014 DSgt 45918, with the assistance of DSgt 50037 (PW25), drew a sketch showing 12 cameras. On returning to his office that day DSgt 45918 made a copy of the sketch (exhibit P16 (b1)), which he placed in the file. 289. That evening DSgt 45918 viewed for the first time the working disc given to him the day before by DSgt 47147. After viewing the CCTV DSgt 45918 realised there were 16 cameras and therefore added cameras 13 & 14 to the original sketch (exhibit P16 (b)). DSgt 45918 did not add cameras 15 & 16 at that time because he was of the view they faced the outside of the report room and not much was captured on the cameras. 290. Only on 3 June 2016 did DSgt 45918 draw the sketch with 16 cameras (exhibit P16 (a)), so as to let the court know there were in fact 16 cameras[206]. 291. At the time of drawing the original sketch DSgt 45918 realised he wrongly drew the door of Room 7 and the position of camera 12. DSgt 45918 therefore used tippex to correct these mistakes. The tippex marks can clearly be seen on the original sketch (exhibit P16 (b)). 292. The evidence of DSgt 45918 was subject to lengthy cross-examination by Mr Chung[207]. DSgt 45918 explained that when he looked at the CCTV system in the police station he saw 12 cameras and that camera 9 was not working. Although a technician from Chubb was present he did not assist DSgt 45918 to locate the cameras. Neither did anyone from the Central Police Station assist in locating the cameras. DSgt 45918 together with DSgt 50037 went round the police station locating the position of the cameras. 293. In cross-examination DSgt 50037 explained that together with CIP Law; DSgt 45918 and DSgt 47147 he went to the Central Police Station on 28 October 2014 to view the scene where the incident was said to have occurred. The team were interested in the CCTV system in particular whether anything showed the complainant was in the police station; what happened to him and the recording of the vehicular entrance/exit of the police station. With the assistance of a technician from Chubb, who was present at the time, the team looked at the CCTV system. 294. DSgt 50037 agreed having found 12 camera images he walked around the police station pointing out the cameras to DSgt 45918, who drew their location on a sketch[208]. DSgt 50037 identified the sketch as (exhibit P16 (b1)). With regard to camera 9, because this was not working and only a blue screen with no image was shown, DSgt 50037 guessed the location of the camera as being between cameras 8 & 10. In re-examination DSgt 50037 said he asked the uniform officers in the report room how many cameras there were but they were not clear about the number and could only say how many cameras by looking at the number of camera images. 295. DSgt 45918 said the copy was placed in the investigation file and the original kept in the drawer of his office. After amending the original DSgt 45918 put the original sketch back in his drawer. In answer to the court DSgt 45918 said that he was negligent in not making a copy of the sketch after amending the sketch to include cameras 13 &14. 296. DSgt 45918 explained that the sketch (exhibit P16 (a)) was drawn to show the court there were 16 cameras. In cross-examination DSgt 45918 said the sketch was drawn at the request of the OC case. DSgt 45918 did not at that time tell the OC case that he had amended the original sketch from 12 cameras to 14 cameras because he had forgotten about this[209]. DSgt 45918 said he only remembered this when he was shown the sketch (exhibit P16 (b)) in court and explained that he handed the sketch to the exhibit room on 6 January 2015. The sketch was entered into the property register (Pol. 69A) and a yellow exhibit label (Pol. 195) filled out and attached to the outside of the envelope containing the sketch[210]. 297. In answer to the court DSgt 50037 explained that the public entrance to the police station was below camera 7 and that the entrance for police officers, arriving at the car park with arrested persons was near cell 3. DSgt 50037 said that either camera 1 or camera 3 showed the corridor leading to this entrance. DSgt 50037 marked in red, on a copy of the sketch (exhibit P16 (b2)), the door of this entrance[211]. In questions arising Mr Leung SC showed DSgt 50037 the 16 cameras as seen on the CCTV recording, exhibit P15 (b). DSgt 50037 explained that camera 1 showed the corridor leading to the entrance marked in red but no camera captured the entrance. DSgt 50037 then marked on the copy a green door representing the door kept open by a cone and a green star representing the end of the corridor shown in camera 1. The copy marked by DSgt 50037 was marked exhibit P16 (b2)). Discussion 298. The evidence of DSgt 45918 was extensively criticised by both Mr Chung and Ms Lam[212]. As noted earlier in cross-examination DSgt 45918 agreed that he omitted to record in the investigation report (Pol. 155) that on 27 October 2014 he had received two discs from DSgt 47147[213]. Similarly, DSgt 45918 did not record in the investigation report that he viewed the discs; he realised there was 16 cameras; and that he amended the original sketch drawn in the Central Police Station. 299. Mr Chung submitted the failure of DSgt 45918 to make a record of amending the sketch is difficult to accept, in particular considering the investigation report was detailed enough to include making a photocopy of a document in the Kwun Tong Police Station. Mr Chung further submitted that the explanation given in cross-examination by DSgt 45918 that he forgot to record this is not believable, Mr Chung pointing to the fact that DSgt 45918 also forgot to make a copy of the sketch after he had added cameras 13 & 14 saying that he was negligent[214]. 300. Ms Lam submitted that the failure to make a record of drawing the sketch, exhibit P16 (b), which sketch had not been disclosed to the defence, was a strong indication that DSgt 45918 was not telling the truth about seeing 16 cameras when he viewed the working disc (exhibit P15 (b)) on 28 October 2014[215]. Ms Lam also submitted that the evidence of DSgt 45918 that he forgot to tell his superior about the sketch, exhibit P16 (b), when asked to draw a sketch showing 16 cameras on 3 June 2016 defied logic and should not be believed[216]. 301. The evidence as to the drawing of the sketch and the failure by DSgt 45918 to make a record of what he did are again properly criticised. DSgt 45918 however did make a record of what he saw when viewing the working copy. This arose in cross-examination when Mr Chung asked DSgt 45918 further questions about the investigation report. This evidence was also subject of criticism by both Mr Chung and Ms Lam[217]. 302. The earlier references made to the investigation report were to a three page extract covering 24–31 October 2014, including entries for 28 October 2014[218]. Later Mr Chung showed DSgt 45918 a one page extract, which had also been written by the sergeant. This extract consists of only two entries for 28 October 2014[219]. Mr Chung submitted copies to the court for the purpose of showing the court the layout of the investigation report, in particular that there were two separate extracts covering entries for 28 October 2014[220]. 303. DSgt 45918 said he should have written the one page extract on 28 October 2014. Asked why this was not included in the three page extract after the entry for 17:00 on 28 October 2014 and before the next entry for 10:45 on 29 October 2014, DSgt 45918 replied that because he executed many things at that time he was confused and did not make the record. Subsequently DSgt 45918 remembered, so he wrote the one page extract. DSgt 45918 agreed that when he wrote the entry for 10:45 on 29 October 2014 he had not yet written the one page extract. On being pointed out the extract could not therefore have been written on 28 October, DSgt 45918 said he could not be sure and could not remember when he wrote the extract. 304. DSgt 45918 confirmed that the entries on the one page extract recited that between 18:10 and 22:00 hours on 28 October 2014 he inspected the CCTV record seized from the Central Police Station and recorded a written transcript. DSgt 45918 explained that when viewing the CCTV, he wrote down on a piece of paper a draft of what he had seen on the CCTV. DSgt 45918 then typed up the draft on his computer and printed out a copy for investigation purpose. The draft was thrown away. 305. DSgt 45918 said a copy of the print out was kept in a file in his office, in which he put material for his own use. In answer to the court DSgt 45918 said this was different to the investigation file. In re-examination Mr Leung SC asked DSgt 45918 to produce this file. 306. Over the lunch adjournment in the presence of a representative from the police and the defence, DSgt 45918 retrieved the file. Leave was given to the defence to further cross-examine DSgt 45918 on the content of the file, the defence having first been supplied with copies of all relevant documents in the file and given the opportunity to go through the file. 307. Mr Chung cross-examined DSgt 45918 on a four page document in the file “Central Police Station Video Images”. DSgt 45918 said this was the print out of what he had viewed on the CCTV. Mr Chung submitted a copy to the court for the purpose of showing the court the layout of the document[221]. 308. DSgt 45918 confirmed the document referred to five persons, two suspects, the victim and two female police officers. DSgt 45918 said he wrote down anything important for example whether the CCTV captured the scene of the incident and if he could see anyone entering or leaving room 7. DSgt 45918 agreed there was no reference to cameras 13 or 14. DSgt 45918 disagreed this was because the disc he viewed only had 12 cameras. 309. Mr Chung submitted that if DSgt 45918 viewed the working disc (exhibit P15 (b)) there was no way he would have missed camera 14, which Tsang Kin Chiu testified showed him entering and leaving the police station[222]. Similarly Ms Lam submitted that the assertions of DSgt 45918 that he found 16 cameras when viewing the working disc should not be believed because he made no mention of camera 14 when recording the details of what he viewed from the disc[223]. 310. I have no hesitation in rejecting the submission that the disc DSgt 45918 viewed did not contain images from camera 14[224]. Although DSgt 45918 said he was instructed to find out if the victim had entered Central Police Station and whether the camera captured room 7, I accept his evidence in cross-examination that he felt it was enough knowing when the victim had entered room 7 and left room 7. Whilst the omission to view camera 14 again reflects poorly on the way in which DSgt 45918 conducted the investigation of the complaint made by Tsang, this does not cause me to doubt he was given the two discs, exhibits P15 (a) & (b). Conclusion 311. I accept the evidence of SPC 16408 that he burnt two discs, exhibits P15 (a) & (b), which were handed to DSgt 47147, who on returning to his office handed them to DSgt 45918. Having carefully considered the evidence and viewed the CCTV recordings, I am satisfied there are no signs of tampering or discontinuity which cause me to doubt the authenticity, accuracy and integrity of the CCTV recordings. I find the prosecution have proved the authenticity of the CCTV recordings beyond reasonable doubt. 312. In addition, I accept the evidence of Tsang that the CCTV recordings showed him being escorted to and from room 7 in the Central Police Station. On this evidence I find the prosecution have proved the authenticity of the CCTV recordings beyond reasonable doubt. Discretion to exclude 313. In ruling the CCTV recordings were prima facie authentic and finding the recordings authentic, I carefully considered the submissions that I should exclude the evidence, including the many criticisms of the evidence of the CAPO officers investigating the complaint made by Tsang, namely DSgt 47147, DSgt 45918; and DSgt 50037, in particular the handling of the CCTV recordings, exhibit P15 (a) & (b) and the drawing of the sketches, exhibit P16 (a), (b) & (b1). 314. Mr Chung submitted that the estimate of time Tsang said he was in room 7 was about half an hour is large enough to be taken into account when assessing whether the contents of the CCTV recording had been tampered with[225]. On the general issue Tsang guessed he was in room 7 for an hour or within an hour. This difference in the time estimate Mr Chung submitted showed the unreliability of Tsang’s evidence[226]. 315. I have no hesitation in rejecting these submissions. The times as Mr Chung accepted were only estimates. The different time estimates as to how long Tsang was in room 7 do not cause me to doubt Tsang’s evidence. The CCTV recordings show Tsang to have been in room 7 just over an hour. Having not long before been assaulted it is not surprising he is not precise when first giving a time estimate. 316. I have no hesitation in rejecting the submission of Ms Lam that the description of the CCTV system by DSgt 50037 was different to that of SPC 16408[227]. In cross-examination DSgt 50037 said he understood there were two machines responsible for recording the CCTV, which he called DVD 1 and DVD 2, as opposed to SPC 16408 who said the system was divided into four areas, each with their own recording machine. This difference was more apparent than real. Both SPC 16408 and DSgt 50037 testified that DVR1/DVD 1 was for the entrance/exit of the car park of the police station and was out of order and that DVR 4/DVD 2 was for the report room. 317. I have also taken into account that no one from Chubb was called to give evidence that the CCTV system was functioning properly [228]; no evidence was called to show those who had access to the CCTV recording had not tampered with the recording[229]; SPC 16408 did not compare what he recorded on the two discs with the CCTV footage[230]; and the location of camera 9 on the sketches was clearly wrong[231]and that none of the police officers were shown the CCTV recordings. 318. Having carefully considered all the evidence and submissions I found no grounds on which to exclude the CCTV recordings on the grounds of unfairness. None of the criticisms made of the evidence of the police officers cause me to doubt the findings I have made. Notwithstanding the justifiable criticisms of the police evidence I remain satisfied on the evidence that the prosecution has proved the authenticity of the CCTV recordings beyond reasonable doubt. Charge 1 – Causing grievous bodily harm with intent Credibility of Tsang Kin Chiu 319. Before considering the evidence against each defendant, I will first consider the credibility of Tsang Kin Chiu. This part addresses the many criticisms made of Tsang’s evidence. Some aspects of Tsang’s evidence for example the injuries sustained by Tsang; the identification of D5 and D6; and what happened in the Central Police Station are dealt with later in my verdict. The apprehension of Tsang for pouring liquid on the police 320. Admitted in evidence is that the police video footage shows Tsang Kin Chiu throwing liquid from a plastic bottle over police officers at approximately 3:25 a.m.[232] 321. As part of the prosecution case on the voir dire, Tsang was not shown and therefore not asked whether he was the person seen pouring liquid on the police in the video footage played to Sgt 47574 (PW1)[233]. Neither was Sgt 47574 asked to identify the male he saw pouring the liquid. Nor were the police officers asked to identify the male who was subdued by them and handed over to the crime officers. 322. In cross-examination Mr Lok SC played to Tsang the police video shown to Sgt 47574 and asked Tsang whether he was the one who was pouring the liquid. After the court gave a warning against self-incrimination[234] Tsang answered that he was the one pouring liquid. Tsang agreed that he was charged in relation to his conduct that evening[235]. 323. Mr Choy submitted that Tsang was not a credible witness by reason of the fact he pleaded not guilty to those charges and at his trial it was put to the witnesses that he may not be the person in black splashing liquid on the police[236]. A similar submission was made by Ms Lam[237]. In cross-examination Tsang disagreed with Ms Lam that he was lying when he said he was the one pouring liquid. 324. I accept the evidence of Tsang that he was the one pouring liquid on the police. Not only was this an admitted fact, the video footage also clearly shows that the person pouring the liquid was Tsang. The video footage shows the person pouring liquid was wearing a black T-shirt with stars on the back, the same as worn by Tsang. This is best seen in the police video, exhibit P8, file 00001 between 06:13 and 06:57 when both the front and back of the T-shirt can be seen. 325. The defence further submitted that not only was the evidence given by Tsang inconsistent with his plea of not guilty but was also inconsistent with the evidence of the police officers as to what happened when the police subdued him for pouring the liquid[238]. 326. The prosecution opened their case on the basis Tsang struggled violently[239]. This is consistent with the evidence of the police officers. In cross-examination by Mr Choy on the voir dire, Tsang disagreed he used a high degree of violence; and agreed with Mr Choy his evidence was he simply grabbed and did not resist at all. In cross-examination by Mr Choy on the general issue Tsang disagreed he aggressively tried to arm lock the neck of a police officer; disagreed in trying to resist arrest he struggled with several police officers banging them on the ground; and disagreed only after he violently struggled did the police use pepper spray. 327. Apart from showing Sgt 47574 grabbing hold of Tsang after Tsang was seen pouring liquid on the police[240], the video footage does not show how Tsang was subdued. The ATV video footage only shows the time when Tsang was finally subdued, handcuffed with zip wire and his face washed[241]. 328. Although in cross-examination by Mr Cheng SC on the general issue Tsang said he did not struggle but resisted; resisted strongly but not violently; and struggled violently when the police tried to handcuff him, his evidence was this was because the police assaulted him. I accept the evidence of the police officers as to how Tsang resisted. I reject that the police assaulted Tsang. 329. The fact I do not accept the evidence of Tsang as to how he was subdued does not cause me to doubt his evidence that after he was subdued he was escorted by other police officers; carried by his arms and legs face down and taken to the substation where he was dumped on the ground and assaulted. 330. Nor does the fact that Tsang denied the charges brought against him arising from his conduct that evening cause me to doubt his evidence as to what happened after he was subdued. Tsang exercised his legal right not to give evidence at his trial and have the prosecution prove the charges against him. 331. Nevertheless, I approach the evidence of Tsang with caution. Does the video footage show Tsang Kin Chiu being subdued, escorted, carried and assaulted? 332. As summarised earlier Tsang identified himself on the video footage and photographs being subdued, escorted, carried by his arms and legs face down and assaulted. Tsang said the TVB footage, exhibit P1 (c), from 10 seconds showed him being thrown to the ground and assaulted. When the footage was played again Tsang was asked to stop the video when he could see himself. The footage was stopped at 26 seconds when Tsang said he did not think he could see himself because the video was very blurred. 333. Tsang said there was a high definition version in which he could see himself. The TVB footage with the brightness enhanced, exhibit P1 (h), was then played to Tsang. When the footage was stopped at 18 seconds Tsang said, “The one being carried is me.” 334. On the voir dire the defence submitted that it was not possible for Tsang to identify himself by looking at the video footage of the assault at the substation. Mr Lok SC submitted that Tsang gave no reason as to why he could recognize himself[242]. Similarly Mr Cheng SC submitted Tsang gave no satisfactory explanation how he was able to recognize himself despite the lack of any determining features[243]. Mr Lam also made a similar submission[244]. Ms Lam submitted Tsang fabricated this evidence because he was convicted in the Magistracy[245]. 335. In making his submission Mr Cheng stated that in cross-examination[246]Tsang again identified himself as “the body of mass” in the footage[247]. Although Tsang agreed with Mr Choy that he identified himself as, “that body of weight being carried by those men” this phrase was not used by Tsang. Tsang said, “The one being carried is me”. After the footage exhibit P1 (h) was played again in cross-examination Tsang disagreed with Mr Choy that he was embellishing his evidence because he could not “possibly see any distinguishing features …in that body of weight being carried by those officers”. This was a phrase used by Mr Choy[248]. 336. Mr Cheng SC submitted that by reason of Tsang’s complete inability to recognize himself in the assault footage his evidence is of very limited value, if any[249]; his recollection of events was extremely weak and of dubious value[250] and that on the face of the evidence adduced the court could not conclude that the video footage captured Tsang being assaulted[251]. In closing submissions Mr Lok SC, Mr Lam, Mr Choy and Ms Lam all submitted that the court could not be satisfied that Tsang was the person seen on the video footage being assaulted at the substation [252]. 337. I have no hesitation in rejecting this submission. The evidence of Tsang was that he was carried by his arms and legs face down to the dark corner where he was dumped on the ground face down and assaulted. In my view Tsang was not saying he could see any particular feature to enable him to recognise he was the person being carried but was identifying “the one being carried is me” because he was carried to the dark corner, dumped on the ground and assaulted. I accept the evidence of Tsang. 338. From a very careful viewing of the video footage, in particular by looking at the clothing worn by Tsang, I am satisfied that Tsang is the person seen being assaulted at the substation. 339. Tsang identified himself by reference to his black T-shirt, with stars on the back. As already seen the police video footage exhibit P8, file 00001 between 06:13 and 06:57 shows both the front and back of the T-shirt at the time when Tsang was being apprehended for pouring liquid on the police. The T-shirt is also seen in the first seventeen seconds of the ATV footage exhibit P3 (2) at the time Tsang was lifted up and his face washed. Screen captures 2 & 4 attached to the prosecution’s closing submissions are from this footage and clearly show the T-shirt[253]. 340. The TVB footage exhibit P1 (b), (d) & (h) between 5-15 seconds and (g) between 02:47-02:55 and the Now TV footage between 40-44 seconds show the front of the T-shirt at the time Tsang was being escorted as seen in screen captures 7, 8, 9 & 11 attached to the prosecution’s closing submissions. The photograph exhibit P12 also shows the front of the T-shirt at the time Tsang was being escorted. The photograph exhibit P13 shows the back of the T-shirt at the time Tsang was being carried face down. 341. Looking at the video footage of the assault at the substation when the person assaulted was lifted up and taken away, the stars on the back of the T-shirt can be seen. This is best seen on the Apple Daily footage exhibit P2 (a) and (e) between 01:50 and 02:07 and (c) between 03:38 and 03:55[254]. Screen capture T1 taken from P2 (e) is attached herewith, showing the back of the T-shirt. The TVB footage, exhibit P1 (g) between 03:40 and 03:46; and exhibit P1 (h) between 00:59 and 01:05 also capture part of the back of the T-shirt just after the male was lifted up. 342. In addition, although Tsang was only asked about his upper clothing, it is clear from the video footage he was wearing a pair of shoes with white soles. The Short Arrest Form, exhibit P56 lists Tsang’s clothing on arrest as a black short-sleeved shirt, black trousers and black-white shoes. PC 9765 (PW46) who filled out the Short Arrest Form said he wrote down the clothing worn by Tsang. The tongue and the laces are the black part of the shoes, which can be seen in the CCTV recording, file 03400, camera 12 between 03:43:40 and 03:43:50, as shown in screen capture 21 attached to the prosecution’s closing submissions. 343. The shoes Tsang was wearing can clearly be seen in the first six seconds of the ATV footage exhibit P3(2). The shoes can be said to be distinctive in that on the soles there are three black spots. Screen capture T2 taken from P3 (2) is attached herewith showing the sole of the shoes. 344. The first nine seconds of the TVB footage P1 (a), (c), (e) & (f) clearly shows the soles of the shoes of the person being carried face down. This is also seen between 02:55 and 03:00 in the TVB footage P1 (g). 345. As the person is carried along the side of the substation one of the shoes can be seen in the TVB footage P1 (g) between 03:00 and 03:15 and P1 (h) between 00:14 and 00:26. Screen capture T3 taken from P1 (h) is attached herewith showing one of the shoes. When the person is dropped to the ground both shoes can be seen as shown in screen captures T4 and T5 taken from P1 (h)[255]. 346. Having carefully considered all the evidence and from a very careful viewing of the video footage, I am satisfied the video footage, whilst not continuous, shows in sequence Tsang being apprehended, escorted and carried to the substation where he was assaulted. 347. In reaching this finding I carefully considered the oral submissions of Mr Lok SC and Mr Choy that the T-shirt and shoes were not unique to Tsang; the clothing was that normally worn by young persons and that many protestors were dressed in dark clothing to hide their identity; and the oral submission of Ms Lam that without the clothing being produced, Tsang having refused to provide his clothing to the police[256], the court cannot be satisfied what clothing Tsang was actually wearing. These submissions do not cause me to doubt that the video footage shows in sequence Tsang being apprehended, escorted and carried to the substation where he was assaulted. How many people assaulted Tsang Kin Chiu? 348. On the voir dire Tsang said that from the time when he was carried face down, until he arrived at the substation, no one in the group left. Only after viewing the video footage did Tsang realise someone else joined the group. On the general issue Tsang said from the time when he was picked up to the time he reached the substation, no one put him down or let go of his limbs. 349. When the group reached the substation the video clearly shows seven persons: one in front leading the way; two in front holding Tsang; two at the back holding Tsang with one walking beside them and one following behind. This is best seen in the TVB footage P1 (g) between 03:00 and 03:10 and P1 (h) between 00:14 and 00:21. 350. In the First Information of Complaints Against Police Report, Pol. 964 (exhibit P26), six police officers are complained about. The evidence of DSPC 50117 (PW49), who completed the report, was read (exhibit P58). Mr Cheng SC submitted that the video footage clearly depicting seven and not six individuals, as suggested by Tsang when first making his complaint of assault, tended to suggest that the footage did not depict Tsang being assaulted, but rather, captured a wholly different episode[257]. 351. Ms Lam also made a similar submission that the evidence of Tsang should be rejected because he never mentioned he was assaulted by seven police officers[258]. Mr Choy also submitted that it would be most unsatisfactory to rely on Tsang’s evidence because he failed to identify the number of persons who assaulted him[259]. 352. When describing the assault on the voir dire, Tsang was not asked how many people assaulted him. What Tsang did say was that from the time he was picked up and carried by his arms and legs face down nobody left the group but having seen the video footage someone else joined the group. The first reference by Tsang to seven people was when he gave evidence that after being assaulted he was taken to a car. When asked how many males left the substation to go to the car Tsang replied seven. 353. In cross-examination[260]Tsang agreed that when the First Information of Complaints Against Police Report was completed, he was accompanied by Ms Tanya Chan, a barrister and member of the Civic Party; in a formal complaint to CAPO (exhibit P38), made later the same day, he said he was assaulted by several persons; in his first witness statement made four days later he said he was assaulted by a group of males and did not know how many kicked and punched him[261]; and later when applying for a Judicial Review he said he was attacked by a group of six. 354. Tsang explained that he told DSPC 50117 that he was assaulted by six or seven persons but DSPC 50117 only wrote six persons in the First Information of Complaint Against Police Report and drew six things below a figure in a diagram. Tsang was not sure whether he corrected this but explained that when he pointed anything out, for example his injuries, the officer would not amend the complaint. When put by Mr Cheng SC that he never told anyone he was assaulted by seven persons until the day before he gave evidence, Tsang disagreed and said that he mentioned this before going to the hospital. 355. Considering Tsang was accompanied by Ms Tanya Chan, a barrister and member of the Civic Party, I have my reservations DSPC 50117 did not write down what Tsang said, in particular if Tsang pointed out the error. Notwithstanding these reservations, the fact that the initial complaint records Tsang was assaulted by six persons and not six or seven as he says and later he said he was assaulted by a group of persons does not cause me to doubt his evidence he was assaulted, which assault was captured on the video footage. For the reasons already given I am satisfied that the video footage shows Tsang being assaulted. Considering Tsang was carried face down and the manner in which he was assaulted, it is not at all surprising he was unable to say exactly how many people assaulted him. Change of guard 356. The video evidence is clear that not only was a seventh person seen when the group were walking alongside the substation but the persons holding the arms and the left leg had changed. In cross-examination Tsang agreed because he was carried face down it was possible that some officers changed, but the people holding him, and as far as he could see, the people on his left and right, had never changed[262]. 357. The defence submitted that Tsang’s assertion there was no “change of guard” was incorrect[263]. Mr Cheng SC submitted that this was significant because the prosecution case on identification rested strongly (if not exclusively) on the premise that there was no change of guard[264]. Similar submissions were made by Mr Lam[265]. 358. I have no hesitation in rejecting this submission. I am satisfied that there was only a change in position and not a change in the group. This is discussed later when considering the case against each defendant. The fact that Tsang did not realise that there had been a change in some of the persons holding him does not cause me to doubt his evidence. Blunt instrument 359. On the voir dire Tsang said he sensed he was beaten with a hard object. In cross-examination on the voir dire Tsang was referred by Mr Cheng SC to his first witness statement[266] in which he says he believed the police used a blunt instrument to hit him about his body and head. Asked by Mr Cheng SC if he could see in any video footage a blunt instrument being used Tsang replied no. 360. Mr Cheng SC submitted because the video footage did not show the assailants using a hard object or blunt instrument, this again tended to suggest that the video footage might not have depicted Tsang being assaulted[267]. 361. For the reasons already given, I am satisfied that the video footage shows Tsang being assaulted. Further, in my view, on a careful look at the video footage a person can be seen at the beginning of the assault holding something in his hand and in a downward motion stabbing Tsang twice. This can be seen on all the TVB footage, for example P1 (g) between 03:11 and 03:13 and P1 (h) between 00:25 and 00:27 as shown in screen capture A1 attached herewith[268]. This part of P1 (h) was played in court during the prosecution’s closing submission. 362. A few seconds later between 03:17 and 03:19 on P1(g) and between 00:31 and 00:33 on P1(h) the same person can again be seen stabbing Tsang twice, on a different part of his body, as shown in screen capture A2 attached herewith[269]. 363. These actions are best seen in slow motion. The fact Tsang agreed with Mr Cheng SC the video footage does not show the assailants using a hard object or blunt instrument does not cause me to doubt his evidence that he was assaulted during which he sensed he was hit by a hard object. Pepper spray 364. On the voir dire Tsang said that after being sprayed with pepper spray his vision was not 100% clear but he could still see things and people around him, for example he could see reporters ahead of him taking photographs. When cross examined by Mr Cheng SC, Tsang agreed his vision was affected by the pepper spray. Asked if it could be said his vision was blurred Tsang replied, “For one moment, yes”. 365. Mr Cheng SC submitted that Tsang’s agreement in cross-examination that his vision was blurred was consistent with Tsang having told a news reporter from AM 730, when giving an exclusive interview four days later, that he could not even recognise his own mother. By reason of his blurred vision Mr Cheng SC submitted that Tsang simply could not have had a clear view of the matters which allegedly occurred when he was escorted and carried towards the substation[270]. 366. I have no hesitation in rejecting this submission. I find inherently improbable Tsang would have no memory of being escorted away by police officers, carried by his arms and legs face down and taken to the substation where he was dumped on the ground and assaulted. 367. I accept the evidence of Tsang that his vision was blurred for one moment. In cross-examination by Mr Lok SC, Tsang said he was still conscious after being sprayed with pepper spray and that he could see clearly things within several meters. The video footage clearly shows Tsang was conscious after his face was washed and when he was escorted[271]. 368. Further Tsang did not agree he told the reporter that he could not even recognise his own mother. In re-examination Tsang explained the reference in the newspaper was to a Chinese idiom which meant because his facial features were so much distorted even his own mother could not recognise him. I accept the explanation given by Tsang. The fact that Tsang whilst being subdued was sprayed with pepper spray did not cause me to doubt his evidence about what happened after he was subdued. Independent memory Influenced by the video 369. It is clear from cross-examination that Tsang had viewed the video footage many times before giving evidence[272]. The first time was on the coach to Wong Chuk Hang when he saw the police officers, who escorted him, viewing the footage. Tsang was not however able to see the footage clearly at this time. The first time Tsang had a clear view of the footage was after he left North Point Police Station on the evening of 15 October[273]. 370. Tsang said he gave CAPO the link to a YouTube video, which video he had viewed a few times before giving his statement to CAPO[274]. Tsang said his recollection of the events was entirely his own personal recollection and that the video footage confirmed his recollection. Asked by Mr Lok SC whether he would accept the proposition that his recollection was partly assisted by what he had seen on the YouTube clip, Tsang replied “Right”[275]. 371. On the general issue Mr Chung cross-examined Tsang with regard to media photographs and video footage he had viewed prior to attending the Tsuen Wan Police Station on 27 January 2015 for identification procedures by way of direct confrontation. This evidence is discussed later when considering the admissibility of the direct confrontations[276]. 372. Mr Cheng SC submitted that Tsang was consciously or otherwise affected by his repeated viewing of the video footage whereby he simply assumed he was the person depicted in the video footage[277]. Citing the Australian case of Alexander v The Queen[278] Mr Cheng SC submitted that Tsang’s memory was displaced by what he had seen in the video footage[279]. 373. The discussion of the “displacement” effect in Alexander v The Queen was said in the context of identification of a stranger from photographs. This, in my view, is very different to someone being asked whether a video recording is consistent with their recollection of the events. 374. I accept the evidence of Tsang that the video footage confirmed his recollection of the events. Nevertheless, the fact that Tsang viewed the video footage many times and agreed that his recollection was partly assisted by the footage I approach his evidence with caution. Was Tsang Kin Chiu subdued by one group and assaulted by another group? 375. On the voir dire Tsang gave evidence that after he was pushed to the ground, assaulted and handcuffed with zip wire he was handed over to other police officers, who carried him to the substation where they assaulted him. 376. In cross-examination[280] Tsang was asked about paragraphs 2 & 3 of his witness statement[281] where he refers to being assaulted after being grabbed; assaulted at the substation and assaulted at the Central Police Station and Q16 which asked whether the same group of males assaulted him throughout. 377. Tsang disagreed that in the statement he clearly stated that the group who assaulted him after he was grabbed was the same group who assaulted him at the substation. Tsang went on to explain that his reply “I believe they were” in answer to Q16 was referring to the same group who assaulted him at the substation and in the police station. 378. Ms Lam submitted that Tsang changed his evidence in cross-examination from that in his first witness statement that the same group who grabbed him assaulted him at the substation because he did not have an independent recollection of what had happened[282]. 379. I find this difference more apparent than real. The fact remains Tsang was subdued by police officers; escorted and carried face down by police officers; and assaulted by police officers. Whether or not Tsang perceived at the time they were all the same group or not, is in my view not material. This difference between his witness statement and his evidence in court does not cause me to doubt his evidence as to what happened after he was subdued. 380. I reject the submission that Tsang had no independent memory of what happened. I find inherently improbable that anyone subdued by being handcuffed at the back with zip ties; escorted; carried by their arms and legs face down and taken to the substation where they were dumped on the ground and assaulted would not know what happened unless they were unconscious. As noted when discussing the use of pepper spray the video evidence clearly shows that Tsang was conscious when he was being escorted[283]. Refusal to co-operate with the police investigation 381. The defence submitted the credibility of Tsang is in doubt by his refusal to co-operate with the police. Tsang having refused to answer police questions whether his recollection of the events was based on his own memory or on the video footage, Mr Cheng SC submitted that Tsang’s evidence that his recollection of events was based on his own memory was not credible[284]. 382. After Mr Cheng SC referred Tsang to his affirmation[285], and paragraphs 26 & 27 of the affirmation of Ms Tanya Chan[286], filed in the judicial review proceedings, Tsang agreed that during the giving of his first witness statement[287] he refused on three occasions to answer questions as to whether his recollection was based on his own memory or based on the video footage. 383. Tsang explained in refusing to answer the questions he was acting on legal advice and not because he did not want to bind himself to evidence he might not be able to change. Tsang agreed the reason given by his legal advisers for refusing to answer the question was as stated in his affirmation, namely that they considered the question inappropriate and calculated to prejudice a possible prosecution. 384. Mr Cheng SC submitted that if Tsang did in fact have any personal recollection of what happened there could have been no possible prejudice in answering the question[288]. 385. Tsang also confirmed that on 16 October 2015, together with his legal representatives, he attended the offices of CAPO to view video footage. After being shown twenty videos (including the video footage shown in court) Tsang was asked to identify himself on the videos. Tsang agreed that he told CAPO he was too tired and refused to give a statement. Tsang said this was not the only reason for not giving a statement and went on to explain that the legal advice was to further prepare before giving a statement. 386. Mr Cheng SC submitted that this act of refusal is most questionable and casts serious doubt as to the ability of Tsang to recognize himself in the video footage. The only plausible explanation Mr Cheng SC submitted was that Tsang was unable to identify himself and feared by providing a statement this would prejudice any subsequent prosecution[289]. 387. Similarly Ms Lam submitted the evidence of Tsang should be rejected by reason of his refusal to answer questions and to give a further witness statement after viewing the videos[290]. Ms Lam further submitted Tsang fabricated this evidence because he was convicted in the Magistracy[291]. 388. Mr Lok SC questioned the credibility of Tsang by submitting that he was not co-operative to give assistance to the police in the investigation; he provided only very limited information to the police regarding the assault incident despite having made a complaint on the very day of the assault and only gave assistance after he was found guilty[292]. 389. I have no hesitation in rejecting these submissions. I accept Tsang’s evidence he was acting on legal advice. The legal advice given to refuse to answer questions can clearly be seen in paragraphs 26 & 27 of the affirmation of Ms Chan, the contents of which Tsang confirmed were accurate. 390. Furthermore, having already been arrested by the police it is not at all surprising Tsang declined to answer any further questions. The fact that Tsang did not answer all questions asked by the police in October 2014 and refused to give a further statement in October 2015, does not cause me to doubt his evidence as to what happened after he was subdued. Failure to complain 391. Submissions were made that Tsang did not make a complaint of assault at the first available opportunity and when he did he provided limited information regarding the assault[293]. Mr Choy submitted that the failure by Tsang to make a timely complaint indicates that he did not have an independent memory of his assault[294]. Ms Lam submitted Tsang did not make an early complaint because he was not certain about when, where, who and how the assault had taken place[295]. 392. Admitted in evidence is that after arriving at the Police College in Wong Chuk Hang, Tsang was informed of his rights at 6:12 a.m. by Inspector Wu (“IP Wu”) (PW48)[296]. Tsang made no complaint at this time[297]. On the general issue Tsang said there was no particular reason for not making a complaint at that time. At 7:40 a.m. PC 9765 (PW46) served a Notice to Persons in Police Custody, exhibit D6 (7) on Tsang, who was given the opportunity to read the Notice. 393. In his application for leave to apply for Judicial Review Tsang made no mention of being informed of his rights by IP Wu and stated, inter alia, that he was only informed of his rights at approximately 7:40 a.m.[298] Ms Lam submitted Tsang deliberately lied in not telling the court in the Judicial Review proceedings about IP Wu advising him of his rights[299]. 394. I have no hesitation in rejecting this submission. I do not accept that in the Judicial Review application Tsang tried to hide the fact he had been advised of his rights by IP Wu. The application was emphasising Tsang was only advised of his rights after arriving at the Police College. The omission that he was first told orally by IP Wu is, in my view, not material and does not cause me to doubt Tsang’s evidence about what happened after he was subdued by the uniform police officers. 395. DSgt 51344 (PW45) testified that at 8:16 a.m. Tsang asked to see a doctor saying he had sustained injury to his head, face, hand and arms when he was arrested by the police. In oral submissions Mr Choy submitted this contradicted the evidence of Tsang that he was assaulted at the substation after he was arrested. 396. I have no hesitation in rejecting this submission. The court was never told when and who arrested Tsang. WSSgt So (PW47), the Duty Officer at the Central Police Station, said she was informed Tsang had been arrested when two plainclothes police officers passed her desk[300]. The only other reference to an arresting officer is in the Short Arrest Form, exhibit P56 produced by PC 9765 (PW46), which states that D6 is the arresting officer. 397. In my view there is no contradiction. The process of arrest clearly covers the time from when Tsang was first apprehended until he was brought to the police station and later to the Police College. 398. Before going to see a doctor Tsang told DSgt 51344 that he wanted to see his lawyer first. At about 8:35 a.m. Tsang was allowed to consult with his lawyer Ms Tanya Chan. Admitted in evidence is that at about 10 a.m. on 15 October 2014 Tsang stated he wanted to lodge a complaint against the police[301]. The First Information of Complaints Against Police Report (Pol. 964), exhibit P26, was then filled out by DSPC 50117 (PW49)[302]. 399. DSPC 50117 did not fill out the summary of allegations because Tsang, on the advice of Ms Chan, provided no information. Tsang, also on the advice of Ms Chan, was not willing to let the police conduct a physical check or take photographs of his injuries. DSPC 50117 therefore recorded the injuries according to a verbal description given by Tsang[303]. 400. On the general issue Tsang testified that from what he saw the police officer did not record what he told him were his injuries. Tsang therefore refused to allow the police to inspect his injuries. Tsang said he was told by the officer that he was only making a simple record of what he saw and then he would go to the hospital for examination by a doctor[304]. Tsang did at one stage pull his shirt up to let the officer look at his injuries. 401. Tsang said he did not provide information of the complaint; consent to a physical examination of his injuries or allow the police to photograph his injuries because he thought it was more appropriate to go to the hospital for examination first and together with Ms Chan they were of the view the police officer had not accurately recorded the injuries, writing only the slightest form of “redness” or 1 cm. In cross-examination Tsang disagreed that the officer recorded what he said about his injuries and disagreed that he signed the First Information of Complaints Against Police Report because he agreed with what was recorded[305]. 402. Admitted in evidence is that at approximately 11:05 a.m. Tsang was examined by Dr Zenith Wu at the Ruttonjee and Tang Shiu Kin Hospital, who also took photographs of the injuries[306]. After medical examination Tsang was taken to North Point Police Station. At the police station DSPC 50117 again asked for details to fill out the summary of allegations. Ms Chan said that Tsang would not give any statement at that time. In evidence Tsang said he needed to rest. 403. Inspector Law (“IP Law”) (PW50) from CAPO was instructed to go to North Point Police station to interview Tsang[307]. IP Law was told by Mr Vidler, solicitor for Tsang, Tsang would only speak with CAPO after he was granted bail. At 23:13 after Tsang was granted bail IP Law interviewed Tsang. IP Law explained the procedure to Tsang. Tsang produced a piece of paper exhibit P38, listing three complaints of assault by police officers. Tsang refused to elaborate further. 404. In cross-examination, on the general issue, Tsang agreed he did not make a complaint to the driver of the car taking him to the Central Police Station; to the Duty Officer at the Central Police Station; or to the driver of the coach to the Police College[308]. Tsang disagreed this was because he did not have an independent recollection of the events. Asked by Mr Choy if the reason he did not complain to the Duty Officer was because he thought the Duty Officer was one of the scumbag police he had described in his evidence, Tsang replied he did not think the drivers and the Duty Officer were reliable or appropriate recipients of the complaint. 405. Tsang also agreed he did not complain to fellow detainees at the Police College. Tsang explained this was because while in detention they were not allowed to talk to each other and he did not know the other detainees. Tsang disagreed he was inventing this excuse to explain away why he did not complain even to his comrade-in-arms. 406. I have no hesitation in rejecting these submissions. I accept the explanation of Tsang that no earlier complaint was made because he considered drivers and the Duty Officer not reliable or appropriate persons to receive his complaint. Further at all these times Tsang was accompanied by two of the police officers he says assaulted him. It is therefore not at all surprising Tsang waited until after he saw his lawyers before making a complaint. I accept Tsang’s evidence why he did not tell fellow detainees about the assault. The failure to complain earlier and provide details of the complaint does not cause me to doubt the evidence of Tsang about what happened after he was subdued. 407. Considering Tsang was accompanied by Ms Chan, a barrister and member of the Civic Party, I do have my reservations about his evidence that DSPC 50117 did not write down what he said about his injuries. The First Information of Complaints Against Police Report, whilst brief, does record that there were injuries to the head, the face, the left arm and left hand, the neck, the chest and the entire back. 408. The description of the injuries in the First Information of Complaints Against Police Report is only a brief description. Whether or not that is what Tsang said to DSPC 50117 is in my view not material. The court relies on the admitted evidence of Dr Zenith Wu, who medically examined Tsang at about 11:05 that morning and found the injuries as particularised in the Medical Examination Form, exhibit P27(a); the medical report, exhibit P27(b) and shown in the photographs taken by Dr Wu, exhibit P28[309]. The injuries were noted by Dr Wu as fresh[310]. 409. The reservations I have in Tsang’s evidence that DSPC 50117 did not write down what he said about his injuries do not cause me to doubt Tsang’s evidence that after he was subdued he was carried to the substation where he was assaulted, which assault was captured on the video footage. Judicial review/political aspirations 410. In cross-examination on the voir dire Mr Lok SC referred Tsang to an application made by him on 22 October 2014 for leave to apply for a judicial review[311] of the conduct of the police on 15 October 2014[312]. Tsang disagreed that part of the reason for making the application for judicial review was to go for money. 411. Tsang explained that he did not know the law well and therefore did not know whether a claim for damages needed to be included in the application. Tsang further explained that the reason for making an urgent application was in the hope the police would disclose the identity of the police officers involved so that they may be interdicted and prosecuted. 412. Mr Lok SC questioned the credibility of Tsang submitting that his motive was open to many interpretations, including that in asking for exemplary damages he was after money and by running for the 2016 Legislative Council Election in September[313] he was seeking to advance his political career[314]. 413. Mr Choy also cross-examined Tsang on his political aspirations. Tsang agreed that by considering standing in the 2016 Legislative Council Election he was an aspiring politician[315]. After playing the ATV footage exhibit P3 (2), which clearly shows Tsang being subdued by police officers, Mr Choy put to Tsang that his evidence he guessed he was subdued by the police because only police use force in demonstrations, was political posturing on his part. Tsang disagreed with the suggestion made by Mr Choy. 414. I accept the evidence of Tsang that the reason for making an urgent application was in the hope the police would disclose the identity of the police officers involved so that they may be interdicted and prosecuted. The fact Tsang applied for leave to apply for a judicial review of the conduct of the police on 15 October 2014, including seeking a claim for damages, does not cause me to doubt his evidence as to what happened after he was subdued. 415. I accept the evidence of Tsang that his evidence was not political posturing. The fact Tsang had political aspirations does not cause me to doubt his evidence as to what happened after he was subdued. Hatred of the police 416. Mr Choy submitted that Tsang’s dislike of the police (possibly as a result of his political stance) greatly impaired his reliability. Mr Choy specifically relied on Tsang calling the police scum[316]. 417. The reference to the police as scum was in cross-examination by Mr Cheng SC on the general issue. Tsang agreed that in postings on Facebook he called the police scum. Tsang explained that this was a conclusion reached after reading many news stories for example where a teenage girl was raped by a police officer in a police station and where a peaceful protestor was hit causing detachment of his retina. 418. Tsang said this was not a view of the police in general but on their individual behaviour and asked Mr Cheng SC to supply the context in which he said the police were scum. Mr Cheng SC did not and instead asked Tsang if he regarded the police officers who subdued him at the flower bed as scum. Tsang replied he was not sure whether they were scum. 419. I have no hesitation in rejecting this submission. Whilst Tsang clearly had a high disregard of the police I accept his explanation that his reference to scum was to individual police officers and not the police force in general. 420. Tsang also agreed with Mr Cheng SC that he posted the following “I’ll lengthen my eyes to see how much longer you people will live a merry life.” Ms Lam submitted this also showed the spite held by Tsang against the police and that Tsang had fabricated his evidence because he had been convicted in the Kowloon City Magistracy[317].I have no hesitation in rejecting this submission. I accept Tsang’s explanation that by posting this he meant he was waiting for justice without specifying which police officers he was referring to. 421. The postings by Tsang in Facebook, whilst giving me cause to approach his evidence with caution, do not cause me to doubt his evidence of what happened after he was subdued. Conclusion 422. In considering Tsang’s evidence I have also considered that in his first witness statement he did not mention the group discussed where to take him; asked him to walk faster; when being assaulted someone mentioned “demonstration” and refused to give details of the building where he says he was assaulted[318]. I do not regard these omissions as material. They do not cause me to doubt Tsang’s evidence about what happened to him after he was subdued. 423. Having carefully considered all the evidence and submissions of counsel, unless otherwise stated, I find I am satisfied so I am sure that I can safely rely on the evidence of Tsang that after he was subdued for pouring liquid on the police he was handed over to other police officers who escorted him, carried him by his arms and legs face down and took him to the substation (the dark corner) where they dumped him on the ground and assaulted him. Identification 424. The prosecution case depending on identification evidence I direct myself in accordance with the principles established in R v Turnbull[319]. I remind myself of the special need for caution before convicting the defendants in reliance on evidence of identification. 425. I remind myself to look at the quality of the identification and all the circumstances in which the identification was made and in so doing look for any specific weaknesses, which have appeared in the identification evidence. I remind myself of the possibility that a mistaken witness can be a convincing witness and that a number of witnesses can all be mistaken. I remind myself that mistakes in recognition of close relatives and friends are sometimes made. 426. When the quality of the identifying evidence is poor unless there is evidence which goes to support the correctness of the identification the accused shall be acquitted. Odd coincidences can, if unexplained, be supporting evidence[320]. The defendants having elected to remain silent there is no evidence from the defence to contradict the prosecution evidence. As was said in Turnbull[321]: “An accused’s absence from the witness box cannot provide evidence of anything and the judge should tell the jury so. But he would be entitled to tell them that when assessing the quality of the identification evidence they could take into consideration the fact that it was uncontradicted by any evidence coming from the accused himself.” 427. The guidelines given in Turnbull equally apply to identification made from film and photographs, whether that identification is made by a witness or by the court[322]. 428. In Attorney General’s Reference (No 2 of 2002)[323] the court said on the authorities there was at least four circumstances in which, subject to appropriate directions, a jury can be invited to conclude that the defendant committed the offence on the basis of a photographic image from the scene of the crime, including: (i) Where the photographic image is sufficiently clear, the jury can compare it with the defendant sitting in the dock; (ii) Where a witness knows the defendant sufficiently well to recognise him as the offender depicted in the photographic image, he can give evidence of this; and (iii) Where a witness who does not know the defendant spends substantial time viewing and analysing photographic images from the scene, thereby acquiring special knowledge which the jury does not have, he can give evidence of identification based on comparison between those images and a reasonably contemporary photograph of the defendant, provided that the images and photographs are available to the jury. 429. The fourth circumstance was a suitably qualified expert with facial mapping skills. No expert evidence has been called. 430. Whilst the identification of each defendant is to be considered separately, the evidence against each defendant is not to be considered in isolation but in the context that the defendants were all on duty in the area of Lung Wo Road[324]and carried out their duties together. 431. Superintendent Ng (“SP Ng”) (PW27), who was in charge of Crime Group A which included eight Quick Response Teams, instructed D1 to lead one Quick Response Team. D2 was the officer-in-charge of QRT A2-2[325]. Including D2 there were seven team members: D3, D5, D6, D7, DPC 8097 (PW28) and DPC 5840 (PW29)[326]. D4 was deployed as a member of a video team under the command of DSgt Ma (PW30) [327]. D2 redeployed D4 to other duties after which DSgt Ma did not see D4 again.[328] 432. Before the clearance operation started SP Ng briefed the members of the Quick Response Teams of their duties and responsibilities, including the processing of persons who had committed offences such as unlawful assembly or obstructing police in the execution of their duty. The plan devised before the operation started was that as far as possible protestors who were arrested would be handed over to crime officers for processing. 433. The crime officers would then take the arrested persons to the escort coaches or cars for transporting to the Central Police Station. The evidence of the drivers of the coaches and the cars was read and show that the coaches and the cars were parked on Lung Wo Road near to the substation[329]. 434. SP Ng saw a uniform police officer trying to stop Tsang pouring water. The uniform police officer pulled Tsang back towards Tamar Park. From where SP Ng was standing he could not see what happened after Tsang was pulled back. SP Ng said that according to the operational plan Tsang should have been arrested and then taken by a crime officer to one of the coaches or cars for transport to the Central Police Station. 435. In cross-examination SP Ng said that the actual decision made as to what to do would be made according to the situation at that time[330]. 436. Before considering the case against each defendant I will first outline the identification evidence relied on by the prosecution. Senior police officers, under whose command D1, D2, D3, D5 and D6 worked[331], looked at part of the video footage and the photographs[332] and identified the defendants on some of the video footage and photographs. Save for the evidence of Senior Superintendent Chan (“SSP Chan”) (PW37), who recognised D1, the witness statements of the senior police officers were read. 437. In respect of all the defendants the prosecution relied on a comparison made by Chief Inspector Chan (“CIP Chan”) (PW42) of the persons seen in the TVB footage P1 (b) and the Apple Daily footage P2 (e) with the warrant card photographs of the defendants. The witness statements of CIP Chan were read into evidence[333]. 438. The recognition of D1, D2, D3, D5 and D6 was from footage and photographs of the apprehension and escorting of Tsang and not from footage of the assault at the substation. The comparison by CIP Chan in respect of D4 was also made from footage showing the escorting of Tsang. Only the comparison by CIP Chan in respect of D7 was made from footage showing the assault at the substation. 439. The prosecution case is that there was no change in the persons escorting Tsang save that there was a change in position and at the substation they were joined by another member of their team, D7. The defence submitted the prosecution failed to prove that the persons seen escorting Tsang were the same persons seen assaulting him at the substation and failed to prove that the seventh person was D7. 440. The prosecution also relied on the evidence of Tsang that D5 and D6 were two of the persons who assaulted him at the substation. The identification of D5 and D6 by Tsang was by way of direct confrontations, the admissibility of which was challenged by both D5 and D6. The prosecution also relied on the CCTV recordings to show that D5 and D6 took Tsang in and out of the Central Police Station. Tsang did not identify D1, D2, D3, D4 or D7. 441. In respect of charge 2 the prosecution relied on the evidence of Tsang that he was slapped on the face by D5 inside room 7 of the Central Police Station. 442. In Annex AI-AVII of their closing submissions the prosecution particularised all the footage and the times and the photographs in which they say each of the defendants are seen. To illustrate submissions which made reference to the video footage, the court requested a screen capture be provided. At the beginning of the second day of submissions the prosecution provided a bundle of 23 screen captures. 443. In reaching my verdict I have carefully considered the video footage, the photographs and the CCTV recordings. The quality and clarity of the video footage varies. I have viewed all the video footage many times in normal speed, slow motion and frame by frame. I have carefully considered the photographs which are of good quality. I have viewed the CCTV recordings many times, in both normal speed and frame by frame[334]. D1 444. SSP Chan testified that between June 2014 and August 2015 he was the superintendent of ‘A’ Division of the Organised Crime and Triad Bureau. D1 was under SSP Chan’s immediate command from June 2014 until his interdiction in October 2014[335]. After D1’s interdiction SSP Chan interviewed D1 on a monthly basis until August 2015 when SSP Chan left his post. 445. On 11 March 2016 SSP Chan was invited by Woman Chief Inspector Hung (“WCIP Hung”) (PW41)[336]to view the TVB footage P1 (a); part of P1 (g) and two parts of the ATV footage P3 (2), to see if he could recognise anyone he knew[337]. SSP Chan recognised D1 on the ATV footage P3 (2) and on the TVB footage P1 (g). SSP Chan marked which person was D1 on screen captures of the shots where he recognised D1[338]. After viewing the TVB footage P1 (a) twice, SSP Chan was unable to recognise anyone on the footage. 446. On 26 May 2016 SSP Chan was invited by WCIP Hung to view two photographs, exhibits P12 and P13. SSP Chan recognised D1 on P13 but was unable to recognise anyone on P12. On a copy of P13 SSP Chan marked which person was D1[339]. 447. SSP Chan also identified D1 in court. 448. I accept the identification evidence of SSP Chan, which evidence was not challenged. 449. The evidence of CIP Chan that he compared the warrant card photograph of D1 with the TVB footage of the escorting is, in my view,no more than the opinion of CIP Chan. There is no evidence that CIP Chan knew D1 or that he had acquired any special knowledge. CIP Chan was therefore in no better position than the court to make such comparison. I place no reliance on the comparison made by CIP Chan. In any event this evidence adds nothing to the identification of D1 by SSP Chan. 450. The identification of D1 by SSP Chan from the ATV footage shows the stage where Tsang was handed over. Although SSP Chan only identified D1 in the second part of the ATV footage, I am satisfied D1 can be seen in the first part of the ATV footage when Tsang was handed over as shown in screen capture 3 attached to the prosecution’s closing submissions. 451. The identification of D1 by SSP Chan from the TVB footage shows the stage where Tsang was escorted. I am satisfied in the Now TV footage (exhibit P4) between 00:40-00:44, which was not shown to SSP Chan, D1 is also seen escorting Tsang as shown in screen capture 14 attached to the prosecution’s closing submissions. I am satisfied in the ATV footage (exhibit P3 (2)) between 01:40 and 01:50 D1 can also be seen at the back escorting Tsang. 452. The identification of D1 by SSP Chan from the photograph P13 shows D1 on the right of Tsang at the stage where Tsang was being carried face down. I am satisfied D1 can also be seen on the right of Tsang in the moving photograph in the Apple Daily footage (exhibit P2(c)), which, for the reasons already given, I am satisfied was taken at the stage Tsang was carried face down[340]. D1 can also be seen in the first nine seconds of the TVB footage (exhibit P1 (a))walking from the right side to the left side of Tsang[341]. 453. From the video footage and in particular the photograph, exhibit P13, the clothing of D1 is clearly seen. D1 was wearing a white shirt with the sleeves rolled up. The shirt was open neck and not tucked in. D1 was also wearing a black OCTB sleeveless jacket; a watch on his left wrist; blue trousers and brown shoes. 454. There has been no identification of D1 from the footage showing the assault at the substation. Mr Lok SC submitted that there is no evidence to connect D1 in the still photographs as one of the group shown in the video footage. In particular Mr Lok SC refers to the evidence of SSP Chan that he could not identify anyone because it was too dark[342]. 455. I have no hesitation in rejecting this submission. This evidence was given in cross-examination when Mr Lok SC asked SSP Chan if the reason why he could not identify anyone on the photograph, exhibit P12, was because it was too dark. This answer must be viewed in the context that the photograph shows the stage where Tsang was escorted. SSP Chan did identify D1 from the TVB footage showing Tsang being escorted. The photograph was almost certainly taken about the same time as the footage. The face of the person on the photograph in the same position as D1 is in the dark. 456. In considering the evidence, I take into account SSP Chan, who knew D1, was unable to identify D1 from the assault footage shown to him. A distinction however is to be drawn between being able to recognise a person and comparing whether the clothing is the same. 457. The TVB footage shows the group arriving at the substation[343]. By following the person carrying the left leg of Tsang, the back of this person is seen as the group turn left at the substation and drop Tsang on the ground. Screen capture ID1 taken from P1 (h) is attached herewith showing the person carrying the left leg of Tsang. This is best seen in slow motion or frame by frame. 458. When this person turns round, his clothing can clearly be seen, in particular a white shirt with the sleeves rolled up. The shirt is the same as seen in the photograph exhibit P13, with an open neck and not tucked in. A watch on the left wrist can also be seen. The colour of the trousers cannot be determined with certainty however the thighs show a faded look on both legs which is consistent with that seen in the photograph exhibit P13. Screen capture ID2 taken from P1 (h) is attached herewith showing the clothing of this person and highlighting the tail of the shirt and the faded look on both thighs. The video needs to be viewed to see the watch. 459. Having carefully viewed all the video footage, I am satisfied so I am sure that when the group arrived at the substation D1 was the person carrying the left leg of Tsang. D1 is seen throughout the assault footage keeping watch and occasionally going up to look at Tsang. D1 was the last person to leave when Tsang was taken away. 460. I have carefully considered the submission of Mr Lok SC that there is a possibility that the persons, including the person being carried, might have changed because there was a change in the number of persons carrying a person to the substation; there were more than two hundred plainclothes officers involved in the clearance operation; and at least forty two arrested persons were searched or processed near to the substation[344]. These submissions do not cause me to doubt the findings I have made. D2 461. Between June 2013 and June 2015 Chief Inspector Chung (PW38) was the Chief Inspector of Crime, Kwun Tong District. D2 was under CIP Chung’s command from the beginning of 2014 until his interdiction in October 2014[345]. After D2’s interdiction CIP Chung would meet D2 when D2 returned to the police station to be interviewed on a monthly basis until June 2015 when CIP Chung left his post[346]. 462. On 14 March 2016 CIP Chung was invited by WCIP Hung to view the TVB footage P1 (a); part of P1 (g) and two parts of the ATV footage P3 (2), to see if he could recognise anyone he knew[347]. CIP Chung recognised D2 on the TVB footage P1 (g) and marked which person was D2 on a screen capture of the shot[348]. After viewing the TVB footage P1 (a) twice CIP Chung was unable to recognise anyone because the video quality was poor and it was too dark. After viewing the ATV footage P3 (2) CIP Chung was also unable to recognise anyone. 463. On 23 May 2016 CIP Chung was invited by WCIP Hung to view the ATV footage P3 (1); the TVB footage P1 (h) and the Now TV footage P4[349]. CIP Chung recognised D2 on the TVB footage P1 (h) which he said was essentially the same as the TVB footage P1 (g) played to him on 14 March. CIP Chung did not recognise D2 on either the ATV or Now TV footage[350]. 464. On 31 May 2016 CIP Chung was invited by WCIP Hung to view the two photographs, exhibits P12 and P13. CIP Chung was unable to recognise D2 on either photograph[351]. 465. In his closing submission Mr Cheng SC stated there was no dispute about the accuracy of CIP Chung’s identification of D2[352]. I accept the identification evidence of CIP Chung. For the reasons given in the case of D1 no reliance is placed on the comparison made by CIP Chan of the warrant card photograph of D2 with the TVB footage. In any event this evidence adds nothing to the identification of D2 by CIP Chung. 466. The identification of D2 by CIP Chung from the TVB footage shows the stage where Tsang was being escorted. Notwithstanding CIP Chung did not identify anyone on photograph exhibit P12 and from the Now TV footage, both of which were taken during the stage Tsang was being escorted, I am satisfied D2 is also seen in the photograph P12 on the far right of Tsang and in the Now TV footage between 00:40-00:44 as shown in screen capture 10 attached to the prosecution’s closing submissions. 467. From the video footage and the photograph, exhibit P12, the clothing of D2 is clearly seen. D2 was wearing a blue T-shirt. The blue colour is quite distinct, which I would describe as azure blue or royal azure blue. D2 was also wearing a black sleeveless jacket with an identification card on the left side of the jacket, most likely his warrant card. The TVB footage shows D2 using a white earpiece as seen in screen capture 7 attached to the prosecution’s closing submissions. The photograph exhibit P12 shows D2 wearing gloves, which appear to be rubber gloves extending beyond D2’s wrist. 468. Notwithstanding CIP Chung, who knew D2, did not identify D2 on the ATV footage, I am nevertheless satisfied that the prosecution are correct in saying the first time D2 is seen is in the ATV footage, exhibit P3 (2) at 01:38 which shows the stage where Tsang was being handed over. This is very easy to miss. The blue T-shirt is clearly visible. The position of D2 is shown in screen capture 6 attached to the prosecution’s closing submissions. 469. The moving photograph in the Apple Daily footage, exhibit P2(c), which I am satisfied was taken at the stage Tsang was carried face down[353], was not shown to CIP Chung. I am satisfied D2 is the person seen on the left of Tsang. The blue T-shirt, the white earpiece and the warrant card are clearly seen. Screen capture ID3 is attached herewith showing the position of D2. 470. Notwithstanding CIP Chung, who knew D2, did not identify D2 from the photograph exhibit P13 showing Tsang being carried face down, I am nevertheless satisfied the prosecution are also correct that D2 is the person in the photograph on the far left of Tsang. Although very little of the T-shirt can be seen the rubber gloves can be seen. D2 is in the same position as in the moving photograph in the Apple Daily footage, exhibit P2(c), as are all the other persons. I am satisfied the facial appearance is the same as in the TVB footage. 471. I am satisfied D2 can also be seen in the first nine seconds of the TVB footage exhibit P1 (a) showing Tsang being carried face down[354]. Although only the back of D2 can be seen, D2 is on the left of Tsang the same position as shown in the photograph exhibit P13 and the moving photograph. The blue T-shirt and the gloves are clearly visible. Screen capture ID4 taken from P1 (a) is attached herewith showing the position of D2. 472. There has been no identification of D2 from the footage showing the assault at the substation. In considering the evidence, I take into account CIP Chung, who knew D2, was unable to identify D2 from the assault footage shown to him. Again a distinction is to be drawn between being able to recognise a person and comparing whether the clothing is the same. 473. The TVB footage shows the group arriving at the substation[355]. By following the person carrying the left arm of Tsang, after the group turn left at the substation and drop Tsang on the ground this person turns round and walks out to the side of the substation next to D1. 474. At this stage the clothing can clearly be seen. I disagree with Mr Cheng SC that the clothing cannot be clearly seen because of the poor lighting conditions[356]. The blue T-shirt, the gloves and the black sleeveless jacket with the warrant card on the left side can be seen as shown in screen capture 19 attached to the prosecution’s closing submissions. 475. Having carefully viewed all the video footage I am satisfied so I am sure that when the group arrived at the substation D2 was the person carrying the left arm of Tsang. 476. D2 is seen at various times throughout the assault footage mostly keeping watch and sometimes looking at Tsang, save for a period when he leaves the substation for about one minute together with one other person[357]. At the time D2 leaves he appears to be talking on the earpiece and at one stage puts the earpiece in his right ear. This is best seen between 00:25-00:42 on the Apple Daily footage P2 (d)[358].After returning D2 remains at the substation until Tsang is taken away at which time he is seen walking behind Tsang. 477. I have carefully considered the submission of Mr Cheng SC as to the quality and/or sufficiency of the identification evidence including that D2’s superior CIP Chung could not identify D2 from the assault footage; that there is no basis for the court to draw the inference D2 was amongst the individuals depicted in the assault footage; no mention was made of D2’s involvement in the short arrest form or Tsang’s complaint; and the court should be slow to draw its own conclusions from the assault footage[359]. These submissions do not cause me to doubt the findings I have made. D3 478. CIP Chung also recognised D3. D3 was under CIP Chung’s command at the beginning of 2013 and June 2013 and again from the beginning of 2014 until his interdiction in October 2014[360]. After D3’s interdiction CIP Chung would meet D3 when D3 returned to the police station to be interviewed on a monthly basis until June 2015 when CIP Chung left his post[361]. 479. On 14 March 2016 CIP Chung recognised D3 on the TVB footage P1 (g) and marked which person was D3 on a screen capture of the shot[362]. After viewing the TVB footage P1 (a) and the ATV footage P3 (2) twice CIP Chung was unable to recognise D3. 480. On 23 May 2016 CIP Chung recognised D3 on the ATV footage P3 (1) and marked which person was D3 on a screen capture of the shot[363]. CIP Chung also recognised D3 on the TVB footage P1 (h) which he said was essentially the same as the TVB footage P1 (g) played to him on 14 March. CIP Chung did not recognise D3 on the Now TV footage[364]. 481. On 31 May 2016 CIP Chung recognised D3 on the photograph exhibit P13 and circled D3 on a copy of the photograph[365]. 482. I accept the identification evidence of CIP Chung. For the reasons given when discussing the case of D1, no reliance is placed on the comparison made by CIP Chan of the warrant card photograph of D3 with the TVB footage. In any event this evidence adds nothing to the identification of D3 by CIP Chung. 483. The identification of D3 by CIP Chung from the ATV footage shows the stage where Tsang was handed over and the TVB footage shows the stage where Tsang was escorted. Notwithstanding CIP Chung, who knew D3, did not identify anyone from the Now TV footage I am nevertheless satisfied D3 can also be seen escorting Tsang at the end of the footage between 00:40-00:44 as shown in screen capture 13 attached to the prosecution’s closing submissions. 484. Notwithstanding CIP Chung, who knew D3, did not identify D3 in the ATV footage, exhibit P3 (2), I am nevertheless satisfied between 01:40 and 01:50 D3 is seen placing his right hand on the left shoulder of Tsang as seen in screen capture ID5 attached herewith. 485. The identification of D3 by CIP Chung from the photograph exhibit P13 shows D3 holding the right arm of Tsang at the stage when Tsang was being carried face down. I am satisfied D3 can also be seen holding the right arm of Tsang in the moving photograph in the Apple Daily footage (exhibit P2(c)), which I am satisfied was taken at the stage Tsang was carried face down[366]. Screen capture ID3 is attached herewith showing the position of D3. 486. From the video footage and the photograph, exhibit P13, the clothing of D3 is clearly seen. D3 was wearing a grey T-shirt with rows of pink horizontal stripes; a black sleeveless jacket; light blue jeans and brown shoes. D3 was also wearing gloves, which also appear to be rubber gloves which extended beyond D3’s wrist. 487. There has been no identification of D3 from the footage showing the assault at the substation. In considering the evidence, I take into account CIP Chung, who knew D3, was unable to identify D3 from the assault footage shown to him, which he said was of poor quality and too dark. Again a distinction is to be drawn between being able to recognise a person and comparing whether the clothing is the same. 488. The left shoulder of D3’s T-shirt can be seen on a careful look at the TVB footage exhibit P1 (g) between 02:55 and 03:00 [367] showing Tsang being carried face down. D3 was in the same position as in the photograph exhibit P13 and the moving photograph in the Apple Daily footage (exhibit P2(c)). 489. Having carefully viewed all the video footage I am satisfied so I am sure the T-shirt of D3 can be seen in the TVB and Apple Daily footage. This is best seen in the Apple Daily footage exhibit P2 (a) between 00:16-00:22 & 01:35-01:41 when D3 is standing in front of the substation. In addition, D3 can be seen wearing gloves in particular when walking in the few seconds before and after the footage 00:16-00:22. 490. By looking at the TVB footage of the full assault, for example exhibit P1 (a) between 00:09 and 01:00, together with P1 (g)[368] and by following the person leading the group, who was the shortest of the group, this person is seen bending over Tsang just after Tsang is dropped on the ground. This is followed by active participation in the assault on Tsang by stabbing Tsang[369]; stamping on Tsang and kicking Tsang. After assaulting Tsang this person walked out and stood in front of the substation in the position seen in the Apple Daily footage. 491. I am satisfied so I am sure that when the group arrived at the substation D3 was the person leading the group. D3 is seen at various times throughout the assault footage, save for a period when he leaves the substation for about one minute together with D2. After D2 and D3 return D3 remains at the substation until Tsang is taken away at which time D3 is seen walking behind Tsang. 492. I have carefully considered the submission of Mr Lam that there was no identification of D3 from the first shot or the second shot of the TVB footage; CIP Chung who had known D3 since the beginning of 2013 was unable to identify D3 from this footage, the quality of which he found poor and too dark; and reliance on clothing as a means of identification is unsafe[370]. These submissions do not cause me to doubt the findings I have made. D4 493. D4 was deployed as a member of a video team under the command of DSgt Ma (PW30) [371]. Sometime in the middle of the period 02:40 and 03:35 D2 redeployed D4 to other duties after which DSgt Ma did not see D4 again.[372] 494. The only identification of D4 is by CIP Chan who compares the appearance of the persons seen in the video footage from TVB and Apple Daily with the warrant card photographs of the defendants. CIP Chan concluded that the facial appearance of D4 in the warrant card resembled the facial appearance of one of the persons seen in the TVB footage escorting Tsang[373]. 495. No police officer recognises D4 on any of the video footage or from the photographs. Mr Choy submitted that the evidence of CIP Chan could not substitute proper identification evidence[374]. For the same reasons given when discussing the case of D1, that CIP Chan did not know the defendants and had not acquired any special knowledge, I am satisfied CIP Chan was in no better position than the court to make such a comparison. The evidence in my view amounts to no more than the opinion of CIP Chan. 496. I place no reliance on the comparison made by CIP Chan. This however does not preclude the court from examining all the video footage; the photographs and the warrant card photograph in determining whether D4 was one of the persons who escorted, carried and assaulted Tsang. 497. The prosecution case is that D4 is first seen standing behind Tsang in the ATV footage exhibit P3 (2) at 00:13 as shown in screen capture 2 attached to the prosecution’s closing submissions. This is also captured in the ATV footage between 01:33 and 01:40[375]. The prosecution submitted that the back of D4 is also seen in the same footage between 01:44 and 01:50 at the time Tsang was taken away[376]. I am satisfied not only the back but the left side of the face of the person can also be seen as shown in screen capture ID6 attached herewith. 498. The prosecution submitted that D4 is next seen escorting Tsang in the TVB footage exhibit P1 (b) between 00:05 and 00:13 as shown in screen capture 9 attached to the prosecution’s closing submissions[377]; on the immediate right of Tsang in the photograph, exhibit P12 and in the Now TV footage (exhibit P4) between 00:40 and 00:44 as shown in screen capture 10 attached to the prosecution’s closing submissions[378]. 499. At the stage when Tsang is being carried face down the prosecution submitted that a partial view of D4 is seen in the photograph, exhibit P13. The prosecution described D4 as being on the right hand side of Tsang near his lower body[379]. In my view this person was carrying the right leg of Tsang. This is also shown in the moving photograph in the Apple Daily footage, exhibit P2(c), which I am satisfied was taken at the stage Tsang was carried face down[380]. Screen capture ID3 is attached herewith showing the position of D4 on the moving photograph. 500. In considering the evidence, I take into account there has been no identification of D4 from any of the footage, including the footage showing the assault at the substation. The video footage and photographs clearly show the face of the person the prosecution say is D4. Having very carefully looked at the photographs and the video footage and had the opportunity to observe D4 during the trial, which lasted 31 days, I am satisfied so I am sure this person is D4. 501. In making this finding I have taken into account that the warrant card was taken on 11 April 2013 just over 18 months prior to 15 October 2014; the hair was longer on 15 October 2014; over two years have now passed; and that in court D4’s hair was also longer and he wore glasses. None of these matters cause me to doubt that the person seen in the video footage and the photographs was D4. 502. From the video footage and the photographs, the clothing of D4 is clearly seen. D4 was wearing a grey T-shirt; a black sleeveless jacket which was zipped up; light coloured trousers and blue and red shoes. In the Now TV footage D4 can also be seen wearing gloves. 503. The trousers are best seen in the photograph exhibit P12 and the ATV and Now TV footage. Screen captures ID7 & ID8 taken from the ATV and Now TV footage are attached herewith showing the colour of the trousers. 504. I acknowledge, depending on the lighting, the light colour of the trousers differs between the photographs and the footage. However, what is clear is D4’s trousers are significantly lighter than the trousers of all the others. This is best seen in the photograph exhibit P13. Whilst only part of D4’s trousers can be seen, the right leg is sufficiently visible to show this difference. 505. As already described D4 is seen in both the photograph exhibit P13 and in the moving photograph in the Apple Daily footage, exhibit P2(c) in the same position, namely carrying the right leg of Tsang. I am satisfied the first nine seconds of the TVB footage exhibit P1 (a) also shows D4 holding the right leg of Tsang. The footage shows only the back of D4, however, the light coloured trousers are clearly seen in the footage[381]. 506. Having carefully viewed all the video footage, I am satisfied so I am sure when the group arrived at the substation D4 was still carrying the right leg of Tsang. This is seen for example in the TVB footage P1 (g) between 03:00-03:08[382]. Screen capture 16 attached to the prosecution’s closing submissions shows the light coloured trousers of D4 when the group were walking alongside the substation. The remaining clothing is also consistent with what D4 is seen wearing in the earlier footage and the photographs. 507. In addition, I find that it is an odd coincidence that D4 who was carrying the right leg of Tsang as seen in photograph P13, the moving photograph and the first nine seconds of the TVB footage exhibit P1 (a), was wearing light coloured trousers and the person seen at the substation carrying the right leg of Tsang was also wearing light coloured trousers. 508. As the group turn the corner and just before they drop Tsang on the ground, D4 is seen holding the right leg of Tsang with the light coloured trousers clearly visible as shown in screen capture ID9 attached herewith, taken from P1 (h). D4 then walks round the back of his colleagues and stands next to Tsang. Between 27 and 29 seconds D4 appears to kick Tsang at least once, although his foot cannot be seen. 509. After D2 and D3 return D4 is seen standing next to D1 and D2. This is best seen in the Apple Daily footage exhibit P2 (d) between 02:40 and 3:00. D4’s light coloured trousers can be seen in this footage. When the group take Tsang away D4 is seen on the right of Tsang. D4’s light coloured trousers can also be seen as shown in screen capture ID10 attached herewith, taken from P1 (g)[383]. 510. I have carefully considered the submissions of Mr Choy that no one has identified D4 including Tsang; and that even if the court was to find D4 was one of the persons escorting Tsang this only shows D4 may have participated in the escorting of Tsang but not the assault as he may have been redeployed to other duties during the forty-nine second blocked shot[384]. These submissions do not cause me to doubt the findings I have made. D5 & D6 511. Tsang testified that after he was assaulted he was taken to the Central Police Station by two of the police officers who assaulted him. The police officers took Tsang to room 7 where one of the police officers, in the presence of the other police officer, slapped him on the face. Direct confrontations 512. On 27 January 2015 Tsang attended the Tsuen Wan Police Station to see whether he could identify the police officer who slapped him. The identification was by way of direct confrontation. In the first direct confrontation Tsang identified D5 as the police officer who slapped him. In the second direct confrontation Tsang identified D6 as the police officer who was present when he was slapped, describing him as one of the escorting officers. Tsang also identified D5 and D6 in court as the persons he identified in the direct confrontations. 513. D5 and D6 objected to the admissibility of the identifications as particularised in the grounds of objection submitted by Mr Chung and Ms Lam[385]. In summary the grounds of objection were that the holding of direct confrontations and the procedure adopted during the confrontations were unfair. 514. These are my reasons for ruling the evidence admissible. 515. Three witnesses gave evidence relevant to the confrontations; Tsang; Chief Inspector Tse (“CIP Tse”) (PW43), who conducted the two confrontations and Superintendent So[386] (“SP So”) (PW51), who was the officer in charge of the complaint made by Tsang and who made the decision to hold direct confrontations. The defence elected to call no evidence. 516. In summary on 14 January 2015 Tsang was invited to take part in identification procedures in connection with the alleged assault at the Central Police Station. Tsang was told the identifications were to be by way of direct confrontation[387]. 517. Tsang attended Tsuen Wan Police Station on 27 January 2015 accompanied by his legal representatives, including Mr Vidler. D5 and D6 were also accompanied by their legal representatives. The two confrontations were recorded on video (exhibits P31 & P32)[388]. 518. The purpose of the confrontations was to see whether Tsang could identify the police officer who slapped him on the face[389]. In each of the confrontations Tsang was told that the person he was asked to look at may or may not have been involved in the case[390]. 519. Notwithstanding the defendants closed their eyes during the confrontations Tsang identified D5 in the first confrontation as the one who slapped him on the face inside room 7[391]and identified D6 in the second confrontation as one of the escorting officers[392]. In cross-examination[393] Tsang said escorting meant escorting into room 7 and disagreed he did not identify D6 as being the police officer present when he was assaulted inside room 7. Discussion 520. Both Mr Chung and Ms Lam referred the court to chapter 46-17 of the Force Procedures Manual[394], which concerns visual identification of suspects. This chapter provides for five types of identification procedures: identification parade; consensual group identification; non-consensual group identification; non-consensual group identification without informing the suspect; and direct confrontation[395]. 521. The first two options require the consent of the suspect. All the defendants refused to participate in identification procedures from a very early stage. In cross-examination[396]SP So said that he was first told about the defendants’ unwillingness to attend any identification procedures on 24 October 2014. On that day SP So gave instructions to ask the defendants, who were being interviewed at the time, whether they were willing to take part in any identification procedures. The defendants all indicated they were not willing. A letter was written on 31 October 2014 to Vidler & Co inviting Tsang to attend a direct confrontation[397]. 522. On 3 November SP So directed a letter to be written to the legal representatives of the defendants seeking written confirmation as to whether they were willing to take part in identification procedures, including group identifications or direct confrontation[398]. Solicitors for the defendants, save for D1, replied the same day saying their clients would not take part in any formal identification parades including group identification or direct confrontation[399]. 523. The consent of the defendants not being forthcoming the holding of identification parades and consensual group identifications were not practicable. Mr Marash SC submitted that the defendants having refused to participate in conventional identification procedures, a direct non-consensual confrontation was the only form of identification procedure left open to the police[400]. Mr Chung and Ms Lam both submitted that non-consensual group identifications should have been held[401]. In reply Mr Marash SC submitted that the police had the discretion whether to hold non-consensual group identifications or direct confrontations. Non-consensual group identification 524. In November and December 2014 the police made three attempts to arrange for Tsang to identify the suspects. The first attempt in early November was cancelled in advance, no reply having been received from Vidler & Co as to whether Tsang would attend. Arrangements were then made for identification on the 26 November. The defendants all attended however Tsang did not. The police next wrote to Vidler & Co on 28 November to invite Tsang to attend identification procedures on 12 December[402]. This was also cancelled as Tsang never agreed to attend. 525. The main reason that identification procedures were not held was because there was disagreement between the police and Vidler & Co as to the use of police officers as actors. Tsang told the court that he did in fact attend the police station on 26 November with the intention of attending the identification procedure. However, having received no satisfactory reply as to why police officers were to be used as actors, Tsang left. SP So agreed that on 7 January 2015 Vidler & Co wrote to the police saying the use of police officers amounted to favourable treatment[403]. 526. The correspondence during this period between the police and Vidler & Co showed that the police in November and December were considering both group identification and direct confrontation[404]. The letter dated 3 December 2014 to Vidler & Co also stated that the police did not rule out the possibility of holding formal identification parades on 12 December[405]. Mr Chung and Ms Lam submitted that the holding of direct confrontations on 27 January 2015 was to overcome the objections made by Vidler & Co[406]. 527. This submission requires a consideration of the reason why SP So gave the instruction for holding direct confrontations. Mr Chung asked SP So the reason why he arranged for a direct confrontation. In reply SP So referred to paragraph 11 of chapter 46-17 and explained that a direct confrontation can only be held where both a normal identification parade and a consensual group identification cannot be conducted. The defendants having indicated that they would not take part in any kind of identification procedure SP So was satisfied the conditions for holding a direct confrontation had been satisfied. 528. Mr Chung then asked SP So why a non-consensual group identification was not adopted. SP So explained that in his view a non-consensual group identification was not to be conducted by prior appointment. On the contrary SP So said this was an option for the Chief Inspector holding the identification procedure where unforeseen circumstances occurred such as a frightened witness or where a suspect objects to a consensual group identification[407]. SP So disagreed with Mr Chung that the reason he held direct confrontations was to avoid the objections raised by Vidler & Co as to the use of police officers as actors. 529. In my view once a suspect has indicated his unwillingness to take part in any form of identification procedure, whether at the parade or as in this case confirmed by letter, consideration should be given to whether any other identification procedure can be held in which the suspect’s consent is not required. The holding of a direct confrontation is the last option to be considered and only where a non-consensual group identification is impractical. This is made clear by paragraph 8 of chapter 46-17, which commences: “Finally the suspect alone may be confronted by a witness…” (my emphasis). 530. Paragraph 6 of chapter 46-17 provides that where neither a parade nor a consensual group identification is practicable, arrangements may be made for the viewing of the suspect in a group of persons without his consent. Although consent is not required and there is no set format for holding non-consensual group identifications, it is clear from paragraphs 6 & 7 of chapter 46-17 that the suspect is to be informed in advance and therefore requires a degree of cooperation from the suspect. 531. Without that cooperation a conventional non-consensual group identification cannot be held. The defendants having said they would not take part in any identification procedure including group identifications, I was satisfied that the holding of a conventional non-consensual group identification was impractical. 532. Even if conventional non-consensual group identifications had been held, I was satisfied the defendants would have refused to cooperate by closing their eyes, just as they did in the direct confrontations. Paragraph 6 of chapter 46-17 specifically gives the covering of the face as an example of a refusal to cooperate. In the circumstances I was satisfied SP So was not at fault in not arranging a conventional non-consensual group identification. 533. Paragraph 7 of chapter 46-17 provides that where the holding of a conventional non-consensual group identification is impractical consideration should be given to conducting a non-consensual group identification without informing the suspect. An example of how to conduct such identification is given for a suspect who is in custody. No guidance is however given as to a suspect who is on bail. Where practical the witness could be taken to the suspect’s place of work or to places the suspect is known to frequent. 534. Clearly the conditions under which a non-consensual group identification without informing the suspect (who is on bail) can be held cannot be controlled to the same degree as identification procedure with the consent of the suspect. The circumstances in which such identification can be held must therefore necessarily be limited. Nevertheless, this option must be considered before deciding whether to hold direct confrontations. 535. I do not agree with SP So’s interpretation of chapter 46-17 that the consideration of holding a non-consensual group identification was only an option for the Chief Inspector holding a conventional identification procedure in the event of unforeseen circumstances occurred during that procedure. I was not satisfied sufficient consideration had been given by SP So to this option before deciding to hold a direct confrontation. 536. I would add that there may well be justification in the objection made by Vidler & Co as to the use of police officers as actors. Although paragraph 31 of chapter 46-17 provides for members of a particular group such as the disciplined services being used as participants in a parade when the suspect is himself a member of that particular group, it is open to argument this would only apply where the suspect is alleged to have committed the offence in uniform. In view of my determination that there had been non-compliance with the Force Procedures Manual on visual identification of suspects by not properly considering whether to hold a non-consensual group identification without informing the suspect, it was not necessary for me to decide this. 537. In HKSAR v Lo Ho Chung the court held that non-compliance with the Force Procedures Manual on visual identification of suspects was not an automatic basis to exclude evidence of the identification. The court said what is important is whether the parade was conducted fairly[408]. The court went on to say whether a breach or non-compliance warrants the evidence inadmissible depends on the nature of the breach and what had actually taken place during the identification[409]. Were the direct confrontations conducted fairly? 538. Admitted in evidence is that on 26 November 2014 both defendants were served with a form entitled “Identification Parade Notice to Suspect”[410]. Paragraph 3 of that form reads as follows: “You are not obliged to attend a parade. If you decline to do so, this fact may be given in evidence at any subsequent Court proceedings at which time a witness may be given the opportunity of identifying you. It will also be open to the Police to make alternative arrangements before any Court proceedings, to test whether the witness does identify you as the person whom he saw on the occasion in question. It should be pointed out to you that neither identification in Court nor identification under whatever other arrangements the Police may make, may be as fair to you as a formal identification parade. You are entitled to request a formal identification parade rather than any other method of identification if you so prefer.” 539. The defendants therefore clearly knew the consequences of refusing to take part in a formal identification procedure. At the beginning of each confrontation the defendants were referred to this notice and again asked if they agreed to take part in a formal identification parade. Both defendants refusing to take part in a formal identification parade the police proceeded to conduct the direct confrontations[411]. 540. Both defendants were legally represented during the confrontations and were explained the purpose of the confrontation. Tsang was told that the person he was asked to look at may or may not have been involved in the case[412]. D5 541. Mr Chung submitted that CIP Tse gave up control of the procedure to Mr Vidler[413]. The basis of this submission arises from the refusal of the defendant to open his eyes. CIP Tse explained to Tsang he could not insist on the defendant opening his eyes because the identification procedure was not held with the consent of the defendant[414]. Between counters 25-38 of the transcript of the confrontation Mr Vidler appeared to object to CIP Tse explaining this to the defendant saying that the defendant had his legal representative present to advise him of his rights[415]. 542. Exactly what Mr Vidler said has not been transcribed. Having carefully viewed the confrontation in an attempt to listen to what Mr Vidler said, whilst it is clear he said more than transcribed, it is not possible to hear clearly what Mr Vidler did say. From the transcript it would appear CIP Tse and the defendant’s legal representative also spoke in English. What they said in English also cannot be heard. 543. The result of the objection by Mr Vidler was that CIP Tse invited the defendant’s legal representative to explain to the defendant what had happened[416]. I do not agree this gives an impression that the identification procedure was dictated by Mr Vidler and is thereby unfair to D5. Quite the contrary. D5 was told he could not be forced to open his eyes. Nothing could have been fairer. 544. Notwithstanding there was no proper consideration of whether to hold a non-consensual group identification without informing the suspect, I was satisfied by refusing to take part in a formal identification procedure, the consequences of which the defendant was fully aware, the holding of a direct confrontation was not unfair. I was satisfied the direct confrontation was conducted fairly. D6 545. Ms Lam submitted that there were irregularities in the conduct of the confrontation which taken cumulatively were unfair to D6[417]. Ms Lam submitted that after Tsang had viewed the defendant for nearly two minutes and before telling CIP Tse he could identify the defendant, there was an exchange of hand gestures between Tsang and Mr Vidler. This Ms Lam submitted was a confirmation or indication by Mr Vidler that Tsang was to identify the defendant. 546. I had no hesitation in rejecting this submission. A careful look at the hand gesture of Mr Vidler, in particular in slow motion, it is clear that all Mr Vidler was doing was indicating Tsang Kin Chiu to sit down. Mr Vidler can be seen to open the palm of his hand and make a gesture towards the table and chair where Tsang had sat when CIP Tse explained the purpose of the identification procedure to him. 547. The inconsistency referred to by Ms Lam in the evidence of CIP Tse and Tsang as to whether CIP Tse had advised Mr Vidler not to make any hand gesture was in my view not material[418]. This inconsistency did not cause me to doubt that Mr Vidler did anything other than indicate to Tsang to sit down. 548. Ms Lam also submitted that the failure by CIP Tse to caution the defendant after informing him that he had been identified was a serious breach of the Force Procedures Manual and a serious violation of the rights of D6[419]. I agree with the submission of Mr Marash SC that the fairness of the confrontation was not affected, the defendant having remained silent after being told he had been identified[420]. 549. Ms Lam further submitted that CIP Tse should have asked Tsang to see if he could identify the defendant based on appearance only before acceding to requests made by Tsang that the defendant stand up and look straight ahead[421]. In support of this submission Ms Lam relies on paragraph 37 of chapter 46-17. Paragraph 37 refers to where a request is made to hear the suspect speak; adopt a specified posture or to move. In those circumstances the witness will first be asked if he can identify on the basis of appearance only. 550. I had no hesitation in rejecting this submission. The requests made by Tsang asking the defendant to stand up and look straight ahead are very different to asking the suspect to speak; adopt a specified posture or to move. In a conventional identification parade the suspect would be standing up and should be looking forward. 551. Notwithstanding CIP Tse agreed with Ms Lam that paragraph 37 applied, in my view paragraph 37 did not apply to the requests made by Tsang. In addition the defendant did not stand up and remained sitting continuously with his eyes closed[422]. Even if there was non-compliance with paragraph 37 the fairness of the confrontation was not affected. 552. Notwithstanding there was no proper consideration of whether to hold a non-consensual group identification without informing the suspect, I was satisfied by refusing to take part in a formal identification procedure, the consequences of which the defendant was fully aware, the holding of a direct confrontation was not unfair. I was satisfied that the direct confrontation was conducted fairly. Media photographs 553. Both Mr Chung and Ms Lam submitted that unfairness arises from the fact Tsang had seen photographs of the defendants which had been exposed by the media[423]. In cross-examination Tsang agreed that one reason he viewed video footage was to find out the identity of the police officers who assaulted him and that in the few days following the incident he paid special attention to the media coverage which showed the faces of the suspects[424]. 554. Mr Chung showed to Tsang photographs of the suspects as appearing in Apple Daily and Ming Pao newspapers on 16 and 17 October 2014 respectively[425]. Tsang said that he had seen the photographs in these newspapers at that time and should have seen the photographs, exhibits P12 & P13 before the confrontation. Tsang said that he might have viewed the media material after being notified on the 14 January 2015 to attend the confrontation and continued to have access to newspaper reports and media coverage after the confrontation. Tsang agreed that he had seen the photographs many, many times. 555. Mr Chung submitted Tsang must have had a deep impression of the facial features of the suspects and therefore the danger that he identified D5 based on this impression was very much alive[426]. Tsang disagreed he identified D5 and D6 because he had seen their photographs in the newspapers and media coverage. 556. In support of their submission both Mr Chung and Ms Lam referred to the case of R v Kwong Yiu Hung[427]where the trial judge excluded evidence of a direct confrontation where the witness had seen a photograph of the suspect released by the police to the press. The defendant was nevertheless convicted and on appeal a re-trial was ordered. Again the trial judge excluded the evidence of the confrontation. 557. The confrontation in Kwong Yiu Hung was held because the other actors were not of the same build and appearance to the defendant and not because the defendant had refused to take part in an identification parade. Similarly HKSAR v Mui Tak Ming[428], also relied on by Ms Lam, is not a case where the defendant refused to take part in an identification parade. 558. In considering whether to exclude the evidence of the confrontation, all the circumstances are to be considered including why a confrontation was held and whether the confrontation was conducted fairly. 559. Mr Chung submitted that D5 was forced to take part in the confrontation and that there was no reason why the other suspects whose photographs had been exposed in the media could not be forced to do the same[429]. 560. I had no hesitation in rejecting this submission. In court Mr Chung explained by force he meant that the confrontation was held against the will of D5. There was no evidence before the court that D5 was forced to go to the police station. D5 was accompanied throughout by his lawyer. D5 was asked if he agreed to a formal identification parade and was explained his rights. 561. I accepted the evidence of SP So that only D5 and D6 were asked to attend the identification procedure on 27 January 2015 because it was believed that only two of the defendants took Tsang to the Central Police Station, which belief was based on the findings of their investigation and not from persons exposed in the newspaper. The CCTV footage shows two police officers escorting Tsang in and out of the police station. I found nothing unfair in the confrontation being limited to D5 and D6. Discretion to exclude 562. Notwithstanding Tsang had viewed the video footage many times and seen photographs of the defendants in the media prior to attending the direct confrontations, I found no grounds to exercise my discretion to exclude the evidence by reason of unfairness. Evidence of identification 563. On the voir dire Tsang testified that after he was assaulted the group lifted him up and took him round the corner of the substation to Lung Wo Road where he boarded a private car. When asked how many males took him to the substation Tsang replied seven. On the general issue Tsang described this as being frogmarched. The TVB and Apple Daily footage show Tsang was taken in the opposite direction to which he was carried to the substation[430]. 564. Two of the seven sat on either side of Tsang in the car. The seven of them discussed with the driver whether it was necessary for the car to wait for others. After waiting for several minutes, Tsang was driven to the Central Police Station.[431] 565. WDPC 3209 (PW33) and DPC 2235 (PW34) were the drivers of two cars. Their evidence was read. DPC 2235 only ferried one suspect Lau Yue Tong [432]. WDPC 3209 made three journeys including one where she ferried two police officers together with one suspect. Although WDPC 3209 did not name the suspect and described the two police officers as police constables, this may have been the car in which Tsang was taken to the Central Police Station. 566. On the general issue Tsang guessed that he was in the car around 30 minutes before reaching the Central Police Station. During this time Tsang said he could clearly see the faces of the two police officers who sat on either side of him in the car. 567. On arrival at the Central Police Station Tsang was escorted to room 7 by the same two police officers who had sat on either side of him in the car. On the voir dire Tsang guessed he was inside room 7 for about half an hour whereas on the general issue he guessed this was an hour or within an hour. The CCTV recordings show that Tsang was in room 7 for just over an hour. 568. Most of the time at least one of the police officers was in the room with the other one standing at the door of the room, which was open. Tsang confirmed room 7 was as seen in photographs 18 & 19, exhibit P25[433] and that the lighting was on in the room. What happened inside room 7 is the subject of charge 2 and will be discussed later. 569. On the voir dire Tsang was shown the CCTV footage, exhibit P15[434]. Tsang identified himself and said file 034000, cameras 12 & 14 showed him being escorted by the two police officers to room 7 and file 045000, cameras 1, 2 & 14 showed him being escorted inside the police station by the same two police officers. 570. On the general issue Tsang said that the two police officers seen on the CCTV escorting him in the police station were the same two police officers who had sat on either side of him in the car. Tsang was not however asked to identify the police officers from the CCTV recordings. 571. The same two police officers escorted Tsang from room 7 to board a coach to go to the Police College in Wong Chuk Hang[435]. On the general issue Tsang said the journey took about 30 minutes. One of the officers sat on Tsang’s right and the other one behind him. The lighting on the coach was dim. 572. Tsang was not asked what opportunities he had of seeing the two police officers after arriving at Wong Chuk Hang. Tsang did however say that the two officers went into hiding. This evidence is considered in more detail later when discussing how Tsang obtained D6’s service number 9008. Discussion 573. I have carefully considered all the evidence and the submissions of Mr Chung and Ms Lam. 574. Mr Chung submitted that however much Tsang wanted to play down the effect of being sprayed with OC foam/pepper spray on his vision by the cleaning of his face, he must still have been affected by the presence of such chemicals on his face and eyes when he was inside the Central Police Station[436]. Ms Lam also submitted Tsang’s vision must have been impaired by the pepper spray[437]. 575. I have no hesitation in rejecting this submission. Although the defence have no burden to prove anything the fact remains there is no medical evidence of the effect of being sprayed with pepper spray for the court to consider. The only evidence before the court is from Tsang. 576. I have already addressed the issue of pepper spray when considering the credibility of Tsang’s evidence[438]. I accept Tsang’s evidence that after being sprayed with pepper spray his vision was not 100% clear but he could still see things and people around him; his vision was blurred for one moment; he was still conscious and could clearly see things within several meters. 577. Mr Chung submitted that there is every possibility that there had been a change in officer(s) handling Tsang at any stage at Lung Wo Road prior to him being taken away to Central Police Station[439]. 578. Notwithstanding I have found Tsang to be mistaken there was no change in the persons who carried him to the substation, I have no hesitation in rejecting this submission. At the end of cross-examination on the general issue Mr Chung put to Tsang that because he had just been assaulted and was very painful and confused he would not have noticed if there was a change of escorting officers whilst waiting in Lung Wo Road. Tsang disagreed saying that in terms of timing it was immediately afterwards. I accept the evidence of Tsang. 579. In oral submissions Mr Chung submitted the estimate given by Tsang of being 30 minutes in the car before reaching the Central Police Station must clearly be wrong. On the available evidence Mr Chung submitted that the assault, which lasted for over four minutes, occurred shortly after 03:30 and that according to the CCTV Tsang arrived at the Central Police Station at 03:43:24. The time spent in the car would therefore have been around 6 minutes. 580. For the reasons already given I do not accept the evidence of David Wong as to the actual times of the assault footage. On the evidence of Mr Sum that the footage from cameras A & B was automatically ingested onto the TVB server roughly between 03:20 and 03:43 and the evidence of David Wong that camera A was broadcast live at around 03:30, I agree with Mr Chung that most likely the assault at the substation occurred shortly after 03:30[440]. 581. As accepted by Mr Chung there is no evidence before the court that the times recorded on the CCTV are the actual times. The Duty Officer in the Central Police Station at that time WSSgt So (PW47) says that at about 03:45 two plainclothes officers and a male suspect walked past her desk[441]. Although WSSgt So did not give the names, ranks or service numbers of the plainclothes officers or the name of the suspect she does describe telling them to go to room 7, the room where Tsang was taken to. 582. WSSgt So’s estimate is very close to the times seen on the CCTV. On the evidence I agree with Mr Chung that the time Tsang spent in the car could not have been 30 minutes. The time would more likely have been around 5 minutes thereby giving Tsang considerably less time to observe the faces of the two police officers whilst he was in the car. Further, although Tsang was sat next to the two police officers in the car, this was night time and the lighting would not have been ideal. 583. The journey to Wong Chuk Hang was estimated to take thirty minutes however again the lighting was not ideal. Tsang described the lighting as dim. 584. Ms Lam submitted that the duration of observation made by Tsang was short and made when he was in a state of shock and pain[442]. I have no hesitation in rejecting this submission. I am satisfied that Tsang had the opportunity of observing the faces of both police officers in the Central Police Station where he was with them for just over an hour in what was clearly very good lighting. 585. Ms Lam submitted that Tsang was clearly uncertain about the identity of D6 because he refused to participate in group identification; only identified D6 after Mr Vidler assisted him and after identifying D6 he asked Chief Inspector Tse, who was conducting the confrontation, if he could walk forward to identify D6 again[443]. 586. I reject that Tsang refused to participate in group identification because he was uncertain about the identity of the two police officers in the Central Police Station. I accept that in refusing to attend a group identification Tsang was acting on the advice of his lawyers who objected to the use of police officers as actors. For the reasons already given, I reject Mr Vidler in any way assisted Tsang to identify D6[444]. 587. The identification of D6 is at counters 25 and 26 of the transcript of the confrontation (exhibit P32C), with the request to walk over there at counter 34. This is followed at counters 35-38 by objection from the legal representative for D6 that it was not necessary. Regrettably the full reasons for the objection are not audible. The video of the confrontation (exhibit P32A) shows that Tsang, apart from looking as he walked out of the room, did not in fact walk over to identify D6 again. 588. In cross-examination Tsang disagreed that the reason he asked to walk forward to identify D6 again was because he was not sure of the identity. In re-examination Tsang explained he made the request because when he made the identification D6’s eyes were closed. Although he could identify D6 with his eyes closed, Tsang wanted to confirm his identity with his eyes open[445]. 589. I accept the evidence of Tsang. The request to walk over to identify D6 again does not cause me to doubt the correctness of Tsang’s identification of D6. 590. I am satisfied so I am sure I can safely rely on the evidence of Tsang that the two police officers who escorted him in the car to the Central Police Station were two of the police officers who assaulted him at the substation. Notwithstanding Tsang had viewed the video footage many times and seen photographs of the defendants in the media prior to attending the direct confrontations, I am satisfied so I am sure I can safely rely on the identification by Tsang that D5 and D6 were these two police officers. 591. The evidence of Tsang does not however stand alone. Police officers who knew D5 and D6 recognised D5 and D6 from video footage and photographs taken prior to Tsang being assaulted. The CCTV recordings show two police officers escorting Tsang in and out of the Central Police Station. I am satisfied this evidence is capable of supporting the identification by Tsang of D5 and D6. Recognition evidence D5 592. Between February 2014 and February 2015 Woman Senior Inspector Wu (“WSI Wu”) (PW39) was the Officer-in-Charge of Regional Special Duty Squad of Kowloon East Region. D5 was under WSI Wu’s command from April 2014 until his interdiction in October 2014[446]. After D5’s interdiction WSI Wu would meet D5 when D5 returned to the police station to be interviewed on a monthly basis until February 2015 when WSI Wu left her post[447]. 593. On 23 May 2016 WSI Wu was invited by WCIP Hung to view the ATV footage P3 (1); the TVB footage P1 (h) and the Now TV footage, P4[448]. After viewing the video footage twice WSI Wu recognised D5 on the TVB footage P1 (h) and marked which person was D5 on a screen capture of the shot[449]. WSI Wu could not recognise any person on either the ATV or Now TV footage[450]. 594. On 25 May 2016 WSI Wu was invited by WCIP Hung to view the two photographs, exhibits P12 and P13. WSI Wu recognised D5 on the photograph exhibit P12 and circled D5 on a copy of the photograph[451]. WSI Wu was unable to recognise D5 on photograph exhibit P13. 595. I accept the identification evidence of WSI Wu. For the reasons given when discussing the case of D1, no reliance is placed on the comparison made by CIP Chan of the warrant card photograph of D5 with the TVB footage. In any event this evidence adds nothing to the identification of D5 by WSI Wu. 596. The identification of D5 by WSI Wu from the TVB footage and the photograph exhibit P12 shows the stage where Tsang was escorted. The group can also be seen escorting Tsang in the ATV footage, exhibit P3 (2) between 01:44 and 01:50. 597. The prosecution case is that not only did D5 escort Tsang but D5 was also present when Tsang was handed over by the uniform police officers and assisted in carrying Tsang face down. 598. The prosecution submitted that D5 is first seen in the ATV footage exhibit P3 (2) at 00:16 as shown in screen capture 4 attached to the prosecution’s closing submissions[452]. This is also captured at the end of the ATV footage between 01:33 and 01:40. The side view only can be seen and for a very short time. This is best seen in slow motion or frame by frame. At normal speed this is very easy to miss. The prosecution submitted D5 is also seen in the photograph exhibit P13 appearing to be holding the left leg of Tsang[453]. 599. Notwithstanding WSI Wu, who knew D5, did not identify D5 from the ATV footage or the photograph, I am nevertheless satisfied D5 can be seen on the ATV footage and is the person holding the left leg of Tsang in the photograph exhibit P13. I am satisfied the facial appearance is the same and the clothing consistent with that seen in the TVB footage and the photograph exhibit P12. D6 600. Between December 2013 and July 2015 Inspector Wong (“IP Wong”) (PW40) was the Officer-in-Charge of District Investigation Team 8 of Kowloon City District. D6 was under IP Wong’s command from December 2013 until his interdiction in October 2014[454]. After D6’s interdiction IP Wong would meet D6 when D6 returned to the police station to be interviewed on a monthly basis until July 2015 when IP Wong left his post[455]. 601. On 11 March 2016 IP Wong was invited by WCIP Hung to view the TVB footage P1 (a); part of P1 (g) and two parts of the ATV footage P3 (2), to see if he could recognise anyone he knew[456]. IP Wong recognised D6 on the second part of the ATV footage and marked which person was D6 on a screen capture of the shot[457]. After viewing the TVB footage twice IP Wong was unable to recognise anyone. 602. On 26 May 2016 IP Wong was invited by WCIP Hung to view the two photographs, exhibits P12 and P13. IP Wong recognised D6 on the photograph exhibit P12 and circled D6 on a copy of the photograph[458]. IP Wong was unable to recognise D6 on photograph exhibit P13. 603. I accept the identification evidence of IP Wong. For the reasons given when discussing the case of D1, no reliance is placed on the comparison made by CIP Chan of the warrant card photograph of D6 with the TVB footage. In any event this evidence adds nothing to the identification of D6 by IP Wong. 604. The identification of D6 by IP Wong from the second part of the ATV footage shows the stage where Tsang was being handed over by the uniform police officers. I am satisfied that D6 is also seen in the first part of the ATV footage exhibit P3 (2) at 00:14 as shown in screen capture 1 attached to the prosecution’s closing submissions[459]. 605. The identification of D6 by IP Wong from the photograph exhibit P12 shows the stage where Tsang was escorted. D6 is seen holding the left arm of Tsang. Notwithstanding IP Wong, who knew D6, did not identify D6 from the TVB footage I am nevertheless satisfied D6 is also seen in the TVB footage P1 (g), between 02:47 and 02:55, holding the left arm of Tsang[460]. I am satisfied the photograph exhibit P12 was taken at the same time as the TVB footage but from a slightly different angle. 606. I am satisfied D6 is also seen in the Now TV footage (exhibit P4) between 00:40-00:44, which was not shown to IP Wong, holding the left arm of Tsang as shown in screen capture 11 attached to the prosecution’s closing submissions. The group can also be seen escorting Tsang in the ATV footage, exhibit P3 (2) between 01:44 and 01:50. 607. The prosecution submitted that photograph exhibit P13 shows D6 carrying Tsang’s left arm[461]. The photograph shows only the left side of the face of the person carrying Tsang’s left arm whereas the moving photograph in the Apple Daily footage, exhibit P2(c)[462], which was not shown to IP Wong, shows the full face of the person carrying Tsang’s left arm. 608. Notwithstanding IP Wong, who knew D6, did not identify D6 from the photograph exhibit P13, I am nevertheless satisfied D6 is the person seen carrying Tsang’s left arm on the photograph exhibit P13 and the moving photograph. I am satisfied the facial appearance is the same and the clothing consistent with that seen in the ATV footage and the photograph exhibit P12. Screen capture ID3 is attached herewith showing the position of D6 on the moving photograph. CCTV 609. Nobody, neither Tsang nor any police officer who knew D5 or D6, was asked to see if they can identify either of the two police officers seen escorting Tsang on the CCTV recordings, exhibit P15. 610. The appearance of the two police officers is best seen from camera 12 between 03:43:40 and 03:43:50; camera 2 between 04:51:20 and 04:51:33; and camera 14 between 04:51:48 and 04:51:58. I am satisfied the CCTV recordings clearly show the same two police officers taking Tsang into the Central Police Station and just over an hour later out of the Central Police Station. 611. I am satisfied the facial features of the two police officers can be seen on the CCTV recordings as for example shown in screen captures 21, 22 & 23 attached to the prosecution’s closing submissions. 612. Having carefully viewed the video footage and looked at the photographs showing D5 and D6 and having carefully viewed the CCTV, I am satisfied that D5 and D6 are the two police officers seen escorting Tsang on the CCTV recordings. 613. I am satisfied the facial appearance is the same in the CCTV recordings as in the video footage and photographs. The clothing of D5 and D6 is consistent with the clothing seen in the video footage and photographs. The T-shirt of D6, as seen for example in screen capture 23 attached to the prosecution’s closing submissions, has the same two rows of writing on the front as seen in photograph, exhibit P12. The clothing is discussed in more detail in the next part of the verdict. 614. I am satisfied the recognition of D5 by WSI Wu and D6 by IP Wong and the CCTV recordings provide supporting evidence for the identification of D5 and D6 by Tsang. Furthermore, taking into account that: (i) D5 and D6 were present when Tsang was handed over by the uniform police officers; (ii) D5 and D6 escorted and carried Tsang prior to Tsang being assaulted; (iii) D5 and D6 escorted Tsang to the Central Police Station after Tsang was assaulted; and (iv) the inherent improbability D5 and D6 would handover Tsang to some other police officers only for them to collect Tsang again and take him to the Central Police Station, I am satisfied so I am sure the only inference to draw is that D5 and D6 were two of the police officers who assaulted Tsang and then escorted him to the Central Police Station. Clothing 615. In addition, I am satisfied supporting evidence is found from the clothing worn by D5 and D6. Unlike the case against D1, D2, D3 and D4 the prosecution do not point to any of the clothing as being distinctive or to any particular part of the assault footage showing the clothing of D5 or D6[463]. 616. From the video footage and the photographs, the clothing of D5 and D6 can be seen. D5 was wearing a grey T-shirt, a black sleeveless jacket and dark trousers. D6 was wearing a light blue/grey T-shirt with two rows of writing on the front, a black sleeveless jacket and blue trousers. 617. In my view the most important feature of D5 and D6’s clothing are the shoes. The shoes of D5 and D6 can be seen in photograph exhibit P13. D5 and D6 are wearing similar shoes. I would describe the shoes as black plimsolls with white laces and a white horizontal band that joins the upper shoe to the sole. The shoes can also be clearly seen in the CCTV recordings, exhibit P15, camera 12 between 03:43:40 - 03:43:50 when Tsang is taken to room 7 and camera 2 between 04:51:20 - 04:51:33 when Tsang is taken from room 7. D6’s shoes can also be seen on the CCTV recording from camera 14 between 04:51:48 - 04:51:58. 618. There has been no identification of D5 or D6 from the footage showing the assault at the substation. In considering the evidence, I take into account WSI Wu, who knew D5 and IP Wong, who knew D6, were unable to identify D5 and D6 from the assault footage shown to them. Again a distinction is to be drawn between being able to recognise a person and comparing whether the clothing is the same. 619. In the photograph, exhibit P13, showing Tsang being carried face down, D5 and D6 are seen holding the left leg and left arm of Tsang. In the first nine seconds of the TVB footage exhibit P1 (a), also showing Tsang being carried face down[464], the persons holding the left arm and the left leg are wearing shoes with the white horizontal band. Screen capture ID11 taken from P1 (a) is attached herewith showing the shoes of these two persons. 620. The TVB footage shows the group arriving at the substation[465]. As already discussed, D1 was holding the left leg of Tsang; D2 the left arm; D4 the right leg and as will be seen later, when discussing the case of D7, the seventh person wearing glasses was holding the right arm. D3 was leading the group. This leaves only two persons, the one walking behind the group and the one walking on the right of D4. 621. As these two persons walk they are both seen wearing shoes with the white horizontal band. Screen capture ID12 taken from P1 (h) is attached herewith showing the shoes of these two persons. Although they have by now changed position and are no longer carrying Tsang they are still together with all their team members with the addition of the seventh person. By following these two persons as the group carry Tsang, they are seen in the position shown in screen capture 18 attached to the prosecution’s closing submissions, which is about ten seconds after Tsang is dropped on the ground. The white horizontal bands on the shoes can also be seen. 622. I find it an odd coincidence that the shoes worn by D5 and D6 are the same as the two persons seen walking with D1, D2, D3, D4 and the seventh person. I am satisfied this is supporting evidence for Tsang’s identification that D5 and D6 were two of the persons who assaulted him. 623. Having carefully considered all the evidence and viewed the video evidence and photographs, I am satisfied so I am sure D5 and D6 are the persons marked by the prosecution on screen capture 18 attached to the prosecution’s closing submissions. From the photographs, video footage and CCTV recordings the hairstyles of D5 and D6 are seen to be very different. The video footage of the assault at the substation is, in my view, sufficiently clear to distinguish between D5 & D6. D5 624. I am satisfied so I am sure when the group arrived at the substation D5 was the one walking on the right of D4. D5 is seen throughout the assault footage and is seen to kick Tsang on a number of occasions, including on one occasion during the time D2 and D3 left the substation. D5 only left when Tsang was taken away. 625. I have carefully considered the submission of Mr Chung that the situation in Lung Wo Road around the time Tsang was arrested was very chaotic; arrested persons were handed over from one officer or team to another; the poor physical condition of Tsang and the quality of the CCTV recordings[466]. These submissions do not cause me to doubt the findings I have made. D6 D6 identified as the arresting officer of Tsang 626. At approximately 5.20 a.m. Tsang together with fifteen other detainees arrived at a temporary detention centre at the Police College in Wong Chuk Hang.[467] After seeing the duty officer, Inspector Wu Zhen Dong (“IP Wu”) (PW48), Tsang waited in a room until he saw his lawyer Ms Tanya Chan. At about 10.00 a.m. in the presence of Ms Chan, Tsang stated he wanted to lodge a complaint of assault against the police[468]. 627. The complaint was recorded in the First Information of Complaints Against Police Report (Pol. 964), exhibit P26. In the report Tsang made complaint against six police officers. Only one police officer was identified DPC 9008, which is the service number of D6[469]. 628. Tsang said he heard the number in the detention room when forms were being filled in. Tsang explained he was dealt with differently to the other detainees. The escorting officers of the other detainees remained with them throughout whereas the two officers escorting Tsang went into hiding. 629. Police officers who were filling out forms, in connection with the handover of detainees, required to know who escorted Tsang. When a police officer asked who escorted Tsang, Tsang heard the number 9008. Tsang did not know who said this but as a result believed one of the two police officers escorting him was DPC 9008. 630. Ms Lam submitted that the evidence of Tsang was different to what Tsang said in the Notice of Application for Leave to Apply for Judicial Review[470]. In cross-examination Ms Lam referred Tsang to paragraph 17 of the grounds on which leave for judicial review was sought[471]. Paragraph 17 stated that Tsang gleaned the police identity number 9008 from some paperwork after he had been transferred to detention in Wong Chuk Hang. 631. Tsang disagreed this was different to his evidence and explained that he heard the number at the same time the police were dealing with the document. Tsang said he did not see the difference between what he said in court and what was in paragraph 17. 632. Tsang was not the only witness to give evidence about this. DSgt 51344 (PW45) and PC 9765 (PW46) also gave evidence about what happened when the detainees were taken to the detention room. DSGT 51344 633. DSgt 51344 was responsible to supervise the detainees brought to the Police College[472]. At 6.10 a.m. sixteen detainees, including Tsang were taken to room 3 where they were each assigned a number. Tsang was assigned number 41. 634. DSgt 51344 explained that when the detainees were brought into room 3 the arresting officer would report to him and reveal his identity. DSgt 51344 ascertained the arresting officer of Tsang was 9008, whose rank he recalled was that of a police constable. DSgt 51344 did not make a record of who was the arresting officer, which was the responsibility of another police officer. 635. In cross-examination DSgt 51344 said before going into the room the arresting officers would pass by his desk at which time they would reveal their identity. 636. In re-examination when asked how he could remember the number of the arresting officer of Tsang was 9008 when so many detainees were processed at the same time, DSgt 51344 replied that the number was uttered and was filled in on the arrest form. DSgt 51344 identified the Short Arrest Form, exhibit P56, as that form. PC 9765 637. PC 9765 was assigned the duty of completing the Short Arrest Forms. When the detainees were taken into room 3, a triage was conducted by plainclothes police officers who recorded information, including the time and date of arrest, on a white board. PC 9765 would copy the information onto the Short Arrest Forms. 638. PC 9765 filled in most of the Short Arrest Form in front of the detainee. On page 1 of the Short Arrest Form exhibit P56, which related to Tsang, PC 9765 wrote down that the AO was DPC 9008. In cross-examination PC 9765 said DPC 9008 was not written on the white board. 639. PC 9765 explained that he asked the plainclothes officer responsible for room 3 who arrested Tsang. Several minutes later a plainclothes police officer claimed he was the arresting officer 9008. PC 9765 therefore wrote this down in the box at the top right of page 1. 640. On page 1 in the box for Escort Officer PC 9765 wrote QRT. PC 9765 did not know who wrote DPC 9008 in this box. PC 9765 also wrote on page 2, under prisoner movement, the entry for 06:10 including that the escorting officer was DPC 9008. The time was written first from the information on the white board and the escorting officer after processing Tsang. 641. In cross-examination PC 9765 agreed that after learning DPC 9008 was the arresting officer, he assumed he was also the escorting officer. PC 9765 therefore wrote DPC 9008 on page 2 but forgot to write this down next to QRT on page 1. In re-examination PC 9765 explained he forgot because he was not clear whether this was required to be written down in the box. In answer to the court PC 9765 said he believed DPC 9008 had not been written next to QRT before he handed over the form to other police officers to process. Discussion 642. Although paragraph 17 of the grounds for leave do not specifically state Tsang heard the number, the difference in my view, is more apparent than real. I accept the evidence of DSgt 51344 and PC 9765 that they were told DPC 9008 was the arresting officer of Tsang. I accept the evidence of Tsang that he heard the number, from which he believed one of the two police officers escorting him was DPC 9008. 643. Even if it can be said this is a material difference, the fact remains that in the Short Arrest Form DPC 9008, the service number of D6, is named as the arresting officer. 644. It was an odd coincidence that D6, who was present when the uniform police officers handed Tsang over, who escorted and carried Tsang prior to Tsang being assaulted and later escorted Tsang to the Central Police Station and Wong Chuk Hang, was recorded as the arresting officer in the Short Arrest Form. I am satisfied this is supporting evidence for the identification of D6 by Tsang. 645. I am satisfied so I am sure when the group arrived at the substation, D6 was the one walking behind the group. D6 is seen throughout the assault footage and is seen to kick Tsang on at least two occasions. D6 only left when Tsang was taken away. 646. I have carefully considered the submission of Ms Lam including that Tsang gave no particulars of DPC 9008 when completing the First Information of Complaints Against Police Report; Tsang was in a state of shock whereby his ability to observe and remember the appearance of various police officers must have been impaired; DSgt 51344 and PC 9765 never checked the warrant card to confirm the identity of DPC 9008; and IP Wong who was very familiar with D6 was unable to identify D6 from the assault footage[473]. Nothing said by Ms Lam causes me to doubt the findings I have made. D7 647. The TVB footage clearly shows that when the group carrying Tsang arrived at the substation there was a seventh person. By following the person holding the right arm of Tsang, this person is seen to turn round after Tsang was dropped on the ground and actively participate in the assault by repeatedly kicking Tsang[474]. This person is seen at various times throughout the assault footage and only leaves when Tsang was taken away when he is seen walking behind Tsang. 648. The clothing of the seventh person can clearly be seen, in particular that he is wearing glasses and a white T-shirt with a distinct pattern as shown in screen captures 17 & 18 attached to the prosecution’s closing submissions. As seen on the video footage and photographs, D1-D6 were neither wearing glasses nor the white T-shirt. I am satisfied so I am sure that when the group arrived at the substation the person holding the right arm of Tsang was the seventh person. 649. The prosecution case is that this person is D7. No police officer recognised D7 on any of the video footage or from the photographs. The only identification of D7 is by CIP Chan who compares the appearance of the seventh person on the TVB and Apple Daily footage of the assault at the substation with the warrant card photograph of D7. Although CIP Chan was unable to see the face of the seventh person on the footage he concluded that the “seventh males’ general descriptions (e.g. wearing spectacles) resembled D7’s general descriptions.”[475] 650. For the reasons already given that CIP Chan did not know the defendants and had not acquired any special knowledge, I am satisfied CIP Chan was in no better position than the court to make such a comparison. The evidence in my view amounts to no more than the opinion of CIP Chan. 651. I place no reliance on the comparison made by CIP Chan. The video footage of the assault, whilst showing the face of the seventh person, is not sufficiently clear for the court to say that person is D7. I therefore turn to consider whether the circumstantial evidence is sufficient from which the court can draw the inference that the seventh person was D7. 652. The circumstantial evidence may be summarised as follows: The operation 653. As summarised at the beginning of this part SP Ng, who was in charge of Crime Group A, briefed the members of Quick Response Teams of their duties and responsibilities, including the processing of arrested persons. 654. SP Ng said that according to the operational plan Tsang should have been arrested and taken by a crime officer to one of the coaches or cars, which were parked on Lung Wo Road near to the substation, for transport to the Central Police Station. D7 was on duty at Lung Wo Road 655. Admitted in evidence is that on 15 October 2014 D7 was on police duty from about 3 a.m. in the area of Lung Wo Road[476]. D7 was a member of the same team as the other defendants 656. Although some of the defendants were attached to different police districts, on that morning they were all part of the same Quick Response Team, A2-2. D1 led the team[477]. D3, D5, D6 & D7 were deployed as members of the team with D2 the officer-in-charge of the team [478]. During the morning D4 was deployed by D2 to assist the team[479]. 657. There were two other team members of QRT A2-2, DPC 8097 (PW28) and DPC 5840 (PW29), whose evidence was read[480]. At the time of the assault on Tsang DPC 8097 and DPC 5840 were processing arrested persons. 658. At about 03:15 at Tamar Park DPC 5840 took over male Lau Sze Yeung from a uniform police officer. DPC 5840 declared arrest on male Lau for unlawful assembly and obstructing police in the execution of duties and took him to the generator room for a quick search. DPC 5840 then escorted male Lau onto coach LZ 4923[481]. 659. At 03:15 something under the flyover (underpass) of Lung Wo Road DPC 8097 took over male Chan Tse Kin. DPC 8097 also took male Chan to coach LZ 4923 where he queued up to wait for details of the arrest to be recorded; photographs to be taken and to conduct a quick search[482]. 660. DPC 5840 and DPC 8097 escorted the arrested persons to the Police College in Wong Chuk Hang. The coach was first driven to the Central Police Station and then to the Police College in Wong Chuk Hang[483]. D7 wears glasses[484] 661. The seventh person was wearing glasses. The warrant card photograph taken in May 2013 shows D7 wearing glasses[485]. DPC 8097 had worked with D7 for about one year prior to 14 October 2014. DPC 8097 stated that D7 sometimes wears glasses but could not remember whether he was wearing glasses on the day of the operation[486]. DPC 5840 stated he had only worked with D7 for two days and had no idea of his appearance and what he was wearing on the day of the operation[487]. 662. Sgt 58332 (PW44) knew D7 since late 2012 when working in different teams of the Kwun Tong Anti-Triad Section. Sgt 58332 stated that D7 normally wears glasses. On 14 October 2014 Sgt 58332 saw D7 on duty but did not talk to him. Sgt 58332 further stated that it seems D7 was wearing glasses that day[488]. 663. DSPC 21077 (PW31) gave evidence that he had also worked with D7 for about one year prior to 14 October 2014. DSPC 21077 also stated that D7 sometimes wears glasses. DSPC 21077 saw D7 twice on the day of the operation but did not pay attention as to whether he was wearing glasses. Presence at the substation 664. D7 assisted DPC 8097 to take male Chan Tse Kin to coach LZ 4923. DPC 8097 estimated that it took 20-25 minutes to walk westbound along Lung Wo Road to the coach and queue for arrested person processing[489]. 665. Coach LZ 4923 was parked on the same side of Lung Wo Road as the substation[490]. Tsang was taken to the north side of the substation, which was at the opposite end of the substation to where the coach was parked. 666. After processing D7 left to join other team members[491]. The prosecution submitted that by leaving to join other team members this is exactly what D7 should have done and was about the time the rest of the team were carrying Tsang to the substation[492]. 667. The estimates given by DPC 8097 would place D7 near the substation between 03:35 and 03:40. To join his team members D7 would need to walk back along Lung Wo Road eastbound. This would take D7 past the substation. 668. Mr Lo submitted that it is doubtful D7 was at the substation when Tsang was assaulted[493]. Time of the assault at the substation 669. Mr Lo submitted the times shown on the TVB footage, exhibit P1 (g) are the actual times Tsang was escorted and assaulted, namely between 03:32 and 03:37[494]. No evidence was however given by TVB as to who inserted the times on the footage. 670. In court Mr Lo submitted that the times on the footage were consistent with the evidence of David Wong. In cross-examination when asked about the blocked shot David Wong confirmed that in his affirmation he said the assault at the substation was filmed by camera B between 03:33:08 and 03:37:48[495]. 671. As explained earlier the evidence of David Wong that the times were real time obtained from blu-ray disc B, which was recorded simultaneously with the live transmission, differed from the evidence of Mr Lam that he burnt both blu-ray discs A & B in the afternoon. Mr Lam being the one who burnt the discs I accept his evidence and do not rely on the times given by David Wong as being the actual times when the assault at the substation was filmed by camera B [496]. 672. What the court can rely on is the evidence of Mr Sum that the footage from cameras A & B was automatically ingested onto the server roughly between 03:20 and 03:43 and the evidence of David Wong that camera A was broadcast live at around 03:30. 673. After Tsang was assaulted he was taken to the Central Police Station. The CCTV recordings show Tsang entering the Central Police Station at 03:43:24. Again as pointed out in court there is no evidence before the court that the times recorded on the CCTV are the actual times[497]. 674. The Duty Officer in the Central Police Station at that time WSSgt So says that at about 03:45 two plainclothes officers and a male suspect walked past her desk[498]. Although WSSgt So did not give the names, ranks or service numbers of the plainclothes officers or the name of the suspect she does describe telling them to go to room 7, the room where Tsang was taken. 675. Allowing for time to travel to the Central Police Station the assault most likely took place shortly after 03:30. Even if the times given by David Wong are the actual times of the assault at the substation, the times given by DPC 8097 remain only estimates. I am satisfied so I am sure on the evidence that D7 was present by the substation at the same time D1-D6 would have been carrying Tsang to the substation. Inherent improbabilities 676. Tsang was not carried direct to where the coaches and cars were parked for processing. Instead D1-D6 carried him to the opposite side of the substation, the north side, away from the road. As seen from the TVB assault footage as soon as Tsang was dropped on the ground the assault immediately started. The only inference to draw is that D1-D6 took Tsang to the opposite side of the substation, from where the coaches and cars were parked, to assault him. This is discussed later when considering the issue of joint enterprise. 677. The assault footage shows that the seventh person was probably the first person to kick Tsang[499]. The only inference to draw is that when the seventh person assisted D1-D6, by holding the right arm of Tsang, he already knew the purpose of taking Tsang to the opposite side of the substation from where the coaches and cars were parked was to assault Tsang. 678. The act of assaulting Tsang was unlawful. I find inherently improbable that D1-D6 would let a police officer from another team assist them to carry Tsang and witness their assault on him. 679. I find inherently improbable a police officer who was not a member of QRT A2-2 assisted D1-D6 to carry Tsang to the substation. By this time Tsang’s hands were handcuffed at the back with plastic zip ties. There were six police officers taking Tsang to the substation. There was no need for a police officer from another team to render assistance. Six police officers were more than sufficient to escort Tsang to one of the coaches or cars for processing. Conclusion 680. Taking into account: (i) D7 was a member of QRT A2-2; (ii) D7 was by the substation when D1-D6 were carrying Tsang to the substation; (iii) the inherent improbability that D1-D6 would let a police officer from another team assist them to carry Tsang and witness their assault on Tsang; (iv) the inherent improbability that a police officer who was not a member of QRT A2-2 assisted D1-D6 to carry Tsang; (v) the only other members of QRT A2-2 DPC 8097 and DPC 5840 were at the time of the assault engaged in processing arrested persons; and (vi) D7 normally wears glasses and seems like he was wearing glasses on the day of the operation; I am satisfied so I am sure the only inference to draw is that: (a) while D7 was walking to join his team members he saw D1-D6 carrying Tsang to the substation; and (b) he assisted his team members to carry Tsang and participated in the assault on Tsang. 681. I have carefully considered the submission of Mr Lo including that no one has identified D7; D7’s deployment to QRTA2-2 cannot be relied on as evidence of D7’s actual involvement in the crime because there were no less than 200 plainclothes police officers on duty and police officers from different teams would lend a helping hand to other teams; there was insufficient basis for the court to say D7 wore spectacles on 15 October 2014; and that the facial features of the assailant in spectacles are exceptionally poor whereby no safe identification can be made. [500]. Nothing said by Mr Lo causes me to doubt the findings I have made. Joint Enterprise 682. The prosecution case is that the defendants were acting in concert pursuant to an express or tacit agreement to cause Tsang serious injury[501]. The prosecution relied on the fact Tsang was carried to the north side of the substation and not to the south side where the waiting coaches and cars were parked along Lung Wo Road[502]. 683. As summarised earlier SP Ng (PW27), who was in charge of Crime Group A, briefed the members of Quick Response Teams of their duties and responsibilities, including the processing of arrested persons[503]. 684. SP Ng said that according to the operational plan Tsang should have been arrested and then taken by a crime officer to one of the coaches or cars for transport to the Central Police Station. In cross-examination SP Ng said that the actual decision made as to what to do would be made according to the situation at that time[504]. 685. The evidence of the drivers of the coaches and the cars was read and show that the coaches and the cars were parked on Lung Wo Road near to the south side of the substation[505]. 686. Mr Cheng SC submitted that it was not uncommon for protestors to be brought to the substation before boarding one of the vehicles parked on Lung Wo Road[506]. In making this submission reliance was placed on the evidence of DPC 5840 (PW29); PC 12232 (PW52) and PC 8382 (PW53)[507]. 687. The evidence of these three witnesses was read[508]. In summary DPC 5840 took an arrested person to the substation (generator room) for a search, describing the substation as a suitable place. PC 12232 arrested a person who had been walking on the rooftop of the substation and searched him at the roadside near the substation. PC 8382 brought an arrested person somewhere near the substation to sit down. All three witnesses described that while waiting to board the coaches, other arrested persons were brought near the substation to wait for transportation. 688. Tsang was not carried direct to where the coaches and cars were parked for processing. Instead Tsang was taken to the north side of the substation. The video evidence is clear that on arrival at the substation Tsang was dropped on the ground and immediately assaulted. I am satisfied so I am sure the only inference to draw is that Tsang was carried to the north side of the substation to be assaulted and not for searching or to wait for transportation. Participation 689. Mr Cheng SC further submitted that on the basis the prosecution was able to prove D2’s presence they were still unable to prove that D2 had taken part in the assault. On the contrary, Mr Cheng SC submitted that the highest the prosecution could take their case was that D2 was present and that the person which vaguely or might reasonably resemble D2 made at least two attempts to stop the others from further perpetrating the assault[509]. 690. Mr Cheng SC citing from Archbold Hong Kong 2017 that mere presence alone is insufficient submitted that the court cannot rule out D2 was simply present without any intention that Tsang would be assaulted. 691. I have no hesitation in rejecting this submission. This was not a case of mere presence. For the reasons already given, I am satisfied D2 was the one carrying the left arm of Tsang as the group arrived at the substation[510]. The only reason Tsang was carried to the north side of the substation was to assault him. 692. Mr Cheng SC submitted that the assault footage clearly showed that at least two persons did not take part in the assault and that if the court recognized D2 as one of the persons depicted in the assault footage then D2 was one of the two who did not take part in the assault[511]. 693. Of these two persons Mr Cheng SC submitted the only one which vaguely or might reasonably resemble D2 is the person circled in red in Annex B to his closing submissions. This person Mr Cheng SC submitted appears to make at least two attempts to stop the others from further perpetrating the assault as particularised in Annex C[512]. 694. The screen captures in Annex B and Annex C are taken from the TVB footage P1 (h). Annex B is the same as the prosecution screen capture 19. For the reasons already given I am satisfied this person is D2[513]. 695. The TVB footage P1 (h) is an edited version of the assault. I have looked at the full assault as shown in the TVB footage P1 (a) to see what occurred before and after each shot relied on by Mr Cheng SC[514]. Annex C (1) is seen between 26-30 seconds and Annex C (2) between 42-46 seconds on the TVB footage P1(a). 696. The TVB footage P1 (a) at 14 seconds shows Tsang being dropped on the ground. Between 14-26 seconds the footage shows Tsang being assaulted by D3, D4, D5 and D7. At 26 seconds D2 is seen to turn round and quickly look at Tsang on the ground. At this time D7 can be seen kicking Tsang. Other than looking at Tsang, D2 appears to do nothing else. I reject that D2 was attempting to stop the assault on Tsang. 697. The assault continued with D3, D5, D6 and D7 assaulting Tsang. At 42 seconds D2 turns round again at which time D3 stands up and walks away from Tsang. The assault on Tsang seems to stop. 698. Whilst D2 may well have said something or pulled D3 away, I reject the submission of Mr Cheng SC that this negates the existence of a criminal joint enterprise[515]. Tsang was carried to the north side of the substation to be assaulted. D2 was the one who held Tsang’s left arm. By the time D3 stands up and walks away from Tsang the assault had lasted for almost thirty seconds during which time D2, the officer-in-charge of the team, had done nothing to stop the assault on Tsang. 699. Every police officer has a duty to intervene to prevent the commission of a crime, even by fellow police officers. A police officer’s duty is to keep the peace. If a police officer stands by and watches his colleague beat up a suspected person, his failure to intervene is evidence of encouragement to carry out the assault[516]. 700. The first thirty seconds of the assault footage shows D3 stabbing Tsang[517]; D3 stamping on Tsang; and D3, D4, D5, D6 & D7 all kicking Tsang. The actions of each defendant are best seen by following one person at a time and viewing the footage in real time, slow motion and frame by frame. During the period that D2 and D3 left the substation, D5 and D7 are seen kicking Tsang. I am satisfied so I am sure D3, D4, D5, D6 and D7 intended to cause Tsang to sustain unlawful personal violence. 701. D1 and D2 are not seen assaulting Tsang. Taking into account that D1 and D2, the leader and the officer-in-charge of the team, both carried Tsang to the north side of the substation so Tsang could be assaulted, I am satisfied so I am sure D1 and D2 were party to the assault on Tsang. By their continued presence I am satisfied so I am sure that D1 and D2 intended to and did encourage and support the other defendants to carry out the assault on Tsang intending Tsang to sustain unlawful personal violence. Grievous bodily harm 702. Tsang was medically examined at approximately 11:05 a.m. on 15 October 2014 by Dr Zenith Wu at the Accident and Emergency Department of the Ruttonjee and Tang Shiu Kin Hospital. The medical examination form completed by Dr Wu (exhibit P27 (a)) and a medical report (exhibit P27 (b)) detailing the injuries of Tsang together with photographs of the injuries (exhibit P28) taken by Dr Wu, have been admitted by agreement[518]. 703. Dr Wu found the following injuries: 1. Swelling and painful reddish bruise on forehead, and bilateral upper face. 2. A linear reddish bruise area on left chin, measuring about 1.5 cm. 3. Multiple reddish bruises over left neck. Linear reddish bruises over lower central neck. Cervical spine not painful. 4. Reddish bruises over left shoulder and clavicle. Range of motion full. 5. Circular reddish bruises over anterior chest wall. Diameter measured around 2 cm. Painful on palpation of the sternum. 6. Reddish bruises over left flank. Bruise over right flank. Painful on palpation. 7. Multiple circular reddish bruise over the back, diameter measured 2 cm. Painful on palpation. 8. Left wrist reddish bruise and painful on palpation. Range of motion full. 9. Abrasion over right forearm and hand. Reddish bruises over right elbow region. 10. Abrasion over left knee. 704. All the injuries were fresh[519]. X-ray performed for the whole spine, chest, right ribs, pelvis, sternum, left hip, bilateral hand and wrist, showed no fracture. CT scan of the brain and orbit showed no intracranial haemorrhage or fracture of the skull and orbit. Tsang was treated and discharged. 705. Admitted in evidence is the expert opinion of Dr Lai Sai Chak (exhibit P29), Senior Forensic Pathologist, on the causation of the injuries[520]. Dr Lai was provided with the medical examination form completed by Dr Wu; the medical report and the photographs. Dr Lai gave his opinion in respect of all the injuries found by Dr Wu. Paragraph 3 of Dr Lai’s report sets out his opinion by reference to fourteen different parts of the body where injuries were found. 706. The prosecution case is that the injuries being numerous and extensive amount to grievous bodily harm[521]. Submissions were made that Tsang sustained all his injuries when he was apprehended and subdued by the uniform police officers, although in court this was modified to the majority were consistent with Tsang being subdued or most, if not all[522]. Evidence 707. On the voir dire Sgt 47574 (PW1); Sgt 34200 (PW2); SIP Fu (PW3); SSgt 52820 (PW4); and SIP Wat (PW5) gave evidence about how they subdued Tsang. This evidence together with the evidence of Tsang and the evidence of SIP Lau (PW32), who was called on the general issue only, has been summarised earlier[523]. 708. On the general issue the prosecution recalled the police officers and Tsang. The evidence largely concentrated on whether in subduing Tsang the police officers used their batons and whether they were wearing kneecap protection kits. I will address this evidence when considering the cause of the injuries to the chest and back of Tsang. 709. The police officers were cross-examined on the degree of violence Tsang used to resist them and how they subdued Tsang[524]. All the police officers said that Tsang struggled violently. At one stage Tsang fell on the ground and banged his head against a wall[525]. 710. On the voir dire Tsang said he was pushed to the ground, assaulted, sprayed with pepper spray and handcuffed with zip wire. In cross-examination on the general issue Tsang described being pushed off the flower bed from behind and pressed down to the ground[526]; that he was struck in the chest and back[527]; and that he believed he was hit on his chest and back with a hard object[528]. Discussion 711. As stated earlier I accept the evidence of the police officers that Tsang struggled violently[529]. On the evidence of the police officers and Tsang there can be no doubt Tsang sustained injuries when he was apprehended and subdued by the uniform police officers. 712. On the general issue Tsang said there were slight scratches, perhaps to his face and that he was not aware of any other scratches. In cross-examination Tsang said he was not seriously injured before arriving at the substation[530]. 713. In cross-examination Sgt 47574 said he did not know if Tsang sustained any injury[531]. Sgt 34200 gave evidence that he believed Tsang suffered injuries. In cross-examination when asked about different stages of the apprehension of Tsang, Sgt 34200 variously said he believed Tsang received injuries; it was possible Tsang received injuries; and that he did not know or pay attention to whether Tsang received injuries[532]. In re-examination Sgt 34200 said that when the OC foam was rinsed off Tsang’s face he saw redness on the face and head of Tsang. 714. In cross-examination SSgt 52820 said after Tsang banged his head against the wall he did not see if Tsang sustained any injuries because he also fell and bumped his head on the wall[533]. SIP Lau, SIP Fu and SIP Wat were not asked whether they saw any injuries on Tsang. 715. Dr Lai’s opinion is that the injuries to the chin and the lower central neck could have been caused when Tsang’s face mask was forcefully removed[534]; the injuries to the left wrist were highly consistent with injuries seen after binding by plastic zip-ties[535]; the injuries to the right arm and elbow and left forearm were probably caused when Tsang was carried by his limbs[536]; and the injury to the right hand[537] was almost certainly caused when someone grabbed his hand[538]. 716. I accept the expert opinion of Dr Lai as to the probable cause of these injuries. I find these injuries were most likely caused when Tsang was being subdued and/or carried and not when he was assaulted at the substation. 717. In the opinion of Dr Lai the injury to the left knee was highly consistent with the result of a fall with the knee hitting the ground[539]. Tsang having fallen to the ground when apprehended and dropped on the ground at the substation, the injury to the left knee could have been caused at either time. 718. This leaves the injuries to the face; the left side of the neck; the left shoulder and clavicle; the chest; the left flank; the right flank; and the back. In the opinion of Dr Lai some of the injuries to the face; the left side of the neck; the left shoulder and clavicle; the left flank and the right flank may have been caused by kicking[540]. 719. There can be no doubt Tsang sustained injuries when he was assaulted at the substation. Viewing the assault footage, in particular the first thirty seconds which show D3 stabbing Tsang; D3 stamping and kicking Tsang and D4, D5, D6 and D7, especially D7, repeatedly kicking Tsang, I have no hesitation in rejecting the suggestion Tsang sustained all or most of his injuries when being subdued by the uniform police officers. During the period that D2 and D3 left the substation D5 and D7 are again seen kicking Tsang. 720. I am satisfied that whilst some of the injuries may have been caused when Tsang was subdued most of the injuries to the face; the left side of the neck; the left shoulder and clavicle; the left flank and the right flank were sustained during the assault at the substation. Chest and back 721. In the opinion of Dr Lai the circular reddish bruises over the anterior chest wall and back were almost certainly caused by a forceful poking (jabbing) action with a fully retracted PPCT Phoenix Baton or a similar instrument with exactly the same configuration[541] . PPCT Phoenix Baton 722. Two police batons an ASP Friction Loc Baton, exhibit P22 (a) and a PPCT Phoenix Baton, exhibit P22 (b) were admitted in evidence[542]. Sgt 47574; Sgt 34200[543], SIP Lau, SIP Fu; SSgt 52820, and SIP Wat all gave evidence that on that night they were issued with the Phoenix brand batons. 723. In cross-examination SSP Chan, who recognised D1 on the video footage, said that a uniform branch baton would not be issued to officers of Organized Crime and Triad Bureau[544]. D1 was a member of the Organized Crime and Triad Bureau[545]. 724. The other defendants were crime officers attached to the Anti Triad Section of either Kwun Tong Police District or Kowloon City Police District except for D5 who was attached to Regional SD Squad of Kowloon East[546]. 725. SP Ng (PW27) gave evidence that during the operation he was in charge of about 150 crime officers called Crime Group A, which consisted of eight Quick Response Teams and fourteen video teams. D2, D3, D5, D6 and D7 were deployed as members of Quick Response Team A2-2 and D4 as a member of a video team[547]. 726. Admitted in evidence is that on 14 October 2014 SPC 22534 issued twelve retractable ASP Friction Loc Batons to D3. The batons were returned to SPC 22534 after the operation by DPC 8097 (PW28).[548] DPC 8097 and DPC 5840 (PW29) were members of the same Quick Response Team as D3 and had been given a baton by D3 for use in the operation[549]. 727. The defendants being crime officers and not uniform officers, the defence submitted that it was farfetched to suggest they borrowed a Phoenix baton to assault Tsang[550]. I have no hesitation in rejecting the submissions that because the defendants were not issued with Phoenix batons they could not have caused the injuries found on the chest and back of Tsang. 728. All the uniform police officers testified that they did not use their police batons in overcoming the resistance put up by Tsang. Sgt 47574 did not pay attention to whether other police officers used their baton; SIP Fu, SSgt 52820 and SIP Lau did not see any police officer use their baton. SIP Wat was not asked if he saw any police officer use their baton. I accept the evidence of the police officers. 729. Mr Cheng SC submitted that if the uniform police officers did not use their batons the only plausible explanation for the injuries is that Tsang was assaulted not by the defendants (who did not possess the PPCT Phoenix Batons) but by another group of police officers who did possess PPCT Phoenix Batons[551]. A similar submission was also made by Mr Lam[552]. 730. I have no hesitation in rejecting this submission. For the reasons already given, I am satisfied the defendants did assault Tsang. Furthermore, the evidence of the police officers does not stand alone. A careful viewing of the video footage in slow motion shows Tsang being stabbed on at least two separate occasions[553]. 731. The TVB video footage P1(g)between 03:11 and 03:13 and (h) between 00:25 and 00:27[554] show D3 holding something in his hand, the shape of which is consistent with a baton. D3 is seen stabbing Tsang twice, which action is consistent with what Dr Lai calls a forceful poking (jabbing) action. Whether D3 stabbed Tsang more than two times cannot be seen. This part of P1 (h) was played in court during the prosecution’s closing submission. Screen capture A1 taken from P1 (h) is attached herewith showing the stabbing action. 732. Six seconds later between 03:17 and 03:19 on P1 (g) and between 00:31 and 00:33 on P1 (h), D3 is seen to move and again stab Tsang twice on a different part of his body, as shown in screen capture A2 attached herewith, taken from P1 (h). Again whether D3 stabbed Tsang more than two times cannot be seen because D1 is standing in front of Tsang and D3. Although on this occasion it is very difficult to see whether D3 is holding something in his hand the action is also consistent with what Dr Lai calls a forceful poking (jabbing) action and not with someone punching Tsang. 733. The video footage together with the evidence of the uniform police officers that they did not use or see any police officer use their baton when subduing Tsang provides cogent evidence that D3 used a Phoenix baton to stab Tsang causing some, if not all, of the circular reddish bruises. Kneecap 734. Sgt 34200; SIP Lau; SIP Fu and SIP Wat all gave evidence they were wearing kneecap protection kits[555] whereas SSgt 52820 said he was not wearing a kneecap protection kit. In cross-examination Sgt 47574 said he did not remember whether he wore a kneecap protection kit[556]. 735. A kneecap protection kit, exhibit P40, was produced by Sgt 34200 in re-examination. In a supplemental report, exhibit P29 (a), Dr Lai examined a police kneecap of the same configuration as exhibit P40 to see if the kneecap could have caused these injuries[557]. Dr Lai was of the opinion that the kneecap did not seem to produce an injury easily, however when the whole body weight was used to exert pressure, especially against resistance, it was conceivable the kneecap could produce braided abrasions at the edge[558]. 736. In conclusion Dr Lai was of the opinion that it was doubtful whether the kneecap had caused all the circular reddish bruises; that his original opinion the circular reddish bruises were caused by a fully retracted PPCT Phoenix Baton still stands; and that as the injury to the central chest seemed rather convincingly (sic), he did not completely exclude the kneecap was the only causative instrument that had caused all the circular reddish bruises found on Tsang’s body[559]. 737. The opinion of Dr Lai that he could not completely exclude the kneecap was the only causative instrument that had caused all the circular reddish bruises found on Tsang’s body was premised on the basis that the injury to the central chest was caused by a kneecap protection kit. 738. I am satisfied this conclusion is not supported by evidence. The evidence of the uniform police officers is that when being subdued Tsang fell on the ground face down and that in subduing Tsang they did not use their knees to press on his chest. Only Sgt 34200 said he used his knee to press on Tsang. 739. In summary Sgt 47574 grabbed Tsang on the planter by putting his arm around Tsang’s neck. Sgt 47574 pressed Tsang down causing Tsang to kneel/squat down. As Tsang resisted, Sgt 34200 and SIP Lau assisted Sgt 47574 by pulling Tsang down from the planter to the pavement. Tsang continued to struggle resulting in Sgt 34200 spraying OC foam in Tsang’s face. When Tsang struggled even more intensely, SSgt 52820 assisted by putting his arm around Tsang. 740. Tsang still continued to resist by swinging his body from side to side causing both SSgt 52820 and Tsang to fall on the ground. Tsang still continued to struggle. At this stage Sgt 34200 said he used his knee for two to three minutes to press against the upper body of Tsang to stop Tsang struggling so intensely. After Tsang was handcuffed with his hands behind his back the police officers helped Tsang to stand up. 741. In cross-examination Sgt 34200 said that when Tsang was on the ground his face and chest were downwards; Tsang faced the ground for more than two to three minutes; and that during this time he used his kneecap protection kit to press against the upper body of Tsang to stop Tsang struggling so intensely[560]. 742. SIP Lau gave evidence that when Tsang was pulled to the ground he was face down. SSgt 52820 gave evidence he only saw Tsang face down and in cross-examination said when colleagues were pressing Tsang down Tsang was facing downwards[561]. SIP Wat gave evidence that when he first saw Tsang on the ground he was facing down which he confirmed in cross-examination saying Tsang was still face down as he continued to struggle[562]. 743. In cross-examination Sgt 47574 said he did not press on Tsang’s back to subdue him; and did not see any other police officers do that.[563] SIP Lau gave evidence that he could not be sure whether he put his knee on the back or chest of Tsang to restrain him. SIP Fu had no recollection of using his kneecap protection kit to press Tsang down or of using his knees to apply force to Tsang. SSgt 52820 gave evidence he did not use his knees to keep Tsang down. SIP Wat gave evidence that apart from grabbing Tsang’s hand he did not use any other force on Tsang. 744. Ms Lam submitted there were material discrepancies in the evidence of the police officers[564]. In a fast evolving situation with police officers arriving at different times, it is not surprising there are differences in their evidence. 745. The differences Ms Lam points to, for example whether Tsang was squatting or had fallen on the ground; whether SIP Lau assisted before or after Tsang was pulled down from the flower bed and how Tsang was subdued, are more apparent than real. 746. SIP Lau’s evidence that he did not see any colleagues use their knees to press against the back of Tsang was confined to the two officers who he first saw struggling with Tsang and did not refer to Sgt 34200[565]. 747. Ms Lam further submitted that the two circular reddish marks on the chest could have been caused by an unknown police officer[566]. The basis of this submission was that in their evidence Sgt 34200, SIP Lau, SIP Fu and SSgt 52820 all referred to a police officer who they did not know or identify. 748. On the voir dire Sgt 34200 referred to two officers assisting him in subduing Tsang, SSgt 52820 and one other who he did not know. In cross-examination on the general issue Sgt 34200 said he could not be sure if this officer was also a PTU officer as he did not pay attention to his number[567]. 749. In his evidence Sgt 34200 made no mention of SIP Wat, who assisted in handcuffing Tsang. In cross-examination on the general issue Sgt 34200 said SIP Wat assisted in handcuffing Tsang[568]. Ms Lam submitted the other officer could not have been SIP Wat because SIP Wat and Sgt 34200 were both in the same team, PTU Training Team 1. 750. Whilst there may be some doubt as to the identity of this officer I have no hesitation in rejecting the submission this officer may have caused injuries to Tsang by the use of his baton or a kneecap protection kit. In cross-examination on the general issue Sgt 34200 said this officer struggled with Tsang during which time the officer tried to grab hold of Tsang with his hands; the officer did not have to use his knees when catching Tsang and that he did not see any officer use their baton[569]. 751. On the general issue SIP Lau gave evidence that when he arrived at the pavement with Sgt 34200, he saw Tsang struggling with two police officers in uniform. In cross-examination SIP Lau said he learnt the identity of one officer after attending court[570]. In re-examination he said this was SSgt 52820. 752. On the voir dire SIP Fu gave evidence that he also saw SSgt 52820 and another officer struggling with Tsang. SIP Fu said SSgt 52820 and the other officer were from Company E. SIP Fu did not know the other officer. 753. In his evidence SIP Lau made no mention of SIP Fu or SIP Wat. Ms Lam submitted the other officer could not have been SIP Fu or SIP Wat because they were all members of the same team, PTU Training Team 1. 754. Again whilst there is some doubt as to the identity of this officer I have no hesitation in rejecting the submission this officer may have caused injuries to Tsang by the use of his baton or kneecap protection kit. 755. I accept the evidence of SIP Fu and SIP Lau that they did not see any officer use their baton when apprehending Tsang. In cross-examination both SIP Fu and SIP Lau said the two officers were not issued with kneecap protection kits[571]. This evidence is consistent with the evidence of SSgt 52820 who said he was not wearing a kneecap protection kit. Further in cross-examination SIP Lau specifically said he did not see the two officers who were struggling with Tsang use their batons or their knees to press against the body of Tsang[572]. 756. On the general issue SSgt 52820 gave evidence that another police officer helped lift Tsang up. SSgt 52820 did not specify who helped him. The ATV video, exhibit P3 (2) clearly shows this was SIP Fu. In explaining this submission in court Ms Lam accepted that the footage shows SIP Fu helping SSgt 52820 to lift Tsang up. 757. I accept the evidence of the police officers. I am satisfied there is no evidential basis to support Dr Lai’s opinion that the injury to the central chest was caused by a kneecap protection kit. 758. I accept Dr Lai’s opinion that it was doubtful whether the kneecap had caused all the circular reddish bruises; and that his original opinion the circular reddish bruises were caused by a fully retracted PPCT Phoenix Baton still stands. I am satisfied that I can safely exclude the kneecap as the cause of all the circular reddish bruises found on Tsang’s body. 759. I accept the evidence of Sgt 34200 that he used his kneecap protection kit to press on Tsang. There is, however, a question mark over whether Sgt 34200 was wearing his kneecap protection kit at that time. The ATV footage, exhibit P3 (2), shows that when SIP Fu and SSgt 52820 were helping Tsang to stand up after Tsang had been subdued, Sgt 34200 was not wearing a kneecap protection kit. 760. When the court drew this to the attention of Sgt 34200, the Sgt explained that he was wearing his kneecap protection kit that night but could not be sure if he had taken the kit off. In re-examination Sgt 34200 said he had no recollection of taking the kit off; the kit could easily come off, especially when there was an intense struggle; and that he had a deep impression he was wearing his kit that night because he had problems with his knees and had to wear the kit when working. 761. Even if Sgt 34200 was wearing his kneecap protection kit when he used his knee for two to three minutes to press against the upper body of Tsang, this does not cause me to doubt that D3 used a PPCT Phoenix Baton to stab Tsang causing some of the circular reddish bruises. Conclusion 762. On the evidence I am satisfied so I am sure that D3 used a PPCT Phoenix Baton to stab Tsang causing some, if not all, of the circular reddish bruises. In making this finding I reject the prosecution submission that it was pertinent to note that during the assault D2 and D3 left for about one minute thereby suggesting the baton may have been obtained at that time[573]. As pointed out in court the video footage shows that when D2 and D3 returned they were not seen carrying anything and there would appear to be no further assault. 763. The fact that D3 was a crime officer and not a uniform officer and had been issued with twelve retractable ASP Friction Loc Batons that day, does not cause me to doubt that D3 used a Phoenix baton to stab Tsang causing some of the circular reddish bruises. As submitted by most counsel there were many uniform police officers involved in the clearance operation. D3 could have obtained a PPCT Phoenix Baton at any time that night. Grievous bodily harm or assault occasioning actual bodily harm? 764. In R v Bollom[574] a seventeen month old baby had suffered extensive bruising and abrasions over the whole of her body. The Court of Appeal held in deciding whether injuries are grievous, an assessment has to be made of the effect of the harm on the particular individual. In determining the gravity of the injuries it was therefore necessary to consider the injuries in their real context. The court said that injuries on a six foot adult in the fullness of health would be less serious than on an elderly or unwell person, someone who was physically or psychiatrically vulnerable or on a very young child. 765. Mr Marash SC submitted Tsang being handcuffed at the back and therefore unable to defend himself was a vulnerable person[575]. I disagree. There is no evidence before the court that Tsang was not a person in the fullness of health. The Court of Appeal in my view when saying a person was physically vulnerable was not referring to a person who was handcuffed but rather to some physical disability. 766. Having carefully considered the medical evidence and the photographs of the injuries, I am not satisfied the injuries to the face; the left side of the neck; the left shoulder and clavicle; the left flank and the right flank and the circular reddish bruises on the back and chest, when viewed collectively, albeit extensive, constitute grievous bodily harm. 767. In Bollom the Court of Appeal held that it was incumbent on the judge to direct the jury that they had to be sure not only that the injuries, viewed collectively, constituted grievous bodily harm, but also that they had been inflicted by the defendant as part of one assault rather than on different occasions as part of a separate assault. 768. The Court of Appeal found the injuries on the baby constituted grievous bodily harm, however, the judge having failed to give this direction, and considering the lack of direct evidence, quashed the conviction for causing grievous bodily harm. The court went on to find that the evidence strongly supported the prosecution case that at least some of the injuries were inflicted by the appellant and substituted a conviction for assault occasioning actual bodily harm. 769. I am satisfied so I am sure that all the injuries to the face; the left side of the neck; the left shoulder and clavicle; the left flank and the right flank and the circular reddish bruises individually amount to actual bodily harm. Assault occasioning actual bodily harm is an alternative verdict open to the court[576]. 770. I am satisfied so I am sure that whilst some of the injuries may have been caused when Tsang was subdued, most of the injuries to the face; the left side of the neck; the left shoulder and clavicle; the left flank and the right flank and some of the circular reddish bruises were sustained during the assault at the substation. 771. I am satisfied so I am sure that the defendants by taking Tsang to the north side of the substation intended to assault Tsang causing him unlawful personal violence. 772. I am satisfied so I am sure the prosecution have proved all the elements of the offence of assault occasioning actual bodily harm against each defendant beyond reasonable doubt, the case against each defendant having been considered separately. The defendants are found not guilty of causing grievous bodily harm with intent and guilty of assault occasioning actual bodily harm[577]. Charge 2 – Common assault 773. In summary the prosecution case is that whilst Tsang was waiting inside room 7 of the Central Police Station, before being taken to the Police College in Wong Chuk Hang, he was slapped on the face by D5. Tsang is the only witness who has given evidence about what happened inside room 7. 774. In summary Tsang testified that after being assaulted at the substation he was frogmarched to Lung Wo Road where he boarded a car. Two of the police officers who assaulted Tsang sat on either side of him in the car. Tsang was taken to the Central Police Station where the two police officers escorted him to room 7. Tsang said that the two police officers seen on the CCTV recordings, escorting him in the police station, were the same two police officers who had sat next to him in the car. 775. Most of the time one of the police officers remained in the room with the other one standing at the doorway. One of the police officers told Tsang to put his belongings in a plastic bag. One of the items Tsang put in his bag was his mobile phone. Tsang then waited in the room. In cross-examination Tsang said the only items he put in the bag were his mobile phone and a packet of tissue paper. After he did this he was searched[578]. 776. At one stage the two officers walked out of the room for a discussion. Tsang took this opportunity to take his mobile phone out of the bag and photograph his face. Tsang then switched off his mobile phone and put the phone back in the bag. In cross-examination Tsang said he took three photographs. 777. Tsang believed the police officers may have seen him taking photographs because they asked him to give them his phone and unlock the phone. Tsang refused. The police officers believing the phone could be unlocked with fingerprints discussed whether to force Tsang to unlock the phone. Tsang responded by telling the police officers not to be so stupid because the phone was switched off and the password was needed to turn on the phone. 778. One of the police officers then used the back of his hand to slap Tsang twice, on the right side of his face. Tsang still refused to unlock his phone. In cross-examination Tsang agreed that while he was in the police station he closed his eyes and tried to get some rest. Tsang disagreed that somebody might have touched his face to wake him up when his eyes were closed, saying if that happened he would have felt that. 779. Tsang then waited in the room until he was taken to the Police College in Wong Chuk Hang. 780. Tsang identified D5 as the police officer who slapped him and D6 as the police officer who was present when he was slapped. The identification of D5 and D6 was by way of direct confrontation. This evidence has been discussed earlier when considering the case of D5 and D6 on charge 1. For the reasons already given, I admitted in evidence the direct confrontations[579]. 781. This leaves for consideration of the court whether Tsang was slapped on his face inside room 7 and if he was, whether D5 was the police officer who slapped him. Discussion 782. Mr Chung submitted that Tsang’s description of being slapped inside room 7 was inconsistent, contrary to common sense and other evidence available and was unreliable whereby the court cannot be satisfied as to what actually happened inside room 7[580]. The three photographs 783. Much of the criticisms of Tsang’s evidence revolve around his assertion that whilst in room 7 he took three photographs of his face. Mr Chung submitted that this assertion was unreasonable, the CCTV recordings showing Tsang’s hands were tied at his back when he entered the police station and when he left room 7[581]. 784. I accept Tsang’s explanation given in cross-examination when asked by Mr Chung how he managed to take the photographs if his hands were tied. Tsang explained that he asked the police officers to release the plastic ties because they were too tight and very painful. After being searched the police officers put new zip ties on him. 785. The CCTV recordings (camera 2 at 04:51:31) show D6 was holding a plastic bag when the police officers and Tsang left room 7. D6 is also seen holding the plastic bag in his left hand in camera 14 at 04:51:31 as shown in screen capture 23 attached to the prosecution’s closing submissions. 786. The CCTV recordings camera 14 between 03:43:20 and 03:43:29 and camera 12 between 03:43:40 and 03:43:50 show that neither D5 or D6 were carrying this bag when they entered the police station with Tsang. I am satisfied the CCTV recordings support Tsang’s evidence that he was told to put his belongings in a plastic bag. The time the photographs were taken 787. In cross-examination Tsang said the three photographs were taken at 4:33 whereas in re-examination when the photographs were shown to the court and the parties, the time shown on the phone was 4:18[582]. Mr Chung submitted that Tsang offered no explanation for this mistake and deliberately avoided the truth about the photographs being revealed[583]. 788. The time the photographs were taken first arose in cross-examination when Mr Chung played part of the CCTV recordings to Tsang showing two female police officers entering room 7. The Duty Officer, Woman Station Sergeant So (“WSSgt So) (PW47), whose evidence was read[584], walked past room 7 when the arrestee called to her. WSSgt So therefore entered the room together with WPC 57215. On being asked by the arrestee how his case would be handled WSSgt So replied that together with forty odd arrestees he would be escorted to a temporary detention centre for handling. 789. Before playing the CCTV recording Mr Chung referred Tsang to the Notice of Application for leave to apply for Judicial Review (Form 86)[585], in particular part A3 which concerned the allegation of assault in room 7. After reciting that Tsang had been slapped twice, paragraph 21 reads “The Applicant (Tsang) yelled in Cantonese that he needed someone urgently as he was being attacked. His request was refused. The officers declined to provide their identities.” In cross-examination Tsang agreed this was true. 790. Asked by Mr Chung if there was any consequence of his yelling, Tsang replied “No.”. Mr Chung then put that two female officers went into room 7 at some stage. Tsang said he remembered that they might have taken a look at the door but did not examine his injuries. Asked if two female officers went into room 7 as a result of his yelling, Tsang again replied “No.”. Mr Chung then played the CCTV recording to Tsang[586]. 791. Tsang said he had no recollection what the two female officers were doing inside room 7 during that time. Tsang first said he had no recollection he called them into the room and then agreed he called them into the room to ask what was going on. Tsang had no recollection he asked how his case was to be handled and that he was told together with forty odd other arrested persons he would be transferred to a temporary detention centre. 792. Mr Chung then asked Tsang if he had any recollection whether the two female police officers came into room 7 after he was slapped. Tsang replied “No.”. Mr Chung next asked Tsang: Q. “So you cannot tell whether it’s before or after the slapping?” A. “If you are only asking me whether I have any recollection as to the time, no, I don’t. But based on the time shown on the CCTV, I believe this period of time was before the occurrence of the slaps. According to the photos I took of myself, the time was 4.33, and it was afterwards that I was slapped.” 793. In re-examination Tsang said the photographs were still on his phone. The photographs were then shown to the court and the parties. After showing the photographs Mr Marash SC asked Tsang if there was a time shown when the photographs were taken. Tsang replied yes and said the time was 4:18. 794. Asked by Mr Marash SC if the earlier time of 4:33 was a mistake, Tsang replied that he was mistaken and that he believed the correct time was 4:18. Tsang explained how to access the time by clicking on the photograph once. 795. I directed copies to be made of the three photographs and further copies to be made showing the time 4:18[587]. There followed a discussion between the court and the parties as to how the copies were to be made. Mr Tang, the junior to Mr Cheng SC, applied for the soft copy of the files so as to disclose when the file containing the photographs was created. 796. Agreement was reached that after the court adjourned the phone would be taken to the Police Headquarters on Arsenal Street for hard copies of the three photographs to be made together with an exact duplicate of the soft copy of the three photographs. Tsang agreed to give his phone to the police for this purpose. I further directed that Tsang and a representative of the defence should be present when the copying was done by the police. 797. PC 8774 (PW56), whose evidence was read[588], was responsible for making the copies. After PC 8774 convinced Tsang that the data in his phone would not be leaked or accessible to a third party, Tsang agreed for his phone to be examined by using a Mobile Device Forensics tool “ACESO KIOSK”. 798. The examination was however unsuccessful due to encryption. Tsang refused to leave his phone for further forensic examination. WSgt 18819 also suggested a further examination by “ACESO KIOSK” after Tsang first removed the password. Tsang again refused because he did not want to waste more time. 799. PC 8774 then examined the phone manually and copied the photographs. Examination of the phone was completed at 20:26 hours. PC 8774 was therefore unable to confirm the three photographs were the original photographs and that the date and time shown on the photographs were accurate. 800. I have no hesitation in rejecting the submission that Tsang deliberately avoided the truth about the photographs being revealed. Tsang was in court when the defence applied for the soft copy/history of the photographs and was fully aware of what the police were going to do when he agreed to hand over his phone to the police for examination. When Tsang was convinced the data in his phone would not be leaked or accessible to a third party he allowed the police to examine his phone. 801. This in my view is not the action of a person who wants to deliberately avoid the truth about the photographs being revealed. The fact that Tsang after being in the police station for over two hours refused further examination by “ACESO KIOSK” and refused to leave his phone with the police overnight for other forensic examination does not cause me to doubt his evidence. 802. I accept the time Tsang gave in cross-examination was a mistake. Tsang was explaining that he believed he was slapped after the two female police officers entered the room. Whether the photographs were taken at 4:18 or 4:33, both times are after the two female police officers had been in the room. This discrepancy in the evidence does not cause me to doubt Tsang’s evidence that he was slapped on the face inside room 7. 803. In addition, I have compared the background of the three photographs with the photographs of room 7, exhibit P25, photographs 17-19. In evidence Mr Marash SC showed Tsang photograph 19. Tsang said he was sat on the black chair in front of the blue wall. Photograph 19 shows that on the blue wall behind Tsang are two notices, most probably informing a suspect of his rights. The notices are covered by either plastic or glass sheets which are screwed to the wall. 804. I am satisfied that the three photographs taken by Tsang are consistent with the appearance of room 7 as shown in photograph 19. A careful look at the three photographs taken by Tsang reveal that to his right is what appears to be the left corner of the plastic/glass sheet showing the top screw, which is nearest to the ceiling and the white coloured wall. This is best seen on photograph P39 (4), which is the photograph with the time 8:33 PM at the top. Failure to disclose the existence of the three photographs 805. Mr Chung submitted that Tsang’s explanations for failing to disclose the existence of the three photographs in the Judicial Review proceedings; to CAPO or to the Department of Justice, were unconvincing and unacceptable and showed that either Tsang was not telling the truth or was very calculated by withholding important evidence to use only when beneficial to him[589]. Ms Lam also made a similar submission[590]. 806. In cross-examination Tsang said he never told any police officer about taking the photographs but told his legal representatives when he first saw them in Wong Chuk Hang. Although Tsang had his phone with him he did not show the photographs to his lawyers at that time because they were busy talking about other things. Tsang said he should have shown the photographs to the lawyers the next morning. 807. Tsang did not disclose the existence of the photographs when he applied for judicial review because he believed other photographs were used which were taken by his lawyers, including showing his front and back. Tsang disagreed it was a deliberate decision not to disclose the three photographs, saying that he was not trying to hide anything. 808. Tsang further explained he did not tell CAPO because neither he nor his lawyers trusted that CAPO would carry out an investigation independently and fairly. Tsang disagreed he deliberately chose not to disclose the existence of the three photographs to the Department of Justice. 809. I accept Tsang’s explanation for not disclosing the existence of the photographs to CAPO. The distrust of CAPO by Tsang and his lawyers is made very clear in the affirmation of Ms Tanya Chan, filed in support of the application for leave to apply for judicial review[591]. 810. I accept the explanation given by Tsang for not disclosing the three photographs in the judicial review proceedings was because there were other photographs taken by his lawyers showing his front and back. In the Notice of Application for Leave to Apply for Judicial Review[592] reference is made to an urgent application for leave being filed on 16 October and that later the same day the Department of Justice asked for clearer copies of the photo exhibits showing the injuries to Tsang’s body sustained in the police beating[593]. 811. Tsang should however have disclosed the existence of the three photographs to the prosecution once a criminal investigation had been commenced. The fact Tsang did not disclose the existence of the three photographs does not however cause me to doubt his evidence that he was slapped inside room 7. Conclusion 812. I have carefully considered all the evidence and the submissions of Mr Chung, including that it is difficult to accept that Tsang had no recollection of why WSSgt So entered room 7 and no recollection other uniform officers entered the room[594]; there was hardly sufficient time for Tsang to take three photographs during the time both officers were out of the room; the duration the police officers were in the room was not enough for all the events described by Tsang to occur, including being slapped on the face; Tsang did not say he yelled after he was slapped until he was confronted with the content of the leave application for Judicial Review; and the discrepancy in Tsang’s evidence on the voir dire and the general issue as to the time he spent in room 7[595]. 813. I have also considered the submissions made by Ms Lam relating to the credibility and reliability of Tsang’s evidence of what happened in room 7, in particular that there were material differences between Tsang’s evidence and the allegations made in the Notice of Application for Leave to Apply for Judicial Review[596]. The differences in how Tsang was asked to unlock his phone are, in my view, more apparent than real. The essence is the same that because Tsang refused to unlock his phone he was slapped on his face. 814. I am satisfied so I am sure I can safely rely on the evidence of Tsang that while he was inside room 7 he was slapped on his face twice. Nothing said by Mr Chung or Ms Lam cause me to doubt that while Tsang was inside room 7 he was slapped on his face twice. Identification 815. Mr Chung submitted that no weight should be given to Tsang’s identification of D5 as the police officer who assaulted him in room 7[597]. I have already dealt with the submissions as to the time Tsang had to observe D5[598] and the effect of being sprayed with OC foam[599]. 816. For the reasons already given, I am satisfied so I am sure I can safely rely on the identification by Tsang that the two police officers who escorted him to room 7 were D5 and D6[600]. I am satisfied so I am sure I can safely rely on Tsang’s identification of D5 as the police officer who slapped him on the face. 817. I am satisfied so I am sure the prosecution have proved all the elements of the charge beyond reasonable doubt. D5 is convicted of common assault. Scanned screen captures T1-5,A1-2,ID1-12 [1] Cap 212. [2] Although called the Lung Wui Road Government Building Pump Station East Substation, the substation is located on Lung Wo Road. The north side of the substation is the opposite side to where the coaches and cars were parked on Lung Wo Road. See photographs 11-24, exhibit P24 together with the plan, exhibit P64 as read with §27 of the admitted facts (4), exhibit P14 (c). [3] See R v Lai Chi Shing [1987] HKLR 422 and HKSAR v Ma Yee Keung CACC 53/2000 §§9-13. [4] Cap 221. A list of prosecution witnesses is attached to the prosecution opening, which list was updated as the trial progressed. The evidence of one additional witness not on the list PC 8774 (PW56) was also read. This relates to charge 2. [5] §20 of the admitted facts (2), exhibit P14 (a). [6] [1997-98] 1 HKCFAR 107. [7] Grounds of Objection to the admissibility of the video footage, photographs and CCTV recordings were marked respectively ‘A (1)’, ‘A (2)’ & ‘E’ for identification. [8] List of witnesses (updated during the voir dire) was marked ‘B’ for identification. At the commencement of the voir dire an outline of the evidence was also submitted and marked ‘C’ for identification. [9] Cap 8. [10] [1990] NI 306 at page 342H. [11] [2003] 3 HKLRD 751 at §34. [12] P4 between 00:31- 00:40. [13] P1(d) between 03:40 and 03:49. [14] P5, file 00001 between 13:48-14:18; P6, file M2U00019 between 23:05-23:25; P7, file M2U00038 between 23:00-23:20; and P8, file 00001 between 06:13 – 06:57. [15] §22 of the admitted facts, exhibit P14. [16] Exhibit P3(2) - the first 18 seconds and between 01:33 and 01:41. [17] Marked ‘C’ and ‘D’ for identification. [18] See pages 3 & 5 of the submissions of D1 on the voir dire. [19] See § 4 of the submissions of D4 on the voir dire. [20] See page 14 of the submissions of D1 on the voir dire. [21] See §§ 10-32 of the submissions of D2 on the voir dire. [22] R v Murphy at 342C. [23] See §5 of the submissions of D5 on the voir dire. [24] §37 of the prosecution submissions on the voir dire. [25] Specific reference is made to the balance of probabilities in §2 of the submissions of D4 and §4 of the submissions of D7 on the voir dire. In his oral submission Mr Lok SC suggested the test was one of “commensurate degree”, which was somewhere in between. Mr Cheng SC also referred to the case of A Solicitor v The Law Society of Hong Kong (2008) 11 HKCFAR 117 where Mr Justice Bokhary said, “Only two standards of proof are known to our law. One is proof beyond reasonable doubt and the other proof on a preponderance of probability. The strength of the evidence needed to establish such a preponderance depends on the seriousness and therefore inherent improbability of the allegation to be proved.” [26] [1972] 1 WLR 651 [27] R v Robson & Harris at 653H-654A [28] R v Robson & Harris at 654C-D. [29] R v Robson & Harris at 653E. [30] Crim LR 2012 486 at 490-491. [31] 12th edition, p 184.This is also the view in Phipson on Evidence, 18th edition, p 1414 fn. 80. [32] Crim LR 2012 486 at 492. [33] In the course of submissions, the court and the parties referred to various Canadian cases including R v Nikolovski [1996] 3 SCR 1197; R v Brown [1999] OTC 213; R v Penney 2002 NFCA 15; R v Jamieson 2004 OTC 369; R v Antone 2015 BCSC 1243; and R v Bulldog 2015 ABCA 251. I considered these cases. Also see R v Mirarchi 2016 QCCS 2531. [34] See §§9-21 of the submissions of D3 on the voir dire. [35] See §11 of the submissions of D3 on the voir dire. [36] 12th edition, page 56. [37] See HKSAR v Lau Shing Chung Simon (2015) 18 HKCFAR 50 §24. [38] This was the evidence of Sgt 47574 (PW1); Sgt 34200 (PW2); SIP Fu (PW3); SSgt 52820 (PW4); SIP Wat (PW5) & Tsang (PW11). [39] See for example §§39-44 of the submissions of D2 and §§53-55 of the submissions of D3 on the voir dire. [40] R v Murphy at 342G. Specific reference is made to this passage in R v Murphy in §39 of the submissions of D2 on the voir dire. [41] R v Murphy at 342E. [42] See for example Blackstone’s Criminal Practice 2017 at F8.62. [43] Sgt 47574 (PW1); Sgt 34200 (PW2); SIP Fu (PW3); SSgt 52820 (PW4) and SIP Wat (PW5). [44] [2011] NICA 19. [45] R v Quinn at §15. Also see Phipson on Evidence, 18th edition at 41-07. [46] See Archbold UK 2017 at §4-360 p 528. [47] R v Murphy at 342F. [48] See §§39, 42-45 & 49 of the prosecution submissions on the voir dire. [49] §§3-9 of the reply submissions of D2 on the voir dire. [50] [2010] NICC 27. [51] [2011] NICA 19. [52] Sgt 47574 (PW1); Sgt 34200 (PW2); SIP Fu (PW3); SSgt 52820 (PW4) and SIP Wat (PW5). [53] Tsang and the five police officers were all recalled on the general issue. The evidence concerned, inter alia, more detail as to how Tsang was subdued and whether Tsang sustained any injuries. This evidence is discussed later, in particular when considering whether the injuries sustained by Tsang amounted to grievous bodily harm. [54] Flower bed was used interchangeably with planter. [55] File 00001 between 06:13- 06:57. Admitted in evidence see §22 of the admitted facts, exhibit P14. [56] Tape played from 31-40 seconds. [57] §22 of the admitted facts, exhibit P14. [58] SIP Lau gave evidence only on the general issue. [59] On the general issue SIP Lau said that after other colleagues came to assist in apprehending the male he went to the rear of the sweeping operation to look for colleagues of the crime department. After bringing the crime colleagues to where the male was being apprehended SIP Lau left. SIP Lau said he brought more than one crime colleague but could not be sure how many he brought. SIP Lau did not know the names or numbers of the crime colleagues [60] By Mr Cheng SC. [61] The evidence of the clothing came after Tsang had described the events that night and before he was shown the video footage. Mr Leung SC only asked Tsang to describe his upper garment. [62] When giving evidence on the general issue Tsang accepted they were police officers. [63] See §§1-8 of the admitted facts, exhibit P14. [64] See §9 of the admitted facts, exhibit P14. [65] By Mr Lok SC. [66] The server was also referred in the evidence as the computer and the computer server. [67] By Mr Cheng SC. [68] This evidence was given in re-examination. [69] By Mr Cheng SC. [70] By Mr Cheng SC. [71] By Mr Cheng SC. [72] See §9 of the admitted facts, exhibit P14. [73] By Mr Lok SC. [74] In examination-in-chief David Wong said the group was blocked by a vehicle and during cross- examination by Mr Lok SC said they were blocked by a van. [75] By Mr Lok SC. [76] The extra four seconds are seen between 13 and 17 seconds. [77] Channel 83. [78] By Mr Lok SC. [79] Tape played from beginning and stopped at 03:12. P1 (d) was downloaded from news.tvb.com. David Wong explained that news.tvb.com was the web address of TVB news. [80] The footage between 03:40-03:49 showing Tsang pouring liquid was not played to David Wong. [81] Tape played from beginning and stopped at 03:48. [82] Channels 81 & 83. [83] Channel 83. [84] Tape played from beginning and stopped at 01:07 seconds. [85] Tsang was shown the first seconds of (b), all of (c) and from 00:14-01:07 of (h). [86] See for example pages 2, 3, 4, 5, 10, 12, 14 & 15 of the submissions of D1 on the voir dire where Mr Lok SC refers to the fragmentary nature of the video footage and §38 (i)-(iii) of the final submissions of D6. [87] See pages 10-13 of the submissions of D1 on the voir dire. [88] See §§37(iii) and 54 of the submissions of D2 on the voir dire. [89] See part E of the submissions of D3 on the voir dire and §5 of the closing submissions of D3. [90] See §§10-19 of the submissions of D7 on the voir dire. [91] See §10 of the closing submissions of D4. [92] By Mr Lok SC. [93] By Mr Cheng SC. [94] The affirmation of David Wong was not shown to the court. [95] See §§37(iii) of the submissions of D2 on the voir dire. [96] See page13 of the submissions of D1 on the voir dire. [97] See §§21 & 22 of the submissions of Mr Lo on the voir dire. [98] See §§20 & 21 of the closing submissions of D2. [99] §§34-40 of the closing submissions of D3. [100] See pages 11-12 of the submissions of D1 on the voir dire. [101] See the first bullet point, page 12 of the submissions of D1 on the voir dire. [102] See between 03:10 and 03:30 which shows the lay-by, the railings parallel with the substation and the end of the railings. [103] See §15 (iv) of the submissions of D6 and §18 of the submissions of D7 on the voir dire. [104] The screen shot at 12:20 seconds also clearly shows the gap at the end of the railings. [105] See §38 The “Blocked Shot” (i)-(iii) of the final submissions of D6. [106] See §§65(iii) and 54 of the submissions of D2 on the voir dire. [107] [2002] NFCA 15. [108] See §55 of the submissions of D2 on the voir dire. [109] See §§61-69 of the submissions of D2 on the voir dire. [110] See for example §4 of the submissions of D6 on the voir dire. [111] See §§43-45 of the submissions of D3 on the voir dire. [112] See §§48 & 49 of the submissions of D3 on the voir dire. [113] (1982) 75 Cr App R 149 at 153. Cited with approval in R v Murphy at page 344B-F. [114] §29 of the judgment. [115] By Mr Lok SC. [116] David Wong referred to this as the 8 seconds from camera A. [117] In cross-examination by Mr Lok SC. [118] By Mr Cheng SC. [119] See §§71-76 of the submissions of D2 on the voir dire. Also see §17 of the submissions of D4 and §15 (vii) of the submissions of D6 on the voir dire. [120] See §37(i) of the submission of D2 on the voir dire. [121] By Mr Cheng SC. [122] In re-examination David Wong said he did not look at the live broadcast footage at the time but later when he retrieved the footage. [123] Ms Lam made this submission when explaining §15(ii) of the submission of D6 on the voir dire. [124] By Mr Cheng SC. [125] By Mr Cheng SC. [126] David Wong explained the second shot was edited for broadcastsee §§146-148. [127] See §108. [128] See for example pages 10-11 of the submissions of D1; §37 (ii) & 45-49 of the submissions of D2; section E2; E3 & E4 of the submissions of D3; §3 of the submissions of D4; §15 of the submissions of D6 and §7 & 8 of the submissions of D7 on the voir dire. [129] See §§10-14 of the admitted facts, exhibit P14. [130] See §15 of the admitted facts, exhibit P14. [131] See §§39 & 42 of the prosecution submissions on the voir dire. [132] This can also be seen in P2 (e). [133] See for example page 14 of the submissions of D1; §79 of the submissions of D2; section E2; E3 & E4 of the submissions of D3; §§3, 6 & 8 of the submissions of D4; §17 of the submissions of D6 and §24 of the submissions of D7 on the voir dire. [134] See §§16 - 19 of the admitted facts, exhibit P14. [135] See §§16 & 18 of the admitted facts, exhibit P14. [136] See §§17 & 18 of the admitted facts, exhibit P14. [137] See §44 of the prosecution submissions on the voir dire. [138] Part of the summary is taken from the answers given by Mr Chim to questions from the court at the end of his evidence clarifying themaking of the master tape. [139] By Mr Lok SC. [140] By Mr Lok SC and Mr Cheng SC. [141] See §17 of the admitted facts, exhibit P14. [142] By Mr Lok SC and Mr Cheng SC. [143] See §§72-75. [144] See for example pages 2-3 of the submissions of D1; §§81-83 of the submissions of D2; section E2; E3 & E4 of the submissions of D3; §7 of the submissions of D6 and §28 of the submissions of D7 on the voir dire. [145] By Mr Cheng SC. [146] See §§6 -10 of the submissions of D4 on the voir dire. [147] See §§27-28 of the submissions of D7 on the voir dire. [148] See §§84-86 of the submissions of D2 on the voir dire. [149] See §8 of the submissions of D6 on the voir dire and §38 (i) of the final submissions of D6. [150] See exhibit P3 (1B). [151] See exhibit P10. [152] Page 338H-339A. [153] By Mr Choy. [154] See for example page 4 of the submissions of D1; section E7 of the submissions of D2 on the voir dire and §§24-26 of the submissions of D7 on the voir dire. [155] See §§20 & 21 of the admitted facts, exhibit P14. [156] See §22 of the admitted facts, exhibit P14. [157] See §68. [158] By Mr Lok SC and Mr Choy. [159] Sgt 47574 was not cross-examined on the voir dire. [160] See for example pages 5-6 of the submissions of D1; sections E2-E5 of the submission of D3; §10 of the submissions of D4; §17 of the submissions of D6 and §§24 & 28 of the submissions of D7 on the voir dire. [161] As particularised in exhibit P17. [162] See §45 as read with §36 (3) & 36(5) of the prosecution submissions on the voir dire. [163] To illustrate submissions which made reference to the video footage the court requested screen captures be provided to show the parts referred to in the video footage. During the second day of his submissions Mr Leung SC provided a bundle of 26 screen captures. Screen captures 13, 14 & 15 are taken from the TVB footage P1 (b), showing the escorting of Tsang. [164] See §§7 & 31-32 of the closing submissions of D3. [165] See for example page 15 of the submissions of D1 on the voir dire; §§88-93 of the submissions of D2 on the voir dire; section E2; E3 & E4 of the submissions of D3 on the voir dire and §§6 & 31 of the closing submissions of D3; §§11-15 of the submissions of D4 on the voir dire; §18 of the submissions of D6 on the voir dire and §29 of the submissions of D7 on the voir dire. [166] See written grounds of objection marked “E” for identification. [167] See list of witnesses marked “B” for identification. [168] On the general issue Tsang’s complaint, exhibit P38, was produced not to prove the contents of the complaint but to show the complaint was made. [169] This evidence is further discussed when considering the case of D5 & D6. [170] Tsang Kin Chiu called the police station the Waterfront Police Station and said this was the only police station he was taken to that night. SPC 16408 (PW21) explained in his evidence that in 2010 the Central Police Station moved from Arsenal Street to the former Waterfront Police Station at No. 2 Chung Kong Road. [171] The Police Training School is also referred to in the evidence and the admitted facts as the Police College. [172] Mr Leung SC submitted by way of aide-memoire a table listing the times of the footage shown to Tsang, marked “F” for identification. [173] PW21 retired from the police force in 2016. [174] By Mr Chung. [175] By Ms Lam. [176] Sgt 14827 was not called to give evidence. [177] By Mr Chung. [178] See §§8-20 of the submissions of D5 on the voir dire. [179] See §23III of the submissions of D6 on the voir dire. [180] When the CCTV footage was played to Tsang neither disc exhibit P15 (a) or P15 (b) was used. Mr Leung SC explained when the footage was shown to Tsang physical exhibits were not used. All the video footage and CCTV recordings (exhibits P1-8 & P15) were copied onto a computer from which the particular footage can be selected and played. Similarly, the court was supplied with a computer on which all the video footage and CCTV recordings had been copied so that the court could access all the footage in chambers. With specific reference to the CCTV recordings Mr Leung SC said the copy played to Tsang was made from the working copy P15 (b). [181] By Mr Chung. [182] The difference in the evidence is referred to at §§13-15 of the submissions of D5 on the voir dire. [183] By Mr Chung. Mr Chung was the only one to cross-examine DSgt 47147. [184] See §23III (iii-vi) of the submissions of D6 on the voir dire. [185] See §37 of the submissions of D5 on the voir dire. [186] See §§46-48 of the prosecution submissions on the voir dire. [187] See §§37 & 38 of the submissions of D5 on the voir dire. [188] This suggestion was put to DSgt 45918 when Mr Chung was cross-examining DSgt 45918 about the Chinese phrase “Chau Luk” used by DSgt 45918 when making the entry at 2000 hours in the investigation report (marked “L” for identification). DSgt 45918 said the entry referred to making a record manually by pen and disagreed this meant making a copy. When Mr Chung suggested that DSgt 45918 recorded that he was reproducing a copy of the CCTV in the Central Police Station, DSgt 45918 replied that he made this record for the sake of convenience for making a record to his superior. The record was produced and marked “M” for identification. Documents “L” & “M” are discussed later in the next section titled “The content issue”. [189] See §39 of the submissions of D5 on the voir dire. [190] See §§16-20 of the submissions of D5 on the voir dire. [191] See §18 of the submissions of D5 on the voir dire. [192] See §23III (ii) of the submissions of D6 on the voir dire. [193] Cap 8. [194] Having ruled the certificate inadmissible, the certificate was returned to the prosecution. No copy was retained on the court file. [195] See §19 of the submissions of D5 on the voir dire. [196] See §23III (i) of the submissions of D6 on the voir dire. [197] See §23III (i) of the submissions of D6 on the voir dire. [198] See §246. [199] See §23III (vii) of the submissions of D6 on the voir dire. [200] See §20 of the submissions of D5 on the voir dire. [201] By Mr Chung. Also see §23III (ii) of the submissions of D6 on the voir dire. [202] See §20 of the submissions of D5 on the voir dire. [203] See §21 of the submissions of D5 on the voir dire. [204] See §§22- 35 of the submissions of D5 on the voir dire. [205] See §23II of the submissions of D6 on the voir dire. [206] The sketch P16 (a) was first shown to SPC 16408 and marked at that time as PP16. In cross-examination a sketch drawn on 28 October 2014 was shown to SPC 16408. At this stage the sketches were marked exhibits PP16 (a) and (b) according to the order shown to SPC 16408. In examination-in-chief when DSgt 45918 was shown the sketch, exhibit PP16 (b) the court noticed that the sketch shown to SPC 16408 was different to the copy submitted to court. The sketch shown to SPC 16408 showed 14 cameras whereas the copy submitted to the court and to the defence showed 12 cameras. The copy submitted to the court and the defence was marked as exhibit PP16 (b1). This explains why the sketches are marked in the reverse order that they were drawn by DSgt 45918. [207] References to cross-examination of DSgt 45918 are by Mr Chung unless otherwise stated. [208] By Mr Chung. [209] In cross-examination by both Mr Chung and Ms Lam. [210] The exhibit label was not produced to court but was shown to the defence before completion of cross-examination. [211] DSgt 50037 marked on a copy of exhibit PP16 (b1). [212] See §§26-35 of the submissions of D5 and §23II of the submissions of D6 on the voir dire. [213] See §283. [214] See §§26- 30 & 33 of the submissions of D5 on the voir dire. [215] See §23II (vi) of the submissions of D6 on the voir dire. [216] See §23II (vii) of the submissions of D6 on the voir dire. [217] See §§31-33 of the submissions of D5 and §23II (v) of the submissions of D6 on the voir dire. [218] Marked “K” for identification. [219] Marked “L” for identification. [220] The purpose being to show the court the layout of the reports only, no translations were provided. [221] Marked “M” for identification. The purpose being to show the court the layout of the document only, no translation was provided. [222] See §§31 & 32 of the submissions of D5 on the voir dire. [223] See §23II (v) of the submissions of D6 on the voir dire. [224] See §34 of the submissions of D5 on the voir dire. [225] See §35 of the submissions of D5 on the voir dire. [226] See §25 of the final submissions of D5. [227] See §§23I (iv) & 23II (ii) of the submissions of D6 on the voir dire. [228] See §23I (i) of the submissions of D6 on the voir dire. [229] See §23I (ii) of the submissions of D6 on the voir dire. [230] See §23I (iii) of the submissions of D6 on the voir dire. [231] See §23II (ix) of the submissions of D6 on the voir dire. [232] §22 of the admitted facts, exhibit P14. [233] P8, file 00001 between 06:13 – 06:57. This can also be seen between 00:31- 00:40 of the Now TV footage, exhibit P4; between 03:40 and 03:49 of the TVB footage, exhibit P1 (d) and the police videos exhibits P5, file 00001 between 13:48-14:18; P6, file M2U00019 between 23:05-23:25; and P7, file M2U00038 between 23:00-23:20. [234] Although Tsang had been convicted, an appeal against conviction was pending at the time he gave evidence. After hearing from all counsel I decided Tsang should be given a warning against self-incrimination. [235] Particulars of the charges were not given to the court. [236] See §19 of the submissions of D4 on the voir dire. [237] See §16 (i) of the submissions of D6 on the voir dire and §31 (8)(ii) of the final submissions of D6. [238] See §20 of the submissions of D4 on the voir dire and §35 of the closing submissions of D1. [239] See §13 of the prosecution opening. [240] P8, file 00001 between 06:13 – 06:57. 06:48 is the point in time when Sgt 47574 identified grabbing hold of the male. [241] See the first 18 seconds and between 01:33 and 01:44 of the ATV footage, exhibit P3(2). [242] See page 8 of the submissions of D1 on the voir dire. [243] See §22 of the submissions of D2 on the voir dire. [244] See §24 (b) of the submissions of D3 on the voir dire and §§28-33 of the closing submissions of D3. [245] See §31 (8) (i) of the final submissions of D6. [246] By Mr Choy. [247] See §22 of the submissions of D2 on the voir dire. [248] Mr Lo, on behalf of D7, also referred to this phrase in his oral submissions. [249] See §20 of the submissions of D2 on the voir dire. [250] See §29 of the submissions of D2 on the voir dire. [251] See §32 of the submissions of D2 on the voir dire and §§19, 21 & 51 of the closing submissions of D2. [252] See §53 of the closing submissions of D1; C.4 of the closing submissions of D3; §10 (3) of the closing submissions of D4; and §§ 31(2)(viii) & 31(8) of the final submissions of D6. [253] At the request of the court on the second day of submissions the prosecution provided a bundle of 23 screen captures. [254] At the end of the submission of Mr Lok SC the court played the Apple Daily footage exhibit P2 (e) and invited the parties to address the court on what was seen in the footage. [255] This is also seen in TVB footage P1 (a), (c), (e) & (f) between 00:09 and 00:18. At the end of the submission of Mr Lok SC the court also played the ATV footage exhibit P3 (2) and the TVB footage (a) and (g) and invited the parties to address the court on what was seen in the footage. [256] See witness statement of IP Law, exhibit P59. [257] See §27 of the submissions of D2 on the voir dire. [258] See §16(ii) of the submissions of D6 on the voir dire and §31(2) (i)-(iii) of the final submissions of D6. [259] See §§27 & 28 of the closing submissions of D4. [260] By Mr Cheng SC and Ms Lam. [261] Marked ‘G’ for identification. [262] By Mr Choy on the general issue. The evidence of Tsang as to whether the group changed is referred to in §§7-14 of the closing submissions of D2. [263] See §§43, 53-54 & 57 of the closing submissions of D1; §§15-18 of the closing submissions of D2; and §31(3) of the final submissions of D6. [264] See §§18-19 of the closing submissions of D2. [265] See §§5 & 41-46 of the closing submissions of D3. [266] Marked ‘G’ for identification. [267] See §28 of the submissions of D2 on the voir dire and §§22 (v) & 23 of the closing submissions of D2. [268] Also seen on P1 (a), (c) & (f) between 00:16 and 00:18; P1 (b) between 00:25 and 00:27; P1 (d) between 02:35 and 02:37; and P1 (e) between 00:21 and 00:23. [269] These actions are further discussed when considering the case of D3 and what injuries Tsang sustained in the assault at the substation. [270] See §23 of the submissions of D2 on the voir dire. [271] See ATV footage P3 (2) between 00:12 and 00:18 & 01:32 and 01:35; TVB footage P1 (b) between 00:05 and 00:013 and the Now TV footage P4 between 00:40 and 00:44. [272] Both on the voir dire and the general issue. [273] This was in cross-examination by Ms Lam. [274] The witness statement is dated 19 October 2014 and is marked ‘G’ for identification. [275] By Mr Lok SC. [276] See §§553-562. [277] See §30 of the submissions of D2 on the voir dire. [278] (1981) 145 CLR 395. [279] See §31 of the submissions of D2 on the voir dire. [280] By Mr Cheng SC on the voir dire. [281] Marked ‘G’ for identification. [282] See §16(iii)-(iv) of the submissions of D6 on the voir dire and §31(2) (iii)-(viii) of the final submissions of D6. [283] See §367. [284] See §24 of the submissions of D2 on the voir dire. [285] Marked ‘H’ for identification. [286] Marked ‘I’ for identification. [287] Marked ‘G’ for identification. [288] See §25 of the submissions of D2 on the voir dire. [289] See §26 of the submissions of D2 on the voir dire. [290] See §16 (iv) of the submissions of D6 on the voir dire. [291] See §31 (8)(i) of the final submissions of D6. [292] See page 7 of the submissions of D1 voir dire and §§38, 39 and 42 of the closing submissions of D1. [293] See §§38 & 39 of the closing submissions of D1. [294] See §§29-32 of the closing submissions of D4. [295] See §31 (1) of the final submissions of D6. [296] The witness statement of IP Wu was read, exhibit P57. IP Wu also gave evidence. [297] §12 of the admitted facts (2), exhibit P14 (a). [298] See §§23 & 24 of the Notice of Application for Leave to Apply for Judicial Review marked ‘H (1)’ for identification. [299] See §31(4) of the final submissions of D6. The difference referred to at §31(4) (ii) is addressed when considering the case of D6. [300] The witness statements of WSSgt 18076 (exhibits P55(a) & (c)) were read. [301] §13 of the admitted facts (2), exhibit P14 (a). [302] The witness statement of DSPC 50117 was read, exhibit P58. [303] See witness statement of DSPC 50117, exhibit P58 and §13 of the admitted facts (2), exhibit P14(a). [304] This evidence is referred to at §31 (5) of the final submissions of D6. [305] By Mr Choy. [306] §§14 & 15 of the admitted facts (2), exhibit P14 (a). [307] The witness statement of IP Law was read, exhibit P59. [308] By Mr Choy. [309] See §§14 & 15 of the admitted facts (2), exhibit P14 (a). [310] Part 11 2b of the Medical Examination Form, exhibit P27 (a). [311] Marked “H (1)” for identification. The application was only marked for identification during cross-examination by Mr Chung on the general issue. [312] The court was not told the outcome of that application. [313] This evidence was given in June 2016 before the September election. [314] See page 7 of the submissions of D1 on the voir dire and §43 of the closing submissions of D1. [315] This evidence was also given in June 2016 before the September election. [316] See §§33-34 of the closing submissions of D4. The reference to the police not being reliable is considered in the previous section on failure to complain. [317] See §31 (7) of the final submissions of D6. [318] See §31 (2) (v) & (vii) of the final submissions of D6. [319] [1976] 3 WLR 445. [320] Turnbull at 229H-230D. [321] Turnbull at 230F. [322] See R v Murphy [1990] NI 306 at 326E & 329B. [323] [2003] 1 Cr App R 21. [324] See §6 of the admitted facts (2), exhibit P14 (a). [325] See §3 of the admitted facts (2), exhibit P14 (a). [326] See §3 of the admitted facts (2), exhibit P14 (a) as read with the witness statements of DPC 8097 (exhibit P48(a)) and DPC 5840 (exhibit P49 (c)). The rank and/or service numbers of the defendants are admitted. See §1 of the admitted facts (2), exhibit P14 (a). [327] See §3 of the admitted facts (2), exhibit P14 (a) as read with the witness statement of DSgt Ma, exhibit P50 (a). The rank and service number of D4 PC 1879 was admitted. See §1 of the admitted facts (2), exhibit P14 (a). [328] See §6 of the witness statement of DSgt Ma, exhibit P50 (a). [329] WDPC 3209 (PW33), exhibit P51; DPC 2235 (PW34), exhibit P52; SPC 1639 (PW35), exhibit P53 and WPC 3087 (PW36), exhibit P54. [330] By Mr Lok SC. [331] D1–SSP Chan (PW37); D2 & D3–CIP Chung (PW38); D5–WSIP Wu (PW39); and D6–IP Wong (PW40). [332] See admitted facts (3), exhibit P14 (b). [333] Exhibits P46 (a) & (b) as read with §§A2 & B14 of the admitted facts, exhibit P14. [334] This is called step forward on the CCTV. [335] Also see §2 of the admitted facts (2), exhibit P14 (a). [336] WCIP Hung (PW41) was not called to give evidence. By agreement of all parties her nine witness statements were not required to be read into evidence. [337] See §1 of the admitted facts (3), exhibit P14 (b) for particulars of the footage shown to SSP Chan. [338] Exhibits P42 (a) & (b). Exhibit P42 (a) shows the time as 00:24. This is seen in the second part of the ATV footage P3 (2) between 01:33 and 01:48. Exhibit P42 (b) shows the time as 00:05. This is seen in the TVB footage P1 (g) between 02:47 and 02:56. The same footage is also seen in TVB footage P1 (b), (d) and (h). [339] Exhibit P42 (c). [340] See §182. [341] This is also seen in the TVB footage P1 (c), (d), (e), (f) & (g) however the quality of (d) is not good. [342] See §57 of the closing submissions of D1. [343] See for example P1 (g) between 03:00 - 03:15. All the TVB footage shows the group arriving at the substation. [344] See §§53 & 54 of the closing submissions of D1. [345] Also see §3 of the admitted facts (2), exhibit P14 (a). [346] See witness statement of CIP Chung, exhibit P43 (a) as read with §1 of the admitted facts (4), exhibit P14(c). [347] See §1 of the admitted facts (3), exhibit P14 (b) for particulars of the footage shown to CIP Chung. [348] Annex A to the witness statement of CIP Chung, exhibit P43 (a). The screen capture shows the time as 00:01. This is seen in the TVB footage P1 (g) at 02:47. The same footage is also seen in TVB footage P1 (b), (d) and (h). [349] See §2 of the admitted facts (3), exhibit P14 (b) for particulars of the footage shown to CIP Chung. [350] See witness statement of CIP Chung, exhibit P43 (b). [351] See witness statement of CIP Chung, exhibit P43 (c). [352] See §28 of the closing submissions of D2. [353] See §182. [354] This is also seen in the TVB footage P1 (c), (d), (e), (f) & (g), however, the quality of (d) is not good. [355] See P1(g) between 03:00 - 03:19. All the TVB footage shows the group arriving at the substation. [356] See §34 (ii) of the closing submissions of D2. [357] The other person is D3 see §491. [358] This is also seen in the TVB footage P1 (a), (c) & (f) between 01:15 – 01:32. [359] See §§25-36 & 57-61 of the closing submissions of D2. [360] Also see §3 of the admitted facts (2), exhibit P14 (a). [361] See witness statement of CIP Chung, exhibit P43 (a) as read with §1 of the admitted facts (4), exhibit P14(c). [362] Annex A to the witness statement of CIP Chung, exhibit P43 (a). The screen capture shows the time as 00:01. This is seen in the TVB footage P1 (g) at 02:47. The same footage is also seen in TVB footage P1 (b), (d) and (h). [363] Annex A to the witness statement of CIP Chung, exhibit P43 (b). [364] See witness statement of CIP Chung, exhibit P43 (b). [365] Annex A to the witness statement of CIP Chung, exhibit P43 (c). [366] See §182. [367] This is also seen in the TVB footage P1 (a), (c), (d), (e) & (f), however, the quality of (d) & (e) are not good. [368] P1 (g) and P1(h) show the extra four seconds when the group arrive at the substation. P1 (g) and P1(h) are clearer than P1 (a), however, they do not show the full assault footage. The actions of the person leading the group are best followed by viewing P1 (a) together with (g) and (h). [369] This is discussed later at §§730-732 when considering what injuries Tsang sustained in the assault at the substation. [370] See §§8-15 & 21-27 of the closing submissions of D3. [371] See §3 of the admitted facts (2), exhibit P14 (a) as read with the witness statement of DSgt Ma, exhibit P50 (a). The rank and service number of D4 PC 1879 was admitted. See §1 of the admitted facts (2), exhibit P14 (a). [372] See §6 of the witness statement of DSgt Ma, exhibit P50 (a). [373] See §3 of the witness statement of CIP Chan, exhibit P46 (a). [374] See §§16-20 of the closing submissions of D4. [375] See §3 of Annex IV of the prosecution’s closing submissions. [376] See §4 of Annex IV of the prosecution’s closing submissions. [377] The screen capture is taken from exhibit P1 (h) which shows the same footage as P1 (b). [378] See §4 of Annex IV of the prosecution’s closing submissions. [379] See §5 (a) of Annex IV of the prosecution’s closing submissions. [380] See §182. [381] This is also seen in the TVB footage P1 (c), (d), (e), (f) & (g). The quality of P1 (d) is not very good. [382] All the TVB footage shows the arrival of the group at the substation. [383] TVB footage P1 (g). Also seen on P1 (a), (c), (f) & (h) and Apple Daily footage P2 (a), (d) & (e). [384] See §§12-15 & 21-24 of the closing submissions of D4. [385] Marked for identification ‘N’ & ‘O’ respectively. [386] Superintendent So was tendered for cross-examination. [387] See letter from the police to Vidler & Co, exhibit D6(4). [388] See §§23 & 24 of the admitted facts (2), exhibit P14 (a). [389] See the brief facts attached to the identification parade books, exhibits D5(4) and D6(6) and the explanations given during the confrontations by CIP Tse at counters 5 & 17 of the transcript, exhibit P31C and counters 3 & 17 of the transcript, exhibit P32C. [390] See counter 17 of the transcripts, exhibit P31C & P32C. [391] See counters 49-51 of the transcript, exhibit P31C. [392] See counters 25-28 of the transcript, exhibit P32C. [393] By Ms Lam. [394] Exhibit D5(3). [395] See §§3-11 of chapter 46-17. [396] By Ms Lam. [397] This letter was not produced. [398] Exhibit P33. [399] Exhibits P34 & P35. [400] See §1 of the prosecution submissions on admissibility of confrontation evidence. [401] See §§ 5-16 of the submissions of D5 and §§11-14 of the submissions of D6 on admissibility of confrontation evidence. [402] Exhibit D6(1). [403] This letter was not produced. [404] See letters exhibits P33; D6(1), D6(2) and D6(3). In cross-examination by Ms Lam SP So agreed in the letter dated 7 November mention was also made of group identification. This letter was not produced. [405] Exhibit D6(2). [406] See §§ 7-9 & 15-16 of the submissions of D5 and §12 of the submissions of D6 on admissibility of confrontation evidence. [407] SP So repeated the same explanation during cross-examination by Ms Lam. [408] [2001] 3 HKLRD 274 at 280I. [409] Lo Ho Chung at 281C. [410] Exhibits P30 (a) & P30 (b). See §22 of the admitted facts (2), exhibit P14(a). [411] See counters 7-9 of the transcript, exhibit P31C and counters 5-7 of the transcript, exhibit P32C. [412] See counter 17 of the transcripts, exhibit P31C & P32C. [413] See §§ 20-21 of the submissions of D5 on admissibility of confrontation evidence. [414] See counters 20-23 of the transcript, exhibit P31C. [415] Mr Leung SC and Mr Chung agreed the person F in the transcript was Mr Vidler. [416] See counters 39-42 of the transcript, exhibit P31C. [417] See §§ 15-21 of the submissions of D6 on admissibility of confrontation evidence. [418] See §19 (iii) of the submissions of D6 on admissibility of confrontation evidence. [419] See §20(i) of the submission of D6 on admissibility of confrontation evidence and §38 of chapter 46-17 Force Procedures Manual on visual identification of suspects. [420] See §9 of the prosecution submissions on admissibility of confrontation evidence. [421] See §20(ii) of the submissions of D6 on admissibility of confrontation evidence. [422] Examination-in-chief of CIP Tse. [423] See §§ 17-19 of the submissions of D5 and §§22-23 of the submissions of D6 on admissibility of confrontation evidence. [424] By Mr Chung. [425] Exhibits D5 (1) & D5(2). [426] See §17 of the submissions of D6 on admissibility of confrontation evidence. [427] [1990] 1 HKC 171. [428] HCMA 1093/2006. [429] See §19 of the written submission of Mr Chung on admissibility of confrontation evidence. [430] For example, see the TVB footage, P1 (a) showing both the arrival at the substation and Tsang leaving together with the map, exhibit P64 showing the location of the substation and Lung Wo Road. [431] Tsang called the police station the Waterfront Police Station and said this was the only police station he was taken to that night. SPC 16408 (PW21) explained that in 2010 the Central Police Station moved from Arsenal Street to the former Waterfront Police Station at No. 2 Chung Kong Road. [432] See witness statements of WDPC 3209 and DPC 2235, exhibits P51 & P52 respectively. [433] See §11 of the admitted facts (2), exhibit P14 (a). [434] The times of the footage are shown in the table marked “F” for identification. [435] The Police College was sometimes translated as the Police Training School. [436] See §29 of the final submissions of D5. [437] See §32 (i) of the final submissions of D6. [438] See §§364-368. [439] See §35 of the final submissions of D5. [440] The time of the assault is further discussed at §§669-675when considering the case of D7. [441] The evidence of WSSgt So was read, exhibit P55 (a). In a second statement exhibit P55 (b) WSSgt So did give the name of Tsang but not the names, ranks or service numbers of the plainclothes officers. [442] See §32 (viii) of the final submissions of D6. [443] See §32 (iv)-(vi) of the final submissions of D6. [444] See §§545-546. [445] In re-examination Mr Marash SC asked Tsang if D6 opened his eyes on the second viewing. Tsang replied no. As pointed out and having viewed the video again, apart from looking as he walked out of the room, Tsang did not in fact walk over to identify D6 again. [446] Also see §3 of the admitted facts (2), exhibit P14 (a). [447] See witness statement of WSI Wu, exhibit P44 (a) as read with §3 of the admitted facts (4), exhibit P14(c). [448] See §2 of the admitted facts (3), exhibit P14 (b) for particulars of the footage shown to WSI Wu. [449] Annex A to the witness statement of WSI Wu, exhibit P44 (a). [450] See witness statement of WSI Wu, exhibit P44 (a). [451] Annex B to the witness statement of WSI Wu, exhibit P44 (b). [452] See §3 of Annex V of the prosecution’s closing submissions. This part is also captured in the ATV footage exhibit P3 (2) between 01:33 and 01:34. The same footage is seen in P3 (1) at 00:57-00:59 and 02:15 -02:16. [453] See §6 of Annex V of the prosecution’s closing submissions. [454] Also see §3 of the admitted facts (2), exhibit P14 (a). [455] See witness statement of IP Wong, exhibit P45 (a). [456] See §1 of the admitted facts (3), exhibit P14 (b) for particulars of the footage shown to IP Wong. [457] Annex A to the witness statement of IP Wong, exhibit P45 (a). The screen capture shows the time as 00:22. This is seen in the ATV footage P3 (2) at 01:35. [458] Annex B to the witness statement of IP Wong, exhibit P45 (b). [459] See §4 of Annex VI of the prosecution’s closing submissions. [460] This is also seen in the TVB footage P1 (b), (d) & (h), however, the quality of (d) is not as good. [461] See §7 of Annex VI of the prosecution’s closing submissions. [462] I am satisfied the moving photograph was taken at the stage Tsang was carried face down. See §182. [463] See §7 of Annex V and §8 of Annex VI of the prosecution’s closing submissions. [464] This is also seen in the TVB footage P1 (c), (d), (e), (f) & (g), however, the quality of (d) is not good. [465] See P1(g) between 03:00 - 03:19. All the TVB footage shows the group arriving at the substation. [466] See §§33-37 of the final submissions of D5. [467] See §12 of the admitted facts (2), exhibit P14 (a). [468] See §13 of the admitted facts (2), exhibit P14 (a). [469] See §1 of the admitted facts (2), exhibit P14 (a). [470] See §31(4) (ii) & (iii) of the final submissions of D6. [471] Form 86 marked ‘H(1)’ for identification [472] DSgt 51344 and PC 9765 referred to the detainees as arrested persons. [473] See §§32 (ii), 32 (iii), 34 and 40 of the final submissions of D6. [474] See for example P1 (g) between 03:00 and 03:19. All the TVB footage shows the arrival of the group at the substation. [475] See §3 of the witness statement of CIP Chan, exhibit P46 (a). [476] See §6 of the admitted facts (2), exhibit P14 (a). [477] Evidence of SP Ng (PW27). [478] See §3 of the admitted facts (2), exhibit P14 (a). [479] See §6 of the witness statement of DSgt Ma, exhibit P50 (a). [480] Exhibits P48 & P49. [481] See witness statements of DPC 5840, exhibits P49 (a) & (c). [482] See witness statements of DPC 8097, exhibits P48 (a) & (c). [483] Also see witness statement of SPC 1639 (PW35) the driver of LZ 4923, exhibit P53 (a). [484] The evidence relating to glasses is set out in §§28-36 of the closing submissions of D7. [485] Exhibit P21(7)(a). [486] See witness statement of DPC 8097, exhibits P48 (c). [487] See witness statement of DPC 5840, exhibits P49 (e). [488] See witness statement of SGT 58332, exhibits P47 (a) as read with §5 of the admitted facts (4), exhibit P14(c). [489] See witness statements of DPC 8097, exhibits P48 (a) & (c). This evidence is referred to at §§25-27 of the closing submissions of D7. [490] See witness statement of SPC 1639 (PW35) the driver of LZ 4923, exhibit P53 (b) as read with exhibit P64 and §27 of the admitted facts (4), exhibit P14 (c). [491] See witness statements of DPC 8097, exhibits P48 (a) & (c) (answer 3). [492] See §10 of the prosecution’s supplemental closing submissions. [493] See §§37-41 of the closing submissions of D7. [494] See §§8 & 37 of the closing submissions of D7. [495] By Mr Cheng. [496] See §158. [497] This was discussed at §579 when considering the case of D5. [498] The evidence of WSSgt So was read, exhibit P55 (a). In a second statement exhibit P55 (b) WSSgt So did give the name of Tsang but not the names, ranks or service numbers of the plainclothes officers. [499] This is just before D3 is seen stabbing Tsang. [500] See §§20-24; 28-36; and 44 of the closing submissions of D7. [501] See §19 of the prosecution opening; §§22 & 24 of the prosecution’s closing submissions and §§12 & 13 of the prosecution’s supplemental closing submissions. [502] See §12 of the prosecution’s supplemental closing submissions. [503] See §§431-435. [504] By Mr Lok SC. [505] WDPC 3209 (PW33), exhibit P51; DPC 2235 (PW34), exhibit P52; SPC 1639 (PW35), exhibit P53 and WPC 3087 (PW36), exhibit P54. [506] See §§42-52 of the closing submissions of D2. [507] This evidence is also referred to in §45 of the closing submissions of D3; §34 of the final submissions of D5; and §38 The Blocked Shot (iii) of the final submissions of D6. [508] Witness statements marked exhibits P49; P60 & P61 respectively. [509] See §§38-41; 53-55 and 66 -69 of the closing submissions of D2. [510] See §§465-475. [511] See §§40 & 53 of the closing submissions of D2. [512] See §55 of the closing submissions of D2. [513] See §§465-475. [514] This is also seen in the TVB footage P1 (c), (e) and (f). The quality of P1 (e) is not good. [515] See §§55-56 of the closing submissions of D2. [516] See R v Forman and Ford [1988] Crim L R 677 cited in HKSAR v Chan Kam Shing FACC 5/2016 at §27. R v Forman and Ford is valuable for the commentary of the late Professor Sir John Smith. [517] This is discussed at §§730-732 when considering what injuries Tsang sustained in the assault at the substation. [518] See §§14 & 15 of the admitted facts (2), exhibit P14 (a). [519] See Part II (b) of the medical examination form, exhibit P27 (a). [520] See §§16 & 17 of the admitted facts (2), exhibit P14 (a). [521] See §36 of the prosecution opening and §§19 & 23 of the prosecution’s closing submissions. [522] See §47 of the closing submissions of D1; and §60 of the closing submissions of D3. [523] See §§68-89. [524] See §§19-31 of the closing submissions of D1; §§51-58 of the closing submissions of D3 and §§6-23 of the final submissions of D6. [525] Evidence of Sgt 34200 (voir dire and cross-examination by Mr Cheng SC on the general issue) and SSgt 52820 (cross-examination by Mr Cheng SC on the general issue). [526] By Mr Lok SC and Mr Cheng SC. [527] By Mr Lok SC. [528] By Mr Cheng SC. [529] See §§325-328. [530] By Mr Cheng SC. [531] By Mr Cheng SC. [532] By Mr Cheng SC. [533] By Mr Cheng SC. [534] §3(ii) & (iv) of Dr Lai’s report. [535] §3(x) of Dr Lai’s report. [536] §3(xi) & (xiv) of Dr Lai’s report. [537] The photographs show the left hand which is consistent with the medical examination form. [538] §3(xii) of Dr Lai’s report. [539] §3(xii) of Dr Lai’s report. [540] §3(i), (iii), (v), (vii) & (viii) as read with paragraph 5 (i) of Dr Lai’s report. [541] §3(vi) & (ix) as read with paragraphs 4 & 5 (ii) of Dr Lai’s report. [542] §5 of the admitted facts (2), exhibit P14 (a). [543] In cross-examination by Mr Cheng SC. [544] By Mr Lok SC. [545] See §2 of the admitted facts (2), exhibit P14 (a). [546] See §3 of the admitted facts (2), exhibit P14 (a) for the attachment of each defendant. [547] See §3 of the admitted facts (2), exhibit P14 (a). [548] §5 of the admitted facts (2), exhibit P14 (a). [549] See witness statements of DPC 8097, exhibit P48 (a) and DPC 5840, exhibit P49 (c). [550] Oral submissions of D1 and D2 (oral submission of D2 made by Mr Tang); §62 of the closing submissions of D3; §§35-41 of the closing submissions of D4 and §26 of the final submissions of D6. [551] See §§22 & 23 of the closing submissions of D2. [552] See §§60-66 of the closing submissions of D3. [553] This evidence has been referred to earlier at §§361-362, 490 & 700. [554] This is seen on all the TVB footage: P1 (a), (c) & (f) between 00:16 and 00:18; P1(b) between 00:25 and 00:27; P1 (d) between 02:35 and 02:37; and P1 (e) between 00:21 and 00:23. [555] Sgt 34200 was in cross-examination by Mr Lok SC. [556] By Mr Lok SC. [557] See §28 of the admitted facts (4), exhibit P14 (c). [558] §5 (iii) of Dr Lai’s supplemental report. [559] §§7 & 8 of Dr Lai’s supplemental report. [560] By Mr Cheng SC. [561] By Mr Cheng SC. [562] By Mr Cheng SC. [563] By Mr Lok 		SC. [564] See §24 of the final submissions of D6. [565] In cross-examination by Mr Lok SC. [566] See §28 of the final submissions of D6. [567] By Mr Lok SC. [568] By Mr Cheng SC. [569] By Mr Lok SC. [570] By Mr Lok SC. [571] By Mr Lok SC. [572] By Mr Lok SC. [573] §14 of the prosecution’s closing submissions. [574] [2004] 2 Cr App R 50. [575] §23 of the prosecution’s closing submissions. [576] See section 51(2) of the Criminal Procedure Ordinance, Cap 221 and §§4-298-299 & 20-198 Archbold Hong Kong 2017. [577] Contrary to Common Law and punishable under section 39 of the Offences Against the Person Ordinance, Cap 212. [578] All the cross-examination of Tsang about what happened inside room 7 was by Mr Chung. [579] See §§511-590. [580] See §5 of the final submissions of D5. [581] See §10 of the final submissions of D5. [582] Exhibit P39. [583] See §§11, 12 & 22 of the final submissions of D5. [584] Exhibit P55. [585] Marked ‘H (1)’ for identification. [586] Mr Chung played camera 2 between 03:58:24 and 04:02:02 (which is contained in two files). [587] Exhibit P39. [588] Exhibit P65. [589] See §§13-15 of the final submissions of D5. [590] See §31 (6) (iv) of the final submissions of D6. [591] The affirmation was marked “I” for identification during the cross-examination of Tsang on the voir dire by Mr Cheng SC. [592] The Notice was marked ‘H (1)’ for identification during the cross-examination of Tsang on the general issue by Mr Chung. [593] See §§31-36 & 39 of the affirmation of Ms Tanya Chan, marked ‘I’ for identification. [594] In cross-examination Mr Chung also played part of the CCTV recordings showing a uniform police officer taking a chair out of room 7. It is not surprising Tsang did not remember this. [595] See §§16-25 of the final submissions of D5. §25 has been addressed earlier at §§314-315 when considering the case of D5 on charge 1. [596] See §31 (6) of the final submissions of D6. [597] See §§26-31 of the final submissions of D5. [598] See §§579-584. [599] See §§574-576. [600] See §§563-590. 
Chief Justice Ma: A INTRODUCTION 1. The plaintiff (the respondent in the present appeal[1]) is a developer and wishes to redevelop five Lots of land at Nam Kok Road in Kowloon City (“the five Lots”). There stands a house on each Lot. The plaintiff wishes to redevelop the five Lots (which are adjacent to one another) by building a 26 storey composite building[2] straddling across all five Lots. The central question in this appeal is whether the relevant Government Leases under which each of the five Lots are held (“the Government Leases”)[3] permit the construction of the proposed building. The issue before the court (as it was before all the courts below) is one of contractual construction. The practical effect of the plaintiff not being permitted to build the composite building under the Government Leases is that they may be required to pay a premium for the privilege. 2. Before identifying the relevant issues to be determined, I should first set out the relevant facts and the relevant contractual documentation. 3. By Conditions of Exchange dated 4 January 1952 (“the Conditions of Exchange”), the then owner of a property on Kennedy Road in Hong Kong surrendered this land to the government in exchange for two Lots of land at Nam Kok Road in Kowloon City on the terms set out in that document. These two Lots[4] were identified as NKIL Nos. 3665 and 3666.[5] As we shall presently see, it was anticipated under the Conditions of Exchange that the lessee[6] would build on the Mother Lots. 4. The Conditions of Exchange contained the following terms:- (1) The General Conditions of Exchange (a) By clause 7(a) thereof, the lessee of the Lots was to build within a period of 18 months from being given possession “one or more good and permanent buildings” on the Lots, spending it least $400,000 for Lot No. 3665 and $160,000 on Lot No. 3666. (b) Where more than one building was built on the Lots, the lessee could be required to take up separate leases in relation to the site of each building and would have to pay a “prescribed fee”[7] for each additional lease required to be taken up (clause 6(b) thereof). It was by virtue of this clause that the five Lots became subject to five individual Government Leases. (2) The Special Conditions (a) Clause (3) restricted the user of the two Mother Lots. It stated :- “The new Lots shall not be used for industrial purposes and no factory building shall be erected thereon”. (b) Clause (6) stated:- “Not more than 20 houses shall be erected on NKIL No. 3665 and not more than eight houses on NKIL No. 3666.” 5. I highlight some aspects of the Conditions of Exchange:- (1) By General Condition 7(a), the lessee was able to build one or more buildings on the Mother Lots. (2) The only restrictions on the buildings to be built were two-fold: first, any building could not be used for industrial purposes, therefore no factory building could be built (Special Condition (3)); and secondly, in the event houses were built, no more than 20 houses could be built on Lot No. 3665 and no more than eight houses could be built on Lot No. 3666 (Special Condition (6)). (3) The Conditions of Exchange use the terms “building” (or “buildings”) and “houses”. In their context, it is difficult to see how they can be used interchangeably. A house may be a type of building, but not all buildings must be houses. 6. As recorded by the relevant Occupation Permits, in about October 1952, 20 houses were erected on Mother Lot No. 3665.[8] These houses, described in the two Occupation Permits as “Chinese type houses”, were identical in design each having five storeys: a ground floor, a cockloft, three floors and a flat roof. The 20 individual, sub-divided Lots, on which the houses stood, each occupied areas of about 1,143 sq. ft. (it will be recalled that the total area of Mother Lot No. 3665 was 22,886 sq. ft.). 7. These houses having been erected in October 1952, it was not, however, until the 1960s that, pursuant to General Condition 6(b) of the Conditions of Exchange[9], individual Government Leases[10] were entered into in relation to each of the Lots on which the houses stood. Accordingly, Mother Lot No. 3665 became sub-divided into 20 separate Lots, each with a house. The five Lots (the subject matter of this appeal) which became governed by Government Leases created on various dates in the 1960s, are:- (1) NKIL No. 3665 (Government Lease dated 11 September 1965).[11] (2) NKIL No. 4036 (Government Lease dated 14 October 1965). (3) NKIL No. 4037 (Government Lease dated 2 October 1965). (4) NKIL No. 4038 (Government Lease dated 2 February 1966). (5) NKIL No. 4039 (Government Lease dated 9 August 1965). 8. The construction of the terms of these Government Leases is crucial to the resolution of the central question in these appeals. The relevant terms are those restricting user and the right to erect houses. The parties have referred to these terms as restrictive covenants. The effect of these restrictive covenants is the same although the wording of them differs in the following way:- (1) In the Government Leases for Lot Nos 3665, 4036, 4038 and 4039, it is stated:- “… AND will not use or allow to be used the demised premises for industrial purposes and will not erect or allow to be erected thereon any factory building AND will not erect or allow to be erected more than one house on the demised premises …” (2) In the Government Lease for Lot No. 4037, the covenant states:- “… AND that the said Lessee will not use or allow to be used the said piece or parcel of ground or any part thereof or any building erected thereon or any part of such building for any purpose other than for non-industrial purposes and will not erect or allow to be erected any factory building on the said piece or parcel of ground AND will not erect or allow to be erected more than one house on the said piece or parcel of ground …” 9. Since 2006, the lessee[12] of the five Lots has wanted to redevelop and for this purpose appointed the plaintiff as the authorized developer. Building plans for the redevelopment being principally the erection of the composite building earlier described were duly submitted to the Lands Department. They were rejected. The position of the Director of Lands (the defendant in the present proceedings and the appellant in the present appeal) was that the proposed redevelopment breached the restrictive covenants set out in para 8 above. The plaintiff disagrees. 10. It is this point that separates the parties and which has led to the present proceedings. B THE ISSUES 11. The parties’ respective stances can be simply stated:- (1) The plaintiff maintains there is no bar to the proposed redevelopment. It is said by Mr Edward Chan SC[13] that nothing in the Government Leases prohibits the building of a 26 storey composite building which straddles across the five Lots. The only prohibition is against industrial user, the building of a factory and, if houses were to be built, building more than one house on each Lot. If the proposed building was a house for the purposes of the restrictive covenants, it straddled all five Lots so that only a part of a house would be on each Lot. Therefore, there would be no breach of the restriction on building more than one house. Alternatively, if the proposed building could not be defined as a house, then, providing it was to be used for non-industrial purposes and was not a factory, then the restrictive covenants had no application either since they only applied to houses; no restrictions were placed on the erection of any other type of building. Mr Chan gave the example in this context that his client could freely build swimming pools, hospitals or schools on the Lots. In support, Mr Chan referred to the Conditions of Exchange[14] and made the point that nothing in those conditions would have prevented the erection of a building such as the proposed building (which could have straddled across the whole of the Mother Lots), the only prohibition being against industrial user, the building of factories and building (on the Mother Lot No. 3665) more than 20 houses. He made the point that nothing in the Conditions of Exchange (referring I assume to General Condition 6(b)) required the imposition of more onerous terms on the lessee in the event any new leases were entered into, than was the case under the Conditions of Exchange themselves.[15] (2) The defendant, on the other hand, submits that the word “house” in the restrictive covenants ought to be given what counsel[16] described as an ordinary and natural meaning. Thus construed, the proposed building was not a house. It was also contended that the effect of the restrictive covenants was that the only redevelopment that could take place on each of the five Lots was the erection of a house, no more and no less. Nothing in the Government Leases permitted the erection of any building other than a house; it would be absurd for restrictions to be placed on the building of houses, yet no restrictions at all on the erection of any other type of building. Further, in any event, nothing in the Government Leases permitted the erection of a single building straddling the five Lots. Finally, it was contended that in the event the restrictive covenants could be regarded as being ambiguous, the court should apply the principle that in any grant made by the Government, covenants should be construed strictly against the grantee: reliance was place on Feather v The Queen.[17] 12. In the Court of First Instance, Deputy Judge Pow SC concluded that the building to be erected under the proposed redevelopment was not a house. He was of the view that the definition of the word “house” in the Government Leases had to be considered in the light of what that term meant at the time the Leases were created. At that time, the learned Deputy Judge observed, houses on the five Lots were already in existence. As he said, “The plaintiff submitted that the building standing on each Lot in 1965 must have been considered as a ‘house’. This must be right.”[18] After examining the plans for the proposed redevelopment, the Judge concluded that it would be a “quantum leap” for the proposed building to be described as a house. He accordingly found in favour of the defendant. 13. The Court of Appeal reversed the Judge’s decision and found in favour of the plaintiff, thus allowing the appeal. Like the judgment of the learned Deputy Judge, the judgment of the Court of Appeal[19] analyzed a large number of authorities (of both the Hong Kong and English courts), although the conclusion reached was quite different. In arriving at the view that the proposed composite building did constitute a “house” under the Government Leases, the Court of Appeal relied much on the effect of General Condition 7(a) and Special Condition (6) of the original Conditions of Exchange. Under those conditions, the original lessee of the Mother Lots could have built a building which occupied the whole of the Mother Lots. It therefore followed that a large composite building could come within the definition of house; in context, the term “house” was synonymous with the term “building”. Straddling caused no difficulties because under the original Conditions of Exchange, a building or buildings could have occupied the whole of the Mother Lots. 14. It is clear from the way the parties and the courts below have approached the case that it becomes necessary in this appeal principally to construe the meaning of the word “house” in the restrictive covenants and to ascertain the intention behind the restrictive covenants. C CONSTRUCTION OF THE RESTRICTIVE COVENANTS AND APPLICATION TO THE FACTS 15. We have been referred to the very well-known statement of principle regarding the construction of contracts contained in the speech of Lord Hoffmann in Investors Compensation Scheme Ltd v West Bromwich Building Society[20], to which can be added the judgment also of Lord Hoffmann NPJ in Jumbo King Ltd v Faithful Properties Ltd[21]. What emerges from these cases – and other authorities on contractual interpretation – is the overall importance of context when construing contractual terms. The statements of principle in Investors Compensation Scheme and in Jumbo King refer time and again to the relevant background against which the relevant contract and contractual terms must be viewed. It is in my view not particularly helpful in most cases to refer to the “ordinary and natural meaning” of words because, as very often experience tells us, there can be much debate over exactly what is the ordinary or natural meaning of words. The surer guide to interpretation is context. Here, I would just add that in the area of statutory and constitutional interpretation, it is context that is key; context is the starting point (together with purpose) rather than looking at what may be the natural and ordinary meaning of words.[22] 16. The word “house” is a good example of where there is no ordinary or natural meaning; at least this is open to much debate. The number of authorities referred to in the judgments below and in the parties’ written Cases before us amply demonstrate this point and, for my part, very little assistance can be derived from these authorities. This Court has already stated that a “search for a free-standing meaning of the word house, valid for all time in all circumstances, is fruitless.”[23] Reference was made in Wah Yick to the judgment of Lawrence J in Annicola Investments Ltd v Minister of Housing and Local Government[24] where he said the word had a “distinct fluidity of meaning, and that it is best construed in relation to the context in which it is found.” 17. In context, it is clear that the meaning of the word “house” under the Government Leases must have reference to those characteristics of the houses which were actually standing[25] at the time the Government Leases were entered into. In this respect, I agree with the learned Deputy Judge. Unlike in the case of the Conditions of Exchange when that document came into existence (when there were no buildings on the Mother Lots that could be described as houses), houses were actually standing on each of the individual, sub-divided Lots at the time the Government Leases were created. In the context of those Leases, the meaning of the word “house” must be taken to mean the type of house existing on the individual Lots. It is significant that the word “house” should be used, rather than the word “building”. In the Conditions of Exchange, as seen earlier, both words are used. They have different meanings. General Condition 7(a)[26] allowed the lessee to build one or more “buildings” on the Mother Lots but if “houses” were to be built, then under Special Condition (6)[27] , this was limited to 20 houses (on the former Mother Lot No. 3665) and eight houses (on the former Mother Lot. 3666). 18. I have not lost sight of the user part of the restrictive covenants. The argument here was that since there was an express prohibition in the restrictive covenants to the erection of a factory building, it therefore followed that any other type of building[28] could be erected. There are two answers to this construction: first, it ignores the meaning of the word “house” in context; secondly, the reference to “factory building” in the restrictive covenants is really to emphasize the prohibition against industrial use. 19. Construed in this way, it is clear that the building that is proposed to be built on the five Lots cannot be regarded as a house for the purposes of the restrictive covenants. 20. What then of the plaintiff’s point that the restrictive covenants only contained a restriction as to the number of houses that could be built (no more than one on each Lot) but had no application to limiting either the type or number of other buildings that could be erected? I cannot agree with this construction of the effect of the restrictive covenants:- (1) First, on a general note, it seems extremely surprising that while a limit is placed on the type and number of houses that can be built on each individual Lot, yet there is no limitation in the event of a redevelopment either on the type or number of other buildings that can be erected. I have already referred to Mr Chan’s submission regarding swimming pools, hospitals and schools. (2) Secondly, the context of the restrictive covenants militates against the construction contended for by the plaintiff. Relevant to this discussion are the Conditions of Exchange. Under the Conditions, as we have seen, the lessee was entitled to build one or more buildings on the Mother Lots (provided they were not for industrial use and were not factory buildings) and within this entitlement, to build houses. The lessee having chosen to build houses on Mother Lot No. 3665[29], it is in this context against which the restrictive covenants have to be viewed. The individual Government Leases entered into in the 1960s therefore concerned houses and nothing else. By then, the Mother Lots had been sub-divided and there was no question of going back to the previous state of affairs where the two Mother Lots had not yet been divided up or built on. Seen in this way, I agree with Ms Eu’s submission that in the case of any redevelopment of the individual, sub-divided Lots, only houses were contemplated, and no other type of building. Such construction is, I believe, the inevitable effect of the restrictive covenants and the intention behind them. I have earlier dealt with the effect of the user part of the restrictive covenants. (3) Where the Court of Appeal principally fell into error was in the treatment of the Conditions of Exchange. While it was right to regard that document and what transpired under it as part of the context of the Government Leases, it was wrong to assume that the state of affairs envisaged under the Conditions of Exchange (namely that there were two Mother Lots on which one or more buildings could be erected) remained the same when the Government Leases were created. The factual circumstances had clearly changed. 21. The only permitted redevelopment under the Government Leases being the building of houses, and no more than one house on each Lot, and no other type of building, the proposed redevelopment by building a 26 storey building is also for this reason prohibited. 22. In view of the construction of the restrictive covenants reached above, it is unnecessary to consider the arguments based on straddling or whether it is permissible to build only a part of a house on each Lot. It does seem to me, however, on straddling it is unlikely that the Government Leases would, without modification, allow for this. Each Government Lease, as stated above, dealt with an individual Lot. The previous state of affairs when the Mother Lots were much larger and before sub-division, no longer exist. 23. It is also unnecessary to deal with the principle (if it exists and if so, its ambit) in Feather v The Queen. D CONCLUSION 24. The proposed redevelopment not being permitted under the Government Leases, this appeal should be allowed and the order made by the learned Deputy Judge, restored. As to costs, I would order that the plaintiff should pay the defendant’s costs in this appeal and in the Court of Appeal, such costs to be taxed if not agreed. The costs of the action shall be as ordered by the learned Deputy Judge. If any party wishes to have a different order for costs, written submissions should be served on the other party and lodged with the Court within 14 days of the handing down of this judgment, with liberty on the other party to lodge written submissions within 14 days thereafter. In the absence of such written submissions, the order nisi will stand absolute at the expiry of the time limited for these submissions. Mr Justice Chan PJ: 25. I agree with the judgment of the Chief Justice. Mr Justice Ribeiro PJ: 26. I agree with the judgment of the Chief Justice. Mr Justice Bokhary NPJ: 27. I agree with Chief Justice Ma’s judgment. Lord Hoffmann NPJ: 28. I agree with the judgment of the Chief Justice. Chief Justice Ma: 29. This appeal is unanimously allowed. The order of the learned Deputy Judge should be restored. The Court also makes an order for costs as set out in para 24 above. Ms Audrey Eu SC & Mr Jenkin Suen, instructed by Department of Justice for the Appellant Mr Edward Chan SC & Mr Desmond Leung, instructed by Iu, Lai & Li for the Respondent [1] From the decision dated 18 January 2012 of the Court of Appeal (Tang VP, Fok and Chu JJA) allowing the appeal from the decision dated 20 April 2011 of Deputy Judge Jason Pow SC. [2] The plans for the proposed development show that there will be shops on the ground floor, utility, sports and function rooms on the first floor, and residential flats from the second to 26th floor. [3] As will be seen, they were Crown Leases entered into on various dates in the 1960s. [4] Ms Audrey Eu SC (for the defendant) described these two Lots as the Mother Lots. For convenience, I shall do the same in order to distinguish them from the individual Lots when the Mother Lots were sub-divided. [5] Lot No. 3665 occupied an area of some 22,886 sq. ft., Lot No. 3666 occupied 9,934 sq. ft.. The Lots were not adjacent to one another; they were on opposite sides of Nam Kok Road. The five Lots with which this appeal is concerned, were carved out of Mother Lot No. 3665. [6] The owner of the Lots is described in the Conditions of Exchange as the lessee, all property held from the Government (formerly the Crown) being held on leasehold. [7] In other words, a premium. [8] At this point, Lot No. 3666 is no longer relevant for the purpose of this appeal. [9] Para 4(b) above. [10] Then called Crown Leases. I shall use the term Government Leases to describe these leases: see definition of Government Leases in section 3 of the Interpretation and General Clauses Ordinance Cap. 1. [11] This Lot has the same number as the Mother Lot of that number but this is purely coincidental as it is quite clearly a different Lot. [12] Sino Plan Corporation Limited. [13] With him, Mr Desmond Leung. [14] Referred to in para 5 above. [15] This was a point which the Court of Appeal also made: see para 5 of the judgment of the Court of Appeal. [16] The defendant was represented by Ms Audrey Eu SC and Mr Jenkin Suen. [17] (1865) 6 B&S 257, at 283-4. [18] Para 42 of the judgment. [19] The judgment of Tang VP (as he then was, now Tang PJ), with which the other Judges agreed. [20] [1998] 1 WLR 896, at 912H-913F. [21] (1999) 2 HKCFAR 279, at 296D-I. [22] See Vallejos Evangeline Banao v Commissioner of Registration, FACV Nos. 19 and 20 of 2012 at para 75-77; HKSAR v Lam Kwong Wai (2006) 9 HKCFAR 574 at para 63. [23] See Wah Yick Enterprises Co. Ltd v Building Authority (1999) 2 HKCFAR 170, at 176H-I per Litton PJ. [24] [1968] 1 QB 631, at 640. [25] These houses had been described as “Chinese type houses”; see para 6 above. [26] Para 4(1)(a) above. [27] Para 4(2)(b) above. [28] Such as, as Tang VP pointed out in para 26 of the judgment of the Court of Appeal, a commercial building. [29] See para 6 above. 
Mr Justice Fok PJ: 1. The appellant was convicted, after re-trial,[1] of murdering his girlfriend. He was granted leave to appeal that conviction on the ground that it was reasonably arguable the judge presiding over his re-trial misdirected the jury in respect of the objective limb of the defence of provocation as it applied to his case. After service of the appellant’s written case, the prosecution, as respondent, indicated that it would concede the appeal on this ground and the parties were directed to file a joint written case for the consideration of the Court in accordance with the established procedure for uncontested appeals.[2] 2. The only contested part of the appeal was whether, if the appeal is allowed quashing the conviction for murder without substituting a conviction for manslaughter, the appropriate consequential order should include a direction for a further re-trial. This was addressed by the parties in separate written cases and orally at the hearing before us. At the conclusion of the hearing, the Court allowed the appeal, quashed the appellant’s conviction and ordered a re-trial. We indicated that we would hand down our reasons for doing so in due course, which we now do. The background facts leading to conviction below 3. The facts out of which the murder charge arose are fully set out in the judgment of the Court of Appeal now under appeal.[3] For present purposes, a short summary of the salient facts will suffice. 4. In 2009, the appellant (then aged 18), was living together with his mother and elder brother in Tuen Mun. He had formed an intimate relationship with the deceased, Ms Lau Shuk Fan (“Lau”), in early 2009 and, by April 2009, she often spent nights at the appellant’s home. On the evening of 29 April 2009, the appellant killed Lau by repeatedly striking her with a chopper. The forensic pathologist testified that there were 64 separate wounds to Lau’s body, including 50 cut wounds, inflicted with a great deal of force. The scientific evidence was that Lau had been struck with the chopper many times whilst she was either sitting or crouching or lying on the lower bunk bed in the flat and that this was the only place where she had been attacked. 5. According to the appellant, whose account is the only version of the events leading to the killing, he was provoked to lose his self-control during a quarrel with Lau which escalated into a violent attack. He harboured suspicions that Lau was conducting an affair with his brother and, on the evening in question, had confronted her about this on the basis of certain evidence he had found at home that day. He said that he brandished a chopper with a view at first to scaring Lau and challenged her to tell him the truth as to whether she was having an affair with his brother. When she denied this and denigrated him as to his sexual abilities, he got the chopper from the kitchen and this led to a struggle with Lau. There was a lull in the struggle, he said, and he put down the chopper which Lau then used to chop him on the back of the head. The appellant claimed this caused him to lose all self-control and to snatch the chopper from her and to chop her indiscriminately. 6. The appellant therefore relied on the partial defence of provocation to the charge of murder provided for in section 4 of the Homicide Ordinance (Cap.339),[4] contending that he should be found guilty of the lesser offence of manslaughter by reason of provocation. For its part, the prosecution sought to argue that the multiple chop wounds were inflicted by the appellant attacking Lau in a vicious rage and not as a result of any provocation. 7. On 15 September 2011, by a majority of 6 to 1, the jury convicted the appellant of murder. They must therefore have been satisfied that the prosecution had made them sure that the appellant was not entitled to the statutory defence of provocation. The intermediate appeal and leave to appeal to the CFA 8. The appellant appealed his conviction for murder on the ground that the trial judge had misdirected the jury by failing to make it sufficiently clear that the defence was available if the provocation would “or may” have caused a reasonable person to lose self-control and do as the appellant did. The Court of Appeal dismissed his appeal on 20 September 2012, on the basis that, contrary to the appellant’s contention, this had been adequately conveyed to the jury.[5] The appellant’s application for a certificate on a point of law for this Court regarding the provocation direction was dismissed by the Court of Appeal on 26 October 2012.[6] 9. There was then a hiatus in the proceedings and it was not until 5 September 2016 that the appellant applied to this Court for leave to appeal out of time. The delay was explained on the basis that the appellant’s initial application for legal aid was refused and it was not until April 2016 that he received advice from the Clinical Legal Education Programme at the University of Hong Kong which led to the grant of legal aid in August 2016 to pursue an appeal to this Court.[7] 10. Ironically, the point on which the appellant was advised to apply for leave to appeal was not one that was reasonably arguable[8] but, by the time of the leave hearing before the Appeal Committee, this Court had handed down its judgment in HKSAR v Liang Yaoqiang (2017) 20 HKCFAR 1 (“Liang Yaoqiang”) and it was thought appropriate in the circumstances to grant leave to appeal to raise a further ground of appeal based on that decision, namely: “It is reasonably arguable that there is a further departure from the accepted norm in that the Trial Judge had failed to give proper direction to the jury on how to address the issue as to whether an ordinary person would have done what the Applicant did when considering the objective limb of the defence of provocation in the particular circumstances of the case, particularly in the light of the Applicant’s frenzied attack on the deceased with a chopper resulting in multiple deep wounds and the closing speech by prosecuting counsel which urged the jury to consider: ‘Would an ordinary person lose control and use a chopper to chop her to death? Chop her over 40 times while she was lying or crouching in the lower bunk.’ (see page 12E-G of the transcript of the Prosecution’s closing speech).”[9] The risk of an impermissible line of reasoning in provocation cases 11. In Liang Yaoqiang, the Court considered the construction of the words “do as he did” in section 4 of the Homicide Ordinance (Cap.339),[10] which relate to the objective limb of the defence of provocation.[11] The Court concluded that those words, properly construed in the light of the context and purpose of section 4 and the underlying policy of the law to mitigate the penal consequences of a killing carried out in the frenzy of a loss of self-control, meant “to kill the victim (i.e. to form the intent to kill or cause grievous bodily harm and act on that intent) by whatever means” rather than to kill the victim in exactly the manner the defendant did.[12] 12. In view of that construction of section 4, it is impermissible for a jury to reason that: (a) notwithstanding the defendant was in fact provoked to lose his self-control, (b) the scale of the retaliation was out of proportion to the provocation offered, so that (c) the objective question to be asked in assessing whether the defence of provocation was made out should necessarily be answered against him.[13] 13. Although this issue does not arise in every case involving provocation (nor indeed in every frenzied attack case[14]), where there is nevertheless a real risk of that impermissible line of reasoning being pursued, the trial judge should give a suitable direction to counteract that risk.[15] Judge should have given special direction on “do as he did” in this case 14. Both the appellant and the respondent jointly submitted that, in the particular circumstances of the present case, there was a real risk of the impermissible line of reasoning being adopted by the jury when they considered the objective limb of the defence of provocation and that the trial judge misdirected the jury in failing to give proper directions to counteract that risk. For the following reasons, we were satisfied that this submission was made out. 15. In Liang Yaoqiang, the Court held that whether a case is one which requires the trial judge to give a special direction depends on whether: “… the nature and extent of the homicidal act in a given case is such that might lead to the impermissible line of reasoning. It also depends on the course of any cross-examination of the defendant and the content of the closing speeches of counsel. This question must therefore be answered on a case by case basis …”.[16] 16. In the present case, the killing of Lau was the result of a frenzied attack with a chopper resulting in at least 50 cut wounds to her body. The prosecution’s cross-examination of the appellant focused on the precise form and extent of his retaliation in the face of Lau’s allegedly provocative actions and words. This theme was revisited in the prosecution’s closing speech, in particular in the following extract from that speech shortly before its conclusion: “However, if despite of what I said above, you still take a view that the accused had been provoked or might have been provoked to kill Miss Lau by things said or done by her and/or by his elder brother, then you have to consider – then now you have to consider the objective element, i.e. whether an ordinary and sober person, free from drug, of the accused’s sex, age and his characteristics, would have behaved as he did. Would an ordinary person choose to break up with Miss Lau if he did believe that she had an affair with his elder brother? Probably yes. Would an ordinary person ask her to leave his place immediately? Probably yes. Would an ordinary person use physical violence on her? Probably yes. Would an ordinary person lose control and use a chopper to chop her to death? Chop her over 40 times while she was lying or crouching in the lower bunk. Use your common sense and your collective experience of life. You would know an ordinary person would not do as he did.”[17] 17. This part of the prosecution’s closing speech clearly invited the jury to reason that a person with ordinary powers of self-control might have had a range of reasonable reactions in the face of Lau’s alleged provocation but would not have chopped her to death by inflicting over 40 cut wounds. The scale of the actual retaliation by the appellant to the provocation was used as a means to suggest that the response to it went beyond that to which a person with ordinary powers of self-control would have resorted. 18. The risk of the jury adopting that impermissible line of reasoning when considering the objective limb of the provocation defence was also exacerbated by defence counsel’s speech to the jury in which he (quite properly)[18] invited it to conclude, in relation to the subjective question of whether the appellant had in fact been provoked into losing his self-control, that the number of chop wounds, their depth and the short period of time in which they were inflicted showed that this was due to a sudden and temporary loss of self-control. 19. The trial judge’s summing up to the jury on the defence of provocation followed the language of section 4 of the Homicide Ordinance (Cap.339). No criticism attaches to the judge in that regard since this Court’s decision in Liang Yaoqiang was not then available and the concern that led to that decision was not raised before him. But it does mean that the judge did not give any particular direction to counteract the risk of the jury pursuing the impermissible line of reasoning. In the particular circumstances of this case, since that was a real risk, it follows that the judge misdirected the jury by omitting the special direction referred to in Liang Yaoqiang at [124] and there was consequently a departure from an accepted norm to the appellant’s disadvantage such as to constitute a substantial and grave injustice. 20. Accordingly, the appellant’s appeal was allowed and his conviction for murder quashed. Appropriate to order a re-trial 21. Where it allows an appeal and quashes a criminal conviction, this Court has the power to order a re-trial.[19] The principles on which the Court acts when considering the issue of whether to order a re-trial are well-established and were common ground between the parties to this appeal: whether or not there should be a re-trial is a matter of discretion, depending on what justice requires (that being the critical question), and this may involve balancing a number of competing factors some weighing for, and some against, a re-trial.[20] 22. If this Court were to order a re-trial, it would mean that the appellant will have to face a third trial in respect of his killing of Lau. His first trial (in HCCC 312/2009 before Wright J and a jury) resulted in a murder conviction on 17 May 2010 that was quashed by the Court of Appeal (in CACC 169/2010) on 21 February 2011. The re-trial ordered by the Court of Appeal was the second trial (in HCCC 69/2011) which resulted in his conviction on 15 September 2011. A further re-trial would also entail considerable delay since Lau was killed in April 2009 and any re-trial of the appellant now would necessarily take place many years after the events in question and after the appellant has already been remanded in custody for over 8 years. 23. These factors, that a re-trial would mean a third trial for the appellant and that there will have been considerable delay in prosecuting that trial during which period the appellant has been remanded in custody, were the principal matters urged upon the Court by Mr Daniel Marash SC on behalf of the appellant in resisting the respondent’s application for a re-trial. 24. Although they are not factors to be lightly dismissed, we were satisfied that it would be in the interests of justice to order a re-trial in the present case. The charge of murder that the appellant will face is a very serious one and the interests of justice include the interests of the victim’s family and, more widely, the interests of the public in Hong Kong that persons guilty of serious crimes should be brought to justice and should not escape it merely because a judge is subsequently found to have misdirected the jury on a point of law.[21] The questions to be addressed in evaluating the defence of provocation are pre-eminently ones for a jury and it would be preferable for a jury to determine if the appellant’s conviction for killing Lau should be for murder or the lesser crime of manslaughter by reason of provocation. Whilst a third trial on any criminal charge must be an unusual course,[22] it is by no means unprecedented.[23] Since the appellant was sentenced to imprisonment for life, this is not a case where he has already served the whole of his sentence as was the case in Mok Kin Kau v HKSAR, where that factor weighed against a further re-trial.[24] Nor is a re-trial of the appellant likely to be lengthy or particularly complex.[25] 25. Although by the time of any re-trial there will certainly have been a considerable period of time since the killing, that delay has fortuitously worked to the appellant’s advantage in that, but for the hiatus during which he was seeking legal aid (referred to at [9] above), he would not have been able to rely on this Court’s decision in Liang Yaoqiang. That is the only basis on which his appeal against conviction for murder has been allowed, since the ground of appeal for which legal aid was eventually granted was determined not to be reasonably arguable. Any difficulties the appellant may have in recalling the events in question or his previous testimony can be mitigated, at least in part, by reference to his police interviews and transcripts of his evidence at his earlier trials. And the trial judge can, if it appears appropriate to do so, direct the jury to make due allowance in the appellant’s favour for the fact that he has to cast his mind back to events which took place a very long time ago. Furthermore, this is not a case in which it has been suggested that the passage of time has resulted in any specific problem for a re-trial, such as a relevant witness not being available. In these circumstances, the delay, whilst doubtless a relevant consideration, is not one which persuaded us to exercise our discretion against an order for a re-trial. Disposition 26. For the above reasons, we concluded that the appropriate orders to make in disposing of this appeal were: (1) to allow the appeal and quash the appellant’s conviction for murder; (2) to order that he be re-tried on a fresh indictment for murder; and (3) to order that the appellant remain in custody pending such re-trial. Mr Daniel Marash SC, instructed by Morley Chow Seto, assigned by the Director of Legal Aid, for the Appellant Mr William Tam SC, DDPP and Ms Sabra Lo SPP, of the Department of Justice, for the Respondent [1] In HCCC 69/2011 before Saw J and a jury. [2] See Mok Kin Kau v HKSAR (2008) 11 HKCFAR 1 and HKSAR v Shum Wan Foon (2014) 17 HKCFAR 303. [3] CACC 405/2011 (Yeung VP, Lunn JA and M Poon J), Reasons for Judgment dated 11 October 2012 (“CA Judgment”) at [3]-[34]. [4] This provides: “Where on a charge of murder there is evidence on which the jury can find that the person charged was provoked (whether by things done or by things said or by both together) to lose his self-control, the question whether the provocation was enough to make a reasonable man do as he did shall be left to be determined by the jury; and in determining that question the jury shall take into account everything both done and said according to the effect which, in their opinion, it would have on a reasonable man.” [5] CA Judgment at [42]-[46]. [6] CACC 405/2011 (Stock and Yeung VPP and Lunn JA), Judgment dated 26 October 2012. [7] Amended Application for Leave to Appeal (Out of Time) dated 8 February 2017 at [31]. [8] See the Reasons for Determination of the Appeal Committee in FAMC 43/2016 dated 17 February 2017 (reported in (2017) 20 HKCFAR 91) at [9]-[12]. [9] Ibid. at [13]-[16]. [10] See FN4 above. [11] As to the ingredients of the defence of provocation, see Liang Yaoqiang at [8]-[11] and Ho Hoi Shing v HKSAR (2008) 11 HKCFAR 354 at [31]. [12] Liang Yaoqiang at [68], [111]-[114]. [13] Ibid. at [115]. [14] FAMC 43/2016, Reasons for Determination dated 17 February 2017 at [16]. [15] Liang Yaoqiang at [124]. [16] Ibid. at [118]. [17] Transcript of Prosecution’s Closing Submissions in HCCC 69/2011. [18] Liang Yaoqiang at [107]. [19] See Hong Kong Court of Final Appeal Ordinance (Cap.484) s.17(2) and Criminal Procedure Ordinance (Cap.221) s.83E(1). [20] See Au Pui-kuen v Attorney General [1979] HKLR 16 at pp.19-20; Ting James Henry v HKSAR (2007) 10 HKCFAR 632 at [49]; and Kissel v HKSAR (2010) 13 HKCFAR 27 at [269]. [21] Au Pui-kuen v Attorney General (supra.) per Lord Diplock at p.20. [22] R v Chau Mei Ling [1981] HKC 542 at p.545B-C. [23] See Nguyen Anh Nga v HKSAR (2017) 20 HKCFAR 149 at [65]; Liang Yaoqiang at [127]; HKSAR v Lo Chun Siu (CACC 90/2013) unrep., Judgment dated 6 June 2014 at [191]; HKSAR v Wong Wai Man & Others (CACC 459/2001) unrep., Decision dated 27 April 2004 at [3] [24] (2008) 11 HKCFAR 1 at [12]. [25] Compare Ting James Henry v HKSAR (supra.) at [50], where the fact that the original trial was long and complex was treated as a factor of particular significance weighing against a re-trial. 
Chief Justice Ma: 1. I agree with the joint judgment of Mr Justice Ribeiro PJ and Mr Justice Fok PJ. Mr Justice Ribeiro PJ and Mr Justice Fok PJ: 2. Having contracted to purchase from the respondent (“Richly Bright”) a property on the 8th Floor of Tower A, New Mandarin Plaza in Kowloon (“the property”), the appellant (“De Monsa”) failed to complete the purchase and summary judgment was entered against it after proceedings before Deputy High Court Judge Le Pichon.[1] After De Monsa’s appeal to the Court of Appeal was dismissed,[2] it obtained leave to appeal from the Appeal Committee.[3] The contract in question was the fourth in a chain of contracts for the sale and purchase of the property and leave to appeal was granted on the basis that the following question of law arises: “Where, at the end of a chain of contracts for the sale and purchase of immovable property, the ultimate purchaser has, in breach of its contract of sale and purchase, failed to complete, what is the proper approach to determining the extent of the purchaser’s liability for such breach in relation to losses attributable to non-completion incurred up the chain of contracts?” 3. Leave was also granted on the “or otherwise” ground to enable De Monsa’s complaint that summary judgment had been granted on the basis of contested findings of fact to be addressed. 4. The issues arising necessitate an examination of the terms of the upstream contracts and the orders made in respect thereof. A. The chain of contracts 5. Under the first contract (“the Head Agreement”) dated 23 October 2007, Win Profit Corporation Limited (“Win Profit”) agreed to sell and World Orient Investment Limited (“World Orient”) to purchase the property for HK$107,200,000 with completion to occur on or before 17 September 2008, deposits totalling $10,720,000 (10%) being paid by World Orient. 6. By the second contract dated 5 November 2007 (“the 1st sub-sale”), World Orient agreed to sell and 823 Investment Ltd (“823”) to purchase the property for HK$123,746,130 with completion to take place on or before 17 September 2008, 3.00 pm, and with deposits totalling HK$18,561,919.50 (15%) being paid by 823. 7. The third contract was entered into on the following day, 6 November 2007 (“the 2nd sub-sale”). 823 thereby contracted to sell and the respondent Richly Bright to purchase the property for HK$133,266,600 with completion to occur on or before 17 September 2008, 1.00 pm, and with deposits totalling HK$19,989,990 (15%) being paid by Richly Bright. 8. The final contract, the subject-matter of the present appeal, was dated 8 May 2008 (“the 3rd sub-sale”) and Richly Bright thereby agreed to sell and De Monsa to purchase the property for HK$135,864,000 with completion to take place on or before 17 September 2008, 11.00 am, and with deposits totalling HK$13,586,400 (10%) being paid by De Monsa. B. The Orders made upon failure to complete 9. After De Monsa failed to complete the 3rd sub-sale, each of the other purchasers up the chain also failed to complete. Proceedings for breach of contract were brought by the respective vendors and judgments entered against the respective purchasers. 10. Win Profit instituted proceedings in HCA 1487/2009 against World Orient in respect of the Head Agreement. It forfeited World Orient’s deposit of $10,720,000 but, since the Court found that on 17 September 2008 the property was worth HK$127,150,000 (considerably more than the contracted sale price), Win Profit suffered no loss and no award was made in its action. 11. World Orient brought proceedings against 823 in HCA 24/2009 in relation to the 1st sub-sale. It forfeited 823’s deposit of HK$18,561,919.50 and obtained an award for damages in the additional sum of HK$8,704,210.50. World Orient claimed such damages as representing (i) its loss of profit in the sum of HK$16,546,130 (being the difference between the price it was to pay Win Profit for the property and the price at which it was to on-sell the same to 823); and (ii) the loss of its deposit forfeited to Win Profit in the sum of HK$10,720,000; after giving credit for 823’s forfeited deposit. 12. 823 in turn forfeited Richly Bright’s deposit of HK$19,989,990 and its action for damages against Richly Bright in HCA 1452/2010 was settled with Richly Bright consenting to (a) an award of damages in the sum of HK$8,092,339.50 (representing (i) loss of 823’s profit in the sum of HK$9,520,470.00 being the difference between 823’s purchase and resale prices; and (ii) loss of 823’s deposit forfeited by World Orient; credit being given for forfeiture of Richly Bright’s deposit); (b) an order that it indemnify 823 in respect of the damages award of HK$8,704,210.50 made against 823 in favour of World Orient in HCA 24/2009; and (c) an order that it pay HK$1,399,298 to estate agents Centaline Property Agency Limited as liquidated damages payable under the 2nd sub-sale agreement. C. The orders made against De Monsa in the Courts below 13. In the first place, the Judge declared that Richly Bright was entitled to forfeit De Monsa’s deposit in the sum of HK$13,586,400. Secondly, her Ladyship ordered De Monsa to pay Richly Bright damages in the sum of HK$9,000,990. This represented damages to compensate Richly Bright for (i) loss of profit in the sum of HK$2,597,400 (being the difference between Richly Bright’s purchase and resale prices); and (ii) loss of the deposit paid by Richly Bright to 823 in the sum of HK$19,989,990. Thirdly, the Judge ordered De Monsa to indemnify Richly Bright “against the sum of HK$8,704,210.50 and any further sums and legal costs due from [Richly Bright] to [823] under the Judgment [in HCA 1452/2010]”. Apart from the said sum of HK$8,704,210.50, damages in the sum of HK$8,092,339.50 were also due from Richly Bright to 823 pursuant to the said Judgment and therefore come within the words of the Order (as the parties accept) although, oddly, the Judgment does not expressly refer to that amount. Fourthly, the Judge ordered De Monsa to indemnify Richly Bright “against the sum of HK$1,399,298.00 ... due from [Richly Bright] to Centaline Property Agency Limited” referred to above. These orders were upheld by the Court of Appeal in dismissing De Monsa’s appeal. 14. The effect of those orders is that De Monsa has been ordered to pay damages totalling HK$40,783,238 plus interest and costs. Apart from the forfeiture of De Monsa’s deposit of $13,586,400, it was required to pay the additional amount of HK$27,196,838 by way of damages for failing to complete the purchase. The award therefore imposes on De Monsa a liability which embraces the indemnities given by Richly Bright to 823, reflecting liabilities established as between parties to the upstream contracts. The question in this appeal is whether such liability is correctly imposed. D. Damages for breach of contract D.1 The basic principle 15. The object of an award of damages for breach of contract is compensatory. It aims to place the innocent party, so far as a monetary award can do so, in the same position as if the contract had been performed in accordance with its terms.[4] This basic principle could, however, if applied without any constraint, result in a defendant being made liable for all losses caused by a particular breach regardless of the improbability or remoteness of any item of loss. 16. Accordingly, in the mid-19th Century landmark case of Hadley v Baxendale,[5]Alderson B formulated the test for remoteness of damage for breach of contract in the following well-known terms:[6] “Where two parties have made a contract which one of them has broken, the damages which the other party ought to receive in respect of such breach of contract should be such as may fairly and reasonably be considered either arising naturally, i.e., according to the usual course of things, from such breach of contract itself, or such as may reasonably be supposed to have been in the contemplation of both parties, at the time they made the contract, as the probable result of the breach of it. Now, if the special circumstances under which the contract was actually made were communicated by the plaintiffs to the defendants, and thus known to both parties, the damages resulting from the breach of such a contract, which they would reasonably contemplate, would be the amount of injury which would ordinarily follow from a breach of contract under these special circumstances so known and communicated. But, on the other hand, if these special circumstances were wholly unknown to the party breaking the contract, he, at the most, could only be supposed to have had in his contemplation the amount of injury which would arise generally, and in the great multitude of cases not affected by any special circumstances, from such a breach of contract. For, had the special circumstances been known, the parties might have specially provided for the breach of contract by special terms as to the damages in that case, and of this advantage it would be very unjust to deprive them.” 17. The limiting purpose of the rule in Hadley v Baxendaleis explained by Lord Pearce in Koufos v C Czarnikow Ltd (The Heron II)[7] as follows: “The underlying rule of the common law is that ‘where a party sustains a loss by reason of a breach of contract, he is, so far as money can do it, to be placed in the same situation with respect to damages, as if the contract had been performed’ (Parke B in Robinson v Harman (1848) 1 Exch 850, 855). But since so wide a principle might be too harsh on a contract-breaker in making him liable for a chain of unforeseen and fortuitous circumstances, the law limited the liability in ways which crystallised in the rule in Hadley v Baxendale.” [8] And: “The whole rule in Hadley v Baxendale limits damages to that which may be regarded as being within the contemplation of the parties.”[9] 18. The two limbs are “the practical expression of a single principle … that parties should only be liable for damages which were when they contracted within their contemplation in the event of a breach”[10] and both limbs turn on an objective assessment of what the contract-breaker knew or ought to have known. As Lord Walker of Gestingthorpe observed in Jackson v Royal Bank of Scotland plc:[11] “The common ground of the two limbs is what the contract-breaker knew or must be taken to have known, so as to bring the loss within the reasonable contemplation of the parties”. 19. The first limb encompasses damages that are “such as may fairly and reasonably be considered [as] … arising naturally, i.e. according to the usual course of things, from such breach of contract”. Since every reasonable person is taken to know of the damage which flows “naturally” from a breach of the contract to which he is a party, this knowledge is imputed to a contract-breaker. In the second limb, actual knowledge is required in respect of special circumstances giving rise to damage which cannot be said to result naturally from such a breach of contract. Where the first limb is relied upon, the horizon of contemplation is confined to loss which arises naturally in the usual course of things and which is therefore presumed to have been within the parties’ contemplation. Under the second limb, the horizon of contemplation is extended to loss that does not arise in the usual course of things but which flows from the circumstances of which the contract breaker had actual knowledge. 20. As Lord Reid said in The Heron II at p 385: “The crucial question is whether, on the information available to the defendant when the contract was made, he should, or the reasonable man in his position would, have realised that such loss was sufficiently likely to result from the breach of contract to make it proper to hold that the loss flowed naturally from the breach or that loss of that kind should have been within his contemplation.” (Emphasis added) 21. As explained below, later authority built upon this statement by Lord Reid to mitigate some of the uncertainties left by The Heron II. While the House of Lords’ decision in that case clarified aspects of the rule in Hadley v Baxendale, it also generated controversial questions. In particular, questions arose as to the degree of likelihood required for the loss to come within the first limb. Thus, on the one hand, Lord Reid used the words “‘not unlikely’ as denoting a degree of probability considerably less than an even chance but nevertheless not very unusual and easily foreseeable.” [12] On the other hand, as we have seen, his Lordship put the requirement in terms of “loss ... sufficiently likely to result from the breach of contract to make it proper to hold that the loss flowed naturally”, emphasising that Alderson B in Hadley v Baxendale was “not distinguishing between results which were foreseeable or unforeseeable, but between results which were likely because they would happen in the great majority of cases, and results which were unlikely because they would only happen in a small minority of cases”.[13] Questions naturally arise as to whether the two formulations are mutually consistent. 22. Various linguistic formulae were suggested in an attempt to capture the desired nuance regarding the likelihood required. For instance, Lord Upjohn stated: “It is clear that on the one hand the test of foreseeability as laid down in the case of tort is not the test for breach of contract; nor on the other hand must the loser establish that the loss was a near certainty or an odds-on probability. I am content to adopt as the test a ‘real danger’ or a ‘serious possibility.’ There may be a shade of difference between these two phrases but the assessment of damages is not an exact science and what to one judge or jury will appear a real danger may appear to another judge or jury to be a serious possibility.”[14] 23. The meaning and application of such verbal formulae will necessarily be subject to debate. They will sometimes fail to provide clear criteria for determining what losses are or are not too remote. D.2 Remoteness in contract and in tort 24. As Lord Upjohn noted in the passage just cited, the rules of remoteness in contract differ from those governing remoteness in tort. The latter rules impose a much wider liability for “any type of damage which is reasonably foreseeable as liable to happen even in the most unusual case, unless the risk is so small that a reasonable man would in the whole circumstances feel justified in neglecting it”[15] or “any damage which [the tortfeasor] can reasonably foresee may happen as a result of the breach however unlikely it may be, unless it can be brushed aside as far fetched”.[16] 25. The reason for this difference is explained by Lord Reid in The Heron II as follows: “In contract, if one party wishes to protect himself against a risk which to the other party would appear unusual, he can direct the other party’s attention to it before the contract is made, and I need not stop to consider in what circumstances the other party will then be held to have accepted responsibility in that event. But in tort there is no opportunity for the injured party to protect himself in that way, and the tortfeasor cannot reasonably complain if he has to pay for some very unusual but nevertheless foreseeable damage which results from his wrongdoing.”[17] 26. For his part, Lord Upjohn explained the reason for the difference in the following terms: “Once an examination of the facts establishes a breach of duty on the part of the tortfeasor, the acts and omissions of the innocent party are irrelevant until the question of contributory negligence comes to be considered. A tortfeasor may and frequently is a complete stranger to the innocent party but he is, however fleetingly in many cases, his neighbour for the purposes of the law and bound to act with due regard to his neighbour's rights whomever he may be. If he fails in such duty the law has rightly laid down a more stringent test for the assessment of damages. But in contract the parties have only to consider the consequences of a breach to the other; it is fair that the assessment of damages should depend on their assumed common knowledge and contemplation and not on a foreseeable but most unlikely consequence. The parties may moreover agree to limit or exclude liability for damage, or agree on a liquidated sum, or one party can disclose to the other special circumstances which will render a breach especially serious to him. So the rules as to the assessment of damages have diverged in the two cases, and nowadays the concept of ‘foreseeability’ and ‘contemplation of the parties’ are different concepts in the law. It is true that as a matter of language there will in many cases be no great difference between foreseeing the possibility of an event happening and contemplating the possibility of that event happening and in some of the cases, from Blackburn J in Cory v Thames Ironworks Co onwards the word foresee or foreseeable is used in connection with contract but it is clear that it has really been used in the sense of reasonable contemplation and in my view it is better to use contemplate or contemplation in the case of contract, leaving foresee or foreseeability to the realm of torts.”[18] 27. The rule in Hadley v Baxendale has been applied by this Court in the context of a sale and purchase agreement in Chen v Lord Energy Ltd.[19] We shall address below the question of whether the losses claimed by Richly Bright against De Monsa are recoverable on Hadley v Baxendale principles. But it is first necessary to consider the effect of the decision of the House of Lords in Transfield Shipping Inc v Mercator Shipping Inc (The Achilleas). [20] D.3 The Achilleas 28. In The Achilleas,a vessel under time-charter was delayed during a legitimate final voyage and, in breach of the charterers’ contract, was re-delivered to the owners nine days late. As a result, the owners were unable to deliver the vessel in time for a follow-on charter lasting four to six months which had been fixed at a particularly lucrative charter rate they had managed to negotiate in an unusually volatile market. Because of the late re-delivery, the owners had to renegotiate the follow-on charter and to agree to a substantially reduced rate of hire. The owners claimed the difference between the original and renegotiated rates of hire for the entire period of the follow-on charter. That claim was allowed by a majority of arbitrators and was upheld at first instance and on appeal to the Court of Appeal. However, the charterers’ appeal to the House of Lords was unanimously allowed and the owners’ award was restricted to the difference in the charter hire for the nine-day overrun. 29. In allowing the appeal, Baroness Hale of Richmond confined herself to the orthodox Hadley v Baxendale approach. Lord Hoffmann and Lord Hope of Craighead, however, took the view that, as Lord Hoffmann put it, the orthodox approach is only: “... a prima facie assumption about what the parties may be taken to have intended, no doubt applicable in the great majority of cases but capable of rebuttal in cases in which the context, surrounding circumstances or general understanding in the relevant market shows that a party would not reasonably have been regarded as assuming responsibility for such losses”.[21] 30. Their Lordships focussed on the contractual underpinnings of the rule in Hadley v Baxendale and emphasised the need for recoverable loss to be loss flowing from the breach which is not only within the parties’ reasonable contemplation under either limb of the rule, but also loss of a type for which the contract breaker could reasonably be regarded as having assumed contractual responsibility. 31. Lord Hoffmann put this as follows: “It seems to me logical to found liability for damages upon the intention of the parties (objectively ascertained) because all contractual liability is voluntarily undertaken. It must be in principle wrong to hold someone liable for risks for which the people entering into such a contract in their particular market, would not reasonably be considered to have undertaken.”[22] 32. This approach is aimed at reflecting the parties’ overall bargain, as His Lordship explains: “The view which the parties take of the responsibilities and risks they are undertaking will determine the other terms of the contract and in particular the price paid. Anyone asked to assume a large and unpredictable risk will require some premium in exchange. A rule of law which imposes liability upon a party for a risk which he reasonably thought was excluded gives the other party something for nothing.”[23] 33. Lord Hope endorsed the requirement of an assumption of responsibility as a limiting principle. He considered it insufficient to hold the charterers liable for loss constituted by the lower charter rates during the entire subsequent fixture simply on the basis that such loss was foreseeable. His Lordship stated at §30: “Both parties were experienced in the market within which they were operating. Late delivery under a time charter is a relatively common situation, and it is not difficult to conclude that the parties must have had in contemplation when they entered into the contract that this might occur. Nor it is difficult to conclude — indeed this was conceded by counsel for the charterers — that in a market where owners expect to keep their assets in continuous employment late delivery will result in missing the date for a subsequent fixture. The critical question however is whether the parties must be assumed to have contracted with each other on the basis that the charterers were assuming responsibility for the consequences of that event.” 34. Because Lord Walker of Gestingthorpe agreed with Lord Hoffmann and Lord Hope as well as with Lord Rodger of Earlsferry (who did not consider it necessary to enter into the discussion of assumption of responsibility[24]), some academic commentators[25] have queried whether Lord Walker accepted the concept of assumption of responsibility as relevant to remoteness of damage in contract. However, in our view, Lord Walker’s reasoning makes it amply clear that he shared the approach of Lord Hoffmann and Lord Hope, emphasising the contractual underpinnings of the rule in Hadley v Baxendale and posing the question whether the contracting parties ought to be taken to have accepted responsibility for the type of loss in question. Thus, referring to the concept of assumption of responsibility, his Lordship stated: “...the underlying idea – what was the common basis on which the parties were contracting? – seems to me essential to the rule in Hadley v Baxendale as a whole. Businessmen who are entering into a commercial contract generally know a fair amount about each other's business. They have a shared understanding (differing in precision from case to case) as to what each can expect from the contract, whether or not it is duly performed without breach on either side. No doubt they usually expect the contract to be performed without breach, but they are conscious of the possibility of breach.”[26] 35. Referring to the speeches in The Heron II, Lord Walker stated: “...their Lordships had well in mind (but did not, perhaps, spell out at length) that it is not simply a question of probability. It is also a question of what the contracting parties must be taken to have had in mind, having regard to the nature and object of their business transaction.”[27] 36. Relating the concept of assumption of responsibility to what Lord Reid had stated in The Heron II, Lord Walker continued: “[The majority of the arbitrators] accepted and applied the owners' submission that ‘what mattered was that the type of loss claimed was foreseeable’ (para 18 of the majority reasons). That was in my opinion too crude a test, and it was an error of law to adopt it. What mattered was whether the common intention of reasonable parties to a charterparty of this sort would have been that in the event of a relatively short delay in re-delivery an extraordinary loss, measured over the whole term of renewed fixture, was, in Lord Reid's words, ‘sufficiently likely to result from the breach of contract to make it proper to hold that the loss flowed naturally from the breach or that loss of that kind should have been within [the defaulting party's] contemplation.’”[28] 37. The assumption of responsibility therefore provides a criterion in appropriate cases for deciding when it is or is not proper to hold a contract breaker liable for loss of a particular type. As Lord Hoffmann pointed out in The Achilleas,[29] it provides the only rational basis for the distinction drawn by the Court of Appeal between losses from “particularly lucrative dyeing contracts” and general loss of profits by the laundry in Victoria Laundry (Windsor) Ltd v Newman Industries Ltd.[30] 38. Whether a contract breaker has assumed responsibility for a particular type of loss is decided by viewing the nature and object of the contract against its commercial background. Adopting an approach similar to that for deciding whether a contractual term should be implied,[31] one ascertains by reference to objective indicia, whether the parties should be taken to have intended that the relevant type of loss flowing from breach of the contract falls within the scope of the contract breaker’s assumption of responsibility. As Lord Walker put it: “...what is most important is the common expectation, objectively assessed, on the basis of which the parties are entering into their contract.”[32] 39. In The Achilleas, objective indicia leading to the conclusion of non-assumption of responsibility by the charterers included the general expectations in the relevant market;[33] the fact that at the time of entering into the contract, the risk was unquantifiable by the charterers;[34] that the terms of the follow-on charter agreed between the owners and the subsequent charterers was something over which the defendant charterers had no control and could not predict;[35] and thus, that the terms of the next charter were, in relation to the charterers, res inter alios acta.[36] 40. As Toulson LJ pointed out in Supershield Ltd v Siemens Building Technologies FE Ltd, the assumption of responsibility, whilst usually a limiting principle, can also operate to include certain heads of loss. Thus, in The Achilleas, the effect was exclusionary since: “...the contract breaker was held not to be liable for loss which resulted from its breach although some loss of the kind was not unlikely. But logically the same principle may have an inclusionary effect. If, on the proper analysis of the contract against its commercial background, the loss was within the scope of the duty, it cannot be regarded as too remote, even if it would not have occurred in ordinary circumstances.” [37] 41. With respect, the analysis in The Achilleas regarding the concept of assumption of responsibility is compelling. It represents a logical extension of the rule in Hadley v Baxendale, building upon Lord Reid’s formulation mentioned above.[38] Being firmly grounded in the contractual principles governing the relationship between the parties, the assumption of responsibility concept provides a principled basis for distinguishing between losses which are or are not too remote. We unhesitatingly adopt it as representing the law in Hong Kong. E. Damages for breaches of sale and purchase agreements 42. As noted in Section D.1 above, the objective of an award of damages for breach of contract is to place the innocent party in the same position as if the contract had been performed in accordance with its terms. In the case of a contract for the sale and purchase of property, where the purchaser breaches the contract by failing to complete, it is often said that the vendor will, in the ordinary case, be entitled to an award of damages equal to the difference between the contract price and the market value of the property at the completion date.[39] 43. Thus, if the contract price is HK$10 million but the market value of the property on the date of completion is only HK$9 million, non-completion by the purchaser causes the vendor a loss of HK$1 million, entitling him to damages in that amount. Since the vendor retains the property after the purchaser’s default, he could if he chooses, sell the property on the market for HK$9 million. His HK$1 million award of damages therefore places him in the same position in monetary terms as if the contract had been duly performed.[40] 44. If, on the other hand, the market value of the property has risen above the contract price, the vendor suffers no loss. On the purchaser’s default, he retains a property worth more than the amount he would have received if the contract had been performed and is able to sell the retained property for more than the contract price. 45. Where it is the vendor who fails to complete, the purchaser is similarly entitled to the difference between the contract price and the market value of the property at the completion date. If the market value has gone up, he will be entitled to damages to compensate him for the loss of his good bargain. If, on the other hand, the market value has fallen, he will not recover any damages since he will have suffered no loss since he could purchase an equivalent alternative property on the market for less than the contract price. 46. The foregoing discussion relates to sales between the owner of the property and his direct purchaser. The position is different where the vendor and purchaser are confirmor and sub-purchaser who are parties to a sub-sale agreement. The present case involves, of course, a string of such sub-sales. 47. A party selling as confirmor does not, if his sub-purchaser fails to complete, stand in the same position as the vendor under the head agreement unless he completes his own contract with his vendor upstream and acquires the property. If he does not complete his purchase, he will not be left holding the property by virtue of his sub-purchaser’s default. His loss in such cases therefore cannot be measured by comparing the contract price with the market value of the property since he does not hold a property which he can sell on the market. Instead, the measure of the confirmor’s loss is the difference between the price at which he had contracted to purchase the property and the on-sale price agreed with his sub-purchaser. 48. Thus, if the confirmor contracts to purchase the property for HK$10 million and to sell to his sub-purchaser for HK$12 million, his loss is HK$2 million on the sub-purchaser’s non-completion. That is the position whether the market value of the property happens to be HK$9 million or HK$13 million at that date. Since the confirmor (who has not completed his own purchase) does not have the property to sell, evidence of the market value of the property is irrelevant in computing his loss. 49. The position outlined above applies to confirmors in a chain of sub-sale agreements. Each confirmor’s loss, upon his sub-purchaser defaulting, is his loss of bargain, being the difference between the price that the sub-purchaser agreed to pay him and the price at which he contracted to purchase the property from his vendor. Only a confirmor who has acquired and is left holding the property due to non-completion by the purchaser, measures his loss in terms of the difference between contract price and market value. F. Deposits in sale and purchase agreements F.1 The legal nature of deposits 50. It is almost universally the practice in Hong Kong for vendors to require purchasers to pay a deposit if there is any appreciable gap in time between contract and completion. The legal nature of a deposit was considered by this Court in Polyset Ltd v Panhandat Ltd.[41] It is a sum representing a percentage of the purchase price provided by the purchaser as an earnest to guarantee his performance of the contract and consideration for the vendor’s withdrawal of his property from the market. Provided the amount paid as a deposit is reasonable, the deposit amount stands to be forfeited to the payee in the event of non-performance by the payer.[42] 51. The vendor’s ability to forfeit a defaulting purchaser’s deposit is thus a highly significant factor in the parties’ contractual relationship. It provides a secure commercial basis for a vendor to enter into a binding contract with a counterparty which may, for instance, be a company with a merely nominal paid up capital. This is well-illustrated in the present case where both 823 and Richly Bright are HK$2 companies but which were nevertheless able to put up deposits in the respective sums of HK$18,561,919.50 and HK$19,989,990. 52. In Polyset, the Court set out the test for determining the reasonableness of a deposit amount in the following terms:[43] “(a) Where (in the absence of fraud or vitiating factors other than excessiveness) the amount of an agreed deposit matches or is less than the conventional amount, its forfeiture will not attract judicial scrutiny, whether or not the innocent party has suffered any loss as a result of the other party’s breach. (b) Where the deposit exceeds the conventional amount, that is, 10% in Hong Kong, forfeiture is only permitted if the party seeking to forfeit can show that exceptional circumstances justify the higher amount. (c) Such exceptional circumstances must relate to a true deposit’s purpose as an earnest of performance and as compensation for the vendor's withdrawal of his asset from the property market pending completion, providing an objective justification for the higher sum. (d) If such justification is not forthcoming, the courts will not recognize the amount as a true deposit but will treat it as an advance payment towards what was payable under the contract and recoverable as such, subject to the innocent party’s entitlement to deduct damages for any actual loss suffered as a result of the other party’s breach.” 53. In the present case, two of the deposits paid were for 15% of the purchase price. They therefore exceeded the conventional 10% amount, but the Court of Appeal held that 15% deposits were justified in the prevailing circumstances involving rapid sub-sales in a volatile market.[44] We see no reason to disagree and will treat the deposits in the present case as forfeitable in principle. 54. A deposit will be liable to forfeiture on breach of contract regardless of whether non-performance by the party in breach has caused the innocent party any, or as much, actual loss.[45] Thus, a vendor who retains a property after his purchaser fails to complete will be entitled to forfeit the deposit even if the market price of the property has risen above the contract price. 55. However, if a vendor has suffered loss which exceeds the deposit amount, he is entitled, after forfeiting the deposit, to sue for damages, giving credit for the amount of the forfeited deposit.[46] So if under a head agreement, the contract price is HK$10 million and the purchaser has paid a deposit of HK$1 million, if the market value of the property should be HK$8 million at the date of the purchaser’s failure to complete, the vendor will have a claim for HK$2 million damages and can recover an additional HK$1 million in damages, giving credit for the HK$1 million deposit forfeited. 56. A confirmor is similarly able to forfeit the deposit paid by his non-completing sub-purchaser, but he will not have a claim for additional damages unless his deposit fails to cover the whole of the profit he would have made on the sub-sale. Thus, if the confirmor has contracted to purchase the property for HK$10 million and to on-sell it for HK$10.5 million, having taken a deposit of HK$1 million from the sub-purchaser, he will have no claim against his sub-purchaser for damages since the deposit he forfeits will exceed the amount of his loss of bargain (HK$500,000). On the other hand, however, if the confirmor had contracted to purchase the property for HK$10 million and to on-sell for HK$13 million but had only taken a deposit of HK$1 million, he will have a claim against his sub-purchaser for damages of HK$3 million, against which he must give credit for the HK$1 million deposit forfeited. F.2 Deletions of standard clauses in the present case 57. The 3rd sub-sale was entered into by the parties signing a “Provisional Agreement for sale and purchase” in a standard form provided by an estate agent. Two clauses in that form were deleted by the parties. They provided as follows: Clause 7: “Should the purchaser fail to complete the purchase in the manner herein contained, the initial deposit shall be forfeited to the Vendor and the Vendor shall then be entitled at his absolute discretion to sell the said premises to anyone he thinks fit and the Vendor shall not sue the Purchaser for any liabilities and/or damages or to enforce specific performance. Clause 8: Should the Vendor after receiving the initial deposit paid hereunder fail to complete the sale in the manner herein contained, the Vendor shall immediately compensate the Purchaser with a refund of the initial deposit together with a sum equivalent to the amount of the initial deposit as liquidated damages and the reimbursement/payment (as the case may be) of stamp duty of the said premises and the Purchaser shall not take any further action to claim for damages or to enforce specific performance.” 58. At various stages, one or other of the parties has sought to rely on this deletion in aid of an argument favouring or countering the recoverability damages for all the heads of loss awarded below. In our opinion, these deletions are of no significance to the issues under discussion. 59. If they had not been deleted, the parties would have put in place an agreed regime for limiting liability in the event that either of them chose not to complete the sale and purchase. If it was the purchaser who chose not to complete, the vendor would have agreed to confine its remedy to forfeiture of the deposit. This would have modified the common law rule which would otherwise have entitled the vendor to pursue the purchaser for any additional loss, not covered by the deposit, incurred as a result of the purchaser’s non-completion. Conversely, if it was the vendor who chose not to complete, the purchaser would have agreed to confine its remedy to a return of the deposit, any stamp duty paid and a sum equivalent to the deposit as liquidated damages, foregoing any potential common law claim for additional loss not covered by the liquidated damages amount. Both parties would also have foresworn an action for specific performance. 60. The effect of deleting these two clauses is therefore simply to remove reliance on the regime for limiting liability in the event of non-completion provided under the standard form agreement. It follows that the usual common law position regarding deposits and damages which has been described in Section F.1 above applies without modification. G. Applying the principles in the present case G.1 The correct measure 61. As set out above,[47] Richly Bright agreed to sell and De Monsa to purchase the property for HK$135,864,000 with deposits totalling HK$13,586,400 being paid by De Monsa. Completion was due on or before 17 September 2008, 11.00 am, but De Monsa failed to complete. De Monsa knew that Richly Bright was selling as a confirmor as was stated expressly in the contract.[48] Richly Bright had itself in fact contracted to purchase the property from 823 for HK$133,266,600 and had paid to 823 deposits totalling HK$19,989,990. Richly Bright did not complete its own agreement with 823 and had its deposit forfeited by 823. Applying the principles discussed above, for what amount of damages should De Monsa properly be held liable? 62. Richly Bright and De Monsa must both be taken to have known that if De Monsa did not duly complete its purchase, Richly Bright stood to lose the anticipated profit on its re-sale as confirmor to De Monsa. Such anticipated profit was in the sum of $2,597,400, which constitutes the applicable measure of Richly Bright’s loss since it did not complete its own contract to purchase the property from 823. Such loss is plainly recoverable under the first limb of Hadley v Baxendale. It may fairly and reasonably be considered as arising naturally, i.e., according to the usual course of things, from De Monsa’s failure to complete its purchase, placing such loss within the parties’ reasonable contemplation. We might add in passing that there has been no suggestion that any communication of special circumstances relevant to engaging the second limb of the rule in Hadley v Baxendale ever passed between Richly Bright and De Monsa. 63. At this point in the analysis, the deposit assumes great importance. As Lord Pearce pointed out in The Heron II,[49] in entering into contracts, parties often “have the opportunity to define clearly in respect of what they shall and shall not be liable”. 64. The parties in the present case have availed themselves of that opportunity. In agreeing that De Monsa should provide Richly Bright with a 10% deposit, the parties were expressly making provision for possible non-completion by De Monsa. They were experienced property traders and clearly can be taken to have known that if such breach should occur, Richly Bright would be entitled to forfeit the deposit and would have a claim in damages if there was any loss flowing from that breach not covered by the deposit. They were making express provision in accordance with the usual practice of the Hong Kong property market, delineating the responsibility assumed by De Monsa in the event of its failure to complete. In the present case, De Monsa assumed responsibility to compensate Richly Bright by agreeing to the forfeiture of its deposit in the sum of HK$13,586,400. It transpires that this amounted to an assumption of responsibility to compensate Richly Bright for an amount considerably exceeding Richly Bright’s actual loss of bargain. As further explained in Sections G.2 to G.4 below, Richly Bright has suffered no recoverable loss over-topping the deposit and thus has no claim for any additional damages. 65. The contract entered into by Richly Bright and De Monsa therefore has as its nature and object, the sale of property by a confirmor to a sub-purchaser who has provided a forfeitable 10% deposit as an earnest of the latter’s performance of the contract. Whether expressed in terms of the parties’ reasonable contemplation or of responsibility assumed in the event of non-completion, De Monsa’s liability is limited to the forfeiture of its deposit. G.2 The additional damages erroneously awarded 66. As we have seen, the Courts below correctly declared that Richly Bright was entitled to forfeit the deposit. However, with respect, they fell into error in: (a) holding that De Monsa was additionally liable for HK$9,000,990, representing damages overtopping the forfeited deposit arrived at by adding together (i) Richly Bright’s abovementioned loss of profit in the sum of HK$2,597,400 and (ii) loss of the deposit paid by Richly Bright to 823 in the sum of HK$19,989,990; and, (b) ordering De Monsa to indemnify Richly Bright against the sum of HK$8,704,210.50 plus HK$8,092,339.50 being sums payable under the Judgment consented to by Richly Bright in favour of 823; and (c) ordering De Monsa to indemnify Richly Bright against the sum of HK$1,399,298.00 due from Richly Bright to Centaline Property Agency Limited arising under the 2nd sub-sale agreement. G.3 The reasoning in the Courts below 67. At first instance, the issues presently under discussion were little explored. The Judge merely agreed with counsel for the plaintiff that since De Monsa knew that Richly Bright was selling as confirmor, “the heads of loss itemized would have been within the reasonable contemplation of the parties” at the relevant time.[50] 68. In the Court of Appeal, referring to this Court’s decision in Chen v Lord Energy Ltd,[51]Kwan JA (with whom the other members of the Court agreed) framed the issues in the following terms: “The two questions framed in Chen v Lord Energy at para 25, as applied to the present situation, would be as follows: (1) whether, on the information available to the parties at the relevant time, the possibility of there being prior sub-sales, and the possibility of corresponding default in completion in the chain of transactions if the ultimate purchaser should fail to complete, was within the reasonable contemplation of Richly Bright and De Monsa; and (2) if it was not within the parties’ reasonable contemplation, whether in the circumstances of this case, knowledge of the possibility of prior sub-sales and corresponding default in completion in the chain of sub-sales if the ultimate purchaser should fail to complete, could be imputed to De Monsa. It is unnecessary for Richly Bright to show that De Monsa had actual knowledge of these matters.”[52] 69. After referring to the parties’ familiarity with investing in the property market; the heated state of the market and the common occurrence of confirmor sales,[53] her Ladyship held Richly Bright’s losses to be recoverable as falling within the rule’s first limb, concluding: “Given the background, knowledge and understanding of the parties, and the market condition at the time, it is clear that the possibility of there being one or more prior sub-sales of the property was nothing unusual and would be easily foreseeable. It would also be easily foreseeable that in such a volatile market, if there should be default by the ultimate purchaser, there might well be a domino effect causing corresponding default in the chain of sub-sales, particularly for a substantial property transaction, as were the transactions in respect of the property, all of which were over $100 million.”[54] (Emphasis supplied) 70. Kwan JA also held that, if necessary she was prepared to “impute knowledge to De Monsa of the possibility of prior sub-sales and corresponding default in the chain of sub-sales in the event the ultimate purchaser should fail to complete”[55] on the basis that the terms of the upstream agreements could have been discovered by a Land Registry search which would have revealed that the respective contracts provided for completion on the same day with two hour gaps between each completion. She held that it “must be easily foreseeable that if the ultimate purchaser should default, corresponding defaults in the chain of sub-sales would be entirely possible.”[56] (Emphasis supplied) 71. Her Ladyship held[57] that The Achilleas did not assist De Monsa because: “... the available evidence plainly indicates that the losses attributable to the sub-sales were within the reasonable contemplation of the parties at the material time, so this is not a situation in which foreseeable losses were of a kind which was completely unpredictable and for which the parties cannot reasonably be presumed to have assumed responsibility, as in the case under discussion in The Achilleas...” Adding that that was a case: “where the court was dealing with a highly specialised area of commercial law”. G.4 Errors in the approach below (a) The wrong test 72. With respect, in reaching the conclusion that all the losses claimed fell within the first limb of Hadley v Baxendale,both the Judge at first instance and the Court of Appeal applied the wrong test. Contrary to the Judge’s conclusion, it was not enough to hold De Monsa liable for all the heads of loss enumerated simply because it “knew that Richly Bright was selling as confirmor”. 73. Equally, it was not correct for the Court of Appeal to hold that such heads of loss were recoverable because the “possibility of there being one or more prior sub-sales of the property was nothing unusual and would be easily foreseeable”; and that “[it] would also be easily foreseeable that in such a volatile market, if there should be default by the ultimate purchaser, there might well be a domino effect causing corresponding default in the chain of sub-sales”.[58] Nor was it correct to use as a basis for liability “imputation” of such knowledge to De Monsa based on “constructive knowledge” of the terms of the upstream agreements.[59] 74. In purporting to adapt paragraph 25 of Chen v Lord Energy to the present case, the Court of Appeal actually misquoted and misapplied what Chan PJ had in fact said. His Lordship stated: “In the present case, it is necessary to decide two questions: (1) whether on the information available to the parties at the relevant time, the possibility of a re-sale by the purchaser with the resultant loss due to a drop in the market value of the property which is being claimed by the purchaser, was within the reasonable contemplation of the parties, that is, whether such loss was liable to result in the ordinary course of things from the stay; and (2) if it was not within the parties' reasonable contemplation, whether in the circumstances of this case, the knowledge that the purchaser was liable to resell the property after assignment can be imputed to the vendors.” (Italics supplied) 75. The Court of Appeal ignored the italicised words and so transposed the test from Chan PJ’s focus in orthodox fashion on the likelihood of the relevant loss[60] to a focus merely on the foreseeable possibility of there being prior sub-sales and corresponding defaults in completion. It thereupon erroneously upheld the award of damages against De Monsa for all the upstream losses simply on the basis that such losses arose out of foreseeable prior sub-sales. 76. Even if the focus had properly been on the types of loss claimed rather than on the mere fact of prior sub-sales, for the reasons set out in Section D.2 above, reasonable foreseeability of such loss is not enough to bring it within the first limb of the rule. As we have seen, the test of reasonable foreseeability applicable in tort cases is not the test for remoteness of damage in contract. In contract cases, recoverable loss is subject to the limiting principle that it must, as Lord Reid stated, be “sufficiently likely to result from the breach of contract to make it proper to hold that the loss flowed naturally”. And applying The Achilleas, it must be loss for which the contract breaker is taken to have assumed responsibility. (b) Wrong to allow Richly Bright to recover its forfeited deposit 77. As noted in Section G.2 above, in awarding damages against De Monsa in the sum of HK$9,000,990, the Courts below were holding De Monsa liable to compensate Richly Bright for the forfeiture by 823 of Richly Bright’s deposit in the sum of HK$19,989,990. 78. On the basis of the analysis set out in Section G.1 above, we consider that award to be incorrect, the proper measure being limited to the forfeiture of De Monsa’s deposit. 79. That conclusion may further be supported as follows. The assessment of damages must focus on the consequences of breach of the contract to which the defendant is a party, not on what flows from breach of someone else’s contract. This was overlooked when the Court of Appeal treated De Monsa’s non-completion as having a “domino effect”, thus sweeping up the entire chain of sub-sales.[61] Its approach ignores the fact that each sub-sale is a separate contract under which each sub-purchaser has contracted to complete its purchase and given a substantial deposit as an earnest of its performance. It plainly does not follow that the failure of each upstream purchaser to complete may “fairly and reasonably be considered as arising naturally, i.e. according to the usual course of things” from De Monsa’s failure to complete its purchase. 80. Each purchaser plainly reaches its own decision whether to complete its sale and purchase agreement, a decision which may be determined by commercial considerations. Thus in the present case, it was common ground between Win Profit and World Orient that the market value of the property at the completion date was $127,150,000. On that basis, it is hard to see any reason to assume that World Orient would fail (as it did) to complete. Objectively, it might readily have been expected to complete so as to avoid forfeiture of its deposit and to acquire a property worth HK$19.95 million more than the contract price. Thus, World Orient’s non-completion cannot be assumed to be part of any domino effect stemming from De Monsa’s non-completion. 81. Of course that valuation does not bind the parties to the other contracts. But, for the sake of argument, if the correct valuation was indeed $127,150,000 at the completion date, 823 would also have had good, objective, commercial grounds – not forfeiting its deposit and acquiring for $123,746,130, a property worth $3.4 million more – for completing its own purchase. On that hypothesis, it is not at all obviously to be assumed that 823’s failure to complete occurred like a falling domino caused by De Monsa’s non-completion. The point is that the state of the market might well provide commercial reasons for independent decisions made by upstream buyers regarding completion or otherwise of their respective contracts. 82. Nor is it naturally to be assumed that the upstream purchasers would be unable to finance acquisition of the property from their upstream vendors unless funded by De Monsa’s completion monies. The property was a commercial unit being purchased and re-sold by professional property traders. As noted in Section A above, whatever the capitalisation of each of the purchaser companies might have been, they each had been able to put up substantial cash deposits ranging from HK$10,720,000 to HK$19,989,990. Kwan JA described the persons behind Richly Bright and De Monsa as seasoned and sophisticated investors with a good understanding of the property market who had frequently traded in properties in a very substantial way. Mr Lauw of De Monsa, for instance, entered into 151 transactions between March 2007 and September 2008 at a total cost of some $7 billion, the most active months peaking at 15 transactions. There is accordingly no basis for assuming that if any of the upstream purchasers had decided to complete notwithstanding De Monsa’s failure to complete, they would have been prevented from doing so for lack of financial resources. 83. Richly Bright plainly made its own decision not to complete its purchase, triggering forfeiture of the deposit that it had provided to 823. It obviously knew that forfeiture would be the consequence. Loss of the deposit did not flow from De Monsa’s breach of the 3rd sub-sale agreement, but from Richly Bright’s own decision not to complete the 2nd sub-sale. Forfeiture of Richly Bright’s deposit was, so far as De Monsa was concerned, res inter alios acta. (c) Wrong to hold De Monsa liable for indemnities consented to by Richly Bright 84. We have also noted in Section G.2 above that the Courts below ordered De Monsa to indemnify Richly Bright against the respective sums of HK$8,704,210.50 and HK$8,092,339.50 payable under the Consent Judgment made against Richly Bright in favour of 823. 85. There is no basis for concluding that at the time of entering into the 3rd sub-sale agreement, it would have been in the parties’ reasonable contemplation that non-completion by De Monsa should result in losses voluntarily incurred by Richly Bright agreeing to indemnify parties further upstream in respect of their losses. There is no basis for taking De Monsa to have assumed responsibility for such voluntarily assumed losses. 86. All the indicia point away from any such assumption of responsibility. As explained in Section F above, the general expectation in the Hong Kong property market is that a purchaser who breaches his contract by failing to complete loses his deposit and, if loss exceeding the amount of the deposit is incurred by the vendor, finds himself liable to damages for the balance. No one familiar with market practices would expect the defaulting purchaser to assume responsibility for more remote losses involving indemnities voluntarily given by the vendor in respect of losses incurred by parties to contracts further upstream. Liabilities so incurred would have been wholly unquantifiable and unpredictable at the time De Monsa entered into the 3rd sub-sale contract. There happened to be three upstream contracts in the present case. But if there is to be liability simply on the basis that prior sub-sales are foreseeable, such liability would in principle cover however many additional sub-sales there might be in the upstream chain. That would make the ultimate purchaser arbitrarily liable for fortuitous matters over which it had no control. (d) Wrong to hold De Monsa liable to indemnify Richly Bright for its liability to the estate agent 87. As noted in Section G.2, De Monsa was also ordered to indemnify Richly Bright against the sum of HK$1,399,298.00 which had been due from Richly Bright to Centaline Property Agency Limited under Richly Bright’s contract with 823 and Centaline. 88. That sum was payable “as liquidated damages” on Richly Bright’s failure to complete. We can see no basis for holding De Monsa responsible for that liability. It was not in the parties’ reasonable contemplation. Nor was it a responsibility that De Monsa should be taken to have assumed. It was a matter agreed to in the tri-partite 2nd sub-sale agreement to which De Monsa was a stranger. H. Disposal of this appeal 89. For the foregoing reasons, we would allow the appeal and set aside the Orders made by the Judge and the Court of Appeal save in so far as they declared Richly Bright entitled to forfeit De Monsa’s deposit. 90. Richly Bright forfeited the deposit at the outset. On the basis of the result we have reached, the entirety of this litigation has ultimately proved fruitless since we have held that it has no further entitlement. In the circumstances, subject to the paragraph which follows, we would make an order nisi that Richly Bright pay De Monsa’s costs here and in both courts below, with liberty to Richly Bright if so advised, to make written submissions as to costs, to be lodged within 14 days of the date of this judgment and liberty to De Monsa to file submissions in reply within 14 days thereafter. We would direct that the order nisi should stand as an order absolute in default of such submissions, without further directions. 91. 823 saw fit to intervene in these proceedings and actively supported Richly Bright in its pursuit of relief beyond forfeiture of De Monsa’s deposit, making and filing submissions of its own. We would accordingly make an order nisi that, with effect from the date of 823’s joinder as an Interested Party, 823 be jointly and severally liable with Richly Bright for the costs payable to De Monsa here and below. We would grant 823 liberty if so advised to make written submissions as to costs to be lodged within 14 days of the date of this judgment, with liberty to De Monsa to file submissions in reply within 14 days thereafter, the order nisi to stand as an order absolute in default of such submissions. Mr Justice Tang PJ: Introduction 92. By a provisional agreement for sale and purchase dated 8 May 2008 (“the De Monsa Agreement”), the plaintiff, Richly Bright International Ltd (“Richy Bright”), agreed to sell and the defendant De Monsa Investments Ltd (“De Monsa”), agreed to buy, the 8th floor of Tower A, the Mandarin Plaza, No 14 Science Museum Road, Kowloon (“the property”), for $135,864,000.00, with completion at 11 am on 17 September 2008. Time was of the essence. Deposits totalling 10% of the purchase price were paid. 93. Two days before 17 September 2008, Lehman Brothers collapsed and completion did not take place. 94. Before this court, it is accepted that De Monsa was in breach of the De Monsa Agreement such that the deposits paid were liable to be forfeited and that it is liable to compensate Richly Bright the difference, if any, between the contract price and the market price at the time of breach. 95. The present proceedings were commenced by Richly Bright against De Monsa in 2012, in which, inter alia, it claimed a declaration that it was entitled to forfeit the deposits paid, which, as I have said, is not contested in this court. However, there was no claim for damages based on the difference, if any, between the contract price and the market price. Instead, Richly Bright claimed damages and indemnities, on the basis that, because Richly Bright was selling as a confirmor[62] in a chain of confirmor sales De Monsa was liable to compensate Richly Bright for all losses incurred by the parties in the chain. That Richly Bright was selling as confirmor was expressly stated in clause 5 of the De Monsa Agreement. Moreover, the rider to the De Monsa Agreement stated that the property was sold subject to the terms and conditions of the “Principal Agreement (‘Head Agreement’)”, insofar as they were consistent with the De Monsa Agreement. It is these claims which led to this appeal. 96. Richly Bright applied for summary judgment and on 15 October 2012, Deputy Judge Le Pichon gave judgment in favour of Rightly Bright as claimed. Her ladyship said these heads of claims were recoverable because they would have been: “within the reasonable contemplation of the parties at the time of the provisional agreement … as the likely consequences of the breach …”[63] Her judgment was affirmed by the Court of Appeal[64] on 22 November 2013. 97. The Appeal Committee granted leave to appeal on 29 September 2014, and formulated the question of great general or public importance as follows: “Where, at the end of a chain of contracts for the sale and purchase of immovable property, the ultimate purchaser has, in breach of its contract of sale and purchase, failed to complete, what is the proper approach to determining the extent of the purchaser’s liability for such breach in relation to losses attributable to non-completion incurred up the chain of contracts?”[65] The Chain 98. Although the head agreement was not identified in the De Monsa Agreement, it was the formal sale and purchase agreement made on 23 October 2007 between Win Profit Corporation Ltd (“Win Profit”), the registered owner of the property, and World Orient investment Ltd (“World Orient”), for $107,200,000.00. 10% deposit was paid and it provided for completion to take place on or before 5 pm on 17 September 2008. Time was made expressly of the essence. 99. Between the Head Agreement and the De Monsa Agreement, there were two further Sub-Sale Agreements.[66] The first Sub-Sale Agreement was the preliminary sale and purchase agreement made between World Orient, as confirmor, and 823 Investment Ltd (“823”)[67] as purchaser made on 5 November 2007 (“the 823 agreement”). The price was $123,746,130.00. A 15% deposit ($18,561,919.50) was paid and it was agreed that completion should take place on or before 3 pm on 17 September 2008. 100. The second Sub-Sale Agreement was the preliminary sale and purchase agreement dated 6 November 2007 made between 823 as confirmor and Richly Bright, as purchaser (“the Richly Bright agreement”). The consideration was $133,266,600.00. A 15% deposit ($19,989,990.00) was paid and completion was to take place at 1 pm on 17 September 2008. 101. None of the earlier Sub-Sale Agreements was completed, which led to proceedings based on the respective agreements. 102. Win Profit took action to claim damages, after forfeiting the deposit[68], representing the difference between the contract price and the price of $80,019,900.00 which Win Profit realized on resale on 17 April 2009.[69] It was agreed by the parties there that the market value of the property as at 17 September 2008 was $127,150,000.00.[70] Win Profit’s claim failed because it was held that it had failed to sell within a reasonable time. 103. In HCA 24/2009, World Orient obtained judgment against 823 in the sum of $8,704,210.50.00 arrived at as follows: 104. In HCA 1452/2010, a consent order was made that Richly Bright indemnifies 823 for the sum of HK$8,704,210.50 due to World Orient in HCA 24/2009 and pays 823 $8,092,339.50, which latter sum represented 823’s loss of profit on its sale to Richly Bright of $9,520,470.00 after netting off the forfeiture of the deposit paid by 823 to World Orient of $18,561,919.50 against the forfeiture by 823 of the deposit paid by Richly Bright of $19,989,990.00. This Action 105. In the present action, Richly Bright claimed against De Monsa damages in the sum of $9,000,990.00 which represented Richly Bright’s profit on its sale to De Monsa of $2,597,400.00 and the difference between the 15% deposit it paid to 823 and the forfeiture of the 10% deposits paid by De Monsa. In addition, Richly Bright sought indemnities against De Monsa in respect of any sum which Richly Bright was liable to pay to 823 as well as the costs payable to 823 in HCA 1452/2010.[71] Hadley v Baxendale 106. As noted, Richly Bright succeeded because the courts below were satisfied that these damages were recoverable because they would have been within the reasonable contemplation of Richly Bright and De Monsa at the time they entered into the De Monsa agreement. For Kwan JA[72] the relevant question was: “… whether, on the information available to the parties at the relevant time, the possibility of there being prior sub-sales, and the possibility of corresponding default in completion in the chain of transactions if the ultimate purchaser should fail to complete, was within the reasonable contemplation of Richly Bright and De Monsa;”[73] 107. Kwan JA said that given that: “40. … the property market was in a heated condition at the time …” 41. … it is clear that the possibility of there being one or more prior sub-sales of the property was nothing unusual and would be easily foreseeable. It would also be easily foreseeable that in such a volatile market, if there should be default by the ultimate purchaser, there might well be a domino effect causing corresponding default in the chain of sub sales, particularly for a substantial property transaction, [which these were] ...” 108. Kwan JA held that losses resulting from the default in the chain of sub-sales were recoverable. 109. The normal measure of damages when a purchaser has failed to complete is “the contract price less the market price.”[74] In order to recover more a plaintiff “has to show that the loss in respect of which he claims damages was caused by the defendant’s wrong, and also that the damages are not too remote to be recoverable. The principle of remoteness of damage is a limiting principle of policy …”[75] The principle was famously expressed by Baron Alderson’s in Hadley v Baxendale (1854) 9 Exch 341 in these words: damages for breach of contract “… should be such as may fairly and reasonably be considered either arising naturally, i.e., according to the usual course of things, from such breach of contract itself, or such as may reasonably be supposed to have been in the contemplation of both parties, at the time they made the contract, as the probable result of the breach of it. Now, if the special circumstances under which the contract was actually made were communicated by the [claimants] to the defendants, and thus known to both parties, the damages resulting from the breach of such a contract, which they would reasonably contemplate, would be the amount of injury which would ordinarily follow from a breach of contract under these special circumstances so known and communicated. But, on the other hand, if these special circumstances were wholly unknown to the party breaking the contract, he, at the most, could only be supposed to have had in his contemplation the amount of injury which would arise generally, and in the great multitude of cases not affected by any special circumstances, from such a breach of contract. For, had the special circumstances been known, the parties might have specially provided for the breach of contract by special terms as to the damages in that case; and of this advantage it would be very unjust to deprive them.” 110. Robert Goff J has explained that the principle is no longer stated in terms of two rules but rather in terms of a single principle and that one decides “each case on the basis of the relevant knowledge of the defendant”.[76] I believe the principle is that only damages which would arise generally and in the great multitude of cases would be recoverable, and such damages may vary according to the defendant’s relevant knowledge including special circumstances known to the defendant at the time of the contract. In other words there is not a different principle according to whether or not special circumstances were known to the defendant. 111. Clear support for this view can be found in the speech of Lord Hodson in the Heron II[77] where he said he found guidance from the use in more than one place in the judgment in Hadley v Baxendale of the expression “in the great multitude of cases” which indicated “the damages recoverable for breach of contract are such as flow naturally in most cases from the breach, whether under ordinary circumstances or from special circumstances due to the knowledge either in the possession of or communicated to the defendants. This expression throws light on the whole field of damages for breach of contract and points to a different approach from that taken in tort cases.” This passage was cited by Lord Rodger of Earlsferry in The Achilleas[78], who also found guidance from the use of the expression “in the great multitude of cases”. 112. Lord Walker of Gestingthorpe also recognised it as one principle and agreed with Robert Goff J.[79] 113. I believe it is also supported by Lord Reid who said in Heron II at p 385 : “In cases like Hadley v Baxendale or the present case it is not enough that in fact the plaintiff's loss was directly caused by the defendant’s breach of contract. It clearly was so caused in both. The crucial question is whether, on the information available to the defendant when the contract was made, he should, or the reasonable man in his position would, have realised that such loss was sufficiently likely to result from the breach of contract to make it proper to hold that the loss flowed naturally from the breach or that loss of that kind should have been within his contemplation.”[80] (Emphasis added) 114. The special circumstances relied on by Kwan JA were that it was a confirmor sale and the market was in a heated condition. The latter added little to the former since it is only in an active market that there are likely to be confirmor sales. More importantly, here, given the special circumstance of a confirmor sale, was it a reasonable and natural consequence of De Monsa’s default that Richly Bright would also default? Is this a result that would flow naturally in the great majority of confirmor sales? I think not. They are not losses which would ordinarily follow from the breach of such a contract. 115. Here, Kwan JA found in favour of Richly Bright because the chain default “would be easily foreseeable.” And in ascertaining what was within the reasonable contemplation of the parties for this purpose, Kwan JA said the test was “a degree of probability considerably less than an even chance but nevertheless not very unusual and easily foreseeable”.[81] It may be that Richly Bright’s failure to complete and, indeed, the default up the chain, were not unforeseeable but, with respect, that is not the test[82]. As Lord Reid explained in The Heron II, Baron Alderson: “was not distinguishing between results which were foreseeable or unforeseeable, but between results which were likely because they would happen in the great majority of cases, and results which were unlikely because they would only happen in a small minority of cases.” and that “… a result which will happen in the great majority of cases should fairly and reasonably be regarded as having been in the contemplation of the parties, but that a result which, though foreseeable as a substantial possibility, would only happen in a small minority of cases should not be regarded as having been in their contemplation...” The Achilleas 116. The Achilleas provides a helpful illustration of the application of this approach. It was concerned with late redelivery of a vessel by the charterers. Under an addendum to the charter-party, redelivery was due on 2 May 2004. In April, market rates having risen, the owners fixed a follow-on time charter with another company at the new market rate of $39,500.00 per day. The company had the right to cancel if the vessel was not available on 8 May. By 5 May the owner realised that the vessel would not be redelivered by 8 May, the market having fallen sharply in the meantime, in order to secure the company’s agreement to extend the cancellation date until 11 May, the revised expected redelivery date, a new daily rate of $31,500.00 was agreed. The owner claimed against the charterer damages for breach of contract for $1,364,584.00, which represented the difference between the 2 rates for the duration of the new charter. The arbitrators, by a majority, decided in favour of the claim. The charterer’s appeal was dismissed by Christopher Clarke J and the Court of Appeal. The appeal to the House of Lords[83] succeeded. Their lordships[84] gave different reasons for allowing the appeal. Baroness Hale of Richmond was doubtful whether the appeal should be allowed but that if it was to be allowed she preferred it to be allowed on: “the narrower ground identified by Lord Rodger[85], leaving the wider ground[86] to be fully explored in another case and another context.” I turn first to the narrower ground. 117. At the arbitration, Counsel for the charterer agreed that the “not unlikely” results arising from the late delivery of the vessel would include missing dates for a subsequent fixture. And the arbitrators so found. I think it was not so much a concession as a matter of common sense. However, the significance of the fact that missing dates for a subsequent fixture was not unlikely is that, notwithstanding, the claim failed because, the loss of a lucrative contract in turbulent market was not the ordinary consequence of a breach of this kind. 118. On the facts Lord Rodger held , at the time of the addendum which provided that redelivery should take place at the latest on 2 May 2004: “60. … neither party would reasonably have contemplated that an overrun of nine days would ‘in the ordinary course of things’ cause the owners the kind of loss for which they claim damages. That loss was not the ‘ordinary consequence’ of a breach of that kind. It occurred in this case only because of the extremely volatile market conditions which produced both the owner’s initial (particularly lucrative) transaction, with a third party, and the subsequent pressure on the owners to accept a lower rate for that fixture.” 119. Here, too, the extraordinary events which preceded the date of completion, including the collapse of Lehman Brothers, (and the attendant global meltdown), which preceded the confirmors’ failure to complete when it was most probably in their interest to do so, given the likely market price at the material time[87], would not reasonably have been in their contemplation at the time of the De Monsa agreement. The loss arising from their failure to complete was not the ordinary consequence of a breach of this kind. In the great run of cases, the normal measure of damages, namely, the difference between the market and contract price, would suffice. Indeed, even on the facts of the present case, had the market price taken a plunge, the normal measure should also supply a ready solution. This Appeal 120. Before us, it was submitted that since completion of the De Monsa agreement was agreed to take place at 11 am whereas Richly Bright’s purchase was to be completed at 1 pm, there was insufficient time to arrange for alternative finance, so it would have been in the reasonable contemplation of the parties at the time of the contract, that the purchasers default would lead to the confirmor’s default. Presumably, this submission is made on the basis that given more time, Richly Bright would have been able to complete. However, as a monograph published by the Estate Agents Authority in 2006 made clear (if clarification be needed) if a purchaser fails to complete, the confirmor “is still obliged to complete under the agreement with the owner.” That is just plain common sense. Thus, if there was little time between the 2 completions it behove the confirmor to make timely preparation. The fact that this was a multi-million transaction in respect of which a substantial deposit had been paid reinforces the view that objectively, the reasonable contemplation was that such preparation would be made[88]. An alternative submission, which was predicated upon Richly Bright’s inability to complete in any event, was that as Richly Bright was a shell company (it had a paid up capital of $2), it must have been in the parties’ contemplation that it could not complete if De Monsa failed to complete. But, this shell company had paid a deposit of $19,989,990 and entered into a binding contract to complete! In the great majority of cases, one would expect Richly Bright to complete. Just as, no doubt when Richly Bright and 823 (another shell company) entered into their respective agreements, it was within their reasonable contemplation that the purchase would complete. 121. We were referred to Chen & Another v Lord Energy Ltd, a decision of this court[89], which was concerned with a claim for damages arising out of an undertaking given in the Court of Appeal in return for a stay of execution of an order of specific performance. There, the Plaintiff might have sold the property but for the stay, and the market had fallen in the meantime. Chan PJ held, with the agreement of the other members of the court that there was ample evidence to justify the Court of Appeal’s conclusion that it was within the reasonable contemplation of the parties at the time the stay was ordered by the Court of Appeal that the purchaser might sell the property and that it might suffer loss if prevented by the stay from doing so. The facts enumerated by Chan PJ amply justified that conclusion. The fact that in Lord Energy it was said purchase for resale was common at the time[90] would not help Richly Bright. In the present case, the fact that it was a confirmor sale was obvious and the decision does not depend on it. What is critical is whether it was within the reasonable contemplation of the parties that De Monsa’s failure to complete would have led to the other default(s) complained of. Lord Energy does not provide any answer. 122. I note however, that, at para 59 in Achilleas, Lord Rodger said: “… the position on damages might also be different, if, for example – when a charter party was entered into – the owners drew the charterers’ attention to the existence of a forward charter of many months’ duration for which the vessel had to be delivered on a particular date. The charterers would know that a failure to redeliver the vessel in time to allow the owners to deliver it under that charter would be liable to result in the loss of that fixture. Then the second rule or limb in Hadley v Baxendale might well come into play. But the point does not arise in this case.” 123. It was not Richly Bright’s case that it had told De Monsa that it would not be able to complete if De Monsa did not complete. Moreover, the effect of the communication of such knowledge is uncertain[91] and is academic in the present case. But I wish to note that Robert Goff J said [92]: “… the test appears to be: have the facts in question come to the defendants knowledge in such circumstances that a reasonable person in the shoes of the defendant would, if he had considered the matter at the time of making the contract, have contemplated that, in the event of a breach by him, such facts were to be taken into account when considering his responsibility for loss suffered by the plaintiff as a result of such breach.” [93] 124. For the above reasons, I believe Richly Bright’s application for summary judgment in respect of these damages was misconceived and should have been dismissed. I would allow De Monsa’s appeal. 125. Although I am able to dispose of the appeal on what Baroness Hale called the narrower ground, we have had the benefit of substantial submissions on the wider ground upon which The Achilleas was decided. I will take this opportunity to express my view on it. 126. There is no inconsistency between the narrower ground and the wider ground. Lord Walker agreed with both. So do I. 127. Lord Hoffmann said and I respectfully agree, since all contractual liability is voluntarily undertaken, liability for damages must be founded upon the intention of the parties, gathered upon the construction of the contract as a whole, construed in its commercial background[94]. And: “12. … It must be in principle wrong to hold someone liable for risks for which the people entering into such a contract in their particular market, would not reasonably be considered to have undertaken.” 128. And that the only rational basis for deciding whether the loss is of the type or kind for which he can be treated as having assumed responsibility and those which he could not is reflected in: “22. … what would reasonably have been regarded by the contracting party as significant for the purposes of the risk he was undertaking.” 129. That the rules in Hadley v Baxendale as explained in The Heron II: “24. ... are intended to give effect to the presumed intentions of the parties and not to contradict them.” 130. Lord Hoffmann’s views are well supported by the authorities, one of which is the Eagle Star[95], where valuers in breach of an implied term to exercise reasonable care and skill had negligently advised their clients that the property which were offered as security for loans were worth a great deal more than their market value. However, the valuers’ liability was confined to losses attributable to the deficient security but not for further losses attributable to a fall in the property market. That is because: “The scope of the duty, in the sense of the consequences for which the valuer is responsible, is that which the law regards as best giving effect to the express obligations assumed by the valuer: neither cutting them down so that the lender obtains less than he was reasonably entitled to expect, nor extending them so as to impose on the valuer a liability greater then he could reasonably have thought he was undertaking.”[96] 131. Moreover as Lord Walker said[97] passages in the speeches in Re Heron II : “78. … show that their Lordships had well in mind (but did not, perhaps, spell out at length) that it is not simply a question of probability. It is also a question of what the contracting parties must be taken to have had in mind, having regard to the nature and object of their business transaction. If a manufacturer of lightning conductors sells a defective conductor and the customer’s house burns down as a result, the manufacturer will not escape liability by proving that only one in a hundred of his customers’ buildings had actually been struck by lightning. The need to take account of the nature and object of the contract is recognised, I think, in the passage from Lord Reid’s speech, at p 385, which I have already quoted; in Lord Morris’s speech, at pp 398-399; in Lord Pearce’s speech, at pp 416-417 (with the example of the court ceiling collapsing during a sitting); and in Lord Upjohn’s speech, at pp 424-425. The need for the loss suffered to be within the horizon of the parties’ contemplation (Lord Pearce, at p 416) makes it less important to define its degree of probability with any precision. Arguably a vague expression (such as ‘real possibility’) is actually preferable, because it is more flexible, once it is understood that what is most important is the common expectation, objectively assessed, on the basis of which the parties are entering into their contract.” 132. I believe the nature and object of a contract can provide an important guide to ascertaining what the parties may be taken to have had in their contemplation as well as what contractual obligations the contracting parties could fairly be said to have undertaken. 133. In Siemens Building Technologies FE Ltd v Supershield Ltd [2010] 1 Lloyds Rep 349, a decision of the English Court of Appeal, Toulson LJ, whose judgment was agreed to by the other members of the court, said: “40. The law on remoteness of damage in relation to claims for breach of contract is grounded on the policy that the loss recoverable by the victim should be limited to loss from which the party in breach may reasonably be taken to have assumed a responsibility to protect the victim. It follows that the question of remoteness cannot be isolated from consideration of the purpose of the contract and the scope of the contractual obligation.” 134. It is not necessary in this appeal to consider the full implication of this approach. It is sufficient to say that on the facts of the present case, it could not fairly be said that De Monsa had undertaken responsibility for the chain default. Neither the nature or object of the De Monsa agreement, nor its purpose or scope, support a contrary conclusion. So, I would also allow the appeal on the wider ground. Costs 135. I agree with Mr Justice Ribeiro PJ and Mr Justice Fok PJ. Lord Walker of Gestingthorpe NPJ: 136. I agree with the joint judgment of Mr Justice Ribeiro PJ and Mr Justice Fok PJ. Chief Justice Ma: 137. The Court unanimously allows the appeal and sets aside the Orders made by the Judge and the Court of Appeal save in so far as they declared Richly Bright entitled to forfeit De Monsa’s deposit. The Court also makes the orders as to costs set out in the final two paragraphs in the joint judgment of Mr Justice Ribeiro PJ and Mr Justice Fok PJ. Mr Lam Chin Ching Gary, instructed by Tsangs, for the Plaintiff/1st Respondent Mr Warren Chan SC and Mr Jean-Paul Wou, instructed by K.C. Ho & Fong, for the Defendant/Appellant Mr Barrie Barlow SC and Mr Calvin Cheuk, instructed by Alfred Lam, Keung & Ko, for the Interested Party/2nd Respondent [1] HCA 548/2012 (15 October 2012). [2] Kwan, Lunn and Barma JJA, CACV 247/2012 (22 November 2013). [3] Ribeiro Ag CJ, Tang and Fok PJJ, FAMV 29/2014 (29 September 2014). [4] Chitty on Contracts (21st Ed.) Vol.1 at para.26-001 [5] (1854) 9 Ex 341; 156 ER 145 [6] At 354-355; 151 [7] [1969] 1 AC 350. [8] At p 414. [9] At pp 415-416. [10] Sempra Metals Ltd v IRC [2007] UKHL 34 per Lord Mance at §215. [11] [2005] UKHL 3 at §48. [12] At p 383. [13] At p 384. See Lord Hodson to like effect at p 411. [14] At p 425. [15] Per Lord Reid in The Heron II at pp 385-386. [16] Per Lord Upjohn, ibid, at p 422. [17] At p 386. [18] At pp 422-423. [19] (2002) 5 HKCFAR 297. [20] [2008] UKHL 48, [2009] 1 AC 61. [21] The Achilleas, at §9. [22] At §12. [23] At §13. [24] At §63. [25] Eg McGregor on Damages (18th Ed) at para 6-173; Edwin Peel, “Remoteness Revisited” (2009) 125 LQR 6. [26] At §69. [27] At §78. [28] At §84. [29] At §22. [30] [1949] 2 KB 528. [31] The Achilleas at §11 and §26. [32] Ibid at §78. [33] Per Lord Hoffmann at §11, §12, §23, §24 [34] “...because although the parties would regard it as likely that the owners would at some time during the currency of the charter enter into a forward fixture, they would have no idea when that would be done or what its length or other terms would be.” Per Lord Hoffmann at §23. See also per Lord Hope at §28. [35] Per Lord Hope at §34, Lord Walker at §86. [36] Per Lord Hoffmann at §23. [37] [2010] 1 Lloyd’s LR 349 at §43. An example of the principle’s potentially inclusive operation is given by Lord Walker at §78 of The Achilleas in relation to the manufacturer of a lightning conductor. [38] The Heron II,at p 385. [39] McGregor on Damages (19th Ed) at para 25-036; Laird v Pim (1841) 7 M & W 474. [40] As pointed out in Strategic Property Ltd v O’Se [2009] EWHC 3512 Ch at §40. [41] (2002) 5 HKCFAR 234. [42] Ibid at §66-§69. [43] Ibid at §90. [44] Court of Appeal at §26 – §30. [45] Polyset at §67. [46] Ibid at §68. [47] Section A. [48] Clause 5. The contract also contained a Rider stating that the sale was subject to the terms and conditions of “the Principal Agreement” (which was not identified) “save and except those varied by or inconsistence [sic] with the other terms of this Agreement”. [49] At p 413. Lord Upjohn similarly noted that, having considered the consequences of breach, the parties to a contract may “... agree to limit or exclude liability for damage, or agree on a liquidated sum, or one party can disclose to the other special circumstances which will render a breach especially serious to him.” (at p 422) [50] Judgment §57. [51] (2002) 5 HKCFAR 297. [52] Court of Appeal §37. [53] Court of Appeal §§39-41. [54] Court of Appeal §41. [55] Court of Appeal §43. [56] Ibid. [57] Court of Appeal §44. [58] Court of Appeal §41. [59] Court of Appeal §43. [60] I.e., arising out of the purchaser being prevented from selling prior to a fall in the market. [61] Court of Appeal §§2 and 41. [62] A sale by a vendor before it had acquired the legal title is called a confirmor sale. At its simplest, the vendor who had entered into an earlier contract of sale and purchase with the owner of the legal title agreed to sell its beneficial interest with more or less simultaneous completion to another purchaser so that on completion the owner of the legal title would assign the legal title directly to this other purchaser, the vendor joining in the assignment as confirmor to confirm the assignment to this purchaser, so as to bar any claim which the vendor might have had in the property. There could be more than one confirmor sale in which event all the confirmors would join in the assignment as confirmors. [63] Para 57. [64] Kwan, Lunn and Barma JJA. [65] Leave was also given on the “or otherwise” basis but it is not necessary to deal with it. [66] There was no reference to these agreements but I do not believe, in principle, it matters whether there was one or more than one confirmor sale. [67] 823 was given leave in the court of appeal to join as an interested party because of garnishee proceedings brought by it against De Monsa and Richly Bright. [68] “Having forfeited the deposit for failure to complete, the vendor remains entitled at common law to sue for damages, giving credit for the forfeited deposit where such damages exceed its amount.” Per Ribeiro PJ in Polyset Ltd v Panhandat Ltd (2002) 5 HKCFAR 234 at 263. [69] Reported [2012] 2 HKLRD 1053. [70] Para 50. If that was the market price at the material time there would have been no loss at all, since the contract price was substantially lower. [71] There were other claims, for example, a claim for $1,399,298.00 being payments to an estate agent. But the claims do not require separate consideration. [72] With whose judgment, the other members of the court agreed. [73] Para 37. [74] McGregor on Damages 4-015. [75] Satef-Huttenes Alberns SpA v Paloma Tercera Shipping Co SA (“The Pegase”) [1981] 1 Lloyd’s Rep 175 at 181. [76] In The Pegase at 182. Even so, the fact that there were 2 limbs in Baron Alderson’s formulation continues to dog discussion. [77] Koufos v C Czarnikow Ltd (The Heron II) [1969] 1 AC 350 at 411. [78] Transfield Shipping Inc v Mercator Shipping Inc (the “Achilleas”) [2009] 1 AC 61 at 77H, para 49. [79] The Achilleas [2009] 1 AC 61 at 82H. [80] This was included in the passage from the judgment of Lord Reid quoted by Lord Walker in his judgment at para 73. [81] Quoting from the judgment of Lord Reid in The Heron II at 383 A-B. [82] See also Lord Hope, Re Achilleas 73G. [83] Lord Hoffmann, Lord Hope of Craighead, Lord Rodger of Earlsferry, Lord Walker of Gestingthorpe, Baroness Hale of Richmond. [84] Lords Hoffmann, Hope, Rodger and Walker. [85] Namely that the loss was not the “ordinary consequence” of a breach of the kind. Para 60 Lord Hope, with the concurrence of Lord Walker. [86] Namely, whether the charter is to be taken to have undertaken contractual responsibility for this kind of loss. Lords Hoffmann, Hope and Walker, each of whom delivered a separate opinion. [87] It was agreed to be $127,150,000.00 in Win Profit’s claim against World Orient. World Orient had no discernible reason to default (indeed, Kwan JA at para 27 referred to Win Profit offering an extension of time to complete, which was not taken up, see para 3, in Win Profit Corp Ltd v World Orient Investment Ltd [2012] 2 HKLRD 1053), since the market price of $127,150,000.00 meant a handsome profit for World Orient. Moreover, not only did it forego the profit, it suffered its deposit of $127,150,000.00 to be forfeited. To different degrees, each of the other confirmors, were likely to have been better off had they completed. Although the market price was not agreed in the present action but the fact that there was no claim for damages based on the difference between contract and market price in any of the actions is telling. [88] In fact, the complaint is not so much against the failure to complete but the failure to give timely notice of the failure. But there was no contractual duty to give notice. [89] (2002) 5 HKCFAR 297, Li CJ, Bokhary, Chan and Ribeiro PJJ and Lord Hoffmann NPJ. [90] Lord Hoffmann at 312D. [91] It is difficult to conceive the purpose of any such communication. A vendor is highly unlikely to disclose that it was in distress and under pressure to sell. Nor is it easy to see the point of disclosure. It would not avail the vendor if the market price had fallen below the cover provided by the deposit. And if it had not, one would naturally expect completion to take place so that at least part of the deposit might be recouped. [92] The Pegase at 183. [93] Quoted by Lord Hoffmann at 69E, The Achilleas. [94] The Achilleas, p 71E, para 25. [95] South Australia Asset Management Corporation v York Montague Ltd (Eagle Star) [1997] AC 191. [96] The Eagle Star at p 212 in the speech of Lord Hoffmann with which the other members agreed. [97] The Achilleas at 86C. 
Mr Justice Chan PJ: 1. I would, for the reasons given in the judgment of Mr Justice Tang PJ and the further reasons as explained in the judgment of Sir Anthony Mason NPJ, allow these appeals and make the order and directions as proposed by Mr Justice Tang in paragraphs 60 and 62. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Sir Anthony Mason NPJ and, subject to that judgment, am also in broad agreement with the judgment of Mr Justice Tang PJ. Mr Justice Tang PJ: Introduction 3. The United Nations Convention Relating to the Status of Refugees (1951) as amended by the Protocol relating to the Status of Refugees(1967) (collectively, the “Convention”) impose important obligations on contracting state parties. Such obligations include an obligation to “facilitate the assimilation and naturalization of refugees” (Art 34)[1]. A refugee is a person who “owing to well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion, is outside the country of his nationality and is unable, or owing to such fear, is unwilling to avail himself of the protection of that country; or who …” (Art 1A(2))[2]. 4. Both the UK and the PRC are contracting statesto the Convention. However,as is permissible under Art 40, neither the UK (prior to 1997) nor the PRC has applied the Convention to Hong Kong Special Administrative Region (“HKSAR”). 5. It is the firm policy of the HKSAR not to grant asylum to refugees. That policy is not challenged. It is not contended that HKSAR Government (“HKSARG”) is obliged to “facilitate the assimilation and naturalization of refugees”. 6. These proceedings are concerned with the principle of non refoulement expressed in Art 33 as follows: “No Contracting State shall expel or return (‘refouler’) a refugee in any manner whatsoever to the frontiers of territories where his life or freedom would be threatened on account of his race, religion, nationality, membership of a particular group or political opinion.”[3] 7. The appellants claimed protection as refugees upon or shortly after their respective arrival in HK. They made their claims[4] to the UNHCR which processedthem in accordance with the procedural standards for Refugee Status Determination(“RSD”) under UNHCR’s mandate. The critical decision for the Director in such cases is whether to order the removal of such claimants and if so, to which country they should be removed. The Director’s practice is that pending RSD by UNHCR, a refugee claimant in Hong Kong, would be permitted to remain and that if the claim succeeds, the refugee would not be repatriated pending resettlement. Background[5] 8. C, the appellant in FACV 18 of 2011 is a national of the Democratic Republic of Congo (Zaire). His claim for refugee status was rejected by UNHCR on 19 March 2004 and his appeal dismissed by UNHCR by letter dated 24 March 2004. The letter stated, inter alia, that he was excluded by virtue of Article 1(F)(a). On 24 March 2004, a Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment (“CAT”) claim was lodged on his behalf. The CAT claim has not been resolved. 9. The home country of KMF, the appellant in FACV 19/2011, is the Republic of Congo (“Congo-Brazzaville”). He made a refugee claim to the UNHCR on 17 November 2004. His appeal was dismissed in July 2006. It was stated in the Form 86A filed on 23 April 2007, he had not “as yet ….made” a CAT claim. 10. BF (FACV 20/2011) is a citizen of Congo-Brazzaville. He arrived in Hong Kong on 10 November 2003 and made a refugee claim with UNHCR on 11 November 2003. His claim and appeal were rejected in early 2006. He has made a CAT claim which is still outstanding. Two grounds of appeal 11. The appellants accept that as the Convention have not been extended to the HKSAR, Art 33 has no direct application. However, they contended, with the support of the intervener,the UNHCR[6], that the principle of non-refoulement (“PNR”) has become a rule of customary international law (“CIL”) as well as a peremptory norm, and as such, has become part of the common law of HKSAR. They also contended, again with UNHCR’s support, that to give effect to such CIL, HKSARG should make its own RSD, and the Director must not return any refugee claimant without appropriate enquiry into their PNR claims. This was the first of the appellants’ two grounds. 12. Although the Court of Appeal only granted leave to appeal on issues covered by the first ground, Mr Michael Fordham QC, for the appellants, relied on a second ground, namely that, in any event, the Director’s decision to return a refugee claimant is subject to judicial review and must satisfy the high standards of fairness required given the gravity and importance of the decision.[7] The Second Ground 13. I will deal with Mr Fordham’s second ground first. 14. If a person is recognized as a refugee by the UNHCR, Hartmann J said: “10. … it is the inevitable practice of the Director not to repatriate that person but to afford him temporary refuge until the UNHCR -notthe Hong KongGovernment - is able to settle that person elsewhere in the world. I have described this practice of the Director as ‘inevitable’ because, during the course of the hearing, it was never suggested that the Director had in fact returned a recognised refugee to a country where there was a real risk he would be persecuted.”[8] Immigration Control 15. This practice has to be considered in the context of immigration control in Hong Kong. 16. Art 154 of the Basic Law (“BL 154”) provides: “The Government of the Hong Kong Special Administrative Region may apply immigration controls on entry into, stay in and departure from the Region by persons from foreign states and regions.” 17. Immigration controls are administered by the Director[9] under powers conferred in the Immigration Ordinance Cap 115 (the Ordinance). Under the Ordinance the Director has wide discretionary powers. He may refuse permission to land. He may give permission to land subject to a limit of stay and impose conditions.[10] Persons refused permission to land may be removed.[11] The Director is also enabled by s13 to authorize a person who landed in Hong Kong illegally (illegal immigrants) to remain in Hong Kong, subject to such conditions of stay as he thinks fit. The Rule of Law 18. The Ordinance is silent on how the Director’s wide powers should be exercised. Mr Yu accepted that the exercise must be rational, such that, for example, the Director may not decide on a toss of a coin. He also accepted that the Director must “administer the scheme of immigration control embodied in the Ordinance fairly and properly.”[12] 19. Mr Yu’s submissions are consistent with the authorities. For example, Wade and Forsyth, Administrative Law, 10th ed, states: “What the rule of law demands is not that wide discretionary power should be eliminated, but that the law should control its exercise” (p 286) “… the truth is that, in a system based on the rule of law, unfettered governmental discretion is a contradiction in terms. The real question is whether the discretion is wide or narrow, and where the legal line is to be drawn…It remains axiomatic that all discretion is capable of abuse, and that legal limits to every power are to be found somewhere.”[13] (pp 296 & 297) 20. An important expression of these principles can be found in Lord Hoffmann’s judgment in A(Alconbury Ltd) v Environment Secretary (HLCE) [2003] 2 AC 295 at 326 that: “73. There is however another relevant principle which must exist in a democratic society. That is the rule of law. When ministers or officials make decisions affecting the rights of individual, they must do so in accordance with the law. The legality of what they do must be subject to review by independent and impartial tribunals… The principles of judicial review give effect to the rule of law. They ensure that administrative decisions will be taken rationally, in accordance with a fair procedure and within the powers conferred by Parliament.” 21. Lord Hoffmann’s observations are not limited to “decisions affecting the rights of individuals” (Emphasis added). In the landmark decision of ex parte Doody 1994 1AC 531 at 560, Lord Mustill said with the concurrence of all their lordships, that “Where an Act of Parliament confers an administrative power, there is a presumption that it will be exercised in a manner which is fair in all the circumstances.” 22. It is beyond doubt, for example, that the Director cannot exercise his power on, for example, racial, colour or gender grounds.[14] The rule of law would not permit it. The legislature would not be taken to have ever intended it. Also, the law requires, and the legislature must have intended, that the Director would take into consideration relevant matters, and ignore irrelevant matters. This is how Lord Greene MR described this requirement: “For instance, a person entrusted with a discretion must, so to speak, direct himself properly in law. He must call his own attention to the matters which he is bound to consider. He must exclude from his consideration matters which are irrelevant to what he has to consider. If he does not obey those rules, he may truly be said, and often is said, to be acting ‘unreasonably’.”[15] 23. The Director’s case stated clearly that: “66. For the avoidance of doubt, it should be stated that even though it is the HKSARG’s position that there is no duty on its part to conduct RSD independently, the Director does independently consider the exercise of his power of removal in each case on its own merits. 67. It is therefore not accepted that if no independent RSD is conducted by the HKSARG, there is no practical and effective protection against refugee non-refoulement.” 24. Since it is rightly accepted, the Director would “independently consider the exercise of his power of removal in each case on its own merits”, what is required of the Director when such decisions are made is of pivotal importance to these appeals. 25. The Director accepted that a claim to refugee status is a relevant humanitarian or compassionate consideration. However, Mr Yu added, the Director is under no duty to have regard to any humanitarian or compassionate reason. This submission has to be taken in the context of Mr Yu’s express agreement that the Director would never say in response to a refugee claim “I don’t care, you will be returned.” 26. What then would the rule of law require of the Director’s decision whether or not to refoule a person who claimed to be a refugee? Prabakar 27. Mr Fordham relied on Secretary of Security v Sakthevel Prabakar (2004) 7 HKCFAR187, an unanimous decision of this Court. At one level, the issue in Prabakar was, in determining whether or not a person faced a well-founded risk of torture in the event of removal to the putative country of torture, whether the Secretary could rely solely on a determination by UNHCR which had rejected a claim to refugee status by that person. The court held that the Director could not do so. 28. LiCJ explained why: “14. A person could of course come within the protection of both Conventions. It could also be that a person is protected by the Refugee Convention but not the Convention Against Torture since a person could be persecuted, the test in the former Convention, in a manner which does not amount to torture as defined in the latter. But more importantly, for the purposes of this appeal, it must be noted that, having regard to their different provisions, a person who is outside the protection of the Refugee Convention may nevertheless be protected by the Convention Against Torture.” 29. But the decision in Prabakar went much further. It decided that the Secretary for Security, who had a policy not to refoule a torture claimant if a claim is well-founded, must determine whether the claim is well-founded. In such determination, Li CJ said: “43. The question in this appeal concerns the standards of fairness that must be observed by the Secretary in determining in accordance with the policy the potential deportee's claim that he would be subjected to torture if returned to the country concerned. One is concerned with procedural fairness and there is of course no universal set of standards which are applicable to all situations. What are the appropriate standards of fairness depends on an examination of all aspects relating to the decision in question, including its context and its nature and subject matter: R v Home Secretary, ex p Doody [1994] 1 AC 531 at p.560D-G. 44. Here, the context is the exercise of the power to deport. The determination of the potential deportee's torture claim by the Secretary in accordance with the policy is plainly one of momentous importance to the individual concerned. To him, life and limb are in jeopardy and his fundamental human right not to be subjected to torture is involved. Accordingly, high standards of fairness must be demanded in the making of such a determination.” 30. Moreover, although ultimately, it was for the Secretary to assess the materials and to come to an independent judgment[16]: “45. … having regard to the gravity of what is at stake, the courts will on judicial review subject the Secretary’s determination to rigorous examination and anxious scrutiny to ensure that the required high standards of fairness have been met. R v Home Secretary, ex p Bugdaycay [1987]] 1 AC 514 at p.531 E-G. If the courts decide that they have not been met, the determination will be held to have been made unlawfully.” 31. Here, Mr Fordham submitted that, in respect of refugee claimants, the Director must also assess the materials and come to an independent judgment whether the appellants’ fear of persecution was well-founded. Also, the Director’s determination must be subjected to “rigorous examination and anxious scrutiny to ensure that the required high standards of fairness has been met”. 32. Mr Yu submitted that Prabakar is distinguishable because, in Prabakar the court was concerned with CAT which, unlike the convention, was applied to HKSAR. 33. But the decision in Prabakar did not depend on the application of CAT to Hong Kong. It is trite that treaties do not form part of domestic law unless domesticated by legislation. The Secretary’s stance was that, “as a matter of Hong Kong domestic law, the Secretary has no legal duty to follow the policy”. 34. Further, although it was the torture claimant’s case that the Secretary was under a legal duty to follow the policy on the following bases: the Basic Law, the Bill of Rights, customary international law and legitimate expectation, Li CJ said: “4. … As the courtindicated at the outset of the hearing, it is unnecessary to decide this issue. For the purposes of this appeal, the court will assume without deciding that the Secretary is under a legal duty to follow the policy as a matter of domestic law. In proceeding on the basis of such an assumption, the Court must not be taken to be agreeing with the views expressed in the judgments below that such a duty exists.” 35. It is clear from the foregoing that it was not critical to the decision inPrabakar whether there was a legal duty. It sufficed that the genuineness of a torture claim was relevant to a determination whether or not to remove a person. 36. Mr Yu relied on Lau Kong Yung & Others v Director of Immigration (1999) 2 HKCFAR 300, where Li CJ, with the concurrence of other members of the Court, said in the context of persons who claimed to be a permanent resident under Art 24(2) of the Basic Law but who were in fact regarded as illegal immigrants under the Ordinance that: “… in relation to such a person the Director has no duty to consider humanitarian grounds in considering the making of a removal order against him. But he can take such grounds into account if he thinks it appropriate in the case in question.” 37. The reliance is misplaced. Lau Kong Yung was concerned with very different circumstances. Lau Kong Yung was not concerned with claims such as refugee claims with their much more serious consequences. There, the Court had matters such as family reunion in mind. In Lau Kong Yung, legislative provisions which had been struck down by this Court as unconstitutional, were restored by an interpretation of the Art 22(4) and Art 24(2)(3) of the Basic Law. Since the effect of the restored provisions was to prevent persons, many of whom, children of one or more Hong Kong parents[17], from rejoining their parents in Hong Kong, humanitarian reasons such as family union has less scope as a basis for resisting a removal order. 38. Mr Yu also sought to distinguish Prabakar because there the Secretary had a policy not to deport a person to a country where that person’s claim that he would be subjected to torture is well-founded. Mr Yu submitted that, here, the Director had no policy, but only a practice. Such labelling was rightly described by Andrew Cheung J (as he then was) in MA v Director of Immigration HCAL 10/2010 (unreported, dated 6 January 2011) as “a matter of semantics”. Since the Director[18] must exercise the powers under the Ordinance in a principled manner, it is surprising indeed if the exercise of such power is not facilitated by a practice or policy. It does not matter what label is used. 39. Mr Yu also submitted: “The concurrent finding of the Court below is that the relevant policy administered by the Director in making removal orders is that he may take into account humanitarian or compassionate grounds…” 40. Indeed, the Director may. To do so fairly and properly, the Director must consider the circumstances of the individual concerned. Since, whether a person is a refugee is a relevant consideration, it follows that if a person claims refugee status, before the Director exercises his/her power, the Director must determine whether the claim is well-founded. Otherwise the power would have been exercised in ignorance of a relevant consideration. That plainly is not permissible. 41. Mr. Yu went on to submit that Hartmann J and the Court of Appeal also made concurrent findings that “The Director does not have a policy which requires him to determine whether a removee is a refugee within the meaning of the 1951 Convention.” I am not persuaded that there was any such finding. But I will deal with Mr. Yu’s submission as stated. With respect,I believe he has confused the disease for the cure. When the complaint is that the Director should determine “whether a removee is a refugee…” because that is relevant to the Director’s exercise of power under the Ordinance, it is not a sufficient answer to say that the Director has deferred to or relied on UNHCR’s RSD. Since a decision of such moment attracts the high standards of fairness identified in Prabakar, the Director’s decision must meet such high standards of fairness. 42. It is unnecessary for present purposes to decide whether the Director when exercising the powers of removal under the Ordinance must have regard to humanitarian reasons when confronted with a refugee claim. It is sufficient that the Director does so in accordance with his practice, and that a relevant humanitarian reason is whether or not a person is indeed a refugee. Nor is it my decision that humanitarian reasons alone must decide the exercise of such power. In a suitable case, for example, where Art 1F or Art 33(2) applies, the Director might well think it right to insist on refoulement. Whether refoulement may be ordered for any other reason is not for decision now. 43. For the sake of completeness, I should mention that although CAT also covered “Cruel, Inhuman or Degrading Treatment or Punishment”(“CIDTP”), Prabakar was concerned with non-refoulement of a torture claimant because CAT only expressly prohibited non-refoulement in relation to torture. 44. In Ubamaka Edward Wilson v Secretary for Security and Another, FACV 15/2011 (unreported, dated 21 December 2012),this court was concerned with the deportation of a Nigerian national who claimed that upon return to Nigeria he would face CIDTP. This Court was concerned with the Director’s exercise of discretionary powers of removal or deportation under the Ordinance, and held that since non-derogable and absolute rights protected by Bill of Rights Art 3 were engaged. “160. …a sufficiently established threat of BOR Art 3 being violated by the receiving country if the deportee should be sent there constitutes a ground for restraining the Hong Kong Government from proceeding with the deportation.” 45. No doubt, following Ubamaka, the Government would adopt procedures which satisfy the high standards of fairness required in determining whether such threat was sufficiently established.[19] Adverse Consequences? 46. Mr Yu submitted that a danger of requiring HKSARG to conduct a RSD is that UNHCR might not assist a person screened in under such circumstances to settle elsewhere,with the result that HKSAR would be forced to allow these refugees to settle in Hong Kong. Mr McCoy, informed us on behalf of UNHCR that,since HKSAR would not grant asylum[20], the UNHCR, faithful to its mandate to assist established refugees, would continue to help them to settle in a safe country. 47. Mr Yu also submitted that a decision adverse to the Director might result in a flood of economic migrants. Mr Chu King Man,Principal Assistant Secretary (Security) said in his affidavit filed on 28 February 2007 at para11: “According to UNHCR HK, they received some 390,800, 1000,2400 refugee status claims in year 2003,2004,2005 and 2006 respectively. The number of outstanding refugee status claims has built up to some 2440 as at the end of January2007. According to a report submitted on 18 July 2006 to theLegCo by the UNHCR HK, only some 10-11% of the asylum seekers had been mandated as refugees.” 48. Mr Chu expressed concern that: “… to institute an RSD mechanism in Hong Kong will likely attract more economic migrants to Hong Kong in the hope of possible changes in the Government’s asylum policy…” 49. But as I have pointed out we are not concerned with the grant of asylum and nothing in this judgment calls into question the Government’s policy not to grant asylum. We are only concerned with potentially returning persons to countries where they have a well-founded fear of persecution on the grounds mentioned above. 50. The UNHCR, the Bar Association and the Law Society have advocated an unified and efficient system consisting of one domestic screening exercise covering torture, CIDTP and refugee claims to avoid duplication and to reduce unmeritorious and protracted claims. Presently, claims could be made sequentially to maximize delay. Their suggestion merits careful consideration. 51. In any event, the solution is not to reduce Hong Kong’s human rights standard. The rule of law has real consequences and effect must be given to them. 52. It was also submitted that to require the Director to conduct RSD would require much expense and expertise. 53. According to a Memorandum of Understanding made between the HKSARG and the UNHCR dated 20 January 2009, HKSARG has agreed, at its own cost, to supply appropriately experienced immigration officials (the secondees) to UNHCR “in order to provide services to UNHCR wholly in connection with refugee status determination related duties(“RSD duties”)”. The secondees’ duty: “13. … include conducting RSD interviews, drafting RSD assessments and formulating recommendations to UNHCR on the merits of refugee status applications, in accordance with UNHCR policies and standards. Recommendations on refugee status determination will be signed by the Secondee and reviewed and signed by UNHCR, which will take and sign the final decision. UNHCR will take full responsibility for these decisions.” 54. Thus, although the final decision is made by the UNHCR, the secondees already play an important role in RSD. So lack of expertise should not be a serious problem. In any event, as Prabakar shows, compliance with the high standards of fairness will require the Director to obtain relevant information and materials. 55. Mr Yu submitted that the Director is entitled to rely on UNHCR’s RSD. I would agree that the Director is entitled to give weight to a RSD by the UNHCR. But it is essential that the determination must be made by the Director and his duly authorized officers and that the determination must satisfy the high standards of fairness required. FBv Director of Immigration [2009] 2 HKLRD346 provides an illustration of what might be required. 56. To conclude, I am of the view that, given it is the practice of the Director, when deciding whether or not to exercise his power under the Immigration Ordinance to remove a refugee claimant to the country of putative persecution, to have regard to humanitarian considerations, and that whether such claim is well-founded, is a relevant humanitarian consideration, the Director must determine whether the claim is well-founded. Moreover, any such determination must satisfy the high standards of fairness required having regard to the gravity of the consequence of the determination. First Question 57. Both Hartmann J and the Court of Appeal were satisfied that refugee non-refoulement as expressed in Art 33 has become a rule of CIL but not a peremptory norm. The Court of Appeal also agreed with Hartmann J: “that the ordinance shows a clear legislative intent to give an unfettered discretion to the Director, sufficient to override the CIL of non-refoulement of refugees.” 58. I am indebted to counsel for their very helpful and interesting submissions on the first question. However, in view of my decision on the second ground, it is unnecessary for me to decide the first question and I express no view on any of the issues raised under the first question. Disposition 59. I would allow the appeal. 60. Since we have not had submissions on any relief which might be granted under the second ground, the parties are invited to agree a draft order for the Court’s approval within 14 days of the judgment. Failing agreement, the parties are at liberty to submit their respective drafts within 10 days thereafter. 61. I have had the benefit of reading the judgment of Sir Anthony Mason NPJ in draft. I wish to say, specifically, that I am in full and respectful agreement with paragraphs 81 & 82 of his lordship’s judgment. Costs 62. I also make an order nisi that the appellants should have the costs here and below (including the trial), such costs to be paid by the respondent and to be taxed if not agreed. If any party wishes to have a different order for costs, written submissions should be served on the other party and lodged with the Court within 14 days of the handing down of this judgment, with liberty to the other party to lodge written submissions within 14 days thereafter. In the absence of such written submissions, the order nisi will stand as absolute at the expiry of the time limited for those submissions. Mr Justice Bokhary NPJ: 63. There are parts of the world where some people are sometimes subjected to persecution on account of their race, religion, nationality, membership of a particular social group or political opinion. In the context of the present case, the word “persecution” means persecution on account of one or more those things. What happens when persons — each of these appellants being such a person — arrive in Hong Kong and ask not to be returned to the place from whence they came, saying that they would be subjected to persecution there if returned? On the evidence before the Court, the answer is that they will not be returned (or “refouled”, to use the legal term) if their fear is well-founded. That does not mean that they will necessarily be permitted to settle in Hong Kong. They may be sent elsewhere. But they will not be returned to the place from which they had fled. They will be permitted to remain in Hong Kong until they are resettled in some country to which they want to go and which is prepared to accept them. 64. So Hong Kong’s practice conforms with the principle of “persecution non-refoulement”. Such a practice necessarily includes — because it would be unworkable unless it includes — determining whether the asserted fear of persecution in the event of return is well-founded. Axiomatically it is for the official or officials through whom the government's “persecution non-refoulement” practice is conducted to make a determination on whether such fear is well-founded. In making such a determination, that official or those officials must proceed fairly, with due regard to the vulnerability of the person concerned and with a degree of care commensurate with what is at stake. If the determination is adverse to the person resisting return, the reasons for it must be given. And the reasons must be sufficient to meet the reasonable needs for judicial review purposes of that person’s legal advisers and of the court. 65. Nothing short of the foregoing would be compatible with the rule of law. And the government's present approach in these matters, though no doubt well-intentioned, falls far short of the foregoing. Although the government rightly recognizes the relevance of a well-founded fear of persecution in the event of return, no government official makes any determination on whether the asserted fear of persecution in the event of return is well-founded. All that the government does, it has told the Court, is to“take into account” the refugee status determinations made by the United Nations High Commissioner for Refugees (“the UNHCR”). The UNHCR’s immunity from legal process in regard to its official functions puts its refugee status determinations beyond the reach of judicial review. Even so, the government says that since the UNHCR’s refugee status determination in respect of each appellant is adverse to him, the way is now legally open to return each of them. No, it is not. 66. In respect of each of these three appellants, the government must, doing so properly as explained above, make its own determination on the question of whether his asserted fear of persecution in the event of return is well-founded and, doing so properly as explained above, give its reasons for any determination adverse to any appellant. I would allow each of these three appeals on the ground of the government’s failure to make its own determination on that question, and call upon the parties to submit an agreed draft order if one can be achieved or rival drafts otherwise. It is of course to be understood that provided that the legal requirements declared by the judiciary are complied with, it is up to the executive to choose what screening method to adopt. One of the choices open to the executive is to screen torture and persecution fears together where both are asserted. 67. Although this appeal can be disposed of as briefly as that, I should say a word further to explain why it has proved unnecessary to address the questions of international law which have been debated before us. Quite simply, this case is in truth covered by the Court’s decision in Secretary for Security v Prabakar (2004) 7 HKCFAR 187. It is immaterial for present purposes that the Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment 1984 extends to Hong Kong while the Convention Relating to the Status of Refugees 1951 does not. Even in Prabakar’s case, which was a torture non-refoulement case, the government did not accept that Hong Kong domestic law prohibits return that would put a person in peril of being tortured. The government did not accept that any such prohibition had become part of domestic law (whether via the Basic Law, the Bill of Rights, statute law, the common law, the application to us of the Convention Against Torture, customary international law, any combination of the foregoing or anything else). 68. As to that, I said in Prabakar’s case at para 68 that the government’s policy of not ordering any return that would put a person in peril of being tortured provided a sufficient basis for classic judicial review, the government having made no determination of its own on the asserted fear of torture in the event of return. There is no material difference between the “policy” admitted and averred by the government in that case and the “practice” which the government cannot deny in the present case. It is to be stressed that no slight whatsoever to the UNHCR is involved in the proposition that the government cannot simply rely on an UNHCR refugee status determination adverse to the person concerned. Indeed, that proposition is a vindication of the UNHCR’s stance, which is that it is for Hong Kong independently to enquire into whether the asserted fear of persecution in the event of return is well-founded. 69. As to costs, I would leave them to be dealt with on written submissions as to which the parties should seek procedural directions from the Registrar. In thanking all the lawyers in the appeal for the arguments prepared and presented to the Court, I make special mention of the fact that counsel and solicitors for the UNHCR have been so public-spirited and generous as to donate their valuable services free-of-charge. Sir Anthony Mason NPJ: 70. I am in broad agreement with the reasons for judgment of Tang PJ. My agreement is subject to such qualifications as appear from my reasons which follow. In these reasons I use the same abbreviations and acronyms as those used by Tang PJ. The threshold question - background 71. The threshold question is whether the relevant decisions of the Director to remove the appellants are subject to judicial review. It is no objection to the general availability of judicial review that the power of removal is expressed in wide and unqualified terms. 72. A statutory discretionary power, no matter how widely expressed, is necessarily subject to some limits. It must, for example, be exercised by the repository of the power. Other limits may arise from context of the power and from the purpose or purposes which it is designed to serve. Or the limits may arise from extraneous considerations giving rise to an abuse of power, such as bias and bad faith, which are naturally presumed to lie outside the scope of the statutory grant of power. Judicial review is available to correct an exercise of power that exceeds the limits set by the statutory grant of power or otherwise constitutes an abuse of power. 73. The exercise of a general statutory power may give rise to judicial review by reason of the way in which the power is exercised. One such example, which is relevant for present purposes, is where the decision-maker adopts a policy in order to provide guidance as to the way in which the power will be exercised and to promote consistency in its exercise. Or when the decision-maker makes a representation as to the way in which the power is exercised, including cases where the representation creates a legitimate expectation. 74. The adoption of a policy by a decision-maker exercising a very general discretion has the advantages of promoting certainty, consistency and administrative efficiency. It is, however, important that the policy adopted, whether general in character or confined to a class of persons, is not so rigid as to exclude the exercise of discretion by the decision-maker to consider the merits of the particular case and a willingness to depart from the policy, if need be, in a particular case, at least in the general run of cases. This is because the exclusion of a residual discretion as a result of a decision-maker applying a rigid policy might well, depending on the circumstances, be at variance with the very discretion created by the statute. However, where the decision-maker purports to exercise that discretion in accordance with the stated policy the manner of that exercise may be reviewed by the courts. The threshold question – what is the principle to be applied? 75. The common law of judicial review of administrative action has travelled a long way from its early beginnings in 1863 in Cooper v Wandsworth Board of Works[21] where a local authority was held to be under a duty to accord natural justice before exercising its power to demolish a person’s house. The history of the common law development of the remedy of judicial review has been marked by a number of abiding concerns, of which two are of immediate relevance. The first was: what is the legal or philosophical foundation for judicial review? The second, which is linked to the first, was: what class of administrative decisions or what class of discretionary powers attracts the remedy of judicial review. 76. One approach to the first of these two questions was the “autonomous” common law approach, encapsulated in the words of Byles J in Cooper v Wandsworth Board of Works: “although there are no positive words in a statute requiring that the party shall be heard, yet the justice of the common law will supply the omission of the legislature.”[22] A second approach was the statutory ultra vires theory based on the proposition that the power has been conferred on the decision-maker on the underlying (and fictional) assumption that the power is to be exercised only within the jurisdiction conferred, in accordance with fair procedures, and reasonably in a Wednesbury sense [23]. The House of Lords endorsed this approach in Reg v Hull University Visitor, Ex parte Page[24] and in subsequent cases[25]. 77. While in particular cases these approaches retain their utility, it is now accepted that the foundation of judicial review is the rule of law. The rule of law as a foundation has the advantage that it extends judicial review to the exercise of non-statutory powers, including prerogative and common law powers[26], and it is capable of extending to administrative powers of non-government agencies[27], these being powers the exercise of which would not necessarily have attracted judicial review on the statutory ultra vires theory. I therefore accept as a correct statement of the principle to be applied in resolving the threshold question in this case the observations of Lord Hoffmann in R (Alconbury) v Environment Secretary[28]: “There is however another relevant principle which must exist in a democratic society. That is the rule of law. When ministers or officials make decisions affecting the rights of individuals, they must do so in accordance with the law. The legality of what they do must be subject to review by independent and impartial tribunals.”[29] 78. His Lordship was then writing with reference to the context of European Law but there is no reason to doubt that his comments apply to English as well as European law. 79. Nor is there reason to doubt that the principle enunciated applies to administrative powers generally and that it is not limited to decisions affecting the rights of individuals. Lord Mustill in R v Home Secretary, Ex parte Doody[30] expressed the reach of the principle more generally when his Lordship, in discussing the standards of fairness, said: “where an Act of Parliament confers an administrative power there is a presumption that it will be exercised in a manner which is fair in all the circumstances.”[31] 80. In earlier times, the availability of judicial review was restricted by reference to the nature of the powers that were exercised or by reference to the character of the decisions that were made – the second of the two questions identified in para 75 above. These restrictions no longer have the force that they once had. So much appears from the statements by Lord Hoffmann and Lord Mustill. The limits of judicial review 81. Although judicial review has expanded beyond its earlier boundaries and now extends to administrative powers generally, it is important to recognize that it is subject to a number of limitations. This is not the occasion to discuss them generally. It is sufficient here to identify as examples two substantial and overlapping limitations, one arising from separation of powers considerations and the other arising from the requirement of justiciability. The separation of powers may deny jurisdiction to the courts when the function involved is exclusively the province of the legislature or the executive. Questions of justiciability may arise in connection with broad issues because they may involve a lack of judicially manageable standards, such as matters of economic or social policy. Questions of justiciability may also arise in connection with issues which have a political character and a high political content where a political rather than a legal solution may be called for. When questions of justiciability arise it may appear that the courts are not institutionally equipped or competent to deal with the issues for determination. 82. Other limitations on judicial review relate not so much to what I have termed the threshold question as to the scope (or grounds), standards and intensity of review and standards of fairness. Some decisions may be reviewable for procedural fairness for example, but not otherwise. And this Court has recognized that the courts should attach particular weight to the views and policies adopted by the legislature in appropriate cases, as it did in Lau Cheong v HKSAR[32]. In Fok Chun Wa v Hospital Authority[33] Ma CJ (in a judgment with which other members of the Court agreed) pointed to the need for the courts to allow a margin of appreciation in relation to legislative and executive decisions affecting macrosocio-economic rights. And the Court has recognized that what fairness requires depends on the circumstances of the case and that the standards will vary according to the circumstances[34]. In Lam Siu Po[35] Ribeiro PJ, in what was the leading judgment, drew attention to the need to avoid standards which involve “over-lawyering” or “over-judicialisation” of procedures in certain administrative and disciplinary tribunals. I mention these instances not because they are relevant to the disposition of the present case, but because they provide some illustrations of the limitations attaching to judicial review. Judicial review in this case 83. Once the principle identified in paragraphs 77-79 above is applied to the facts of the present case, the Director’s decisions to remove the appellants are subject to judicial review. The power to remove is a far-reaching administrative power. Moreover, the Director’s decisions are clearly of momentous importance to the appellants. The decisions deny the appellants’ claims for protection from what may be a well-founded fear of persecution in the state of their nationality and expose them to refoulement to that state. The decisions therefore affect the appellants in a very material and serious way. 84. This Court’s decision in Secretary for Security v Sakthevel Prabakar[36] supports this view. In that case this Court held that judicial review was available in relation to the Secretary for Security’s deportation of Prabakar who fled from Sri Lanka as a refugee and claimed that he was tortured in Sri Lanka and was in danger of being tortured again if he were to return there. It was the Secretary’s policy not to deport a person to a country where that person’s claim that he would be subjected to torture was well-founded. The policy reflected the safeguard in art 3(1) of the Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment which applied to Hong Kong. Without deciding whether the Secretary was under a legal duty to follow the policy as a matter of law, the Court subjected the Secretary’s decision to judicial review and applied high standards of fairness, an aspect of the matter to which I shall return shortly. Although Prabakar was a case in which Hong Kong was subject to a Convention obligation reflected in art 3 of the Bill of Rights unlike the present case, it was a decision on the statutory power of deportation where the Secretary’s decision, like the Director’s decisions here, involved his adherence to a policy which he formulated in order to provide guidance and consistency in the exercise of the statutory power. In the context of the availability of judicial review in this case, it is the relation between the exercise of the statutory power and the application of the policy that is of crucial importance rather than the legal foundation for the policy, though that is a relevant factor. 85. Reg v Home Secretary, Ex parte Bugdaycay[37] likewise supports the availability of judicial review in the present case. There a question arose whether the administrative decision to remove the appellant Musisi to Kenya could properly be challenged in judicial review proceedings[38]. The House held that the Secretary of State had failed to consider the evidence that there was a danger that his removal would result in his removal to a third country where he had a well-founded fear of persecution. Lord Bridge of Harwich (with whom Lord Brandon of Oakbrook, Lord Griffith and Lord Goff of Chieveley concurred) said: “The action of an authority entrusted by Parliament with decision-making can be investigated by the court: ‘with a view to seeing whether they have taken into account matters which ought not to take into account, or, conversely, have refused to take into account or neglected to take into account matters which they ought to take into account’: per Lord Greene M.R. in Associated Provincial Picture Houses Ltd. v Wednesbury Corporation [1948] 1 K.B. 223, 233-234. In my opinion where the result of a flawed decision may imperil life or liberty a special responsibility lies on the court in the examination of the decision-making process.”[39] See also R v Home Secretary, Ex parte Doody[40]. Standard of Review 86. In the light of the nature of the Director’s decision and its drastic consequences for the appellants, it follows from Prabakar in particular and other cases mentioned that the Court is bound to: “subject the [Director’s] determination to rigorous examination and anxious scrutiny to ensure that the required high standards of fairness have been met …. If the courts decide that they have not been met, the determination will be held to have been made unlawfully”, in the words of Li CJ in Prabakar[41]. The Director’s policy 87. What is the Director’s policy? In answering this question I put aside as an unnecessary distraction the Director’s submission that, whatever it may be, it is a practice not a policy. In the context of this case, the submission amounts to a distinction without a difference. 88. Mr Benjamin Yu, SC for the Director, then submitted that there was a concurrent finding by the courts below that the relevant policy administered by the Director in making a removal in the case of claimants for refugee is that he may take into account humanitarian or compassionate grounds. I agree that the Director may do so. But that seems to be neither a policy nor a practice. It is simply a statement about the scope of his power of removal under the Ordinance. 89. Further, this is not a case in which this Court’s practice of not interfering with concurrent findings of fact in the courts below is engaged. That practice was comprehensively reviewed by Ribeiro PJ in Chinachem Charitable Foundation Limited v Chan Chun Chuen[42]. Here the findings of fact were made by the courts below on different materials. The Memorandum of Understanding (MOU) between the Government of the HKSAR (HKSARG) and the Office of the UN High Commissioner for Refugees dated 20 January 2009, which provided for the UNHCR to make RSDs, only came to light in the proceedings in the Court of Appeal. So the findings of fact made by Hartmann J at first instance were made in the absence of knowledge of the MOU which is of great materiality to the existence of the Director’s policy. Hence this is not a case of concurrent findings of fact which is an expression applied to findings of fact based on substantially the same materials. This conclusion also answers another submission of the Director that there were concurrent findings of fact that the Director had no policy requiring him to determine whether a claimant is a refugee within the meaning of the Convention. 90. Once regard is brought to the MOU, which I shall examine shortly, the Director’s policy is to be properly understood as a policy of not returning to their country of nationality claimants for refugee status whose fear of persecution is well-founded. It is that policy that enables the Director to say that, although he is under no obligation to comply with the Convention, he voluntarily complies with its requirements. 91. The processing of claimants for refugee status in Hong Kong is conducted in accordance with the MOU. The MOU recites that the UNHCR maintains a sub-office in the HKSAR to conduct “refugee status determinations” (RSDs) under its UN mandate. Clause 21 acknowledges the position of the HKSARG when it provides: “The HKSARG considers that it has no obligation to grant asylum and process refugee status claims under the … Convention … and its 1967 Protocol or otherwise and that it is under no obligation to investigate, consider or determine any refugee status claims.” 92. The MOU provides that the HKSARG will second appropriately experienced personnel to UNHCR to perform RSD duties (cl 2), their remuneration to be paid by the HKSARG (cl 7). The Secondees will comply with all rules, regulations and instructions issued by UNHCR regarding their duties made available to them (cl 11(d)). The Secondees are to perform RSD duties, conducting RSD interviews, drafting RSD assessments and “formulating recommendation to UNHCR on the merits of refugee status applications in accordance with UNHCR policies and standards” (cl 13). Recommendations on refugee status determination “will be signed by the Secondee and reviewed and signed by UNHCR which will take and sign the final decision. UNHCR will take full responsibility for these decisions” (cl 13). 93. The MOU contains provisions dealing with the provision of information by the UNHCR to the HKSARG. The information which can be provided, set out in Appendix B to the MOU, relates to “assessment of refugee status application of asylum seekers” who have authorized the sub-office in writing to release and the HKSARG to receive such information (cl 22). The information to be provided includes case summaries, the grounds for rejection of refugee status, information relating to the claimant’s credibility and whether the claim is considered well-founded, as well as “the final decision” of the UNHCR on the claim. The Director’s decisions 94. It follows from the provisions of the MOU that, as between the HKSARG and the UNHCR, that the UNHCR takes the “full responsibility” for making the RSDs of refugee status claimants in Hong Kong. As far as the MOU is concerned, the Director’s participation is limited to the part played by his “secondees” in the processing of applications. No doubt it was in the interests of both the HKSARG and the UNHCR to vest the responsibility for making the RSDs in the UNHCR. 95. Nonetheless the Director’s case filed in this Court asserts that: “the Director does independently consider the exercise of his power of removal in each case on its own merits.” (para 66) This assertion calls for several comments. First, it is a statement only and is not supported by evidence. Secondly, it is a statement that does not tell us what is involved in the consideration of each case “on its own merits”. Thirdly, we were informed that in all the cases where the UNHCR has made an RSD in favour of a claimant, namely that he has a well-founded fear of persecution in his country of nationality, the Director has not returned him to that country and has allowed him to remain in Hong Kong pending re-settlement. 96. In a case of this kind, one would have expected the Director to adduce evidence showing in detail the consideration he gives to claims for refugee status after the UNHCR makes its “final” determination of refugee status, including his receipt of the UNHCR case summaries and the materials identified in the Appendix to the MOU. No such evidence has been forthcoming, even in the matters relating to the appellants. 97. There is, of course, no basis for saying that the Director has failed to exercise his statutory power of removal, but there are very strong reasons for concluding that the Director has either failed to apply his mind independently to the correctness of the determinations made by the UNHCR or, if he has done so, he has done so in a way that falls short of the anxious scrutiny and high standards of fairness required by Prabakar. It is not sufficient for the Director simply to rely on the UNHCR determinations, as his counsel contended. It is, of course, legitimate for the Director to give weight to the UNHCR determination but not to simply rely on it. 98. It is no answer to the Director’s failure to make an independent assessment to say that the power of removal is broad and unqualified and that it imposes upon him no duty to make an RSD. The fact is that the Director has, under statutory authority, adopted a policy the object of which is to exercise his power of removal according to a determination of the refugee status of a claimant to that status. Indeed, the HKSARG asserts publicly that, although not bound by the Convention, it nonetheless voluntarily complies with its requirements. Having adopted that policy in these circumstances, no doubt by reason of the powerful humanitarian considerations which are involved in RSD determinations and the consequences they may entail, the requirement of fairness, arising from the adoption by the Director of a policy under the authority of the statute, calls for him to make an independent assessment of the UNHCR determination, especially in those cases where the UNHCR determination is adverse to the claimant. In making that assessment, the Director must observe high standards of fairness. Conclusion 99. I conclude therefore that the appellants succeed on what Tang PJ has identified as their second ground of appeal. To that end I would allow their appeal. 100. For the sake of clarification, I point out that my reasons and my conclusion relate only to the exercise of the Director’s power of removal pursuant to the policy which I have identified. If the Director were to put an end to the policy or adopt another policy, different considerations would apply. Orders 101. I agree with the order proposed by Tang PJ as to costs. Mr Justice Chan PJ: 102. The Court unanimously allows the appeals and makes the order and directions as proposed in paragraphs 60 and 62. Mr Michael Fordham, QC, Ms Gladys Li, SC & Mr Simon Young, instructed by Barnes & Daly assigned by the Legal Aid Department, for the appellants Mr Benjamin Yu, SC, Professor Vaughan Lowe, QC, Mr Anderson Chow, SC & Ms Grace Chow, instructed by the Department of Justice, for the respondents Mr Gerard McCoy, SC & Mr Timothy Parker, instructed by Baker & McKenzie, for the intervener [1] References are to articles in the Convention unless otherwise stated. [2] The word refugee is used in this sense throughout the judgment. [3] The prohibition against refoulement applies to a person who has the requisite fear of persecution who has entered a receiving state. R (European Roma Rights Centre) v Immigration Officer at Prague Airport [2005] 2AC 1. However, under Art 1F and Art 33(2), a person is not entitled to the protection of the Convention if, for example, that person has committed a crime against humanity or is a person in respect of whom there are reasonable grounds for regarding as a danger to the security of the receiving country. A refugee may be removed to a safe third country. R v Home Secretary, ex p Bugdaycay [1987] 1 AC 514. [4] It is the Director’s case that persons who claim protection as refugees would be told that they should apply to the UNHCR. [5] Greater details can be founded in the judgment of Hartmann J (as he then was) at first instance. [6] The UNHCR was given leave to intervene, and was represented by Mr Gerard McCoy SC, and Timothy Parker, both of whom appeared on a pro bono basis. [7] Mr Benjamin Yu SC for the Director correctly raised no procedural objection to this submission, which was fully argued before us. I note that in Bugdaycay, the sole successful appellant, Musisi, succeeded on a ground which was raised for the first time in the House of Lords. (P.526H). [8] As noted in para 7 above, it was also the Director’s practice not to return refugee claimant pending determination by the UNHCR. [9] And his duly authorized officers. [10] eg s11. [11] eg under s18 & s19. [12] Per Litton JA (as he then was) in R v Director of Immigration, ex p Chan Heung-mui and Others (1993) 3 HKPLR 533 at 545. [13] A similar and longer passage in the 5th edition was cited with approval in R v Tower Hamlets London Borough Council Ex parte Chetnik Developments Ltd [1988] 1 AC 858 at 872, by Lord Bridge of Harwich in a judgment which had the concurrence of all their lordships. [14] This is so, even in the absence of any constitutional or human rights consideration. Such grounds would be as irrational as the red-haired teacher of Wednesbury fame. Warrington LJ in Short v Poole Corporation [1926] Ch 66 at 91 gave the example of the red-haired teacher, dismissed because she had red hair. [15] Associated Provincial Picture Houses Ltd v Wednesbury Corporation [1948]1KB 223, at 229. [16] At para 60 per Li CJ. [17] But neither parent was at the time of these persons’ birth entitled to a right of abode under Article 24(2)(2) or (3). [18] And his duly authorized officers. [19] In a press release dated 18 February 2013 in response to statements issued by the Hong Kong Bar Association and the Law Society, a spokesman on behalf of the Immigration Department said “The Immigration Department “is actively exploring the various possible options to ensure the relevant duties are carried out in accordance with the CFA’s judgment” in Ubamaka. The Immigration (Amendment) Ordinance 2012 has introduced a new Part VII C to the Immigration Ordinance to deal with torture claimants as a result of Prabakar. [20] It is not contended that the HKSAR is obliged to offer asylum. [21] (1863) 14 CB(NS) 180; 143 ER 414. [22] Ibid at 194 [6]. [23] See Reg v Hull University Visitor, Ex parte Page [1993] AC 682 at 701. [24] Ibid. [25] See Boddington v British Transport Police [1999] 2 AC 143 at 173; Reg v Home Secretary, Ex parte Doody [1994] 1 AC 531 at 560. [26] Council of Civil Service Unions v Minister for the Civil Service [1985] 1 AC 374 (the action under review was an executive act but it was enough that it had its source in the common law and its exercise gave rise to justiciable issues). [27] See, for example, R v Panel on Take-overs and Mergers; Ex parte Datafin Plc [1987] 1 QB 815. [28] [2003] 2 AC 295. [29] Ibid at 326B. [30] [1994] 1 AC 531 at 560. [31] Ibid at 560D-E. [32] (2002) 5 HKCFAR 415 (where the Court, after giving particular weight to the legislative judgment, concluded that the imposition of a mandatory life sentence for murder based on the infliction of grievous bodily harm was not “arbitrary” for the purposes of art 28 of the Hong Kong Bill of Rights. [33] FACV 10/2011 (2 April 2012), paras 10, 61-81. [34] Stock Exchange of Hong Kong v New World Development Co Ltd (2006) 9 HKCFAR 234 at 271 para 109 (where, for example, the risk of “over-lawyering” was taken into account in arriving at what was “fair” in disciplinary proceedings. [35] (2009) 12 HKCFAR 237 at paras 69-71. [36] (2004) 7 HKCFAR 187. [37] [1987] 1 AC 514. [38] Ibid at 526G-H. [39] Ibid at 537F-H. [40] [1994] 1 AC 531 at 560D-G. [41] [2004] 7 HKCFAR at 204H. [42] Miscellaneous Proceedings No 20 of 2011 (Civil), decision delivered on 24 October 2011, reasons handed down on 28 October 2011. 
Press Summary (English) Press Summary (Chinese) FAMV No. 591 of 2022 [2022] HKCFA 23 IN THE COURT OF FINAL APPEAL OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION MISCELLANEOUS PROCEEDINGS NO. 591 OF 2022 (CIVIL) (ON APPLICATION FOR LEAVE TO APPEAL FROM CACV NO. 425 OF 2022) Appeal Committee: Chief Justice Cheung, Mr Justice Ribeiro PJ and Mr Justice Fok PJ Date of Hearing: 25 November 2022 Date of Determination: 28 November 2022 DETERMINATION THE APPEAL COMMITTEE: A. The decision at first instance 1. On 19 October 2022, Poon CJHC[1] granted the application of Mr Timothy Wynn Owen KC (“Mr Owen KC”) for ad hoc admission to represent Mr Lai Chee Ying (“Mr Lai”) at a trial[2] in which Mr Lai faces four charges involving a conspiracy in relation to seditious publications, contrary to sections 10(1)(c), 159A and 159C of the Crimes Ordinance[3]; and conspiracies to collude with a foreign country or external elements to endanger national security, contrary to Article 29(4) of the National Security Law[4] (“NSL”). The trial is due to commence on 1 December 2022 and is scheduled to last until 20 January 2023. 2. The power to grant or refuse such an admission arises under section 27(4) of the Legal Practitioners Ordinance[5] which confers a broad discretion on the court. 3. In deciding whether to admit the applicant, Poon CJHC applied the principles developed by the courts over the years,[6] as summarised in Re Perry QC.[7] The overriding consideration is whether it is in the public interest to admit the overseas counsel in question and, as Poon CJHC held, factors relevant to that consideration are “(1) the importance of the legal issues to Hong Kong’s jurisprudence; (2) the complexity and difficulty of the issues; (3) whether the overseas counsel would add [a] significant dimension to the case; (4) the availability of suitable local counsel”.[8] 4. Applying each of those factors, taking into account the issues which the parties had indicated were likely to arise at the trial,[9] Poon CJHC exercised his discretion in favour of granting the admission. B. The Court of Appeal’s decision 5. On the appeal of the Secretary for Justice (“SJ”) to the Court of Appeal,[10] leading counsel appearing for the SJ[11] recognised that it was an appeal against the Judge’s exercise of discretion and acknowledged the principles applicable on such appeals.[12] As summarised by the Court of Appeal, those principles are that: “The appeal court must defer to the judge’s exercise of discretion and must not interfere with it merely on the ground that it would have exercised the discretion differently. The function of the appeal court is initially one of review. It can only interfere in three instances: (1) where the judge has misdirected himself with regard to the principles in accordance with which his discretion had to be exercised; (2) where the judge, in exercising his discretion, has taken into account matters which he ought not to have done or failed to take into account matters which he ought to have done; or (3) where his decision is plainly wrong. It is only if and after the appeal court has reached the conclusion that the judge’s exercise of discretion must be set aside for one or more of these reasons that it becomes entitled to exercise an original discretion of its own.”[13] 6. After examining each of the ways in which it was alleged that the Judge had erred in the exercise of his discretion, the Court of Appeal observed that the SJ’s challenge “essentially goes to the weighing exercise of the judge of the relevant aspects of the public interest engaged in striking a proper balance in the exercise of his discretion.”[14] The Court was not persuaded by counsel’s arguments and dismissed the appeal, concluding: “There is no valid basis to interfere with the exercise of the judge’s discretion. If we were to exercise the discretion afresh, we would have reached the same conclusion.”[15] C. The application to the Court of Appeal for leave to appeal 7. The SJ applied to the Court of Appeal for leave to appeal to this Court, represented by Mr Rimsky Yuen SC who appeared in place of his two predecessors. In that application, the SJ sought to advance very different arguments in the Court of Final Appeal from those made in the Courts below. 8. In particular, as set out in the notice of motion (“NM”), the SJ contended that the established principles for dealing with ad hoc admissions were “inapposite” for applications involving NSL cases and that a radically different approach should be adopted.[16] The principle he contended for was that in NSL cases: (a) ad hoc admissions of overseas counsel should generally be refused “save in exceptional circumstances”; (b) the burden of establishing such exceptional circumstances is on the applicant; and (c) unless that burden is discharged, admission should be refused.[17] In other words, it was argued that there should be a blanket ban on ad hoc admissions of overseas counsel in NSL cases, subject only to undefined “exceptional circumstances”. This would effectively mean that the court should be deprived of its statutory discretion. 9. The SJ’s main arguments for adopting this fundamentally different approach were: (a) That the NSL is unique, being a PRC law from a continental legal system where the emphasis is on the need to protect the PRC’s sovereignty and security and HKSAR’s interests within the principle of one country two systems;[18] (b) That overseas counsel from England have little to offer in terms of contributing to the development of the HKSAR’s NSL jurisprudence given that their experience is essentially confined to the common law;[19] (c) That the NSL addresses inter alia “interference in the HKSAR’s affairs by foreign or external forces” and ad hoc admissions in NSL cases “would generally tend to defeat that aim in that, among others, there is no meaningful or effective enforcement of overseas counsel’s confidentiality obligations under NSL 63 over State secrets and other confidential information which he or she comes to know in the practice of law”;[20] and (d) That the duty to prevent, etc, any act or activity endangering national security “includes [preventing] any possible attempt to use the legal process to compromise the protection of national security… and that such need [is] of paramount importance in considering the public interests in an application of this nature”.[21] 10. Those are indeed novel arguments. The SJ accepted, however, that ad hoc admissions of overseas counsel involve an exercise of the court’s discretion. He also accepted that “the arguments advanced in support of the intended appeal go beyond those made before the Court of Appeal”. The SJ recognised that the Court, applying the Flywin principle,[22] might refuse to entertain an entirely new case not explored below, but argued that there was no issue as to the state of the evidence and submitted that an exception to the Flywin principle should be made because it was in the public interest to have the Court of Final Appeal deal with the points of principle sought to be raised.[23] D. The Court of Appeal’s decision on leave to appeal 11. The Court of Appeal handed down its judgment refusing leave to appeal on 21 November 2022.[24] It held that in so far as the SJ sought to raise new points, the circumstances were not of such an exceptional nature as to justify departing from the Flywin principle.[25] 12. The Court of Appeal also held that in any event, two of the new points sought to be raised were not reasonably arguable.[26] 13. The first such point concerned the SJ’s suggestion that the ad hoc admission might tend to defeat the aim of preventing “interference in the HKSAR’s affairs by foreign or external forces”.[27] Mr Yuen SC pointed to Articles in the NSL[28] designed to protect State secrets and other sensitive information. He also drew attention to the duty of confidentiality imposed by NSL 63 on defence lawyers and others who acquire knowledge of State secrets, trade secrets or personal information while handling the case and submitted that an important consideration was that such duty could not be enforced against overseas counsel after their leaving Hong Kong. The Court of Appeal held however, that there was no indication that State secrets or such sensitive information were involved in the case, that English counsel were subject to discipline under their own code of conduct in respect of practice abroad and that no basis had been advanced for suggesting that Mr Owen KC might breach the confidentiality requirement.[29] 14. Secondly, the Court of Appeal held that it was not reasonably arguable that the radically different approach to ad hoc admissions referred to above should be adopted, since it would fetter the discretion granted to the Court by statute and prevent it from approaching such admissions in a flexible and sensible manner.[30] E. The present application to the Appeal Committee 15. On 22 November 2022, the SJ lodged a Notice of Motion (“Form B”) seeking leave to appeal from the Appeal Committee on an urgent basis against the Orders for ad hoc admission granted by the Courts below. 16. The questions enumerated in Form B are set out in the Annex to this Determination. Subject to the “or otherwise” addition mentioned below, it is in the same terms as the notice of motion filed in the Court of Appeal, referred to in Section C of this Determination. Thus, the SJ repeats his submission that the radically different approach to ad hoc admissions involving NSL cases should be adopted (Annex Q2 §2, see §8 above). He also repeats his main arguments for adopting that fundamentally different approach (Annex Q1 §1, see §9 above). 17. The SJ additionally seeks leave to appeal on the “or otherwise” basis (Annex §3(1)-(4)) on the grounds (i) that an important question is raised on this first application for ad hoc admission in a case involving national security; (ii) that the Courts below failed to appreciate the unique features of the NSL making such admissions incompatible with the objective and design of the NSL (repeating in substance the submissions contained in Annex Q1 §1); (iii) that the applicant’s allegation that refusal of admission would undermine the perception of fairness in the criminal trial wrongly suggests that the HKSAR courts assisted by the local bar will not be able to secure a fair trial for Mr Lai; and (iv) that the overall circumstances of the case do not justify ad hoc admission. Subject to what is said below in connection with the Flywin doctrine, the SJ’s skeleton argument essentially reiterates the points made in the application for leave made before the Court of Appeal. 18. The skeleton submissions lodged on the applicant’s behalf, reiterated by Mr Robert Pang SC in his oral submissions, rely on the Flywin principle as a basis for dismissing the leave application and contend that various points made by the SJ are not reasonably arguable. 19. The Bar Council’s brief submissions support the SJ’s argument that the unique status of the NSL is a cogent factor weighing against admission. At the same time, they recognise that the proposed appeal is against an exercise of discretion which faces significant hurdles. The Bar states that it “will leave it to the SJ to persuade the Court”. F. The Flywin principle 20. It is well-established that when an application is made for leave to appeal on a new point which has not been considered in the courts below, the Flywin doctrine[31] applies as a discretionary principle.[32] That doctrine has two aspects which bear on fairness to the other party and on the court’s ability properly to adjudicate upon the matter. 21. The first aspect has been referred to as the “state of the evidence” bar. As the Court explained: “... where a party has omitted to take a point at the trial and then seeks to raise that point on appeal, the position is as follows. He will be barred from doing so unless there is no reasonable possibility that the state of the evidence relevant to the point would have been materially more favourable to the other side if the point had been taken at the trial.”[33] 22. That consideration therefore arises where the new point involves factual issues which have not been explored below because, the matter not having been raised, potentially relevant evidence has not been adduced and the issue has not been addressed in argument. To allow such a new point to be taken on appeal is obviously likely to entail unfairness to the other party. 23. Such issues of fairness are not confined to the absence of relevant evidence. They also arise where failure to take the point may affect the way the case was conducted, the arguments advanced and the course taken by the proceedings in the courts below. As was explained in Lehmanbrown Ltd v Union Trade Holdings Inc: “... it is ... clearly and firmly established that new points which are fact sensitive or otherwise affect the course of evidence or conduct of the case at the hearing below should not be allowed. Though this principle is usually applied in situations where the new points necessitate further evidence to be adduced, it is not confined to such scenarios. Very often, the raising of new point by one party may lead to the other party raising counter arguments and the consideration of such counter arguments may involve factual assessment in a different light from that undertaken by the court below. Sometimes, it may involve a different weighing of factors in the exercise of discretion. Alternatively, the other party may embark on a different course of forensic conduct if the new point were taken earlier. The appellate court, in considering whether the new point would be entertained, is entitled to take these matters into account in order to avoid unfairness to the other party.”[34] 24. In such cases, failure to take the point below has an impact on the fairness of the proceedings. It also hampers the Court’s ability to adjudicate effectively on the appeal. 25. The second aspect of the Flywin doctrine has been referred to as “the not considered on intermediate appeal” hurdle. As explained in Flywin, it is only in the most exceptional circumstances that the Court of Final Appeal will entertain an appeal on a new issue, not fully explored and argued below, when it involves a major development of the law: “Even where a new point is not barred [by the state of the evidence bar], there is still a hurdle facing a party who seeks to raise in the final appellate court a point which was not pursued in the intermediate appellate court. As Chief Justice Li said in Wong Tak Yue v Kung Kwok Wai & Another (No 2) (1997-98) 1 HKCFAR 55 at p.66E-G: ‘The Court of Final Appeal as a court of final appeal should be very reluctant to consider an issue which was not duly raised and considered in the Court of Appeal. The circumstances must be very exceptional.’ This is especially so when entertaining the new point and accepting it would constitute a major development of the law. In such an exercise, it is to be expected that having the intermediate appellate court’s view on the point of law involved would be of assistance to the final appellate court. Therefore where the question of a major development of the law is involved, the foundational consideration underlying the ‘not considered on intermediate appeal’ hurdle, as I propose to call it, is that when the courts develop the law, it is best that they do so with all their intellectual resources fully deployed.”[35] 26. Applying this aspect of the doctrine, the Court of Appeal in The Law Society of Hong Kong v A Solicitor,[36] refused leave to appeal to the Court of Final Appeal, stating: “As explained by Kwan JA in Ageas Insurance Co Ltd v Lam Hau Wah Inneo CACV 65 of 2014, 19 May 2015 at [7] to [9] (footnote omitted), there is also a practice of this court not granting leave for new points (not restricted to new points involving fact-sensitive issues) which had not been canvassed in the substantive appeal before us. ... The rationale for this practice, as explained by Bokhary PJ in Flywin Co Ltd v Strong Associates Ltd (2002) 5 HKCFAR 356 at [39], stems from the undesirability and disadvantages of asking the Court of Final Appeal, as the highest judicial authority in this jurisdiction, to determine a point of great general or public importance without the benefit of earlier rounds of full argument and judgments of the lower courts.” G. The Appeal Committee’s Determination 27. In our judgment, the SJ’s application for leave to appeal, seeking to raise radically new points which were not mentioned or explored either before Poon CJHC or the Court of Appeal, notwithstanding their obvious importance, clearly fails to surmount the well-established hurdles posed by Flywin. 28. The Questions set out in the Annex (including the matters raised on the “or otherwise” basis) self-evidently give rise to a host of further issues which have not been examined factually nor canvassed in argument on intermediate appeal. For instance, the new principle advocated in Question 2 §2, prompts the questions: Why should there be effectively a blanket ban on ad hoc admissions of all overseas counsel in relation to all NSL cases without differentiation? What kinds of matters constitute “exceptional circumstances” which would justify admissions? How would the applicant be required to discharge the burden of establishing the exception? At the hearing, Mr Yuen SC declined the invitation to indicate the nature of the “exceptional circumstances” contemplated. 29. Aspects of Question 1 raise factual issues that have not been explored in argument or supported by evidence. Thus, Question 1 §1(2) contains the suggestion that admission of any overseas counsel would tend to defeat the aim of countering “interference in the HKSAR’s affairs by foreign or external forces”. It is a suggestion which cries out for elaboration and evidential support: How would any particular ad hoc admission result in such interference? The Question as formulated also raises concerns as to the lack of effective enforcement of overseas counsel’s confidentiality obligations under NSL 63 “over State secrets and other confidential information”. If the case does involve State secrets and such confidential information (which has not been suggested in the present case), that would obviously be a highly important, and possibly crucial factor which the Court would be bound to take into account in the exercise of its discretion. The involvement of any such sensitive materials ought to have been properly raised and fairly explored factually and as a matter of law in the Courts below, rather than being first raised as an unsubstantiated new point when applying for leave to appeal to the Final Court. 30. Similarly, the suggestion in Question 1 §1(3) regarding some “possible attempt to use the legal process to compromise the protection of national security” cries out for elaboration, factual support and a fair exploration of the issue in the Courts below if it is to provide a basis for leave to appeal at the present stage. 31. In his skeleton argument,[37] the SJ submits that the Court of Appeal has in substance dealt with the ‘new’ points in its leave judgment[38] so that, he argues, the CFA “does have the benefit of the views of the intermediate appellate court”. With respect, that is untenable and at the hearing, Mr Yuen SC recognised that the points touched on by the Court of Appeal in its leave decision did not reflect a full exploration of the issues. As indicated in the foregoing paragraphs, the new points generate numerous issues that have not been explored below, whether factually or as a matter of law. The Court of Appeal made it clear that it was dismissing the leave application applying Flywin principles and that the matters raised by the SJ did not justify permission exceptionally to be granted.[39] The Court of Appeal added that it considered two of the new points as advanced in any event not to be reasonably arguable. 32. Accordingly, we hold that the SJ has not made out a proper case for the grant of leave to appeal in respect of the radically new points he seeks to advance and his application must be dismissed. We note that the Court of Appeal held that certain questions raised are not reasonably arguable. As we have declined leave to appeal on the Flywin principle, it is unnecessary for us to enter into that discussion. We consider the concern expressed by the SJ[40] that the Court of Appeal’s judgment will become an undesirable precedent is unfounded. That Court’s refusal of leave to appeal is entirely case-specific and does not constitute a precedent. 33. The courts of the HKSAR are of course fully committed to safeguarding national security and to acting effectively to prevent, suppress and impose punishment for any act or activity endangering national security as required by NSL 3.[41] That duty would unfailingly be carried out whenever national security issues are properly raised and duly explored, enabling the courts to undertake a proper adjudication of those issues. Thus, in relation to ad hoc admissions, where national security considerations properly arise, such considerations are plainly of the highest importance to be taken into account. In the present case, however, the SJ has fundamentally changed his case only at the stage of seeking leave to appeal to this Court, raising undefined and unsubstantiated issues said to involve national security which were not mentioned or explored in the Courts below. No appropriate basis has been made out for the grant of leave to appeal. 34. Accordingly, we dismiss the application and make an Order nisi that the SJ pay the applicant’s costs of this application with liberty to the SJ to lodge written submissions as to costs within 7 days from the date of this Determination and liberty to the applicant to lodge written submissions in reply, if any, within 7 days thereafter. We make no order as to the Bar Council’s costs. Mr Rimsky Yuen SC and Mr Martin Ho, instructed by the Department of Justice, for the Secretary for Justice (Applicant) Mr Robert Pang SC, Mr Steven Kwan, Mr Ernie Tung and Mr Colman Li, instructed by Robertsons, for the Respondent Mr Jat Sew Tong SC and Mr Jonathan Fung, instructed by Anthony Chiang & Partners, for the Bar Council of the Hong Kong Bar Association (Interested Party) ANNEX (Form B) The SJ’s Formulated Questions “Question 1 1. In cases concerning national security or the application of the [NSL] (collectively ‘cases concerning national security’), what is the proper approach in considering the question of public interests in the ad hoc admission of overseas counsel under section 27(4) of the Legal Practitioners Ordinance (Cap 159)? In particular, whether the public interests arising from the uniqueness of NSL should generally take precedence and/or override other public interest considerations (though not necessarily conclusive): (1) (a) First, national security law is inherently unique by nature. In the present case, the uniqueness of the NSL as a piece of legislation drafted: (i) as a national law of the People’s Republic of China (“PRC”) under the legal system of the PRC which is a continental legal system, and (ii) in the specific and unique circumstances cumulating to and prevailing at the time which gave rise to the need to safeguarding national sovereignty and security of the PRC as well as the interests in the Hong Kong Special Administrative Region (“HKSAR”) against the background of the ‘one country, two systems’ principle. (b) Considerations involved in the ad hoc admission of overseas counsel in cases concerning the interpretation and/or application of provisions contained in the Basic Law of the HKSAR (“Basic Law”) are generally inapposite and do not support ad hoc admission of overseas counsel in cases concerning national security, since: (i) the background, nature, objectives and characteristics of the Basic Law and those of the NSL are very different notwithstanding that both ultimately seek to ensure the implementation of “one country, two systems” principle; (ii) the NSL contains inter alia substantive PRC criminal law provisions concerning national security; (iii) an additional and overriding consideration in cases concerning national security is that no institution, organisation or individual in the HKSAR shall contravene the ‘one country, two systems’ principle in exercising their rights and freedoms (NSL 2), whereas this unique consideration is absent from past cases involving ad hoc admission of overseas counsel in cases concerning the interpretation and/or application of the Basic Law outside the context of PRC’s national security; and (iv) NSL is expressly accorded precedence over local laws of the HKSAR under NSL 62. (c) Accordingly, that the contribution of overseas counsel (whose expertise and experience are exclusively based on common law, statutes drafted against common law background, international human rights conventions or European Union laws, as opposed to the continental legal system of the PRC) in the proper development of HKSAR’s national security-related jurisprudence is necessarily limited if not negligible or de minimis. (2) Second, the admission of overseas counsel in cases involving national security is incompatible with the overall objective and design of the NSL. NSL was enacted as a result of a need to address, inter alia, interference in the HKSAR’s affairs by foreign or external forces, and that admitting overseas counsel on an ad hoc basis in cases involving national security would generally tend to defeat that aim in that, among others, there is no meaningful or effective enforcement of overseas counsel’s confidentiality obligations under NSL 63 over State secrets and other confidential information which he or she comes to know in the practice of law. (3) Third, the fact that it is the obligation of the executive authorities, legislature and judiciary of the HKSAR, under NSL 3, to effectively prevent, suppress and impose punishment for any act or activity endangering national security, which includes any possible attempt to use the legal process to compromise the protection of national security (as exemplified in the various powers conferred under NSL, such as NSL 41 to NSL 47), and that such need should be of paramount importance in considering the public interests in an application of this nature. (4) Fourth, the suggestion that the public perception of fairness of criminal trial would or might be undermined simply because overseas counsel is not admitted on an ad hoc basis is inappropriate, unjustified, illogical and in any event wholly without evidential basis. Question 2 2. Whether, by reason of the key considerations of public interests outlined above: (1) ad hoc admission of overseas counsel in cases concerning national security should generally be refused (save in exceptional circumstances); (2) that the burden of establishing the existence of exceptional circumstances rests with the applicant; and (3) unless and until the said burden of establishing exceptional circumstances is duly discharged, such ad hoc admission application of overseas counsel should be dismissed. 3. Insofar as may be necessary, the SJ also relies on the “or otherwise” limb in seeking leave to appeal: (1) Given: (a) this is the first application for ad hoc admission of overseas counsel in cases involving national security and (b) the important questions involved, guidance by CFA is plainly justified. (2) The courts below have failed to appreciate: (a) the uniqueness of NSL as a piece of national law, rooted in substantive PRC criminal law provisions on national security, and which have to be interpreted against the Working Principles set out in the Explanation on the “Draft Law of the People’s Republic of China on Safeguarding National Security in the Hong Kong Special Administrative Region” addressed to the 19th Session of the 13th Standing Committee of the National People’s Congress by the Responsible Official of the Legislative Affairs Commission under the Standing Committee of the National People’s Congress dated 18 June 2020; and (b) the inherent incompatibility of the ad hoc admission of overseas counsel in cases concerning national security with the overall objective and design of NSL. (3) CA also relied on the ground that: (i) the forthcoming criminal trial is a high profile case attracting substantial publicity locally and abroad; and (ii) the public perception of fairness of the criminal trial would or might be undermined if the Applicant is not admitted on an ad hoc basis (CA Judgment §51). Such a reason or ground: (a) is wrong in principle and sends a wrong signal to the general public that the HKSAR courts assisted only by the local Bar will not be able to secure a fair trial for Mr Lai, (b) damages public confidence in the administration of justice, (c) is not supported by any evidence, and (d) is in any event contrary to the public interest if stand uncorrected. (4) Considering the overall circumstances of this case (including the nature of the issues involved, the unique nature of the NSL and the Applicant’s lack of expertise on national security law of the PRC), the ad hoc admission of the Applicant in any event cannot be justified.” [1] [2022] HKCFI 3233. [2] HCCC 51/2022. [3] Cap 200. [4] The Law of the People’s Republic of China on Safeguarding National Security in the Hong Kong Special Administrative Region (Promulgated 30 June 2020). [5] (Cap 159) section 27(4): “Notwithstanding that a person does not satisfy all the requirements specified in subsections (1) and (2)(b) [for admission generally as a barrister in the HKSAR], where the Court considers that he is a fit and proper person to be a barrister and is satisfied that he has (a) the qualification acquired outside Hong Kong to engage in work that would, if undertaken in Hong Kong, be similar to that undertaken by a barrister in the course of ordinary practice as a barrister in the High Court or Court of Final Appeal; and (b) substantial experience in advocacy in a court, the Court may admit such person as a barrister under this section for the purpose of any particular case or cases and may impose such restrictions and conditions on him as it may see fit.” [6] Relevantly going back to Re Flesch QC [1999] 1 HKLRD 506, where it was noted that the admission of overseas counsel was consistent with the policy of the Basic Law (BL94: “On the basis of the system previously operating in Hong Kong, the Government of the Hong Kong Special Administrative Region may make provisions for local lawyers and lawyers from outside Hong Kong to work and practise in the Region.”) [7] [2016] 2 HKLRD 647. [8] At §10. [9] Identified at §§5 and 6. [10] [2022] HKCA 1689, Kwan VP, Chu VP and Au JA (9 November 2022). [11] Mr Benjamin Yu SC, in place of Mr Jin Pao SC who appeared before Poon CJHC. [12] CA§18, citing Re Mably [2014] 1 HKLRD 627 at §38; Hadmor Productions Ltd v Hamilton [1983] AC 191 at 220; and The Abidin Daver [1984] 1 AC 398 at 420. [13] Ibid. [14] CA§29. [15] CA§51. [16] NM §1(1)(b), cf Annex Q1 §1(1)(b). [17] NM §2, cf Annex Q2 §2. [18] NM §1(1)(a) and (b), cf Annex Q1 §1(1)(a) and (b). [19] NM §1(1)(c), cf Annex Q1 §1(1)(c). [20] NM §1(2), cf Annex Q1 §1(2). [21] NM §1(3), cf Annex Q1 §1(3). [22] As set out in Flywin Co Ltd v Strong & Associates Ltd (2002) 5 HKCFAR 356. [23] SJ Skel in CA leave application §1. [24] [2022] HKCA 1751, Kwan VP, Chu VP and Au JA (“CA leave decision”). [25] CA leave decision §16. [26] CA leave decision §17. [27] NM §1(2), cf Annex Q1 §1(2). [28] NSL 41, NSL 46 and NSL 47. [29] CA leave decision §§18-21. [30] CA leave decision §23 [31] Set out in Flywin Co Ltd v Strong & Associates Ltd (2002) 5 HKCFAR 356. [32] While the citation refers to new points taken at a civil trial, it is a principle of general application. It has been applied in criminal cases (eg, HKSAR v Yam Chim Kwan [2013] HKEC 1302, FAMC 57/2012 at §9; HKSAR v Mark Richard Charlton Sutherland [2019] HKCFI 140 at §10); and in public law cases (eg, Chee Fei Ming v Director of Food and Environmental Hygiene (No 2) [2016] 3 HKLRD 412 at §21; K v Commissioner of Police [2021] 2 HKLRD 645 at §37). [33] At §38. [34] [2015] HKEC 1139; HCMP 977/2015 (17 June 2015), at §10. [35] Flywin Co Ltd v Strong & Associates Ltd (2002) 5 HKCFAR 356 at §39. [36] CACV 78/2014 (10 July 2015), at §6. [37] At §3(4). [38] [2022] HKCA 1751. [39] CA leave decision §16. [40] SJ’s skeleton argument §3(5). [41] NSL 3: The Central People’s Government has an overarching responsibility for national security affairs relating to the Hong Kong Special Administrative Region. It is the duty of the Hong Kong Special Administrative Region under the Constitution to safeguard national security and the Region shall perform the duty accordingly. The executive authorities, legislature and judiciary of the Region shall effectively prevent, suppress and impose punishment for any act or activity endangering national security in accordance with this Law and other relevant laws. 
Press Summary (English) Press Summary (Chinese) CCDI 346-354/2013 (PW) IN THE CORONER’S COURT THE HONG KONG SPECIAL ADMINSTRATIVE REGION (Case No. CCDI 346-354 of 2013) HO Oi-hing HO Oi-ying HO Oi-ming TANG Yuk-ling SIU Chi-man KWAN Pui-man Eleni POON Lau-tim TO Sau-ching POON Tak-sze Coram: June Cheung, Coroner Dates of hearing: 15-26 Feb, 2 March, 9 March, 31 March and 23 May 2016 Date of delivery of inquisition: 7 June 2016 I N Q U I S I T I O N Introduction 1. I have mentioned at the commencement of and during this inquest, but I will reiterate here: that the purpose of this Inquest is to inquire into the cause of the death and the circumstances connected with the death. We are not concerned with the fault, civil liability or compensation here. Although we have looked into matters concerning travel insurance as well as responsibility of travel agent at different locality, those, as I have mentioned more than once, were to facilitate me in making practicable recommendation to prevent future tragedy. In fact, the law expressly prohibits any conclusion being framed in such a way as to determine any question of civil liability. Therefore the question as to whether or not there was negligence such as to give rise to an action for damages or compensation is not a matter for this court; - that is a matter for the civil courts. My role here is to determine simply: what happened and what did not happen as well as making practicable recommendation having the hard lesson learnt. 2. The standard of proof I apply in this inquest, unless I specify otherwise, when deciding upon all matters is what known is as “on a balance of probabilities”. In other words, is it more likely than not that a fact existed or an event occurred. 3. Despite the difficulties encountered during the investigation, namely the fact that all eye witnesses of incident were not available, the court was not able to examine the real exhibits and that Hong Kong itself does not have experience or expertise in hot-air ballooning, this court has fortunately obtained two detailed reports from the Egyptian authority which were admitted as evidence, namely, 1) the investigation report issued by the Egyptian Ministry of Civil Aviation in December 2013 (“AAIC report”) governed by Annex 13 to the Convention on International Civil Aviation and 2) the Public prosecution report signed by the “President of the Appeals, Acting Chairman of the Technical Officer to the Officer of the Attorney General of the Arab Republic of Egypt (“Prosecution report”) containing the transcript and memorandum from the General Prosecutor’s office. The two reports are in my view thorough, detailed and professionally produced by persons with the necessary skills, knowledge and expertise. We also have the benefit of having Mr. Chadwick, General Aviation Flight Standard Officer (Balloon Operations) from the UK Civil Aviation Authority, to assist this court to understand the two reports and operation of hot air balloon (“HAB”) in general. I consider Mr. Chadwick’s expert evidence was most helpful, impartial and professional. 4. During the inquest, we have called 9 witnesses: they were Mr Wilson AU (CW1) (“AU”) and LAU Tak ping (CW3) (“LAU”) from Kuoni Travel (“Kuoni”); Mr Wong Kong Sing (CW2), one of the tour member who did not join the optional hot air balloon ride; Mr Joseph Tung, Chairman of the Travel Industry Council (“TIC”) (CW4), Mr Chin Shing (CW5) who was responsible for making the HAB model for the purpose of this inquest and two Arabic translators for clarification of some of the contents of the reports (CW6 and CW7); Mr Chadwick (CW8), the HAB expert mentioned above and finally Miss Scarlett SO Lai-kuen (CW9), person in charge in Hong Kong of Paradise Travel (“Paradise”) which was the Destination Management Company (“DMC”) engaged by Kuoni for this Egypt tour. Their evidence covered a number of issues in great details that apparently might not be strictly relevant to the findings at inquest set out in Form 12 but essential for this court to understand the background of the case and finally making recommendation to prevent future similar tragedy. General course of event leading to the deaths 5. The present case may be one of the most tragic incidents happened while Hong Kong citizens were travelling abroad. As a matter of fact, it was the deadliest hot air ballooning disaster in history resulted in 19 deaths, 9 of which were the deceased in the present inquest. The general course of event was largely undisputed. All the nine deceased, together with 6 other tour members who did not join the HAB ride, were joining a 10 days’ tour to Egypt departing on 22 February 2013 organized by Kuoni. The HAB ride in question took place in Luxor city and was one of the self-pay optional activities offered in the tour. Kuoni has engaged Paradise as their DMC responsible for the arrangement in Egypt and Paradise has chosen Egyptian Airship and Balloon-Sky Cruise Company (“Sky Cruise”) as the local HAB service provider. 6. At small hours on 26 Feb 2013, the 9 deceased, AU, tour escort of Kuoni, and Mostafa, assistant from Paradise, set off from hotel to the place where the balloon would take off. It was around 0615 hours local time when the balloon took off with 20 passengers. When the balloon had flown for more than 35 minutes, the pilot instructed the ground crew to receive the drop line starting the landing procedure. The HAB suddenly caught fire in the air which caused major injury to the pilot who exited the basket followed by a few passengers. While the balloon was still burning with less weight on the basket, it kept ascending until the envelope was so badly damaged, incapable of supporting the basket and fell on the field. As a result of the incident, 19 passengers were dead, 9 of which were the deceased of the present inquest. The pathologists found that the injury causing each of their death to be multiple injuries and burns. Cause of the accident 7. The HAB in question was manufactured in 2004 by a Spanish balloon company called Ultramagic and registered as SU 283. The balloon type was “Ultramagic N425” and Sky Cruise bought it in 2005 from a Sweden operator. The balloon was operated by wind action effects for horizontal movements while vertical movements control was made through the control of the burners by the balloon pilot. 8. On the evidence, the take-off procedures seemed to be normal and the balloon had flown for about 35 minutes before the fire started in the course of landing. Evidence from the Pilot 9. The only surviving eye-witness of the event who has given evidence to the Egyptian authority was the pilot of the HAB in question, Mr Momen Murad Ali Hussein, Egyptian, aged 29. The other surviving passenger was not willing to give evidence. The pilot has held a balloon pilot commercial licence issued by the Egyptian Civil Aviation Authority since 2006. He was firstly interrogated by the authority when he was badly injured in bed after the accident. He was further taxed for the purpose of prosecution in April 2013 and he gave a statement to the investigation authority again on 17 June 2013[1]. The prosecution was not proceeded with finally[2]. His evidence was contained in the AAIC report[3] and prosecution report[4]. 10. He stated that because there was no maintenance manager available at the time, he took a look at the balloon and ensured the hoses were safely installed using sense of smell before taking off. At the time of taking off, he contacted the control tower but no one was there to answer him. He said however the take-off procedures went smoothly and the balloon kept flying like usual until it started to land at around 10 meters height from the ground. At the beginning of the landing, the fuel quantity for both aft cylinders showed ¼ capacity, whilst the forward burners were not used except for the “pilot light”. After about 35 minutes, the pilot used the forward cylinders. All of a sudden, a fire hit him on his back and ejected him from the balloon. He was not able to take any measures to control the fire, neither using the extinguisher nor turning off the valve of the gas cylinder. 11. The pilot also stated that at the beginning of the landing, he dropped the rope out of the basket before he heard a loud sound of a fire flame. He did not notice any interference between the rope and any of the hoses. The location of the drop line rope is far from the hoses, he believed it cannot be scrambled around them. Neither did he smell any gas nor hear any explosion sound of any part. All he heard was the sound of a fire flame. The fire sound was similar to the spraying sound coming out of an “insect repellant pressurized can”. He said the fire came from the forward burners but was not able to identify the exact location of the fire source. The fire was from left side and his injuries mainly on the left side. The fire was in one direction of yellow color. 12. He could not conclusively identify the source of the flame. He believed that the cause of accident could be as a result of defect of [5]one of the hoses connecting the cylinder to the burner. 13. The Egyptian investigation[6] also showed that the surviving pilot was suffering from burn wounds scattered on his head, his face, his back, his left upper limb and right upper arm, and upper chest and abdomen and lower limbs. Skin grafting operations were conducted. Maintenance of the hoses 14. The maintenance manual states the need for a special inspection of the hoses every ten years and to check if there are any cuts or damage.[7] It is said that the maintenance manager conducted a 100 hour annual inspection dated 5 October 2012 under the supervision of Egyptian Civil Aviation Authority and the Validity Certificate was renewed for the period of 13 October 2012 until 12 October 2013[8]. 15. Within the AAIC analysis[9], it is stated that maintenance carried out by Sky Cruise did not always refer to the appropriate part number or serial number, as per the full maintenance procedures.[10] 16. On the other hand, Hasan Muhammad Kinnawi, the assistant of the Ground Team Head for the HAB who was responsible for supervising the workers in taking off and landing, stated that gas leakage was a usual occurrence in his job, and they usually noticed the leakage by smell it and would change it without mentioning of any reporting system. He also said that the operation manager (ground team head) and the maintenance engineer were both absent on the day of the incident for unknown reasons[11]. The operation manager of Viking Company, another HAB company in Luxor, also said that gas leakage of HAB happened a lot.[12] 17. Mr Chadwick commented[13]that the hose that was attached to the burner that was made in 2005 and may have accumulated high flight hours may not necessarily be directly a contributing factor to the accident. Hoses with many more confirmed flying hours and much older are being regularly used in the world hot air balloon fleet. At the time of this report, the UK CAA are unaware of another hot air balloon accident that is attributed to an in-flight hose failure of the nature specified in the report, this being largely due to the stringent inspection and maintenance requirements in of HAB activities in UK and the implementation of it. Conclusions from the Egyptian authority 18. It is stated in AAIC report (Exhibit C30), at page 147, that: “The probable cause for the accident as seen by the investigation committee is due to a (1) hose fuel leak at the upper portion of the forward right hose connected to burner number 193 capturing its ignition source from burner’s fire causing a fire that caused a major and direct injury to the (2) balloon captain” (my emphasis) 19. Nevertheless, it could not be ascertained from the investigation carried out by the Egyptian authority if the hose fuel leak was as a result of internal malfunction of the hose due to lack of inspection and/or maintenance or if there was an outer force pulling off the hose from the burner, e.g. tangling by drop line at the time of landing. 20. The AAIC report also considered two contributing factors to the accident at page 147 of the report: “1. Maintenance actions that were carried out on the hoses could not indicate the need to replace the hose that was the cause of the accident” 2. The part number of the hose connected to the burner 193 was for a hose made in 2005 and therefore it has accumulated high flight hours and sometimes under adverse conditions” 21. On the other hand, the Spanish Civil Aviation Commission of Accidents (CIAIAC) made their point in their letter[14], as the accredited representative from Spain in the Egyptian investigation, that: “the simultaneity or coincidence in time of the action of pulling down the balloon by the ground crew through the drop line rope (handling line) and the start of the gas leak and a further fire should be considered as a significant condition for the accident scenario under which the fire broke out” 22. Further conclusion by CIAIAC[15] was that: “- Operator maintenance discrepancy issues and certain lack of maintenance control were evident during the AAIC investigation; - With regard to this particular flight, the pre-departure check had not been fully completed by the balloon pilot.” Finding at inquest in Form 12 23. Having heard all the evidence I accept and find that: Injury causing death The medical causes of death for all of the deceased were multiple injuries and burns. Time, place and circumstances at or in which injury was sustained All nine deceased died in the morning of 26 Feb 2013 during a hot-air balloon ride over Luxor City when the balloon caught fire during the flight after the landing procedure had started. The fire was due to hose fuel leak at the upper portion of the forward right hose connected to the burner number 193. Such fire also caused serious and direct injury to the balloon pilot who was unable to do anything to control the fire. Conclusion as to the deaths I find the conclusion as to the deaths of all deceased to be deaths by accident. 24. I now also read out the Particulars for the time being of each deceased required by the Birth and Deaths Registration Ordinance (Cap 174) to be registered concerning the death (See form 12 attached for each deceased). Consideration of recommendation 25. Under section 44(2) of the Coroners Ordinance (Cap 504), the coroner at an inquest may make recommendation designed: (a) to prevent the recurrence of fatalities similar to that in respect of which the inquest is being held; (b) to prevent other hazards to life disclosed by the evidence at the inquest; (c) to bring to the attention of a person who may have power to take appropriate action any deficiencies in a system or method of work which are disclosed by the evidence at the inquest and which are of public concern. 26. The following are some features of the evidence which are helpful to me in considering recommendations to be made after the inquest. Risk involved in HAB ride in Egypt 27. HAB is an activity not available in Hong Kong and has become a popular activity around the world and welcomed by Hong Kong tourists travelling abroad. As HAB was operated under apparently simple mechanics and the speed of the same is generally quite slow and steady for commercial/tourism purpose, the risk of the same was generally not recognized by tourists. Although Mr LAU of Kuoni testified that he, as a member of senior management responsible for designing tour, viewed that HAB ride involved moderate degree of risk, Kuoni has not conducted any risk assessment on this activity before offering them and there was no particular attention, guidelines or information concerning risk of taking part in this activity provided to their employees, including front desk receptionists or tour escorts nor their customers. LAU said that they basically relied on the fact that other companies were also offering HAB activities and they were aware that HAB was offered in many other countries and considered to be safe in general. 28. HAB is a type of aircraft and a highly regulated industry. Of course, each jurisdiction has its own civil authority to regulate this activity. It was emphasized at times by Mr Ozorio, SC, representing Kuoni, that HAB ride is a safe activity by quoting Mr Chadwick’s report: “the advancement in fabric technology and envelope rapid deflation systems has facilitated larger balloons to be operated with a high degree of safety and reliability.”[16] However, since HAB ride is taken place in the air with considerable height, it undoubtedly involves a certain degree of risk. By giving a simple click in the internet will show a number of serious fatal HAB accidents occurred in the past around the world. That is why hot air ballooning needs stringent regulation on the airworthiness and operation of the balloon and pilot licensing scheme. In fact, as testified by Mr Chadwick, UK has developed a set of stringent regulations to ensure safety of air ballooning activities especially the commercial passenger carrying balloon has been vastly developed over the years around the world[17]. The safety of and risk involved in this activity actually depend on how stringent such regulations are and how well the same are implemented. 29. It is of note that AAIC report did make a recommendation, at p.149 of the report,[18] that: “Issue necessary instructions to balloon companies to assure informing passengers when they reserve their flights that this kind of sports included some degree of risk and for the purpose of adventure in order that passengers would make their decisions from the beginning .” (my emphasis) 30. In relation to that, Mr Chadwick responded in his report[19] that: “Many hot-air balloon passenger flights are bought by a third party as a present, with the actual passenger not finding out about their flight until the day of the flight. However, it is accepted that flying as a passenger in a hot-air balloon involves certain different threats to that of flying on a commercial airliner, and the AAIC is correct that information as to the physical requirements of flying in a hot-air balloon needs to be given at the point of reservation. This would therefore have implications to tour operators and local representatives at resorts, etc.” (my emphasis) Information about HAB ride provided by Kuoni to its staff 31. The present tour was the fifth Egypt tour Mr AU led as he testified. He started escorting Egypt tour since Sept 2012. He testified that before he led the first Egypt tour it was the experienced colleagues sharing to him local information about Egypt and he has not received any formal training or briefing relating to safety of leading tour in Egypt or taking part in hot air balloon activities. About half to one month prior to the 1st Egypt tour escorted by him in Sept 2012, experienced colleagues did mention to him briefly and casually in office how HAB was operated by wind, the flight distance and landing posture. Apart from that, no other information or training was given to him in respect of the risk of taking HAB ride. He was also told that people who had heart condition, hypertension and fear of height should not take part in HAB. Information about HAB provided to tour members 32. In early 2013, Mr AU was assigned by Kuoni to host the “tea party” on 7 Feb 2013 for the present tour. By then, he did not know if he would be the tour escort of the tour. About a week before the departure, Mr AU was informed by Kuoni to escort the tour in question. 33. There is no dispute that the need for customers to purchase comprehensive travel insurance before departure is generally set out in the tour booking form[20] and the receipts issued by Kuoni to the deceased[21]. It was also reminded in the receipts, though in small prints, under the heading of “Important Notice” that customers are suggested to read carefully the coverage and relevant clauses in their insurance policies to ensure sufficient coverage in case of accident. I am also well aware that Kuoni was not allowed by the TIC[22] nor did they put pressure upon or force customers to purchase Kuoni’s own Generali insurance. The question is whether the customers are provided with clear and sufficient information concerning the activity they are joining and the risk involved in it before they made an informed decision of participation. 34. AU said that during the tea party, among other information, there was a note named “Safety guidance of activities in Middle East and African area”[23], provided to tour members. On which there was a paragraph concerning taking HAB rides which stated to the effect that participants shall follow staff’s safety guidance and rules of activity. Participants shall only leave the balloon as directed by the staff upon landing of the balloon.[24] 35. At the tea party, tour members were also provided with a Price list of Self-pay Activities in Middle East and Africa [25] on which it was said relating to HAB ride: “Go on a hot air balloon ride in Luxor to view the beauty of Nile and Luxor. This activity includes breakfast and insurance. An aviation certificate will be issued: Approx. 45 minutes (the entire trip takes approx. 2.5 hrs: USD 190 for all age group” (my emphasis) 36. The information given to customers in the above price list: “this activity includes insurance” was ambiguous. When Senior Manager LAU was asked as to what it exactly meant, he said vaguely that might mean this activity included flying insurance taken by Sky Cruise and customers were covered by insurance when they take part in this HAB activity. LAU said they in fact just translated and copied the information stated on the Sky Cruise’s leaflet[26]. In my view, this clause may let the customers misunderstand that their joining of this activity was covered by insurance which turned out to be not the case, as revealed from the evidence of this inquest. Should the members be warned more clearly and specifically that their insurance may not cover this HAB activity, their decision to join the activity may be changed. 37. This court is well aware of its limited jurisdiction. I am not concerned about why 6 deceased out of 9 did not buy Kuoni’s travel insurance but bought their own Comprehensive Travel Insurance from China Merchants Insurance Company Limited. Nor is this court attempting to explore the reasons why they were not insured. My concern is whether practicable recommendation could be made to prevent recurrence of similar fatalities. The ambiguous wording on the above optional self-pay activity price list[27] about the HAB ride and the lack of sufficient information provided to the customers joining the HAB ride so that they could make better assessment of the risk may affect their choices of participating the activity. 38. On AU’s evidence, he said that when tour members were paying the fees the day before the ride, he had told them briefly information about the duration of the flight, what scenery to be seen and advised them not to join if they have heart condition, hypertension and fear of height. He has also shown pictures taken about the ride before to the tour members. Upon arrival at the taking-off place before they were approaching the balloon in question, he gave his group of 9 members some information as to the duration of the flight and landing posture. He also told them he would be taking pictures for them and taking care of their belongings. While the pilot of the HAB from Sky Cruise was conducting briefing session in English to 20 passengers before taking off, no instant interpretation was given by AU because he saw no need as his had given his own briefing to the deceased earlier on. Information concerning HAB ride acquired by Kuoni 39. In fact, after hearing the evidence, it was not surprising why not much information relating to HAB ride was given to the employees and customers by Kuoni. On LAU’s evidence, even the senior management designing the tour did not have much information about the HAB ride. Apart from having been told by Paradise that Sky Cruise was licensed, insured and experienced, Lau basically told the court that he had never discussed the HAB ride among the senior management before offering this activity. Neither did he nor Kuoni as far as he knew conduct any formal evaluation after this incident. 40. It was submitted by Mr Ozorio, SC for Kuoni that there were no discussion nor query raised among tour members concerning risk of HAB ride and that they could easily do researches on their own should they wish so. That, as submitted, showed that the deceased were not concerned with the risk of HAB ride. I am afraid I cannot agree with that. It is not hard to observe that people who tend to join travel tour, as opposed to planning their own trip, are those who are either more senior in age or do not prefer to plan by their own. Participants of travel tours tend to leave the planning and entrust all the arrangement to the travel agent company and the tour escort. That may require higher responsibility from the travel company to conduct sufficient risk assessment before offering the activities and to provide sufficient information to the customers. 41. It is of note that this case happened in Egypt which is not a developed and advanced country like the United Kingdom, Australia or New Zealand where HAB industry is long established and governed by stringent regulatory system. It is common sense that an activity is considered to be safe in one country does not necessarily mean that it is safely operated in another. 42. As a matter of fact, it was not the first HAB accident occurred in Luxor. It is not disputed that there was a major HAB accident happened in Luxor as a result of which 16 passengers were injured which caused temporary suspension of the activity. I consider that Hong Kong travel companies apart from designating a DMC to appoint and liaise with the local service providers should also acquire sufficient information of such activity before offering the activities that involved a certain degree of risk. 43. Instead of relying on the fact that other companies are also providing the same activity, travel companies should acquire sufficient information for its own risk assessment before providing such activity. The information is also essential for training the frontline staff. More importantly, customers should be well informed of the information before taking part in such activity involving certain degree of risk. In my view, the list of basic information about the HAB ride that should be acquired by the travel company and provided to the customers includes but not limited to: - Height and duration of the HAB flight; - Pictures of HAB showing the envelope, basket and burners of the HAB; - Size of the HAB; - Brief information about material and mechanics of operation of the HAB; -Information about experience and reputation of the company providing the HAB ride; - Any HAB accidents happened before in the destination and any of such related to the service provider; - The licensing system in the country as compared with other worldly recognized country. 44. In fact, it is stated at paragraph 9.3 of Code of Conduct for Outbound Tour Escort issued by TIC in October 2007[28] that: “Prior to arranging self-pay activities, tour escort shall explain clearly to tour participants the content, duration, fee, safety and responsibility of such activities.” Checklist 4.2 attached to the same Code also provides a few items for checking in respect of optional activities. Selecting and supervising the local destination management company and/or service provider 45. From the evidence of LAU, Senior manager of Kuoni responsible for product designing and tour planning, Kuoni started to offer HAB ride in 2008. Kuoni stopped offering HAB immediately after this accident, but still offer Egypt tour until June 2013. He said that Kuoni relied upon the fact that other Hong Kong travel companies were also offering HAB ride before they were offering the same. He was aware that HAB ride was also provided in other countries around the world and was considered to be safe by other companies. He said that some of their colleagues also had the experience of HAB ride in other companies before they joined Kuoni. 46. When asked if there was any risk assessment of HAB ride in Egypt done by him or Kuoni as far as he knew before the accident, LAU replied negative and said that they usually relied on the local licensing system and if the company was insured. He was informed by Paradise that Sky Cruise was licensed and insured (being one of criteria of obtaining license). Surprisingly, Kuoni has not even obtained a copy of those documents but just relied on Paradise words until the accident occurred. He was told by Paradise that Sky Cruise was an experienced operator of HAB and started to operate HAB activities in 90s. Kuoni basically relied on the choice and information given by Paradise. LAU said that they also relied on feedback from tour escorts and customers when considering the quality of service provided by local company. 47. Despite the absence of any risk assessment done by LAU/Kuoni before offering HAB ride, he agreed that HAB contained certain degree of risk. He considered HAB ride as an activity having moderate degree of risk. As to what training, if any, was given to tour escort or front desk staff, he said there would only be casual sharing given by experienced colleagues to junior ones. Kuoni did not have particular safety guidelines or briefing of HAB ride provided to staff. During the tour in Egypt, AU basically was responsible for liaisoning with the local guide/assistant there called Mostafa and relied on him for all the local arrangements. 48. Miss Scarlet So, the person in charge of Paradise in Hong Kong testified that she had not received any training on travel industry. She worked for Osama, founder of Paradise in Egypt, since 2005 through an introduction by a common friend. She did not have any travel or working experience in Egypt. She, although being the person in charge of Paradise in Hong Kong and having the title as Regional Manager, basically said that she was not more than a mail box transferring message between Kuoni and Osama. In relation to the choosing Sky Cruise, she said it was Osama’s choice and she had no idea why it was chosen. She said there was no discussion between Kuoni and her on HAB activities. Before the accident, she has never given any copy of Sky Cruise’s license or insurance but just told Kuoni that Sky Cruise was chosen as the HAB service provider upon Osama’s instruction. 49. A Risk Solutions Manual for Travel Industry in Hong Kong was issued by TIC in February 2008[29]. Paragraph A1.1.1 of the Manual provided a Supplier selection criteria checklist. However, the checklist does not cover present situation in which the Hong Kong travel company has engaged a DMC to select and arrange with the local service providers. 50. Despite the fact that Kuoni was told by Paradise that Sky Cruise was an experienced service provider with reputation, it was revealed from the Egyptian investigation that the maintenance carried out at Sky cruise did not always refer to the appropriate part numbers or serial numbers, as per the full maintenance procedures[30]; for example: absence of record of the 5th tank (for inflating the balloon) and unclear basket number[31]. Further, gas leakage appeared to be a common problem as testified by the staff of balloon companies, which did not appear to cause serious concern by the company, which seemed to be quite contrary to the stringent UK reporting system as testified by Mr. Chadwick. 51. Regarding operation of the present HAB, it was also revealed that the pilot was allowed to take off, despite absence of the maintenance engineer and the operation manager and that he was not able to contact the control tower until 10 minutes after taking off[32]. When the pilot was asked by the investigation team about the dissimilarity between his pre-departure check signature and his previous signature, he claimed that he would use several signatures in order not to be imitated! Mr. Chadwick could not help expressing his concern as to why anyone would want to imitate a pre-departure check signature during normal operation.[33] Also, no summary or translation of pilot briefing was supplied to the passengers before taking off. It was however recommended by Mr. Chadwick that the use of laminated passenger safety briefing cards should complement the pilot’s briefing to passengers. The use of diagrams and briefing cards in major world language should be provided.[34] 52. My pointing out the above observation made after Egyptian investigation was not because I wanted to discuss or determine the fault or liability, but to show the danger of Hong Kong travel companies simply replying on words of DMC without conducting its own risk assessment. It is equally important for Hong Kong travel companies to properly select and supervise the service providers even after a DMC has been engaged. 53. Given the highly technical nature of HAB operation and the fact that the inquest does not have the advantage of hearing direct evidence, this court is not in the best position to make any conclusion or comments on the requirement of a safe HAB maintenance or operation. 54. Upon hearing this inquest, I reckon that there is at present a lacuna in TIC’s guidance on Hong Kong travel companies’ duty and responsibility on local service providers when DMC is engaged. I note that there is quite a detailed list for TIC members to select local service providers[35], however such manual does not cover situation when travel agent companies have entrusted a local DMC. 55. Apart from telling me that we should trust the licensing system of each country and it is very difficult for Hong Kong travel company to monitor local service providers as well as it is a worldwide common practice to engage a DMC to do the work for travel companies, Mr. Tung of TIC basically gave me no suggestion nor assistance as to how the service of Hong Kong travel companies can be improved and safety of Hong Kong customers be better safeguarded. 56. To my surprise, both Kuoni and TIC have never conducted any evaluation of the case until the present inquest. Neither has TIC done anything to review the existing guidelines. More than three years have lapsed since the date of this tragic event. It is disappointing to learn that not only the travel company and TIC were not able to realize the risk involved in this particular activity showed by the lack of risk assessment, they were also not alerted after loss of nine precious lives. That in a way shows how insensitive the Hong Kong travel industry is to the well-being of customers. Regardless of whether this tragedy could possibly be avoided, the cursory attitude in assessing risk and reflecting after the accident as showed from the testimony given by LAU being senior management of Kuoni, given its reputation as one of the most high-end travel companies in Hong Kong, was unsatisfactory in my view. Similarly, I consider TIC’s reaction after the accident to be too slow and attitude too passive. The AAIC report was issued in December 2013 and I expect TIC should have looked into the matter thoroughly and reviewed the present guidelines on outbound activities involving risk much earlier instead of waiting for the result of this inquest. 57. Further, I consider that information or assistance currently provided by TIC to travel companies and more importantly to customers as to risks involved in various activities operated at various countries to be insufficient. Recommendations to Travel Industry Council (TIC) 58. Analyzed as the above, I now recommend: 1) TIC to conduct a thorough research and collect data from the industry so as to list out popular activities operated in different countries offered by travel companies involving certain degree of risk. TIC to categorize the degree of risk in respect of different activities provided in different countries for industry as reference. Such list should be updated from time to time; 2) TIC to set out clear guidelines for travel companies as to the necessary information of activities involving certain degree of risk to be provided to customers before taking part in such activities; 3) TIC to set out clear and specific guidelines for travel companies that customers should be reminded that their travel insurance may not cover activities involving certain degree of risk; 4) TIC to look into travel companies’ duty and responsibility in selecting and supervising the local service providers and review the existing guidelines when destination management companies are engaged; 5) TIC to set out new guidelines on travel companies’ duty and responsibility to supervise destination management companies; 6) TIC to set out safety guidance and advertise to customers as to the importance of obtaining details of travel policy insurance and information on activities involving risk before enrolment. Recommendations to Kuoni 59. I also recommend: 1) Kuoni to obtain sufficient information from local service providers and conduct thorough risk assessment on activities involving certain degree of risk provided to customers before offering the same; 2) Kuoni to conduct thorough evaluation after every major accident taking place during tours provided by them so as to find out the cause of the accident and consider if there is anything that can be done to improve the quality and safety of service provided to customers. 3) Kuoni to provide sufficient information to customers about activities involving certain risk and remind them in more specific and clear terms that activities involving risk may not be covered by their travel insurance; 4) Kuoni to provide more training and information to front desk staff, including receptionists and tour escorts, so that they can be better equipped to explain the risk involved in activities provided by the company; 5) Kuoni to set out clear criteria for selecting destination management companies and local service providers with first hand information and to closely monitor the service provided by destination management companies and local service providers. 60. Both TIC and Kuoni need to reply the Coroner in writing in three months’ time as to 1) whether the above recommendations are accepted and if yes 2) what steps have been taken to implement the recommendations and 3) reasons if the recommendations are not accepted. 61. Lastly, let me offer my deepest condolence again to the next of kin for their loss of loved ones and also thank you all for attending this inquest. I cannot even try to imagine the difficult journey you have gone through and dare not expect that by having this inquest you will be soothed in anyway. What I sincerely hope is that you will find this inquest helpful in some way having been informed of more details about the incident. I am indebted to the Hong Kong police, Mr Brook and Mr IP, SPP of Department of Justice, for their unfailing support in this investigation despite all the difficulties. I also thank counsel’s assistance from all parties. Lastly, I need to deliver my gratitude to the senior interpreters, Mr Wong and Mr Lai and the Clerk to Coroner, Miss On for their invaluable assistance throughout the hearing. 62. If parties and the press wish, a copy of this inquisition can be obtained from my clerk after the court rises. A press summary is also available for the press. The inquisition and press summary will also be uploaded to the Judiciary website very soon after today’s hearing. The inquest is hereby closed. June CHEUNG	Coroner Representation: Mr Michael Ozorio, SC instructed by Messrs Paul C. K. Tang & Chiu for KUONI Travel (China) LTD. Mr Bernard Murphy of Messrs Howse Williams Bowers for Ultramagic S.A. Mr Foster Yim instructed by Messrs Wan Yeung Hau & Co. for next of kin of Ho Oi Hing, Ho Oi Ying, Ho Oi Ming, Tang Yuk Ling, Siu Chi Man & Kwan Pui Man Eleni Mr Edward Brook and Mr Cliff Ip of Department of Justice as Coroner’s Officer [1] Pp. 17-20 of the AAIC report (Exhibit C30) [2] P.12 of the prosecution report (Exhibit C32A) [3] Paras.1.5.1, pp.16-20, AAIC report (Exhibit C30) [4] Pp.74-84, prosecution report (Exhibit C32A) [5] P.17 of AAIC report (Exhibit C30) [6] P.5 of the investigation report (Exhibit C30) [7] P.6.8 of Maintenance Manual of Ultramagic HAB (Exhibit C50) [8] P.7 of the prosecution report (Exhibit C32A) [9] Paragraph 2.6, page 140 of AAIC report (Exhibit C30) [10] Paragraph 3.4.6 of Chadwick’s report (Exhibit C46a) [11] pp.48-59 of Prosecution report (Exhibit C32A) [12] p.66 of Prosecution report (Exhibit C32A) [13] Page 8 of Chadwick’s report (Exhibit C46a) [14] P.219 of the AAIC report (Exhibit C32a) [15] Page 217 and 218 of AAIC report and page 10, para 3.3.3 of Chadwick’s report (Exhibit C46a) [16] Para.2.12, page 2 of Chadwick’s report (Exhibit C46a) [17] Paragraphs 6.1-6.8 of Chadwick’s report (Exhibit C46a) [18] Paragraph 4.1.4, page 149 of AAIC report (Exhibit C30) [19] Para. 4.14, page 20 of Chadwick’s report (Exhibit C46a) [20] Exhibit C1 [21] Exhibits C8 and C9 [22] Exhibit C35 [23] Exhibit C5 [24] Exhibit C5 [25] Exhibit C6 [26] Sky Cruise HAB leaflet (Exhibit C11) [27] Exhibit C6 [28] Exhibit C34 [29] Exhibit C 36 [30] page 140 of AAIC report, comment 13 of Chadwick report (Exhibit C46a) [31] comments 12 of Chadwick report (Exhibit C46a) [32] pp.75-86 of the prosecution report (Exhibit C32a) [33] Paragraph 3.4.4, p.11 of Chadwick’s report (Exhibit C46a) [34] Paragraph 5(g), pp. 22 and 23 of Chadwick’s report (Exhibit C46a) [35] Paragraph A1.1.1 if Risk Solution Manual (Exhibit C36 and 36A) 
The Court: A. Background 1. These proceedings raise questions of the constitutionality and legality of the Emergency Regulations Ordinance (Cap 241) (“ERO”) and the Prohibition on Face Covering Regulation (Cap 241K) (“PFCR”) made thereunder. 2. The ERO is a law of some antiquity, dating back to 1922, but the PFCR is a very recent creature, made by the Chief Executive in Council (“CEIC”) on 4 October 2019 following months of protests and civil unrest and against the background of scenes of escalating violence and danger on the streets of Hong Kong not seen in half a century. The protests began in opposition to the Fugitive Offenders and Mutual Legal Assistance in Criminal Matters Legislation (Amendment) Bill 2019 (“Bill”), which was proposed for the purpose of amending the Fugitive Offenders Ordinance (Cap 503) as well as the Mutual Legal Assistance on Criminal Matters Ordinance (Cap 525). The unrest has however continued notwithstanding the Government’s decision in June 2019 to suspend the legislative process for the Bill, the acknowledgement since 9 July that the Bill was “dead”, the announcement on 4 September that the Bill would be formally withdrawn, and the actual withdrawal of the Bill in the Legislative Council of Hong Kong (“LegCo”) on 16 October. 3. From 9 June to 4 October 2019, according to the Government’s records, over 400 “public order events” (referring to public assemblies or processions) arising out of the Bill and other matters had been held, with a significant number of them ending up in outbreaks of violence. 4. The degree of violence had escalated and, in particular, on 29 September and 1 October, gatherings took place in many districts in which certain protesters blocked major thoroughfares, vandalised Mass Transit Railway stations and facilities, government offices and selected shops, and hurled petrol bombs at police officers, vehicles and police stations. 5. The use of facial covering was not prohibited in Hong Kong before the PFCR. Having regard to the widespread and escalating danger posed by the situation, the Security Bureau recommended to the CEIC that there was an urgent need to introduce the PFCR to facilitate police investigation and to serve as a deterrent against the violent and illegal acts of masked perpetrators. 6. On 4 October 2019, the CEIC, at a special meeting of the Executive Council, formed the view that the violence and rampage had placed Hong Kong in a state of public danger and that it was necessary in the public interest for the PFCR to be made with a view to restoring law, order and public peace. Accordingly, the PFCR was made on that day pursuant to the ERO, was gazetted on the same day,[1] and came into operation about nine hours later at midnight on 5 October 2019. 7. Also on 4 October 2019, the Chief Executive, the Secretary for Justice and the Secretary for Security spoke at a press conference to explain the Government’s decision and the operation of the PFCR. The Government made clear that the power under the ERO was being invoked on the “public danger” ground only, not the “emergency” ground and that Hong Kong was not being proclaimed to be in a state of emergency. 8. The PFCR has led to a number of applications to the court for leave to apply for judicial review filed between 4 and 10 October 2019. 9. On 11 October 2019, the PFCR was considered by the House Committee of the LegCo and a subcommittee on the PFCR was appointed. At the LegCo meeting of 16 October 2019, the PFCR was laid on the table of the LegCo. B. The Applications and Grounds for Judicial Review 10. The application in HCAL 2945/2019 (“HCAL 2945”) is one made by 24 Members of the LegCo, first lodged in draft on 5 October. The putative respondents are the CEIC and Secretary for Justice.[2] On 6 October, directions were given for an early “rolled‑up” hearing[3] of that application for leave and (if leave be granted) the judicial review itself. The application in HCAL 2949/2019 (“HCAL 2949”), filed on 8 October, is an application made by Mr Leung Kwok Hung, a former LegCo Member, with the same putative respondents. On 11 October, directions were given for a rolled‑up hearing of his application (but limited to two of the grounds thereof [4]) at the same time as HCAL 2945. The conjoined hearing of these two matters took place before us on 31 October and 1 November. We shall refer to the putative respondents below simply as the “respondents” or “Government”. Four recent applications by other applicants relating to the ERO and PFCR have been directed to await the outcome of these proceedings.[5] 11. In summary, the grounds that have been put forward[6] as the basis for judicial review and argued at the hearing may be broadly identified as follows: (1) Ground 1 — The ERO is unconstitutional because it amounts to an impermissible grant or delegation of general legislative power by the legislature to the CEIC and contravenes the constitutional framework under the Basic Law. We shall refer to this as the “delegation of legislative power ground”. (2) Ground 2 — The ERO was impliedly repealed in 1991 by s 3(2) of the Hong Kong Bill of Rights Ordinance (Cap 383) (“HKBORO”) either entirely or to the extent it is inconsistent with s 5 of that latter Ordinance; alternatively, it was impliedly repealed in 1997 by Art 4 of the ICCPR as applied through Art 39 of the Basic Law. We shall refer to this as the “implied repeal ground”. (3) Ground 3 — The ERO, to the extent that it empowers the CEIC to make regulations restricting fundamental rights protected by the Basic Law and the Hong Kong Bill of Rights (“Bill of Rights”), falls foul of the “prescribed by law” requirement in Art 39 of the Basic Law. We shall refer to this as the “prescribed by law ground”. (4) Ground 4 — By reason of the principle of legality, the general words in s 2(1) of the ERO are not to be read as allowing the Government to adopt measures that infringe fundamental rights of the individual in circumstances far removed from emergency situations. The PFCR is therefore ultra vires — beyond the power conferred on the CEIC by the ERO. We shall refer to this as the “principle of legality ground”. (5) Ground 5A — Section 3 of the PFCR amounts to a disproportionate restriction of a person’s liberty and privacy, freedom of expression and right of peaceful assembly under Arts 5, 14, 15, 16, 17 of the Bill of Rights and Art 27 of the Basic Law. We shall refer to this as the “section 3 proportionality ground”. (6) Ground 5B — Section 5 of the PFCR constitutes a disproportionate interference with the rights and freedoms protected by Art 27, 28 and 31 of the Basic Law and Art 5(1), 8, 14 and 16 of the Bill of Rights. We shall refer to this as the “section 5 proportionality ground”. 12. At the hearing, Ms Gladys Li, SC addressed us on Ground 1 and Mr Johannes Chan, SC addressed us on Grounds 2 to 5B on behalf of the applicants in HCAL 2945; Mr Hectar Pun, SC addressed us on Grounds 3 and 5B on behalf of the applicant in HCAL 2949; and Mr Benjamin Yu, SC addressed us on behalf of the respondents on all the grounds argued in both applications. 13. As can be seen, Grounds 1 to 3 seek to impugn the ERO itself, whereas Grounds 4, 5A and 5B attack the validity of the PFCR. All of the grounds are targeted at the constitutionality and validity of the ERO or the PFCR; none of them seeks to impugn the good faith of the Government in invoking the ERO to enact the PFCR or the reasonableness of the decision to do so in the classic administrative law sense. We shall discuss these grounds separately in turn below, after examining the provisions of the ERO and the PFCR. We shall deal with each ground without assuming that the others have been established. We bear in mind the duty of the court to adjudicate in accordance with the law including the Basic Law as we find them, and to exclude politics from our consideration. C. The Emergency Regulations Ordinance 14. The ERO is a short Ordinance with only three substantive sections, which provide as follows: “ 2. Power to make regulations (1) On any occasion which the Chief Executive in Council may consider to be an occasion of emergency or public danger he may make any regulations whatsoever which he may consider desirable in the public interest. (2) Without prejudice to the generality of the provisions of subsection (1), such regulations may provide for— (a) censorship, and the control and suppression of publications, writings, maps, plans, photographs, communications and means of communication; (b) arrest, detention, exclusion and deportation; (c) control of the harbours, ports and waters of Hong Kong, and the movements of vessels; (d) transportation by land, air or water, and the control of the transport of persons and things; (e) trading, exportation, importation, production and manufacture; (f) appropriation, control, forfeiture and disposition of property, and of the use thereof; (g) amending any enactment, suspending the operation of any enactment and applying any enactment with or without modification; (h) authorizing the entry and search of premises; (i) empowering such authorities or persons as may be specified in the regulations to make orders and rules and to make or issue notices, licences, permits, certificates or other documents for the purposes of the regulations; (j) charging, in respect of the grant or issue of any licence, permit, certificate or other document for the purposes of the regulations, such fees as may be prescribed by the regulations; (k) the taking of possession or control on behalf of the Chief Executive of any property or undertaking; (l) requiring persons to do work or render services; (m) payment of compensation and remuneration to persons affected by the regulations and the determination of such compensation; and (n) the apprehension, trial and punishment of persons offending against the regulations or against any law in force in Hong Kong, and may contain such incidental and supplementary provisions as appear to the Chief Executive to be necessary or expedient for the purposes of the regulations. (3) Any regulations made under the provisions of this section shall continue in force until repealed by order of the Chief Executive in Council. (4) A regulation or any order or rule made in pursuance of such a regulation shall have effect notwithstanding anything inconsistent therewith contained in any enactment; and any provision of an enactment which may be inconsistent with any regulation or any such order or rule shall, whether that provision shall or shall not have been amended, suspended or modified in its operation under subsection (2), to the extent of such inconsistency have no effect so long as such regulation, order or rule shall remain in force. (5) Every document purporting to be an instrument made or issued by the Chief Executive or other authority or person in pursuance of this Ordinance or of any regulation made hereunder and to be signed by or on behalf of the Chief Executive or such other authority or person, shall be received in evidence, and shall, until the contrary is proved, be deemed to be an instrument made or issued by the Chief Executive or that authority or person. 3. Penalties (1) Without prejudice to the powers conferred by section 2, regulations made hereunder may provide for the punishment of any offence (whether such offence is a contravention of the regulations or an offence under any law applicable to Hong Kong) with such penalties and sanctions (including a maximum penalty of mandatory life imprisonment but excluding the penalty of death), and may contain such provisions in relation to forfeiture, disposal and retention of any article connected in any way with such offence and as to revocation or cancellation of any licence, permit, pass or authority issued under the regulations or under any other enactment as to the Chief Executive in Council may appear to be necessary or expedient to secure the enforcement of any regulation or law or to be otherwise in the public interest. (2) Any person who contravenes any regulation made under this Ordinance shall, where no other penalty or punishment is provided by such regulations, be liable on summary conviction to a fine of $5,000 and to imprisonment for 2 years. 4. Declaratory provision as to effect of an amending Ordinance For the purpose of removing doubts it is hereby declared that the words in subsection (1) of section 2 “he may make any regulations whatsoever which he may consider desirable in the public interest” shall be deemed always to have included power to make such regulations as are mentioned in paragraph (g) of subsection (2) of section 2 and it is further declared that the provisions of subsection (4) of section 2 shall be deemed always to have been incorporated herein.” 15. The ERO[7] was placed before the LegCo as the Strike Legislation Bill and enacted in the course of one sitting of the LegCo on 28 February 1922 during the height of a general strike called in support of seamen in their dispute with shipping companies over wages. One of the first regulations made under the new law was to criminalise trade unions as unlawful societies. An amendment to the ERO to provide for more severe punishment was passed in one sitting on 15 July 1925. 16. More substantial amendments to the ERO were made in March 1949,[8] including the removal of three paragraphs under s 2(2) and the addition of paragraphs (g) to (n) as well as s 2(4) and (5). The limit on penalties in s 3 was increased from a fine of $1,000 and imprisonment for one year to a fine of $5,000 and imprisonment for two years. In quick succession, on 31 August 1949, the ERO was amended again,[9] by the substitution of a new s 3 and s 4, to empower regulations to be made with such penalties and sanctions as appear to the Governor in Council necessary or expedient, including the death penalty subject to the LegCo’s approval. 17. After the Bill of Rights came into operation, the ERO was amended in 1993,[10] when the death penalty in s 3(1) was substituted by mandatory life imprisonment.[11] 18. As part of the general adaptation exercise for pre‑existing laws, amendments were made in 1999 to the terminology used in the ERO, with, for example, the substitution of “Chief Executive in Council” for “Governor in Council”.[12] 19. In the course of the many years since its enactment, various regulations had been made under the ERO, such as in the 1920s during general strikes, in 1929 during a severe drought, in 1931 during the anti‑Japanese riots, in the 1930s during an outbreak of cholera, in 1935 when there was a case of rabies, in the late 1930s to mid‑1940s during the Second World War, in 1950 to impose the death penalty for the possession of any bomb, grenade, mine, or like apparatus, in 1950 to deal with the shortage of small coins, in 1956 during the Tsuen Wan riots, and in 1967 during the widespread riots in Hong Kong. By the 1940s most of the then existing regulations were consolidated into the Emergency (Principal) Regulations. 20. Before the making of the PFCR, the ERO had not been invoked to make any new regulation since the 1970s, although some very old regulations continued to be in the statute books. In 1995, all the remaining extant regulations made under the ERO, including the Emergency (Principal) Regulations, several regulations relating to deportation and detention, and regulations relating to requisition of land for use by British military forces, were revoked by the Governor in Council.[13] D. The Prohibition on Face Covering Regulation 21. The PFRC is a short piece of regulation, with six sections. Section 2 contains the definitions, and s 6 provides for the time in which prosecution may be brought. 22. Section 3 of the PFCR imposes a prohibition on the use of and makes it an offence to use facial covering in certain circumstances. It reads: “ (1) A person must not use any facial covering that is likely to prevent identification while the person is at‑ (a) an unlawful assembly (whether or not the assembly is a riot within the meaning of section 19 of Cap 245[14]); (b) unauthorised assembly; (c) a public meeting that‑ (i) takes place under section 7(1) of Cap 245; and (ii) does not fall within paragraph (a) or (b); or (d) a public procession that‑ (i) takes place under section 13(1) of Cap 245; and (ii) does not fall within paragraph (a) or (b). (2) A person who contravenes subsection (1) commits an offence and is liable on conviction to a fine at level 4[15] and to imprisonment for 1 year.” 23. “Facial covering” is defined in s 2 to mean a mask or any other article of any kind (including paint) that covers all or part of a person’s face. 24. Section 3 prohibits the use of facial covering likely to prevent identification in four specified situations. These situations require some elaboration. 25. The first, referred to in s 3(1)(a), is an “unlawful assembly” (非法集結). This has the same meaning as in s 18 of the Public Order Ordinance (“POO”). Section 18(1) & (2) provide: “ (1) When 3 or more persons, assembled together, conduct themselves in a disorderly, intimidating, insulting or provocative manner intended or likely to cause any person reasonably to fear that the persons so assembled will commit a breach of the peace, or will by such conduct provoke other persons to commit a breach of the peace, they are an unlawful assembly. (2) It is immaterial that the original assembly was lawful if being assembled, they conduct themselves in such a manner as aforesaid.” 26. Secondly, s 3(1)(b) refers to “unauthorised assembly” (未經批准集結). This has the same meaning as in s 17A(2) of the POO, which provides: “ (2) Where— (a) any public meeting[16] or public procession[17] takes place in contravention of section 7 or 13; (b) 3 or more persons taking part in or forming part of a public gathering[18] refuse or wilfully neglect to obey an order given or issued under section 6; or (c) 3 or more persons taking part in or forming part of a public meeting, public procession or public gathering, or other meeting, procession or gathering of persons refuse or wilfully neglect to obey an order given or issued under section 17(3), the public meeting, public procession or public gathering, or other meeting, procession or gathering of persons, as the case may be, shall be an unauthorized assembly.” 27. Paragraph (a) of this definition refers to ss 7 and 13 of the POO. Under s 7 of the POO, a public meeting may take place if the Commissioner of Police is notified of the intended meeting and has not prohibited it, but a meeting of not more than 50 persons, a meeting in private premises (whether or not the public are permitted to attend) with no more than 500 persons, and certain meetings in schools, are exempted. Similarly, under s 13 of the POO, a public procession may take place if the Commissioner of Police is notified of the intended procession and does not object to its taking place, but a public procession consisting of not more than 30 persons and a public procession not on a public highway or thoroughfare or in a public park are exempted. Paragraphs (b) and (c) of the definition refer to ss 6 and 17(3). These provisions allow the police to give orders to regulate, stop or disperse public gatherings, public meetings and public processions in specified circumstances or for specified purposes. 28. The third situation is set out in s 3(1)(c). It essentially refers to a public meeting (公眾集會) which has been notified to and not prohibited by the Commissioner of Police and which is not an unlawful or unauthorised assembly. Since meetings of no more than 50 persons, meeting in private premises with no more than 500 persons, and certain meetings in schools do not have to be notified, they also fall outside s 3(1)(c). 29. The fourth situation, set out in s 3(1)(d), essentially means a public procession (公眾遊行) which has been notified to and is not objected to by the Commissioner of Police and which is not an unlawful or unauthorised assembly. Public processions of no more than 30 persons and public processions not on a public highway or thoroughfare or in a public park, which do not have to be notified, fall outside s 3(1)(d). 30. In short, broadly speaking, the prohibition in s 3(1) applies to persons at unlawful assemblies, unauthorised assemblies, public meetings notified and not prohibited, and public processions notified and not objected to, and does not prima facie apply to public meetings or processions that do not need to be notified, although such meetings or processions may turn into unauthorised assemblies (eg if 3 or more persons taking part in them refuse or wilfully neglect to obey an order given under ss 6 or 17(3) of the POO) or unlawful assemblies. 31. Section 4 of the PFCR sets out a defence to the offence under s 3(2). It provides: “ (1) It is a defence for a person charged with an offence under section 3(2) to establish that at the time of the alleged offence the person had lawful authority or reasonable excuse for using a facial covering. (2) A person is taken to have established that a person had lawful authority or reasonable excuse for using a facial covering if‑ (a) there is sufficient evidence to raise an issue that the person had such authority or reasonable excuse; and (b) the contrary is not proved by the prosecution beyond reasonable doubt. (3) Without limiting the scope of the reasonable excuse referred to in subsection (1), the person had a reasonable excuse if, at the assembly, meeting or procession concerned‑ (a) the person was engaged in a profession or employment and was using the facial covering for the physical safety of the person while performing an act or activity connected with the profession or employment; (b) the person was using the facial covering for religious reasons; or (c) the person was using the facial covering for a pre‑existing medical or health reason.” 32. In other words, the defence is one of lawful authority or reasonable excuse. An evidential burden is placed on an accused to raise the defence but it is the prosecution’s legal burden to disprove the defence. The scope of reasonable excuse is not exhaustively defined, but three grounds are specifically included, namely, professional or employment reasons, religious reasons and pre‑existing medical or health reasons. 33. Section 5 concerns police powers in relation to facial covering. “ (1) This section applies in relation to a person in a public place who is using a facial covering that a police officer reasonably believes is likely to prevent identification. (2) The police officer may‑ (a) stop the person and require the person to remove the facial covering to enable the officer to verify the identity of the person; and (b) if the person fails to comply with a requirement under paragraph (a)– remove the facial covering. (3) A person who fails to comply with a requirement under subsection (2)(a) commits an offence and is liable on conviction to a fine at level 3[19] and to imprisonment for 6 months.” 34. “Public place” (公眾地方) has the meaning given by s 2(1) of the POO, that is to say: “ any place to which for the time being the public or any section of the public are entitled or permitted to have access, whether on payment or otherwise, and, in relation to any meeting, includes any place which is or will be, on the occasion and for the purposes of such meeting, a public place.” E. Ground 1 — the delegation of legislative power ground (1) The parties’ contentions 35. Under Ground 1, the argument of the applicants in HCAL 2945 is in essence this. The constitutional order of Hong Kong is founded on the Basic Law. Under the Basic Law, the LegCo is the legislature of the Hong Kong SAR in which the region’s legislative power is vested. The CE and the Executive Branch are given executive powers and in addition the power to make subordinate legislation. On a proper understanding and construction of the framework of the Basic Law, the LegCo is constitutionally precluded from granting or delegating general legislative power to the CEIC. The ERO purports to do this and is therefore inconsistent with the Basic Law and unconstitutional. 36. The respondents submit that there is nothing in the Basic Law which provides whether expressly or by implication that the LegCo cannot authorise the CEIC to make regulations during an occasion of public danger or emergency. The ERO was in force before 1997 and was therefore part of the laws that were carried over into the Hong Kong SAR under Art 8 of the Basic Law, unless it contravenes the Basic Law. The ERO had been twice held to be valid by the Full Court. There is nothing that indicates any intention that the arrangements in relation to the ERO were to change upon the resumption of exercise of sovereignty by the People’s Republic of China over Hong Kong. Regulations made under the ERO are not intended to be permanent and are subject to negative vetting by the LegCo. 37. Our decision below is confined to the public danger ground in the ERO and to the powers it confers when the CEIC considers there to be an occasion of public danger, which is the ground on which the PFCR has been made. There may be considerations of necessity arising from real emergencies that may affect the proper analysis on which we do not think we have been fully addressed and on which therefore we should express no concluded opinion. (2) Provisions of the Basic Law 38. We shall first turn to the relevant provisions of the Basic Law that are relied on in this context. In Chapter I on General Principles, Art 2 makes clear that the executive, legislative and judicial power exercised in Hong Kong is by authority from the National People’s Congress (“NPC”) and is enjoyed in accordance with the provisions of the Basic Law.[20] Art 8 provides for continuity of the laws previously in force in Hong Kong (ie before 1 July 1997) subject to any amendment by the legislature.[21] 39. In Chapter II, which concerns the relationship between the Central Authorities and the Hong Kong SAR, Art 17 states that the Hong Kong SAR “shall be vested with legislative power” and that laws enacted by the legislature of the Hong Kong SAR must be reported to the Standing Committee of the National People’s Congress (“NPCSC”) for the record. Under prescribed conditions, the NPCSC may return the law, which becomes invalidated. 40. Art 18 provides: “ The laws in force in the Hong Kong Special Administrative Region shall be this Law, the laws previously in force in Hong Kong as provided for in Article 8 of this Law, and the laws enacted by the legislature of the Region. National laws shall not be applied in the Hong Kong Special Administrative Region except for those listed in Annex III to this Law. The laws listed therein shall be applied locally by way of promulgation or legislation by the Region. … In the event that the Standing Committee of the National People’s Congress decides to declare a state of war or, by reason of turmoil within the Hong Kong Special Administrative Region which endangers national unity or security and is beyond the control of the government of the Region, decides that the Region is in a state of emergency, the Central People’s Government may issue an order applying the relevant national laws in the Region.” 41. Chapter IV deals with political structure. Section 1 (Arts 43 to 58) concerns the CE. Art 48 sets out the CE’s powers and functions (行使下列職權) as follows: “ Article 48 The Chief Executive of the Hong Kong Special Administrative Region shall exercise the following powers and functions: (1) To lead the government of the Region; (2) To be responsible for the implementation of this Law and other laws which, in accordance with this Law, apply in the Hong Kong Special Administrative Region; (3) To sign bills passed by the Legislative Council and to promulgate laws; To sign budgets passed by the Legislative Council and report the budgets and final accounts to the Central People’s Government for the record; (4) To decide on government policies and to issue executive orders; (5) To nominate and to report to the Central People’s Government for appointment the following principal officials: Secretaries and Deputy Secretaries of Departments, Directors of Bureaux, Commissioner Against Corruption, Director of Audit, Commissioner of Police, Director of Immigration and Commissioner of Customs and Excise; and to recommend to the Central People’s Government the removal of the above-mentioned officials; (6) To appoint or remove judges of the courts at all levels in accordance with legal procedures; (7) To appoint or remove holders of public office in accordance with legal procedures; (8) To implement the directives issued by the Central People’s Government in respect of the relevant matters provided for in this Law; (9) To conduct, on behalf of the Government of the Hong Kong Special Administrative Region, external affairs and other affairs as authorized by the Central Authorities; (10) To approve the introduction of motions regarding revenues or expenditure to the Legislative Council; (11) To decide, in the light of security and vital public interests, whether government officials or other personnel in charge of government affairs should testify or give evidence before the Legislative Council or its committees; (12) To pardon persons convicted of criminal offences or commute their penalties; and (13) To handle petitions and complaints.” 42. Section 2 (Arts 59 to 65) of Chapter IV concerns the Executive Authorities, and contains the following provisions among others: “ Article 59 The Government of the Hong Kong Special Administrative Region shall be the executive authorities of the Region. Article 60 The head of the Government of the Hong Kong Special Administrative Region shall be the Chief Executive of the Region. … … Article 62 The Government of the Hong Kong Special Administrative Region shall exercise the following powers and functions: (1) To formulate and implement policies; (2) To conduct administrative affairs; (3) To conduct external affairs as authorized by the Central People’s Government under this Law; (4) To draw up and introduce budgets and final accounts; (5) To draft and introduce bills, motions and subordinate legislation; and (6) To designate officials to sit in on the meetings of the Legislative Council and to speak on behalf of the government.” 43. In relation to the making of laws, therefore, the powers and functions of the Executive Authorities under the Basic Law are to draft and introduce bills, (in the case of the CE) to sign bills after they are passed by the LegCo and to promulgate laws, and to make and introduce subordinate legislation. Art 56 states that except for the adoption of measures in emergencies, the CE shall consult the Executive Council before, inter alia, introducing bills to the LegCo and making subordinate legislation. 44. Section 3 (Arts 66 to 79) of Chapter IV deals with the Legislature. Art 66 states: “ The Legislative Council of the Hong Kong Special Administrative Region shall be the legislature of the Region.” 45. Art 72 sets out the powers and functions of the President of the LegCo: “ The President of the Legislative Council of the Hong Kong Special Administrative Region shall exercise the following powers and functions: (1) To preside over meetings; (2) To decide on the agenda, giving priority to government bills for inclusion in the agenda; (3) To decide on the time of meetings; (4) To call special sessions during the recess; (5) To call emergency sessions on the request of the Chief Executive; and (6) To exercise other powers and functions as prescribed in the rules of procedure of the Legislative Council.” 46. Art 73 sets out the powers and functions (行使下列職權) of the LegCo itself: “ The Legislative Council of the Hong Kong Special Administrative Region shall exercise the following powers and functions: (1) To enact, amend or repeal laws in accordance with the provisions of this Law and legal procedures; (2) To examine and approve budgets introduced by the government; (3) To approve taxation and public expenditure; (4) To receive and debate the policy addresses of the Chief Executive; (5) To raise questions on the work of the government; (6) To debate any issue concerning public interests; (7) To endorse the appointment and removal of the judges of the Court of Final Appeal and the Chief Judge of the High Court; (8) To receive and handle complaints from Hong Kong residents; (9) If a motion initiated jointly by one-fourth of all the members of the Legislative Council charges the Chief Executive with serious breach of law or dereliction of duty and if he or she refuses to resign, the Council may, after passing a motion for investigation, give a mandate to the Chief Justice of the Court of Final Appeal to form and chair an independent investigation committee. The committee shall be responsible for carrying out the investigation and reporting its findings to the Council. If the committee considers the evidence sufficient to substantiate such charges, the Council may pass a motion of impeachment by a two-thirds majority of all its members and report it to the Central People’s Government for decision; and (10) To summon, as required when exercising the above-mentioned powers and functions, persons concerned to testify or give evidence.” 47. In addition, Arts 49, 50 and 52 contain provisions that deal with, inter alia, conflicts between the CE and the LegCo in relation to bills. (1) Under Art 49, if the CE considers that a bill passed by the LegCo is not compatible with the overall interests of Hong Kong, she may return it to the LegCo within three months for reconsideration. If the LegCo passes the bill again by not less than a two‑thirds majority, the CE must sign it or dissolve the LegCo under Art 50. By Art 52(2), if the new LegCo again passes the bill by a two‑thirds majority, the CE must resign if she still refuses to sign it. (2) Conversely, under Art 50, if the LegCo refuses to pass an important bill introduced by the government, and if consensus cannot be reached after consultation, the CE may dissolve the LegCo. If the new LegCo still refuses to pass the bill, the CE must resign under Art 52(3). The CE may dissolve the LegCo only once in each term of her office (Art 50(2)). (3) The powers of the LegCo and the CEIC 48. This examination of the Basic Law shows that the LegCo is constituted as the legislature of the Hong Kong SAR. It is given not only the power but also the function to enact, amend and repeal laws “in accordance with” the Basic Law and “legal procedures” (Art 73(1)). These procedures would seem to include the rules of procedure of the LegCo which it is empowered by Art 75 to make on its own, provided they do not contravene the Basic Law. Legislative procedures, characterised by the exchange and interaction of different points of view visible to the public, generally include consultation with the relevant Panel of the LegCo,[22] the introduction of bills (generally a function of the Executive (Art 62(5)) though certain bills may be introduced by Members of the LegCo (Art 74)), the readings of and votings on the bills including committee stage proceedings, and, finally, the signing of the bills (which is a function of the CE under Art 48(3) but regulated by Arts 49, 50 & 52), and the promulgation of the laws. 49. Although not expressly set out amongst the powers of the LegCo, the existence of the power for the LegCo by statute to authorise subordinate legislation to be made is necessarily implied, not only because it had long been the custom and usages of the system previously in place, without which the multitudinous matters that need to be legislated for would be beyond the work capacity of the LegCo, but also because the Basic Law expressly includes as part of the powers and functions of the Executive the making of subordinate legislation. This necessarily envisages the Executive being authorised by the LegCo by statute to make subordinate legislation, although, as is well known, other bodies may also be so authorised in relation to specific matters, such as the Rules Committee which is authorised to make rules of court. 50. In relation to the making of laws, the Executive is not vested with any general legislative power or the general power to enact, amend or repeal laws, but only the power to sign or refuse to sign bills (with the attendant consequences as between the CE and the LegCo in the case of refusal) and the power to make subordinate legislation. 51. The legislative power enjoyed by the Hong Kong SAR (Arts 2 & 17) is thus allocated by the Basic Law. The relevant provisions of the Basic Law show a marked contrast with the pre‑1997 position where the Governor, “by and with the advice and consent of” the LegCo, was empowered to “make laws for the peace, order, and good government of the Colony”[23] — a formula recognised as conferring a plenary legislative power, and where the colonial legislature was regarded as having “plenary powers of legislation, as large, and of the same nature, as those of [the Imperial] Parliament itself” (R v Burah (1878) 3 App Cas 889, 904). Under the new constitutional order, the LegCo has the powers and functions vested in it by the Basic Law expressly or by implication. The legislature can no longer claim supremacy but is subject to the Basic Law. Art 2 lays down that legislative power is to be exercised “in accordance with” the provisions of the Basic Law. Art 73(1) gives the LegCo the function to enact, amend or repeal laws “in accordance with” the provisions of the Basic Law. 52. It has been argued by the respondents that there is nothing in the Basic Law expressly preventing delegation of legislative powers. It has to be recalled, however, that the LegCo is designated as the legislature and is given not only powers but functions which in Chinese is “職” in the phrase “職權” and connotes that it is LegCo’s “job” in common parlance. It seems to us that this constitutional scheme does not permit the LegCo to grant and the CEIC (or, for that matter, any other body) to receive and be vested with what is essentially the LegCo’s own constitutional power and function as the legislature of the Hong Kong SAR to enact, amend or repeal laws, except for an authorisation of subordinate legislation. The question is whether the ERO should be regarded as thus contravening the Basic Law. If it contravenes the Basic Law, then being a law previously in force in Hong Kong, it did not become the law of the Hong Kong SAR through Art 8 of the Basic Law. (4) Nature of the ERO and regulations made thereunder 53. If, as we think to be the case, a line is to be drawn between general legislative power and the power to make subordinate legislation, it is necessary to turn to the question of what subordinate legislation means. The term “subordinate legislation” is used in Arts 8, 56(2) and 62(5) of the English version of the Basic Law but not defined. The equivalent Chinese term used is not uniform: Art 8 refers to “附屬立法”; Arts 56(2) and 62(5) both refer to “附屬法規”, but no party has suggested that there is any significance in the difference between the two Chinese terms. 54. According to s 3 of the Interpretation and General Clauses Ordinance (Cap 1) (“IGCO”), “subsidiary legislation” and “subordinate legislation” (附屬法例、附屬法規、附屬立法) mean any proclamation, rule, regulation, order, resolution, notice, rule of court, bylaw or other instrument made under or by virtue of any Ordinance and having legislative effect. However, the provisions of the IGCO apply to Ordinances and to instruments made or issued under or by virtue of any Ordinance (see s 2(1) of the IGCO). They do not directly apply to the Basic Law. Indeed, the phrase “subordinate legislation” and the equivalent Chinese terms (附屬法規、附屬立法) were only added to the above definition in s 3 of the IGCO in 1998, after the Basic Law had come into effect.[24] The previous version[25] defined the terms “subsidiary legislation” (附屬法例) and “regulations” (規例) only. The meaning of the expression “subordinate legislation” in the Basic Law must depend on an autonomous interpretation rather than on the subsequent definitions in the IGCO. 55. It would be a simplistic approach to say that, since any regulations made under the ERO are made by the CEIC as “regulations” pursuant to the power conferred by the ERO, the ERO does no more than permit the CEIC to make subordinate legislation and is therefore wholly unobjectionable. If that were correct, an Ordinance that delegated power to the CEIC to make regulations generally for the peace, order, and good government of Hong Kong would be valid as authorising no more than subordinate legislation even though it would in truth enable government by proclamation generally. Instead, it seems to us the matter has to be approached in substance rather than by labels: Yau Kwong Man v Secretary for Security [2002] 3 HKC 457, §43 & §67. Approached as a matter of substance, there is in our view very little that characterises the regulations made under the ERO as subordinate legislation and much to indicate that the ERO confers general legislative powers on the CEIC. 56. The first thing that strikes one is that the ERO is not a statute that legislates on a subject matter in principle leaving another body to devise the detailed legal norms that elaborate or put flesh on the broad matters laid down in the primary legislation. The long title of the ERO specifies that its object is to confer on the CEIC power to make regulations on occasions of emergency or public danger. But it gives no shape or direction of what the regulations that may be made are to be about. For example, the PFCR was enacted under the ERO not to work out and fill in the details for certain broad norms established by primary legislation, but as the very first piece of legislation in Hong Kong that has anything to do about face covering. This is fundamentally different from one’s ordinary conception of subordinate legislation. 57. Secondly, associated with the above point is that the powers conferred by the ERO are, without any doubt, of the widest possible nature. The authority under s 2(1) is to make any regulations whatsoever the CEIC may consider desirable in the public interest. There is no limit on the subject matter or the field of legislation. While a large number of matters are enumerated in s 2(2)(a) to (n) — which are very broad in themselves — for which provisions may be made by regulations, the chapeau in s 2(2) states that they are without prejudice to the generality of s 2(1). That generality is subject only to the phrase “which he may consider desirable in the public interest”. The touchstone notably is “desirable”, not “necessary”. That this hardly sets out any limit at all on the power may be seen from the decision of the Full Court in R v To Lam Sin (1952) 36 HKLR 1 and R v Li Bun & Others [1957] HKLR 89. 58. In To Lam Sin, the court said (at p 11): “ In the present case, there is power to make regulations if the Governor in Council is of a certain opinion upon two matters. Does not the fact that the regulations were made, and expressed to be made under that particular power, equally imply that the Governor in Council considered that an occasion of emergency or public danger had arisen and that the regulations were desirable in the public interest? Both are essentially matters of opinion and whether in fact such an occasion had arisen is no concern of the Courts. … [T]here being nothing on the face of the regulations to indicate that the proper matters have not in fact been considered the Court feels … that as a matter of construction it is a necessary implication that they have.” 59. In Li Bun, Hogan CJ also stated (at p 105): “ I think that in the absence of bad faith … It would be virtually impossible to separate in one’s mind the decision to make these regulations from a conclusion that they were desirable in the public interest.” 60. Thirdly, the powers may be invoked on any occasion which the CEIC “may consider to be an occasion of emergency or public danger”, neither of which is defined in the statute. In making the PFCR, the CEIC proceeded on the basis of an occasion of public danger and not emergency having arisen in Hong Kong, but the meaning of public danger can potentially be very broad. The statute does not state a “reasonable grounds” test and the words used have been construed by the Full Court to confer a wide and virtually unreviewable discretion, as shown in the passage from To Lam Sin quoted above. In Li Bun, Hogan CJ did not regard the mere making of the regulations as necessarily implying a conclusion that there is a state of emergency or public danger (see p 105), but thought that it provided prima facie proof that the Governor in Council considered that there was an occasion of emergency or public danger (p 109). Gould J considered that the very enactment of the regulations implied that the mind of the Governor in Council had been addressed to the consideration concerning emergency or public danger as well as to the desirability of the regulations (p 113). Reece J took a similar view (at pp 117‑120). Further, Hogan CJ held (at p 112), with the agreement of the other two judges: “ If the Governor in Council came bona fide to the conclusion that such an occasion existed, then that is an end of the matter. Whether the conclusion was justified is not a matter into which the Courts can inquire.” 61. Even if one were to inquire into the question, with the strict confidentiality and public interest immunity attaching to information placed before the CEIC, the scope of that inquiry is necessarily highly limited. 62. We shall presently return to these two cases, but it may be noted here that the width of the power as evidenced by these passages leaves precious little room for the doctrine of ultra vires to operate. 63. Fourthly, by s 2(2)(g), regulations made under the ERO may amend or suspend the operation of any enactment or apply any enactment with modification. It is not disputed that “enactment”, as defined in the IGCO, means “Ordinance”, and that “amend” includes “repeal, add to or vary”. By s 2(4), a regulation made under the ERO, and even an order or rule made in pursuance of such a regulation, has effect notwithstanding anything inconsistent with it contained in any enactment. The relevant provision in such enactment, to the extent of the inconsistency, shall “have no effect” so long as the regulation, order or rule remains in force. 64. As we have seen above, there is no provision in the Basic Law that authorises the CEIC by herself to amend or repeal primary legislation. That power and function is given to the LegCo, to be exercised “in accordance with the provisions of [the Basic] Law and legal procedures” (Art 73(1)). By these provisions, the ERO not only gives such power to the CEIC to be exercised by regulation, but also gives the power to unspecified bodies through an “order or rule” made in pursuance of a regulation for so long as such order or rule remains in force. 65. The respondents submit that such provisions as s 2(2)(g) of the ERO, conferring power on the Executive by regulation to repeal or alter primary legislation (sometimes known as “Henry VIII clauses”), are not regarded in the United Kingdom and Australia as impermissible or unconstitutional: see Thoburn v Sunderland City Council [2002] 4 All ER 156, §13; R (on the application of the Public Law Project) v Lord Chancellor and Secretary of State for Justice [2016] UKSC 39, §28; The Victorian Stevedoring and General Contracting Co Pty Ltd v Dignan (1931) 46 CLR 73, 125; ADCO Constructions Pty Ltd v Goudappel (2014) 308 ALR 213, §61; Capital Duplicators Pty Ltd v Australian Capital Territory (No 1) (1992) 66 ALJR 794, 798. It seems to us, however, that the foundation of the validity of such clauses (albeit treated as exceptional) is Parliamentary sovereignty. As Laws LJ said in Thoburn at §13: “by force of its very sovereignty, Parliament may delegate the power of amendment or repeal.” In Public Law Project at §20, Lord Neuberger referred to the UK system of “parliamentary supremacy”, in which “it is not open to a court to challenge or refuse to apply a statute, save to the extent that Parliament authorises or requires a court to do so”. 66. Quite apart from the fact that the LegCo does not enjoy such supremacy in the Diceyan sense[26] under the Basic Law, such clauses are antithetical to the norm of subsidiary legislation as understood in Hong Kong under s 28(1)(b) of the IGCO, which provides that “no subsidiary legislation shall be inconsistent with the provisions of any Ordinance” — a provision in place since the coming into operation of the Interpretation and General Clauses Ordinance 1966.[27] In the ordinary case, a piece of subsidiary legislation that is inconsistent with a primary statute, such as where it purported to remove a right arising under a statute, would be ultra vires as being contrary to s 28(1)(b): Gohind Mohan & Anor v Brian Shane McElney & Others [1983] HKLR 308, 312; Mita Kogyo Kabushiki Kaisha v Mitac Inc [1993] 2 HKLR 466, 470. 67. Fifthly, s 3(1) confers the widest powers for regulations made under the ERO to provide for the punishment of any offence with up to mandatory life imprisonment, in addition to forfeiture of articles and revocation or cancellation of licences and permits. This again contradicts the norm for subsidiary legislation as set out in s 28(1)(e) of the IGCO, according to which subsidiary legislation may provide that its contravention is an offence punishable by a fine not exceeding $5,000 or imprisonment for up to 6 months. 68. Sixthly, there is no time limit on the validity and force of the regulations made under the ERO, nor any mechanism for constant review. The regulations made simply become part of the general law and by s 2(3) they “shall continue in force until repealed by order of the [CEIC]”. We shall come to the question of “negative vetting” below, but it is to be noted here that once vetted and passed, the regulations would continue until they are repealed by order of the CEIC. We assume that the regulations may also be repealed by an Ordinance subsequently enacted, but bills introduced by Members of the LegCo are subject to special limitations and voting procedures (see Art 74 and Annex II of the Basic Law). 69. Seventhly, the respondents submit that despite the power given to the CEIC to make regulations under the ERO, the LegCo retains a role through the “negative vetting” of any regulations so made. Section 34(1) of the IGCO provides that all subsidiary legislation has to be laid before the LegCo at its next sitting after its publication in the Gazette. Section 34(2) provides: “ Where subsidiary legislation has been laid on the table of the Legislative Council under subsection (1), the Legislative Council may, by resolution passed at a sitting of the Legislative Council held not later than 28 days after the sitting at which it was so laid, provide that such subsidiary legislation shall be amended in any manner whatsoever consistent with the power to make such subsidiary legislation, and if any such resolution is so passed the subsidiary legislation shall, without prejudice to anything done thereunder, be deemed to be amended as from the date of publication in the Gazette of such resolution.” While the period of vetting is 28 days, under s 34(4) the LegCo may by resolution extend it by 21 days. 70. As defined in s 3 of the IGCO, to “amend” includes to “repeal, add to or vary”. However, the applicants contend that s 2(3) of the ERO precludes repeal of any regulations by negative vetting (and therefore precludes even amendment because amendments are mostly carried out by repeal and substitution). In To Lam Sin, surpa, at p 14, the Full Court seemed to think that any regulation made under the ERO could be amended or repealed by the LegCo through negative vetting. In Li Bun, supra, at p 97, however, Hogan CJ, with whom Gould J agreed,[28] was reluctant to place reliance on the negative vetting procedure as ensuring the control of the Legislature over regulations made under the ERO, because he took the view that it could be said that the then equivalent of s 34 of the IGCO had been “pro tanto repealed” by s 2(3) of the ERO. 71. Section 2(3) of the ERO was included in the original Ordinance as enacted in 1922. The procedure of negative vetting by the LegCo was not introduced until 1937,[29] although it appears that regulations made under the ERO had in practice been tabled in the LegCo since the earliest days. At the time in 1922, s 41 of the then Interpretation Ordinance[30] provided for vetting of subsidiary legislation by the Governor in Council. It could not therefore have been the purpose of s 2(3) to exclude the LegCo’s negative vetting. Nor could it have the purpose of laying down the only way in which a regulation may cease to have effect, for, first, s 2(3) uses the word “until”, not “unless and until”, and, secondly, as is common ground, an Ordinance can be passed at any time to repeal a regulation made under the ERO (and indeed the ERO itself). Rather, it would appear that the purpose of s 2(3) was to specify when the regulations made under the ERO would come to an end, perhaps to avoid any doubt as to whether they would cease to be in force when the occasion of emergency or public danger relied upon for their enactment had vanished. 72. Thus analysed, s 2(3) should in our view be construed to mean that regulations made under the ERO continue in force — in the sense that they do not lapse — until repealed by the CEIC. That section does not itself prevent such regulations from repeal by resolution of the LegCo during negative vetting under s 34 of the IGCO or from repeal by a subsequent Ordinance. 73. There may be a question as to whether negative vetting under s 34 is applicable if the regulations made under the ERO are not properly regarded in substance as subordinate or subsidiary legislation. We have not, however, been addressed on this point and we shall therefore, for present purposes, assume that s 34 applies to such regulations. 74. In any event, negative vetting is available (if at all) only insofar as it is left intact by the exercise of power under the ERO, for it is open to the CEIC, as part of the regulation made, to amend or suspend the operation of s 34 of the IGCO or to render it inapplicable to the regulation in question, as long as the CEIC considers that to be desirable in the public interest. In this regard the ERO may be contrasted with the legislation considered in Dignan, supra, which empowered the making of regulations subject to the Acts Interpretation Acts. 75. Furthermore, despite negative vetting, a regulation made under the ERO can come into operation immediately upon being made (as the PFCR did within 9 hours of its enactment). Even if a resolution for the amendment or repeal of a regulation is passed by a majority vote of both the Members of the LegCo returned by functional constituencies and those returned by geographical constituencies (as is required under Annex II of the Basic Law for such a resolution), as set out in s 34(2) that resolution only takes effect prospectively from the date of gazettal of the resolution. Such repeal does not operate to unwind the legal effect the regulation has already brought about. In particular, where a regulation has upon its commencement amended or repealed existing legislation, the subsequent repeal of that regulation by resolution does not “undo” the regulation’s effects and revive the original legislation. In other words, the “Henry VIII power” of a regulation is unchecked by negative vetting. 76. A number of authorities from overseas jurisdictions on this area of the law have been drawn to our attention in the course of submissions, but both sides accept that one must be exceedingly careful before placing reliance on overseas jurisprudence in the present context because they are based on those jurisdictions’ constitutional set‑up and their conception of separation of powers which cannot be simply transplanted to Hong Kong. The differences in the language and structure of the constitutional provisions, as well as in the social, historical and political contexts, require caution, as advised by Sir Anthony Mason NPJ in The Place of Comparative Law in Developing the Jurisprudence on the Rule of Law and Human Rights in Hong Kong (2007) 37 HKLJ 299, 304-305. 77. We also bear in mind the counsel of A Cheung J (as Cheung PJ then was) in Luk Ka Cheung v The Market Misconduct Tribunal & Anor [2009] 1 HKC 1 at §36, concurred in by Hartmann JA, that the traditions of Hong Kong and the theme of continuity may call for “a flexible and realistic, as opposed to an idealistic, approach to the doctrine of separation of powers, and a purposive and contextualised interpretation of the scope and meaning of ‘judicial power’ in the Basic Law”. There are cases illustrating both sides of the line in relation to judicial power. Thus, the determination of a fixed minimum term of imprisonment for young persons convicted of murder and made subject to an indefinite sentence has been held to be a judicial power which could not be vested in the CE, so that a statutory provision that purported to do so[31] was unconstitutional and invalid: Yau Kwong Man, supra.[32] On the other hand, the vesting in an administrative tribunal of the function of determining whether insider dealing had taken place and imposing sanctions therefor was held to be within the Basic Law; the function in question involved neither a determination of criminal guilt nor civil liability: Luk Ka Cheung, supra. 78. In Australia, which takes a relatively broad view of the power to delegate legislative authority, there are nevertheless limits to such delegation. In Dignan, supra, section 3 of the Transport Workers Act 1928‑1929 conferred power on the Governor-General of Australia to make regulations, not inconsistent with that Act, with respect to the employment of transport workers, and in particular for regulating the engagements, service and discharge of transport workers, and for regulating or prohibiting the employment of unlicensed persons as transport workers, and for the protection of transport workers. While Dixon J held that such a conferment of power was not unconstitutional, his Honour said (at p 101): “ This does not mean that a law confiding authority to the Executive will be valid, however extensive or vague the subject matter may be, if it does not fall outside the boundaries of Federal power. There may be such a width or such an uncertainty of the subject matter to be handed over that the enactment attempting it is not a law with respect to any particular head or heads of legislative power. Nor does it mean that the distribution of powers can supply no considerations of weight affecting the validity of an Act creating a legislative authority.” Earlier in that case (at pp 95‑96), Dixon J also drew attention to R v Burah, supra, and In re The Initiative and Referendum Act [1919] AC 935, saying it should be noticed that the Privy Council had held that a general power of legislation belonging to a legislature constituted under a rigid constitution does not enable it by any form of enactment to create an arm with general legislative authority and new legislative power not created or authorised by the instrument by which it is established. It seems to us that nothing in Dignan, which illustrates the more generous approach in Australia on this subject, suggests that a devolution of legislative power as complete and unfettered as in the ERO would be permissible. 79. In Capital Duplicators Pty Ltd v Australian Capital Territory (No 1) (1992) 66 ALJR 794, Brennan, Deane and Toohey JJ (whose decision together with Gaudron J’s was the majority) said (at p 800) that although s 122 of the Australian Constitution “confers on the Parliament a power not expressly limited as to its subject matter, it does not follow that that power authorises the Parliament to vest a general legislative power in a legislature created to receive and exercise it.” 80. By arming the CEIC with such general legislative power, the ERO, once invoked, seems to us to create in Hong Kong a separate source of laws that are primary legislation in all but name, but which are not made by the legislature in accordance with legal procedures (Art 73(1)) or reported to NPCSC (Art 17), and are not subjected to the scrutiny concomitant with the normal legislative process. Whenever the CEIC considers an occasion falling within the ERO has arisen, the CEIC becomes a legislature. Instead of the Government drafting and introducing bills (Art 62(5)) and the LegCo (by passing such bills) enacting, amending or repealing laws (Art 73(1)), the CEIC enacts, amends or repeals laws by the power given under the ERO. The LegCo, without bills introduced by the government, is left with a diminished role. 81. The conclusion seems to us to be inevitable that by the ERO, powers to make laws generally rather than merely subordinate legislation are conferred on the CEIC, unless one takes the view that subordinate legislation means no more and no less than laws made pursuant to powers conferred by a primary statute — a view to which we are unable to subscribe. (5) The two Full Court’s decisions and the theme of continuity 82. The ERO had come before the courts of Hong Kong before 1997, particularly in two cases in the 1950s. In To Lam Sin, supra, a person was indicted for being in possession of hand grenades in contravention of reg 116A of the Emergency (Principal) Regulations 1949 made under the ERO. The offence created was punishable with death. The accused moved to quash the indictment on the ground, inter alia, that the ERO was ultra vires the LegCo. The Full Court (Howe CJ, Gould and Scholes JJ), to which the relevant point of law was referred, did not find the “effacement” test used in Ping Shek & Anor v The Canossian Institute (1949) 33 HKLR 66 — which asked whether by the delegation the Legislature had wholly or partly effaced itself — to be of assistance (p 12). They considered that the trend of modern opinion was “to regard a Colonial legislature as being, not mere delegates of Imperial power, but supreme within their own limits, and within the powers conferred by the Letters Patent”. They relied on the Privy Council’s decision in Hodge v The Queen (1883) 9 App Cas 117, where it was said at p 132 that a colonial legislature was conferred “powers not in any sense to be exercised by delegation from or as agents of the Imperial Parliament, but authority as plenary and as ample within the limits prescribed by sect. 92 [of the British North America Act 1867] as the Imperial Parliament in the plenitude of its power possessed and could bestow”. At p 14 the Full Court held: “ …The power of a colonial legislature to delegate is a full one, limited only by the necessity not to go outside the powers conferred by or contravene the rights reserved by the Letters Patent or other constitutional document. As is well known, delegation of powers almost parallel with those given by the Emergency Regulations Ordinance has been resorted to frequently in England under the various Emergency Powers Acts. If the legislature of Hong Kong is supreme (subject to its constitution) in its own area there can be no reason why it should not act similarly — it is not and cannot be suggested that the law is not one for the ‘peace, order, and good government’ of the Colony. Even by the ‘effacement’ test, we would not hold that the delegation of the powers is ultra vires. Wide though the powers may be, the Legislative Council retains a very firm and close control by virtue of Section 14 of the Interpretation Ordinance (Cap. 1). No regulation involving the imposition of the death penalty can become of force or effect without the prior approval of the Legislative Council — this is provided specifically by the Emergency Regulations Ordinance as well. All other regulations must be laid on the table at the first meeting of the Legislative Council after their publication in the Gazette and the Council may repeal or amend any of them. There is in addition the overriding power to repeal or amend the Ordinance itself. We see nothing there which can be called effacement as we understand it.” 83. Pausing here, we note that an examination of the (UK) Emergency Powers Act 1920, for example, shows that it was in fact far narrower than the ERO. It deals with situations where action has been taken or is threatened of such a nature and scale as to be calculated to deprive the community of the essentials of life by interfering with the supply and distribution of food, water, fuel, or light, or with the means of locomotion. It provides for the Monarch to declare, by proclamation, a state of emergency which can be in force for no more than a month (unless proclaimed again), whereupon Parliament has to be summoned to meet within five days. It enables the Executive during a proclaimed emergency to make regulations for “securing the essentials of life to the community”. Any regulation so made shall immediately be laid before Parliament, and shall not continue in force after the expiration of seven days therefrom unless a resolution is passed by both Houses for its continuance. In short, we do not think it provides support for the validity of the ERO beyond showing that a sovereign parliament may by legislation confer on the executive government some sort of regulation‑making power to cope with emergencies. 84. The other case is Li Bun, supra, which was an appeal by way of case stated against convictions for attempting to export motor vehicles without a licence contrary to the Importation and Exportation Ordinance, as amended by the Emergency (Importation and Exportation) (Amendment) Regulations 1953 and 1954 made under the ERO. Specifically, those regulations sought to amend the Importation and Exportation Ordinance by making a particular contravention an offence and imposing the attendant penalty. The question submitted to the Full Court was whether the regulations themselves were ultra vires the Governor in Council, but it was treated as involving the question whether the ERO was ultra vires the legislature of Hong Kong. The Full Court (Hogan CJ, Gould and Reece JJ) again held that it was not. 85. Like the Full Court in To Lam Sin, Hogan CJ also did not find it useful to approach the question by reference to the effacement test. Instead, having referred to R v Burah, supra, and In re The Initiative and Referendum Act, supra, he examined the question whether the legislature had created a co‑ordinate legislative power of a concurrent or alternative character. He considered that the ERO did confer, but for one limitation, “general legislative powers” on the CEIC (p 100), and that although the position of the LegCo was not directly affected by the ERO, “it ha[d] not … an effective role to play in the enactment of Emergency Regulations which may range over virtually the whole field of legislation” (p 102). Although his Lordship was plainly much exercised by the width of the power given to the Governor in Council, he eventually concluded: “ It seems to me that a significant difference [from the case of In re The Initiative and Referendum Act] lies in the fact that the powers conferred by the Hong Kong Ordinance are limited to those occasions which in the opinion of the Governor in Council are occasions of public danger or emergency. This, it seems to me, is the one factor or at any rate the principal factor which prevents them from being regarded as that arming, by a dependent Legislature, of another authority with general legislative authority similar in complexity to its own, on which the Privy Council has frowned. It may be argued that to hold so ample a power as falling below the line which cannot be crossed by the Legislative Council, is to push that line so high as to make it almost meaningless; nevertheless it is a real limitation and since it is the principal reason for not treating the ordinance as ultra vires ab initio, a limitation which must, I think, be strictly observed and strictly enforced, since any tendency to regard it as a mere formality would tend to diminish the importance of the principal, if not indeed the only, factor which saves this ordinance from being ultra vires.” 86. Gould J concurred in the reasoning and conclusions of Hogan CJ except on one matter which is not relevant for present purposes (see p 113). Reece J agreed with the Full Court’s decision in To Lam Sin (see pp 117, 125). Like the court in that case, he placed emphasis on the fact that the LegCo was “invested with the fullest powers of making laws” and had been “held to be, within its limits, … a sovereign body” (p 126) and that “subject to the limitations imposed by the Letters Patent or other constitutional documents, the Legislative Council has full power to delegate its powers” (p 127). 87. On behalf of the respondents, Mr Yu has understandably placed reliance on these two cases and on the “theme of continuity” (Luk Ka Cheung, supra, at §32). The theme of continuity is of course an important theme in the Basic Law. This is particularly prominent in its provisions relating to the courts and the legal system: Art 8 provides that subject to exceptions, the laws previously in force in Hong Kong shall be maintained (see also Art 18); Art 81 provides that the judicial system previously practised in Hong Kong shall be maintained subject to changes consequent upon the establishment of the Court of Final Appeal; Art 87 provides that in criminal or civil proceedings, the principles previously applied in Hong Kong and the rights previously enjoyed by parties to proceedings shall be maintained; Art 91 states that the previous system of appointment and removal of judicial officers shall be maintained; Art 94 provides for the admission of lawyers based on the system previously operating in Hong Kong. 88. On the other hand, there is also obviously a change in the constitutional order brought about by the resumption of exercise of sovereignty by the People’s Republic of China over Hong Kong and the coming into effect of the Basic Law as the foundation of the Hong Kong SAR’s constitutional and legal system. As Ms Li points out on behalf of the applicants in HCAL 2945, under Art VII of the Letters Patent, it was the Governor who, by and with the advice and consent of the LegCo, might make laws for the peace, order, and good government of colonial Hong Kong. Under Art X, the Governor had a discretion to assent or refuse assent to a bill passed by the LegCo or reserve it for the Crown’s signification; and if the Governor refused to give assent, there was nothing the LegCo could do about it. Under the Basic Law, the LegCo alone is the legislature of Hong Kong (Art 66), and while a bill passed by the LegCo may take effect only after it is signed and promulgated by the CE (Art 76), there are carefully calibrated provisions requiring the CE either to sign bills or refuse to do so in which case a chain of events would be set in motion that could result in the resignation of the CE (Arts 49, 50 & 52). There is basis for the observation in A Companion to the history, rules and practices of the Legislative Council of the Hong Kong Special Administrative Region at §2.6 that the law‑making power of the LegCo under the Basic Law is “in substance different from that enjoyed by the pre‑1997 Legislative Council whose constitutional role was to provide advice on, and give consent to, bills which the Governor then enacted into law in exercise of the law‑making powers conferred on him under the Letters Patent”. 89. Further, the notion that the colonial legislature was sovereign and supreme within its province, which underpinned the reasoning of the Full Court in To Lam Sin and Li Bun (especially Reece J), is no longer an apt description of the LegCo. The legislature was conceived by the court in To Lam Sin as supreme in its own area in the way the Imperial Parliament was treated under English law as having power that was “absolute and without control”.[33] But under the new constitutional order, within the Hong Kong SAR it is the Basic Law that is supreme, and even the legislature cannot act contrary to a requirement under the Basic Law: Chief Executive of the HKSAR v President of the Legislative Council [2017] 1 HKLRD 460, §§24‑25, 86‑87. The legislature no longer has the plenary power enjoyed by the Imperial Parliament but that which is conferred expressly or by implication on it under the Basic Law. The Basic Law devotes 14 articles (Arts 66‑79) in Chapter IV to the LegCo and specifies the method of its formation and its voting procedures in an annex (Annex II). The method for forming the first LegCo (1997‑1998) was separately the subject of a decision of the NPCSC dated 4 April 1990. The LegCo is constituted by and subject to the Basic Law, derives its powers from the Basic Law and has to exercise them in accordance with the provisions of the Basic Law. 90. In our opinion, there is a difference between a constitutional order which prescribes the legislature’s authority to make certain laws and binds the legislature to legislate according to certain procedures, and one which treats the legislature as supreme: see The Executive Council of the Western Cape Legislature & Others v The President of the Republic of South Africa & Others 1995 (4) SA 877, §59. 91. The theme of continuity has also to be assessed in the context of the constitution of the bodies in question. The ERO was enacted at a time when the LegCo consisted of the Governor, the Official Members (ie members that were officials of the Government) and not more than six Unofficial Members, and was presided over by the Governor.[34] Even by the 1950s, at the time when To Lam Sin and Li Bun were decided, there were ten Official Members and only eight Unofficial Members, and the Unofficial Members were then of course appointed by none other than the Governor. This had enabled a submission to be made in Ping Shek, supra (which was recorded at p 72 and not expressly rejected by the court), that the Governor in Council was a body that “had actually, through its members, a controlling voice in the Legislature itself”. As Evatt J said in Dignan, supra, at p 114, in dealing with the “separation” of legislative and executive powers, the underlying framework of government needs to be borne in mind. In Australia there is the notion of the British system of an Executive which is responsible to Parliament. In the United Kingdom the Executive is represented in and usually control the majority of the Parliament. 92. Now under the Basic Law, the transfer of general legislative power by the ERO has to be examined in the context of a constitutional framework that seeks to ensure that laws are enacted, amended or repealed by a legislature which is constituted by election and whose composition is carefully prescribed (Art 68 and Annex II). The LegCo, moreover, is quite separate from the Government. There may be overlap in the Executive Council in that Members of the LegCo may be appointed to serve on the Executive Council (Art 55), but it should be noted that under the Basic Law (Art 56), the CE need only consult the Executive Council and is not obliged to accept its majority opinion provided the specific reasons are put on record. 93. The reasoning of Hogan CJ in Li Bun, supra, at pp 100 & 102 actually supports the conclusion that inasmuch as the ERO confers “general legislative powers” on the CEIC (subject only to a limitation as to the occasions of public danger or emergency), the LegCo is deprived of any “effective role” to play in the making of regulations which may range over virtually the whole field of legislation. His Lordship’s ultimate conclusion that the ERO did not cross the fine line under the then constitutional set‑up is in our view not applicable to the constitutional order under the Basic Law. (6) Art 160 of the Basic Law 94. Mr Yu also placed reliance on Art 160 of the Basic Law. That article makes provision for the NPCSC to declare any laws previously in force in Hong Kong to be in contravention of the Basic Law and therefore to be excluded from the laws to be adopted as laws of the Hong Kong SAR. It is submitted that the constitutionality‑check was conducted prior to 1997 and the NPCSC did not consider the ERO to be in contravention of the Basic Law. The respondents only rely on this as a factor in favour of validity, and do not submit that the absence of rejection by the NPCSC has the effect of precluding any future finding that the ERO contravenes the Basic Law. Indeed, Art 160 envisages that laws may subsequently be discovered to be in contravention of the Basic Law. In the absence of any further information about that screening process and the reasoning and materials involved, however, we are respectfully unable to place overriding weight on this factor. (7) Arts 18(4) and 72(5) of the Basic Law 95. The respondents have stressed that the ERO is intended to deal with occasions of emergency or public danger, and that regulations made thereunder are intended to be temporary measures necessitated by the exigencies of the situation. In response, the applicants point to the existence of provisions in the Basic Law that deal with emergencies, including Art 18(4) which provides for the application of relevant national laws and Art 72(5) which enables the President of the LegCo to call emergency sessions on the request of the CE. It is fair to say, however, that Art 18(4) does not cater for all emergencies but only for situations where a state of war is declared or where there is such turmoil within Hong Kong as to endanger national unity or security and to be beyond the control of the local government. As to legislation by the LegCo in an emergency, while there were examples in the past (the last of which took place on 9 July 1997 in the Provisional LegCo) where a bill went through all three readings to become law within a single day, there are in practice and reality likely to be difficulties that stand in the way of swift and decisive action. Yet as a broad statement of approach, we agree with the view of the Supreme Court of Ireland in Bederev v Ireland [2016] IESC 34 (at §25) that the need for an urgent response is no justification for departing from or impugning the constitutional scheme. 96. Having said this, we do not wish it to be thought to be our opinion that the Basic Law categorically precludes any emergency powers from being given to the Executive. Rigidity is not a virtue in constitutional interpretation, and one recalls the adage that a constitution that will not bend will break. We have not been addressed on the possibility that states of emergency necessitating urgent action can occur from which an implication can arise out of necessity that the LegCo can in wide terms authorise the Executive authorities to take necessary action: see eg The Executive Council of the Western Cape Legislature, supra, §§62, 140; Cheng Kar Shun v Li Fung Ying [2011] 2 HKLRD 555 at §203, and consequently we ought not to express any view on it. (8) Conclusion 97. It is the power and function of the LegCo as the designated legislature of the Hong Kong SAR to legislate. Other bodies cannot consistently with the constitutional framework be given general legislative power but only the power to make subordinate legislation. It may be a matter of degree whether a power granted is in truth general legislative authority rather than the acceptable power to make subordinate legislation. But insofar as the public danger ground is concerned, the ERO is so wide in its scope, the conferment of powers so complete, its conditions for invocation so uncertain and subjective, the regulations made thereunder invested with such primacy, and the control by the LegCo so precarious, that we believe it is not compatible with the constitutional order laid down by the Basic Law having regard in particular to Arts 2, 8, 17(2), 18, 48, 56, 62(5), 66 and 73(1) of the Basic Law. We do not consider that, within the proper limits of remedial interpretation as set out in HKSAR v Lam Kwong Wai (2006) 9 HKCFAR 574 at §66 and Keen Lloyd Holdings Ltd v Commissioner of Customs and Excise [2016] 2 HKLRD 1372 at §97, the ERO in relation to the public danger ground could be made compatible with the Basic Law without introducing changes that the court is ill‑equipped to decide on or producing something wholly different from what the legislature originally intended. F. Ground 2 — the implied repeal ground 98. Section 3(1) and (2) of HKBORO provided, prior to July 1997 (when they were not adopted as part of the laws of the Hong Kong SAR), as follows: “ (1) All pre‑existing legislation that admits of a construction consistent with this Ordinance shall be given such a construction. (2) All pre‑existing legislation that does not admit of a construction consistent with this Ordinance is, to the extent of the inconsistency, repealed.” 99. Section 5 of HKBORO provides: “ (1) In time of public emergency which threatens the life of the nation and the existence of which is officially proclaimed, measures may be taken derogating from the Bill of Rights to the extent strictly required by the exigencies of the situation, but these measures shall be taken in accordance with law. (2) No measure shall be taken under subsection (1) that— (a) is inconsistent with any obligation under international law that applies to Hong Kong (other than an obligation under the International Covenant on Civil and Political Rights); (b) involves discrimination solely on the ground of race, colour, sex, language, religion or social origin; or (c) derogates from articles 2, 3, 4(1) and (2), 7, 12, 13 and 15.” 100. This section reflects, mutatis mutandis, the provisions of Art 4 of the ICCPR.[35] 101. The argument on this Ground, advanced by the applicants in HCAL 2945, is in summary as follows. (1) Section 5 of the HKBORO permits derogation from the Bill of Rights only in the exceptional case of a public emergency, where the “life of the nation” is at risk, and puts such power of derogation under strict control, for example, by requiring an official proclamation of a public emergency. (2) Section 2(1) of the ERO, which empowers the CEIC to make regulations that derogate from fundamental rights not in times of public emergency and without any safeguards, is inconsistent with s 5 of the HKBORO and, therefore, is to be treated as having been automatically repealed by s 3(2) of the HKBORO when the HKBORO came into effect on 8 June 1991.[36] (3) Having been thus repealed, the ERO is not a law previously in force in Hong Kong as referred to in Art 8 of the Basic, that is to say, in force as at 30 June 1997: see HKSAR v Ma Wai Kwan David & Others [1997] HKLRD 761, 777B. (4) Notwithstanding that s 3(1) and (2) of the HKBORO was in turn effectively repealed on 1 July 1997 as previous laws not adopted as provisions of laws of the Hong Kong SAR,[37] the ERO was not thereupon revived: see s 24 of the IGCO.[38] (5) Accordingly, s 2(1) of the ERO was repealed either in its entirety, or insofar as it enabled the CEIC to make regulations other than in times of public emergency which threatens the life of the nation, and the existence of which is officially proclaimed. 102. There is an alternative, but materially identical, argument that the ERO is inconsistent with Art 4 of the ICCPR and was therefore likewise implicitly repealed when the Basic Law commenced operation on 1 July 1997. As this argument adds nothing, we shall concentrate on the argument based on s 5 of the HKBORO. 103. We acknowledge that a public emergency threatening the life of a nation is a much narrower concept than “public danger” in the ERO. The phrase has been said in the Syracusa Principles on the Limitation and Derogation Provisions in the ICCPR, at §39, to refer to “a situation of exceptional and actual or imminent danger which threatens the life of the nation. A threat to the life of the nation is one that: (a) affects the whole of the population and either the whole or part of the territory of the state; and (b) threatens the physical integrity of the population, the political independence or the territorial integrity of the state or the existence or basic functioning of institutions indispensable to ensure and protect the rights recognised in the Covenant.” 104. The ERO is by its terms intended to apply not only in emergencies that fall within the meaning of s 5 of HKBORO. But we do not accept there is therefore any incompatibility with s 5. With respect, the flaw in the applicants’ argument is the conflation of the concepts of derogation from the Bill of Rights itself and restriction of non‑absolute rights under and in compliance with the Bill of Rights. Derogation provisions such as found in s 5 of the HKBORO and Art 4 of the ICCPR allow for a state lawfully to suspend the human rights guarantees in order to respond to extraordinary circumstances that threaten the life of the nation: see R Burchill, When does an Emergency threaten the Life of the Nation? Derogations from Human Rights Obligations and the War on International Terrorism (2005) 8 Year of New Zealand Jurisprudence 99, at 99‑100. They are to be distinguished from provisions within the human rights norms that do not suspend such norms but instead permit proportionate restrictions or limitations of the rights in question in accordance with law. 105. The respondents accept, rightly, in our view, that except in times of public emergency officially proclaimed as referred to in s 5 of the HKBORO, the powers under the ERO may not be exercised with the effect of derogating from — or, in other words, suspending — the Bill of Rights. But this does not mean that, in situations not amounting to a public emergency within the meaning of s 5, measures may not be taken under the ERO which have an effect of restricting the rights protected by the Bill of Rights, provided the restriction is prescribed by law and compliant with the principle of proportionality. 106. The distinction between derogation in an emergency on the one hand and restriction or limitation of non-absolute rights in ordinary situations on the other is explicitly drawn, in relation to the ICCPR, in UN Human Rights Committee, CCPR General Comment No 29: Article 4: Derogations during a State of Emergency, 31 August 2001, CCPR/C/21/Rev.1/Add.11, at §4: “ Derogation from some Covenant obligations in emergency situations is clearly distinct from restrictions or limitations allowed even in normal times under several provisions of the Covenant. Nevertheless, the obligation to limit any derogations to those strictly required by the exigencies of the situation reflects the principle of proportionality which is common to derogation and limitation powers.” 107. Indeed, numerous laws enacted by the LegCo itself both before and after the HKBORO, not in any emergency, make provisions that impinge upon fundamental rights — see, for example, the POO and Defamation Ordinance (Cap 21). These laws do not thereby “derogate from the Bill of Rights” and violate s 5 of the HKBORO and became impliedly repealed in June 1991. Instead, they continue in force subject to the Bill of Rights and, if challenged, will have to be tested by reference to the usual principles. 108. Thus analysed, the proper approach to the ERO is that: (1) In times of a public emergency officially proclaimed and in accordance with the other requirements of s 5 of the HKBORO, measures may be adopted under the ERO which derogate from the Bill of Rights (even so, excepting the specified non‑derogable provisions and discrimination on the prohibited grounds). Subject to the conditions of s 5 (including that the derogations are limited to those strictly required by the exigencies of the situation), this may have the effect of temporarily suspending the relevant human rights norms. (2) In other situations, measures adopted under the ERO may not derogate from the Bill of Rights, which means that if any such measure has the effect of restricting fundamental rights, then like any other restriction in normal times, it has to satisfy the twin requirements that the restriction is prescribed by law and meets the proportionality test: Leung Kwok Hung & Others v HKSAR (2005) 8 HKCFAR 229, §§16-17. 109. Accordingly, the challenge based on Ground 2 is rejected. G. Ground 3 — the prescribed by law ground 110. The second sentence of Art 39 of the Basic Law provides: “ The rights and freedoms enjoyed by Hong Kong residents shall not be restricted unless as prescribed by law.” The phrase “prescribed by law”, and the cognate concepts of “provided by law” in Art 16 and “in conformity with the law” in Art 17 of the Bill of Rights, import the same principle. The principle is that of legal certainty and accessibility: Hysan Development Co Ltd v Town Planning Board (2016) 19 HKCFAR 372, §30. 111. As understood in Hong Kong law, this principle not only requires the restriction to have a basis in law but entails two further requirements, namely, that (1) the law must be adequately accessible: the citizen must be able to have an indication that is adequate in the circumstances of the legal rules applicable to a given case; and (2) the norm is formulated with sufficient precision to enable the citizen to regulate his conduct so that he is able — if need be with appropriate advice — to foresee, to a degree that is reasonable in the circumstances, the consequences which a given action may entail: see Shum Kwok Sher v HKSAR (2002) 5 HKCFAR 381, §§62‑63; Leung Kwok Hung & Others v HKSAR (2005) 8 HKCFAR 229, §27; Hong Kong Television Network Ltd v Chief Executive in Council [2016] 2 HKLRD 1005, §84; Hysan Development Co Ltd v Town Planning Board (2016) 19 HKCFAR 372, §§30-32. 112. The argument of the applicants in both HCAL 2945 and HCAL 2949 is this. The ERO contemplates and enables the making of regulations which can severely curtail fundamental rights. Section 2(1) is, however, couched in wide terms and lacks any express limit or guidance on the exercise of the power by the CEIC in making regulations that may affect fundamental rights. As such, s 2(1) of the ERO violates the principle of legal certainty mandated by Art 39 of the Basic Law. What the argument focuses on is the certainty (or the lack of it) in the regulation‑making power in the ERO, rather than that in respect of any regulations made under the ERO. 113. We do not accept this argument. Legal certainty is not a notion existing in the abstract and in a vacuum. Art 39 of the Basic Law provides that the provisions of the ICCPR as applied to Hong Kong shall remain in force, and states that those rights and freedoms shall not be restricted unless as prescribed by law. As such, the “prescribed by law” requirement applies to the restraints on the rights and freedoms of the individual. It is the “norms” which purport directly to restrict the citizen’s freedom that must be sufficiently precise to enable the citizen to conduct himself accordingly. This is how the requirement has been applied in the decisions of the Court of Final Appeal. Thus in Shum Kwok Sher v HKSAR (2002) 5 HKCFAR 381 and Mo Yuk Ping v HKSAR (2007) 10 HKCFAR 386, the requirement was applied to the common law offence of misconduct in public office and conspiracy to defraud respectively. In Leung Kwok Hung & Others v HKSAR (2005) 8 HKCFAR 229, it is the statutory scheme in the POO restricting the right of assembly and procession that was called into question by reference to the requirement of “prescribed by law”. In Hysan Development Co Ltd v Town Planning Board (2016) 19 HKCFAR 372, the point made was that property rights are to be protected by clear and accessible laws. In HKSAR v Fong Kwok Shan Christine (2017) 20 HKCFAR 425, the question was whether a provision in the administrative instructions issued by the President of the LegCo to regulate behaviour within the precincts of the LegCo’s chamber was sufficiently certain. See also Hong Kong Television Network Ltd v Chief Executive in Council [2016] 2 HKLRD 1005 (CA), §90. 114. In contrast, the ERO does not itself purport to limit any fundamental rights. It does not lay down any norms that curtail any right of an individual. Although regulations enacted under it may purport to do so, it seems to us that it is those regulations, if and when enacted, that have to meet the principle of legal certainty, not the enabling Ordinance in itself which has no direct effect on any individual right or freedom. If such regulations are themselves laws of general application, accessible to all residents, and sufficiently well defined, then they cannot be said to fall foul of the requirement of accessibility and foreseeability which is the essence of the principle. 115. The ERO, as the source of power for making regulations, cannot be attacked on its own under the “prescribed by law” requirement. This is not to say that it can never be a matter of concern that executive authorities are given ill‑defined powers to make laws that may restrict fundamental rights, but this seems to us to raise the analytically separate and different point in relation to delegation of legislative power, which we have dealt with under Ground 1 above. In addition, the laws thus made will themselves have to possess the quality of accessibility and to afford sufficient safeguards against arbitrary application by indicating with sufficient clarity the scope of any discretionary power conferred. 116. The case of Leung Kwok Hung & Another v Chief Executive of the Hong Kong SAR (HCAL 107/2005, 9 February 2006) relied on by the applicants can be readily distinguished.[39] There, s 33 of the Telecommunications Ordinance (Cap 106) provided that the Chief Executive or authorised public officers could order that any message or any class of messages should not be transmitted or should be intercepted or detained or disclosed to the Government. It was therefore a law that directly authorised executive measures which had the effect of directly restricting the freedom and privacy of communication protected by Art 30 of the Basic Law and Art 14 of the Bill of Rights, and was found to be unconstitutional for falling foul of the principle of legal certainty. In our view, the case provides no support for suggesting that an empowering statute for making subsidiary legislation without itself purporting to limit any rights can be struck down under the “prescribed by law” requirement. The same may be said in relation to Malone v United Kingdom (1984) 7 EHRR 14 which concerned the lawfulness of interception of communications effected by the UK police with reference to the right to respect for private life and correspondence. 117. As to the case of Dawood & Anor v Minister of Home Affairs & Others 2000 (3) SA 936, Mr Pun for the applicant in HCAL 2949 relies on footnote 74 in paragraph 54 of the judgment of O’Regan J (with which the other members of the court agreed). Her Honour there referred to s 56(1)(f) of the Aliens Control Act 96 of 1991 which provided that the Minister “may make regulations relating to … the conditions subject to which such permits or certificates may be issued …”, and said that affording the Executive power to regulate such matters is not sufficient; the Legislature must take steps to ensure that appropriate guidance is given. With respect, Mr Pun’s reliance on that sentence is misplaced as it involves taking a footnote completely out of context. What was in issue in that case was that the discretionary power under s 25(9)(b) of the Act to issue a permit for a person to temporarily sojourn in South Africa engaged fundamental rights but was too vague and without guidance. It was in this context that the court said that guidance must be given and that it was not sufficient that s 56(1)(f) stated regulations may be made to provide guidance, if they were not in fact made. Properly read, the footnote provides no assistance to the applicants on this Ground. 118. Mr Chan submits that the restriction on rights must be prescribed by laws passed and scrutinised by elected representatives in the legislature. The only authority cited is Advisory Opinion OC‑6/86 (9 May 1986) of the Inter-American Court of Human Rights on the word “Laws” in Art 30 of the American Convention on Human Rights (at §22). We are unable to accept this contention. The Advisory Opinion was given in the context of a very different instrument in a very different region with certain assumptions about the member States’ constitutions including the existence of representative democracy. Such an approach cannot be mechanistically transplanted to Hong Kong, where many of our statutes were enacted in an era with no elected representative in the legislature at all. It could not have been the intention in enacting the Bill of Rights and the Basic Law that restrictions placed upon fundamental rights under those statutes are, for that reason alone, not to be regarded as prescribed by law. 119. Furthermore, it is well established that for the purpose of the requirement of “prescribed by law”, law does not necessarily mean only statute law. In A v Director of Immigration [2008] 4 HKLRD 752, the Court of Appeal plainly considered that an accessible and sufficiently certain policy on detention would be sufficient to satisfy the requirement under Art 5 of the Bill of Rights that no one shall be deprived of his liberty except on such grounds and in accordance with such procedures as are established by law (see §§16, 33-37, 42-43, 63), although the Director of Immigration failed in that case because he did not have an adequately accessible and complete policy on detention. Similarly, in R (Munjaz) v Mersey Care NHS Trust [2006] 2 AC 148, where the question was whether guidance given by the Secretary of State and written guidelines given by a hospital were sufficient to justify certain interference with the right to private life in accordance with law, Lord Bingham said that the requirement of “in accordance with the law” is directed to substance and not form (§34); Lord Hope said that “law” in this context is not limited to statutory enactment or to measures that have their base in a statute (§91); and Lord Scott also said that the “law” for the purposes of article 8 of the European Convention for the Protection of Human Rights and Fundamental Freedoms (“ECHR”) does not consist only of statutes, directives, statutory codes and the like (§103). The cases of Shum Kwok Sher, supra, and Mo Yuk Ping, supra, show that common law offences can also constitute the “law” that prescribes an acceptable restriction on fundamental rights. It is plain from all these authorities that “prescribed by law” does not refer only to laws passed and scrutinized by elected representatives in the legislature. 120. For the above reasons, Ground 3 is rejected. H. Ground 4 — the principle of legality ground 121. The fourth ground impugns the PFCR as being ultra vires the ERO. The essence of the argument is that, properly construed in accordance with the principle of legality, the ERO does not, either expressly or by necessary implication, empower the CEIC to make regulations that impose restrictions on fundamental rights of the kind and to the extent found in the PFCR. 122. The principle of legality, supported by overseas authorities such as R v Secretary of state for the Home Department, ex p Simms [2000] 2 AC 115, 131 and Coco v The Queen (1994) 179 CLR 427, 437, has been acknowledged by the Court of Final Appeal in A v Commissioner of Independent Commission Against Corruption (2012) 15 HKCFAR 362 at §§28‑29, 67‑71 as applicable in cases where the question arises as to whether certain legislation is intended to override or constrain fundamental rights. It is a principle of statutory construction which requires that any abrogation or restriction of fundamental rights by statute should be done unmistakably, ie expressly or by necessary implication. In other words, the court has to be satisfied that the legislature had its attention properly drawn to the abrogating provision and consciously enacted legislation to such effect: HKSAR v Yeung Ka Sing Carson (2016) 19 HKCFAR 279, §54. 123. The applicants in HCAL 2945 argue that s 2(1) of the ERO is broad and general and does not advert to any specific fundamental right. Neither s 2(1) nor s 2(2) deals specifically with the subject of assemblies and processions, let alone facial covering during such activities. Further, the examples in s 2(2) do not show a general intention to interfere with fundamental rights in any circumstances. Instead, they are restricted to a number of strategic areas in time of emergency. Section 2(1) should not be construed to confer power to legislate wider than in those areas. 124. In response, the respondents contend that it is plain that the legislature must have had the restriction of the freedom of individuals in mind when enacting the ERO. Section 2(2) refers to censorship, arrest, detention, appropriation and forfeiture of property, entry and search of premises, taking of possession or control of any property or undertaking, and requiring persons to do work or render services. Section 3(1) enables the regulations made to provide for the punishment of offences. Reading the ERO as a whole, the respondents say, there can be no doubt that the legislature did have in mind the potential restriction of rights and freedoms by the making of regulations, including but not limited to the specific measures set out in s 2(2)(a) to (n). 125. A tension can be detected between Ground 1 and Ground 4 as advanced by the applicants. Under Ground 1, the applicants submit that s 2(1) of the ERO is of the widest scope, essentially conferring an unrestricted and unfettered legislative power. Ground 4, as we see it, is essentially an alternative ground, contending instead that s 2(1) is to be read as not authorising any regulation to be made that would restrict fundamental rights. Having upheld the applicants’ contention on Ground 1, it is not necessary for us to deal with Ground 4. I. Ground 5A — the section 3 proportionality ground (1) The rights engaged 126. Section 3 of the PFCR has been set out in §22 above. Essentially it prohibits a person at the types of gatherings specified from using any facial covering that is likely to prevent identification, and makes it an offence punishable with a fine of $25,000 and imprisonment for 1 year. 127. There is no dispute that a number of rights are engaged by the restrictions imposed by s 3 of the PFCR, including the freedom of assembly, procession and demonstration (Art 17 of the Bill of Rights;[40] Art 27 of the Basic Law[41]), the freedom of speech or expression (Art 16 of the Bill of Rights[42]; Art 27 of the Basic Law), and the right to privacy (Art 14 of the Bill of Rights[43]). These rights are not absolute and may be subject to lawful restrictions. 128. There is equally no dispute that whether or not the restrictions are legally valid is to be determined by a 4‑step proportionality analysis, asking: (1) does the measure pursue a legitimate aim; (2) if so, is it rationally connected with advancing that aim; (3) whether the measure is no more than reasonably necessary for that purpose; and (4) whether a reasonable balance has been struck between the societal benefits promoted and the inroads made into the protected rights, asking in particular whether pursuit of the societal interest results in an unacceptably harsh burden on the individual: Hysan Development Co Ltd v Town Planning Board (2016) 19 HKCFAR 372, §§134‑135; HKSAR v Choi Wai Lun (2018) 21 HKCFAR 167, §68. 129. We shall analyse the restrictions imposed by s 3 and s 5 of the PFCR separately. (2) Step (1): legitimate aims 130. On behalf of the Government, Mr Yu submits that the aims sought to be pursued by the PFCR are two‑fold, namely: (i) deterrence and elimination of the emboldening effect for those who may otherwise, with the advantage of facial covering, break the law, and (ii) facilitation of law enforcement, investigation and prosecution. It cannot be disputed that these are legitimate aims which the Government may lawfully pursue, both of which would serve to promote the interests of public order and public safety. In this regard, it is relevant to note that many of the provisions of the Bill of Rights expressly permit restriction or limitation of the relevant rights for the purpose of protection of public order and/or public safety: see Art 15(3) (freedom of thought, conscience and religion), Art 16(3)(b) (freedom of opinion and expression), Art 17 (right of peaceful assembly), and Art 18(2) (freedom of association), of the Bill of Rights. 131. In SAS v France (2015) 60 EHRR 11, the law in question provided that: “No one may, in public places, wear clothing that is designed to conceal the face”. The Grand Chamber of the European Court of Human Rights held that the right to respect for private life and the freedom to manifest one’s belief under Arts 8 and 9 of the ECHR were engaged.[44] The court also accepted that in adopting the ban, the State sought to advance the legitimate aim of “public safety” within the meaning of the Convention.[45] We shall come back to this case later in our discussion of whether the restrictions imposed by s 3 of the PFCR are proportionate to the pursuit of the legitimate aims as identified by the Government. 132. In this connection there is evidence before us of the enormity of the damage and danger created by some of the protesters. In the few months leading to 4 October 2019, Hong Kong has witnessed numerous instances where certain protesters charged police cordon lines with weapons, blocked public roads and tunnels with a variety of large and heavy objects, attacked drivers who voiced complaints at such blockades, vandalised public facilities and buildings, burned public property, hurled inflammable liquid bombs at the police and at and inside Mass Transit Railway stations, damaged shopping malls, shops, banks and restaurants (with reports of looting and theft in some of the damaged shops), damaged residential quarters of the disciplined forces, crippled the operations of transport infrastructure, and harassed and attacked ordinary citizens holding different political views. These acts of violence and vandalism had increased in intensity and frequency, with the incidents on 1 October 2019 being especially serious. The more violent protesters were often all suited up and masked by facial covering such as surgical masks, balaclavas and gas masks which concealed their identity. A particularly worrying trend is the apparent increasing number of young persons and students taking part in what appear to be riotous gatherings and criminal acts of violence and vandalism. By 4 October 2019, a total of 223 persons below the age of 18 had been arrested out of a total of 2,135 persons arrested in these events, compared to 67 out of 1,110 as at 1 September 2019. (3) Step (2): rational connection 133. As we understand it, the Government’s arguments that the measure adopted under s 3 of the PFCR promotes or advances the first legitimate aim (ie deterrence and elimination of the emboldening effect for those who may otherwise, with the advantage of facial covering, break the law) run along the following lines: (1) Protesters are able to avoid identification by wearing masks. Those protesters who are bent on resorting to violence rely in substantial part on the support of other protesters. Even though the hardcore radical and violent protesters may continue to flout the law, those protesters who are not prepared to break the law may comply with the PFCR and this would generally result in lessening the support for the more radical and violent protesters. The making of the PFCR will signal a clear disapproval by the law, not merely of acts of violence and vandalism, but also of the illegitimate use of face covering to conceal one’s identity while breaking the law. (2) Face covering clearly emboldens protesters to engage in violent or unlawful acts which they may not otherwise perform without concealment. Masked protesters mix themselves into larger groups and instigate violence and vandalism. Presence of non‑violent protesters also wearing facial coverings in such circumstances makes identification of the violent protesters much harder because they can easily slip away amid chaos aroused by them. Their clean getaways embolden and allow them to redouble their efforts to break the law. (3) The prohibition of face covering will help ensure peaceful protests and demonstration will stay that way. Non‑radical protesters will be less likely to be influenced by or emulate their violent peers and will think twice before emulating them when they know their identity is not concealed. While there is no guarantee that the PFCR will stop all acts of violence and vandalism, it is incorrect to assume that the PFCR will definitely be ignored. At least some protesters will be discouraged from instigating or joining violent or riotous behaviour. (4) In view of the alarming surge in the number of students participating in unlawful activities, the PFCR can act as an effective deterrent against at least some students from wearing masks when joining a protest (lawful or unlawful), which thereby substantially reduces the chance that they will be induced to break the law. 134. In relation to the second legitimate aim (ie facilitation of law enforcement, investigation and prosecution), the Government’s arguments are as follows: (1) Violent protesters often deploy what has been described as the “black bloc” tactic[46] to make it harder for them to be identified, arrested, and successfully prosecuted, and prohibition of face covering would make such tactic much less effective. (2) The PFCR provides an extra tool to the Police in maintaining law and order (including making necessary arrests), particularly in relation to groups of masked radical and violent protesters who might have mingled with non‑radical protesters or otherwise dispersed into the crowd, thereby making it harder to distinguish between violent and non-violent protesters at the scene, and thus harder to make arrests. (3) Police officers need to be able to identify the protesters engaged in unlawful or criminal activities to make arrests and restore public order amid chaos, but the concealment of identity by face covering is a major impediment to law enforcement. As police officers are unable to quickly and effectively identify the protesters, those who manage to escape from the scene may change their clothes soon afterwards and remove their facial coverings in order to avoid arrest. (4) Identification based on the evidence collected at the scene or in the vicinity (like CCTV footage) is crucial in subsequent investigation and prosecution. The “black bloc” tactic and face covering make identification difficult. Prohibiting face covering will help identification of the violent protesters. (5) The PFCR also aims to make the crowd dispersal tools of the Police, such as tear gas and pepper spray, more effective. Past experience shows that masks are being worn by protesters to shield themselves from the effect of tear gas and pepper spray, rendering those measures much less effective, with the result that the Police encountered much more difficulty in dispersing crowds. 135. The question of whether a particular measure is rationally connected to an identified aim is essentially a matter of logic and common sense. The fact that the measure may subsequently prove to be ineffective to achieve the aim does not in itself disprove rational connection, although it would have a bearing on the issue of whether a measure adopted is proportionate to the pursuit of a legitimate aim, and also the issue of reasonable balance. 136. At the rolled‑up hearing, Mr Chan made it clear that he would be concentrating on steps (3) and (4) of the proportionality analysis, and did not make any oral submissions on step (2). Nevertheless, for the sake of completeness, we shall briefly deal with the arguments raised in the written submissions of the applicants in HCAL 2945 that the restrictions under s 3 of the PFCR are not rationally connected to the legitimate aim of the protection of public order and public safety. Essentially, four arguments are raised:[47] (1) Section 3 covers all public meetings, lawful and unlawful, and peaceful or otherwise, and thus it does not in fact pursue the legitimate aim of reducing violence (because in peaceful assemblies there is no violence), or facilitate police investigation and administration of justice (because there is no violence to investigate to begin with). (2) Insofar as authorised and unauthorised but peaceful assemblies or processions are concerned, there is no reason why wearing a mask would turn them into something other than lawful or peaceful that calls for restoration of peace and order. (3) The suggestion that because a lawful and peaceful assembly or demonstration can develop into an unlawful one by the acts of some violent protesters, all law‑abiding citizens should not wear masks in the first place is based on the logical fallacy or false equivalence that because all violent protesters wear masks, all masked protesters are violent. (4) The suggestion that the wearing of a mask would encourage or embolden protesters to engage in unlawful activities that they would otherwise not be willing to without a mask is not supported by evidence. 137. As rightly submitted by Mr Yu, the first argument fails to take into account the evidence that many public assemblies or processions in the past months which took place lawfully and peacefully at the beginning turned into unauthorised or unlawful ones with some radical protesters resorting to violence. In such cases, the prohibition of face covering would deter some protesters, without the advantage of face covering, from committing acts of violence or breaking the law, and facilitate law enforcement, investigation and prosecution for the reasons given by the Government. In our view, the argument is in truth an objection to the width of the measure, which is a matter for consideration under steps (3) and (4) of the proportionality analysis. The fact that a measure may be wider than is reasonably necessary to achieve a legitimate aim does not mean that it is not rationally connected to that aim. Thus, while s 3 of the PFCR may be objected to for its width in that it would also criminalise protesters wearing facial coverings in peaceful and lawful assemblies, it does not mean that the measure adopted under s 3 does not pursue the legitimate aims of deterring violent protests, or facilitating law enforcement, investigation and prosecution of violent protesters. 138. The second argument can be disposed of shortly on the ground that it is no part of the Government’s argument that mask-wearing would necessarily turn a peaceful assembly or procession into an unlawful or violent one. The Government’s case is that many recent public assemblies or processions which took place lawfully and peacefully at the beginning turned, as a matter of fact, into unlawful or violent ones, and the prohibition of face covering at such initially lawful and peaceful assemblies or processions would, or at least may, reduce the risk or likelihood of them being turned into unlawful or violent ones. There is force in Mr Yu’s criticism that the applicants have presumed a simple dichotomy of peaceful and violent protesters. Human nature being what it is, there is likely to be a range of attitudes and predispositions among different people. Furthermore, people’s behaviour may change depending on the circumstances and the influence from others around them. There is no real challenge from the applicants that face covering makes law enforcement, investigation and prosecution more difficult. As has been observed by Fish J (dissenting in the result) in R v Cornell [2010] 2 SCR 142 at §118 (cited in Villeneuve v Montreal (City of), 2016 QCCS 2888 at §486): “Just as anonymity breeds impunity, so too does impunity breed misconduct”. The court in Villeneuve (which is described in greater detail below) also took the view (at §489) that “[l]ogic and common sense suggest that [prohibiting facial covering] has a deterrent effect on persons who choose to cover their faces in order to engage in acts of violence and vandalism under the cover of anonymity during demonstrations” and that “a reasonable inference may be drawn that the measure will aid in realising the objective being pursued”. 139. The above comments apply equally to the third argument. Again, it is no part of the Government’s argument that all masked protesters are violent. 140. In so far as the fourth argument is concerned, the Government has produced an affidavit by Dr Tsui Pui Wing Ephraem, a clinical psychologist, dated 23 October 2019 to explain how the wearing of facial covering affects a person’s psychology and emboldens the wearer to commit acts which he or she might otherwise not commit. In his affidavit, Dr Tsui made clear that he had not personally conducted any study or research on the emboldening effect of face covering, but had carried out a search of the relevant literature bearing on the issue that he was asked to opine on, using online databases such as Google Scholar and the Electronics Resources of The University of Hong Kong Libraries. He identified three papers to be of particular relevance, and presented his literature review on them in his affidavit. At paragraph 17 of the affidavit, Dr Tsui set out his conclusion as follows: “ People make rational decisions or intention about wearing a mask or not or a particular mask. It maybe for physical protection, psychological protection of their identities, or to display their identification with a particular group. An individual wearing a mask may provide him or her a stronger sense of safety or group identification, but that alone does not lead to a loss of self‑awareness or loss of self-regulation as proposed by Silke (2003). It seems that it is when the person joins a group which uses wearing masks as group identification that triggers the deindividuation effect. Research has shown that the larger the group size, the stronger the deindividuation feeing becomes (see Postmes & Spears (1998)). The person will identify more with the group values, situation norms and behaviour, and will begin to set his own values and self-regulations aside. He is not relinquishing his own values and self-regulation permanently because they can return to him after the group action. Mask functions as a facilitator of anonymity. When anonymity joins with group action, participants’ responsibilities become easily diffused or shared. Individuals tend to feel they are being supported by a lot of people. This has an emboldenment effect. If the dominant group value or purpose in the situation is pro‑social, the individual will conform to it and act pro‑socially. However, if the dominant group value or purpose in the situation is anti-social, the individual will conform to that and be more likely to act antisocially. The word ‘antisocial’ refers to acting against larger social norms which, within the group, can be their own emerging group norm. It implies that the person is not really losing his self in the group, but is just following to the group norm.” 141. The applicants in HCAL 2945 object to the admission of the evidence of Dr Tsui on the grounds of: (1) irrelevance; (2) lack of expertise; and (3) procedural unfairness, and invites the court to exercise its case management power to exclude the evidence. 142. On the issue of relevance, it seems to us that Dr Tsui’s evidence supports the proposition that face covering would, or at least could, embolden protesters to commit violent or unlawful acts which they might not otherwise commit without concealment of their identities. The evidence of Dr Tsui is, in our view, relevant to the issue of rational connection between the measure adopted under s 3 of the PFCR and the first legitimate aim relied upon by the Government. For the purpose of showing rational connection, it is not necessary for the Government to prove that face covering would necessarily have the emboldenment effect contended for. It suffices for the Government to show that it could have such effect. 143. On the question of Dr Tsui’s expertise, the applicants in HCAL 2945 argue that Dr Tsui is in no position to offer anything more than comment, and point to the fact that Dr Tsui has not performed any research on the psychological profile specific to Hong Kong and the extent of any emboldening effect in Hong Kong with masks and protesters. In our view, it is not necessary for an expert to have personally carried out some study or research on an issue before he can give expert evidence on it. One of the skills possessed by an expert which the court or an ordinary lay person does not have is the ability to look for and identify relevant research papers and literature in his field of expertise, and explain, interpret and comment on the findings and conclusions in plain language. The fact that an expert may not have personally carried out the relevant study or research is a matter going to the weight (rather than admissibility) of the opinion given by the expert. 144. Lastly, in so far as procedural fairness is concerned, the question of the need for evidence on the issue of the emboldening effect of face covering was mooted at the hearing on 6 October 2019, and the Government took out a summons for admission of Dr Tsui’s evidence on 23 October 2019. In view of the urgency of this application and the short time frame between the commencement of the proceedings and the date of the rolled‑up hearing, we do not consider the Government to have delayed in producing the affidavit of Dr Tsui. The applicants in HCAL 2945 have not put forward any evidence of their own to counter the evidence of Dr Tsui or to support the proposition that face covering does not produce any emboldening effect. While they complain about the absence of any opportunity to cross‑examine Dr Tsui, they did not make any application to the court for cross-examination. Overall, we see no reason, from the perspective of procedural fairness, to exclude the evidence of Dr Tsui. 145. For the above reasons, we would admit Dr Tsui’s affidavit as evidence in these proceedings. As earlier mentioned, we consider Dr Tsui’s evidence to be supportive of the proposition that face covering would, or at least could, embolden protesters to commit violent or unlawful acts which they might not otherwise commit without concealment of their identities. In any event, we consider this proposition to be a matter of common sense which we would readily accept even without the evidence of Dr Tsui. 146. In all, we are satisfied that the measure adopted under s 3 of the PFCR is rationally connected to the two legitimate aims identified by the Government mentioned in paragraph 130 above. (4) Steps (3) and (4): no more than reasonably necessary and reasonable balance 147. The following principles are relevant in step (3) of the proportionality analysis. (1) The test is one of “reasonable necessity”, not “strict necessity”. This is plain from the judgment of Ribeiro PJ in Hysan Development Co Ltd v Town Planning Board (2016) 19 HKCFAR 372, at §§83‑88 and §§119‑122. The question is whether there is “a reasonable relationship of proportionality between the means employed and the aim sought to be realised” (James v United Kingdom (1986) 8 EHRR 123, at §§50‑51). While the court should consider whether “some less onerous alternative would have been available without unreasonably impairing the objective” (R (Lord Carlile of Berriew) v Secretary of State for the Home Department [2015] AC 945, at §34), it does not mean that “the restriction must be the very least intrusive method of securing the objective which might be imagined or devised” (Official Receiver v Zhi Charles (2015) 18 HKCFAR 467, at §53). (2) The yardstick of reasonable necessity is not a strict, bright line, but occupies a continuous spectrum which should be viewed as a “sliding scale” in which the cogency of the justification required for interfering with a right will be proportionate to its perceived importance and the extent of the interference (Hysan, at §§83 and 86). (3) A wide margin of discretion should be given to the Government in the assessment of the necessity in taking measures to restrict unlawful and/or violent conduct disrupting the ordinary life and activities of the majority, law‑abiding, citizens in Hong Kong (Kudrevičius v Lithuania (2016) 62 EHRR 34, at §156), just as the Government is afforded a wide margin of discretion in the choice of reasonable and appropriate measures to enable lawful assemblies to take place peacefully, which, the respondents emphasise, has been said to be a positive duty on the part of the Government (Leung Kwok Hung v HKSAR (2005) 8 HKCFAR 229, at §22). 148. The importance of the rights and freedoms in question hardly needs elaboration. Freedom of expression “lies at the heart of civil society and of Hong Kong’s system and way of life” and its constitutional guarantee must be given a generous interpretation, even where the majority of people find the ideas expressed disagreeable or offensive: HKSAR v Ng Kung Siu & Anor (1999) 2 HKCFAR 442, 455H. As the Court of Final Appeal stated in Leung Kwok Hung v HKSAR, supra, at §2, the freedom of speech and the freedom of peaceful assembly are “of cardinal importance for the stability and progress of society”. As regards the extent of the interference with these rights and freedoms, it will be noted that the PFCR does not restrict public assemblies and processions as such, but only the use of facial covering at such gatherings. Nevertheless, there is no dispute that the freedom of peaceful assembly extends to the manner of its exercise, and that a restriction such as that imposed by the PFCR is an inroad into the relevant right. As has been recognised in many jurisdictions and is, we believe, common ground, some participants in demonstrations may wish to wear facial covering for legitimate reasons, such as to avoid retribution. The restriction imposed by the PFCR is therefore not to be trivialised as a minor inhibition on mask-wearing during demonstrations but, depending on the context, can have a significant impact on the freedom of expression in peaceful public meetings and processions. The question here is whether the inroad made by the PFCR is proportionate. (a) Section 3(1)(a) of the PFCR 149. At the rolled‑up hearing, Mr Chan informed the court that the applicants in HCAL 2945 do not challenge the lawfulness of s 3(1)(a) of the PFCR under Ground 5A (without prejudice to the other grounds of judicial review concerning the constitutionality or lawfulness of the ERO and PFCR). It will be recalled that s 3(1)(a) prohibits any person from using any facial covering that is likely to prevent identification while that person is at an “unlawful assembly”, which expression is defined, by reference to s 18 of the POO to mean the following – “ When 3 or more persons, assembled together, conduct themselves in a disorderly, intimidating, insulting or provocative manner intended or likely to cause any person reasonably to fear that the persons so assembled will commit a breach of the peace, or will by such conduct provoke other persons to commit a breach of the peace.” 150. We consider Mr Chan’s concession to be rightly made. It is well established that there is a clear distinction drawn between peaceful assemblies and those which threaten life and physical integrity of the person or involve damage to property. In Kudrevičius, the Grand Chamber of the European Court of Human Rights states that Art 11 of the ECHR, which concerns the right to freedom of peaceful assembly, does not cover a demonstration where the organisers and participants have violent intentions, and that the guarantees of Art 11 do not apply to those gatherings where the organisers and participants have such intentions, incite violence or otherwise reject the foundations of a democratic society (§92). The Hong Kong Court of Appeal has also emphasised, in its recent judgment in Secretary for Justice v Wong Chi Fung [2018] 2 HKLRD 699, the importance of preserving public order as the foundation for the exercise and enjoyment of fundamental human rights by the ordinary citizens of Hong Kong. At §118 of the judgment of the Court of Appeal, Poon JA states as follows: “ Society is prone to descend into anarchy if public order is not preserved; once such a situation arises, the harm done to both the society and its citizens cannot be understated. For the society as a whole, preserving public order is indispensable to societal safety and public peace. Lawlessness in anarchic situations undermines social stability and hampers continuous development of a society. For the general public, preserving public order helps create a safe and stable social environment to enable individuals to exercise their rights (including human rights of which the freedom of assembly and expression is one), express their views and pursue their goals. In fact, the above‑mentioned rights themselves will be lost in situation of anarchy if public order is not preserved. That is exactly the rationale underlying Article 17 of the Hong Kong Bill of Rights in only safeguarding peaceful assembly: the legal protection of the right of assembly is effective only in a society where public order is preserved. Because preserving public order is so important to the society and the general public, the law must always remain vigilant to ensure that the public order in Hong Kong is not under threat. That does not mean that the law is only concerned about public order, or that it will ignore the rights and freedoms enjoyed by citizens in accordance with law, lest the society is likely to descend into a suppressed state, which would impede Hong Kong’s development and progress and deprive its citizens of their various freedoms and rights. The law must give consideration to both, and to strike a balance between the right of assembly and the need to preserve public order. That balance is embodied in the basic premise that assemblies must be held peacefully without disrupting or threatening to disrupt public order, or without involving any violence or threat to use violence.” 151. Such being the importance attached to the preservation of law and order in Hong Kong, we consider that the prohibition against the use of facial covering by any person who is at an unlawful assembly (which, by definition, would not be a peaceful assembly) imposed by s 3(1)(a) of the PFCR falls within the wide margin of discretion that the law affords to the Government to devise and implement measures to restrict unlawful and/or violent conduct. We also do not consider that the pursuit of the societal interest of law and order by the measure adopted under s 3(1)(a) would result in an unacceptably harsh burden on the individual. No person should take part in an unlawful assembly in the first place, it being a criminal offence to do so under s 18(1) of the POO, and the additional prohibition against the wearing of facial covering by a person at such assembly cannot be said to be unduly harsh on that person. In relation to an assembly which is initially lawful and peaceful but subsequently develops into an unlawful or violent one, we consider Mr Yu’s submission to have force that in such a case, the participants should distance themselves from the assembly as soon as possible rather than to continue to participate in it and seek refuge behind the facial covering. (b) Section 3(1)(b), (c) and (d) of the PFCR 152. Different considerations, however, apply to the prohibition against the use of facial covering in the situations under s 3(1)(b), (c) and (d) of the PFCR. When considering whether the restrictions of rights imposed by s 3(1)(b), (c) or (d) of the PFCR are proportionate to the legitimate aims sought to be achieved by the Government, the following features of the prohibition imposed by these sub‑paragraphs are of note. 153. First, s 3(1)(c) and (d) relate to public meetings and processions which may remain authorised and peaceful from beginning to end, and in which the participants behave lawfully and in good order throughout. The prohibition against the use of any facial covering imposed by s 3(1)(c) and (d) would directly interfere with these participants’ right of privacy and/or freedom of expression while taking part in perfectly lawful activities in the exercise of their right of peaceful assembly. 154. Second, s 3(1)(b) relates to “unauthorised” assemblies as explained in §§26 and 27 above. An assembly which is “unauthorised” may yet be entirely peaceful, without any violence being used or threatened by anyone participating in that assembly, eg a large scale public procession may become an unauthorised procession as a result of the failure by some participants to comply with a condition as regards the route of the procession imposed by the Commissioner under s 15(2) of the POO but the participants may continue to proceed with the procession in an entirely peaceful and orderly manner. 155. Third, the prohibition applies to any assembly, meeting or procession for whatever causes; it is not restricted to assemblies, meetings or processions arising from the now withdrawn Bill. Many assemblies, meetings or processions for different causes, such as LGBT, labour or migrant rights, take place in different parts of Hong Kong throughout the year, and traditionally these gatherings have been orderly and peaceful. It cannot be disputed that participants in such gatherings may have perfectly legitimate reasons for not wishing to be identified, or seen to be supporting such causes. Nevertheless, the effect of s 3(1)(b), (c) or (d) is to impose a near‑blanket prohibition against the wearing of facial covering by the participants, without any mechanism for a case‑by‑case evaluation or assessment of the risk of any specific gathering developing or turning into a violent one such as would make it desirable or necessary to impose the prohibition in relation to that gathering only. 156. Fourth, the prohibition applies to any person while he or she is “at” any unauthorised assembly, public meeting or public procession referred to in s 3(1)(b), (c) or (d). It is not clearly stated whether, to be caught by the prohibition, the person must be a participant in the relevant gathering, or whether it suffices for that person to be merely present at the gathering, eg a person who goes to the scene for the purpose of taking photographs, or giving first-aid to persons in need of help, or even a mere passer‑by who has stopped to observe the gathering. The wording of s 3(1) may be contrasted with (i) s 17(3) of the POO, which makes it an offence for any person who, without lawful authority or reasonable excuse, “knowingly takes or continues to take part in or forms or continues to form part of” any unauthorized assembly, and (ii) s 18(3) of the POO, which makes it an offence for any person who “takes part in” an unlawful assembly. It is uncertain whether the restrictions imposed by s 3(1)(b), (c) or (d) cover not only participants of the types of gathering referred to in those sub‑paragraphs, but also any person who is physically present (other than perhaps for a fleeting moment) at the gathering in question. 157. Fifth, the prohibition applies to facial covering of any type and used for whatever reason, including those worn for religious, cultural, aesthetic or other legitimate reasons, for example, to avoid reprisals or unpleasant consequences as a result of being seen to support some particular cause. There is no requirement that the facial covering is used by a person for the purpose of preventing identification, or is designed to have that effect. It suffices that the facial covering is “likely” to prevent identification. Although s 4(1) of the PFCR provides for a “lawful authority or reasonable excuse” defence to a charge under s 3(2), the scope of the defence is not clearly defined. A non‑exhaustive list of situations deemed to be reasonable excuses is given in s 4(3), namely, (i) a person engaging in a profession or employment and using the facial covering for the physical safety of that person while performing an act or activity connected with the profession or employment, (ii) a person using the facial covering for religious reasons, and (iii) a person using the facial covering for a pre‑existing medical or health reason, leaving other situations to be argued or determined on a case‑by‑case basis. 158. In SAS v France, mentioned in §131 above, the prohibition in question (namely, anyone wearing clothing that was designed to conceal the face in public places) was held by the European Court of Human Rights to engage the right to respect for private life under Art 8, and the right to freedom of thought, conscience and religion and to manifest one’s religion or belief under Art 9, of the ECHR. Furthermore, the prohibition was held to be disproportionate to the legitimate aim of prevention of danger for the safety of persons and property and the combat of identity fraud, because it was not shown that the prohibited conduct constituted a general threat to public safety. At §139 of the judgment of the European Court of Human Rights, it was stated that the blanket ban imposed could not be said to be necessary, in a democratic society, for public safety within the meaning of Arts 8 and 9 of the Convention when “the objective alluded to by the Government could be attained by a mere obligation to show their face and to identify themselves where a risk for the safety of persons and property has been established, or where particular circumstances entail a suspicion of identity fraud”. The ban was, however, saved as being proportionate to another legitimate aim sought to be pursued by the French Government (not relevant to our present case), namely, to ensure the observance of the minimum requirements of life in society, or the preservation of the conditions of “living together” as an element of the protection of the rights and freedoms of others (see §§140‑142 and 157 of the judgment). 159. In Yaker v France, Communication No 2747/2016 (17 July 2018), the United Nations Human Rights Committee had occasion to consider the same ban that was considered in SAS v France. The Human Rights Committee considered that the ban constituted an infringement of the complainant’s freedom of thought, conscience and religion protected by Art 18 of the ICCPR (corresponding to Art 15 of the Bill of Rights) which could not be justified. The reasoning for this conclusion was set out in §§8.7 and 8.8 of the decision, as follows: “ 8.7 With respect to protection of public order and safety, the State party contends that it must be possible to identify all individuals when necessary to avert threats to the security of persons or property and to combat identity fraud. The Committee recognizes the need for States, in certain contexts, to be able to require that individuals show their faces, which might entail one‑off obligations for individuals to reveal their faces in specific circumstances of a risk to public safety or order, or for identification purposes. The Committee observes, however, that the Act is not limited to such contexts, but comprehensively prohibits the wearing of certain face coverings in public at all times, and that the State party has failed to demonstrate how wearing the full‑face veil in itself represents a threat to public safety or order that would justify such an absolute ban. Nor has the State party provided any public safety justification or explanation for why covering the face for certain religious purposes — i.e., the niqab — is prohibited, while covering the face for numerous other purposes, including sporting, artistic, and other traditional and religious purposes, is allowed. The Committee further observes that the State party has not described any context, or provided any example, in which there was a specific and significant threat to public order and safety that would justify such a blanket ban on the full‑face veil. No such threats are described in the statement of purpose of Act No. 2010‑1192 or in the National Assembly resolution of 11 May 2010, which preceded the adoption of the Act. 8.8 Even if the State party could demonstrate the existence of a specific and significant threat to public safety and order in principle, it has failed to demonstrate that the prohibition contained in Act No. 2010‑1192 is proportionate to that objective, in view of its considerable impact on the author as a woman wearing the full‑face veil. Nor has it attempted to demonstrate that the ban was the least restrictive measure necessary to ensure the protection of the freedom of religion or belief.” 160. It can be seen that the Human Rights Committee reached its conclusion that the ban constituted a disproportionate interference with the complainant’s right under Art 18 of the ICCPR for essentially the same reasons relied upon by the European Court of Human Rights. For the sake of completeness, we should point out that, unlike the European Court of Human Rights, the Human Rights Committee further held that the ban could not be justified by the aim of preserving or promoting the values of the Republic and the requirements of “living together”.[48] 161. In Villeneuve v Montreal (City of), 2016 QCCS 2888, the relevant law (Art 3.2 of the Regulation for the Prevention of Disturbances of the Peace, Public Safety and Public Order) provided that: “No person who participates in or attends an assembly, parade or gathering on public property may cover their face without a reasonable motive, namely, using a scarf, hood or mask”.[49] The target of this provision was the phenomenon of masked persons engaging in acts of vandalism and violence during demonstrations on public roads.[50] Its objective was to prevent disturbance of public order, in particular to reduce the risks that legitimate parades or demonstrations taking place on public roadways would degenerate because of acts of vandalism and violence committed under the cover of anonymity, and also to promote the freedom of expression and assembly by aiming at preserving the peaceful nature of such parades or demonstrations, and the protection of peaceful participants.[51] The Superior Court of Quebec accepted that the above objectives were important, real and urgent on the evidence before it, and also that there was rational connection between those objectives and the measure adopted.[52] However, the restriction was held not to satisfy the “minimum infringement” requirement, or “what was necessary” to achieve the objectives because the measure adopted went far beyond the context of demonstrations that blocked public roads, but would prohibit face covering on the occasion of any assembly, parade or gathering on public property, including innocent activities (such as moving about in a group during Halloween or a group of people in a park throwing snowballs on a cold winter day).[53] Further, even if Art 3.2 were to be read down to restrict its application to only those demonstrations that blocked vehicular traffic, the provision would still not pass the minimum infringement test because (i) it permitted police officers to question a person exercising his freedom of political expression without requiring that there be reasonable grounds to believe that the person intended to disturb the peace or wished to disguise his or her identity from police officers, (ii) Art 3.2 had the effect of a near-absolute, if not absolute prohibition giving police wide latitude and leading to risks of abuse, and (iii) persons having the face covered for perfectly legitimate reasons, including that of expressing themselves without fear of reprisals, saw their freedoms of expression and assembly violated by Art 3.2, even in the case of peaceful assemblies.[54] In the result, Art 3.2 was declared null and void as being incompatible with the freedoms of expression and assembly protected by both the Canadian Charter and the Quebec Charter. 162. As rightly pointed out by Mr Yu, one cannot directly apply the judgment of the Quebec court in Villeneuve to the present case because the test of proportionality applicable in that jurisdiction appears to be different from that adopted in Hong Kong. At §§441‑442 of that judgment, the test of proportionality applicable in Quebec was explained as follows (footnotes omitted): “ [441] In essence, this is the test set forth in R v Oakes. In analysing proportionality, the courts must exercise some deference to the position of the legislature, proportionality not requiring perfection but only that the limits placed on the rights and liberties be reasonable. [442] Thus, in connection with the criterion of minimal infringement, for example, it is sufficient for the challenged measure to lie within a range of reasonable measures, and the courts will not conclude that it is disproportional solely because a better-suited alternative solution can be envisaged. However, a strict standard of justification should be applied when the freedom of political expression is infringed upon or when the State plays the role of unique adversary of the individual, principally in criminal matters.” [emphasis added] 163. This having been said, the reasoning of the Quebec court in coming to the conclusion that Art 3.2 was disproportionate to the pursuit of the legitimate aims of prevention of disturbance of public order as well as the promotion of the freedom of expression and assembly and the protection of peaceful participants seems to us to have force even for the application of the “reasonable necessity” test in step (3) of the proportionality analysis in Hong Kong. 164. As regards the effect of the PFCR, Mr Yu argues that although there is no guarantee that it will stop all acts of violence and vandalism, it is incorrect to assume that the PFCR will definitely be ignored; that at least some protesters will be discouraged from instigating or joining violent or riotous behaviour; and that it is believed that the PFCR can act as an effective deterrent against at least some students from wearing masks when joining a protest (lawful or unlawful) thereby substantially reducing the chance that they will be induced to break the law. On the other hand, Mr Chan points to the fact that there has been and continues to be massive defiance of the law since the coming into effect of the PFCR, and argues that the PFCR has not deterred violent protesters, but instead has caused further confusion and led to more outbursts of violent protests; he submits that in short the PFCR has not been shown to be effective in achieving its justification of reducing violence and crimes but has caused disproportionate adverse effect and restrictions on the rights of the freedom of the person, freedom of expression and freedom of peaceful assembly, and generated widespread uncertainty and resentment in a large number of law‑abiding citizens. We consider it to be self‑evidently correct that the court should not assume that a law will not be observed or will be flouted with impunity. Nevertheless, the evidence before us is far from clear that the PFCR has achieved to any substantial degree the intended aims of deterrence and elimination of the emboldening effect for those who may otherwise, with the advantage of facial covering, break the law, or facilitation of law enforcement, investigation and prosecution. 165. Mr Yu has impressed upon us the grave public danger that the recent violence used by some protesters has caused to Hong Kong and its inhabitants and the dire situation in which Hong Kong has currently found itself. These are matters of which the court can readily take judicial notice. However, we consider that even in these challenging times, and particularly in these challenging times, the court must continue to adhere to and decide cases strictly in accordance with established legal principles. 166. In our view, having regard to the reach of the impugned restrictions to perfectly lawful and peaceful public gatherings, the width of the restrictions affecting public gatherings for whatever causes, the lack of clarity as regards the application of the restrictions to persons present at the public gathering other than as participants, the breadth of the prohibition against the use of facial covering of any type and worn for whatever reasons, the absence of any mechanism for a case‑by‑case evaluation or assessment of the risk of violence or crimes such as would justify the application of the restrictions, the lack of robust evidence on the effectiveness of the measure, and lastly the importance that the law attaches to the freedom of expression, freedom of assembly, procession and demonstration, and the right to privacy, we do not consider the restrictions of rights imposed by s 3(1)(b), (c) and (d) to be proportionate to the legitimate aims sought to be achieved by the imposition of those restrictions. 167. Having reached this conclusion, it is not necessary for us to consider step (4) of the proportionality analysis. If it is necessary to do so, we would conclude, for essentially the same reasons, that s 3(1)(b), (d) and (d) have failed to strike a reasonable balance between the societal benefits sought to be promoted and the inroads made into the aforesaid protected rights having regard in particular to the burden placed on those who wish for wholly legitimate reasons to wear facial covering at peaceful assemblies. (5) Conclusion 168. For the above reasons, we consider that while the measures introduced by s 3(1) of the PFCR are rationally connected to the pursuit of legitimate societal aims, sub-paragraphs (b), (c) and (d) of s 3(1) go beyond what is reasonably necessary and therefore do not pass the proportionality test. J. Ground 5B — the section 5 proportionality ground (1) The rights engaged 169. We can deal Ground 5B more briefly. Art 28 of the Basic Law provides: “ The freedom of the person of Hong Kong residents shall be inviolable. No Hong Kong resident shall be subjected to arbitrary or unlawful arrest, detention or imprisonment. Arbitrary or unlawful search of the body of any resident or deprivation or restriction of the freedom of the person shall be prohibited…” 170. Similarly, Art 5(1) of the Bill of Rights provides: “ Everyone has the right to liberty and security of person. No one shall be subjected to arbitrary arrest or detention. No one shall be deprived of his liberty except on such grounds and in accordance with such procedure as are established by law.” 171. Section 5 of the PFCR has been set out in §33 above. Its effect is to empower a police officer to stop any person in any public place who is using a facial covering and to require that person to remove it so that his or her identity may be verified, if the officer reasonably believes the facial covering is likely to prevent identification. If the person fails to remove the facial covering pursuant to the police officer’s requirement, the officer may remove it and the person commits an offence punishable by a fine of $10,000 and imprisonment for 6 months. 172. It is clear that s 5 engages the freedom of the person or right to liberty protected by Art 28 of the Basic Law and Art 5 of the Bill of Rights. In DPP v Avery [2002] 1 Cr App R 31, the English Divisional Court had to consider the legality of s 60(4A) of the Criminal Justice and Public Order Act 1994, which conferred on any police constable in uniform the power to (a) “require any person to remove any item which the constable reasonably believes that person is wearing wholly or mainly for the purpose of concealing his identity” and (b) “seize any item which the constable reasonably believes any person intends to wear wholly or mainly for that purpose” in certain specified circumstances which we shall further discuss below. Newman J considered that subsection (4A) created a significant power to interfere with the liberty of the subject and was thus required to be justified.[55] 173. Some other protected rights are also relied upon by the applicants to challenge the legality of s 5 of the PFCR, including the right to privacy (Art 14 of the Bill of Rights), the freedom of expression (Art 27 of the Basic Law and Art 16 of the Bill of Rights), and the freedom of movement (Art 31 of the Basic Law and Art 8 of the Bill of Rights). None of these rights are absolute. Since the analysis for considering the proportionality of the restriction imposed by s 5 on (i) the freedom of the person or right to liberty protected by Art 28 of the Basic Law and Art 5 of the Bill of Rights on the one hand, and (ii) the other rights relied upon on the other hand are effectively the same, it is not necessary to separately consider those other rights for the purpose of the present discussion. (2) Step (1): legitimate aim 174. According to the Government, the aim of s 5 of the PFCR is to assist in law enforcement, investigation and prosecution by enabling police officers to verify the identity of all masked individuals not only during assemblies or processions but also in public places in the prevailing circumstances of public danger in Hong Kong. The aim of law enforcement, investigation and prosecution of offenders is undoubtedly a legitimate aim in itself. 175. Mr Pun argues, however, that this is not what the Government stated to be the aim of s 5, and relies on paragraph 16 of the LegCo Brief (File Reference: SCBR 3/3285/57) titled “Emergency Regulations Ordinance (Cap 241) Prohibition on Face Covering Regulation” dated October 2019 as “exhaustively” setting out the Government’s case on justification.[56] Paragraph 16 states as follows: “ As regards the proposal on the Police’s power to remove facial covering in a public place, it is only reasonable that a police officer should be empowered to require a person to remove the person’s facial covering in order to verify the person’s identity, as a police officer is authorized under various laws to demand proof of identity. The person in question will only be stopped and asked to remove the facial covering for a short period of time, and may wear the facial covering after the officer has completed the verification process. Such minor interference with the person’s right to privacy guaranteed by Article 14 of the BOR is justifiable. It is also a proportionate measure to make non‑compliance with the requirement an offence, given that refusal to comply with the requirement in such circumstances may, under existing law, amount to the offence of resisting or obstructing a police officer in the due execution of the officer’s duty.” 176. Based on this paragraph, Mr Pun argues that the Government’s aim of enacting s 5 is merely to “verify a person’s identity”, which he submits cannot by itself be a legitimate aim.[57] In our view, this is a distorted way of reading the LegCo Brief. The Government’s purposes or aims of the measures adopted by the PFCR are spelt out in paragraph 3 of the LegCo Brief, under the heading “Justifications”, as follows: “ Due to the widespread and imminent public danger posed by the violent and illegal acts of masked protesters, there is an urgent need to consider introducing legislation to prohibit face covering to enable the Police to investigate into such acts and to serve as a deterrent against such behaviour. To restore public order, prohibition on facial covering in public assemblies, lawful and unlawful, would be necessary as it would effectively reduce act of violence and facilitate police investigation and administration of justice. The prohibition would be essential in public interest in restoring public peace, and is rationally connected to protecting public order and public safety.” 177. Paragraph 4 of the LegCo Brief goes on to explain why further powers are required by the Police to handle the illegal and violent acts of radical protesters: “ We have critically considered the existing powers of the Police and relevant laws. We are of the view that legislation has to be enacted urgently to enable the Police to handle further illegal and violent acts of radical protesters more effectively so as to restore law and order, and to prevent serious public disorder, as well as to apprehend the offenders and bring them to justice. The proposal has taken into account the regulatory tools required to protect public safety and order having regard to the practical experience in handling protests in the past few months, with due regard for fundamental rights even in times of public danger.” 178. It is clear that the Government’s aims or purposes in adopting the measures under the PFCR are all to do with law enforcement, investigation and prosecution in the light of the prevailing circumstances in Hong Kong. Section 5 of the PFCR forms part of the measures brought in by the PFCR with a view to achieving such aims or purposes. Paragraph 16 of the LegCo Brief goes to explaining why the Government considers that the obligation imposed on a person in a public place, by requiring that person to remove his or her facial covering to enable a police officer to verify his or her identity, constitutes only a minimal interference of that person’s right and is a proportionate measure. 179. In any event, as submitted by Mr Yu, correctly in our view, the question of whether a measure which restricts a fundamental right is lawful or not lawful is a question of law for the court. The relevant question is not whether the decision‑maker had properly considered whether the measures served a legitimate aim and was rationally connected to that aim and did not constitute a disproportionate interference. In short, the question is not whether the decision‑maker thought the decision was lawful, but whether it is in fact lawful (see Belfast City Council v Miss Behavin’ Ltd [2007] 1 WLR 1420, at §§88‑90). (3) Step (2): rational connection 180. As we understand it, the Government’s case on rational connection is that, at this time of public danger when the use of face covering is prevalent among vandals and violent protesters, for the purposes of law enforcement, investigation and prosecution, police officers have a heightened need to see a person’s face in order to verify his or her identity not only during assemblies or processions but also in public places. 181. Under the existing law, police officers already have various powers to demand verification of a person’s identity, in particular: (1) Section 54(1)(a) of the Police Force Ordinance (Cap 232) empowers a police officer, if he finds any person in any street or other public place, or on board any vessel, or in any conveyance, at any hour of the day or night, who acts in a suspicious manner, to stop the person for the purpose of demanding that he produces proof of his identity for inspection; (2) Section 49 of the POO empowers a police officer, if he reasonably believes that it is necessary for the purpose of preventing, detecting or investigating any offence for which the sentence is fixed by law or for which a person may (on a first conviction for that offence) be sentenced to imprisonment, to require any person to produce proof of his identity for inspection; and (3) Section 17C(2) of the Immigration Ordinance (Cap 115) empowers (inter alios) any police officer who is in uniform or who produces, if required to do so, documentary identification officially issued to him as proof of his appointment as a police officer, to demand that any person, who is required under subsection (1) to have with him proof of his identity, produce the same for inspection. Although this sub‑section does not stipulate any condition for the exercise of the power by a police officer to demand proof of identity, it would appear, from the context of the provision, that the power could only be exercised for purposes connected with immigration control under the Immigration Ordinance. 182. It is unclear whether these powers are wide enough to authorise a police officer to require a person to remove any facial covering for the purpose of inspection or verification of that person’s identity. The applicants in HCAL 2945 contend that, by necessary implication, police officers already have such power. Be that as it may, the evidence suggests that, without a specific power authorising the police to require persons to remove their facial covering, police officers would face practical difficulties, including verbal challenges and insults, when exercising the power of checking identity under the existing law.[58] 183. Mr Yu submits that the measure adopted by s 5 of the PFCR is rationally connected to the legitimate aim of law enforcement, investigation and prosecution of violent protesters in that: (1) The aim and effectiveness of the PFCR should be considered as a whole. In view of the fact that many radical protesters committing criminal acts wear face covering when attending both lawful and unlawful meetings and processions, as well as in public places before joining such meetings or processions, if the police has power to require a person in a public place to remove his/her facial covering to verify that person’s identity, it would have a deterrent effect against that person subsequently committing violent or other criminal acts, knowing that his or her identity may already have been revealed or exposed. (2) The information obtained from verification of identity of persons wearing face covering in public places would assist the police in identifying the persons who committed criminal acts during lawful or unlawful meetings or processions wearing the same or similar face covering. (3) The measure under s 5 of the PFCR would help “water down” the emboldening effect of face covering. Many protesters wear face covering at all times in public, including in public places prior to and after joining public meetings or processions. If the protesters know that they can be asked to remove their face covering in public places, this would minimise the sense of anonymity created by the wearing of face covering at all times in public. (4) Lastly, in situations where it is not clear‑cut if a person is attending a public meeting or procession, the police can at least exercise the power under s 5 of the PFCR to identify the person, as a precursor to exercising further powers where appropriate or required in the circumstances. This gives more flexibility and options to the police in enforcing the PFCR as a whole.[59] 184. We consider Mr Yu’s submission to have force, and accept that the measure adopted under s 5 of the PFCR is rationally connected to the legitimate aim identified by the Government. (4) Steps (3) and (4): no more than reasonably necessary and reasonable balance 185. The following features of s 5 of PFCR are notable: (1) It applies to any public place, which, as stated above, broadly means any place to which the public or any section of the public are entitled or permitted to have access, whether on payment or otherwise — not necessarily one where a public meeting or public procession is taking place or about to take place or even a neighbouring area. Nor is there any provision for a senior police officer to designate particular places where the section applies, based on actual circumstances.[60] (2) The power may be exercised by any police officer, not only by or with the authorisation of an officer of or above a certain rank. (3) It applies to facial covering of any type. (4) The only condition for the exercise of the power is that the officer reasonably believes that the facial covering is likely to prevent identification. There is no requirement that the person is using the facial covering for the purpose or with the intention of preventing identification. There is equally no requirement for the officer to believe that it is necessary to exercise the power for the purpose of preventing, detecting or investigating any offence. (5) It applies to any person who is using a facial covering. It does not require that the officer should have any suspicion or ground for suspicion that the person has committed or is about to commit an offence or is acting in a suspicious or otherwise objectionable manner. 186. A similar, but much more limited, power exists in the UK in the form of s 60(4A) of the Criminal Justice and Public Order Act 1994. Under that subsection, a police constable may require any person to remove a face covering which the constable reasonably believes is worn by that person wholly or mainly for the purpose of concealing his identity. However, such power can only be exercised where a police officer of or above the rank of inspector reasonably believes (i) that incidents involving serious violence may take place in any locality in his police area, and that it is expedient to give an authorisation under this section to prevent their occurrence, or (ii) that persons are carrying dangerous instruments or offensive weapons in any locality in his police area without good reason, and has given an authorisation that the powers conferred by the subsection are to be exercisable at any place within that locality for a specified period not exceeding 24 hours.[61] In DPP v Avery, supra, Newman J held that the interference with the liberty of the subject sanctioned by subsection (4A) was justified for the following reasons: “ (I) The powers conferred by section 60, of the 1994 Act arise only in anticipation of violence and after deliberation and a decision of a senior police officer. (II) It is well recognised that the wearing of masks to conceal identity in the course of violent demonstrations serves two aims for an offender and could serve to defeat two legitimate objectives of the enforcement of the law: (i) to impede arrest and to facilitate escape from the commission of an offence in the course of a demonstration; (ii) to impede proper measures by way of control in connection with future demonstrations. The control of the movement of troublemakers and offenders as a preventative measure is a legitimate objective. (III) The common law has not upheld an unconditional right to be informed of the reason for an interference with liberty. It is not a rote or incantation which is required but a reason, which gives rise to an opportunity to decline the request… A request to a person to remove a mask is, for all material purposes, self explanatory. (a) It is a request to the person to reveal his physical identity. Nothing material would be gained by the person being told by a constable that he believed the person was concealing his identity. (b) The existence of a requirement for a subjective belief on the part of a constable is a legislative restraint upon the power being exercised oppressively or arbitrarily.”[62] 187. It seems to us to be clear that the limitation of the circumstances and period in which the power to require removal of a facial covering could be exercised by a police constable was an important consideration which led the court in that case to come to the conclusion that the interference with the liberty of the person was justified. 188. As earlier noted, in SAS v France, the European Court of Human Rights considered that the legitimate aim of prevention of danger for the safety of persons and property and the combat of identity fraud “could be attained by a mere obligation to show their face and to identify themselves where a risk for the safety of persons and property has been established, and where particular circumstances entail a suspicion of identity fraud”.[63] Similarly, in Yaker v France (discussed in §§159‑160 above), the Human Rights Committee also recognized “the need for States, in certain contexts, to be able to require that individuals show their faces, which might entail one-off obligations for individuals to reveal their faces in specific circumstances of a risk to public safety or order, or for identification purposes”.[64] These cases support the view that, to be proportionate, the power to require an individual to show his or her face should not be exercisable generally, but only where the circumstances pertaining to any specific case give rise to a risk of public safety or order. 189. The remarkable width of the measure under s 5 of the PFCR has already been explained in §185 above. There is practically no limit on the circumstances in which the power under that section can be exercised by a police officer, save the requirement that (i) the person is in a public place, and (ii) the facial covering used by that person is reasonably believed by the police officer to be likely to prevent identification. The power can be exercised irrespective of whether there is any public meeting or procession taking place in the vicinity, and regardless of whether there is any risk of outbreak of violence or other criminal acts, at the place where the person is found, or in the neighbourhood, or indeed anywhere else in Hong Kong. The power may, on its face, be used by a police officer for random stoppage of anyone found wearing a facial covering in any public place. We consider it to be clear that the measure adopted by s 5 of the PFCR exceeds what is reasonably necessary to achieve the aim of law enforcement, investigation and prosecution of violent protesters even in the prevailing turbulent circumstances in Hong Kong, and that it fails to strike a reasonable balance between the societal benefits promoted and the inroads made into the protected rights. 190. We should mention that the applicant in HCAL 2949 also contends that s 5 of the PFCR infringes the “prescribed by law” requirement in that it authorises a police officer to interfere with fundamental rights arbitrarily. In our view, this argument is in truth part of the argument on the proportionality of s 5. It adds nothing and requires no separate treatment. (5) Conclusion 191. For the above reasons, while there are legitimate societal objects to the pursuit of which the measure introduced by s 5 is rationally connected, s 5 represents a more serious inroad into protected rights than is reasonably necessary, and therefore fails the proportionality test. 192. We have considered whether s 3(1)(b), (c) and (d) and/or and s 5 could be saved by a process of “reading in” or “reading down” of the relevant provisions so as to make them proportionate to the interferences with the protected rights. Mr Yu has not made any relevant submission on this point. Having given this matter careful thought, we consider that any such “reading in” or “reading down” would require a substantial re‑writing of the legislation in a manner which would effectively be a fresh legislative exercise involving fundamental changes to the substance of the provisions, and is a task which ought to be undertaken by the legislature itself rather than by the court. We should also make it clear that it is not our judgment that “anti‑mask” law is generally objectionable or unconstitutional. Its validity must, however, depend on the details of the legislation and the particular societal aims sought to be pursued by the measure being brought in through the legislation. Generalisation is neither possible, nor appropriate, in this matter. K. Conclusion and Disposition 193. In summary, our conclusions on each of the grounds of judicial review may be briefly stated as follows: (1) The ERO, insofar as it empowers the CEIC to make regulations on any occasion of public danger, is incompatible with the Basic Law, having regard in particular to Arts 2, 8, 17(2), 18, 48, 56, 62(5), 66 and 73(1) thereof. We leave open the question of the constitutionality of the ERO insofar as it relates to any occasion of emergency. (2) The ERO was not impliedly repealed by s 5 of the HKBORO. Insofar as it is invoked in situations not falling within the kind of public emergency referred to in the HKBORO, the Bill of Rights is not suspended and the measures adopted will have to comply with it. (3) The ERO does not in itself fall foul of the “prescribed by law” requirement (ie the principle of legal certainty). Where regulations and measures are adopted under the ERO that curtail fundamental rights, the entire relevant body of law including the regulations and measures have to be taken together to see whether they meet the requirement of sufficient accessibility and certainty. (4) It is not necessary to deal with the argument based on the principle of legality. (5) The provisions in s 3(1)(a), (b), (c) and (d) of the PFCR are rationally connected to legitimate societal aims that the respondents intend by those measures to pursue but the restrictions that sub‑paragraphs (b), (c) and (d) impose on fundamental rights go further than is reasonably necessary for the furtherance of those objects. (6) The measure introduced by s 5 of the PFCR is rationally connected to the legitimate societal aims pursued but the restrictions it imposes on fundamental rights also go further than is reasonably necessary for the furtherance of those objects. 194. We have not heard submissions on the question of relief and the respondents have asked for an opportunity to address the court should that question arise. In the light of our conclusions above, we shall convene a hearing on the appropriate relief and costs. 195. As to the procedural applications, the applicants in both proceedings have applied to amend their Form 86. The amendments are not opposed. We give leave for the proposed amendments to be made. We also grant leave for the applicants in HCAL 2945 to file the 2nd affidavit of Kwok Wing Hang dated 23 October 2019, for the applicant in HCAL 2949 to file the 2nd affirmation of Leung Kwok Hung dated 29 October 2019, and for the respondents to file the affidavit of Dr Tsui Pui Wang Ephraem dated 23 October 2019. Ms Gladys Li SC, Mr Johannes Chan SC (Hon), Mr Earl Deng, Mr Jeffrey Tam, Mr Geoffrey Yeung and Ms Allison Wong, instructed by Ho Tse Wai & Partners, for the 1st to 24th Applicants in HCAL 2945/2019 Mr Hectar Pun SC, Mr Lee Siu Him and Mr Anson Wong Yu Yat, instructed by JCC Cheung & Co, assigned by the Director of Legal Aid Department, for the Applicant in HCAL 2949/2019 Mr Benjamin Yu SC, Mr Jenkin Suen SC, Mr Jimmy Ma and Mr Mike Lui, instructed by the Department of Justice, for the Putative Respondents in both HCAL 2945/2019 and HCAL 2949/2019 [1] LN 119 of 2019. [2] Another putative respondent in the original Form 86, the Commissioner of Police, was removed by amendment. [3] For an explanation of the nature of a “rolled-up hearing”, see the Court of Appeal’s judgment in Chee Fei Ming & Another v Director of Food and Environmental Hygiene & Another [2016] 3 HKLRD 412 at paras 5‑7. [4] Corresponding to Ground 3: the “prescribed by law ground”, and Ground 5B: the “section 5 proportionality ground”, as described below. [5] Namely, HCAL 2929/2019, HCAL 2930/2019, HCAL 2942/2019 and HCAL 2993/2019. HCAL 2993/2019 has since been withdrawn with leave of the court. [6] In the Amended Form 86 in HCAL 2945 and in the Amended Form 86 in HCAL 2949. [7] The history of the ERO and more generally of emergency powers in Hong Kong has been described in two academic articles: Norman Miners, The Use and Abuse of Emergency Powers by the Hong Kong Government (1996) 26 HKLJ 47 and Max W L Wong, Social Control and Political Order — Decolonisation and the Use of Emergency Regulations in Hong Kong (2011) 41 HKLJ 449. [8] By the Emergency Regulations (Amendment) Ordinance 1949, (Ord No 8 of 1949). [9] By the Emergency Regulations (Amendment) (No 2) Ordinance 1949, (Ord No 40 of 1949). [10] By the Crimes (Amendment) Ordinance 1993, (Ord No 24 of 1993), section 24. [11] Section 3(3), which required the LegCo’s approval for the death penalty, was also repealed at the same time. [12] Adaptation of Laws (No 32) Ordinance 1999, (Ord No 71 of 1999), Schedule 8. [13] LN 251‑255 of 1995. [14] Cap 245 means the Public Order Ordinance (Cap 245). [15] This means $25,000: see s 113B of and Schedule 8 to the Criminal Procedure Ordinance (Cap 221). [16] “Public meeting” means any meeting held or to be held in a public place. [17] “Public procession” means any procession in, or from a public place. [18] “Public gathering” means a public meeting, a public procession and any other meeting, gathering or assembly of 10 or more persons in any public place. [19] This means $10,000: see s 113B of and Schedule 8 to the Criminal Procedure Ordinance (Cap 221). [20] Art 2 provides: “The National People’s Congress authorizes the Hong Kong Special Administrative Region to exercise a high degree of autonomy and enjoy executive, legislative and independent judicial power, including that of final adjudication, in accordance with the provisions of this Law.” [21] Art 8 provides: “The laws previously in force in Hong Kong, that is, the common law, rules of equity, ordinances, subordinate legislation and customary law shall be maintained, except for any that contravene this Law, and subject to any amendment by the legislature of the Hong Kong Special Administrative Region.” [22] See s 22(q) of the House Rules of the LegCo. [23] Letters Patent, Art VII. [24] Adaptation of Laws (Interpretative Provisions) Ordinance (26 of 1998), s 4. [25] The previous version provided: “ ‘subsidiary legislation’ (附屬法例) and ‘regulations’ (規例) mean any proclamation, rule, regulation, order, resolution, notice, rule of court, by-law or other instrument made under or by virtue of any Ordinance and having legislative effect.” [26] AV Dicey, The Law of the Constitution (1885), pp 39-40. [27] No 31 of 1966. [28] Except on one point which is not material for present purposes; see p 113. [29] Ordinance No 26 of 1937. [30] Ordinance No 31 of 1911. [31] Namely, s 67C of the Criminal Procedure Ordinance (Cap 221). [32] An appeal from the decision of Hartmann J to the Court of Appeal was dismissed but without consideration of the merits: see CACV 377/2002, 2 July 2003. [33] Blackstone’s Commentaries on the Laws of England, vol I, p 157. [34] Art XIII and XXI of the Royal Instructions 1917. [35] Art 4 of the ICCPR provides: “1. In time of public emergency which threatens the life of the nation and the existence of which is officially proclaimed, the States Parties to the present Covenant may take measures derogating from their obligations under the present Covenant to the extent strictly required by the exigencies of the situation, provided that such measures are not inconsistent with their other obligations under international law and do not involve discrimination solely on the ground of race, colour, sex, language, religion or social origin. 2. No derogation from articles 6, 7, 8 (paragraphs 1 and 2), 11, 15, 16 and 18 may be made under this provision. 3. Any State Party to the present Covenant availing itself of the right of derogation shall immediately inform the other States Parties to the present Covenant, through the intermediary of the Secretary-General of the United Nations, of the provisions from which it has derogated and of the reasons by which it was actuated. A further communication shall be made, through the same intermediary, on the date on which it terminates such derogation.” [36] The doctrine of automatic repeal is not in dispute: see R v Sin Yau Ming (CACC 289/1990, 30 September 1991). [37] Pursuant to the Decision of the NPCSC on 23 February 1997, which came into effect on 1 July 1997. [38] This provides: “Where any Ordinance repealing in whole or in part any former Ordinance is itself repealed, such last repeal shall not revive the Ordinance or provision previously repealed, unless provision is made to that effect”. [39] The case went on appeal to the Court of Appeal (CACV 73 & 87/2006, 10 May 2006) and Court of Final Appeal (2006) 9 HKCFAR 441, but there was no appeal by the respondents from the declaration that s 33 of the Telecommunications Ordinance was unconstitutional in so far as it authorised access to or disclosure of the contents of any message or any class of messages. [40] Art 17 of the Bill of Rights provides: “The right of peaceful assembly shall be recognized. No restrictions may be placed on the exercise of this right other than those imposed in conformity with the law and which are necessary in a democratic society in the interests of national security or public safety, public order (ordre public), the protection of public health or morals or the protection of the rights and freedoms of others”. [41] Art 27 of the Basic Law provides: “Hong Kong residents shall have freedom of speech, of the press and of publication; freedom of association, of assembly, of procession and of demonstration; and the right and freedom to form and join trade unions, and to strike”. [42] Art 16(2) of the Bill of Rights provides: “Everyone shall have the right to freedom of expression; this right shall include freedom to seek, receive and impart information and ideas of all kinds, regardless of frontiers, either orally, in writing or in print, in the form of art, or through any other media of his choice”. [43] Art 14 of the Bill of Rights provides: “No one shall be subjected to arbitrary or unlawful interference with his privacy, family, home or correspondence, nor to unlawful attacks on his honour and reputation”. [44] Paras 107 and 108. [45] Para 115. [46] In para 9 of the Affirmation of Cheung Tin Lok, the “black bloc” tactic is explained to mean protestors appearing in groups and wearing black clothing with little or no distinguishing feature and covering the whole or a substantial part of their faces with sunglasses, goggles, masks, respirators, etc, which not only prevent identification of those who commit criminal acts but also shield them from tear gas and pepper spray deployed by the Police. [47] Paras 6.12 to 6.15 of the Skeleton Submissions dated 24 October 2019. [48] See §§8.9-8.12 of the decision. [49] See §263 of the judgment of Superior Court of the Quebec. The original judgment is in French; the references here are to an unofficial translation provided by the applicants in HCAL 2945. [50] See §§481-483 of the judgment. [51] See §§484-485 of the judgment. [52] See §§487-489 of the judgment. [53] See §§490-491 of the judgment. [54] See §§493-495 of the judgment. [55] See §18 of the judgment of the Divisional Court. [56] See §55 of the Skeleton Argument of the applicant in HCAL 2949 dated 24 October 2019. [57] See §56 of the Skeleton Argument of the applicant in HCAL 2949 dated 24 October 2019. [58] See §26 of Cheung Tin Lok’s Affirmation dated 18 October 2019. [59] See §143(4) of the Skeleton Submissions for the Putative Respondents dated 28 October 2019. [60] Cf s 60 of the Criminal Justice and Public Order Act 1994 (as amended), considered in Director of Public Prosecutions v Avery [2002] 1 Cr App R 31. [61] See s 60(1) of the Criminal Justice and Public Order Act 1994. [62] See §18 of the judgment of the Divisional Court in that case. [63] See §139 of the judgment. [64] See §8.7 of the decision. 
Chief Justice Ma: A. INTRODUCTION 1. This appeal involves principally an analysis of the appropriate test to be used when dealing with alleged discrimination under the Sex Discrimination Ordinance[1] and of its application to the facts of the case. The guarantee of equality is found in the Basic Law.[2] Article 22 of the Hong Kong Bill of Rights[3] states:- “Equality before and equal protection of law All persons are equal before the law and are entitled without any discrimination to the equal protection of the law. In this respect, the law shall prohibit any discrimination and guarantee to all persons equal and effective protection against discrimination on any ground such as race, colour, sex, language, religion, political or other opinion, national or social origin, property, birth or other status.” 2. The present case is concerned with custodial discipline in penal establishments. Section 9 of the Hong Kong Bill of Rights Ordinance states:- “Armed forces and persons detained in penal establishments Members of and persons serving with the armed forces of the government responsible for the foreign affairs of Hong Kong and persons lawfully detained in penal establishments of whatever character are subject to such restrictions as may from time to time be authorized by law for the preservation of service and custodial discipline.” 3. A constitutionality issue also arises in this appeal under Article 25 of the Basic Law, but as we shall see, this is of less importance than the issue of discrimination under the SDO. A.1 The facts 4. The facts are straightforward. In March 2012, the appellant, a politician and activist who is widely known as “Long Hair”, was convicted in the Magistrates’ Court of charges of criminal damage (2 charges) and disorderly behaviour (2 charges) and sentenced concurrently to a total term of imprisonment of 2 months. On appeal, in June 2014, the conviction for one of the charges of criminal damage was quashed and the totality of his sentence was reduced to 4 weeks. At Lai Chi Kok Reception Centre, he was required to have his hair (then about 80 cm long) cut pursuant to Standing Order 41-05 (“SO 41‑05”). This standing order,[4] which appears to be the only provision dealing with the length of hair of prisoners, applies to both male and female prisoners:- “1. The hair of all male convicted prisoners will be kept cut sufficiently close, but not close clipped, for the purposes of health and cleanliness unless the prisoner himself requests it. 為保健康及清潔，所有已男子定罪囚犯的頭髮須盡量剪短，但不用剪陸軍裝，除非囚犯本身要求如此。 2. Upon request, female prisoners will have their hair cut especially before discharge or production in court. Except as recommended by MO, a female prisoner’s hair shall not be cut shorter than the style on admission without her consent. 若女子囚犯申請剪髮，須為其作出安排，特別是在獲釋前或到法庭應訊前。未經囚犯同意，不可把其頭髮剪至較進入院所時的髮型更短，但如醫生建議這樣做，則屬例外。” 5. It was the decision to require the appellant to cut his hair (the Decision) that formed the basis for his application for leave to apply for Judicial Review. Essentially, the appellant’s complaint is that he was discriminated against on account of his sex: while the hair of male convicted prisoners like him had to be kept sufficiently close, by contrast female prisoners had a freer choice and, except as recommended by a Medical Officer, their hair could not be cut shorter than the style on admission to prison without their consent. In the language of the relevant provision of the Ordinance with which this appeal is concerned, the appellant says he has been treated less favourably than female prisoners. A.2 The proceedings below 6. Before Au J, the Judicial Review challenge by the appellant was based on 4 grounds, namely, discrimination under the SDO, breach of the equality provisions under Article 25 of the Basic Law, Wednesbury unreasonableness and breach of the dignity provision in Article 6(1) of the Bill of Rights. The judge held[5] in favour of the appellant on the first 2 grounds, did not think it was necessary to deal with Wednesbury unreasonableness and rejected the 4th ground. Declarations were made that the Decision constituted direct sex discrimination under the SDO and also violated Article 25 of the Basic Law. An order was also made quashing the Decision. 7. On the respondent’s appeal, the Court of Appeal[6] allowed the appeal and set aside Au J’s order. There being no Respondent’s Notice, the Court of Appeal only dealt with the sex discrimination and Article 25 issues. Like the CFI Judgment, the Court of Appeal concentrated mainly on the sex discrimination issue. These are the same issues before us. A.3 The certified question 8. Leave to appeal to the Court of Final Appeal was refused by the Court of Appeal,[7] but on 13 August 2019, the Appeal Committee[8] granted leave to appeal to the appellant on the following question of law:- “Whether the Standing Order 41-05 issued by the Commissioner of Correctional Services requiring all male prisoners but not female prisoners to have their hair cut ‘sufficiently close’ (‘盡量剪短’): (1) constitutes direct discrimination under section 5(1)(a) of the Sex Discrimination Ordinance, Cap 480 (‘SDO’) and is therefore unlawful under section 38 of the SDO; and/or (2) is inconsistent with Article 25 of the Basic Law of the HKSAR and is therefore unconstitutional?” 9. It is convenient to deal first with the issue of sex discrimination under the SDO before dealing with the constitutionality issue under Article 25 of the Basic Law. I start by discussing the approach in discrimination cases under the SDO before applying the relevant principles to the facts in the present case. I will then deal with the Article 25 issue. B. THE APPROACH IN DISCRIMINATION CASES UNDER THE SDO B.1 Relevant statutory provisions 10. The SDO is fairly comprehensive in dealing with many aspects of sex discrimination including sexual harassment. The Ordinance also makes provision for the establishment and operation of the Equal Opportunities Commission. In relation to sex discrimination, there are set out detailed provisions relevant to employment (Part 3) and to other miscellaneous fields. The basic concept of sex discrimination is defined in Part 2 of the Ordinance in s 5(1):- “Sex discrimination against women (1) A person discriminates against a woman in any circumstances relevant for the purposes of any provision of this Ordinance if— (a) on the ground of her sex he treats her less favourably than he treats or would treat a man; or (b) he applies to her a requirement or condition which he applies or would apply equally to a man but— (i) which is such that the proportion of women who can comply with it is considerably smaller than the proportion of men who can comply with it; (ii) which he cannot show to be justifiable irrespective of the sex of the person to whom it is applied; and (iii) which is to her detriment because she cannot comply with it.” Although this provision refers to discrimination against women, it equally applies to discrimination against men: see s 6(1). 11. Section 5(1)(a) of the SDO deals with what is known as direct discrimination. Section 5(1)(b) (in particular sub‑paragraphs (i) and (iii)) deal with indirect discrimination. These two types of discrimination were discussed by this Court in QT v Director of Immigration.[9] Further, in R (E) v Governing Body of JFS,[10] the difference was put in the following way:- “56. The basic difference between direct and indirect discrimination is plain: see Mummery LJ in R (Elias) v Secretary of State for Defence [2006] 1 WLR 3213, para 119. The rule against direct discrimination aims to achieve formal equality of treatment: there must be no less favourable treatment between otherwise similarly situated people on grounds of colour, race, nationality or ethnic or national origins. Indirect discrimination looks beyond formal equality towards a more substantive equality of results: criteria which appear neutral on their face may have a disproportionately adverse impact upon people of a particular colour, race, nationality or ethnic or national origins.” We are in this appeal only concerned with direct discrimination.[11] 12. The discrimination provisions are relevant to the facts of the present case by reason of s 38(1) of the Ordinance which states that it is unlawful for the Government to discriminate in the performance of its functions or in the exercise of its powers. B.2 The approach 13. In the context of the present case, what must be demonstrated by the appellant is that on the ground of his sex he has been treated by the respondent less favourably than the respondent has treated or would treat a female prisoner in similar circumstances. I will refer to a person in the position of the appellant as “the complainant” and the person who is alleged to have discriminated simply as “the discriminator”. 14. In Secretary for Justice v Chan Wah,[12] a case involving alleged sex discrimination in the context of elections for village representatives, this Court adopted the “but for” test articulated by Lord Goff of Chieveley in R v Birmingham City Council, ex p Equal Opportunities Commission:[13] “There is discrimination under the statute[14] if there is less favourable treatment on the ground of sex, in other words if the relevant girl or girls would have received the same treatment as the boys but for their sex”. The test was reaffirmed in James v Eastleigh Borough Council and stated to be an objective, not subjective, one.[15] 15. This test, simply stated, however requires a little further elaboration in terms of setting out the necessary steps before a finding of sex discrimination can be established. Here, the 4 step approach contained in the speech of Baroness Hale of Richmond in R (European Roma Rights) v Prague Immigration Officer[16] is helpful. Notwithstanding the fact that was a case involving discrimination under the Race Relations Act 1976, the approach is logical and, translated to the context of sex discrimination, can be stated as follows:- (1) There must be a difference in treatment between one person, the complainant (in our case male prisoners; more specifically the appellant) and another person, real or hypothetical, from a different sex group, the compared person (in our case, female prisoners). (2) The relevant circumstances between the complainant and the compared person are the same or at least not materially different. (3) It must then be shown that the treatment given to the complainant is less favourable than that given to the compared person. (4) The difference in treatment is on the basis of sex. Although there may be some overlap in these steps, it is useful to break down the analysis in this way. B.3 Difference in treatment 16. As to para 15(1), this can be identified in most cases, by referring to the essential complaint made by the complainant. Thus, in the present case, the appellant’s complaint is that he was treated differently to female prisoners, not being given the choice they had and accordingly being required to have his hair cut. B.4 Comparing like with like 17. In relation to the requirement in para 15(2), what is needed to be shown is that when embarking on the comparison exercise, it is important to compare like with like and to view the matter in proper context. It is only when one is comparing like with like that one can then move to the next critical inquiry of asking whether there is less favourable treatment that has been accorded by the discriminator to the complainant. This reflects s 10 of the SDO that when making comparisons for the purpose of the s 5(1) exercise, “the relevant circumstances in the one case are the same, or not materially different, in the other.” In a nutshell, “there must be no less favourable treatment between otherwise similarly situated people on grounds of”[17] sex. In the present case, the appellant is comparing himself with female prisoners. B.5 Less favourable treatment 18. This is the requirement in para 15(3) above. It is expressly referred to in s 5(1)(a) of the SDO and will almost invariably be the focal point in discrimination cases. The mere fact there is a difference in treatment between the complainant and the compared person will not of itself amount to unlawful discrimination; there must be shown less favourable treatment given to the complainant. Less favourable treatment is the essence of discrimination. As Lord Walker of Gestingthorpe put it in R (Carson) v Secretary of State for Work and Pensions,[18] “Discrimination on the ground of sex or race demeans the victim by using a sexual or racial stereotype as a sufficient ground for unfavourable treatment.” While in many situations (obviously when one is comparing like with like), such as in the cases dealing with selection requirements for secondary schools,[19] it may be relatively easy to identify less favourable treatment, this is not always so and a careful evaluative assessment will sometimes be required. Of course, in the assessment of this factor, context is all important. In addition, the question should be answered objectively and not subjectively from the point of view of the complainant. As was put by Dillon LJ in the Court of Appeal in R v Birmingham City Council,[20] “[T]he loss, because of sex, of the chance of getting something which is reasonably thought to be of value is enough to constitute sex discrimination.” Lord Goff of Chieveley in the House of Lords in the same case also referred to “reasonable grounds”.[21] That the test is objective was, as stated earlier, also confirmed in James v Eastleigh Borough Council.[22] The reference to “reasonable grounds” highlights the requirement in discrimination cases, once different treatment is shown, to evaluate objectively the grounds and factual criteria explaining why the discriminator acted as he or she did: see JFS.[23] 19. In some cases (such as the present) the discriminator will refer to some underlying objective, policy or reason to demonstrate that the difference in treatment does not amount to less favourable treatment. Apart from of course articulating such objective, policy or reason, it will be necessary to demonstrate that the difference in treatment is logically and reasonably connected to it. Only then will it be possible to take such objective, policy or reason into account in order to determine whether or not there has been less favourable treatment. I have found useful the analysis of Weatherup J in Re McMillen, a case involving length of hair restrictions in schools:-[24] “[21] Secondly, the approach to less favourable treatment. There are interrelated questions as to whether the restriction on appearance could properly be justified by the objective of the code and whether the restriction, in the context of the code as a whole, resulted in less favourable treatment for the boys. The initial assessment is made with reference to the objectives of the code in question. In other words, it is not a question of making comparisons with other people who would adopt the dress or appearance that is prohibited, but rather it is a question of assessing the restrictions that have been introduced against the objectives of the code. That, of course, requires one to identify the objectives. This approach is not to judge the quality of the reasons or the motives for introducing the code, because there can be no justification if there is direct discrimination, but it is to consider whether there is less favourable treatment by establishing a legitimate objective for the code and then assessing the treatment of those affected against the standard that has been adopted by the code to determine whether males or females are disadvantaged compared to each other. Thus in the employment context the objective of the restrictions may be to promote a concept of smartness on commercial grounds and that has been accepted as a legitimate objective. The code might define the particulars of smartness for commercial reasons, for example by no ponytails for men or collar and tie for men. The issue is not concerned with the extent of the general use of the prohibited items but whether a package that includes requirements which differ between men and women, is directed at the objective of the code and whether it imposes a particular disadvantage on one or other sex. In schools the context is different and the objective will be different.” This approach also has relevance when considering what is known as packaging.[25] 20. It is, however, important here to point out that while it is crucial to look at the grounds and factual criteria explaining the difference in treatment, it is irrelevant to consider the motive of the discriminator; in other words, in cases of direct discrimination, justification in this sense is irrelevant.[26] As Lord Kerr of Tonaghmore JSC said in JFS,[27] there is “the need to recognise the distinction between, on the one hand, the grounds for the decision (what was the basis on which it was taken) and on the other, what motivated the decision‑maker to make that decision.” 21. In the present case, as we shall see, the respondent explains the different treatment of male and female prisoners by reference to the policy of custodial discipline. It is said that custodial discipline requires the imposition of reasonable uniformity and conformity in appearance among inmates, and in this context, the respondent adopts the requirements stated in SO 41‑05. The difference in treatment of men and women in that standing order is explained by the underlying fact that it is said to reflect conventional standards of appearance in society. Accordingly, it is said there is no less favourable treatment of male prisoners. 22. The respondent’s arguments will have to be looked at more closely later, but it can be accepted that in some situations, resort to societal or conventional standards may be legitimate. One example of this is where for reasons of privacy and decency, different rules may exist between men and women. The segregation of men and women in different toilets provides a common scenario. So too the difference in cross gender rub down searches in prisons. In R (On the Application of James Dowsett) v Secretary of State for Justice[28], a policy of rub down searches in prisons was challenged as being directly discriminatory. In that case, male prisoners were subject to searches by both male and female prison officers whereas female prisoners could only be searched by female officers. The challenge in the case was made by a male prisoner. Silber J held this difference not to be discriminatory because on the facts there had to be considerations of privacy and decency.[29] In Conway v Canada,[30] in relation to frisk searches in circumstances similar to Dowsett, the Supreme Court of Canada held there was no violation of s 15 of the Canadian Charter of Rights and Freedoms.[31] It was said by La Forest J,[32] “the historical trend of violence perpetrated by men against women is not matched by a comparable trend pursuant to which men are the victims and women the aggressors”. 23. It is unnecessary (and unwise) to attempt to define the limits of the factor of conventional standards when considering the aspect of less favourable treatment, but it can be accepted that in certain circumstances it may be a relevant consideration. Much of course depends on the precise circumstances of any given case and the evidence before the Court. 24. There are also raised by the parties aspects of what are known as packaging and stereotyping, and it is to these facets I now turn. B.6 Packaging 25. It is the respondent’s[33] contention that when determining discrimination, it is wrong merely to compare one item or aspect with another without taking into account the whole context in which the relevant item appears. In the present case, the respondent contends that it is wrong merely to compare the length of hair provisions regarding male and female prisoners. It is important, the respondent says, to have regard to the whole context, this being custodial discipline in prisons.[34] Put another way – and I would have no quarrel with this on a general level – it is important to examine what is complained about as discrimination in context. After all, the comparison exercise between the complainant and the compared person must, as stated before, be a comparison between like and like. Section 10 of the SDO states that relevant circumstances must be taken into account. This taking into account of the whole is what is involved in the so‑called package approach. 26. This term finds its origin, certainly its articulation, in Smith v Safeway PLC.[35] The case concerned a complainant who was an assistant in the delicatessen department of a well‑known supermarket. He was dismissed on account of his refusal to cut his hair; his employers had a dress and appearance code which required male, but not female, staff to have hair not below collar length. The complainant claimed compensation against the employers on the basis of sex discrimination. On appeal from the Employment Appeal Tribunal, the Court of Appeal held against the complainant on the basis that it was necessary to look at the matter in proper context and the context of that case was a dress and appearance code. Accordingly, in this context there had to be adopted a package approach whereby it was accepted that men and women were not necessarily to be treated identically. Phillips LJ said this:-[36] “In my judgment, a package approach to the effect of an appearance code necessarily follows once one accepts that the code is not required to make provisions which apply identically to men and women. Phillips J. held that this was the approach more likely to lead to a sensible result in that case and in cases like it. I agree. This is not to say that when applying the test, the requirement of one particular item of a code may not of itself have the effect that the code treats one sex less favourably than the other. But one has to consider the effect of any such item in the overall context of the code as a whole. Mr Bibby submits that if such an approach is legitimate, it must be confined to dress and cannot be extended to hair. He makes the particular point that employees only have to comply with requirements as to dress while they are in the workplace, but can change their clothes so as to dress as they please once they leave work. As to this I accept Mr Elias’s argument that, while this may be a relevant consideration when applying the appropriate test, it does not affect the test itself. Appearance depends in part on ephemera, clothes, rings and jewellery worn, but it depends also on more permanent characteristics, tattoos, hair style, hair colouring and hair length. The approach adopted in the Schmidt case can, in my judgment, properly be applied to both types of characteristic.” The reference to the Schmidt case is a reference to Schmidt v Austicks Bookshops Ltd,[37] in which the Employment Appeal Tribunal also approached the matter on a broader basis, rather than using an item by item comparison. There, the alleged discrimination consisted of female employees not being permitted to wear trousers. 27. The respondent points to a number of cases, all involving dress or appearance codes, to demonstrate that the package approach has been commonly applied. Thus, for example, the requirement to wear a uniform for nurses did not mean that male and female nurses had to be treated identically.[38] 28. It is important of course to reiterate that the package approach ought to be treated as an exercise in putting matters in proper context, applying some common sense as well, so as to enable a proper comparison to be made for the purpose ultimately of discovering whether or not less favourable treatment has been accorded to the complainant in the particular aspect about which he or she has complained as being discriminatory. To take another example, where an appearance code required employees “to present a clean and tidy appearance and to dress in a businesslike way”, it was held[39] that a male employee who was required to wear a collared shirt and tie at work (whereas female employees were not) did not necessarily mean there was less favourable treatment. Keith J, who chaired the Employment Appeal Tribunal in the case, referred to the need to look at context and then continued:-[40] “Thus, an even‑handed approach does not necessarily mean that members of one sex are treated less favourably than members of the other simply because members of one sex are required to wear clothing of a particular kind but members of the other are not. It will depend on ‘the overall context of the code as a whole’. The relevant context when considering the code introduced by Jobcentre Plus was its overarching requirement for its staff to dress in a professional and businesslike way. The issue which the employment tribunal should have addressed, therefore, was whether the requirement for male members of staff to wear a collar and tie whereas no particular form of dress was required for female members of staff meant that the male members of staff were being treated less favourably than female members of staff in the context of that overarching requirement which applied to all members of staff. We do not think that the employment tribunal addressed that issue.” I would add here that one would almost invariably need to look at the objectives of the code, or underlying policy or reason in question in order to picture the whole context. 29. However, care needs to be taken when adopting the package approach to ensure that it does not become an exercise of merely comparing features applicable to the complainant with separate features belonging to the compared person and then asking whether overall, the two persons have been treated equally. The approach in discrimination cases is not a “tit for tat” or “swings and roundabouts” approach. As has been said,[41] “In deciding whether a woman who is prevented from wearing trousers to work is being less favourably treated on the ground of her sex than a comparable man, courts and tribunals should be wary of arguments which seek to balance the fact that a woman is denied opportunity X with the fact that men are denied opportunity Y”. This passage was referred to in the article by Robert Wintenute Recognising New Kinds of Direct Sex Discrimination: Transsexualism, Sexual Orientation and Dress Codes[42] who says this: “Sex distinctions applying to different choices cannot be lumped together and their net effect examined. Courts must look instead at their effect on the ability of individuals to make each specific choice. For the woman who wants badminton at the same price as a man, free swimming is no consolation. For the man who wants to wear a pony‑tail or a skirt, it is no consolation that women are prohibited from wearing short hair or trousers.” It is always necessary to examine whether the particular act or policy said to be discriminatory does in context result in less favourable treatment. Moreover, sometimes it may be that the particular act or policy, even though part of a whole, treats the complainant less favourably to such an extent that it would have the effect of rendering the whole policy as being discriminatory: see Smith v Safeway PLC.[43] 30. It is again unnecessary to define the limits of the package approach. Apart from anything else, much depends on the facts of any given case. I also add this. The way that past cases have been dealt with on their facts, provides no sure indication that seemingly similar factual situations will nowadays still produce the same outcome. 31. One final point on packaging. In some of the cases,[44] the approach has been to examine the aspect of less favourable treatment by reference to convention or conventional standards. I have touched upon this earlier.[45] In Smith v Safeway PLC, Phillips LJ stated[46] that a dress code was not rendered discriminatory if it “applies a standard of what is conventional”. 32. What may constitute convention or conventional standards will vary depending on the context and facts of any given case. For example, in a dress code which has its objective dressing in a businesslike way,[47] there may be conventional standards in this context which differ from how people may dress in everyday, social contexts. On the assumption this factor is relevant in the first place, what is required, however, is some factual basis to support what is asserted to be convention or conventional standards. A mere assertion or subjective belief without more will be unlikely to overcome the evidential hurdle. 33. What if asserted conventions or conventional standards involve stereotyping? It is to this facet I now turn. B.7 Stereotyping 34. Generally speaking, there will be found discriminatory conduct if there has been stereotyping. Stereotyping occurs when an individual or a group is treated in a different way to others in the same position by reason of a stereotype that has been applied to that individual or group. In the Roma case,[48] the claimants in that case, travelling from the Czech Republic to the United Kingdom, were subject to much tighter, longer and more intrusive immigration questioning than others simply because they were Roma (that is of Romani ethnic origin). They complained of less favourable treatment being accorded to them on the basis of their race. Even though there was some evidence to suggest that Roma persons had in the past advanced false claims to enter the United Kingdom, it was held by the House of Lords to be unacceptable to apply stereotypes. Baroness Hale of Richmond said this:[49] “The person may be acting on belief or assumptions about members of the sex or racial group involved which are often true and which if true would provide a good reason for the less favourable treatment in question. But ‘what may be true of a group may not be true of a significant number of individuals within that group’: see Hartmann J in Equal Opportunities Commission v Director of Education[50] …. The object of the legislation is to ensure that each person is treated as an individual and not assumed to be like other members of the group.” 35. Equal Opportunities Commission v Director of Education provides another example of discrimination by stereotyping, this time on account of sex. That was a case in which the Government’s then education policy of entry from primary school to secondary school was found to be discriminatory against girls purely on the assumption that at the relevant age, girls were more academically developed than boys.[51] 36. Where reliance is placed on convention or conventional standards, but they reflect stereotyping, a question may arise as to the appropriateness of relying on such convention or conventional standards for the purpose of defeating an argument of less favourable treatment. This is the point made by the author in Sex Discrimination Law[52] in criticising the Schmidt case: “Is an employer’s policy of refusing women to wear trousers to work sex discrimination under the 1975 Act?[53] The EAT decision in Schmidt, validating such policy if analogous, albeit different, restrictions are imposed on men, is not convincing. That an employer is requiring men to do what they normally do – wear trousers – and asking women to do what they normally do – wear skirts – cannot, of itself, amount to a non‑discriminatory policy. Otherwise employers would always have a defence if their practices mirrored social behaviour. This would clearly conflict with a major objective of the 1975 Act, which was introduced precisely because aspects of social behaviour were unfair to women”. 37. As will be seen later, it is unnecessary on the facts of the present case to resolve some of the potential difficulties, especially in the context of convention or conventional standards, that can arise in relation to stereotyping. This should be left to a case in which the point properly arises. It suffices to say that the point is capable of many nuanced arguments, and is certainly not without difficulty both conceptually and as applied in practice. B.8 The difference in treatment on the basis of sex 38. This is the 4th requirement stated in para 15(4) above. It is the crucial question. As Lord Nicholls of Birkenhead said in Nagarajan v London Regional Transport,[54] “Thus, in every case it is necessary to inquire why the complainant received less favourable treatment. This is the crucial question.” However, in asking the question “why?”, it is important to distinguish finding out what caused the alleged discriminatory treatment in question (this is permissible, indeed critical) from looking at justification or motive (this is not permissible).[55] It is also an inquiry based on substance and not form. In James v Eastleigh Borough Council,[56] the plaintiff complained of discrimination in a leisure centre operated by the defendant council in which free admittance was available only to those persons who had reached pensionable age. The plaintiff and his wife were both 61 years of age. While his wife gained entrance free of charge, the plaintiff was charged admission. The pensionable age for women was 60, for men it was 65. In the Court of Appeal, it had been held that the less favourable treatment given to the plaintiff was not on the basis of sex but on the basis of pensionable age, a criterion that applied equally to men and women. The majority of the House of Lords[57] held otherwise and analysed the problem as a matter of substance. Lord Bridge of Harwich, with whose speech Lord Ackner and Lord Goff of Chieveley agreed, said of the approach of the Court of Appeal:[58] “The fallacy, with all respect, which underlines and vitiates this reasoning is a failure to recognise that the statutory pensionable age, being fixed at 60 for women and 65 for men, is itself a criterion which directly discriminates between men and women in that it treats women more favourably than men ‘on the ground of their sex’”. 39. The above being the applicable legal approach, I now turn to its application to the facts. C. APPLICATION TO FACTS 40. There is no controversy between the parties regarding the requirements in paras 15(1) and (4) above: the respondent accepts there is a difference in the treatment between male and female prisoners, and that this difference is on the basis of sex. Further, as I understand it, there is no real controversy regarding the para 15(2) requirement either: it is accepted that male and female prisoners are, for the purposes of hair length, in comparable positions. The real difference dividing the parties has been over the factor whether less favourable treatment, seen in proper context, has been accorded to male prisoners like the appellant, para 15(3) above.[59] Here, it should first be noted that on its face, without more, SO 41‑05 suggests that male prisoners are not only treated differently to female prisoners but are also, on account of the absence of choice, treated less favourably. The burden is then passed onto the respondent to say why this is not so. 41. The respondent’s approach has been inconsistent in the lower courts. Before Au J, it was the respondent’s case “that SO 41-05 is a preventive operational measure which has the effect of reducing the number of risks which threaten the individual rights of prisoners while in custody.”[60] The risks were enumerated as follows:-[61] “These risks are in summary (a) the inherent vulnerability of prisoners with long hair in the case of an attack by another inmate involving violence; (b) the risk that long hair is used as a potential method of concealing prohibited items (eg, a razor blade or self-made weapons) which could be used by a prisoner both to attack other inmates, to inflict self-harm, or for suicide; (c) the risk of violence associated with triad or gang affiliation; and (d) the risk of prisoners using long hair as a readily accessible tool (as opposed to a method of concealment) for self-harm and suicide. Importantly, it is the Commissioner’s evidence that, from experience (partly backed by some statistics), these risks are significantly higher in the male inmates population than the female inmates population. It is therefore the Commissioner’s position that the real ground for the hair cutting requirement is not on the sex but on the maintenance of prison security and discipline.” Although this passage indicates that the respondent’s position was that it was disputing whether the different treatment was on the ground of sex, it clearly was since the asserted difference lay in the contention that security risks were higher in the case of male prisoners than female. Be that as it may, the real argument of the respondent was that by reason of these enhanced risks, male prisoners had to be treated differently, but they did not receive less favourable treatment. The respondent seemed also to be arguing that the relevant circumstances between male and female prisoners were not alike.[62] 42. Au J held against the respondent on the basis of stereotyping[63] in that the various risks involved stereotypes of male prisoners “as a gender as a whole”. Whether or not this analysis is sound in the context of security risks is immaterial because there is a more fundamental objection. The evidence adduced by the respondent in the proceedings to prove the existence of the said risks does not demonstrate any real link between the length of a prisoner’s hair and the asserted risks. The evidence is contained in the affirmation of Leung Kam Yan[64] dealing with the aspect of security concerns.[65] Although certain statistics are given, nowhere is it compellingly explained, other than by way of conjecture, why there is a link between these concerns and the length of hair. Ultimately, the respondent’s evidence does not prove or explain (as distinct from merely assert) why male and female prisoners should be treated differently in the context of security concerns relating to the length of hair. 43. The respondent’s stance shifted substantially in the Court of Appeal. The emphasis was instead placed on custodial discipline. His position, accepted by Lam VP (who gave the main judgment), was this: “[T]he common underlying objective of the policy in issue is to foster custodial discipline by imposing reasonable uniformity and conformity among the inmates (both male and female inmates) and reasonable restrictions are set by reference to the respective conventional standards for appearance for male and female inmates.”[66] The reference to conventional standards was a reference to “the conventional standards of appearance for men and women in our society”.[67] The conventional standards, according to the Court of Appeal, amounted to these: “The conventional hairstyle of male persons in Hong Kong is a short hairstyle, male inmates are therefore required to wear a similar short hairstyle. On the other hand, as the conventional hairstyle of Hong Kong women is that it may either be long or short (putting the matter in a simplified way), female inmates are therefore allowed under the female haircut requirement to keep their own hairstyles.”[68] 44. The respondent maintains this position before us and argues essentially that in the context of custodial discipline, there is no less favourable treatment given to male prisoners compared with female prisoners. As stated above, the respondent contends that he is entitled, as part of custodial discipline, to impose reasonable uniformity and conformity in appearance among inmates, whether male or female.[69] In the present case, the respondent has done so by reference to SO 41‑05, notwithstanding that that standing order deals only with health and cleanliness at least as far as male prisoners are concerned. 45. The respondent’s position is contained in the affirmation of Supt Leung:- “5. It is part of the CSD’s overall mission and objectives to provide a secure, safe, humane, decent and healthy environment for people in custody. In order to achieve the mission, maintaining prison security and custodial discipline are of paramount importance whereas preserving health and cleanliness (including the requirement for male prisoners to have reasonably uniform haircut), amongst numerous stipulations and orders, is only one of the requirements under the regime within prison context for the purpose of maintaining prison security and custodial discipline. Custodial Discipline Concerns 6. The CSD considers that a fundamental part of custodial discipline is reasonable uniformity within correctional facilities. Custodial discipline ensures proper administration of correctional facilities by implementing uniform rules on all matters for the inmate population. Within a prison context, reasonable uniformity is important because it fosters both a sense of collective conformity and respect for prison authority among inmates during their remedial terms. As a matter of proper administration in an environment that is meant, in part, to instill law‑abiding concepts and to correct past behavior, individuality is de‑emphasized to give way to custodial discipline. 7. Custodial discipline is enforced by a set of Standing Orders issued by the Commissioner. The Standing Orders cover a variety of matters including, but are not limited to: diet, clothing, work hours, exercise routine, sleep hours, as well as hair cutting (SO 41‑05) for reasonable uniformity and maintaining custodial discipline. The Commissioner has general authority under Prison Rule 77 to make such standing orders which are subject to review semi‑annually. Therefore, while SO 41‑05 can be said to be authorized under Prison Rules 34 and 119, it is also authorized under Prison Rule 77 as part of the Commissioner’s general powers of custodial administration to maintain custodial discipline in all correctional facilities.” 46. As pointed out earlier, SO 41‑05 prescribes different treatment for male and female prisoners. The respondent seeks to explain this difference in treatment by reference to the underlying conventional standards of appearance in society for men and women. So analysed, it is argued it cannot be said that there is any less favourable treatment as far as male prisoners are concerned because the length of hair requirements contained in the standing order treat men and women in accordance with conventional standards. Accordingly, whether one adopts the item by item approach (simply comparing hair length requirements) or the package approach (this is said to be the need for uniformity and conformity in appearance as a matter of custodial discipline), the respondent contends that the appellant has simply not made out a case of less favourable treatment.[70] 47. I am unable to accept these submissions. 48. It can of course be readily accepted that custodial discipline is a legitimate factor to be considered, as can the proposition that a person can expect to be deprived of certain privileges or freedoms normally enjoyed outside of incarceration. But we are considering and comparing the respective situations of male and female prisoners in determining the issue of discrimination. It is necessary not to lose focus of this. SO 41‑05 clearly sets out a different treatment for male and female prisoners. 49. The stated objective, policy or reason for the difference in treatment being custodial discipline as earlier set out,[71] the next step is to analyse the difference in treatment by reference to it. For my part, it is difficult to see how a difference in treatment regarding the length of hair between male and female prisoners based on asserted conventional standards has any reasonable connection with custodial discipline. The reliance on conventional standards amounts to no more than saying[72] that the respondent wished, as a matter of custodial discipline, to have inmates in prison have the same length of hair as persons outside of prison. It is not readily apparent, and no explanation was provided by the respondent, as to why this had any reasonable connection with custodial discipline. Mr Wong emphasised the point that in prison, there has to be a de‑emphasis on a prisoner’s individuality. As Supt Leung said in his affirmation,[73] “individuality is de‑emphasized to give way to custodial discipline”. This point can perhaps as a general statement be accepted but in the present case, it is difficult to accept, without a proper explanation, why individual choice should be denied to male prisoners but not female ones, and what this selective denial of choice has to do with a de‑emphasis on individuality anyway. 50. Without a reasonable connection to the stated objective, policy or reason, the respondent does not really begin to explain the difference in treatment and why there has not been less favourable treatment. 51. There is a further impediment to the respondent’s argument. The respondent’s submissions depend on the factual premise that there exist conventional standards of appearance for men and women in society in relation to the length of hair. The respondent refers to these standards in implementing the policy on hair lengths required by SO 41‑05. Does the evidence support this contention? Throughout the respondent’s submissions, there are references to the evidence in support being unchallenged. However, the evidence far from establishes this and it is difficult to accept the evidence at face value. The respondent relies on[74] para 36 of the affirmation of Supt Leung:- “Finally, I would like to point out that it is not factually correct to say that female inmates do not have any hair restrictions. For example, in Lo Wu Correctional Institution, female inmates with long hair are required to tie back their hair in accordance to hairstyle requirement issued by the institution. Attached and shown to me is a copy of the “Hairstyle Requirement of Male and Female Prisoners”, which shows the photos of required standard of hairstyle in Stanley Prison and Lo Wu Correctional Institution, marked as Exhibit “LKY-17”. The hairstyle requirements for both male and female prisoners are specified by making reference to the hairstyle of male and female persons in Hong Kong society without carrying any stigmatization. It is not as if male prisoners are required to shave off all their hair, they simply must keep their hair about the length of a standard hair cut. Such length is objectively reasonable and does not impose an undue burden upon prisoners, whose rights and privileges are more limited as a result of lawful incarceration.” (Emphasis added) 52. It may be that what was said in this paragraph was not challenged by the appellant as such but its contents do not provide a sufficient factual basis for the conventional standards relied on by the respondent. First, SO 41‑05 refers on its face to “health and cleanliness”. No explanation is provided as to how these aspects are related to conventional standards for hair length and it is difficult to see how they can. The respondent has simply used it as a reference point. Secondly, this passage from Supt Leung’s affirmation does not mention conventional standards as such although admittedly reference was made to the hairstyle of men and women in Hong Kong society. Thirdly and most important, no details are provided and no basis is given for saying that in our society, the conventional hairstyle for men is a short one whereas for women, hair can be long or short.[75] There must be sufficient evidence going to these matters; they cannot simply be asserted. In his submissions to us, Mr Wong seemed to suggest it was sufficient for Supt Leung merely to state this fact as though it was some sort of factual finding on his part and that it was for the appellant to disprove it. The old adage of “he who asserts, must prove” applies. It is for the respondent to prove the conventional standards he relies on. Moreover, these are not matters of which judicial notice can be taken. On the contrary, one would have thought judicial notice would be taken, if anything, of the fact that the hairstyles for men and women in our society would be quite diverse. Mr Wong sought to reinforce his submissions by referring to the fact that SO 41‑05 had been subject to semi‑annual reviews since its introduction (in the mid‑1950s we are told). If this were so, it seems somewhat unrealistic, if not bizarre, to suggest that SO 41‑05, if it was intended to reflect conventional standards in hair length in society, had not been amended in any way since that time. Fourthly, there is some inconsistency in the way the Court of Appeal treated the aspect of conventional standards. Its view (as stated above) as to the conventional standards for hair length for men and women is to be contrasted with the reference later in its judgment to there being “different conventional standards for appearance in our society.”[76] 53. It is unnecessary to deal with the further point raised by the appellant[77] as to whether in any event, conventional standards regarding the length of men and women’s hair in society involved stereotyping. Apart from anything else, until one knows the exact extent of the evidence regarding this, one is not in a position to embark on a meaningful analysis. As just discussed, the respondent’s evidence is unsatisfactory and such conventional standards are not made out. Accordingly, without such evidence, in the present case one is simply left with his view, based on SO 41-05, of what hair lengths for men and women ought to be in society, and this amounts to stereotyping on the respondent’s part. 54. For these reasons, I am of the view that less favourable treatment was given to the appellant compared with female prisoners. Accordingly, all four requirements as stated in para 15 above are satisfied in the present case for the purposes of the SDO. There has been discrimination on the basis of sex. 55. I should finally in the discussion in the present section just make this point. The respondent has the responsibility for and expertise in administering prisons. His duties are set out in the Prisons Ordinance. In this respect, he should be given a wide margin of discretion to adopt appropriate policies. However, he must do so consistently with applicable laws and, on the facts of the present case, he has discriminated against the appellant in violation of s 5(1)(a) of the SDO. D. THE CONSTITUTIONAL ISSUE UNDER ARTICLE 25 OF THE BASIC LAW 56. In view of the conclusion reached on the previous issue, it is unnecessary to deal with this issue. However, on the facts of the present case, the outcome would not be different under Article 25 of the Basic Law. E. CONCLUSION 57. For the above reasons, the appeal is allowed. As to the exact form of the order, the respondent asked that the Court should entertain further submissions in the event the appeal was to be allowed. I would therefore order that submissions on the form of order and on costs be made in writing. I would direct that the respondent file and serve submissions within 14 days of the handing down of this judgment. The appellant will have 14 days from being served with the respondent’s submissions to file and serve submissions in reply. The respondent will have 7 days from being served to file and serve any submissions in reply. Mr Justice Ribeiro PJ: 58. I agree with the judgment of the Chief Justice. Mr Justice Fok PJ: 59. I agree with the judgment of the Chief Justice. Mr Justice Chan NPJ: 60. I agree with the judgment of the Chief Justice. Lord Collins of Mapesbury NPJ: 61. I agree with the judgment of the Chief Justice. Mr Hectar Pun SC, Ms Annie Leung and Mr Anson Wong Yu Yat, instructed by Ho Tse Wai & Partners, assigned by the Director of Legal Aid, for the Appellant Mr Stewart Wong SC, Mr Jin Pao SC and Mr John Leung, instructed by the Department of Justice, for the Respondent [1] Cap 480 (“the SDO”). We are concerned in particular with section 5(1)(a) of that Ordinance (set out in para 10 below). [2] Article 25 states that “All Hong Kong residents shall be equal before the law”. [3] As set out in the Hong Kong Bill of Rights Ordinance Cap 383 (“the HKBORO”). This article incorporates under Hong Kong law the provisions of the International Covenant on Civil and Political Rights (see Article 39 of the Basic Law). [4] Issued by the Commissioner of Correctional Services pursuant to Rule 77(4) of the Prison Rules Cap 234A under the Prisons Ordinance Cap 234. Further, the reference in the Standing Order to health and cleanliness can be traced to Rules 34 and 119 of the Prison Rules. These provide as follows:- (1) Rule 34 “34. Obedience to directions as to cleanliness Every prisoner shall obey such directions as regards washing, bathing, shaving and hair cutting as may from time to time be prescribed.” (2) Rule 119 “119. Duty as to personal cleanliness of prisoners The Chief Officer shall ensure that the prisoners’ clothing is kept in good state of repair, their hair kept trimmed, and their washing, hair clipping and personal ablutions attended to.” [5] In a judgment dated 17 January 2017 (“CFI Judgment”). [6] Cheung CJHC, Lam VP and Poon JA in a judgment dated 30 April 2018 (“CA Judgment”). [7] In a judgment dated 22 January 2019. [8] Ma CJ, Fok PJ and Chan NPJ. [9] (2018) 21 HKCFAR 324, at para 33. [10] [2009] UKSC 15, [2010] 2 AC 728, at para 56 (per Baroness Hale of Richmond JSC). [11] In the Court of Appeal, Cheung CJHC dealt with the appeal also on the basis of indirect discrimination, but this has not been pursued before us. [12] (2000) 3 HKCFAR 459, at 476A-D. [13] [1989] 1 AC 1155, at 1194A-B. [14] Section 1(1) of the Sex Discrimination Act 1975 has the same wording as s 5(1) of the SDO. [15] [1990] 2 AC 751, at 765D. [16] [2004] UKHL 55, [2005] 2 AC 1, at para 73. [17] See JFS, at para 56 quoted in para 11 above. [18] [2005] UKHL 37, [2006] 1 AC 173, at para 49 (in a passage quoted by the Court in QT v Director of Immigration (2018) 21 HKCFAR 324, at para 27). [19] Birmingham City Council v Equal Opportunities Commission: see para 14 fn 13 above; Equal Opportunities Commission v Director of Education [2001] 2 HKLRD 690. [20] At 1176B-C. [21] At 1193G-H. [22] At 765D (Lord Bridge of Harwich). [23] At paras 16, 22, 23 (per Lord Phillips of Worth Matravers PSC). [24] [2008] NIQB 21, at para 21. [25] See Section B.6 below. [26] See JFS, at paras 14, 17, 20, 22, 57, 59, 113, 115, 116, 132 and 141. [27] At para 116. [28] [2013] EWHC 687 (Admin). [29] At para 64. It is interesting to note that the consideration of “privacy and decency” features in the SDO in relation to communal accommodation: see s 52(1). [30] [1993] 2 SCR 872. [31] The equality provision of the Charter. [32] At 877. [33] The respondent is represented in this appeal by Mr Stewart Wong SC, Mr Jin Pao SC and Mr John Leung. [34] This, after all, is expressly referred to in s 9 of the Hong Kong Bill of Rights Ordinance: see para 2 above. [35] [1996] ICR 868. [36] At 877C-F. [37] [1978] ICR 85. [38] See Burrett v West Birmingham Health Authority [1994] IRLR 7. [39] In Department for Work and Pensions v Thompson [2004] IRLR 348. [40] At para 27. [41] David Pannick: Sex Discrimination Law (1985) at Pg 187. [42] (1997) 60 MLR 334, 355. [43] At 877C-D, see para 26 above. [44] See for example Re McMillen at para 22 (“a code applying conventional appearance is not of itself discriminatory”); G v The Head Teacher and Governors of St Gregory’s Catholic Science College [2011] EWHC 1452 (Admin), at para 56 (“What is to be regarded as conventional may well vary as time goes by and will depend on the facts of a particular case”) and para 59 (“Furthermore, although cornrows for boys are by no means unusual, I am not persuaded that the defendants are wrong to regard them as not being conventional.”). [45] Paras 22 and 23 above. [46] At 877H – 878A. [47] Such as in Thompson. [48] Para 15 fn 16 above. [49] At para 82. [50] See para 18 fn 19 above. [51] The policy effectively boosted the academic scores of boys and in turn downgraded those of girls in primary schools such that boys then stood a better chance of gaining entry into the secondary school of their choice. [52] See para 29 fn 41 above, at Pg 187. [53] The Sex Discrimination Act 1975 which corresponds in material respects with the SDO. [54] [2000] 1 AC 501, at 510H-511A. [55] Para 20 above; see also JFS at para 64 (per Baroness Hale of Richmond JSC). [56] See para 14 fn 15 above. [57] The decision was a 3:2 one. The dissenting speeches of Lord Griffiths and Lord Lowry largely adopted the analysis of the Court of Appeal that it was pensionable age, and not sex, that led to the difference in treatment. [58] At 763H. [59] See Respondent’s written Case, at para 12(1). [60] Para 27 CFI Judgment. [61] Para 28 CFI Judgment. [62] The factor mentioned in para 15(2) above. [63] Para 40 CFI Judgment. [64] A superintendent of the Penal Administration Section, Operations Division of the Correctional Services Department. [65] See paras 9 to 20 of the affirmation. [66] At para 52 CA Judgment. [67] At para 66 CA Judgment (Lam VP); see also para 17 CA Judgment (Cheung CJHC). [68] Para 9 CA Judgment. [69] Rule 77(4) of the Prison Rules refers to the power of the respondent to issue orders for the discipline of persons. And, as stated earlier, s 9 of the Hong Kong Bill of Rights Ordinance refers also to “custodial discipline”. [70] I have condensed the respondent’s submissions which are largely to be found in paras 4 to 7, 27 to 65 of his written Case. [71] Paras 21, 43 to 45 above. [72] This was confirmed by Mr Wong in argument. [73] See para 45 above. [74] In para 6 of the respondent’s written Case. [75] See para 43 above. [76] See para 75 CA Judgment, in the context of stereotyping. [77] The appellant is represented by Mr Hectar Pun SC, Ms Annie Leung and Mr Anson Wong. 
Mr Justice Fok PJ: 1. This uncontested appeal arises as a consequence of the Court’s judgment in HKSAR v Wan Thomas,[1] which was handed down on 14 May 2018. Based on that judgment, the appellant now appeals against her conviction on one count of conspiracy to defraud arising out of her employment by a company which offered visiting services to prisoners on remand awaiting trial. 2. The facts out of which the conviction arose are set out in HKSAR v Wan Thomas at [9] to [13]. The appellant was one of the nine defendants jointly charged with the offence, being the 8th defendant (D8) at trial. As noted in HKSAR v Wan Thomas, three of the defendants pleaded guilty before trial and the trial proceeded against six of the defendants.[2] The appellant, together with the five other defendants who stood trial, were convicted by the magistrate on 13 September 2013.[3] The appeals to the Court of Appeal by four of the defendants, including the appellant, against conviction were dismissed on 17 October 2016.[4] 3. Thereafter, two of the original defendants (D1 and D2 at trial) further appealed to the Court of Final Appeal in the appeals which led to the judgment of the Court in HKSAR v Wan Thomas. Neither of the other two appellants before the Court of Appeal (D8 and D9) sought to appeal to this Court, until the application (by D8) leading to the current appeal. 4. The prosecution case at trial was that the defendants charged had conspired with other unknown person to defraud officers of the Correctional Services Department (“CSD”) by dishonestly and falsely representing to those officers that each of them was a “friend” of an inmate remanded at Lai Chi Kok Reception Centre. The purpose of the misrepresentation, it was alleged, was to induce the officers to act contrary to their public duty, namely to grant the defendants permission to visit the relevant inmates under remand at the Reception Centre which the officers would not otherwise have granted. 5. The Court held, in HKSAR v Wan Thomas, that the appeals of D1 and D2 at trial should be allowed and their convictions quashed. Essentially, the Court’s reasons for so holding were that: (1) The appellants were “visitors” under rule 203 of the Prison Rules[5] and “visitors” under rule 203 is to be construed to mean “relatives and friends” as provided in rule 48 of the Prison Rules;[6] (2) The word “friends” in rule 48 is to be construed to include a person: “… (a) who has been requested to visit the prisoner, either directly by the prisoner himself or indirectly through a relative or personal acquaintance of the prisoner; (b) who wishes to visit the prisoner in order to provide him with some moral or material benefit consistent with the statutory purposes of visits to that category of prisoner; and (c) by whom the prisoner is willing to be visited. If those conditions are satisfied, the visitor should, objectively, be regarded as a ‘friend’ of the prisoner.”[7] (3) Since, so understood, the appellants were “friends” of the prisoners whom they visited, they did not make misrepresentations to the CSD officers as to their relationship to those prisoners and therefore those officers were not induced to act contrary to their duty to admit persons who were not entitled to visit the prisoners in question;[8] (4) In any event, given the broad meaning of the word “friend” and the absence of guidelines from the CSD as to its meaning, the evidence at trial had not been sufficient to establish the prosecution case that the appellants had made any misrepresentation dishonestly;[9] and (5) Similarly, merely because they were all from the same company, there was insufficient evidence to establish the prosecution case that the appellants had all agreed together to induce the CSD staff to admit them as visitors by selecting the category “friend” in the relevant Visit Request Slips.[10] 6. In the light of this Court’s judgment in HKSAR v Wan Thomas, by a notice dated 10 September 2018, the appellant applied to the Appeal Committee for leave to appeal out of time. She explained that she did not appeal within the prescribed time after her appeal to the Court of Appeal because she did not have the means or resources to pursue a further appeal against her conviction. After reading the news about the Court’s judgment handed down on 14 May 2018, she made inquiries which eventually led to her applying for legal aid to commence an appeal to this Court. 7. The respondent, being prepared to concede the appeal, did not oppose the grant of leave to appeal and so, by an order dated 11 October 2018, the Appeal Committee granted the appellant leave to appeal out of time on the ground that substantial and grave injustice had been done to her by reason of her conviction. The exceptional grant of leave to appeal out of time was justified in the present case for reasons similar to those that applied in HKSAR v Wong Wai Kwong David.[11] The appellant’s conviction arose from the same case, involved the same offence and was based on the same facts and evidence as that leading to the judgment in HKSAR v Wan Thomas. 8. In granting leave, the Appeal Committee gave directions for the filing of a joint case by the parties and for the appeal to be disposed of on the papers in the event the Court was satisfied the appeal should be allowed.[12] 9. In the parties’ Joint Case, filed on 11 December 2018, it was contended that the appellant’s conviction is unsustainable as a matter of law for the same reasons that led to the appeals of D1 and D2 being allowed in HKSAR v Wan Thomas. The parties jointly invited the Court to allow the appellant’s appeal on the basis that she is in the same position as the appellants in that case. That submission is clearly correct and her conviction in these circumstances has resulted in substantial and grave injustice to her. 10. In the Joint Case, the parties also drew to the attention of the Court that D4, D5 and D6, who pleaded guilty before trial and did not appeal to the Court of Appeal, have now, since the grant of leave to appeal in this case, also appealed against their convictions, relying on this Court’s judgment in HKSAR v Wan Thomas, and their uncontested appeals[13] were allowed on 6 November 2018. Accordingly, of the original nine defendants, five have already had their convictions quashed. This reinforces the fact that substantial and grave injustice would result from the appellant remaining convicted. 11. For these reasons, we allow the appeal and quash the appellant’s conviction. The parties being in agreement as to the issue of costs, we also order that the respondent shall pay to the appellant an agreed sum of HK$5,760, being the appellant’s legal aid contribution in respect of this appeal. Joint Written Submissions by: Mr Eric T.M. Cheung, Solicitor Advocate (with HRA in criminal cases), instructed by ONC Lawyers, assigned by the Director of Legal Aid, and Mr Jevons C.H. Chan, instructed by ONC Lawyers, acting on a pro bono basis, for the Appellant Mr David Leung SC, DPP and Ms Audrey Parwani, SPP, of the Department of Justice, for the Respondent [1] [2018] HKCFA 15, (2018) 21 HKCFAR 214 (“HKSAR v Wan Thomas”). [2] D4, D5 and D6 had pleaded guilty before trial and so the trial proceeded against the defendants who were D1, D2, D3, D7, D8 and D9: ibid. at [5]. [3] KTCC 2097/2013. [4] HCMA 700/2013, reported in [2016] 5 HKLRD 656: these were the appeals of D1, D2, D8 and D9 at trial. D3 at trial did not appeal. The appeal of D7 at trial was dismissed, upon abandonment, by an order dated 25 July 2014. [5] (Cap.234A). [6] [2018] HKCFA 15, (2018) 21 HKCFAR 214 at [23] to [30]. [7] Ibid. at [45]. [8] Ibid. at [50]. [9] Ibid. at [52] to [54]. [10] Ibid. at [55]. [11] (2015) 18 HKCFAR 29 at [15] to [16]. [12] In accordance with the procedure for uncontested appeals laid down in Mok Kin Kau v HKSAR (2008) 11 HKCFAR 1 and HKSAR v Shum Wan Foon (2014) 17 HKCFAR 303. [13] HCMA 369/2018, [2018] HKCFI 2648. 
Mr Justice Ribeiro and Mr Justice Fok PJJ : 1. The offence of money laundering generally involves a person dealing with money or property which he knows or has reasonable grounds to believe represents the proceeds of an indictable offence with a view to concealing the criminal origins of such proceeds. In this appeal, the question arises whether, by virtue of a statutory definition, the offence can be committed where a person deals with funds that are known not to derive from any offence but which are intended to be used as part of a fraudulent conspiracy. 2. The Appeal Committee granted leave to appeal on the basis of a question formulated as follows: “Whether the expression ‘proceeds of an indictable offence’ in s 25(1) of the Organized and Serious Crimes Ordinance, Cap 455, is confined to money gained from the commission of an indictable offence or, instead, extends to money used in the furtherance of such an offence.” A. The parties 3. The appellants are four of the six defendants who appeared at the trial. Two of them, namely, Wayland Tsang Wai Lun (“Tsang”) and his wife Nancy Kwok Wai Man (“Kwok”), controlled a listed company called Grand Field Group Holdings Limited (“Grand Field”), both being directors and shareholders, and Tsang being its chairman. They were named as 1st and 2nd defendants[1] and, under what was Charge 3, convicted of having conspired with an immunised prosecution witness named Steve Au Yeung Keung (“Au Yeung”) to defraud the Stock Exchange of Hong Kong (“Stock Exchange”) and Grand Field’s shareholders (“the Charge 3 conspiracy”). 4. The other two appellants, namely, Charles Cheng Kai Ming (“Cheng”) and George Li Kwok Cheung (“Li”), were executive directors of another listed company called Upbest Group Limited (“Upbest”), the holding company of a group providing financial and securities broking services. They were named as the 4th and 5th defendants[2] and alleged to have implemented a scheme involving circular payments aimed at concealing the Charge 3 conspiracy. They were convicted of having conspired with Tsang, Kwok and others[3] to commit a money laundering offence contrary to section 25(1) of the Organized and Serious Crimes Ordinance (“OSCO”).[4] The present appeal relates solely to the appellants’ convictions on this charge. B. The factual background 5. The case was tried in the District Court before HH Judge A Wong[5] whose findings relevant to this appeal may be summarised as follows. 6. In mid-March 2002, the price of Grand Field’s shares had fallen from $0.68 to $0.12. In an effort to bolster that price, Tsang and Kwok hatched a scheme to give a false impression of profitable activity involving a Mainland joint venture to construct a natural gas pipeline near Chongqing. A company called Sino Richest Limited (“Sino Richest”) was formed with shareholders, including a company called Logistic China Enterprises Limited (“Logistic China”), who were ostensibly unconnected with Grand Field. On 25 May 2002, Sino Richest purported to sign a joint venture agreement with the Mainland party[6] and shortly thereafter, Grand Field purported to acquire 75% of Sino Richest in consideration of Grand Field issuing to the shareholders of Sino Richest 315 million new Grand Field shares. On 4 June 2002, a bullish public announcement was made about this joint venture which was said to be subject to government licences and approvals. 7. The public announcement was false, it never having been the intention of Tsang and Kwok to proceed with the joint venture. However, Grand Field came under increasing pressure from the Stock Exchange to provide further details relating to the joint venture and the needed government approvals. As Tsang and Kwok were unable to comply, hoping to put an end to such pressure, they sought to pretend with Au Yeung’s help that Grand Field had sold its interest in the joint venture (held via Sino Richest) back to Logistic China. The purported sale was announced on 12 August 2003. This pretence led to their conviction of the Charge 3 conspiracy. Charge 3 alleged that Tsang, Kwok and Au Yeung had conspired to defraud the Stock Exchange and the shareholders of Grand Field by dishonestly “... concealing that there had been no genuine acquisition of a business to develop and construct a gas pipeline in Chongqing, the Mainland; and falsely representing that there was a genuine disposal of the aforesaid business in Chongqing”. 8. To give credence to the re-acquisition by Logistic China, the four appellants (and others) set up a scheme with a view to providing evidence that Logistic China had duly paid Grand Field (through its subsidiary called Ka Fong Industrial Limited (“Ka Fong”)) $32 million to re‑acquire the Sino Richest shares. This was where Cheng and Li came into the picture. On 31 July 2003, advancing funds which originated from Upbest, they caused a series of payments to be made (with Cheng signing a series of Upbest cheques) which resulted in a circular flow of funds bringing the money back to Upbest on the same day, after having passed through a web of local and offshore companies.[7] Payments to and by Logistic China, Ka Fong and Tsang formed part of these circular payments. Logistic China received $32 million from a company upstream in the circular flow and then, purporting to pay for the re-acquired Sino Richest shares, made a payment of $32 million to Ka Fong (Grand Field’s subsidiary). Ka Fong (by a cheque signed by Kwok) then paid the $32 million to Tsang who then passed the money onwards so that it was eventually returned, after certain disguising complications, to Upbest. On the following day, 1 August 2003, Grand Field faxed to the Stock Exchange a copy of the payment-in slip showing that Logistic China had paid $32 million to Ka Fong on the previous day. 9. These circular payments were the subject of Charge 4 which alleged a “conspiracy to deal with property known or believed to represent the proceeds of an indictable offence contrary to ... sections 25(1) and (3) of [OSCO]”, particularised as follows: “[Tsang, Kwok, Cheng, Li (and others[8])] ... knowing or having reasonable grounds to believe that property, namely $32 million ... in whole or in part directly or indirectly represented proceeds of an indictable offence, conspired together with [Au Yeung] to deal with the said property.” C. The prosecution’s case on the main OSCO provisions 10. The provision which creates the substantive offence in respect of which the conspiracy under Charge 4 was charged is section 25(1) of OSCO which states: Section 25(1) “Subject to section 25A, a person commits an offence if, knowing or having reasonable grounds to believe that any property in whole or in part directly or indirectly represents any person’s proceeds of an indictable offence, he deals with that property.” 11. There was no suggestion that the $32 million which was advanced by Upbest consisted of funds derived from the commission of an indictable offence, hence the point of law raised in the present appeal. The prosecution’s contention that those funds are nonetheless in law to be regarded as the “proceeds of an indictable offence” within the meaning of section 25(1) relies on section 2(6)(a) of OSCO which provides: Section 2(6)(a) “For the purposes of this Ordinance ... a person’s proceeds of an offence are – (i) any payments or other rewards received by him at any time … in connection with the commission of that offence; (ii) any property derived or realised, directly or indirectly, by him from any of the payments or other rewards; and (iii) any pecuniary advantage obtained in connection with the commission of that offence...” 12. The prosecution argues that for the purposes of the conspirators’ intended section 25(1) substantive offence, the $32 million advanced by Upbest would come within section 2(6)(a)(i) because it would be a payment received by Tsang “in connection with” the commission of the Charge 3 conspiracy in that it would be used in furtherance of that offence. By receiving the funds and using them to make an onward payment, Tsang would (so the prosecution alleges) be dealing with such proceeds with the requisite knowledge and so committing the section 25(1) offence. The conspiracy consisted of an agreement by the four appellants to commit the substantive offence so constituted. 13. Accordingly, the way that the prosecution put its case against the appellants on this charge was succinctly summarised in its opening submissions as follows: “In respect of Charge 4 the underlying indictable offence is the conspiracy to defraud, the subject of Charge 3. The HK$32 million Ka Fong ultimately received from Logistic China and paid into [Tsang’s] bank account (the cheque having been signed by [Kwok]) were the proceeds of that conspiracy. [Tsang] dealt with the proceeds by receiving it through [his] personal bank account. It was [Tsang’s] proceeds as it was a payment received by him in connection with the conspiracy to defraud by falsely representing that there was a genuine disposal of the joint venture to develop and construct a gas pipeline in Chongqing.” 14. The prosecution’s construction of those provisions is therefore that section 2(6)(a)(i) widens the concept of “proceeds of an indictable offence” in section 25(1) to extend to money or property received in circumstances in which such receipt may be said to be “in connection with” the commission of an indictable offence even if that money or property is known not to be tainted as a benefit received on account of the commission of such an offence. Such money or property, the prosecution contends, includes “clean” money which is intended to be used as an instrument for committing an indictable offence. It includes, in the present case, the funds advanced by Upbest intended to be used as a vehicle for deceiving the Stock Exchange and Grand Field’s shareholders in furtherance of the Charge 3 conspiracy. That construction was accepted by the trial Judge and the Court of Appeal[9] which upheld the appellants’ conviction. D. The proper construction of the OSCO provisions 15. With respect, for the reasons which follow, we are unable to agree with the conclusion reached in the Courts below. In our view, on the true construction of the relevant sections, the prosecution’s wide interpretation is erroneous. D.1 The elements and structure of the main OSCO provisions 16. Before seeking to construe the relevant sections, it is worthwhile examining the elements of the offence and the structure of sections 25(1) and 2(6)(a)(i) when read together. 17. The section 25(1) offence is committed by the act of dealing[10] with certain property which has specified characteristics. It must be property which a person knows or has reasonable grounds to believe “in whole or in part directly or indirectly represents any person’s proceeds of an indictable offence”. As the Court held in Oei Hengky Wiryo v HKSAR (No 2),[11]section 25(1) does not define the actus reus as dealing with the proceeds of an indictable offence. It defines it as dealing with “property” which the defendant knows or has reasonable grounds to believe represents the proceeds of an indictable offence. The quality of the property being such proceeds is therefore an element in the mens rea but not the actus reus. It is nevertheless a necessary ingredient of the offence that the characteristics of the property known to the defendant or giving reasonable grounds for belief must be such as to qualify the property in law as the “proceeds of an indictable offence” as an element of the mens rea. The discussion which follows must be understood to proceed on that basis. It must moreover be borne in mind that in the present appeal, the underlying offence relied on by the prosecution is a conspiracy to commit money laundering and not the substantive offence. 18. The concept of dealing with “proceeds of an indictable offence” envisages two separate transactions. First, that an underlying indictable offence (often referred to as “the predicate offence”) has been committed. Secondly, that there has been an act amounting to “dealing” with the proceeds of that predicate offence. It is possible that the predicate offender and the dealer are the same person. But often, they will be different persons. In any case, at the time when the money or property is dealt with, it must have the prescribed characteristic of representing the proceeds of the predicate offence. 19. As noted previously, the prosecution relies on section 2(6)(a)(i) to contend that the concept of “proceeds of an indictable offence” extends to cover the untainted money advanced by Upbest which Tsang was intended to deal with. Section 2(6)(a) specifies certain characteristics which operate to qualify particular property as a person’s proceeds of an indictable offence. If property is received as a payment or other reward in connection with the commission of the predicate offence, such property constitutes in law the recipient’s proceeds of that offence. It may be noted that the recipient may, but does not have to be the predicate offender. But in every case where section 2(6)(a)(i) is relied on, the property in question must have the characteristic of a payment or other reward received in connection with commission of the predicate offence. 20. If those criteria are met, the dealer who subsequently deals with the property with the requisite mens rea is deemed to be dealing with property which he knows or has reasonable grounds to believe represents the proceeds of the relevant indictable offence and the “proceeds” ingredient is established. D.2 The ordinary meaning of “proceeds of an indictable offence” 21. Leaving aside section 2(6)(a) for the moment, if one were to give the word “proceeds” its ordinary meaning in the phrase “represents any person’s proceeds of an indictable offence” in section 25(1), such proceeds would be taken to mean money or property which is derived from or results from the commission of the relevant indictable offence. 22. This corresponds with the word’s dictionary meaning. Thus, the Oxford English Dictionary defines “proceeds” as: “That which proceeds, is derived, or results from something; that which is obtained or gained by any transaction; produce; outcome; profit”. 23. That must be the starting-point in the process of construction. It is against that background that one must scrutinise the prosecution’s contention that “clean” money which does not proceed from, is not derived from and is not the result of, some person’s commission of an indictable offence is nonetheless to be treated as representing the proceeds of an indictable offence. D.3 OSCO section 25A 24. While the word “proceeds” in its ordinary meaning requires the property dealt with to be something which is generated by the predicate offence, the prosecution submits that section 2(6)(a) removes that limiting requirement. All that is required (so the prosecution argues) is that the property should be a payment received “in connection with” the predicate offence. It contends that the words “in connection with” cast a net that is sufficiently wide to catch “clean” money which is intended to be used as an instrument in furtherance of a conspiracy to defraud. 25. In our view, section 25A(1) of OSCO suggests that the prosecution’s wide interpretation is wrong. Section 25A imposes a duty to report suspected money laundering offences in the following terms: Section 25A(1) “Where a person knows or suspects that any property- (a) in whole or in part directly or indirectly represents any person's proceeds of; (b) was used in connection with; or (c) is intended to be used in connection with, an indictable offence, he shall as soon as it is reasonable for him to do so disclose that knowledge or suspicion, together with any matter on which that knowledge or suspicion is based, to an authorized officer.” 26. Section 2(6)(a) applies “for the purposes of this Ordinance” and therefore applies as much to section 25A(1) as it does to section 25(1). If, as the prosecution seeks to argue, section 2(6)(a) has the effect of widening the meaning of the phrase “any person’s proceeds of an indictable offence” in section 25(1) to embrace payments which are not “proceeds” as ordinarily understood, but which (like the Upbest advance) are payments used or intended to be used in connection with the commission of an indictable offence, section 2(6)(a) ought to have the same effect in relation to the identical phrase which is contained in section 25A(1)(a). 27. But the legislature has found it necessary to go on to make express provision in paragraphs (b) and (c) of section 25A(1) to cover property which “was used in connection with” or which “is intended to be used in connection with” an indictable offence. If the prosecution’s construction of section 2(6)(a) were correct, paragraphs (b) and (c) in section 25A(1) would be redundant. On its argument, by reading section 25(1) together with section 2(6)(a), section 25A(1)(a) would already cover the content of paragraphs (b) and (c). 28. Section 25A(1) therefore provides a clear indication that it is not the legislative intent to give section 2(6)(a) the wide interpretation contended for by the prosecution. Unfortunately, the significance of section 25A(1) was not brought home to the Court of Appeal which expressed the view that section 25A is of no relevance to this case.[12] D.4 “Payments or other rewards” 29. As noted above, section 2(6)(a) is concerned with specifying certain characteristics which operate to qualify particular property as a person’s proceeds of an indictable offence – a necessary mens rea ingredient of a section 25(1) offence. It provides that “a person’s proceeds of an offence are … any payments or other rewards received by him at any time … in connection with the commission of that offence”. 30. If section 2(6)(a) is to apply, the payment must be envisaged to be received “in connection with the commission of” the predicate offence. But, contrary to the prosecution’s submission, the section requires more than just some general, unspecified connection between the payment and the offence. Due weight must be given to the words “or other rewards” in the phrase “payments or other rewards”. As a matter of language, those words make it clear that the contemplated payment must be in the nature of a reward. It must in particular be a payment having the quality of a reward received in connection with commission of the relevant offence. 31. In R v Allpress, Toulson LJ, giving the judgment of the English Court of Appeal sitting as a panel of five judges, stressed the need to have regard to the whole phrase and its effect as a matter of language: “The phrase ‘payment or other reward’ implies that the payment must be in the nature of a reward in order to fall within the relevant section, and that is consistent with the rationale of the confiscation scheme as explained by the House of Lords. The final sentence of [17] in May would make no sense if physical receipt of a sum of cash by D constitutes ipso facto the receipt of a payment or other reward, whether payment is for himself or not.”[13] (Emphasis in original) 32. A relationship of “reward” linking the payment and the commission of the offence must therefore exist before the payment qualifies as “proceeds of an indictable offence” under section 2(6)(a). For a payment to have this character, it has, for instance, to be a recompense or return or remuneration or incentive for or in connection with doing or having done something.[14] In the context of section 2(6)(a), that “something” is the commission of the predicate offence. It follows that the payment must derive from or be generated by, or be received on account of, commission of the offence. Money so received is tainted by its criminal origins and unsurprisingly treated as the proceeds of an indictable offence. 33. Section 2(6)(a) therefore only applies to tainted payments. “Clean money” which is not paid or received in the nature of a reward in connection with commission of the predicate offence does not qualify as the proceeds of such an offence for the purposes of section 25(1). So construed, section 2(6)(a) does not assist the prosecution. The sum of $32 million advanced by Upbest does not qualify as the “proceeds” of the Charge 3 conspiracy either on the ordinary meaning of the word or by virtue of section 2(6)(a). It was “clean money” advanced by Upbest intended to be used to deceive the shareholders and regulators. Without section 2(6)(a), the prosecution’s case on Charge 4 fails. 34. The foregoing argument was made in the Court of Appeal but rejected on the basis, as Stock VP put it, that it “sits ill with the decision of the Court of Appeal of England and Wales in R v Gertrude Osei[15] and of this Court in R v Lo Chak Man and another”.[16] We shall return to deal more fully with those cases later.[17] For present purposes, it suffices to note that, citing Glidewell LJ in Osei,[18] the Court of Appeal adopted the view that the word “payment” in the relevant phrase meant “any payment” including a payment which is not in the nature of a reward but “which may be in some other way in connection with” the predicate offence (drug trafficking in that case).[19] Lo Chak Man was cited as a case decided in our Court of Appeal which endorsed the Osei approach.[20] In order to place the Court of Appeal’s view based on those two cases in its proper context, the purposive construction of the relevant provisions advanced by Mr David Perry QC[21] should first be considered. D.5 A purposive construction of the OSCO provisions 35. Mr Perry QC submits that the fundamental purpose of OSCO is to deprive drug traffickers, organized gangsters and other serious offenders of their ill-gotten gains. It does so by giving the courts power to make confiscation orders stripping such offenders of the economic benefits of their crimes and by making it difficult for them to get help in dealing with and concealing those gains by criminalising money laundering activities. To those ends, OSCO focusses on the “proceeds of an indictable offence” as representing such ill-gotten gains. The Ordinance empowers the courts to make confiscation orders depriving the predicate offender of the proceeds of his predicate offence and depriving those who subsequently deal with those gains of their respective proceeds to the extent to which they may have benefited from their money laundering activities. 36. Mr Perry submits that when section 25(1) and section 2(6)(a) are construed in the light of that purpose, the inevitable conclusion is that the word “proceeds”, both for the purpose of establishing the section 25(1) offence and of making confiscation orders, has to be construed as referring only to money or property which represents an economic benefit gained by the relevant defendant in connection with the commission of the predicate offence. The entire scheme of the Ordinance is aimed at depriving defendants of such ill-gotten benefits and criminalising the laundering of those gains. He argues that the $32 million Upbest advance clearly was not a benefit received in connection with the Charge 3 conspiracy as it passed through the respective accounts of Logistic China, Ka Fong and Tsang. The circular payments were merely the instrument by which the conspiracy to deceive was carried out and were never intended to benefit or to confer any beneficial interest in the funds on the relevant appellants (or other participants in the circular flow). 37. In HKSAR v Lam Kwong Wai[22] and HKSAR v Cheung Kwun Yin,[23] the Court affirmed the need generally to interpret statutory language in the light of its context and purpose, and not only when an ambiguity may be thought to arise. And as Li CJ pointed out,[24] the context of a statutory provision is to be taken in its widest sense and certainly includes the other provisions of the statute and the existing state of the law. His Lordship also noted that: “The purpose of a statutory provision may be evident from the provision itself. Where the legislation in question implements the recommendations of a report, such as a Law Reform Commission report, the report may be referred to in order to identify the purpose of the legislation. The purpose of the statutory provision may be ascertained from the Explanatory Memorandum to the bill. Similarly, a statement made by the responsible official of the Government in relation to the bill in the Legislative Council may also be used to this end.”[25] 38. Adopting that approach, in our view, the purposive interpretation advanced by Mr Perry is made good. It is supported by the other provisions of OSCO (especially those concerning confiscation orders). It also receives support from the context in which OSCO came to be enacted, namely from the United Kingdom legislation upon which OSCO was modelled and decisions in the House of Lords and United Kingdom Supreme Court on materially comparable provisions. D.5a The purpose appearing from the other provisions of OSCO 39. The Long Title demonstrates that a central purpose of OSCO is to target the proceeds of crime. OSCO is: “An Ordinance to create new powers of investigation into organized crimes and certain other offences and into the proceeds of crime of certain offenders; provide for the confiscation of proceeds of crime; make provision in respect of the sentencing of certain offenders; create offences relating to the proceeds of crime or property representing the proceeds of crime; and for ancillary and connected matters.” 40. The provisions governing confiscation orders are aimed at depriving a defendant of his proceeds of the relevant offence to the extent that he has benefited from such offence. Such an order can be made against a defendant who has been convicted of and is to be sentenced for a specified offence.[26] But before the power can be exercised, section 8(4) requires the court first to determine that he “has benefited from the specified offence” and that “his proceeds of that specified offence or offences are in total at least $100,000.” Benefit is therefore fundamental to the confiscation jurisdiction. If satisfied as to such benefit, the court proceeds to determine the amount to be recovered from that defendant.[27] Section 11(1) prescribes as the recoverable amount “the amount [the court] ... assesses to be the value of the defendant’s proceeds of the specified offence to which section 8(4) applies.” The court may order[28] the defendant to pay that assessed amount or, if it is satisfied that the amount which can be realised is less than the assessed value of the defendant’s proceeds, a lesser sum.[29] The power is therefore aimed at stripping the defendant of his proceeds of the offence to the extent to which he has benefited from it, within his realisable means. 41. The statutory purpose of targeting persons to the extent that they have benefited from crime is also evident from the ancillary powers conferred in aid of the confiscation jurisdiction. Thus, special powers of investigation including the power to order material to be produced;[30] to issue certain search warrants;[31] and to make restraint and charging orders[32] are all subject to the condition that “there are reasonable grounds for suspecting that the person has benefited from organized crime or [a particular] specified offence”. Similarly, special procedures have been laid down to help to determine whether a defendant has benefited from the crime and the value of his proceeds, including statutory assumptions that the court is permitted to make[33] and a procedure for the prosecutor to tender statements as a means of establishing such benefit and value.[34] 42. Section 25(1) does not directly refer to “benefit” as an element of the money laundering offence. However (as we have seen), read together with section 2(6)(a), receipt of a payment or other reward – plainly a form of economic benefit – in connection with commission of an indictable offence qualifies such payment or other reward as the “proceeds of an indictable offence” for the purposes of section 25(1). That the legislative intent is that such proceeds should reflect the “benefit” gained by the defendant from the offence is put beyond doubt by section 2(8) which provides: “For the purposes of this Ordinance, a person who has ... received any payment or other reward in connection with the commission of an offence or an organized crime has benefited from that offence or organized crime, as the case may be.” 43. Thus, by the combined effect of sections 2(6)(a) and 2(8), the concept of “proceeds of an indictable offence” is equated with the receipt of “payments or other rewards” in connection with a predicate offence and the latter expression is deemed a benefit to the recipient for the purposes of the entire Ordinance, including sections 8 (dealing with confiscation) and 25 (constituting the offence). Benefit is therefore of basic importance in both contexts. 44. The purposive interpretation is accordingly supported by the confiscation provisions and the overall scheme of OSCO: payments received in the circular flow of funds which were never intended to benefit the appellants but were merely instrumentalities of the conspiracy to defraud do not qualify as proceeds of an indictable offence to constitute the Charge 4 conspiracy. D.5b The United Kingdom legislation 45. In the written cases lodged on behalf of the prosecution and of Cheng, the legislative history of OSCO section 25(1) and section 2(6)(a) is traced in detail. Its local precursors involved an Ordinance targeting the proceeds of drug trafficking,[35] later supplemented by an Ordinance targeting the proceeds of organized crime.[36] We arrived at the current edition of OSCO after amendments enacted in 1995.[37] OSCO is modelled on United Kingdom legislation, especially the Drug Trafficking Offences Act 1986 (“DTOA 1986”) which created a money laundering offence in relation to drug trafficking and conferred powers to make confiscation orders and is an Act which (as later amended[38]) continues to be reflected in the current provisions of OSCO. 46. Thus, section 8(4) of OSCO which requires the court when considering whether to make a confiscation order first to determine that the defendant “has benefited from the specified offence” and that “his proceeds of that specified offence or offences are in total at least $100,000”, was modelled on section 1(2) of DTOA 1986 which had made such benefit a necessary and basic condition.[39] Section 2(1)(a) of DTOA 1986 equates “payments or other rewards” received in connection with drug trafficking as the person’s “proceeds” of that offence and is reflected in OSCO’s section 2(6)(a).[40] OSCO’s section 2(8) (which links “benefit” and “proceeds” for the purposes of both confiscation and the offence) was copied from section 1(3) of DTOA 1986.[41] The Act also prescribes as the amount recoverable on a confiscation order, the value of the defendant’s proceeds of drug trafficking.[42] And OSCO’s ancillary provisions in aid of investigation and proof of benefit and value of proceeds, and so forth, are all modelled on similar provisions in DTOA 1986.[43] 47. It is therefore clear that OSCO’s fundamental requirement that “proceeds of an indictable offence” should involve “benefit” to the defendant as discussed above[44] derives from the DTOA 1986 scheme. That Act was subsequently amended, and in important respects, Hong Kong followed suit. For instance, the money laundering offence as originally enacted in DTOA 1986, section 24 involved an offence of assisting another to retain the proceeds of drug trafficking, reproduced in section 25 of the 1994 edition of OSCO. When the offence was progressively refined and extended in the United Kingdom,[45] Hong Kong followed suit, resulting in the current section 25(1) offence. 48. Subsequent amending legislation in the UK retained the feature of benefit to the defendant as a central requirement in the confiscation scheme which continued to be linked through the concept of “proceeds” to the money laundering offence.[46] 49. The significance of the United Kingdom legislation for present purposes is that it accords basic importance to the element of benefit to the defendant and links such benefit to the concept of “proceeds of an indictable offence” in relation to both the offence and confiscation orders. Modelling itself on such legislation, OSCO has done the same. Of course there are differences in the various generations of the UK legislation but, regarding the “benefit” feature, OSCO is closely comparable to the United Kingdom legislation, making it helpful to examine some of the key judicial decisions concerning the place of “benefit” in the statutory scheme as an aspect of the purposive interpretation under discussion. D.5c The United Kingdom cases 50. The relevant United Kingdom cases demonstrate the importance of “benefit to the defendant” by treating that requirement not only as essential but as one which limits the scope of confiscation orders which can be made. They do not deal directly with the elements of the money laundering offence but, as previously explained, the applicable legislation both in the UK and Hong Kong establishes a link between “benefit” and “proceeds” in relation to both confiscation and the offence, making the UK case-law on confiscation helpful in the present discussion. (i) R v May[47] 51. The leading decision is that of the House of Lords in R v May, where the facts (involving a VAT “carousel” or “missing trader” fraud) are not presently relevant. However, a valuable analysis and statement of principle are to be found in the speech of Lord Bingham of Cornhill.[48] Having referred to a series of statutes directed at confiscation of the proceeds of crime, his Lordship noted that: “...despite much refinement and differences between the 1986 and 1994 Acts on the one hand and the 1988, 1993 and 1995 Acts on the other, the essential structure of the 1986 regime has been retained.”[49] 52. His Lordship explained that that regime (upon which OSCO is based): “...requires the court, before making a confiscation order, to address and answer three questions: ... The first question is: has the defendant (D) benefited from the relevant criminal conduct? If the answer to that question is negative, the inquiry ends. If the answer is positive, the second question is: what is the value of the benefit D has so obtained? The third question is: what sum is recoverable from D?”[50] 53. Prominence was accordingly given to the need for the defendant to have benefited from the offence and for the value of that benefit to be assessed. Elaborating upon the first of his three questions, Lord Bingham surveyed certain decisions where the lower courts had made confiscation orders in amounts which went beyond the actual benefit gained by the defendant and reflected instead the much greater value of property dealt with in the criminal activity in question. 54. His Lordship provided some examples. The first was R v Osei,[51] which involved “a drug courier armed, it seems, with a sum of cash to enable her to show that she could support herself in order to enter the country”, where the court held that “payment” in the statutory formula “payment or other reward” was “apt to cover not merely a profit or fee but also a payment of this kind”.[52] This was the decision relied on by the Court of Appeal below to which we shall return.[53] 55. Lord Bingham pointed out that Osei had been applied in R v Smith (Ian),[54] “where ‘any payments’ were held to mean any payment in money or in kind, not net profit after deduction of expenses.”[55] 56. His Lordship went on next to consider R v Simons,[56] which involved a middle-man who had bought and resold consignments of drugs. Applying R v Smith (Ian), the Court of Appeal held that the confiscation order should not be confined to his profit but should be in the amount of the sale price, commenting: “... where you have a chain of contracts, each purchase price is a ‘payment’. That result cannot be avoided simply by the entrepreneur or middle man ensuring that the contractual situation was that he merely passed on as postman a payment from the consignee to the consignor and that what in the ordinary course of events would have been his profit was paid to him as a handling fee or something of that sort. As Parliament deprives drug dealers of the proceeds of their sale, namely the payments made to them and not profits made from them, there is no reason why they should treat conduits any differently.”[57] 57. Stressing the importance of the defendant obtaining a benefit, Lord Bingham described the approach in Simons (and by implication in Smith (Ian) and Osei)as problematical, stating: “The court went on to observe, obiter (p 104), that this result could not be avoided by treating the intermediary as a postman, and those acting as a conduit should not be treated differently. But this, with respect, is more problematical: under the 1986 Act the first question was always whether, on the facts (and allowing permissible inferences) the defendant had benefited by receipt of any payment or other reward, which a mere intermediary might possibly not. It does not necessarily follow from the mere possession of drugs that a person is not a mere minder or custodian...”[58] 58. In an “Endnote”, Lord Bingham summarised the broad principles applicable in confiscation cases. The first principle supports the purposive interpretation being discussed as it identifies the object of the legislation in terms of stripping away the benefit gained by defendants: “(1) The legislation is intended to deprive defendants of the benefit they have gained from relevant criminal conduct, whether or not they have retained such benefit, within the limits of their available means. It does not provide for confiscation in the sense understood by schoolchildren and others, but nor does it operate by way of fine. The benefit gained is the total value of the property or advantage obtained, not the defendant's net profit after deduction of expenses or any amounts payable to co-conspirators.”[59] 59. After having re-iterated the three questions (two of which relate to the existence and value of benefit to the defendant) mentioned above and having made certain other points, Lord Bingham added: “(5) In determining, under the 2002 Act, whether D has obtained property or a pecuniary advantage and, if so, the value of any property or advantage so obtained, the court should (subject to any relevant statutory definition) apply ordinary common law principles to the facts as found. The exercise of this jurisdiction involves no departure from familiar rules governing entitlement and ownership. ... (6) D ordinarily obtains property if in law he owns it, whether alone or jointly, which will ordinarily connote a power of disposition or control, as where a person directs a payment or conveyance of property to someone else. He ordinarily obtains a pecuniary advantage if (among other things) he evades a liability to which he is personally subject. Mere couriers or custodians or other very minor contributors to an offence, rewarded by a specific fee and having no interest in the property or the proceeds of sale, are unlikely to be found to have obtained that property. It may be otherwise with money launderers.”[60] (ii) Jennings v Crown Prosecution Service[61] 60. The decision in Jennings v CPS was published on the same day as R v May. It arose in connection with the making of a restraint order in aid of confiscation proceedings where the central issue concerned the construction of section 71(4) of the Criminal Justice Act 1988. That section materially provides: “... a person benefits from an offence if he obtains property as a result of or in connection with its commission and his benefit is the value of the property so obtained”. 61. The focus was therefore again on the benefit gained by a defendant as a requirement of making a confiscation order. Jennings had been convicted on a charge of conspiracy to defraud. The fraud had involved the extraction of “advance fees” from victims totalling over £584,000 and the prosecution argued that that amount should be treated as the property obtained by the defendant and adopted in the confiscation order. His case was that he had at most obtained £50,000 by way of salary and a few minor payments. The Court of Appeal had ruled in favour of the prosecution’s approach holding that it was only necessary to show “that the defendant's acts should have contributed, to a non-trivial (that is, not de minimis) extent, to the getting of the property”, viewing it as essentially a question of causation. Lord Bingham, again giving the opinion of the Committee, considered this erroneous and stressed the importance of bearing in mind the purpose of the legislation (which his Lordship held to be the same as the purpose of the drug trafficking legislation): “It is, however, relevant to remember that the object of the legislation is to deprive the defendant of the product of his crime or its equivalent, not to operate by way of fine. The rationale of the confiscation regime is that the defendant is deprived of what he has gained or its equivalent. He cannot, and should not, be deprived of what he has never obtained or its equivalent, because that is a fine. This must ordinarily mean that he has obtained property so as to own it, whether alone or jointly, which will ordinarily connote a power of disposition or control, as where a person directs a payment or conveyance of property to someone else.”[62] 62. Lord Bingham pointed out that someone’s acts “may contribute significantly to property ... being obtained without his obtaining it” and emphasised that under section 71(4), “a person benefits from an offence if he obtains property as a result of or in connection with its commission, and his benefit is the value of the property so obtained, which must be read as meaning ‘obtained by him’.”[63] Accordingly, the obtaining had to involve the gaining of a benefit and not just participation in the fraud. Such benefit was not only essential to making a confiscation order, the extent of such benefit limited the amount of the confiscation order. Not being in the nature of a fine, such an order had to be limited to depriving the defendant of the benefit he had gained from commission of the offence. The Committee held that the Court of Appeal’s approach was wrong but dismissed the appeal because the available material justified the making of a restraint order in any event. (iii) R v Waya[64] 63. The essential legislative purpose of stripping away the benefit gained from the commission of crime was also stressed in R v Waya,a decision of a panel of nine judges in the UK Supreme Court. Giving the judgment of the majority, Lord Walker of Gestingthorpe JSC and Hughes LJ stated: “The purpose of the legislation is plainly, and has repeatedly been held to be, to impose upon convicted defendants a severe regime for removing from them their proceeds of crime. It is not to be doubted that this severe regime goes further than the schoolboy concept of confiscation, as Lord Bingham explained in R v May [2008] AC 1028 . Nor is it to be doubted that the severity of the regime will have a deterrent effect on at least some would-be criminals. It does not, however, follow that its deterrent qualities represent the essence (or the ‘grain’) of the legislation. They are, no doubt, an incident of it, but they are not its essence. Its essence, and its frequently declared purpose, is to remove from criminals the pecuniary proceeds of their crime.”[65] 64. Their Lordships added: “...the scheme of the Act, and of previous confiscation legislation, is to focus on the value of the defendant's obtained proceeds of crime, whether retained or not. It is an important part of the scheme that even if the proceeds have been spent, a confiscation order up to the value of the proceeds will follow against legitimately acquired assets to the extent that they are available for realisation.”[66] (iv) R v Mackle[67] 65. The benefit theme is also prominent in R v Mackle, a case where the defendants had been convicted on their own pleas for evading duty and value added tax leading to confiscation orders against them on the basis that they had obtained a pecuniary advantage by evading such tax and duty. The orders were made in amounts equal to aggregate duty and tax allegedly evaded. It was however subsequently discovered that they had not in fact been liable to pay such tax and duty. They had therefore not evaded such liability and the pecuniary advantage relied on as the benefit they had obtained did not exist. 66. Giving the opinion with which the other members of the Court agreed, and after referring to the principles set out in R v May,Lord Kerr of Tonaghmore JSC stated: “The focus must be, as Lord Bingham has said, on what benefit the defendant has actually gained. Simply because someone has embarked on a joint criminal enterprise, it does not follow that they have obtained an actual benefit. Being engaged in a conspiracy does not, of itself, establish that each conspirator has obtained the property which is the product of the conspiracy.”[68] D.5d Applying the purposive interpretation 67. In our view, given that the United Kingdom’s confiscation regime employs closely comparable concepts, the abovementioned decisions of the House of Lords and Supreme Court provide persuasive and helpful guidance in the purposive construction of the relevant concepts in OSCO and in particular, the interpretation of the phrase “proceeds of an indictable offence” both for the purposes of section 25(1) and the confiscation regime. 68. The central proposition of the purposive interpretation is that, in line with the pervasive object of stripping away the economic benefits obtained by defendant in connection with the commission of the predicate offence, property ought not to be held to be a particular defendant’s “proceeds” unless that defendant has gained an economic benefit from such property. 69. As we have been at pains to point out, the section 25(1) offence is committed where a person deals with property knowing or having reasonable grounds to believe that the same represents the proceeds of the predicate offence. The predicate offender is benefited by obtaining those proceeds. The property he obtains are his payment or other reward received in connection with his commission of the predicate offence. But, as we have seen, a further aspect of “benefit” arises in the context of the confiscation regime as applied to persons who subsequently deal with the proceeds of the predicate offence. Before a confiscation order can be made against such persons, they must also be shown to have benefited from those proceeds as required by section 8(4). Such benefit need not be shown to prove liability under section 25(1) but is a condition of making a confiscation order. The following example illustrates this: (a) X is a drug trafficker who has just sold a consignment of drugs to a buyer and received a suitcase containing $3 million in banknotes in payment. (b) X asks his friend Y (who knows where the money comes from) to deliver the suitcase to Z, a professional money launderer. Y receives nothing for doing this, merely acting because of his friendship with X. (c) Z receives the money and is allowed to keep $100,000 as his fee after having dealt with the balance of the $3 million on X’s behalf in ways designed to conceal its criminal origins. (d) X commits the predicate offence of drug trafficking and the $3 million constitutes its proceeds with X obviously benefiting to the extent of $3 million. A confiscation order in the amount of $3 million can in principle be made against X. (e) Y commits a section 25(1) offence since he deals with the $3 million which he knows or has reasonable grounds to believe represent X’s proceeds of the predicate drug trafficking offence. However, since Y has not himself benefited, no part of the $3 million constitutes “his proceeds” of the predicate offence and the section 8(4)(a) requirement of a benefit to Y is not satisfied. No confiscation order can be made against Y although he can be otherwise sentenced[69] for the section 25(1) offence. (f) Z commits a section 25(1) offence since he deals with the $3 million which he knows or has reasonable grounds to believe represent X’s proceeds of the predicate drug trafficking offence. His benefit from those proceeds is his fee of $100,000 and a confiscation order can be made against him to that extent. He can of course also be otherwise sentenced for the section 25(1) offence. (g) We might add that a confiscation order can be made against X in respect of the $3 million proceeds, permitting restraint and receivership orders, etc, to be made to secure his property for eventual confiscation by invoking the ancillary provisions of OSCO where, for instance, the money is intercepted in Y’s or Z’s possession but X has absconded. D.6 The Court of Appeal’s construction 70. As we previously noted,[70] the Court of Appeal rejected the argument that “payments” can only qualify by virtue of section 2(6)(a) as “proceeds of an indictable offence” if they are in the nature of a reward because the section refers to such payments as “payments or other rewards”. Stock VP considered that argument to be inconsistent with the decisions of the English Court of Appeal in R v Gertrude Osei[71] and the Hong Kong Court of Appeal in R v Lo Chak Man”[72] which had adopted the Osei approach.[73] He also took the view that the decisions in R v Allpress[74] and R v May[75] do not militate against that conclusion. There appear to be two main strands in the reasoning of the Court of Appeal. 71. First, their Lordships agreed with Glidewell LJ’s view in Osei that the word “payment” in the relevant phrase meant “any payment” including a payment which is not in the nature of a reward but one which “may be in some other way in connection with” the predicate offence.[76] This involves, as Mr Perry QC puts it, the uncoupling of “payment” on the one hand and “reward” or “benefit” on the other. The Court of Appeal held that “the crediting of the funds passing though the accounts of Ka Fong and Tsang and then on through the accounts of the other companies in the convoluted chain are properly classified as ‘payments’”[77] so that under section 2(6)(a) they were to be treated as the appellants’ proceeds received in connection with the predicate Charge 3 conspiracy. 72. We are unable to agree with that reasoning. Osei[78] preceded the House of Lords’ decisions in R v May[79]and Jennings v CPS[80]as well asthe later decisions of the UK Supreme Courtexamined above.[81] In the light of the importance that they attach to the requirement of benefit to the defendant, both as a condition of making confiscation orders and as an assessed value limiting the amount of such orders, we do not think that Glidewell LJ’s uncoupling of payment from reward or benefit is consistent with those decisions. 73. R v Osei, it will be recalled, involved a drug courier who pleaded guilty to trafficking and was found to have £2,500 in her possession. The question was whether that sum should be treated as her proceeds of the trafficking offence and taken into the valuation for the purposes of a confiscation order. She had told inconsistent stories about the money, one version being that it had been given to her by a stranger on the plane so that she would have enough money to get her through Immigration. It was said that she herself had stood to make only £1,000 to £1,500 for acting as courier. It is unfortunately unclear what if any findings were made in Osei as to whether the whole or part of the £2,500 sum represented a reward for acting as courier, but it certainly appears likely that her benefit did not extend to the whole of that amount. To the extent that the trial judge’s decision to order confiscation of the whole of the £2,500 was upheld by the Court of Appeal irrespective of any benefit to the courier, that judgment is inconsistent with the later House of Lords and UK Supreme Court decisions. The Court of Appeal relied on R v Lo Chak Man[82] because of its adoption of the Osei approach.[83] It does not call for separate treatment. 74. As indicated above,[84] our view is that on the true construction of section 2(6)(a), the payments in question must be “in the nature of a reward” received in connection with the commission of the predicate offence. The “uncoupling” approach should in any event not be adopted in Hong Kong. 75. The second strand of the Court of Appeal’s reasoning for upholding the Charge 4 convictions involved taking note of the “reward” requirement. Stock VP cited Toulson LJ’s statement in R v Allpress[85] that “The phrase ‘payment or other reward’ implies that the payment must be in the nature of a reward in order to fall within the relevant section” but did not consider it necessary to decide whether that cast doubt on Lo Chak Man.[86] 76. His Lordship cited the following passage from R v May:[87] “D ordinarily obtains property if in law he owns it … which will ordinarily connote a power of disposition or control, as where a person directs a payment or conveyance of property to someone else. … Mere couriers or custodians or other very minor contributors to an offence, rewarded by a specific fee and having no interest in the property of the proceeds of sale, are unlikely to be found to have obtained that property. It may be otherwise with money launderers.” 77. It was suggested that Lord Bingham had there stated that “It may be otherwise with money launderers”, because “money launderers are likely, on the facts, to enjoy a power in law of disposition of control”.[88] This, the Court of Appeal concluded, applies in the present case: “Mr Marash SC for the respondent submits, correctly in our view, that Tsang was not a mere nominee or trustee of the funds he received; he possessed a chose in action whilst the sum of $32 million was in his account. Whether this is classified as a payment or a pecuniary advantage matters not since the latter is also a person’s proceeds as defined by section 2(6) of the Ordinance.”[89] 78. In other words, the Court of Appeal was suggesting that insofar as benefit to the appellants had to be shown, this was achieved on the facts of this case. This was because upon receiving payment from Ka Fong, Tsang “possessed a chose in action whilst the sum of $32 million was in his account”; Tsang thereby obtained a power of disposition or control and was not a mere nominee or trustee of those funds. 79. We do not accept that second strand of reasoning. It is, with respect, entirely contrary to the evidence to suggest that the $32 million received in Tsang’s account from Ka Fong was a genuine benefit received by him. It is of course true that for a very short period of time, as the funds washed around the circular flow of payments, a chose in action arose as a matter of law representing a debt owed by the bank to Tsang. It disappeared once the funds were passed along the chain of payments. The very essence of the Charge 3 conspiracy was that the payments were a sham. It was a conspiracy to defraud the Stock Exchange and the shareholders of Grand Field by dishonestly concealing the absence of any genuine acquisition of an interest in the Mainland joint venture and pretending to have effected a disposal of that interest. There was no question of the payment conferring upon Tsang a genuine power of disposition or control over the funds which briefly transited his bank account. It was essential to the conspiracy that the funds belonged to Upbest and that they would complete the circle and return to Upbest after having effected the deception. The whole case is premised on the payment of $32 million being an instrument used in furtherance of the conspiracy, which is fundamentally inconsistent with Tsang personally benefiting from the receipt of that amount. 80. In the prosecution’s written case,[90] the following submission is made, seeking to establish benefit to Tsang on the basis of his having obtained a new form of a pecuniary advantage: “ ... the Court of Appeal’s conclusion can also be supported on the narrower basis of evading a personal liability. ... In receiving the $32 million in his personal bank account as part of the money circle, A3 (Tsang) obtained a pecuniary advantage by evading potential civil liability to shareholders resulting from the conspiracy to defraud offences particularised in Charges 1 and 3. ... It is unnecessary to quantify this advantage for purposes of money laundering liability. It is submitted that the $32 million received as a means to evade such liability must, at least in part, directly or indirectly represent the pecuniary advantage obtained.” 81. This argument is not open to the prosecution as it involves fact-sensitive issues which were not raised or explored at the trial. We would comment in passing that it is in any event a highly tenuous submission. It is not suggested that there was any existing civil liability but only some unparticularised potential liability to shareholders, apparently incapable of quantification, which was allegedly evaded. It is highly questionable, whether, if the matters alleged had been explored, they would have been capable of constituting a pecuniary advantage. D.7 Other statutory formulations 82. A final point we would make on construction is that where the legislature has sought to regulate dealings with property used or intended to be used as an instrument of or in furtherance of crime, as distinct from property representing the proceeds of crime, it has done so in unequivocal terms. We have already seen an example of this in section 25A(1) of OSCO discussed above.[91] A further example can be found in the Anti-Money Laundering and Counter-Terrorist Financing (Financial Institution) Ordinance[92] which has a definition of money laundering referring to the proceeds obtained from commission of an indictable offence which is quite distinct from the definition of “terrorist financing” which refers to the provision or collection of property “with the intention that the property be used; or knowing that the property will be used ... to commit ... terrorist acts...” [93] Similarly, section 7 of the United Nations (Anti-Terrorism Measures) Ordinance[94] prohibits the provision or collection of any property “with the intention that the property be used; or ... knowing that the property will be used, in whole or in part, to commit one or more terrorist acts ...” E. Policy consequences 83. In our view, adoption of the wide interpretation proposed by the prosecution would be likely to have highly detrimental consequences. Money laundering would become an offence of great and uncertain width. Payments would constitute the offence without having to be payments which the defendant knows or has reasonable grounds to believe are in the nature of a reward received in connection with the commission of an offence. The prosecution would merely have to prove receipt of a payment and show reasonable grounds to believe that such payment has an undefined and uncertain “connection” with an indictable offence. In jurisdictions having legislation aimed at catching payments beyond those representing the proceeds of crime, definitions are laid down to identify the necessary connection, thus avoiding such uncertainty. For instance, in Australia,[95] the Commonwealth Criminal Code catches money or property which “will become an instrument of crime”. 84. Given the low threshold presented by the mental element in section 25(1) as held by the Court in Oei Hengky Wiryo v HKSAR (No 2),[96] the wide interpretation would place lenders at risk and impose onerous burdens on them. They would be at risk if, unknown to them, loan moneys were used by a customer as an instrument of the customer’s criminal enterprise. In a hypothetical case similar to the present, a customer might, for instance, seek a bridging loan plausibly explaining that it was to enable him to consummate a deal with repayment swiftly to follow. If it turns out that, unknown to the lender, the customer uses the funds in furtherance of a conspiracy to defraud, the lender could be at risk if it had missed the significance of aspects of the transaction which the prosecution might (with hindsight) allege were such as to give the lender reasonable grounds to believe that the money was intended to be used “in connection with” the hypothetical conspiracy, even though the lender knew full well that the funds it was lending were its own entirely legitimate funds unconnected with any crime. An onerous duty of due diligence would be imposed on lenders fearful of missing what might later be said to be clues about intended criminal conduct. It is one thing to criminalise dealing with funds where the dealer knows or has reasonable grounds to believe that they are the proceeds of crime, it is quite a different matter to stigmatise as a money launderer, a lender dealing with its own “clean” funds because of what the borrower does or intends to do with them. 85. As the appellants point out, the wide interpretation would be likely to distort the practice of prosecutors. They would understandably find it hard to resist the temptation of bringing money laundering charges made easy to prove whenever it was possible to prove a payment and some broad connection with some indictable offence, whatever might be the underlying criminality involved. Indeed, in the present case, for unknown reasons, Cheng and Li were not charged as parties to the Charge 3 conspiracy. That is somewhat surprising since, as the Court of Appeal pointed out, the prosecution appears in its closing submissions to have asserted that their participation in the Charge 3 conspiracy had been established on the evidence: “[Prosecuting counsel] said that although it was not alleged that Cheng and Li were parties to the Charge 3 conspiracy, ‘the prosecution case is that pursuant to their own agreement they knew or had reasonable grounds to believe that a sum of money to be advanced by Upbest was for the purpose of falsely representing to the regulator and investors that Grand Field had genuinely sold out their interest in the joint-venture to Logistic China. ... It is submitted the evidence is sufficient to establish that the defendants [by which he was referring to them all] believed the sum of $32 million was not to be a legitimate payment by Logistic China to Grand Field but was to be used to dishonestly represent that Grand Field had genuinely disposed of their interest in the joint venture. With that belief they agreed that [Tsang] would deal with the money in a manner designed to return the entire sum to Upbest.’”[97] 86. The trial Judge apparently agreed. As the Court of Appeal noted,[98] his Honour found that knowledge of the conspiracy to defraud was acquired by Cheng and Li at a meeting with Au-Yeung[99] and that Li “knew full well that the arrangement was for D1 to receive the money after it had been used to defraud the Stock Exchange and then dispose of it as arranged.”[100] F. Conclusion and disposal 87. We conclude that the Courts below were wrong to construe the expression “proceeds of an indictable offence” in s 25(1) of OSCO as extending to cover money used as an instrument furthering the Charge 3 conspiracy where there was no evidence that the appellants knew or had reasonable grounds to believe that such money represented the proceeds of an indictable offence. It was on the contrary clear that all concerned knew that the payments and receipts of the sum of $32 million relied upon by the prosecution involved dealing with funds provided by Upbest, there being no evidence that such funds were other than legitimate. It was erroneous to construe section 2(6)(a) as widening the concept of “proceeds of an indictable offence” to cover such payments. 88. The Charge 4 conspiracy was therefore in our view not established and we would accordingly allow the appellants’ appeal in respect of that charge and quash the appellants’ convictions thereunder. We would direct that any party wishing to make submissions as to costs, should lodge such submissions in writing with the Registrar within 21 days of the date of this judgment and that any submissions in reply should be lodged within 14 days thereafter. Mr Justice Tang PJ : 89. I agree with the joint judgment of Mr Justice Ribeiro and Mr Justice Fok PJJ. Mr Justice Bokhary NPJ : 90. These appeals succeed if the expression “proceeds of an indictable offence” in s.25(1) of the Organized and Serious Crimes Ordinance, Cap.455, is confined to what is derived from the commission of an indictable offence. They fail if the expression extends to resources used in the furtherance of its commission. On an ordinary reading, the word “proceeds” points to what is derived from rather than to what is used in the furtherance of. That understanding of the expression in question is fortified by the interpretation clause of the statute concerned. Section 2(6)(a)(i) thereof provides that a person’s proceeds of an offence extends to any “payments or other rewards” received by him at any time in connection with the commission of that offence. The words “or other rewards” show that the “payments” covered are ones in the nature of a reward. And that points to what is derived from rather than what is used in the furtherance of. 91. The foregoing is sufficient for the appellants’ purposes. But as it happens, their position is supported by the highly persuasive authority of the decisions of the House of Lords in R v May [2008] 1 AC 1028 and Jennings v Crown Prosecution Service [2008] 1 AC 1046 and of the United Kingdom Supreme Court in R v Waya [2013] 1 AC 294 and R v Mackle [2014] 2 WLR 267. 92. For the brief reasons which I have given and for the reasons stated more fully in the joint judgment of Mr Justice Ribeiro PJ and Mr Justice Fok PJ, with which joint judgment I agree, I would allow these appeals to quash the appellants’ convictions under Charge 4. I concur in the direction as to costs made in that joint judgment. And I conclude with an expression of my indebtedness to all counsel for their helpful and succinct arguments. Lord Collins of Mapesbury NPJ : 93. I agree with the joint judgment of Mr Justice Ribeiro and Mr Justice Fok PJJ. Mr Justice Ribeiro PJ : 94. The Court unanimously allows the appeal, quashes the appellants’ convictions in respect of Charge 4 and gives the directions as to the lodging of submissions regarding costs set out in the last paragraph of the joint judgment of Mr Justice Fok PJ and myself. Mr S.K. Khattak (Kelly Lam), instructed by Lennon & Lawyers, for the Appellant in FACC No.4 of 2013 Mr David Perry QC, Mr Michael Blanchflower SC, Ms Maggie Wong and Ms Tanie Toh, instructed by Michael Li & Co., for the Appellant in FACC No.5 of 2013 Mr Collingwood Thompson QC, Mr Andrew Bruce SC and Mr Felix CY Hoe, instructed by Wat & Co., for the Appellants in FACC No.6 of 2013 Mr Keith Yeung SC, DPP of the Department of Justice, Mr Simon NM Young on fiat for, and Mr Anthony Chau, SPP of, that Department, for the Respondent [1] Presently the 3rd and 4th appellants. [2] Now the 2nd and 1st appellants respectively. [3] Including the 3rd and 6th defendants with whom we are not concerned. [4] Cap 455. [5] DCCC No 24 of 2008 (1 March 2010). [6] Chongqing Wansheng Coal Carbonization Gas Company Limited. [7] Upbest paid out a total of $45,749,500 and received back $45,400,000. [8] The 6th defendant and Au Yeung. [9] Stock VP, Yeung VP and Lunn JA, CACC 96/2010 (28 November 2012), Stock VP giving the judgment of the Court. [10] Defined in section 2 to include “receiving or acquiring” and “disposing of” the property. [11] (2007) 10 HKCFAR 98 at §99, endorsing the Appeal Committee’s view in HKSAR v Wong Ping Shui (2001) 4 HKCFAR 29 at 31. [12] Court of Appeal §130. [13] [2009] 2 Cr App R (S) 58, p 399 at §58. The House of Lords decision in R v May referred to is dealt with in Section D.5c(i) of this judgment. [14] Or possibly for not doing or having not done something. [15] (1988) 10 Cr App R (S) 289. [16] Cr App 744 of 1995, 7 November 1996, unreported [17] Section D.6 below. [18] At p 293. [19] Court of Appeal §154. [20] Court of Appeal §155. [21] Appearing with Mr Michael Blanchflower SC, Ms Maggie Wong and Ms Tanie Toh on behalf of Cheng. [22] (2006) 9 HKCFAR 574 at 606. [23] (2009) 12 HKCFAR 568 at §12. [24] Cheung Kwun Yin at §13. [25] Ibid, §14. [26] Being one of the offences listed in Schedules 1 or 2 of OSCO: section 8(1)(a)(i). It is unnecessary to deal with the variation involving persons who have died or have absconded: section 8(1)(a)(ii). [27] OSCO section 8(6). [28] OSCO section 8(7). [29] OSCO section 11(3). [30] OSCO sections 4(2), 4(4)(b)(ii). [31] OSCO sections 5(1), 5(3)(b)(ii) and 5(4)(b)(ii). [32] OSCO sections 14(1)(c)(ii). [33] OSCO section 9(1) and (2). [34] OSCO sections 10(1)(b) and 10(3)(c). [35] Drug Trafficking (Recovery of Proceeds) Ordinance, enacted 13 July 1989. [36] Organized and Serious Crimes Ordinance, enacted 20 October 1994. [37] Organized and Serious Crimes (Amendment) Ordinance, enacted on 3 August 1995. The Drug Trafficking (Recovery of Proceeds) (Amendment) Ordinance was enacted on the same day. [38] By the Criminal Justice Act 1993 and the Proceeds of Crime Act 2002. [39] Section 1(2): “The court shall first determine whether he has benefited from drug trafficking”. [40] DTOA 1986, section 2(1)(a): “For the purposes of this Act any payments or other rewards received by a person at any time ... in connection with drug trafficking carried on by him or another are his proceeds of drug trafficking”. It goes on in paragraph (b) to say that the “value of his proceeds of drug trafficking is the aggregate of the values of the payments or other rewards”, as reflected in OSCO’s section 2(6)(b)(i). [41] DTOA 1986, section 1(3): “For the purposes of this Act, a person who has at any time ... received any payment or other reward in connection with drug trafficking carried on by him or another has benefited from drug trafficking”. [42] DTOA 1986, section 4(1). There is likewise a discretion to reduce that amount to what is realisable: section 4(3). [43] Including section 2(2)-(5) and section 3. [44] In Section D.5a. [45] For example the Criminal Justice Act 1993 introducing sections 93A et seq into the Criminal Justice Act 1988. [46] See for example the Criminal Justice Act 1988, sections 71(2)(b)(i), 71(4), 71(6)(a); the Criminal Justice Act 1993, sections 27(2), 28 and 29(2); the Drug Trafficking Act 1994, sections 2(2), 2(3), 2(4), 2(8), 3(1), 5, 6; and the Proceeds of Crime Act 2002 sections 76(4), 76(7), 340(5), 340(8), 340(10). [47] [2008] 1 AC 1028. [48] Giving the opinion of the Appellate Committee. [49] [2008] 1 AC 1028 at §8. [50] Ibid, authorities cited have been omitted. [51] (1988) 10 Cr App R (S) 289. [52] [2008] 1 AC 1028 at §15. [53] See Section D.6 below. [54] [1989] 1 WLR 765, 769; (1989) 89 Cr App R 235. [55] [2008] 1 AC 1028 at §15. [56] Ibid. [57] (1994) 98 Cr App R 100 at 104. [58] [2008] 1 AC 1028 at §15. [59] Ibid at p 1045. [60] Ibid. [61] [2008] 1 AC 1046. [62] At §13. [63] At §14. [64] [2013] 1 AC 294. [65] At §21. [66] At §27. [67] [2014] 2 WLR 267 (SC(NI)). [68] At §64. [69] Confiscation orders constitute sentences on defendants: OSCO section 8(8A). [70] Section D.4 above. [71] (1988) 10 Cr App R (S) 289. [72] Cr App 744 of 1995, 7 November 1996, unreported [73] Court of Appeal §153. [74] [2009] 2 Cr App R (S) 399. [75] [2008] 1 AC 1028. [76] Court of Appeal §154. [77] Court of Appeal §151. [78] Which concerned the making of a confiscation order under the DTOA 1986. [79] [2008] 1 AC 1028. [80] [2008] 1 AC 1046. [81] Section D.5c above. [82] Cr App 744 of 1995, 7 November 1996, unreported [83] Court of Appeal §155. [84] Section D.4. [85] [2009] 2 Cr App R (S) 58, p 399. [86] Court of Appeal §158. [87] [2008] 1 AC 1028 at §48(6), in the Endnote. [88] Court of Appeal §158, citing in support R v Allpress at §85. [89] Court of Appeal §159. [90] At §75. [91] Section D.3. [92] Cap 615. [93] Schedule 1, Part 1, section 1: “money laundering” and “terrorist financing”. [94] Cap 575. [95] Criminal Code Act 1995 (Cth) s 400.3. [96] (2007) 10 HKCFAR 98 at §99. [97] Court of Appeal §136. [98] Court of Appeal §140. [99] Judge §704. [100] Judge §744. 
Hon Poon CJHC (giving the judgment of the Court): 1. The main and indeed only issue arising from these appeals from the judgment of Wilson Chan J dated 30 August 2022[1] is whether on a proper interpretation, “specified evidence” in section 1 of Schedule 1 of the Implementation Rules for Article 43 of the Law of the People’s Republic of China on Safeguarding National Security in the Hong Kong Special Administrative Region (“IR”)[2] covers journalistic material. A. The proceedings below 2. Pursuant to a search warrant dated 6 August 2020 (“the 2020 Warrant”) issued under the Police Force Ordinance (“PFO”),[3] the police searched the plaintiff’s residence and seized, among other things, his two iPhones. The 2020 Warrant did not authorize the search and seizure of journalistic material. So on 13 August 2020, the plaintiff commenced HCMP 1218/2020 for directions under the court’s inherent jurisdiction for determining if the seized materials are subject to legal professional privilege or contain journalistic material, and if so, for return of the same to him. By an order of the Judge dated 19 November 2020, varied on 26 February 2021 and 4 August 2022, a protocol was put in place for sealing of the phones and the determination of the plaintiff’s claims for legal professional privilege and journalistic material.[4] 3. On 8 July 2022, as part of the ongoing criminal investigation and based on the latest circumstances and evidence available to the police, they obtained from a designated magistrate a search warrant (“the 2022 Warrant”) under section 2 of Schedule 1. In contrast to the 2020 Warrant, the 2022 Warrant specifically authorizes the search of any parts of the digital contents of the phones seized and their copies, including such digital contents which are subject to claims of journalistic material in HCMP 1218/2020 (“the Digital Contents”). It further provides that the phones and their copies may only be unsealed pursuant to the Judge’s further order. The case of the Commissioner of Police (“the Commissioner”) is that the magistrate had conducted a balancing exercise based on public interest when issuing the 2022 Warrant. 4. By a summons dated 21 July 2022 (“the Summons”), the Commissioner applied for directions that the Digital Contents, which are sealed, including those parts on which claims for journalistic material have been made, but excluding those on which legal professional privilege is claimed, be made available to the police. It would appear that the purpose of the Summons is to give effect to the 2022 Warrant. At the hearing on 5 August 2022, the plaintiff invited the Judge to conduct the balancing exercise afresh while the Commissioner submitted that he could apply either to set aside or judicially review the 2022 Warrant. In the end, the plaintiff commenced HCAL 738/2022 for leave to apply for judicial review against the validity of the 2022 Warrant (“the Leave Application”). The only ground raised by the plaintiff was that as a matter of construction, “specified evidence” as defined in section 1 of Schedule 1 does not cover journalistic material, so that the magistrate did not have the power to issue the 2022 Warrant. 5. The Judge heard both the Leave Application and the Summons at a rolled-up hearing on 22 August 2022. By the judgment, the Judge dismissed the Leave Application. After rejecting the plaintiff’s construction of “specified evidence” as wholly untenable, the Judge held that his intended judicial review is bound to fail. The Judge allowed the Summons, noting that the directions sought are simply to give effect to the 2022 Warrant. After hearing the parties, he granted an interim stay of his order until 6 September 2022, giving the plaintiff time to apply for urgent appeal to the Court of Appeal, which he subsequently did. 6. Pursuant to the order of this Court dated 5 September 2022,[5] we fixed the hearing on 28 September 2022 with an interim stay pending the outcome of the appeals. We did so with HCCC 51/2022 in mind. On 1 December 2022, the plaintiff, together with other defendants, will stand trial in those criminal proceedings of the offence of conspiracy to print, publish, sell, distribute, display and/or reproduce seditious publications, contrary to sections 10(1)(c), 159A and 159C of the Crimes Ordinance,[6] and the offences of conspiracy to commit collusion with a foreign country or with external elements to endanger national security, contrary to article 29(4) of The Law of the People’s Republic of China on Safeguarding National Security in the Hong Kong Special Administrative Region (“NSL”). We expedited the hearing because depending on the outcome, the journalistic material in the Digital Contents might be available to the police, which in turn might potentially have an impact on the conduct of the trial in HCCC 51/2022. B. The parties’ contentions in a nutshell 7. The term “specified evidence” is defined in section 1 of Schedule 1 to mean: “anything that is or contains, or that is likely to be or contain, evidence of an offence endangering national security.” 8. Before us, as was before the Judge, the plaintiff contended that journalistic material is essential to the freedom of expression and the freedom of the press as guaranteed under the Basic Law and the Hong Kong Bill of Rights. There are procedural safeguards provided for in Part XII of the Interpretation and General Clauses Ordinance (“IGCO”)[7] to protect journalistic material from search and seizure based on public interest. When Schedule 1 was drafted, the principle that search and seizure of journalistic material must be protected by those procedures had already been firmly entrenched in our law. In order to interfere with the freedom of the press and/or deviate from those procedural safeguards, there must be express language or necessary implication. If there is no intention to deviate from that principle, then “specified evidence” in section 1 cannot be construed to include journalistic material from its definition. Access to journalistic material would be left to the process in Part XII of the IGCO. 9. The Commissioner countered by submitting that there is no reason not to give the words defining “specified evidence” their natural and ordinary meaning, which clearly cover journalistic material. Further, Part XII of the IGCO is not the only lawful regime to protect journalistic material from search. Under section 2 of Schedule 1, the magistrate may issue a warrant. The legislative intention is that in exercising his jurisdiction to issue a warrant, the magistrate has to conduct a balancing exercise based on public interest as required under the common law. The interpretation advocated by the plaintiff is therefore based on a false premise and cannot stand. C. Construction of “specified evidence” 10. Schedule 1 of the IR relevantly provides: “1. Interpretation In this Schedule – … specified evidence (指明證據) means anything that is or contains, or that is likely to be or contain, evidence of an offence endangering national security. 2. Magistrate’s warrants (1) A police officer may, for investigation of an offence endangering national security, apply to a magistrate by information on oath for a warrant under this section in relation to the place specified in the information. (2) A magistrate may issue a warrant authorizing a police officer with such assistance as may be necessary to enter and search any place if the magistrate is satisfied by information on oath that there is reasonable ground for suspecting that any specified evidence is in the place. (3) A warrant issued under subsection (2) authorizes the policer officer to – … (b) inspect, examine, search, seize, remove and detain anything in the place that the officer reasonably believes to be specified evidence; and …” C1. General approach of construction 11. Within the framework of the NSL, the IR are delegated legislation made pursuant to the authorization under NSL 43(3), which provides: “The Chief Executive shall be authorized, in conjunction with the Committee for Safeguarding National Security of the Hong Kong Special Administrative Region, to make relevant implementation rules for the purpose of applying the measures under [NSL 43(1)].” The IR contain seven schedules corresponding to and augmenting each of the measures listed in the seven subparagraphs of NSL 43(1). 12. As a general rule, delegated legislation is interpreted in the same way as the primary legislation, with the additional consideration that since delegated legislation derives its authority from the enabling legislation, it must be interpreted in light of the latter: see Bennion, Bailey and Norbury on Statutory Interpretation, Eighth Edition, §3.17, at p 114. Applying the general rule, the approach of construction for the NSL as expounded by the Court of Final Appeal in HKSAR v Lai Chee Ying (2021) 24 HKCFAR 33 is equally apposite for the construction of the IR. 13. As is now well-known, the Court of Final Appeal in Lai Chee Ying distilled from the legislative history of the NSL and NSL 1, 3, 4 and 5 the obvious legislative intention for the NSL to operate in tandem with the laws of the HKSAR, seeking convergence, compatibility and complementarity with local laws, subject to NSL 62 which gives priority to NSL in case of inconsistency: [29]. Based on that legislative intention, the Court laid down a purposive and contextual construction to an NSL provision in its application to the HKSAR by examining the matrix in which it exists, consisting of the relevant provisions of the Basic Law and the NSL, the applicable corpus of local laws including human rights and rule of law principles, the statutory norms and the common law rules, with a view to ascertaining how it is intended to operate in that context: [42] and [45]. C2. The context of “specified evidence” 14. The NSL being the enabling legislation forms a crucial part of the legislative context in which the IR must be construed: Bennion, ibid. Reading the NSL and the IR as a coherent whole, the court proceeds on the basis that the legislative purpose underlying section 1 of Schedule 1 is consistent with the NSL. 15. The following provisions in the NSL and the IR are of immediate relevance. 16. First, NSL 43(1) mandates that the NSL and the laws of the HKSAR shall apply to procedural matters including criminal investigation in respect of cases concerning offence endangering national security over which the Region exercises jurisdiction in these terms: “When handling cases concerning offence endangering national security, the department for safeguarding national security of the Police Force of the Hong Kong Special Administrative Region may take measures that law enforcement authorities, including the Hong Kong Police Force, are allowed to apply under the laws in force in the Hong Kong Special Administrative Region in investigating serious crimes, and may also take the following measures: (1) search of premises, vehicles, vessels, aircraft and other relevant places and electronic devices that may contain evidence of an offence; …” 17. Two points are noteworthy: (1) One of the primary objectives of the NSL is to effectively suppress, prevent and punish offences and acts endangering national security. That objective is repeatedly emphasized in NSL 1, NSL 3(1) and (2), NSL 5(1), NSL 8 and NSL 42(1). Effective investigation by the police is crucial to achieving that objective. To that end, the police must have sufficient powers to take all necessary measures in carrying out investigation, as is the case for other serious crimes. Hence the provision in NSL 43(1). Thus understood, the legislative purpose of NSL 43(1) is evidently to give effect to the above objective of the NSL. Derived from and in augmenting subparagraph (1) of NSL 43(1), Schedule 1 clearly shares the same legislative purpose. A proper construction of “specified evidence” in section 1 must be consistent with and give effect to that purpose. (2) As mandated by NSL 43(1), both the NSL and local laws on search apply in an offence endangering national security, indicating the legislative intention that Schedule 1 and the local laws on search are to work in tandem as a coherent whole. It means that the local laws on search powerfully informs the construction of “specified evidence” in section 1. 18. Second, NSL 4 requires the protection of fundamental rights thus: “Human rights shall be respected and protected in safeguarding national security in the Hong Kong Special Administrative Region. The rights and freedoms, including the freedoms of speech, of the press, of publication, of association, of assembly, of procession and of demonstration, which the residents of the Region enjoy under the Basic Law of the Hong Kong Special Administrative Region and the provisions of the International Covenant on Civil and Political Rights and the International Covenant on Economic, Social and Cultural Rights as applied to Hong Kong, shall be protected in accordance with the law.” 19. Journalistic material concerns the freedom of the press, which is a fundamental right guaranteed by article 27 of the Basic Law: “Hong Kong residents shall have freedom … of the press …” The freedom of the press is also protected as a facet of the freedom of expression by article 16 of the Hong Kong Bill of Rights via article 39 of the Basic Law: HKSAR v Ng Kung Siu (1999) 2 HKCFAR 442, per Li CJ at p.455; see also Comilang v Director of Immigration (2019) 22 HKCFAR 59, per Ribeiro and Fok PJJ at [25]. 20. Journalistic material is essential to the freedom of the press. As Ma CJHC (as he then was) observed in So Wing Keung v Sing Tao Ltd [2005] 2 HKLRD 11, at [36]: “Journalistic material forms the backbone of the freedom of the press. As a general rule, such material must be given the greatest possible protection from seizure or public exposure; otherwise the press may become inhibited in informing the public of matters it is entitled to know.” In a similar vein, the European Court of Human Rights in Financial Times Ltd v United Kingdom [2010] EMLR 21 emphasized at [59]-[63] that the protection of journalistic sources is one of the basic conditions for the freedom of the press; and where there is a possibility of disclosure of such a source, there is potentially a chilling effect undermining the “public watchdog-role” of the press. 21. The immense importance of journalistic material to the freedom of the press which is to be protected in safeguarding national security in the HKSAR, as explicitly required by NSL 4, is centrally important to the construction of “specified evidence” in section 1 of Schedule 1. 22. Third, NSL 5 emphasizes the adherence to the principle of the rule of law by stipulating: “The principle of the rule of law shall be adhered to in preventing, suppressing, and imposing punishment for offences endangering national security.” 23. As an important facet of the rule of law, the principle of legality requires that if a statute does not expressly or by necessary implication override or restrict fundamental rights, it will not be construed as doing so: R v Home Secretary, ex p Simms [2000] 2 AC 115, per Lord Hoffmann at p.131; HM Treasury v Ahmed [2010] 2 AC 534, per Lord Hope at [45]-[46], per Lord Walker at [112]; A v Commissioner of ICAC (2012) 15 HKCFAR 362, per Bokhary and Chan PJJ at [24]-[29]; and per Ribeiro PJ at [68]-[70]. This principle of construction is also highly relevant to the construction exercise at hand. 24. Fourth, section 2(2) of Schedule 1 confers on the magistrate a discretion in issuing a warrant. In exercising that discretion, the magistrate is guided by the above primary objective of the NSL; the above legislative purpose of Schedule 1; the requirements in NSL 4 and NSL 5 for the protection of the freedom of the press and the adherence to the principle of the rule of law; and the local laws on search, which should operate as a coherent whole with Schedule 1. 25. Arising from the local laws on search are the following features which shed considerable light on the construction of “specified evidence”. 26. First, unless there is express provision to the contrary, any statutory provisions conferring on or authorizing the issue of a search warrant shall not be construed as conferring or authorizing the search of journalistic material: section 83 of the IGCO.[8] The practical effect of section 83 is to subject the search and seizure of journalistic material by warrant authorized by legislation generally to the regime in Part XII of the IGCO. 27. Part XII was introduced in 1995. It is underpinned by the competing public interests in the freedom of the press and law enforcement. As Keith JA explained in Apple Daily Ltd v Commissioner of the ICAC (No 2) [2000] 1 HKLRD 647, p.654C-F: “The rationale underlying Pt.XII, I believe, relates to the important role played by a free and independent press as public watchdog. The press should be able to speak out on matters of public interest without fear of reprisal, and journalists need to protect the confidentiality of the sources of the information they receive. On the other hand, the legitimate requirements of law enforcement agencies may in exceptional cases make it necessary for journalistic materials to be the subject of seizure and inspection. In this sensitive area, Pt.XII of the IGCO requires a judge of the Court of First Instance or the District Court to hold the balance between these competing interests.” 28. Three broad points emerge from the scheme in Part XII: (1) Journalistic material is not immune from search and seizure for the purpose of criminal investigation. (2) Applications for a warrant for journalistic material require the court to consider the public interest: section 89(2). A person from whom journalistic material has been seized can apply to the court for the return of such material: section 87. At a section 87 hearing, the court has to consider whether it would be in the public interest that the seized material should be made use of for the purpose of the relevant investigation. As Ma CJHC pointed out in So Wing Keung, at [36(9)], this, together with section 89(2), permits the court to look at all the circumstances of the case. (3) Public interest is relevant at both the issue and execution stages of the warrant. 29. Second, as agreed between the parties, the common law also recognizes the need for the court, in dealing with a search warrant for journalistic material, to balance the competing public interests in (a) protecting such material and (b) crime prevention and law enforcement objectives: R (Bright) v Central Criminal Court (DC) [2001] 1 WLR 662, per Judge LJ at p 681C-G. His Lordship emphasized: “Premises are not to be entered by the forces of authority or the State to deter or diminish, inhibit or stifle the exercise of an individual’s right to free speech or the press of its freedom to investigate and inform, and orders should not be made which might have that effect unless [the court] is personally satisfied that the statutory preconditions to the making of an order are established and, as the final safeguard of basic freedoms, that in the particular circumstances it is indeed appropriate for an order to be made.” 30. Third, it is not suggested by the plaintiff, nor do we think it can be seriously suggested, that Part XII of the IGCO had expressly or by implication abrogated the above common law rule. It follows that Mr Suen SC, for the Commissioner is correct in his submission that Part XII of the IGCO is not the only lawful regime to address claims based on journalistic material for warrants. 31. Fourth, under the common law, the court performs the judicial gatekeeping role against unlawful and arbitrary interference with fundamental rights that a search warrant might entail: Keen Lloyd Holdings Ltd v Commissioner of Customs and Excise [2016] 2 HKLRD 1372, at [71]-[75]. The court performs that role at different stages of the warrant before it becomes functus officio. For example, it may impose conditions when issuing a warrant. It may set aside or vary a warrant upon the application by a person affected by it. 32. In performing its judicial gatekeeping role in connection with journalistic material, the court has to balance the public interest to decide whether to issue the warrant with or without conditions. Bearing in mind the application is made ex parte, if the court has some concerns about the public interest but is satisfied that the warrant should nonetheless be issued because of, say, the urgency of the matter, it may impose conditions to ensure that the material seized is properly sealed pending any claim of journalistic material or determination of such claim based on public interest at an inter partes hearing. Lastly, the court may set aside or vary the warrant based on public interest upon the application by the person affected by it. As to the exact procedure to be adopted for the balancing exercise, it is a matter for the magistrate to decide having regard to the actual circumstances of the case. C3. Construing “specified evidence” 33. Construing “specified evidence” in the above context, it is clear that as borne out by its natural and ordinary language, it covers journalistic material. 34. First, despite its importance to the freedom of the press, the protection afforded to journalistic material is not absolute. Although always subject to the protection and procedural safeguards based on public interest and vigilant judicial scrutiny, journalistic material is not immune from search and seizure in the investigation of any criminal offence. As a matter of principle, the same must be true for offences endangering national security. 35. Second, to serve the legislative purpose of furthering the primary objective of the NSL to effectively suppress, prevent and punish offences endangering national security as identified at [17(1)] above, the police must be able to carry out effective search on anything, including journalistic material, that contains or is likely to contain evidence of an offence endangering national security. Were such material excluded from the definition of “specified evidence”, it would unduly limit the scope and hence reduce the effectiveness of police investigation. That would not be conducive to the said legislative purpose. 36. Third, such a construction does not diminish the protection afforded to the freedom of the press by the local laws or violate the principle of legality. For although Part XII of the IGCO has not been incorporated within its framework, Schedule 1 operates in tandem with the local laws on search as a coherent whole. The same protection and safeguards based on public interest for journalistic material under the common law equally apply to a warrant under Schedule 1. The magistrate will perform the same judicial gatekeeping role as detailed above in exercising his discretion under section 2 of Schedule 1 in ensuring that the search and seizure of journalistic material is justified in the public interest. 37. We digress to deal with Mr Pang’s complaint that, even if the magistrate had the jurisdiction to set aside or vary the 2022 Warrant based on public interest, such remedy is illusory because any such application taken by the plaintiff now would have been time-barred under section 104(1) of the Magistrates Ordinance.[9] We disagree. 38. Section 104(1) provides: “Within 14 clear days after the determination in any manner by a magistrate of any matter which he has power to determine in a summary way it shall be lawful for either party thereto to apply to the magistrate to review his decision in the matter.” It would apply only if the decision to issue a warrant is a matter which a magistrate has power to determine in a summary way. 39. As explained by this Court in Kwok Tak Ying v HKSAR & Secretary for Justice (Interested Party) [2021] 4 HKLRD 841, at [59] – [62], the reference to the power of a magistrate to determine in a summary way denotes his summary jurisdiction to hear and determine a matter brought before him by way of a complaint, information or charge. When a magistrate deals with an application for a warrant, he is exercising the statutory power that authorizes him to do so. He is not exercising his summary jurisdiction. His decision to issue a warrant is therefore not covered by section 104(1). 40. The discussion above has covered the parties’ main submissions. So we do not propose to separately address them point by point in detail. For completeness, we only need to deal with two new points taken by Mr Pang, SC for the plaintiff, which do not concern construction. 41. First, he argued that there is no evidence to show that the magistrate had performed the balancing exercise based on public interest in issuing the 2022 Warrant. However, as pointed out by Mr Suen, this complaint did not appear in the Form 86. Nor was it raised before the Judge. In the circumstances, Mr Pang cannot run it before us. 42. Second, Mr Pang submitted the police ought to have applied for a production order under Schedule 7 of the IR instead of the 2022 Warrant. Again this point was neither pleaded nor raised below. More importantly, it is entirely a matter for the Commissioner to decide how best to carry out the investigation, whether by a warrant or a production order. There is no merit in this point. 43. For the above reasons, like the Judge, we hold that the plaintiff’s construction of “specified evidence” is untenable. D. Dispositions 44. Accordingly, the plaintiff’s intended judicial review against the 2022 Warrant is doomed to fail. The Judge was correct in refusing leave. As to the Summons, as the plaintiff has not advanced any other separate ground in opposing it, the Judge was also correct in allowing it. 45. In consequence, we dismiss both appeals. We further make an order nisi that the plaintiff do pay the Commissioner the costs of the two appeals, to be taxed if not agreed with a certificate for three counsel. Mr Robert Pang SC, Mr Steven Kwan, Mr Albert N B Wong and Ms Samantha Lau, instructed by Robertsons, for the plaintiff/applicant Mr Jenkin Suen SC, Mr Michael Lok and Ms Ellen Pang, instructed by the Department of Justice, for the defendant/putative respondent [1] [2022] HKCFI 2688. [2] The IR came into force on 7 July 2020. [3] Cap 232. [4] The plaintiff’s claims were disposed of by the Judge by a judgment (not open to public) handed down on 30 September 2022: [2022] HKCFI 3003. [5] Poon CJHC and Kwan VP. [6] Cap 200. [7] Cap 1. [8] The statutory provisions which expressly cover search of journalistic material are section 21 of the Drug Trafficking (Recovery of Proceeds) Ordinance, Cap 405 and section 5 of the Organized and Serious Crimes Ordinance, Cap 455, which already contain their own elaborate and stringent safeguards. [9] Cap 227. 
Chief Justice Cheung: 1. I agree with the judgment of Mr Justice Lam PJ. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Mr Justice Lam PJ. Mr Justice Fok PJ: 3. I agree with the judgment of Mr Justice Lam PJ. Mr Justice Lam PJ: 4. In HKSAR v Lo Kin Man[1], this Court expounded at length on the law relating to the offences of unlawful assembly and riot under Sections 18 and 19 of the Public Order Ordinance (“the POO”) Cap 245. In respect of the “taking part” in an unlawful assembly, it was held that a defendant must have a participatory intent[2]. The present appeal addresses how such requirement is to be applied when a defendant was one of the persons who committed the prohibited conduct specified in Section 18(1) which rendered the assembly unlawful[3]. The incident 5. The relevant parts of the incident leading to the arrest and prosecution of the Respondent were captured on video recordings produced at the trial and there is no serious dispute on the facts. It took place at a time when the doxxing of police officers was a serious problem and the courts had to intervene by granting injunctions against such activities[4]. 6. On 8 March 2020, a police officer (“PW1”) was on plainclothes duty near Tai Po Mega Mall. At around 9:40 p.m. he was approached by a person[5] demanding that PW1 reveal his identity as a police officer. PW1 did not reply. Thereafter, that person spoke with a co-defendant, viz the 4th defendant, informing the latter that PW1 was a police officer[6]. PW1 was then tailed by the 4th defendant and several other persons. Some of them came close to PW1 from time to time whilst most of them kept their distance. There were people shouting his name and the names of his wife and daughter. There were also repeated accusations that he had previously beaten up a youngster with a baton. Denigrating and threatening remarks were made against him. PW1 left the mall and walked in the direction of Tai Po Police Station. 7. After PW1 had come out from the mall, four people including the Respondent rushed to where he was and trailed closely behind him. As he did so, the Respondent used a video camera continuously to make a recording. There were other people following a further distance away. Amongst the four who were trailing closest to PW1, two of them flashed their torches at him. At one stage, the flashes aimed at PW1’s head and the Respondent overtook PW1 to film his face whilst the others shouted PW1’s name and accused him of the beating of a youngster. In addition to those four persons, those following behind also came closer and the group of people around PW1 increased. 8. Subsequently, other police officers arrived and arrests were made. The Respondent and four other persons were charged with the offence of unlawful assembly. The Respondent pleaded not guilty. His defence was that there was no unlawful assembly and that he did not participate in any assembly. His counsel submitted that the Respondent only filmed the incident spontaneously without any ill intent on his part. The proceedings below and leave to appeal 9. After trial, Principal Magistrate Don So (“the magistrate”) found the Respondent guilty and imposed a sentence of 3 months’ imprisonment on him. The magistrate decided the case without the benefit of this Court’s judgment in HKSAR v Lo Kin Man as that appeal was only heard after his decision. Although PW1 was first approached by a person asking if he was a police officer about 6 minutes prior to the commencement of the filming by the Respondent, the magistrate focused on the conduct of the parties, which lasted about 2 minutes, as shown in the recording made by the Respondent. The magistrate found that the Respondent and three other defendants[7] had assembled together[8] and their acts insulted and provoked PW1. Such acts included the Respondent’s filming which targeted PW1 most of the time and were likely to cause other persons reasonably to fear that those assembled would commit a breach of the peace or provoke other persons to commit a breach of the peace[9]. He also found that cumulatively these acts had a greater intimidating effect. He held that the respective act of the 2nd to 4th defendants, and that of the Respondent, constituted prohibited conduct under Section 18(1) of the POO. 10. In rejecting the Respondent’s defence, the magistrate found that he had consciously assembled with the other three defendants. He noted that there was no suggestion that the Respondent was a reporter but that he was able to film the incident without fear of assault or other hostile responses by the 4th defendant. He inferred that there was a consensus between the Respondent and the 4th defendant that the recording would not be voluntarily produced as incriminating evidence. He further found that the Respondent was aware of the harassment, threat and provocation directed towards PW1 at the scene and that he deliberately participated in the assembly[10]. 11. On appeal to the Court of First Instance, A Wong J (“the Judge”) allowed the Respondent’s appeal against conviction. The Judge rejected the contention that the filming of PW1 by the Respondent did not come within the scope of prohibited conduct under Section 18(1) of the POO[11]. Having regard to the circumstances surrounding the Respondent’s filming, the Judge upheld the magistrate’s finding that such conduct was provocative. 12. The Judge also agreed with the magistrate in finding that the Respondent was aware of the acts of the other defendants[12] and that the conduct of the Respondent had the effect of procuring or encouraging the other defendants to commit their prohibited acts[13]. He noted that there was an unlawful assembly with more than 3 persons (including the Respondent) committing prohibited conduct and the Respondent’s filming took place against such backdrop when he was “well aware of the actual situation at the scene”[14]. 13. The Judge affirmed the finding that the 4th defendant permitted or acquiesced in the Respondent’s filming which recorded the commission of the offence by the 4th defendant. However, he did not accept that the only reasonable inference was that there was a consensus between the Respondent and the 4th defendant. The Judge also declined to assess the matter on the basis that the Respondent had targeted PW1 in his filming though he considered the magistrate was justified in so holding[15]. 14. By the time the Judge heard the appeal, this Court had delivered its judgment in HKSAR v Lo Kin Man. The Judge referred to the law set out in that judgment[16] and he discussed the issue of participatory intent in light of it [17]. After considering various matters which he regarded as relevant, the Judge concluded that though it was a reasonable inference supported by the evidence that the Respondent had the requisite participatory intent, it was not the only reasonable inference[18]. He therefore allowed the appeal by quashing the conviction. 15. The prosecution sought leave to appeal against the Judge’s decision. On 11 July 2022, the Appeal Committee granted leave as it was reasonably arguable that the Judge had misapplied this Court’s judgment in HKSAR v Lo Kin Man and that substantial and grave injustice had been done. Lo Kin Man’s exposition of unlawful assembly 16. Section 18 of the POO codified the offence of unlawful assembly. In HKSAR v Lo Kin Man[19], the elements of the offence were set out as follows: “[1] When 3 or more persons, [2] assembled together, [3] conduct themselves in a disorderly, intimidating, insulting or provocative manner [4] intended or likely to cause any person reasonably to fear that the persons so assembled will commit a breach of the peace, or will by such conduct provoke other persons to commit a breach of the peace, [5] they are an unlawful assembly. [section 18(1)] [6] It is immaterial that the original assembly was lawful if being assembled, they conduct themselves in such a manner as aforesaid. [section 18(2)] [7] Any person who takes part in an assembly which is an unlawful assembly by virtue of subsection (1) shall be guilty of the offence of unlawful assembly ... [section 18(3)]” 17. The offence-creating provision is Section 18(3) and the essence of the offence is the “taking part” in an unlawful assembly[20]. Such “taking part” may or may not involve the same acts as, and should not be confused with, the prohibited conduct of the constituent offenders referred to in elements [3] and [4]. 18. Constituent offenders are those who assemble together and conduct themselves in the manner described in element [3]. Because of their conduct, the assembly is unlawful[21]. For these offenders, their “taking part” is self-evident: “If the defendant was one of the constituent offenders, by engaging in the prohibited conduct, he or she will (along with other constituent offenders) have taken part in the unlawful assembly that they will together have brought into being.”[22] “As a matter of textual analysis, it is implicit that if the defendant is one of the constituent offenders whose conduct falls within [3] and [4] he or she would ‘take part’ since it is by such acts that the unlawful assembly is established, being of the essence of the offence.”[23] 19. For persons other than the constituent offenders, viz those who took part later, they could participate by acting in the manner prohibited by elements [3] and [4][24]. Alternatively, they could also participate by facilitating, assisting or encouraging the performance of such conduct by others in the assembly[25]. In so doing, they acted in furtherance of the prohibited conduct[26]. 20. This Court also alluded to element [6] which underscores that an assembly could be transformed from a lawful one to an unlawful one[27]. It is only the participation in an unlawful assembly which constituted an offence. The participation in a lawful assembly is not an offence as freedom of assembly is a fundamental right protected under Article 27 of the Basic Law. 21. As discussed in earlier cases[28], the fluidity of an assembly has to be borne in mind. People who participate in a lawful assembly which subsequently develops into an unlawful one should not be found guilty if they do not take part in the unlawful assembly. Thus, this Court emphasized: “It is obviously important to avoid treating innocent passers-by who find themselves caught up in an unlawful assembly or riot as guilty of an offence just because they were present at the scene. Presence at the scene in itself is not enough to constitute ‘taking part’ or aiding and abetting. As the Queensland Court of Appeal held in R v Cook, at common law, mere presence in an unlawful assembly or riot is generally insufficient to found liability. It has traditionally been considered necessary that there be some intentional activity in furtherance of the riot.”[29] (footnotes omitted) 22. On the other hand, if some latecomers to a lawful assembly conduct themselves in the manner prohibited by Section 18(1), they could turn the assembly into an unlawful one and they would be the constituent offenders. Those in the assembly who continue to take part in it after such transformation would be guilty under Section 18(3). Moreover, even if the constituent offenders leave the scene after committing the prohibited conduct, the unlawful assembly continues so long as three or more participants in the assembly remain actively engaged in such conduct[30]. 23. Whether those who remain in the assembly after its transformation into an unlawful assembly can be regarded as taking part in the unlawful assembly depends on their continued participation in the assembly with a participatory intent. This Court stressed that mere presence without more would not be regarded as taking part[31]. 24. The requirement of participatory intent is neatly summarized in HKSAR v Lo Kin Man: “The defendant must accordingly intend to take part in, that is, become part of, the unlawful assembly, being aware of the related conduct of other participants and intending, while assembled together with them, to engage in or act in furtherance of the prohibited conduct. The defendant must, in other words, have what we will call a ‘participatory intent’.”[32] 25. Two ingredients were set out in that paragraph: (a) the defendant intended to become part of the assembly; and (b) while he or she assembled together with these other participants and was aware of the related conduct of other participants, he or she intended to engage in or act in furtherance of the prohibited conduct. 26. Ingredient (a) refers to the intention to become part of an assembly, whilst ingredient (b) refers to the mental state of the defendant in acting in the way he did when he participated. 27. The intention to become part of an assembly corresponds with the actus reus of the defendant in acting as part of the relevant assembly[33]. Thus, as held in Secretary for Justice v Leung Kwok Wah: “... if three persons in a lawful assembly committed acts of the prescribed nature at different parts of the place of assembly for different purposes, sparking off different incidents, involving and affecting an entirely different mix of persons, there would not be a sufficient nexus to turn these independent acts into an unlawful assembly of those three persons.”[34] (footnote omitted) 28. Conversely, if a defendant was in proximity to the other participants who committed prohibited acts and, whilst being aware of such conduct, the defendant joined in by committing further act or acts, a sufficient nexus could readily be established. The prohibited acts of different participants could be different in nature but they must satisfy the requirement of element [3]. Since they committed these acts in the same assembly, the intended or likely fear of a breach of the peace by mob disorder, intimidation or violence under element [4] is enhanced. As observed in R v Caird[35], the gravity of this type of offence stems from the number of people involved as a mob engaged in a crime against the peace and it is the degree of mob disorder, intimidation or violence that matters. It is a wrong approach to treat different acts of provocation or intimidation in an assembly as isolated conduct in such circumstances. 29. As far as ingredient (a) is concerned, it suffices that a defendant has the intent to become part of the assembly. It is not necessary to show that such defendant has the intention to take part in each and every act of the other participants. The intent to participate in specific act or acts in the assembly is to be considered under ingredient (b). 30. In respect of ingredient (b), a defendant has to be aware of the prohibited conduct which rendered the assembly unlawful or the further prohibited conduct which continued the unlawful character of the assembly. With such knowledge, the requisite mental state can be satisfied in two different ways: by showing either (i) an intention to engage in the prohibited conduct or (ii) an intention to act in furtherance of the prohibited conduct. Such alternatives in establishing the mental element mirror the conduct of a defendant that constitute “taking part” in the assembly, viz by either engaging in prohibited conduct or acting in furtherance of such conduct by facilitating, assisting or encouraging the conduct of the others in the assembly. 31. This Court also held that the prosecution needs not establish any extraneous common purpose amongst the participants[36]. In so holding, this Court specifically rejected an additional “mutual assistance” element and answered Question 1(d) in the negative[37]. 32. Further, this Court held that it would be confusing to apply the doctrine of basic joint enterprise in the context of unlawful assembly. It emphasized that it was not necessary to show a prior agreement between the participants: “Their participatory intent would be inferred from their prohibited conduct while assembled together, so from the prosecution’s viewpoint, invoking the BJE doctrine would add the unwarranted burden of showing a prior agreement. From the Trial Judge’s point of view, jettisoning BJE in this context would avoid possibly confusing the jury by trying to explain two layers of ‘taking part’ – first in the joint enterprise and then in the unlawful or riotous assembly.”[38] The constituent offenders in the present case 33. Though PW1 was first approached about 6 minutes prior to when the Respondent started his filming of the incident, there was no suggestion or finding that there was an unlawful assembly during those 6 minutes. In my view, the magistrate correctly focused on the events captured in the Respondent’s recording. That was the stage at which 3 or more persons acted together in the close vicinity of PW1 and behaved in the manner prohibited by Section 18(1). Whilst there were earlier acts of flashing light and accusations directed at PW1 with another person making a video recording, such footage did not evidence conduct committed by 3 or more persons acting together in those instances. 34. In short, for present purposes, the unlawful assembly only came into being after the Respondent had joined the other defendants in the group trailing closely behind PW1. The Respondent was one of the constituent offenders in the unlawful assembly. 35. The finding of the magistrate was that the Respondent had committed the prohibited conduct by filming when he was with the other defendants. The Judge upheld such finding as he rejected the ground of appeal concerning the nature of his conduct. Thus, the Respondent participated in the assembly by committing a prohibited act. Ingredient (a): the Respondent’s intent to become part of the assembly 36. The magistrate specifically found that the Respondent had assembled with the other defendants. Though the Judge reversed the magistrate on participatory intent, he did not do so on the basis that there was an insufficient nexus between the Respondent and the other defendants. On the contrary, he found that the Respondent had run to the place where others were assembled[39] and had engaged in the filming, well aware of the actual situation at the scene[40]. 37. These findings are well-supported by the video recordings produced as evidence at the trial. The Respondent was seen to be rushing towards PW1 together with others in the group[41] shortly after the 4th defendant flashed his torch at PW1 inside the mall. Moreover, as found by the magistrate (and affirmed by the Judge), the Respondent was able to do the filming without objection or fear of reprisal by the others in the group. Thus, the magistrate was justified in drawing the inference that the Respondent was acting together with the other defendants. In other words, he had the intent to become part of the group of people who pestered PW1 at close distance. 38. Mr Choy SC[42] submitted on behalf of the Respondent that the Judge had adopted the approach suggested by this Court in HKSAR v Lo Kin Man[43] and came to the conclusion that he could not draw an irresistible inference that the Respondent had the intention to participate after considering a list of 19 factors. Counsel said the Judge was obliged to examine the matter by way of rehearing and this Court should not interfere with his assessment. 39. As explained in HKSAR v Lo Kin Man, circumstantial matters should be taken into account in the determination of whether a defendant had taken part in an unlawful assembly or a riot. However, the relevance of each matter must depend on the facts of each case. 40. The Judge did not reverse the finding that the Respondent had joined the others in the assembly. He also accepted that the Respondent was aware of the related conduct of the other defendants and that his provocative filming of PW1 was done against the backdrop of the unlawful assembly. 41. In the consideration of his list of 19 factors[44], the Judge veered away from asking whether the Respondent had the intent to become part of the assembly. Instead, he proceeded to examine if the Respondent had the requisite intent of procuring or encouraging the other defendants to perform their prohibited conduct[45]. Thus, he laid emphasis on certain factors including his reservation regarding the finding of the magistrate as to consensus with the other defendants, the different nature of the conduct of the Respondent and his lack of communication with the others. 42. With respect, in doing so in a case where the participation was by committing a prohibited act, the Judge failed to have regard to the following propositions laid down in HKSAR v Lo Kin Man: (a) The prohibited act of one participant could be different from that committed by another participant; (b) There is no need to prove an agreement or consensus on the part of a defendant with the other participants in the commission of their prohibited conduct; (c) There is no need to prove an intention to offer mutual assistance to one another in the commission of the prohibited conduct. 43. Had the Judge properly focused on the Respondent’s intent to become part of the assembly, notwithstanding those factors he alluded to, the inescapable conclusion would have been that the Respondent had joined the assembly for the reasons already canvassed above[46]. Ingredient (b): the intent of the Respondent in his acts of participation 44. As the Respondent participated by committing a prohibited act, his participatory intent can be established by reference to his intention to engage in such act while assembled with the other defendants. There is no doubt that he was aware of the related conduct of the other participants. Likewise, there is no doubt that he intended to engage in his filming of PW1. As held in HKSAR v Leung Chung Hang Sixtus[47], it is not necessary to prove in addition that the Respondent intended his act to be provocative in nature. The submission to the contrary by Mr Choy SC is without merit. 45. Though the Judge had previously acknowledged that the magistrate had found the Respondent guilty on the basis that he was a constituent offender[48], it is not clear whether the Judge had regard to such finding in his assessment of participatory intent. 46. As explained above[49], it is not necessary in the present case for the prosecution to prove an intent on the part of the Respondent to act in furtherance of the prohibited conduct of the other defendants. In other words, there is no need to show that by his filming of PW1 the Respondent intended to facilitate or encourage the other defendants in committing their prohibited conduct. 47. Had the Judge properly directed himself as to the Respondent’s participation by committing a prohibited act and the requirement of participatory intent as discussed above, the matters which he relied upon should not have precluded him from drawing an irresistible inference that the Respondent had the intent to commit the prohibited act. This is clearly not a case of mere presence or spontaneous filming by an innocent bystander. Conclusion 48. The Judge misapplied the law as stated in HKSAR v Lo Kin Man and the quashing of the conviction therefore resulted in substantial and grave injustice. I would allow the appeal and restore the conviction and sentence against the Respondent. Lord Hoffmann NPJ: 49. I agree with the judgment of Mr Justice Lam PJ. Chief Justice Cheung: 50. Accordingly, the Court unanimously allows the appeal and restores the conviction and sentence against the Respondent. Mr Ivan Cheung, ADPP (Ag) and Ms Jennifer Tsui, SPP (Ag), of the Department of Justice, for the respondent (appellant) Mr Edwin Choy, SC, Mr Brian Tsui and Mr Jonathan Lin, instructed by Chak & Associates LLP, for the appellant (respondent) [1] (2021) 24 HKCFAR 302. [2] Ibid, [17], quoted at [24] below. [3] Ibid, [9]. [4] See Secretary for Justice v Persons Unlawfully and Wilfully Conducting Etc (1957/2019) [2019] 5 HKLRD 500; Junior Police Officers’ Association of the Hong Kong Police Force v Electoral Affairs Commission [2019] 5 HKLRD 291 and Secretary for Justice v Persons Unlawfully and Wilfully Conducting Etc (1847/2020) [2020] 5 HKLRD 638. Even after the grant of injunctions, there were people continuing in such activities and contempt proceedings had to be pursued, see Secretary for Justice v Cheng Lai King [2020] 5 HKLRD 356; Secretary for Justice v Chan Kin Chung [2021] 1 HKLRD 563; Secretary for Justice v Yiu Ka Yu [2021] 1 HKLRD 607. [5] That person was the 1st defendant in the trial before Principal Magistrate Don So in which the Respondent was the 5th defendant. The 1st defendant did not walk alongside PW1 during the last two minutes of the incident and he was acquitted after trial. [6] Statement of Findings at [11]. [7] The 2nd to 4th defendants in the case. [8] Statement of Findings at [14], [18] to [20]. [9] Statement of Findings at [15] and [16]. [10] Statement of Findings at [18] to [20]. [11] CFI judgment at [37] to [44]. He reiterated this at [45] as the factual basis for his consideration of participatory intent. [12] CFI judgment at [49]. [13] CFI judgment at [47]. [14] CFI judgment at [54(1) to (4) and (10)]. [15] CFI judgment at [51] to [52]. [16] In the CFI judgment, the Judge referred to the judgment as HKSAR v Leung Tin Kei. [17] CFI judgment at [33]. See also [48] where the Judge reiterated that he would rely on his [33] as the basis for his consideration of participatory intent. [18] CFI judgment at [49] to [55]. [19] Supra. at [8]. [20] Supra. at [11]. [21] Supra. at [9]. [22] Supra. at [12]. [23] Supra. at [13]. [24] Supra. at [13]. [25] Supra. at [14]. [26] Supra. at [14] and [109(d)]. [27] Supra. at [10]. [28] Kwok Wing Hang v Chief Executive in Council (2020) 23 HKCFAR 518 at [91] and [92]; Leung Kwok Hung v Secretary for Justice (No 2) [2020] 2 HKLRD 771 at [14]; Secretary for Justice v Tong Wai Hung [2021] 2 HKLRD 399 at [56]. The point was reiterated in HKSAR v Lo Kin Man, supra. at [75] to [78]. [29] Supra. at [81]. [30] Supra. at [77]. [31] Supra. at [85] and [86]. [32] Supra. at [17]. [33] Supra. at [16]. [34] [2012] 5 HKLRD 556 at [21]. Cited in HKSAR v Lo Kin Man at [42]. [35] (1970) 54 Cr App R 499 at 506 to 508. [36] Supra. at [38] to [50]. [37] Supra. at [34] to [35] and [50]. Question 1(d) is as follows: “whether it is a distinct element of riot under Section 19 of the POO that the defendants must have the mutual intention to assist each other, by force if necessary, against any person who might oppose them in the execution of the common purpose?” [38] Supra. at [67]. Basic joint enterprise was defined as BJE and referred to as such in the passage cited. [39] CFI judgment at [54(9)]. [40] CFI judgment at [54(10)]. [41] P18(D) at 21:45:46; P17A at 01:16-01:19; P18(C) at 21:45:51 to 21:46:10. [42] Appearing together with Mr Tsui and Mr Lin for the Respondent. [43] Supra. at [78]. [44] Set out at CFI judgment [54]. [45] CFI judgment at [47]. The original Chinese version of the judgment is clearer than the English translation in this respect. [46] See [36] and [37] above. [47] (2021) 24 HKCFAR 164. [48] CFI judgment at [36]. [49] See [29] to [30] and [44] above. 
Press Summary (English) Press Summary (Chinese) FACV Nos 20 & 21 of 2013 FACV No. 20 of 2013 IN THE COURT OF FINAL APPEAL OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION FINAL APPEAL NO. 20 OF 2013 (CIVIL) (ON APPEAL FROM CACV NO. 48 OF 2012) Between FACV No. 21 of 2013 IN THE COURT OF FINAL APPEAL OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION FINAL APPEAL NO. 21 OF 2013 (CIVIL) (ON APPEAL FROM CACV NO. 48 OF 2012) Between : Before: Chief Justice Ma, Mr Justice Ribeiro PJ,Mr Justice Tang PJ, Mr Justice Bokhary NPJ and Mr Justice Gummow NPJ Dates of Hearing: 23-24 June 2014 Date of Judgment: 17 July 2014 Chief Justice Ma: 1. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Ribeiro PJ: 2. This appeal arises out of an application for financial provision in matrimonial proceedings. It raises questions concerning the proper approach to a discretionary trust set up by the husband and issues concerning separation and estoppel in the context of ancillary relief where profits earned by the husband’s business substantially increased during the final years of the marriage. A. The course of events 3. Kay Kan Lai Kwan (“W”) was born in 1939 and Otto Poon Lok To (“H”) in 1940. They married on 6 January 1968 in England where they were each working and pursuing further qualifications, W as a nurse and H as an engineer. After returning to Hong Kong in 1969, they found employment and had three children, Karen,[1] Richard[2] and Heather.[3] In 1977, H started his own engineering business while W continued to work as a nurse. 4. After a slow start, H’s business enjoyed some success but then had a major setback in the mid-1980s when it almost failed. It was a very difficult time for them since most of their savings were lost in a stock market crash. They decided to emigrate to Canada in 1988. W remained in Canada with the children for several years while H returned to Hong Kong and started rebuilding the business. They became Canadian citizens in 1992. 5. H’s engineering business began to prosper and in 1994 it consisted of a number of operating companies based in Hong Kong. H decided to restructure the group and obtained advice from Messrs Deacons, a firm of Hong Kong solicitors. The plan adopted was to set up an off-shore corporation which would act as the holding company for the operating companies and to settle the shares of the holding company on the trustee of an off-shore discretionary trust. In a letter dated 21 March 1995 covering a draft of the trust deed, the solicitors recorded H’s purpose in the following terms: “As previously discussed, the principal purpose of establishing the trust is to provide for continuity in the ownership of the Analogue group of companies for the benefit of you and your family, to avoid a requirement for probate in respect of your interests in the Analogue group following your death and to relieve your estate of the liability to estate duty that would arise in respect of your shareholdings in the Analogue group.” 6. On 18 July 1995, Analogue Holdings Ltd (“Analogue”) was incorporated in Bermuda. H was its Chairman and Analogue became the holder of all the shares in the operating companies. 7. The trustee selected was HSBC International Trustee Limited (“the trustee”) and by a deed dated 25 July 1995 entered into between H as settlor and the trustee, “The Otto Poon Family Trust” (“the trust”) was established under the laws of Jersey. The property settled on trust consisted solely of shares in two private companies, Analogue and Realty Limited. The latter company initially owned a flat in Fulham Gardens, Pokfulam, which was later replaced by a house at Twin Bay Villas in Sai Kung, being residences where the parties lived. 8. Unfortunately, two tragedies befell the family. On 7 October 1995, the younger daughter Heather was killed in a road accident in the United States. Karen’s boyfriend, who was driving, was also killed and Karen was injured. H stated in evidence that he and W were devastated by their loss. W described herself as “totally heart-broken”.[4] According to H: “... Kay and I found it impossible to talk to each other or to turn to each other in our moment of grief. We both somehow bottled it up and I again threw myself completely and with renewed energy in my work and my projects.”[5] 9. Five years later, in September 2000, they suffered the second tragedy when Richard took his own life, leaping from the balcony of the family home in Fulham Gardens. H described the consequences of this event affecting the parties’ relationship as follows: “After Richard’s death, the rift between the parties simply became too wide to be bridged. Still we did not talk about our pain, our anger, our grief or our overwhelming and irreplaceable loss. Still we were unable to turn to each other in our darkest moments and by then we probably both had changed into virtual strangers to each other. There was simply nothing that we could say to each other anymore.”[6] 10. Reflecting this sad situation, the trial Judge, Deputy High Court Judge Carlson, stated: “As I have already indicated, Richard died in September 2000. The husband says, and I am satisfied that he is right, that this tragic event really marked the end of the marital relationship. There was no more warmth or connection between the parties notwithstanding which they continued to live together.”[7] 11. The Fulham Gardens flat, with its unhappy associations, was sold and on 22 January 2001, the parties purchased and moved into the house at Twin Bay Villas. H’s case is that while he and W lived there under the same roof, their marriage was (as the Judge found[8]) “a bare shell and nothing more”. As discussed below,[9] W successfully contested this finding in the Court of Appeal. 12. The Analogue Group’s development is reflected in its annual net profits which were as follows: 13. After strong initial results, profits fell during 2003 and 2004 (according to H due to the SARS epidemic) followed by 2005 when the Group made a significant loss. Thereafter, especially as from 2007, there was a steep increase in its profits. 14. In 2008, matters came to a head in the parties’ relationship. W discovered and confronted H about a long-term relationship he had maintained with Queenie Law, a younger woman employed in the Group. On 3 July 2008, H moved out of the Twin Bay Villas property. On 6 November 2008, W petitioned for divorce on the basis of one-year separation and consent.[10] And on 23 December 2008, the Trustee, at H’s request, transferred to W the shares which it held in Realty Limited, thereby transferring to her ownership of the Twin Bay Villas house. B. The proceedings and awards in the Courts below 15. W did not proceed with her petition. It was dismissed by consent and, on 6 February 2009, H issued his petition based on two years’ separation.[11] He affirmed that the parties had lived apart since February 2001.[12] W did not defend the proceedings and on 26 May 2009 and 1 September 2010 respectively, the decree nisi and the decree absolute were pronounced by the Court on the basis of H’s petition. The February 2001 separation date was also deposed to by both parties in their Form E financial statements filed in the ancillary relief proceedings. However, shortly before the start of the hearing, W made an affirmation[13] stating that the February 2001 date was incorrect and that separation had in fact occurred “some time in 2008”. The date of the parties’ separation and the extent to which W should be awarded a share of the profits of the business accruing after 2001 therefore became contentious issues. 16. Also controversial was the prior question of how the trust ought to be approached. H contended that the Court ought to treat Karen as if she had a “one-third interest” in the trust estate which should be protected from W’s claims. His case was that only two-thirds of the value of the shares should be treated as matrimonial assets. The trustee was joined as a party to the proceedings and adopted a position which was consistent with H’s contentions. B.1 The award of Deputy High Court Judge Carlson 17. The hearing before DHCJ Carlson took place on dates in October and December 2011.[14] The Judge accepted a valuation of $1,560,686,000 for the shares held by the trust, representing 84.63% of the issued shares in Analogue.[15] His Lordship found that additionally, H had assets to the value of $46,052,707 and W, assets worth $58,259,660.[16] He also found that they jointly owned a flat in Westland Gardens (H agreeing to transfer his 50% share valued at $7,500,000 to W). Those figures are not in dispute. It was agreed that each party would retain other properties held in their respective names without bringing such properties into account. 18. The Judge held that the trust assets were not a “resource” of H[17] but nevertheless decided that $1,040,457,300 representing two-thirds of the value of the Analogue shares held on trust should be ascribed in equal shares to the parties as forming the major part of their matrimonial assets available for distribution.[18] Netting off the parties’ respective assets, the Judge concluded that if the principle of equal sharing[19] were to be applied without modification, H would be required to pay $508 million to W.[20] 19. However, for a combination of reasons involving estoppel, a perceived need to preserve the Analogue Group’s liquidity and a finding that the parties had lived separate lives during the last ten years of the marriage, his Lordship held that there ought to be a departure from the equal sharing principle. In the result, he ordered H to pay W a lump sum of $370 million representing 72.83% of the sum of $508 million he had prima facie arrived at by applying the equality principle.[21] B.2 The award of the Court of Appeal[22] 20. Cheung JA, with whom the other members of the Court agreed, upheld the Judge’s decision that only two-thirds of the value of the shares held on trust should be regarded as matrimonial assets available for distribution.[23] However, the Court of Appeal disagreed with the Judge’s reasons relating to liquidity and separation, substituted a separation date of February 2007 and held that there was no basis for departing from the equality principle.[24] 21. Taking the parties’ matrimonial assets as totalling $1,144,800,000,[25] their Lordships awarded to W 50% of that amount, resulting (after netting off adjustments) in a lump sum award of $510,400,000 in W’s favour, an increase of $140,400,000 over the Judge’s award of $370 million. 22. The Court of Appeal ordered the award to be paid in three tranches: $250 million within one month and two tranches of $130 million each to be paid on or before 1 March 2014 and 1 March 2015 respectively. The Court was told that the first two tranches have duly been paid. B.3 This appeal 23. Pursuant to leave granted by the Appeal Committee,[26] both parties have brought appeals against the Court of Appeal’s judgment. W challenges its decision upholding the Judge’s conclusion that only two-thirds of the value of the shares held on trust should be regarded as a financial resource available to H. W contends that the entire value of those shares constitutes such a resource and that the “yardstick of equality” should be applied to the full value and not merely to two-thirds of that amount. 24. H, on the other hand, disputes the Court of Appeal’s substitution of a February 2007 separation date for the Judge’s finding that separation occurred in February 2001. He challenges the Court of Appeal’s conclusion that there was no basis for departing from the equal sharing principle and argues that W should be excluded from sharing in the increased profits accruing after February 2001. C. Is the discretionary trust is a financial resource available to H? C.1 The legal test 25. The Court’s jurisdiction to grant ancillary relief, including its power to make financial provision orders,[27] arises under the Matrimonial Proceedings and Property Ordinance (“MPPO”).[28] Section 7(1)(a) provides that in deciding whether and in what manner to exercise those powers: “It shall be the duty of the court ... to have regard to the conduct of the parties and all the circumstances of the case including the following matters, that is to say – (a) the income, earning capacity, property and other financial resources which each of the parties to the marriage has or is likely to have in the foreseeable future...” 26. The terms of the trust are considered more closely below.[29] For the present, it suffices to note that apart from being the settlor, H was named as protector of the trust and also as a member of the class of potential beneficiaries.[30] The question which arises is whether the trust and its assets should be regarded as a “financial resource” which H has or is likely to have in the foreseeable future within the meaning of section 7(1)(a). In addressing that question, the Court is engaged in taking the first step in deciding an ancillary relief claim. As pointed out in LKW v DD,[31] the first step is: “... to ascertain the financial resources of each of the parties calculated as at the date of the hearing. ... At this stage, the court need not attempt to distinguish between matrimonial and non-matrimonial property, that being an exercise best undertaken (if necessary) when considering distribution of the assets.” 27. Section 7(1)(a) is closely modelled on the equivalent English provision, making English decisions on section 25(2)(a) of the Matrimonial Causes Act 1973[32] helpful and persuasive. The approach adopted in England and Wales was laid down by the Court of Appeal in Charman v Charman,[33] a case concerning a discretionary trust situated in Bermuda. Wilson LJ formulated the test as follows: “Superficially the question is easily framed as being whether the trust is a financial ‘resource’ of the husband for the purpose of section 25(2)(a) of the Matrimonial Causes Act 1973, as substituted by the Matrimonial and Family Proceedings Act 1984, section 3 . But what does the word ‘resource’ mean in this context? In my view, when properly focused, that central question is simply whether, if the husband were to request it to advance the whole (or part) of the capital of the trust to him, the trustee would be likely to do so.”[34] 28. This was confirmed to be the test in Charman v Charman (No 4),[35] and has been treated as established in English law. A helpful elaboration can be found in the judgment of Lewison J in the English Court of Appeal in Whaley v Whaley:[36] “... a discretionary beneficiary has no proprietary interest in the fund. But under section 25 the court looks at resources; not just at ownership. Thus whether a beneficiary under a discretionary trust has a proprietary interest is not relevant. The resource must be one that is ‘likely’ to be available. This is the origin of the ‘likelihood’ test. No judge can make a positive finding about the future: the best that can be done is to assess likelihood. What is relevant is the likelihood of the trust fund or part of it being made available to him, either by income or capital distribution. If the husband were to ask the trustees to advance him capital, would the trustees be likely to do so: Charman v Charman [2006] 1 WLR 1053; A v A [2007] 2 FLR 467, 499? The question is not one of control of resources: it is one of access to them.” 29. Charman v Charman was cited with approval by this Court in KEWS v NCHC.[37] That was a case involving parental financial support rather than a discretionary trust in which a similar test (asking what is the likelihood of such financial assistance continuing) was adopted. It is right that the Charman test should be adopted in this jurisdiction since the issue arising in cases like the present and since the Hong Kong provision are materially identical. To decide whether a discretionary trust is a financial resource of one of the parties, the Court asks whether, if that party were to request the trustee to advance the whole or part of the capital or income of the trust to him or her, the trustee would, on the balance of probabilities, be likely to do so. 30. The Judge was therefore wrong to hold that Charman was inapplicable – a conclusion which he reached apparently because: “Mr Charman’s behaviour was completely different to that of this husband who has always viewed his responsibilities to his wife (even after their divorce) and his children, now only Karen, with the utmost concern for their best interests.”[38] 31. The Charman test provides the approach to deciding whether a discretionary trust forms part of a party’s resources. A comparison of H’s conduct with Mr Charman’s conduct is not relevant to that question. 32. The Court of Appeal accepted the applicability of the Charman test.[39] However, the way in which it actually applied that test requires further discussion below.[40] C.2 Acceding to the wishes of a beneficiary, settlor or protector 33. It should be clearly understood that the Charman test does not postulate any impropriety on the trustee’s part. This was noted in Charman v Charman itself where Wilson LJ stated: “A trustee – in proper ‘control’ of the trust – will usually be acting entirely properly if, after careful consideration of all relevant circumstances, he resolves in good faith to accede to a request by the settlor for the exercise of his power of advancement of capital, whether back to the settlor or to any other beneficiary.”[41] 34. In like vein, Lloyd LJ stated: “There is no doubt that trustees can properly take into account any expression of wishes, formal or otherwise, on the part of the settlor as to how they should exercise their discretionary powers, and indeed that they should have regard to any such wishes expressed to them.”[42] 35. In In Re Esteem Settlement,[43] the Jersey Royal Court[44] provided a helpful explanation of how trusts like the present are generally administered and how trustees may properly arrive at distribution decisions on the basis of requests received: “[165] ...In our judgment there is nothing untoward in beneficiaries making requests of a trustee as to the investment of the trust fund, the acquisition of properties for them to live in or for the refurbishment of properties in which they already live. In our judgment many decisions of this nature are likely to arise because of a request by a beneficiary rather than because of an independent originating action on the part of a trustee. The approach that a trustee should adopt to a request will depend upon the nature of the request, the interests of other beneficiaries and all the surrounding circumstances. Certainly, if he is to be exercising his fiduciary powers in good faith, the trustee must be willing to reject a request if he thinks that this is the right course. But when a trustee concludes that the request is reasonable having regard to all the circumstances of the case and is in the interests of the beneficiary concerned, he should certainly not refuse the request simply in order to assert or prove his independence. His duty remains at all times to act in good faith in the interests of his beneficiaries, not to act against those interests for improper reasons. [166] In our judgment, where the requests made of trustees are reasonable in the context of all the circumstances, it would be the exception rather than the rule for trustees to refuse such requests. Indeed, as Mr Journeaux accepted, one would expect to find that in the majority of trusts, there had not been a refusal by the trustees of a request by a settlor. This would no doubt be because, in the majority of cases, a settler would be acting reasonably in the interests of himself and his family. This would particularly be so where there was a small close-knit family and where the settlor could be expected to be fully aware of what was in the interests of his family. Indeed, in almost all discretionary trusts, the settlor provides a letter of wishes which expresses informally his desires in relation to the administration of the settlement. Furthermore he may change his wishes from time to time. In our judgment it is perfectly clear that trustees are entitled ... to take account of such wishes as the settlor may from time to time express provided, of course, that the trustees are not in any way bound by them. The trustees must reach their own independent conclusion having taken account of such wishes. [167] ... A lack of any refusal may of course be indicative of the fact that trustees have abdicated their fiduciary duties and are simply following the wishes of the settlor without further consideration. But, as mentioned above, a lack of any refusal may be equally consistent with a properly-administered trust where the trustees have in good faith considered each request of the settler, concluded that it is reasonable and concluded that it is proper to accede to such requests in the interests of one or more of the beneficiaries of the trust. But one does not start, as at times seems to have been the plaintiffs’ case, with an attitude that it is very surprising and worthy of criticism that the trustee acceded to all Sheikh Fahad’s requests. On the contrary, as the Privy Council said in Letterstedt v Broers (9 App Cas 371), trustees exist for the benefit of beneficiaries and it is in our judgment very common that trustees will have perfectly properly acceded to all the requests of a settlor without in any way abdicating their fiduciary duties and responsibilities.” 36. Of course there may be cases where the trust is a sham and a settlor controls the purported trustee who does not act in good faith as a trustee when acceding to the settlor’s or beneficiary’s wishes. Or a trust may be constituted with provisions which result in the settlor being found not to have effectively divested himself of the trust property. That occurred for instance in Tasarruf Mevduati Sigorta Fonu v Merrill Lynch Bank and Trust Company (Cayman) Limited,[45] where the settlor retained a power to revoke the trust and was held by the Privy Council to have thereby retained rights tantamount to ownership, enabling a judgment creditor to levy execution (via a receiver by way of equitable execution to whom the power of revocation was ordered to be delegated) on the trust assets.[46] 37. In the present case, W accepts that there is no question of the trust being a sham and there is no suggestion of any impropriety on the part of the trustees. The question is whether, looking at all the evidence, the Court should be satisfied that if H were to request the trustee to advance the whole or part of the capital or income of the trust to him, the trustee, acting in accordance with its duties and after having regard to all relevant considerations, would on the balance of probabilities be likely to do so. C.3 The judgments below on the discretionary trust 38. Unfortunately, DHCJ Carlson appears to have fallen into error in considering it somehow necessary for W to show that the trustee was prepared to act in breach of trust before the Court would treat the trust estate as a resource likely to be available to H. 39. After acknowledging that on the evidence the trustee had shown H “great deference”, treating him and the Analogue Group as valued customers of HSBC;[47] that H “does have a very powerful position within the Trust by having retained the power to remove the trustee”;[48] and that Mr Lynn[49] had been able “to point to various aspects of the relationship of the trustee with the husband which may be said to demonstrate his strong influence over the affairs of the trust and of the way that the trustee has in the past done what he has asked of it”, his Lordship continued: “... what Mr Lynn has signally failed to achieve is to show that the trustee has acted in breach of trust in the way that it has treated the rights of the other beneficiaries compared to its treatment of the husband.”[50] 40. The Judge appears erroneously to have thought that Mr Lynn’s submissions, advocating application of the Charman test, amounted to an argument that the trust was a sham – which had been disavowed on W’s behalf. Thus, his Lordship observed: “Mr Lynn really has had a very difficult row to hoe. He has disavowed the notion of a sham trust and yet he has had to make submissions which plainly have their place in support of an argument that the Trust is indeed a sham.”[51] 41. The learned Judge evidently misunderstood the Charman test, leading him to conclude that “it would be wrong to hold that this shareholding (the trust assets) can be viewed by me as a resource of the husband” and to distinguish Charman (as I have previously indicated) on the basis of that the quality of H’s conduct differed from that of Mr Charman.[52] 42. His Lordship’s subsequent reasoning is not easy to follow. Notwithstanding his rejection of the Charman test and his decision that the trust assets were not a resource, he nevertheless took two-thirds of the value of the shares held on trust as representing the major part of the matrimonial assets of H and W. That decision appears to be explained by his Lordship’s statement that while it was incorrect to regard the “three discretionary beneficiaries” – that is, H, W and Karen – as each having a fixed one-third share, it was nevertheless his view that: “... the trustee in exercising its duty of safeguarding the interests of all the beneficiaries fairly as between themselves would not ... countenance any disposition which results in Karen’s interest being diluted to below something in the order of 1/3.”[53] 43. It therefore appears that the Judge again fell into error in treating Karen as “a beneficiary” with an “interest” to be protected as against the other beneficiaries. In the joint judgment of Gummow and Hayne JJ in Kennon v Spry,[54] the description of the position of the wife in that case is apposite for describing Karen’s position in the context of the present trust: “The wife was an eligible object of benefaction of the Trust. She was one of the class of ‘beneficiaries’ identified in [the Instrument]. The use in that document of the term ‘beneficiaries’ was inapt insofar as it suggested the existence of any vested beneficial interest in the assets held on the trust of [the Instrument]. Dr Hardingham correctly identified the wife as one of the class of objects of the discretionary power conferred upon the trustee ... Furthermore, as an object of these powers the wife had a right in equity to due administration of the Trust. The existence of such a right did not depend upon entitlement to any fixed and transmissible beneficial interest in the trust fund. The right of the wife was accompanied at least by a fiduciary duty on the part of the trustee, the husband, to consider whether and in what way he should exercise the power conferred ...” 44. Karen, like the wife there mentioned, did not have any fixed beneficial interest in the trust fund, but was merely owed limited fiduciary duties of the kind described. Moreover, as we shall see,[55] the trust deed expressly authorizes the trustee in its absolute discretion to appoint capital and income to any one member of the class of eligible objects to the exclusion of the others. It was therefore erroneous to think that the trustee had a duty to safeguard the “interests” of “all the beneficiaries fairly as between themselves” so that the trustee “would not countenance” a disposition diluting Karen’s “interest” to the stated extent. It was evidently on that incorrect basis that the Judge ascribed only two-thirds of the value of the trust shares to the matrimonial estate. 45. Even though the Court of Appeal acknowledged the applicability of Charman v Charman,[56]Cheung JA, writing for the Court, appears to have accepted the Judge’s erroneous approach: “In my view the Judge was correct when he only took into account a two-third interest in the Family Trust as the resource that would likely be available to the husband for the purpose of assessing the matrimonial assets. In so doing the Judge had expressly recognized that the husband’s view that the three of them each has a one-third share is not a correct appreciation of the position. However, the Judge considered (and I agree) that the Trustee in exercising its duty of safeguarding the interest of all the beneficiaries would not countenance any disposition which results in K’s interest being diluted below one-third.”[57] His Lordship added: “This is not a case where the husband said that the whole of the Family Trust should not be taken into account. On the contrary he accepted that it should be taken into account but only in a way that will not adversely affect K’s interest.”[58] 46. The Court of Appeal identified as a further ground for supporting the Judge’s conclusion “the stand taken by the Trustee and also by the Jersey law experts”[59] and proceeded on the basis “that the matrimonial assets include two-thirds and not the whole of the value of the Family Trust”.[60] 47. With respect, I can see nothing in the Jersey law opinions to suggest that the trustee is bound to protect a one-third “interest” in the trust attributable to Karen. I turn then to deal with “the stand taken by the trustee”. C.4 Submissions of counsel 48. The trustee has instructed solicitors and counsel to appear throughout these proceedings.[61] However, no evidence was tendered on the trustee’s behalf. Instead, at the hearing before DHCJ Carlson, two notes[62] containing opening and closing submissions of counsel setting out the trustee’s position were provided to the Court. Two paragraphs, repeated in each of the two notes, were relied on in particular, as is re-iterated in the trustee’s printed case[63] lodged in this appeal. They were also given prominence on H’s behalf. 49. The first paragraph addresses W’s alternative claim for the trust to be varied pursuant to MPPO section 6(1)(c),[64] which I have not so far mentioned, but which relevantly indicates the view taken by the trustee of Karen’s “interest”. The submission states: “Although the Court does have the power to vary the Trust under s 6 MPPO, in so far as the 1st Respondent [W] seeks to have 50% of the shares held under the Trust transferred to her, the Trustee has obvious concerns that this will not be in the interests of the beneficial class as a whole, particularly Karen as an innocent third party, and also the Polytechnic and other potential beneficiaries. This should also be something that the Court is concerned about and gives careful consideration to before exercising its discretion to vary. See C v C (Ancillary Relief : Nuptial Settlement) [2004] EWHC 742.”[65] 50. This paragraph obviously does not purport to address the Charman test but merely expresses reservations against any variation order that the Court might make to vest half of the trust fund in W. It also introduces the curious description of Karen as “an innocent third party”. She is, of course, properly seen as a member of the class of eligible objects of the trustee’s discretionary powers of appointment with no vested beneficial interest in the trust fund, as described above. 51. The second paragraph, heavily relied on by Ms Anita Yip SC[66] on H’s behalf, states as follows: “Should the Court consider that this is an appropriate case where ‘judicious encouragement’ should be applied the Trustee’s position on this is a neutral one (the Court has seen that there have been various distributions of dividends to the Petitioner in the past) but only to the point as set out above where innocent third party beneficiaries will be adversely affected. Accordingly if an Order applying judicious encouragement is made as part of the final AR [ancillary relief] Order made for the 1st Respondent [W], the position of the Trustee is that no more than 1/3 of the value of the Trust should be realized to meet such an order. This will ensure that the interests of all beneficiaries are preserved.”[67] 52. Four points may be made regarding that paragraph. The first is that the trustee has not answered the Charman question. It has not addressed the likelihood of its acceding to a hypothetical request by H to advance to him the whole or part of the capital or income of the trust. Instead, the trustee has taken it upon itself to engage in advocacy as to the kind of order that the Court should avoid making. 53. Secondly, Karen is once again apparently assumed to have some sort of accrued one-third interest on the basis that she is an “innocent third party beneficiary” whose interest must be preserved, a proposition which is unwarranted. 54. Thirdly, the paragraph focuses on the award which the Court might make on the basis of “judicious encouragement”. But the question we are concerned with does not relate to the Court’s award. It involves ascertaining the extent of the parties’ matrimonial assets as the first step in determining ancillary relief. The Court is asking whether the trust fund should be regarded for the purposes of section 7(1)(a) as a “financial resource” of H. If the Court answers “Yes”, it does not thereby decide to award the whole of the fund to W. It proceeds next to consider the subsequent steps in the ancillary relief exercise, including the question whether and to what extent it should depart from the equal sharing principle. 55. That leads to the fourth point. A decision by the Court that the trust fund is a resource available to H does not disable the trustee from appointing one-third of that fund in favour of Karen. Even if there was to be no departure from the sharing principle, satisfying the award in W’s favour out of the trust assets would still leave half of the fund intact, enabling the trustee if so advised to make a one-third appointment in Karen’s favour out of that balance. 56. The decision to recognize only two-thirds of the trust fund as H’s financial resource therefore does not, in my view, find any support in “the stand taken by the trustee” which, with respect, is legally flawed and not to the point. 57. The foregoing reasons also undermine the central argument advanced on this issue on H’s behalf. His written case recalls that: “Husband asserted that given the presence of the Trustee, there was no need for the court to debate hypothetically what was the ‘likelihood’ of the Trustee advancing all trust assets to the Husband, granted that Wife accepted the Trustee was acting properly. In this case, the Trustee could and did unequivocally articulate the way they would exercise the discretion on the question of extent of availability.”[68] 58. It goes on to state: “With the participation of the Trustee in the present case, given Wife’s concession that this was not a sham trust and the Trustee was acting properly in discharging its duty, the ‘likelihood test’ is simply not engaged. The Trustee has answered the question: if requested, would the trustee advance the whole of the trust assets to H? The answer was ‘no’, they were not going to countenance an advancement which would result in a termination of Karen’s interests. As submitted, this was well within the terms of the trust and within the trustee’s powers.”[69] 59. Those propositions are incorrect. The trustee did not in fact address the Charman test and the premises underlying those propositions are legally and logically flawed. C.5 The proper application of the Charman test 60. As the Charman test has not been properly applied, it falls to this Court to decide whether it should be satisfied that if H were to request the trustee to advance the whole or part of the capital or income of the trust to him, the trustee, acting in accordance with its duties, would on the balance of probabilities be likely to accede to that request. In making its assessment the Court is able to consider the creation and terms of the trust; H’s letters of wishes; the nature of the trust assets; and previous distributions made by the trustee. As was held in KEWS v NCHC,[70] the Court looks at the reality of the situation and regards past conduct as a useful guide. C.5a The creation and terms of the trust 61. As pointed out above, when H created the trust as settlor, he caused himself to be named as protector and also as a potential beneficiary. 62. The deed authorizes the trustee to act through managers and agents[71] and in practice, the trust is operated on the Jersey trustee’s behalf by HSBC entities in Hong Kong.[72] The trustee is given power in its absolute discretion from time to time to appoint capital and to distribute income to any eligible object of its discretion to the exclusion of the others.[73] It also has power in its absolute discretion to appoint additional persons to become beneficiaries.[74] 63. As settlor and protector, H’s wishes, conveyed from time to time in letters of wishes, are taken into account by the trustee in considering the exercise of its discretion. As protector, H has the power to remove the trustee and appoint a new trustee.[75] Furthermore, certain powers conferred on the trustee are only exercisable with H’s consent as protector. These include the power to remove potential beneficiaries[76] and the power to alter, revoke or add to any provisions of the deed.[77] 64. While the settlement constitutes the trustee the 84.63% majority shareholder of Analogue, the deed authorizes the trustee to leave the administration, management and conduct of Analogue’s business to its directors and managers, relieving the trustee of any responsibility in that connection.[78] 65. Relevant to the Charman test, the following features may therefore be noted: (i) In making himself protector, H has reserved to himself important powers, including the power to remove the trustee, obviously intending his views as to the administration of the trust to be given great weight. (ii) In making himself a potential beneficiary, he intended to benefit from the trustee’s distributions of capital and income. (iii) He intended the trustees to have only the passive role of a shareholder, leaving it to him to run the Analogue Group. (iv) Karen is one of the discretionary objects and has no vested beneficial interest in the trust assets. (v) There is no impediment to the trustee appointing the whole or part of the capital or income to a single beneficiary to the exclusion of all the others. C.5b H’s power to replace the trustee 66. One controversy developed in argument concerns the significance of H’s power as protector to replace the trustee. In answer to W’s argument that such power indicates a high degree of control over the trustee, H replied that the power is “toothless” because (as W’s own Jersey law expert states) “the office of protector [is] fiduciary” and not personal to the protector.[79] 67. It is incorrect to state categorically that a protector must be a fiduciary or that a power to replace a trustee is necessarily fiduciary. The position is explained in an article by Matthew Conaglen and Elizabeth Weaver entitled “Protectors as fiduciaries: theory and practice”.[80] As the learned authors point out, the term “protector” is not a term of art and generally “signifies little more than that a person who is not the (or a) trustee has been granted a power affecting the operation of the trust”.[81] One cannot assume that such a power is held in a fiduciary capacity requiring it to be exercised only in the interests of others to the exclusion of the donee of the power. Whether a power to replace a trustee is fiduciary depends on the construction of the trust instrument.[82] 68. In the present case, H’s power as protector to replace the trustee[83] is not subject to any express limitations. Even if the deed were construed to require that power to be exercised in the interests of the potential beneficiaries, that class includes H himself. That makes it highly debatable whether its exercise is subject to fiduciary duties and, if so, what such duties are. As the learned authors of the article point out: “Where the purpose and intention of the settlor was that the protector was also to be able to benefit under the trusts, the courts will usually respect that intention and not find fiduciary obligations which would disable the protector from acting in his own interest, although they might still hold that the protector owes limited or qualified fiduciary duties to consider the exercise of his powers on a regular basis. On the other hand, the cases show that powers which impinge upon the trustees’ position as ‘ultimate guardians of the trust’ are likely to be treated as fiduciary, to some degree at least, so that the court can retain a supervisory jurisdiction.”[84] 69. The question whether the power of replacement is fiduciary does not directly arise in the present case. It is only likely to become relevant in a practical sense if a conflict or lack of harmony were to develop between H and the trustee, such discord often being a legitimate ground for replacing a trustee.[85] Everything indicates that there is and has always been the fullest cooperation between the trustee and H so that the power is of no immediate significance. Nonetheless, it is relevant to note for the purposes of the Charman test that H (as the trustee undoubtedly is aware) has the ultimate power to replace the trustee which, in the event of a serious disagreement as to the operation of the trust, might well be lawfully exercisable. C.5c H’s letters of wishes 70. As was pointed out in the Esteem Settlement case, it is usual for the settlor of a discretionary trust to provide letters of wishes in relation to the administration of the trust, changing his wishes from time to time. Trustees are entitled to take account of such wishes while not being bound by them. In the present case, prior to the hearings below, H had issued five letters of wishes, each superseding the previous.[86] They all followed the same pattern. 71. Thus, the letter of wishes dated 14 May 2010 has the following aspects. Having identified the Analogue shares as constituting the trust fund, it goes on express H’s desire: (a) that “following [his] death, Mr Fong Chun Yau shall be appointed as the Managing Director of [Analogue] and the Analogue Group ... for a period of two years after which the position of Managing Director shall be decided by the shareholders at their discretion;” (b) that after his death, Karen should be appointed protector; (c) that the trustee should continue to hold its present interests in Analogue and that “following [his] death the balance of the Trust Fund ... be divided into three equal parts or shares” one for the benefit of “my soon-to-be former wife, Kay”, one for Karen and one for three named colleagues in the Analogue Group, with The Hong Kong Polytechnic University as the residuary beneficiary if Kay and Karen should predecease H without issue from Karen. 72. H added a general direction that: “In considering whether and how to exercise your powers and discretions, it is my wish that you should consult with me during my lifetime and thereafter with my daughter, Karen. I would also like you to consider any suggestions put to you by me or, after my death, by Karen and, if you consider them wise, to act upon such suggestions.” 73. It is clear that the trustee has accorded H’s desires great weight. When in a letter of wishes signed on 22 January 1996, H indicated a wish that The Hong Kong Polytechnic University should be a potential beneficiary, the trustee executed a deed four days later appointing the University to the class of eligible objects. Then, after H moved out of Twin Bay Villas and W issued her divorce petition on 6 November 2008, the trustee, at H’s request, transferred all its shares in Realty Limited (which owned Twin Bay Villas) to W by an instrument dated 23 December 2008. And when applying for directions from the Jersey Court[87] regarding its participation in these proceedings, the trustee referred to the letter of wishes of 14 May 2010. 74. The letters of wishes confirm that the trust is principally intended to function as a substitute for H’s will in relation to the Analogue shares. They focus on what is to happen following his death. Just as a testator is free to change his will, it was evidently understood that any changes desired by H would be treated with respect by the trustee. 75. What also emerges from the letters is that it is plainly not H’s intention that the trust assets should be divided up or that he should relinquish his position of influence during his lifetime. He only contemplates such matters occurring after his death. He makes it clear that he should be consulted by the trustee and that it should consider any suggestions which he might put to it during his lifetime. Karen is only to be consulted and her suggestions given weight after his death. There is no suggestion that Karen or anyone else should acquire any vested beneficial interest in the trust assets during H’s lifetime.[88] 76. In the context of Charman v Charman, the terms of the trust and the letters of wishes indicate that H has always intended to occupy and in fact occupies a dominant position in relation to the administration of the trust. The trustee is plainly likely to treat any request he might make for a distribution with great deference. C.5d The nature of the trust assets and distributions made by the trustee 77. It is important to note that the sole asset of the trust is the 84.63% parcel of shares in Analogue.[89] The underlying group of companies was described by the expert valuer as consisting of “13 electrical and mechanical engineering companies employing around 1,500 staff and achieving a turnover in excess of HKD2.5 billion a year”.[90] 78. It is not surprising that the trustee is assigned an entirely passive role as shareholder and plays no part in the management of such a group. It is even relieved of any obligation to exercise the voting powers or rights conferred by its shareholding.[91] A trustee in such a situation is obviously in no position to second guess any requests made by H as to how those shares should be voted or his wishes touching on the operations or resources of the Group. The trustee would be entitled to assume that acceding to such wishes would generally be in the best interests of the underlying companies and thus of the potential beneficiaries. H is thus not only the Group’s managing director, but also acts effectively as its controlling shareholder. Indeed, the letters of wishes show that he seeks to influence the management of the Group even after his death by stipulating who should then be appointed managing director. 79. This is not a trust in which the trustee manages and invests the trust fund to produce income. The only income of the trust consists of dividends declared by Analogue and the decision whether, when and how much to declare by way of dividend is taken by the Analogue board controlled by H. Unless a dividend is declared, the trust has no income. And when a dividend has been declared, the trustee has, quite understandably, invariably complied with H’s wishes as to how the money which he has sent its way should be applied – namely, distributed to H himself. 80. The following distributions, all made to H personally following declarations of dividends, occurred prior to the ancillary relief proceedings: 81. The modus operandi is revealing. On 26 March 2001 for example, H wrote to the trustee as Analogue’s managing director announcing that a dividend of $833.3333 per share had been declared and sent it a cheque for $8,463,333 being its entitlement as holder of 10,156 shares. On the same day, the trustee wrote to H stating: “... we intend to exercise our trustee power to make a distribution from the trust fund to you as beneficiary thereof” and on 29 March 2001, the distribution of $8,463,333 was credited to H’s account. On 9 April 2001, H paid the trustee $5,000 as its fee for “arranging the distribution”. 82. The procedure was later streamlined. By letter dated 10 July 2008, Analogue again informed the trustee that a dividend of $833.3333 per share had been declared, entitling the trustee to payment of $8,463,333 as shareholder. The letter went on to state: “As Mr Otto Poon wishes to have the said dividend to be received by HSBC International Trustee Ltd distributed to him, we would therefore appreciate if you could kindly send us your letter of instructions to pay the said dividend directly into Mr Poon’s bank account.” 83. The trustee obliged. By letter dated 16 July 2008 it instructed Analogue to pay the dividend into H’s account. H therefore caused the dividend to be declared and notionally distributed to him by the trust without the money ever passing through the hands of the trustee. C.5e Satisfying the awards of the courts below 84. A crucial indication relevant to the Charman test is the glaringly important fact that H has indeed been able to draw on the trust as his resource for the purpose of satisfying the awards made in favour of W. 85. On 10 February 2012, DHCJ Carlson awarded $370 million to W and ordered H to pay a first instalment of $250 million. This led H to cause Analogue to declare a dividend on 12 April 2012 of which $262 million was received by the trust and, according to the trustee,[92] “advanced as a loan” to H to enable him to satisfy the Court’s order which he did on 30 May 2012. 86. On 25 March 2013, as we have seen, the Court of Appeal increased the award in W’s favour and ordered H to pay a lump sum of $510 million by an initial instalment of $250 million followed by two equal instalments of $130 million on or before 1 March 2014 and 2015 respectively. The Court was informed that the first two instalments have duly been paid. 87. The trustee also recently disclosed that a fresh letter of wishes dated 12 November 2013 had been received from H. It follows the pattern of identifying H’s choice of managing director and protector following his own death and stating his wishes as to how the balance of the fund should then be divided. H goes on to request the trustee to advance to him from the trust assets the sums needed to satisfy the Courts’ awards in the following terms: “My former wife Kay should receive in aggregate an amount equal to the amount awarded by the Hong Kong Court under [DHCJ Carlson’s judgment] (“the Judgment”) as varied and finally determined on appeal by the Court of Appeal or the Court of Final Appeal (as the case may be) less any amounts already advanced or paid to her under or in respect of these proceedings. For the purpose of calculating amounts paid to Kay under the Judgment, amounts advanced to myself by way of distribution or loan and paid by me to Kay shall be deemed as amounts paid by the Trustee to Kay. On payment of the amount awarded to Kay under the Judgment (as varied and finally determined on appeal as aforesaid) it is my request that Kay should be excluded from further benefit under the Trust.” 88. Given that the payments already made by H were funded by the trustee’s distributions, there is every reason to believe that the trustee will comply with H’s request to meet the Court’s award as determined on final appeal out of the trust assets (no doubt in turn funded by dividends which H will cause Analogue to declare). There could, in my view, be no clearer evidence of the overwhelming likelihood that the trustee, acting in accordance with its duties, would if requested by H, advance the whole or part of the capital or income of the trust to him. C.6 Conclusion as to the discretionary trust 89. It follows that the entire trust fund valued at $1,560,686,000 should be regarded as a financial resource available to H for the purposes of section 7(1)(a). The Judge and the Court of Appeal were wrong to confine the parties’ matrimonial assets represented by the trust shares to two-thirds of their assessed value. 90. The combined matrimonial assets of the parties to be taken into account in assessing ancillary relief therefore consist of (i) the aforesaid sum of $1,560,686,000, plus (ii) H’s assets of $46,052,707, plus (iii) W’s assets of $58,259,660, plus (iv) $15 million representing the value of the flat in Westland Gardens. Those figures total $1,679,998,367. Applying the “yardstick of equality”, 50% of that sum comes to $839,999,183.50, which may be rounded up to $840 million as the starting-point. If the equal sharing principle were to be applied without modification W would be entitled to payment of that sum less the value of the matrimonial assets already in her possession (her own assets and the Westland Gardens flat totalling $73,259,660). That would leave her an entitlement to $766,740,340 of which H has already paid $380 million in partial satisfaction of the awards below. If these tentative calculations are correct, that would mean that a balance of $386,740,340, in round figures $386,700,000, remains to be paid unless the sharing principle is departed from resulting in a lower award. 91. Mr Lynn devoted considerable efforts on the alternative argument that the Court should exercise its powers under MPPO section 6(1)(c) to vary the trust, settling half of its assets on W in case the Court decided that the trust fund is not a financial resource available to H. In the light of the conclusion I have reached, that question does not arise. D. Should there be a departure from the principle of equality? 92. I turn then to consider the second main question in this appeal: Should there be a departure from the equality principle on the basis that the parties’ separation ought to deprive W of an equal share in the increased Group profits which accrued after February 2001? 93. Although he was later persuaded to depart from equality, DHCJ Carlson began his judgment by describing this marriage as a paradigm case for adopting the sharing principle: “On any view, the ‘yardstick of equality’ as a starting point, which Hong Kong matrimonial courts are now required to give the most serious consideration in applications such as this, must have particular resonance in this case where the wife, after a very long marriage in which she has fully played her part as wife, mother and supporter of the husband in his business successes, comes to court as worthy of the highest consideration. Prima facie, if ever the yardstick of equality is to be wielded it must be in a case such [as] this.”[93] 94. His Lordship was plainly right to take that view, at least for the greatest span of the marriage. W’s contribution to the family, bringing up the children, settling in Canada to acquire Canadian citizenship, making economies and taking up arduous nursing jobs (sometimes working double shifts) while offering support to H through good and bad times over decades, presents a classic example of a marital partnership that justifies equal sharing of the matrimonial assets upon the marriage coming to an end. 95. H recognizes this, at least to some extent. In his first affirmation,[94] he states: “It was undoubted that during the first 10 years of the marriage both parties contributed equally and unstintingly towards the marriage and the family. Kay worked, looked after the children and achieved savings. I worked hard on building up the business and the first two years were really scary. There was nothing happening and no business coming in for the first two years. Then everything started coming together, the business took off and we began to enjoy some stability. .... We were even able to send Karen and Richard to boarding schools in England.” 96. The reference to “the first 10 years” must surely be an understatement. As we have seen, the business only really began to take off in the 1990s, so that H appears to be referring to a period which runs into the third decade of the marriage. Later in the same affirmation,[95] he states more broadly: “... I fully recognise Kay’s full contribution to the marriage and her entitlement to a fair share of all my assets including the companies” in the context of a request that the Court should not make an award which involves transferring Analogue shares to W. 97. This is therefore a case where the equality principle should be adopted not merely notionally, but as a starting-point which reflects a fair basis for sharing the matrimonial assets upon dissolution of a 40 year marriage. As was stressed in LKW v DD,[96] the “yardstick of equality” is applied “with a view to eliminating insidious discrimination and promoting fairness” and “should be departed from only for good, articulated reasons”. D.1 The Judge’s reasons for departing from equality 98. The course of the divorce proceedings has been described in Section B above. It will be recalled that they were undefended and that the decree was granted on the basis of H’s petition grounded on two years’ separation. It will also be recalled that both parties affirmed that separation took place in February 2001. In considering whether there should be a departure from the equality principle, the Judge held that W was estopped from denying that separation had occurred in 2001. He held that W could not be heard to say that they had only been separated since 2008 because that was less than two years immediately preceding presentation of the petition.[97] 99. On that basis, his Lordship concluded that there ought to be a departure from equality for two main reasons. First, he held that since the parties had lived separate lives as from February 2001, “some adjustment should be made in the husband’s favour” so that W was not to share equally in the substantial profits of the Analogue Group which had accrued after that date.[98] His Lordship did not quantify that adjustment since he regarded his second reason for departing from the sharing principle, namely, “the issue of the company’s liquidity” as “a very much stronger factor in potentially driving the court away from equality”; so much so that “any reduction due to post-separation accruals is likely to be theoretical rather than real and which will be subsumed by the issue of the company’s liquidity”.[99] 100. I shall deal with the estoppel question later.[100] However, the main reason for departure articulated by the Judge – fears for the Group’s liquidity – was rejected as unfounded by the Court of Appeal.[101] Cheung JA points out that the expert valuer testified that as at 30 December 2010, the company had $446 million which was not needed for the ongoing operations of the business and that, according to the Group’s financial statements for the following year, bank balances and cash increased by some $227 million to $749,593,757. The liquidity argument is in my view unsound and is in any event no longer supported by H in this Court. The principal reason articulated by the Judge for departing from equality was therefore unsound. D.2 The evidence as to separation 101. Although DHCJ Carlson ultimately excluded such evidence on the basis of estoppel, there was a considerable body of evidence before the Court regarding the parties’ relationship between 2001 and 2008. As noted above, the tragic deaths of Heather and Richard had had a devastating effect on the parties’ relationship, leaving them emotionally numb and mutually unable to communicate. As the Judge put it, there was no more warmth or connection between the parties. 102. However, after leaving the Fulham Gardens flat following Richard’s death, they did not go their separate ways but moved together to the Twin Bay Villas house in February 2001. As to their relationship in the period which followed, his Lordship found as follows: “What is agreed is that they had separate bedrooms. ... The wife says that whilst this may well have by then become an unhappy and unfulfilling marriage, it remained a marriage. She would polish his shoes, see to it that his laundry was done and if she did not always prepare his meals herself, they had a maid, she saw to it that his meals were prepared and kept warm for him if he returned home late. They would sometimes have meals at home together and she continued to be a ‘corporate wife’, attending functions and entertaining, as his wife, clients from overseas as well as Hong Kong government officials and mainland officials. They travelled together to the mainland on company business where she was seen to be his wife. They even went on a cruise together to Greece and shared the same cabin. For the first time in court she said that on one occasion they even had sexual intercourse which he has denied.[102] On one or perhaps two occasions she performed the ‘wifely’ task of driving him to the airport when he had to travel on business. Usually he employed a driver to do this sort of thing, who I presume on this occasion, was not available. They even entertained together at home. He would have barbecues for friends and for young engineers who the husband wished to encourage in their careers. This ‘modus vivendi’ continued until the husband finally left the home and went to live elsewhere in 2008.”[103] 103. Ms Anita Yip SC acknowledged that W’s evidence to the aforesaid effect was uncontradicted. Indeed, although H suggested that theirs was “a mere shell of a marriage”, he added that it had been “preserved for convenience, appearances, for friends and relatives and for the odd family [or] business occasion”.[104] He elaborated on this in cross-examination, stating: “... I think we all have face, we, you know, Chinese people are so face conscious, I think a divorced person would be looked upon by his friends, his business associates and the community as an outclassed [sic] person. And as a businessman running a business with a number of quite important public service on my shoulder, I do not wish to be seen by the community or by my business friends whatever to be divorced or to be separated physically because when I move out, everybody knows. ... I could have left but I did not, I decided not to leave because of those reasons I’ve just mentioned.” 104. The Judge evidently recognized that all this militated against there having been a separation in 2001 but apparently felt unable to act on such evidence because of the conclusion he had reached on estoppel. His Lordship observed: “If I were to accept the wife’s evidence on the way that they ordered their lives until he finally departed in 2008, it could not be said, on any view of the matter, that these parties were living separate lives. This would need to be viewed as a single household within an unhappy marriage that was, no doubt, in terminal decline. The conclusion that I would need to come to is that after so many years together, the parties continued to live within the same single household almost as a matter of habit. Whilst there was no tenderness, or feeling they still presented the outward appearance of a married couple appearing together at functions, travelling together on business, once going on a cruise and entertaining at home together. This apart from the performance of wifely services like seeing to laundry, polishing shoes and on occasions having meals together and sometimes watching television programmes together. The husband has even signed a statement at the wife’s insistence affirming their marriage.”[105] 105. The signed statement referred to was made in circumstances explained in W’s 3rd affirmation.[106] She recounted how on 4 May 2008, she discovered H’s personal effects in a flat occupied by Queenie Law. H had previously admitted to having had an extra-marital relationship in 1988 with Ms Law, an employee at Analogue, but he had assured W that the affair had been terminated. It therefore appeared to W that H had maintained a clandestine relationship with Ms Law for some twenty years. She states: “I was very unhappy and cried, and he asked me what happened. I told him I knew he had a woman. I said I want to have a complete family. He said he had done so. I said he always did not keep promise. Then I asked him to write a note to me.”[107] 106. The note, dated 10 May 2008 and signed by H, is exhibited. In translation, it states: “My wife Kan Lai Kwan, I want a complete family forever harmonious and happy”. 107. A week later, H and W went on a trip to France together. It was intended to be a two-week trip but, when H said he would return to Hong Kong first, W insisted on returning with him since she was worried about Queenie Law.[108] After returning, W had an emotional confrontation with Ms Law. H moved out on 3 July 2008 but they later travelled to Singapore together only to have another row there.[109] W states that a final attempt was made at reconciliation when in October 2008, H asked her to book a cruise holiday but, when that did not materialise, W lodged her divorce petition on 6 November 2008. 108. The learned Judge stated his factual conclusions as follows: “Taking all of the evidence as a whole I come to the following conclusions: Firstly, any tenderness or intimacy had ended by 2001. I reject the wife’s evidence that the parties had engaged in sexual intercourse as recently 2008 as she has suggested in her oral evidence. She had in fact said in cross-examination that the last time they had sex was in 2001, contradicting her evidence in chief. Secondly, they occupied separate bedrooms. Thirdly, communication was very limited. Fourthly, they had occasional meals together at home. These features lead me to the overall conclusion that none of the indicia of a matrimonial relationship were present. This was a bare shell and nothing more. Accordingly, I find as a fact that they were separated, as the husband alleges, from February 2001.”[110] 109. With great respect, those conclusions are unconvincing. The four matters mentioned – lack of sexual intimacy, separate bedrooms, very limited communication and occasional meals together – appear insignificant when contrasted with the many facets of their ongoing relationship. 110. As the Judge recognized, the evidence tended to show that while the marriage was unhappy, with “no tenderness or feeling”, the parties continued their habitual relationship founded on their very long marriage, with W continuing to do domestic chores for H; entertaining at home and accompanying H to business events to keep up appearances as a “corporate wife” to save H’s face in front of his associates and friends; going on trips together; and having a row on W discovering H’s continued relationship with his mistress – W reacting very much as a jealous wife and H acting like a husband seeking to placate her with a note which acknowledges her as his wife and records a desire for a harmonious and complete family life in May 2008. In my view, the evidence unmistakeably points to the marriage (unhappy though it was) having continued until the parties finally separated when H moved out of the house in mid-2008. D.3 The Court of Appeal’s decision 111. The Court of Appeal was also of the view that the evidence had “simply not shown that there is the necessary degree of separateness”.[111] Cheung JA pointed to the matters to which attention has been drawn above and concluded that the parties did not separate in 2001.[112] But because of the doctrine of estoppel, the Court of Appeal felt unable to find that separation had only occurred in 2008 even though that was very much what the evidence indicated. Cheung JA held that: “... the wife certainly cannot rely on separation that only began in 2008 because it will destroy the very foundation of the decree, namely, there was a two-year separation.”[113] 112. His Lordship’s solution was to hold that the separation had occurred in 2007. He reasoned that while W was estopped from asserting a 2008 separation, she was not precluded from denying that the separation occurred in February 2001 and could validly assert that separation had occurred in February 2007.[114] Understandably, objection was taken on H’s behalf since there was no factual foundation for such a finding. Cheung JA’s response was as follows: “Certainly there is no reference in the evidence to this period of time but in my view the reliance on this time is really a practical solution in order to reconcile the requirement of honouring the integrity of the decree and at the same time allowing the wife to challenge in the ancillary relief proceedings that the separation already began in 2001.”[115] 113. Having found that the separation occurred only in 2007, the Court of Appeal held that there should be no departure from the equality principle on the basis of post-separation accruals to the value of the trust estate. The period between 2001 and 2007 was irrelevant and post-2007 accruals were seen to be “based on the business foundation that the company has built up from its earlier years” to which W had equally contributed over the life of the lengthy marriage.[116] 114. With respect, I am unable to accept the adoption of a February 2007 separation date as a “practical solution”. Whether parties have separated is a question of fact and the evidence simply does not support a finding that separation occurred in February 2007. The real question is whether the Judge and the Court of Appeal were right to hold that they were estopped from finding that separation in fact occurred in mid-2008. D.4 Estoppel D.4a The authorities 115. In approaching the question of estoppel in a case like the present, it is essential to recognize that the Court is exercising a statutory jurisdiction under MPPO section 7(1)(a) which states: “It shall be the duty of the court ... to have regard to the conduct of the parties and all the circumstances of the case ...” Estoppel is a doctrine which regulates the relations between parties. It precludes one party from unconscionably contradicting certain facts or certain proprietary expectations bearing upon the position assumed by another party. But the Court remains subject to its statutory duty to have regard to all the circumstances of the case, regardless of the position assumed as between the parties. 116. This was acknowledged by the English Court of Appeal in Thompson v Thompson,[117] a divorce case which came before the court after fully contested maintenance proceedings involving cross-charges of cruelty, in which the husband’s version alleging the wife’s cruelty was accepted. When the husband petitioned for divorce, the wife sought to rely on the same allegations of cruelty that she had unsuccessfully made against him in the maintenance proceedings. The husband argued that she was estopped from doing so. However, the divorce proceedings were subject to section 4 of the Matrimonial Causes Act 1950 which provided that: “On a petition for divorce it shall be the duty of the court to inquire, so far as it reasonably can, into the facts alleged”. Denning LJ dealt with the estoppel argument, stating: “There is no doubt, to my mind, that if the doctrine of res judicata applies in its full force to the Divorce Division of the High Court, the wife is so estopped. ... The question in this case is, however, whether those ordinary principles do apply to the Divorce Division. The answer is, I think, that they do apply, but subject to the important qualification that it is the statutory duty of the divorce court to inquire into the truth of a petition - and of any countercharge - which is properly before it, and no doctrine of estoppel by res judicata can abrogate that duty of the court. The situation has been neatly summarized by saying that in the divorce court ‘estoppels bind the parties but do not bind the court’: but this is perhaps a little too abbreviated. The full proposition is that, once an issue of a matrimonial offence has been litigated between the parties and decided by a competent court, neither party can claim as of right to reopen the issue and litigate it all over again if the other party objects ...: but the divorce court has the right, and indeed the duty in a proper case, to reopen the issue, or to allow either party to reopen it, despite the objection of the other party... If the court does decide to reopen the matter, then there is no longer any estoppel on either party. Each can go into the matter afresh.”[118] 117. Unfortunately, in Hull v Hull,[119] the import of Thompson v Thompson was, in my view, unjustifiably narrowed. Mr Hull was seeking custody of the children after his wife had been granted a decree nisi of divorce in an undefended suit on the ground of his desertion. In the custody proceedings, he sought to allege that it was in fact his wife who had deserted him and also that she had committed undisclosed adultery. Sachs J held that the husband was estopped from making those allegations, distinguishing Thompson v Thompson on the ground that a different statutory provision was involved. That was surprising since the relevant words of section 1 of the Guardianship of Infants Act, 1925 provided as follows: “Where in any proceeding before any court ... the custody ... of an infant, ... is in question, the court, in deciding that question, shall regard the welfare of the infant as the first and paramount consideration, and shall not take into consideration whether from any other point of view the claim of the father, ... in respect of such custody, ... is superior to that of the mother, or the claim of the mother is superior to that of the father.”[120] 118. Sachs J construed those words as dealing “simply with the weight to be attached to facts once they are proved”.[121] But, with respect, I do not think that in a custody case where the court is under a duty to “regard the welfare of the infant as the first and paramount consideration”, the court should exclude itself from considering evidence which might have a material bearing on the child’s welfare on the ground of an estoppel, especially where the estoppel is based on earlier undefended proceedings where the conduct of the parents had not been investigated. 119. It is noteworthy that nine years later in Porter v Porter,[122] a case about how discretion in awarding maintenance should be approached, Sachs LJ recognized that the courts’ attitude had changed: “The practice as to discretion has thus naturally varied on this matter as on many others — such as the discretion exercised when granting a decree, where the ambit of the discretion has fundamentally altered in the past 25 years. In the exercise of any such discretion the law is a living thing moving with the times and not a creature of dead or moribund ways of thought.”[123] 120. In marked contrast with the Hull v Hull approach, Sachs LJ cited with approval what Denning LJ had stated back in 1950 in Trestain v Trestain,[124] moving away from being constrained by the implications of earlier proceedings: “I desire to say emphatically that the fact that the husband has obtained this decree does not give a true picture of the conduct of the parties. I agree that the marriage has irretrievably broken down and that it is better dissolved. So let it be dissolved. But when it comes to maintenance, or any of the other ancillary questions which follow on divorce, then let the truth be seen.” 121. In Porter v Porter,[125]decided two years later, Ormrod J pointed to two competing policy interests affecting the approach to estoppel in matrimonial cases: “This then is yet another case on the much argued problem of estoppel in matrimonial causes. It is a great pity that it has not yet been finally laid to rest. It arises from a conflict between two issues of public policy; on the one hand, the desirability of finality in litigation, which means the very proper and reasonable wish to prevent the same parties litigating the same issues of fact in the suit, and again in chambers on ancillary applications; and, on the other hand, the importance in the interests of justice to the individuals concerned, that the discretionary powers of the court in ancillary matters should be exercised with a full knowledge of all the relevant facts, rather than on a basis, partly of fact and partly of assumptions, arising from such rules as estoppel. It is particularly difficult to do justice in so personal a field as matrimonial cases if the realities of the situation are allowed to be obscured by the application of rules or principles which in other situations assist the cause of justice.” D.4b The proper approach to estoppel in matrimonial cases 122. In my view, the approach of Denning LJ in Thompson v Thompson should be adopted. It provides a measured and flexible approach which goes a long way towards reconciling those two competing policy interests, giving priority to the latter where appropriate. The parties are bound by the estoppel but, where the circumstances demand the Court’s intervention, it is free to override that estoppel in exercising its statutory jurisdiction and to act upon evidence which is material to its determination. 123. Ms Yip sought to argue that since the power to grant ancillary relief arose “on granting a decree of divorce”,[126] W was bound to adhere to the ground on which the decree was granted or else, as she puts it: “... the integrity of the decree of divorce would fall apart. In turn, this would mean that no ancillary relief could be obtained.”[127] I do not accept that argument. In performing its statutory duty of having regard to all the circumstances of the case, including circumstances which may be inconsistent with the basis upon which the undefended decree was obtained, the Court is not involved in setting aside the decree. No one has sought to have it rescinded and jurisdiction to deal with ancillary relief on the basis of an extant decree is unaffected. 124. The Thompson v Thompson approach accords with the recognized quasi-inquisitorial role of the Judge in a matrimonial case. As Thorpe LJ put it in Parra v Parra:[128] “The quasi-inquisitorial role of the judge in ancillary relief litigation obliges him to investigate issues which he considers relevant to outcome even if not advanced by either party. Equally he is not bound to adopt a conclusion upon which the parties have agreed.” 125. As Lord Sumption JSC explained in Prest v Prest:[129] “... claims for ancillary financial relief in matrimonial proceedings ... have some important distinctive features. There is a public interest in the proper maintenance of the wife by her former husband, especially (but not only) where the interests of the children are engaged. Partly for that reason, the proceedings although in form adversarial have a substantial inquisitorial element.” 126. Similarly, Baroness Hale of Richmond JSC stated: “I would ... emphasise the special nature of proceedings for financial relief and property adjustment under the Matrimonial Causes Act 1973 ... There is a public interest in spouses making proper provision for one another, both during and after their marriage, in particular when there are children to be cared for and educated, but also for all the other reasons explored in cases such as McFarlane v McFarlane [2006] 2 AC 618. This means that the court’s role is an inquisitorial one.”[130] D.4c Conclusion as to estoppel 127. In my view, the Judge and the Court of Appeal were wrong in law to hold that the Court was precluded by the doctrine of estoppel from finding that separation occurred only in mid-2008. The present case is one calling for the Court to override the estoppel created as between the parties. The suit was undefended and the facts relating to separation were not investigated at any stage prior to the ancillary relief proceedings. The evidence referred to in Section D.2 above compellingly points to separation occurring only as from mid-2008. Those are circumstances to which the Court ought properly to have regard if it is to make fair financial provision orders for distribution of the matrimonial assets on dissolution. D.5 Conclusion as to departure from equality 128. When considering ancillary relief, the financial position is generally approached on the basis of the values existing at the date when the hearing takes place. [131] 129. Where, however, there has been a substantial period of separation prior to the hearing and where during that period, there has been a steep increase in the value of the matrimonial assets attributable to the independent business or professional efforts by one spouse, unmatched by any contribution from the other spouse, grounds may exist for departing from equality.[132] In some such cases, fairness may dictate that the non-contributing spouse has no claim to share equally in the post-separation accrual to the matrimonial assets. 130. There are opposing arguments as to whether a spouse should be excluded in such cases. As Nicholas Mostyn QC explained in Rossi v Rossi:[133] “...it can legitimately be argued that the party in question has traded with the other party's undivided share and so should share with that party the profit that has been generated. On the other hand it can equally convincingly be said that the second party has not contributed to the industry or endeavour that gave rise to the profit or growth and so it is unfair that the second party should share to the same extent in that profit as the first who made all the effort....” 131. In Cowan v Cowan,[134] Thorpe LJ favoured the former approach and visualized only rare and exceptional departures from equality by reason of post-separation accruals: “The assessment of assets must be at the date of trial or appeal. The language of the statute requires that. Exceptions to that rule are rare and probably confined to cases where one party has deliberately or recklessly wasted assets in anticipation of trial. In this case the reality is that the husband traded his wife’s unascertained share as well as his own between separation and trial ... The wife's share went on risk and she is plainly entitled to what in the event has proved to be a substantial profit. If this factor has any relevance it is within the evaluation of the husband's exceptional contribution.” 132. His Lordship’s reference to “exceptional contribution” was a reference to cases where it can be established that the increase is only attributable to what has been called one spouse’s “stellar” contribution. As discussed in LKW v DD cases in that class are necessarily rare and exceptional.[135] H makes no claim to “stellar contribution” in respect of the increased profits of the business in the present case. 133. The summary of the principles provided in Rossi v Rossi[136] is broader than Thrope LJ’s stricter approach and is, in my view, preferable. It points to various factors relevant to deciding whether a post-separation accrual justifies departure from equality, including the length of the marriage and separation, the nature of the property accruing and the means or efforts by which it was acquired, and so forth. Of particular present relevance is the following passage: “Assets acquired or created by one party after (or during a period of) separation may qualify as non-matrimonial property if it can be said that the property in question was acquired or created by a party by virtue of his personal industry and not by use (other than incidental use) of an asset which has been created during the marriage and in respect of which the other party can validly assert an unascertained share. Obviously, passive economic growth on matrimonial property that arises after separation will not qualify as non-matrimonial property.”[137] 134. In my view, the increased Analogue Group profits do not provide a ground for departure from the equal sharing principle in the present case. The parties married in January 1968 and separated in mid-2008, over 40 years later. The period of separation prior to the hearing date was relatively insignificant. The profits accruing to the Analogue Group during the post-separation period arose out of the business which had been built up in the course of the marriage, in respect of which W can legitimately assert an unascertained share on the principles accepted in LKW v DD.[138] 135. As no other ground for departing from equality is contended for, I conclude that the Court of Appeal was right to hold that there should be no such departure in the present case. E. Disposal of this appeal 136. For the foregoing reasons, I would allow W’s appeal and dismiss H’s appeal. E.1 Quantum 137. In Section C.6 above, I arrived at a tentatively calculated balance of $386.7 million remaining to be paid by H on the footing that there is no departure from equality. That represents an increase of $156.7 million over the Court of Appeal’s award (in respect of which an instalment of $130 million payable on or before 1 March 2015 is outstanding). The calculation is necessarily tentative since the parties have not been heard as to whether that calculation correctly reflects the conclusions reached in this judgment. Miss Yip also requested an opportunity to make submissions regarding instalments and other payment arrangements in the event that the award was increased. The parties should plainly be afforded the chance to make such submissions. I would accordingly direct that the parties be at liberty to lodge written submissions as to the quantum of the award and as to directions for its payment. E.2 Costs E.2a Costs at first instance 138. In W’s printed case, she seeks an order for “costs here and below”.[139] However, the position in relation to the proceedings at first instance is unclear. In his formal judgment in HCMC 2/2010 dated 10 February 2012, DHCJ Carlson directed that there should be no order as to costs as between H and W. However, in the Court of Appeal’s judgment on costs,[140] Cheung JA states that the Judge had ordered H to pay W the costs of the ancillary relief proceedings and that there was no appeal against that decision, adding that the Court had been told that the costs had been paid on 20 August 2012. It is therefore unclear whether the first instance order is in issue. I would direct that the parties be at liberty, if so advised, to address that question in written submissions. 139. As to the trustee’s costs at first instance, the Court of Appeal noted that DHCJ Carlson had ordered such costs to be borne by H and W equally on an indemnity basis, but that H had taken it upon himself to pay the trustee’s costs in full (save for the costs incurred on 27 February 2012 on W’s unsuccessful application to vary the order nisi regarding the trustee’s costs). The Court of Appeal directed that “the parties” should pay the trustee’s costs of 27 February 2012. I would direct that the parties and the trustee be at liberty, if so advised, to address the question of the trustee’s costs of 27 February 2012 in written submissions. E.2b Costs in the Court of Appeal 140. The Court of Appeal ordered that as between H and W, W should be paid two-thirds of her costs of the appeal on the footing that she had been unsuccessful in claiming that the whole trust fund was H’s resource. As the Court of Appeal has been reversed on that point, I would set aside the aforesaid order of the Court of Appeal and make an order nisi that W should have all her costs of the appeal in the Court of Appeal to be paid by H. 141. As to the trustee’s costs in the Court of Appeal, the Court of Appeal ordered that W should be solely responsible for the trustee’s costs in that Court on an indemnity basis on the basis that its presence “was solely on the issue of the treatment of the trust assets” and that W failed on that issue. Since W has now succeeded on that issue, I would set aside the aforesaid order of the Court of Appeal and make an order nisi that H should pay the trustee’s costs. The parties are obviously entitled to make submissions on the order nisi. I would direct that the trustee should also be at liberty to lodge submissions on its own behalf. E.2c Costs in this Court 142. Regarding costs as between H and W in this Court, I would make an order nisi that H should pay W the costs of both appeals. 143. As to the trustee’s costs in FACV 20/2013, I note that the trustee was joined on W’s application, presumably because this was thought necessary in the light of the alternative claim for a variation of the trust. However, the trustee participated by making submissions not just in relation to variation, but also regarding the discretionary trust as a possible financial resource in the ancillary relief proceedings, as it had done in the Court of Appeal. In doing so, it aligned itself with H, arguing for only two-thirds of the trust fund to be treated as an available asset on a basis which I have rejected. I would accordingly make an order nisi that H do pay the trustee’s costs and direct that the trustee be at liberty to lodge submissions on its own behalf in that regard. E.3 Directions regarding written submissions 144. I would direct that any written submissions to be lodged in relation to the abovementioned orders nisi and regarding matters on which liberty to make submissions has been granted must be lodged with the Registrar within 14 days of the date of this judgment and any written submissions in reply within 14 days thereafter. No further submissions should be accepted without the leave of a single Permanent Judge. I would furthermore direct that each set of submissions and of any submissions in reply must not exceed 10 single-sided A4 pages of ordinarily legible 14 point print. Non-compliant submissions should not be accepted. 145. In the proceedings below the parties were anonymized, being referred to by their initials. The parties informed the Court that there was no reason for anonymity in the present case and that they had no objection to being named. They are accordingly referred to by name in this judgment. It is consistent with open justice that anonymity should be maintained only if there is good reason to follow that course. Mr Justice Tang PJ: 146. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Bokhary NPJ: 147. This case is very unfortunate in its immediate circumstances and, for reasons which are emphatically not the fault of either party, utterly tragic in its background. Despite the arguments so ably advanced by Ms Anita Yip SC on the husband’s behalf, I would allow the wife’s appeal and dismiss the husband’s appeal, doing so in terms of the orders and directions proposed by Mr Justice Ribeiro PJ. There are only two matters on which I will add something of my own. 148. In connection with the question of what financial provision ought to be made for the wife, there arose an issue as to when the parties are to be treated as having started to live separately. How this issue is now to be decided goes to the proper role of each of the three court levels which make up a legal system like ours: first instance, intermediate appeal and final appeal. 149. The husband said that it was in 2001 that the parties started to live separately while the wife said that it was not until 2008 that they started to do so. After conscientious consideration, the trial judge concluded that the parties had started to live separately, albeit under the same roof, as from February 2001. There being no basis for doing so in the circumstances, the Court of Appeal rightly refrained from disturbing any of the trial judge’s findings of primary fact on which he based that conclusion. What the Court of Appeal did – and was within their province to do – was to review the conclusion at which the trial judge arrived on the basis of those findings of primary fact. Upon such review and on a correct understanding of what constitutes a married couple living separately, the Court of Appeal felt unable to support the trial judge’s conclusion as to when this married couple started to live separately. And they arrived at a very different conclusion in that regard. Their conclusion was, in my view, supported by the trial judge’s findings of primary fact viewed in the context presented by the record realistically approached. That being so, their conclusion arrived at on intermediate appeal ought not to be disturbed by us on final appeal. 150. It was pointed out on the husband’s behalf that the date upon which the Court of Appeal proceeded, namely February 2007, is not consistent with the evidence either of the husband (who said that the separation had begun as long ago as 2001) or of the wife (who said that it did not begin until 2008). That is true in the sense that the Court of Appeal adopted the date February 2007 on the footing that although the separation had begun only in 2008 as the wife said, the parties’ divorce having been decreed on the ground of two-year’s separation prior to the presentation of the petition on 6 February 2009, the wife is estopped from relying on any date earlier than two years prior to such presentation. That is, I think, the limited extent to which an estoppel would have operated if one had arisen. As it happens, for the reasons given by Mr Justice Ribeiro PJ, no estoppel arose. 151. I understand that anonymity in family cases is the subject-matter of on-going consultation, and I am content to say nothing on the subject at this stage. Mr Justice Gummow NPJ: 152. I agree with the judgment of Mr Justice Ribeiro PJ. Chief Justice Ma: 153. The Court unanimously : (a) Allows W’s appeal and dismisses H’s appeal. (b) Directs that the parties be at liberty to lodge written submissions as to the quantum of the award and as to directions for its payment. (c) Directs that the parties be at liberty to address the question of costs at first instance in written submissions. (d) Directs that the parties and the trustee be at liberty to address the question of the trustee’s costs of 27 February 2012 in written submissions. (e) Sets aside the Court of Appeal’s order that W should be paid two-thirds of her costs in the Court of Appeal and makes an order nisi that all of W’s costs in the Court of Appeal be paid by H. (f) Sets aside the order of the Court of Appeal that W should be solely responsible for the trustee’s costs in the Court of Appeal on an indemnity basis and makes an order nisi that the trustee’s costs be paid by H, with liberty to the trustee to lodge submissions on its own behalf in that regard. (g) Makes an order nisi that H do pay W the costs of both appeals in this Court. (h) Makes an order nisi that H do pay the trustee’s costs and directs that the trustee be at liberty to lodge submissions on its own behalf in that regard. (i) Directs that any written submissions to be lodged in relation to the abovementioned orders nisi and regarding matters on which liberty to make submissions has been granted must be lodged with the Registrar within 14 days of the date of this judgment and any written submissions in reply within 14 days thereafter, with no further submissions to be accepted without the leave of a single Permanent Judge. (j) Directs that that each set of submissions and of any submissions in reply must not exceed 10 single-sided A4 pages of ordinarily legible 14 point print and that non-compliant submissions will not be accepted. Mr Andrew Lynn and Mr Patrick Siu, instructed by C.Y. Lam & Co., for the Wife (Appellant in FACV 20/2013, 1st Respondent in FACV 21/2013) Ms Anita Yip SC and Mr Eric Leung, instructed by Foo & Li, for the Husband (1st Respondent in FACV 20/2013, Appellant in FACV 21/2013) Mr Victor Joffe QC and Ms Mairead Rattigan, instructed by Withers, for the Trustee (2nd Respondent) [1] Born on 16 February 1971. [2] Born on 10 March 1972. [3] Born on 10 July 1974. [4] W’s third Affirmation 14.9.11, §94. [5] H’s first Affirmation 18.6.10, §18. [6] Ibid, §21. [7] HCMC 2/2010 (10 February 2012) at §33. [8] Judgment §94. [9] Section D.3 of this judgment. [10] Under section 11A(2)(c) of the Matrimonial Causes Ordinance (Cap 179). [11] Under section 11A(2)(d) of Cap 179. [12] H’s Form 21(4) dated 6 April 2009. [13] W’s third affirmation dated 14 September 2011, §164. [14] 10-14 October and 19-22 December 2011. Judgment was handed down on 10 February 2012. [15] Judgment §114. [16] Judgment §120. [17] Judgment §69. [18] Judgment §121. [19] Explained in LKW v DD (2010) 13 HKCFAR 537. [20] Judgment §125. [21] Judgment §130. [22] Cheung and Fok JJA and Macrae J, CACV 48/2012 (25 March 2013). [23] Court of Appeal §§41-43. [24] Court of Appeal §71. [25] $1,040,457,300 + $46,052,707 + $58,259,660 rounded up. [26] Ma CJ, Tang PJ and Litton NPJ, FAMV 27/2013 (1 November 2013). [27] MPPO sections 4 to 6A. [28] Cap 192. [29] Section C.5a of this judgment. [30] Initially along with W, Karen and her now deceased siblings, as well as any as yet unborn lineal descendants of H. [31] (2010) 13 HKCFAR 537 at §71. [32] Originally section 5 of the English Matrimonial Proceedings and Property Act 1970, then substituted by the Matrimonial and Family Proceedings Act 1984, section 3. Section 25(1) and (2) of the 1973 Act relevantly state: “(1) It shall be the duty of the court in deciding whether to exercise its [ancillary relief] powers ... and, if so, in what manner, to have regard to all the circumstances of the case, first consideration being given to the welfare while a minor of any child of the family who has not attained the age of eighteen. (2) As regards the exercise of the [relevant] powers of the court ... in relation to a party to the marriage, the court shall in particular have regard to the following matters – (a) the income, earning capacity, property and other financial resources which each of the parties to the marriage has or is likely to have in the foreseeable future ...” [33] [2006] 1 WLR 1053. [34] At §12. [35] [2007] 1 FLR 1246 at §48: “...we agree with both counsel that, before he attributed all the assets of [the trust] to the husband, the judge had to be satisfied that, if so requested by the husband, [the trustee] would be likely to advance them to him: in the judgments in this court on the husband’s appeal against the order for issue of the letter of request, in particular at [12], such had been confirmed as the central question generally arising in such cases.” [36] [2011] EWCA Civ 617, (2011-2012) 14 ITELR 1 at §113. The test is also acknowledged by Black LJ at §40. [37] (2013) 16 HKCFAR 1 at §49. [38] Judgment §69. [39] Court of Appeal §§28-34. [40] In Section C.3. [41] [2006] 1 WLR 1053 at §12. [42] Ibid at §61. [43] In re Esteem Settlement, Grupo Torras SA v Al Sabah and Four Others (2003) JLR 188; cited with approval in Charman v Charman at §71 and in Whaley v Whaley [2011] EWCA Civ 617 at §54. [44] Birt, Deputy Bailiff and Jurats de Veulle and Bullen. [45] [2012] 1 WLR 1721 (PC). [46] Per Lord Collins of Mapesbury at §59. [47] Judgment §61. [48] Judgment §65. [49] Mr Andrew Lynn who appeared for W at first instance and appeared with Mr Patrick Siu in the Court of Appeal. Mr Lynn and Mr Siu also appear for W in this Court. [50] Judgment §66. [51] Judgment §68. [52] Judgment §69 and §71. [53] Judgment §121. [54] (2008) 238 CLR 366 at §125 (footnotes omitted); [2008] HCA 56. [55] Section C.5a below. [56] Court of Appeal §§28-33. [57] Court of Appeal §41. [58] Court of Appeal §42. [59] Ibid. [60] Court of Appeal §43. [61] Ms Mairead Rattigan appeared for the trustee at first instance and in the Court of Appeal. The trustee is represented by Mr Victor Joffe QC and Ms Rattigan before this Court. [62] Dated 7 October 2011 and 16 December 2011 respectively. [63] At §72. [64] Section 6(1)(c) relevantly provides: “On granting a decree of divorce ... or at any time thereafter ... the court may ... make any one or more of the following orders, that is to say ... an order varying for the benefit of the parties to the marriage ... or either ... of them any ... post-nuptial settlement ... made on the parties to the marriage.” [65] At §11 in the submissions dated 7 October 2011 and §14 in the submissions dated 16 December 2011. [66] Appearing with Mr Eric Leung for H. At first instance and in the Court of Appeal, H was represented by Mr Gerard McCoy SC together with Ms Anita Yip and Mr Wilson Leung. [67] At §13 in the submissions dated 7 October 2011 and §16 in the submissions dated 16 December 2011. [68] At §50. [69] At §62. [70] (2013) 16 HKCFAR 1 at §§37 and 38. [71] Clause 10. [72] Such as HSBC Trustee (Hong Kong) Ltd, which was described in correspondence with Analogue (dated 16 July 2008) as acting as administrative assistant of the Trustee. [73] Clauses 5 and 6. [74] Clause 5(b). [75] Clause 18(a). [76] Clause 3(a). [77] Clause 22. [78] Clause 26(b) and (c). [79] At §80. [80] Trusts & Trustees, Vol 18, No 1, January 2012, pp 17-35. [81] Ibid. [82] At p 24. [83] Under clause 18(a) of the trust deed. [84] At p 35. [85] Letterstedt v Broers (1884) 9 App Cas 371 (PC) at 386-387. [86] Dated 17 January 1996, 17 October 1997, 6 November 2000, 1 March 2005 and 14 May 2010. A further letter of wishes was issued on 12 November 2013 disclosed shortly before the hearing in this Court. That is dealt with it later in this Section. [87] On 29 July 2011. [88] Subject to what is said about the recently disclosed letter of wishes dated 12 November 2013 discussed in Section C.5e below. [89] After the Realty Limited shares were transferred to W. [90] Mr John Utting’s report dated October 2011. [91] Trust deed, clause 26(b)(i). [92] Printed case at §64. [93] Judgment §3. [94] Dated 18 June 2010, §9. [95] At §25. [96] (2010) 13 HKCFAR 537 at §58. [97] Judgment §§75-80. [98] Judgment §99. [99] Judgment §99 and see also §132. [100] Section D.4 below. [101] Court of Appeal §§73-78. [102] The Judge rejected W’s evidence on this point. [103] Judgment §§83-85. [104] H’s 5th Affirmation §41. [105] Judgment §§86-88. [106] Dated 14 September 2011. [107] 		3rd Affirmation at §104. [108] Ibid at §106. [109] At §117. [110] Judgment §94. [111] Court of Appeal §61. [112] Court of Appeal §64. [113] Court of Appeal §48. [114] Court of Appeal §§48 and 64. [115] Court of Appeal §49. [116] Court of Appeal §§71 and 72. [117] [1957] P 19. [118] At pp 28-29. Hodson and Morris LJJ agreed. [119] [1960] P 118; followed by a number of cases including Field v Field [1964] P 336. [120] At p 124. [121] Ibid. [122] [1969] 1 WLR 1155. [123] At p 1159. [124] [1950] P 198 at 202; cited by Sachs LJ at p 1160. [125] [1971] P 282 at 284. [126] MPPO sections 4 and 6. [127] H’s printed case in FACV 21, §45. [128] [2003] 1 FLR 942 at §22, cited with approval in LKW v DD (2010) 13 HKCFAR 537 at §69. [129] [2013] 2 AC 415 at §45. [130] At §85. [131] Rossi v Rossi [2007] 1 FLR 790 at §24.1; Cowan v Cowan [2002] Fam 97 at §70. [132] LKW v DD (2010) 13 HKCFAR 537 at §94. [133] [2007] 1 FLR 790 at §15. [134] [2002] Fam 97 at §70. [135] At §§112-118. [136] [2007] 1 FLR 790 at §24. [137] At §24.3. [138] (2010) 13 HKCFAR 537, adopting the principles laid down in White v White [2001] 1 AC 596 and in Miller v Miller and McFarlane v McFarlane [2006] 2 AC 618. [139] Printed case in FACV 20/2013, §115(5). [140] CACV 48/2012 (10 May 2013) at §§1 and 2. 
Chief Justice Ma: 1. I agree with the judgment of Mr Justice Cheung PJ. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Mr Justice Cheung PJ. Mr Justice Fok PJ: 3. I agree with the judgment of Mr Justice Cheung PJ. Mr Justice Cheung PJ: 4. This appeal concerns an application to reverse a guilty plea before sentence and raises an important procedural matter in such context that should be clarified. The facts 5. The appellant was the driver of a private motor vehicle involved in a traffic accident that happened on 3 November 2016, in which a pedestrian was injured whilst crossing the road from a safety island. Arising from the accident, the appellant was charged with the offence of careless driving, for which he appeared before a deputy magistrate[1] on 1 June 2017. Unrepresented, the appellant pleaded guilty to the charge and agreed with the police brief facts which were read out in court: “The incident took place at 2:28 pm on 3 November 2016. The Defendant was driving private car KE3888 along Sheung Yuet Road in the east-bound direction, turning left into the north-bound lane of Wan Kwun Road. PW1 was crossing the north-bound lane from east to west at a safety island in the middle of Wan Kwun Road. Without due care and attention, the Defendant caused the right front wheel of the vehicle to press against PW1’s left foot. PW1 was injured in her left foot and was sent to the hospital for treatment. …” 6. Accordingly, the deputy magistrate convicted the appellant of the offence. During mitigation, the following exchange took place between the deputy magistrate and the appellant: “COURT: Anything to say in mitigation? DEFENDANT: Actually, I want to say that if the same thing happens, she was not hit by the front of my car, she had her foot stuck out at the roadside, the wheel of my car came into contact with her, I could not stop. COURT: That’s right, the wheel of your car hit …… DEFENDANT: That’s right. COURT: Pressed her foot. DEFENDANT: Right. COURT: And that place was a safety island. DEFENDANT: That’s right, but I was actually very far away from the safety island, she had her foot stuck out, I -- I mean I didn’t know how it could be avoided under the circumstances. COURT: Safety island, if there are pedestrians standing in the safety island, is it right that you should give way -- give way to her? DEFENDANT: That’s right, that’s not right, she -- at that time there was actually a lot of people because it happened to be lunch time, so …… COURT: Why did you not look carefully, and not give way to the pedestrians and let them cross the road first? DEFENDANT: No, all those people had stopped there, only the miss was at the last, the farthest back on the right hand side of the safety island. She was playing with her phone when she walked out. I had a witness by my side. Nonetheless, I admit that I was careless. But she dashed out like that from the side of my car, I actually could not avoid her. I already stopped my car immediately. Therefore, my car had not run over her, it only came into contact – you said that the front wheel of the car had come into contact with her foot. COURT: But the facts you admitted just now stated that the front wheel of your car had pressed her foot. DEFENDANT: Right, but what I meant to say was I already knew that her foot was at the wheel of my car. I already stopped the car immediately.” 7. After hearing mitigation, the deputy magistrate, probably in view of the appellant’s previous conviction record, adjourned the hearing to 29 June 2017 to obtain a community service report before sentencing. 8. Before the adjourned hearing, the defendant obtained legal advice and made an application to the court to reverse his plea of guilty. This was reflected in the community service report dated 28 June 2017: “5. Regarding the present offence, Defendant indicated reservations towards the content of the Police’s Brief Facts. He recalled that on the material day, he drove his cousin’s car for test drive to a potential buyer and passed the location-in-question. When the car he drove was passing the pedestrian crossing area in a speed around 20 to 30 km/hour, he noticed there were lots of pedestrian standing on the pavement and nobody crossing the road. Unexpectedly, the victim (PW1), focusing on her mobile phone, stepped out from the pavement without noticing the road situation. While PW1’s mobile phone hit the car’s side mirror, Defendant stopped the car. He found the PW1 had fallen near the rear wheel. Subsequently, he was arrested for the present offence. 6. During interviews, Defendant insisted that he was innocent and considered the fault was not primarily out of his carelessness. Reportedly, he pleaded guilty in order to shorten the court proceeding time, but he had underestimated the legal consequences and the seriousness of the offence. Meanwhile, Defendant had sought legal advice and prepared to appeal for the case. Yet, upon lengthy discussion, Defendant also shared with the Investigating Probation Officer (IPO) that he would further heighten his awareness of the traffic condition, especially in congested area and followed traffic regulation strictly in future. 7. Defendant stated that his children were in young age that he was in heavy childcare work. In addition, though he showed understanding on the nature of Community Service Order (CSO), taking into consideration of the context and his interpretation of the present offence as well as time constraint, he indicated his limitation and hesitation in completing those unpaid work under CSO. In view of Defendant’s limited capacity in performing unpaid under CSO, Community Service Order is NOT recommended in this case.” 9. At the adjourned hearing on 29 June 2017, the deputy magistrate heard the application to reverse plea. On 6 July 2017, she rejected the application. On the same day, she sentenced the appellant to 150 hours of community service. The deputy magistrate’s reasons 10. In her statement of findings dated 28 July 2017, the deputy magistrate explained that the application to reverse plea was made on the basis that the guilty plea was an equivocal one and therefore the conviction could not stand.[2] The deputy magistrate understood the essential basis of the application to be that the injured pedestrian was herself at fault in the accident – she was not paying attention to the road condition when she stepped out from the safety island.[3] 11. The deputy magistrate agreed with the defence’s submission that the place of the accident was not a zebra crossing and pedestrians did not have priority in crossing the road.[4] However, she took the view that it did not mean that this was inconsistent with the appellant’s guilty plea. The fact that the injured pedestrian might also have been partly responsible for the accident did not mean that the appellant was not driving carelessly at the time.[5] The deputy magistrate emphasised that the appellant understood the charge, admitted the brief facts and made an admission to the elements of the offence.[6] 12. The deputy magistrate noted that the appellant claimed in the community service report that he chose to plead guilty to save time but had underestimated the seriousness of the offence and the legal consequences.[7] However, the deputy magistrate observed that the defence had confirmed that the application to reverse plea did not concern an unequivocal plea. Therefore, there was no need for the court to consider “the circumstances where unequivocal pleas shall be treated as nullities”.[8] The appeal before the deputy judge 13. The appellant appealed to the Court of First Instance against his conviction. Originally, the respondent was prepared to concede the appeal before Deputy High Court Judge S T Poon on the basis that the appellant’s guilty plea was equivocal and should not have been accepted. However, the deputy judge asked the respondent to conduct further legal research on the applicable legal principles and adjourned the hearing of the appeal. At the adjourned hearing, the respondent changed its position and opposed the appeal. After hearing arguments, by a judgment dated 23 March 2018, the deputy judge dismissed the appeal. 14. In paragraphs 12 to 14 of the judgment, the deputy judge set out his understanding of the relevant legal principles: “12. Prior to the passing of sentence, the magistrate has the power to exercise her discretion to allow a defendant to reverse his guilty plea irrespective of whether the defendant’s guilty plea is equivocal or not. Such discretion should not be exercised lightly and such discretionary power is one which should only be exercised in clear cases and very sparingly. 13. In the present case, the trial magistrate refused to exercise her discretion to allow the Appellant to reverse his guilty plea at the review hearing. The Appellant is now asking this Court to overrule the trial magistrate’s decision on exercising her discretionary power. 14. Generally, the appeal court will not interfere with the way a trial court exercises its discretion unless the decision made by the trial court in exercising its discretion is obviously unreasonable or unfair. Absenting which, the appeal court generally will not interfere with the magistrate’s decision on exercising her discretion.” 15. The deputy judge also referred to case law and observed that for a plea to be equivocal, the defendant must add to the plea of guilty a qualification which amounts to a defence.[9] 16. The deputy judge then turned to the facts and essentially agreed with the respondent’s counsel as well as the deputy magistrate below that the fact that the injured pedestrian might have been partly responsible for the accident did not mean that the appellant was not driving carelessly. The fact remained that the appellant had clearly admitted he had been careless. What he said subsequently could merely serve to mitigate the gravity of the offence without adding any qualification to his plea of guilty.[10] 17. As regards what the appellant had said to the probation officer as stated in the community service report, the deputy judge took the view that it had no relevance in considering whether the appellant’s plea was equivocal.[11] 18. Treating the appeal as one against the exercise of the deputy magistrate’s discretion on whether to allow the application to reverse plea, the deputy judge concluded: “In my judgment, the trial magistrate made no obvious error in exercising her discretionary power and I shall not interfere.”[12] 19. He dismissed the appeal accordingly. 20. With leave granted by the appeal committee, the appellant now brings his case to this court on the substantial and grave injustice ground. The law 21. For reasons that will become apparent, it is necessary first to set out and clarify some relevant legal principles on reversal of guilty pleas. 22. First, there is a crucial distinction between an equivocal plea and an unequivocal plea of guilty. A plea is equivocal if the defendant adds to his plea of guilty a qualification which, if true, may show that he is not guilty of the offence charged. 23. The distinction is crucial because only an unequivocal plea can be accepted by the court. The court cannot accept an equivocal plea; it has no discretion in the matter. 24. The position is best explained by O’Connor J (as he then was) in his lead judgment in P Foster (Haulage) Ltd v Roberts:[13] “In my judgment, a clear distinction must be drawn between the duties of a court faced with an equivocal plea at the time it is made and the exercise of the court’s jurisdiction to permit a defendant to change an unequivocal plea of guilty at a later stage of the proceedings. A court cannot accept an equivocal plea of guilty: it has no discretion in the matter; faced with an equivocal plea the court must either obtain an unequivocal plea of guilty or enter a plea of not guilty. For a plea to be equivocal the defendant must add to the plea of guilty a qualification which, if true, may show that he is not guilty of the offence charged. An example of this type of qualification is found where a man charged with handling a stolen motor car pleads ‘guilty to handling but I didn’t know it was stolen’. It is not every qualification which makes a plea of guilty equivocal; for example, the burglar charged with stealing spoons, forks and a camera, who pleads ‘guilty but I did not take the camera’ is making an un-equivocal plea to burglary. Once an unequivocal plea of guilty has been made, then the position is entirely different. From this stage forward until sentence has been passed the court has power to permit the plea of guilty to be changed to one of not guilty, but the exercise of this power is entirely a matter of discretion.” 25. Secondly, whether a plea is equivocal is to be determined “at the time it is made”.[14] If the guilty plea is accompanied by a qualification of the type described above, it is an equivocal plea, and the court cannot and must not accept it. If it is not, it is an unequivocal plea, based on which the court is entitled to convict the defendant of the offence charged. If, after conviction, the defendant says to the court during mitigation, or to a probation officer when preparing a report for sentencing purposes, something that if it had been said at the time the plea was taken would have amounted to a qualification of the type described above, that does not turn the unequivocal plea into an equivocal one. The plea remains an unequivocal one. 26. Thirdly, a “conviction” is not complete, and the court does not become functus, until sentence is passed. It follows that in the scenario described in the preceding paragraph, where something emerges after conviction but before sentence, which, if true, may show that the defendant is not guilty of the offence charged, the court has a discretion to allow a change of plea. In fact, given that the discretion is one ultimately based on the interests of justice, the court should, whether on application or of its own initiative, consider exercising its discretion to reverse the plea once it becomes aware of the full picture. 27. However, the case law does not always speak consistently with the propositions stated above. In particular, there are cases which speak of an otherwise unequivocal plea becoming or taking on the character of an equivocal plea by reason of what transpires at a later stage of the proceedings. 28. Amongst the cases cited to this court, Lee Fu-yuen v The Queen[15]was a case concerning the employment of under-age persons, which the magistrate thought constituted an absolute offence. The defendants pleaded guilty and were convicted accordingly. In mitigation, the defendants said they thought and believed that the girls in question were over 18. On appeal, in relation to the jurisdiction and discretion to allow a reversal of plea, the court rightly said: “The learned Magistrate is not functus officio unless and until sentence has been passed. Even when the plea in mitigation reveals that an offence has not been committed the learned Magistrate, was still in a position, if he thought fit, if he saw any justification, to reverse a plea of guilty to one of not guilty and enter a plea of not guilty. If the Magistrate considered that the offence charged was not one of an absolute offence then he would, and I think he should, have entered a plea of not guilty for the Appellants at the time.”[16] However, later on in the judgment, after holding that the offence was not an absolute one but required proof of knowledge of the age of the girls, the court treated the defendants’ guilty pleas as “equivocal” ones,[17] and held that the trial before the magistrate was a “nullity”.[18] This terminology was, with respect, wrong. 29. Likewise, in HKSAR v Wang Jing-yun,[19] the applicant pleaded guilty to attempted arson with intent and was convicted accordingly. During mitigation and from the reports obtained by the court for sentencing purposes, it emerged that the applicant’s case was that actually she had no intention to set fire to the premises in question; rather, what she intended was simply to frighten her husband (the victim). The Court of Appeal treated the case as one involving an “equivocal plea”,[20] by reason of what was said in mitigation and in the reports. It said: “A conviction is not complete until sentence has been passed … and in our view the court below ought, at the latest by the time the reports had been read, have been alerted to the fact that what was being said by the applicant was inconsistent with her plea, and ought then to have raised the matter with the solicitor for the applicant … Ms Wong, for the respondent, very properly concedes that the conviction should not be allowed to stand, and we agree. She accepts that even at the initial stage when it was asserted in mitigation that the idea on the applicant’s part was merely to threaten, the plea took on the character of an equivocal one.”[21] (emphasis added) 30. In Fong Loy v The Queen,[22] the defendant, on a charge of assisting an offender, pleaded guilty and admitted the brief facts. He added immediately a qualification that he was threatened by the offender with assault to assist him – thus suggesting a defence of reasonable excuse. He was nonetheless convicted. On those facts, the court, on appeal, rightly concluded that the plea was equivocal, and could not be allowed to stand. However, it went on to give guidance on situations where the qualification only came into the picture during mitigation: “Of course I am aware that a person pleading guilty often in purported mitigation says something that he hopes will be accepted as a mitigating factor, but which he does not wish to be taken as being in derogation of his plea. Where this occurs the magistrate is quite entitled to inquire from him whether or not he is serious in making such an allegation and furthermore the magistrate would be entitled to tell him that such an allegation is not consistent with a plea of guilty or with the agreed facts. On that being pointed out to a defendant he not infrequently indicates that the matter in derogation of plea is not put forward seriously. However he is under no obligation to elaborate and if he stands by such a remark at that stage, a plea of not guilty should be entered.”[23] 31. In view of some of the terminology employed by the courts, there is a need to clarify the relevant law. On first principles, a guilty plea is an answer of admission of guilt by a defendant given in reply to the court’s demand to plead. Based on the guilty plea, and very often together with the defendant’s agreement with the police brief facts,[24] the court may make a finding of guilt and enter a conviction against the defendant of the offence charged, without requiring proof of the offence by evidence at trial. 32. By definition, an equivocal plea, sometimes referred to as a “guilty but …” plea,[25] that is, a guilty plea accompanied by a qualification of the sort described above, is an ambiguous, “yes and no” type of answer. Unless clarified, it does not possess the necessary quality in terms of an admission of guilt that would entitle a court to make a finding of guilt and dispense with a trial to prove the offence charged by evidence. For this reason, it cannot be accepted and form the basis of a conviction. And for this reason, an equivocal plea or a conviction resulting from the erroneous acceptance by the court of such a plea is often described as a “nullity”. And also for this reason, when the true picture is subsequently revealed, the conviction which is founded on such an insufficient basis must be set aside; no discretion is involved. 33. However, if the guilty plea is unequivocal when it is made, the court is quite entitled to make a finding of guilt on the basis of the admission. Whatever happens at a later stage of the proceedings before sentence cannot alter the prior, historical fact that there has been made by the defendant an unequivocal admission of guilt. In other words, what happens subsequently cannot change the nature or character of the defendant’s earlier admission of guilt by his guilty plea. Whether the court has the power or discretion to allow a withdrawal of the guilty plea by reason of the subsequent development is quite another matter. An unequivocal guilty plea, once given, is a historical fact. It cannot “become” or be “turned into”, nor can it “take on the nature of”, an equivocal plea by what happens afterwards. 34. As may be gleaned from the House of Lords’ decision in S (An Infant) v Recorder of Manchester,[26] the confusion probably crept in because prior to that decision, it had been thought that in summary proceedings in the English magistrates’ courts, once a conviction, in the sense of a finding of guilt, was entered by the court following a guilty plea, the court had no power to allow a reversal of plea; the court had become functus so far as conviction was concerned. Not surprisingly, this functus rule could lead to injustice in circumstances where after conviction it emerged that the defendant might not be guilty of the offence charged or for some other reason should be given a chance to contest the proceedings. That led the courts, over the years, to adopt some “rather artificial practices”[27] to get around the difficulty created by the supposed functus rule. One of these was to label an unequivocal plea when made an equivocal one by reason of what transpired at a later stage of the proceedings, in which event, so it was reasoned, the plea or the conviction would become a “nullity” and the conviction could be quashed on that basis. The situation was by no means satisfactory, as Lord Upjohn explained: “These cases lead understandably enough, in order to do justice to the accused, to some rather artificial practices such as accepting a plea of guilty provisionally, as explained by Widgery J. in Reg. v. Blandford Justices [1967] 1 Q.B. 82; or in the ‘guilty but ...’ cases, an expression used by Lord Goddard C.J. in Reg. v. Durham Quarter Sessions, Ex parte Virgo [1952] 2 Q.B. 1 to describe the type of case where the accused pleaded guilty but then or at some later stage of the trial showed that he misunderstood the nature of the plea for his explanation showed that he should have pleaded not guilty.”[28] 35. Lord MacDermott expressed a similar sentiment when he said: “I think this is all too confusing and difficult to be sound. And the confusion becomes worse confounded if, as I am inclined to think may have happened in some of the cases, the ‘guilty but ...’ or equivocal factor was only revealed to the court by statements made in mitigation during the sentencing stage of the proceedings. These fine distinctions between what pleas are acceptable and what not, between the equivocal and the unequivocal, between provisional acceptance and final acceptance of the plea made, and between one stage and another of the same proceedings owe much to the introduction of what I have called the functus doctrine and suffice in themselves to cast a grave doubt on its validity.”[29] 36. After reviewing the authorities and overruling some key decisions, the House of Lords clarified that as a matter of law, a magistrates’ court does not become functus until the passing of sentence. “Conviction” – in the sense of the final disposal of a case – is not complete at the stage of “conviction” – in the sense of a finding of guilt; it is only complete when sentence is passed. [30] The important consequence of this clarification, for our present purpose, is that henceforth, there is no longer any need or indeed justification for continuing with the previous artificial practices to get around a non-existent rule and do justice. 37. This was made clear by Lord Upjohn when he said: “The court, whether High Court, quarter sessions or a court of summary jurisdiction, retains full jurisdiction over all matters before it until sentence, that is, until the final adjudication of the matter; and the reasoning in Sheridan’s case and the cases of Grant [1936] 2 All E.R. 1156; Guest [1964] 1 W.L.R. 1273, and Gore Justices [1966] 1 W.L.R. 1522, which followed that reasoning must be treated as overruled. In future it will be quite unnecessary to accept a provisional plea or to resort to the ‘guilty but ...’ artifice. If the court upon all the facts before it, thinks it is proper to accept a plea of guilty then the court may permit that plea to be withdrawn and a plea of not guilty accepted at a later stage up to sentence, that is, until the complete adjudication of conviction.”[31] 38. The true position has been made doubly clear by O’Connor J, following S, in the passage from Foster already cited above. 39. In other words, for our present purpose, an equivocal plea is a guilty plea which when made is accompanied or immediately followed by a qualification of the type described in Foster. Absent such a qualification, the plea is an unequivocal one. Anything that is said or comes to light after conviction cannot and does not turn the unequivocal plea to an equivocal one. Rather, its significance lies in that it may form the basis of an application to invoke the court’s discretion to allow a reversal of plea. 40. Continuing with my review of the relevant legal principles, fourthly, the discretion to allow the reversal of an unequivocal plea after conviction (but before sentence) is an unfettered one, although it has been said that it should be “exercised in clear cases and very sparingly”.[32] 41. This suggestion of a cautious approach is not difficult to understand. A plea of guilty is, as it should be so regarded by all defendants, a serious plea. It must not be lightly made without full consideration. The policy of certainty and finality dictates against allowing a reversal of an unequivocal plea without good reason. As McCowan LJ observed in R v Croydon Youth Court:[33] “… the interests of justice also include the interests of the courts and the public that people who have pleaded guilty with the advice of counsel should continue to be regarded as guilty and that there should be certainty and an end to litigation.” 42. Moreover, where sentencing is adjourned, any possible prejudice to the prosecution due to any change of circumstances following conviction must be borne in mind. Furthermore, time and costs in preparation of reports may be wasted. The court’s diary position may have changed. Other court users’ timetables could be affected. 43. However, ultimately, how the court’s discretion should be exercised must turn on the facts. As O’Connor J said in Foster, “the exercise of this power is entirely a matter of discretion”.[34] More recently, Lord Phillips of Worth Matravers CJ (as his Lordship then was) said in the English Court of Appeal in Revitt v Director of Public Prosecutions:[35] “16. What principles should govern allowing an application to withdraw a guilty plea? In S (An Infant) v Recorder of Manchester [1971] AC 481, 507 Lord Upjohn observed that the discretionary power was one which should ‘only be exercised in clear cases and very sparingly’. That guidance is not of great practical assistance. Better general guidance appears from the comments of Lord Morris of Borth-y-Gest, at p 501: ‘The duty of a court to clear the innocent must be equal or superior in importance to its duty to convict and punish the guilty. Guilt may be proved by evidence. But also it may be confessed. The court will, however, have great concern if any doubt exists as to whether a confession was intended or as to whether it ought really ever to have been made.’ 17. If after an unequivocal plea of guilty has been made, it becomes apparent that the defendant did not appreciate the elements of the offence to which he was pleading guilty, then it is likely to be appropriate to permit him to withdraw his plea – see R v South Tameside Magistrates’ Court, Ex p Rowland [1983] 3 All ER 689, 692, per Glidewell J. Such a situation should be rare, for it is unlikely to arise where the defendant is represented and, where he is not, it is the duty of the court to make sure that the nature of the offence is made clear to him before a plea of guilty is accepted. 18. It may happen, and again this is likely to be rare, that the court hearing an application to withdraw a guilty plea will or should appreciate that the facts relied upon by the prosecution do not add up to the offence charged. In such circumstances, justice will normally demand that the defendant be permitted to withdraw his plea. R v Bournemouth Justices, Ex p Maguire [1997] COD 21 appears to have been such a case.” 44. In my view, the overriding consideration, in the exercise of the discretion, must be the interests of justice. As Lord MacDermott said in S:[36] “Once made, a mistaken plea may be properly accepted and the mistake may never stand revealed. But if, as can happen, the truth comes to light during the second stage of the proceedings, when the question of what to do with the accused is under consideration, why should it not be acted upon and a changed plea of not guilty allowed where the interests of justice so require? There is no good reason for thinking that such a course would create an administrative problem or open the door to a widespread abuse of process.” 45. In the type of case under consideration, where there is material appearing in mitigation or the reports, which tends to suggest that the defendant may not be guilty of the offence charged, the interests of justice would, in a normal case, weigh heavily in favour of allowing a reversal of plea. Of course, where, for instance, the court has a real doubt that what is subsequently asserted is only a recent fabrication to avoid the consequence of, say, a heavy sentence following a guilty plea, the court may certainly probe deeper into the matter before deciding how its discretion should be exercised. 46. It must also follow from the above discussion that once one reserves the label of an “equivocal” plea to a plea that is accompanied or immediately followed by a qualification of the type described in Foster at the time the plea is made, leaving all post-conviction qualifications as cases seeking to invoke the court’s discretion to allow the change of an unequivocal plea, one is no longer constrained to look for a qualification of the type mentioned in Foster in these post-conviction cases, when considering how the court’s discretion should be exercised. In other words, whilst Foster describes, for the purpose of deciding whether a plea is equivocal, the requisite qualification as one which, if true, may show that the defendant is not guilty of the offence charged, this need not be the threshold that a defendant in a post-conviction case must reach in order to successfully invoke the court’s exercise of discretion. Depending on the facts, even in a case of doubt about the merits of the defendant’s case (as per his story), the court may still think that its discretion should be exercised in favour of allowing a reversal of plea, so that the matter can be properly tried at trial, bearing particularly in mind that in deciding an application to change plea, the court should avoid conducting a mini-trial on the merits of the case on the basis of what the defendant now asserts before it. 47. Needless to say, the circumstances that may give rise to an application to reverse an unequivocal plea are not limited to those involving a qualification of the type discussed above. 48. As this court has explained in HKSAR v Shum Wan Foon,[37] where the unequivocal plea of guilty was in fact entered into as a result of duress, inducement or misrepresentation, the plea is in substance a nullity, and the court’s discretion may only be judicially exercised by allowing a change of plea.[38] This court has also emphasised that where any of these vitiating factors is suspected, the court must make sufficient inquiries to determine whether there is any truth in the suspicion, without which there would not be a sufficient basis to decide how the court’s discretion should be exercised.[39] 49. Examples of applications based on circumstances other than post-conviction qualifications of the type discussed are many. In R v Bow Street Metropolitan Stipendiary Magistrate, ex p Roche[40] and R v South Tameside Magistrates’ Court, ex p Rowland,[41]the respective defendants both applied (unsuccessfully) to reverse their guilty pleas for fear of a custodial sentence. The defendant in Foster, on the other hand, based his application on an (alleged) mistake of law that his lawyer had made. The approaches below 50. It should be apparent from the discussion of the law above that the approaches of the deputy magistrate and the deputy judge in the present case were, for understandable reasons, flawed. 51. For the deputy magistrate, what she had taken from the appellant was an unequivocal plea. What the appellant relied on to reverse his plea only emerged after the plea was made and conviction entered, that is, during mitigation and in the community service report. Although the application to reverse plea was made to her on the basis that the plea was equivocal, it was nonetheless an application to reverse that plea, on which the deputy magistrate had a discretion to exercise. She was correct in deciding that the plea before her was unequivocal. To decide that, all she ought to have considered was the circumstances at the time the plea was taken. What happened during mitigation or was said in the report was quite irrelevant to the question of whether the plea was equivocal. 52. However, the deputy magistrate’s decision that the plea was unequivocal did not, contrary to what she thought, dispose of the application to reverse plea. By that stage, she had assertions before her which if true would suggest that the appellant was not guilty of careless driving at all, even though the application to reverse plea was made on the wrong footing that the plea was equivocal. The deputy magistrate was therefore wrong, as a matter of approach, to simply refuse the application to reverse the guilty plea once she ruled that it was not equivocal,[42] without considering how her discretion should be exercised in light of the material emerging during mitigation and from the community service report. 53. As for the deputy judge’s approach to the appeal against conviction, again understandably, he shared the mistake of the deputy magistrate in thinking that the only issue he had to deal with was whether the plea was equivocal.[43] Like the deputy magistrate, he also came to the right conclusion that the plea was unequivocal. He was right in thinking that in deciding whether the plea was equivocal, what was said in the community service report was irrelevant,[44] but was wrong in taking into account what was said during mitigation in determining the question. Although, as explained, the deputy magistrate had never gone on to consider whether she should exercise her discretion to allow the reversal of plea once she had ruled that the plea was unequivocal, somehow the deputy judge wrongly thought that she had, and concluded his judgment by saying that he could not discern any obvious errors on the part of the deputy magistrate and therefore was not prepared to interfere with her exercise of discretion.[45] The appellant’s version of events 54. Turning now to what emerged after conviction in the present case, both the deputy magistrate and the deputy judge were right to say that the fact that the injured pedestrian may have been partly responsible for the accident does not necessarily mean that the appellant has not driven carelessly. But this begs the question as to whether the appellant was really giving a version of events in mitigation and in the community service report, under which he was not driving carelessly at all. 55. The deputy magistrate and the deputy judge were also right in saying that the defendant had repeatedly admitted that he was careless in the accident. However, the fact remains that he was never asked to elaborate on the reasons why he considered himself to have driven carelessly, and he never did. As such, his admissions added nothing to what he had already admitted to when he agreed with the brief facts which stated, again without elaboration, that he had driven “without due care and attention”. 56. In my view, what the deputy magistrate and deputy judge have both failed to appreciate was that according to the appellant’s story, when the collision occurred, the front of the appellant’s vehicle had already travelled past the injured pedestrian. It was the injured pedestrian who, without paying attention to the road condition and whilst playing with her mobile phone, stepped out from the safety island unexpectedly and ran into the side of the vehicle and got hit by its offside mirror and front wheel. The appellant was in effect saying that it was the pedestrian who hit the side of his vehicle, rather than his vehicle hitting her. That was why he repeatedly said he did not know what he could have done to avoid the accident. 57. The deputy magistrate said to the appellant during the course of his mitigation that when approaching the safety island, he ought to have slowed down his vehicle. However, there was nothing in the brief facts to say that the appellant was driving at an excessive speed at the time. According to the appellant, he managed to stop his vehicle immediately after collision. Moreover, as the deputy magistrate recognised in her statement of findings, the place of the accident was not a zebra crossing and pedestrians had no priority to use the road. 58. Furthermore, the appellant also told the magistrate during mitigation that in fact, his vehicle was driving past the safety island at some distance (measured sideways) from it, so it was not a case of his driving too close to the safety island when driving past it. Rather, according to the appellant, it was the pedestrian who, unlike all other pedestrians who remained standing at the safety island at the time, suddenly stepped out from the safety island and collided with the side mirror of his vehicle and got hit by its front wheel. 59. Thus understood, what has been described by the appellant, if true, may show that he was not driving carelessly in the accident at all. But even if there was any doubt about his guilt on the basis of his own version of events, the application to reverse plea was not an occasion for the deputy magistrate to conduct a mini-trial on bare assertions. In my view, what had transpired should have been sufficient to alert the deputy magistrate to the possibility that the appellant was not guilty as charged, and in the interests of justice, he should not have been denied his day in court – notwithstanding his ill-considered guilty plea, particularly when no real prejudice would appear to result to the prosecution or others by a reversal of plea. 60. As mentioned, the deputy magistrate did not exercise her discretion given her view that her ruling that the plea was unequivocal concluded the application to reverse plea against the appellant. Therefore, on appeal, there was no bar to the deputy judge exercising the court’s discretion in the matter, which, unfortunately, he did not do. If he had realised the true position, he ought to have exercised his discretion to allow the change of plea and quashed the conviction accordingly. 61. However, Mr Sean, for the respondent, argues that as things now stand, the matter should be remitted to the magistrates’ court “for inquiries” to be made regarding whether the appellant should be allowed to reverse his plea, apparently basing his argument on what this court has said in Shum Wan Foon. With respect, that represents a misreading of this court’s judgment. In Shum Wan Foon, the basis for the application to reverse plea was that the defendant was misled by his former lawyers into pleading guilty. It was thus incumbent upon the magistrate to make sufficient inquiries to ascertain the basis of the defendant’s application and to decide if it was sound in fact and in law, before he could properly exercise his discretion on the application. The duty to make sufficient inquiries was said in that context. Here, the appellant’s story has emerged sufficiently clearly from the mitigation and the community service report, based on which the court’s discretion can be properly exercised. Whether the appellant should be believed on his story would be a matter for trial. Indeed, as I have explained, the court’s discretion should have been exercised in favour of granting the application to reverse the plea. There is therefore no point in remitting the matter back to the magistrates’ court to conduct any further inquiries. Rather, the application should be allowed and the conviction quashed. Disposition 62. For these reasons, I would allow the appeal, set aside the order of the deputy judge, and order that the appellant’s appeal against conviction be allowed and his conviction quashed. Given that the appellant has already served the sentence of 150 hours community service, which is by no means a light sentence for a careless driving conviction, I would, exceptionally, not order a retrial.[46] 63. As for costs, I would make an order nisi that the respondent pay to the appellant his costs before us and before the deputy judge, to be taxed if not agreed. Lord Phillips of Worth Matravers NPJ: 64. I agree with the judgment of Mr Justice Cheung PJ. Chief Justice Ma: 65. The court unanimously allows the appeal and makes the order referred to in paragraph 62 of this judgment. As to costs, we make an order nisi that the respondent pay the costs of the appellant in this appeal and the appeal before the deputy judge, such costs to be taxed if not agreed. Should any party seek a different order as to costs, written submissions should be lodged with the Registrar (and served on the other party) within 14 days of the handing down of this judgment, with liberty on the other party to lodge and serve written submissions in reply within 14 days thereafter. If no written submissions are received seeking a different order as to costs before the expiry of the relevant period, the order nisi will become absolute. Mr Edward M H Chan, instructed by N K Lee & Co, for the appellant Mr Eddie Sean, SADPP and Ms Elisa Cheng, PP of the Department of Justice, for the respondent [1] Ms Chiu Wai‑yee. [2] Para 4. [3] Para 7. [4] Para 9. [5] Paras 10 and 13. [6] Para 14. [7] Para 8. [8] Para 12. [9] Para 18. [10] Paras 19-21. [11] Para 24. [12] Para 25. [13] [1978] 2 All ER 751, 754j to 755c. It was cited and applied by the Court of Appeal in HKSAR v Ng Chi Wai [2012] 3 HKLRD 356, para 19. [14] Foster, 754j. [15] [1978] HKLR 522. [16] Page 524. [17] Page 529. [18] Page 529. [19] CACC 326/2005, 30 May 2006. [20] Paras 1 & 15. [21] Paras 14 & 15. [22] Criminal Appeal No 856 of 1978, 20 September 1978. [23] Page 2. [24] Which, when properly drafted, should cover all elements of the offence charged and contain nothing which would suggest a defence. [25] R v Durham Quarter Sessions, ex p Virgo [1952] 2 QB 1, 7. [26] [1971] AC 481. [27] Page 507C. [28] Page 507C-D. [29] Page 496B-C. [30] Pages 489C-E, 498G-499E, 504D-E, 504F & 507E-F. [31] Page 507E-G. [32] S, 507G/H, per Lord Upjohn. [33] [1997] 2 Cr App R 411, 417. [34] Page 755b/c. [35] [2006] 1 WLR 3172, 3179. [36] Page 493F-G. [37] (2014) 17 HKCFAR 303. [38] Paras 11, 12 & 15. [39] Paras 13-17. [40] Unrep, The Times, 5 February 1987. [41] [1983] 3 All ER 689. [42] Statement of findings, para 15. [43] Judgment, para 17. [44] Para 24. [45] Para 25. [46] Similarly, in Shum Wan Foon, this court, exceptionally, did not order a remitter. One significant reason was that the appellant had already served his 6-month term of imprisonment (para 33). 
Mr Justice Ribeiro PJ and Mr Justice Chan NPJ: 1. This appeal raises questions concerning the proper approach to statements relied on by the prosecution as admissions made by a defendant where the statements are equivocal or ambiguous. A. The relevant events 2. In the afternoon of 15 November 2012, the appellant arrived in Hong Kong from Kuala Lumpur. At the airport, an X-ray examination of her suitcase raised suspicions and a Customs Officer, Mr Chan Wai-kei, searched it in her presence. After emptying the suitcase and unzipping the lining, he found concealed within it two packets wrapped in cardboard, tinfoil and paper. He sliced open the tinfoil and some white powder fell out. A rapid test-tube test revealed that it was heroin. 3. At that point, having confirmed that the appellant understood the local dialect, Officer Chan arrested and cautioned her. In his testimony at the trial, he said that the appellant responded in Cantonese, saying: “This suitcase is not mine. It belongs to an African male called Ah Sam.” And when asked what the white powder was, she said in Cantonese: “我諗呢一啲係毒品啩”, translated as “I suppose this is dangerous drug”. 4. Later that day, Officer Chan prepared a post-record, accepted as accurate by the appellant, in which her response was stated to be: “我諗係毒品”, translated as “I think it is drug”. This differed from what the Officer stated at the trial in that the final particle “啩”[1] was omitted, a matter to which we will return. 5. Some hours later, the appellant took part in a video-recorded interview in which she recounted in detail the circumstances leading to her arrival in Hong Kong with the suitcase. When she was asked why she had answered “我諗係毒品”, translated as “I think it is drug” as mentioned above, her explanation was as follows: “Er, it was because, because, because those, those people were checking with a straw on the spot. He said er, it was positive. He said this was, was, was, was, was drug. Well, I, well, well, I myself was stupefied then. He said it was dangerous drug”. 6. She added: “Mm, mm. It was because, it was because the officer on the spot took out, took out a few bottles. There was also, also some colour in there. There was some colour, or, or it could be seen that he was also there --- you, it was because you spoke Cantonese. I could also understand the Cantonese spoken by all of you.” 7. The appellant was charged with a single count of unlawful trafficking in 1.79 kilogrammes of a mixture containing 0.80 kilogrammes of heroin hydrochloride, contrary to section 4(1)(a) and (3) of the Dangerous Drugs Ordinance, Cap 134. B. The Trial Judge’s summing-up 8. At her trial, in a careful summing-up, Deputy High Court Judge A Wong[2], told the jury that the most important factual issue in the case was whether the appellant knew of the existence of dangerous drugs in the suitcase and pointed out that the defence case was that she had no such knowledge.[3] 9. He gave the jury a detailed summary of the evidence regarding the circumstances leading up to and surrounding the appellant’s arrest. This included evidence of the appellant’s background and her account of how she had met an African man in Guangzhou where she resided and owned three small shops selling cosmetic products; of how she was persuaded by him to go first to Nigeria and subsequently to Kuala Lumpur to bring back sample goods; of her having been given the suitcase containing such samples by another African man in Kuala Lumpur and how she landed in Hong Kong on the way back to the Mainland. An important part of the evidence related to the search of her suitcase, the discovery of the concealed packages, the testing of the white powder, her being cautioned and arrested, followed by the verbal exchanges between herself and Officer Chan, her acknowledgment of the accuracy of the post-record and the explanations given by her in the video-recorded interview. 10. As the Judge explained to the jury, the prosecution placed substantial reliance on the appellant’s answers as constituting a confession: “[Officer Chan] told you, when he testified, that the defendant said, ‘I suppose this is dangerous drug’ - ‘我諗呢一啲係毒品啩’. Later, [Officer Chan] made a post-record of the statement made by the defendant. ... In this document, it was recorded that the defendant said, ‘I think it is drugs’ - ‘我諗係毒品’. The prosecution asks you to treat this as a confession of the defendant, that she knew the substance was dangerous drug, and asks you to give this confession full weight.”[4] 11. His Lordship instructed the jury that they had accordingly to decide three matters: “(1) what exactly did the defendant say; (2) what was the effect of her statement; and (3) is what she said true”.[5] 12. The jury were told that, having decided what the appellant had actually said, when they came to consider the effect of her statement, they had to: “... exercise care and examine whether it is a confession on the part of the defendant in the sense that she all along knew the substance was a dangerous drug or whether, having regard to the circumstances, in particular at that time - the substance in powder form had been found and a rapid test yielding a positive result of heroin had been conducted - whether it would have been possible that the defendant made the statement only as a response to her understanding of the finding of the test, only as a reaction to the situation rather than reflecting that she all along had the knowledge that the dangerous drug was inside her suitcase. [Recalling that defence counsel had pointed out that] ... when the defendant made the statement, the officer had already told her that she was arrested for an offence contrary to the Dangerous Drugs Ordinance.”[6] 13. The jury were thus left to decide whether the appellant’s answers amounted to a confession that she had all along known of the drugs found in her suitcase. If the answer was “Yes”, they had to decide whether what she said was true and then decide, in the light of the whole of the evidence, whether she was guilty of trafficking. 14. The appellant was convicted on a five to two majority verdict. C. The decision of the Court of Appeal[7] and leave to appeal to this Court 15. The first two grounds of appeal advanced in the Court of Appeal are relevant for present purposes. They were that there had been a material irregularity in that the trial judge had erroneously failed to hold a voir dire to determine the admissibility of the appellant’s answers to Officer Chan and had wrongly left it to the jury to determine their legal effect when admissibility ought to have been a matter for the Judge.[8] As Poon J,[9] giving the judgment of the Court, noted, counsel then appearing for the appellant: “... placed much emphasis on the Chinese word ‘啩’ used in the oral statement, which he said, indicated suspicion rather than knowledge on the applicant’s part. Thus what the applicant said orally, Mr Wong reasoned, did not amount to a confession.”[10] 16. That argument was rejected, Poon J pointing out that: “... the Judge’s directions to the jury...were: (1) to have regard to the possibility that the statement was said in response to the positive result of the rapid test; (2) the fact that Officer Chan told the applicant that she was arrested for an offence contrary to the Dangerous Drugs Ordinance; (3) to take into account the explanation given in the VRI; (4) to exercise care before treating the statement as a confession; (5) only consider the third question (‘is what she said true?’) if they were sure that the applicant meant by her statement that she all along knew dangerous drug was contained in the Suitcase; and (6) otherwise, they should ignore this evidence altogether.”[11] 17. The Court of Appeal concluded that: “... the way in which the Judge dealt with the oral statement and the post-record was entirely correct. It was plainly a matter for the jury to decide which version they accepted, and having made that decision, to give effect to the oral statement as they found to have been made by the applicant according to the natural meaning of the words used. Contrary to Mr Wong’s argument, no voir dire was required to deal with these matters which fell squarely within the province of the jury. There is no substance in Grounds 1 and 2.”[12] 18. Leave to appeal to this Court was granted by the Appeal Committee[13] on the basis that “it is reasonably arguable that a substantial and grave injustice arose from the appellant being treated as having made an admission regarding knowledge of the presence of dangerous drugs found on the search conducted on her baggage at the airport”. D. The applicable principles 19. The hearsay rule makes out-of-court statements inadmissible as evidence of any fact or opinion stated. But if such a statement[14] constitutes an admission by the accused of some fact in issue, it is admissible in evidence against him as an exception to the hearsay rule. D.1 Where not reasonably capable of being an admission 20. Where it is disputed whether a statement does or does not amount to an admission, the Judge must determine whether the statement is reasonably capable of being an admission of a relevant fact in issue. As pointed out by the Court of Appeal of Victoria in Patrick v R,[15]a statement can only be an admission if the maker “intended by the admission to convey what the admission appears to assert”. If it is not capable of being an admission, then the statement is excluded. Thus, in R v Schofield,[16] when the accused was charged by the police with stealing tarpaulin, he said: “Just my luck”. The magistrates were held to have wrongly treated this as an admission since it might well merely “be taken to be an expression of disappointment” on the defendant’s part at being charged with any offence. 21. To take another example, in R v Toka,[17] the accused, who was suspected of murder, was asked by his partner “if he did it” and replied that he didn't want to discuss it until he had seen his lawyer. The Court rejected the suggestion that this was capable of being understood as an admission “that he had some guilty connection with the crime and needed legal advice”. It was an inference that the jury could not legitimately draw, especially given the defendant’s right of silence. 22. The context of the statement is obviously very important. In the New Brunswick case of R v Taylor,[18] evidence was given by a police officer that the accused had said: “I feel like I’m going to jail”. This was excluded as inadmissible, William T Grant J pointing out that: “It was not said in the context of any conversation but rather when Constable Lyons was seizing clothing from him, his striped shirt, his tee shirt, his jeans, his belt, his socks and his sneakers. In that context it was more likely, in my opinion, that he was verbalizing what was happening to him at that particular time rather than making an admission of guilt. In my view, then, the statement is innocuous and not probative of any fact in issue. I therefore find that it is irrelevant and inadmissible.”[19] 23. Where the question arises as to whether a statement ought to be excluded as not reasonably capable of being an admission, it is obviously desirable that the issue be resolved in the absence of the jury. D.2 Where reasonably capable of constituting an admission D.2a The jury’s task and the directions required 24. Where an equivocal statement relied on by the prosecution is reasonably capable of being an admission, it is left to the jury to decide whether they are satisfied beyond reasonable doubt first, that the defendant did in fact make that statement; secondly, that it does in fact constitute an admission by the accused, probative of a fact in issue; and thirdly, that it is true.[20] In leaving these questions to the jury, it may be necessary to permit evidence to be called to enable them to consider the whole of what transpired so as to place the statement in its proper context. This would require appropriate directions to be given as to how the evidence is to be used, guarding against potential dangers that may arise in the process. 25. The Queen v Marcantelli,[21] where the defendant was convicted of indecently assaulting a female child aged four years and ten months, provides a good illustration. The child’s mother had complained to the police and recounted to an officer what the child had told her about the defendant’s acts. Giving evidence in court, the officer testified that he had informed the defendant of the allegations against him and that the defendant had replied: “Yes, that is correct”. In cross-examination, it was suggested to the officer that the answer “Yes, that is correct” meant only that it was correct that the child's mother had made an allegation against him of indecent assault, but did not mean that the appellant agreed with the allegation that he had committed an indecent assault. It was, as the Judge recognized, an equivocal answer which was reasonably capable of amounting to a confession by the defendant. It was therefore necessary for the jury to decide what meaning to give to it. 26. To enable them to do so, the Crown sought to elicit details of the mother's statement to the police which had thus been narrated to the appellant. After considering the proposed evidence in the absence of the jury, the Judge ruled that it was proper to let the whole conversation between the officer and the accused go to the jury so that they could properly consider what the answer meant. The report goes on to describe the evidence consequently given: “The witness then deposed that he had read to the appellant quite a large section of the mother's statement in the course of which she had stated the child had told her in the presence of the appellant, that the appellant had been touching her bottom with his fingers and also that she had said ‘He kissed my bottom’; that the mother had said the appellant had denied it. The discussion with the accused was terminated by the detective alleging that in fact the appellant indecently assaulted her, told him that those were the allegations made by the mother of the girl, and ‘when I concluded that I informed him he had indecently assaulted the girl’. It was at that stage the accused replied ‘Yes, that is correct’.”[22] 27. It was the Judge’s ruling that such evidence was admissible that the defendant challenged on appeal. Plainly, that evidence was necessary to place the defendant’s statement in its proper context. But it was equally obvious that a careful direction was required since the girl’s complaint as reported by her mother, if treated as evidence of the truth of her allegations, was inadmissible hearsay and potentially highly prejudicial to the defendant. 28. The New Zealand Court of Appeal upheld the Judge’s ruling: “In our opinion the whole of what transpired at the interview ... was admissible and properly admitted for the purpose of enabling the jury to determine what the appellant had accepted as being correct. ... If an answer given amounts to an admission of the truth of the whole or a part of what is related in the complaint against him, it is admissible. If the answer be such as to be evidence from which an acknowledgment might be inferred, the whole of what transpired is properly admissible in order that the question whether or not and to what extent it constitutes an admission may be determined by the jury whose function it is to determine whether his words, action, conduct, or demeanour at the time amounted to an acceptance in whole or in part. The narration to which the answer is made is not admissible to prove the truth of what is narrated, and the jury must be directed as to the use they may properly make of it.”[23] 29. The Court held that the Judge’s directions as to how the contextual evidence could be used were adequate, noting that the summing-up “strongly emphasised that the statements attributed to the child were not in themselves evidence at all, and could be regarded only if the appellant were found to have accepted the statements by the answer he made.”[24] 30. As The Queen v Marcantelli shows, careful directions are required where a jury is asked to decide whether an equivocal statement is in fact an admission. They must clearly be told what their task is and instructed that if they have any reasonable doubt as to whether the statement was in fact made or whether it is an admission, that they should ignore it altogether. Where evidence which is otherwise inadmissible is allowed to be adduced to provide the context of the statement, the jury must be directed as to how they may or may not use such evidence. They must also be told that if they do find that it is an admission, they must decide whether they are satisfied beyond reasonable doubt that it is true in substance. 31. As Barwick CJ, Gibbs and Mason JJ stated in their joint judgment in R v Burns:[25] “It is clear and elementary law that once a confessional statement has been admitted into evidence its weight and probative value are matters for the jury. It is for the jury to determine whether the alleged confession was made and whether it was true in whole or in part. Unless the jury are satisfied that so much of a confession as tends to show the guilt of the accused was true they cannot treat it as a proof of guilt. However, a confessional statement may be only one piece of the evidence against the accused and the jury are entitled to consider all the relevant evidence together in deciding upon their verdict. The nature of the direction necessary to be given properly to instruct the jury as to the use of evidence of an alleged confession must depend on all the circumstances of the case.” D.2b The use that can be made of the admission 32. The relevant fact in issue should be identified so that the jury know what it is the prosecution is seeking to prove that the defendant has admitted and so that they can focus on whether the defendant’s statement actually amounts to an admission of that fact (and not of some irrelevant fact). Thus, for instance, in R (on the application of Wright) v Crown Prosecution Service,[26] the fact in issue was whether the mushrooms in the defendant’s possession were a Class A drug, namely psilocybin mushrooms (referred to as “magic mushrooms”). The prosecution had failed to adduce scientific evidence as to the type of mushroom in question and sought instead to rely on an alleged admission by the defendant. When interviewed, the defendant had admitted that he had wanted to find magic mushrooms but his statements were held to have fallen “a long way short of an admission that he had actually picked magic mushrooms”. It was an admission that did not relate to the fact in issue. 33. In this context, it should be borne in mind that admissions may be used to prove facts in issue which differ in their relative importance and probative value in relation to the outcome of the case. Thus, in The Queen v Marcantelli, a finding by the jury that the appellant had in fact agreed with the allegation that he had committed an indecent assault would be of great probative value, since it would amount to a finding that he had confessed to the crime. In other cases, an admission may be relied on only to establish a fact which is not decisive, concerning for instance the defendant’s or the complainant’s credibility or to prove a fact from which the jury might be invited to draw an inference bearing on the defendant’s guilt. 34. An instance where an admission was used to undermine the credibility of the defendant’s denials and to support the evidence of the complainant may be found in AM v Western Australia.[27] The defendant was a bar manager and the complainant a girl just under 15 years of age who worked at the bar. She gave evidence that after the restaurant had closed one Friday night, leaving her and the defendant alone, he had given her two Southern Comforts and Coke and then, behind a dividing wall towards the front of the restaurant, he had forced sexual intercourse with her. The defendant denied this and testified that when the restaurant was closed, he was always accompanied by his wife or his mother or father. However, a witness testified that he and the defendant had subsequently had a conversation in which the witness said to the defendant: “... ‘What's this I hear about you shagging young waitresses,’ or something along those lines, and he replied that, you know, a few beers and away you go sort of thing. Along those lines ... … I sort of asked him what had happened and he sort of said, ‘Behind the counter’.”[28] 35. While such evidence did not identify the complainant or the time when sexual intercourse allegedly occurred, the Court of Appeal of Western Australia upheld the trial judge’s decision to admit it as capable of being a relevant admission. Steytler P held that: “It amounted to evidence of admissions that were adverse to the appellant's case. The evidence, if accepted, established that the appellant had admitted that he had, on one occasion at least, been on his own with a waitress, who was not a family member, at the restaurant. It also established that the waitress had been young, that the appellant had sex with her after a couple of drinks and that the sex had taken place behind a counter (although the complainant described it as having been behind a dividing wall). Although falling well short of a confession that he had committed the offence charged, or any offence, the evidence had sufficient probative value to justify its admission.”[29] And Murray AJA stated: “...it was by no means a confession that the appellant sexually penetrated the 14-year-old complainant in the restaurant on a night in about early October 2001 ... Much less was it a confession that sexual penetration, always denied by the appellant, occurred without the complainant's consent. ... The importance of the evidence was that it contradicted the appellant's denial, supported to some extent by the evidence of his mother, that any such activity, consensual or otherwise, involving the appellant and the complainant, or any other young waitress, in the restaurant, with or without consent, ever occurred. But its probative value was certainly limited in that way.”[30] 36. R v SJRC,[31] is an example ofa case where ambiguous statements were relied on as admissions permitting inferences adverse to the defendant to be drawn. He was charged with indecent assault and sexual intercourse without the complainant’s consent and the prosecution successfully argued that two text messages he had sent were capable of amounting to an admission that sexual acts had occurred between himself and the complainant on the day in question. The fact that the defendant was expressing regret in those messages was also capable of permitting an inference to be drawn that he knew that the complainant had not consented to the sexual acts.[32] D.2c Whether prejudicial effect outweighs probative value 37. As Li CJ noted in Secretary for Justice v Lam Tat Ming:[33] “The Judge has the overriding duty to ensure a fair trial for the accused according to law. For this purpose, he has what should be regarded as a single discretion to exclude admissible evidence, including a voluntary confession, whenever he considers it necessary to secure a fair trial for the accused.” In particular: “The Judge may in his discretion exclude admissible evidence where its prejudicial effect is out of proportion to its probative value.”[34] 38. The circumstances may make it incumbent on the Judge to consider exercising that residual exclusionary discretion where the probative value of admitting an equivocal statement into evidence as an admission may be outweighed by its prejudicial effect. Thus, the statement in question may be so equivocal that it borders on being incapable of amounting to an admission, making its probative value so slender as to be outweighed by its prejudicial effect.[35] 39. In some cases (of which The Queen v Marcantelli[36] provides an illustration), the need to adduce evidence to enable the alleged admission to be assessed in its proper context may involve exposing the jury to highly prejudicial material. Having considered in the absence of the jury the extent to which such prejudice might effectively be dealt with by appropriate directions, the Judge might conclude that the probative value of the alleged admission is insufficient to justify putting the fairness of the trial at risk by introducing the prejudicial matter. 40. And in gauging the probative value of the alleged admission, its scope and relative importance should be borne in mind. As illustrated by AM v Western Australia and R v SJRC,[37] it may, at one end of the spectrum, potentially constitute a full confession of guilt while, at the other end, it may merely provide corroborative evidence or supply one piece of evidence amongst others, permitting a relevant inference to be drawn. 41. Where the trial judge has exercised his or her discretion by declining to exclude the equivocal admission, the appellate court’s role is limited, as stated in the joint judgment in Kissel v HKSAR:[38] “The Judge’s decision not to exclude the relevant evidence as being disproportionately prejudicial involved the exercise of a judicial discretion which, as is well-established, will only be interfered with on appeal in limited circumstances. The principles are clearly stated in the joint judgment of Dixon, Evatt and McTiernan JJ in House v The King[39] as follows: ‘The manner in which an appeal against an exercise of discretion should be determined is governed by established principles. It is not enough that the judges composing the appellate court consider that, if they had been in the position of the primary judge, they would have taken a different course. It must appear that some error has been made in exercising the discretion. If the judge acts upon a wrong principle, if he allows extraneous or irrelevant matters to guide or affect him, if he mistakes the facts, if he does not take into account some material consideration, then his determination should be reviewed and the appellate court may exercise its own discretion in substitution for his if it has the materials for doing so. It may not appear how the primary judge has reached the result embodied in his order, but, if upon the facts it is unreasonable or plainly unjust, the appellate court may infer that in some way there has been a failure properly to exercise the discretion which the law reposes in the court of first instance. In such a case, although the nature of the error may not be discoverable, the exercise of the discretion is reviewed on the ground that a substantial wrong has in fact occurred.’” 42. But where the lower court ought to have, but failed to, exercise its discretion, the appellate tribunal may itself exercise the residual discretion.[40] D.2d The principles summarised 43. The aforesaid principles may be summarised as follows: (a) Where the prosecution seeks to rely on an equivocal statement as an admission, the Judge should consider whether it is reasonably capable of constituting an admission probative of a relevant fact in issue. (b) If it is so capable, subject to exercise of the court’s residual exclusionary discretion, it is admissible and should be left to the jury to decide if it does in fact constitute an admission, to be relied on as an exception to the hearsay rule. If it is not reasonably capable of being an admission, it is inadmissible. (c) If the statement is left to the jury, they should be permitted to consider the whole of what transpired, placing the statement in context. (d) The jury should be told that they can only rely on the statement against the defendant if they find beyond reasonable doubt: (i) that the defendant in fact made the claimed admission; (ii) that, viewed in context, it was indeed intended to be an admission probative of a fact in issue; and (iii) that the substance of the admission is truthful. (e) The judge should give appropriate directions as to how to deal with the statement, drawing attention to its ambiguity and other possible meanings and indicating any matters that the jury may take into account in deciding whether an admission was in fact intended. The judge should also give appropriate directions in respect of how any contextual evidence may or may not be used. (f) The Judge should also identify for the jury, if they find that the statement is an admission, what its scope or limits are; ie, what it is that the defendant may be found to have admitted. (g) Since an equivocal statement may have little probative value, the judge has to consider whether to exercise his discretion to exclude the statement on the ground that its probative value is outweighed by a risk of unfair prejudice. The more equivocal the statement, the less may be its probative value. Depending on its content and relevance, the fact in issue provable by the statement may be of greater or lesser probative value in relation to the offence as a whole. Evidence admitted to establish the statement’s context may be highly prejudicial. E. The principles applied to the present case 44. The learned Judge’s direction to the jury was meticulous and fair. However, with respect, in our view, he erred in failing to consider whether, examined in its context, the defendant’s response to Officer Chan was reasonably capable of constituting an admission of the relevant fact in issue. Moreover, if the appellant’s response was to be treated as admissible, his Lordship erred in failing to consider whether its prejudicial effect was disproportionate to its probative value. E.1 Not reasonably capable of being a relevant admission 45. The fact in issue which the prosecution sought to prove in reliance on the appellant’s response was that “she all along had the knowledge that the dangerous drug was inside her suitcase”.[41] In considering whether the appellant’s response was reasonably capable of being an admission that she had such knowledge, the context in which her statement was made is all important. It was a response given after her suitcase had been searched in her presence, having been emptied of its contents and its lining having been unzipped to reveal the two packets concealed within; after one of the packets had been sliced open and white powder extruded; after the test-tube test had been conducted and the Customs Officers heard by the appellant to say that it was positive; and after Officer Chan had arrested and cautioned the appellant telling her that she was suspected of having committed an offence under the Dangerous Drugs Ordinance. Only then was she asked what the white powder was, and only then did she say: “我諗呢一啲係毒品啩”, translated as “I suppose this is dangerous drug”, or in the post-record version: “我諗係毒品”, translated as “I think it is drug”. 46. Although there is a material difference in the meaning of the two versions (dealt with below), given the abovementioned context of the appellant’s response, made only after the Customs Officers had graphically demonstrated to her that the substance was heroin, we do not consider her statement reasonably capable of being an admission that she knew “all along” that she was carrying dangerous drugs in her suitcase. This case is akin to R v Taylor,[42] where it was held that the defendant’s statement: “I feel like I’m going to jail”, made when his clothing was being seized from him, could fairly only be understood as him “verbalizing what was happening to him at that particular time rather than making an admission of guilt”. 47. This conclusion is reinforced by the content of the appellant’s response in the version to which Officer Chan testified, namely, where the appellant said: “我諗呢一啲係毒品啩”, translated as “I suppose this is dangerous drug.” The addition in this version of the final particle “啩”significantly qualifies the preceding words, as pointed out in a linguistic study of how such Cantonese final particles are translated in bilingual court proceedings published by Ester Leung and John Gibbons:[43] “Usually 啩/gwa/ is used to signify a degree of uncertainty about a preceding utterance, which allows the speaker to be non-committal concerning what s/he has said. The speaker is indicating compliance in general but at the same time reflecting some reservation. This meaning resource may be important in the testimony of witnesses who wish to register their uncertainty about the propositional content, particularly when a counsel puts a proposition to them for confirmation.” 48. The injection of this sense of uncertainty and the indication that hers was a non-committal response by the addition of the “啩” final particle reinforces the conclusion that the appellant’s statement was not intended to be an admission that she knew “all along” that she had dangerous drugs in her suitcase. E.2 Probative value outweighed by prejudicial effect 49. As indicated above, our view is that the appellant’s response is incapable of constituting an admission of the fact of knowledge sought to be proved by the prosecution. Even if one were to take a different view and treat her statement as possibly being capable of being understood as such an admission, it would, given its context – especially in the version with the final particle “啩” attached – be so equivocal and qualified that its probative value would be extremely limited and would plainly be outweighed by the risk of unfair prejudice which would flow from leaving it to the jury as a possible admission on her part. 50. The trial judge not having considered exercising his residual discretion, we consider that, in so far as necessary, this Court should itself exercise such discretion and exclude that statement. F. Retrial 51. It was held in Lau Ka Yee v HKSAR,[44] that a conviction which is based on an equivocal admission with no other evidence in support will be quashed. It cannot however be said that the present case is one where the prosecution is able to tender no other evidence in support of its case against the appellant. Accordingly, while we are of the view that her appeal should be allowed and the conviction quashed because her purported admission was erroneously left to the jury, we would order that there be a re-trial. That is a course which counsel appearing on her behalf accepted would be justified in the event that she was to succeed on this appeal. Mr Justice Tang PJ: 52. I agree with the judgment of Mr Justice Ribeiro PJ and Mr Justice Chan NPJ. Mr Justice Fok PJ: 53. I agree with the judgment of Mr Justice Ribeiro PJ and Mr Justice Chan NPJ. Lord Millett NPJ: 54. I agree with the judgment of Mr Justice Ribeiro PJ and Mr Justice Chan NPJ. Mr Justice Ribeiro PJ: 55. The appellant’s appeal is unanimously allowed. We order that her conviction be quashed and that there be a retrial. Mr Shahmim K. Khattak and Ms Holly Cheng, instructed by Annie Leung & Co., for the Appellant Ms Anna YK Lai SC, DDPP (Ag.), of the Department of Justice, for the Respondent [1] Phoneticized as “gwaa3”. [2] As Mr Justice A Wong then was. [3] Summing-up pp 2 and 3. [4] Summing-up p 12. [5] Ibid. [6] Summing-up pp 12 and 13. [7] Lunn VP, Poon and D Pang JJ, CACC 81/2014 (31 March 2015). [8] Court of Appeal §15(1) and (2). [9] As Poon JA then was. [10] Court of Appeal §16. [11] Court of Appeal §21. [12] Court of Appeal §22. [13] Ribeiro, Tang and Fok PJJ, FAMC 35 of 2015 (26 May 2016). [14] Here, “statement” is used to include conduct or a combination of words and conduct. [15] (2014) 42 VR 651 at 663. [16] (1917) 12 Cr App R 191. [17] (1994) 1 NZCrimC 275. [18] 2008 NBQB 340. [19] At §29. [20] Choudhary v R [2013] VSCA 325 at §50 (citing R v MMJ (2006) 166 A Crim R 501 at §70) and at §56. [21] [1962] NZLR 974. [22] At 977. [23] Ibid. [24] At 978. [25] (1975) 132 CLR 258 at 261. [26] [2015] EWHC 628 (Admin). [27] [2008] WASCA 196. [28] At §5. [29] At §14. [30] At §191 and §195. [31] [2007] NSWCCA 142. [32] At §30. [33] (2000) 3 HKCFAR 168 at 178-179. [34] At 179. [35] Lau Ka Yee v HKSAR (2004) 7 HKCFAR 510 at §53. [36] [1962] NZLR 974, discussed in Section D.2a above. [37] [2008] WASCA 196 and [2007] NSWCCA 142, discussed in Section D.2b above. [38] Li CJ, Chan and Ribeiro PJJ and Sir Anthony Mason NPJ (2010) 13 HKCFAR 27 at §120. [39] (1936) 55 CLR 499 at 504-505. [40] R v Cook (1959) 2 QB 340; Wong Ching Chu v R [1957] 2 HKLR 61. [41] Summing-up p 13. Italics supplied. [42] 2008 NBQB 340, discussed in Section D.1 above. [43] Ester Leung and John Gibbons, “Interpreting Cantonese utterance-final particles in bilingual courtroom discourse” in “Interpreting Chinese, Interpreting China”, Ed Robin Setton (John Benjamins Publishing company) at p 97. [44] (2004) 7 HKCFAR 510 at §53. 
Chief Justice Ma: 1. I agree with the joint judgment of Mr Justice Ribeiro PJ and Mr Justice Spigelman NPJ. Mr Justice Ribeiro PJ and Mr Justice Spigelman NPJ: 2. The appellant was convicted of offences under the Elections (Corrupt and Illegal Conduct) Ordinance (“ECICO”)[1] in connection with the District Council Election held on 22 November 2015. In particular, he was convicted after trial before HH Judge CP Pang[2] on charges of engaging in corrupt conduct by corruptly offering an advantage consisting of sums of money to another person (i) as an inducement for that person to stand as a candidate at the election (contrary to section 7(1)(a)); and (ii) as an inducement for that person to get, or try to get, a third person to stand as a candidate at the election (contrary to section 7(1)(c)). He was also convicted of conspiracy to engage in corrupt conduct contrary to section 7(1)(g).[3] A. The proceedings below 3. The Judge found that the appellant had approached certain persons associated with “localist” political agendas, offering them cash in sums of “up to $250,000” as an inducement either to stand themselves, or to get others to stand, for election in designated constituencies with a view to diverting or “undercutting” votes which might otherwise be cast in favour of candidates with “pan-democratic” sympathies.[4] The charges concerned the advantage offered by the appellant to Ku Ka Ho (“Ku”) to arrange for Chan Kin Loong (“Chan”) to stand as a candidate in a designated constituency. Ku and Chan accepted the offer and were charged as 2nd and 3rd defendants alongside the appellant. 4. At the trial, the appellant admitted having made the approaches and having offered what was referred to as a “sponsorship plan” to potential localist candidates. His defence was that he did not genuinely intend to induce them to stand as candidates but that his offer was a pretence to enable him to uncover and expose on his online radio station “shady affiliations” between localist organizations and pan-democratic political parties.[5] 5. The Judge rejected his evidence, finding that “... the election proposal mentioned by [the appellant] was not a pretence as he claimed” but that “[his] real intention was to use the sponsorship to induce others to stand for the election in designated constituencies.”[6] His Honour held that in offering the advantage, the appellant had “... used money to get people to stand for elections in order to manipulate the result of elections”, being conduct which “would prevent elections [from being] conducted fairly, openly and honestly”.[7] 6. The appellant was sentenced to a total term of imprisonment of 4 years. The Court of Appeal[8] dismissed the appellant’s appeal against conviction but reduced his sentence to a total term of 3 years and 3 months’ imprisonment. B. The issue on this appeal 7. The offence with which we are concerned is created by ECICO section 6 which states: “A person who engages in corrupt conduct at an election commits an offence”. If tried summarily, the defendant is made liable on conviction to a fine of $200,000 and to imprisonment for 3 years; and liable to a fine of $500,000 and to imprisonment for 7 years if tried on indictment. 8. The scheme of the ECICO is more closely examined in Section C of this judgment, but for immediate purposes, the relevant variant of the section 6 offence created by section 7(1)(a) is materially defined as follows: “(1) A person engages in corrupt conduct at an election if the person corruptly— (a) offers an advantage to another person as an inducement for the other person— (i) to stand, or not to stand, as a candidate at the election...” 9. Likewise, the pertinent variant of the section 6 offence defined by section 7(1)(c) relevantly provides: “(1) A person engages in corrupt conduct at an election if the person corruptly— (c) offers an advantage to another person as an inducement for the other person to get, or try to get, a third person— (i) to stand, or not to stand, as a candidate at the election ...” 10. In the Courts below, the true construction of section 7(1) was considered. In particular, the meaning and effect of the word “corruptly”, qualifying the words “offers an advantage” was in issue. As the trial Judge and the Court of Appeal adopted different approaches and as disparate authorities relied upon in the debate appear to be of questionable relevance, the Appeal Committee[9] granted leave to appeal, certifying as a suitably important question of law, the following: “What is the meaning of the word ‘corruptly’ in section 7(1) of the [ECICO]?” 11. The determination of this question also involved considering whether the word had relevance to the mens rea element of the offence created under section 7(1). 12. For present purposes, unless otherwise indicated, the discussion can proceed in relation to section 7(1) without distinguishing between section 7(1)(a) and section 7(1)(c) since the issues arising generally apply to both provisions. C. Construction of section 7(1) 13. It is well-established that the proper approach to statutory construction is contextual and purposive. As Sir Anthony Mason NPJ stated in HKSAR v Lam Kwong Wai:[10] “The modern approach to statutory interpretation insists that context and purpose be considered in the first instance, especially in the case of general words, and not merely at some later stage when ambiguity may be thought to arise ...” C.1 ECICO section 3: the statutory purpose 14. The purpose of the ECICO in general and of section 7(1) in particular must be examined, placing that section, including the word “corruptly”, in its appropriate context. In undertaking that exercise, ECICO section 3 is of prime importance as an aid to construction since it expressly sets out the objects of the Ordinance, relevantly stating: “The objects of this Ordinance are— (a) to ensure that, as far as practicable, [the relevant elections[11]] ... are conducted fairly, openly and honestly and are free from corrupt conduct and illegal conduct ...” 15. Thus, section 3 identifies as a positive objective, conducting elections “fairly, openly and honestly”; and as a negative objective, keeping elections “free from corrupt conduct and illegal conduct”. This strongly indicates that construed purposively, conduct which the Ordinance qualifies as “corrupt and illegal” is intended to be conduct which is antithetical to the achievement of “fair, open and honest” elections. Section 3 therefore gives the inherently vague concept of “corrupt” a purposive and contextual framework: conduct which is prohibited as “corrupt” is activity which has a tendency to subvert “fair, open and honest” elections. C.2 Context: the structure of Part 2 of the ECICO 16. The structure of Part 2 of the ECICO headed “Corrupt Conduct” is somewhat unusual. As pointed out above, section 6 creates the offence of engaging in corrupt conduct at an election and prescribes the maximum sentences. Sections 7 to 21 respectively then proceed to specify different forms of conduct constituting separate variants of that offence. 17. An examination of those variants reveals that the term “corrupt conduct” in Part 2 is used in a special, expanded way. While some instances of the offence do involve “corruption” in the usual sense of involving some form of “bribery and corruption”, ie, illicit conduct calculated to subvert or deflect an official or agent from the proper exercise of his or her duties, “corrupt conduct” in the context of Part 2 clearly covers a broader range of behaviour. 18. The activities prohibited as “corrupt conduct” may be placed roughly in four groups: (i) conduct involving some form of “bribery” (sections 7, 11, 12 and 21);[12] (ii) conduct involving use of force or duress (sections 8 and 13);[13] (iii) conduct involving deception (sections 9, 14 and 15);[14] and (iv) conduct involving interference with the election process (sections 10, 16, 17, 18 and 20).[15] Conduct under the last three headings would not normally be called “corrupt”, but in the context of Part 2 and viewed purposively in the light of section 3, the unifying theme of the variants of the offence is evidently the promotion of fair, open and honest elections and the prohibition of contrary conduct, indicating the special sense in which “corrupt conduct” is used. 19. It is in this context that a peculiarity of section 7(1) comes to the fore. It is the only variant of the offence of engaging in corrupt conduct, created by ECICO sections 6 and 7, in which the word “corruptly” is used to qualify the prohibited conduct in question. As we have seen, section 7(1)(a) materially provides: “(1) A person engages in corrupt conduct at an election if the person corruptly— (a) offers an advantage to another person as an inducement for the other person— (i) to stand, or not to stand, as a candidate at the election [etc]; 20. In each of the paragraphs of section 7(1), the forbidden behaviour – consisting of offering, soliciting or accepting an advantage – is defined as the offence of “engaging in corrupt conduct”. The question therefore arises as to why section 7(1) introduces the concept of corruption a second time by applying the word “corruptly” to each paragraph. Why is this needed in relation to offering, soliciting or accepting an advantage but not needed in respect of the other variants of the offence, involving bribes, force or duress, deception and interfering with the election process? 21. In our view, the answer lies in the nature of the section 7(1) offences involving the offering, etc, of an “advantage” which is a very broadly defined concept, and the need to set the boundaries of those offences by reference to the purposive and contextual construction here adopted. 22. As noted above, on a purposive construction of Part 2, the unifying purpose of the variants of the offence of engaging in corrupt conduct (including the section 7(1) instances) is the promotion of fair, open and honest elections and the prohibition of antithetical conduct. The variants other than those defined by section 7(1), specific in nature, [16] may be thought self-evidently to promote that purpose. However, by contrast, the section 7(1) variants are all concerned with “advantages”, a concept given a very broad definition which may well encompass innocuous conduct as well as matters or transactions which are properly prohibited as detrimental to fair elections. ECICO section 2(1) materially defines “advantage” as follows: “(1) In this Ordinance, unless the context otherwise requires— advantage means— (a) any valuable consideration, gift or loan; or (b) any office, employment or contract; or (c) the full or partial payment, release, discharge or liquidation of an obligation; or (d) the exercise of or forbearance from exercising a right or power; or (e) the performance of or forbearance from performing a duty; or (f) any favour, including— (i) giving protection from a liability incurred or anticipated; and (ii) giving protection from proceedings or possible proceedings of a disciplinary, civil or criminal nature; or (g) any other service (other than voluntary service or the provision of entertainment), but does not include an election donation if particulars of the donation are given in an election return that has been lodged with the appropriate authority ...” 23. One can well conceive of conduct which may fall within the abovementioned definition (especially paragraphs (f) and (g)) but which does not affect the integrity of an election. A hypothetical example was discussed at the hearing: D, a member of the Legislative Council, considers a young member of the same party as a promising potential candidate for election to the legislature. He offers that person employment as his research assistant at a modest salary in order to give him insights into the work of a legislator in order to prepare him for possible candidacy in elections due to be held a year later. Such employment qualifies as an “advantage”. And while the offer of such employment might be thought to fall within section 7(1)(a) as an inducement for the young person to stand as a candidate at the election, it may be difficult to see how criminalisation of such an arrangement would be justified. Much of course depends on the facts of any given case. 24. In our view, the insertion of “corruptly” in section 7(1) indicates a legislative intention to introduce a purposive limit on the broad definition of “advantage”. The offer, etc, of the advantage must be made “corruptly” before an offence is committed. “Corruptly” in this context should be understood to be an invocation of the statutory purpose laid down by ECICO section 3. As noted above,[17] giving effect to that purpose involves understanding “corrupt” activity to be conduct having a tendency to subvert “fair, open and honest” elections. 25. Thus, construed purposively in the context of section 3, the word “corruptly” in section 7(1) operates to confine the offence to conduct which has an objective tendency to undermine “fair, open and honest” elections. Accordingly, if a defendant is charged with offering an advantage to another person as an inducement for that person to stand or not to stand as a candidate at an election, one has to examine the alleged advantage and ask whether it was “corruptly offered”, meaning, an advantage of such a nature and offered in such circumstances as to have a tendency to undermine fair, open and honest elections. If not, the alleged conduct does not qualify as “corrupt” and the offence is not committed. 26. We have expressed the view that the other variants of the offence self-evidently align with the statutory purpose. However, to the extent that doubt exists in a specific case, the same approach is properly applicable. Sections 7 to 21 are to be construed purposively as creating offences which advance and are confined by the statutory objectives, both positive and negative, set out in section 3. D. The Judge’s approach to construction of section 7(1) 27. Important aspects of the Judge’s approach reflect the reasoning adopted in this judgment. Thus he noted the importance of ECICO section 3 in a contextual and purposive construction[18] and recognised the need to avoid criminalising “ordinary, reasonable and even necessary conduct of political parties and election campaigns”.[19] As His Honour rightly held: “For the purpose of section 7 offences, what the Ordinance expressly means to forbid is conduct which will undermine the objects of the Ordinance as stipulated in section 3. One of the objects of the Ordinance is to ensure that elections are conducted fairly, openly and honestly and are free from corrupt conduct and illegal conduct (see Mok Charles Peter v Tam Wai Ho FACV 2/2012). In other words, the conduct which [it] is meant to forbid is conduct which will hinder elections to be conducted fairly, openly and honestly.”[20] 28. However, with respect, we part company with the Judge in two major respects. First, we do not consider that decided cases, especially those in the 19th Century,[21] touching on use of the word “corrupt” in English legislation which is very differently structured and worded, provide any guidance in the interpretation of the relevant provisions of the ECICO. Nor do we consider it helpful to refer to case-law[22] on the Prevention of Bribery Ordinance[23] or its predecessor the Prevention of Corruption Ordinance,[24] which pursue different statutory purposes and raise very different issues. It is, however, entirely understandable that His Honour should have made reference to these materials as they have been relied on in previous decisions of the Hong Kong courts and also relied on by the present parties and the amicus in this Court. 29. Secondly, we do not agree with the Judge’s conclusion that on its true construction, section 7(1) imposes what he calls a requirement of “double intent”. His Honour held as follows: “The prosecution is required not only to prove the intention of the defendant to induce others or be induced by others to stand or not to stand for election, it is also required to prove the intention of the defendant to hinder elections to be conducted fairly, openly and honestly. In other words, the prosecution is required to prove ‘double intent’.”[25] “The word ‘corruptly’ is injected with subjective element. The prosecution shall prove beyond reasonable doubt that the defendant has the intention to corrupt. If a defendant may not purposely perform an act which is forbidden by law, he shall be acquitted.”[26] “I consider that the factors which the court should take into consideration include: what is the value of the advantage offered or accepted by the defendant, is he in breach of his duty of a voter or a candidate, is he in conflict of interest, what are his objects and intentions, is his conduct open or secretive, is he in breach of any rules and what is his influence. The court should also consider all the circumstances in the case.”[27] 30. In our view, the mens rea requirement under section 7(1) is an intention to do the prohibited acts which constitute the actus reus of the offence of engaging in corrupt conduct. Thus, a defendant charged with an offence contrary to section 7(1)(a)(i) must be proved to have intentionally offered an advantage to the relevant person and intended it to be an inducement for that person to stand, or not to stand, as a candidate at a particular election. There is no need for the prosecution to prove a specific intention on the defendant’s part to “hinder elections to be conducted fairly, openly and honestly” and the court is not called upon to assess the various factors enumerated by the Judge.[28] As explained above,[29] the word “corruptly” functions in section 7(1) to confine dealings involving the offering, soliciting or receipt of “advantages” to conduct which carries an objective tendency to undermine “fair, open and honest” elections. It does not introduce an additional mens rea requirement. E. The Court of Appeal’s approach 31. The Court of Appeal rejected the Judge’s “dual intent” construction[30] and adopted a very different approach of its own. 32. We have emphasised the statutory purposes expressed in section 3(a) as follows: “The objects of this Ordinance are— (a) to ensure that, as far as practicable, [the relevant elections] ... are conducted fairly, openly and honestly and are free from corrupt conduct and illegal conduct ...” 33. However, the Court of Appeal[31] emphasised paragraph (c) which provides: “The objects of this Ordinance are ... (c) to ensure that candidates properly account for the expenditure of money at elections and the soliciting and receipt of election donations and that they do not exceed the prescribed levels of expenditure.” 34. Embarking on their construction with their focus on election donations and the statutory regime that limits candidates to using election donations only towards election expenses, their Lordships developed an interpretation of “corruptly” in section 7(1) revolving around the prohibition of personal gain from standing for election. 35. Yeung VP held as follows: “The legislative intent of the Ordinance is clear: a candidate is entitled to accept election donations, but such donations can only be used towards election expenses as prescribed in the Ordinance. Where election donations exceed election expenses, the former must be disposed of in accordance with the mechanism set out in section 19 of the Ordinance. In short, a candidate must not have any gains by reason of standing for an election. Section 7(1) of the Ordinance also aims to prohibit candidates from having any personal gains as a result of standing for an election.”[32] “In my judgment, where a person intentionally engages in conduct prohibited under the ‘Ordinance’, i.e. conduct characterized by the Ordinance as ‘corrupt’, and one of the purposes of doing so is for personal gain of a particular individual or organization, then that person is ‘corruptly’ engaging in such conduct, which constitutes an offence prohibited under the Ordinance.”[33] 36. Applying this to the appellant: “... One of the conditions Cheng put forward was to cause the person who stood for the election in accordance with his instruction to obtain personal gain. The above conduct is plainly ‘corrupt conduct’ which the Ordinance is meant to prohibit.”[34] 37. Poon JA[35] followed the same course, centring on the limits of permissible use of election donations[36] and reaching the same conclusion as to the statutory purpose of forbidding personal gain: “As this is the legislative purpose of the Ordinance vis-à-vis election donations,[37] the statutory definition of ‘advantage’ encompasses virtually all sorts of benefits, but at the same time does not include election donations properly declared in accordance with the law, the purpose being to prohibit candidates from obtaining any personal gain out of the election. In the context of section 7(1), its purpose is to prohibit anyone at the election from obtaining any personal gain out of the election.”[38] 38. While holding that “corruptly” involves a mens rea requirement, his Lordship held that a “dual intent” was unnecessary given the “personal gain” construction adopted: “The offences under section 7(1) aim to prohibit anyone from obtaining personal gain from an election. Any act of offering, soliciting or accepting an advantage goes against the legislative purpose of enacting the section 7(1) offences and will certainly prevent an election from being conducted fairly, openly and honestly. In other words, the section 7(1) offences must be inherently culpable. Therefore, as long as the offender had the intent to commit the corrupt conduct in question, he must have acted ‘corruptly’. The dual intent mentioned by the trial judge is unnecessary.”[39] 39. With respect, we are unable to agree with the Court of Appeal’s construction of section 7(1) and the word “corruptly” which it contains. 40. In the first place, the Court of Appeal’s identification of personal gain as the touchstone of “corruptly” appears to have been driven by its focus on the ECICO’s provisions limiting the use of election donations. But it is hard to see why the present case should be thought to have anything to do with election donations. The money offered and handed over by the appellant to Ku and Chan did not constitute an “election donation” or an offer of an election donation. Such donations are defined to involve money, goods or services given to or in respect of a candidate towards meeting election expenses.[40] But neither Ku nor Chan were “candidates” since, as Yeung VP noted, “they had not stood nominated as candidates and had never publicly declared an intention to stand as candidates”[41] and so did not come within the relevant statutory definition.[42] The situation with which we are concerned – offering an advantage to non-candidates to stand – is specifically dealt with by section 7(1) and not by the provisions concerning election donations. 41. Secondly, the Court of Appeal’s emphasis on “personal gain” appears to elide the concept of “advantage” with the concept of making an offer “corruptly”. 42. To constitute the section 7(1) offence, it must be proved both that the defendant offered an advantage and that he did so corruptly. The Court of Appeal has centred on personal gain for the offeree. An offer to confer such a benefit no doubt constitutes an offer of an “advantage” within the definition set out above.[43] Proof of an offer of personal gain would therefore establish the “advantage” ingredient of the offence. But to assert that proof of an offer to confer a personal gain on the candidate establishes not merely the offer of an “advantage” but also that such personal gain establishes that the offer was “corruptly” made elides the two concepts of “advantage” and “corruptly”, treating them as having no meaning independent of one another. It is therefore unsatisfying to interpret “corruptly” as meaning the conferment of “personal gain”. 43. Thirdly, the offence as defined by section 7(1)(a) to (d) focuses on the conduct of the offeror and not on the benefit offered to or conferred on the offeree. The Court of Appeal’s approach erroneously concentrates on the corruption involved in the offeree making a personal gain rather than on the offeror’s corrupt act of making the offer of an advantage which is the act prohibited by the statute. If the offeror’s act of offering an advantage to a person as an inducement for standing, or not standing, as a candidate at the election has a tendency to subvert “fair, open and honest” elections, whether or not he thereby acquires personal gain, his offer is made “corruptly”, construing section 7(1) purposively in the light of section 3. F. The appellant’s submissions 44. The appellant placed before the Court a volume of material in which the word “corruptly” had been used or discussed. This included religious as well as legal material. The thrust of his submission was that the Court should not apply a line of precedent stretching from 19th Century English authority, particularly Cooper v Slade.[44] He submitted that this case law was influenced by its feudal and religious origins and was inappropriate for application in contemporary Hong Kong, with its different cultural traditions. 45. These submissions are of little assistance. The task of the Court is to interpret the language adopted by the legislature in accordance with the well-established principles of statutory interpretation. A word such as “corruptly” must take its colour from its context. Case law on other legislation can provide guidance, but cannot determine the issue before us. Non-legal texts, such as The Bible, are of limited, if any, relevance. 46. The appellant drew attention to the different approaches to the meaning of “corruptly” adopted by the trial judge and the Court of Appeal. The former treated the word as an “independent element” of the offence. The latter treated the word as equivalent to “advantage…as an inducement.” As discussed above, in this regard he was correct. His submissions do not, however, suggest how that assists his case. 47. The appellant relied on a number of references which focused on the obligations of candidates with respect to money. He submitted that only candidates were “accountable” in this regard. However, these sections were concerned with the distinct, comprehensive regime for “election donations”. The references have no relevance to that part of section 7 with which we are concerned, with its focus on an offer made to an actual or potential candidate by an inducer. 48. The appellant also invoked an express statement of intent by the Administration as to the object of section 7(1), which appears in the Report of the Bills Committee on Elections (Corrupt and Illegal Conduct) Bill (“Report”)[45] in the following terms: “The object of clause 7(1) is to prevent any persons or organizations from offering an advantage to candidates at an election so that, once elected, they will protect or promote the interests of these persons or organizations.” 49. As he put it in oral submissions, the legislative intent was to confine corrupt conduct to the behaviour of an elected person who would “vote in favour of the sponsor”. And that section 7(1) was directed to “behaviour of elected members, not targeted at the process of elections”. 50. The first thing to note is that this cannot possibly be a complete statement of the “object” of section 7(1). The section applies expressly to offers of advantage to induce a person not to stand as a candidate. A person persuaded not to stand will not be in a position to vote at all. 51. Reading the sentence relied on in its full context of paragraphs 30 and 31 of the Report leads to the conclusion that the use of the definite article in the sentence gives a completely wrong impression. “Corrupt conduct 30. Corrupt conduct refers to conduct which involves intentional acts, and which if allowed would produce a direct effect on an election. Members have sought detailed clarifications from the Administration on whether certain conduct could be caught by the Bill. Specific examples have been quoted for illustration purpose. Corrupt conduct to bribe candidates (clause 7) 31. Members have requested the Administration to explain the rationale for treating an act to induce a person to stand as a candidate at an election as corrupt conduct under clause 7(1). They note that there is no similar provision in the UK Representation of the People Act 1983. According to the Administration, section 8A(1A) of the CIPO was added in 1984 as suggested by an interdepartmental working Group after reviewing the arrangements for the 1982 District Board elections and 1983 Urban Council elections. The Working Group considered that as there were provisions aimed at preventing any person from bribing or intimidating any elector to vote or not to vote for a particular candidate, additional provisions should be made to prevent any person from bribing or intimidating any other person to stand as a candidate or to withdraw his candidate as well. The object of clause 7(1) is to prevent any persons or organizations from offering an advantage to candidates at an election so that, once elected, they will protect or promote the interests of these persons or organizations. The Administration considers that the clause is necessary to maintain clean and honest elections in Hong Kong.” 52. As appears from the first sentence of paragraph 31 of the Report, the Administration was asked a question limited only to “inducing a person to stand”. The sentence relied on, with its use of the definite article, was directed to that specific question. The word “the” in the sentence relied on, properly understood, should be read as “an” or preceded by the word “relevantly”. 53. Furthermore, the full text of paragraph 30 of the Report makes it clear that this answer has nothing to do with the word “corruptly” with which this appeal is concerned. The Administration's answer relied on the reasoning of the Working Group for the previous Corrupt and Illegal Practices Ordinance, which was being superseded, to justify maintenance of the existing prohibition. The word “corruptly” did not appear in that Ordinance. 54. The addition of the word “corruptly” in ECICO was, eventually, accepted by the Administration in response to a different question from members of the Bills Committee as set out at paragraphs 24-26 of the Report. 55. The Court rejects the appellant's submissions that section 7(1) is concerned only with conduct of elected candidates. G. Conclusion and disposal 56. On the Judge’s findings, the appellant’s conduct clearly fell within section 7(1). He had as his objective, diverting votes from targeted candidates and thus sought to manipulate the outcome of the election against them. He induced or sought to induce Ku to procure Chan to stand in part for their own personal gain rather than any proper purpose. Their candidature would not have been honest or open. His offer of an advantage to Ku and Chan plainly had a tendency to undermine a fair, open and honest election. 57. For the foregoing reasons we would dismiss this appeal. Our answer to the question upon which leave to appeal was granted is set out in Section C above. Mr Justice Fok PJ: 58. I agree with the joint judgment of Mr Justice Ribeiro PJ and Mr Justice Spigelman NPJ. Mr Justice Cheung PJ: 59. I agree with the joint judgment of Mr Justice Ribeiro PJ and Mr Justice Spigelman NPJ. Chief Justice Ma: 60. The Court unanimously dismisses this appeal. The appellant, appeared in person Mr Jonathan Man SADPP, Ms Samantha Chiu SPP and Mr Ivan Cheung SPP, of the Department of Justice, for the respondent Mr Derek Chan SC and Ms Vivian Henrietta Ho, amici curiae [1] Cap 554. [2] DCCC 1084/2015 (24 October 2016). References to the Judge’s Reasons for Verdict are to the English translation. [3] ECICO section 7(1)(g): “A person engages in corrupt conduct at an election if the person corruptly solicits or accepts an advantage as an inducement to get, or try to get, another person – (i) to stand, or not to stand, as a candidate at the election…” [4] Reasons for Verdict §§77, 137-142. [5] Reasons for Verdict §26. [6] Reasons for Verdict §77(9). [7] Reasons for Verdict §141. [8] Yeung VP, Poon and Pang JJA [2018] HKCA 45 (8 March 2018). References to the Court of Appeal’s judgment are references to the English translation. [9] Ma CJ, Fok and Cheung PJJ [2019] HKCFA 29 (13 August 2019). [10] (2006) 9 HKCFAR 574 at §63 (cited authorities omitted). This has consistently been followed in numerous authorities. [11] “[Elections] to elect the Chief Executive and persons to membership of the Legislative Council, District Councils and certain other public bodies and to be Rural Representatives...” [12] As indicated in the headings of these sections: section 7: to bribe candidates or prospective candidates; section 11: to bribe electors and others at elections; section 12: to provide others with refreshments and entertainment at election; section 21: to withdraw election petition or election appeal for a bribe. [13] Section 8: to use or threaten to use force or duress against candidates or prospective candidates; section 13: to use or threaten to use force or duress against electors. [14] Section 9: to engage in certain deceptive behaviour in relation to candidates and prospective candidates; section 14: to engage in certain deceptive behaviour in relation to electors; section 15: to impersonate another at election. [15] Section 10: to deface or destroy nomination papers; section 16: [to do certain acts] with respect to voting at elections; section 17: to destroy or deface ballot papers; section 18: to make improper use of election donations; section 20: to lodge false or misleading election return. [16] By, for example, section 8 prohibiting the use or threat of force against candidates or electors; or section 9 prohibiting deceptive behaviour regarding candidates, etc. [17] Section C.1 of this judgment. [18] Reasons for Verdict §§100-101. [19] Reasons for Verdict §112. [20] Reasons for Verdict §125. [21] Notably Cooper v Slade (1858) 6 HL Cas 746; The Bewdley Case (1869) 1 O’M & H 16; The Bradford Case (1869) 1 O’M & H 35; and The Brecon Case (1871) 2 O’M & H 43. [22] Such as The Queen v Robert Eli Low [1961] HKLR 13. [23] Cap 201. [24] Cap 215. [25] Reasons for Verdict §126. [26] Reasons for Verdict §133. [27] Reasons for Verdict §134. [28] Reasons for Verdict §134. [29] Section C of this judgment. [30] Court of Appeal §§110-112, 188. [31] Court of Appeal §102. [32] Court of Appeal §105. [33] Court of Appeal §115. [34] Court of Appeal §126. [35] As Poon CJHC then was. [36] Court of Appeal §§166-174. [37] “... namely to ensure that, as far as practicable, elections are conducted fairly, openly and honestly” Court of Appeal, Poon JA, §174. [38] Court of Appeal §175. [39] Court of Appeal §188. [40] ECICO section 2. [41] Court of Appeal §118. [42] ECICO section 2: “‘candidate’ (a) means a person who stands nominated as a candidate at an election; and (b) also means a person who, at any time before the close of nominations for an election, has publicly declared an intention to stand as a candidate at the election...” [43] Section C.2 of this judgment. [44] (1858) 6 HL Cas 746. [45] LC Paper No. CB(2)1026/99-00. 
Press Summary (English) Press Summary (Chinese) FACV Nos. 3 and 5 of 2019 [2020] HKCFA 32 FACV No. 3 of 2019 IN THE COURT OF FINAL APPEAL OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION FINAL APPEAL NO. 3 OF 2019 (CIVIL) (ON APPEAL FROM CACV NO. 158 OF 2012) FACV No. 5 of 2019 IN THE COURT OF FINAL APPEAL OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION FINAL APPEAL NO. 5 OF 2019 (CIVIL) (ON APPEAL FROM CACV NO. 158 OF 2012) (HEARD TOGETHER) J U D G E M E N T Mr Justice Ribeiro PJ: 1. The present cross-appeals arise out of the same transaction. The first appeal, FACV3 of 2019, is brought by Eton Properties Limited and Eton Properties (Holdings) Limited, both Hong Kong companies. I will refer to them as D1 and D2 respectively and jointly as D1-D2, this being how they were referred to as defendants in the first instance proceedings and throughout. Their appeal is brought against the Plaintiff, Xiamen Xinjingdi Group Co Ltd,[1] a PRC corporation. 2. The second appeal, FACV5 of 2019, is brought by the Plaintiff as appellant against D1 and D2 as well as three others who were also named as defendants at first instance. They are Eton Properties Group Limited, a BVI company (D3), Legend Properties (Xiamen) Company Limited, a Hong Kong company (D4) and Mr Lucio C Tan (D6), an individual who established and controlled the Eton Group. 3. Another entity and five other individuals were previously joined as defendants but are no longer parties. Two of them will, however, be mentioned and can conveniently continue to be referred to as D5 (Legend Properties (Xiamen) Company Limited, a foreign-owned enterprise incorporated in the PRC) and D10 (Cheung Chi Ming, D6’s brother-in-law, a director of D4 and D5 and described as one of the key senior management personnel of the group). 4. These are appeals from the judgment of the Court of Appeal[2] which allowed in part an appeal from DHCJ Stone QC who had entirely dismissed the Plaintiff’s claims.[3] 5. In this judgment, I set out the somewhat lengthy course of events leading to these appeals and deal with the issues arising in FACV3, the first appeal. Lord Sumption NPJ deals with FACV5, the second appeal. I have read his judgment in draft and respectfully agree with his reasons and conclusion. A. The transactions between the parties A.1 The Agreement 6. On 4 July 2003, D1-D2 and the Plaintiff entered into an agreement which lies at the root of these proceedings (“the Agreement”). In it, D1 and D2 declared that each owned one share in D4, comprising the whole of D4’s issued capital and that D4 had incorporated and owned D5 which had the right to develop and use Lot 22, a piece of land in Xiamen comprising an area of 11,994.59 sq m. D1-D2 warranted that they had “absolute control” over D4 and D5. The Plaintiff agreed to purchase those development and use rights by acquiring the shares in D4 and thus D5. Article 3 provided: “[The Plaintiff] for the purpose of obtaining the right to develop [Lot 22] in the name of [D5] and obtaining the right to profits, and at the same time for the purpose of compensating [D1-D2] the investment which it has put into the two companies [D4] and [D5] and [Lot 22], agrees to pay [D1-D2] the transfer price of RMB 120,000,000. After [the Plaintiff] has settled the transfer price in full, [D1-D2] agree to transfer all the shares in [D4] for HK$2 to a legal entity or individual outside the territory of China designated by [the Plaintiff] and exempt repayment of all shareholders’ loans advanced to [D4] and [D5].” 7. The Agreement provided for the Plaintiff’s payment of a deposit of RMB5,000,000 on signing (which was duly paid) and payment of the balance of the transfer price by instalments. Delivery of the land was to take place within 6 months of the date of the Agreement,[4] after current residents had moved out and certain structures had been demolished. The Agreement provided for a significant degree of supervision by D1-D2 in the event that the development process should occur. Their consent would, for instance, be required for construction plans, “project finance index” and “change of land use area”.[5] Pending payment of the transfer price in full, D1-D2 would have the right to “supervise [the Plaintiff’s] development and operation activities”, in respect of a series of specified matters.[6] 8. As provided by Article 8, after full payment, D1-D2 were to transfer their shares in D4 to the Plaintiff’s nominee: “When [the Plaintiff] has settled the total transfer price and fulfilled the obligations according to the Agreement, [D1-D2] will agree to transfer all the shares of [D4] to an approved individual or an approved legal entity outside the territory of China designated by [the Plaintiff], the two parties shall jointly sign the ‘Sale and Purchase Agreement for Transfer of Shares and Debts of [D4] [annexed].” 9. Importantly for present purposes, Article 13 provided for arbitration and choice of law: “1. Parties shall make every effort to negotiate to resolve the disputes arising from the performance of this Agreement. In the case where no agreement is reached through negotiation, any party has the right to submit the dispute to the China International Trade Arbitration Commission [“CIETAC”] in Beijing for arbitration, the arbitration rules of the Arbitration Commission effective at the time when the dispute occurs should be adopted. 2. The Agreement is governed by the laws of the People’s Republic of China, but the procedure and validity related to the transfer of shares of [D4] is governed by the laws of Hong Kong Special Administrative Region.” A.2 Renunciation of the Agreement 10. D6 did not give evidence. However, D10 testified that in September 2003, a mere two months or so after concluding the Agreement, D6 decided to terminate it and instructed D10 to implement his decision.[7] D10 also testified that at around this time, he proposed setting up a real estate investment holding company in the PRC and D6 agreed, leaving it up to D10 to determine when this should be done.[8] 11. On 14 November 2003, D1-D2 wrote to the Plaintiff giving notice that performance of the Agreement would be discontinued, alleging that issues of illegality had arisen and seeking to return the deposit. The Plaintiff replied on 8 December 2003, refusing to accept the termination or return of the deposit. 12. The land should have been delivered to the Plaintiff by 4 January 2004, being six months after the date of the Agreement. That did not occur and, on 17 January 2004, the Plaintiff wrote making a demand for delivery which was not met.[9] 13. Instead, the Defendants took steps to develop Lot 22 themselves, obtaining regulatory and town planning approvals between November 2004 and February 2005. In March 2005, D10 set about incorporating a real estate holding company as discussed with D6 back in September 2003. This led to D3 being incorporated and introduced into the scene. A.3 The restructuring 14. On 8 August 2005, the Plaintiff commenced the CIETAC arbitration in Beijing against D1 and D2.[10] And on 24 October 2005, the tribunal served notice that the first hearing would take place on 23 November 2005. 15. On 16 November 2005, a week before that hearing, the restructuring of the corporate shareholdings relating to Lot 22 began. D3, Eton Properties Group Limited, a BVI company, was brought into the picture and on that date, 9,998 newly-issued shares in D4 were allotted to D3. The restructuring was completed when, on 6 April 2006, D1 transferred its single share in D4 to D3 and D2 declared itself trustee of its single share in D4 in favour of D3. Consequently, D4’s 10,000 issued shares were held beneficially by D3 (in place of D1-D2) so that D3, via D4, held D5 which owned the rights to Lot 22. It followed that the promise contained in the Agreement that D1-D2 would, at completion of the development, transfer their shares in D4 to the Plaintiff could not be performed. The Plaintiff was unaware of the restructuring which was not disclosed by D1-D2 until 2 January 2008. B. The arbitration – the First Award 16. The first arbitration hearing took place as scheduled on 23 November 2005 and a second hearing was held on 18 May 2006. Even though by the start of the arbitration, D1-D2’s interest in D4 had been diluted to 0.02% after the allotment of shares to D3, they referred to themselves as the parent companies of D4 and D5.[11] On 30 June 2006, D5 started pre-sales of the units being developed on the site. 17. On 27 October 2006, the tribunal made its First Award.[12] It noted the terms of the Agreement including D1-D2’s warranty that they had absolute control over D4 and D5; the Plaintiff’s agreement to pay RMB120,000,000 to obtain the right to develop Lot 22 in D5’s name; the undertaking to deliver Lot 22 to the Plaintiff within 6 months from the date of the agreement; the obligation on D1-D2, after full payment to transfer all the shares in D4 to the Plaintiff for HK$2; and the Plaintiff’s freedom to decide on the details of the development while D1-D2 maintained certain rights of supervision pending settlement of the full transfer price. 18. The tribunal recorded that D1-D2 had repudiated the Agreement and that the Plaintiff refused to accept termination. It noted that D1-D2 asserted that the Agreement was illegal and invalid and claimed that even if it was valid, it had become impossible to perform. This was because construction activities on the ground and sales of the residential units meant that they were unable to transfer Lot 22 to the Plaintiff for development. 19. The tribunal held that PRC law was applicable. It described the Agreement as “neither an agreement of share transfer nor an agreement of transfer of land-use right”. It held, rather, that – “The subject matter of the Agreement is the contractual right to buy and sell the shares in [D4] that has an indirect effective control over [Lot 22]. The amount of RMB 120 million is the consideration for [the Plaintiff] to obtain the contractual right to acquire all the shares in [D4] for HK$2, and also the consideration for [D1-D2] to obtain the contractual right to sell all the shares in [D4] for HK$2.” 20. So characterised, the tribunal concluded that D1-D2’s challenge to the legality and validity of the Agreement failed. It also rejected the argument that performance was impossible: “Even though any change in circumstances makes it difficult to perform the agreement during its performance, the parties shall exert reasonable efforts in good faith to perform the Agreement completely and fully other than purely emphasize external causes. In this case, as stated by [D1-D2], the Agreement is a framework agreement, whose performance may be difficult due to various uncertainties. This needs close cooperation between the parties and reasonable efforts to seek alternative approaches to meet the purpose of the Agreement. [D1-D2’s] allegations cannot constitute justifiable reasons for impossibility to perform the Agreement and discontinuing performance of the Agreement without the consent of [the Plaintiff].” 21. The tribunal decided that the Plaintiff’s claim for continued performance of the Agreement should be supported and dismissed D1-D2’s counterclaim seeking confirmation that the Agreement was invalid. The tribunal awarded the Plaintiff damages for breach of contract of RMB 1,275,000 (representing damages for delay in delivering the land) and ordered that “[D1-D2] shall continue to perform the Agreement…” C. Attempts at enforcement of the First Award 22. The Plaintiff applied to enforce the First Award in Xiamen but, on 30 July 2007, the Xiamen Municipal Intermediate Court dismissed the application principally because D1-D2 were Hong Kong companies whose assets were outside the jurisdiction.[13] 23. The Plaintiff then applied ex parte to the Hong Kong Court (in Action HCCT 54/2007) for leave to enforce the award, invoking the statutory procedure provided by section 2GG of the Arbitration Ordinance (“the Ordinance”).[14] Adopting such procedure, an arbitral award is enforceable with the court’s leave in the same way as a judgment of the court and, if leave is given, the court may enter judgment in terms of the award. By order dated 31 October 2007, Mr Justice A Cheung[15] granted leave to enforce the award and entered judgment in its terms, ordering that D1-D2 “shall continue to perform the [Agreement]…” 24. On 2 January 2008, D1-D2 applied to set aside the aforesaid order and judgment. As the trial Judge noted,[16] it was only in an affirmation filed on that date in support of D1-D2’s application that it was first disclosed that the restructuring had taken place. While the Plaintiff had been aware of D5’s construction activities and sales of units in the development, this was the first that the Plaintiff had learned that D1-D2 had divested themselves of their shareholding in D4 and of D3’s acquisition of that company.[17] 25. With that realisation, on 27 May 2008, the Plaintiff started a fresh action (initially listed as HCA 961/2008 and re-listed as HCCL 13/2011 – “the Action”) from which the present proceedings flow. It was a common law action to enforce the First Award. Additionally, it was brought also against D3, D4 and D5 who were not parties to the Agreement or the arbitration. The Plaintiff sought declarations that D3 and D2 each held the D4 shares registered in their names on constructive trust for the Plaintiff and obtained interim injunctions restraining the defendants from further changing the shareholding structure of D4 and restraining D5 from disposing of or diminishing the value of its assets. 26. Meanwhile, the application by D1-D2 to set aside the order granting leave to enforce the award and the statutory judgment came on for hearing before Reyes J.[18] The applicants’ main argument was that the outstanding non-monetary obligations imposed on them by the First Award could no longer be performed so that it was contrary to public policy to enforce that part of the award.[19] They contended that performance had become impossible because some 99% of the residential units built by D5 on the Lot had been sold and consequently, that neither D1-D2 nor D5 could deliver possession of the land to the Plaintiff. Furthermore, D1-D2 could no longer transfer the D4 shares to the Plaintiff since the structure of the shareholding had changed and they no longer held those shares, all of D4’s 10,000 shares now being held by D3, save one held by D2 on trust for D3.[20] 27. Reyes J rejected the impossibility argument and dismissed the application. He was not persuaded that the Agreement was substantially incapable of performance.[21] Although the viability and merits of the Action were not issues before him, his Lordship took notice of the Action and considered it arguable that execution of the Agreement and payment of the deposit may have given rise to an equitable interest in the shares in favour of the Plaintiff, resulting in a constructive trust being imposed on the D4 shares in the hands of D3 and D2.[22] He noted the Plaintiff’s willingness to pay the RMB120,000,000 and the HK$2 nominal fee and also to bear the properly audited construction costs of the development in seeing through enforcement of the award.[23] 28. Dissatisfied with Reyes J’s decision, D1-D2 lodged an appeal to the Court of Appeal where once again, their main argument was that the award had become impossible to perform so that enforcement was contrary to public policy. Le Pichon JA, writing for the Court,[24] held that the impossibility argument was misguided since there was no insuperable impediment to the transfer of the shares registered in D3’s name to the Plaintiff and since the “impossibility (if any) [was] self-inflicted”. D. The Second Award 29. Having failed before Reyes J, D1-D2 returned to the tribunal, applying for a determination that the Agreement could no longer be performed and seeking a ruling that the parties be discharged therefrom. The hearing was held on 21 January 2009 and the tribunal gave its Second Award on 22 April 2009.[25] 30. D1-D2 argued that the Agreement had lawfully terminated in accordance with its terms because there had been a substantial change involving completion of the development; sale of more than 90% of the units; and a change in the shareholding of D4, making the transfer of the shareholding impossible as a matter of fact. This was rejected by the tribunal which held that the change of shareholding was a breach of the Agreement and not a ground for lawful termination. The unilateral notice of termination was invalid and the tribunal was not satisfied that the objectives of the Agreement were incapable of being met. It reiterated that it had made an order for continued performance. The Second Award therefore dismissed D1-D2’s application. 31. D1-D2 made one further approach to the tribunal. It will be recalled that in the First Award, the tribunal had pointed out that the Agreement “... needs close cooperation between the parties and reasonable efforts to seek alternative approaches to meet the purpose of the Agreement”. Seizing on this, by letter dated 19 June 2009, D1-D2 wrote to the tribunal seeking a ruling on those “alternative approaches”. They returned to their familiar theme, saying that directions from the tribunal as to how to proceed were needed because “… [the] objective circumstances of this case have changed significantly, the two contracting parties are unable to continue to perform the Agreement according to the original provisions”. 32. The tribunal gave this application short shrift, stating: “The arbitral tribunal adjudicated the case fully... After investigation, the [First Award] is final, there was no mistake in relation to writing, typing or calculation which needs to be corrected. Also the arbitral tribunal in the Award did not leave out any matter which would need to be further determined according to the PRC Arbitration Law and the Arbitration Rules of the Arbitration Commission. Accordingly, the arbitral tribunal does not accept the abovementioned application of [D1-D2]”. E. Interlocutory rulings in the present proceedings 33. The focus of the litigation switched to the Action. Fok J (as Fok PJ then was) noted in his judgment dated 16 March 2010[26] that the Plaintiff asserted proprietary rights over the D4 shares in the hands of D3. Those rights were said to have arisen out of alleged breaches of fiduciary duty by D1-D2 committed in effecting the restructuring, with D3 the recipient of the shares knowing that they were the product of those breaches and thus subject to a constructive trust in the Plaintiff’s favour. A series of interlocutory applications were placed before his Lordship. 34. He allowed the Plaintiff’s application for leave to amend the Writ and Statement of Claim, thereby expanding the Action to add claims in tort for inducing breach of contract and conspiracy against six additional individuals named as D6 to D11, who were associated with D1-D2 and the defendant companies. He also permitted the Plaintiff to add a plea of fraud in connection with the allotment of D4 shares to D3. 35. Fok J dismissed the Defendants’ applications to strike out the Action, holding that it was not plain and obvious that the constructive trust claim was bound to fail. However, he granted the application of D1-D2 to stay the constructive trust and conspiracy claims to arbitration holding that they were disputes that fell within the arbitration clause in the Agreement.[27] He dismissed the application for a stay in the Court’s inherent jurisdiction mounted by D3-D5.[28] 36. The Plaintiff’s appeal against Fok J’s order staying the Action against D1-D2 to arbitration succeeded in the Court of Appeal (“the Stay Appeal”).[29] Rogers VP considered the effect of the arbitration awards and held that the time and scope for arbitration had clearly passed and that it was now for the Hong Kong courts to enforce the award.[30] 37. Kwan JA elaborated upon this, holding that “the claim against the 1st and 2nd defendants in this action is primarily an action on an arbitration award, not an action on the performance and obligations of the Agreement”.[31] As her Ladyship explained: “The plaintiff has brought this action against the 1st and 2nd defendants on the basis that these defendants were in breach of their implied promise to perform a valid arbitration award. This is within the enforcement jurisdiction and powers of the court ...”[32] 38. Kwan JA did not consider the statutory judgment obtained by the Plaintiff under sections 2GG and 40B of the Ordinance an obstacle, holding that the Plaintiff “is at liberty to bring an action on the award against the 1st and 2nd defendants, combined with other claims against other entities and individuals.”[33] She also held that an action on the award did not need to be “mechanistic in nature”, the remedies available depending on the scope of the issues raised.[34] 39. The applications of D1-D2 and D3, D4 and D5 respectively for leave to appeal to this Court against the Court of Appeal’s decision in the Stay Appeal were dismissed by the Appeal Committee,[35] Ma CJ holding that, so far as D1-D2 were concerned, the Action was brought by way of enforcement of the award and that the arbitrators had treated the arbitration agreement as spent.[36] Since the application of D3, D4 and D5 for leave was contingent on the stay of D1-D2’s proceedings being upheld, it was also dismissed.[37] 40. Having staved off the Defendants’ attempts to strike out or stay the Action, the Plaintiff sought to expand its scope further. On 21 December 2011, it obtained leave from Reyes J to re-amend its pleadings to add an alternative claim (in paragraph 35(5)) for damages or equitable compensation in the event that the relief it had been pressing for, namely, transfer of the D4 shares and through them, acquisition of the benefits earned by D4, should prove unachievable. Reyes J also gave directions for expert evidence regarding the valuation of D4 to be adduced in connection with that plea. 41. On the Defendants’ appeal against Reyes J’s orders, the Court of Appeal[38] noted that the loss and damage pleaded involved “the difference between (i) the existing value of the entire shareholding in [D4] and (ii) the sum of RMB120,000,000 which the Plaintiff agreed to pay under the agreement for acquisition of the same”, the Plaintiff alleging that “each of [D1 to D11] is jointly and severally liable whether by means of an award of equitable compensation or by means of an award of damages” in respect of such loss and damage. 42. The Court of Appeal held that those amendments were properly allowed by Reyes J, as they raised arguable issues as to whether such relief was available. However, because of the lateness of the amendments, they ordered a split trial with the issues of quantum raised by paragraph 35(5) to be dealt with, in so far as necessary, after the outcome of the trial was known. F. The judgment of DHCJ Stone QC[39] 43. The Action finally came on for trial before Deputy High Court Judge William Stone QC on 8 March 2012. The hearing lasted 22 days and, on 14 June 2012, his Lordship delivered a meticulous and comprehensive judgment running to 135 pages. 44. The Judge dealt first with what he called the “derivative claim” whereby the Plaintiff sought to enforce the Award against D1-D2. Secondly, he addressed what he called the “original action” involving causes of action and equitable claims in the expanded Action against various combinations of the Defendants based on the torts of conspiracy, inducing breach of contract and constructive trust.[40] F.1 The enforcement claim against D1-D2 45. In approaching the enforcement claim, his Lordship was troubled by the existence, alongside the Action, of the statutory judgment entered pursuant to leave granted to enforce the Award on 31 October 2007.[41] He described the Award and judgment as languishing in “procedural limbo” and as the “elephant in the courtroom”.[42] 46. He was puzzled at the Plaintiff’s persistence in pursuing continued performance of the Agreement when, as he saw it, after its breach, which had occurred when it was “torn up” some 21 months before the arbitration was commenced,[43] such performance had become impossible. The Agreement: “... had been rendered physically incapable of continued performance by reason of the plain fact that site at Lot 22 by then had been fully developed by the 5th defendant – which, as earlier noted, had been built in a configuration strikingly different from the development plan as earlier proposed by the plaintiff – and in turn the vast majority of the units as erected had been on-sold to third party purchasers.”[44] 47. It may be observed that his Lordship’s view as to impossibility ran counter to the finding of the tribunal in the First and Second Awards. It was also at odds with the views of Reyes J and the Court of Appeal in their judgments refusing to set aside the statutory judgment on the basis of impossibility and public policy, even after the restructuring had been disclosed. Le Pichon JA could see “[no] insuperable impediment to the transfer of the shares registered in D3’s name to [the Plaintiff]”.[45] 48. While DHCJ Stone recognised that the Action was brought to enforce the implied promise to honour the Award at common law, he regarded this as being “no different in specie” from, and permitting no greater latitude in terms of remedy than,the statutory procedure. He held that on either approach, the Court is limited to “mechanistically” converting the award into a judgment in terms of the award.[46] 49. In bringing the Action, his Lordship held that the Plaintiff was seeking to go far beyond such mechanistic enforcement. It sought: “... to substitute for the ‘continuous performance’ award of the CIETAC tribunal a new claim for damages/equitable compensation, which not only goes beyond the scope of the award as it now exists, but never was contemplated by the tribunal, which was not asked even to consider/countenance such a remedy.”[47] 50. This, he held, was an attempt to “subvert the integrity of the arbitral process” and to get the Hong Kong Court “to usurp the designated dispute resolution body and impose its own remedy under the purported guise of ‘enforcement’” by asking it “to rule upon an issue relating to breach of the Agreement which must lie solely within the jurisdiction of the Beijing arbitral tribunal (or, if ultimately became necessary, in the Beijing court of supervisory jurisdiction)”.[48] The enforcement claim against D1-D2 was accordingly held to fall outside the Court’s jurisdiction.[49] 51. The Judge went on to hold that even if, contrary to his conclusion, one were to assume that jurisdiction exists, the enforcement claim would still have to be dismissed since the proper law of the Agreement is PRC law which has no concept of the constructive trust.[50] Moreover, even if one were to assume that Hong Kong law applies, the Agreement was not specifically enforceable because it was a contract requiring constant supervision and conditional in nature. Thus, no equitable interest in the D4 shares passed to the Plaintiff on execution of the Agreement, so that no basis existed for any constructive trust to be imposed on those shares in the hands of D1-D2 and D3.[51] 52. On the assumption that Hong Kong law applied and that the Award could be treated as if it were an order for specific performance, the Judge regarded the principle in Johnson v Agnew[52] as a further obstacle to the Plaintiff’s claim against D1-D2. He accepted the argument that since the Award and statutory judgment were obtained on the footing that the contract remained alive, a claim for damages or equitable compensation was excluded by that principle unless the Plaintiff first applied for discharge of the order for continued performance so as to terminate the Agreement.[53] 53. The enforcement action was therefore held to fail.[54] The Judge observed that it was up to the Plaintiff to return to the tribunal and, if necessary, to the Beijing supervisory court, to try to obtain substitute orders providing for the relief it sought in the Action.[55] F.2 The other claims asserted in the Action 54. The Judge proceeded also to dismiss the Plaintiff’s claims against all the other Defendants. 55. As to the claims based on constructive trust or fiduciary duty (against D3-D6), his Lordship held that those concepts were unknown to PRC law which he found to be applicable.[56] On the alternative basis that the issue was governed by Hong Kong law, he held that the Agreement was not specifically enforceable, requiring constant supervision and being conditional in character, so that no equitable interest passed upon its execution and no constructive trust arose.[57] 56. It was an important premise of the Judge’s approach to the Plaintiff’s economic tort claims that the only breach of contract committed by D1-D2 was their renunciation of the Agreement in November 2003. He rejected the suggestion that the restructuring which took place two years later entailed any breach or other unlawfulness: “... I am unwilling to characterize such restructuring as ‘unlawful’ or ‘illegal’ or ‘fraudulent’. It certainly was not a contractual breach, which breach itself had taken place some two years earlier with the tearing up of the Agreement by the 1st and 2nd defendants.”[58] 57. One might pause to observe that this was at odds with the tribunal’s ruling in the Second Award that the changes in shareholdings were a breach of the Agreement, referred to above. 58. Nevertheless, in the context of inducing breach of contract, the Judge held that no overt acts of inducement and no causal link to the November 2003 breach had been proved against D3-D5 or any of the individual Defendants (D6-D11).[59] 59. Similarly, the Judge’s rejection of the restructuring as involving any unlawfulness led to his decision that neither the alleged conspiracy by unlawful means nor the alleged conspiracy to injure were established for want of any unlawful means and want of a predominant motive to injure.[60] 60. His Lordship also held that since the overt acts relied on had occurred on the Mainland and, since the economic torts in question were not known to PRC law, the Plaintiff failed to meet the double actionability requirement and the causes of action in tort were not actionable in Hong Kong.[61] 61. The Judge further accepted that the principle in Said v Butt[62] excluded the individual Defendants, acting as directors and company officials, from being held personally liable for the tort of procuring the relevant company’s breach of the Agreement.[63] 62. And quite apart from the aforesaid legal arguments held to have barred the Plaintiff’s claims, the Judge also found fatal evidential weaknesses in the respective conspiracy and inducement cases, especially regarding the elements of intention and purpose.[64] 63. Accordingly, DHCJ Stone dismissed the Action in its entirety. G. Judgment of the Court of Appeal[65] G.1 The common law action on the award 64. Giving the judgment of the Court of Appeal, Yuen JA allowed the Plaintiff’s appeal against DHCJ Stone’s dismissal of the common law action on the award against D1-D2. 65. Her Ladyship noted that an arbitration award might be enforced by invoking the statutory procedure (under section 2GG of the Ordinance) or by bringing a common law action and proceeded to examine the juridical nature of such an action. 66. Yuen JA pointed out that an action on an award could be brought at common law where there had been a valid submission of a dispute to arbitration; an award had been made in favour of the plaintiff and the defendant had failed to honour that award.[66] Hong Kong law (following English law) implied a mutual promise to honour the award. Failure to do so constituted a fresh cause of action separate and independent from an action based on breach of the underlying contract (which contained the arbitration clause).[67] 67. In reaching her conclusion, Yuen JA had to deal with three issues which are presently relevant. First, D1-D2 sought to argue that in a common law action, the court has no jurisdiction to order damages where the arbitrators had not made any award of damages. Her Ladyship did not agree, holding that, unlike in cases involving the statutory procedure, the Court was not limited in a common law action to granting relief which simply mirrors the terms of the award.[68] Accordingly, the Court had jurisdiction to order damages or equitable compensation even though the tribunal had only made an award for “continued performance”. 68. Secondly, D1-D2 contended that the Action was an invalid attempt at circumventing the arbitral process so that the Hong Kong Court ought to refuse relief and stay the Action to arbitration. Yuen JA left open the question whether an issue estoppel, arising out of the decision of the Court of Appeal in the Stay Appeal,[69] prevented D1-D2 taking this point. She held that even if an issue estoppel did not arise, the present action operated at the enforcement phase and was distinct from any dispute falling within the arbitration clause so that no stay was called for.[70] 69. Finally, D1-D2 sought to argue that, by invoking the statutory procedure and entering judgment on the award, the Plaintiff had elected for a remedy involving “continued performance” of the Agreement which was premised on the continued existence of that contract, a position inconsistent with the present claim for damages or equitable compensation which had to be premised on the Agreement having come to an end.[71] 70. Yuen JA held that in entering judgment under the statutory procedure, the Plaintiff had not made an informed irrevocable election since it had done so while material information concerning the restructuring had been withheld by D1-D2.[72] However, her Ladyship decided that the inconsistent remedies of ordering continued performance of the Agreement on the one hand, and damages or equitable compensation on the other, could not stand together. She therefore gave directions that the Plaintiff, when in a position to make an informed choice, was required to make an election which would determine what orders the Court of Appeal might make as to the further conduct of these proceedings. G.2 The other claims in the Action 71. The Court of Appeal dismissed all the other claims. Her Ladyship held that whether breach of the Agreement was constituted by its renunciation or by the restructuring, the Plaintiff had failed to prove that any of D3-D5 or D6 had caused such breach so as to constitute the tort of inducing breach of contract.[73] Moreover, since such a tort was not known to PRC law, the claim fell foul of the double actionability requirement.[74] 72. Similarly, the conspiracy claim involved alleged overt acts occurring on the Mainland where, under PRC law, tortious liability for conspiracy was unknown so that by reason of the double actionability rule, the tort was not actionable in Hong Kong.[75] Furthermore, the Judge had found that there was no common intention to injure by the restructuring, which was fatal to the claim.[76] 73. Yuen JA held that the constructive trust claim failed because, there being no concept of constructive trust in PRC law, execution of the Agreement gave rise to no equitable interest in the Plaintiff’s favour and no constructive trust. She held that even if Hong Kong law were applicable, the claim would fail since the Agreement, requiring constant supervision and mutual cooperation at various stages, was “only a framework agreement” and not susceptible to specific performance and thus provided no basis for imposing a constructive trust on D1-D2 or D3-D5. 74. Having allowed the Plaintiff’s appeal in relation to the common law action on the award, but having held that an informed election had to be made between an order for “continued performance” and an order awarding damages (or equitable compensation), the Court of Appeal directed that the Plaintiff should make up its mind and then propose a suitable order after consulting D1-D2 or restore the matter before the Court of Appeal.[77] 75. D1-D2 issued a motion applying for leave to appeal to this Court. The Plaintiff notified the Court of its election and, on 20 October 2017, the Court of Appeal ordered that the statutory judgment[78] be set aside and that judgment be entered in favour of the Plaintiff for damages against D1-D2 for breach of the implied promise to honour the Award under HCCL 13/2011. Assessment was to await the quantum stage of the split trial. H. Application for leave to appeal to this Court 76. Having unsuccessfully sought leave to appeal from the Court of Appeal,[79] leave to bring the present cross-appeals was granted by the Appeal Committee[80] as follows. 77. D1-D2 were granted leave to appeal[81] confined to the following questions of law, namely: “Whether the jurisdiction of the Hong Kong court as an enforcement court under a common law action to enforce a Mainland arbitral award is circumscribed or otherwise affected by the statutory regime for enforcement of international awards covered by the New York Convention and Mainland awards; and if the answer is ‘yes’, what should the approach of the court be? Whether the rule in Johnson v Agnew [1980] AC 367 and/or legal principle prevents an enforcing court from awarding damages on the basis that the underlying contract was no longer subsisting when there was an extant award for continued performance of the underlying contract?” 78. As for the cross-appeal, the plaintiff was granted leave to appeal[82] in respect of the following questions of law, namely: “Under the tort of inducing a breach of contract, where the inducement takes the form of the conclusion of a dealing inconsistent with an existing contract, what is the requirement of causation as regards the breach of contract? In particular, where the contracting party is determined to breach the contract with or without the inducement of the defendant, what is the requirement of causation? Where the obligation owed by a defaulting vendor to a purchaser to continue to perform an agreement for the sale and purchase of shares in a private company is governed by a foreign law, is the Hong Kong court entitled to grant the remedy of a constructive trust pursuant to its equitable jurisdiction over such obligation notwithstanding that the foreign law which governs the agreement does not recognise the concept of constructive trust? If so, what is the correct approach of the Hong Kong court in granting such remedy? Where an agreement for the sale and purchase of shares in a Hong Kong private company is by agreement governed by foreign law, does that preclude the law of property and the law of trust under Hong Kong law from operating to transfer the equitable title or interest in the shares to the purchaser upon the making of the agreement or upon the agreement becoming specifically enforceable? 79. Leave to appeal was also granted on the “or otherwise” basis in connection with the following question: “In circumstances where an agreement for the sale and purchase of the entire issued shares in a private company which ultimately holds land or development rights in land situated outside Hong Kong contains provisions to the effect that: (a) pending completion of the sale, possession of the land be given to the purchaser, with liberty to, but no obligation upon, the purchaser to develop or sub-sell interests in the land in the meantime; and (b) in the event that prior to completion of the sale, the purchaser does take possession of the land and does carry out development of the land, express powers be reserved to the vendors to regulate the development activities, but the purchaser does not opt to carry out any development or sub-sale activities prior to completion, is the agreement one which is amenable to specific performance?” I. The appeal brought by D1-D2 80. D1-D2’s basic contention on the present appeal is that the Court of Appeal were wrong because, in awarding the Plaintiff damages to be assessed, they went beyond the relief that can properly be awarded in a common law action on the award. In support, they advance three main arguments which may be referred to as (1) the “mechanistic argument”; (2) the “outflanking argument”; and (3) the “extant award argument”. At the hearing, the submissions advanced by Mr Steven Gee QC[83] predominantly involved variants of the outflanking argument. J. The mechanistic argument 81. D1-D2 argue that the Court of Appeal acted beyond its jurisdiction in awarding the Plaintiff damages because its power to enforce the arbitration award, whether pursuant to the statutory summary procedure or to a common law action on the award, is subject to the same restriction: A court may only enforce an award “mechanistically” by making an order which mirrors and does not go beyond the award.[84] In the present case, the First Award stipulates that D1-D2 “shall continue to perform the [Agreement]…” It says nothing about damages so, the argument runs, awarding damages against D1-D2 exceeded the Court’s powers. It will be recalled that this is a proposition that found favour with DHCJ Stone[85] but not with Yuen JA and the Court of Appeal.[86] 82. This argument is advanced as one of statutory construction. Having referred to the statutory procedure, D1-D2 submit that: “... as a matter of proper statutory interpretation and having regard to the public interest in giving effect to the parties’ choice of arbitration for dispute resolution, the same mechanistic approach must apply to a common law action on an award.”[87] It is therefore necessary to examine the applicable provisions. J.1 The statutory procedure 83. The scheme for the enforcement of Mainland awards in the HKSAR, including CIETAC awards, is the result of the Arrangement concerning Mutual Enforcement of Arbitral Awards between the Mainland and the HKSAR[88] entered into pursuant to Article 95 of the Basic Law[89] (“the Arrangement”). 84. The Arrangement provides that a party may enforce an award with which the other party has failed to comply.[90] It specifies that the enforcing court “shall handle the application and enforce the award according to the legal procedure of the place of enforcement”.[91] It also provides that the limitation period is governed by the law of the place of enforcement.[92] 85. The Arrangement was implemented by amendments to the Ordinance[93] with section 40B providing for two alternative methods of enforcement, namely, either by a court action or by the summary procedure. It states: “(1) A Mainland award shall, subject to this Part, be enforceable in Hong Kong either by action in the Court or in the same manner as the award of an arbitrator is enforceable by virtue of section 2GG. (2) Any Mainland award which is enforceable under this Part shall be treated as binding for all purposes on the persons between whom it was made, and may accordingly be relied on by any of those persons by way of defence, set off or otherwise in any legal proceedings in Hong Kong, and any references in this Part to enforcing a Mainland award shall be construed as including references to relying on a Mainland award”.[94] 86. Section 2GG referred to in section 40B(1) provides for conversion of the award into a judgment of the court so that it becomes enforceable “in the same way as a judgment”: “(1) An award, order or direction made or given in or in relation to arbitration proceedings by an arbitral tribunal is enforceable in the same way as a judgment, order or direction of the Court that has the same effect, but only with the leave of the Court or a judge of the Court. If that leave is given, the Court or judge may enter judgment in terms of the award, order or direction. (2) Notwithstanding anything in this Ordinance, this section applies to an award, order and direction made or given whether in or outside Hong Kong.” 87. As the Rules of the High Court provide, the party seeking to invoke section 2GG may apply ex parte for leave to enforce the award,[95] supporting the application with an affidavit which exhibits the arbitration agreement and the original award, states the addresses of the parties and deposes that the award has not been complied with.[96] Upon leave being granted, judgment is entered “in terms of the award”. After being served with the order granting leave, the other party may apply to set the order aside, but only on limited grounds which bear directly on the validity or enforceability of the arbitration agreement or the award itself. Grounds for refusing to enforce an award include incapacity of a party to the arbitration agreement; invalidity of the arbitral agreement under its proper law; lack of notice to the other party of the hearing; the award or the composition of the tribunal falling outside the scope of the arbitration agreement; the award not being binding; and the existence of public policy grounds for refusing enforcement.[97] 88. The section 2GG procedure is therefore summary in nature. The object of the exercise is for the enforcing court to endow the award with the status of a compulsorily executable judgment without itself scrutinising the merits of the arbitrators’ award and only entertaining challenges within the limits laid down in the Ordinance. 89. It is therefore unsurprising that the enforcing court is enjoined by section 2GG to confine itself to entering judgment “in terms of the award”. Thus, for example, where the award did not grant the successful claimant post-award interest, it was held that under the summary procedure, the enforcing court could not enter judgment including such a sum of interest.[98] It is in that context that the court’s task has been said to be “as mechanistic as possible”.[99] As Cheung CJHC (as Cheung PJ then was) put it, the “plain intent behind the relevant provisions [is] to make awards to which they apply enforceable with ease, subject to the narrowly confined exceptions, almost ‘as a matter of administrative procedure’”.[100] J.2 Is the common law action on the award similarly restricted? 90. The alternative modes of enforcing awards provided for by section 40B(1) of the Ordinance – by court action or under section 2GG – are quite separate and procedurally very different. Unlike under section 2GG, an action on the award does not involve the plaintiff obtaining a judgment after an ex parte application. The plaintiff must sue on the award and prove his case. As MacKinnon J explained in Norske Atlas Insurance Co Ltd v London General Insurance Co Ltd:[101] “In order to sue on an award, it is, I think, necessary for the plaintiffs to prove, first that there was a submission; secondly, that the arbitration was conducted in pursuance of the submission; and, thirdly, that the award is a valid award, made pursuant to the provisions of the submission, and valid according to the lex fori of the place where the arbitration was carried out and where the award was made.” 91. The rationale for restricting the court to entering judgment mechanistically therefore does not apply to the common law action. There is no reason for thinking that the statutory intent is that remedies which may be granted in a common law action on the award must be circumscribed by requirements designed for judgments entered under the summary procedure. 92. The object of the Ordinance is, as section 2AA states, “to facilitate the fair and speedy resolution of disputes by arbitration without unnecessary expense”. The intent is to facilitate and not to place obstacles in the way of the arbitral process, including the process of enforcing arbitral awards. 93. Indeed, the purpose of the summary procedure itself is to facilitate enforcement – “ensuring the effective and speedy enforcement” of arbitration awards.[102] It is incongruous to construe a provision forming part of such a facilitative scheme as having a constricting effect on an alternative mode of enforcement which is quite separate from, and not regulated by, a summary procedure. 94. While no authorities supporting the mechanistic argument have been put forward, various decisions can be found which demonstrate that, in a common law action on the award, the Court is not constrained by the requirement that the judgment must be “in terms of the award”. 95. A clear illustration is provided by the Dalmia judgments of Kerr J, upheld by the English Court of Appeal.[103] The plaintiff Dalmia, an Indian company, obtained arbitration awards which it sought to enforce against the defendant Bank in London, invoking the statutory summary procedure.[104] The awards directed the Bank to pay the amounts to the plaintiff “in India”. In his first judgment, Kerr J posed the central question as follows: “... whether [the Act’s] summary procedure for enforcement is available in relation to awards which specifically require payment in a foreign country, or whether the plaintiff's only remedy in these circumstances is to bring actions on the awards because the bank has failed to pay in India.”[105] 96. He held that the summary procedure was not available. Entering judgment in terms of the award requiring payment in India would not be a judgment that an English court would issue so that it would not be a judgment enforceable in the same manner as a judgment of the High Court. However, Kerr J held that in these circumstances, “[the] appropriate remedy of the plaintiff ... would be an action for the payment of damages here on account of the defendant’s breach in not making payment abroad”.[106] That would obviously involve bringing an action seeking relief going beyond the terms of the award. That is precisely what Dalmia then did. It brought an action for damages against the Bank, “present and served within the jurisdiction of [the English] Court, for non-compliance with the awards.”[107] In contrast with the summary procedure, those proceedings involved a trial which lasted over two months[108] where various arguments attacking the validity of the award were advanced. Dalmia’s action succeeded and the judgment was upheld in the Court of Appeal.[109] 97. For the aforesaid reasons, I would reject the mechanistic argument. K. The outflanking argument K.1 Action on implied promise caught by arbitration clause 98. Mr Gee’s outflanking argument has two versions. The first is an argument in favour of a stay. The contention is that, instead of awarding damages against D1-D2, the Court of Appeal ought to have stayed the Action to arbitration because the claim is in truth a claim to enforce the matrix agreement and is caught by the arbitration clause contained in Article 13 of the Agreement. Two reasons are advanced for this. First, it is said that suing for breach of an implied promise to perform the award is an action based on and arising out of the Agreement and so is caught by the arbitration clause. Secondly, D1-D2 argue that the damages claimed in the Action are actually damages for loss of the profits of the development flowing from breach of the Agreement and so that the claim comes within Article 13. To award the Plaintiff such damages would be to allow it to outflank the arbitral regime agreed to by the parties: “... any argument that P’s claim for damages falls outside the ambit of the Agreement overlooks the principle that an essential element in a common law action on the award is the implied promise to perform the award. That promise is contractual and the cause of action is based on and arises from nothing but the contract ... The fact that P chooses to advance its claim as damages in an ‘action on the award’ concerns the cause of action, and does not prevent the damages claim from being within the arbitration agreement ... The damages awarded by the CA are closely connected with and overlap what would be the damages for breach of the Agreement for its non-performance or repudiation. Payment of one would be payment on account of the other. Both are subject to the arbitration agreement.”[110] K.1a The arbitration clause 99. I shall deal first with the proposition that the Action falls within the arbitration clause because it is based on and arises out of the Agreement. This requires consideration of the scope of Article 13 which relevantly provides: “Parties shall make every effort to negotiate to resolve the disputes arising from the performance of this Agreement. In the case where no agreement is reached through negotiation, any party has the right to submit the dispute to [CIETAC] in Beijing for arbitration, the arbitration rules of the Arbitration Commission effective at the time when the dispute occurs should be adopted.” 100. What the parties have thus agreed to refer to arbitration are “disputes arising from the performance of the Agreement”. Is an action to enforce the implied promise to honour the Award a “dispute arising from the performance of the Agreement”? 101. The language of Article 13 suggests that the answer is “No”. The disputes referred to the tribunal arose from D1-D2’s renunciation of the Agreement on 14 November 2003, their failure to deliver the land to the Plaintiff by 4 January 2004 and their failure to transfer their shares in D4 to the Plaintiff.[111] Those were the disputes that arose from D1-D2’s refusal and failure to perform the Agreement. The First Award was issued on 27 October 2006, representing the outcome of the arbitration. When D1-D2 failed to comply with it, a fresh dispute giving rise to a fresh cause of action came into existence. But that was a dispute stemming from breach of the implied promise to comply with the Award, rather than a dispute “arising from the performance of the Agreement”. K.1b The separate causes of action 102. It is clear that as a matter of law, the implied promise to honour the Award exists as a contractual obligation separate and distinct from the obligations created by the underlying contract. As Otton J put it in Agromet Motoimport v Maulden Engineering Co (Beds) Ltd:[112] “In my judgment, the action on the award and the action to enforce an award is an independent cause of action. It is distinct from and in no way entangled with the original contract or the breach occurring from it, as reflected in the award.” 103. This is evident when one considers what must be proved to sustain such a claim. As we have seen, MacKinnon J pointed out in Norske Atlas Insurance Co Ltd v London General Insurance Co Ltd[113] that the requirements for suing to enforce an award are that there was a submission; that the arbitration was conducted in pursuance of the submission; and that the award is a valid award, made pursuant to the provisions of the submission, and valid according to the lex fori of the place where the arbitration was carried out and where the award was made. These are obviously elements which are quite different from the matters which must be proved to establish breach of the matrix contract – in the present case, proof of the renunciation, the failure to deliver the land, the restructuring and the failure to transfer the shares in the light of the Agreement. 104. The dissimilarity between the two actions is also clearly brought out by the different limitation periods applicable. As we have seen, the Arrangement provides that the limitation period for enforcement of Mainland awards is governed by the law of the place of enforcement, ie, Hong Kong law.[114] That period is laid down by section 4(1)(c) of the Limitation Ordinance:[115] “The following actions shall not be brought after the expiration of 6 years from the date on which the cause of action accrued, that is to say ... actions to enforce an award, where the submission is not by an instrument under seal”. 105. The cause of action for breach of the implied promise accrues at the date when the losing party ought to have complied with the award. The limitation period therefore begins to run at a date which will necessarily be later than the date of accrual of the cause of action which led to the arbitration. In the present case, the underlying breaches relied on occurred variously in 2003, 2004 and 2005 (and at the latest on 6 April 2006). The First Award was issued on 27 October 2006, which is the earliest date from which the limitation period relevant to the Action began to run. 106. The relevant limitation period was recognised in Agromet Motoimport v Maulden Engineering Co (Beds) Ltd,[116] where Otton J adopted the approach of the learned authors of Mustill & Boyd, Commercial Arbitration,[117] who stated: “... for the purposes of the Limitation Act 1980, it is necessary to classify an action on an award either as ‘an action to enforce an award, where the submission is not by an instrument under seal,’ for which the limitation period is six years, or as ‘an action upon a specialty,’ for which the limitation period is 12 years. But we submit that time begins to run from the date on which the implied promise to perform the award is broken, not from the date of the arbitration agreement nor from the date of the award.” 107. In National Ability SA v Tinna Oils & Chemicals Ltd (“The Amazon Reefer”),[118] Thomas LJ adopted the same approach. He regarded enforcement using the statutory procedure and by a common law action as two procedurally different methods of enforcing the same contractual cause of action, both triggered by breach of the implied contractual promise, to which the same limitation period applies. 108. Moreover, in the present case, the two claims are governed by different systems of law. Suing for breach of the implied promise is an aspect of enforcement (to which I shall return), regulated by the enforcing court according to the lex fori (here, Hong Kong law), to be distinguished from the arbitration concerning performance of the Agreement which was governed by PRC law, expressly chosen by the parties. K.1c The Bremer decision and the importance of context and purpose 109. In making the argument that the Action must be stayed because the implied promise to perform the award is “based on and arises out of” the Agreement, D1-D2 place reliance on Bremer Oeltransport GmbH v Drewry,[119] in particular, on the passage in Slesser LJ’s judgment where his Lordship states: “It would appear, therefore, that the greater weight of authority is in favour of the view that in an action on the award the action is really founded on the agreement to submit the differences of which the award is the result.”[120] 110. Bremer involved a dispute between parties to a charterparty entered into in London with an arbitration clause providing for arbitration in Hamburg. A dispute arose and the arbitration duly took place, leading to an award made in Hamburg. The plaintiff issued a writ in England seeking to enforce that award and sought leave to serve it on the defendant, a British subject resident in Paris. The question at issue was whether the English court could assume jurisdiction by permitting service of the writ out of the jurisdiction on the ground that the action was brought “to enforce a contract made within the jurisdiction” pursuant to what was then Order XI r. 1(e).[121] 111. Bremer was therefore not a case about staying to arbitration an action to enforce an award. Slesser LJ’s characterisation of the action as “founded on” the agreement in the charterparty was made for the purpose of assuming jurisdiction. It was in that context that the action on the implied promise was considered as “founded on” the arbitration agreement in the chaterparty and thus treated as enforcing a contract made within the jurisdiction. 112. Subsequent cases have stressed the importance of context and purpose when characterising the action on the implied promise. Thus, in FJ Bloemen Pty Ltd v Council of the City of Gold Coast,[122] the issue was whether certain sums could be claimed in a common law action to enforce an arbitration award as sums which were, on the contract’s true construction, payable by the defendant under the contract. Lord Pearson, giving the advice of the Privy Council, stated: “It is true ... that when an arbitrator fixes a sum to be paid by one party to the submission by way of damages for breach of contract the award creates a fresh cause of action superseding that arising out of the breach. But it does not appear to their Lordships to follow from that that the cause of action which comes into existence when the award is made cannot be said to arise under the contract which contains the submission. The award of an arbitrator differs materially from a judgment. The plaintiff's right to sue and the court's right to give judgment for him if he proves his case are not derived from the agreement of the parties and the judgment when given is an entirely fresh departure. The award of an arbitrator on the other hand cannot be viewed in isolation from the submission under which it was made. It was this sort of consideration which led the Court of Appeal in the case of Bremer Oeltransport G.m.b.H. v. Drewry [1933] 1 K.B. 753 to hold that an action brought to recover a sum awarded by an award made in Hamburg under a submission contained in a contract made in London was an action brought to enforce a contract made within the jurisdiction and their Lordships think that the same reasoning applies here. The distinction between an award which merely establishes and measures a liability under the contract and so does not create a fresh cause of action and an award of damages which supersedes the liability under the contract and creates a fresh cause of action, whatever its validity in other contexts, does not, in their Lordships' opinion, govern the broad question at issue here.”[123] 113. His Lordship therefore recognised that while, for certain purposes, the distinction between an arbitration award “which merely establishes and measures a liability” under the matrix contract and an action to enforce the award in pursuance of a fresh cause of action, may be important, in other contexts, that difference is immaterial. In FJ Bloemen the fact that the claim for payment was pursued in enforcing the arbitration award did not exclude the amounts claimed from being amounts payable under the contract. Bremer is an example where, for Order XI purposes, the enforcement action was treated as arising out of the charterparty contract made within the jurisdiction. 114. The Bumbesti[124] fell on the other side of the line with the enforcement action characterised as separate and distinct from the matrix contract. There, an attempt was made to establish Admiralty in rem jurisdiction on the basis that an action to enforce an arbitration award arising out of a bareboat charterparty was a claim arising out of an agreement “relating to the use or hire of a ship”.[125] It failed. Aikens J held: “The arbitration agreement is, at least, one step removed from the ‘use or hire’ of a ship. The breach of contract relied upon to found the present claim has nothing to do with the use or hire of the ship: it concerns the implied term to fulfil any award made pursuant to the agreement to refer disputes.”[126] 115. In Harbour Assurance Co (UK) Ltd v Kansa General International Insurance Co Ltd,[127] while discussing the separability of arbitration agreements,[128] Hoffmann LJ (as Lord Hoffmann then was) explained how different characterisations may be appropriate in different contexts for different purposes. The pleaded issues in that case raised the question whether the matrix contract was void for illegality. It gave rise to the collateral question whether, if it was void, it had to follow that the arbitration agreement was also void and of no effect. So characterisation of the arbitration agreement had to be considered: Was it part of the matrix contract or was it separable and capable of continued vigour despite the matrix contract’s illegality? Parallel issues are raised by D1-D2 in their outflanking argument. Hoffmann LJ’s observations are equally applicable to those issues: “Mr. Longmore's argument is extremely simple. He says that the question raised on the pleadings is whether the retrocession agreement was void ab initio. The arbitration clause formed part of the retrocession agreement. Therefore the issue must involve the validity of the arbitration clause itself. Mr. Longmore calls this logic. I call it over-simplification. The flaw in the logic, as it seems to me, lies in the ambiguity of the proposition that the arbitration clause ‘formed part’ of the retrocession agreement. In one sense of course it did. It was clause 12 of a longer document which also dealt with the substantive rights and duties of the parties. But parties can include more than one agreement in a single document. They may say in express words that two separate agreements are intended. Or the question of whether the document amounts to one agreement or two may have to be answered by reference to the kind of provisions it contains. In any case, it is always essential to have regard to the reason why the question is being asked. There is no single concept of ‘forming part’ which will provide the answer in every case. For some purposes a clause may form part of an agreement and for other purposes it may constitute a separate agreement. One must in each case consider the terms and purpose of the rule which makes it necessary to ask the question.”[129] 116. Adopting Hoffmann LJ’s approach, while it may be correct to hold that for certain purposes (such as founding jurisdiction) the action on the implied promise is “based on and arises out of” the underlying contract so as to be treated in the same manner as the matrix agreement, in the present case there is no justification for lumping the action on the implied promise together with the arbitration for breach of the matrix contract for the purpose of compelling a stay of the common law proceedings. They are plainly separate causes of action with different elements and features. More fundamentally, they are causes of action that arise at different phases of the arbitral process in which jurisdiction is allocated differentially to the arbitral tribunal and to the enforcing court, as discussed in the section which follows. K.1d The nature of the claim for damages in the Action 117. I turn next to Mr Gee’s proposition that the damages claimed in the Action should be viewed as damages for loss of the profits of the development flowing from breach of the Agreement and so caught by the arbitration clause. It is a suggestion which fails to recognise the fundamental difference between proceedings at the arbitration as opposed to the enforcement phase, and the range of remedies open to the enforcing court. 118. Lord Hobhouse of Woodborough[130] highlighted the different spheres of arbitrators and the enforcing court: “The essential purpose of arbitration is to determine disputes between the parties to the arbitration. Historically this was what the function of arbitrators was - to say who was right. The decision of the arbitrators could, as a result of the authority given to the arbitrator by the parties’ agreed submission to arbitration, declare what were the rights and liabilities of the parties and bind the parties by that declaration. Enforcement lay with the courts. Common law remedies were available besides statutory ones. It is possible to sue on the award or for damages for failing to honour the award; or to rely upon the award as having conferred a right or determined a fact. ... Statutes and International Conventions have since facilitated the direct enforcement of awards with a minimum of formality but still ultimately requiring the assistance of the judicial system. But the situation remains that the foundation of arbitration is the determination of the parties’ rights by the agreed arbitrators pursuant to the authority given to them by the parties. As section 58 of the United Kingdom Arbitration Act 1996 says, ‘an award made by the tribunal pursuant to an arbitration agreement is final and binding ... on the parties’.[131] It is an implied term of an arbitration agreement that the parties agree to perform the award.” 119. The line drawn is therefore between the determination of the parties’ mutual rights and liabilities, being a matter for the arbitrators, and enforcement of that award, being a matter for the court. 120. Mr David Joseph QC[132] referred to a number of legal texts elaborating upon the two phases of the process.[133] Conceptually, the parties’ submission to arbitration is contractual, giving a mandate to the arbitrators to resolve the disputes specified and containing a mutual promise, usually implied, that they will comply with the award. Once the final award is made, the arbitrators’ mandate is exhausted and, if the award is not complied with, a fresh cause of action arises for breach of the implied promise to honour the award. That cause of action lies within the enforcing court’s jurisdiction and is not within the power of the tribunal. Procedurally, enforcement may be pursued summarily (under section 2GG) or by a common law action on the award, as provided for by section 40B(1). As Thomas LJ noted in National Ability SA v Tinna Oils & Chemicals Ltd (“The Amazon Reefer”) regarding the common law action: “Enforcement of an award by action is by an ordinary action brought in the High Court. The procedure is not subject to any statutory provision, but it has long been established at common law as an action founded upon the implied promise to pay the award.”[134] 121. The split functions of the tribunal and the enforcing court were graphically described extra-curially by Lord Mustill as follows: “Ideally, the handling of arbitral disputes should resemble a relay race. In the initial stages, before the arbitrators are seized of the dispute, the baton is in the grasp of the court; for at that stage there is no other organisation which could take steps to prevent the arbitration agreement from being ineffectual. When the arbitrators take charge they take over the baton and retain it until they have made an award. At this point, having no longer a function to fulfil, the arbitrators hand back the baton so that the court can in case of need lend its coercive powers to the enforcement of the award.”[135] 122. At the enforcement stage, in an action on the implied promise, the enforcing court may grant relief appropriate to the award. If it is a monetary award for payment within the jurisdiction, it may simply be a judgment enforcing the award as a debt. If it is a non-monetary award which has not been complied with, the court may fashion an apt remedy chosen from the full range of remedies available in an ordinary common law action. Thus, in Agromet,[136] Otton J approved the approach of Mustill and Boyd who stated: “Parties to an arbitration agreement impliedly promise to perform a valid award. If the award is not performed the successful claimant can proceed by action in the ordinary courts for breach of this implied promise and obtain a judgment giving effect to the award. The court may give judgment for the amount of the award, or damages for failure to perform the award. It may also in appropriate cases, decree specific performance of the award, grant an injunction preventing the losing party from disobeying the award, or make a declaration that the award is valid, or as to its construction and effect.” 123. The cases involving non-monetary awards granting declarations of rights or specific performance illustrate the remedial flexibility of the enforcing court. Thus, in Selby v Whitbread Co,[137] when the defendant company demolished a building, withdrawing support from the plaintiff’s adjoining building and rendering it dangerous, surveyors appointed under an applicable statute made an award which required the defendant to erect a substantial pier to give support to the plaintiff’s structure and to do certain minor works. The defendant refused to comply and the plaintiff brought the action to enforce that award. McCardie J held that while “... a decree of specific performance can in some cases appropriately be granted in order to carry out the terms of an award”, in the instant case, such a decree should not be made because of difficulties of enforcement and because damages were an adequate remedy. So by way of relief, his Lordship granted the plaintiff a declaration that the award was valid and binding and awarded damages in respect of the erection of the pier and other works.[138] 124. In Birtley and District Cooperative Society Ltd v Windy Nook and District Industrial Cooperative Society Ltd (No 2),[139] an arbitration held pursuant to certain co-operative union rules resulted in an award declaring that the defendant co-operative trading society should not trade in certain geographical areas where the plaintiff was exclusively to provide services. When the defendant refused to be bound by the award, the plaintiff successfully brought an action for a declaration that the award was binding on the defendant, an injunction restraining them from trading contrary to the terms of the award and also an inquiry as to damages. 125. It has been held that a purely declaratory award, including a declaration of non-liability,[140] could be enforced both by a common law action and under the summary procedure. A judgment reflecting such a declaratory award could be relied on by way of an issue estoppel and, if necessary, enforced by suitable methods of execution, including in exceptional cases, a writ of sequestration.[141] 126. In the present case, the tribunal made a non-monetary award requiring continued performance of the Agreement. When it was discovered that the possibility of compelling such performance by requiring transfer of the shares had been frustrated as a result of the restructuring, the enforcing Court granted relief in the form of an award of damages. The fact that there is very likely to be a significant overlap between such damages and whatever damages might have been awarded by the tribunal for breach of the Agreement, does not mean that the Court’s order “usurps” the function of the tribunal. It is an order made at the enforcement phase, exercising the Court’s jurisdiction with a view to fashioning an appropriate remedy to give effect to the award, distinct from any remedy that might have been claimed in the arbitration. K.2 Enforcement action bounded by the scope of the Award 127. The second version of Mr Gee’s outflanking argument takes as its premise what he submits is the limited scope of the Award. He contends that the only issue referred to and decided by the tribunal, concerned the validity of the Agreement. The Award confirmed merely that the Agreement was binding. Referring to Lord Mustill’s analogy, Mr Gee submitted that the parties were only part of the way through the relay race and had not even entered the enforcement phase. Accordingly, the common law action on the award could not legitimately be employed to obtain from the Court of Appeal an order for damages to be assessed. That remedy went far beyond the scope of the Award and fell outside the Court’s enforcement jurisdiction. The judgment entered by the Court of Appeal therefore usurped the function of the arbitrators. If the Plaintiff wanted such damages, he argued, it had to return to the tribunal and persuade it to make such an award. Only then could the enforcing Court enter a judgment for such damages by way of enforcement. 128. In my view, the premise of that argument is untenable. The First Award was not merely about the validity of the Agreement. The tribunal certainly did decide that the Agreement was valid and not illegal. But it went much further. It held that D1-D2 were in breach of the agreement in several ways and rejected the submission that continued performance was impossible, re-affirming the relief awarded. It made it clear that it had fully adjudicated the case and that the First Award was final, recognising that the matter had reached the enforcement phase before the Hong Kong Court. 129. Thus, one sees from the First Award dated 27 October 2006 that the challenge to the Agreement’s validity was unsuccessfully raised by D1-D2 in a counterclaim. The Plaintiff’s claims for relief were based on their allegation that D1-D2 had breached the Agreement. It complained about the renunciation, arguing that “[the] contractual breach of [D1-D2] in unilaterally terminating the Agreement [was] against the principle of good faith and [had] no legal basis”, submitting that they “should continue performing the Agreement”. Such an order, the Plaintiff pointed out, was one form of relief provided for by Article 107 of the PRC’s Contract Law which: “... stipulates that ‘Where one party to a contract fails to perform the contractual obligations or its performance fails to satisfy the terms of the contract, it shall continue to perform its obligations, to take remedial measures, or to bear responsibility for breach of contract such as paying damages’.” 130. In support of an award for such continued performance (bearing in mind that at the stage of the first arbitration, D1-D2 had concealed the fact that they had divested themselves of control by allotting 9,998 shares in D4 to D3), the Plaintiff submitted: “Since the [D1-D2] have absolute control over [D5] and development and operation of [Lot 22], they have the ability to actually perform the agreement”. 131. The Plaintiff also complained about D1-D2’s breach by their failing to deliver the land within six months of the execution of the Agreement and claimed damages accruing at a daily rate totalling RMB 1,275,000 calculated up to 1 June 2005, reserving the right to claim further damages from 2 June 2005 until the date of actual delivery. 132. The Plaintiff argued that the challenge to validity was just an excuse and that D1-D2 had “wilfully breached the Agreement for the purpose of exclusively seizing the commercial benefits”. 133. Those arguments were accepted by the tribunal which held that D1-D2 had no justification for discontinuing performance of the Agreement unilaterally, rejecting the suggestion that performance was impossible notwithstanding the site’s development and the unit sales which had occurred. It also awarded the damages claimed for failure to deliver the land in time. 134. The tribunal saw the Agreement as essentially: “... a share transfer arrangement ... for the contractual purpose of assigning and obtaining the right to develop and obtain earnings from the development of [Lot 22].” To this end, the [Plaintiff] agrees to pay [D1-D2] RMB 120 million by instalments and progressively obtain the right to develop and make earnings from [Lot 22]. After the amount is paid in full, the [Plaintiff] is further entitled and obligated to purchase all the shares in [D4] for HK$2 ...” 135. It was evidently on the basis of a prospective transfer of all the shares in D4 in exchange for payment of RMB 120 million that the tribunal made the award requiring D1-D2 to continue to perform the Agreement. 136. Almost two years later, by an application dated 18 August 2008, D1-D2 returned to the tribunal, asking it “to order termination of the Agreement” on the ground that continued performance was impossible and claiming an award of RMB 6 million against the Plaintiff as compensation for alleged financial loss. This resulted in the tribunal’s Second Award dated 22 April 2009 which underlines the substantive effect of the tribunal’s award disposing of the arbitration. 137. At that stage, the restructuring had been revealed and D1-D2 put this forward as a reason (along with completion of the physical development and sale of most of the units) why continued performance was impossible. They sought to persuade the tribunal that they were entitled to terminate pursuant to Article 11 of the Agreement which specified certain circumstances which would justify termination. 138. The tribunal noted that in the First Award, completion of the development and the unit sales did not satisfy the conditions justifying termination. As to the restructuring, the Second Award stated: “... [D1-D2] by the representations and warranties in Article 2 of the Agreement ... expressly guarantee that they have the absolute control right of [D4]. [D1-D2’s] changes to the shareholding of [D4] during the arbitration procedure of the earlier case have breached the Agreement ... The [tribunal] considers that the change of shareholding of [D4] likewise does not satisfy the condition set out in Clause 2 of Article 11... and cannot be a ground for [D1-D2] to terminate the Agreement.” 139. The Second Award also noted that “the recognition and enforcement of the earlier arbitral award” was being carried out in the Hong Kong Court in accordance with Hong Kong legal procedures and refrained from commenting on the earlier award and the Hong Kong Court’s judgment. 140. Finally, it will be recalled that, by letter dated 19 June 2009, D1-D2 made one last attempt to get the tribunal to accept that continued performance was not possible, asking for directions “on how it should be performed and/or what relief (including damages or other appropriate reliefs)” the Plaintiff should have to meet the purpose of the Agreement. The tribunal replied on 27 July 2009 stating that it had “adjudicated the case fully” that the First Award was final and that the tribunal “did not leave out any matter which would need to be further determined”. 141. It is clear from the foregoing that Mr Gee’s suggestion that the awards were solely concerned with the validity is wholly unsustainable. Quite apart from finding that the Agreement was binding, the tribunal held that D1-D2 were in breach in the abovementioned respects. It awarded damages for the delay in delivering the land and made the order for continued performance as relief for the breaches. The Hong Kong Court translated the award into a judgment under the summary procedure and, after it was discovered that the restructuring had occurred, the Plaintiff started the common law action and sought damages as an alternative remedy in the enforcement proceedings. After failing against D1-D2 at trial but succeeding in the Court of Appeal, the Plaintiff was put to its election and elected for damages which were granted as a remedy within the power of the enforcing court. 142. Mr Gee’s argument that the arbitration was incomplete and that the Plaintiff ought to return to the arbitrators to seek an award of damages ignores the tribunal’s letter 27 July 2009 firmly refusing to entertain further proceedings (including proceedings for “damages or other appropriate reliefs”), asserting that the arbitration was final and comprehensive. 143. While it is conceivable that cases could arise where a remedy sought in a common law action may be refused on the ground that such remedy is unwarranted given the limited scope of the award, the present appeal comes nowhere near to such a case. L. The extant award argument 144. D1-D2 contend that the Court of Appeal was wrong to award the Plaintiff damages because such relief “is fundamentally inconsistent with, and barred by, the extant Award (which requires parties to continue to perform the agreement)”.[142] They argue that: “... according to [Johnson v Agnew], when the decree of specific performance had not been complied with, the innocent party may ‘apply to the court to put an end to the contract’ and the court may, in the exercise of its equitable jurisdiction, order the contract to be dissolved and award damages. In this case, P never applied to the tribunal to give up any of its rights under the Award or for permission to treat the Agreement as terminated. In the circumstances, the enforcing court should not award damages on the basis that the underlying contract was no longer subsisting when there was still an extant award for continued performance of such contract.”[143] 145. The argument is that to enable the enforcing court to grant them damages, the Plaintiff must first return to the tribunal and ask it to rescind its order that “[D1-D2] shall continue to perform the agreement…” Unless that is done, the enforcing Court exceeds its powers in awarding damages which are inconsistent with the extant award. 146. The argument as presently formulated is a modified and diminished version of the argument that had been made before the Court of Appeal. There, the issue was whether the plaintiff was entitled to claim damages in the common law action on the award when it had already entered judgment under the statutory process.[144] D1-D2’s argument was that in entering judgment, the Plaintiff had made its election and could not subsequently seek inconsistent relief in the form of damages. As noted above,[145] Yuen JA held that there had been no election and that the Plaintiff was entitled to an award of damages if a properly informed election for such relief was made. The Plaintiff duly made such election and, on 20 October 2017, the Court of Appeal set aside the statutory judgment and entered judgment in the Action in favour of the Plaintiff for damages against D1-D2 for breach of the implied promise to honour the Award, with quantum to be assessed later. 147. Events have thus overtaken that earlier argument and the present proposition is that there is an inconsistency, not with the statutory judgment (which has been set aside), but with the terms of the award which, D1-D2 argue, is still extant. Johnson v Agnew[146] is relied on by analogy. 148. I do not accept that argument. It fails to recognise that the relief granted in the common law action falls within the enforcement jurisdiction of the Hong Kong Court and is neither a matter for the tribunal nor relief which in some way requires adjustment to achieve consistency with the terms of the award. 149. Johnson v Agnew does not assist D1-D2’s argument. That was a decision concerned with consistency of remedies claimed in the same action and when a party may be held to have irrevocably elected for a particular remedy. It was not concerned with any purported requirement of consistency between relief granted in a common law enforcement action and the content of an arbitral award. In any event, the enforcement proceedings in the present case are perfectly in tune with Johnson v Agnew. Thus, (in relation to remedies sought in a vendor and purchaser action) Lord Wilberforce noted that the vendor may proceed against the defaulting purchaser by action for specific performance or damages in the alternative but would have to elect which remedy to pursue at trial.[147] He went on to point out that “if the order for specific performance is not complied with by the purchaser, the vendor may either apply to the court for enforcement of the order, or may apply to the court to dissolve the order and ask the court to put an end to the contract.”[148] That is essentially the course followed in the present enforcement proceedings. 150. The so-called “extant award” presents no obstacle to the Hong Kong enforcing Court awarding the Plaintiff damages for breach of the implied promise to honour the Award. M. Conclusion as to D1-D2’s appeal 151. For the abovementioned reasons which agree in large measure with the reasons of the Court of Appeal, I would dismiss D1-D2’s appeal. Mr Justice Fok PJ: 152. I agree with the judgments of Mr Justice Ribeiro PJ and Lord Sumption NPJ. Mr Justice Cheung PJ: 153. I agree with the judgments of Mr Justice Ribeiro PJ and Lord Sumption NPJ. Mr Justice Bokhary NPJ: 154. I agree with the judgments of Mr Justice Ribeiro PJ and Lord Sumption NPJ. Lord Sumption NPJ: 155. I gratefully adopt the account of the facts in the judgment of Mr Justice Ribeiro PJ, which is concerned with the arbitration issues in this complex and long-running dispute. I shall also adopt the terminology that he uses. I agree with the order which he proposes in Appeal FACV 3 of 2019, and with his reasons. The present judgment is concerned with the Plaintiff’s appeal (No. 5 of 2019). The object of that appeal is to challenge the comprehensive dismissal by the courts below of the Plaintiff’s non-contractual claims against the D3-D11, who are associated companies of D1-D2 or involved in the management of the group, and were concerned to a greater or lesser degree in the events which have given rise to this dispute. Leave to appeal was granted by this Court, limited to two of those claims. 156. The first is a claim in tort for inducing D1-D2’s breaches of contract. It is nominally directed against all nine of the D3-D11, but in this Court has been pursued only against D3-D6. In this Court, this claim is limited to liability said to arise out of the 2005 restructuring of the group to which D1-D5 belonged. D3-D5 participated in the group restructuring that resulted in D1-D2 ceasing indirectly to control Xiamen Legend (D5), the owner of the land use rights for Lot 22. Mr Tan (D6) was the founder and effective controller of the group. The main issues under this head concern the nature of the knowledge and intent required in order to make an ancillary party liable for inducing a breach of contract. 157. The second claim, which is now pursued against D3 alone, has been given the general label “constructive trust” in these proceedings. What is said is that the contractual obligation of D1-D2 to transfer to a company nominated by the Plaintiff the shares in D5’s parent company Hong Kong Legend (D4) was amenable in Hong Kong law and practice to specific performance. Therefore, it is argued, the Plaintiff acquired by virtue of the Agreement an immediate beneficial interest in the D4 shares which a Hong Kong Court should recognise. When, in the course of the group restructuring of 2005 D1-D2 transferred the shares in D4 to D3, that beneficial interest subsisted and became enforceable against D3, which had notice of the facts giving rise to it. The main issues under this head are whether Hong Kong law is the relevant law for the purpose of deciding whether such a beneficial interest existed and, if so, whether D1-D2’s contractual obligation to transfer the shares was specifically enforceable as a matter of Hong Kong law. 158. Both of these claims were rejected in the courts below. They also rejected a third non-contractual claim of conspiracy to injure the Plaintiff by the use of unlawful means. Leave to appeal was sought from this Court on a limited issue arising under this head, but was refused. We are not therefore concerned in this appeal with the cause of action for unlawful means conspiracy. Inducing a breach of contract 159. At common law, an act sued on in Hong Kong but committed in another jurisdiction can be tortious only if it is actionable under the law of both jurisdictions. The rule, commonly called the “double actionability rule”, is subject to an exception for cases in which the act, wherever committed, is sufficiently closely connected with another jurisdiction. The double actionability rule has been discarded in a number of common law jurisdictions including Australia and Canada, and abolished by statute in others, including England and New Zealand. Its application in Hong Kong may one day need to be reconsidered, but for the moment it is accepted on all sides that the rule, together with the exception, remains the law in Hong Kong. The Judge dismissed the entire claim for inducing a breach of contract on the ground that no such action was recognised by the law of the PRC. The Court of Appeal dismissed the claim on the same ground so far as it arose from the renunciation of the Agreement, and that part of the judgment is no longer in issue. But so far as the cause of action arose from the reconstruction, the Court of Appeal dealt with it purely as a matter of Hong Kong law, presumably on the basis that since the reconstruction was achieved by transactions in the shares of a company incorporated in Hong Kong it happened in Hong Kong. 160. The modern origin of the tort of inducing a breach of contract is the famous decision of the Court of Queen’s Bench in England in Lumley v Gye (1853) 2 E&B 216. The effect of this decision was to make a person who procured a breach of contract liable jointly with the contract-breaker as an accessory to his wrong. The tort, like other economic torts, gives rise to difficult questions of legal policy which are rarely overtly recognised but never far from the surface. The dilemma is how to protect contractual rights against interference from non-parties without unduly limiting the commercial opportunities of businessmen. In the nature of things, commercial competition causes damage to other traders. What limits can the law legitimately place upon it? 161. Since Victorian judges in England first grappled with these questions, most of the world has adopted the statutory regulation of competition, taking its cue from the case-law on the Sherman and Clayton Acts of the United States. Hong Kong has done so since 2012. Statutory regulation has thus come to assume part of the burden of resolving the economic dilemma. But private law, and in particular the economic torts, still has an important role to play. The main difference is that in private law the relationship giving rise to a duty owed by the Defendant to the particular Plaintiff is of critical importance. It commonly arises from the Defendant’s knowledge and intentions in acting as he did. By comparison, public law competition regulation is more concerned with consequences and with objective standards of conduct. 162. The elements of the tort of inducing a breach of contract are (i) a contract, (ii) known to a third party who (iii) does something which induces or persuades a contracting party to break it, (iv) intending to bring about the breach and (v) thereby causing loss. Most of the problems about this area of law have concerned the kind of acts which will be regarded as having “induced” the breach of contract, and the necessary mental state of the alleged inducer. The tendency of much of the case-law until recently was to systematise and assimilate the various economic torts. This may perhaps have made the law more coherent and consistent, but in the process it lost sight of important differences between the various torts, in particular differences between their economic functions and the control mechanisms required in order to keep them within proper limits. In dealing with the tort of inducement, they also overlooked the accessory nature of the liability which had been a cardinal feature of the tort ever since Lumley v Gye. In OBG Ltd v Allan [2008] 1 AC 1 the House of Lords, after reviewing a large body of authority, unbundled the principal economic torts and drew attention to the distinct elements of each. It should now be unnecessary to refer to any older authority on the points which OBG decided. 163. The principal speeches in OBG were delivered by Lord Hoffmann and Lord Nicholls of Birkenhead. They were agreed on the elements of liability for inducing a breach of contract. Lord Walker of Gestingthorpe, Baroness Hale of Richmond and Lord Brown of Eaton-under-Heywood agreed with both of them on that tort, and proceeded to discuss the points on which they differed, all of which related to other torts. 164. Lord Hoffmann dealt separately with knowledge and intent. He dealt with knowledge at §39: “To be liable for inducing breach of contract, you must know that you are inducing a breach of contract. It is not enough that you know that you are procuring an act which, as a matter of law or construction of the contract, is a breach. You must actually realize that it will have this effect. Nor does it matter that you ought reasonably to have done so.” It follows that the alleged inducer must know not just the facts but their legal consequence. This is apparent from Lord Hoffmann’s citation of the House of Lords’ decision in British Industrial Plastics Ltd v Ferguson [1940] 1 All ER 479 as an illustration of his statement of principle. The Defendant was held not to be liable for inducing a breach of an employee’s duty of confidentiality to his employer, because he held the mistaken, indeed unreasonable (“eccentric”) view that there was no such breach as the information disclosed belonged to the employee. 165. The test of intention is related to that for knowledge: “42. … It is necessary for this purpose to distinguish between ends, means and consequences. If someone knowingly causes a breach of contract, it does not normally matter that it is the means by which he intends to achieve some further end or even that he would rather have been able to achieve that end without causing a breach. Mr Gye would very likely have preferred to be able to obtain Miss Wagner's services without her having to break her contract. But that did not matter. Again, people seldom knowingly cause loss by unlawful means out of simple disinterested malice. It is usually to achieve the further end of securing an economic advantage to themselves. As I said earlier, the Dunlop employees who took off the tyres in GWK Ltd v Dunlop Rubber Co Ltd 42 TLR 376 intended to advance the interests of the Dunlop company. 43. On the other hand, if the breach of contract is neither an end in itself nor a means to an end, but merely a foreseeable consequence, then in my opinion it cannot for this purpose be said to have been intended. That, I think, is what judges and writers mean when they say that the claimant must have been ‘targeted’ or ‘aimed at’. In my opinion the majority of the Court of Appeal was wrong to have allowed the action in Millar v Bassey [1994] EMLR 44 to proceed. Miss Bassey had broken her contract to perform for the recording company and it was a foreseeable consequence that the recording company would have to break its contracts with the accompanying musicians, but those breaches of contract were neither an end desired by Miss Bassey nor a means of achieving that end.” 166. Lord Nicholls’ analysis was the same. He spoke at §191 of the inducer’s “intentional causative participation” in the breach. He warned against equating liability under this head with negligence. Ignorance, however careless, is a defence. Lord Nicholls identified the element of intent at §192 as follows: “He is liable if he intended to persuade the contracting party to breach the contract. Intentional interference presupposes knowledge of the contract. With that knowledge the defendant proceeded to induce the other contracting party to act in a way the defendant knew was a breach of that party's obligations under the contract. If the defendant deliberately turned a blind eye and proceeded regardless he may be treated as having intended the consequence he brought about. A desire to injure the claimant is not an essential ingredient of this tort.” To the same effect is the brief statement of Baroness Hale at §306: “[T]he common thread is striking through a third party who might otherwise be doing business with your target, whether by buying his goods, hiring his barges or working for him or whatever. The refinement proposed by my noble and learned friend, Lord Hoffmann, is entirely consistent with the underlying principles to be deduced from the decided cases. It is also consistent with legal policy to limit rather than to encourage the expansion of liability in this area.” 167. The effect of the judgments in OBG is that the act said to constitute the inducement must be targeted at the Plaintiff, in the sense the breach of his contract is either an end in itself or the means to an end. The trade union cases in which the union procures a breach of its members’ contracts of employment are generally cases in which breach of the contract is an end in itself. But, as Lord Hoffmann pointed out, it is more usual for the inducer to be out to serve some commercial interest of his own. He is liable because he knows and intends the breach of contract to be a means to that end. 168. The Agreement included, at Article 2.3, a warranty that D1-D2 “has absolute control over [D4] and [D5]” and an undertaking (not in the event honoured) that after the Agreement had been concluded they would convene a board meeting of those two companies to approve its terms. The purpose of this warranty was manifestly to ensure that D1 was in a position to procure compliance with the terms. It makes sense only on the footing that it related to the position not just at the time of the Agreement but throughout the period when the contract was executory. Moreover, even without such a warranty, it must necessarily be implied that D1 would not put it out of his power to perform. 169. The Judge held that the reconstruction of 2005 was not a breach of the Agreement although it involved D1 parting with control of D4-D5. This was because the relevant breach of contract had already occurred two years before when the Agreement was renounced. This conclusion cannot be correct, and no attempt has been made to defend it in this Court. The renunciation was a repudiation of the Agreement, but the Plaintiff were not obliged to accept it and declined to do so. The contract therefore continued to bind and both sides remained obliged to perform it. That was the overt position of the Plaintiff, who maintained it throughout the two references to arbitration in Beijing and indeed still does. It follows that the reconstruction was a breach of contract. The Court of Appeal accepted that but thought that the prior renunciation of the Agreement by D6 was relevant for a different reason. It explained why at the time of the reconstruction D1-D2 regarded the Agreement as having already been brought to an end for all practical purposes. It followed that the claim for inducing a breach of contract had “no ‘causative’ element” (§249). This was no doubt an echo of Lord Nicholls statement in OBG (§191) about “intentional causative participation” in the breach. 170. Although put as a matter of causation, the Court of Appeal’s point would be better described as a matter of intent. It is, however, the same point albeit with a different label. In cases like British Motor Trade Association v Salvadori [1949] Ch. 556 where the inducement consisted of persuasion, or D.C. Thomson & Co. Ld. v Deakin [1952] Ch. 646 where it consisted of threats, difficult questions arose about what kind of act could be regarded as “inducing” a breach of contract. These difficulties do not arise on the present appeal. The reconstruction undoubtedly caused D1-D2 to be in breach of contract because it was itself a breach of warranty and because it made it impossible to procure the transfer of the shares in accordance with the Agreement. Whether that caused loss or damage so as to complete the tort is a different question with its own difficulties. But this appeal is not concerned with the character of the act constituting the inducement or the question what loss can be said to flow from it. It is concerned with the mental element required of the inducer. The Court of Appeal held that the relevant Defendants cannot have intended the reconstruction to result in a breach of contract because they believed, albeit wrongly, that the contractual obligation to transfer the shares was already at an end once they had already made the decision not to perform it. 171. In my judgment the Court of Appeal was correct. The Judge found that at an early stage, well before the first arbitration award, D6 had directed his brother-in-law Mr. Cheung (D10), a senior member of the group management, to set up an intermediate holding company in the PRC for the property interests of the group in the PRC, in the same way as had been done for its brewery interests in 1995. There were important commercial reasons for doing this, unrelated to the dispute about the Agreement relating to Lot 22. It was necessary in order to make possible intra-group loans and cross-guarantees in a manner consistent with PRC law. The proposal to restructure the group on these lines was set out in a memorandum of 25 August 2005 of Mr. Mok, the group Financial Controller. Mr. Mok’s evidence, which the Judge accepted, was that no one had told him to make the transfer of the D4 sharesimpossible and that this thought never occurred to him. This was mainly because he had read the Agreement and had noted Article 11.3, which provided for purely financial remedies in the event that it was terminated as a result of a breach of the warranty in Article 2. It provided that D1-D2 would be required to refund twice the amount of the deposit, the transfer price paid, and “default compensation” equal to the transfer price paid. D1-D2 disputed the validity of the Agreement in the arbitration proceedings. But even on the footing that it was valid and wrongfully repudiated, Mr. Mok did not believe that there could now be any question of D1-D2 having to transfer the shares. He left the details of the reconstruction to be worked out by Ms. Wong, a senior manager in the company secretarial department. It was she who proposed the form which the reconstruction ultimately took, namely the allotment of the D4 shares to D3 so as to make the latter its majority shareholder. For her, this was a routine transaction of a kind which she had often set up before. The Judge accepted her evidence that her reason for carrying it out in the way she did was that it made for a simpler transaction, with fewer complications as regards stamp duty and accounting. 172. For present purposes it is irrelevant that because of the Plaintiff’s refusal to accept the repudiation the Agreement was still open, because on the findings of fact this was no part of the Defendants’ state of mind. Mr. Mok can fairly be convicted of not having appreciated that an unaccepted repudiation is “writ in water”, or the implications of the House of Lords’ decision in White and Carter (Councils) Ltd v McGregor [1962] AC 413. He has that much in common with many reputable common lawyers. I do not know whether there is a corresponding rule in the law of the PRC, which governed the Agreement. But whether there is or not, Mr. Mok’s view of the effect of Article 11.3 was not so strange as to suggest that he could not have held it. For present purposes it is enough that both the Judge and the Court of Appeal accepted that the group management regarded the contractual obligations as being at an end. That being so, their state of mind was not significantly different from that of the Defendant in British Industrial Plastics Ltd v Ferguson who wrongly believed that in law a trade secret was the employer’s property to dispose of, and was therefore rightly acquitted of inducing its unlawful disclosure. 173. It follows that none of the relevant participants in the reconstruction intended to frustrate the contractual obligation to procure the transfer of the shares, even if that was its effect. In fact, however, it was not its effect. The group had already firmly decided not to transfer the shares and (for reasons which I shall explain below) could not have been compelled to transfer them. The reconstruction did not therefore frustrate the transfer. The question whether that was a foreseeable consequence therefore does not arise on the facts, quite apart from being irrelevant in law. Nor is it relevant that (as the Judge speculated) D6 was likely to have been aware that the reconstruction would make the transfer of the shares to the Plaintiff’s nominee impossible, and to have found this “not unwelcome”. Knowledge of or satisfaction at the result is not the same as intending it. As far as D6 was concerned, the fate of the Agreement was already sealed. Constructive trust 174. A constructive trust can exist only if there is a relevant fiduciary relationship. It may arise on either of two juridical bases. It may be derived from the character of the relationship between the beneficiary or the trustee, where the court recognises that relationship as fiduciary. In such a case the trust gives rise to a true proprietary interest in equity. Or it may be imposed by the court on a purely remedial basis in a case where there is no subsisting proprietary interest (usually because it has been extinguished) but the relevant party is a fiduciary and is before the court. I sought to set out the conceptual framework in Akers v Samba Financial Group [2017] AC 424 at §82. The Plaintiff in this case claims to derive a fiduciary relationship from the executory contract for the sale of the shares. The principle, which was said in 1876 to have been settled law for at least two centuries, is that “the moment you have a valid contract for sale the vendor becomes in equity a trustee for the purchaser of the estate sold”: Lysaght v Edwards (1876) 2 Ch. D. 499, 506 (Jessel MR). This well-known statement of principle related to land. But the same rule has long been applied to contracts for the sale of unquoted shares, which are regarded as unique in the same way as land. It gives rise to a true proprietary interest in equity, arising from the fiduciary character of the relationship between vendor and purchaser under an executory contract of sale. The basis of the rule in both cases, and the essential condition for its operation, is that the contractual obligation to transfer the property should be specifically enforceable. This means that the purchaser’s beneficial interest arises as soon as the contract is made or (if later) as soon as a relevant promissory condition precedent to the transfer has been satisfied. 175. The first question for us is by what law the existence of such beneficial interest is to be determined. The existence and incidents of an interest in property is governed by the lex situs. In the case of shares, this is the law of the jurisdiction in which the company is incorporated: Akers v Samba Financial Group at §19 (Lord Mance), §84 (Lord Sumption). The position is, however, complicated where there is a contract governed by the law of some other jurisdiction. The lex situs determines whether an equitable proprietary interest in the shares is legally capable of existing. If it is legally capable of existing, the next question is whether it has actually been created. Where this is said to be the effect of a contract, the lex situs refers to the proper law of the contract to determine what it means. The question in such a case is not whether its proper law recognises constructive or indeed any trusts. The question is whether the characteristics which the proper law of the contract treats it as having, are such as the lex situs would regard as giving rise to a proprietary interest in equity: Kuwait Oil Tanker Co SAK v Al-Bader (No. 3) [2000] 2 All ER (Comm.) 271, at §§190-193 (Nourse LJ); First Laser Ltd v Fujian Enterprises (Holdings) Co Ltd (2012) 15 HKCFAR 569 at §§66-70 (Lord Collins of Mapesbury NPJ). As applied to the present case, the result of these principles is that the law of Hong Kong, as the situs of the shares, determines whether an equitable interest in shares of a Hong Kong company is capable of subsisting. The answer to that question is that it is, if the contract is specifically enforceable. It is then necessary to refer to the law of the PRC, as the proper law of the Agreement, to determine whether the contract has the characteristics which make it specifically enforceable in the law of Hong Kong. For this purpose, the arbitration award is a decisive source. The D3 was not party to the arbitration agreement. But the D3’s liability as a constructive trustee depends on the position as between D1-D2 and the Plaintiff, all of whom are bound by the arbitration agreement. 176. The Judge and the Court of Appeal found that the law of the PRC did not recognise the concept of trusts. Both of them thought that that was the end of the matter. For the reasons which I have given I do not agree. They should have gone on to ask themselves whether the characteristics of the Agreement in PRC law were such as to make it specifically enforceable in Hong Kong law. But their decision to dismiss the claim in constructive trust was nevertheless correct, because the Hong Kong courts would not specifically enforce a contract with those characteristics. There are, as it seems to me, two related reasons for this. 177. First, subject to limited exceptions, it is a bar to specific performance of a contract that the obligation to be enforced has not arisen because it is subject to a condition precedent which has not been satisfied. Until it is, there is no equitable proprietary interest: Wood Preservation Ltd v Prior [1969] 1 WLR 1077, 1092 (Goff J). The exceptions follow from the logic of the rule. Specific performance may nevertheless be ordered and the purchaser’s beneficial interest will subsist if either (i) it is wholly within the power of the obligee to satisfy or waive the condition, or (ii) the obligor is bound to satisfy the condition and that obligation is itself specifically enforceable. 178. Secondly, it is a bar to specific performance that it would require what the courts have called “supervision”. This is the ground on which the courts refuse to enforce a contract to carry on a business or to build a house. Of course, the court does not literally supervise anything. It makes orders which are enforceable by proceedings for contempt. What is meant by references to supervision in this context is that specific performance will not be ordered if the relevant obligation depends on disputable questions, such as whether a given standard of performance has been achieved, which might have to be referred for decision to a court before it could be known whether the obligation was enforceable. As Lord Lane, delivering the advice of the Privy Council on appeal from Hong Kong, said in Anstalt Nybro v Hong Kong Resort Co. Ltd. [1980] HKLR 76, 82: It is a vain exercise to try to force one person against his will to co-operate with another and it would not be proper to attempt by order to produce that effect. 179. The terms of the Agreement give rise to both problems. The critical feature of the Agreement was that the land comprised in Lot 22 was to be delivered to the Plaintiff well before the shares were to be transferred. The terms contemplate, although they do not expressly require, the development to be begun between delivery of the land and the transfer of the shares. Most of the terms are concerned with the rights and obligations of the parties in relation to the development during that period, the satisfaction of which was a condition precedent to the transfer of the shares. Article 5 provided that the land was to be delivered to the Plaintiff after all the current residents had been moved out, but in any event within six months of the date of the Agreement. D1-D2 were to carry out the demolition of the Legend Commercial Building. It provided that the Plaintiff was to carry out the demolition of other buildings on the site with D1-D2’s assistance. Articles 6 and 7 provided that the Plaintiff was to decide how the land should be developed and the development marketed, but subject to D1-D2’s right to approve certain aspects of the development plans and the financing, and to “supervise” the development and operations on the site. D1-D2 also had a veto on the Plaintiff’s choice of builder “if it has proper ground”. Until the transfer price of the shares was paid there were restrictions on the proportion of the development that the Plaintiff could dispose and the disposal of any proceeds of sale. Article 4 dealt with the payment of the transfer price of the shares. It was to be paid in installment, on signature of the Agreement, on delivery of the land, within a year after the delivery of the land and within two years after the delivery of the land. Under Article 8, the transfer of the shares was to occur “when [the Plaintiff] has settled the total transfer price and fulfilled the obligations according to the Agreement.” The arbitrators, who analysed the terms in detail in the First Award, did not regard the Agreement as a straightforward contract for the transfer of shares or land use rights. They described it as a “complicated arrangement of rights and obligations closely related to the transfer of shares,” and a “framework agreement whose performance may be difficult due to various uncertainties”. For that reason, they considered that it “needs close cooperation between the parties”. 180. Against that background, I return to the question of specific performance. It will be apparent that there were a number of conditions to be satisfied before the shares could be transferred. The satisfaction of those conditions was not within the Plaintiff’s sole control. Nor was the Plaintiff in a position to waive them, since they were inserted at least in partly for the protection of the unpaid vendor. Some of the conditions related to demolition, design and construction and had to be satisfied before the transfer of the shares. They called, as the arbitrators observed, for a high level of cooperative performance, and were inherently disputable. I would not go as far as the courts below did when they described the Agreement as a joint venture for the development of Lot 22. But it clearly had much in common with a joint venture during the period of up to two years which would intervene between the delivery of the land and the transfer of the shares. In my judgment, an agreement with these features was not one which a Hong Kong court applying its own law would specifically enforce. Conclusion 181. For these reasons I would dismiss the Plaintiff’s cross-appeal FACV 5 of 2019. Mr Justice Ribeiro PJ: 182. The Court unanimously dismisses both appeals. The parties are at liberty to lodge written submissions as to costs within 21 days after the date of this judgment, to be dealt with on the papers. FACV 3/2019 Mr Steven Gee QC, Mr Benjamin Yu SC, Mr Richard Khaw SC and Ms Bonnie Y K Cheng, instructed by Mayer Brown, for the 1st and 2nd Defendants (1st and 2nd Appellants) Mr David Joseph QC, Mr Edward Chan SC, Mr Bernard Man SC, Mr Lee Tung Ming, Mr Keith Lam, Mr Justin Ho and Mr James Man, instructed by Anthony Siu & Co., for the Plaintiff (Respondent) FACV 5/2019 Mr David Joseph QC, Mr Edward Chan SC, Mr Bernard Man SC, Mr Lee Tung Ming, Mr Keith Lam, Mr Justin Ho and Mr James Man, instructed by Anthony Siu & Co., for the Plaintiff (Appellant) Mr Steven Gee QC, Mr Benjamin Yu SC, Mr Richard Khaw SC and Ms Bonnie Y K Cheng, instructed by Mayer Brown, for the 1st and 2nd Defendants (1st and 2nd Respondents) Mr Benjamin Yu SC, Mr Paul Shieh SC and Ms Elizabeth Cheung, instructed by Wilkinson & Grist, for the 3rd and 4th Defendants (3rd & 4th Respondents) Mr Ronny Tong SC and Ms Kate Poon, instructed by Baker & McKenzie, for the 6th Defendant (5th Respondent) [1] The name in Chinese: 廈門新景地集團有限公司. [2] Yuen, Barma and Poon JJA, CACV 158/2012, [2016] 2 HKLRD 1106 (15 April 2016). Unless otherwise indicated, references in this judgment to paragraphs (§§) in the judgment of the Court of Appeal are references to the aforesaid judgment. [3] HCCL 13/2011 (14 June 2012). [4] Agreement, Article 5. [5] Ibid, Article 6.1. [6] Ibid, Article 7. [7] Court of Appeal §23. [8] Ibid §35. [9] Ibid §30. [10] The Plaintiff nominated Mr Cheng Dejun and D1-D2 jointly nominated Mr Gu Gongyun. The director of CIETAC appointed Mr Liu Yuwu as chief arbitrator. Subsequently, on 3 March 2006, Mr Liu disqualified himself on the ground of conflict of interest and on 7 March 2006, Mr Kou Liyun was appointed as chief arbitrator in his stead. [11] Court of Appeal §56.4. [12] Award [2006] No 0389. References in this judgment to the content of the tribunal’s awards are to English translations thereof. [13] Court of Appeal §64. [14] Then Cap 341, since replaced by the Arbitration Ordinance (Cap 609). Relevant provisions are set out in Section J.1 of this judgment. [15] As Cheung PJ then was. [16] DHCJ Stone §§41 and 105. [17] Court of Appeal §71. [18] In HCCT 54/2007, [2008] 4 HKLRD 972 (24 June 2008). [19] Relying on section 40E(3) of the Ordinance. Reyes J §§2-3. [20] Reyes J §§86-88. [21] Ibid §§95, 111. [22] Ibid §109. [23] Ibid §133. [24] Rogers VP, Le Pichon and Hartmann JJA, cacv 106 and 197/2008, [2009] 4 HKLRD 353 at §31. [25] Award [2009] No 0144. [26] While the Action was listed as HCA 961/2008. [27] Fok J §§137, 149-150. [28] Ibid §§151, 154. [29] Rogers VP, Le Pichon and Kwan JJA, CACV 88 and 89/2010, [2011] 1 HKLRD 781 (15 December 2010). [30] CA Stay Appeal §§15-16. [31] Ibid §29. [32] Ibid §30. [33] Ibid §31. [34] Ibid §32. [35] Ma CJ, Bokhary and Ribeiro PJJ, FAMV 4 and 5 of 2011 (4 May 2011). [36] Determination §§4-5. [37] Ibid §7. [38] Hon Cheung CJHC (as Cheung PJ then was) and McWalters J, HCMP 13, 15, 18, 21/2012 (14 February 2012) at §7. [39] HCCL 13/2011 (14 June 2012). [40] DHCJ Stone §80. [41] By A Cheung J in Action HCCT 54/2007. [42] DHCJ Stone §§51 and 129. [43] Ibid §114. [44] Ibid §117. Italics in the original. [45] cacv 106 and 197/2008, [2009] 4 HKLRD 353 at §31. [46] DHCJ Stone §§140-141, 144. [47] Ibid §145. [48] Ibid §§145-147. Italics in the original. [49] Ibid §170. [50] Ibid §§191 and 195. [51] Ibid §§205, 208-210, 217-219. [52] [1980] AC 367. [53] DHCJ Stone §§220-223, 228. [54] Ibid §§229, 236. [55] Ibid §234. [56] Ibid §§195, 200, 270-271, 327-329. [57] Ibid §§205, 207-211, 331-333. [58] Ibid §114. [59] Ibid §§290, 304, 366-367. [60] Ibid §§253-254, 291, 296, 298, 317-318, 345-346 [61] Ibid §§258-261, 275, 279. [62] [1920] 3 KB 497. [63] DHCJ Stone §§335, 338-340, 354, 356, 358, 365. [64] Ibid §§248, 250-251, 304-306, 364. [65] Yuen, Barma and Poon JJA, CACV 158/2012, [2016] 2 HKLRD 1106 (15 April 2016). [66] Court of Appeal §114. [67] Ibid §§104, 106-108, 111-112. [68] Ibid §§165-167, 177. [69] Rogers VP, Le Pichon and Kwan JJA, CACV 88 and 89/2010, [2011] 1 HKLRD 781 (15 December 2010). [70] Court of Appeal §§179-182. [71] Ibid §§190-192. [72] Ibid §§193-195. [73] Ibid §§213-216, 249, 253. [74] Ibid §§217, 220-233. [75] Ibid §262. [76] Ibid §§264-265. [77] Ibid §278. [78] In HCCT 54/2007. [79] Yuen, Barma and Poon JJA, CACV 158/2012, [2018] HKCA 940 and [2018] HKCA 942 (10 December 2018). [80] Cheung PJ, Bokhary and Chan NPJJ, FAMV 6 and 7/2019, [2019] HKCFA 21 (3 June 2019). [81] In what is now listed as FACV 3 of 2019. [82] In what is now listed as FACV 5 of 2019. [83] Appearing for D1-D2 with Mr Benjamin Yu SC, Mr Richard Khaw SC and Ms Bonnie Y K Cheng. [84] D1-D2 Written Case §50. [85] DHCJ Stone §§140-141, 144. [86] Court of Appeal §§165-167, 177. [87] D1-D2 Written Case §50. The construction argument is developed at §§43-53. See also D1-D2 Supplemental Case §§9-12. [88] People’s Republic of China Supreme Court Notice of 24 January 2000. [89] “The Hong Kong Special Administrative Region may, through consultations and in accordance with law, maintain juridical relations with the judicial organs of other parts of the country, and they may render assistance to each other.” [90] Article 1. [91] Article 6. [92] Article 5. [93] Arbitration (Amendment) Ordinance 2000, entering into force on 1 February 2000. [94] Now section 92 of Cap 609. [95] See Order 73, r 10(1)(b). [96] Order 73, rr 10(3)(a)(ii), 3(b) and 3(c). [97] Ordinance section 40E. [98] Walker v Rome [1999] 2 All ER (Comm) 961 at 968. It should, however, be noted, that post-judgment interest stands on a different footing: Gater Assets Ltd v Nak Naftogaz (No 2) [2009] 1 All ER (Comm) 667. [99] Norsk Hydro ASA v State Property Fund of Ukraine [2009] Bus LR 558 at §12, per Gross J. [100] Petrochina International (Hong Kong) Corp Ltd [2011] 4 HKLRD 604 §12. [101] (1927) 28 Ll L R 104 at 106-107. [102] Per Gross J in Norsk Hydro ASA v State Property Fund of Ukraine [2009] Bus LR 558 at §17; cited in PetroChina International (Hong Kong) Corp Ltd [2011] 4 HKLRD 604 at §12. [103] Dalmia Cement Ltd v National Bank of Pakistan [1975] 1 QB 9; Dalmia Dairy Industries v National Bank of Pakistan [1978] 2 Lloyd’s Rep 223. Other illustrations include Selby v Whitbread & Co [1917] 1 KB 736 and Birtley and District Cooperative Society Ltd v Windy Nook and District Industrial Cooperative Society Ltd (No.2) [1960] 2 QB 1. [104] Under section 26 of the Arbitration Act 1950, our equivalent being section 2GG(1). [105] [1975] 1 QB 9 at 23-24. [106] Ibid at 24. [107] [1978] 2 Lloyd’s Rep 223 at 225. [108] Ibid. [109] Ibid at 277 et seq. [110] D1-D2 Written Case§74. Authorities cited in this passage have been omitted. Material decisions are discussed below. [111] The First Award was made without the tribunal or the Plaintiff being aware of the restructuring which occurred between 16 November 2005 and 6 April 2006. [112] [1985] 1 WLR 762 at 772. [113] (1927) 28 Ll L R 104 at 106-107. [114] Article 5. [115] Cap 347. [116] [1985] 1 WLR 762 at 770-771. [117] 1982 Edition, at p 368-369, cited by Otton J in [1985] 1 WLR 762 at 771. [118] [2010] 1 Lloyd’s Rep 222 at §14. [119] [1933] 1 KB 753. [120] Ibid at 764. [121] The current Hong Kong equivalent is Order 11 r 1(e) which is in the same terms. [122] [1973] AC 115. [123] Ibid at 126. [124] [2000] QB 559. [125] Jurisdiction pursuant to Supreme Court Act 1981, section 20(2)(h). [126] [2000] QB 559 at 571. [127] [1993] QB 701. [128] As to which see Heyman v Darwins Ltd [1942] AC 356; and Fiona Trust v Privatov [2008] 1 Lloyd’s Rep 254. [129] [1993] QB 701 at 722. [130] Associated Electric & Gas Insurance Services Ltd v. European Reinsurance Company of Zurich [2003] 1 WLR 1041 at §9. [131] The equivalent for present purposes is section 40B(2) of the Ordinance, set out in Section J.1 above. [132] Appearing for the Plaintiff with Mr Edward Chan SC, Mr Bernard Man SC, Mr Lee Tung Ming, Mr Keith Lam, Mr Justin Ho and Mr James Man. [133] Quintin McGarel Hogg, The Law of Arbitration (Butterworth, 1936) pp 120-122, 159; Mustill and Boyd, The Law and Practice of Commercial Arbitration in England (Butterworths, 2nd Ed, 1989) pp 416-418. [134] [2010] 1 Lloyd’s Rep 222 at §5. [135] Cited in Redfern and Hunter on International Arbitration (OUP, 6th Ed), §7.07. [136] Agromet Motoimport v Maulden Engineering Co (Beds) Ltd [1985] 1 WLR 762 at 770, approving the Mustill and Boyd approach at 772. See also Merkin Op cit at §19.6, citing authorities illustrating the grant of a range of remedies. [137] [1917] 1 KB 736. [138] Ibid at 753-754. [139] [1960] 2 QB 1. This decision was held by Bellshill and Mossend Co-operative Society Ltd v Dalziel Co-operative Society Ltd [1960] AC 832 to be wrong on an unrelated ground. [140] West Tankers Inc v Allianz SpA (“The Front Comor”)[2012] 2 All ER (Comm) 113. [141] African Fertilizers & Chemicals NIG Ltd (Nigeria) v BD Shipsnavo GmbH & Co Reederei KG [2011] 2 CLC 761 at §18. [142] D1-D2 Written Case §2(4). [143] Ibid §2(6). [144] Court of Appeal §190. [145] Section G.1 above. [146] [1980] AC 367. [147] Ibid at 392. [148] Ibid at 394. 
Chief Justice Cheung: 1. I agree with the judgment of Lord Neuberger of Abbotsbury NPJ. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Lord Neuberger of Abbotsbury NPJ. Mr Justice Fok PJ: 3. I agree with the judgment of Lord Neuberger of Abbotsbury NPJ. Mr Justice Lam PJ: 4. I agree with the judgment of Lord Neuberger of Abbotsbury NPJ. Lord Neuberger of Abbotsbury NPJ: A. Introduction 5. The basic issue of principle raised on these appeals is whether a person other than the Secretary for Justice (“the SJ”) who wishes to bring proceedings for criminal contempt of court (an “applicant”) must obtain the consent of the SJ before commencing such proceedings. The Court of Appeal (Cheung JA, Yuen JA and Barma JA) reversing the decision of DHCJ Saunders, although agreeing with him on this point, held that such consent is not required[1]. 6. The appellants, Sun Min (“Sun”), Chang Dafa (“Chang”), and Pacific Bulk Shipping (Cayman) Ltd (“PBSC”) (together “the appellants”), contend that the courts below were wrong, and that such consent is required. More particularly, as Mr Yu SC refined his argument orally, the appellants’ contention was this, that an applicant must inform the SJ of the alleged criminal contempt, whereupon the SJ has the following options, namely (i) to decide to bring contempt proceedings himself, or (ii) to decline to do so and either (a) refuse or (b) grant consent to the applicant to start such proceedings; and it is only if option (ii)(b) applies that the applicant can bring contempt proceedings without joining the SJ. 7. The respondent, Chu Kong (“the respondent”), with the support of the SJ (who was given leave to intervene by Lam PJ on 12th October 2022), argues that the courts below were right, and that, although (as is common ground) an applicant requires the leave of the court to bring criminal contempt proceedings, such proceedings require the consent of no other person. 8. After setting out the factual background, I will consider the nature of contempt proceedings, and will then address the basic issue on these appeals as a matter of principle. I will then turn to consider whether there are any domestic statutory or regulatory provisions, or any decided cases or other material which bear on the issue, following which I will address the case law and other relevant material in other common law jurisdictions. I will then express my conclusions and answer the four questions which are said to be raised on these appeals. B. The factual background 9. For the purpose of this judgment, it is unnecessary to set out the detailed history leading up to these appeals: it is described in paragraphs 8 to 31 of Barma JA’s judgment. The following is therefore a brief, simplified summary. 10. The action underlying the instant contempt application[2] (“the underlying action”) arose from a dispute concerned with the control of the motor vessel “Grain Pearl” (“the Vessel”), which was owned by Joint Silver Limited (“JSL”), a company (i) 50% owned by the respondent and two associates, and (ii) 50% ultimately owned by Lau Wing Yan and his associates, Sun and Chang (“Lau and associates”). 11. The respondent fell out with Lau and associates, and the disagreements between them included a dispute as to what, if anything, had been orally agreed as to how decisions relating to JSL would be made, as to the role of PBSC in managing the Vessel, and as to PBSC’s fee entitlement. Lau and associates contended that the respondent had wrongly (i) appointed himself sole director of JSL, and (ii) caused JSL (a) to terminate PBSC’s Vessel management role, and (b) to issue a lien notice over certain cargo on board the Vessel. 12. These disagreements led to the commencement of the underlying action by Lau and associates and PBSC (“the plaintiffs”) against the respondent, and in the course of those proceedings, the plaintiffs applied ex parte for, and obtained, an injunction against the respondent. 13. In support of their application, the plaintiffs relied, inter alia, on three emails (“the emails”) which were exhibited to Sun’s first affirmation. The emails had been altered by one Yan Donghai (“Yan”) so as to bolster the case of Lau and associates as to PBSC’s role and fee entitlement. The respondent contends that Sun and PBSC were aware of these alterations, whereas Sun, while accepting that the emails were altered by Yan, says that she was unaware of this until January 2016. 14. After making an unsuccessful application to cross-examine Sun and Yan, the respondent applied on 17th August 2016 for leave to commence contempt proceedings (“the Contempt Proceedings”) against the appellants and Yan, based on the alteration of the emails, and also on allegedly false statements in affirmations by Sun and Yan, and an invoice allegedly fabricated by Chang on behalf of PBSC. 15. DHCJ Kent Yee granted the respondent leave to bring the Contempt Proceedings on 20th December 2016, and the appellants then applied to set aside the grant of leave. On 5th July 2017, DHCJ Saunders set aside the grant of leave on the ground of material non-disclosure, based on his view that there had been a failure to explain to DHCJ Kent Yee “the context in which the emails were used”. However, DHCJ Saunders also held that it would have been unnecessary for the respondent to obtain the SJ’s consent to the Contempt Proceedings. 16. On 25th July 2018, the Court of Appeal allowed the respondent’s appeal against DHCJ Saunders’s setting aside of leave, for reasons contained in a judgment given by Barma JA, with which Cheung JA and Yuen JA agreed. The Court of Appeal overruled DHCJ Saunders’s finding of material non-disclosure, but agreed with him that the SJ’s consent to the bringing of the Contempt Proceedings was not needed, although they considered that those proceedings would not be for civil contempt, but for criminal contempt. 17. After the Court of Appeal had refused them leave to appeal on 31st January 2022, the appellants issued notices of motion applying for leave from this Court on 28th February 2022, contending that the projected appeals would raise four “Questions of great general or public importance”. Those Questions were as follows: 1) Does the [SJ] have the exclusive right to bring proceedings against an alleged contemnor in respect of criminal contempt? 2) If the answer to the first question is in the negative, then, if a private litigant seeks to commit another for criminal contempt under Order 52 of the Rules of the High Court (Cap 4A), is he or should he be required (in the absence of exceptional circumstances such as urgency) to consult the [SJ] before he can bring an application for leave to commit under Order 52 rule 2? 3) If the answer to the second question is in the affirmative, and, assuming the [SJ], having been so consulted, declines to bring proceedings in respect of the said criminal contempt, is the private litigant required to join the [SJ] as party, and/or lay the relevant facts before the Court including any expressed views of the [SJ], when bringing proceedings for criminal contempt under Order 52? 4) In the event that the answer(s) to any of the above questions is/are in the affirmative, should the leave granted to Mr Chu to commence committal proceedings against the Defendants herein (by Deputy High Court Judge Kent Yee on 20th December 2016) be set aside and these contempt proceedings be dismissed on the basis that they are procedurally defective, and/or that there has been material non-disclosure in respect of Mr Chu’s failure to consult the [SJ] before commencing these proceedings and/or his failure to join the [SJ] to these proceedings. 18. On 18th July 2022, this Court (Fok PJ, Lam PJ and Tang NPJ) granted the appellants’ applications for leave to appeal on Questions 1 to 4. 19. Meanwhile, to complete the story, the underlying action came on for trial in April 2021, and, on the seventh day of the hearing, the plaintiffs applied for and obtained leave to discontinue the action. C. Civil and criminal contempt of court Contempt of court 20. There is no more important aspect of a civilized society than an effective, independent judiciary whose orders are respected and obeyed. The public interest in the law being properly administered by the courts is plain and well-established. In the 1974 House of Lords case of Attorney-General v Times Newspapers Ltd[3],Lord Cross of Chelsea said that “the due administration of justice is something which all citizens, whether on the left or the right or in the centre, should be anxious to safeguard”. And to much the same effect, some two years earlier, Salmon LJ had said in the English Court of Appeal that “[t]he public at large no less than the individual litigant have an interest and a very real interest in justice being effectively administered”[4]. 21. An essential ingredient of “justice being effectively administered” is the ability of a court to entertain applications for contempt of court (“contempt”) and the power to punish those who commit contempts. This power exists to enable judges to maintain the authority of, and public confidence in, the courts, by preventing and punishing abuse or obstruction of the process of the courts. This power represents a very well-established aspect of the court’s inherent jurisdiction, and it was described by Sir Jack Jacob in an article on “The Inherent Jurisdiction of the Court”[5], as part of “the authority of the judiciary to uphold, to protect and to fulfil the judicial function of administering justice according to law in a regular, orderly and effective manner”. However, it is always salutary for judges to bear in mind that, rather than existing “to protect the dignity of the judges”, the function of the contempt jurisdiction is “to prevent or punish conduct which tends to obstruct, prejudice or abuse the administration of justice” – per Lord Ackner in the 1992 case, Attorney-General v Times Newspapers Ltd[6]. 22. The importance of the contempt jurisdiction was described in ringing terms by Wilmot J in R v Almon[7], where he wrote that it was “a necessary incident to every Court of Justice … to fine and imprison for a contempt to the Court”, and such a power was “intrinsic in a superior court; it is its very life-blood, its very essence, its immanent attribute”.[8] As Lord Simon of Glaisdale explained in the 1974 Times Newspapers case[9], there is “public interest in due administration of justice – that is, in the resolution of disputes … by independent adjudication in courts of law according to an objective code”, as the “alternative is anarchy”. Unless court orders are, and are seen to be, enforced, and those who disobey court orders are, and are seen to be, punished, the independent adjudication in courts to which Lord Simon referred will cease to be, or to be regarded as, effective. For this reason, Lord Steyn, speaking for the Privy Council in Ahnee v DPP[10], an appeal from Mauritius, described the court’s “power and [duty] to enforce its orders and to protect the administration of justice against contempts which are calculated to undermine it” as “an integral part of [the court’s] constitutional function”. 23. In the light of these observations, it is clear that the law on contempt is founded entirely on public policy as was rightly stated by Poon JA giving the judgment of the Court of Appeal in Secretary for Justice v Wong Ho Ming[11], citing Lord Reid in the 1974 Times Newspapers case[12] to the same effect. The two types of contempt 24. The authorities also establish that there are two categories of contempt, civil contempt and criminal contempt. 25. A civil contempt involves the breach of a court order (or of an undertaking to the court), and the sanction for such a contempt is primarily directed towards ensuring that court orders are obeyed (and undertakings to the court are observed). The civil contempt procedure therefore exists ultimately for the benefit of the party in whose favour the order was made or undertaking given. Thus, a civil contempt can therefore, at least normally, be effectively waived by the person for whose benefit the order was made (or undertaking was given) [13]. Further, as Sir Gerard Brennan NPJ explained in Kao, Lee & Yip v Koo Hoi Yan[14],a civil contempt can be committed without the contemnor having the intention of breaching a court order or otherwise interfering with the due administration of justice. 26. A criminal contempt, by contrast, is one which “so threatens the administration of justice that it requires punishment from the public point of view” - per Arlidge, Eady & Smith on Contempt (“Arlidge”)[15].Accordingly, a criminal contempt engages a much wider public interest, and it cannot be waived by anyone. As the Court of Appeal (Cheung CJHC, Lam V-P, and Macrae JA) said in Secretary for Justice v Cheung Kai Yin (No 2)[16], a criminal contempt has been accurately described as being “conduct which goes beyond mere non-compliance with a court order or undertaking and involves a serious interference with the administration of justice”[17].And, before a person can be liable for criminal contempt, it must be established that he intended to interfere with or impede the due administration of justice. 27. The difference between the two types of contempt was described in the recent UK Supreme Court judgment in Attorney General v Crosland[18], which distinguished between “a case involving a breach of an order by a party to litigation where the order has been made at the instance of an opposing party and its purpose is simply to protect the private rights of that other party”, and a case involving breach of an order made “to protect the administration of justice and its breach involves a general interference from which the administration of justice must be safeguarded”. 28. In practice, the distinction between criminal and civil contempts is rather less clear than may appear from the above discussion. Thus, as Poon JA said in Secretary for Justice v Wong Ho Ming[19], it would be “inappropriate to regard an order for committal in civil contempt as no more than a form of execution. For the court still has a very substantial interest in seeing that its orders are upheld”. And it appears that, even where a civil contempt is “waived” by the party with the benefit of the order or undertaking, the court still has jurisdiction to take action of its own motion in respect of the contempt in exceptional circumstances[20]. 29. As Arlidge rightly records[21], “the two types of contempt … overlap to a considerable degree”, a view supported by the Court of Appeal in Cheung Kai Yin (No 2)[22]. Indeed, the distinction between the civil contempt and criminal contempt has been seen by a number of judges as being unsatisfactory. Thus, in the 1988 case of Attorney-General v Newspaper Publishing Plc, Lloyd LJ said that the distinction led to differences which represented “the sort of nonsense which does no credit to the law”[23], and Sir John Donaldson MR suggested that the distinction “now tends to mislead”[24]. And in the 1992 Times Newspapers case[25], Lord Oliver of Aylmerton mentioned that the distinction had been “variously described as ‘unhelpful’ or ‘largely meaningless’”. Further, as pointed out by Hayne J in Re Colina ex p Torney[26]¸ “four members of the [Australian High] court [in Witham v Holloway[27]] concluded that the distinction between civil and criminal contempt was based upon differences that are ‘in significant respects, illusory’”.It is also worth mentioning that the Phillimore Committee[28] doubted the value of the distinctions between the two types of contempt and recommended the abolition of those distinctions. However, the distinction is well-established, and it has (understandably) not been challenged on these appeals. Criminal contempt proceedings 30. Although criminal contempt constitutes a common law misdemeanour, it does not fall within the scope of the ordinary criminal law. The point was fully considered by the Court of Appeal in Secretary for Justice v Cheung Kai Yin (“Cheung Kai Yin (No 1)”)[29].Relying on a number of Hong Kong, UK, and Australian cases, Lam V-P (as he then was) concluded that: “Notwithstanding that the proceedings are ‘prosecutions’ for criminal contempt, we have come to the conclusion that these appeals are properly characterised as civil appeals. In other words it is correct to have the appeals intituled as CACVs and they are civil causes or matters”[30] As he went on to say: “[T]he jurisdiction of the court to punish contempt is sui generis and in many respects principles applicable to criminal cases are applicable to contempt proceedings (both proceedings for civil as well as criminal contempt)”[31] 31. The fact that criminal contempt proceedings are civil in nature is also illustrated by the following observation of the Court of Appeal in Cheung Kai Yin (No 2)[32]: “Criminal contempt proceedings brought under Order 52 are the means for the court to ensure that conduct or activities calculated to undermine the administration of justice are properly addressed. In terms of the time limit for commencement of proceedings, these considerations are not present in a criminal prosecution of an offence under s.23 of the SOO or other offences in the Magistrates’ Courts” The Court of Appeal went on to say[33] that they did not accept, “for the same reasons, that the sentencing court on a committal for contempt is confined or constrained by the sentence which would be appropriate to the statutory offence”. 32. Among a number of authorities which Lam V-P cited in Cheung Kai Yin (No 1) in support of his conclusion was a decision of the High Court of Australia in a 2015 case[34], in which French CJ, Kiefel, Bell, Gageler and Keane JJ said this: “In Witham v Holloway[35], the plurality expressly noted that the process whereby a contempt proceeding is resolved is a civil ‘hearing’ not a criminal ‘trial’. McHugh J also expressed the view that proceedings for contempt of court to punish a respondent are ‘civil and not criminal proceedings’.” Equally clearly, in a concurring judgment, Nettle J expressed himself as follows[36]: “A proceeding for punishment for contempt constituted by disobedience of an injunction granted in a civil proceeding is not part of the criminal justice system in the [normal sense]. Although ‘all proceedings for contempt “must [now] realistically be seen as criminal in nature”’, not all contempts are criminal. Failure to obey an injunction is not a criminal offence unless the failure to comply is defiant or contumacious. A proceeding for contempt is not a proceeding for criminal contempt if the proceeding appears clearly to be remedial or coercive in nature as opposed to punitive. A criminal contempt is a common law offence, albeit not part of the ordinary common law. But even a proceeding for criminal contempt is not a criminal proceeding” 33. The fact that criminal contempt proceedings are not criminal proceedings is underlined by the very well-established fact that a criminal contemnor is liable to be dealt with both in criminal proceedings for the crime involved in the criminal contempt, as well as in civil proceedings for the contempt involved in the criminal contempt. This leads to potential tensions in a case where there are two sets of proceedings, and two English Court of Appeal judgments on the topic are usefully discussed in Cheung Kai Yin (No 2)[37]. However, the essential point for present purposes is that criminal contempt proceedings cannot be criminal proceedings, as otherwise it would be impossible to justify both criminal proceedings in the criminal courts and criminal contempt proceedings in the civil courts being conducted against the same party in respect of the same action or conduct. 34. Thus, the only appropriate way to bring an alleged contempt before the court is through civil proceedings. Prosecuting contempts by way of indictment has become obsolete, as Hartmann and Suffiad JJ explained in Secretary for Justice v Choy Bing Wing[38]. The same is true in England: as Lord Judge LCJ explained in Attorney General v Dallas[39]: “It should now be clearly understood that trials for contempt of court on indictment are obsolete, and of historical interest only. Neither the Attorney General, nor the alleged contemnor, has any right to seek trial by jury on indictment” 35. It therefore seems clear that criminal contempt proceedings, while sui generis, are ultimately civil in nature, and therefore, as Arlidge points out[40], the initiation of contempt proceedings should not be described as a “prosecution”. However, a party bringing an alleged contempt before the court is often referred to as a “prosecutor” (and it is right to bear in mind that civil proceedings can be correctly described as being “prosecuted”). Whatever descriptive noun or adjective one adopts to describe criminal contempt proceedings, the basic function of a party bringing such proceedings is to draw the attention of the court to the facts and matters which are said to give rise to the contempt, leaving it to the court to decide whether there was a criminal contempt, and, if so, what is the appropriate sanction. 36. Thus, in the 1974 Times Newspapers case, Lord Cross said that the role of the Attorney General (“the AG”) in English contempt applications was similar to that of an “amicus curiae” who brings “to the notice of the court some matter of which he considers that the court shall be informed in the interests of the administration of justice”[41]. It is right to mention that Lord Diplock (with whom Lord Morris agreed) put it slightly differently when he described the AG’s role[42] as “[acting] on behalf of the Crown as ‘the fountain of justice’ and not in the exercise of its executive functions”, adding that this involved “the exercise of his personal judgment” as to whether “the conduct … falls into the category of contempts which the court would regard as deserving of some punishment”. 37. While it is open (and indeed normal) for criminal contempt proceedings to be initiated by a prosecutor, commonly an individual interested in the proceedings or the responsible member of the executive (in Hong Kong the SJ, in England the AG), it is very well established that the court can punish a person for criminal contempt of its own motion – see for instance per Rigby LJ in Seaward v Paterson[43], per Lord Denning MR in Balogh v St Albans Crown Court[44] (a most unusual case), and per Mustill LJ in R v Griffin[45]. The most obvious examples of such cases would be contempts in the face of the court or actions such as impeding or attacking officers of the court. 38. Having considered the background, it is now appropriate to address the central issue in this case, namely whether criminal contempt proceedings can only be brought by, or with the consent of, the SJ, or whether they can be brought simply with the leave from the court. It seems appropriate to consider this question as a matter of principle, and then to turn to the case law and other relevant material in this and other common law jurisdictions. D. The central issue as a matter of principle 39. Given that the contempt jurisdiction (i) goes to the heart of the court’s function and role, (ii) is an aspect of the court’s inherent jurisdiction, and (iii) has been authoritatively described as “constitutional” in nature, it would be surprising if any entity other than the court itself was able to fetter the ability of any person to bring an alleged contempt, whether civil or criminal, to the attention of the court. 40. The SJ obviously has a vital and fundamental role in supporting the rule of law, and therefore in supporting the role and authority of the courts. However, it seems highly questionable whether he, as a member of the executive, should have a power which enables him to prevent the court from hearing a criminal contempt application, and therefore from upholding its own authority. The notion that the SJ could stop a criminal contempt being brought before the court thus seems wrong in principle. 41. Even more surprising is the notion, which was advanced as part of the appellants’ case, that the SJ can prevent a court from initiating and prosecuting criminal contempt proceedings of its own motion, even when the court considers it right to do so. That is, to put it politely, a very ambitious submission, which runs wholly counter to the important constitutional principle of separation of functions. Even if this extreme aspect of the appellants’ case is ignored, however, it still seems inappropriate that the SJ should be able to prevent an applicant from drawing the court’s attention to an alleged criminal contempt, as I have just explained. 42. This view is in no way undermined by the fact that the SJ can initiate or take over the conduct of criminal contempt proceedings, and that he may play an important part in connection with some contempt applications – e.g. by intervening in any criminal contempt proceedings, or as a result of the court referring to him matters arising in connection with criminal contempt proceedings. In bringing criminal contempt proceedings the SJ is acting like any other citizen – although, obviously, with greater authority, and taking over criminal proceedings is an ancillary aspect of that function. And in playing a part in contempt proceedings brought by another, the SJ is assisting the court (as on these appeals), but that in no way supports the notion that he should be able to prevent an alleged contempt being put before the court. 43. It is fair to say that there is some initial attraction in the notion of the SJ being able to weed out misconceived or vengeful contempt applications before the court is troubled by them. However, the court, which will often have had the conduct of the underlying action, and which would be determining the alleged contempt if the application succeeds, would normally be in at least as good a position as the SJ to decide whether to let an application proceed, and the court indeed has that role as its leave is needed to bring contempt proceedings. And, as already mentioned, if the court considers that it would benefit from the assistance of the SJ (including at the leave stage), it can request it. 44. The notion that criminal contempt applications require the consent of the SJ also appears to sit a little oddly with the fact that the Court can initiate criminal contempt proceedings of its own motion. Judges generally have no power to initiate civil, or indeed criminal, proceedings themselves, and the reason why they have this unusual power when it comes to criminal contempt is because of the nature and purpose of the contempt jurisdiction as discussed above. This point reinforces the notion that the only gatekeeper controlling access to the court to a party seeking to bring criminal contempt proceedings should be the court itself. 45. Further, the fact that (unless of course there are constitutional or legislative rules which provide otherwise) civil contempt proceedings can freely be brought to court without the consent of the SJ or any other third party also supports the proposition that the same should be true for criminal contempt proceedings. Given that the contempt jurisdiction is based on the court’s duty to protect its own efficacy and authority, it would be strange if the court was free of any third party’s constraint to entertain civil contempt proceedings, but was more restricted when it came to entertaining the more serious criminal contempt proceedings. It would perhaps be more understandable if criminal contempt proceedings were truly criminal in nature, as the SJ has a central role when it comes to bringing criminal proceedings, but as explained, criminal contempt proceedings are ultimately civil in nature. 46. The appellants justify a restriction on criminal, as opposed to civil, contempt proceedings on the ground that civil contempt proceedings are (at least normally) intended to be for the benefit the prosecutor, who should therefore be free to issue civil contempt proceedings, whereas the criminal contempt jurisdiction exists for the public benefit, and so it is appropriate that a prosecutor needs the consent of the SJ. However, that argument suffers from the problem that the bringing of criminal contempt proceedings does need consent in the form of leave from the court. (The fact that civil contempt proceedings may also need such leave does not undermine the force of the point). Quite apart from this, the somewhat unsatisfactory nature of the distinction between civil and criminal contempts discussed above[46] would render it difficult in some cases to decide whether the SJ’s consent was needed, and could lead to arid arguments. 47. In addition to these points, the fact that it is very well established that all members of the public have an interest in the proper administration of justice tends to support the contention that there should be no fetter imposable by any entity other than the court itself on a member of the public seeking to bring criminal contempt proceedings. Seeking the leave of the court can involve no or little loss of time or extra effort, as an application for leave can no doubt be made “ex parte on notice”, so that, if the court gives permission, it can give directions immediately thereafter. If consent has to be sought from the SJ by contrast, it would involve extra delay and expense. 48. Further, given that contempt proceedings require the leave of the court, it would seem somewhat surprising if the consent of a third party was also required. It is not easy to see why there should be two gatekeepers as a matter of course in relation to every potential criminal contempt proceedings. In an appropriate case, the SJ could make representations at the leave stage, and, if he did not do so, then, as already mentioned, the court, if it thought fit to do so, could no doubt consult the SJ, or ask if he had any submission to make. 49. Quite apart from these points, as well as being inappropriate from the perspective of the court and an applicant, it would represent an unnecessary burden on the SJ if he had to vet every proposed contempt application. In that connection, it is relevant to invoke what, when considering the role of the SJ in charity proceedings, Lam V-P (as he then was) referred to as “[t]he stark reality” that “under our present system, those in the Department of Justice have no resource or power to carry out an investigation” [47]. Indeed, what was effectively that very point was made in relation to the issue on these appeals by Mr Dawes SC on behalf of the SJ. 50. As the arguments developed, the appellants put forward as their “core proposition”, the point that “there is an inherent conflict between the conduct of criminal contempt proceedings which are public law proceedings concerned solely with the administration of justice and public interest, and the notion that a private party with his own agenda and motive can bring such proceedings”. The answer is that there is no such conflict as a private party can only bring criminal contempt proceedings with the leave of the court, and all judges are well aware of the need to scrutinise applications to bring such proceedings by a litigant who may want to put pressure on an alleged contemnor for his own commercial benefit or to get revenge on the alleged contemnor[48]. The appellants contend that this is a weaker protection than would be accorded to the accused in a criminal court. Even if that is right, it is well established that many of the protections accorded in criminal cases are not available in contempt proceedings[49]. However, an alleged contemnor is not without substantial protection: as Mr Mok SC submitted, on behalf of the respondent, quite apart from being able to contest criminal contempt proceedings, and to rely on the criminal burden of proof[50], an alleged contemnor can apply to have an ex parte grant of leave set aside if there was abuse[51], and at any stage, the court may of its own motion, or at the instigation of a party, seek assistance from the SJ. 51. Analogies are always dangerous, and that must be particularly true in relation to criminal contempt proceedings which have been consistently described as sui generis[52]. However, it is not uninstructive to consider privately brought criminal proceedings. The appellants’ submission that the consent of the SJ is required before a private person can bring criminal contempt proceedings is hard to reconcile with the fact that a private person is free to bring a private prosecution, a proposition recently confirmed in clear terms by Poon CJHC giving the judgment of the Court of Appeal in Kwok Tak Ying v HKSAR[53]. Whether a criminal prosecution may be proceeded with is a matter ultimately for the SJ (pursuant to Article 63 of the Basic Law, as referred to below), whereas whether criminal contempt applications can be brought is ultimately a matter for the court. 52. An analogy which seems to me to be misleading for present purposes is that of a relator action, which was pressed by Mr Yu SC. He relied on the reasoning of the House of Lords in Gouriet v Union of Post Office Workers[54], where it was held that in England and Wales only the AG could bring proceedings to restrain the commission of an offence (unless the action or inaction complained of also represented a breach of a private party’s rights). A relator action is an action on behalf of the public, and the law clothes the AG with the authority to decide whether to bring, and actually to bring, proceedings on behalf of the public, whereas a private citizen, however well intentioned, can have no such authority. By contrast, all that a private citizen is doing by initiating criminal contempt proceedings is drawing an alleged contempt to the attention of the court, whereupon it is for the court to decide whether to allow him to proceed, and, if it does allow him to do so, to decide how matters should proceed. In any event, if the logic of Gouriet applies in relation to criminal contempts, it would mean that a party to the underlying action could not bring contempt proceedings, and that is plainly not the law. 53. The appellants also contended that, if criminal contempt proceedings could be brought without the consent of the SJ, then the system would be open to abuse. To the extent that that is right, it is equally true of civil contempt proceedings, and indeed any other proceedings. More importantly, if a filter is required to consider whether contempt applications should be allowed to proceed, it is more appropriate that it is the court itself rather than the SJ, which applies the filter. As already explained, (i) contempt is very much a matter for the court and not for the executive, (ii) in many cases, the court will have a much better idea of the background, as there will almost always be an underlying action, and (iii) all citizens have an interest in courts and court orders being respected, and there should be as few as possible barriers in the way of bringing alleged contempts to the attention of the court. 54. Accordingly, it appears to me that, if these appeals are to succeed, the appellants would have either (i) to identify a legislative provision which imposes a requirement that an individual seeking to bring criminal contempt proceedings must first obtain the consent of the SJ, or else (ii) to establish that the case law has developed in such a way as to demonstrate that the courts themselves have recognised that such a requirement should be recognised or imposed. 55. It is right to say that the point made in paragraph 54 above would be appropriate even in the absence of the discussion in the preceding fifteen paragraphs, on the basis that, once it is accepted that any person can bring criminal contempt proceedings with the leave of the court, it is for the party asserting that there is a further fetter on the bringing of such proceedings to show that such a fetter exists. However, the preceding discussion underscores the force of that point in this particular case, and suggests that any alleged statutory or judicial fetter should be considered with particular scrutiny. 56. Accordingly, I turn first to consider whether there is any domestic legislative provision or case law which mandates or indicates that the SJ’s consent is required before a private person can initiate criminal contempt proceedings. E. Domestic constitutional, legislative, and other material Rules of the High Court 57. As one would expect, the procedure to be adopted on contempt applications is dealt with in The Rules of the High Court (Cap 4A) (“The Rules”). Many of The Rules which apply to applications generally apply to contempt applications, but it has not been suggested that there is anything in those generally applicable rules which assist for present purposes. 58. However, and again as one would expect, The Rules include an Order, namely Order 52 (“O.52”), which deals specifically with applications for committal for contempt. It is headed “Committal”, and is based very closely on the English Order 52 of the Rules of the Supreme Court (“RSC Order 52”). 59. O.52 rule 1 provides that “a single Judge or … a single justice of appeal” has the power “to punish for contempt of court … by an order of committal”. O.52 rule 2(1) states that “[n]o application for an order of committal” can be made “unless leave to make such an application has been granted in accordance with this rule”. Rule 2 goes on to provide that such an application should initially be “made ex parte to a judge”, and should comply with certain other formalities, and that the judge may dispose of the application without a hearing. O.52 rule 3 deals with the procedure to be adopted by the applicant after an application is granted under rule 2, and O.52 rule 6 is concerned with the hearing of the contempt application. There is no rule 4, but O.52 rule 5 makes it clear that nothing in O.52 affects the court’s power “to make an order of committal of its own motion”. 60. It is to be noted that, in the light of its title and the provisions of rules 1 and 2, O.52 appears to be limited to contempt applications seeking committal. However, O.52 rule 9 provides: “Nothing in the foregoing provisions of this Order shall be taken as affecting the power of the Court to make an order requiring a person guilty of contempt of court … to pay a fine or to give security for his good behaviour, and those provisions, so far as applicable, and with the necessary modifications, shall apply in relation to an application for such an order as they apply in relation to an application for an order of committal.” Accordingly, it seems clear that, if a fine or some form of security, rather than committal, is the sanction sought in a contempt application, the court’s leave would still be required under O.52 rule 2. And if sequestration is the sanction sought, then the court’s leave would be required under Order 45 rule 5(i)(b)(i) and (ii) of The Rules. 61. It is also to be noted that O.52 applies to civil as well as criminal contempt applications, as Hartmann and Suffiad JJ observed in Choy Bing Wing[55]. Committal is almost always sought in criminal contempt applications, and, where it is not sought, the application would presumably seek a fine, security or sequestration, and therefore would require the court’s leave. Accordingly, it seems clear that the respondent’s contention (not challenged by the appellants) that criminal contempt applications would always need leave from the court before being commenced, is well founded. 62. Committal is often, and in my experience usually, sought in civil contempt applications, although, as recorded in Arlidge[56], while “[i]mprisonment has always been a sanction in cases of contempt”, it is regarded by the courts today “as a matter of last resort, especially in cases of civil contempt”. In civil contempt applications where committal is not sought, one would expect some other sanction, such as a fine, security or sequestration to be sought, in which case, as just explained, the court’s leave would be required. However, given the purpose of civil contempt proceedings, it is possible to envisage the making of an application which sought a different remedy, at least initially – e.g. simply requiring the alleged contemnor to appear before the court. Thus, while the leave requirement under O.52 will apply to the great majority of civil contempt applications, it may be that it would not necessarily apply to them all. 63. The way in which O.52 is drafted has avoided parties or the court having to delve into the sometimes difficult question whether the alleged contempt is civil or criminal. Rather, the sole question is whether the application seeks the imprisonment, the imposition of a fine, the provision of security or sequestration: if it does, then whether the alleged contempt is criminal or civil, the court’s leave must be obtained. If it does not, then it would seem that leave would not be needed, unless it was required under another rule of court, or another legislative provision. 64. Clearly, there is not even a hint or suggestion in O.52 that an applicant seeking to commit someone for contempt needs to apply to the SJ, or indeed that the SJ has any gatekeeping function in respect of contempt applications. And, particularly given that O.52 expressly provides for the obtaining of leave from the court, it is a reasonable inference that, unless there is any other legally enforceable rule which clearly so provides, there is no requirement for the consent of the SJ before any contempt proceedings, whether criminal or civil, are commenced. 65. The appellants also drew attention to Order 41A rule 9 of The Rules (“O.41A rule 9”), paragraph (1) of which permits proceedings for contempt against a person who makes or causes to be made “a false statement in a document verified by a statement of truth without an honest belief in its truth”. Paragraph (2) of O.41A rule 9 provides that: “(2) Proceedings under this rule may be brought only – (a) by the Secretary for Justice or a person aggrieved by the false statement; and (b) with the leave of the Court.” 66. It is common ground that O.41A rule 9 does not apply in this case, and I do not see how it helps the appellants. Indeed, I consider that, if anything, it is actually unhelpful to their case. O.41A rule 9 is a provision which deals with contempt applications, and which not only specifically requires in sub-paragraph (b) an applicant to obtain leave of the court while saying nothing about the SJ’s consent, but it also has the SJ well in mind as shown by sub-paragraph (a). Constitutional provisions 67. Casting their sights more widely, the appellants relied on Article 63 in Section 2 of Chapter IV of the Basic Law (“Article 63”) which states that “[t]he Department of Justice … shall control criminal prosecutions, free from any interference”. Although the appellants are right in arguing that the cases show[57] that this important provision should be given a purposive, rather than a narrow, effect, it seems to me clear that the Court of Appeal was right to reject their contention that Article 63 extends to applications for criminal contempt. The constitutional basis of the criminal contempt jurisdiction as explained above (protecting the efficacy and authority of the court) is different from that of the criminal jurisdiction, as is perhaps most graphically illustrated by the court’s undoubted power to initiate and prosecute contempt proceedings of its own motion, but is also clear from the differences in procedure between criminal contempt proceedings and criminal proceedings in the criminal courts. 68. It was the appellants’ reliance on Article 63 which led Mr Yu SC to argue, as he logically had to, that the SJ could prevent the court from considering an alleged contempt, whether of its own motion or otherwise, even if the court was of the view that it should do so. I mentioned earlier that any legislative or other provision which was said to cut down the court’s freedom to consider any alleged contempt should be considered with “particular scrutiny”[58]. I would go further in relation to this particular argument: very clear and unambiguous words would be required before it would be appropriate to conclude that a legislative provision, or even a provision in the Basic Law, had the effect of enabling the SJ to prevent the court from considering an alleged contempt which the court wished to consider. The wording of Article 63 falls a very long way short of satisfying that requirement. To put the point the other way round, as observed by Lam V-P in Cheung Kai Yin (No 1)[59], the reason that the requirement for leave of the court in O.52 is not incompatible with the Basic Law is that Article 63 does not apply to the proceedings covered by O.52. 69. As was pointed out on behalf of the SJ, this conclusion is supported by the fact that the Basic Law, as explained by Ribeiro PJ[60], “aims to provide for continuity between the pre-existing and the present courts and judicial systems”. And the role of the AG, the SJ’s effective predecessor before the Basic Law, was discussed in paras 3 and 4 of Appendix 1 to an authoritative 1987 Final Report[61] which led to the drafting of some provisions of the Basic Law, including Article 63. In paragraph 3, the Report referred to the “overall authority for the initiation of criminal proceedings” being vested in the AG, and the fact that “[h]is specific consent is required under a number of statutes”. After summarising his role in relator actions, paragraph 4 stated (reflecting the views expressed in the House of Lords in the 1974 Times Newspapers case[62]): “The [AG] also has a more general public interest role as amicus curiae, the most important example of which is the bringing of alleged contempts to the notice of the courts” The reference to “criminal proceedings” in paragraph 3 of the 1987 Final Report was thus clearly not envisaged as extending to criminal contempt proceedings, and it is significant “criminal proceedings” is the relevant expression in Article 63. 70. Of course, some changes would have had to be made as a result of the introduction of the Basic Law, but there appears to be no conceivable reason why it would have been thought to be appropriate to transfer control over criminal contempt proceedings to the Department of Justice, and there is no trace of any suggestion that such a transfer was contemplated, let alone intended. 71. It is right to mention that, in relation to this argument, the appellants placed some reliance on the drafting history of the Basic Law. I mean no disrespect to them by not addressing the argument other than to say that there is nothing in that history which casts doubt on the views expressed in the preceding four paragraphs. Accordingly, I conclude that Article 63 does not apply to criminal contempt proceedings. Legislative provisions 72. The appellants also sought to draw some support for their argument from section 3(3) of the Judicial Proceedings (Regulation of Reports) Ordinance (Cap 287) (“JPRRO”). Subject to certain specified exceptions, section 3(1)(b) of JPRRO forbids the publication of information “in relation to any judicial proceedings for dissolution of marriage, for nullity of marriage or for judicial separation”, and section 3(2) states that anyone who “contravenes the provisions of [section 3(1)] shall be guilty of an offence”. Section 3(3) provides: “A prosecution under this section shall not be instituted except by or with the consent of the Secretary for Justice” 73. The appellants argue that a breach of section 3(1) of the JPRRO is a species of contempt, and that section 3(3) shows that contempt applications need the consent of the SJ. I disagree. A breach of section 3(1) of the JPRRO is a breach of an Ordinance, which is rendered a crime by section 3(2), and it is not a contempt of court – unless, of course, the publication concerned also breaches a court order. However, in agreement with the respondent’s argument, I would go further and say that section 3(3) is actually unhelpful to the appellants’ case, as it shows that, where the SJ’s consent is required, it is actually spelled out in the relevant legislation; in other words, the contrast between section 3(3) of JPRRO and O.52, so far as the need for the consent of SJ is concerned, reinforces the notion that such consent is not needed for an application to which O.52 applies. Case law and other domestic material 74. So far as other relevant domestic material is concerned, it is noteworthy that in this case the Court of Appeal considered that there was no domestic case which assisted the appellants’ argument, and that, treating the issues as one of principle, they held that[63]: “[T]here [is] no good or compelling reason why it should not be open to a litigant in civil proceedings in which the opposing party has committed a criminal contempt … to pursue the matter by way of criminal contempt proceedings. … The existing procedure under [O.52], with a requirement that leave be obtained, provides a suitable and time-tested process for the vetting of intended claims for contempt, such that obviously unsustainable or abusive proceedings will be weeded out at an early stage.” 75. The appellants point out that in Choy Bing Wing,Hartmann and Suffiad JJ said[64] that a “criminal contempt is a matter for the [SJ] to raise, acting as the guardian of the public interest”. But this statement is not expressed in exclusive terms; and it cannot have been so meant, as it would exclude applications for criminal contempt being made by any party other than the SJ, and, as the appellants rightly accept, private parties can bring contempt proceedings: the question at issue is whether they need the SJ’s consent. Indeed, a little later in the same judgment[65] the court said that “an applicant (be it an aggrieved litigant or the [SJ]) may seek committal for contempt … whether the contempt is … civil or criminal”. Although that statement could just about be said to be neutral on the question at issue, it seems to me that it implies that no consent from the SJ is needed, as “an aggrieved litigant” and the SJ are referred to as if they are in the same position, and anyway if the SJ’s consent was required, one would have expected the judgment to have said so. 76. So far as other domestic material is concerned, in its 1986 Report on Contempt of Court, the Hong Kong Law Reform Commission stated[66] that “in general, contempt proceedings … may be instituted by a private individual”. When one reads those words together with the recommendation[67] that “it should be a requirement before proceedings alleging a contempt other than one arising out of a breach of an order of the court are begun, that notice of the intention to proceed be given to the [AG]”, it is clear that the members of the Commission (who included the then-Chief Justice and the then-AG) considered, indeed took it for granted, that the consent of the AG to the bringing of criminal contempt proceedings was not required. And, as explained above[68], the Basic Law was intended generally to continue the previous law. F. Case Law and other material from other jurisdictions 77. All three parties cited a number of judicial observations in cases decided in England, Scotland, Australia, Ireland, Uganda, and Singapore to support their respective positions on the central issue on these appeals, and they have also cited other material. In a case such as this where a specific point of practice is raised, care must be taken before one gives weight to a statement in a judgment which appears to deal with the point, but which was given in a case where the point was not in fact in issue, and may well not have been the subject of much, if any, argument or citation of authority. In some cases, the statement may be intended to be no more than a very short summary of the relevant law, with irrelevant over-simplification, and in other cases the statement may simply be wrong. Of course, it is very different in a case where the point was fully argued, or it was in issue, or where the statement is consistent with statements in other judgments. 78. Turning to the English cases, the question whether criminal contempt proceedings could only be brought by the AG was considered by the Law Lords in the 1974 Times Newspapers case[69]. Although the appellants suggested that some of the speeches indicated the criminal contempt proceedings could not be brought freely by a private party, it seems to me that all five members of the panel accepted that this was the case. 79. After saying that the AG “has a right to bring before the court any matter which he thinks may amount to a contempt”, Lord Reid said that the “party aggrieved has the right to bring before the court any matter which he alleges amounts to a contempt but he has no duty to do so”[70]. While he thought that “[i]n the situation which presented itself … it was in accordance with desirable practice for the [AG] to be concerned and to consider what course in the public interest he should follow”, and whether to bring contempt proceedings, Lord Morris of Borth-y-Gest added “though it would have been open to [a party in the underlying action on whom pressure was being put by the alleged contemnor] to initiate proceedings had they so decided”[71]. Lord Diplock considered that the AG was “the appropriate public officer to represent the public interest in the administration of justice”[72], he said in terms that if the AG “declines to move, the party complaining can bring the motion on his own behalf”, and, in the immediately preceding paragraph he referred to the court being “vigilant to see that the procedure for committal is not lightly invoked” partly because “it is a procedure which if instituted by one of the parties to litigation is open to abuse”[73]. Lord Simon did not expressly address the point, but he began his speech by saying that he “admiringly agree[d]” with Lord Diplock[74]. Lord Cross said[75] that, save in exceptional cases (such as those involving urgency) it was “most desirable” that a private party should first give the AG the opportunity of applying, thereby strongly implying that a private party could proceed without first doing so. 80. It is fair to say that all these observations were obiter, but they are five consistent views from the United Kingdom’s highest court, and they all support the contention that, in England, the consent of the AG need not be sought by a private person before issuing criminal contempt proceedings. 81. The appellants point out that in an unreported Court of Appeal case in 2000, Pelling v Hammond[76] Laws LJ (sitting on his own) said that it was “certainly desirable and perhaps necessary for the [AG] to be approached to see whether he should bring contempt proceedings”. Given the word “perhaps”, the observation can be of no assistance to the appellants. But, quite apart from this, the observation was made in the context of an application for permission to appeal which Laws LJ dismissed as being “totally without merit”. Accordingly, this is a classic case of an obiter observation which, despite the eminence of the judicial source, cannot be given any weight, because the point was not in issue and apparently not argued, and the judge was not referred to any relevant authority. 82. Of much more relevance is the 2009 case of KJM Superbikes Ltd v Hinton[77], in which the Court of Appeal reversed the refusal of the trial judge to grant leave to the claimant to issue criminal contempt proceedings[78] against the defendant under CPR 32.14 (the English equivalent of O.41A rule 9), on the ground that the defendant had provided a dishonest witness statement. Having held that the claimant had locus to bring the proceedings, Moore-Bick LJ (with whom Mummery and Arden LJJ agreed) went on to address a submission very close indeed to that raised by the appellants in this case, namely[79]: “[the] submission that the court should generally decline to give permission to a private litigant to pursue proceedings for contempt in cases of this kind and should instead direct that the matter be referred to the [AG] for her to consider whether proceedings should be instituted. Such a course was said to be more likely to promote consistency of approach and thus greater confidence in the administration of justice.” While accepting that consistency of approach was “highly desirable” Moore-Bick LJ said[80]: “I do not think that to refer all cases of this kind to the [AG] is the only way in which it can be achieved. Cases are bound to differ widely, … and whether the matter is determined by the [AG] or the court, each will have to be considered on its own facts. In practice complete consistency is unlikely to be unattainable, but it is possible for the courts through individual decisions to establish and develop a body of principles which will provide guidance to judges who have to deal with applications of this kind and which will by their nature promote the necessary degree of consistency. …The court is free to take whichever course appears most appropriate in the circumstances and I can see no good reason for saying that the most appropriate course is normally to direct that the matter be referred to the [AG].” 83. The appellants point out that CPR 32.14 is a separate code from the normal English contempt rules (originally RSC Order 52 and now CPR rule 81). I do not accept that as a relevant distinction for present purposes; CPR 32.14, like O.43A rule 9, refers to “[p]roceedings for contempt of court”, and therefore the usual principles applicable to contempt applications apply, and in any event Moore-Bick LJ’s judgment plainly proceeded on that basis. 84. As to the English rules of court, since October 2020, RSC Order 52 – on which O.52 is based, as mentioned above – has been replaced by CPR Part 81, which permits civil contempt proceedings to be brought as of right, while criminal contempt proceedings require the permission of the court.[81] Like RSC Order 52, CPR Part 81 contains no suggestion that criminal contempt proceedings brought by a private person require the consent of the AG. 85. The Phillimore Report[82] stated that “[i]n general, contempt proceedings, like most other proceedings, civil or criminal, may be instituted by a private individual”. When read together the recommendation that “the normal practice should be … that the attention of the [AG] should be drawn to the matter before any private proceedings are begun”[83], it seems clear that the Committee believed that the AG’s consent need not be sought before criminal contempt proceedings are instituted. 86. Turning to English textbooks on the issue, Borrie & Lowe: The Law of Contempt[84] states that “the consent of the [AG] is not required for the institution of contempt proceedings at common law (although the usual practice is for the complaining party to refer the matter and for the [AG] to initiate contempt proceedings)”. As the authors go on to point out, section 7 of the Contempt of Court Act 1981 specifically requires a party to obtain the AG’s consent to “[p]roceedings for a contempt of court under the strict liability rule”. It is hard to make much of that provision in this case with any confidence as its ambit is unclear, as the discussion in Borrie & Lowe explains. However, the fact that the section specifically requires the consent of the AG in the types of contempt applications to which it applies, strongly implies that such consent is not needed for other types of contempt applications (unless they are subject to some other similar legislative provision). 87. Arlidge is somewhat more elliptical on the point. It describes “institution of proceedings in relation to criminal contempts” as being “primarily a matter for the [AG], or for the court of its own motion”[85]: this suggests that others have the right too, which presumably would include an aggrieved party in the underlying litigation. That the authors consider that such a person could bring proceedings without first informing the AG is I think confirmed by a later passage which states that “it is therefore desirable that the aggrieved party should place the facts before the [AG] prior to instituting proceedings himself”[86]. Other common law jurisprudence 88. As to Scots law, the point at issue was considered in Robb v Caledonian Newspapers Ltd, where Lord Justice-General Hope, sitting with Lord Cowie and Lord Wylie considered an argument that as an application for criminal contempt by a private party was made “without the concurrence of the Lord Advocate the application was incompetent”[87]. The court rejected the submission, saying that, in addition to being able to deal with such a contempt “at its own hand”, a court “may also deal with a contempt which is brought to its notice by means of a petition and complaint by the Lord Advocate or by any other interested party”[88]. 89. In Australia, it was said by Street CJ in DPP v Australian Broadcasting Corporation[89], that “[T]he only way in which[a person] can ensure that [an interference with the administration of justice] is brought to the notice of the court and the integrity of the administration of justice in respect of the litigation maintained is by having the power himself to institute proceedings for contempt” And in Witham v Holloway[90], McHugh J sitting in the High Court stated that “proceedings for criminal contempt could be initiated by the [AG], by the court of its own motion or any person with an interest in the subject matter of the proceedings”. And in the later High Court case, Re Colina ex p Torney[91], Hayne J said that “[t]he function that is exercised when a court proceeds against an alleged contemnor is not one to be exercised or controlled by the executive”. 90. In Canada, Allen CJ said in R v Ellis[92], “I do not find it laid down anywhere that the [criminal contempt] application must be made by the [AG]”, and accordingly a party in the underlying litigation could make the application. 91. The same conclusion was reached by McKechnie J sitting in the High Court of Ireland in Murphy v British Broadcasting Corporation[93]. 92. I also note that in the Ugandan High Court case AG v Kiwanuka[94]Ssekaana Musa J said that: “[w]hereas contempt proceedings are between the Court and alleged contemnor, the contemptuous acts can be brought to the attention of the Court by any person, including a person who is not a party to the court proceedings from which the contempt arose.” 93. Finally (not least because the position there is more nuanced), I turn to Singapore. In 2013, the Singapore Court of Appeal decided Aurol Anthony Sabastian v Sembcorp Marine Ltd[95]. In his judgment, Menon CJ concluded[96] that “given the sui generis nature of criminal contempt, there is a compelling interest in the AG being consulted before proceedings are commenced by a private party”, noting Lord Diplock’s observation in the 1972 Times Newspapers case[97] that it was “a procedure which if instituted by one of the parties to litigation is open to abuse”. Menon CJ then went on to reject the applicant’s reliance on “the absence of specific statutory provisions in Singapore that confine the power to initiate certain types of contempt proceedings to the AG”, on the ground that: “In Singapore, as reflected in Art 35(8) of the Constitution, the AG plays a unique and integral role as guardian of the public interest vis-a-vis the institution and conduct of all criminal proceedings. This includes proceedings for criminal contempt.” 94. Accordingly, the Singapore Court of Appeal concluded that the proper course for any applicant was first to consult the AG. As Menon CJ put it, if “the AG then decided to take up the matter”, he would be doing so pursuant to his constitutional power, but if the AG decided not to proceed, the applicant “could then have commenced proceedings of its own accord but in doing so it would have been obliged to join the AG as a party and lay all the facts before the court including any expressed views of the AG”.[98] He then held that the failure to consult the AG in that case was fatal to the criminal contempt proceedings, but nonetheless went on to consider in some detail the conduct of the alleged contemnor, and concluded that it justified his being deprived of any costs, despite succeeding in his appeal.[99] 95. In the course of the judgment in Aurol, reference was made to the US Supreme Court’s decision in Young v United States ex rel Vuitton et Fils SA[100], where it was held to be inappropriate for the legal representatives of the party in the underlying action (who was the party pressing for the contempt application against the other party in the underlying action) to have been charged by the first instance court to prosecute the contempt application. It would, said Brennan J, have “the potential for private interest to influence the discharge of public duty” and “[create] opportunities for conflicts to arise, and [create] at least the appearance of impropriety”[101]. 96. As Menon CJ recognised[102], “the appointment of [one party’s] legal representatives [in Young] as private prosecutors [of the other party] entailed also conferring certain investigatory powers upon them, over and above the power to bring proceedings”. However, he was not impressed by this distinction, although, as Brennan J said in Young[103], as private prosecutors, the legal representatives had “a terrible array of coercive methods”, including “police investigation and interrogation, warrants, … [and] authorized wire-tapping …”. G. Conclusion 97. For the reasons given in Section D of this judgment, it appears to me that, in the absence of any relevant legislative provision or case law to the opposite effect, the answer to the basic issue on these appeals should be that an applicant need not apply to, or otherwise involve, the SJ before issuing an application for leave to proceed with criminal contempt proceedings under O.52 rule 2. 98. Consideration of domestic and other common law jurisprudence on the topic (in Sections E and F of this judgment) seems to me clearly to confirm this preliminary view, in that there is no Hong Kong legislation or case law which has the effect of cutting down an applicant’s right of access to the court in such a case, other than the need to obtain the court’s leave. With one exception, in all the other common law jurisdictions on which we have been addressed, the law seems to be well established that, subject to obtaining the leave of the court, a private party is free to initiate criminal contempt proceedings. 99. The only exception appears to be in Singapore, where the effect of the Aurol case[104] is that a private party cannot initiate criminal contempt proceedings without first giving the AG the opportunity to initiate the proceedings, on the basis that, if the AG elects not to do so, the private party would be free to initiate such proceedings, subject to the leave of the court. But that is not of assistance to the appellants. First, even in Singapore, the AG does not have the power to prevent a private party from bringing criminal contempt proceedings[105]. Secondly, there was no suggestion in Aurol that the AG would not want all projected criminal contempt proceedings brought to his attention, whereas here the SJ has made it clear that that is his position. Thirdly, the court’s reasoning in Aurol was at least partly based on Article 35 of the Constitution of the Republic of Singapore, which of course has no part to play in Hong Kong. And I should add that the powers of lawyers conducting a contempt investigation in the United States means that the decision in the Young case[106] is of no assistance to the appellants. 100. In these circumstances, I have no hesitation in rejecting the appellants’ case. Not only does it conflict with principle, but there is nothing in our legislation or case law to support it. On the contrary, all the indications in The Rules and cases are against it. And when one turns to the position in other common law jurisdictions, the position appears exactly the same – with the exception of Singapore, which is only marginally different. 101. It is worth briefly addressing the question whether, reflecting the recommendations of the Phillimore Report[107] and the Hong Kong Law Reform Commission Report[108], and in accordance with the approach in Singapore and approved in Arlidge[109], it would be appropriate to adopt a practice whereby a private party wishing to bring criminal contempt proceedings should first be required to notify the SJ. If this course were adopted, the SJ would automatically be aware of every criminal contempt application that is to be made, so that he could decide whether to begin the proceedings himself, and, if he decided against, he could make sure that, to the extent that he so wished, he was represented in any proceedings from the start, i.e. from the leave application. 102. While I see the force of that argument (which is a much more limited version of the appellants’ case), I would not favour it. First, I would deprecate imposing any burden on an applicant who wishes to bring criminal contempt proceedings, especially if it could add to cost and delay. Secondly, it is always open to the SJ to apply to take over, or to be a party to, criminal contempt proceedings, and it accords better with the fact that contempt proceedings exist to protect the authority of the courts and are based on the inherent jurisdiction of the court, that it should be for the court to invite the assistance of the SJ when it appears to the court to be appropriate. As Mr Dawes SC said in his submissions, the SJ would be more likely to have a role in some criminal contempt proceedings (e.g. those involving allegedly contemptuous publications giving rise to constitutional rights issues) than in others. Thirdly, the SJ would not be particularly well equipped to deal with every prospective contempt application. 103. As well as being more consistent with principle, it would, I think, also involve a better allocation of resources for the court to act as the initial filter, while seeking the SJ’s assistance if wanted, then for every criminal contempt application to be initially referred to the SJ. And, of course, if the SJ wished to be heard on the leave application, he could always apply. Indeed, if his attitude were to change and he wished to be informed of every application for leave to bring criminal contempt proceedings, he could so inform the court, which would then no doubt comply with his request. H. Disposal of the appeals 104. In these circumstances, I would answer the four Questions[110] as follows: 1) No; 2) No; 3) Not applicable, but no; 4) Not applicable, but no. 105. Accordingly, I would dismiss these appeals. 106. The appellants should pay the respondent’s costs, and there should be no order in respect of the SJ’s costs. Chief Justice Cheung: 107. Accordingly, these appeals are unanimously dismissed and the Court makes an order nisi in terms of paragraph 106 above. Absent an application to vary, the order nisi shall become absolute after the lapse of 14 days from the date hereof. Mr Benjamin Yu SC, Ms Bonnie Y K Cheng and Mr Keith Chan, instructed by DLA Piper Hong Kong, for the 1st to 3rd Appellants in both appeals Mr Johnny Mok SC, Mr Michael Lok and Mr Charlie Liu, instructed by Howse Williams, for the Respondent in both appeals Mr Victor Dawes SC and Mr Cedric Yeung, instructed by Department of Justice, and Mr William Liu, Acting Deputy Law Officer (Civil Law), of the Department of Justice, for the Intervening Party in both appeals [1] [2021] HKCA 1580 [2] HCA 2562/2014 [3] [1974] AC 273, at p. 322 [4] Jennison v Baker [1972] 2 QB 52, at p. 61 [5] (1970) 23 Current Legal Problems 23, at p. 28 [6] [1992] 1 AC 191, at pp. 207-208, quoting from the 1974 Report of the Committee on Contempt of Court (the Phillimore Committee), Cmnd 5794 [7] (1765) Wilm 243, at p. 254 [8] See footnote 5, at p. 27 [9] See footnote 3, at pp. 315-316 [10] [1999] 2 AC 294, at p. 303 [11] [2018] HKCA 173, at para 74 [12] See footnote 3, at p. 294 [13] Roberts v Albert Bridge Co (1873) LR 8 Ch App 753 [14] (2009) 12 HKCFAR 830, at para 45 [15] 5th edition, 2017, at para 3-1 [16] [2016] 5 HKLRD 370, at para 42 [17] Ibid, quoting from Director of the Serious Fraud Office v O’Brien [2014] AC 1246, at para 39, per Lord Toulson [18] [2021] 4 WLR 103, at para 23. Decision affirmed [2022] 1 WLR 367 [19] See footnote 11, at para 82 [20] See Accent Foundation Ltd v Lee [2008] HLR 3, at paras 17 and 18 [21] See footnote 15, at para 3-26 [22] See footnote 16, at para 45 [23] [1988] Ch 333, at p. 377 [24] Ibid, at p. 362 [25] See footnote 6, at pp. 217 [26] (1999) 200 CLR 386, at para 110 [27] (1995) 183 CLR 525 [28] See footnote 6, at para 169 [29] [2016] 4 HKLRD 367, at paras 5 to 24 [30] Ibid, at para 5 [31] Ibid, at para 7 [32] See footnote 16, at para 46 [33] Ibid, at para 47 [34] Construction, Forestry, Mining and Energy Union v Boral Resources (Vic) Pty Ltd (2015) 256 CLR 375, at para 45 [35] See footnote 27 [36] Ibid, at para 65 [37] See footnote 16, at paras 47 to 51 [38] [2005] 4 HKC 416, at pp. 441-442 [39] [2012] 1 WLR 991, at para 7 [40] See footnote 15, at para 2-216 [41] See footnote 3, at p. 326 [42] Ibid, at pp. 311-312. Arlidge (see footnote 15) at paras 2-213 to 2-216 does not suggest that there is any conflict between these descriptions, apparently regarding them as complementary. The Court of Appeal in Re Mahesh J Roy (Intervener: Secretary for Justice) [2017] 2 HKLRD 935 at para 17 favoured the view of Lord Diplock and Lord Morris over that of Lord Cross. In so far as there is a difference, it seems to be more one of emphasis than principle, and at any rate, it does not detract from the characterization of criminal contempt proceedings as civil in nature. Indeed, such characterization was acknowledged by the court in Re Mahesh J Roy (No 2) (Intervener: Secretary for Justice) [2017] 5 HKLRD 830 at para 55 [43] [1897] 1 Ch 545, at pp. 559-560 [44] [1975] QB 73, at p. 84 [45] (1989) 88 Cr App R 63, at p. 67 [46] See paras 28 and 29 above [47] Sik Chiu Yuet v Secretary for Justice [2018] 4 HKLRD 194, at para 35 [48] See e.g. the English case Navigator Equities Ltd v Deripaska [2022] 1 WLR 3656 and the cases cited therein [49] See e.g. per Mustill LJ in R v Griffin see footnote 45, at p. 67 [50] See e.g. Re Bramblevale Ltd [1970] Ch 128, at p. 137 and Dean v Dean [1987] 1 FLR 517, at pp. 521 and 522 [51] See e.g. Tiong King Sing v Sam Boon Peng Yee [2015] 1 HKLRD 981; Pan Chung Pat Wo Tong (Hong Kong) Ltd v Law Yan Wai (unreported, HCA 1719/2010, 5th May 2017); China Metal Recycling (Holdings) Ltd v Chun Hei Man [2018] 1 HKLRD 455 [52] See e.g. in Cheung Kai Yin (No 1), footnote 29, at para 7 [53] [2021] 4 HKLRD 841 [54] [1978] AC 435 [55] See footnote 38, at p. 441 [56] See footnote 15, at para 14-5 [57] See e.g. per Li CJ in Ng Ka Ling v Director of Immigration (1999) 2 HKCFAR 4, at p. 28 [58] See para 55 above. In this connection, there is some congruity with the approach of the House of Lords’decision in Anisminic Ltd v Foreign Compensation Commission [1969] 2 AC 147, affirmed by a majority of the UK Supreme Court in R (on the application of Privacy International) v Investigatory Powers Tribunal [2020] AC 491. Anisminic was analyzed in Privacy International by Lord Carnwath (writing for the majority) as establishing a “strong” and “fundamental” presumption against statutory exclusion of the supervisory role of the High Court over other adjudicative bodies (paras 43 and 99), which can only be displaced by “the most clear and explicit words” (paras 37, 77 and 111). The supervisory jurisdiction of the High Court is an inherent one (paras 38 and 60), so there is hence much to be said of the view that such a presumption should also be applicable to any alleged statutory exclusion or restriction of the court’s contempt jurisdiction, given that it is also an inherent jurisdiction of the court [59] See footnote 29, at para 37 [60] Stock Exchange of Hong Kong Ltd v New World Development Co Ltd (2006) 9 HKCFAR 234, at para 43 [61] Final Report on Some Aspects of Final Adjudication and the Judicial System of the SAR, and the Role of an Independent Prosecuting Authority of 12th June 1987 [62] See paras 35 and 36 above [63] See footnote 1, at para 42 [64] See footnote 38, at p. 441 [65] Ibid, at para 13 [66] At para 9.1 [67] Ibid, at para 9.3 [68] See para 69 above [69] See footnote 3 [70] Ibid, atp. 293 [71] Ibid, at p. 306 [72] Ibid, at p. 311 [73] Ibid, atp. 312 [74] Ibid, at p. 314 [75] Ibid, at p. 326 [76] unreported, (C/2000/2363), [77] [2009] 1 WLR 2406 [78] It is not entirely clear whether they were criminal contempt proceedings, but they were described as such in the subsequent Court of Appeal Navigator case (footnote 48) at para 109 [79] See footnote 77, at para 14 [80] Ibid, at para 15 [81] See Navigator footnote 48, at para 81 [82] See footnote 6, at para 184 [83] Ibid, at para 187 [84] 4th edition (2010), at para 13.13 [85] See footnote 15, at para 3-183 [86] Ibid, at para 3-188 [87] 1994 SCCR 659, at p. 664 [88] Ibid, at p. 665 [89] (1987) 7 NSWLR 588, at p. 595 [90] See footnote 27, at p. 540 [91] See footnote 26, at para 112 [92] (1889) 28 NBR 497, at p. 519 [93] [2005] 3 IR 336, at pp. 354 to 356 [94][2022] UGHCCD 46, at p. 14 [95] [2013] 2 SLR 246 [96] Ibid, at para 51 [97] See footnote 3, at p. 312 [98] See footnote 95, at para 54 [99] Ibid, at para 115 [100] 481 US 787 (1987) [101] Ibid, at p. 806 [102] See footnote 95, at para 50 [103] See footnote 100, at p. 811 [104] See footnote 95 [105] Ibid, at para 66 [106] See footnote 100 [107] See footnote 6 [108] See footnote 67 [109] See footnote 86 [110] See para 17 above 
Chief Justice Ma: 1. I agree with the judgment of Lord Walker of Gestingthorpe NPJ. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Lord Walker of Gestingthorpe NPJ. Mr Justice Tang PJ: 3. I agree with the judgment of Lord Walker of Gestingthorpe NPJ. Mr Justice Fok PJ: 4. I agree with the judgment of Lord Walker of Gestingthorpe NPJ. Lord Walker of Gestingthorpe NPJ: Introduction 5. The issue in this appeal is whether the Respondent, CLP Power Hong Kong Limited (“CLP”) is liable to be assessed under the Rating Ordinance, Cap 116, in respect of certain fixed equipment at its generating stations in the New Territories and at its stand-by generating station on Lantau. The Appellant is the Commissioner of Rating and Valuation (“the Commissioner”). 6. The appeal is from the judgment of the Court of Appeal (Cheung, Kwan and Barma JJA) given on 2 February 2016, remitting the matter to the Lands Tribunal for further consideration of the evidence. The nature of the fixed equipment, and the reasons for the Court of Appeal’s decision to make a remission, are considered below. Much turns on the relationship between two sections of the Rating Ordinance, and on the meaning, in the context of those sections, of the words “machinery” and “plant”. 7. This matter has a long history. In 2008 there were no fewer than eight outstanding appeals by CLP against assessments under the Rating Ordinance and the Government Rent (Assessment and Collection) Ordinance, Cap 515, the earliest relating to the 1999-2000 assessment year. But seven of the appeals were stayed by an order of Lam J made on 3 November 2008. The exception was CLP’s appeal against the rating assessment for 2004-5, which was chosen as a suitable “reference year”. Since then further assessments have been made for later years. 8. The appeal in relation to the 2004-5 assessment was heard by the Lands Tribunal (Au J and Mr W K Lo) over 44 days between February 2010 and February 2012, with evidence from numerous expert witnesses. Judgment was given on 24 April 2013. The written judgment extends to 391 paragraphs, of which only three relatively short groups of paragraphs are directly relevant to the issue in this appeal. Paragraphs 28 to 43 summarise the relevant statutory provisions in the context of CLP’s undertaking (part of which is in mainland China). Paragraphs 84 to 93 contain a finding (in the very different context of an argument about “interdependency”, which concerned the relative bargaining positions of the hypothetical landlord and the hypothetical tenant) that the hypothetical landlord’s fixed equipment (the rateability of which is in dispute) and the hypothetical tenant’s machinery (the value of which, as is common ground, is not to be taken into account for rating purposes) “have been designed and built as an integrated system”. Paragraphs 227 to 234 set out the competing arguments on the disputed items of fixed equipment and decide the point in favour of the Commissioner, and paragraph 384 records Mr Lo’s concurrence on that point. 9. However the Lands Tribunal decided other points of more central importance in favour of CLP and the appeal relating to the 2004-5 assessment was allowed to the extent summarised in paragraphs 235 and 387 of the Lands Tribunal’s judgment. On 3 January 2014 its decision was confirmed after a review. On 9 April 2015 the Lands Tribunal refused leave to appeal on any of three grounds relied on by CLP, but on 28 August 2015 the Court of Appeal (Kwan and Barma JJA) granted leave to appeal on one of the three grounds, that is the issue relating to the items of fixed equipment. The facts 10. CLP has four generating stations in Hong Kong, on three different sites. The following summary is based on paragraphs 28 to 37 of the judgment of the Lands Tribunal. (1) Station A at Castle Peak Power Station (“CPPS ‘A’”) consists of four generating units. Their primary fuel is coal but their design is such that they can also burn oil. (2) Station B at Castle Peak Power Station (“CPPS ‘B’”) has four generating units similar to those at CPPS ‘A’, except that two of the units have been modified to burn natural gas. (3) Black Point Power Station (“BPPS”) is on the coast four miles north of Castle Peak Power Station. It has eight generating units with a combined cycle gas-fired system. (4) Penny’s Bay Power Station (“Penny’s Bay”) on Lantau Island has three 100MW gas turbines. It provides stand-by capacity to CLP’s network, to Hong Kong airport and to some other areas. CLP’s assets and interests in mainland China are described in paragraphs 38 to 43. 11. The items of equipment in issue in this appeal are described as follows in paragraph 227 of the Lands Tribunal’s first judgment: “(a) the boilers and supporting steelwork at CPPS ‘B’ and BPPS; (b) the cooling water circuits at CPPS ‘B’ and BPPS; and (c) ancillary pipework and electrical cables within the power stations.” This summary contains at least one, and possibly two, serious errors. It is common ground between the parties that the reference in (a) above to “boilers” is mistaken. The reference should be to steelwork supporting the boilers, which are themselves machinery coming within section 8(b) of the Rating Ordinance (set out in paragraph 14 below). It is regrettable that this serious mistake was not formally corrected with the agreement of the parties when the Lands Tribunal reviewed its first judgment. It is even more regrettable that the other possible error (asserted by CLP and denied by the Commissioner) was not brought to the attention of this Court until the close of counsel’s oral submissions. The parties cannot even agree as to whether there is an issue as to the supporting steelwork at BPPS. It will therefore be necessary to return to this (paragraphs 66 to 69 below). The Rating Ordinance 12. The issue in this appeal turns on the correct construction of sections 8 and 8A of the Rating Ordinance, and their application to the facts as found by the Lands Tribunal. The basic unit of assessment in Hong Kong is a “tenement”, defined in section 2 of the Rating Ordinance as - “any land (including land covered with water) or any building, structure, or part thereof which is held or occupied as a distinct or separate tenancy or holding or under any licence.” 13. In English rating law the corresponding expression is “hereditament”. In Hong Kong, where almost all land is held under a Government lease, the word “tenement” (and its definition as set out above) express the importance of the notion of a holding of land. It has been a basic principle of rating law in Hong Kong that a rateable tenement is held and occupied by one person only. That principle now has two exceptions, one in section 8A (added by amendment in 1991) and the other in section 9 (which relates to advertising stations, and was amended in 1981). As explained below, the main (but not the sole) purpose of section 8A is to keep up with advances in equipment related to information technology, such as fibre-optic cables. 14. Section 8 has been part of the Rating Ordinance since its enactment in 1973 (as a consolidation, with amendments, of legislation going back to the 19th century; there is a summary of the history in the judgment of Cons JA in Commissioner of Rating and Valuation v Yiu Lian Machinery Repairing Works Ltd & Ors [1985] 2 HKC 517, 521-2 – “Yiu Lian”). The section was amended in 1981, but only by adding a reference to the new section 7A. It provides, “For the purpose of ascertaining the rateable value of a tenement under section 7 and 7A - (a) subject to paragraph (b), all machinery (including lifts) used as adjuncts to the tenement shall be regarded as part of the tenement, but the reasonable expenses incurred in working such machinery shall be allowed for in arriving at the rateable value of the tenement; (b) no account shall be taken of the value of any machinery in or on the tenement for the purpose of manufacturing operations or trade processes.” 15. Section 8A provides, “(1) Where any land (including land covered with water) or any building or structure is occupied by a person by means of any plant, such land, building or structure shall, to the extent that the land, building or structure is so occupied, be deemed for rating purposes to be a separate tenement, whether or not such land, building or structure is otherwise a tenement and that person shall be deemed for rating purposes to be the occupier of such tenement and liable for payment of rates assessed thereon. (2) For the purposes of ascertaining the rateable value of such tenement, the plant by means of which the person is occupying the tenement shall be regarded as part of the tenement. (3) In this section “plant” (工業裝置)　includes cables, ducts, pipelines, railway lines, tramway lines, oil tanks, settings and supports for plant or machinery.” 16. Section 10, by an amendment made in 1991, modifies the effect of the deeming provision in Section 8A by enabling the Commissioner to treat a number of separate tenements as one single tenement by making what is called an in cumulo assessment. That method of assessment is regularly adopted in the rating of utility companies. 17. Both leading counsel (Mr Benjamin Yu SC for the Commissioner, and Mr David Elvin QC for CLP) made detailed submissions as to the correct construction of sections 8 and 8A, in the context of the Rating Ordinance as a whole, and their submissions are considered below. But it may be useful to make some fairly simple preliminary points about the language of sections 8 and 8A, since the two sections, standing side by side but without any direct reference to each other, present some unusual features. 18. It is to be noted that section 8 refers only to machinery (not to plant), and does not define the term “machinery” (except for indicating that it includes lifts in a building); and conversely section 8A refers only to plant, and not to machinery (except for the final words of subsection (3) referring to “settings and supports for plant or machinery”). Section 8A(3) does not contain a full and comprehensive definition of the term “plant”. It does no more than list a number of items which are to be included in the statutory meaning. 19. That sort of inclusive definition is not necessarily bad drafting practice, especially when the term in question is an ordinary word with a wide but imprecise meaning. Henry Thring, who is often regarded as having been the founder of modern statutory draftsmanship during his time as first parliamentary counsel in the days of Gladstone, gave a warning to over-ambitious draftsmen in his short work, Practical Legislation, first published in 1877 (3rd ed, 2015, p 102): “The proper use of definitions is to include or exclude something with respect to the inclusion or exclusion of which there is a doubt without such definition. No attempt should be made to make a pretence of scientific precision by defining words of which the ordinary meaning is sufficiently clear and exact for the purpose of the Act in which they are used.” 20. Nevertheless the drafting of sections 8 and 8A is rather unusual in that there is no clear explanation of the relationship between the two sections. That omission is particularly surprising because for well over a century the phrase “plant or machinery” (or sometimes “plant and machinery”) has often been used in statutes, in different common law jurisdictions and in different legal contexts (including rating and taxation), as a composite expression similar to other familiar composite expressions such as “repair or maintenance”, “fit and proper”, “health and safety” or “sports and pastimes”. In ACT Construction Limited v Customs and Excise Commissioners [1981] 1 WLR 1542, 1546 Lord Roskill said of “repair or maintenance”, “The two words are not used in antithesis to one another. The phrase is a single composite phrase ‘repair or maintenance’ and in many cases there may well be an overlap between them.” 21. Even when the two words “plant” and “machinery” are not used in a single composite phrase, there is a considerable overlap between them. They are both ordinary words of wide but imprecise meaning, but “plant” normally has a wider meaning than “machinery”. In normal usage “plant” refers to a wide variety of tangible property, other than land or buildings, used as fixed assets (rather than circulating assets) of a business. That point is clearly made in the well-known definition given by Lindley LJ in Yarmouth v France (1887) 19 QBD 647, 658, in which the expression was, at first glance surprisingly, held to include a horse pulling a trolley of goods unloaded from a ship: “There is no definition of plant in [the Employers’ Liability Act 1880]: but in its ordinary sense, it includes whatever apparatus is used by a business man for carrying on his business - not his stock-in-trade which he buys or makes for sale; but all goods and chattels, fixed or moveable, live or dead, which he keeps for permanent employment in his business: see Blake v Shaw.” The vicious nature of the horse which kicked the driver, breaking his leg, was held to be a defect in the employer’s plant within the meaning of the Employers’ Liability Act 1880. 22. Blake v Shaw (1860) Johns 732 is a decision of Page-Wood V-C holding that a large stock of ready-made clothes was not included in a bequest of “plant” in the will of a wholesale clothier. He said at p 734, “In most cases the word ‘plant’ is used to describe something which, if not in direct contrast to stock, is at any rate of an entirely different nature. All the matters permanently used for the purpose of trade, as distinguished from the fluctuating stock, are commonly included in the word ‘plant’. It consists sometimes of things which are fixed, as, for example, counters, heating, gas and other apparatus and things of that kind, and in other cases of horses, locomotives and the like, which are in this sense only fixed that they form part of a permanent establishment intended to be replaced when dead or worn out, as the case may be.” 23. This passage is useful in clarifying two different meanings of “fixed”. Old as it is, it remains relevant today, at any rate if the reference to horses is replaced by a reference to motor vehicles of all sorts. There is of course no question of horses, motor vehicles or locomotives coming within the scope of section 8A, but that is not because “plant” has a restricted meaning. It is because the section applies only to plant by means of which a person occupies some land, building or structure. 24. “Machinery” is also an ordinary word of wide but imprecise meaning. Not all machinery is used for manufacturing purposes, or indeed for any business purpose, as the reference to lifts in section 8(a) reminds us. A passenger lift in a private house or block of flats is machinery but would not normally be described as plant. In this respect, therefore, but only in this respect, the normal meaning of “machinery” covers a larger field than the normal meaning of “plant.” In his judgment in the Court of Appeal Cheung JA cited an instructive case from New Zealand concerned with gasometers and their ancillary pipes and mains, Auckland City Corporation v Auckland Gas Co Ltd [1919] NZLR 561, 586 (“Auckland”), in which Sim J said that “machinery” “... means primarily a number of machines taken collectively, and a machine in its popular sense is a piece of mechanism which, by means of its interrelated parts, serves to utilise or apply power, but does not include anything that is merely a reservoir or conduit, although connected with something which is without doubt a machine.” The majority of the New Zealand Court of Appeal held that each gasometer was to be regarded as machinery, because it was a sort of giant pump, but the pipes and mains were not regarded as machinery. Only Stout CJ followed an old Massachusetts decision that an entire gasworks should be considered to be “one great integral machine”. Each member of the Court recognised that the decision must turn not on dictionary definitions but on the context of the particular statute (see at pp 564, 583, 586-7 and 588-9). The Hong Kong Electric case 25. This appeal is not the first occasion on which this Court has had to consider the rating of a power station generating electricity. In Hong Kong Electric Co Ltd v Commissioner of Rating and Valuation (No 2) (2011) 14 HKCFAR 579 (“Hong Kong Electric”) this court heard an appeal concerned with the rating of the power station on Lamma Island owned by Hong Kong Electric Co Ltd (“HKE”). That appeal might have raised contentious issues about the proper treatment of plant and machinery, and the correct construction of sections 8 and 8A of the Rating Ordinance, such as have been raised in this case. In the event the issues do not seem to have been raised, which is no doubt the reason why the case received little attention in the Court of Appeal. 26. The leading judgment was given by Lord Millett NPJ, with whom the other members of the court agreed. Ribeiro PJ gave a separate judgment dealing with assets under construction, with which the other members of the court (including Lord Millett NPJ) agreed. Lord Millett NPJ discussed the meaning and effect of sections 8 and 8A in some detail, and his considered view, concurred in by the whole court, must carry great weight, even though it was not a contentious issue in that appeal. 27. Cheung JA did cite (and adopt) paragraph 136 of the judgment of Lord Millett NPJ, which is as follows: “Section 8 provides that, for the purpose of ascertaining the rateable value of the tenement, (a) machinery used as an “adjunct” to the tenement is to be regarded as part of the tenement, but (b) no account is to be taken of the value of any machinery in or on the tenement required for the purpose of manufacturing operations or trade processes. Section 8A provides that plant, together with land, buildings or structures occupied by means of any plant, is deemed for rating purposes to be a separate tenement. Cables, ducts, pipelines, oil tanks, and settings and support for plant or machinery fall within the definition of ‘plant’. While plant and fixed machinery both form part of the tenement, therefore, the value of such machinery used for manufacturing operations or trade processes is to be disregarded in ascertaining the rateable value of the tenement.” 28. Cheung JA did not however cite the following paragraphs, which amplify and apply the general description in paragraph 136. They also refer to the official practice of in cumulo assessment under section 10 of the Rating Ordinance. It is useful to see how Lord Millett NPJ continued: “137. It follows that the assets employed by the Company in its undertaking are divided for rating purposes into two categories. The first consists of the rateable assets (‘RA’) being those assets which form part of the tenement and whose value is to be taken into account in ascertaining its rateable value. The second comprises those assets which do not form part of the tenement or whose value is to be disregarded in ascertaining its rateable value and are accordingly non-rateable assets (‘NRA’). 138. The tenement with which the present case is concerned consists of the land, buildings and structures occupied and used on 1 April 2004 by the Company for the generation, transmission, distribution and supply of electricity on Hong Kong Island, Ap Lei Chau and Lamma Island. These comprise a large number of separate tenements but the Commissioner exercised her power to treat them as one single tenement. Such an assessment is described as an in cumulo assessment and its adoption in the present case is not controversial. 139. The assets employed by the Company for the purposes of its undertaking are described at length in the decision of the Lands Tribunal. It was based upon detailed evidence which was not in dispute. On the relevant date: (a) Electricity was generated by means of eight coal fired generating units, five gas turbines and one combined cycle gas turbine installed on a 62 hectare site on Lamma Island. A 22 hectare extension to the site, consisting of reclaimed land, was under construction but not in use. While the land, buildings, structures, plant and some of the machinery are part of the tenement and rateable, the generating station houses a huge amount of machinery and apparatus which, while part of the tenement, fall to be disregarded in assessing its rateable value and are NRA. While the turbines which generate the electricity are NRA, the specially designed reinforced concrete foundations to which they are bolted are rateable. Similarly, while the boilers are NRA, they are held in place within specially designed structures which are part of the tenement and rateable. The moving parts of the coal conveyor system are NRA but the structure which supports them is part of the tenement. The switchgear and transformers are NRA but the cables which are specially designed to connect to them in order to transmit, distribute and supply the electricity are plant and form part of the tenement. (b) Electricity was transmitted at very high voltages by cables which are largely underground or in tunnels, although some are overhead supported by pylons, to 32 substations where it was stepped down for distribution. The cables comprise plant and form part of the tenement. The substation buildings are also part of the tenement but the transformers, switch gear and other apparatus which they contain are NRA. (c) Electricity was then distributed in cables to over 3,000 sub-stations where it was further stepped down to the voltages used by consumers. Most of these sub-stations were located within buildings owned and occupied by third parties. Again, the cables comprise plant and so form part of the tenement. The space within buildings where the sub-station equipment is located is also part of the tenement but the transformers, switch gear and other apparatus installed there are NRA.” 29. This analysis by Lord Millett NPJ, concurred in by the other members of the court, clearly proceeds on the basis that the items of plant specifically brought in by section 8A (“cables, ducts, pipelines” etc) were not “machinery” and so could not obtain the benefit of section 8(b). Machinery which was within the protection of section 8(b) was not deprived of that protection by section 8A merely because it might also be described as plant. The only effect of section 8A on such machinery would be to attract the deeming provision (as to being a separate tenement) in section 8A(1), but that would in practice be undone by an in cumulo assessment under the power conferred by the amended section 10. The Yiu Lian Case 30. The conclusion reached by Cheung JA, and concurred in by the other members of the Court of Appeal, differed from the view which Lord Millett NPJ expressed (I stress again, without it having been a contested issue) in Hong Kong Electric. Cheung JA referred at some length to the case of Yiu Lian [1985] 2 HKC 517. That was a very unusual case in that the Commissioner was seeking to charge rates on a floating dry dock which was attached to the land only by chains attached to concrete blocks on the sea bed. In the Court of Appeal (which reversed the Lands Tribunal’s decision in favour of the Commissioner) there was a good deal of discussion about differences between rating law in Hong Kong and England, and on how far those differences depended on the different concepts of “tenement” and “hereditament”. 31. I am doubtful whether the unusual facts of Yiu Lian would have led to a different result in England. But the case was regarded in Hong Kong as having wider implications for the rating of equipment which could not be classified as machinery, but could be classified as plant. In this way Yiu Lian contributed indirectly to the legislative changes made in 1991, but a more immediate impetus for the changes was a dispute about the rating of fibre-optic cables and ducts installed by the Hong Kong Telephone Company, which in 1990 reached a negotiated settlement with the Commissioner of an appeal commenced in 1988. These cables and ducts formed a very extensive network, passing through public and private property, and presented a new problem as to the reach of the rating legislation. 32. All this is set out in some detail in the legislative history contained in section V (paragraphs 5.1 to 5.6) of the judgment of Cheung JA. Section 5.1 contains an illuminating extract from a work by Mrs Mimi Brown, written in 2013 when she was Commissioner, and paragraph 5.6 contains an extract from the 1991 Legislative Council briefing paper (produced to the Court of Appeal without objection) which also explains (in its paragraphs 5 and 7) the reasons for the amendment of section 10 of the Rating Ordinance. Principles of statutory construction 33. There was little difference between counsel as to the principles of statutory construction to be applied (but a great deal of difference as to the conclusions that should be reached). The classic statement is that of Li CJ in HKSAR v Cheung Kwun Yin (2009) 12 HKCFAR 568, paragraphs 11 to 14, of which there is a full summary in paragraph 9.1 of the judgment of the Court of Appeal. 34. Mr Yu also referred, however, to the warning against an exorbitantly purposive approach given by Lord Millett NPJ in China Field Ltd v Appeal Tribunal (Buildings) (No 2) (2009) 12 HKCFAR 342, paragraph 36: “There can be no quarrel with the principle that statutory provisions should be given a purposive construction; but there has been a distressing development by the courts which allows them to distort or even ignore the plain meaning of the text and construe the statute in whatever manner achieves a result which they consider desirable. It cannot be said too often that this is not permissible. Purposive construction means only that the statutory provisions are to be interpreted to give effect to the intention of the legislature, and that intention must be ascertained by a proper application of the interpretative process. This does not permit the Court to attribute to a statutory provision a meaning which the language of the statute, understood in the light of its context and the statutory purpose, is incapable of bearing: see HKSAR v Lam Kwong Wai (2006) 9 HKCFAR 574 per Sir Anthony Mason NPJ at paragraph 63.” 35. The general principles of statutory construction are indeed general principles which apply even in specialised fields such as taxation, rating and compulsory purchase of land. This has a bearing on what Cheung JA (in paragraph 9.14 of his judgment) rightly called “the so-called ‘piecemeal’ principle in rating law”. In a compulsory purchase case which was one of the last appeals heard by the House of Lords before its jurisdiction passed to the Supreme Court, Transport for London Ltd v Spirerose Ltd [2009] UKHL 44, paragraph 25, I cited (with the approval of the other members of the Committee) some observations of Kirby J in a dissenting judgment in Commissioner of Taxation v Ryan (2000) 201 CLR 109, 146: “It is hubris on the part of specialised lawyers to consider that ‘their Act’ is special and distinct from general movements in statutory construction which have been such a marked feature of our legal system in recent decades. The Act in question here is not different in this respect. It should be construed, like any other federal statute, to give effect to the ascertained purpose of the Parliament”. The principal issues in the appeal 36. When leave to appeal to this Court was granted by the Court of Appeal by an order dated 24 May 2016, the order identified six questions of great general or public importance. The first two questions relate to section 8A and are framed as alternatives, and the third question also relates to section 8A. The fourth question relates to the so-called ‘piecemeal’ principle. The fifth question relates to the interaction of section 8 (and especially section 8(b)) with section 8A. The sixth question (if it arises) relates to the issue of remission to the Lands Tribunal. 37. The first three questions can conveniently be considered together, and I think it is more logical to consider the fifth question before the fourth. With that rearrangement, there are four principal issues in the appeal: (1) the correct construction of section 8A, and in particular its statutory purpose and its reach (that is, the extent of its application); (2) the interaction of section 8A with section 8, and especially the relieving provision in section 8(b); (3) the so-called “piecemeal” principle; and (4) the issue of remission. Those issues are considered below in that order. (1) The Construction of Section 8A 38. In paragraph 9.4 of his judgment Cheung JA identified the key issue as: “...whether section 8A is indeed a deeming provision as submitted by CLP. If it is, then, in my view, section 8A has to be discarded in its application in the present case because the disputed items which are said to be rateable are already part of CLP’s tenements and as such it is not necessary to rely on a deeming provision to bring them into the fold of tenements for rating purposes.” 39. Plainly section 8A(1) is in some sense a deeming provision. It uses the words “shall ... be deemed for rating purposes.” But deeming provisions are not all of a kind. They do not all have the same force. As Lord Radcliffe said in St Aubyn v Attorney General [1952] AC 15, 53, the expression may be “used to give a comprehensive description that includes what is obvious, what is uncertain and what is, in the ordinary sense, impossible”. Moreover in paragraphs 9.5 and 9.9 of his judgment Cheung JA seems to me, with great respect, to have interpreted the words “whether or not such land, building or structure is otherwise a tenement” in section 8A(1) in a sense opposite to their natural meaning in this context. The natural meaning of “otherwise”, in this context, is “in the absence of this section” or “but for this section”. The section is to apply to any equipment which is plant by means of which any land, building or structure is occupied by a person “whether or not such land, building or structure is otherwise a tenement. It achieves that result in two stages. Subsection (1) deems the land, building or structure (to the extent of its being occupied by the relevant plant) to be a separate tenement, and subsection (2) provides that the plant is to be regarded as part of that separate tenement for the purposes of ascertaining its rateable value. The word “regarded” in subsection (2) indicates a less emphatic type of deeming. 40. That is not to say that section 8A is without its difficulties. It leaves the reader to puzzle out that “plant” has, in the context of business premises, a wider meaning than “machinery”, even though the very wide meaning given to it in Yarmouth v France (1887) 19 QBD 647 is restricted by the requirement in section 8A(1) that some land, building or structure “is occupied by means of” the item of plant in question. The reader also has to puzzle out that although the principal and immediate legislative purpose of the amendments made in 1991 was to ensure that “plant” should cover the network of fibre-optic cables and ducts of the Hong Kong Telephone Company (cables and ducts being the first two items specified in section 8A(3)), it was also intended (in view of Yiu Lian) to apply to other ancillary equipment that was not naturally described as machinery, especially equipment used for the transmission of electrical current or substances in gaseous, liquid or solid form (that being a common feature of cables, ducts, pipelines, railway lines and tramway lines, the first five items in the list). Sim J may have had the same approach in mind in his comment in Auckland (at p 587) “They are thus conduits and reservoirs, but nothing more.” 41. Land occupied by any of these items is to be treated as a separate tenement, partly to avoid difficulty with the long-standing principle of rating law that a tenement can have only one person at a time in rateable occupation of it. A telephone company’s cables and ducts within (say) an office block are to be regarded as in that company’s occupation, even though the office block is occupied by others. The same treatment can be applied to electricity cables and other equipment forming part of the distribution network of a company generating electricity. As a matter of practical administration, however, the theoretical “separateness” of all the company’s tenements would be overridden by an in cumulo assessment under section 10. 42. Mr Yu challenged the Court of Appeal’s interpretation of section 8A both on linguistic and on contextual grounds. In the first place he drew attention to this passage in paragraph 9.5 of the judgment of Cheung JA: “To emphasise this point, this section further provides that the land is to be deemed to be a separate tenement ‘whether or not such land, building or structure is otherwise a tenement’. Mr Yu submitted the word ‘otherwise’ is quite different from ‘not otherwise’. As will be discussed later, I disagree with this view.” 43. With great respect to Cheung JA, I cannot follow the reasoning in this passage. The words emphasised make clear that some land, etc is to be deemed to be a separate tenement, but the wording of section 8A read as a whole also make clear that that the plant occupying that land, etc may be either plant (for instance, a fibre-optic cable in a duct) that would not otherwise be recognised as a tenement at all, or plant (for instance, a steel support for a boiler) that would otherwise be treated as part of a larger tenement. The later discussion in paragraph 9.9 of the judgment does not to my mind take the point any further. 44. Nor did Mr Elvin provide a satisfactory answer. His written submissions state (at paragraph 23) that in the case of CLP’s power stations there is “no need to apply section 8A to deem a tenement because one already exists”, and he developed the point in his oral submissions. But the argument ignores the “whether or not” words which Cheung JA emphasized. No doubt there were other ways, and perhaps simpler ways, for the legislature to achieve its legislative intention, especially as the separateness of all these tenements is in practice undone by an in cumulo assessment under section 10. But on the plain meaning of the words of the statute Mr Yu’s argument on this point must prevail. 45. In paragraphs 9.12 and 9.13 of his judgment Cheung JA referred to the judicial definition of “plant” in Yarmouth v France and observed that under it machinery comes within the definition of plant. He went on to say, without further explanation, that the Commissioner’s argument would create “... an extremely unsatisfactory situation. This will render section 8 devoid of any true meaning and impose such an absurd result that the Court should strive to avoid giving section 8A such a meaning.” He treated this as confirmation of the view that he had already expressed, that section 8A has only a limited application. But the reasoning depends on an unexplained and questionable assumption about the relationship between section 8 and section 8A, the topic dealt with in the next section of this judgment. 46. As a footnote to the discussion of the first issue, I note that the Lands Tribunal dealt with it in paragraph 233 of their first judgment. The Lands Tribunal did not accept the submissions of leading counsel then appearing for CLP. I agree with their conclusion but not with all their reasoning. In my view paragraph 233(1) overstates the differences between what section 8 is dealing with, and what section 8A is dealing with, by overlooking section 8A(2), which (like section 8) refers to ascertaining the rateable value of a tenement. (2) The relationship of section 8 and section 8A 47. Section 8A does not expressly amend the text of section 8, which must be taken to remain in force unless and except so far as section 8A necessarily alters its effect. Section 8A does alter the effect of section 8, but only in the rather technical way mentioned in paragraphs 40 and 41 above. Machinery which is an adjunct to a tenement within the meaning of section 8(a) (which I shall call “adjunct machinery”) may also be plant occupying some land, building or structure within the meaning of section 8A (which I shall call “occupying plant”). Occupying plant is by section 8A(2) to be regarded as part of the separate tenement established by the deeming provision in section 8A(1). If the occupying plant happens to be adjunct machinery then the words “shall be regarded as part of the tenement” in section 8(a) must be displaced; the plant cannot simultaneously be a separate tenement, and part of another tenement. But that is a technicality with little or no practical significance, since in practice the Commissioner will make an in cumulo assessment under section 10. 48. In other respects section 8 remains in full force. In particular, the relieving provision in section 8(b) continues to apply in respect of “any machinery in or on the tenement for the purpose of manufacturing operations or trade processes.” Section 8(b) does not say that such machinery is to be deemed not to exist, since that might have had unintended implications for the valuation of the tenement as a whole. It simply provides that no account is to be taken of the value of the machinery. Section 8A(2) applies “for the purpose of ascertaining the rateable value of such tenement”, but there that is no more inconsistent with the relieving provision in section 8(b) than the identical opening words of section 8. In my view there is absolutely nothing in section 8A to produce the “extremely unsatisfactory situation” or the “absurd result” referred to in paragraph 9.13 of the judgment of Cheung JA (see paragraph 45 above). I respectfully consider that the Court of Appeal has adopted an unnatural construction of section 8A in order to avoid a supposed problem which does not really exist. 49. The overlap created by the possibility (indeed, the strong probability) that most adjunct machinery falling within section 8(a) will also be occupying plant falling within section 8A could be avoided only by treating the inclusive definition of “plant” in section 8A(3) as limited to ancillary equipment which is not adjunct machinery falling within section 8(a), and is one of (or perhaps is of a kind with) the seven types of ancillary equipment specifically mentioned in the definition. But it would be a strong thing to treat those specific cases, mentioned in an inclusive definition, as if they effectively constituted a full, comprehensive definition (see paragraph 19 above). It is not necessary to go that far in order to make sense of the two sections as a whole (nor would it alter what I regard as the proper outcome of this appeal, although it might affect the outcome of other cases on different facts). (3) The so-called ‘piecemeal’ principle 50. Mr Yu has in his written case (paragraph 37) and orally submitted that it is a rule in the law of rating that in deciding whether an item is plant, it is not permissible to have regard to its function as part of a larger whole. The authorities on which he relies for this proposition are Shell-Mex and BP Ltd v Holyoak [1959] 1 WLR 188 (“Shell-Mex”) and Manchester Marine Ltd v Duckworth [1973] 1 WLR 1431 (“Manchester Marine”). He goes so far as to say (paragraph 38) that in the construction of section 8A of the Rating Ordinance, “the legislature should be assumed to know the law, including the aforesaid principle.” 51. In my opinion there is no such principle. I would not of course accuse Mr Yu of hubris, but the notion that there is some sort of unwritten common law of rating known only to rating specialists must be dismissed for the reasons stated by Kirby J in Commissioner of Taxation v Ryan (paragraph 35 above). It is not a matter of the legislature having to know the law, but of the legislature making the law. If there were anything like the principle for which Mr Yu contends, it would be a principle derived from statute law construed in accordance with the general principles of statutory construction. Despite their common roots, the English statute law of rating is very different, particularly in relation to machinery and plant, from the rating law of Hong Kong set out in the Rating Ordinance. 52. The process of statutory construction can be seen in several of the authorities cited to this Court. In Auckland the Court of Appeal of New Zealand was concerned with that country’s Rating Act 1908, which exempted various categories of property including “machinery, whether fixed to the soil or not”, with no further definition. The majority of the Court decided, on the normal meaning of the word “machinery”, that the gas mains and pipes were not exempted: see Chapman J at p585, Sim J at p587, and Hosking J at p591. 53. In contrast to the simplicity of the New Zealand law a century ago, English statute law on rating has for almost a century adopted a much more detailed and prescriptive approach to machinery and plant. This approach was first introduced by the Rating and Valuation Act 1925, followed by the Plant and Machinery (Valuation for Rating) Order 1927. This primary and secondary legislation followed an unusual technique, which was described and commented on by the English Court of Appeal in BP Refinery (Kent) Ltd v Walker [1957] 2 QB 305 (“BP (Kent)”), a case concerned with BP’s large oil refinery at the Isle of Grain. I am reluctant to go much further into this and later English legislation, since it is so different from the Rating Ordinance. But in view of Mr Yu’s reliance on Shell-Mex and Manchester Marine I feel obliged to do so. 54. The 1925 Act contained a Third Schedule, incorporated into the Act by section 24(1), which set out four classes of plant or machinery, described in some technical detail, which were to be treated as part of the rateable hereditament, though with an exception for machinery or plant “for the purpose of manufacturing or trade processes” (the same wording as is used in section 8(b) of the Rating Ordinance). However (and this is the unusual legislative technique) section 24(3) was in these terms: “For the purpose of enabling all persons concerned to have precise information as to what machinery and plant falls within the classes in the said Third Schedule, there shall be constituted a committee of five persons to be appointed by the Minister...” The committee of experts were to advise the Minister in even more detail about the contents of the Third Schedule, and the Minister had power to reproduce their advice, with or without modification, in a statutory instrument. That was the origin of the 1927 Order. 55. As Lord Evershed MR dryly remarked in BP (Kent) at p315, the parliamentary intention was undoubtedly laudable, but the hope of achieving precise certainty was somewhat over-sanguine. In fact one of the issues in the case was whether the 1927 Order was ultra vires. Nevertheless the same technique, with some modifications, has been used in a succession of later English statutes, including the General Rate Act 1967 (section 21 and the Third Schedule) and the Plant and Machinery (Rating) Order 1960. The distinction drawn in paragraph 39.1 of Mr Yu’s written case is therefore mistaken, but the mistake is understandable since Lord Denning MR’s unreserved judgment in Manchester Marine refers only to section 21. 56. Shell-Mex was a borderline case on relatively simple facts. It was concerned with the rating of a newly-built filling station, which had underground tanks for the storage of petrol. There were four 3000-gallon cylindrical metal tanks, each weighing about two tons when empty. For safety reasons they were housed in rectangular concrete and brick silos, with the space between the silo walls and the tanks filled with a special type of sand. Under the 1925 Act and the 1927 Order this facility was to be treated as part of the rateable hereditament only if it met the two-part test of (a) being a “tank” (which was not in dispute) and (b) being within the words “is, or is in the nature of, a building or structure” (this was the essential issue). The borderline nature of the point is apparent from the course of the litigation. The Lands Tribunal decided in favour of the ratepayer. The Court of Appeal unanimously reversed that decision. The House of Lords, by a bare majority of three to two, restored the decision of the Lands Tribunal. 57. All three law lords in the majority treated the issue as a short (though difficult) question of fact. For instance Viscount Simonds said, at p 193, “... it is not a case of that which was a tank being turned into a structure but of that which was a tank remaining a tank and being housed in a structure made for that purpose.” So the structure of the silo was rateable but the metal cylinder was not. The only point of law in the case was the House's disapproval of a test depending on the notion of a “functional entity,” which had been used as a test in earlier cases, and also alternatively the notion of a “physical entity”. Viscount Simonds referred to the latter test, at p 194, as “obscurum per obscurius,” (the obscure through the more obscure), a Latin phrase which is a shorthand way of describing the futility of trying to explain or define an obscure concept by the use of even more obscure language. Lord Reid, at p199, did say that “the purpose of this legislation is to split up for rating purposes installations which are for practical purposes indivisible,” but I do not read that as laying down any general principle of the need for a “piecemeal” approach. It was his conclusion as to how the detailed provisions of the 1925 Act and the 1927 Order should be applied to the facts of the particular case. 58. In Manchester Marine Lord Denning MR did, in an unreserved judgment, express himself in wide general terms, as he was sometimes inclined to do. At p1435 he said, “Rating law is different from revenue law”, which is a surprisingly broad proposition, not least because rating law can be seen as a branch of revenue law. On the same page he cited the passage from Lord Reid that I have already mentioned, and said that it is legitimate in rating cases to consider the subject piecemeal and to split it up in a way that is not permissible in revenue cases. That is so, but the reason is not because of some quasi-constitutional principle about construing statutes dealing with different fields of law (the strict construction of penal statutes and statutes restricting civil liberties does, by way of contrast, have a constitutional basis: Gurung Kesh Bahadur v Director of Immigration(2002) 5 HKCFAR 480, para 24). The reason is simply that the English rating legislation contains very detailed provisions set out in statutory instruments, whereas the English law on capital allowances for purposes of income tax (or now corporation tax) refers to plant and machinery without further explanation or elaboration (as does the Hong Kong legislation concerned with profits tax: see Inland Revenue Ordinance, Cap 112, sections 36A to 39E). 59. For these reasons Cheung JA was right to refer to the supposed principle as “so-called”. What is needed is reliance on the ordinary general principles of statutory construction. In my view they lead irresistibly to the conclusions set out in the earlier sections of this judgment. (4) The issue of remission 60. The Court of Appeal directed that the Lands Tribunal’s judgment of 24 April 2013 and 3 January 2014 shall be remitted to the Lands Tribunal for consideration of three categories of equipment, referred to as “the disputed items”. These were the items referred to in paragraph 11 above, but (as explained in that paragraph) there is an issue as to whether the disputed items include the supporting steelwork for the boilers at BPPS. That issue should not have been left unacknowledged and unresolved until the close of argument in this Court. 61. I shall first consider items of equipment agreed to be disputed items. CLP had not asked the Court of Appeal to remit the matter to the Lands Tribunal. CLP’s case before the Court of Appeal was that section 8A did not, as a matter of law, apply to any equipment on any land, building or structure which was already (that is, apart from section 8A) a rateable tenement. It accepted, as Mr Elvin has confirmed to us, that the disputed items did come within the definition of “plant” in section 8A(3). The Court of Appeal recognised (para 9.35 of the judgment of Cheung JA) that there are serious practical difficulties about a remission, a full seven years after the beginning of the first Lands Tribunal hearing. 62. As already mentioned (paragraph 8 above) the Lands Tribunal made a finding that the hypothetical landlord’s assets and the hypothetical tenant’s non-rateable assets “had been designed and built as an integrated system”. But the same was no doubt true of the gasworks in Auckland, the petrol tanks and their silos in Shell-Mex, and the dry dock in Manchester Marine. Moreover this discussion was in the different context of “interdependency” and the relative bargaining positions of the hypothetical landlord and tenant. It was not expert evidence as to whether all or any of the disputed items were “machinery” falling within section 8(b), as the question- begging reference to non-rateable assets indicates. 63. If expert evidence on that point was required, it was for CLP, rather than for the Commissioner, to adduce it. But I doubt whether much if any expert evidence about the disputed items was required. Supporting steelwork for boilers falls squarely within the definition in section 8A(3) and cannot reasonably be described as a machine, or part of a machine, even though the steelwork was carefully designed to be suitable for the boilers. The same is true of the ancillary pipework and electrical cables. 64. As to the water-cooling systems at CPPS B and BPPS, the Lands Tribunal held (paragraph 234) that they fall within the definition of “plant”, and the fact that they are (as was said by Mr Jupp, one of CLP’s experts) designed to suit the capacity of the steam turbines does not make them part of the steam turbines. It appears from Mr Jupp’s report, mentioned in paragraph 9.30 of the judgment of Cheung JA, that each system includes a pump house with a pump extracting water from the sea. The pumps are no doubt machines, but the Order of the Court of Appeal (paragraph 2(2)) indicates that the pump houses are not treated as part of the systems. In any event, the water-cooling systems as a whole cannot be regarded as a machine, or part of a machine, falling within section 8(b). 65. Apart from the issue of the supporting steelwork at BPPS, therefore, I would not make a remission to the Lands Tribunal. It is unfortunate that because of their construction of sections 8 and 8A (see paragraph 46 above) the Lands Tribunal did not make clear, specific findings on these points. But the answers are clear and in all the circumstances a remission would not be appropriate. 66. It remains to consider the supporting steelwork at BPPS. I have already described how the issue was left unresolved before the Lands Tribunal, even on review. The matter came before the Court of Appeal under an order (at A6 of the Record) which refers in general terms to “non-rateable machinery”. CLP’s notice of appeal (at A8 of the Record) refers to paragraph 227 of the Lands Tribunal’s first judgment without drawing attention to any special problem about steelwork at BPPS. The Commissioner’s respondent’s notice (at A9 of the Record) refers to (inter alia) “the boiler supporting steelwork at CPPS ‘B’ and BPPS.” 67. It is correct (as stated in a letter dated 3 March 2017 from CLP’s solicitors) that CLP’s skeleton argument in the Court of Appeal (at B5 of the Record, and apparently objected to by CLP) referred in paragraph 3(1) only to “the integral supporting steelwork from which the boilers hang at Castle Point (sic) Power Station B” and footnote 2 stated that “the Commissioner has not contested that at BPPS, the heat recovery boilers and their supporting steelwork are all non-rateable.” (This sentence is misquoted in the solicitors’ letter of 3 March 2017, possibly through confusion of “contested” and “contended”.) But this assertion was directly challenged in paragraph 1 and footnote 1 of the Commissioner’s skeleton argument (not included in the Record, but annexed to a letter dated 3 March 2017 from the Department of Justice). 68. Despite this direct conflict between the parties’ skeleton arguments, the Court of Appeal appears to have been unware that the identification of the disputed items was itself a contentious issue. Paragraph 1.2 of the judgment (although purporting to set out the Commissioner’s view) seems to have been taken word for word from CLP’s skeleton argument. The difference between the parties on this point is not mentioned in paragraph 1, paragraphs 9.30 to 9.32, or elsewhere in the judgment. 69. The solicitors’ letter of 3 March 2017 raises various points of fact about coal-fired and gas-fired boilers. There are no findings of fact about these points in either of the Lands Tribunal’s judgments and it is not appropriate for this Court to enquire into them now, not least because the reference year 2004-5 is now so long ago. Moreover, the Court of Appeal’s order at A10 of the Record, drafted (as the Court has been informed in the Department of Justice’s letter of 3 March 2017) by CLP’s solicitors without reference to the Commissioner, does not set aside or direct remitter of the Lands Tribunal’s judgment so far as it affects the structural steelwork at BPPS. The judgment of the Lands Tribunal on that issue has remained in force throughout. The right course is for this Court to leave it in force. 70. I would therefore allow this appeal and make an order nisi for CLP to pay the Commissioner’s costs of this appeal. Any submissions as to costs or as to the form of the order should be made in writing to be served and lodged with the Registrar within 21 days after the handing down of this judgment, and any submissions in reply should be made in writing, to be served and lodged within 14 days afterwards. In the absence of any submissions as to costs, the order nisi will become absolute. Chief Justice Ma: 71. For the reasons contained in the judgment of Lord Walker of Gestingthorpe NPJ, this appeal is allowed. The orders as to costs and the form of the formal order are as set out in paragraph 70 above. Mr Benjamin Yu SC and Ms Yvonne Cheng SC, instructed by the Department of Justice, for the Appellant Mr David Elvin QC and Mr John Litton, instructed by Holman Fenwick Willan, for the Respondent 
A. INTRODUCTION 1. A member of the Legislative Council (“LegCo”) is constitutionally mandated under Article 104 (“BL104”) of the Basic Law (“BL”), upon assuming office, to take an oath to swear to uphold the BL and to swear allegiance to the Hong Kong Special Administrative Region of the People’s Republic of China. 2. For the purpose of BL104, a LegCo member is required under section 19 of the Oaths and Declarations Ordinance (Cap 11) (“ODO”) to take the Legislative Council Oath (“LegCo Oath”)in the form as prescribed under section 16(d) and Schedule 2 thereof. Part IV of Schedule 2 prescribes the form of the LegCo Oath as follows: “THE LEGISLATIVE COUNCIL OATH I swear that, being a member of the Legislative Council of the Hong Kong Special Administrative Region of the People’s Republic of China, I will uphold the Basic Law of the Hong Kong Special Administrative Region of the People’s Republic of China, bear allegiance to the Hong Kong Special Administrative Region of the People’s Republic of China and serve the Hong Kong Special Administrative Region conscientiously, dutifully, in full accordance with the law, honestly and with integrity. (name of person making the oath)” (emphasis added) 3. Section 21 of the ODO further provides that if a LegCo member “declines or neglects” to take the LegCo Oath “when requested” to do so, he “shall” (meaning he must) vacate the office if he has already entered on it, or he “shall” be disqualified from entering on the office if he has not. 4. In the present cases, Mr Leung and Ms Yau were elected in their respective constituencies in the general election held in September this year to be a LegCo member. As mandated under BL104 and section 19 of the ODO, they were asked to take the LegCo Oath before the Clerk to the LegCo (“the Clerk”) at the first meeting of the LegCo on 12 October 2016. 5. There is no dispute that they purported to take the LegCo Oath before the Clerk on that occasion in the following way and manner: (1) Each of them used the term “Hong Kong nation” right at the outset of oath-taking: (a) Mr Leung declared in open public that he shall keep guard over the interest of the Hong Kong nation; (b) Ms Yau declared in open public that she will be faithful and bear true allegiance to the Hong Kong nation. (2) The Clerk interrupted each of them and said he could not administer their respective oath-taking as that was not taken in compliance with the LegCo Oath. (3) Each of them then purported to take the oath again. (4) In doing so, each of them mis-pronounced the word “China” consecutively for three times, as “Geen-na” or “Sheen-na” (“支那”). (5) Further, Ms Yau mis-pronounced “People’s Republic of China” as “the People’s Refucking of Sheen-na” consecutively for three times. (6) Each of them also intentionally unfolded and displayed a blue banner bearing the words “HONG KONG IS NOT CHINA”. (7) Mr Leung adopted a contrast in the tone of his voice between his initial words before the interjection by the Clerk and his subsequent words after such interjection (which shows a dismissive and not serious attitude). He further crossed the index and middle fingers of his right hand over the Bible in seeking to take the oath again after the initial interjection by the Clerk. (8) Ms Yau emphasized “Hong Kong” with a distinctly loud tone of voice but adopted a lower voice and hurried manner for the rest of the oath. 6. Moreover, Mr Leung said to the press immediately after the LegCo meeting on 12 October 2016 as follows: “今日我哋 [referring to himself and Ms Yau] 成個誓詞嗰個部份我哋有，我自己啦，我有3 part係做咗啦，呀，有個手勢啦，有一個係呀披肩咁樣嘅野啦，亦都有一個說法，咁而3樣野係一個組合嚟嘅”. 7. In light of the above undisputed way and manner in which Mr Leung and Ms Yau purported to take the oath, on 18 October 2016, the President of the LegCo (“the President”) decided (on legal advice) that the oath purported to be taken by each of them on that day is invalid. However, the President then further decided (also on legal advice) to allow each of them to re-take the LegCo Oath at the next LegCo meeting if they made a written request to do so. I would call this second part of the decision for convenience as “the President’s Decision”. 8. Mr Leung and Ms Yau later on that day did make a written request to re-take the oath. 9. The Chief Executive of the HKSAR (“the CE”) and the Secretary for Justice (“the SJ”) (collectively, “the applicants”) however contend that Mr Leung and Ms Yau already declined or neglected to take the LegCo Oath on 12 October 2016. As a result, under section 21 of the ODO, as a matter of law, they must be regarded as having vacated their office of a LegCo member or alternatively be disqualified from entering on it. On the same basis, the CE and the SJ also say that the President misdirected himself on the law in making the President’s Decision as there is no question that Mr Leung and Ms Yau could re-take the oath again after they have so vacated the office or having been disqualified from entering on it. 10. Thus, on the same day of the President’s Decision (ie, 18 October 2016), the CE and the SJ commenced the present action under HCMP 2819/2016 against Ms Yau, Mr Leung and the President; and this judicial review against the President (joining Mr Leung and Ms Yau as interested parties).[1] 11. In HCMP 2819/2016, the applicants seek against Ms Yau and Mr Leung (respectively as the 1st and 2nd defendants) the principal reliefs of (a) a declaration that the oaths taken by them on 12 October 2016 contravened BL104 and the relevant provisions of the ODO, and are invalid and void as the LegCo Oath as required under the law; (b) a declaration that they are disqualified from assuming or entering on the office of a LegCo member, or have vacated that office, and are not entitled to act as a LegCo member; and (c) injunctions to restrain them from acting or claiming to be entitled to act as a LegCo member. 12. In the judicial review, the applicants seek against the President the principal reliefs of (a) a declaration that the President has no power to re-administer or allow for re-administration of any oaths to be taken by Mr Leung and Ms Yau as they are disqualified from assuming or entering on the office of a LegCo member or have already vacated it; (b) a declaration that the office of member of the LegCo previously occupied by each of Mr Leung and Ms Yau is now vacant; (c) an order of certiorari to quash the President’s Decision; and (d) an injunction to restrain the President from administering or allowing to be administered the making of oaths of Mr Leung and Ms Yau under section 19 of the ODO. 13. It is important to note that, notwithstanding the CE and the SJ’s above forefront contentions made in these applications, Mr Leung and Ms Yau have not put forward any positive case by way of submissions (primary or alternative) or evidence that the oaths they purported to take on 12 October 2016 complied with BL104 or the requirements under the ODO. They also have not put forward any positive arguments by way of submissions or evidence that they did not “decline or neglect” to take the LegCo Oath when requested to do so on that day. 14. Instead, Mr Leung and Ms Yau oppose these applications on two principal grounds. They are that (a) matters concerning the oath taking by a LegCo member and the validity thereof are “internal business” of the LegCo and, under the non-intervention principle in common law, the court should and could not intervene in these matters or any decisions made by the President or the Clerk relating to them; and (b) the words spoken in relation to the “oaths” taken by Mr Leung and Ms Yau are protected by the immunity provided under BL77 and sections 3 and 4 of the Legislative Council (Powers and Privileges) Ordinance (Cap 382) (“LCPPO”), and they are therefore immune from suit, including the present ones. Mr Leung and Ms Yau have also raised some other ancillary arguments in aid of the above principal grounds, including an objection that the CE has no locus to bring these proceedings. I would elaborate on all these grounds and arguments in greater detail later. 15. Insofar as the President is concerned, his only objection to these proceedings is that he should not be joined as a party. It is submitted that he has not made any substantive decision as to whether Mr Leung and Ms Yau have declined or neglected to take the LegCo Oath, and the President’s Decision is only a procedural decision in nature. There are thus no extant substantive decisions that are amenable to judicial review. The judicial review should therefore not be brought against him. Further, in relation to HCMP 2819/2016, it is said that there are no suggestions that the President would not abide by the ruling of this court if it decides to grant the reliefs as sought against Mr Leung and Ms Yau. In that case, there is also no basis for the CE and the SJ to join the President in this action. 16. Before I deal with all these grounds of opposition, it is important for me to first look at the relevant constitutional and statutory frameworks and their proper meanings which provide for the legal context mandating a LegCo member to take the LegCo Oath and as to when he would be disqualified in failing to do so. B. THE RELEVANT CONSTITUTIONAL AND STATUTORY FRAMEWORKS B1. The constitutional framework 17. The BL is the written mini-constitution for Hong Kong. BL104 provides as follows: “When assuming office, the Chief Executive, principal officials, members of the Executive Council and of the Legislative Council, judges of the courts at all levels and other members of the judiciary in the Hong Kong Special Administrative Region must, in accordance with law, swear to uphold the Basic Law of the Hong Kong Special Administrative Region of the People's Republic of China and swear allegiance to the Hong Kong Special Administrative Region of the People's Republic of China.” (emphasis added) 18. It is thus a constitutional requirement under BL104 that LegCo members (as well as the CE, principal officials, Executive Council members and judges of the Hong Kong Special Administrative Region) are mandatorily required to take an oath when assuming office. 19. On 7 November 2016, the Standing Committee of the National People’s Congress of the People’s Republic of China (“NPCSC”) in the exercise of its power under BL158(1) pronounced an interpretation (“the Interpretation”) of the meaning of BL104. The Interpretation is as follows: (1) Oath-taking is the legal prerequisite and required procedure for public officers specified in the Article to assume office. No public office shall be assumed, no corresponding powers and functions shall be exercised, and no corresponding entitlements shall be enjoyed by anyone who fails to lawfully and validly take the oath or who declines to take the oath. (2) Oath-taking must comply with the legal requirements in respect of its form and content. An oath taker must take the oath sincerely and solemnly, and must accurately, completely and solemnly read out the oath prescribed by law, the content of which includes “will uphold the Basic Law of the Hong Kong Special Administrative Region of the People’s Republic of China, bear allegiance to the Hong Kong Special Administrative Region of the People’s Republic of China”. (3) An oath taker is disqualified forthwith from assuming the public office specified in the Article if he or she declines to take the oath. An oath taker who intentionally reads out words which do not accord with the wording of the oath prescribed by law, or takes the oath in a manner which is not sincere or not solemn, shall be treated as declining to take the oath. The oath so taken is invalid and the oath taker is disqualified forthwith from assuming the public office specified in the Article. (4) The oath must be taken before the person authorized by law to administer the oath. The person administering the oath has the duty to ensure that the oath is taken in a lawful manner. He or she shall determine that an oath taken in compliance with this Interpretation and the requirements under the laws of the Hong Kong Special Administrative Region is valid, and that an oath which is not taken in compliance with this Interpretation and the requirements under the laws of the Hong Kong Special Administrative Region is invalid. If the oath taken is determined as invalid, no arrangement shall be made for retaking the oath. 20. Although the courts in Hong Kong are authorised by BL158(2) and (3) to interpret provisions of the BL in adjudicating cases, the NPCSC has the final power of interpretation of the BL as provided under BL158(1) and (3) and Article 67(4) of the Constitution of the People’s Republic of China. An interpretation made by the NPCSC is binding on all the courts of Hong Kong, and the courts are under a duty to follow it. This has been well recognised by the Court of Final Appeal in Director of Immigration v Chong Fung Yuen (2001) 4 HKCFAR 211, where Li CJ said at pp 222G and H as follows: “…where the [NPCSC] has made an interpretation of the [BL] pursuant to its power under art 67(4) of the Chinese Constitution and art 158 of the Basic Law, the courts in Hong Kong are under a duty to follow it. … That power of the [NPCSC] extends to every provision in the Basic Law and is not limited to the excluded provisions referred to in art 158(3).” 21. In the premises, the Interpretation is binding on this court. In this judgment, unless otherwise stated, whenever I refer to the meaning of BL104, it is a reference to the meaning as set out in the Interpretation. 22. Hence, essentially for the present purposes, under BL104, a LegCo member when assuming office must take the LegCo Oath as prescribed under the ODO (being the laws of the Hong Kong Special Administrative Region).[2] He shall take it solemnly and sincerely and in compliance with the LegCo Oath both in substance and in form. If he intentionally declines or fails to so take the LegCo Oath, whether in form or in substance, the oath taken is invalid and he shall be disqualified from assuming the office. 23. On the other hand, as submitted by Mr Yu SC for the CE and the SJ, independent of the Interpretation, the laws of Hong Kong as set out in the relevant provisions ODO, when properly construed, indeed carry effectively the same meanings and legal effects as these essential meanings of BL104. This is what I would turn to next. B2. The ODO 24. Sections 16 and 19 of the ODO provide as follows: “Section: 16 Forms of Oaths The Oaths referred to in this Ordinance as- … (d) the Legislative Council Oath; … shall be in the respective forms set out in Schedule 2. … Section: 19 Oath of Legislative Councillors A member of the Legislative Council shall, as soon as possible after the commencement of his term of office, take the Legislative Council Oath which- (a) if taken at the first sitting of the session of the Legislative Council immediately after a general election of all members of the Council and before the election of the President of the Council, shall be administered by the Clerk to the Council; (b) if taken at any other sitting of the Council, shall be administered by the President of the Council or any member acting in his place.” (emphasis added) 25. Section 21 of the ODO further provides for the legal consequence of non-compliance by a person who is requested to take an oath as required under the ODO as follows: “Section: 21 Consequence of non-compliance Any person who declines or neglects to take an oath duly requested which he is required to take by this Part, shall- (a) if he has already entered on his office, vacate it, and (b) if he has not entered on his office, be disqualified from entering on it.” (emphasis added) 26. The construction of the meaning of these provisions of the ODO is for this court, and under common law the court should adopt a purposive construction, read in their proper context of the whole statue.[3] 27. In this respect, as rightly submitted by Mr Yu, on a proper purposive construction of sections 16, 19 and 21 of the ODO, they have the following meanings and legal effects in relation to the taking of the LegCo Oath by a LegCo member. 28. First, the relevant provisions in the ODO are to reflect and underline the constitutional requirements in BL104. If a LegCo member swears his oath in a manner or form that is inconsistent with the requirements set out in the ODO, his oath offends the constitutional requirements under BL104: Leung Kwok Hung, supra, at paragraph 22, per Hartmann J. 29. Second, a LegCo member has to take the LegCo Oath “as soon as possible” after the commencement of his term of office, and that must be done before he assumes his office. This is borne out by the plain and express words of BL104 and section 16 of the ODO, and supported by Leung Kwok Hung, supra, at paragraph 22. 30. Third, a LegCo member must take the oath in the same form, manner and substance as the LegCo Oath as expressly prescribed under section 19 and Schedule 2 of the ODO. It is not any oath that may be devised by a LegCo member provided that the “oath” meets the two obligations set out in BL104: Leung Kwok Hung, supra, at paragraphs 24 - 27. A fortiori, if the “oath” taken does not even meet any of the two obligations set out in BL104, it is not a valid oath. 31. Fourth, an oath is a solemn declaration. In its original form it was invariably a promise to one’s deity. The ODO does not allow for any real difference in the form and substance of the oath itself as to how a person wishes to take it. Hence, a LegCo member who seeks to alter the form, manner or substance of the oath when taking it will offend BL104 and therefore be unlawful and of no effect: Leung Kwok Hung, supra, paragraphs 36 - 40. 32. An oath must be taken solemnly and sincerely is consistent with the trite position under common law that taking an oath is a form of attestation by which a person signifies that he is bound in conscience to perform an act faithfully and truthfully. Similarly an affirmation in lieu of oath binds a person to live by what he undertakes. In particular, an oath of allegiance or loyalty means that a person promises and binds himself to bear true allegiance to a particular sovereign and government and to support its constitution.[4] The obvious purpose is to ensure the person concerned (such as a member of legislature) makes a commitment to live by the constitutional process; he has to owe allegiance to the constitution, and he has to uphold the sovereignty and integrity for the country. It is not a mere formality or empty form of words. The purpose of prescribing the form is to induce a sense of subordination to the constitution in men of all faiths. It has to be followed “in letter and spirit”. See: Haridasan Palayil v The Speaker, Kerala Legislative Assembly AIR 2003 Ker 328, 2003 (3) KLT 119 at paragraphs 21, 24, 30 and 31; AG v Bradlaugh (1885) 14 QBD 667 at 696 - 698, per Brett MR, at 707 - 709, per Cotton LJ, and at 716 - 717, per Lindley LJ. 33. In the premises, the fundamental and essential question to be answered in determining the validity of the taking of an oath is whether it can be seen objectively that the person taking the oath faithfully and truthfully commits and binds himself or herself to uphold and abide by the obligations set out in the oath. 34. Fifth, read together with the above provisions and context, section 21 of the ODO when properly construed means that if a LegCo member “declines or neglects” to take the LegCo Oath whether in form or in substance when he is requested to do so, he “shall” (thus must)vacate his office if he has entered upon it, or must be disqualified from entering upon his office if he has not done so. In other words, under the clear and express words of section 21 of the ODO, once it is shown that a LegCo member declines or neglects to take the LegCo Oath when requested to do so, he can no longer continue to be qualified as a LegCo member and act as such; in which case, there is no question that he could seek to take the oath again. 35. In this respect: (1) The ordinary and natural meaning of the word “decline” is to refuse or object.[5] This is consistent with the natural meaning of its Chinese translation “拒絕”. Read in the context of section 21 of the ODO, which entails the serious consequence of disqualification, the word refers to an intentional act to refuse or object to the taking of the oath as prescribed by law. As submitted by Mr Yu, such an act would also include conduct which objectively manifests itself an intentional refusal. (2) The word “neglect” has been defined as “an omission to perform a duty which the person owing the duty is able to perform”.[6] Again, when properly read in the context of section 21 of the ODO, it should mean a deliberate or wilful (in contrast to an inadvertent or accidental) omission to perform the duty to take the oath as prescribed when requested to do so. 36. The above proper meanings and effects of the relevant ODO provisions are reached by application of common law principles of construction and supported by authorities, independent of the Interpretation. They are in any event not disputed or challenged in any material way by counsel for the President, Mr Leung and Ms Yau whether by way of primary or any alternative submissions. C. WHETHER MR LEUNG AND MS YAU DECLINED TO TAKE THE LEGCO OATH WHEN REQUESTED TO DO SO OR AS REQUIRED UNDER BL104 37. As mentioned above, Mr Leung and Ms Yau in these proceedings have not put forward any positive arguments that they did not decline or neglect to take the LegCo Oath when requested to do so on 12 October 2016. 38. The applicants submit that objectively viewed at, it must be clear and beyond debate that Mr Leung and Ms Yau have declined to take the LegCo Oath when requested to do so. I agree with the applicants as that is incapable of dispute in light of the following. 39. The objective and undisputed evidence shows that Mr Leung and Ms Yau had been duly requested to take the LegCo Oath at the first meeting of the LegCo on 12 October 2016. They had also been reminded repeatedly that they must take the oath in accordance with the form as prescribed by the ODO.[7] 40. Further, objectively viewed against the undisputed acts and manner in how Mr Leung and Ms Yau purported to take the oath on 12 October 2016 as summarised in paragraphs 5 and 6 above, they must be regarded as having manifested an intentional refusal (and thus their “decline”) to take the LegCo Oath, whether in form or in substance. My reasons are these. 41. It is plain that, in referring to “Hong Kong nation” when they first sought to take the oath, and later in mis-pronouncing the words “Republic” and “China”, they did not take the oath in the same form of the LegCo Oath as prescribed by law, and they did that intentionally. Mr Leung and Ms Yau have not suggested otherwise. 42. Further, as submitted by Mr Yu, they also did not objectively intend to take the LegCo Oath in substance as they did not seriously, faithfully and truthfully intend to abide by a declaration to pledge allegiance to the Hong Kong Special Administrative Region of the People’s Republic of China and to uphold the BL. It is so as in adopting the way and manner they purported to take oath, they expressed the position that they did not recognise the fundamental constitutional model of “one country, two systems”, which underpins and underlines the BL, and under which the Hong Kong Special Administrative Region was established. 43. As recognised by the Court of Final Appeal in Ng Ka Ling v Director of Immigration (1999) 2 HKCFAR 4 at 28G-H, the purpose of the BL is to establish the Hong Kong Special Administrative Region being an inalienable part of the People’s Republic of China under the principle of “one country, two systems” in accordance with China’s basic policies regarding Hong Kong as set out and elaborated in the Joint Declaration. The Court of Final Appeal also recognises the fundamental importance of the “one country” in this model as observed by Li CJ in HKSAR v Ng Kung Siu (1999) 2 HKCFAR 442, at 460C-E as follows: “… In these circumstances, the legitimate societal interests in protecting the national flag and the legitimate community interests in the protection of the regional flag are interests which are within the concept of public order (ordre public). As I have pointed out, the national flag is the unique symbol of the one country, the People’s Republic of China, and the regional flag is the unique symbol of the Hong Kong Special Administrative Region as an inalienable part of the People’s Republic of China under the principle of ‘one country, two systems’. These legitimate interests form part of the general welfare and the interests of the collectivity as a whole.” 44. Indeed, as emphasized by Mr Yu, the theme and policy of “one country, two systems” runs throughout the BL, as reflected in particular in the following articles: (1) The Preamble states that Hong Kong has been part of the territory of China since ancient times. The resumption of the exercise of sovereignty by China over Hong Kong with effect from 1 July 1997 fulfils the long cherished common aspiration of the Chinese people for the recovery of Hong Kong. Upholding national unity and territorial integrity, and maintaining the prosperity and stability of Hong Kong, the People’s Republic of China has decided that upon China’s resumption of the exercise of sovereignty over Hong Kong, a Hong Kong Special Administrative Region will be established and that under the principle of “one country, two systems”, the socialist system and policies will not be practiced in Hong Kong. The BL was enacted by the NPCSC in accordance with the Constitution of the People’s Republic of China. (2) The Preamble, BL1 and 12 all state that the Hong Kong Special Administrative Region is an inalienable part of the People’s Republic of China. It shall be a local administrative region of the People’s Republic of China which shall enjoy a high degree of autonomy and come directly under the Central People’s Government (“CPG”). (3) BL 16, 17 and 19 provide the constitutional foundation for Hong Kong Special Administrative Regions to enjoy executive, legislative and judicial power; and BL66 provides the constitutional foundation that the LegCo of the Hong Kong Special Administrative Region shall be the legislature of the Region. (4) The importance of territorial integrity is also underscored by BL23, which imposes a constitutional obligation on the Hong Kong Special Administrative Region to enact laws on its own to, among others, prohibit secession, sedition, subversion against the CPG. 45. Bearing the above in mind, again as rightly submitted by Mr Yu, the manner and way in which Mr Leung and Ms Yau purported to take the LegCo Oath on 12 October 2016 demonstrate objectively and clearly their intention not to recognise this fundamental constitutional model of “one country, two systems” and the importance under this model of the “one country”: (1) In the case of Ms Yau, she used an “f” word to replace the word “Republic”. The inevitable inference is the contempt she showed for the People’s Republic of China as the “one country” in the “one country, two systems” concept which is fundamental to the BL and her repudiation of any allegiance to the Hong Kong Special Administrative Region as an inalienable part of the People’s Republic of China. She has not sought to suggest otherwise in these proceedings. (2) The deliberate mis-pronunciation of China as “Geen-na” or “Sheen-Na” (“支那”) and the use of the banner bearing “Hong Kong is NOT China” by both Mr Leung and Ms Yau assumes particular significance: (a) The unchallenged evidence is that the meaning of “支那” has a historical background and is widely or commonly understood to carry a derogatory, disparaging meaning and is used in mockery when referring to China and the People’s Republic of China; was used by the Japanese to refer to China during the Japanese invasion; and was used to mock China in an offensive way.[8] (b) By seeking to make a mockery of China and the People’s Republic of China in a derogatory and humiliating manner, it is objectively plain that Mr Leung and Ms Yau refused to pledge allegiance to the Hong Kong Special Administrative Region as an inalienable part of the People’s Republic of China. (c) The unchallenged evidence is also that the term “支那人” is commonly used by anti-Chinese organizations such as “台獨”, “港獨” groups to refer derogatorily or disparagingly to Chinese people.[9] Mr Leung and Ms Yau thus conveyed the message that they advocated independence of Hong Kong (港獨), as exemplified in their reference and pledge to the “Hong Kong nation”. (d) The unchallenged evidence is further that “Hong Kong is NOT China” also has reference to the use of the same banner at a local football match between Hong Kong and China held on 17 November 2015, which expresses the meaning, in context, that the Hong Kong Special Administrative Region is not part of the People’s Republic of China.[10] (3) The expression “支那” was uttered three times, by each of Mr Leung and Ms Yau. It cannot be the result of mere inadvertence, ignorance or mistake. Objectively looked at, and coupled with the absence of any explanations (let alone any credible explanations) by them, it was part of the wilful and deliberate attempt by both of them (sharing the same political affiliation) to insult China and the People’s Republic of China, advocate the political message of “港獨” and make a mockery of the contents of the LegCo Oath, by praying in aid the derogatory meaning of the expression “支那” commonly subscribed to it by the general public and the overseas Chinese community.[11] Mr Leung and Ms Yau have not sought to suggest otherwise in these proceedings. (4) Mr Leung in crossing his index and middle fingers with his hand placed over the Bible in purporting to take the oath must objectively be regarded as intending to send the message of not taking the oath seriously or with a clear conscience, and that he was intending to tell a lie or to signify that the oath he was making should be invalidated.[12] Mr Leung has not suggested otherwise. (5) It is also obvious that Mr Leung and Ms Yau acted in concert and did so deliberately: see paragraphs 6 and 39 above. (6) In the midst of repeated reminders, the manner in which Mr Leung and Ms Yau took their oath demonstrates that they not only refused and deliberately failed to take the LegCo Oath as an oath, with due solemnities of the mode of taking an oath,[13] but also manifested a clear intention not to be bound in conscience to perform faithfully and truthfully the oath as required by BL104 and the ODO. Again, Mr Leung and Ms Yau have not sought to suggest otherwise in these proceedings. 46. In the circumstances, as submitted by the applicants, Mr Leung and Ms Yau and each of them manifestly refused (and thus declined) to solemnly, sincerely and truly bind themselves to uphold the BL or bear true allegiance to the Hong Kong Special Administrative Region of the People’s Republic of China. Alternatively, at the least, they must have wilfully omitted (and hence neglected) to do so. 47. Section 21 of the ODO should therefore apply and operate to disqualify Mr Leung and Ms Yau from continuing to be a LegCo member, and the court should proceed to declare as such. 48. However, Mr Pun SC and Mr Dykes SC respectively raise a number of grounds for Mr Leung and Ms Yau to submit that the court should not and cannot intervene in these matters. I will now look at these grounds in turn. D. MR LEUNG AND MS YAU’S GROUNDS OF OPPOSITION D1. The court should not intervene in these cases under the non-intervention principle 49. Under this ground, Mr Pun argues that the taking of the LegCo Oath by LegCo members and the President’s Decision to allow Mr Leung and Ms Yau to take the oath again belong to the “internal business” of the LegCo. In the premises, under the established non-intervention principle, the court should and could not intervene in them. In support of these submissions, Mr Pun relies heavily and principally on the authority of Bradlaugh v Gosset (1884) 12 QBD 271. 50. With respect, I am unable to agree with Mr Pun for the following reasons. 51. The non-intervention principle has its origin in common law and is premised on the doctrine of separation of powers in England, where there is no written constitution, and where there is supremacy of the Parliament. However, the scope of this principle as applied in a different jurisdiction must be understood in and limited to the proper context of that jurisdiction, in particular where there is a written constitution.[14] Thus, in present day Hong Kong, where there is the written constitution of the BL and where the BL is supreme instead of the legislature, the court does have jurisdiction under the BL to determine, by declaratory relief, questions such as whether the internal Rules of Procedure enacted by the LegCo (which would be regarded as “the internal matters” of the Parliament under the non-intervention principle in the UK) are consistent with the BL. This has been explained by A Cheung J (as the learned CJHC then was) in Cheng Kar Shun v Li Fung Yung [2011] 2 HKLRD 555 at paragraph 217 as follows: “217. After 1997, in Leung Kwok Hung v President of Legislative Council [2007] 1 HKLRD 387, a case concerning the prohibition of members of the Legislative Council from introducing bills to the Legislative Council which have a charging effect under art.74 of the Basic Law, Hartmann J (as he then was) noted that being subordinate to the Basic Law, the Legislative Council must act in accordance with that law. His Lordship pointed out that in the United Kingdom, Parliament is supreme. The courts there are confined to interpreting and applying what Parliament has enacted. Parliament has exclusive control over the conduct of its own affairs. The courts will not permit any challenge to the manner in which Parliament goes about its business. If there are irregularities, that is a matter for Parliament to resolve, not the courts. However, in Hong Kong, the Basic Law is supreme. But subject to that, the Basic Law recognises the Legislative Council to be a sovereign body under that law. In setting Rules of Procedure to govern how it goes about the process of making laws, provided those rules are not in conflict with the Basic Law, the Legislative Council is ‘answerable to no outside authority’. The learned Judge concluded that so far as jurisdiction is concerned, the courts of the Hong Kong Special Administrative Region do have jurisdiction under the Basic Law to determine, by way of declaratory relief, whether Rules of Procedure enacted by the Legislative Council are consistent with the Basic Law. Yet, it is a jurisdiction which, having regard to the sovereignty of the Legislative Council under the Basic Law, ‘should only be exercised in a restrictive manner’. See p.390 para.5, p.391 paras.9-10, p.393 para.24, p.394 paras.28, 31.” (emphasis added) 52. In this respect, the scope of the non-intervention principle as applied in Hong Kong in the context of BL has recently been authoritatively explained by the Court of Final Appeal in Leung Kwok Hung v The President of the Legislative Council (No 1) (2014) 17 HKCFAR 689 (“Leung Kwok Hung CFA”), where the Court said theseat paragraphs 28 - 32 and 39: “28. In construing and applying the provisions of the BL, it is necessary not only to apply common law principles of interpretation but also principles, doctrines, concepts and understandings which are embedded in the common law. They include the doctrine of the separation of powers and, within it, the established relationship between the legislature and the courts. This relationship includes the principle that the courts will recognise the exclusive authority of the legislature in managing its own internal processes in the conduct of its business, in particular its legislative processes. The corollary is the proposition that the courts will not intervene to rule on the regularity or irregularity of the internal processes of the legislature but will leave it to determine exclusively for itself matters of this kind (‘the non-intervention principle’). 29. The strength of this proposition rests not only on principle and authority but also on public policy. In Hong Kong, LegCo has as its primary responsibility its law-making function. It also has vested in it other important powers and functions under art 73, for example: ‘(2) To examine and approve budgets introduced by the government; (3) To approve taxation and public expenditure; (4) To receive and debate the policy addresses of the Chief Executive; (5) To raise questions on the work of the government; (6) To debate any issue concerning public interests;’ 30. The important responsibilities of LegCo, notably its law-making function, require, as with other legislatures, that it should be left to manage and resolve its own internal affairs, free from intervention by the courts and from the possible disruption, delays and uncertainties which could result from such intervention. Freedom from these problems is both desirable and necessary in the interests of the orderly, efficient and fair disposition of LegCo’s business. 31. The adoption of the principle of non-intervention by the courts will reduce, if not eliminate, the prospect of pre-enactment challenge to proceedings in LegCo. It will also reduce, if not eliminate, post-enactment challenges to the validity of laws made by LegCo based on irregularity in its proceedings, unless such an irregularity amounts to non-compliance with a requirement on which the validity of a law depends. 32. In this respect it is important to recognise that the principle of non-intervention is necessarily subject to constitutional requirements. The provisions of a written constitution may make the validity of a law depend upon any fact, event or circumstance they identify, and if one so identified is a proceeding in, or compliance with, a procedure in the legislature the courts must take it under its cognizance in order to determine whether the supposed law is a valid law. In Australia, Cormack v Cope was such a case. There s 57 of the Australian Constitution provided a means of resolving a deadlock between the two Houses of Parliament culminating in a joint sitting of the two Houses to deliberate and vote upon a proposed law. But the section prescribed a procedure to be followed and compliance with that procedure was a condition of the validity of the proposed law when enacted. … 39. This qualification arises from the circumstance that, in the case of a written constitution, which confers law-making powers and functions on the legislature, the courts will determine whether the legislature has a particular power, privilege or immunity. In R v Richards, ex p Fitzpatrick and Browne,Dixon CJ, speaking for the High Court of Australia and with reference to the two Houses of the Australian Parliament, said: it is for the courts to judge of the existence in either House of Parliament of a privilege, but, given an undoubted privilege, it is for the House to judge of the occasion and of the manner of its exercise.” (emphasis added) 53. Distilled from these principles as laid down by the Court of Final Appeal are the following ones which are particularly relevant for the present purposes: (1) The principle of non-intervention as applied in Hong Kong is necessarily subject to the constitutional requirements of the BL. The provisions of the BL may make the validity of a law depend on any fact, event or circumstance they identify, and if one so identified is a proceeding in, or compliance with, a procedure in the legislature, the court must take it under its cognizance in order to determine whether the supposed law is a valid law. See Leung Kwok Hung CFA, paragraph 32. (2) Further, in the case where the written constitution which confers (as the BL in Hong Kong does) law-making powers and functions on the legislature, the court will determine whether the legislature has a particular power, privilege or immunity. See Leung Kwok Hung CFA, paragraph 39. (3) In the premises, what can be properly regarded as the “internal business” or “internal process” of the LegCo must be viewed under the above caveat prescribing the non-intervention principle in Hong Kong. 54. Bearing these principles in mind, it is clear to me that the non-intervention principle as applied in Hong Kong does not prohibit the court from determining the matters under challenge in these proceedings. 55. The constitutional requirement to take the oath under BL104 is of fundamental importance. All the most important public officials under the executive, legislative and judicial branches of the Hong Kong Special Administrative Region must take the oath as prescribed and in accordance with the laws set out in BL104 and the ODO before each of them can assume his or her office. As I have explained above, an intentional failure to do so would result in his or her disqualification from assuming the office. In the premises, given that BL104 (incorporating the ODO) has identified and provided for the legal parameters under which the oath-taking requirement is to be complied with, under the principle of non-intervention as explained in Leung Kwok Hung CFA, the court must still“take it under its cognizance in order to determine whether”that act of the oath-taking is valid and compliant with BL104 (together with the ODO). 56. In the same vein, the court must also have the jurisdiction to determine whether the Clerk or the President has the power under BL104 (and the ODO) to make a decision (such as the President’s Decision in the present case) to allow a LegCo member to take the LegCo Oath again if and when it is challenged that the member has already been disqualified in law from doing so under BL104 and the ODO. 57. When considered in the above context, the oath-taking by a LegCo member as mandated by BL104 and a decision made by the Clerk or the President relevant to that act also cannot be properly regarded as the “internal business” of the LegCo for the purpose of the non-intervention principle. 58. In respect, it must also be remembered that BL104 mandates constitutionally not just LegCo members but also the CE, principal officials, Executive Council members and judges of the Hong Kong Special Administrative Region to take the oath before each of them can assume the office. Further, section 21 of the ODO also provides the same consequence for non-compliance of the oath-taking for all these categories of persons, and not just the LegCo members. As such, it does not make sense that somehow only LegCo members (but not all the others) would be excluded from the courts’ jurisdiction to determine whether an oath is taken in compliance with the constitutional requirement implemented through the provisions of the ODO. 59. The authority of Bradlaugh v Gossett relied on by Mr Pun does not assist him. 60. In Gossett, Mr Bradlaugh was returned by election as a member of the House of Commons (ie, the Parliament). Under the Parliamentary Oaths Act 1866, he had to take a parliamentary oath in the House before the Speaker as prescribed by that Act. However, in light of the disruptive conducts he had had before in the House, the House passed a resolution “that the Serjeant-at-Arms do exclude Mr Bradlaugh from the House until he shall engage not further to disturb the proceedings of the House”. In other words, under the resolution, he could not enter the House to take the oath unless he had stopped any acts disturbing the proceedings of the House. Mr Bradlaugh applied to the court for an injunction to restrain the Serjeant-at-Arms from carrying out the resolution. In support of the application, Mr Bradlaugh argued that he had a legal right conferred by the Act to take the oath before the Speaker, and the resolution deprived him of that right. The court therefore should intervene and grant the injunction to protect his legal right provided by the Act. 61. However, the court (Coleridge CJ, Mathew J and Stephen J) refused to grant the injunction on the basis that the resolution was a matter relating to the internal management of the House, and thus the court had no power to intervene whether that resolution was valid or invalid in law. 62. Mr Pun submits that the circumstances of Gossett are similar to the present one. Leading counsel therefore submits that if, under the non-intervention principle, the court cannot intervene in the Parliament’s decision to prevent a member from exercising his statutory right to take the oath, a decision of the President to allow a member to take the LegCo Oath (as in the present case) must similarly be regarded under that principle as a matter relating to the internal management of the LegCo. The court therefore has no power to intervene. 63. I am unable to agree. Gossett is clearly distinguishable and does not apply to Hong Kong directly for two reasons. First, in Hong Kong, the binding authority on the scope and applicability of the non-intervention principle is Leung Kwok Hung CFA. I have already explained above why under the principles laid down in Leung Kwok Hung CFA, the courts have jurisdiction to intervene in the present matters. Second, and in any event, Gossett was decided in the context of parliamentary supremacy and the absence of a written constitution in England. These constituted an important rationale underlying the English court’s decision in that case. As explained above, Hong Kong is very different as we have a written constitution of the BL and legislature is not supreme. These are important material differences in determining the scope and extent of the non-intervention principle. Hence, one cannot simply apply that authority to Hong Kong. 64. Mr Pun also relies on the authorities of R v Chaytor [2011] 1 AC 684 and Baron Mereworth v Ministry of Justice [2012] Ch 325 to support his submissions for the application of the non-intervention principle in the present cases. These cases are similarly concerned with the discussions and determinations of the application of that principle in the UK. For the same reasons I have explained above, they are not directly applicable to Hong Kong and must be read subject to the principles laid down in Leung Kwok Hung CFA. They therefore also do not take Mr Pun’s case any further. 65. Finally, as an ancillary and supporting argument, Mr Pun says, given the non-intervention principle, on a proper interpretation of BL104 and sections 19 and 21 of the ODO, the person administering the oath is the final arbiter of the questions of (a) whether an oath taken is in compliance with the legal requirements; and (b) whether the oath taker has declined or neglected to take the oath when requested to do so. This reinforces, says Mr Pun, the submissions that these matters and questions are related to the “internal business” of the LegCo and hence not subject to the court’s scrutiny under the non-intervention principle. After the issue of the Interpretation, Mr Pun further submits that paragraph (4) of the Interpretation supports this construction. 66. There is no substance in these submissions. 67. Mr Pun’s submissions are in fact a circular one, which is premised on the correctness of his contention that the non-intervention principle applies in the present case. For the reasons I have explained above, that contention is incorrect. The submissions that the Clerk or the President is the final arbiter in relation to these questions therefore also fall away. 68. Further and in any event, it is plain and obvious that the courts of law in the society are regarded generally and in principle as the final arbiter in adjudicating questions of law, including important questions of constitutional compliance with the provisions of the BL and the questions of compliance with statutory provisions. This is also consistent with our constitutional model where our courts are given the powers and authority to adjudicate cases in accordance with the law: see BL81 - 85.[15] In the premises, a statutory provision could not be construed with the intention to oust the courts’ role in being the final arbiter of questions of law, unless it is clearly and expressly provided for. 69. In this respect, BL104 and sections 19 and 21 of the ODO simply have not provided in any express and clear way that the person who administers the oath is the final arbiter of the questions of the validity of the oath taken and whether someone has declined or neglected to take the oath and therefore has to vacate his office or to be disqualified from entering on it under section 21. Of course, as a matter of necessary implication, that person would have the power incidental to his duty to administer the oath to make a decision on those questions as and when circumstances may practically require. But that is different from saying that it is intended by BL104 or the provisions of the ODO that the person administering the oath is the final decision-maker on those questions. 70. Paragraph (4) of the Interpretation does not assist Mr Pun. It only provides that the person administering the oath has the power to determine whether “an oath taken is in compliance” with BL104 and laws of Hong Kong. It does not provide expressly or impliedly that that decision is a final one. It is consistent with the court’s above observation. 71. On a proper analysis and construction, the courts in Hong Kong must be objectively intended in BL104 and the provisions of the ODO to be final arbiter of those questions. 72. For all the above reasons, I am of the clear view that the non-intervention principle as applied in Hong Kong does not cover the matters under the present challenges in these proceedings. The courts therefore have jurisdiction to adjudicate them. 73. Alternatively, and in any event, I accept Mr Yu’s submissions that section 73 (“section 73”) of the Legislative Council Ordinance (Cap 542) (“LCO”) confers an express jurisdiction to the court to determine matters such as whether a LegCo member has been disqualified from being a LegCo member or has ceased to be one. In the premises, the non-intervention principle in any event should give way to this express jurisdiction conferred to the court by statute in relation to these matters. 74. Section 73 provides as follows: “Section: 73 Proceedings against persons on grounds of disqualification (1) An elector, or the Secretary for Justice, may bring proceedings in the Court against any person who is acting, claims to be entitled to act, as a Member on the ground that the person is disqualified from acting as such. (2) Proceedings under this section may not be brought after 6 months from the date on which the person concerned acted, or claimed to be entitled to act, as a Member. (3) If, in proceedings brought under this section, it is proved that the defendant acted as a Member while disqualified from acting in that office, the Court may- (a) make a declaration to that effect; and (b) grant an injunction restraining the defendant from so acting; and (c) order the defendant to pay to the Government such sum as the Court thinks appropriate, not exceeding $5000 for each occasion on which the person so acted while disqualified. (4) If, in proceedings brought under this section, it is proved that the defendant claimed to be entitled to act as a Cap 542 - Legislative Council Ordinance 47 Member while disqualified from acting in that office, the Court may- (a) make a declaration to that effect; and (b) grant an injunction restraining the defendant from so acting. … (7) Proceedings against a person on the ground that the person has, while disqualified from acting as, or claimed to have been entitled to act, as a Member may be brought only in accordance with this section. (8) For the purposes of this section, a person is disqualified from acting as a Member if the person- (a) is not qualified to be, or is disqualified from being, a Member; or (b) has ceased to hold office as a Member.” 75. As submitted by Mr Yu, section 73 therefore expressly gives jurisdiction to the court to determine proceedings brought by the SJ or an elector against a LegCo member who has been disqualified or who has ceased to be one but continues to act or claims to be entitled to act as a member. In the section 73 proceedings, the court can declare that the member so acts or seeks to act while disqualified, and restrain him from doing so. In considering granting those reliefs, the court must be entitled to determine the underlying essential question as to whether the member has been so disqualified.[16] This includes the question of whether a member has been so disqualified under section 21 of the ODO. 76. Thus, even if (which I disagree) the non-intervention principle were initially to cover the questions raised under the present actions, by section 73, the LegCo has expressly conferred jurisdiction to the court to determine them in the proceedings brought thereunder. The non-intervention principle must hence be subject and limited to that extent, and would not prevent the court from determining such a question under section 73 proceedings. 77. Mr Dykes and Mr Pun however argue that section 73 is objectively intended to apply only to those circumstances of disqualification as provided under section 15(1) of the LCO. To supplement this submission, Mr Pun also says section 15 intends to provide exhaustively the circumstances when a member can be disqualified from his office. 78. Section 15(1) provides as follows: “Section: 15 When Member ceases to hold office (1) A Member’s office becomes vacant if the Member- (a) resigns in accordance with section 14 or is taken to have resigned from that office in accordance with section 13; or (b) dies; or (c) subject to subsection (2), alters either the Member’s nationality or the fact as to whether the Member has a right of abode in a country other than the People’s Republic of China as declared under section 40(1)(b)(ii); or (d) is the President and has been found under the Mental Health Ordinance (Cap 136) to be incapable, by reason of mental incapacity, of managing and administering his or her property and affairs; or (e) is declared in accordance with Article 79 of the Basic Law to be no longer qualified to hold that office.” 79. With respect, I am unable to agree with Mr Dykes and Mr Pun: (1) Section 15 of the LCO does not provide expressly that the circumstances listed thereunder are all and the only circumstances when a member can be disqualified. There is also nothing in the submissions that can suggest why necessarily it should be implied that the provision intends to exhaustively provide for the circumstances of disqualification. In the premises, I cannot see how it could be construed objectively and purposively that this is intended to provide exhaustively all the circumstances of disqualification of a LegCo member. (2) Section 73 also does not provide expressly that the circumstances of disqualification are limited to only those under section 15 of the LCO. It is not even provided that section 73 is subject to section 15. Again, nothing has been submitted to show why it is necessary to imply that section 73 is intended only to cover disqualification circumstances provided under section 15 of the LCO. I therefore also cannot see how it could be construed objectively that section 73 is intended to cover only section 15’s disqualification circumstances. 80. I note that Mr Dykes and Mr Pun have also raised the point that even if section 73 proceedings can be brought against Mr Leung and Ms Yau in relation to the present challenges, the CE has no locus to bring a section 73 action in HCMP 2819/2016 when suing in his capacity as the CE. I think counsel is right in this part of the submissions. However, this cannot constitute a valid ground to oppose the section 73 action as the SJ (being a party entitled to bring the action under section 73) is also a plaintiff in the action. 81. In the premises, notwithstanding the non-intervention principle, the court must alternatively in any event have jurisdiction to look at and determine the underlying question of whether a LegCo member has been disqualified to be a member under section 21 of the ODO when section 73 proceedings are brought against him, as in the present case.[17] D2. Mr Leung and Ms Yau’s oath-taking conducts are immune from suit 82. Under this ground, Mr Dykes submits that what were said by Mr Leung and Ms Yau in the oath-taking process are protected by the immunity provided under BL77 and sections 3 and 4 of the LCPPO. Hence, Mr Leung and Ms Yau cannot be sued upon in relation to these matters, and the court cannot even look at them. 83. With respect, I do not accept these submissions. 84. BL77 provides as follows: “Article 77 Members of the Legislative Council of the Hong Kong Special Administrative Region shall be immune from legal action in respect of their statements at meetings of the Council.” 85. Sections 3 and 4 of the LCPPO provide as follows: “Section: 3 Freedom of speech and debate There shall be freedom of speech and debate in the Council or proceedings before a committee, and such freedom of speech and debate shall not be liable to be questioned in any court or place outside the Council. Section: 4 Immunity from legal proceedings No civil or criminal proceedings shall be instituted against any member for words spoken before, or written in a report to, the Council or a committee, or by reason of any matter brought by him therein by petition, Bill, resolution, motion or otherwise.” (emphasis added) 86. In its proper context and read together with BL104, the plain and ordinary meaning of the word “statements” used in BL77 is to mean statements made by a LegCo member in the course of official debates on the floor of the LegCo when exercising his powers and discharging his functions as a LegCo member. See also: A v The United Kingdom (App No 35373/97, 17 March 2003), ECHR, at paragraph 84. It cannot be seriously suggested that a member can enjoy such immunity even before he has validly assumed the office of a LegCo member, as it is only after the assumption of office that he could validly exercise his powers and functions as a LegCo member. 87. Thus, the word expressed by a LegCo member in purporting to take an oath (when he has not yet validly assumed office) cannot be properly regarded as falling within the meaning of “statements” in BL77. 88. Once understood that way, sections 3 and 4 of the LCPPO do not add anything further, as they must be read consistently with the constitutional provision of BL77. In fact, in my view, the plain words of sections 3 and 4 of the LCPPO[18] (as italicized above) make it even clearer that the immunity provided attaches only to words and speeches (spoken or written) in relation to debates in the LegCo meetings. 89. Moreover and in any event, as submitted by Mr Yu, section 73 also constitutes an alternative answer to Mr Dykes’ arguments based on the immunity provided under BL77 and sections 3 and 4 of the LCPPO. 90. By granting jurisdiction to the court to adjudicate proceedings brought under section 73, LegCo members must have regarded to have consented to subject themselves to the court’s jurisdiction in circumstances where someone who has been disqualified as LegCo member but continues to act as such. Thus, even if (which I disagree) the word “statements” used in BL77 and the phrases “speeches and debates” and “words” used respectively in sections 3 and 4 of the LCPPO were intended to cover the reading out of the LegCo Oath in the process of taking an oath, the court would still have jurisdiction and power to look into them in a section 73 proceedings to determine the question of whether someone has been disqualified as a member of the LegCo under section 21 of the ODO or otherwise. 91. I therefore also reject this ground of opposition. D3. Section 21 of the ODO does not operate automatically as a matter of law 92. Finally, Mr Pun argues that on a proper construction of section 21(a) of the ODO, even if a person has declined or neglected to take the LegCo Oath as prescribed and that he “shall” vacate the office as a member of the LegCo, the vacating of the office does not operate automatically as a matter of law as contended for by the SJ and the CE. This is so because the provision is phrased in active voice in that the person “shall vacate” the office, instead of passive voice such as that the person “shall be vacated from his office”. Mr Pun therefore submits that although the person must vacate the office, it can only take effect either by the person resigning under section 14(1) of the LCO[19] or, if he fails to do so, by the President exercising his power under BL73(1) or (7) to declare that person to be disqualified. 93. There are no merits in this argument. 94. In light of the use of the word “shall”, Mr Pun accepts, as he must, that once the condition of “decline or neglect” under section 21 of the ODO is established, the legal effect intended in section 21 is that the person must vacate the office or be disqualified from entering on it, with the clear intended inevitable consequence that he cannot continue to act as a member of the LegCo. In the premises, I see no basis at all (let alone any reasonable basis) to construe section 21 to require any further intervening steps to be taken to achieve that inevitable effect and result. 95. Further, as pointed out by Mr Yu, a LegCo member who tenders a resignation under section 14(1) of the LCO is entitled to specify in the resignation notice the date upon which he intends the resignation to take effect. See: section 14(3)(b).[20] In other words, if Mr Pun’s above submissions are correct, it would be entirely up to the disqualified member to decide when he would effectively vacate the office by resigning. That cannot be consistent with the above objective meaning of section 21 of the ODO. 96. BL79(1) and (7) provide as follows: “Article 79 The President of the Legislative Council of the Hong Kong Special Administrative Region shall declare that a member of the Council is no longer qualified for the office under any of the following circumstances: (1) When he or she loses the ability to discharge his or her duties as a result of serious illness or other reasons; … (7) When he or she is censured for misbehaviour or breach of oath by a vote of two-thirds of the members of the Legislative Council present.” 97. It is plain that BL79(1) provides for the situation where the President shall declare a LegCo member to be disqualified when the member is either physically or mentally unable to discharge his duties as a member. This is supported by the Chinese version of BL79(1), which states: “因嚴重疾病或其他情況無力履行職務”, in particular the words “無力履行職務”. The plain and ordinary meaning of these words refers to the circumstances where, although it is open to the member concerned to carry out the duties, he is unable to do so. This does not cover the circumstances where the member is mandated to vacate his office under section 21 of the ODO, and hence it is no longer open to the member to carry out the duties as a LegCo member. 98. BL73(7) similarly does not cover a section 21 circumstance as it is again plain that the a person who declines or neglects to take the LegCo Oath cannot be regarded as to be “in breach of the oath”. 99. The word “misbehaviour” (“行為不檢”) under BL73(3) also cannot be objectively intended to cover the conduct of intentionally refusing or failing to take the LegCo Oath. It is so as, given the context that such “misbehaviour” has to be “censured” for by two-thirds of the LegCo members before the President could declare the member to be disqualified: (1) It objectively makes no logical or common sense to require such a procedure of censure to disqualify someone from being a LegCo member when he or she is not even capable of assuming the office given BL104. (2) It would be up to the other LegCo members to decide whether this would happen, which is by no means certain. This again cannot be and is clearly inconsistent with the above objective intention of section 21 of the ODO. 100. In the premises, on a proper construction of section 21 of the ODO, a LegCo member who has already entered on the office but declined or neglected to take the LegCo Oath shall, by the operation of law, be regarded as having vacated his office. Similarly, the LegCo member will be as a matter of law disqualified from entering that office if he has not entered on it. This does not require any further steps to be taken by that person himself, the President or any other person. D4. The CE’s locus 101. Finally, it has been raised by Mr Leung and Ms Yau that the CE has no locus to bring either the judicial review or HCMP 2819/2016. 102. Mr Yu however submits that the CE clearly has a locus to bring these proceedings given his constitutional role and duty under BL48(2). 103. Under BL48(2) provides as follows: “Article 48 The Chief Executive of the Hong Kong Special Administrative Region shall exercise the following powers and functions: (1) … (2) To be responsible for the implementation of this Law and other laws which, in accordance with this Law, apply in the Hong Kong Special Administrative Region; …” 104. Thus, under BL48(2), the CE has a constitutional role and duty to implement the BL and other laws of the Hong Kong Special Administrative Region. 105. Given this constitutional role, I agree with Mr Yu that the CE has a proper locus to bring these proceedings to implement BL104 and the relevant provisions of the ODO and to ensure that they are complied with. In this respect, it must be noted that the action brought under HCMP 2819/2016 is not premised only on a section 73 cause of action. 106. I therefore also reject Mr Leung and Ms Yau’s contention on locus. D5. Conclusion under Mr Leung and Ms Yau’s grounds of opposition 107. For all the above reasons, I do not accept any of the grounds of opposition advanced by Mr Leung and Ms Yau. 108. Thus, as a matter of law under BL104 and section 21 of the ODO, they shall be regarded as having already vacated their office in declining to take the LegCo Oath in form and in substance on 12 October 2016.[21] Their respective office as a member of the LegCo has become vacant. E. THE PRESIDENT’S ARGUMENTS 109. As mentioned above, the President’s only objection in these proceedings is that he should not be joined as the respondent in the judicial review application and as a defendant in HCMP 2819/2016. 110. The principal arguments advanced by Mr Jat SC for the President are these: (1) The President’s Decision is in nature not a substantive decision determining whether Mr Leung and Ms Yau declined or neglected to take the LegCo Oath on 12 October 2016 and whether therefore they had vacated their office as a LegCo member. It is also not a substantive decision to say whether the re-taking of the oath again by them, if it is to proceed, would be a valid one. The President has simply not yet made any decisions of substance. The President’s Decision is therefore simply a procedural one, allowing Mr Leung and Ms Yau to seek to procedurally retake the oath, which is by itself not amenable to judicial review. There is no extant substantive decision that is amenable to judicial review. The judicial review should be dismissed. (2) Insofar as the HCMP 2819/2016 is concerned, the substantive reliefs sought therein are only against Ms Yau and Mr Leung (as the 1st and 2nd defendants). There are no reliefs sought against the President (who has been joined as the 3rd defendant). However, there is nothing to suggest that there is a likelihood that the President would not abide by the decision of and the reliefs granted by the court under this action. In the premises, there are no reasons and basis for the CE and the SJ to join the President as one of the defendants in HCMP 2819/2016. 111. I do not accept the President’s submissions. 112. I agree with Mr Yu that in making the President’s Decision, the President impliedly has decided that Mr Leung and Ms Yau did not decline or neglect to take the LegCo Oath on 12 October 2016 and thus shall not be regarded as having disqualified under section 21 of the ODO to continue to act as a LegCo member. This is so since there would be no practical and useful purpose to be served to allow them to re-take the oath in another occasion if the President has not already effectively come to the view under the President’s Decision that they have not been so disqualified. Whether or not they already declined or neglected to take the LegCo Oath on 12 October 2016 should not be dependent on how they would take the oath again on a subsequent occasion. If they declined or neglected to take the LegCo Oath on 12 October 2016, as a matter of law, they shall be regarded as having vacated their office or been disqualified from that. There is no question that they can do it again as a member of the LegCo. 113. In the premises, I am satisfied that the President’s Decision is amenable to judicial review, and the President has been properly joined as a respondent. 114. Given my above conclusion that in law Mr Leung and Ms Yau should be regarded as having already vacated their office as a member of the LegCo on 12 October 2016 in declining or neglecting to take the LegCo Oath, I agree with Mr Yu that the President had misdirected himself on the law in making the President’s Decision. On that basis, I would quash the President’s Decision. 115. In relation to HCMP 2819/2016, given the President’s Decision and its effect as explained above, I am similarly satisfied that it is reasonable and necessary for the CE and the SJ to join the President as the 3rd defendant to ensure that he would be bound by the ruling in that case. F. SUPPLEMENTAL SUBMISSIONS RELATING TO THE INTERPRETATION 116. As the Interpretation was issued by the NPCSC on 7 November 2016 after the hearing of the present proceedings, the court subsequently directed the parties to provide supplemental written submissions, if they so wished, on the effect of the Interpretation relevant to the present cases or arguments already advanced by the parties. The parties provided their written supplemental submissions on 10 November 2016. 117. The CE and the SJ submit in their supplemental submissions that the Interpretation is binding on this court. 118. However, Mr Yu for the CE and the SJ emphasises in their supplemental written submissions that the Interpretation does not affect their original submissions made to this court on the proper purposive construction of BL104 and sections 16, 19 and 21 of the ODO insofar as the time and manner of oath-taking and the consequences of declining to take the oath are concerned (which have been accepted by this court above in paragraphs 24 - 36 and 92 - 100 above). 119. In this respect, it must be noted, as this court has repeatedly emphasised above, Mr Leung and Ms Yau have not advanced any arguments by way of submissions or evidence seeking to rebut the applicants’ submissions on the proper meaning and effects of BL104 and the provisions in the ODO, construed independently of the Interpretation.[22] They have also not advanced any submissions or evidence to say that they did not decline or neglect to take the LegCo Oath falling within the meaning of section 21. Their principal grounds of opposition raised in these cases based on the non-intervention principle and the immunity also have nothing to do with, and thus would not be affected in any material way by, the Interpretation. 120. Thus, I agree with Mr Yu’s submissions that the outcome of this case as regards Mr Leung and Ms Yau is the same with or without referring to the terms of the Interpretation. 121. Insofar as Mr Leung and Ms Yau are concerned, their supplemental submissions also say that the Interpretation does not affect and has no impact on their arguments raised under the non-intervention and the immunity grounds. This is correct. I have for the above reasons rejected these grounds without reference to or relying on the Interpretation. 122. Mr Leung and Ms Yau in their supplemental submissions further raise two main observations concerning the Interpretation itself. 123. First, Mr Pun for Mr Leung submits that paragraph (4) of the Interpretation supports his submissions that the person administering the oath is the final arbiter to determine the taken oath’s validity and compliance with BL104 and the ODO, hence also in support of his submissions under the non-intervention principle. I have rejected this submission above in paragraph 70. 124. Second, Mr Dykes for Ms Yau submits that, construed in the eyes of common law, the Interpretation goes further than merely as an interpretation of the meaning of BL104 and thus amounts to effectively a legislative act to amend BL104. As such, it is not made in compliance with BL158 and thus not binding on this court. In any event, given that it is effectively an amendment, it also has no retrospective effect under common law. 125. Given my acceptance of the CE and the SJ’s above submissions that the court would reach the same above conclusion for reasons I have explained above with or without referring to the terms of the Interpretation, this submission is irrelevant to the present proceedings and it is unnecessary for me to determine it. I however think that it may be open to arguments that whether the Interpretation falls within the proper meaning and scope of “interpretation” as intended under BL158 is a final matter for the NPCSC, given that the NPCSC has the ultimate power to interpret the BL. However, since this issue has not been argued substantively before me, I would refrain from saying anything further. 126. Finally, the President accepts that the Interpretation is binding on this court but also submits that it does not affect their submissions that the President should not be joined in these proceedings. For the reasons stated above, I have already rejected those submissions. 127. The President further asks this court to clarify the date from which the vacation or disqualification took effect if the court shall declare that Mr Leung or Ms Yau have vacated their office or disqualified from taking office. 128. In my view, given the plain meaning of section 21 of the ODO, the vacation took effect from the time when Mr Leung and Ms Yau declined to take the LegCo Oath when requested to do so. That would be 12 October 2016. 129. This is also consistent with BL104 and the Interpretation. Under BL104 and the Interpretation, a person can only “assume the office” of a LegCo member after he has validly taken the LegCo Oath, but would be disqualified from “assuming” the office after he has intentionally declined to take the LegCo Oath. Insofar as the present cases are concerned, Mr Leung and Ms Yau could only seek to “assume office” on 12 October 2016 when they purported to take the LegCo Oath. As they declined to take the LegCo Oath on that day, they have thus been disqualified from assuming the office since that day. G. DISPOSITIONS 130. For all the above reasons, the CE and the SJ succeed in the judicial review and HCMP 2819/2019. I would grant the following reliefs:[23] (1) Under HCMP 2819/2016: (a) A declaration that the oaths that Mr Leung and Ms Yau purported to take on 12 October 2016 contravened BL104 and/or sections 16(d) and 19(a) and Schedule 2 of the ODO and are invalid and void and have no legal effect as an oath of a member of the LegCo required by BL104 and as a LegCo Oath required by the ODO. (b) Further: (i) a declaration that the Mr Leung and Ms Yau have been disqualified from assuming the office of a member of the LegCo, and have vacated the same since 12 October 2016, and are not entitled to act as a member of the LegCo; (ii) an injunction restraining Mr Leung and Ms Yau from acting as a member of the LegCo; (iii) a declaration under section 73 of the LCO that Mr Leung and Ms Yau claimed to be entitled to act and/or acted as a member of the LegCo while disqualified from acting in that office since 12 October 2016; and (iv) an injunction under section 73 of the LCO to restrain Mr Leung and Ms Yau from claiming to be entitled and/or acting as a member of the LegCo. (2) Under the judicial review: (a) A declaration that the President has no power to re-administer or allow for re-administration of any further oath(s) to be taken by Mr Leung and Ms Yau under section 19 of the ODO on the basis that Mr Leung and Ms Yau have already vacated the office as a member of the LegCo, and are not entitled to act as a member of the LegCo. (b) A declaration that the office of member of the LegCo previously occupied by each of Mr Leung and Ms Yau is now vacant. (c) An order of certiorari to quash the President’s Decision. (d) An injunction that the President be restrained from administering or allowing to be administered the making of oaths of Mr Leung and Ms Yau under section 19 of the ODO. 131. Finally, I make an order nisi that costs of these applications be to the CE and the SJ, to be taxed if not agreed, with certificate for three counsel. Given the limited scope of the opposition raised by the President in these proceedings, the President shall bear one-fifth of the applicants’ costs, while Mr Leung and Ms Yau shall bear four-fifth of the costs. 132. Lastly, I must thank counsel for their helpful assistance in these matters. High Court Mr Benjamin Yu SC, Mr Johnny Mok SC, Mr Jimmy Ma and Mr Jenkin Suen, instructed by Department of Justice, for the 1st and 2nd applicants in HCAL 185/2016 and the 1st and 2nd plaintiffs in HCMP 2819/2016 Mr Jat Sew-tong SC and Mr Anthony Chan, instructed by Lo and Lo, for the respondent in HCAL 185/2016 and the 3rd defendant in HCMP 2819/2016 Mr Hectar Pun SC and Mr Anson Wong Yu-yat, instructed by Ho Tse Wai and Partners, for the 1st interested party in HCAL 185/2016 and the 2nd defendant in HCMP 2819/2016 Mr Philip J Dykes SC and Mr Jeffrey Tam, instructed by Khoo & Co, for the 2nd interested party in HCAL 185/2016 and the 1st defendant in HCMP 2819/2016 [1] The applicants also sought on the evening of 18 October 2016 urgent interim injunctions to restrain the President from allowing Mr Leung and Ms Yau to proceed to re-take the oath, and to also restrain them from proceeding to do so. I refused to grant the interim injunctions after hearing the parties for the reasons now recorded in the written decision dated 18 October 2016. [2] It has been held in Leung Kwok Hung v Clerk to the Legislative Council (HCAL 112/2004, 6 October 2004, Hartmann J) at paragraph 22 that the phrase “in accordance with the law” under BL104 refers to the legal requirements provided in the relevant provisions of the ODO. [3] See: T v Commissioner of Police (2014) 17 HKCFAR 593, at paragraphs 4 and 48; Vallejos v Commissioner of Registration (2013) 16 HKCFAR 45 at paragraphs 75 - 77; Town Planning Board v Society for the Protection of the Harbour Ltd (2004) 7 HKCFAR 1 at 13I-J. [4] As submitted by the CE and the SJ, similar constitutional requirement to swear an oath of allegiance to the sovereign and/or the constitution exists in other jurisdictions like the, Australia, Canada and New Zealand. In the UK, members of both Houses of Parliament are required by section 3 of the Parliamentary Oaths Act 1866 to swear an oath of allegiance to the Queen. [5] See: The New Shorter Oxford Dictionary. [6] See: Jowitt’s Dictionary of English Law (4th Edn). [7] See: the circular paper issued by the Clerk to the LegCo on 20 September 2016; circular paper issued by the Clerk dated 7 October 2016 reminding members to take the oath in the form and manner as prescribed by ODO; circular paper dated 11 October 2016 issued by the Clerk once again reminding members to take the LegCo Oath in the form and manner prescribed by the ODO; on 12 October 2016, they were requested by the Clerk to walk up to the table at the centre of the LegCo Chamber to take the Legislative Council Oath in accordance with the pre-determined order (see Affirmation of Wong Hwa Yih filed in HCMP 2819/2016 at paragraph 7); while Mr Leung and Ms Yau were at the centre table, the Clerk orally requested and reminded them to take the Legislative Council Oath in accordance with the statutory form of words. [8] See: paragraphs 11 - 12 of the Affirmation of Wong Hwa Yih in HCMP 2819/2016, paragraphs 10 - 11 of the Affirmation of Wong Hwa Yih in HCAL 185/2016 and paragraph 20 of the 2nd Affirmation of Rosanna Law in HCMP 2819/2016. [9] See: paragraph 13 of the Affirmation of Wong Hwa Yih in HCMP 2819/2016 and paragraph 12 of the Affirmation of Wong Hwa Yih in HCAL 185/2016 [10] See: paragraphs 14 - 19 of Affirmation of Rosanna Law. [11] See: a public statement (“嚴正聲明”) jointly published by a group of scholars and workers in the field of history, culture and education in various newspapers on 18 October 2016, exhibited as “WHY-7” to the Affirmation of Wong Hwa Yih in HCMP 2819/2016. [12] See: paragraphs 8 - 10 of the Affirmation of Wong Hwa Yih in HCMP 2819/2016; paragraphs 7 - 9 of the Affirmation of Wong Hwa Yih in HCAL 185/2016. [13] See also AG v Bradlaugh, supra, at 701 - 702, 719. [14] See also Sir Anthony Mason’s extra-judicial views in “The place of comparative law in developing the jurisprudence on the rule of law and human rights in Hong Kong” (2007) HKLJ299 at 304 - 305, where it is expressed that, given the obvious differences in the meaning and constitutional model of the doctrine of separation of powers between UK and other common law jurisdictions with a written constitution, such as Australia, Canada and the Untied States, “judicial decisions on the separation of powers need to be treated with great care before they can be imported from one jurisdiction to another. This proposition has significance for Hong Kong… It would not follow that the Basic Law, when construed in the light of its context and the preservation of the English common law by Article 8 of the Basic Law, necessarily mandates a separation of powers that confirms either to the United States or Australian model.” [15] Subject to BL158 where the power of interpretation of the BL is vested in the NPCSC, where the Hong Kong courts have been authorised by the NPCSC to interpret the BL when adjudicating cases. [16] Cf: AG v Bradlaugh, supra, is in relation to a statutory action brought under the Parliamentary Oath Act 1866 by the Attorney General against Mr Bradlaugh as a member of parliament for a statutory penalty for an offence in voting and setting in any debate of the Parliament “without having made and subscribed” to the parliamentary oath in the form as required under the Parliamentary Oath Act 1868. Although the 1866 Act only gives the AG the right to bring an action against such a member to recover the penalty, in determining whether Mr Bradlaugh was liable for the penalty, and notwithstanding the non-intervention principle as considered in the earlier related case of Gossett,the court had to and did determine the underlying necessary question as to whether Mr Bradlaugh as an antitheist could validly take and subscribe to the oath. [17] There is no question that Mr Leung and Ms Yau did purport to act or claim to be entitled to continue to act as a LegCo member. See: Affirmation of Law Shuk Pui Rosanna filed under HCMP 2819/2016, at paragraphs 29 - 31. Further, the fact that they sought to proceed to take the oath again after the President’s Decision must also amount to their acting or claiming to be entitled to act as a LegCo member. [18] The Chinese version of section 4 of LCPPO is even clearer in this meaning. It states “不得因任何議員曾在立法會或任何委員會席前發表言論，或在提交立法會或委員會的報告書中發表的言論，或因他曾以呈請書、條例草案、決議、動議或其他方式提出的事項而對他提起民事或刑事法律程序。”. [19] Section 14(1) of the LCO provides “A Member may, at any time, resign from office as a Member by giving written notice of resignation to the Clerk to the Legislative Council.” [20] Section 14(3)(b) of the LCO provides “ A notice of resignation takes effect – (a) …; or (b) if a later date is specified in the notice, on that later date.” [21] In light of sections 4 and 13(1) of the LCO, the CE and the SJ in these proceedings proceed primarily on the basis that Mr Leung and Ms Yau, for the purpose of section 21 of the ODO, had entered on their respective office of a member of the LegCo by the time when they purported to take the LegCo Oath on 12 October 2016. No one has submitted or argued otherwise in these proceedings. The court would therefore also proceed on that basis. However, to avoid any doubts, it must be noted that, to be consistent with BL104, the meaning of “entered on the office” in section 21 of the ODO, and the meaning of “having accepted office” under section 13(1) of the LCO should not be treated as the same meaning of “assuming office” under BL104. [22] Save as to Mr Pun’s contention that section 21of the ODO does not operate automatically as a matter of law to vacate a member’s office after the member has declined or neglected to take the LegCo Oath. A contention I have rejected at paragraphs 92 - 100 above. [23] For completeness, I would also grant leave to the CE and the SJ (a) to amend respectively the Form 86 and the Originating Summons in the form of the draft attached to the respective summons issued under HCMP 2819/2016 and HCAL 185/2016 both dated 24 October 2016; and (b) to file and rely on the 2nd Affirmation of Law Shuk Pui Rosanna affirmed on 27 October 2016 in both actions. All these have been considered at the hearing on an debene esse basis. The proposed amendments and the further evidence are clearly relevant to the present proceedings and do not cause any real prejudice to the respondents. No real objections have been taken by the respondents. The costs of all these summonses shall be in the cause. 
Chief Justice Ma: 1. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Ribeiro PJ: 2. At the hearing, the Court dismissed this appeal with reasons to follow. These are my reasons. 3. The District Court Ordinance (“DCO”)[1] gives a limited right of appeal from decisions of the District Court in civil cases. Section 63(1)[2] provides that such appeals can only proceed with leave of a judge or the Court of Appeal. By section 63A(2),[3] such leave shall not be granted unless the appeal has a reasonable prospect of success or there is some other reason in the interests of justice for hearing the appeal. And section 63B[4] provides that no appeal lies from the Court of Appeal’s decision to refuse (or grant) leave. 4. The appellant seeks to challenge the constitutionality of section 63B in respect of a refusal of leave, contending that it is inconsistent with Article 82 of the Basic Law which materially states: “The power of final adjudication of the Hong Kong Special Administrative Region shall be vested in the Court of Final Appeal of the Region...” A. How the issue arises 5. The appellant is the owner of a shop on the ground floor of Po Hang Building, of which the respondent is the Incorporated Owners’ corporation (“IO”). The appellant erected a metal fence and door enclosing a service lane running adjacent to the building and forming a portion of its common parts. The IO brought proceedings against the appellant in the District Court alleging breach of the Deed of Mutual Covenant (to which the appellant is a party)[5] and seeking mandatory injunctions requiring the fence and door to be removed and the lane reinstated. 6. The appellant failed to file a defence in time, leading to the IO’s application for judgment in default. This was met by the appellant’s application for an extension of time to file its defence and counterclaim, contending that the IO’s claim was defeated by limitation and that the appellant had acquired title to the service lane by adverse possession. 7. His Honour Judge Kent Yee[6] refused the appellant leave to file its pleadings out of time and entered judgment in favour of the IO. He held that on the relevant authorities,[7] the defence of adverse possession was not reasonably arguable because breach of the DMC remained actionable by the IO notwithstanding the appellant’s purported acquisition of a possessory title. 8. The appellant then applied to his Honour for a stay of execution and for leave to appeal to the Court of Appeal. That application was dismissed[8] on the basis that an appeal had no reasonable prospect of success and there was no other reason why the appeal should be heard in the interests of justice. 9. The appellant next applied to the Court of Appeal for leave to appeal against the Judge’s decision refusing leave to file out of time and entering judgment in the IO’s favour. This was dealt with on the papers and Chu JA, writing for the Court,[9] held that the threshold criteria for granting leave to appeal were not made out and refused leave. The application was renewed at an oral hearing[10] but the Court of Appeal remained unpersuaded that the appellant’s case was reasonably arguable. 10. The appellant proceeded to seek the Court of Appeal’s leave to appeal to this Court against that Court’s refusal to grant the appellant leave to appeal to itself against Judge Kent Yee’s decision. As Lam VP (giving the judgment of the Court) pointed out,[11] given the finality provision in DCO section 63B, this was on its face an incompetent application. However, the appellant sought to contend that section 63B is inconsistent with Article 82 of the Basic Law and thus unconstitutional. Applying an earlier Court of Appeal decision[12] and the reasoning of the Appeal Committee in HLF v MTC,[13] Lam VP upheld the constitutionality of section 63B as a proportionate restriction and dismissed the application. 11. Leave to appeal to this Court was sought from the Appeal Committee[14] on both the constitutionality and adverse possession issues. It granted leave only on the following questions: (1) Is s 63B of the District Court Ordinance, Cap 336 inconsistent with Article 82 of the Basic Law of the Hong Kong Special Administrative Region and thus unconstitutional? (2) Does the CFA have jurisdiction to entertain an appeal from a judgment of the CA refusing leave to appeal to it? The application seeking to raise issues regarding adverse possession and advanced on the “or otherwise” basis was adjourned until after determination of the aforesaid questions. B. The effect of Article 82 12. In most cases, constitutional challenges are founded on an applicant’s claim that his or her constitutional rights have been violated. Thus, the analysis usually begins by identifying the constitutional rights engaged.[15] However, this approach is inapplicable in the present case. This is because Article 82 of the Basic Law operates to vest the power of final adjudication in the Court of Final Appeal. It does not confer on parties to litigation any constitutional right of appeal to the final court. As Li CJ pointed out: “... it is obvious that the intent of the Basic Law was not to give every party to every dispute a right to have the dispute resolved by final adjudication by the Court. By its very nature, the Court's power of final adjudication vested by art.82 calls for and indeed requires regulation, which may include limitation. Such limitation is permitted by implication, having regard to the nature of the power. It may be dealt with by the enactment of statutes by the legislature or it may be dealt with by rules of court made by the rules committee exercising subordinate legislative powers.”[16] 13. Article 83 reflects this by providing that “the structure, powers and functions of the courts of the Hong Kong Special Administrative Region at all levels shall be prescribed by law”. 14. However, any restrictions on rights of appeal (eg, by finality provisions confining appeals to intermediate courts) have a limiting effect upon the Court’s constitutional power of final adjudication and cannot be arbitrarily imposed. As was held in Solicitor v Law Society:[17] “... The limitation imposed must pursue a legitimate purpose and there must be reasonable proportionality between the limitation and the purpose sought to be achieved. ... it is the duty of the courts to review any legislation enacted which seeks to impose any limitation on the power of final adjudication vested in the Court by art.82 and to consider whether the limitation satisfies the proportionality test.” 15. This approach is well-established and was endorsed by the Court in Mok Charles v Tam Wai Ho.[18] To apply it in the present case, one must first ascertain precisely what (if any) restriction the legislature has enacted affecting the Court’s constitutional function of final adjudication. This requires the relevant provisions to be construed. C. Construction of the statutory provisions C.1 The appellant’s argument 16. The appellant seeks to argue that on the true construction of the relevant statutory measures, the finality provision in DCO section 63B does not take effect: “... as a matter of statutory construction, a Refusal of Leave, being a ‘decision’ or ‘order’ of the CA, is a ‘judgment’ as defined by s.19 of the Hong Kong Court of Final Appeal Ordinance, Cap 484... , and is therefore within the jurisdiction of the CFA by virtue of s.22(1)(b) of the HKCFAO ...; and thus BL82 is not engaged.”[19] 17. The argument runs as follows: (a) Section 22(1)(b) of the Court’s statute (the “HKCFAO”)[20] confers a discretionary jurisdiction on the Court of Final Appeal to hear appeals in the following terms: “An appeal shall lie to the Court ... at the discretion of the Court of Appeal or the Court, from any judgment of the Court of Appeal in any civil cause or matter, whether final or interlocutory, if, in the opinion of the Court of Appeal or the Court, as the case may be, the question involved in the appeal is one which, by reason of its great general or public importance, or otherwise, ought to be submitted to the Court for decision...” (b) HKCFAO section 19 provides: “In this Part, unless the context otherwise requires ... ‘judgment’ includes decree, order or decision...” (c) DCO section 63B purports to exclude appeals “from a decision of the Court of Appeal as to whether or not leave to appeal to it should be granted.” (d) But such a “decision” comes squarely within HKCFAO section 22(1)(b) read together with section 19, and therefore is a decision falling within the Court’s discretionary jurisdiction, notwithstanding what DCO section 63B says. 18. This argument involves two main propositions: (i) that section 63B is inconsistent with section 22(1)(b); and (ii) that section 22(1)(b) is the controlling provision which trumps section 63B. If it is accepted, then the first question on which leave was given does not arise and the second question must be answered “Yes”. I am, however, unable to accept this argument. C.2 Later law prevails 19. Assuming, for present purposes, that section 63B is inconsistent with sections 19 and 22(1)(b) read together, it does not follow that section 63B is overridden by the latter sections. 20. The scope of the Court’s discretionary jurisdiction was established by sections 19 and 22(1)(b) upon their enactment in HKCFAO on 1 July 1997. The DCO amendments came later, enacted on 5 February 2008.[21] They include section 63(1), laying down the requirement for leave to appeal to the Court of Appeal, replacing an earlier provision; and adding section 63B which excludes any appeal from a decision of the Court of Appeal on the grant or refusal of leave to appeal. Those amendments thus expressly exclude that class of “decision” by the Court of Appeal from the ambit of section 22(1)(b). 21. There is no basis for thinking that section 63B has no effect. In so far as it cannot be reconciled with section 22(1)(b), the rule is that the earlier provisions give way to the later. Bennion states the principle as follows: “Where a later enactment does not expressly repeal an earlier enactment which it has power to override, but the provisions of the later enactment are contrary to those of the earlier, the later by implication repeals the earlier in accordance with the maxim leges posteriores priores contrarias abrogant (later laws abrogate earlier contrary laws).”[22] 22. It follows that section 63B is not somehow trumped by section 22(1)(b) but (subject to its constitutional validity) operates as a finality provision qualifying the latter section. It is true that section 22(1)(b) was amended in 2014.[23] That was, however, a minor amendment to reflect abolition of appeals to this Court as of right.[24] It merely substituted “any judgment” for “any other judgment” since there was no longer a separate class of judgment entitling the parties to appeal as of right. The 2014 amendment is not in any way inconsistent with section 63B and does not impliedly override or qualify that section. C.3 The reasoning in Lane v Esdaile 23. Quite apart from the lex posterior rule, a line of authority beginning with Lane v Esdaile[25] leads compellingly to the conclusion that sections 63(1), 63A(2) and 63B read together should be construed as investing with finality a decision of the Court of Appeal refusing leave to appeal to itself from a first instance judgment of the District Court. 24. Lane v Esdaile was concerned with a rule of court which relevantly provided that no appeal lay to the Court of Appeal after the expiration of one year without special leave of the Court of Appeal.[26] After a tortuous procedural history, the appellants sought leave to appeal against the order of Kay J at first instance made some three years earlier. The Court of Appeal refused leave.[27] The appellants applied to the House of Lords for leave to appeal against such refusal, relying on a statutory provision which stated: “... an appeal shall lie to the House of Lords from any order or judgment of … [the] Court of Appeal”,[28] arguing that refusal of leave was such an “order or judgment”. 25. The House of Lords held that on a purposive construction of the provisions, given the requirement to obtain the Court of Appeal’s leave to appeal from the first instance judgment, that Court’s refusal of leave was final and there was no jurisdiction to entertain the appeal. Lord Halsbury LC stated: “... when I look not only at the language used, but at the substance and meaning of the provision, it seems to me that to give an appeal in this case would defeat the whole object and purview of the order or rule itself, because it is obvious that what was there intended by the Legislature was that there should be in some form or other a power to stop an appeal—that there should not be an appeal unless some particular body pointed out by the statute... [that is, the Court of Appeal[29] ], should permit that an appeal should be given.”[30] 26. His Lordship pointed out[31] that the rule was “intended as a check to unnecessary or frivolous appeals” and held that allowing an appeal from a refusal of leave to appeal would render the leave requirement “absolutely illusory”. Lord Bramwell held that if a refusal of leave could be called an “order” at all, “it is not such an order as it was intended by the Statute of 1876 should be appealable”.[32] As Lord Macnaghten put it: “... according to the true construction of the Judicature Act and Orders, the Court of Appeal are constituted the sole and final judges of the question whether an appeal to them should or should not be admitted when the proposed appellant has allowed the prescribed period to elapse, and therefore that there can be no appeal from the grant or refusal of that indulgence.”[33] 27. The House of Lords’ decision was succinctly summarised a year later by Lord Esher MR in Ex p Stevenson[34] who stated: “...the very nature of the thing really concludes the question; for, if, where a legal authority has power to decide whether leave to appeal shall be given or refused, there can be an appeal from that decision, the result is an absurdity, and the provision is made of no effect.” 28. As Lord Dyson MR pointed out in Sarfraz v Disclosure and Barring Service,[35] the reasoning of Lane v Esdaile has repeatedly been applied for more than 100 years. Thus, giving the advice of the Privy Council, Lord Hoffmann explained Lane v Esdaile’s approach to construction as follows: “Their Lordships consider that the principle in [Lane v Esdaile as explained in Ex parte Stevenson], is that a provision requiring the leave of a court to appeal will by necessary intendment exclude an appeal against the grant or refusal of leave, notwithstanding the general language of a statutory right of appeal against decisions of that court. This construction is based upon the ‘nature of the thing’ and the absurdity of allowing an appeal against a decision under a provision designed to limit the right of appeal.”[36] 29. The aforesaid approach was adopted by the Appeal Committee in HLF v MTC,[37] where the Court of Appeal had refused to give the applicant leave to appeal (required by DCO section 63) against ancillary relief orders made at first instance in matrimonial proceedings. The Court of Appeal also refused leave to appeal to this Court against its aforesaid decision, prompting the applicant to apply to the Appeal Committee for leave to appeal. DCO section 63B had not yet been enacted, so that no express finality provision was in existence. Just as in the House of Lords in Lane v Esdaile, the question was whether, given the requirement for leave to appeal laid down by section 63(1) and the Court of Appeal’s refusal of leave, “an application for leave to appeal to the Court of Final Appeal against the refusal by the Court of Appeal to give leave to appeal from a District Court judgment to itself [could] be entertained.”[38] 30. Adopting the reasoning in Lane v Esdaile, and noting that no constitutional challenge was being mounted, the Appeal Committee concluded that no appeal lay to this Court. In so holding, I stated: “Where the Court of Appeal refuses leave to appeal, it declines to hear the appeal at all and no judgment of that court, whether final or interlocutory, comes into being. Therefore no appeal lies under s.22 to the Court of Final Appeal ...”[39] It would have been preferable to state (as their Lordships did in Lane v Esdaile) that “no appealable judgment” comes into being. 31. In my view, the approach to construction in Lane v Esdaile is entirely apt in the present case. Not only does the DCO (by section 63(1)) lay down a requirement of obtaining the Court of Appeal’s leave to appeal from the first instance judgment, section 63B expressly provides for finality where such leave is refused. It is thus unnecessary in the present case to imply a finality provision by a process of purposive construction. Sections 63(1), 63A(2) and 63B are plainly intended, inter alia, to enable the Court of Appeal to filter out unnecessary, unmeritorious or frivolous would-be appeals. For this Court to entertain an appeal against the Court of Appeal’s refusal of leave to appeal to itself would be to render those sections illusory and would result in absurdity. Such a construction of HKCFAO section 22(1)(b) cannot have been intended. 32. For the foregoing reasons, I conclude that on its true construction, DCO section 63B operates as a finality provision excluding appeals to this Court against a refusal by the Court of Appeal to grant leave to appeal from a decision of a District Judge. I turn then to the question whether such a finality provision is constitutionally valid. D. Constitutional validity - Proportionality 33. A finality provision which prevents a class of cases from reaching the Court of Final Appeal limits the Court’s exercise of the power of final adjudication vested in it by Article 82 of the Basic Law. The combined effect of DCO sections 63(1), 63A(2) and 63B is to create such a limitation. It is a constraint which has to be justified on a proportionality analysis, as held in Solicitor v Law Society and Mok Charles.[40] 34. This Court’s approach to proportionality was recently reviewed in Hysan Development Co Ltd v Town Planning Board.[41] In summary, the proportionality test in the present case requires the limitation on this Court’s function of final adjudication (i) to pursue a legitimate aim; (ii) to be rationally connected to advancing that aim; (iii) to be no more than is necessary to accomplish that aim;[42] and (iv) to strike a fair balance between the general interest and any individual rights intruded upon. D.1 Steps (i) and (ii): Legitimate aims and rational connection 35. In identifying the aim of the restriction with a view to considering its legitimacy and the rationality of the measures adopted to achieve it, section 63B should not be viewed in isolation. The statutory purpose emerges from the scheme created by sections 63(1), 63A(2) and 63B in the context of other relevant provisions of the DCO.[43] 36. The scheme has two broad aims. First, in line with the Lane v Esdaile line of authorities, by having the Court of Appeal screen out cases which have no reasonable prospects of success on appeal, it promotes the proper and efficient use of judicial resources and the avoidance of oppressive and unproductive appeals. It avoids the squandering of resources by the Court of Appeal or this Court on hearing appeals which cannot be expected materially to benefit either party, merely causing delays to others in the queue waiting for suitable appeals to be dealt with. By filtering out unfit cases, the scheme aims to enable this Court to play its proper role. As Li CJ pointed out in Solicitor v Law Society:[44] “The limitations serve a legitimate purpose namely, to prevent the Court at the apex of the judicial system from being unduly burdened with appeals so as to enable it to focus on appeals, the judgments on which will be of importance to the legal system.” 37. Secondly, in the context of a court of limited jurisdiction, the statutory scheme aims to maintain reasonable proportionality between litigation costs and the amounts at stake by restricting the available tiers of appeal. 38. The aim of economic proportionality in litigation is generally recognized. Thus, the Rules of the District Court state as two of their underlying objectives, promotion of “a sense of reasonable proportion and procedural economy in the conduct of proceedings” and ensuring “that the resources of the Court are distributed fairly.”[45] As part of the scheme for regulating appeals, the DCO empowers the Judge or the Court of Appeal to subject grant of leave to appeal “to such conditions as ... [they consider] necessary in order to secure the just, expeditious and economical disposal of the appeal.”[46] It should be noted in this context that the District Court’s general civil jurisdiction is limited to claims which do not exceed $1 million.[47] 39. The statutory restrictions seek to avoid the kind of situation that arose in Piglowska v Piglowski,[48] where the modest amount of a couple’s joint matrimonial assets in issue was wholly exceeded by the costs of successive appeals. Lord Hoffmann described the situation as follows: “... there is the principle of proportionality between the amount at stake and the legal resources of the parties and the community which it is appropriate to spend on resolving the dispute. In a case such as the present, the legal system provides for the possibility of three successive appeals from the decision at first instance. The first is as of right and the second and third are subject to screening processes which themselves may involve more than one stage. If one includes applications for leave, the facts of this case, by the time it reached the Court of Appeal, had been considered by five differently constituted tribunals. This cannot be right. To allow successive appeals in the hope of producing an answer which accords with perfect justice is to kill the parties with kindness.”[49] His Lordship added: “... even if a case does raise an important point of practice or principle, the Court of Appeal should consider carefully whether it is fair to have it decided at the expense of parties with very limited resources or whether it should wait for a more suitable vehicle.”[50] 40. Similar points were made in Hong Kong Housing Society and Secretary for Justice v Wong Nai Chung,[51] and in HLF v MTC.[52] 41. It is plain that the two broad aims discussed above are legitimate aims and that the restriction of rights of appeal by sections 63(1), 63A(2) and 63B is rationally connected to their achievement. D.2 Step (iii): No more than necessary 42. Two main arguments have been advanced on the appellant’s behalf to contend that the finality provision in the present case goes disproportionately beyond what is necessary. 43. The first is the suggestion that section 63B imposes an absolute ban so that, in line with this Court’s decisions in Solicitor v Law Society of Hong Kong[53] and Mok Charles v Tam Wai Ho,[54] it ought to be held to fail the proportionality test. That argument must be rejected. 44. In Solicitor v Law Society, the restriction contained in section 13(1) of the Legal Practitioners Ordinance[55] as it then stood was indeed absolute. It provided that an appeal would lie from the Solicitors Disciplinary Tribunal to the Court of Appeal and that “the decision of the Court of Appeal on any such appeal shall be final”.[56] As Li CJ pointed out,[57] the exclusion of further appeals was absolute and precluded any appeal to this Court even if the discretionary criteria under HKCFAO section 22(1)(b) were satisfied. That total ban was held to be disproportionate: “... s.22(1)(b) permits an appeal from the Court of Appeal only in narrowly defined circumstances: where the question is one which should be submitted to the Court by reason of its great general or public importance, or otherwise. The total ban imposed by the finality provision where questions of this order of importance arise cannot, in my view, be said to be reasonably proportionate to any legitimate purpose which may underlie the finality provision.”[58] 45. The same is true of the restriction imposed by section 67(3) of the Legislative Council Ordinance (“LCO”)[59] as it stood when its proportionality was examined in Mok Charles v Tam Wai Ho. That section provided that at the end of the trial of an election petition in the Court of First Instance, that Court: “... must certify the determination of the Court in writing. ... The determination as certified is final as to the matters at issue concerning the election petition.”[60] 46. Section 14(3)(c) of the High Court Ordinance[61] provides that no appeal shall lie to the Court of Appeal “from a judgment or order of the Court of First Instance, where it is provided by any Ordinance or by rules of court that the same is to be final”. Thus, the effect of section 14(3)(c) in combination with LCO section 67(3) was to exclude appeals from decisions of the Court of First Instance regarding election petitions. As was the case in Solicitor v Law Society, that provision was absolute, there having been no mechanism to vet the arguability or importance of any potential appeal. It was held to be a disproportionate restriction and thus constitutionality invalid. 47. Ma CJ stated: “... it is important to understand the nature of the restriction contained in s.67(3). That provision not merely restricts an appeal, it actually eliminates it: under no circumstances can an appeal be launched.”[62] His Lordship added: “... s.67(3) goes much further than is necessary to deal with the said aim of speedy determination in election petitions. ... the nature of the restriction is absolute: there is simply no avenue of appeal, however much in error the Court of First Instance may have been. Further, the following point was raised by Sir Anthony Mason NPJ in the course of submissions that, for my part, has considerable significance: it is perhaps easy to see the possibility of points of constitutional importance being raised in the course of an election petition and yet the effect of a provision such as s.67(3) is that no appellate court (and in particular the Court of Final Appeal) will have an opportunity to deal with them.”[63] 48. In the present case, the DCO provisions limiting the right of appeal plainly do not erect a total ban on appeals. The Court of Appeal is entrusted with vetting the prospects of a potential appeal and enjoined to refuse leave unless the criteria specified in section 63A(2) are met. Conversely, if the application relates to an appeal which does have a reasonable prospect of success or in respect of which there is some other reason in the interests of justice for hearing the appeal, the Court of Appeal may be expected to grant leave. If leave is granted and the appeal is determined, the parties could, if so advised, apply for leave to appeal to this Court. If the section 22(1)(b) criteria are satisfied, leave could be expected to be granted and the final appeal duly heard by the Court in the exercise of its power of final adjudication. 49. The appellant’s second argument takes objection to the fact that it is the Court of Appeal rather than the Court of Final Appeal itself which decides what cases should be excluded as having no reasonable prospects of appeal. It argues that this Court has its own filtering rules comprising HKCFAO section 22(1)(b) and rule 7 of its Rules[64] which the Court ought itself to operate in the exercise of its power of final adjudication. The contention is that allocating the screening process to the Court of Appeal in itself constitutes an unnecessary and disproportionate restraint on the Court’s power of final adjudication. 50. Mr Martin Lee SC[65] went so far as to argue that any rule which permits any case (save for decisions of a purely interlocutory nature) to be screened out as unfit for appeal by an intermediate court would constitute a disproportionate derogation from the Court’s power of final adjudication. He did not shrink from the logic of this argument and made the extravagant submission that the Court of Final Appeal, by its Appeal Committee, was bound to vet for itself every application for leave to appeal, including applications originating in decisions of tribunals like the Small Claims Tribunal. 51. That argument involves a complete departure from the proportionality analysis. The DCO’s restriction of the Court’s power of final adjudication by assigning the filtering process to the Court of Appeal is the starting-point in the proportionality analysis, raising questions as to the aims, rationality and proportionality of that measure. It is not, as Mr Lee SC submitted, the end of the inquiry, in itself justifying a conclusion of unconstitutionality. 52. The appellant’s argument thus ignores or loses sight of the legitimate aims identified above as the first step of the analysis, namely, the aims of promoting the proper use of judicial resources, the proper role of the Court of Final Appeal and economic proportionality in litigation. And in extending the argument to cover applications for leave to appeal from all judicial tribunals, it ignores other important legitimate aims. For example, rules which limit rights of appeal from tribunals like the Small Claims and Labour Tribunals are aimed in part at fostering an equality of arms between parties such as well-resourced employers or businesses on the one hand and employees and consumers with modest means on the other. Such rules seek to provide a cheap and quick means of resolving small claims. Access to justice afforded by such tribunals would be wholly undermined if a well-resourced litigant were able to drag poorer opponents up successive appellate levels all the way to this Court’s Appeal Committee, requiring unaffordable costs to be incurred and greatly delaying resolution of their claims. 53. Mr Lee SC’s argument fails to address the crucial question at step (iii), that is, whether the relevant statutory measures go beyond what is reasonably necessary to accomplish the legitimate aims identified. It fails in particular to examine that question with regard to the legitimate aim of preventing the apex Court from being unduly burdened with appeals so as to enable it to concentrate on appeals of importance to the entire legal system. The appellant’s contention that all applications for leave to appeal should be allowed to proceed unrestricted from the courts or tribunals below to be adjudicated upon by this Court necessitates abandonment of that aim. The appellant’s objection to the appellate process being halted at the level of the Court of Appeal is thus not an argument about the proportionality of the statutory measures designed to achieve the aforesaid aim but an argument which disavows that legitimate aim itself. It misapprehends the issue at the core of the proportionality analysis. D.3 Step (iv) the overall balance 54. The fourth step in the proportionality analysis is not of direct significance in the present case. No individual constitutional rights are infringed. The filtering mechanism, while a restriction on the Court’s power of final adjudication, is beneficial since it screens out unfit applications for leave to appeal, helping to ensure that the Court of Final Appeal is able to exercise that power effectively. It is in the general interest to avoid the waste of judicial resources and to promote economy in litigation. It is beneficial both to the parties and to the courts that appeals which have no reasonable prospects of success should not be allowed to proceed. E. Conclusion 55. For the aforesaid reasons, I conclude that the restrictions in question do not go beyond what is reasonably necessary for the achievement of the legitimate aims identified. They are proportionate and constitutionally valid limitations on the Court’s power of final adjudication. The questions on which leave was given should therefore be answered in the negative. The appeal was accordingly dismissed and the appellant ordered to pay the respondent’s costs.[66] It follows that restoring the adjourned aspects of the leave application would serve no purpose. Mr Justice Tang PJ: 56. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Fok PJ: 57. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Gleeson NPJ: 58. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Martin Lee SC, Ms Queenie Ng and Mr Lee Siu Him, instructed by K. H. Lam & Co., for the Defendant (Appellant) Mr Lawrence Ng, instructed by Lam & Partners, for the Plaintiff (Respondent) Mr Wong Yan Lung SC, instructed by the Department of Justice, and Mr William Liu SGC, of that Department, for the Intervener (Intervener) [1] Cap 336. [2] Section 63(1): “Subject to subsection (3) [presently not relevant], an appeal can, with leave of a judge or the Court of Appeal, be made to the Court of Appeal from every judgment, order or decision of a judge in any civil cause or matter.” [3] Section 63A(2): “Leave to appeal shall not be granted unless the judge ... or the Court of Appeal hearing the application for leave is satisfied that (a) the appeal has a reasonable prospect of success; or (b) there is some other reason in the interests of justice why the appeal should be heard.” [4] Section 63B: “No appeal lies from a decision of the Court of Appeal as to whether or not leave to appeal to it should be granted.” [5] As well as breach of section 34I(1) and (2) of the Building Management Ordinance (Cap 344). [6] DCCJ 2259/2013 (23 December 2013). [7] Incorporated Owners of Man Hong Apartments v Kwong Yuk Ching [2001] 3 HKC 116; Incorporated Owners of No 27A Chatham Road, Kowloon v Lee Kai Kong [2001-2003] HKCLRT 273; Incorporated Owners of Mountain View Mansion v Heart Cuisine and Ors. [2012] 4 HKLRD 628; and Wong King Lim v Incorporated Owners of Peony House [2013] 4 HKC 295. [8] DCCJ 2259/2013 (7 February 2014), §§23-24. [9] Lam VP and Chu JA, HCMP 415/2014 (6 August 2015), §32. [10] Lam VP and Chu JA, HCMP 415/2014 (18 September 2015), referred to in the Court of Appeal’s judgment dated 8 April 2016. [11] Hon Cheung CJHC, Lam VP and Chu JA, HCMP 415/2014 (8 April 2016) at §3. [12] Hong Kong Housing Society and Secretary for Justice v Wong Nai Chung HCMP 880/2009 (unreported, 22 September 2010) (CA). [13] (2004) 7 HKCFAR 167. [14] Ma CJ, Tang and Fok PJJ, FAMV No 21 of 2016 (19 August 2016). [15] See Catholic Diocese of Hong Kong v Secretary for Justice (2011) 14 HKCFAR 754 at §65. [16] Solicitor v Law Society of Hong Kong (2003) 6 HKCFAR 570 at §30. [17] At §§31-32. [18] (2010) 13 HKCFAR 762 at §§21-29, per Ma CJ. [19] Appellant’s Case §9 (Emphasis in the original). [20] Hong Kong Court of Final Appeal Ordinance (Cap 484). [21] As part of the Civil Justice (Miscellaneous Amendments) Ordinance (Ord No 3 of 2008). [22] Oliver Jones, Bennion on Statutory Interpretation,6th Ed (Lexis Nexis, 2013), p 279. The qualification stated in the text regarding the maxim generalia secialibus non derogant (a general provision does not derogate from a special one) is not applicable. [23] Administration of Justice (Miscellaneous Provisions) Ordinance 2014, section 8(2). [24] By section 8(1). [25] [1891] AC 210. [26] Order LVIII r 15. [27] (1889) 40 Ch D 520. [28] Appellate Jurisdiction Act 1876, section 3. [29] See 213. [30] At 211-212. [31] At 212. [32] At 213. Lord Herschell took the same view (at 214). [33] At 215-216. “... the Legislature intended that the matter should not go beyond the Court of Appeal, and therefore they have used the words that leave is to be given ...” per Lord Field at 216. [34] Re Housing of the Working Classes Act 1890 Ex p Stevenson [1892] 1 QB 609 (CA) at 611. [35] [2015] 1 WLR 4441 (CA) at §25. [36] Kemper Reinsurance Co v Minister of Finance (Bermuda) [2000] 1 AC 1 (PC) at 13. [37] (2004) 7 HKCFAR 167. Followed by the Appeal Committee in Lau Luen Hung Thomas v Insider Dealing Tribunal (2009) 12 HKCFAR 955 at §5. [38] At §15. [39] At §23. [40] Discussed in Section B above. [41] (2016) 19 HKCFAR 372 in Sections E, F and G of the judgment. [42] This is not a case in which the “manifestly without reasonable foundation” standard should be applied, as discussed in the Hysan case at Section G of that judgment. [43] In particular DCO sections 32 and 63A(1)(b) and Order 1A r 1(c) and (d) of the Rules of the District Court discussed below. [44] At §36. [45] RDC Order 1A r 1(c) and (f). [46] DCO section 63A(1)(b). [47] DCO section 32. [48] [1999] 1 WLR 1360 (HL). [49] At 1373. [50] At 1373-1374. [51] HCMP 880/2009 (unreported, 22 September 2010) (CA) at §§25 and 31. [52] (2004) 7 HKCFAR 167 at §26. [53] (2003) 6 HKCFAR 570. [54] (2010) 13 HKCFAR 762. [55] Cap 159. [56] (2003) 6 HKCFAR 570 at §5. [57] Ibid at §39. [58] Ibid at §40. [59] Cap 542. [60] (2010) 13 HKCFAR 762 at §14(7). [61] Cap 4. [62] (2010) 13 HKCFAR 762 at §52. [63] Ibid at §§61-62. The Court also noted that no such finality provision existed in respect of the elections under the Chief Executive Election Ordinance (Cap 569). [64] Hong Kong Court of Final Appeal Rules (Cap. 484A) Rule 7 provides for disposal of leave applications on the papers: “(1) Where the Registrar is of the opinion either on the application of the Respondent or of his own motion that an application discloses no reasonable grounds for leave to appeal, or is frivolous or fails to comply with these Rules, he may issue a summons to the applicant calling upon him to show cause before the Appeal Committee why the application should not be dismissed. (2) The Appeal Committee may, after considering the matter, order that the application be dismissed or give such other directions as the justice of the case may require.” [65] Appearing with Ms Queenie Ng and Mr Lee Siu Him for the appellant. [66] It was agreed that there would be no order as to the Intervener’s costs. 
The Court: A. Introduction 1. These appeals were against sentences imposed by the Court of Appeal following a review of sentence pursuant to section 81A of the Criminal Procedure Ordinance.[1] This Court has recently considered the Court of Appeal’s powers on, and the proper approach to, such a review of sentence in Secretary for Justice v Wong Chi Fung (“Wong Chi Fung”)[2]and, as will be seen, the Court’s judgment in that case governs the disposition of these appeals. 2. The appellants in these appeals were convicted of the offence of unlawful assembly in the circumstances described below. As will be seen, it was, on any view, a violent incident. Upon conviction, the appellants were sentenced by the magistrate to community service orders. The Secretary for Justice applied to review those sentences on the basis that they were wrong in principle or manifestly inadequate. The Court of Appeal acceded to the Secretary for Justice’s application and substituted sentences of imprisonment in respect of each of the appellants. 3. The appellants sought leave to appeal against the sentences imposed by the Court of Appeal upon review of the magistrate’s original sentences. In the light of Wong Chi Fung (CFA), the points of law sought to be raised by way of appeal had already been answered and so leave to appeal to this Court was granted by the Appeal Committee on the limited basis described below. 4. At the conclusion of the hearing, indicating that we would hand down our reasons for doing so in due course, the Court allowed the appeals and made orders in respect of each of the appellants, substituting, for the Court of Appeal’s sentences of imprisonment, sentences of imprisonment corresponding to the period of time already served in prison by each. These are the Court’s reasons for allowing the appeals and making the orders we did. B. The offences and proceedings below B.1 The unlawful assembly 5. The facts out of which the appellants’ convictions arose have already been the subject of a previous appeal to this Court (FACC 5/2017) on a discrete point of law.[3] 6. To re-state the facts briefly, on 13 June 2014, about 300 to 400 protesters objecting to development works in the North East New Territories New Development areas were engaged in a demonstration outside the Legislative Council (“LegCo”) Chamber at a designated demonstration area. At that time, LegCo’s Finance Committee was in session to discuss a funding application relating to advanced works at the development areas. At around 8.45pm that day, protesters in the demonstration area started to rush towards the entrances of the LegCo building in an attempt to gain entry through the glass doors. These doors, in front of which Mills barriers had been set up, had been locked to prevent unauthorised entry. The protesters started to pull away the Mills barriers and to use them, as well as bamboo poles and metal bars, which were wielded with considerable violence, to try to force their way into the LegCo complex by prising open or battering the doors. The situation was sufficiently serious to cause the President of LegCo to request police assistance to maintain the order and safety of the LegCo complex. The police had to use riot shields and pepper-spray to prevent the protesters from gaining entry to LegCo. The break-in attempt, which was recorded on CCTV footage, was sustained, involved “hundreds of demonstrators”[4] and lasted for nearly half an hour. As a result, damage was caused to a number of LegCo facilities, costing over HK$400,000 to repair. One of the security officers was injured by a falling Mills barrier and sustained broken toes, for which he was granted 85 days’ sick leave. 7. The acts of each of the appellants were summarised in tabular form in paragraph [26] of the judgment of the Court of Appeal, which we set out below: B.2 The proceedings before the magistrate 8. Each of the appellants was charged with unlawful assembly, contrary to section 18(3) of the Public Order Ordinance[5] based on the facts summarised above (Charge 1). Based on the same facts, they were also each charged with attempted forcible entry, contrary to section 23(1) of the Public Order Ordinance and section 159G of the Crimes Ordinance[6] (Charge 2). The 1st and 2nd appellants were each additionally charged with a further offence of obstructing an officer of the Legislative Council in the execution of his duty, contrary to section 19(b) of the Legislative Council (Powers and Privileges) Ordinance[7] (Charges 3 and 4 respectively).[8] 9. The 12th appellant pleaded guilty to Charges 1 and 2, whilst the charges against the other appellants proceeded to trial at the Eastern Magistrates’ Court.[9] After a trial lasting 13 days, on 30 December 2015, the magistrate found the appellants guilty of Charge 1 and not guilty of Charge 2.[10] 10. On 19 February 2016, the magistrate sentenced the appellants on their convictions in respect of Charge 1: the 1st to 9th and 11th appellants were sentenced to 120 hours of community service[11] and the 10th and 13th appellants were sentenced to 150 hours of community service; the 12th appellant, having pleaded guilty, was sentenced to 80 hours of community service.[12] Of relevance to one of the issues in these appeals is the fact that the 5th and 6th appellants were both under the age of 21 as at the date of the unlawful assembly and the 6th appellant was also under 21 as at the dates of conviction and sentence by the magistrate.[13] B.3 The review of sentence by the Court of Appeal 11. The Secretary for Justice was dissatisfied with the sentences imposed and applied for leave to appeal to the Court of Appeal for a review of those sentences pursuant to section 81A of the Criminal Procedure Ordinance. The procedural course of the application was somewhat convoluted. On 11 March 2016, Cheung CJHC granted such leave but the review could not proceed for the time being because, on 2 March 2016, the 1st to 3rd, 6th and 13th appellants had appealed to the Court of First Instance against their convictions.[14] 12. On 25 January 2017, Albert Wong J dismissed the appeals against conviction. This led to an application by the 1st and 2nd appellants for a certificate of points of law in order to seek leave to appeal to this Court. On 23 February 2017, Wong J refused that application and so, on 6 April 2017, the Court of Appeal directed that the review of sentence should be heard on 21 June 2017.[15] On that date, however, the hearing of the review as a whole was adjourned, since the 1st and 2nd appellants had applied to this Court, on 16 June 2017, for leave to appeal against their convictions and so the review of their sentences could not, in any event, proceed.[16] 13. On 10 August 2017, the Appeal Committee granted leave to the 1st appellant to appeal against his conviction on Charge 3[17] but dismissed his and the 2nd appellants’ applications for leave to appeal against their convictions on Charge 1. At that point, the review of sentence in respect of Charge 1 could proceed against all the appellants as a whole and so, on 14 August 2017, the hearing before the Court of Appeal resumed.[18] 14. At the conclusion of the hearing of the review of sentence on 15 August 2017, the Court of Appeal allowed the Secretary for Justice’s application, set aside the community service orders that had been imposed by the magistrate and substituted, for them, (i) in respect of Charge 1, on the 1st to 11th and 13th appellants, 13 months’ imprisonment, and (ii) in respect of both Charges 1 and 2, on the 12th appellant, 8 months’ imprisonment for each offence, to run concurrently. The Court of Appeal subsequently handed down its Reasons for Judgment (“CA Reasons”) on 11 September 2017. 15. It is relevant to note that, shortly prior to the hearing of this review of sentence on 15 August 2017, the same division of the Court of Appeal heard the review of sentence in Wong Chi Fung on 9 August 2017,[19] at the conclusion of which judgment was reserved. The Court of Appeal handed down its reserved judgment in Wong Chi Fung on 17 August 2017 (“Wong Chi Fung (CA)”) by which it increased the sentences of the three appellants in that case. 16. In paragraph [67] of the printed case for the 2nd and 6th appellants, it was submitted that the Court of Appeal failed to take into account that almost 18 months (from 19 February 2016 to 15 August 2017) had passed between the dates of sentence and review. As the above chronology shows, however, the review could not proceed until after the dismissal on 10 August 2017 of the 1st and 2nd appellants’ applications to the Appeal Committee for leave to appeal against their convictions. So the hiatus during the period from 11 March 2016 until 10 August 2017 was effectively the result of the statutory bar on the review proceeding pending an appeal against conviction. Since those appeals, which were consistently held to be unmeritorious, were instituted by the appellants themselves, the submission that there was undue harshness to them in the imposition of custodial sentences so long after their original sentences were imposed is a point of little merit. B.4 The application for leave to appeal to this Court 17. The 13 appellants all applied to the Court of Appeal for certification of various questions of law for the purpose of an application for leave to appeal to this Court. Before that application was resolved by the Court of Appeal, the Appeal Committee granted leave to appeal in Wong Chi Fung on 7 November 2017.[20] 18. In the light of that grant of leave by the Appeal Committee, the Court of Appeal certified, by its judgment dated 14 November 2017 (“CA Certification Judgment”),that four questions of great and general importance were involved in the proposed appeal. They were of the view that these four questions were the same as issues (1), (2) and (3) for which leave to appeal had been granted by the Appeal Committee in Wong Chi Fung.[21] 19. By its Determination dated 21 March 2018, after judgment in Wong Chi Fung was handed down by this Court, the Appeal Committee granted leave to appeal to this Court on a limited basis, stating: “2. We grant leave to appeal to each of the applicants confined to the ground that it is reasonably arguable that substantial and grave injustice has been done to the applicants by reason of the Court of Appeal’s departure from the principle that an offender is to be sentenced on the existing or prevailing guideline or tariff of sentence existing at the time of the commission of the offence. 3. We also grant leave to appeal to the 5th and 6th applicants on the additional ground that it is reasonably arguable that substantial and grave injustice has been done to them by reason of the Court of Appeal’s failure properly to consider section 109A of the Criminal Procedure Ordinance. 4. We are otherwise satisfied that the four questions of law certified by the Court of Appeal have already been fully answered in the Court’s judgment in Secretary for Justice v Wong Chi Fung & Others and, in the circumstances, we therefore refuse leave to appeal in respect of those questions. 5. We also refuse leave to appeal on the various other grounds advanced in support of the application under the substantial and grave injustice limb since they lack sufficient merit.” B.5 The appellants’ status as at the date of these appeals 20. It is material to note that, immediately following the review of sentence in the Court of Appeal, each of the appellants commenced their sentences of imprisonment on 15 August 2017 and served part of their sentences until granted bail by a single permanent judge of this Court. For convenience, the following table sets out, in respect of each appellant, (a) the length of sentence imposed, (b) the date on which they commenced their sentence, (c) the date on which each was granted bail and (d) the period of imprisonment served: * The 13th appellant was found to be in breach of the community service order imposed on him and so, on 15 December 2016, his original sentence was substituted with a sentence of imprisonment of 3 weeks which he served. It is therefore necessary to add, in respect of his time served, an additional 21 days, making a total of 3 months 30 days. C. The issues in this Court 21. In light of the limited basis on which leave to appeal to this Court was granted, there are three principal issues for determination in these appeals, namely: (1) Did the magistrate make a relevant error so as to justify the Court of Appeal exercising its jurisdiction under section 81A of the Criminal Procedure Ordinance to review the sentences of community service orders imposed by the magistrate? (2) In substituting sentences of imprisonment for the community service orders imposed by the magistrate, did the Court of Appeal retrospectively apply the Wong Chi Fung sentencing guidelines?[22] (3) Did the Court of Appeal fail properly to consider section 109A of the Criminal Procedure Ordinance in respect of the 5th appellant and/or the 6th appellant before imposing custodial sentences on them? C.1 Issue (1): Was the Court of Appeal’s jurisdiction under section 81A engaged? 22. In Section D of Wong Chi Fung (CFA), this Court addressed the jurisdiction of the Court of Appeal under section 81A of the Criminal Procedure Ordinance. We held, at [47]: “47. As the wording of section 81A shows, the grounds on which the Court of Appeal may interfere with a sentence passed by a lower court are restricted to the four grounds specified, namely that it is (i) not authorised by law, (ii) wrong in principle, (iii) manifestly excessive, or (iv) manifestly inadequate. If any of these grounds exists, the original sentence may be set aside, increased or reduced. Plainly, therefore, a review of sentence may be either advantageous or disadvantageous to the respondent to the application.” 23. In asking whether the magistrate made a relevant error so as to justify the Court of Appeal exercising its jurisdiction under section 81A of the Criminal Procedure Ordinance to review the sentences of community service orders imposed by the magistrate, the relevant inquiry is to ascertain whether the magistrate could be shown to have erred in principle or to have imposed sentences that were manifestly inadequate and “outside the ambit of appropriate sentences which would be reasonably imposed by a judge having considered all the relevant factors”. This was the basis of the Secretary for Justice’s application for review.[23] 24. The Court of Appeal analysed the magistrate’s sentencing in Section H of the CA Reasons. It rejected the argument advanced by the appellants[24] to the effect that the Court of Appeal was limited only to those facts found by the magistrate (Section H2). The Court of Appeal found (in Section H4) that the magistrate made a number of errors in law and principle. These were, in summary, as follows: (1) The magistrate erred in saying that the appellants did not use violence nor did they inflict any injuries. In so saying, the magistrate ignored the fact that the appellants intentionally and deliberately charged at and tried to force their way into the LegCo Complex by violent means.[25] (2) The magistrate failed to take into account that charging at the LegCo Complex with violence was an aggravating factor warranting more weight being given to deterrence.[26] (3) The magistrate failed to give due consideration to deterrence in accordance with applicable sentencing principles and gave unduly heavy weight to the appellants’ personal circumstances, their motive and rehabilitation. This was reflected in the magistrate’s view expressed at the initial sentencing hearing that he would not consider a deterrent sentence.[27] (4) The magistrate erred in taking the view that community service orders were appropriate despite none of the appellants (apart from the 12th appellant who pleaded guilty) showing remorse and his understanding on genuine remorse was seriously flawed.[28] (5) The magistrate’s reasoning for imposing community service orders on the 10th and 13th appellants, namely “so that they could through their act demonstrate that they were willing to shoulder legal responsibility” was thoroughly unsound.[29] (6) “All in all, the trial magistrate erred in not following the applicable approach to sentencing as explained by us above.” [30] 25. Leaving to one side points (3) and (6) in the above list for the time being, the other reasons given by the Court of Appeal for concluding that the magistrate erred in principle and imposed manifestly inadequate sentences are, in our view, sound and therefore engaged the jurisdiction of the Court of Appeal to increase the sentences on review pursuant to section 81A. 26. In Wong Chi Fung (CFA), we made it clear, in Section D.1, that the Court of Appeal is permitted, in a review of sentence, to have regard to all the evidence available to the sentencing court. It may not give different weight to relevant factors taken into account by the sentencing court unless that court has made an error in principle or imposed a manifestly inadequate sentence. But if the sentencing court has done so, then the Court of Appeal is entitled to assign a different weight to a particular relevant factor. 27. Thus, as regards point (1) above, although in his Reasons for Sentence the magistrate nominally referred to the fact that violence was involved in the unlawful assembly resulting in injuries,[31] he was wrong, as a matter of fact, in discounting that factor and, in effect, giving it no weight at all. This flies in the face of the clear evidence that the appellants were themselves violent and intentionally and deliberately charged at and tried to force their way into the LegCo complex by violent means. It is also inconsistent with the finding of injuries sustained by one of the security guards and the physical damage caused to the LegCo premises. Accordingly, as regards point (2) above, the magistrate having wrongly effectively proceeded on the footing that the appellants did not use violence, he did not give sufficient weight to the need for a deterrent sentence. 28. In Wong Chi Fung (CFA) at [122], we held that the Court of Appeal was justified in clarifying the principles on which it would be appropriate to impose a community service order and the factors to be considered in determining if an offender is genuinely remorseful. These principles and factors were set out by the Court of Appeal in Wong Chi Fung (CA) in Section H6 and especially at [140] to [141] and [147] respectively. Those principles and factors were also set out in the CA Reasons in this case at [100] and [102] in virtually the same terms. It should be noted, for the avoidance of doubt, that the Court of Appeal did not there say that genuine remorse must always be present in every case before a community service order may be imposed. Rather, the Court of Appeal was stating that, normally, one would expect genuine remorse to be present before a community service order is imposed. Obviously, there may be exceptions. 29. As regards points (4) and (5) above, in the CA Reasons at [29], the Court of Appeal recorded the stance taken by each of the appellants when the magistrate heard their preliminary mitigation on the date of conviction (30 December 2015), as follows: “(1) The 1st, 4th and 10th respondents did not have anything to say in respect of mitigation; (2) Save and except stating some basic background information, the 3rd and 5th respondents also had nothing to say in mitigation; (3) The 2nd respondent said, ‘I would say clearly to you at this moment that I do not feel any remorse’; (4) The 6th and 9th respondents indicated through their lawyers that ‘they would not have any regrets’; (5) The 7th respondent said, ‘I have no remorse for what I did on that day’; (6) The 8th respondent indicated through his lawyer that he ‘did not feel regret’; (7) Save and except stating some basic background information, the 11th respondent had no submission in mitigation; and (8) The 13th respondent told the magistrate that, if he was sentenced to imprisonment, ‘there would be serious consequences’. He also told the magistrate that the verdict of his conviction would ‘make all people angry’.” 30. The Court of Appeal also set out, at some length in Section C2 of the CA Reasons, a summary of the contents of the community service order suitability reports that were obtained in respect of the appellants and then made the following observations in respect of the magistrate’s reasons for sentence: “72. In respect of both respondents’ mitigation and indication in the Community Service Order Suitability Reports that they had no remorse, the trial magistrate considered that they should continue with their pursuit of justice or to voice for the needy and the neglected, and should not have any regrets about that, otherwise they would be shown as being unprincipled. However, he considered that the respondents should reflect on their means of resistance, so that their views could be expressed without sacrificing other people or putting them in danger. 73. The trial magistrate considered that most of the respondents did not deny their acts and were willing to accept the consequences and responsibility. That was considered to be part of the remorse. 74. The trial magistrate considered that the aggressive acts in this case were not the most serious of its kind. Having considered all the circumstances, he took the view that, although the most appropriate sentence for Charge (1) was a short term of immediate imprisonment, community service order in lieu of imprisonment was also a sentencing option. 75. In respect of the 10th and 13th respondents, although the Community Service Order Reports did not recommend that they receive community service orders, and at first, the trial magistrate also considered imposing custodial sentences, he eventually decided to give them a chance to demonstrate that they would bear the legal consequences through their action and sentenced them to 150 hours of community service. Subsequently, the 13th respondent was found to be in breach of the community service order, his sentence was substituted with an imprisonment of 3 weeks on 15 December 2016.” 31. In the circumstances, and with regard to the applicable principles regarding the relevance of genuine remorse referred to above, the lack of remorse on the part of the appellants was plainly a relevant consideration as to whether to impose a community service order rather than a sentence of imprisonment. In the present case, the absence of genuine remorse was plainly a relevant factor weighing against the imposition of community service orders in respect of the appellants. This was all the more so, in the case of the 10th and 13th appellants, who were not recommended for community service orders. We are therefore satisfied that the magistrate erred in principle in the manner set out by the Court of Appeal in points (4) and (5) above. In his Reasons for Sentence, the magistrate simply omitted to take into account the fact that the appellants were not remorseful.[32] 32. In any event, we accept the submission of the Secretary for Justice that, given the scale of the unlawful assembly and the degree of violence involved, the community service orders imposed by the magistrate in the present case were manifestly inadequate sentences for the appellants in respect of Charge 1. We are satisfied that, even disregarding the sentencing guidance laid down by the Court of Appeal in Wong Chi Fung, a custodial sentence of imprisonment was called for. In Wong Chi Fung (CFA) at [95], we gave illustrations of the range of sentences imposed for the offence of taking part in an unlawful assembly. As a review of those cases shows, unlawful assemblies involving the scale of participants and the degree of violence involved in this case generally result in sentences of imprisonment rather than community service orders. The case of The Queen v To Kwan-hang and Another[33] in which community service orders were apparently imposed[34] does not affect this conclusion: although also a case of unlawful assembly involving some violence, it is distinguishable on the facts as to the scale of violence; moreover, that decision was an appeal against conviction and the Court of Appeal did not address the question of the appropriateness of the particular sentence imposed. We are therefore satisfied that, in the present case, given the scale of the unlawful assembly and the violence deployed, the appropriate sentence was a custodial sentence rather than community service. We shall address below whether sentences of the length imposed by the Court of Appeal were appropriate but, for the purposes of the review of sentence under section 81A, we are satisfied that the Court of Appeal was entitled to substitute increased sentences in the form of sentences of imprisonment. 33. In conclusion on this issue, we consider that the Court of Appeal was right to conclude that the magistrate erred in principle and that he imposed sentences that were manifestly inadequate in the circumstances. Having reached the conclusion that the magistrate did so err and that his sentences were manifestly inadequate even by reference to the pre-Wong Chi Fung approach because they were, as contended by the Secretary for Justice, outside the ambit of appropriate sentences which would be reasonably imposed by a judge having considered all the relevant factors,[35] the Court of Appeal was entitled to exercise its discretion to review the sentences. 34. It was submitted in the printed case of the 3rd, 5th, 10th, 11th and 12th appellants[36] that the Court of Appeal is not required to exercise its discretion to increase an original sentence where it is wrong in principle unless it is also manifestly inadequate and that there was no good reason for the Secretary for Justice to apply for a review of sentence simply on the ground that the sentence is wrong in principle if it is not also manifestly inadequate.[37] We reject this submission which flies in the face of the plain language of section 81A which clearly states that the jurisdiction to review a sentence exists where any of the four grounds there set out is established. It is only necessary for one of those grounds to be established in order for the jurisdiction to be engaged: see Wong Chi Fung (CFA) at [47] and [53]. 35. It is convenient at this point of the judgment to address a number of discrete points raised by various of the appellants which the Court determined were not open to them on these appeals in view of the limited basis on which leave to appeal to this Court was granted. 36. It was submitted in Section D of the printed case for the 1st, 4th, 9th and 13th appellants,[38] that the Court of Appeal’s review of the appellants’ sentences was contrary to the principle of double jeopardy. The argument sought to be advanced was that, since the community service orders had been completed by the appellants (or, in the case of the 13th appellant, since he had already served the 3-week term of imprisonment ordered in place of community service upon his failure to perform the community service order imposed on him), the Court of Appeal’s imposition of terms of imprisonment amounted to punishing the appellants twice for the same offence, contrary to Article 11(6) of the Hong Kong Bill of Rights. 37. This point was not open to the 1st, 4th, 9th and 13th appellants to argue on this appeal for a number of reasons: (1) This point was not argued on behalf of the appellants below, as it could have been: see the CA Reasons at [81] et seq. describing the issues that were argued. (2) This point did not form the basis of any question of law sought to be certified by the Court of Appeal (or the Appeal Committee) for determination by this Court, nor the basis of any substantial and grave injustice argument in this Court: see the CA Certification Judgment at [5(1)]. (3) Question 3 raised before the Appeal Committee loosely raised the question of the court’s approach in terms of whether the CA “should take into account the fact that the original sentence has already been fully served”. That, however, is not the same as the double jeopardy point now sought to be argued. In any event, the CA’s Certification Judgment at [18] explained its reasoning (in our view, correct) for refusing to certify Question 3. (4) This argument was newly introduced into the submissions before the Appeal Committee in the submissions for these appellants before the Appeal Committee (at pp.13-17). (5) The Appeal Committee was not minded to accede to an application for leave to appeal on this ground, either as a point of law or as a basis for arguing substantial and grave injustice. 38. It was submitted that the probation officer responsible for seeing that the 1st, 4th and 9th appellants completed their community service orders satisfactorily should have suspended the order pending the hearing of the review of sentence, under section 6(1)(a) of the Community Service Orders Ordinance[39], to avoid the possibility of double punishment. This submission was also not open to the appellants since it was part and parcel of the double jeopardy argument in Section D of the printed case for the 1st, 4th, 9th and 13th appellants. However, in not addressing it, we should not be taken to accept the correctness of the submission as to the scope of section 6(1)(a). 39. In Section E.3 of the printed case of the 2nd and 6th appellants,[40] it was submitted that, when an offender’s criminal conduct was “intertwined with his exercising his fundamental rights to expression or assembly, the Court should consider if the sentence imposed is proportionate to the legitimate aim of protecting public order”. These submissions on proportionality sought to re-argue the same point that was raised before the Appeal Committee and which, including the reference to Barabanov v Russia,[41] addressed Question 2 and Question 3 for which leave to appeal was sought, the Court of Appeal having certified those questions. The Appeal Committee having refused leave to appeal in respect of those questions, it was not open to the appellants to pursue this submission. C.2 Issue (2): Did the Court of Appeal retrospectively apply the Wong Chi Fung guidelines? 40. In Wong Chi Fung (CFA), the Court addressed the approach when the Court of Appeal gives guidance for future cases in Section D.3 of its judgment. We held (at [77]) that: “77. As a reflection of the principle of legal certainty, it is settled law that the sentence for an offence should be in accordance with the practice prevailing at the time of the commission of the offence: see HKSAR v Tsoi Shu & Ors [2005] 1 HKC 51 at [39], citing R v Chan Ka Wai, CACC 530/1988, unrep., 9 May 1989 at [6]-[7].” 41. Although this Court endorsed the guidance for future cases laid down by the Court of Appeal in Wong Chi Fung, we held (at [126]) that: “126. In accordance with the principles discussed in Section D.3 above, it would not, however, have been appropriate to apply the Court of Appeal’s guidance to the appellants here. The increase in sentences intimated by the Court of Appeal represented a sentence significantly more severe than the range established by the courts’ existing sentencing practice and so, to avoid retrospectively imposing a more severe sentence based on a new sentencing guideline, the new level of sentence should not have been applied to them.” 42. The appellants contended that the Court of Appeal had, in this case, applied the guidance laid down in Wong Chi Fung (CA) and that, in accordance with the above statements of principle in Wong Chi Fung (CFA), this was a departure from an accepted norm occasioning substantial and grave injustice by (to quote the submission made in the printed case of the 7th and 8th appellants[42]): “(a) subjecting them to a substantially more severe term of imprisonment than the range of sentences existing at the time of their offences; as well as (b) denying them the protection under the principles of legal certainty and non-retroactivity.”[43] 43. Reference to points (3) and (6) of the list of points set out above (at [24]), summarising the Court of Appeal’s conclusions as to the magistrate’s errors in law and principle, demonstrates that there is substance in this complaint. 44. In referring to the “applicable sentencing principles” in point (3) (CA Reasons at [140]), the Court of Appeal did not expressly state that it was referring to the post-Wong Chi Fung principles. However, in point (6) (CA Reasons at [144]), the Court of Appeal clearly referred to the magistrate being in error “in not following the applicable approach to sentencing as explained by us above”. Read in context, we are satisfied that that must be a reference to the sentencing principles addressed in Section G1 of the CA Reasons. In that section, the Court of Appeal clearly applied the post-Wong Chi Fung principles applicable to unlawful assemblies involving violence. 45. Indeed, the following passages in the CA Reasons clearly support the contention that the Court of Appeal must have applied the new Wong Chi Fung guidelines in this case. In the CA Reasons at [87], the Court of Appeal said: “87. On 17 August 2017, this Court handed down the judgment in Wong Chi Fung & Others, CAAR 4/2016. In part H of that judgment, we expounded the legal propositions and sentencing principles applicable to unlawful assembly involving violence similar to the present case. Those propositions and principles are equally applicable here. They may be reiterated as follows.” Moreover, in Section G1 of the CA Reasons (entitled “Applicable legal propositions and sentencing principles”), at sub-section (8) (in turn entitled “Applicable sentencing principles for unlawful assembly involving violence”), the Court of Appeal set out at [106] to [108] a list of principles in virtually identical terms to those set out in Wong Chi Fung (CA) at [151] to [153]. 46. Thus, when the Court of Appeal referred in the CA Reasons at [140] to “the applicable sentencing principles” and at [144] to “the applicable approach to sentencing as explained by us above”, these are clearly, in our view, references to those passages in Wong Chi Fung (CA) we have identified in the preceding paragraph. 47. Furthermore, although we are satisfied that the Court of Appeal would have been entitled, on the existing authorities on unlawful assembly, to conclude that it was necessary in this case to impose a term of imprisonment as a deterrent sentence, we do not think (for the reasons set out in Section D below) that the starting point of 15 months’ imprisonment can be justified on the state of previous sentences imposed for this offence (see CA Reasons at [145]). None of the various earlier cases to which reference was made in Wong Chi Fung (CFA) at [95], and which represented a summary of the range of sentences previously imposed (prior to Wong Chi Fung (CA)) for that offence, approached (save for that arising in the context of the 1967 riots) a starting point of 15 months’ imprisonment. The case of HKSAR v Tai Chi Shing & Ors,[44] which involved acts described as “‘riotous’ in nature, if not a ‘riot’ by legal definition” and very similar to the facts of the present case, attracted sentences based on a starting point of 6 months’ imprisonment and therefore resulted in significantly shorter sentences than those imposed in the present case. The facts of HKSAR v Yip Po Lam,[45] involving a group of persons charging at a police cordon, resulted in a sentence of four weeks’ imprisonment suspended for 12 months. 48. Given the previous range of sentences for this offence, in arriving at the 15 month starting point here, it is reasonably clear, in our view, that the Court of Appeal must have been applying the new guidelines that it laid down (or was very shortly to lay down) in Wong Chi Fung (CA). Those principles, it will be recalled, were stated by the Court of Appeal in Wong Chi Fung (CA) “to provide guidance to the sentencing courts in the future”.[46] The Director of Public Prosecutions, Mr David Leung SC,[47] in his oral submissions on behalf of the Secretary for Justice, fairly accepted that this must have been the case. Given the chronology of the respective proceedings, when imposing the increased sentences on these appellants on 15 August 2017, the Court of Appeal would have had in mind the guidelines that it was very shortly going to hand down in its written judgment in Wong Chi Fung (CA). The relevant dates are as follows: (1) The substantive hearing of the review of sentence in this case took place on 14 and 15 August 2017. (2) That was after the hearing of the review of sentence in Wong Chi Fung, which took place on 9 August 2017, at the conclusion of which judgment was reserved. The Court of Appeal’s reserved judgment in Wong Chi Fung (CA) was handed down on 17 August 2017. (3) However, two days before that, on 15 August 2017, the Court of Appeal gave its decision in these cases and increased the sentences imposed by the magistrate. It then subsequently handed down the CA Reasons on 11 September 2017. 49. That being the case, and applying the reasoning in Section D.3 and at [126] of Wong Chi Fung (CFA), the sentences imposed by the Court of Appeal in this case were wrong in principle and the appellants’ appeals against them must be allowed. This is not on the basis that the Court of Appeal should not have interfered with the magistrate’s sentences of community service orders at all (for the reasons set out in Section C.1 above) but, rather, on the basis that, although it was justifiable for the Court of Appeal to substitute sentences of imprisonment, they should not have applied the new guidelines in Wong Chi Fung (CA) in fixing the length of the terms of imprisonment. 50. We shall address at Section D below the appropriate disposition of these appeals. C.3 Issue (3): Did the Court of Appeal fail properly to consider section 109A? 51. In Section D.4 of the judgment in Wong Chi Fung (CFA), this Court addressed the relevance of youth in sentencing and section 109A of the Criminal Procedure Ordinance. We held that the age of an offender, whether youth or advanced age, is always a relevant mitigating factor in sentencing and that the purpose of section 109A was to ensure that, save for excepted offences, the imprisonment of young persons between the ages of 16 and 21 is a sentencing measure of last resort.[48] 52. We concluded that, on its plain wording, section 109A placed a duty on a sentencing court when considering the appropriate sentence for an offender aged between 16 and 21, to obtain and consider information about the circumstances of the offender, the offence, his suitability for particular types of punishment, his character and his physical condition, to determine whether a non-custodial sentence was appropriate. Nevertheless, the requirement to obtain information was not absolute and the sentencing court could determine, without resort to obtaining information pursuant to section 109A, that the only appropriate sentence was imprisonment.[49] 53. Section 109A, it will be recalled, provides as follows: “No court shall sentence a person of or over 16 and under 21 years of age to imprisonment unless the court is of opinion that no other method of dealing with such person is appropriate; and for the purpose of determining whether any other method of dealing with any such person is appropriate the court shall obtain and consider information about the circumstances, and shall take into account any information before the court which is relevant to the character of such person and his physical and mental condition.” (Emphasis added.) 54. The two appellants in this case in respect of whom section 109A might be relevant are the 5th and 6th appellants. As noted above, the 5th appellant was born on 18 July 1994 and the 6th appellant on 30 October 1995. Therefore: (1) At the date of the offence, on 13 June 2014, the 5th appellant was aged 20 and the 6th appellant was aged 18. (2) At the date of conviction, on 30 December 2015, the 5th appellant was aged 21 and the 6th appellant was aged 20. (3) At the date of sentence by the magistrate, on 19 February 2016, the 5th appellant was aged 21 and the 6th appellant was aged 20. (4) At the date of the review of sentence by the Court of Appeal, on 15 August 2017, the 5th appellant was aged 23 and the 6th appellant was aged 21. 55. In the CA Reasons, there is no reference to section 109A at all and therefore no indication that the Court of Appeal took that section into account in respect of either the 5th appellant or the 6th appellant before increasing their respective sentences from community service orders to terms of imprisonment. Was there, therefore, a departure from an established norm giving rise to substantial and grave injustice to them? 56. The critical issue here is the date by reference to which the words “No court shall sentence a person of or over 16 and under 21 years of age to imprisonment” in section 109A is to be measured. Is this the date of offending, the date of conviction or the date of sentence? This is a question of construction of the relevant statutory language. 57. For the 5th appellant, Mr Martin Lee SC submitted that the operative date for determining whether a person is under the age of 21 for the purpose of section 109A is the date of commission of the offence. As a fallback, Mr Lee contended that, if the operative date is some later date, such as conviction or sentence, the likely sentence that the defendant would have received at the date of the offence should still be “a powerful factor” to be taken into account by the sentencing court.[50] For the 6th appellant, Mr Hectar Pun SC submitted that the relevant date for the purposes of section 109A is the date of offending, alternatively the date of conviction.[51] 58. The Director of Public Prosecutions submitted on behalf of the Secretary for Justice that “the relevant date for determining age is clearly the date of sentencing by the trial magistrate in the first instance; by the [Court of Appeal] on a sentence review; and indeed by [this Court] if it decides to sentence a person afresh”.[52] 59. For the following reasons, we conclude that the relevant date for the purposes of section 109A is the date on which a sentence of imprisonment is passed. 60. The construction of section 109A requires the Court to have regard to the language used having regard to its context and purpose. The purpose of section 109A is clearly rehabilitative in that it makes the imposition of a custodial sentence of imprisonment a last resort for a person aged between 16 and 21. This purpose is borne out by the speech of the Director of Social Welfare in moving the first reading of the bill by which section 109A was introduced into the Criminal Procedure Ordinance.[53] In that speech, he said: “The new provision follows generally a provision made in the United Kingdom Criminal Justice Act, and its simple purpose is to secure, without unduly fettering the discretion of the court, that imprisonment is used in relation to such young persons only where it is absolutely necessary to do so. We are … proposing this amendment … because we believe – and our belief is backed by experience – that it is true for Hong Kong that young offenders, involved even in serious crime, may be more effectively rescued from embarking upon a lifetime of crime if contact with hardened criminals through imprisonment is avoided, and if, in association with such curtailment of their liberty as is necessary, constructive measures are taken to help them to adjust to the requirements of society. Whether the young person is sent to a training centre or to a reformatory school, or is placed under the supervision of a probation officer his liberty is in a greater or lesser degree curtailed and it is on this basis of discipline that the constructive element of rehabilitation through training, counselling and education is founded.”[54] 61. This purpose is reflected in a passage from the judgment of Dyson LJ (as he then was) for the English Court of Appeal in R v Imran Hussain Ghafoor[55] at [31]: “… the philosophy of restricting sentencing powers in relation to young persons reflects both (a) society’s acceptance that young offenders are less responsible for their actions and therefore less culpable than adults, and (b) the recognition that, in consequence, sentencing them should place greater emphasis on rehabilitation, and less on retribution and deterrence than in the case of adults.” 62. The UK provision on which section 109A was modelled was, as noted in Wong Chi Fung (CFA) at [86], section 17(2) of the Criminal Justice Act 1948, which provided that: “No court shall impose imprisonment on a person under twenty-one years of age unless the court is of opinion that no other method of dealing with him is appropriate; and for the purpose of determining whether any other method of dealing with any such person is appropriate the court shall obtain and consider information about the circumstances, and shall take into account any information before the court which is relevant to his character and his physical and mental condition.” (Emphasis added.) 63. That provision in the UK has been superseded by subsequent legislation. Section 17(2) of the Criminal Justice Act 1948 became section 19(2) of the Powers of Criminal Courts Act 1973, which was then repealed and replaced by section 1 of the Criminal Justice Act 1982. The material part of section 1 of the Criminal Justice Act 1982[56] was replaced by section 1(3A) of the Criminal Justice Act 1982 (enacted by the Criminal Justice Act 1988)[57] and is now found in section 89(1) of the Powers of Criminal Courts (Sentencing) Act 2000.[58] Thus, it is clear that in the UK, the statutory provisions now benchmark the relevant age to the date of conviction. For that reason, reference to cases from England and Wales in respect of current legislation is of limited assistance to the construction of section 109A. 64. As a matter of language, the wording of section 109A is clear. It places a restriction on a court sentencing a person of or over 16 and under 21 years of age to imprisonment unless the court is of opinion that no other method of dealing with such person is appropriate. This refers, in our view, to a restriction by reference to the age of a defendant at the time when he is sentenced by a court. It is his remand into prison on the passing of a sentence of imprisonment that would frustrate the statutory purpose of the section of dealing with him by some other means so as to facilitate his rehabilitation and so it is the date of the imposition of that sentence that is material under section 109A. 65. This conclusion is also reinforced by reference to another relevant statutory provision of like character. Section 4 of the Training Centres Ordinance[59] provides expressly that a sentence of detention in a training centre may be possible if an offender is, in the opinion of the court, not less than 14 but under 21 years of age “on the day of his conviction”. That express reference to the date of conviction may be contrasted with the restriction, in section 109A, on sentencing a person over 16 and under 21 years of age to imprisonment. 66. Although the editors of Archbold Hong Kong 2018 state, at [5-143], that the date for determining a person’s age for the purpose of section 109A is the date of conviction, that statement is supported by reference to R v Danga,[60] a case decided by the Court of Appeal of England and Wales. That decision is, however, based on a statutory provision different to section 109A. In Danga, the relevant provision was section 1(3A) of the Criminal Justice Act 1982, as amended by the Criminal Justice Act 1988. That section was expressed in terms of a restriction on custodial orders that could be made “where a person under 21 years of age is convicted or found guilty of an offence” (emphasis added). Those italicised words plainly point to the date of conviction as the relevant date for consideration. Given the difference in the statutory language between section 1(3A) of the 1982 Act and section 109A, it is doubtful that the statement in Archbold can be supported by reference to Danga. We shall, however, return to Danga later in this judgment as it provides useful guidance as a matter of principle to the approach where an offender reaches a particular threshold age between the dates of offending, conviction and sentence. 67. Mr Pun referred to the decision of the Court of Appeal of R v Hor Wai Ming & Anor[61] as supporting his submission that the date of conviction is the relevant date for section 109A. However, the question of construction of section 109A before this Court was not before the Court of Appeal in that case. Instead, the appellants there were 15 and 16 at the date of conviction (29 September 1984) so that, when the case was before the Court of Appeal (16 January 1985), they were still clearly under the age of 21 and so the issue of the relevant date for section 109A did not arise. There is no discussion in the judgment of Kempster JA of the proper construction of section 109A and so the decision does not, in our view, assist on the issue before us. 68. Mr Lee also referred to and relied on the dissenting opinion of Lord Salmon in Baker v The Queen,[62] a Privy Council case on appeal from Jamaica concerning the construction of section 29(1) of the Juveniles Law.[63] There, the appellants had committed murder when aged 17½ but were convicted when aged 18. The majority, in an opinion delivered by Lord Diplock, concluded that the wording of sub-section (1) made it plain that the time for ascertaining whether the appellants were to be treated as juveniles was the date on which the sentence was passed and not the date of the offence. However, in a powerful dissent, Lord Salmon considered that the language of the sub-section was capable of two meanings and preferred the construction which referred to the age of the offender at the date of the offence. He considered that adopting a meaning which made a death sentence mandatory for a person who committed murder between the ages of 8 and 18 so long as he was sentenced after he reached the age of 18 would lead to “shocking and indeed barbarous results”,[64] such as where two boys of the same age committed murder but were tried and convicted on different dates so that one was sentenced before 18 but the other after. 69. Disregarding the fact that Lord Salmon’s was a lone dissenting opinion in Baker, we do not consider that case provides assistance in relation to section 109A. It concerns a case of capital punishment and that context therefore provides the basis for Lord Salmon’s dissent. It was thus a case not so much about promoting the offender’s rehabilitation as about sparing young offenders from being executed. That is simply not a concern with which we are faced in construing section 109A. We shall, however, address below the approach that courts should adopt when a defendant crosses a relevant age threshold between the dates of offending, conviction and sentence. 70. Mr Lee and Mr Pun both relied on the decision of the New Zealand Supreme Court in R v Mist,[65] in which Lord Salmon’s dissenting judgment in Baker was endorsed. That case concerned the proper construction of section 75 of the Criminal Justice Act 1985.[66] In their joint judgment, Elias CJ and Keith J wrote (at [39]): “[39] We prefer the position taken by Lord Salmon. It is supported by the considerations of fairness, predictability and due process discussed earlier. Further, as indicated, the reasons given by the majority for rejecting the argument that their interpretation was irrational and unjust, relating as they do to the Judicial Committee’s lack of familiarity with the Jamaican situation, provide a basis for distinguishing the decision. And judicial attitudes to the protection of human rights have undergone a change over the last 30 years in response to national legislative and constitutional changes and the development of international human rights law among other matters.” 71. Mr Lee and Mr Pun also relied on Article 12 of the Hong Kong Bill of Rights, which materially provides that “Nor shall a heavier penalty be imposed than the one that was applicable at the time when the criminal offence was committed.” This is in the same terms as Article 15 of the International Covenant on Civil and Political Rights and was part of the body of human rights law which led the New Zealand Supreme Court to construe section 75 of the Criminal Justice Act 1985 as requiring to be read with section 4(2) of the same act[67] and therefore as referring to the offender’s age on the date of the offence.[68] 72. We do not think that the decision in Mist provides a basis for concluding that section 109A means otherwise than what its clear wording indicates, namely that no court may sentence a person over the age of 16 but under 21 to imprisonment unless of the opinion that no other method of dealing with him is appropriate. The statutory provisions under consideration in Mist were in different terms to section 109A and clearly related the relevant age to the date of conviction. In contrast, for the purposes of section 109A, a court in Hong Kong takes the offender at his age when he appears before the court for sentence and at the time when that court might be minded to impose a sentence of imprisonment on him. 73. Similarly, we do not consider that Article 12 of the Hong Kong Bill of Rights supports the appellants’ submissions as to the construction of section 109A. The prohibition on a heavier penalty than one applicable at the time when the offence is committed is not infringed by construing the age in section 109A by reference to the date of sentence. The maximum sentence that can be imposed is that existing at the date of the offence. In the case of taking part in an unlawful assembly that maximum is 5 years’ imprisonment (if convicted on indictment) or a fine at level 2 and 3 years’ imprisonment (if summarily convicted).[69] Construing section 109A by reference to the age of the offender at the date of sentence does not increase the potential penalty for the offence. 74. We turn now to the question of the approach that courts should adopt when a defendant crosses a relevant age threshold between the dates of offending, conviction and sentence. As the Court stated in Wong Chi Fung (CFA) at [84], the age of an offender is always a relevant mitigating factor in sentencing. Where, therefore, a defendant is just under the age of 21 when he commits the offence, the fact that he has turned 21 shortly thereafter and before conviction or sentence should be taken into account by the sentencing court. As the court held in Danga: “It is, however, important to say that the broad conceptual approach of a sentencing court does not undergo a fundamental change simply because the offender passes his twenty-first birthday. If all factors were identical an offender aged 21 years and a few days is likely to receive in substance much the same punishment as one who is 20 years and 11 months, and the court will in substance take account, for instance, of the criteria of section 1(4) and (5) of the Act of 1982 for the slightly older person although not in terms strictly required by statute to do so.”[70] 75. This approach has consistently been adopted in England and Wales, in particular in relation to sentencing an offender for an historic offence prosecuted long after its commission: see, for example, R v Cuddington[71] where the fact that but for the passage of time between offence and conviction a lesser punishment would have been imposed is described as “a powerful factor” to be taken into account in sentencing. Other examples illustrating the same approach include: R v Dashwood[72] and R v Imran Hussain Ghafoor.[73] 76. For the above reasons, we conclude that, for the purposes of section 109A, it is the defendant’s age at the date of sentence that is relevant. However, if a young person has turned 21 years of age between the date of offending or conviction and the date of sentence, the fact of his youth will be a powerful factor in determining the appropriate sentence for him. All the more so should this be the case where an offender aged under 21 years of age at the date when an original sentence is imposed turns 21 before a later review of that sentence by the Court of Appeal. Although the provisions of section 109A do not strictly apply to such a person (so that the court is not strictly obliged to obtain the prescribed information), the sentence he should receive will, in most cases, be the same as if section 109A had applied to him. There will, of course, be exceptions (and this Court has already noted that section 109A may, in an appropriate case, be dispensed with)[74] but, in practical terms, this will mean that a young offender who has turned 21 by the time of sentencing should only be sent to prison as a matter of last resort and, for that purpose, the court will have to be alive to the possibility it may need to obtain reports on the young person in question. This approach reflects a wider principle that a court should have regard to common sense when sentencing a young person: see, in this context, Wong Chun Cheong v HKSAR.[75] 77. Returning to the present case, the provisions of section 109A did not, in our view, apply to the 5th appellant who was already 21 at the date of sentence by the magistrate. Nor, strictly, did they apply to the 6th appellant at the time of the passing of sentence by the Court of Appeal on the review of his sentence, since by then had reached the age of 21. However, applying the approach set out in Danga (see above), we consider that, conceptually, the Court of Appeal should have imposed sentences on the 5th and 6th appellants that were in substance much the same as those which would have been imposed on an offender who was just under 21 years of age. In practice, this means that the Court of Appeal should only have imposed sentences of imprisonment if of the opinion that there was no other appropriate means of dealing with them. Since there was no reference in the Court of Appeal’s judgment in this case to section 109A, or indeed to the fact that the 5th and 6th appellants had passed the threshold age of 21 years between the date of the offence and the review of sentence, we are satisfied that there was a departure from an established norm. 78. In view of our conclusion (in Section C.2 above) that the Court of Appeal erred in retrospectively applying the Wong Chi Fung (CA) guidelines to the appellants, the appeals must be allowed and the sentences of imprisonment imposed by the Court of Appeal set aside. For this reason, it is unnecessary to determine whether the Court of Appeal’s failure to have regard to the conceptual approach laid down in section 109A as a relevant factor in sentencing the 5th and 6th appellants is a sufficient reason on its own for allowing their appeals on the basis of substantial and grave injustice. D. The appropriate disposition of these appeals 79. In the light of our conclusions on Issues (1) and (2) above (see Sections C.1 and C.2 of this judgment), the appeals must be allowed and the sentences of imprisonment imposed by the Court of Appeal set aside. This is not, however, on the basis that the Court of Appeal should not have interfered with the magistrate’s sentences of community service orders but, rather, on the basis that, although it was justifiable for the Court of Appeal to substitute sentences of imprisonment, they should not have applied the new Wong Chi Fung (CA) guidelines in fixing the length of the terms of imprisonment. 80. Instead, having concluded that the magistrate had erred in principle and imposed manifestly inadequate sentences, it would be appropriate for the appellants to be re-sentenced on the basis of the sentencing regime for the offence of taking part in an unlawful assembly prior to the new guidelines in Wong Chi Fung (CA). 81. In Wong Chi Fung (CFA), we made the observation at [115] that the Court of Final Appeal is not a sentencing court and that the function of sentencing is primarily that of the convicting court of trial, subject to review by the Court of Appeal. That would militate in favour of a remitter to the Court of Appeal to pass the appropriate sentences. However, it is clearly within the Court’s jurisdiction to make such orders as the Court of Appeal could on the review of sentence[76] and, given the history of this case and the fact that the relevant events took place over 4 years ago, rather than subject all concerned to yet another hearing (which may even give rise to a further appeal), we took the view that the better course was for the Court to resolve the question of the appropriate sentence rather than remit the matter to the Court of Appeal. 82. For the following reasons, we concluded that we should substitute, in place of the Court of Appeal’s sentences of imprisonment, sentences of imprisonment corresponding to the period of time already served in prison by each prior to the grant of bail. 83. In arriving at this result, we took into account that sentencing is an art and not a science and that the determination of a suitable sentence for taking part in an unlawful assembly is not something that can be arrived at by the application of any formula. Previous sentences in other cases have inevitably varied because of the fact sensitive nature of the sentencing process. We refer again to the illustrations of the range of sentences for this offence set out in Wong Chi Fung (CFA) at [95]. 84. As we have already indicated (in Section C.1 above), we consider that the degree of violence involved in the present case, the number of persons involved, the duration of the unlawful assembly and the injuries and damage inflicted merited immediate custodial sentences. Having regard to the previous cases, we consider the closest factual parallel to be the case of HKSAR v Tai Chi Shing & Ors.[77] This would suggest an upper end of appropriate sentence for this offence, prior to the guidelines laid down in Wong Chi Fung (CA) of 6 months. On that basis, allowing for time off for good behaviour (approximately one-third), the maximum period of incarceration would be around 4 months for each of the appellants. 85. Since five of the appellants have served more than 4 months,[78] 		there is no realistic possibility they would be sentenced to a term of 		imprisonment that would require them to be remanded into custody again. 		The remaining eight appellants who have served 3 months and 9 days[79] include the 12th appellant who pleaded guilty and who should therefore only have received a sentence of no more than 4 months in any event, so he would not receive a sentence that would require him to go back to prison. The other appellants have therefore served terms of imprisonment which are less than 3 weeks’ shy of the period which would entitle them to be released (assuming an original sentence of 6 months). Since some discount should have been afforded for the fact that this was a sentence review, they should also not be sentenced to terms that would require them to return to prison. In this regard, it is relevant to note that the Secretary for Justice, in his written submissions before the Court of Appeal on the review of sentence, initially asked that court to substitute “a short custodial sentence” in place of the community service orders.[80] This request is relevant to the question of any discount that would have been afforded for the fact that the sentences were being increased on review. 86. Ordinarily, of course, the appellants having taken part in the same unlawful assembly and having been regarded by both the magistrate and Court of Appeal to be equally culpable, they could have expected to receive the same sentence. The fact that they have ultimately served different lengths of sentence simply reflects the fact that their release from prison pending this appeal occurred on different dates because of the timing of their applications for bail. E. Conclusion 87. For these reasons, the Court allowed each of these appeals and made the orders set out in paragraph [4] above. 88. We would finally just emphasise that, as Wong Chi Fung (CFA) made clear, guidance for the future was provided by the Court of Appeal in that case in respect of offences involving unlawful assembly. We reiterate the much stricter view to be taken when sentencing in this context where disorder or violence is involved. Mr Martin Lee SC and Mr Jeffrey Tam, instructed by Ho Tse Wai & Partners, assigned by the Director of Legal Aid, for the 3rd, 5th, 10th, 11th & 12th Appellants (D3, D5, D11, D12 & D14) in FACC 9, 10, 11, 12 & 13/2018 Mr Philip Dykes SC and Mr Douglas Kwok, instructed by Bond Ng Solicitors, assigned by the Director of Legal Aid, for the 1st, 4th, 9th & 13th Appellants (D1, D4, D10 & D15) in FACC 3, 5, 7 & 8/2018 Mr Hectar Pun SC and Mr Anson Wong Yu Yat, instructed by Kenneth Lam Solicitors, for the 2nd & 6th Appellants (D2 & D6) in FACC 4 & 6/2018 Mr Osmond Lam and Ms Jacquelyn Ng, instructed by Sanny Kwong & Co., for the 7th & 8th Appellants (D7 & D8) in FACC 14 & 15/2018 Mr David Leung SC, DPP, Mr Jonathan Man, SADPP, Mr Derek Lau, SPP and Ms Annie Li, PP, of the Department of Justice, for the Respondent in FACC 3-15/2018 [1] (Cap.221). [2] [2018] HKCFA 4; (2018) 21 HKCFAR 35; Judgment dated 6 February 2018 (“Wong Chi Fung (CFA)”). [3] Namely, whether a police officer carrying out duties within the precincts of LegCo is an “officer of the Council” for the purposes of the Legislative Council (Powers and Privileges) Ordinance (Cap.382): see HKSAR v Leung Hiu Yeung [2018] HKCFA 2; (2018) 21 HKCFAR 20. [4] CAAR 3/2016, Reasons for Judgment (11 September 2017) at [129(1)]. [5] (Cap.245). [6] (Cap.200). [7] (Cap.382). [8] Charge 3, of which the 1st appellant was convicted and the 2nd appellant acquitted, was the subject of the appeal in FACC 5/2017 (see FN 3 above) and it is not necessary to consider Charges 3 and 4 further. [9] In ESCC 3658/2014, before Mr Jason Wan Siu-ming. [10] Statement of Findings (“SoF”) at [36]; the magistrate concluded that, since Charge 2 was based on the same facts as Charge 1, a conviction on Charge 2 would be repetitive and unfair to the appellants. [11] The 1st appellant was also sentenced to 120 hours of community service on Charge 3, which sentence was ordered to run concurrently with the sentence on Charge 1. [12] The 12th appellant received the same sentence in respect of Charge 2, to run concurrently with his sentence on Charge 1. [13] The 5th appellant was born on 18 July 1994 and the 6th appellant was born on 30 October 1995. [14] By reason of section 81C(1)(a) of the Criminal Procedure Ordinance, the review could not proceed until the conclusion of the appeals against conviction. [15] In CAAR 3/2016, before Yeung VP, Poon and Pang JJA. [16] Again, by reason of section 81C(1)(a) of the Criminal Procedure Ordinance. [17] This was the appeal in FACC 5/2017 (see FN3 and FN7 above). [18] The review of sentence in respect of the 1st appellant’s conviction on Charge 3 was adjourned pending the judgment of this Court in FACC 5/2017. [19] In CAAR 4/2016, in which the Secretary for Justice was seeking to review sentences of community service orders imposed on the three defendants in that case. [20] In FAMC 43-55/2017 (Ma CJ, Ribeiro & Fok PJJ). [21] See Wong Chi Fung (CFA) at [43]. [22] Being the guidelines laid down by the Court of Appeal and approved by in Wong Chi Fung (CFA). [23] CA Reasons at [76]. [24] Although they were respondents in the review, we shall for consistency refer to them as the appellants in this judgment. [25] CA Reasons at [138]; in this regard, as to the facts, see also CA Reasons at [25] to [27]. [26] CA Reasons at [139]. [27] CA Reasons at [140]. [28] CA Reasons at [142]. [29] CA Reasons at [143]. [30] CA Reasons at [144]. [31] ESCC 3658/2014, Reasons for Sentence, 19 February 2016 (“RS”) at pp.3S-4R and 5O-5T. [32] RS at pp.4T-5N. [33] [1995] 1 HKCLR 251. [34] Case for the 2nd and 6th Appellants at [47(5)], referring to a newspaper report of the sentences. [35] CA Reasons at [76] to [80] recording the Secretary for Justice’s grounds of application. [36] Represented by Mr Martin Lee SC and Mr Jeffrey Tam. [37] Case for the 3rd, 5th, 10th, 11th and 12th Appellants at [32]. [38] Represented by Mr Philip Dykes SC and Mr Douglas Kwok. [39] (Cap.378). [40] Represented by Mr Hectar Pun SC and Mr Anson Wong Yu Yat. [41] An unreported decision of the European Court of Human Rights (Applications Nos. 4966/13 & 5550/15), decided on 30 January 2018 and therefore after Wong Chi Fung (CFA). [42] Represented by Mr Osmond Lam and Ms Jacquelyn Ng. [43] Case for the 7th and 8th Appellants at [26]. [44] [2016] 2 HKC 436. [45] [2014] 2 HKLRD 777. [46] Wong Chi Fung (CA) at [18]. [47] Appearing with Mr Jonathan Man, SADPP, Mr Derek Lau, SPP, and Ms Annie Li, PP. [48] Wong Chi Fung (CFA) at [84] and [86]. [49] Wong Chi Fung (CFA) at [87] to [90]. [50] 5th Appellant’s printed case at [83] and [84]. [51] 6th Appellant’s printed case at [35] to [36] and [34] respectively. [52] Printed Case for the Respondent at [63]. [53] Young Offenders (Miscellaneous Provisions) Bill 1967. [54] Hansard (1 November 1967) at p.448. [55] [2003] 1 Cr. App. R. (S.) 84, a case concerning the Powers of Criminal Courts (Sentencing) Act 2000. [56] Which referred to “Where a person under 21 years of age is convicted or found guilty of an offence …”. [57] Similarly referring to a person under 21 years of age being “convicted or found guilty of an offence”. [58] Also referring to a person “aged under 21 when convicted of the offence …”. [59] (Cap.280). Section 4(1) materially reads: “Where a person is convicted of an offence punishable with imprisonment, then if on the day of his conviction he is in the opinion of the court not less than 14 but under 21 years of age, …, the court may, in lieu of any other sentence, pass a sentence of detention in a training centre.” [60] [1992] 1 QB 476. [61] [1985] 1 HKC 30. [62] [1975] AC 774. [63] Relevantly providing: “Sentence of death shall not be pronounced on or recorded against a person under the age of 18 years, but in place thereof the court shall sentence him to be detained during Her Majesty’s pleasure, …”. [64] [1975] AC 774 at 790H. [65] [2006] 3 NZLR 145. [66] Section 75 provided that it would apply to a person who was not less than 21 years of age and who was “convicted of an offence”. [67] Section 4(2) materially provided that: “no court shall have power, on the conviction of an offender of any offence, to impose any sentence or make any order in the nature of a penalty that it could not have imposed on or made against the offender at the time of the commission of the offence, except with the offender’s consent”. [68] R v Mist [2006] 3 NZLR 145 at [5], [62], [88] and [107]. [69] See Wong Chi Fung (CFA) at [93]. [70] R v Danga [1992] 1 QB 476 at 481D-E. The restrictions in section 1(4) and (5) of the Criminal Justice Act 1982 apply to a person under 21 but not to a person over 21. [71] (1995) 16 Cr. App. R. (S.) 246 at 250. [72] (1995) 16 Cr. App. R. (S.) 733 at 736-737. [73] [2003] 1 Cr. App. R. (S.) 84 at [31]. [74] Wong Chi Fung (CFA) at [90]. [75] (2001) 4 HKCFAR 12, a case concerning a “plainly trivial” offence (participating in a lion dance in a public place without a permit) in which, by reason of the training centre order imposed by the magistrate, the appellant was exposed to the risk of detention for up to five times the statutory maximum prescribed for the offence. [76] Court of Final Appeal Ordinance (Cap.484), section 17(2). [77] [2016] 2 HKC 436. [78] The 1st, 4th, 6th, 9th and 10th appellants. [79] The 2nd, 3rd, 5th, 7th, 8th, 11th, 12th and 13th appellants. [80] CA Reasons at [80]. 
The Court: 1. The appellants were convicted of a conspiracy the alleged object of which was for the 1st appellant to engage in misconduct in public office.[1] Immediately before, and in anticipation of, his appointment to a senior public office, the 1st appellant received, from the other appellants, large payments of money. The prosecution case was that these were bribes, not in contemplation of any specific act or omission, but in order to secure an improper inclination and that, in return for the payments, the 1st appellant agreed to be or remain favourably disposed, in office, to commercial interests associated with the other appellants. The certified question[2] involves the resolution of the issue whether such an agreement is capable of answering the description of an agreement on the part of the 1st appellant to misconduct himself in the course of or in relation to his public office. The appellants contend that no relevant act of misconduct is involved. For the reasons that follow that contention fails. A. Background 2. The factual background to these appeals, helpfully set out at length in the judgment of Lunn VP in the Court of Appeal,[3] lies in the relationship between a senior public servant employed by the Government of the Hong Kong Special Administrative Region (“HKSAR”) and senior executives of a large Hong Kong property developer. For the purposes of this judgment, a summary of that background will suffice. A.1 The individuals concerned 3. Hui Si-yan Rafael (“Rafael Hui”) joined the Civil Service in 1970 and rose to become, in 1995, the Secretary for Financial Services of the HKSAR Government. He resigned from that position in 2000 and, on 7 June 2000, was appointed Managing Director of the Mandatory Provident Fund Scheme (“MPFS”) Authority for a period of four years. Although he resigned with effect from 1 January 2003, his service was extended until 14 August 2003. Having worked as a consultant in the private sector for a brief period (described below), Rafael Hui re-entered public service on 30 June 2005 and was appointed Chief Secretary of the HKSAR Government. He served as Chief Secretary from 1 July 2005 until 30 June 2007 and, as such, was an Official Member of the Executive Council (“ExCo”). Following the period of his appointment as Chief Secretary, Rafael Hui became a Non-Official Member of ExCo from 1 July 2007 to 20 January 2009. Rafael Hui was 1st defendant at trial and 1st appellant below and in this Court. 4. Sun Hung Kai Properties Limited (“SHKP”) is a major property developer and publicly listed company in Hong Kong under the control and management of members of the same family. At the time of the events in question, Kwok Ping-kwong Thomas (“Thomas Kwok”) and his brother Kwok Ping-luen Raymond (“Raymond Kwok”), were its Co-Chairmen and Managing Directors. Thomas Kwok was 2nd defendant at trial and 2nd appellant below and in this Court; Raymond Kwok was 3rd defendant at trial. 5. Chan Kui-Yuen (“Thomas Chan”) was an executive director of SHKP and was 4th defendant at trial and 3rd appellant below and in this Court. 6. Kwan Francis Hung-sang (“Francis Kwan”) was a close friend of Rafael Hui and was 5th defendant at trial and 4th appellant below and in this Court. A.2 Rafael Hui’s relationship and dealings with SHKP 7. In 2003, Rafael Hui began to work in the private sector as a consultant to SHKP and Thomas Kwok and Raymond Kwok, as well as their elder brother Walter Kwok. Subsequently, a written service agreement was entered into between a company controlled by Rafael Hui, Top Faith Enterprises Limited (“Top Faith”), and Sun Hung Kai Real Estate Agency Limited (“SHKR”) under which Top Faith would provide consultancy services and professional advice to SHKR from 1 March 2004 for two years. The agreement provided that SHKR would pay Top Faith an annual consultancy fee of HK$4.5 million and provide it with an office. Although the agreement was terminated in March 2005, the full HK$9 million for two years’ service was paid on the authorisation of Raymond Kwok. An office was provided to Top Faith in One IFC, a prime commercial building in Central owned by a company in which SHKP was a principal shareholder. Rafael Hui used this as his office and the expenses for it were paid by SHKR and Raymond Kwok and/or Thomas Kwok. 8. From 2000, a number of loans were advanced to Rafael Hui and Top Faith by Honor Finance Company Limited (“Honor Finance”), a SHKP subsidiary whose business was principally to provide mortgage loans to purchasers of properties developed by SHKP. A first loan was granted to Rafael Hui on 31 March 2000 for HK$900,000 for 38 months at prime lending rate and this was repaid in full with interest on the due date. A second loan was granted to Rafael Hui on 18 December 2001 for HK$1.5 million and this was also repaid in full with interest. A third loan was granted to Top Faith in May 2004 for HK$3 million for 12 months. Although interest was specified at 3% per month on overdue sums, this was never paid and the loan was extended annually until 2010. 9. On 14 February 2003, having previously lived in government quarters, Rafael Hui and his wife moved into two conjoined flats in SHKP’s Leighton Hill development.[4] The two flats were respectively owned by a Kwok family private company and a SHKP subsidiary. Subsequently, on 26 November 2003, two tenancy agreements in respect of the flats were executed, backdated to 1 April 2003, for a period of one year and at a monthly rental of HK$55,000. A.3 Dealings between SHKP and the MPFS Authority/the Government 10. The MPFS Authority rented substantial parts of One IFC, pursuant to three leases which all expired on 31 March 2004. One of those leases was signed by Rafael Hui as Managing Director of the MPFS Authority. At the end of 2002, FPD Savills recommended to the MPFS Authority that it should renew its existing leases at One IFC for six years after their expiration. This recommendation was endorsed by a Joint Committee of the MPFS Authority, of which Rafael Hui was a member. He did not attend the meeting at which the renewal decision was made but was circulated with the papers and a voting slip and approved the lease renewals. In contrast, the Chief Executive of the Hong Kong Monetary Authority (“HKMA”) abstained from voting due to his potential conflict by reason of the HKMA’s leasing options at Two IFC. On 10 June 2003, Rafael Hui signed the letter of acceptance on behalf of the MPFS Authority in respect of the lease renewals. 11. During the time when Rafael Hui was Chief Secretary, there were substantial commercial negotiations between SHKP and the HKSAR Government. These included negotiations relating to two major property development projects, namely the Ma Wan Park Project and the West Kowloon Cultural District (“WKCD”) Project, in which SHKP had a significant interest. In particular, as Chief Secretary, Rafael Hui was the Chairman of the Steering Committee of the WKCD Project. A.4 The HK$8.5m and HK$11.182m payments made to Rafael Hui 12. In the period between 27 and 30 June 2005, immediately before Rafael Hui took the oath of office as Chief Secretary, eight payments totalling HK$8.5 million were made into his account with the Standard Chartered Bank. These were made either directly or on the direction of Francis Kwan and came from a sum of HK$10.8 million which Francis Kwan had received, in the form of a cashier’s order, from an account in the name of Villalta, a company controlled by Thomas Chan and his family. 13. The payments of HK$8.5 million to Rafael Hui came from and were made on the ultimate directions of Thomas Kwok, who paid HK$4 million and HK$4.8 million to Thomas Chan by cheques dated 20 and 28 June 2005 respectively. 14. In the period between 21 November and 12 December 2007, payments totalling HK$10.182 million were made by Francis Kwan from his account with the Hang Seng Bank to the account of Rafael Hui with the Standard Chartered Bank. Further, on 21 November 2007, a payment of HK$1 million was made by Francis Kwan from his account with the Hang Seng Bank to the account of Top Faith. 15. The prosecution alleged that the source of the HK$10.182 million was ultimately also Thomas Kwok. On 9 November 2007, HK$12 million was remitted from an account in the name of Villalta, operated by Thomas Chan and his family, to an account with DBS Bank in Singapore in the name of Wedingley Limited (“Wedingley”), a BVI company. On 20 November 2007, US$1.543 million was remitted from Wedingley’s DBS account to an account of Francis Kwan with DBS Bank in Hong Kong, where it was split into two time deposits. Loans raised on the security of those deposits were the source of the HK$10.182 million paid to Rafael Hui. Subsequently, on 22 and 30 April 2008, Thomas Kwok paid two cheques to Thomas Chan for HK$5 million and HK$7 million respectively and, on 30 April 2008, Thomas Chan transferred HK$13 million to the account of Villalta. On 7 November 2008, Thomas Kwok paid Thomas Chan a further HK$6 million. A.5 The non-disclosures 16. In his capacity as Managing Director of the MPFS Authority and as a member of ExCo, both during his tenure as Chief Secretary and thereafter as a Non-Official Member, Rafael Hui was subject to duties[5] to make disclosures of any interests that might give rise to conflicts between his personal interests and his public duties. In addition, any member of ExCo with interests in a specific topic to be discussed in an ExCo meeting was required to give notification and whether he should participate in the meeting would be determined by the Chief Executive. 17. At no time did Rafael Hui disclose to the MPFS Authority his relationship with SHKP, including in particular the first and second loans from Honor Finance to him or the fact that he had been occupying the two Leighton Hill flats without payment of rent since February 2003. 18. When he was Chief Secretary and during his time as a member of ExCo, Rafael Hui made no express declaration or disclosure of interest arising from his relationship with SHKP. His only declarations of interests were limited to the ownership of a flat in Wanchai, his connection with the Hong Kong Jockey Club and his status as a prospective or existing civil service pensioner. B. The proceedings below 19. The appellants, together with Raymond Kwok, were tried together on an indictment containing eight counts before Macrae JA and a jury.[6] 20. Rafael Hui was charged with: three offences of misconduct in public office (Counts 1, 6 and 8); three offences of conspiracy to commit misconduct in public office (Counts 2, 3 and 5); one offence of furnishing false information (Count 4); and one offence of conspiracy to offer an advantage to a public servant (Count 7). 21. Thomas Kwok was charged with: two offences of conspiracy to commit misconduct in public office (Counts 2 and 5); and one offence of conspiracy to offer an advantage to a public servant (Count 7). 22. Thomas Chan and Francis Kwan were each charged with: one offence of conspiracy to commit misconduct in public office (Count 5); and one offence of conspiracy to offer an advantage to a public servant (Count 7). B.1 The indictment 23. Count 1 alleged that, between 7 June 2000 and 13 August 2003, Rafael Hui, as the Managing Director of the MPFS Authority, without reasonable excuse or justification, wilfully misconducted himself in relation to his public office, by failing to declare or disclose to the MPFS Authority, that he had accepted the rent-free use of the Leighton Hill flats, the unsecured first and second loans from Honor Finance, and his negotiations for a consultancy agreement with SHKP.[7] 24. Count 2 alleged that Rafael Hui and Thomas Kwok conspired together, between 1 March 2005 and 30 June 2007, that Rafael Hui, whilst Chief Secretary of the HKSAR Government, without reasonable excuse or justification, would wilfully misconduct himself in the course of or in relation to his public office, by being or remaining favourably disposed to SHKP[8] in return for the payment of HK$5 million from Thomas Kwok.[9] 25. Count 3 alleged that Rafael Hui and Raymond Kwok conspired together, between 1 March 2005 and 30 June 2007, that Rafael Hui, whilst Chief Secretary of the HKSAR Government, without reasonable excuse or justification, would wilfully misconduct himself in the course of or in relation to his public office, by being or remaining favourably disposed to SHKP in return for the payment of HK$4.125 million from SHKP.[10] 26. Count 4 alleged that Rafael Hui and Raymond Kwok furnished a false invoice, with a view to gain for themselves or another or with intent to cause loss to another, which purported to show that the payment of HK$4.125 million was in respect of Rafael Hui’s services for the period from April 2005 to February 2006.[11] 27. Count 5 alleged that Rafael Hui, Thomas Kwok, Raymond Kwok, Thomas Chan and Francis Kwan conspired together, between 1 March 2005 and 30 June 2007, that Rafael Hui, whilst Chief Secretary of the HKSAR Government, without reasonable excuse or justification, would wilfully misconduct himself in the course of or in relation to his public office, by being or remaining favourably disposed to SHKP in return for the sum of HK$8.5 million through a series of payments from Thomas Kwok, Raymond Kwok, Thomas Chan and Francis Kwan.[12] 28. Count 6 alleged that Rafael Hui, whilst Chief Secretary of the HKSAR Government, between 30 June 2005 and 30 June 2007, without reasonable excuse or justification, wilfully misconducted himself in public office, by failing to declare or disclose to the Government of the HKSAR the provision to him of the annual extensions of the unsecured third loan from Honor Finance.[13] 29. Count 7 alleged that Rafael Hui, Thomas Kwok, Raymond Kwok, Thomas Chan and Francis Kwan conspired together, between 30 June 2005 and 20 January 2009, without lawful authority or reasonable excuse, to offer an advantage, namely HK$11.182 million, to Rafael Hui as Chief Secretary of the HKSAR Government and then a Non-Official Member of ExCo, as an inducement or reward for or otherwise on account of Rafael Hui performing or abstaining from performing or having performed or abstained from performing an act in his capacity as a public servant, namely being or remaining favourably disposed to SHKP.[14] 30. Count 8 alleged that, between 1 July 2007 and 20 January 2009, Rafael Hui, without reasonable excuse or justification, wilfully misconducted himself in public office as a Non-Official Member of ExCo, by failing to declare or disclose to the Government of the HKSAR his receipt of HK$11.182 million through a series of payments from Thomas Kwok, Raymond Kwok, Thomas Chan and Francis Kwan.[15] 31. Although the other charges on the indictment are relevant by way of background to this judgment, the issue of law raised in these appeals arises only in relation to Count 5 on the indictment. B.2 The prosecution case on Count 5 32. The prosecution case on Count 5 alleged that the payments totalling HK$8.5 million to Rafael Hui in June 2005, shortly before he became Chief Secretary of the HKSAR Government, were made pursuant to a conspiratorial agreement for Rafael Hui to misconduct himself in public office. It was known by the time the payments were made that Rafael Hui would be appointed Chief Secretary. The payments were secret and unexplained and elaborate lengths were taken to ensure that they were concealed and not recorded in any documents. The prosecution alleged that the payments to Rafael Hui were made as a “general sweetener” made to secure his “favourable disposition”, as Chief Secretary, to SHKP and that they were, in effect, “bribes”. It was further contended that Thomas Chan contributed his own monies to the payments and that Francis Kwan was rewarded for his part in the transactions. B.3 The no case submission and ruling 33. At the close of the prosecution case at trial, the appellants’ counsel made submissions that there was no case for the appellants to face on Count 5. These submissions were made on the basis that the prosecution case had shifted in the course of its presentation from one alleging that Rafael Hui was paid the sum of HK$8.5 million to be the “eyes and ears” of SHKP and that he had showed favour in some way to the company to one alleging that he was paid “general sweetners” for doing nothing more than his normal duty. It was submitted that, as a matter of law, this could not constitute misconduct in public office.[16] 34. The trial judge rejected the submissions that there had been any shift in the prosecution’s position and, on its validity as a matter of law, concluded: “Their position, as it was explained to the jury, and as I have just referred, is that it is not necessary for the prosecution to prove that D1 was in fact favourable to SHKP Ltd. Their case is that D1 was paid sweeteners so that he would be or remain favourably disposed to SHKP Ltd. He thereby became their man in government, regardless of whether he did in fact use his eyes and ears, or do anything favourable to SHKP Ltd. That, as I understand it, has always been their position, and I see no shift in position that would warrant my intervention at this stage upon a submission of ‘no case to answer’. It is a legitimate way of putting the prosecution case in law, and I find a case to answer on the evidence in relation to those counts. That is my ruling.”[17] B.4 The defence case 35. Although each of the appellants advanced his own case to the various charges, there were common elements of their defences. In summary, when Rafael Hui was intending to leave the MPFS Authority and moving to the private sector, he was concerned with remuneration and accommodation. He eventually agreed to work as a consultant for SHKP because Thomas Kwok offered to pay him HK$15 million per year and to provide living quarters as well as reimbursing his office expenses. Although the remuneration stated in the written service agreement was understated at HK$4.5 million per year, Rafael Hui signed the agreement on the understanding that Thomas Kwok would make up the difference and pay for his accommodation. 36. Upon accepting the invitation of the Chief Executive-elect[18] to become Chief Secretary, Rafael Hui terminated the service agreement and sought payment of the outstanding sums due to him from SHKP. Taking into account the payments made up to April 2005, there was a sum of HK$10.8 million due. The HK$8.5 million which featured in Count 5 was the payment to Rafael Hui of that outstanding amount, with the balance of HK$2.3 million being retained by Francis Kwan in order for the latter to carry out foreign exchange transactions for Rafael Hui. The payment of the HK$8.5 million was made in a convoluted manner due to a family dispute within the Kwok family concerning Walter Kwok. In short, the defence in relation to the payment which formed the subject of Count 5 was that this was the balance of monies legitimately owing to Rafael Hui from SHKP and the Kwoks. It was therefore denied that the payment was a bribe. 37. So far as the payment of HK$11.182 million in 2007 was concerned, different cases were advanced but with the common thread that this was not a payment from SHKP or the Kwoks to Rafael Hui at all. It was the case of Francis Kwan and Rafael Hui that this sum was paid by Francis Kwan to Rafael Hui on behalf of a Mainland third party and its purpose was to alleviate Rafael Hui’s financial difficulties caused by his expensive lifestyle and with a view to his remaining Chief Secretary. Thomas Kwok’s case was that the monies paid by him to Thomas Chan were bonuses to the latter. Thomas Chan’s case was that Villalta paid the sum of HK$12 million to Wedingley as part of an arrangement for Francis Kwan to manage this money on behalf of Thomas Chan. 38. It was also the defence case that Rafael Hui did not show any favour to SHKP whilst he was in office as Chief Secretary or as a Non-Official Member of ExCo, in particular in relation to either the Ma Wan or WKCD Projects. B.5 The trial judge’s summing up on Count 5 39. The trial judge instructed the jury that in order to find a conspiracy it was necessary for them to find that there had been an agreement as alleged by the prosecution. His summing up on the charge of conspiracy to commit misconduct in public office in Count 5 reflected the prosecution case as to the law of misconduct in public office. The material part of that summing up where the judge dealt with the third element of the offence as identified by Sir Anthony Mason NPJ in Sin Kam Wah & Another v HKSAR,[19] namely “wilfully misconducts himself; by act or omission, for example, by wilfully neglecting or failing to perform his duty” was in the following terms: “The act of wilful misconduct alleged by the prosecution in these three counts, Counts 2, 3 and 5, and this is the third requirement the prosecution must prove, is that D1, as Chief Secretary, would be or remain favourably disposed to SHKP’s interests in return for certain specified payments. Although the particulars of each count do not mention the word, the prosecution say that these payments were, in effect, bribes. The payments are, of course, different in each count. In Count 2, it is $5 million; in Count 3, $4.125 million; and in Count 5, $8.5 million. But the objective of the conspiracy alleged, namely, that D1 should misconduct himself as Chief Secretary by being favourably disposed to SHKP’s interests in return for money is the same. So, members of the jury, what does ‘being or remaining favourably disposed to SHKP Limited and/or its subsidiaries or associated companies and/or named conspirators’ mean? What does it mean? The words ‘being or remaining favourably disposed to’ reflect the prosecution allegation that the payments made to D1 were general goodwill payments. In other words, they were sweeteners, payments made to sweeten a public official. The counts do not allege that the payments were made in return for any specific identifiable favour which D1 was to perform or had performed. And the law does not require that any particular favour be specifically identified. Indeed, in the context of paying any public official a sweetener, which you might think would necessarily involve subtlety and secrecy, it may be very difficult to show that any favour was in fact performed. A public official may have acted in exactly the same way, or objectively made the correct decision, or done the right thing, whether he was sweetened or not. The reason the law does not require a particular favour to be identified is, you may think, obvious. The acceptance of money by a public official in return for him in a general way, rather than in a specific way, being favourably disposed to the person or persons giving him the money, is itself capable of amounting to misconduct by virtue of the breach of the duties and obligations he owes to the public as a public official. The evil or the vice of these kinds of payments is that no one could have confidence in the acts of a public official who, through the offer and acceptance of money, has been kept sweet by private interests. And the confidence which the public are entitled to have in the fair and impartial performance of a public official’s duties and obligations is thereby eroded or destroyed.”[20] B.6 Verdicts and sentences 40. In respect of the appellants, the jury returned verdicts of guilty on some of the charges against them and not guilty on the others: thus, Rafael Hui was found guilty of Counts 1, 5, 6, 7 and 8 and acquitted of Counts 2, 3 and 4; Thomas Kwok was found guilty of Count 5 and acquitted of Counts 2 and 7; and Thomas Chan and Francis Kwan were both found guilty of Counts 5 and 7. The jury acquitted Raymond Kwok of the four charges on which he stood trial, namely Counts 3, 4, 5 and 7. 41. The judge imposed sentences on the appellants as follows: Rafael Hui was sentenced to a total term of imprisonment of 7 years and 6 months and ordered to pay the HKSAR Government the sum of HK$11.182 million;[21] Thomas Kwok was sentenced to 5 years’ imprisonment and fined HK$500,000 and disqualified from acting as a director of any company for a period of 5 years; Thomas Chan was sentenced to a total term of imprisonment of 6 years and fined HK$500,000 and disqualified from acting as a director of any company for a period of 6 years; and Francis Kwan was sentenced to a total term of imprisonment of 5 years.[22] B.7 The Court of Appeal decision 42. The appellants each appealed their convictions to the Court of Appeal and, in addition, Thomas Chan appealed against his sentence. The appeals against conviction were each dismissed for the reasons set out in a lengthy judgment.[23] Although the Court of Appeal allowed Thomas Chan’s appeal against sentence, the totality of sentence imposed in respect of his convictions remained the same, namely 6 years’ imprisonment. 43. In respect of the appeal against conviction on Count 5, Yeung VP noted that the jury had clearly rejected the defence case and must have concluded that the HK$8.5 million was paid to Rafael Hui because of his appointment as Chief Secretary and that in return for the payment Rafael Hui had agreed to be or to remain favourably disposed towards SHKP.[24] Lunn VP observed that, in convicting, the jury rejected the evidence of Rafael Hui and Thomas Kwok that the payments of HK$8.5 million to Rafael Hui were legitimate payments arising out of their undocumented oral agreement, by which he was to be compensated in addition to the payments made pursuant to the written service agreement by which, through Top Faith, he provided consultancy services and advice to SHKP.[25] 44. Both Yeung VP and Lunn VP, for the reasons set out in their respective judgments (with both of which Pang JA agreed), considered the validity of the prosecution case on Count 5 and concluded that, on the facts alleged and accepted by the jury to be proved, the offence of conspiracy to engage in misconduct in public office was established.[26] C. Misconduct in public office and bribery in Hong Kong C.1 Misconduct in public office 45. For the purposes of the law of Hong Kong, the elements of the common law offence of misconduct in public office were stated by Sir Anthony Mason NPJ in Sin Kam Wah & Another v HKSAR[27] as follows: “The offence is committed where: (1) a public official; (2) in the course of or in relation to his public office; (3) wilfully misconducts himself; by act or omission, for example, by wilfully neglecting or failing to perform his duty; (4) without reasonable excuse or justification; and (5) where such misconduct is serious, not trivial, having regard to the responsibilities of the office and the officeholder, the importance of the public objects which they serve and the nature and extent of the departure from those responsibilities.”[28] That was a re-formulation, in a presently immaterial respect, of what Sir Anthony Mason NPJ had earlier said in Shum Kwok Sher v HKSAR.[29] The re-formulation was to take account of developments in the law in respect of the concepts of wilfulness and recklessness.[30] The statement has since been applied in a number of cases, including HKSAR v Wong Lin Kay,[31] and was accepted in argument in the present appeals.[32] 46. In R v Boulanger[33] the Supreme Court of Canada, having pointed out[34] that the Canadian statutory offence of breach of trust by a public officer can be traced to the common law offence of misconduct in public office, said[35] that the actus reus of the offence “defies precise definition because of the range of conduct that it is designed to cover”. That it may include solicitation or acceptance of a bribe is clear. Professor Finn, in his article on “Official Misconduct”[36] gives “entering into secret commission agreements while acting in an official agency capacity” as an example of the “fraud in office” type of misconduct, and notes that it may also constitute an offence under secret commissions legislation. It can be said that the receipt of bribes is a quintessential example of misconduct in public office. 47. For behaviour to qualify as relevant misconduct it may, but need not, involve a contravention of a statute. It must, however, have the necessary link to official powers, duties or responsibilities. Yet, not every breach of the law by a person when he or she is a public official is in the course of or in relation to the office held. In Sin Kam Wah and Anor v HKSAR[37] the relevant conduct was not in the performance of the police officer’s duties, but was found to have such a relation with his public office as to bring that office into disrepute. In R v Quach[38] Redlich JA, with whom the other members of the Victorian Court of Appeal concurred, approved[39] Professor Finn’s statement that “the kernel of the offence is that an officer, having been entrusted with powers and duties for the public benefit, has in some way abused them, or has abused his official position”, and said:[40] “In my opinion the relevant misconduct need not occur while the officer is in the course of performing a duty or function of the office. Certain responsibilities of the office will attach to the officer whether or not the officer is acting in the course of that office. Where the misconduct does not occur during the performance of a function or duty of the office, the offence may be made out where the misconduct is inconsistent with those responsibilities. It may be connected to a duty already performed or to one yet to be performed or it may relate to the responsibilities of the office in some other way. The misconduct must be incompatible with the proper discharge of the responsibilities of the office so as to amount to a breach of the confidence which the public has placed in the office, thus giving it its public and criminal character.” 48. The characterisation of the misconduct alleged may involve both a descriptive element (of the facts of the transaction) and a value judgment (of its effect). For example, the Australian case of R v Boston[41] concerned a conspiracy to make a corrupt payment to a member of the New South Wales Parliament to induce him to use his official power in an improper way. The criminality of the alleged agreement lay in its tendency to produce a public mischief.[42] Members of the High Court characterised the public mischief. Knox CJ said:[43] “Payment of money to a member of Parliament to induce him to persuade or influence or put pressure on a Minister to carry out a particular transaction tends to the public mischief in many ways, irrespective of whether the pressure is to be exercised by conduct inside or outside Parliament. It operates as an incentive to the recipient to serve the interest of his paymaster regardless of the public interest, and to use his right to sit and vote in Parliament as a means to bring about the result which he is paid to achieve. It impairs his capacity to exercise a disinterested judgment on the merits of the transaction from the point of view of the public interest, and makes him a servant of the person who pays him, instead of a representative of the people.” This, of course, is a reference to a specific transaction. But Isaacs and Rich JJ put the public mischief in wider terms in the following passage:[44] “[The member] has … placed himself in a situation embarrassing and inconsistent with that independence to criticize or censure which he is bound to preserve; he has fastened upon himself golden fetters which preclude his freedom of action. The natural fear of exposure or reproach, or the sense of personal obligation, must inevitably operate to dissuade him from fearlessly pursuing the path of true service …”. 49. The public mischief that was the object of the conspiracy in Boston did not lie either in the bare financial transaction involved in the payment or in the making by a parliamentarian of representations about government action. It lay in the connection between the two: the representations were to be made in return for the payment. The payment took its character from the purpose for which it was made, and by accepting it the recipient placed himself in a situation incompatible with the responsibilities of his office. 50. The present appeals, like Boston, involve a charge of conspiracy. Section 159A of the Crimes Ordinance (Cap.200) provides: “159A. The offence of conspiracy (1) Subject to the following provisions of this Part, if a person agrees with any other person or persons that a course of conduct shall be pursued which, if the agreement is carried out in accordance with their intentions, either— (a) will necessarily amount to or involve the commission of any offence or offences by one or more of the parties to the agreement; or (b) would do so but for the existence of facts which render the commission of the offence or any of the offences impossible, he is guilty of conspiracy to commit the offence or offences in question.” C.2 Bribery at common law 51. The common law offence of bribery is described in Russell on Crime[45] as “receiving or offering any undue reward by or to any person whatsoever, in a public office, in order to influence his behaviour in office, and incline him to act contrary to the known rules of honesty and integrity.” The purposive element of that formula is stated from the point of view of the offeror rather than the recipient. The motive of the recipient may be simple greed, but it is the offeror’s purpose which characterises the payment. C.3 Statutory formulations of bribery 52. As the statement in Russell on Crime indicates, the common law was concerned with bribery of persons in public office (originally, it appears, judicial officers[46]), but concern with the practice of payment of secret commissions to “agents” in the private sector resulted in legislation which extended the reach of anti-corruption law. The history of United Kingdom legislation on the topic, culminating in the Bribery Act 2010 (UK), was examined by the Court of Appeal in 2013 in R v J (P) and others[47] and in 2016 in R v AIL (a company) and others.[48] Originally, three Acts were of particular significance: the Public Bodies Corrupt Practices Act 1889, the Prevention of Corruption Act 1906 and the Prevention of Corruption Act 1916. The 1889 Act provided: “(1) Every person who shall by himself or by or in conjunction with any other person, corruptly solicit or receive, or agree to receive, for himself, or for any other person, any gift, loan, fee, reward, or advantage whatever as an inducement to, or reward for, or otherwise on account of any member, officer, or servant of a public body as in this Act defined, doing or forbearing to do anything in respect of any matter or transaction whatsoever, actual or proposed, in which the said public body is concerned, shall be guilty of a misdemeanor. (2) Every person who shall by himself or by or in conjunction with any other person corruptly give, promise, or offer any gift, loan, fee, reward, or advantage whatsoever to any person, whether for the benefit of that person or of another person, as an inducement to or reward for or otherwise on account of any member, officer, or servant of any public body as in this Act defined, doing or forbearing to do anything in respect of any matter or transaction whatsoever, actual or proposed, in which such public body as aforesaid is concerned, shall be guilty of a misdemeanor.” 53. The 1906 Act dealt with what are often referred to as secret commissions, and extended to private enterprise. The 1916 Act addressed payments to public employees responsible for awarding contracts. 54. The first Hong Kong legislation on the subject was the Misdemeanors Punishment Ordinance 1898 (No.1 of 1898) which used language closer to that of Russell on Crime (although, it should be added, the word “corruptly” in the United Kingdom legislation served a similar purpose). The long title was “An Ordinance for the more effectual punishment of bribery and certain other misdemeanors”. Sections 3 and 4 were as follows: “(3) Every public servant who accepts, or obtains, or agrees to accept, or attempts to obtain, or causes or procures to be obtained, from any person … any bribe, with a view to influence his conduct as such public servant or to incline him, contrary to his duty as such public servant or contrary to the rules of honesty and integrity, to do or to omit to do any act, shall be guilty of a misdemeanor … (4) Every person who gives or offers, or causes or procures to be given or offered, to any public servant any bribe, for himself or for any other person, with a view to influence his conduct as such public servant or to incline him, contrary to his duty as such public servant or contrary to the rules of honesty and integrity, to do or to omit to do any act, shall be guilty of a misdemeanor …”. 55. “Bribe” was defined in section 2 to include “any fee, perquisite, reward, or gratification, whether pecuniary or otherwise, not payable or receivable by law”. Again, the expression of the purposive element was somewhat more natural in (4) than in (3), but obtaining improper influence or securing a disloyal inclination was the essence of the purposive element of the offence. It is the tendency of a payment or other advantage which stamps it as corrupt and a bribe. 56. Neither the United Kingdom legislation nor the Hong Kong Ordinance addressed an issue identified, but neither pursued nor resolved, in a 1901 Scottish case of HM Advocate v Dick,[49] that is, the case of a payment, otherwise having the characteristics of a bribe, to a person before and in anticipation of that person’s assuming public office. Anti-bribery legislation, whether directed towards the conduct of or in relation to people in public office or extended to cover people in other positions, typically relates to the current position of the office-holder, (see for example, the definition of “public servant” in the 1898 Ordinance), at the time of the payment or other advantage. If a legislature were to take up the matter of corrupt investment in potential future benefits by pre-office payments or other advantages, it would no doubt be confronted by some policy choices. 57. The next Hong Kong legislation was the Prevention of Corruption Ordinance 1948.[50] The Acting Attorney General said the purpose of the law was both to cover corruption by “agents” and to make the offence of corruption in office more comprehensive. The legislation followed the language of the earlier United Kingdom legislation. Section 3 dealt with corruption by or in relation to public servants. Section 4 dealt with corruption by or in relation to agents. Sub-sections (1) and (2) of section 3 substantially followed sub-sections (1) and (2) of the 1889 United Kingdom Act, making the offences corruptly soliciting or receiving an advantage or corruptly giving, promising or offering an advantage. As in the United Kingdom legislation, the proscribed advantage was as an inducement to, or reward for, or otherwise on account of doing or forbearing to do something. 58. The current Hong Kong legislation, the Prevention of Bribery Ordinance 1970 (Cap.201) widened the kinds of advantage that were proscribed,[51] and, so far as presently relevant, expressed the offence of bribery in relation to a public servant as follows: “4.(1) Any person who, without lawful authority or reasonable excuse, offers any advantage to a public servant as an inducement to or reward for or otherwise on account of that public servant’s — (a) performing or abstaining from performing, or having performed or abstained from performing, any act in his capacity as a public servant; (b) expediting, delaying, hindering or preventing, or having expedited, delayed, hindered or prevented, the performance of an act, whether by that public servant or by any other public servant in his or that other public servant’s capacity as a public servant; or (c) assisting, favouring, hindering or delaying, or having assisted, favoured, hindered or delayed, any person in the transaction of any business with a public body, shall be guilty of an offence. (2) Any public servant who, without lawful authority or reasonable excuse, solicits or accepts any advantage as an inducement to or reward for or otherwise on account of his — (a) performing or abstaining from performing, or having performed or abstained from performing, any act in his capacity as a public servant; (b) expediting, delaying, hindering or preventing, or having expedited, delayed, hindered or prevented, the performance of an act, whether by himself or by any other public servant in his or that other public servant’s capacity as a public servant; or (c) assisting, favouring, hindering or delaying, or having assisted, favoured, hindered or delayed, any person in the transaction of any business with a public body, shall be guilty of an offence.” 59. In its current form, section 4 has been given an extra-territorial reach, which is presently immaterial, but otherwise it is substantially the same. The Ordinance goes on to deal with more specific kinds of corruption, some of which involve public servants and others of which may not. It also establishes a special investigative regime. By comparison with the 1948 Ordinance it replaces “corruptly” with “without lawful authority or reasonable excuse”. This was unlikely to have been intended to have a narrowing effect.[52] The pattern of legislative development in Hong Kong, as in the United Kingdom and elsewhere, has been to widen the anti-corruption net, and to deal with specific forms of conduct thought to require particular attention, but there is no indication of an intention to abandon the original concept of purchasing improper influence or an inclination to act contrary to duty. It is to be noted that the concept of inclination had been in Hong Kong’s statute book since 1898 (see [54] above). C.4 Is a specific act or omission required? 60. In 1978 there arose an issue as to the effect of the 1970 Ordinance in a case where the facts were relatively minor (an oblique solicitation of a small gift by a postman) but which was thought to involve an important principle. The question was whether, in order to establish a contravention of section 4(2)(a), it is sufficient to show that a public servant, in the course of or in relation to his duty, solicited an advantage or whether it was also necessary to prove that either the payer or the payee had in contemplation some specific act or omission on the part of the payee. The case was Attorney General v Chung Fat Ming.[53] McMullin J said:[54] “[T]he distinction which has been argued before us is between the advantage which is seen to be solicited or accepted as a ‘quid pro quo’ for some particular act or abstention identifiable as to place and time on the one hand and, on the other, an advantage solicited or accepted as a general earnest of good relations-the ‘keeping sweet’ situation.” The argument which the Court of Appeal accepted was “that section 4 does not require the Crown to prove any specific act in contemplation of either party as being related to the particular advantage sought [or offered]”.[55] That is the proposition for which the case is authority. It is a negative proposition and invites the further question: what, then, is it necessary or sufficient for the prosecution to prove? A person who pays a bribe may be taken to expect, or at least hope, for some kind of value for money, but there can be an almost limitless variety of circumstances bearing upon the extent to which such hope or expectation takes any concrete form. In a case of extortion, the payer may have nothing to hope for, but in the case of a bribe the benefit in prospect may (perhaps deliberately) be left completely undefined. McMullin J’s reference to “an advantage solicited or accepted as a general earnest of good relations” was to an advantage solicited or accepted by a public servant in the course of or in relation to his duty. That context gives colour to the concept of “good relations”. 61. In Attorney General v Chung Fat Ming,[56] and in an earlier case of Chan Wing Yuen v The Queen,[57] McMullin J said that the phrase “otherwise on account of” (which was in the original United Kingdom legislation) goes beyond “inducement” and “reward”, which relate to specific acts, and covers “an act” which may denote the entire range of a public servant’s duties. In a passage from Chan Wing Yuen v The Queen which Leonard J quoted and adopted in Chung Fat Ming,[58] McMullin J said: “The substance of the offence created by section 4(2) is in the soliciting or accepting of an advantage by one acting in the capacity of a government servant under circumstances which make it clear that the advantage is solicited or offered in reference to his having such capacity. It is a feature of the particular evil at which those provisions are aimed that the purpose may be generally apparent without being entirely explicit. In a given case it may be that the evidence available to the prosecution is not sufficiently specific to indicate whether the advantage has been solicited or offered as an ‘inducement’ to do a particular act or as a ‘reward’ for having done it. In such a case the charge should no doubt properly speaking be one of offering or accepting the advantage ‘on account of’ some prospect of favour not particularized but discernible among a variety of possible acts within the public capacity of the accused.” Following an expression of his agreement with that passage, Leonard J went on to say[59] that, in a case of solicitation contrary to section 4(2)(a), the solicitation must be on account of performance of an “act” in the person’s capacity as a public servant, and that he would regard being or remaining favourably disposed to the person solicited as sufficient to amount to an “act”. In the context, he was evidently referring to “some prospect of favour not particularized but discernible among a variety of possible acts within the public capacity of the accused” on account of which the advantage was solicited. 62. The same question arose in the New South Wales Court of Criminal Appeal in 1992 in R v Allen,[60] a case concerning the common law misdemeanour of bribery. The appellant, an Assistant Commissioner of Police, had formed an association with an operator of licensed nightclubs in Sydney’s Kings Cross area. Evidently placed in funds by his associate, the appellant set out to corrupt a junior police officer, whose responsibilities covered that area, by making a number of unsolicited cash gifts. Ultimately, the officer reported the matter. The case was tried before a judge sitting without a jury and the appellant was convicted of bribery. On appeal it was argued that there was an absence of a necessary quid pro quo because the recipient of the payments was never asked to do anything, and never agreed or intended to do anything, in consideration for them. The Court of Criminal Appeal said that a corrupt intent on the part of the appellant had been shown and that was sufficient. The size and regularity of the payments suggested an intention to place the junior officer under a sense of obligation to the appellant, and the systematic nature of the dealings suggested that the junior officer was being placed on some kind of payroll. The Court said:[61] “In its application to circumstances such as the present, the gravamen of the offence of bribery is the making or offering of a payment with an intent to incline a person in public office to disregard his duty. The occasion for the disregard of duty need not have arisen at the time of the offence, and it need never arise. Nor is it necessary that the particular kind of contemplated breach of duty be specified at the time of the payment or inducement.” 63. The risk associated with euphemisms such as “sweetener”, and “being … favourably disposed”, while they are understandable because the argot of corruption commonly employs understatement and evasion, is that they may be taken as signifying a purely emotional state, of a kind that could be benign or even involuntary. Protection ordinarily denotes care and concern, but in the context of extortion it means something different. Similarly, goodwill is a sentiment that a public servant may feel towards honest citizens generally, but in the context of what may be purchased by a bribe it means something different. It was pointed out in argument that, in one sense, a person may be favourably disposed to another for all manner of reasons that are innocent, or at least legally neutral. Furthermore, the trial was conducted on the basis that if the defence explanation of the reason for the payment of HK$8.5m were accepted (at least to the level of raising a reasonable doubt), then the prosecution must fail; yet, even on that explanation, the new Chief Secretary might be expected to have felt general goodwill towards a former employer or client by whom he had been well remunerated, and with whom he was and would be likely to remain on cordial terms. In relation to the payment of HK$8.5 million there was no charge of a contravention of section 4 of the Prevention of Bribery Ordinance because of a matter of timing. On the facts alleged by the prosecution, however, if the payment had been made a week later there would have been an offence against the Ordinance. In that event, what would have given the payment the character of a bribe would have been its tendency to incline the recipient to disregard his public duty; not his emotional state. 64. To return to the offence of misconduct in public office, which may overlap with bribery, in Sin Kam Wah & Another v HKSAR, Sir Anthony Mason NPJ, with whom the other members of the Court agreed, after explaining that the trial in that case had been conducted on the basis that the relevant misconduct lay in the first appellant’s part in accepting sexual services from women over whom he knew the second appellant was exercising control, direction or influence, added:[62] “I should make it clear, however, that acceptance of a ‘general sweetener’ by a public officer can, in appropriate circumstances, amount to misconduct in public office.” D. The arguments on appeal D.1 Background 65. To understand the competing arguments presented to this Court, and therefore the issue at hand, it is convenient to record first what was not asserted by the respondent to be the allegation advanced by the particulars of offence in Count 5. 66. It was not alleged by the prosecution at trial and it is not contended before this Court: (1) that Rafael Hui’s act of misconduct in public office was his receipt and acceptance in June 2005 of the sums amounting to HK$8.5 million. That is because, in June 2005, he did not occupy public office; (2) that of itself, the fact of favourable disposition on Rafael Hui’s part towards SHKP, after he assumed office as Chief Secretary, was an act of misconduct. That is because disposition, whether favourable or unfavourable, is a common and largely unavoidable incident of every day experience which may or may not bear an impermissible consequence. Whereas the failure by a public officer to disclose a bias might, in certain circumstances, constitute misconduct, it seems obvious that the mere existence of a disposition, in a vacuum as it were, does not; or (3) that the agreement was that upon and after the assumption of office, Rafael Hui would, in return for the sum of HK$8.5 million, show favour to SHKP if and when that should become necessary or desirable. 67. The last of these three points, in particular the phrase “if and when that should become necessary or desirable” requires some explanation. The case for the appellants at trial was that if, contrary to the defence case, the payments to Rafael Hui in June 2005 were not monies due for his work as a consultant, it was incumbent upon the prosecution to prove that there was agreed an act or acts to be performed by Rafael Hui, once in office, as the quid pro quo for these payments and that mere favourable disposition was not an act. It was however conceded by the appellants that if an agreement were proved that in return for the payments, Rafael Hui would show favour should that become necessary or, put another way, as and when opportunity presented itself, the agreement necessarily envisaged an act or acts, and that it could not then validly be argued that that anticipated conduct did not amount to misconduct. 68. The point was raised specifically by Mr Ian Winter QC, for Thomas Chan, in submissions made to the trial judge about proposed directions to the jury. What was at issue, said Mr Winter, was whether it was alleged that the agreement was one which contemplated Rafael Hui acting “should it become necessary to do so” or was it “just an agreement that [Rafael Hui] would remain favourably disposed”?[63] The prosecution’s response at trial was that the agreement alleged was no more than an agreement to be or to remain favourably disposed. Whilst it was open to the jury to conclude that acts of specific favour had been carried out by Rafael Hui, and that such acts, if proved, would support the allegation of favourable disposition, the long and the short of the prosecution case was favourable disposition in return for money and that that sufficed. 69. Mr David Perry QC explained to this Court why the prosecution chose not to tie its colours to the mast of a conditional conspiracy, meaning a conspiracy to show favour if and when the occasion presented itself. Amongst the several considerations advanced was that such a charge would raise the question whether the course of conduct contemplated would necessarily result in the commission of a criminal offence; would generate arguments about what was meant by “necessary” or “appropriate”; and would give rise to a danger that focus would shift from the character of the payments to the merits of decisions taken, thereby creating unnecessary complexity. 70. The point was urged upon us that, by framing its case as it did, the prosecution was seeking to widen the reach of the offence of misconduct in public office because it perceived a gap in the law’s armoury against corruption in the realm of payments made in anticipation of office.[64] This as a matter of principle was impermissible but there was, in any event, no such gap, so the argument ran, because it would have been open to the prosecution to allege as the act of misconduct the non-disclosure by Rafael Hui, to whomsoever potential conflicts of interest fell to be disclosed, the fact of the June payments. Mr Perry explained that whilst such a charge could properly have been laid, it would have failed adequately to reflect the criminality of the case, for the criminality did not merely lie in the failure to disclose but in the corrupt payments and receipt of monies and in the impermissible use of office for private gain. D.2 The appellants’ argument 71. The essence of the appellants’ argument is at a fundamental level. It is to say that no criminal offence is committed without an actus reus, an external conduct element, and that the actus reus of the offence of misconduct in public office is either misfeasance, in other words, an act of misconduct, or non-feasance, an act of omission amounting to misconduct. For the purpose of the present case, we are concerned only with misfeasance, yet one looks in vain, according to this argument, for the actus reus, the act of misconduct, since being or remaining favourably disposed is not an act. It is said that the case was not left to the jury on the footing that the favourable disposition was intended to lead to action or that the payments in June were to secure a disposition which would necessarily and materially impair Rafael Hui in the performance of his public duties but was instead left on the basis that what was intended was no more than the creation in Rafael Hui of goodwill towards SHKP. 72. Viewed in that way, what was alleged to constitute the intended misconduct was nothing other than a state of mind, a mental disposition, whereas the criminal law does not recognise mere motive, intention, disposition or inclination as sufficient for the commission of a criminal offence. It is the principle which was articulated by Lord Mansfield that “so long as an act rests in bare intention, it is not punishable by our laws”,[65] and Professor Glanville Williams has suggested that the reasons for the rule are, first, “the difficulty of distinguishing between day-dream and fixed intention in the absence of behaviour tending towards the crime intended” and, second, “the undesirability of spreading the criminal law so wide as to cover a mental state that the accused might be too irresolute even to begin to translate into action. There can hardly be anyone,” he added, “who has never thought evil. When a desire is inhibited it may find expression in fantasy; but it would be absurd to condemn this natural psychological mechanism as illegal.”[66] So at the root of the appellants’ argument is that favourable disposition is itself a natural psychological mechanism which only results in criminality if the disposition is translated into action proscribed by criminal law, whereas the result of the judge’s ruling and that of the Court of Appeal, if left undisturbed, is the creation of an Orwellian thought crime. 73. The strength of the appellants’ argument is said to be rendered all the more apparent in the light of section 159A of the Crimes Ordinance which provides that conspiracy to commit an offence is established only if there was an agreement to pursue a course of conduct which, if the agreement be carried out in accordance with the intentions of the parties to the agreement, will necessarily amount to the commission of an offence by one or more parties to the agreement.[67] The argument is that, as framed, the prosecution case was unable to identify any course of conduct that would necessarily amount to the offence of misconduct in public office. The adoption of a state of mind is not sensibly categorised as the carrying out of a course of conduct; and the alternative assertion in the particulars of offence that it was agreed that Rafael Hui would “remain” favourably disposed sits ill with the tenor of section 159A which envisages an agreement that a future crime be committed. 74. The appellants contend that what is proposed is an unprecedented and unprincipled expansion of the boundaries of the offence of misconduct in public office. This common law offence has a long history[68] and the appellants say that it is no accident that no example has been discovered of any prosecution mounted on the basis now advanced. 75. The ramifications are said to be chilling, for an expansion of the offence’s parameters to embrace favourable disposition would render susceptible to prosecution those in the private sector who enter public life and carry with them natural feelings of goodwill. Rules for disclosure of potential conflicts of interest and the present law which proscribes external acts of misconduct suffice and just as “it [is] not for this Court to create a new offence as an answer to a perceived problem of imprecise definition or accessibility,”[69] it is not for this Court to relax the limits of the offence so as to cater for a perceived lacuna in the law relating to pre-office payments. 76. The appellants next contend that the suggestion by Leonard J in Chung Fat Ming that he “would regard being or remaining favourably disposed to the person solicited as sufficient to amount to an ‘act’” within the meaning of section 4(2)(a) of the Prevention of Bribery Ordinance, which is the genesis of the phrase used in Count 5 (as well as in Count 7), is of no avail to the prosecution. The point is made that the base criminality or actus reus of the section 4(2)(a) offence is not the suggested favourable disposition to which Leonard J referred but the solicitation or acceptance of an advantage. Under section 4(2)(a), an offence may be committed absent any breach of duty by the public servant: it is enough that he accepts or solicits the advantage merely on account of his doing his job and what Leonard J was saying, so the argument goes, was that if he solicited or accepted an advantage – the actus reus – for being favourably disposed, he could properly be said to have committed the proscribed conduct on account of performing an act in his capacity as a public servant. What has happened, therefore, is that the prosecution in the present case has lifted an analysis not directed at the actus reus of section 4(2)(a) of the Ordinance and turned it into the actus reus of the common law offence with which these appellants were charged. 77. Finally, the appellants say that the remark by Sir Anthony Mason NPJ in Sin Kam Wah & Another v HKSAR (at [64] above) that the “acceptance of a ‘general sweetener’ by a public officer can, in appropriate circumstances, amount to misconduct in public office”[70] is also not to the point since the comment relates to the act of acceptance, which is not the act of misconduct in the present case. D.3 The respondent’s argument 78. The respondent’s position was explained by Mr Perry as follows. As is obvious from the verdicts returned by the jury, the defence offered by the appellants at trial was rejected and the jury concluded that the June 2005 payments were bribes paid to Rafael Hui in relation to the performance by him of his public office and were paid in return for him being or remaining favourably disposed to the interests of SHKP. That was a finding that he was paid to be favourably disposed to a private interest group. It was a finding that he was to be, and was in fact, on the payroll of that private interest group. What therefore was envisaged by the conspirators was a continuing act of partiality in breach of Rafael Hui’s duty to serve the public interest and not to serve his private interests and those of a secret paymaster. The misconduct in this case was a continuing act of disloyalty, since the performance of a public office in golden fetters is to act disloyally and without integrity. It is a continuing breach of the public officer’s obligation to act with integrity in the public interest and is at the heart of what the offence of misconduct in public office is designed to deter. 79. There is no need, he suggested, to identify a particular act or acts contemplated by the conspirators. That is not only because favourable disposition to private interests in return for a bribe is itself misconduct but because the law recognises the practical realities of proof, in that conspirators are unlikely in such cases to spell out a specific act; indeed no specific act may be in contemplation and showing favour may be understood to mean no more than avoiding fault picking or obstruction.[71] 80. Nor are the merits of decisions taken by the public officer thus corrupted of any relevance for he has consciously and deliberately put himself into a position in which his duty to act in the public interest is compromised. That being so, the question whether a particular policy decision or transaction is beneficial to the public is immaterial. 81. It is, says Mr Perry, fallacious in the context of this case to categorise favourable disposition as a mere thought. The error underlying that contention is to be found in placing favourable disposition in isolation, divorced from the payments made to secure it. The favourable disposition thus secured is the description of the misconduct itself, and the mens rea is the wilfulness of that conduct. E. Addressing the issue raised in the appeals E.1 The many forms of misconduct in public office 82. The decided cases show that a broad range of different acts and omissions can constitute the relevant conduct element of the offence. In Shum Kwok Sher v HKSAR, Sir Anthony Mason NPJ observed (at [69]): “The difficulty which has been experienced in defining with precision the elements of the offence stem not so much from the various ways in which they have been expressed as from the range of misconduct by officials which may fall within the reach of the offence. This is because, to quote the words of PD Finn, ‘Public Officers: Some Personal Liabilities’ (1977) 51 ALJ 313 at p.315: The kernel of the offence is that an officer, having been entrusted with powers and duties for the public benefit, has in some way abused them, or has abused his official position. It follows that what constitutes misconduct in a particular case will depend upon the nature of the relevant power or duty of the officer or of the office which is held and the nature of the conduct said to constitute the commission of the offence.” 83. This dictum was adopted by the Supreme Court of Victoria in R v Quach[72] and is reflected in the passage in the judgment of the Supreme Court of Canada in R v Boulanger quoted in [46] above. 84. Thus, the following acts and omissions have been held to constitute the offence: failing to disclose a relationship with a company and showing preferential treatment to that company by permitting it to tender for Government contracts despite lacking the requisite experience;[73] accepting free sexual favours provided by prostitutes controlled by the owner of a nightclub;[74] obtaining and using the personal particulars of patients of a public hospital to advertise the commencement of a private medical practice.[75] But these are merely specific instances of the offence and they are illustrative rather than definitive of the ways in which it can be committed. As Sir Anthony Mason NPJ said in Shum Kwok Sher v HKSAR, the offence “is necessarily cast in general terms because it is designed to cover many forms of misconduct on the part of public officers.”[76] 85. That there must be a relevant relationship between the act or omission constituting the misconduct and the public office was confirmed by this Court’s decisions in Sin Kam Wah & Another v HKSAR[77] and HKSAR v Wong Lin Kay.[78] Nevertheless, as Sir Anthony Mason NPJ noted in Shum Kwok Sher v HKSAR: “… the essence of the offence is that an officer who has been entrusted with powers and duties for the public benefit has abused them or his official position. Abuse of such powers and duties may take various forms, ranging from fraudulent conduct, through nonfeasance of a duty, misfeasance in the performance of a duty or exercise of a power with a dishonest, corrupt or malicious motive, acting in excess of power or authority with a similar motive, to oppression. In all these instances the conduct complained of by the public officer takes place in or in relation to, or under colour of exercising, the office.”[79] E.2 The responsibility of the office and the office-holder 86. That, as noted at [47] above, the essence of the offence is the abuse of public trust by the officer is also amply supported by the authorities cited by Ribeiro PJ in his judgment in HKSAR v Wong Lin Kay[80] at [26] to [33]. And as Chan ACJ put it, in HKSAR v Ho Hung Kwan Michael, “this offence is aimed at punishing an abuse by a public officer of the power and duty entrusted to him for the public benefit or of his official position”.[81] 87. In this area of the law, as elsewhere, context is all important. Because the essence of the offence is the abuse of office it is necessary that the misconduct be “serious, not trivial” and, therefore, when asking whether a public officer has misconducted himself in office, it is essential to put that inquiry into its proper context by identifying “the responsibilities of the office and the officeholder”.[82] 88. In the present case, we are concerned with the office of the Chief Secretary for Administration of the HKSAR Government. The Chief Secretary is a Principal Official under the Basic Law of the Hong Kong Special Administrative Region (“the Basic Law”)[83] and the second most senior officer in the HKSAR Government. In the event the Chief Executive is not able to discharge his or her duties for a short period, such duties shall be temporarily assumed by the Chief Secretary, the Financial Secretary or the Secretary for Justice, in that order of precedence.[84] 89. The Chief Secretary is also an Official Member of ExCo, which assists the Chief Executive in policy-making.[85] As the prosecution pointed out in its Case Summary at trial, ExCo is consulted by the Chief Executive before he or she makes important policy decisions, introduces bills to the Legislative Council, makes subordinate legislation or dissolves the Legislative Council. ExCo normally meets once a week and its proceedings are confidential. It is also the fact (and this can be assumed without evidence) that the Chief Secretary exercises statutory functions vested in him by law, such as those concerning the handling of appeals and certain public bodies. 90. Specifically, in the present case, whilst he was Chief Secretary and later as a Non-Official Member of ExCo, Rafael Hui was involved in important matters of policy concerning the WKCD Project and the Ma Wan Project, two large property developments which concerned also the interests of SHKP. During his time as Chief Secretary, in particular, Rafael Hui was the Chairman of the Steering Committee of the WKCD Project. 91. As a Principal Official, in addition to his oath of office, Rafael Hui was subject to Government Regulations relating to conduct and conflicts of interests. He was also subject, in particular, to the Code for Principal Officials. This Code contains the basic principles governing the performance by Principal Officials of their duties, which include the following: to be dedicated to their duties and be responsible to the Government of the HKSAR; to uphold the law, abide by the law, and protect the integrity of the public office; to observe the highest standards of personal conduct and integrity at all times; to ensure that no actual or potential conflict of interest arises between their public duties and their private interests; to promote and support these principles by leadership and example. In relation to the prevention of conflicts of interests, the Code specifically lays down the principles that Principal Officials should: avoid putting themselves in a position where they might arouse any suspicion of dishonesty, unfairness or conflict of interest; refrain from handling cases with actual or potential conflict of interest; and report to the Chief Executive any private interests that might influence, or appear to influence, their judgment in the performance of their duties. 92. In short, as the prosecution asserted in its case at trial, as Chief Secretary, Rafael Hui would have had an important role in the development of Government policy, and be privy to highly confidential information, in relation to matters in which SHKP had substantial interests. And it is plain that, as Chief Secretary, Rafael Hui was “appointed to be a sentinel of the public welfare”.[86] E.3 The offence alleged in Count 5 93. As already noted, Count 5 charged a conspiracy to commit misconduct in public office. As in any charge of conspiracy, it is necessary to focus on the character of the conspiratorial agreement alleged by the prosecution. It is also necessary, for the statutory offence of conspiracy to be established, that the conspirators’ agreement must be “that a course of conduct shall be pursued which, if the agreement is carried out in accordance with their intentions, either … will necessarily amount to or involve the commission of any offence … by one or more of the parties to the agreement”.[87] 94. The agreement here was made shortly before Rafael Hui assumed his position as a public officer. It was an agreement under which Rafael Hui was to be paid HK$8.5 million in return for his favourable disposition while in office and was clearly therefore a corrupt bargain. Had that payment been made after he assumed the office of Chief Secretary, the prosecution case could have been put simply on the basis of an agreement to offer an advantage to a public officer since the acceptance of the payment would have constituted the substantive offence of misconduct in public office.[88] Similarly, had the conspiratorial agreement been made when Rafael Hui was already Chief Secretary, the prosecution might alternatively have been put simply on the basis that the making of the corrupt agreement itself was a relevant act of serious misconduct sufficient to constitute the offence. However, the timing in this case was such that this could not be alleged. 95. Nevertheless, from Rafael Hui’s perspective, the agreement alleged in Count 5 was concerned with his position in public office and, from the other conspirators’ perspective, their payment of HK$8.5 million to Rafael Hui was referable to his position as Chief Secretary and was made in anticipation of his assuming that office. The purpose of the payment, from their perspective, was to put the incoming Chief Secretary on their payroll and thus inclined to show them favour or prefer their interests. One might ask, rhetorically, if Rafael Hui had not taken the oath of office and become Chief Secretary, would not the other conspirators have sought the repayment of the HK$8.5 million from Rafael Hui? As Mr Perry put it in his closing speech to the jury, those who were paying Rafael Hui that sum were not “running a charity”.[89] 96. As such, the agreement alleged in the indictment, although entered into before Rafael Hui assumed public office, was made in anticipation of his entering into public office as Chief Secretary, and was wholly concerned with his anticipated holding of that public office. It was an agreement that, in the language of section 159A of the Crimes Ordinance, involved the commission by him of misconduct in public office because, by making the agreement, Rafael Hui would (to use the language of Knox CJ in R v Boston, see [48] above) “[impair] his capacity to exercise a disinterested judgment” or (to use the language of Isaacs and Rich JJ in the same case) “[place] himself in a situation embarrassing and inconsistent with that independence to criticize or censure which he [was] bound to preserve” (or, more colourfully, would “[fasten] upon himself golden fetters which preclude[d] his freedom of action”). Once he had accepted the sum of HK$8.5 million in relation to his holding of public office, Rafael Hui’s independence when he assumed office would be hopelessly compromised and he could not properly discharge the duties of Chief Secretary nor be trusted to do so. This involved a continuing offence that commenced from the time he entered into his public office as Chief Secretary and infected the entire period he held that office under the influence of the payment made to him. 97. It is correct, of course, that Rafael Hui might have been charged with the substantive offence of misconduct in public office for failing to disclose his receipt of the payment when acting as Chief Secretary, but, as Mr Perry rightly observed, such a charge would not properly reflect the criminality involved in this conspiracy. E.4 Sufficient allegation of misconduct in public office in Count 5 98. So analysed, it will be apparent that the conspiracy alleged was an agreement under which, in return for a substantial payment of money, Rafael Hui would incline himself in a manner inconsistent with his duty as a public officer. That inclination was improper since it was wholly inimical to his duties as Chief Secretary, described above, and involved a serious abuse of office and abuse of public trust. Having accepted the payment of HK$8.5 million in return for his improper inclination, Rafael Hui could not properly discharge the duties of Chief Secretary. 99. The abuse of public trust contemplated by the conspirators in the present case is therefore clear and, by agreeing to place himself in such a compromised state, Rafael Hui made an agreement which contemplated a continuing act of misconduct whilst he was Chief Secretary. The fact that the payment made to induce that compromised state was made before he assumed his public office does not mean that the abuse of trust occurred at the time of the making of the payment. That payment was made to secure an ongoing inclination on the part of Rafael Hui towards SHKP once he assumed the office of Chief Secretary and it is by so agreeing to act as Chief Secretary whilst he was in the “golden fetters” constituted by that payment that he conspired to commit an act of misconduct sufficient to satisfy the conduct element of the offence of misconduct in public office. 100. The above conclusion on the issue raised in these appeals is, as the prosecution contended, consistent with the core of the common law offence of misconduct in public office and the mischief it is designed to address. If the expression “being … favourably disposed”, the risks of which are alluded to in paragraph [63] above, is used in the context of a case of bribery (see Section C above) it is an adequate description of an act of misconduct. F. Suggested misdirections and non-directions 101. Given our conclusion upon the central issue of law, that the agreement alleged by the particulars of offence in Count 5 was an agreement to commit misconduct in public office, the argument advanced to this Court that the judge erred in not directing the jury that it was necessary to prove that specific acts of favour were contemplated by the appellants or that at least it was agreed that favour would be shown if and when that should become necessary, is an argument which necessarily fails. 102. A number of other complaints were advanced. These included, first, the contention that the manner in which the respondent has sought before this Court to categorise the misconduct alleged by the particulars is neither the way it was put to the jury by the prosecution at trial nor the sense in which it was explained to the jury by the judge. Secondly, it was said that the directions created a danger of conviction on Count 5 from the mere fact of such favourable disposition towards the Kwoks and SHKP as naturally resulted from their long relationship. Thirdly, that the judge erroneously left the jury with the impression that it was the acceptance of the payments that was the relevant alleged misconduct. Fourthly, it was suggested that it was wrong at trial to refer to the June 2005 payments as alleged bribes since Count 5 did not charge either common law or statutory bribery; and, further, that the judge erred in failing to direct the jury that evidence of conduct by Rafael Hui which enured to the disadvantage of SHKP was relevant in deciding whether the appellants had entered into the impugned agreement. 103. Some of the complaints strayed somewhat from the ground on which leave was given but, nonetheless, we are satisfied that none of the complaints is made out. We have already set out the material part of the judge’s directions at [39] above. As to the criticism of the description of the payments as bribes, that description was used by counsel on all sides as a convenient way by which to highlight the core issue, namely whether the payments were for a corrupt or for an innocent purpose. G. Conclusion and disposition of appeals 104. For the above reasons, the appeals are dismissed. The appellants were each properly convicted of the offence of conspiracy to commit misconduct in public office charged in Count 5 of the indictment. Mr Edwin Choy and Mr Joe Chan, instructed by Tang, Lai & Leung, for the 1st Defendant/Appellant in FACC 12/2016 (1st Appellant) Ms Clare Montgomery QC, Mr Gary Plowman SC and Mr Benson Tsoi, instructed by Davis Polk & Wardwell, for the 2nd Defendant/Appellant in FACC 14/2016 (2nd Appellant) Mr Ian Winter QC, Mr Selwyn Yu SC and Mr Isaac Chan, instructed by Morley Chow Seto, for the 4th Defendant/Appellant in FACC 15/2016 (3rd Appellant) Mr Hugo Keith QC, Mr Charles J. Chan and Mr Billy Kwan, instructed by Simon Ho & Co., for the 5th Defendant/Appellant in FACC 13/2016 (4th Appellant) Mr David Perry QC, Mr Joseph Tse SC and Ms Maggie Wong, on fiat, for the Department of Justice, for the Respondent [1] Misconduct in public office is a common law offence. [2] FAMC 8, 9, 10 & 11/2016, Determination dated 12 July 2016, at [2]. [3] CACC 444/2014, Judgment dated 16 February 2016, at [59] to [83]. [4] Flats 20A and 20B, Tower 6, Leighton Hill. [5] Addressed in more detail at [91] below. [6] In HCCC 98/2013; Macrae JA tried the case sitting as an additional judge of the Court of First Instance. [7] Contrary to common law and s.101I(1) of the Criminal Procedure Ordinance (Cap.221). [8] The particulars of offence for Counts 2, 3, 5 and 7 alleged favourable disposition to SHKP and/or its subsidiaries or associated companies and various named defendants. In this judgment, we shall simply refer to the allegation of favourable disposition to SHKP in relation to those various counts. [9] Contrary to common law and ss.159A and 159C of the Crimes Ordinance (Cap.200) and s.101I(1) of the Criminal Procedure Ordinance (Cap.221). [10] Contrary to common law and ss.159A and 159C of the Crimes Ordinance (Cap.200) and s.101I(1) of the Criminal Procedure Ordinance (Cap.221). [11] Contrary to s.19(1)(b) of the Theft Ordinance (Cap.210). [12] Contrary to common law and ss.159A and 159C of the Crimes Ordinance (Cap.200) and s.101I(1) of the Criminal Procedure Ordinance (Cap.221). [13] Contrary to common law and s.101I(1) of the Criminal Procedure Ordinance (Cap.221). [14] Contrary to ss.4(1)(a) and 12 of the Prevention of Bribery Ordinance (Cap.201) and ss.159A and 159C of the Crimes Ordinance (Cap.200). [15] Contrary to common law and s.101I(1) of the Criminal Procedure Ordinance (Cap.221). [16] HCCC 98/2013, transcript of Ruling of Macrae JA on Friday, 12 September 2014, at pp.7-8. [17] Ibid. at pp.10-11. [18] Mr Donald Tsang Yam-kuen. [19] (2005) 8 HKCFAR 192 at [45]. [20] HCCC 98/2013, Summing-up on 8 December 2014, at pp.46-47. [21] Pursuant to s.12(1) of the Prevention of Bribery Ordinance (Cap.201). [22] The disqualification orders against Thomas Kwok and Thomas Chan were made pursuant to s.168D of the Companies Ordinance (Cap.32). [23] CACC 444/2014 (Yeung VP, Lunn VP and Pang JA), Judgment dated 16 February 2016. [24] Ibid. at [10]. [25] Ibid. at [157]. [26] Ibid. at [12]-[44] per Yeung VP and [173]-[226] per Lunn VP. [27] (2005) 8 HKCFAR 192. [28] Ibid. at [45]. [29] (2002) 5 HKCFAR 381. [30] Re A-G’s Reference (No 3 of 2003) [2004] 2 Cr App R 23, 366. [31] (2012) 15 HKCFAR 185. [32] For present purposes it is not materially different from the law as recently stated in cases in England and Wales (R v Chapman & Ors [2015] 2 Cr App R 10, 161) and in Australia (R v Quach [2010] VSCA 106; (2010) 201 A Crim R 522; Obeid v R [2015] NSWCCA 309). [33] [2006] 2 SCR 49. [34] [2006] 2 SCR 49 at 55. [35] [2006] 2 SCR 49 at 69. [36] (1978) 2 Crim LJ 307 at 314 fn 71. [37] (2005) 8 HKCFAR 192. [38] [2010] VSCA 106. [39] [2010] VSCA 106 at [37]. [40] [2010] VSCA 106 at [40]. [41] (1923) 33 CLR 386. [42] (1923) 33 CLR 386 at 392. [43] (1923) 33 CLR 386 at 393. [44] (1923) 33 CLR 386 at 404. [45] 12th Ed. p.381. [46] Russell on Crime, 12th Ed., p.381 fn 1. [47] [2014] 1 WLR 1857. [48] [2016] QB 763. [49] (1901) 3 F (Ct of Session) 59. [50] (Cap.215). [51] Hong Kong Hansard, Prevention of Bribery Bill 1970 (First Reading, 21 October 1970). [52] See, in a related context, Secretary for Justice v Chan Chi Wan Stephen & Another, FACC 11 & 18/2016, unrep., 14 March 2017 at [56] to [58] and [140] to [146]. [53] [1978] HKLR 480. [54] Ibid. at 482. [55] Ibid. at 482. [56] Ibid. at 486. [57] [1977] HKLR 186. [58] [1978] HKLR 480 at 495. [59] Ibid. at 497. [60] (1992) 27 NSWLR 398. [61] Ibid. at 402. [62] (2005) 8 HKCFAR 192 at [54]. [63] Trial transcript, Day 105, p.9, lines 12-19. [64] In India, the Prevention of Corruption Act 1988, section 7 renders it an offence for “whoever, being, or expecting to be a public servant” accepts pecuniary gratification, other than legal remuneration, for doing or forbearing to do any official act. In the United States of America, it is an offence corruptly to give anything of value to any public official or “person who has been selected to be a public official” and for such a person corruptly to demand or receive anything of value, where the intent is to influence the performance of an official act: 18 U.S. Code, section 201. [65] Scofield (1784) Cald 397 at 402. [66] Criminal Law – The General Part (2nd Ed.), p.2. [67] The terms of s.159A are set out at [50] above. [68] Sir Anthony Mason NPJ in Shum Kwok Sher stated at [66] that it went back at least to 1704; whereas in a recent article, “Revival of the common law offence of misconduct in public office”, David Lusty (2014) 38 Crim LJ 337, the suggestion is made that it can be traced back to the 13th century. [69] Shum Kwok Sher v HKSAR (2002) 5 HKCFAR 381 at [98]. [70] (2005) 8 HKCFAR 192 at [54]. [71] HKSAR v Cheung Koon Chee, CACC 356/2010, unrep. at [40], a decision of the Court of Appeal in relation to s.9 of the Prevention of Bribery Ordinance. [72] [2010] VSCA 106; (2010) 201 A Crim R 522; per Redlich JA at [15]. [73] Shum Kwok Sher v HKSAR (2002) 5 HKCFAR 381. [74] Sin Kam Wah & Another v HKSAR (2005) 8 HKCFAR 192. [75] Chan Tak Ming v HKSAR (2010) 13 HKCFAR 745. [76] (2002) 5 HKCFAR 381 at [91]. [77] (2005) 8 HKCFAR 192 at [47]. [78] (2012) 15 HKCFAR 185 at [17]. [79] (2002) 5 HKCFAR 381 at [81]. [80] (2012) 15 HKCFAR 185: viz. R v Bremridge (1783) 3 Doug KB 327, 99 ER 679; R v Dytham [1979] QB 722; R v Whitaker [1914] 3 KB 1283; Re A-G’s Reference (No 3 of 2003) [2005] QB 73; Russell on Crime (12th ed., 1964) p.361 para.32; Three Rivers District Council v Governor and Company of the Bank of England (No 3) [2003] 2 AC 1, 235; Henly v Mayor and Burgesses of Lyme (1828) 5 Bing NC 91 at 107, 130 ER 995 at 1001; Northern Territory of Australia v Mengel (1995) 185 CLR 307. [81] (2013) 16 HKCFAR 525 at [26]. [82] Element (5) of Sir Anthony Mason NPJ’s re-formulation of the offence set out at [45] above. [83] As such he would be required to take the Oath of the Principal Officials under the Oaths and Declarations Ordinance (Cap.11), s.16B and Sched.2, Part II. [84] Basic Law, Article 53(1). [85] Basic Law, Article 54. As a member of ExCo, he would be required to take the Oath of Fidelity and the Executive Council Oath: Oaths and Declarations Ordinance (Cap.11), s.18 and Sched.2, Parts III and VI. [86] R v Boston (1923) 33 CLR 386 per Isaacs and Rich JJ at p.403. [87] Crimes Ordinance (Cap.200) s.159A. [88] The prosecution could also have charged the offence under s.4 of the Prevention of Bribery Ordinance: see [63] above. [89] Trial transcript, Day 113, p.107, lines 11-12. 
Mr Justice Ribeiro PJ: 1. These appeals, which arise out of the same transaction and which have been heard together, raise questions concerning the elements of the offence created by section 9 of the Prevention of Bribery Ordinance[1] (“POBO”). A. The relevant events 2. The appellant Stephen Chan Chi Wan (“Chan”) was at the material time employed as General Manager (Broadcasting) by Television Broadcasts Ltd (“TVB”) under a contract designed for administrative personnel. He was not required to appear in front of the cameras but had voluntarily hosted a talk show screened on a TVB pay channel where he interviewed various celebrities, receiving no additional remuneration for so doing. 3. The show has been referred to as the “Be My Guest” show but its Chinese name “志雲飯局” translates literally as “Chi Wan’s Dinner Party”, “Chi Wan” being Chan’s given name. It gained considerable popularity, more than 150 episodes having been aired, so that Chan himself acquired a certain celebrity status. This led to his being engaged by third parties to appear at various external events, the evidence being that between June 2008 and January 2010, this occurred 20 odd times, Chan receiving remuneration on 18 of those occasions. 4. The appellant Tseng Pei Kun (“Tseng”) was the sole director and shareholder of Idea Empire Advertising & Production Company Limited (“IEAP”). He and IEAP acted as Chan’s agent in respect of his outside commercial engagements. 5. Olympian City is a shopping mall operated by Olympian City 2 Management Co Ltd (“OC”). For several years, OC had collaborated with TVB, sponsoring a New Year’s Eve Countdown show produced by TVB and broadcast from Olympian City as a means of promoting the shopping complex. In November 2009, OC and TVB reached agreement for the Countdown show to be presented on 31 December 2009. OC agreed to pay $1.3 million as the sponsorship fee without having discussed any of the programme’s contents. 6. Coincidentally, the parties came up with the idea of adding, as a segment ancillary to the main Countdown presentation, a live, staged edition of the “Be My Guest” show. It was accepted that the $1.3 million fee would not cover the cost of this additional segment and there was no suggestion that TVB would be responsible for arranging for Chan’s participation. The evidence was that OC knew from previous experience that TVB would not provide what were referred to as their “top-tier artistes” for the Countdown event, so OC would independently invite any personalities they wished to have appearing on the show. 7. OC accordingly contacted Tseng, knowing that IEAP acted as Chan’s agent, and entered into an agreement with IEAP for Chan to appear on a live “Be My Guest” segment in the Countdown show, interviewing as his guest Lai Yiu Cheung (“Lai”). It was agreed that they would wear stickers promoting Olympian City, a requirement that did not apply to the artistes supplied by TVB. 8. OC agreed to pay IEAP $160,000 and in turn, IEAP agreed to pay $112,000 to Chan and $20,000 to Lai, keeping $28,000 for itself. Chan did not seek or obtain permission from TVB to appear on or to be paid for hosting the show although he had a contractual obligation to seek approval for outside work. However, since TVB produced the show, it was generally known by persons at TVB that Chan had been engaged to take part in the Countdown event. It was, as its Chinese name indicates, Chan’s show obviously to be aired with his participation. The special edition of “Be My Guest” was duly broadcast and IEAP, Chan and Lai duly received the agreed remuneration. B. The charges and the course of the proceedings 9. The appellants were charged with conspiring for an agent to accept an advantage contrary to section 9(1)(a) (Charge 1) and, alternatively, with the substantive offences of an agent accepting an advantage (Charge 2, in Chan’s case) and offering an advantage to an agent (Charge 3, in Tseng’s case). 10. The charge against Chan as particularised was that he, being an employee and thus an agent of TVB, without lawful authority or reasonable excuse, accepted from Tseng acting on behalf of IEAP, an advantage consisting of the $112,000 fee as an inducement or reward for or otherwise on account of Chan’s doing or having done an act in relation to his principal’s affairs or business “namely participating and performing in a side-show entitled ‘Be My Guest’ in the ‘New Year Eve Countdown Programme at Olympian City for 2010’ which was produced and broadcast by TVB” on 31 December 2009. 11. Tseng was charged with offering that amount to Chan as an inducement, etc, for Chan so to act in relation to the affairs or business of his principal TVB. The alternative charge that they both faced was that they had conspired for Chan to accept that advantage as such an inducement, etc, in relation to his principal’s affairs or business. 12. On 2 September 2011, His Honour Judge S T Poon, AgCDJ (as he then was) acquitted the appellants.[2] The prosecution appealed by way of case stated[3] and the Court of Appeal[4] allowed the appeal, set aside the not guilty verdicts and remitted the case to the trial judge to consider whether the appellants could rely on the defence of reasonable excuse for Chan’s conduct. 13. On 7 March 2013,[5] S T Poon CDJ ruled that the reasonable excuse defence availed both appellants and dismissed the charges afresh. However, the prosecution once again appealed by way of case stated[6] and, once again, the Court of Appeal[7] allowed the appeal and directed the Judge to convict Chan and Tseng on the conspiracy charge and to sentence them accordingly. Chan was fined $84,000 and Tseng fined $28,000. 14. The Court of Appeal[8] refused a certificate sought by the appellants for the purposes of appeal, but the Appeal Committee granted each appellant leave to appeal,[9] certifying the following questions as being of sufficiently great and general importance, namely: (1) In the context of a section 9 offence under the Prevention of Bribery Ordinance (Cap 201) (“POBO”): (i) what is the mens rea required of an agent to constitute him guilty of an offence under that section; and (ii) how should the element “in relation to the principal’s affairs or business” be interpreted having regard to the determination of the Privy Council in Commissioner of the ICAC v Ch’ng Poh [1997] HKLRD 652? (“Questions 1(i) and 1(ii)”) (2) In the context of a section 9 offence under POBO, what is the proper approach the Court should adopt in considering the defence of reasonable excuse? (“Question 2”) 15. Leave was also granted on the ground that it was reasonably arguable that a substantial and grave injustice had been done to the appellants in the present case. C. Question 1(i): Elements of section 9 and mens rea 16. POBO section 9 is concerned with transactions between third persons and agents in relation to the affairs or business of their principals. Its offence-creating provisions materially state as follows: Corrupt transactions with agents (1) Any agent who, without lawful authority or reasonable excuse, solicits or accepts any advantage as an inducement to or reward for or otherwise on account of his – (a) doing or forbearing to do, or having done or forborne to do, any act in relation to his principal's affairs or business; or (b) showing or forbearing to show, or having shown or forborne to show, favour or disfavour to any person in relation to his principal’s affairs or business, shall be guilty of an offence. (2) Any person who, without lawful authority or reasonable excuse, offers any advantage to any agent as an inducement to or reward for or otherwise on account of the agent's – (a) doing or forbearing to do, or having done or forborne to do, any act in relation to his principal's affairs or business; or (b) showing or forbearing to show, or having shown or forborne to show, favour or disfavour to any person in relation to his principal’s affairs or business, shall be guilty of an offence. 17. Section 9(4) excludes from liability an agent who obtains the informed permission of his principal before or as soon as reasonably possible after the conduct which would otherwise constitute a section 9 offence. As permission was neither sought nor given in the present case, section 9(4) is of no application. 18. Addressing Question 1(i), there is no doubt that section 9 is an offence which requires proof of mens rea in respect of all its essential actus reus ingredients. It should be borne in mind that there are several variants of the offence. Thus, section 9(1) covers agents who solicit or accept an advantage while section 9(2) deals with persons who offer an advantage to the agent. The advantage may be an inducement to do or forbear from doing some act in the future. Or it may be a reward for having already done or forborne to do something. The act or forbearance must in each case be in relation to the affairs or business of the agent’s principal, a phrase which is examined in detail below.[10] The mens rea requirement of any particular variant is necessarily adapted to the elements of that variant and may differ from the mens rea requirement of some other variant of the offence. 19. The importance of the mens rea or mental dimension of the offence looms larger in cases involving the offering, soliciting or accepting of an advantage as an inducement for a future act or forbearance. This is because the offence is committed upon the making of the offer, solicitation or acceptance, it being no defence that the contemplated act or forbearance was not or could not be performed by the agent.[11] In such cases, it is the state of mind of the person making the offer to the agent or of the agent soliciting the advantage, or where the agent accepts the advantage, the mental state of the agent in accepting the advantage, that is the major component of the offence. Where the induced or offered act or forbearance has taken place or where the case involves a reward for past conduct, that conduct is available to be examined as part of the actus reus with the accused’s accompanying mental state. 20. It has been stressed that the liability of each person involved in the transaction must be independently assessed according to his own mens rea. Thus it has been held that an agent accepting an advantage with the requisite corrupt intent may be convicted even though the person furnishing the advantage acted innocently. Stuart-Moore VP put it thus in HKSAR v Wong Yuk Sim:[12] “...where a provider of a reward thinks he is providing something innocently but the receiver believes that the provider is providing it as a corrupt reward and he accepts it as a corrupt reward, believing it to have been provided as a corrupt reward, then the receiver behaves corruptly even if the provider behaves innocently.” 21. It follows that in cases involving the offering or solicitation of an advantage, the prosecution must prove that the accused intended that, if provided, the advantage should be accepted as an inducement or reward for or otherwise on account of the agent’s act or forbearance in relation to his principal's affairs or business (in the sense discussed below[13]). 22. And in cases involving the accused agent’s acceptance of an advantage, the prosecution must prove that he knew or believed it to have been provided as an inducement or reward for or otherwise on account of his act or forbearance in relation to his principal's affairs or business (in the aforesaid sense). D. The Judge’s reasoning 23. The Judge referred to Commissioner of ICAC v Ch’ng Poh,[14] (to which I shall return[15]) and recognized that the prosecution did not have to establish that Chan was acting in his capacity as TVB’s agent either when accepting the money or when appearing on the show. His Honour noted, relying on Ch’ng Poh, that: “What the prosecution needs to prove is only that the relevant act is an act done in relation to his principal’s affairs or business. However, this phrase should not be given a limitlessly wide meaning. Its meaning should be kept within a specific parameter, which is: when the defendant did the act in question, he must have intended the act to ‘influence or affect’ his principal’s affairs or business. In other words, his act was ‘aimed at’ his principal.”[16] 24. He concluded that Chan’s appearance in the “Be My Guest” segment “was purely an act of moonlighting in his capacity as a ‘celebrity’ or an ‘artiste’, and such an act was not intended to influence or affect the affairs or business of TVB, nor was it aimed at TVB”[17] so that it fell outside the ambit of section 9. 25. In his 2nd Case Stated,[18] the Judge elaborated on his views, pointing out that Chan had not used any powers of his office to benefit himself and had acted openly, his participation having been discussed at TVB production meetings, so that there was no question of him making a secret profit. He held that Chan’s participation in the show “would do no harm but good to TVB” and that “[his] performance and acceptance of the reward did not conflict with any interest of TVB.” He commented that Chan’s mistake merely involved his failure to “follow the procedures prescribed in the employment contract” regarding approvals of outside work and remuneration. 26. The Judge therefore concluded that Chan’s act of accepting $112,000 as an inducement or reward for appearing on the show was not an act done “in relation to his principal's affairs or business” within the meaning of section 9 or (in his 2nd Case Stated) was amenable to the defence of reasonable excuse. He therefore acquitted the appellants. E. The Court of Appeal’s reasoning 27. The Court of Appeal reversed the Judge’s verdict of acquittal holding that he had been wrong to find that Chan’s act fell outside section 9. In the joint judgment of Yeung VP and Yuen JA, it was held that Chan’s appearance in the show “related to TVB’s affairs or business”; would “influence or affect TVB”; was “aimed at TVB” and would “involve TVB”.[19] The reasoning behind those conclusions is contained in the following paragraphs of that judgment (translated from the original Chinese): “107. Although Chan’s chief position at TVB was General Manager (Broadcasting) and he was not contractually obliged to perform in front of the camera, the undisputed evidence shows that he had performed in more than 150 episodes of ‘Be My Guest’, which was a well-known and very favourably received talk show on TVB’s pay channel. 108. It must have been because of the popularity of ‘Be My Guest’ that Olympian City would like to have the additional ‘Be My Guest’ show added within the main show. 109. This request made by Olympian City met with approval from TVB. TVB allowed Chan and another artiste Lai Yiu-cheung to perform together in the additional ‘Be My Guest’ show, which was to be produced and broadcast live by TVB. 110. Had Chan refused to perform in the additional ‘Be My Guest’ show, it might have been necessary for TVB to arrange another programme as a substitute. This might affect the contractual relationship between TVB and Olympian City and would significantly influence the audience ratings of the live broadcast of the main show. 111. Judging from the undisputed and/or indisputable background facts of the present case, the propositions that Chan performed in the additional ‘Be My Guest’ show merely in the capacity of a celebrity, that his performing in that show was not related to TVB’s affairs or business, and that it would not influence or affect TVB, would not involve TVB and was not aimed at TVB, are all artificial and contrived. These propositions not only deviate from the crux of the issues in the present case and defy common sense, but also represent an approach which is neither realistic nor reasonable.” 28. These are, with respect, decidedly odd reasons for labelling Chan’s conduct as criminal. What is asserted in those paragraphs is that TVB wanted Chan’s show to be part of its Countdown event because its popularity would boost audience ratings. Chan in fact obliged without being asked to do so by TVB and at no cost to the television station, having been engaged and paid to appear on the show by OC via IEAP. Thus, on the Court of Appeal’s reasoning, Chan’s conduct was wholly in line with and beneficial to TVB’s interests. 29. However, the joint judgment holds that Chan’s conduct constituted the section 9 offence by saying that it “related to”, would “influence or affect”, was “aimed at” or would “involve” TVB’s business or affairs in that TVB’s business would have been prejudicially affected if Chan had refused to perform (as he was entitled to under his contract) because audience ratings would have been adversely affected, alternative programming arrangements might have been necessary and TVB’s contractual relationship with OC might have been affected. This is to assert that Chan committed the offence because TVB would have been prejudiced if Chan did not act in the beneficial way in which he was induced to act by remuneration from OC. Such reasoning appears to turn the policy of section 9 on its head. 30. The joint judgment did not regard this as a difficulty, taking the view that lack of prejudice to the principal’s affairs or business was irrelevant: “What section 9(1)(a) requires is that the act of the agent is done in relation to his principal’s affairs or business. There is no requirement that the agent’s act has to cause prejudice to his principal’s affairs or business. Ch’ng Poh further explained that the phrase ‘in relation to his principal’s affairs or business’ meant that the agent’s act would influence or affect his principal’s affairs, would involve his principal, and was aimed at his principal. Ch’ng Poh did not say that the ‘influence’ or ‘effect’ must be adverse, and the parties agreed that the term ‘aimed at’ does not carry a negative meaning.[20] In support of this view, reliance was placed on the Court of Appeal’s earlier decision in HKSAR v Fung Hok Cheung,[21] to which I shall return.[22] 31. Cheung JA, in a concurring judgment, gave different reasons for holding that Chan’s performance was an act in relation to his principal's affairs within the meaning of section 9. His Lordship stated: “... Judge Poon overlooked the fact that besides being a celebrity or an artiste, the 1st respondent was also General Manager (Broadcasting) of TVB, and therefore his acts were certainly related to TVB’s business. This is because, on the facts, when TVB and Olympian City Management Ltd were planning the countdown event, they happened to put forward the same idea, to which they then agreed, of adding in a shopping mall version of ‘Be My Guest’ as part of the event. It follows that, if the 1st respondent participated in that show and accepted money therefor, he would be doing an act in relation to TVB’s business.”[23] 32. With respect, the proposition that Chan was General Manager (Broadcasting) and “therefore” his acts “were certainly related to TVB’s business” is hard to follow. Chan’s appearance on the talk show had nothing to do with his being General Manager (Broadcasting). Whether his hosting of the show did or did not qualify as an act “in relation to his principal's affairs or business” required examination of the relationship between that act and the principal’s affairs or business. It was not determined simply by pointing to his managerial position in TVB. 33. Cheung JA added (in the context of his discussion of mens rea): “Judge Poon considered that when the 1st respondent agreed to perform in the ‘Be My Guest’ show, he had no intention to influence or affect TVB’s business and his act was not intended to be aimed at TVB, because the idea of adding in the show was not conceived by either of the two respondents, who merely played a passive role. However, as stated above, the most important consideration is that the 1st respondent’s act was related to the TVB programmes, and this precisely indicates a causal relationship between his act and his principal’s business, on which basis the court is entitled to find that the 1st respondent had the mens rea for the section 9(1)(a) offence.”[24] (Italics supplied) 34. To regard the required relationship between the agent’s act and the principal’s affairs or business as established simply because Chan’s show “related to TVB programmes”, gives the section an extremely broad interpretation. It was an approach supported by Mr Jonathan Caplan QC[25] who submitted that it was enough that Chan’s show was part of a programme broadcast by TVB. 35. For the reasons developed below, I am respectfully of the view that the Court of Appeal’s judgments are erroneous and based on a misconstruction of section 9. F. Ch’ng Poh 36. The leading authority in this context is Commissioner of the ICAC v Ch’ng Poh[26] in which the Privy Council considered the nature and limits of the offence created by section 9. Mr A, a partner in a solicitors’ firm referred to as X & Co, was acting for Ch’ng Poh who had been convicted of fraud. To bolster his prospects of appeal, Ch’ng Poh caused A to offer a bribe to a corrupt former prosecutor, Warwick Reid, who was asked to swear an affidavit discrediting an accomplice and key prosecution witness, one C H Low, and to try to persuade Low not to cooperate with the prosecuting authorities. A warrant authorizing a search of X & Co’s premises which was purportedly issued by the magistrate under POBO section 9(1)(a) was quashed for want of jurisdiction. The Privy Council held that on the facts, A’s acts had nothing to do with the affairs or business of his principal X & Co, so that section 9 was not engaged. 37. In reaching that conclusion, Lord Lloyd of Berwick held that the words “in relation to his principal's affairs or business” had a restricting purpose and that it was not enough that the recipient of a bribe should merely be shown to be someone’s agent: “Section 9 is the only section in the Ordinance dealing with persons other than public servants. It is confined to agents. It does not say, like section 4, that the agent must have been acting in his capacity as an agent. Instead the act done (or not done) by the agent must be an act done or not done ‘in relation to his principal's affairs’. But as Keith J pointed out in R v Ng Man Ho [1993] 1 HKC 632 at p.638 the alternative words serve much the same purpose. They are clearly intended to be restrictive. It is not enough that the recipient of the bribe should be an agent in fact. Otherwise any partner in a firm of solicitors, accepting an advantage without authority or reasonable excuse, would be caught by the section. This would be much too wide.”[27] 38. Turning to the meaning of the restricting phrase, his Lordship stated: “So what do the limiting words mean? They mean that, for the section to apply, the person offering the bribe, must have intended the act or forbearance of the agent to influence or affect the principal's affairs. Accordingly s.9 would apply if Ch’ng Poh had bribed Mr A to secure him a benefit at X & Co's expense, for example, to arrange a reduction in X & Co's ordinary professional fees; or if X & Co were induced to act in a way in which they would not otherwise have acted. Thus it is an essential ingredient of the offence under s.9 that the action or forbearance of the agent should be aimed at the principal. If it is sufficient for the purposes of the person offering the bribe that the agent should act on his own without involving his principal, then, whatever other offence may have been committed, it is not a corrupt transaction with an agent for the purposes of s.9.”[28] 39. In giving this meaning to the restrictive words, Lord Lloyd was elaborating upon both the actus reus and mens rea elements of the offence. In variants of the offence where the agent has already done the act or forborne to act, his Lordship’s dictum explains the quality of the relationship between the performed act or forbearance on the one hand and the principal’s affairs or business on the other which must be proved: the act or forbearance must have been aimed at and intended to influence or affect the latter’s affairs or business. In cases where the act or forbearance have yet to occur, Lord Lloyd identifies the mental element required: the advantage offered or solicited must be intended as an inducement or reward for an act or forbearance aimed at the principal’s business, with a view to influencing or affecting the same. 40. Reliance on section 9(1)(a) by the Commissioner in the Ch’ng Poh case was plainly misconceived. There was no advantage offered to or solicited or accepted by the only relevant agent, Mr A. And although he was a partner in X & Co and was in general an agent of the firm, his act of conveying a bribe to Reid was not related to X & Co’s affairs or business. The importance of Lord Lloyd’s judgment is in its explication of the required nexus between act and business which was plainly missing in Ch’ng Poh. He construed the requirement that the agent’s act or forbearance should be “in relation to his principal's affairs or business” as a requirement that the act or forbearance be “aimed at” and intended to “influence or affect” the principal's affairs. G. Question 1(ii): The Court of Appeal and Ch’ng Poh 41. Although the Court of Appeal referred extensively to Ch’ng Poh, I do not, with respect, think that that authority was correctly analysed and applied. To take Cheung JA’s judgment first, his Lordship, as we have seen, adopted an extremely wide interpretation of the relevant phrase, holding that it was enough simply to prove that Chan’s show “related to TVB programmes” in some undefined way. This is inconsistent with Ch’ng Poh in that it fails to assign to that phrase the restrictive function referred to above. 42. It appears that Cheung JA may have adopted his wide interpretation because he was not persuaded that Lord Lloyd had meant the words to operate restrictively. Cheung JA stated: “The words ‘influence’, ‘affect’ and ‘aimed at’ do not appear in section 9. The Privy Council used these words merely because this would facilitate explanation of the causal relationship between the act of the offending agent and his principal’s business. The Court used these words for the purpose of emphasizing the need for the prosecution to prove such a causal relationship in a given case, but this does not mean that, apart from being related to the principal’s business, the agent’s act must also have been deliberately intended to influence, affect or be aimed at the principal’s business. Put another way, these words do not connote a positive or negative impact of the agent’s act on the principal’s business or any good or bad consequences of such act upon the principal’s business. I think that this is the principle which the courts should apply when considering the agent’s intention.”[29] 43. Two main points are made in that paragraph, the important proposition being that the words “influence”, “affect” and “aimed at” do not appear in section 9. That is certainly true and it is correct that in the relevant phrase, words with a general, non-specific meaning – “in relation to his principal's affairs or business” – are used. However, I am unable to agree with Cheung JA’s statement that “this does not mean that, apart from being related to the principal’s business, the agent’s act must also have been deliberately intended to influence, affect or be aimed at the principal’s business” in so far as his Lordship may have intended to question the correctness of Lord Lloyd’s construction. 44. Lord Lloyd elaborated upon the meaning of those words by reference to the mischief of the legislation (which is further discussed below[30]) as indicated by the examples he gave. In so doing, he was reflecting the Court’s duty to give effect to the legislative intention when construing general words, as explained by Lord Millett NPJ in Ho Choi Wan v Hong Kong Housing Authority:[31] “In construing the language of a statute, it is the task of the court to ascertain and give effect to the intention of the legislature. But that does not mean that the Court must give a literal construction to every word or phrase in the statute. As Lord Bingham of Cornhill said in R (Quintavalle) v. Secretary of State for Health[2003] 2 AC 687 at p.695: ‘The court’s task, within the permissible bounds of interpretation, is to give effect to Parliament’s purpose.’ Whenever the legislature enacts or amends an Ordinance, its purpose is to remedy a perceived mischief or defect in the pre-existing legislation. It is to be presumed that it did not intend the statute to go wider in its operation than is necessary to remedy the mischief or defect in question. If it has inadvertently employed general words which, if given their fullest effect, are wider than necessary, the court not only may but must restrict them by construing them in a narrower sense which, while still falling within the ordinary meaning of the words, gives effect to the legislative intent but does not go beyond it, still less frustrate it.” 45. I am less clear about the meaning of Cheung JA’s second point regarding a “causal relationship.” The suggestion that Lord Lloyd was “emphasizing the need for the prosecution to prove [some] causal relationship” is puzzling especially since it is juxtaposed with the statement that the words in question “do not connote a positive or negative impact of the agent’s act on the principal’s business”. If this is a reference to the rule laid down by POBO section 11 that inability or failure to carry out the corrupt purpose is not a defence, then it is uncontroversial. Otherwise, its relevance is not evident. 46. Turning to the joint judgment, it held, as we have seen, that the appellants’ conduct was caught by the relevant phrase because TVB’s enthusiasm for having Chan’s segment as part of the Countdown show to achieve higher audience ratings meant that his participation was an act “in relation to his principal's affairs or business”. That is not an approach consistent with Ch’ng Poh. It appears that in arriving at its decision, the joint judgment sought to distinguish the Privy Council’s decision. First, it treats Ch’ng Poh as authority merely for the proposition that where an “agent’s act has nothing to do with his principal’s affairs or goes beyond the agent’s official powers and duties and is not authorized by his principal” no section 9 offence is committed.[32] This was certainly the position on the facts of Ch’ng Poh[33] and may be said to be its ratio decidendi. However,as noted above,[34] the importance of Lord Lloyd’s judgment lies in its elaboration of the phrase currently under discussion. 47. The joint judgment seeks next to distinguish Lord Lloyd’s explication by asserting that: “The intention with which Ch’ng Poh was concerned is the intention in relation to the offering of the bribe, not the intention of the agent who accepts the advantage. This point is not difficult to understand. The person offering the bribe will not give an advantage to the agent for no reason. He certainly wants the agent who has accepted the advantage to do certain acts which will influence or affect his principal and which will benefit him. It is possible that an agent accepts an advantage solely out of greed, with no intention to actually influence or affect his principal’s affairs by doing his act, but it does not follow that he has not committed the offence under section 9(1)(a) of the Prevention of Bribery Ordinance.”[35] 48. In Ch’ng Poh, Lord Lloyd was analysing the position from the point of view of the offeror, giving hypothetical examples involving Ch’ng Poh offering a bribe to Mr A. But his Lordship was not suggesting that the need to establish the relationship in question only arises in the variant of the offence involving the offering of a bribe. Lord Lloyd stated generally that “the act done (or not done) by the agent must be an act done or not done ‘in relation to his principal's affairs’” and that it was “an essential ingredient of the offence under s.9 that the action or forbearance of the agent should be aimed at the principal”.[36] The section itself employs the same restrictive phrase in both subsections, dealing first with the variant involving an agent who solicits or accepts the advantage and secondly with the variant involving a person who offers the advantage. 49. But even on the narrow view of Ch’ng Poh taken by the joint judgment, the question which had to be – but was not – posed was whether Tseng, acting on behalf of IEAP and OC, in offering remuneration of $112,000 to Chan as an appearance fee, intended his performance in the “Be My Guest” segment to be an act aimed at and intended to influence or affect TVB’s business. I find it impossible to see how the answer could be in the affirmative. Tseng’s position was in principle no different from that of OC which had provided the fee to secure Chan’s performance as OC’s guest in a show promoting its shopping mall. It cannot plausibly be suggested that OC committed a section 9 offence involving TVB’s affairs or business. 50. Properly applying Ch’ng Poh,the joint judgment should also have asked whether, in accepting the fee and doing the show, Chan knew or believed that his fee had been provided by Tseng as an inducement or reward for appearing on the show as an act aimed at TVB’s affairs or business and intended to influence or affect the same. As discussed above[37], this represents the agent’s mens rea requirement in a section 9 offence involving solicitation or acceptance. Again, I can see no basis for an affirmative answer. 51. It follows, in my view, that the Court of Appeal misconstrued section 9 and applied the wrong test for determining whether the necessary relationship between the agent’s act or forbearance and the principal’s affairs or business was proved. On this basis alone, the Court of Appeal’s decision to reverse the trial Judge’s acquittals cannot stand. However, one other aspect of the Court of Appeal’s decision, also bearing on that relationship, is of general importance and ought to be addressed. This involves the question whether, on the true construction of the section, the contemplated act or forbearance by the agent has to be adverse to the principal’s interests. H. The need for conduct adverse to the principal’s interests 52. As noted above,[38] the joint judgment was not troubled by the fact that Chan’s appearance on the show would only have beneficial consequences for TVB and involved no conflict of interest or other detriment to the principal’s interests. It took the view that “[There] is no requirement that the agent’s act has to cause prejudice to his principal’s affairs or business” and that “Ch’ng Poh did not say that the ‘influence’ or ‘effect’ must be adverse”, the term ‘aimed at’ not carrying a negative meaning.[39] 53. I respectfully disagree. In my view, on a proper construction of section 9 in the light of its mischief, the induced or rewarded conduct “aimed at the principal’s business” has to be conduct which subverts the integrity of the agency relationship to the detriment of the principal’s interests. It is not the legislative intent to stigmatize as criminal, conduct of an agent which is beneficial to and congruent with the interests of the principal (as in the present case). 54. I hasten to add that the prejudice to the principal’s interests to which I refer does not need to involve immediate or tangible economic loss to the principal or benefit to the agent at the principal’s expense. Of course, it will frequently (or indeed, usually) do so, but that is not essential on the true construction of the section. The agent may, for instance, be induced to act prejudicially to the reputation of the principal’s business or to divulge confidential information without any immediately palpable loss to the principal. Where the offering, solicitation or acceptance of an advantage is of such a nature as to undermine the integrity of the agency relationship, that is, of such a nature as to injure the relationship of trust and loyalty that a principal is entitled to expect from his agent, this in itself is capable of constituting the necessary detriment. This reflects increasing recognition in the field of employment law, of the importance of the reciprocal duties of trust and confidence in the relationship of employer and employee – perhaps the most common agency relationship – with the evolution of remedies for damage to that relationship.[40] 55. Corrupt transactions with agents which undermine that relationship of trust and loyalty may well give rise to serious reputational damage to the principal even if no direct economic detriment is suffered. This may be illustrated by some hypothetical examples: (a) An employee, A, is a manager with authority to select subordinate employees in his department for promotion. He accepts gifts from a subordinate’s parents who make it plain that they expect their daughter to get promotion, as she does. The employer suffers no detriment (if she is up to the job) but the corrupt transaction, if revealed, will damage its reputation with its other employees. (b) An employee, B, is in charge of a confidential tendering process. Shortly before the closing date B accepts a bribe from X to reveal the lowest tender price so far submitted (say $900,000), thus enabling X to submit a successful tender of $850,000. B has apparently saved his employer $50,000 (if X does the job properly) but the breach of confidence has impaired the relationship of trust and loyalty and the corrupt transaction, if revealed, will damage the employer’s reputation with its regular suppliers. (c) An employee, C, runs the department selling electronic toys in a large store. At Christmas there is intense demand, exceeding supply for a particular toy. C takes a bribe of $500 from a customer to jump the queue on the waiting list. The store suffers no financial loss but the corrupt transaction, if revealed, would damage the store’s reputation with its regular customers. H.1 The mischief of the Ordinance and of section 9 56. At the risk of stating the obvious, the object of the POBO is the suppression of bribery and corruption. The long title makes this clear[41] and the purpose of the legislation has been acknowledged in many reported cases.[42] In Ch’ng Poh, Lord Lloyd noted that “section 9(1)(a) criminalises corrupt transactions with agents ...”[43] In the example he gave of a case which would fall within the section, he spoke of Ch’ng Poh hypothetically bribing Mr A “to secure him a benefit at X & Co’s expense”.[44] 57. The respondent sought to make something of the fact that the word “corruptly”, used in the equivalent provision previously in force,[45] has been removed from the current Ordinance. However, its removal obviously does not change the mischief at which the Ordinance is aimed. In a debate on the Bribery Bill 1970, Mr Oswald Cheung, a Legislative Councillor, expressed doubts as to removal of the word and suggested an amendment to restore it.[46] The Attorney-General responded that it was removed lest inclusion of a reference to corruption “would oblige the Crown, before an accused had a case to answer, to establish that the soliciting or acceptance of a bribe was attributable to a corrupt motive”, experience showing that this had caused difficulties.[47] He went on to explain that, as drafted, “the Crown will need to prove only a payment to the public servant, or agent, or other person. It will then be for the accused to show that he received it for reasons which are unconnected with bribery...” 58. With respect, it was an over-simplification for the Attorney-General to suggest that the Crown would need “to prove only a payment”. As is evident from the structure and elements of the section 9 offence discussed in Section H.2 below, while the need to prove a corrupt motive was avoided by removing a reference to corruption in the wording, the approach adopted involves creating offences requiring proof of actual or proposed transactions involving particular classes of defendants (including agents in the private sector) which are of such a nature as to savour implicitly of corruption. Thus proof of a payment (or an offer or solicitation) meeting the conditions defined by section 9 implicitly establishes a corrupt transaction, subject to the defences of lawful authority and reasonable excuse. H.2 The structure of section 9 59. As Li CJ stated in HKSAR v Cheung Kwun Yin,[48]the purpose of a statutory provision may be evident from the provision itself. The structure and elements of the section 9 offence give a strong indication of the legislative intent. Thus, the section concerns transactions between third persons and agents who have been entrusted with power to act on their principals’ behalf. It focuses on an advantage solicited or accepted by, or offered to, the agent as an inducement for him to act or forbear to act in some way in relation to his principal’s affairs or business. If the agent were duly acting in the ordinary course of his principal’s business, one would expect no such inducement to enter the picture. The situation targeted by section 9 is thus evidently one where the agent is being offered or solicits without permission, lawful authority or reasonable excuse, an inducement to do (or forbear from doing) something “in relation to his principal's affairs or business”. It is a situation which is implicitly likely to involve some deviation from his normal duties and is likely to undermine the integrity of the agency relationship. This is reflected in paragraph (b) of each of sections 9(1) and 9(2)[49] which, by focussing on an agent’s “showing or forbearing to show ... favour or disfavour to any person in relation to his principal’s affairs or business” as the basis of guilt, indicate that section 9 targets transactions tending to undermine the integrity of the agency relationship. 60. As Blair-Kerr J, in R v R E Low,[50] put it in relation to an earlier version of section 9:[51] “It is the inducing influence of the gift in relation to the recipient's performance or non-performance of his duty which is the essence of the offence.” H.3 Fung Hok Cheung 61. In rejecting the argument that the “influence” or “effect” must be adverse, the Court of Appeal relied on HKSAR v Fung Hok Cheung.[52] The facts were described as follows: “... the applicant had access to a corrupt jockey, one Munce, who was prepared to provide tips as to the prospects of success of horses in respect of which he had peculiar information as a result of having ridden them at the request of various trainers; the applicant would pass this information on to PW1 [an ICAC undercover agent] who would place bets in his own name but on Munce's behalf; if a bet was successful a proportion of the winnings would be paid over to Munce, via the applicant who would retain a share, whereas if it was unsuccessful the loss would be borne by, ostensibly, PW1.”[53] 62. The jockey was held to be the agent of the trainers for whom he was riding and Wright J, giving judgment for the Court of Appeal, described the case as one in which the agent “surreptitiously provides confidential information concerning his employer's business to a third party in return for payment”.[54] The applicant was convicted of conspiracy to offer an advantage to an agent, contrary to POBO section 9(2)(a), read together with sections 159A and 159C of the Crimes Ordinance (“CO”).[55] In the passage relied on below, Wright J stated: “The section does not require, nor should the phrase ‘aimed at’ in Commissioner of the Independent Commission Against Corruption v Ch’ng Poh be read as importing, any direct adverse proprietary consequence to the principal's affairs or business. What it requires is that the person offering the inducement intended that the act to be carried out by the agent itself in some way would influence or affect the affairs or business of the principal.”[56] 63. His Lordship added: “The Deputy Judge was fully justified in finding on the evidence before her that the use by the applicant, for financial gain, of the confidential information obtained by Munce as jockey for and agent of the trainer, potentially would affect the trainer's reputation. It requires nothing more than commonsense to appreciate that such a situation would affect the trainer's affairs or business, from which it is self-evident that Munce's conduct was an act in relation to his principal's affairs or business and that the applicant intended that consequence.”[57] 64. It is clear from these two paragraphs that Fung Hok Cheung is not authority for the proposition that a person can be guilty of a section 9 offence even where the contemplated act or forbearance has no potentially adverse impact on the principal’s interests. Wright J was stating that there was no need to show any “any direct adverse proprietary consequence to the principal's affairs or business”. In the next paragraph he endorsed the trial judge’s finding that Munce’s acts potentially damaged the trainer’s reputation and thus would affect his affairs or business. Such reputational damage is not a direct “proprietary” or “economic” consequence, but it is plainly a consequence inimical to the integrity of the agency relationship between Munce and the trainer. It might moreover lead to indirect economic harm to the trainer’s business. 65. I pause to note that in the leading textbook in this jurisdiction on the subject,[58] the learned author asserts the need for prejudice to the principal’s interests: “Section 9 is the only offence in the POBO dedicated solely to corruption in the private sector but it is limited in its scope by the requirement that there exists a principal/agent relationship and that the principal’s interests are prejudiced or at risk of being prejudiced by the secret actions either of his corrupt agent or by a person with whom he is having dealings who seeks to corrupt his agent. This is the traditional narrow view of corruption which is typically found in anti-corruption legislation and is even reflected in Article 21 of the United Nations Convention Against Corruption (UNCAC) which proscribes the offering solicitation or acceptance of undue advantages to or by the employees of private sector entitles “in order that he or she, in breach of his or her duties, act or refrain from acting”. Thus the focus of the UNCAC and other similar offences is on conduct which undermines the integrity of the agent in his relationship with his principal ...” I. Conclusions on section 9 summarised 66. There are variants of section 9 offences differing according to whether they are committed by the agent soliciting or accepting the advantage or by the person offering the same; and whether the act or forbearance is contemplated or already performed. 67. The mens rea requirements of knowledge, belief and intention adhere in their appropriately adapted forms to the essential actus reus ingredients of each of these variants, applied independently to each person allegedly involved in the transaction. 68. The reference in section 9 to the agent’s act or forbearance being “in relation to his principal's affairs or business” is properly construed to mean that the agent’s act or forbearance must be aimed at and intended to influence or affect the principal’s affairs or business in a manner that undermines the integrity of the agency relationship by injuring the bond of trust and loyalty between principal and agent. 69. In offering cases, the prosecution must prove that the offeror intended that the advantage would be accepted as an inducement or reward for or otherwise on account of the agent’s act or forbearance which is aimed at and intended to influence or affect the principal’s affairs or business. 70. In soliciting or accepting cases, the prosecution must prove that the accused agent knew or believed that the advantage was provided as an inducement or reward or otherwise on account of his actual or contemplated act or forbearance as conduct aimed at or intended to influence or affect the principal’s affairs or business. J. Question 2: reasonable excuse 71. Section 9 creates an offence based on conduct performed without lawful authority or reasonable excuse and section 24 lays it down that these are matters of defence which the accused bears the onus of proving: “In any proceedings against a person for an offence under this Ordinance, the burden of proving a defence of lawful authority or reasonable excuse shall lie upon the accused.” 72. In HKSAR v Ho Loy,[59] Fok PJ pointed out that a defence of reasonable excuse involves (i) identifying the matters relied on as constituting the excuse; (ii) determining the genuineness of the excuse; and (iii) assessing “... whether that excuse is reasonable, which the court will do on an objective standard depending on the particular facts of the case”. 73. An issue as to reasonable excuse only arises if there is proof of conduct which prima facie constitutes the section 9 offence so that an assessment of the facts relied on as an excuse is required. Since I have concluded that the appellants do not fall within section 9, there is no realistic basis for assessing the defence in the present case. The prosecution has not proved that they respectively offered and accepted an advantage to induce or reward conduct by Chan aimed at TVB which prima facie undermined the integrity of the agency relationship to TVB’s detriment. It is thus not a meaningful exercise to search for and assess the reasonableness of matters relied on to excuse a state of affairs that never came into existence. This applies, for instance, to Chan’s reliance on his belief that he effectively had TVB’s permission to perform versus the prosecution’s assertion that permission would have been refused. What divided the parties on that issue was whether Chan’s act could be regarded as aimed at and intended to influence or affect TVB’s business – questions bearing on prima facie guilt rather than on reasonable excuse. I shall accordingly confine myself to a discussion of one question of principle relating to reasonable excuse that was canvassed at the hearing. K. Conspiracy, section 24 and the burden of proof 74. As noted above, section 24 provides that the “burden of proving a defence of lawful authority or reasonable excuse shall lie upon the accused”. If that section is engaged and a constitutional challenge mounted on the basis that it infringes the presumption of innocence, a proportionality analysis may be required. I do not, however, intend to enter into that discussion. 75. The point of principle which I wish to address arises out of the Court of Appeal’s direction requiring the Judge to find the appellants guilty solely on the charge of conspiracy for an agent to accept an advantage.[60] The question is whether, in cases of conspiracy, section 24 is engaged so that a reverse onus to prove lawful authority or reasonable excuse is placed on the accused. 76. In HKSAR v Ng Po On,[61] the effect of section 24 was considered in relation to POBO section 14 which made it an offence to fail “without reasonable excuse” to comply with a notice requiring the person served to furnish information relevant to an ICAC investigation. The prosecution argued that section 24 made it clear that reasonable excuse was a matter of defence to be proved by the respondent on the balance of probabilities. This Court held that on a proportionality analysis, the validity of the section 14 offence could be upheld as consistent with the presumption of innocence provided that section 24 was read down to impose merely an evidential burden on the person failing to comply with the notice to raise the issue of reasonable excuse while the prosecution bore throughout the persuasive burden of proving non-compliance and of negating reasonable excuse. 77. In connection with inchoate offences including the offence of conspiracy, I stated:[62] “The precise offence charged has to be borne in mind. Thus, if a defendant is charged with an inchoate offence of conspiracy, attempt or incitement relating to a statutory offence, it does not matter that the substantive statutory offence may, on construction, be found to contain a reverse onus provision. The inchoate offence exists at common law and, in accordance with the general common law rule, does not involve any reverse onus. This is unaffected by s.94A.[63] Thus, in HKSAR v Lam Yuk Fai (2006) 9 HKCFAR 281, §33, a count alleging a conspiracy to commit the statutory offence of transferring a passport to another person without reasonable excuse was held not to involve the imposition of any persuasive burden on the accused and the presumption of innocence was not engaged. Lord Woolf NPJ stated: As Count 3 alleges a conspiracy, the task of the prosecution was to prove in the normal way the nature of the conspiracy; namely that (a) there was an agreement alleged between the appellant and Philip Chu and; (b) the purpose of the agreement, which on the case for the prosecution, was that the passport should be used for an unlawful purpose and so without reasonable excuse (see for example Yip Chiu Cheung v R [1995] 1 AC 111). This conspiracy could not either under S.94A or at common law give rise to any reverse burden. In view of the ingredients of the offence of conspiracy it is unlikely that any charge of conspiracy should give rise to any burden being placed on a defendant.” 78. It may also be noted that CO section 159A(2),[64] which places the offence of conspiracy on a statutory footing, retains the common law approach by providing that even where the substantive offence is one of strict liability in relation to any factual ingredient, a conspiracy to commit that offence nevertheless requires proof of intention or knowledge in relation to that fact or circumstance. 79. It was however made clear in Ng Po On,[65] that the Court was concerned with section 24 exclusively as read in conjunction with, and affecting the meaning of, section 14(4). It was observed that section 24 is also relevant to other provisions of the POBO which involve “reasonable excuse”, including section 9, and noted that “[its] interaction with such other provisions raises separate considerations which fall outside the scope of this judgment”. 80. Mr Caplan QC submitted that, by virtue of POBO section 12A(1), the present case is indeed one where section 24 operates differently. Section 12A(1) provides: “Any person convicted of conspiracy to commit an offence under this Part shall be dealt with and punished in like manner as if convicted of such offence and any rules of evidence which apply with respect to the proof of any such offence shall apply in like manner to the proof of conspiracy to commit such offence.” 81. The argument is that section 24 is a rule of evidence which applies with respect to the proof of a section 9 offence so that section 12A(1) causes it to operate in conspiracy cases just as it does to proof of substantive section 9 offences. 82. In response, Mr David Perry QC[66] relies on CO section 159A(1) which materially provides: “Subject to the following provisions of this Part, if a person agrees with any other person or persons that a course of conduct shall be pursued which, if the agreement is carried out in accordance with their intentions, ... (a) will necessarily amount to or involve the commission of any offence or offences by one or more of the parties to the agreement ... he is guilty of conspiracy to commit the offence or offences in question.” 83. Mr Perry QC’s argument runs as follows: (a) The words “without ... reasonable excuse” appear in the body of the offence-creating provisions of section 9 and form an element of the offence which the prosecution has the burden of proving in the usual way, negating reasonable excuse. (b) To prove a conspiracy to commit a section 9 offence, section 159A(1)(a) makes it necessary as a matter of substantive criminal law for the prosecution to prove that pursuit of the course of conduct agreed to by the alleged conspirators would necessarily amount to or involve the commission of a section 9 offence by at least one of them. (c) If, contrary to Tseng’s submission, POBO section 24, read with POBO section 12A(1), has the effect of reversing the onus for the purposes of conspiracy, it would mean, inconsistently with CO section 159A(1)(a), that the prosecution would not have to prove that pursuit of the alleged conspiratorial agreement would necessarily result in commission of the section 9 offence since it would be relieved from having to prove that pursuit of the agreement would not involve conduct exonerated by a reasonable excuse. (d) Mr Perry QC argued that this could not be the intended effect of section 24 since that section is concerned with rules of evidence and is not intended to modify the substantive law defining the elements of criminal conspiracy as laid down by CO section 159A(1)(a). Accordingly, he submitted that section 24 does not apply to the conspiracy charge and the prosecution’s failure to negate reasonable excuse presents another reason for allowing the appeal. 84. I have difficulty accepting Mr Perry QC’s argument. Its premise is that the words “without reasonable excuse” constitute an element of the offence to be proved by the prosecution. But if section 24 is engaged, it clearly stipulates that “reasonable excuse” is a matter of defence and expressly reverses the onus. Thus an argument founded on the premise that reasonable excuse is not a matter of defence but an ingredient of the offence as the basis for contending that section 24 is not engaged appears to beg the question. 85. However, I arrive at the same conclusion, but by a different route. Section 24 on its face only relates to “any proceedings against a person for an offence under this Ordinance”. A conspiracy to commit a section 9 offence is not an offence under the POBO. It is either a common law offence or, having been put on a statutory footing, an offence contrary to CO section 159A, so that on its face, section 24 does not apply to such conspiracies. 86. To make section 24 apply, the prosecution has to rely on section 12A(1), arguing that it makes section 24’s reverse onus, treated as a “rule of evidence” applicable to the proof of conspiracies to commit section 9 offences in like manner as it applies to proving the substantive offence. 87. But in my view, on its true construction, section 12A(1) is inapplicable to section 24. Section 12A(1) extends “rules of evidence which apply with respect to the proof of any ... offence” [67] under Part II of POBO (including section 9) to conspiracies to commit such offences. It does not cover rules of evidence (assuming a reverse onus provision to be such a rule) regarding available defences. Section 24 stipulates how the defences of lawful authority and reasonable excuse are to be established. It is not concerned with the proof of any offence. 88. The position is therefore as stated in Ng Po On set out above. To prove conspiracy, the prosecution cannot rely on section 12A or section 24 but must, in accordance with the usual common law rule and section 159A(2), discharge the burden of proving that pursuit by the alleged conspirators of their agreement would necessarily result in the commission of a section 9 offence without any (lawful authority or) reasonable excuse availing the accused. 89. Mr Caplan QC acknowledged that if the reverse onus provision did not avail the prosecution, the conspiracy conviction directed to be entered by the Court of Appeal could not stand since reasonable excuse had not been negated by the prosecution. However, he invited the Court in such event to exercise its discretion[68] to order that there be substituted in respect of each appellant, a conviction on the respective alternative charges of offering and accepting an advantage which, he submitted, are clearly within the ambit of section 24. 90. Since I have in any event held that liability cannot be established on the substantive charges because the appellants’ conduct was not caught by section 9, there is no question of any such substitution. L. Disposal of the appeals 91. For the foregoing reasons, I would allow both appeals and quash the appellants’ convictions. I would direct that any submissions as to costs should be lodged within 21 days of the date of the handing down of this judgment with any submissions in reply to be lodged within 14 days thereafter. Mr Justice Tang PJ: Introduction 92. I have had the advantage of reading Ribeiro PJ’s judgment in draft and agree that the appeals should be allowed. The background and facts have been stated by Ribeiro PJ and I gratefully adopt them. However, I am of the view that Chan’s participation in the 2010 countdown edition of “Be My Guest” was an act “in relation to his principal’s affairs or business” but that he had a reasonable excuse to accept OC’s payment within the meaning of s 9(1)(a) of the Prevention of Bribery Ordinance Cap 201 (“Cap 201”). 93. I will state the essential facts which, in my view, support my view that he had a reasonable excuse. 94. It is not disputed that OC and TVB agreed that, a live, staged edition of “Be My Guest” should form part of the countdown programme. Mr Caplan QC accepted that TVB knew that Chan would appear as a guest of OC. The learned trial judge said it would be naive to think that Chan would have done so gratuitously. With respect, I agree. Not only is this plain common sense, the evidence was that OC had to procure Chan’s attendance through Tseng because as the learned trial judge said “everybody knew that to invite (Chan) to take up a job had to proceed through (Tseng)”. Tseng, and his company, IEAP “acted as Chan’s agent in respect of his outside commercial engagements”.[69] The Court of Appeal said: “14. … For this reason they approached Tseng, and later Olympian City concluded a contract with him (on behalf of IEAP), under which he was to procure Chan and Lai Yiu-cheung [Lai][70] to perform in the additional ‘Be My Guest’ show, and Olympian City was to pay IEAP $160,000 as consideration.”[71] “It was agreed that they would wear stickers promoting Olympian City, a requirement that did not apply to the artistes supplied by TVB.”[72] 95. On these facts, I am satisfied that whether Chan had the evidential burden or persuasive burden to prove reasonable excuse,[73] the burden has been discharged.[74] 96. What is a reasonable excuse “depends not only on the circumstances of the individual case but also on the purpose of the provision to which the defence of ‘reasonable excuse’ is an exception.”[75] In order to determine the ambit of the defence of reasonable excuse one needs to take a view on the offence created by s 9.[76] Section 9(1) provides: “Any agent who, without lawful authority or reasonable excuse, solicits or accepts any advantage as an inducement to or reward for or otherwise on account of his - (a) doing or forbearing to do, or having done or forborne to do, any act in relation to his principal’s affairs or business; or (b) showing or forbearing to show, or having shown or forborne to show, favour or disfavour to any person in relation to his principal’s affairs or business. Shall be guilty of an offence.”[77] 97. I would just note[78] that s 9(4) and (5) provide that if the advantage was solicited or accepted with the permission of the principal which complied with s 9(5), neither the agent nor the person who offered the advantage would be guilty of an offence under s 9(1) or (2). 98. To constitute an offence under s 9, there must be a payment (without lawful authority or reasonable excuse) as an inducement to or reward for or otherwise on account of the agent doing or forbearing to do any act in relation to his principal’s affairs or business (the forbidden purpose). In the present case it was not disputed that the payment to Chan was made as an inducement to or reward for him to take part in the countdown edition of “Be My Guest” (Chan’s performance). What was disputed was whether the performance was in relation to TVB’s affairs or business and if so, whether he had a reasonable excuse. 99. Agent is defined to include “a public servant and any person employed by or acting for another”. It was not disputed that Chan was TVB’s agent since he was employed by TVB. TVB was a public body and Chan as its General Manager (Broadcasting) was a public servant.[79] As the learned trial judge pointed out, Chan could have been prosecuted under s 4. However, the prosecution relied on s 9 which covered dealings with agents as well as public servants. Section 4(2) makes it an offence by a public servant to solicit or accept any advantage as an inducement to or reward for or otherwise on account of his “performing … or abstained from performing, any act in his capacity as a public servant”.[80] However, s 9(1)(a) does not require that the agent acted or forborne from acting in his capacity as agent. It is engaged if the act or forbearance was “in relation to his principal’s affairs or business”.[81] The prosecution case was that Chan’s performance was an act in relation to TVB’s affairs or business. Ch’ng Poh 100. Lord Lloyd of Berwick said in Commissioner of ICAC v Ch’ng Poh [1997] HKLRD 652 at 657, a decision of the Privy Council on appeal from Hong Kong: “… the alternative words serve much the same purpose. They are clearly intended to be restrictive. It is not enough that the recipient of the bribe should be an agent in fact.” 101. Ch’ng Poh arose out of an application to quash a search warrant under s 10B of the ICAC Ordinance Cap 204 which was issued based on the commission of an offence under s 9. Briefly, a witness, Reid, had been bribed by Mr A, a solicitor, to made a false affidavit favouring Ch’ng Poh who was appealing his conviction. Mr A was a partner in the firm which acted in Ch’ng Poh’s appeal. Ch’ng Poh provided the money for the bribe. 102. Lord Lloyd who delivered the advice of the Privy Council was of the view that since the case being made by ICAC was that Reid had been bribed by Mr A: “… s.9(1)(a) of the Bribery Ordinance was simply not engaged. As Litton V-P put it: Section 9(1)(a) criminalises corrupt transactions with agents: not dishonest acts by agents.”[82] 103. However, it was argued on behalf of the ICAC in the Privy Council that the argument had been misunderstood and that its case was that Mr A must have received a bribe to do an act in relation to X & Co’s business. Their lordships did not accept that was the case relied on in the courts below but in light of it Lord Lloyd said of the words “in relation to his principal’s affairs” at 657: “ … mean that, for the section to apply, the person offering the bribe, must have intended the act or forbearance of the agent to influence or affect the principal’s affairs. Accordingly s.9 would apply if Ch’ng Poh had bribed Mr A to secure him a benefit at X & Co’s expense, for example, to arrange a reduction in X & Co’s ordinary professional fees; or if X & Co were induced to act in a way in which they would not otherwise have acted. Thus it is an essential ingredient of the offence under s.9 that the action or forbearance of the agent should be aimed at the principal. If it is sufficient for the purposes of the person offering the bribe that the agent should act on his own without involving his principal, then, whatever other offence may have been committed, it is not a corrupt transaction with an agent for the purposes of s.9. If the facts put before the magistrate had been that Ch’ng Poh offered Mr A a bribe in order to secure the filing by X & Co of an affidavit which they both knew to be false, then a case might just have been made out. But as already mentioned, that is not a fair reading, or indeed a possible reading, of the information sworn by Mr Osborn.” Section 9(1)(a) 104. I will consider s 9(1)(a) first and return to Ch’ng Poh later. Section 9(1) and (2) was first enacted as s 4 of the Prevention of Corruption Ordinance Cap 215 (“Cap 215”) in 1948.[83] Section 4 of Cap 201 was s 3 in Cap 215. When enacted, those sections required that the advantage be offered or accepted corruptly. As Bribery and Corruption Law in Hong Kong 3rd edition by Ian McWalters SC (as McWalters JA then was), explained the word “corruptly” was understood to require “purposely doing an act which the law forbids as tending to corrupt”.[84] 105. When Cap 201 was enacted in 1970 and ss 3 and 4 of Cap 215 were replaced by s 9 and s 4 of Cap 201 the word “corruptly” was removed and substituted by “without lawful authority and reasonable excuse”. 106. The reason for the change was explained by the Attorney General[85] in response to the remarks by Hon Mr Oswald Cheung that he was “staggered” that “the word corruption” was not used in the proposed Cap 201. The Attorney said the word was not used: “… since it would oblige the Crown, before an accused had a case to answer, to establish that the soliciting or acceptance of a bribe was attributable to a corrupt motive. Unfortunately, past experience has been that this is difficult, and many accused persons have not been charged, and some have been charged but acquitted, because the prosecution was not able to establish that a payment of money in suspicious circumstances was corrupt but had to leave it to the court to draw an inference of corruption, which it was sometimes reluctant to do.”[86] 107. The mischief against which the amendments were made is clear, namely, that corrupt motive was not to be an essential element of the offences under s 4 or s 9 of Cap 201. The Attorney General went on to say: “As the bill is now drafted, the Crown will need to prove only a payment to the public servant, or agent, or other person.” 108. Section 9 is straightforward. It makes it an offence for an agent to solicit or accept any advantage as an inducement to or reward for or otherwise on account of his doing or forbearing to do “any act in relation to his principal’s affairs or business” in the absence of lawful authority or reasonable excuse.[87] In other words, if an advantage was solicited or accepted in respect of an act or forbearance which related to the principal’s affairs or business, absent lawful authority or reasonable excuse or permission under s 9(4) and (5), an offence is committed. The deliberate removal of the word “corruptly” is significant and I do not believe s 9 requires that any payment for a forbidden purpose must be made with a corrupt motive. With respect, the Attorney General was exactly right when he said: “As the bill is now drafted, the Crown will need to prove only a payment to the public servant, or agent, or other person.” Influence or Affect 109. Ch’ng Poh is authority that for an act or forbearance to relate to the principal’s affairs or business it must have been intended to “influence or affect the principal’s affairs.” And it would not so relate if the agent “should act on his own without involving his principal”. With respect, I agree. Thus, a moonlighting employee would not fall foul of s 9 even if moonlighting was strictly forbidden by his contract of employment.[88] But Lord Lloyd went on to say: “Thus it is an essential ingredient of the offence under s.9 that the action or forbearance of the agent should be aimed at the principal.”[89] Aimed At 110. It is not clear what if anything “aimed at the principal” additionally requires. The expression has been considered by the English Court of Appeal in R v Majeed and Another [2013] 1 WLR 1041 when they considered s 1 of the 1906 Act on which s 9 of Cap 201 was based. The only material difference between s 1 of the 1906 Act and s 9 is that section 1 had not replaced “corruptly” with “without lawful authority or reasonable excuse.” 111. Majeed was concerned with cricketers who were bribed to play in a certain way to facilitate betting on cricket matches where one could bet not only on the outcome but also, for example, “on individual events during the course of an over or passage of play”.[90] One of the arguments relied on was that the actions of the cricketers on the field were not aimed at their principals, the Pakistan Cricket Board or the Essex Cricket Club (“the boards”). Relying on the passage in Ch’ng Poh quoted above,[91] counsel submitted: “The boards did nothing, and nothing was expected of them. The bribes were not intended to and did not influence them in any way. … He suggested that, like the dishonest solicitor, ‘A’, in the Ch’ng Poh case, the way in which [the cricketers performed in their matches], was personal to them. Like the firm in which ‘A’ was a partner, the boards did not and were not expected to do anything at all by way of response or as a reaction to the giving and the acceptance of the bribes.” 112. Lord Judge CJ who delivered the judgment of the Court[92] said: “21. … The problem for [counsel] is that [the passage] was addressing the stark reality that ‘A’s’ firm was remote from the bribes paid by Ch’ng Poh, and wholly unaffected by them. They were not paid to ‘A’ for any purposes associated, directly or indirectly, with the business or affairs of his firm. In fact they were not paid to ‘A’ at all. He simply organised their onward transmission to the intended recipients. It is however clear from the judgment that the transaction would have fallen within the ambit of the legislation if the bribes had been accepted by ‘A’ with the intention or for the purpose of influencing or affecting the affairs of the firm. This is why Lord Lloyd went on to acknowledge that if the bribe had been offered to ‘A’ by Ch’ng Poh in a way which would have led the firm to file an affidavit which both ‘A’ and Ch’ng Poh knew to be false, the warrant, as issued, might have been appropriate. This would have been because, although Lord Lloyd did not need to spell it out, the bribe would then have involved the firm, albeit wholly innocently on the part of the firm, in the corrupt activity. 22. We cannot find anything in the judgment in the Ch’ng Poh case … which lends support to [counsel]’s argument, and if it did, the submission would be wholly inconsistent with the clear statutory language. 23. … looking at the realities of the situation, there could on the evidence have been nothing closer to the heart of the affairs or business of a cricket board than the performance of the players selected by them to represent their and his country, or their and his county.” 113. With respect, like the English Court of Appeal, I believe the language of s 9 does not require that the act or forbearance should, apart from influencing or affecting the principal’s affairs or business thus involving them be also intentionally directly or indirectly aimed at the principal though one would expect that in many cases they would do so. Nor do I understand Lord Lloyd to have said otherwise. 114. So the question here is whether Chan’s performance related to TVB’s affairs or business. With respect, I agree with Ribeiro PJ’s statement of the elements and mens rea at paras 20 and 21 of his judgment. However, it was TVB’s countdown which was televised by TVB and the “Be My Guest” portion was part of the programme. So TVB was very much involved. Suppose TVB had requested Chan to take part in this edition of “Be My Guest” and Chan had agreed. Undoubtedly his participation would have related to TVB’s affairs or business. I am not sure the fact that he had appeared as OC’s guest and paid by them to do so would necessarily result in his not being involved in the business or affairs of TVB. Morgan 115. DPP v Morgan [1970] 3 All ER 1053 was also concerned with the interpretation of s 1 of the 1906 Act, Lord Parker CJ in the Divisional Court[93] was quite satisfied that these words “fall to be widely construed”.[94] Morgan was an employee of Rover’s as well as the convenor of shop stewards. He solicited a bribe from H who worked as a subcontractor for Rover’s who had been blacked by the union, telling H that he could possibly make it easier for H at the management/union meeting by suggesting that there was no objection to H doing the work. Morgan is often cited for the proposition that an agent can have more than one principal. 116. But it decided more than that. It was submitted on behalf of Morgan that: “… if the £25 was obtained as an inducement for doing an act, ie for procuring that there would be no objection by the union to Mr Hurford doing the work, the act was not in relation to Rover’s affairs but in relation to those of the union; that, if the £25 was so obtained, it was not obtained by the appellant as agent of Rover’s but as agent of the union.”[95] 117. Of course, the fact that Morgan was the agent of Rover’s was not to the point unless the act “related to the affairs” of Rover’s as well. Hence, Lord Parker CJ said at page 1057: “The real point is whether, there being an obtaining of money as an inducement for the doing of some other act, that act was to be in relation to his principal’s affairs, namely Rover’s affairs.” (my emphasis) 118. Lord Parker went on to say at pages 1057 to 1058: “ … It is counsel for the appellant’s argument that the words “any act in relation to his principal’s affairs” in s 1(1) must mean in direct relation to his principal’s affairs or, put another way, in relation to matters concerning his principal where he owes a duty as an agent. Read in that way, it can be said that, while the appellant was an agent of Rover’s, nevertheless the act in question was in relation to union affairs albeit against the background of the business that Rover’s carried on. For my part, I am quite satisfied that those words “in relation to his principal’s affairs” fall to be widely construed, as indeed they were in the only case to which the court has been referred, R v Dickinson, R v De Rable,[96] where Pritchard J, in giving the judgment of the Court of Criminal Appeal, said: “In the judgment of the court the words of s 1 of the Act of 1906 are designedly very wide, and it would be undesirable in the extreme to narrow their meaning in the way which would be necessary if the argument on this first point were held to be valid.” It seems to me that the conditions precedent to an offence being proved are present here; the appellant was an agent, and what was done here, albeit it was in relation to the union affairs, was also in relation to his principal’s affairs, namely Rover’s affairs.” 119. So here, the fact that Chan performed as the guest of and was paid exclusively by OC is not a sufficient answer to a charge under s 9 unless Chan’s performance did not also relate to TVB’s affairs or business. I am also of the view that it would be undesirable in the extreme to narrow down the meaning of those words. It is of course a question of fact whether the agent’s act or forbearance relates to the affairs or business of his principal. On the facts of this case, I am of the view that Chan’s performance related to the affairs or business of TVB. Prejudice to the Principal 120. Ribeiro PJ is of the view that the agent’s act or forbearance must be adverse to the principal’s interest,[97] in the sense, I think, that it must “prima facie undermined the integrity of the agency relationship to [the principal’s] detriment.”[98] I regard any advantage solicited or accepted without the knowledge of the principal to have such effect. Naturally, in most cases under s 9, the act or forbearance would be adverse to the principal’s interest which was the reason for the advantage. But as Lawton LJ said in connection with s 1 of the 1906 Act: “The mischief aimed at by the modern statutes dealing with corruption is to prevent agents and public servants being put in positions of temptation.”[99] 121. I do not accept that s 9 requires any other detriment. The acceptance of a payment to favour one of two equal competitors is no less an offence under s 9. Nor would it have made any difference in Morgan, if Morgan’s promised effort would have stopped a crippling strike. 122. Ribeiro PJ said[100] that on the Court of Appeal’s reasoning, namely that Chan’s performance would boost audience rating at no cost to TVB, “Chan’s conduct was wholly in line with and beneficial to TVB’s interest”. In my view such benefits have no bearing on the offence. Fiduciary obligation of agents 123. Cap 201 was enacted to make “further and better provision for the prevention of bribery” and in construing s 9 one naturally has such purpose in mind. Lord Thomas of Cwmgiedd CJ, in R v J(P) and others [2014] 1 WLR 1857[101] said of the 1906 Act that: “14. … It is clear from the materials before us that the mischief at which the provision was principally aimed was criminalising the bribery of agents in commercial transactions, particularly commissions paid to the agent without the knowledge and informed consent of his principal. These were and are commonly referred to as secret commissions.”[102] 124. And that the aim and purpose of the 1906 Act: “17. … was entirely in accordance with the underlying law relating to the fiduciary obligations of agents established in the cases to which we have referred.” 125. One of those cases was Parker v McKenna (1874-75) LR 10 Ch App 96 where James LJ at 124-125 said: “ … I do not think it is necessary, but it appears to me very important, that we should concur in laying down again and again the general principle that in this Court no agent in the course of his agency, in the matter of his agency, can be allowed to make any profit without the knowledge and consent of his principal; that that rule is an inflexible rule, and must be applied inexorably by this Court, which is not entitled, in my judgment, to receive evidence, or suggestion, or argument as to whether the principal did or did not suffer any injury in fact by reason of the dealing of the agent; for the safety of mankind requires that no agent shall be able to put his principal to the danger of such an inquiry as that.” 126. In my opinion, once it is established that an advantage was offered or accepted by an agent for a forbidden purpose then absent a reasonable excuse[103] an offence under s 9 would have been committed. Furthermore, I would not regard absence of prejudice to the principal, or that the act was beneficial to the principal to be a reasonable excuse. I see no reason why criminal law should differ from civil law in this respect. Just as in civil law the principal should not be put to the danger of such an inquiry, to permit such inquiries in criminal prosecution would set back the fight against corruption. It is well to remember what the Attorney General said in moving the second reading of the Bill: “Sir, it is impossible to assess with any accuracy the extent to which society in Hong Kong is affected by corruption … But … corruption does exist here to an extent which not only justifies, but demands, that the utmost efforts be made to eradicate it from our public and business affairs.”[104] Since 1970, corruption has become less of a problem in Hong Kong but it has not been defeated. 127. Section 4 which covers public servants may provide a useful comparison. There, I do not believe it is suggested that prejudice to public service is required or that its absence might provide a defence. Under s 4(2) “any public servant who, whether in Hong Kong or elsewhere, without lawful authority or reasonable excuse, solicits or accepts any advantage as an inducement to or reward for or otherwise on account of his – (a) performing or abstaining from performing, or having performed or abstained from performing, any act in his capacity as a public servant; … shall be guilty of an offence.” 128. As I understand s 4(2)(a), a public servant, say a postman, who accepted any advantage from a householder would be guilty of an offence under s 4 even if the householder had offered it in appreciation of the fact that he had gone out of his way to deliver a letter or parcel. It is important, given the scourge of corruption under which Hong Kong had suffered for so long,[105] that it should be clearly understood that under no circumstances should any public servant accept any advantage for acting or forbearing to act in his capacity as a public servant. The corrupting tendency of any offer or acceptance of advantage is too obvious to require elaboration. Nor would I leave it to be determined on a case by case basis whether the tip was deserved and whether there was any prejudice. 129. If prejudice to public service is not required under s 4, I see no reason for requiring it for an offence under s 9. Reasonable excuse 130. I turn now to consider reasonable excuse. 131. R v J (P) was also concerned with s 1 of the 1906 Act and involved the allegation that the defendants had conspired corruptly to give agents of the tax authorities of a state in the Commonwealth a substantial sum of money as an inducement to show favours to a company in relation to the calculation of tax owed by that company to the tax authorities. The issue before the court was whether the word “corruptly” implied that the payment for the prohibited purpose had to have been paid or received secretly and without the knowledge and informed consent of the principal. 132. The defendant’s contention was: “8(ii) … that the term ‘corruptly’ connoted secrecy. As the 1906 Act had been formulated in terms of principal and agent, it must follow that a payment could not be secret if it was made with the knowledge and consent of the principal. Thus it must be for the prosecution to prove, as part of its case, the specific ingredient of lack of knowledge or informed consent by the principal; in essence the element of corruption was the doing of the act prohibited without having first made full disclosure to the principal and obtaining his informed consent.” 133. Given that the 1906 Act required that the giving be made corruptly, Lord Thomas said: “28. There might well be cases where the defendant who is a commercial agent avers that his principal knew of the payment as he had made full disclosure and gave his informed consent; or the defendant might aver that he honestly believed that the agent’s principal knew and gave his informed consent to the receipt of the payment by his agent. It would in such a case be necessary when deciding whether the payment was or was not made or received corruptly for the jury to consider that evidence. However, there is no requirement that the prosecution specifically prove the lack of knowledge and informed consent of the principal separately.” 134. Had the word corruptly been retained in Hong Kong, I would have concluded that on the facts of this case the payment did not fall within s 9. However, the word “corruptly” having been replaced, the payment being for a forbidden purpose, it fell within s 9, and it is necessary to consider whether there was any reasonable excuse. The Court of Appeal was of the view that there was no reasonable excuse. 135. In the second judgment of the Court of Appeal[106] it said at para 88: “We agree that permission[107] and reasonable excuse are two separate defences provided for by the Ordinance. We also agree that reasonable excuse is wider in scope than permission.” With respect, I agree. 136. The Court of Appeal went on to say: “88. … Under some circumstances, it is possible that an agent who does not have his principal’s permission but accepts an advantage which is related to his principal’s business can use reasonable excuse as his defence. Take for example, an agent who has from time to time applied to his principal for permission to accept an advantage on account of his doing an act in relation to his principal’s business, and to whom the principal has always granted permission, will reasonably believe that under the same circumstances he can obtain his principal’s permission. However, on one occasion, when the agent applied for permission from his principal pursuant to section 9(5)(b) of the Ordinance after he accepted an advantage, his principal, without informing him of any change, suddenly, for no reason, refuses to give him permission. The agent cannot legitimize his acceptance of the advantage by the permission provided in the Ordinance, but it is highly likely that he can successfully convince the court that he has a reasonable excuse and thereby legitimizing his acceptance of the advantage.” 137. With respect, on these facts I feel sure that the defence of reasonable excuse has been established because they showed that the agent honestly believed that his principal would not object to his accepting the advantage. In other words, he had the consent of his principal to accept the payment, the paradigm of a reasonable excuse. In my view, that is precisely Chan’s position. TVB accepted it knew that Chan would appear in the programme as a guest of OC. On the evidence the inescapable conclusion was that Chan would be paid by OC through IEAP.[108] It is true that there was no evidence that TVB knew how much Chan would be paid by OC. It was suggested that absent such knowledge there was no informed consent. What constitutes a fully informed consent is a question of fact and “there is no precise formula which will determine all cases”.[109] In this case I do not believe it mattered. On the facts TVB must be taken to have consented to Chan performing in the programme as OC’s guest and be paid for it. The exact amount was of no consequence to TVB. 138. For the above reasons, I am of the view that no offence was committed under s 9 because Chan had a reasonable excuse in appearing as OC’s guest and be paid for it. It follows that both he and Tseng should be acquitted of all the charges relating to this incident. I would allow their appeals. Mr Justice Fok PJ: 139. I have had the benefit of reading in draft the judgments of Ribeiro PJ and Tang PJ. I agree with them in their respective conclusions that these appeals should be allowed. As will be apparent, however, Ribeiro PJ arrives at his conclusion by a very different route to that of Tang PJ and, for that reason, I add this short judgment to indicate my agreement with the reasoning of Ribeiro PJ and to explain my preference for his reasoning in arriving at the conclusion that the two appellants were not guilty of an offence under section 9 of the POBO. 140. Ribeiro PJ’s analysis of the need, if guilt under section 9 is to be established, for there to be conduct adverse to the principal’s interests (Section H of his judgment) is, in my opinion, sound and consistent with the language of section 9 properly construed in accordance with its context and purpose. 141. I do not, with respect, share the view of Tang PJ that the omission of the word “corruptly” in section 9 (which had previously been found in section 4 of the Prevention of Bribery Ordinance (Cap.201)) means that, for an offence under section 9 to be made out, the prosecution need prove only a payment to an agent of which his principal does not know and to which he does not consent. 142. The section does not, in my view, criminalise any and all payments of money by a third party to an agent made without the principal’s knowledge and consent. If it had been intended to cast the offence in such wide terms (and they would be very wide indeed), section 9 would have been very differently and much more simply worded. 143. Instead, the Legislature required that the advantage solicited, accepted or offered must have been as “an inducement to or reward for” the agent’s doing (or forbearing to do) an act “in relation to his principal’s affairs or business” (in sub-sections 9(1)(a) and 9(2)(a)) or showing or forbearing to show “favour or disfavour” to someone “in relation to his principal’s affairs or business” (in sub-sections 9(1)(b) and 9(2)(b)). The words “inducement” and “reward” require the advantage solicited, accepted or offered be invested with some quality of purpose. This is reinforced by the use of the words “favour” and “disfavour” in sub-sections 9(1)(b) and 9(2)(b). 144. Likewise, the Legislature required that the advantage be “in relation to his principal’s affairs or business”. Section 9 is not the private sector mirror image of the section 4 offence. The offence in the latter section is established on proof of an advantage solicited or accepted by a public servant for an act “in his capacity as a public servant”. It is the nature of the act performed (or not) as a public servant in return for the advantage that invests the advantage with its criminality. In contrast, for section 9, it is not enough that the advantage is solicited or accepted by the agent in his capacity as such. It must additionally be “in relation to his principal’s affairs or business”. 145. Those words in section 9 are critical. They were construed by Lord Lloyd of Berwick in Commissioner of the ICAC v Ch’ng Poh [1997] HKLRD 652 at pp.656-657 as meaning that the action or forbearance of the agent must have been intended to “influence or affect the principal’s affairs” and be “aimed at the principal”. It would be surprising, to say the least, for Lord Lloyd to have expressed himself as he did if he thought that those words simply meant that the payment must be “related to” the principal’s affairs or business in the broad sense urged by the prosecution in these appeals. 146. I do not therefore agree that the Attorney General’s statement that “the Crown will need to prove only a payment to the … agent” is a correct or complete statement of the elements of the offence under section 9. It is unlikely that this was his intention given the language of section 9 as a whole. Rather that statement, seen in context, was commenting on the wording of the then proposed section 9 “[a]s … now drafted” and addressing the concern he expressed as to what the Crown would have to establish “before an accused had a case to answer”. Removing the word “corruptly” from section 9 did not mean that there was no need for a qualitative assessment of the payment and, on its wording, the section clearly requires an examination of the nature, purpose and context of the payment. This is referred to, in Tang PJ’s judgment, as “a forbidden purpose”. I respectfully agree with the need for there to be such a “forbidden purpose” and the real debate in these appeals has been directed to the search for a proper definition of that prohibited purpose. 147. In my view, that definition is to be found in Sections H and I of the judgment of Ribeiro PJ. I also agree in all other respects with that judgment. Mr Justice Stock NPJ: 148. I agree with the judgments of Mr Justice Ribeiro PJ and Mr Justice Fok PJ. Lord Walker NPJ: 149. I agree with the judgments of Mr Justice Ribeiro PJ and Mr Justice Fok PJ. Mr Justice Ribeiro PJ: 150. The appeals are unanimously allowed and the appellants’ convictions are quashed. Any submissions as to costs should be lodged within 21 days of the date of the handing down of this judgment with any submissions in reply to be lodged within 14 days thereafter. Mr Joseph Tse SC, Mr Charles J. Chan and Mr Byron C.Y. Tsang, instructed by Ivan Tang & Co., for the Appellant in FACC 11/2016 Mr David Perry QC, Ms Maggie Wong, Mr Derek C.L. Chan and Mr Samuel Sung, instructed by Angela Lau Law Office, for the Appellant in FACC 18/2016 Mr Jonathan Caplan QC and Mr Eric Kwok SC, on fiat, and Mr Felix Tam SPP, of the Department of Justice, for the Respondent in FACC 11/2016 and FACC 18/2016 [1] Cap 201. [2] DCCC 1214/2010. [3] The Case Stated by the Judge is dated 5 March 2012, CACC 103/2012 (“1st Case Stated”). [4] Yeung VP, Cheung and Yuen JJA, CACC 355/2011 and CACC 103/2012 (21 November 2012) (“1st CA Judgment”). [5] DCCC 1214/2010. [6] The second Case Stated by the Judge is dated 17 September 2014, CACC 183/2014 (“2nd Case Stated”). [7] Yeung VP, Yuen and Pang JJA, CACC 92/2013 and CACC 183/2014 (26 October 2015) (“2nd CA Judgment”). [8] Yeung VP, Yuen and Pang JJA, CACC 103/2012 and CACC 183/2014 (26 January 2016). [9] Tang and Fok PJJ and Stock NPJ, FAMC 4/2016 (29 June 2016) in Chan’s case; and Ribeiro, Tang and Fok PJJ, FAMC 40 of 2016 (4 October 2016) in Tseng’s case. [10] In Sections F, G and H of this judgment. [11] POBO section 11. See HKSAR v So Kam Tim [1997] HKLRD 1123 at 1126. [12] (CACC 497/2002, 24 August 2004, unreported) at §44. Citing R v Li Fook Siu Ronald [1991] 2 HKLR 288, see per Fuad VP at 298. [13] That is, an act “aimed at” and “intended to influence or affect” the principal’s affairs or business. See Section F of this judgment. [14] [1997] HKLRD 652. [15] In Section F of this judgment. [16] 1st Case Stated §11. [17] Ibid §12(6). [18] At §§32-35. [19] 1st CA Judgment §§105-106. [20] 1st CA Judgment, §128. [21] [2008] 5 HKLRD 846. [22] In Section H.3 of this judgment. [23] 1st CA Judgment, §153. [24] 1st CA Judgment §154. [25] Acting for the respondent with Mr Eric Kwok SC and Mr Felix Tam SPP. [26] [1997] HKLRD 652. [27] At 656-657. [28] At 657. [29] 1st CA Judgment, §150. [30] In Section H.1 of this judgment. [31] (2005) 8 HKCFAR 628 at §109. This duty was reiterated by Lord Woolf NPJ in HKSAR v Lam Yuk Fai (2006) 9 HKCFAR 281. [32] 1st CA Judgment, §101-104. [33] As noted in R v Majeed, R v Westfield [2013] 1 WLR 1041 at §§19 and 21. [34] Section F of this judgment. [35] 1st CA Judgment, §122. [36] Ch’ng Poh at 657. [37] Section C of this judgment. [38] Section D of this judgment. [39] 1st CA Judgment, §128. [40] See for example Mahmud v Bank of Credit and Commerce International SA [1998] AC 20, per Lord Nicholls of Birkenhead at 35 and Lord Steyn at 45-46. [41] “To make further and better provision for the prevention of bribery and for purposes necessary thereto or connected therewith”. [42] Including in this Court in HKSAR v Ng Po On (2008) 11 HKCFAR 91 at §§51 and 69. [43] Commissioner of ICAC v Ch’ng Poh [1997] HKLRD 652 at 656. [44] Ibid at 657. Italics supplied. [45] Prevention of Corruption Ordinance, Cap 215, section 4(a) (enacted in 1948), referring to any agent who “corruptly accepts or obtains [etc] ... any gift or consideration as an inducement or reward [etc]”. [46] 18th November 1970, Resumption of debate on second reading (21.10.70), Hansard p 196-197. [47] Mr Denys Roberts, ibid, p 201. [48] (2009) 12 HKCFAR 568 at §14. [49] Set out in Section C above. [50] [1961] HKLR 13 at 94 (Full Court). [51] Section 4(a) of the Prevention of Corruption Ordinance (Cap 215) which provided that if any agent corruptly accepts any gift as an inducement or reward for doing or forbearing to do or for having done or forborne to do any act in relation to his principal's affairs, he shall be guilty of an offence. [52] [2008] 5 HKLRD 846. [53] Ibid at §3. [54] Ibid at §37. [55] Cap 200. [56] [2008] 5 HKLRD 846 at §39. [57] Ibid at §40. [58] Ian McWalters SC, Bribery and Corruption Law in Hong Kong (Lexis Nexis 3rd Ed), §9.5. [59] (2016) 19 HKCFAR 110 at §36. [60] 2nd CA Judgment, §93. [61] (2008) 11 HKCFAR 91. [62] At §37. [63] Section 94A of the Criminal Procedure Ordinance (Cap 221) which relieves the prosecution of negativing exceptions or exemptions or qualifications to the operation of a law creating an offence. [64] Cap 200. Section 159A(2): “Where liability for any offence may be incurred without knowledge on the part of the person committing it of any particular fact or circumstance necessary for the commission of the offence, a person shall nevertheless not be guilty of conspiracy to commit that offence by virtue of subsection (1) unless he and at least one other party to the agreement intend or know that that fact or circumstance shall or will exist at the time when the conduct constituting the offence is to take place.” [65] (2008) 11 HKCFAR 91 at §11. [66] Appearing with Ms Maggie Wong, Mr Derek C L Chan and Mr Samuel Sung for Tseng. [67] Such as the rules of evidence contained in POBO sections 10, 20, 21 and 21A. [68] Pursuant to section 83A of the Criminal Procedure Ordinance (Cap 221) in conjunction with section 17(2) of the Hong Kong Court of Final Appeal Ordinance (Cap 484). [69] Para 4 above. [70] A popular artiste in the employ of TVB. [71] Yeung VP, Cheung and Yuen JJA, CACC 355/2011 and CACC 103/2012 (21 November 2012). [72] Para 7 above. [73] The issue is complicated by the fact that we are concerned with the charge of conspiracy. See section K of Ribeiro PJ’s judgment. [74] Like Ribeiro PJ, I will not consider whether s 24 infringes the presumption of innocence. [75] Taikato v R (1996) 186 CLR 454 at 464. [76] All references are to Cap 201 unless otherwise stated. [77] Section 9(2) provides a corresponding offence for the offeror. It is not necessary to deal separately with him. [78] No reliance on such permission. [79] DCCC 1214/2010, para 29. [80] Section 4(2)(a). Section 4(1) contains the corresponding offence for the offeror, with which we are not concerned. [81] It was not the prosecution case that Chan was acting in his capacity as a public servant. [82] At p 656. [83] Section 9(4) and (5) were introduced by amendment in 1980. Section 4 of Cap 215 was based on s 1 of the Prevention of Corruption Act 1906 (“the 1906 Act”). [84] Para 1.31, following Cooper v Slade (1858) 6 HLC 746, Willes J and R v Smith [1960] 2 QB 423. [85] During the second reading. [86] Hansard 18 November 1970 at 201. [87] However, if the soliciting or acceptance was made with permission by the principal in compliance with s 9(4) and (5), then neither the agent nor the offeror would be guilty of an offence under s 9(1)(a) or (b). [88] There was a discussion at the hearing over whether if a junior clerk employed by TVB were to return to work after office hours for a contracted cleaner to clean TVB’s premises whether he could come within s 9. I feel sure the defence of reasonable excuse would be available. It would no more affect his employer if he was engaged after work to make sandwiches for the caterer who supplied them to his employer. But, the matter is fact sensitive. The situation of a manager who owns an outside catering service and sells sandwiches to his employer without any disclosure may be more problematical. Here Chan’s contract of employment with TVB did not permit outside work without the written consent of TVB. Mr Caplan QC rightly accepted that this has no bearing on whether Chan’s performance in “Be My Guest” related to TVB’s affairs or business. But, whether his appearance in the “Be My Guest” related to the affairs or business of TVB is not easily resolved. [89] P 657. [90] 1043G. [91] Para 103 above. [92] Lord Judge CJ, Openshaw, Irwin JJ. [93] Together with Ashworth and Browne JJ. [94] With the agreement of Ashworth and Browne JJ. [95] P 1056e. [96] (1948) 33 Cr App Rep 5 at 9. [97] Paras 61-65 above. [98] Para 73 above. [99] R v Wellburn and Others (1979) 69 Cr App Rep 254 at 265, CA, Lawton LJ, Cantley and Willis JJ. [100] Para 28 above. [101] Rafferty LJ, Henriques J. [102] Lord Thomas also pointed out that s 1 also covered public servant which has a bearing on its construction. [103] I ignore lawful authority or permission under subsections (4) and (5) which are irrelevant for the present purpose. [104] Second reading, Attorney General, Hansard 21 Oct 1970, p 131. [105] For a long time, tipping was expected for many public service, including service in public hospitals. [106] Yeung VP, Yuen and Pang JJA. [107] Complying with ss (4) and (5). [108] Para 94 above. [109] Bowstead and Reynolds on Agency 20th edition 6 – 039. 
Mr Justice Ribeiro PJ: 1. I agree with the Judgment of Mr Justice Tang PJ. Mr Justice Tang PJ: 2. Part X of the Insurance Companies Ordinance, Cap 41 (“the Ordinance”) was added by amendment in 1994 “to put into place a scheme for the supervision of self regulation by the insurance industry of insurance agents and brokers.”[1] The Secretary for Financial Services in moving the second reading of the Bill, explained: “The main objective of the Bill is to define the role of agents and brokers. … It is also an objective of the Bill to strengthen protection for policy holders by requiring agents and brokers to meet specified standards before being appointed or authorized. Insurers are required to comply with the Code of Practice drawn up by the Hong Kong Federation of Insurers and approved by the Insurance Authority for the appointment and administration of agents.” 3. Part X requires the Hong Kong Federation of Insurers (“HKFI”), with the approval of the Insurance Authority, to issue a Code of Practice for the Administration of Insurance Agents (“the Code”).[2] An insurer is required to comply with the Code.[3] Failure to do so carries potential criminal consequences.[4] An insurance agent[5] is also required by the Ordinance to comply with the Code of Practice. Under s 67(5), the Insurance Authority may require “an insurance agent to supply information that verifies … the insurance agent’s, compliance with the code of practice” and it is an offence under s 77(6) if an insurance agent fails “to supply information that is required under s 67(5) that verifies compliance with the code of practice”. 4. A Code of Practice was duly issued and the Insurance Agents Registration Board (“IARB”) was established by the HKFI to administer the Code of Practice. We are concerned with the June 2004 version of the Code of Practice. Under para 8 of the Code: “The IARB may issue Guidance Notes from time to time as to how it intends to exercise its powers and fulfil its responsibilities under this Code. Such Guidance Notes shall not form part of this Code.” We are concerned with the Guidance Note revised on 13 June 2005. 5. The Code also contained provisions for the regulation of a technical representative which is defined to mean: “a person (not being an insurance subagent who is classified as an insurance agent for the purpose of this Code) who provides advice to a policy holder or potential policy holder on insurance matters for such insurance agent, or arranges contracts of insurance in and from Hong Kong or behalf of that insurance agent.” 6. A technical representative is not an insurance agent but they share many common characteristics[6] and there is little difference in the regulation of insurance agents and technical representatives under the Code.[7] Indeed, given their common qualification, a technical representative is also qualified to work as an insurance agent. As indeed was the case with Mr Singh, the appellant who was the defendant at trial. For the purpose of this appeal, it is sufficient to note first that a person could not be appointed as a technical representative by an insurance agent unless the IARB had confirmed his registration as such in its sub-register of technical representatives on application made by the insurance agent.[8] That upon the cessation of appointment, the insurance agent should inform the IARB of that fact within 7 days of such cessation[9] and upon such notification the IARB shall remove the technical representative from the sub-register relating to that insurance agent. 7. Secondly, a technical representative is required to undergo Continuing Professional Development (“CPD”).[10] Under the Guidance Note,[11] it was the responsibility of the insurance agent to collect evidence of compliance with the CPD requirements from its technical representatives and supply them to the IARB. Should such evidence not be provided in accordance with the Note, the technical representative would suffer a mandatory de-registration as well as being barred from registration for 3 months. The detailed provisions on compliance with the requirements of the CPD program in relation to technical representatives and the responsibilities of insurance agents who have appointed them are stated in clause 6(b) of the Guidance Notes as follows: “Responsibilities of insurance agents who have appointed TRs: (i) dispatch copies of Declaration Form[12] to all of their TRs; (ii) arrange sufficient CPD training for their TRs, if necessary; (iii) issue evidence/record of attendance to prove a TR who has attained a course sponsored by the insurance agent with the number of CPD credits/hours and type of credits printed on evidence/record of attendance; (iv) monitor the compliance of their TRs with the CPD requirements; (v) collect Declarations Forms from their TRs by 14 August; (vi) file the Annual Return to the IARB and report on those who fail to achieve the CPD credits/hours by 15 September; and (vii) offer help to their TRs in maintaining proof of compliance with CPD requirements.” 8. The consequence of non-compliance is stated in clause 7. For brevity, I call this, the duty to report CPD credits to IARB. 9. The appellant (“Mr Singh”) was appointed a technical representative by the respondent, Dah Sing Insurance Services Limited (“Dah Sing”), an insurance agent, by a contract dated 23 January 2007 (“the contract”).[13] The contract provided expressly that it should take effect from 1 January 2007. However, at trial it was agreed that the appointment only took effect from 19 January 2007 which was the date when Dah Sing obtained confirmation of the IARB.[14] Under the contract, his appointment could be terminated by a month’s notice. On 23 August 2007, his employment was duly terminated by notice. In his counterclaim against Dah Sing, he complained first that because Dah Sing had failed to inform the IARB of the cessation of his appointment, he could not work as a technical representative. In respect of this claim, he was awarded damages equivalent to one month’s loss of earning.[15] It was not disputed at trial and was the evidence of Dah Sing’s witness, that unless, the cessation of his appointment had been reported to IARB, he could not work for another insurance agent. His other complaint was that, although he had complied with the CPD requirements and supplied the necessary information to Dah Sing, Dah Sing had failed to report his CPD credits to IARB, as a result he suffered 3 months’ compulsory de-registration. He was awarded damages equivalent to 3 months’ income in respect of this claim.[16] 10. At trial Mr Singh relied on breach of contract, negligence and breach of statutory duty. However, counsel for Dah Sing at trial, who did not appear in the subsequent appeals, limited the dispute to whether as a matter of fact there was any breach.[17] It was not disputed in relation to the deregistration that if that was the result of Dah Sing’s fault, 3 months’ loss of income should be awarded. 11. On appeal, the District Judge’s judgment was reversed. The Court of Appeal took the view that there was no actionable breach of statutory duty and that Dah Sing owed Mr Singh no duty of care at common law. It also held that notwithstanding Dah Sing’s failure to report the cessation of his appointment as a technical representative, because Mr Singh could work as an insurance agent for more than one insurer, he had not suffered any loss. For this reason it also allowed the appeal in relation to this claim. 12. The Appeal Committee gave leave to appeal, first on: “whether the relationship between insurers and insurance agents of the one part and insurance agents [and] technical representatives of the other part can give rise to a duty of care owed by the former to the latter to comply with the mandatory requirements in the code of practice and guidance note, for the breach of which, a claim for damages may arise.”[18] 13. Leave was also granted on the “or otherwise” basis so that we could consider whether in the circumstances of this case, the Court of Appeal should have, on its own motion, refused to accept or abide by the concession made by the plaintiff that before the cessation of his appointment was reported he could not work for another insurance agent. Although Mr Singh also sought leave to argue that he had a cause of action for breach of statutory duties, leave was not granted. 14. In this appeal the focus of the argument of Ms Eu SC, for Mr Singh, was on tortious duty of care to which I now turn. Duty of Care 15. Ms Eu SC submitted that Dah Sing owed Mr Singh a duty of care to comply with the requirement of the Code to report the cessation of Mr Singh’s appointment within 7 days of the cessation as well as to report his CPD credits to the IARB. She relied in particular on these words of Lord Bridge of Harwich in Caparo Industries Plc v Dickman [1990] 2 AC 605 at 617 that: “… in addition to the foreseeablity of damage, necessary ingredients in any situation giving rise to a duty of care are that there should exist between the party owing the duty and the party to whom it is owed a relationship characterised by the law as one of ‘proximity’ or ‘neighbourhood’ and that the situation should be one in which the court considers it fair, just and reasonable that the law should impose a duty of a given scope upon the one party for the benefit of the other.” 16. Foreseeability of loss is straightforward. In the case of failure to report the cessation of appointment, it was accepted that Mr Singh could not work as a technical representative for another insurance agent until cessation of his appointment had been reported by Dah Sing.[19] In the case of failure to report the CPD credits, the penalty was mandatory and it was not disputed that he could not work during the 3 months when he was de-registered. 17. “Proximity” or “neighbourhood” are not the only relevant concepts in cases of omissions involving pure economic loss. The House of Lords in Henderson v Merrett Syndicates Ltd [1995] 2 AC 145 preferred the concept of “assumption of responsibility”. Lord Goff of Chieveley[20] said the concept of assumption of responsibility appeared, at 181: “… to have been adopted, in one form or another, by all of their Lordships in Hedley Byrne [1964] A.C. 465.” and that: “Furthermore, especially in a context concerned with a liability which may arise under a contract or in a situation ‘equivalent to contract,’ it must be expected that an objective test will be applied when asking the question whether, in a particular case, responsibility should be held to have been assumed by the defendant to the plaintiff: see Caparo Industries Plc v Dickman [1990] 2 A.C. 605, 637, per Lord Oliver of Aylmerton.” 18. But, whether the applicable concept is “proximity”, “neighbourhood” or “assumption of responsibility”, Lord Hoffmann explained in Customs and Excise Commissioners v Barclays Bank plc [2007] 1 AC 181 at 198 and 199: “the answer does not depend upon what the defendant intended but, as in the case of contractual liability, upon what would reasonably be inferred from his conduct against the background of all the circumstances of the case. The purpose of the inquiry is to establish whether there was, in relation to the loss in question, the necessary relationship (or ‘proximity’) between the parties and, as Lord Goff of Chieveley pointed out in Henderson v Merrett Syndicates Ltd [1995] 2 AC 145, 181, the existence of that relationship and the foreseeability of economic loss will make it unnecessary to undertake any further enquiry into whether it would be fair, just and reasonable to impose liability.” 19. Earlier at 190, Lord Bingham of Cornhill regarded an assumption of responsibility which is to be applied objectively: “as a sufficient but not a necessary condition of liability, a first test which, if answered positively, may obviate the need for further enquiry.” 20. In the present appeal, I prefer to approach the question of liability by considering whether, viewed objectively, Dah Sing could be said to have assumed the responsibility to report the cessation of his appointment or report his CPD credits. The answer depends on what could be inferred from the circumstances of this case. I have already mentioned the contract and said that although it expressly stated that its effective date was 1 January 2007, it did not become effective until 19 January 2007 when Dah Sing obtained confirmation from IARB of the registration. There were two other contractual documents between the parties, both dated 14 February 2007. Although they were dated later, there was no suggestion that they were not agreed contemporaneously with the contract. First, an offer letter signed by the parties under which Mr Singh was appointed “senior district manager” with effect from 1 February 2007 and which required that Mr Singh be registered with the HKFI as an agent[21] representing the company. Under the offer letter Mr Singh was paid an one off sign-on fee of $150,000. By a separate letter of understanding signed by Mr Singh and Dah Sing, he undertook to perform all his duties and to remain as a bona fide insurance agent[22] of the company for a minimum of 5 years and to repay the sign-on fee if his contract was terminated in the first 3 years.[23] Mr Singh’s counter claim was made in response to Dah Sing’s claim for a refund of the sign-on fee because his appointment was terminated in the first year. It is clear that the foundation of all these contractual documents was the registration of Mr Singh as a technical representative of Dah Sing which required the application by Dah Sing and his continued registration which depended on Dah Sing reporting his CPD credits to the IARB.[24] 21. Mr Tong SC relied on the fact that the contract of appointment expressly placed upon Mr Singh the responsibility to comply with the Code[25] whereas it was silent so far as Dah Sing was concerned. I would not read too much into this. As I have said,[26] in my view, Dah Sing was obliged by the Ordinance to comply with the provisions of the Code, that being the case, the requirement for compliance by Dah Sing could be taken for granted and no express provision was required. 22. In Henderson, Lord Goff of Chievely spoke of the importance of the possession of special knowledge or special skill in the person said to have assumed responsibility.[27] Here, we are not concerned with special knowledge or special skill, but Dah Sing was in an analogous position because of its role under the Code in reporting the cessation of appointment and reporting of CPD credits. Moreover, it was obvious that Mr Singh relied on Dah Sing to perform its role. The provisions set out in para 7 above outlined the many duties which Dah Sing as an insurance agent had to perform to enable its technical representatives to obtain the CPD credits including the supply of a copy of the necessary declaration which is essential for reporting the CPD credits. Also, the reporting could only be done by Dah Sing. It was obvious that Mr Singh would rely on Dah Sing. In these circumstances, I believe it is clear that Dah Sing must be taken to have assumed the responsibility to comply with these requirements. 23. The Code provided that the appointment of a technical representative could not take effect before registration by the IARB on the application of Dah Sing. I asked Mr Tong SC who appeared for Dah Sing whether it was his submission that, although under the Code Mr Singh could not begin work until his registration as a technical representative with the IARB upon application by Dah Sing had been confirmed, Dah Sing was not obliged to apply for registration. Mr Tong said it was. With respect, it seems to me clear that Dah Sing was under a duty to report the appointment of Mr Singh to IARB. The duty may be both contractual as an implied term of the contract to give it business efficacy as well as a tortious duty arising out of the relationship between the parties under which Dah Sing must be taken to have assumed such responsibility. As for the report of cessation of appointment and the CPD credits, given the clear foreseeability of loss to Mr Singh in the event of Dah Sing’s failure and Dah Sing’s assigned role under the Code, viewed objectively, Dah Sing must be taken to have assumed responsibility to do so. 24. Kwan JA in the Court of Appeal referred to and relied on what Ribeiro PJ said in Leung Tsang Hung v Incorporated Owners of Kwok Wing House (2007) 10 HKCFAR 480 at paras 49 to 53. But, with respect, what his Lordship said at para 54 is more apposite. His Lordship said[28] in cases where one has: “… actually done acts or entered into relationships or undertaken responsibilities which give rise to a common law duty of care. … The duty rests upon a solid, orthodox common law foundation and the question is not whether it is created by the statute but whether the terms of the statute (for example, in requiring a particular thing to be done or conferring a discretion) are sufficient to exclude it. The law in this respect has been well established since Geddis v Proprietors of the Bann Reservoir (1878) App Cas 430.” 25. Given the assumption of responsibility, the foreseeability of loss and Mr Singh’s reliance on Dah Sing to report the cessation of his appointment and his CPD credits in accordance with Dah Sing’s assigned role under the Code, I believe one could also conclude that it would be fair, just and reasonable to impose a duty of care on Dah Sing. 26. Mr Tong also relied on some of the provisions of the contract of appointment to show that Dah Sing had not assumed any contractual responsibility to report termination of appointment. Assumption of responsibility which would otherwise give rise to tortious liability may be excluded by contract if the latter is inconsistent with it.[29] 27. First, clause 13.4.6, which provided that: “Upon termination of this Agreement and subject to the provisions of Clause 13.5: ‘the Company, shall, if termination is for a cause relating to a breach of the Code or relevant Ordinance, notify the IARB of the relevant particulars and notify the Technical Representative that it has done so.’” 28. And clause 13.7 of the contract which provided that Dah Sing: “… shall not be liable to pay any compensation to the Technical Representative for any loss or damages howsoever arising from or in connection with the termination, for whatever reason, of this Agreement.” 29. Clause 13.7 does not help, the losses complained of flowed not from “the termination, for whatever reason” but failure to report cessation of his appointment or report of CPD credits. Nor would clause 13.4.6 help Mr Tong. Clause 13.4.6 expressly required Dah Sing to notify IARB and the technical representative of the termination of the appointment for a cause related to a breach of the Code or relevant Ordinance. We are not concerned with notification of any breach and I cannot read clause 13.4.6 as being inconsistent with a tortious liability to report cessation of appointment which is unrelated to any breach at all. 30. Mr Tong also sought to argue that Mr Singh had suffered no loss as a result of the deregistration because it was said that Mr Singh was not aware of the deregistration until after it was over. This is a point which had not been taken in any of the courts below nor had leave been given by us to raise the point. We would not entertain it. Or otherwise 31. We also gave leave on the or otherwise basis.[30] It is here that I turn to an important difference between an insurance agent and a technical representative which I can deal with briefly. Under clause 19 of the Code, an insurance agent could work for 4 principals provided he has the consent of the principal for whom he acts as an insurance agent.[31] However, in the case of a technical representative, clause 33 provided that he could not act as a technical representative for more than one insurance agent. At trial there was no dispute that unless the cessation of appointment was reported by Dah Sing, Mr Singh could not work for another insurance agent and it was Mr Singh’s claim that he suffered loss of income as a result. It was never Dah Sing’s case that even if they were in breach Mr Singh could have worked. It was Mr Singh’s clear evidence that he couldn’t work. His evidence was not contradicted. Indeed, Ms Kan Chuy Hant who gave evidence for Dah Sing, said during cross-examination: “Q: Right, so the code de-registration[32] is for enabling the agent[33] concerned to work in another company, right? If the code is not de-registered, he will not be able to work in another company, do you agree? A: Agree” 32. In the teeth of such evidence, the Court of Appeal of its own initiative,[34] referred to Mr Singh’s “Insurance Agents Registration History” and found against Mr Singh on this part of his claim. Kwan JA said: “78. This was not a point taken by the plaintiff at the trial. Mr Li [leading counsel for Mr Singh] has referred us to the evidence of the plaintiff’s witness who agreed in cross-examination with the suggestion that if an agent was registered with one insurer, he could not be appointed as agent by another insurer. But this concession of the witness was plainly incorrect. Under clause 19 of the Code, an agent could work for not more than four principals simultaneously, of whom no more than two shall be long term insurers. This was borne out by the Insurance Agents Registration History of the defendant, there were two periods (from July 1995 to December 1996, and from December 1996 to May 1998) when he was registered as an agent of two insurers at the same time. There were other instances when his appointment as agent was subsisting with more than one insurer. 79. On this evidence which cannot be refuted, the defendant had failed to establish that any loss of income was caused by the failure of the plaintiff to report his termination of employment. So for this reason as well, the award of damages of one month’s loss of income should also be set aside.” 33. I have quoted at length from the judgment because, with respect, I find these paragraphs difficult. Mr Singh was a technical representative and under clause 33 he could not act as a technical representative for more than one insurance agent. The Code is silent on whether whilst still registered as a technical representative, he could work as an insurance agent. I am unable to tell whether it was Kwan JA’s view that Mr Singh could at the same time be both a technical representative and an insurance agent. But the fact that in the past he had been insurance agent for 2 insurers was not to the point. At para 7 of her judgment, she said Mr Singh was registered as a “technical representative of insurance agency” of Dah Sing but at para 5 she said “He was appointed as the plaintiff’s insurance agent …” which is wrong. I believe her Ladyship might have been misled by the fact that in his defence and counterclaim, Mr Singh said he entered into an “agent contract”. Furthermore, Ms Eu told us it was only after the receipt of Court of Appeal’s judgment that Mr Singh’s lawyers realized that the court regarded the fact that he had once worked for more than one insurer significant. 34. With respect, I do not believe the Court of Appeal was at liberty to take a new point on appeal against the clear evidence at trial. The parties could be expected to know, as a matter of common understanding within the profession and practical reality, whether Mr Singh could work before the cessation of his appointment by Dah Sing had been reported. His clear and uncontradicted evidence was that he could not. Moreover Mr Singh in his second affirmation has set out the prima facie relevant evidence which he could have adduced had the point been raised at first instance. In these circumstances, I would also allow Mr Singh’s appeal on this point. Mr Justice Fok PJ: 35. I agree with the Judgment of Mr Justice Tang PJ. Mr Justice Stock NPJ: 36. I agree with the Judgment of Mr Justice Tang PJ. Mr Justice Spigelman NPJ: 37. I agree with the Judgment of Mr Justice Tang PJ. Mr Justice Ribeiro PJ: 38. The appeal is unanimously allowed. We make an order nisi for costs to be paid by the respondent to the appellant here and below, with liberty to the parties, if so advised, to lodge written submissions on costs within 14 days of the date of the handing down of this judgment. In default of such submissions, the order nisi is to stand as an order absolute without further direction. Mr Ronny Tong SC, Mr Norman Nip and Ms Prisca Cheung, instructed by Keith Lam, Lau & Chan, for the plaintiff (respondent) Ms Audrey Eu SC and Mr Kelvin Leung, instructed by T C Lau & Co, assigned by the Director of Legal Aid, for the defendant (appellant) [1] Explanatory Memorandum to the Insurance Companies (Amendment)(No 3) Bill 1993. [2] s 67(1). [3] Para 67(4). [4] s 67(5), s 77(6)(a). [5] Defined in the Ordinance to mean “a person who holds himself out to advise on or arrange contracts of insurance in or from Hong Kong as an agent or sub-agent of one or more insurers”. [6] But there are also important differences, for example, an insurer cannot exclude or limit its liability for the actions of its appointed insurance agents, s 68(2). This is not the case with technical representatives and this and other measures meant to protect policy holders or potential policy holders can be circumvented by an insurer appointing an “inhouse” insurance agent. [7] The Court of Appeal in its judgment mainly referred to provisions in the Code which concern insurance agents. This, as will be seen, has given rise to some confusion, because there is one important difference in the Code concerning them, namely, under clause 33 a person cannot act as a technical representative for more than one insurance agent but an insurance agent could work for as many as 4 insurers. See para 31 below. [8] Clauses 26 and 27. [9] Clause 31. [10] Clauses 65 and 67. [11] The Guidance Note on compliance with CPD Programme issued on 30 November 2004 and revised on 13 June 2005. [12] Under clause 6(a)(iii), a technical representative has to complete this as part of the reporting process. [13] In his defence and counterclaim, he called this “an agent contract”. Be that as it may, the learned trial judge was aware that he was a technical representative. See para 1 of his judgment. However, in the Court of Appeal, although at para 7 of Kwan JA’s judgment, it said Mr Singh was “registered as ‘a technical representative of insurance agency’ of [Dah Sing]”. At para 5, it said that “He was appointed as [Dah Sing’s] insurance agent …”. This confusion may be important when we come to deal with the appeal on the “or otherwise” ground. [14] Clause 26 provided that an insurance agent shall obtain IARB’s confirmation before confirming the appointment of a technical representative. [15] Para 91, DC judgment. [16] Para 92, DC judgment. [17] Para 50, DC judgment. [18] Determination, 4 August 2015. [19] I will deal with the Court of Appeal’s contrary view below at para 32. [20] With the concurrence of all their Lordships. [21] In context as a technical representative and not an insurance agent or sub-agent. But in both Mr Singh and the evidence of Ms Kan Chuy Hant, the only witness of Dah Sing, he was referred to as “agent”, a term used, no doubt, colloquially. [22] See para 4 but Mr Singh was not an insurance agent. Dah Sing as an insurance agent could not appoint an insurance agent, only an insurer could appoint an insurance agent. [23] 100% in the event of termination during the first year, 66.66% the second year and 33.34% the third year. [24] See para 7 above. [25] Clause 6. See para 7 above. [26] See para 3 above. [27] Henderson at 180. [28] Quoting from Gorringe v Calderdale Metropolitan Borough Council [2004] 1 WLR 1057 at 1068. [29] Henderson at 193C. [30] See footnote 7 above. [31] Clause 21. [32] This was the expression used to refer to the report of cessation of appointment. [33] I believe this was also used colloquially and not a reference to “insurance agent” as defined in the Ordinance or the Code. [34] See para 11 of the court’s decision refusing leave to appeal. The point was not mentioned in the skeleton submissions before the Court of Appeal. 
The Court: 1. At the conclusion of the oral argument the Court ordered that the appeal be dismissed. What follows are our reasons for supporting that order. The Trial 2. Part XII (ss 117-159) of the Crimes Ordinance (Cap 200) is headed “Sexual and Related Offences”. Section 118(1) states: “A man who rapes a woman shall be guilty of an offence and shall be liable on conviction on indictment to imprisonment for life.” Section 4B of the Evidence Ordnance (Cap 8) abrogates any requirement for the judge at a trial of an offence under Part XII to give the jury a warning about convicting the accused on the uncorroborated evidence of the complainant. 3. The appellant stood trial in the High Court before Barnes J and a jury of 7 on five counts of rape of a girl (X), born in 2003. The particulars of the offences specified rental premises in the New Territories (“Room D”) of which the appellant was the landlord. The appellant contended that the whole case against him was fabricated, that X’s mother (Y) was his “kept woman”, that he had been supporting Y and X, and that Y had abetted X to accuse him falsely of rape with a view to extorting further money from him. 4. For counts 1 and 2 the particulars included the phrase “on an unknown date in March 2014”, count 3 “an unknown date between the 1st day of April 2014 and 31st day of May 2014, both dates inclusive” and for count 4 “an unknown date in May 2014”. Count 5 was unique in identifying a specific date, 4 June 2014. At the trial on 22, 23 June 2016 the appellant was convicted on counts 1, 2, 3 and 4 by a majority verdict of five to two. The appellant was sentenced to imprisonment for a total of 10 years. He was, significantly, unanimously acquitted of count 5. 5. Barnes J had given her directions to the jury in Chinese and references in these reasons are to the English translation. She observed in her directions to the jury that “the evidence is most basically the oral evidence of only two witnesses…just X and the Defendant”. With respect to count 5 the appellant had given evidence of a specific alibi. He contended that at the time of the offence charged in count 5, around 1 pm on 4 June 2014, he could not have been in Room D. This was because at that time he was negotiating with one Madam Lee for the renting of another room, he signed a lease with her, and, during this period, he had 11 records of phone conversations with other people. Appeal to the Court of Appeal 6. Section 82(2)(b) of the Criminal Procedure Ordinance (Cap 221) (“the Procedure Ordinance”) provides for an appeal, with the leave of the Court of Appeal on, inter alia, any ground which appears to the Court “to be a sufficient ground of appeal”. An application by the appellant in respect of his four convictions was dismissed by the Court of Appeal (Yeung VP, Poon JA, Chan J) in a comprehensive judgment, dated 8 December 2017 and extending to 80 paragraphs. 7. Section 83(1)(a) of the Procedure Ordinance provides for a conviction to be set aside “on the ground that under all the circumstances of the case [the conviction] is unsafe or unsatisfactory”. The Court of Appeal refused to grant leave to appeal because it did not find anything unsafe or unsatisfactory in the convictions of the appellant. 8. In particular, the Court of Appeal rejected the appellant’s submissions that because the jury found him not guilty on count 5 this rendered his conviction on the counts 1-4 unsafe and unsatisfactory. Appeal to this Court 9. On the application to this Court for leave particular reliance was placed upon what appeared to be a direction required by R v Markuleski[1], and upon the outcome in Jones v The Queen[2] where, the High Court of Australia held that, given the acquittal on one count of sexual intercourse with a female child, it was not open to the jury on the whole of the evidence to be satisfied beyond reasonable doubt of guilt on the other two counts. 10. On 15 November 2018, this Court granted leave to appeal from the Court of Appeal on two questions of law: (1) Although the standard direction to the jury is that they should consider each count in the indictment separately and may accept part of a witness’s evidence and not accept other parts of the same witness’s evidence, should that direction be qualified in cases of sexual offences where the only direct evidence of commission of the offences is that of the complainant and, if so, how? (2) What is the proper approach of an appellate court to inconsistent verdicts in cases of sexual offences where, although not logically inconsistent, they are based on the uncorroborated evidence of the complainant? 11. This Court also granted leave “on the basis that it is reasonably arguable that the different verdicts of the jury on the indictment against the [appellant] resulted in substantial and grave injustice being done to him”. 12. On his appeal, the appellant sought an order quashing his conviction on counts 1-4, having regard to what he submitted was the correct approach to the impact of acquittal on count 5. Question 1 - Markuleski Direction 13. With respect to Question 1, in his Written Case the appellant relied in particular on what was said by Spigelman CJ in R v Markuleski[3]. The passages in question read: “184. … I express considerable reluctance to add to the number of directions and warnings which trial judges are already required to give. McHugh J noted in KRM v The Queen [(2001) 206 CLR 221 at 37]: ‘The more directions and warnings juries are given the more likely it is that they will forget or misinterpret some directions or warnings.’ 185. Nevertheless, the case law on inconsistent verdicts, particularly in the context of sexual assault cases but not limited to that context, indicates that there is a recurring difficulty in this respect. It may appear to be obvious that a reasonable doubt about one aspect of a complainant's evidence ought to be taken into account when assessing that witness's evidence on other matters. However, there have been a significant number of cases in which courts of criminal appeal have acted on the basis that the jury may have failed to do so. 186. In the light of the number of cases it is desirable that the traditional direction as to treating each count separately is supplemented in a word against word case. Some reference ought to be made to the effect upon the assessment of the credibility of a complainant if the jury finds itself unable to accept the complainant's evidence with respect to any count. … 187-190 … 191. The precise terminology must remain a matter for the trial judge in all the particular circumstances of the specific case. The crucial matter is to indicate to the jury that any doubt they may form with respect to one aspect of the complainant's evidence, ought be considered by them when assessing the overall credibility of the complainant and, therefore, when deciding whether or not there was a reasonable doubt about the complainant's evidence with respect to other counts.” (italics added) 14. In his judgment in KRM McHugh J (with the other members of the Court) rejected the submissions that on the trial of multiple counts of sexual offences against the same complainant the judge was obliged to direct the jury that if they found the accused guilty on one or more counts they could not use that finding to reason that the accused had a propensity making him likely to commit the offences charged in the other counts. While Markuleski is not immediately directed to a propensity warning, McHugh J’s following observations are in point. He said[4]: “It has become the standard practice in cases where there are multiple counts, however, for the judge to direct the jury that they must consider each count separately and to consider it only by reference to the evidence that applies to it (a ‘separate consideration warning’). The universal giving of a separate consideration warning and the omission of a universal propensity warning indicates that the giving of a separate consideration warning is ordinarily sufficient to avoid miscarriages of justice in cases such as the present. This indication is confirmed by the many cases where juries acquit accused persons of some charges and convict them of others where the presentment contains multiple counts involving the same or similar offences.” 15. The nature and effect of the “separate consideration warning” will be considered below when dealing with Question 2. 16. The italicised passage from Markuleski set out above at para 13 above may suggest that the supplementary direction is mandatory. That is not how the matter has been considered in subsequent appellate decisions in Queensland, New South Wales, Victoria and Western Australia which were collected and discussed by Keane JA (as he then was) in R v Ford[5]. Rather, as his Honour pointed out, it is not the position that “the Markuleski direction is always necessary, or even desirable, as a counterweight to the ‘separate offences’ direction”; it must first be clear that the risk of unfairness to the accused “has truly arisen”. The New South Wales Court of Criminal Appeal in DF v The Queen[6] recently referred with approval to the judgment of Keane JA. 17. It may be noted that in Ford[7] Keane JA had approved a statement by Buchanan JA in R v PMT[8] (with citations omitted): “I think it unlikely that a jury given a separate consideration direction will be entirely uninfluenced by the impressions they derive from the evidence of a witness taken as a whole; I doubt that such a natural tendency needs judicial encouragement in the form of a Markuleski direction. Further, I am of the opinion that the proposed direction is likely to promote propensity reasoning and produce confusion rather than assist a jury to properly evaluate the evidence. In my view, in this case it was well within the ability of the jury to assess the evidence of the complainant in the light of their own experience and with the benefit of the addresses of counsel, without the necessity of the warning advocated by counsel for the applicant.” 18. Further, in England the Court of Appeal in R v Fanning[9], with reference to MacKenzie v The Queen[10], reaffirmed that the constitutional position of the jury includes an ameliorative aspect which allows for a “merciful” view of the facts on one or more of a plurality of counts. 19. We are of the view that Ford and Fanning put Markuleski properly in context. 20. Several points are to be made respecting what on this appeal was the appellant’s approach to Markuleski. First, correctly, the appellant did not contend that the extended direction is necessarily required in word against word cases. Rather, it was submitted in the Written Case (para 20) that: “[The] approach suggested in Markuleski allows the trial judge to tailor the direction to suit the individual circumstances of each particular case, and if necessary, to give strong direction to the jury that if they acquit on one count, it is very difficult for them to convict on other counts. This allows the judge to give suitably strong direction to the jury to acquit without usurping the function of the jury, while avoiding the danger of asking the jury to use the guilt in one count to add to the propensity of the defendant in committing the offences in the other counts.” (italics added) 21. Secondly, at the trial the jury sought assistance with several questions and counsel for the appellant saw this as indicating that the jury was not sure as to on how many of the charges they were inclined to return a guilty verdict. Counsel sought from the trial judge a direction that: “If they have a reasonable doubt about the girl’s evidence on any of the rape charges, they must take that reasonable doubt into consideration when [they] consider the other charges.” Her Ladyship responded with a proposed direction and counsel reacted: “Right, [they] have to consider whether the evidence that makes them cast doubt on the girl would affect her evidence in respect of the other charges.” Her Ladyship then gave a direction with particular reference to count 5: “Whether the situation of the fifth count is slightly different from those of the first, second, third and fourth, you people will have to consider whether it will affect the evidence in the case; [whether] it might not necessarily affect the first, second, third and fourth. This is for you to decide. But what is the most important thing: Most importantly, what [you] have to bear in mind is that there is no other evidence to support [the version of] this witness.” 22. Thirdly, in the appellant’s Written Case counsel contends that the direction did not go far enough. Paragraphs 27 and 28 state: “27. It is respectfully submitted that the learned judge should have directed the jury that: - (i) They may think that there is nothing to distinguish the evidence of [X] on the 5th count and the other counts. (ii) They may think that the Appellant’s alibi evidence is so credible that it is difficult to convict on the 5th count. (iii) Because [X] did not give specific dates of the offences in counts 1-4, it is impossible for the Appellant to call alibi evidence on counts 1-4, which is a serious disadvantage to the defence. (iv) If they find a reasonable doubt in [X’s] evidence in relation to count 5 they must take that into consideration of X’s credibility in relation to counts 1-4, and they may believe that it is difficult to see how [X’s] evidence could be accepted in relation to counts 1-4. (v) If they find the Appellant not guilty on the 5th count, they may consider it very difficult to find him guilty on any of the other counts. 28. In the circumstances, because of the inadequate direction to the jury as indicated above, the Appellant has not been given a fair trial and therefore the conviction should be quashed.” 23. At the trial counsel had not sought to invite the trial judge to expand the direction she did give by a further detailed direction as set out above. 24. Given these circumstances is there substance in the above submission that the appellant was denied a fair trial? The answer must be in the negative. 25. Question 1 asks whether the “separate offences” direction should be qualified in cases of sexual offences where the only direct evidence of commission of the offences is that of the complainant, and, if so, how? The qualification, if necessitated by the state of the evidence, would be a Markuleski direction, usually in terms to the effect of the directions sought and given in the present case. 26. The necessity for such a direction may be urged by counsel, or, the issue may be raised by the trial judge for consideration and submissions by counsel. The direction is not lightly to be given and the use of the term “necessary” indicates the overarching criterion of the conduct of a fair trial. Whether the state of the evidence raises that necessity is determinable only on the specific circumstances. The approach of the trial judge in this case provides a positive example of the necessity. The judge gave the direction to the jury that was requested of her. 27. The terms in which the direction is given will be indicated by the state of evidence in the case. Whether in this case, if then sought, it would have been appropriate for the trial judge to give a direction in the detailed terms now sought by the appellant’s counsel is a question upon which, at this distance from the trial, this Court should not enter. It was not obvious that such a direction should have been given. Question 2 – Inconsistent Verdicts 28. We turn now to Question 2 and inconsistent verdicts in sexual cases. 29. In MacKenzie v The Queen[11] the High Court of Australia observed: “A distinction must be drawn between cases of legal or technical inconsistency and cases of suggested factual inconsistency. The former will generally be easier to resolve. On the face of the court's record there will be two verdicts which, in law, cannot stand together. Examples include the case where the accused was convicted both of an attempt to commit an offence and the completed offence[12] or of being, in respect of the same property and occasion, both the thief and the receiver[13] . There are other like cases[14]. Where technical or legal inconsistency is established, it must be inferred that the jury misunderstood the judge's directions on the law; compromised disputes amongst themselves; or otherwise fell into an unidentifiable error. The impugned verdict or verdicts must be set aside and appropriate consequential orders made.” 30. The issue on this appeal concerns “factual inconsistency” which is said to arise between different verdicts against the same accused. 31. In R v Fanning[15] the English Court of Appeal reaffirmed, with respect to an argument on an appeal that the jury had returned inconsistent verdicts, the requirement associated with remarks of Devlin J (as he then was) in R v Stone[16] and their adoption by the Court of Appeal in R v Durante[17]. The criterion is that no reasonable jury which had applied their minds properly to the facts of the case could have returned the verdict called into question. In Fanning the Court of Appeal rejected those English decisions given after Stone which indicated that verdicts could be treated as inconsistent because the jury was satisfied about some parts of the evidence given by a witness but unable to be sure to the requisite standard as to other parts of that evidence.[18] 32. In Hong Kong Durante was adopted in R v Cheng Man To[19], as explained by the Court of Appeal in HKSAR v Wu Wing Hong[20]. It follows from the disposition of the present appeal that this remains the position in Hong Kong and Question 2 is to be answered accordingly. 33. In Cheng Man To[21] the appellant was convicted on two counts of incest with his daughter and acquitted on six counts in the one indictment. It appeared to the Court of Appeal that “being very careful” the jury had not been prepared to convict on charges not relating to specific dates and had convicted only where the complainant alleged specific dates. However, cases will arise where different verdicts on counts in sexual offences cases are held to be irreconcilable and “an affront to logic and common sense”.[22] 34. In Australia, the decision in Jones v The Queen[23] was read by the New South Wales Court of Criminal Appeal in R v RAT[24] as adding to the “separate offences” direction referred to above at para 14 in cases of multiple counts of sexual assault where the only direct evidence was that of the complainant. In RAT the Court held that in such cases, if the jury is not satisfied to the criminal standard upon one or more counts, and there is no independent evidence respecting the other counts, they must acquit on all counts. However, subsequently in MFA v The Queen[25] the High Court held that it was erroneous to treat Jones as authority for the proposition that not guilty verdicts on some counts necessarily support a conclusion that the complainant was untruthful and unreliable and so was of damaged creditability. Jones was to be treated as deciding that on its facts the necessary justification in logic and reasonableness was missing. 35. In MFA[26] Gleeson CJ, Hayne and Callinan JJ emphasised of a verdict of not guilty in a case of sexual offences: “It may simply reflect a cautious approach to the discharge of a heavy responsibility. In addition to want of supporting evidence, other factors that might cause a jury to draw back from reaching a conclusion beyond reasonable doubt in relation to some aspects of a complainant's evidence might be that the complainant has shown some uncertainty as to matters of detail, or has been shown to have a faulty recollection of some matters, or has been shown otherwise to be more reliable about some parts of his or her evidence than about others.” 36. On the hearing of the present appeal it became clear that counsel for the appellant accepted Stone and Durante, and founded the appellant’s case on the submission that, given the acquittal on count 5, no reasonable jury could have returned a guilty verdict on counts 1-4. 37. The appellant pointed to several matters which together were said to provide a sufficient basis for that conclusion, but the respondent cogently countered these submissions. First, the absence of corroboration of the evidence of the complainant was a feature of many sexual offence cases and while a relevant factor could not of itself be decisive. Secondly, there had been no undue delay, the alleged offences being committed in 2014, the indictment preferred on 22 October 2015 and the evidence of X at the trial being given in June 2016. Thirdly, there had been an apparent difference in the quality of X’s evidence; counts 1-4 were included in the letter written by X to herself, referring to events of April-May, but count 5, placed in June was not mentioned; count 5 was pleaded with reference to a specific date, which had been contradicted by the appellant’s asserted alibi; the alibi and the omission from the letter were considered in the summing up and should be taken to have been considered by the jury. 38. The acquittal on count 5 may reflect the cautious approach described in the passage from MFA quoted above in para 35. Viewed overall the differing verdicts in the present case were not inconsistent such as to necessitate intervention on appeal. Substantial And Grave Injustice Not Established 39. There being no departure from any norm of substantive or procedural law, it follows that the appellant failed to establish that his conviction resulted in substantial and grave injustice being done to him. Mr Wong Man Kit SC and Mr Joe Chan, instructed by Foo & Li, for the Appellant Mr Martin Hui SC, Consultant Counsel and Mr Ivan Cheung, SPP, of the Department of Justice, for the Respondent [1] (2001) 52 NSWLR 82 at 121-122. [2] (1997) 191 CLR 439. [3] (2001) 52 NSWLR 82 at 121-122. [4] (2001) 206 CLR 221 at [36]. [5] [2006] QCA 142 at [123]–[126]. [6] (2012) 222 A Crim R 178 at [28]. [7] [2006] QCA 142 at [126]. [8] (2003) 8 VR 50 at 59. [9] [2016] 1 WLR 4175 at [11], [16], followed in R v Fletcher [2017] EWCA Crim 1778 at [26]; R v Electricity North West Ltd [2018] 4 WLR 148 at [41]. [10] (1996) 190 CLR 348 at 366-368. [11] (1996) 190 CLR 348 at 366. [12] R v Roach [1948] NZLR 677. [13] R v Irvine [1976] 1 NZLR 96. [14] See, eg, R v Mitchell [1971] VR 46 at 51; Connecticut v Hinton (1993) 630 A 2d 593. [15] [2016] 1 WLR 4175. [16] [1955] Crim LR 120. [17] [1972] 1 WLR 1612 at 1617. [18] [2016] 1 WLR 4175 at 4188. [19] [1987] 2 HKC 261 at 265. [20] [1998] 1 HKLRD 508 at 511. [21] [1987] 2 HKC 261 at 267-268. [22] For example, BG v Western Australia (2005) 152 A Crim R 207 at [53]-[54]; R v Bond [2018] QCA 130 at [100]-[105]. [23] (1997) 191 CLR 439. [24] (2000) 111 A Crim R 360 at 371. [25] (2002) 213 CLR 606 at [35], [89], [97]. [26] (2002) 213 CLR 606 at [34]. See also MC v R [2017] NSWCCA 274 at [83]-[92] per Hoeben CJ at CL; Cabot (pseudonym) v R [2018] NSWCC A 265 at [54]-[58] per Leeming JA. 
Mr Justice Ribeiro PJ: 1. I agree with the judgment of Lord Neuberger of Abbotsbury NPJ. Mr Justice Tang PJ: 2. I agree with the judgment of Lord Neuberger of Abbotsbury NPJ. Mr Justice Fok PJ: 3. I agree with the judgment of Lord Neuberger of Abbotsbury NPJ. Mr Justice Bokhary NPJ: 4. I agree with the judgment of Lord Neuberger of Abbotsbury NPJ. Lord Neuberger of Abbotsbury NPJ: Introductory 5. This is an appeal brought by Moody’s Investors Service Hong Kong Limited (“Moody’s”) against a decision of the Court of Appeal (Hon Lam VP, and Yuen and Kwan JJA) given on 8 June 2017, [2017] 3 HKLRD 565. At the end of the hearing, we indicated that we would dismiss the appeal, and give our reasons later. These are our reasons. 6. By their decision, the Court of Appeal substantially upheld a determination of the Securities and Futures Appeals Tribunal (“the Tribunal”, Hon Hartmann NPJ, Chairman, and Dr Billy Mak Sui-choi and Ms Ding Chen, Members), given on 31 March 2016. In that determination, the Tribunal upheld in part a Decision Notice (“the Notice”) issued by Securities and Futures Commission (“the Commission”) dated 3 November 2014. 7. The Notice stated that, pursuant to section 194 of the Securities and Futures Ordinance Cap 571 (“the Ordinance”), Moody’s had failed to meet the standards, or to comply with the practices, expected of a corporation licensed under the Ordinance, in its preparation and publication of a report published on 11 July 2011 (“the Report”). After Moody’s had made representations as to the appropriate sanction, the Commission went on to decide to impose a public reprimand and a financial penalty of HK$23 million. 8. On Moody’s application for a review, the Tribunal upheld some, but not all, of the criticisms in the Notice, and, while maintaining the public reprimand, it reduced the financial penalty to HK$11 million. Although the Court of Appeal disagreed with the Tribunal on one aspect, it made no difference to the outcome, and so Moody’s appeal was dismissed. Moody’s now appeals to the Court of Final Appeal. The regulatory background 9. Moody’s is the Hong Kong arm of the eponymous global credit rating agency network, which trades under the names Moody’s Investors Services (“MIS”) and Moody’s Analytics. The network’s website claims that its services are “an essential component of the global capital markets, providing credit ratings, research, tools and analysis that contribute to transparent and integrated financial markets”. The website also states that MIS is “a leading provider of credit ratings, research, and risk analysis”, whose ratings cover the great majority of sovereign nations, and tens of thousands of entities which issue bonds. 10. The influence of credit ratings agencies (“CRAs”), and in particular the largest two or three CRAs (which on any view include MIS), on global capital markets became apparent following the worldwide financial crisis of 2007-2008. In a nutshell, it was thought that a large proportion of investors and traders in government, corporate and other debt and debt-related instruments placed great weight on the ratings issued by the CRAs. As a result, in many jurisdictions where the CRAs were not subject to specific regulatory control, they were brought within the regulatory system. Hong Kong was one such jurisdiction, and CRAs were specifically brought within the ambit of the Ordinance in 2011. 11. The Ordinance in its present form contains provisions for the licensing, regulating and disciplining of any person who wishes to carry on specified “regulated activities” in financial markets. Part 1 of Schedule 5 currently sets out twelve types of such activities. Type 10 is “providing credit rating services”. 12. Part 2 of the same Schedule contains two relevant definitions. The first is “credit ratings” which are defined as: “opinions, expressed using a defined ranking system, primarily regarding the creditworthiness of – (a) a person other than an individual; (b) debt securities ….”. The second relevant definition is “providing credit rating services”, which is as follows: “(a) preparing credit ratings- (i) for dissemination to the public, whether in Hong Kong or elsewhere; … or … (b) preparing credit ratings- (i) for distribution by subscription, whether in Hong Kong or elsewhere; … … but does not include – (c) preparing, pursuant to a request made by a person, a credit rating which is exclusively prepared for, and provided to, the person, and that is neither intended for dissemination to the public or distribution by subscription …”. 13. Part IX of the Ordinance is concerned with “Discipline, etc”. The first section of this Part, section 193, includes in subsection (1) a definition of “misconduct”, which has five paragraphs. Paragraphs (a) to (c) cover contraventions of provisions of the Ordinance, of provisions of any relevant licence or registration, or of any other condition. Paragraph (d) – “section 193(1)(d)” - is in these terms: “an act or omission relating to the carrying on of any regulated activity for which a person is licensed or registered which, in the opinion of the Commission, is or is likely to be prejudicial to the interest of the investing public or to the public interest”. Section 194 is concerned with “Disciplinary action [by the Commission] in respect of licensed persons, etc”, and it sets out the powers of the Commission when, inter alia, it concludes that a regulated person “is, or was at any time, guilty of misconduct”. 14. Three other sections of the Ordinance should be mentioned. Section 169(1) empowers the SFC to publish codes of conduct: “for the purpose of giving guidance relating to the practices and standards with which intermediaries ... are ordinarily expected to comply in carrying on the regulated activities for which the intermediaries are licensed”. Section 217 provides that a person who has been found to be guilty of misconduct by the Commission can apply to the Tribunal for a review, and section 229 enables a person who wishes to challenge a decision of the Tribunal on a point of law to appeal to the Court of Appeal. The relevant substantive facts 15. Prior to 1 June 2011, when the regime regulating CRAs came into force, Moody’s operated in Hong Kong, as it was entitled to do, as an unlicensed credit ratings agency. In that capacity, Moody’s prepared and circulated a number of reports which were undoubtedly credit ratings, and, of course, it continues to carry out those activities to this day. Then, as now, Moody’s reports covered a large number of different types of companies and other institutions. Those companies included a total of 61 high-yield non-financial Chinese companies (“the 61 Companies”). Inevitably, its rating assessment of any of those companies would vary from time to time as the information about a company, and the environment in which it operated, changed and more information came to light. 16. On 1 June 2011, when the regime regulating CRAs came into force, Moody’s became licensed under the Ordinance to carry on Type 10 regulated activity. 17. Some six weeks later, Moody’s issued the Report, the subject of these proceedings. The Report described itself as a “Special Comment Report”, and was entitled “Red Flags for Emerging-Market Companies: A Focus on China”. It ran to 25 pages, and began with a section headed “Overview”, which explained that: “Red flags as a screen. In rapidly developing emerging markets, the use of frameworks to assess elements of credit risk provides consistency in identifying relative strengths and weaknesses across a growing pool of rated issuers. In this report, we look at 20 red flags, grouped into five categories, that highlight issues meriting scrutiny to identify possible governance or accounting risks for non-financial corporate issuers in emerging markets”. The five categories of red flags were then identified, and in summary terms they were (i) weakness in corporate governance, (ii) riskier and more opaque business models, (iii) fast-growing businesses strategies, (iv) poor quality of earnings or cash flow, and (v) concerns over auditors and quality of financial statements. The Overview then continued by explaining that in “this inaugural report”, the framework is applied to “61 rated Chinese entities”, viz. the 61 Companies. 18. The Overview then explained that “for non-property firms” there was “a degree of correlation” between previously issued ratings and the number of red flags, but that this did not appear so true for property companies. The Overview then stated that “[t]hese findings show that screens for governance or accounting risks can help identify areas to investigate but cannot serve as mechanisms to rank order credit risk.” The Overview also pointed out that Moody’s “ratings already factor in the inherent challenges of analyzing young, fast-growing Chinese companies”, but added that “[t]he red flags provide further clarity and detail, but do not represent a change in our rating methodologies”. 19. The next section of the Report was headed “Red Flags An Interesting Screen”, and it began by referring to recent public expressions of concern about Chinese companies’ financial reporting. The Report then stated that in order “[t]o address investors’ concerns and provide transparency on our approach to ratings, this report identifies warning signs – so-called ‘red flags’ – for our rated, high-yield …, non-financial Chinese companies”. The point was then made that “[t]he identified issues that we flag do not represent a change to our analytical approach”. It was then stated that “[o]ur ratings already account for the inherent challenges in assessing Chinese companies”. 20. The following section of the Report was headed “Framework Focuses on Five Key Categories”, namely those very briefly summarised in [17] above. After discussing the numbers in very general terms, the Report set out in Figure 1 the three “rating categor[ies]” Ba, B, and Caa, of 49 of the 61 Companies (“the 49 Companies”), broken down into “Chinese property” and “Chinese non-property” companies. Figure 2 set out the “No. of red flags tripped by rating category”, and showed the “average no. of red flags tripped” against each rating categories of “Investment Grade”, Ba, B, and Caa. After some further discussion, the Report stated that a company’s “tripping of many red flags does not represent an immediate rating concern”. The Report then turned to discuss six of the 49 Companies individually. 21. In Appendix 1 to the Report each of the five categories of warning signs was discussed, and, in relation to each category, a Figure was included showing the red flag scores for each category of company - ie listed by reference to their respective credit ratings. Appendix 2 described certain adjustments which are irrelevant for present purposes. Appendix 3 set out in two Figures (one for property companies the other for non-property companies) against the names of each of the 49 Companies, its current credit rating, the number of red flags tripped, and the nature of the company’s business. Appendix 4 set out against each of the 49 Companies, its current credit rating and, in greater detail than Appendix 3, a breakdown of the red flags for each company. 22. In an Announcement which accompanied the publication of the Report (“the Announcement”), Moody’s stated that the “framework” disclosed by the Report “adds consistency in approach to challenging and often non-transparent situations that are part of the credit decision-making process in new markets”. The Announcement also described the framework as “supplemental to Moody’s methodological approach to rating non-financial corporates in the emerging markets …”. 23. As the Tribunal recorded, the Report was issued at a time when there was considerable concern as to the accounting and corporate governance standards of mainland Chinese companies. Accordingly, “not unsurprisingly”, the Report “received extensive local and international media attention”, and it “had a material impact on the market”. The prices of shares in many of the 49 Companies, particularly those with a large number of tripped red flags, fell substantially (well over 10% in some cases) on the date of the Report’s publication. While acknowledging that many factors can affect share prices, the Tribunal was satisfied that these falls were “principally occasioned by the contents of the Report”, and that Moody’s “must have appreciated … that in all probability the Report would have a material impact on the market”. The procedural history 24. The Commission formed the view that, in preparing and subsequently circulating the Report, Moody’s had not lived up to the standards to be expected of a licensed CRA. In particular, the Commission considered that Moody’s had “failed to have the required procedural safeguards in place to ensure the integrity of the Report and that the Report itself was in a number of material respects misleading, confusing and inaccurate to the extent that the publication was, or was likely to be, prejudicial to the interests of the investing public, including Moody’s own clients, and prejudicial also to the integrity of the market” – to quote the Tribunal’s “broad description” of the Commission’s views. 25. More particularly, the Commission was of the view that the Report (i) had caused confusion in the market because its nature and purpose were unclear; (ii) failed to disclose justifications for the red flags; (iii) awarded a number of inappropriate red flags; (iv) identified six companies as “negative outliers” since they attracted the most flags, even though Moody's acknowledged that the number of flags did not correlate with a higher credit risk; (v) contained 12 factual errors; and (vi) was published without adequate internal control procedures. 26. The Commission concluded that, as a result, Moody's had failed to comply with three provisions of the Code of Conduct for Persons Licensed by or Registered with the Commission (“the Code of Conduct”), which had been issued pursuant to section 169 of the Ordinance. In summary terms, those failures were as follows: a. Three breaches of General Principle 1 (“GP1”), which requires licensed persons to conduct their business “honestly, fairly and in the best interests of their clients and the integrity of the market”; b. A breach of General Principle 2 (“GP2”), which requires licensed person conducting their business to “act with due skill, care and diligence, in the best interests of their clients and the integrity of the market”; c. A breach of paragraph 4.3 (“para 4.3”), which requires licensed persons to have appropriate “internal control procedures and operational capabilities”. 27. Accordingly, the Commission issued a Notice of Proposed Disciplinary Action against Moody’s on 14 February 2013, in answer to which Moody’s made representations, which did not persuade the Commission to change its view. Accordingly, as mentioned above, the Commission issued the Notice in November 2014 fining Moody’s and declaring it unfit. 28. On Moody’s application for review, the Tribunal: a. Upheld two of the three breaches the Commission had found of GP1, but overturned one finding of breach; b. Upheld the breach the Commission had found of GP2; c. Overturned the Commission’s finding of a breach of para 4.3. Accordingly, the Tribunal reduced the financial penalty, but maintained the declaration, and ordered Moody’s to pay 60% of the Commission’s costs. 29. More centrally for present purposes, the Tribunal rejected Moody’s argument that the Commission (and therefore the Tribunal) had no jurisdiction to discipline Moody’s under Part IX of the Ordinance because the publication of the Report could not constitute “misconduct” because it did not “relat[e] to the carrying on of any regulated activity”. 30. In that connection, the Tribunal primarily held that the publication of the Report amounted to the “carrying on of a regulated activity”, because, in the light of its contents, the Report constituted the provision of credit rating services, and therefore it fell within the ambit of the section. As the Tribunal put it, “the red flag framework constituted a well-defined system or mechanism for judging levels of credit risk and, as such, constituted a credit rating” and was “therefore a regulated activity”. Alternatively, the Tribunal held that, if that was wrong, then in the light of Moody’s existing activities, which undoubtedly involved the provision of credit rating services, the publication of the Report amounted to “an act or omission relating to the carrying on of [a] regulated activity” within the ambit of section 193(1)(d). As the Tribunal expressed it, the Report “was intended to be read as amplifying and supplementing Moody’s ratings, as being so intimately attendant upon them that it constituted more than mere comment and became part and parcel of Moody’s ratings themselves”. 31. On Moody’s appeal, the Court of Appeal upheld the decision of the Tribunal. However, while the Court of Appeal did not differ from the Tribunal in so far as the criticisms of Moody’s were concerned, they took a narrower view than the Tribunal on the jurisdiction issue, namely as to how the preparation and publication of the Report fell within the ambit of section 193(1)(d). In particular, the Court of Appeal rejected the view that the Report constituted an exercise of credit rating. They decided that, since it addressed only two elements bearing on credit risk, namely corporate governance and accounting risks, the Report did not come within the definition of “credit rating”, because it could not be regarded as expressing an opinion primarily on creditworthiness of the companies concerned. 32. However, the Court of Appeal agreed with the Tribunal on its alternative ground for accepting that the Commission had jurisdiction. In that connection, the Court said at [31] that: “[E]ven though the red flag framework was not part of the methodology in arriving at Moody’s credit ratings of a classic kind (a point which the Tribunal fully acknowledged …), the Report did constitute additions and clarifications which were meant to be read together with such classic ratings and as such the publication of the Report was an activity relating to the ratings within the meaning of section 193.” 33. It is against that conclusion that, with leave from the Appeal Committee, Moody’s appeals to this Court. The meaning and effect of the phrase “relating to” 34. On behalf of Moody’s, Mr Paul Shieh SC argued that, in the context of section 193(1)(d), the preparation and publication of a document such as the Report, which did not itself involve the provision of credit rating services, could only be said to “relat[e] to” the provision of such services if it had been (or was understood, or would reasonably have been understood to have been) involved in the preparation of credit ratings. Unless such a clear and limited meaning is given to the phrase, he contended, the law would be uncertain, which is particularly inappropriate in the context of Part IX of the Ordinance, given that it creates an offence and involves curtailing freedom of expression. 35. This interpretation of the section involves giving the phrase “relating to” an inappropriately narrow and specific effect. At any rate as a general proposition, it is a phrase with a wide and broad import. Like the words “in respect of”, the phrase could be said to be “colourless” and to have “the widest possible meaning of any expression intended to convey some connection … between the two subject-matters to which the words refer” - see the authorities cited by Fok PJ in Securities and Futures Commission v Pacific Sun Advisors Ltd (2015) 18 HKCFAR 138, at [23]. And it is worth noting that in Mariner International Hotels Ltd v Atlas Ltd (2007) 10 HKCFAR 1 at [51], after observing that the word “involving” was “one of the broadest words of association known to the English language”, Bokhary PJ said that “[p]erhaps only phrases like ‘in relation to’, ‘relating to’ and ‘with respect to’ are wider”. 36. Mr Shieh referred to a decision of the High Court of Australia, Tooheys Ltd v Commissioner of Stamp Duties (1961) 105 CLR 602, and in particular observations in the judgment of Taylor J at pp 620-621 about the meaning of the expression “relating to” in an Australian taxing statute. He said that “the expression ‘relating to’ is extremely wide but it is also vague and indefinite”. And he went on to say that the expression “predicates the existence of some kind of relationship but it leaves unspecified the plane upon which the relationship is to be sought and identified”. Those observations appear to be entirely consistent with the passages cited in [35] above and to be of general application. 37. While the phrase “relating to” has a naturally wide meaning, like any word or set of words it must, in any particular case, take its meaning from its context. Accordingly, a particular context can justify giving a narrower meaning to a term with a generally wide meaning. However, a court should be cautious before holding that, where the legislature has used an expression with a naturally wide meaning, it is nonetheless appropriate to give it a limited effect. 38. To depart from the naturally wide meaning of an expression in a statute requires cogent grounds, because otherwise one may be subverting the expressed aims of the legislature. For the same reason, when the legislature has used words which are very general in their natural ambit, such as “relating to” it is inappropriate, in the absence of a cogent reason, to attribute a relatively specific meaning to those words. Yet that is what Mr Shieh is arguing for in this case. 39. In order to support his case in this connection, Mr Shieh relied on two propositions. The first was the principle that legislation, especially if it creates an offence, should be clear, and in cases of real doubt should be construed beneficially to alleged offenders. Secondly, Mr Shieh argued that legislation which cuts down freedom of expression should be given a narrow effect. 40. The proposition raised in the first point in [39] above will rarely justify, on its own, giving an artificially narrow meaning to a term in a statute. It is more usually invoked to justify not giving a provision an unusually wide meaning, or favouring a narrower meaning when the provision is equally capable of bearing a wider meaning. Further, when one considers the purpose of section 193(1)(d), it appears, if anything, to point away from giving its provisions a narrow meaning. The section is in a Part of the Ordinance which is concerned with regulating and sanctioning “regulated activity” in financial markets, by licensed persons. It therefore should be interpreted bearing in mind that it was enacted as a part of a scheme introduced to protect members of the public and financial markets against inappropriate or substandard behaviour, and which is directed to sophisticated people, expert and experienced in financial markets, who will, as Mr Shieh acknowledged, be in a privileged position as a result of being licensed, and who will often have ready access to legal advice, and some of whom will be (in many cases perfectly properly) keen to find ways of avoiding or minimising any control over their activities. 41. As to the second point made in [39] above, an Ordinance which is aimed at encouraging high standards, and which penalises substandard work, when carrying out activities which can affect financial markets and investors, is clearly in the public interest. And, in order to be effective, such legislation has to include provisions for disciplining and penalising those who do not live up to the requisite standards. It is hard in those circumstances to see much mileage for Moody’s argument based on freedom of expression. The correct viewpoint for interpreting the Report 42. In the present context, we are concerned with the publication of a report which is said to relate to the provision of credit rating services. It is clear both from the nature of those services and from the statutory definition of “providing credit rating services” that such services involve communication of information to potential investors and dealers in debt and debt-related instruments, either to the market generally or to subscribers. The Report itself was published to the market generally. Indeed, its publication was essential to the complaint against Moody’s: if the Report had not been published, there would have been no Notice, and publication to the market is an inherent feature in paragraphs (a), (b), and (c) of the statutory definition of “providing credit rating services”. 43. It follows that the question whether the Report can be said to have been “relating to” the provision of credit rating services must, at least primarily, be assessed by reference to how it was, or could reasonably have been expected to be, understood by the people to whom it was addressed – viz. investors and traders in the market for debt and debt-related instruments. That is of some significance not merely of itself, but also because Mr Shieh relied quite heavily on the fact that Moody’s did not regard the Report as part of, or connected with, its credit rating service. He contended that Moody’s personnel regarded the Report as something of a pilot or new experiment. Although such evidence could conceivably be relevant when it comes to assessing the culpability of Moody’s conduct, it is inadmissible on the issue we have to resolve, in the same way that a party’s subjective understanding or assumption as to the effect or meaning of a contract is inadmissible on the issue of its meaning or effect (it is also by no means clear that, if this evidence was admissible, it would help Moody’s case, but it is unnecessary to address that point further). 44. Strictly speaking, it is unnecessary to decide whether the issue in this case should be determined by reference to how the market actually treated the Report, or how the market could reasonably have been anticipated to treat the Report. That is because the Tribunal concluded that Moody’s ought reasonably to have anticipated that the market would react to the publication of the Report in the way that it actually did react. However, it is right to say that the correct approach is the latter formulation, namely how the market would reasonably have been anticipated to treat the Report. That is because one is ultimately judging the actions of Moody’s in preparing and publishing the Report, and the market reaction occurred after publication. However, that does not mean a tribunal deciding the issue cannot take into account the way the market actually reacted when determining what someone (perhaps particularly someone with considerable experience of the market) should have expected of the market. 45. This is scarcely a surprising conclusion. We are here concerned with a document which was intended to be read by people in the debt and debt-related instrument market. Accordingly, the question whether it related to other services provided to people in that market by the compiler of the document, ought, on normal principles of interpretation of documents, to be determined by how it would have been understood by a reasonable person in the position of the people to whom it was addressed, and not by reference to the private intentions of its compiler. And that involves considering the document by reference to what it says, reading it in a practical and realistic way in its commercial context. Did the Report relate to Moody’s credit rating services? 46. Having indicated the correct approach to the meaning of “relating to” and to the interpretation of the Report, it is now appropriate to turn to consider the contents of the Report coupled with Moody’s relevant prior activities. Once one does so, it appears to be an inescapable conclusion that, even accepting that it did not itself involve the provision of credit rating services, the Report related to the provision of such services by Moody’s. 47. Moody’s is renowned for providing credit rating services, and indeed that is the only one of the twelve Types of regulated activity for which it is licensed under the Ordinance. Even in the absence of any of the other factors mentioned in [48] to [54] below, there must at least be a substantial risk that many people in the market will assume that any report emanating from Moody’s, especially if it is concerned with the soundness of the accounting and governance processes and the quality of earnings of specific companies, is in some way related to its credit rating service. Of course, a fair reading of such a report might satisfy a reader that it is not in fact so related. However, when one examines the Report, far from negativing such an assumption, it is possible to identify a number of significant features which would have served to confirm it. 48. Thus, the Report covered 61 Chinese high yielding non-financial companies, all of which were the subject of Moody’s existing published credit ratings. Indeed, they were the only Chinese high yielding non-financial companies for which Moody’s provided credit ratings. This fact on its own seems to tie in the Report with Moody’s previous credit rating reports on those companies. A clear and convincing disclaimer might have discharged that view, but there was none. 49. Additionally, in the text of the Report, there were frequent references to the credit ratings of the 49 Companies, albeit mostly in general terms. More specifically, in the introductory Overview, Moody’s stated that the red flag results, with which the Report is concerned, are shown “by rating category”, which strongly suggests a connection. 50. Quite apart from these points, reflecting what was said in the Overview, in all of the eight Figures, and in three of the four Appendices of the Report, the “tripped” red flags in relation to companies, or categories of companies, were compared with the credit rating of the companies or categories respectively. 51. It is true that the Report emphasised that the red flag system which it introduced did not represent or justify a departure from the previous ratings accorded by Moody’s to any of the 61 companies in its previous credit rating reports. But the obvious message to any remotely acute trader or investor was that the red flags attributed to a particular company in the Report should at least be borne in mind when considering that company’s existing credit rating. Revealingly, the Report stated that a large number of red flags “does not represent an immediate rating concern”, which was a pretty clear hint that the rating in question may very well change negatively in the future. 52. It is also true that the Report largely concentrated on corporate governance and accounting risks which are but two of the many factors which a credit rating takes into account. While this represents a powerful reason for concluding (as the Court of Appeal did) that the Report did not constitute a credit rating service, this point goes nowhere in the context of the present issue. Apart from anything else, those two factors play an important part in assessing credit risk, and it cannot seriously be suggested that a document cannot relate to credit rating reports unless it covers every factor which is taken into account when arriving at such a rating. 53. Another significant factor which tells against Moody’s arguments is the Announcement accompanying the Report, which described the contents of the Report as being “supplemental to Moody’s methodological approach to rating non-financial corporates in the emerging markets”. If a document is “supplemental” to something, one would have thought that, while it ultimately depends on the context, it would also “relate to” that thing. And, while it is true that “Moody’s methodological approach to rating” is not the same as its actual rating reports, it is little more than playing with words to invoke the difference in the present context. Of course, Moody’s own description of the Report cannot be determinative of the nature or character of the Report as a matter of law, but, especially when it is a contemporaneous published description, it is a telling feature. 54. In addition to these points, there is the external factor of the market reaction to the Report, a reaction which should, according to the Tribunal, have been foreseen by Moody’s. Given the importance accorded by the market to the ratings attributed to companies or to company debt by the two or three main rating agencies, the only sensible interpretation of the market reaction to the Report is that it was indeed seen as providing a negative sort of qualification to the existing ratings of the 61 Companies, or at any rate at least the great majority of the 49 Companies. Conclusion 55. For these reasons, Moody’s appeal must be dismissed. 56. The Commission argued in its written submissions that we should reverse the conclusion of the Court of Appeal overturning the Commission’s primary reason for holding that section 193(1)(d) applied, namely that the production and publication of the Report constituted the “carrying on of [a] regulated activity”, namely the “preparing [of a] credit rating…for dissemination to the public”. After we had indicated that we would be dismissing Moody’s appeal, the Commission did not press us to decide the point, and we saw no benefit in considering or determining it in the context of the facts of this case. 57. Finally, we invited submissions on costs at the end of the hearing and Mr Shieh realistically conceded that he could not oppose the Commissioner’s application for the costs of this appeal, which we accordingly grant. Mr Paul Shieh SC and Ms Zabrina Lau, instructed by Linklaters, for the appellant Mr Benjamin Yu SC and Mr Laurence Li, instructed by Securities and Futures Commission, the respondent 
Hon Cheung CJHC: The facts 1. These appeals are from the judgment of Au J dated 15 November 2016. 2. The controversy leading to these appeals is so great and so widely reported that it is unnecessary, particularly in view of the urgency of the matter, to give any detailed account of the facts. Suffice it to say, Sixtus Leung Chung Hang (“Leung”) and Yau Wai Ching (”Yau”) were elected in their respective geographical constituencies in the general election for the Legislative Council (“LegCo”) held in September this year. Their terms of office as members of the LegCo started on 1 October 2016. 3. As stipulated in section 19 of the Oaths and Declarations Ordinance (Cap 11) (“the Ordinance”): “ A member of the Legislative Council shall, as soon as possible after the commencement of his term of office, take the Legislative Council Oath which – (a) if taken at the first sitting of the session of the Legislative Council immediately after a general election of all members of the Council and before the election of the President of the Council, shall be administered by the Clerk to the Council; (b) if taken at any other sitting of the Council, shall be administered by the President of the Council or any member acting in his place.” 4. Section 16(d) and Schedule 2, Part IV of the Ordinance require the LegCo Oath to be in the following terms: “ I swear that, being a member of the Legislative Council of the Hong Kong Special Administrative Region of the People’s Republic of China, I will uphold the Basic Law of the Hong Kong Special Administrative Region of the People’s Republic of China, bear allegiance to the Hong Kong Special Administrative Region of the People’s Republic of China and serve the Hong Kong Special Administrative Region conscientiously, dutifully, in full accordance with the law, honestly and with integrity.” 5. The first meeting of the LegCo was held on 12 October 2016. On that day, both Leung and Yau were duly requested to take the LegCo Oath before the Clerk to the LegCo, as the election of the President of the Council had yet to take place. Both Leung and Yau purported to do so, but in ways and manners, detailed in paragraph 5 of the judgment below, which departed substantially from the statutory contents of the LegCo Oath, and evinced objectively an intention on their respective parts not to be bound by it. 6. In particular, it is plain, as the learned judge below analysed in paragraphs 45 and 46 of his judgment, that neither Leung nor Yau intended to uphold the Basic Law of the Hong Kong Special Administrative Region of the People’s Republic of China, or bear allegiance to the Hong Kong Special Administrative Region of the People’s Republic of China. Both elements are mandatory parts of the LegCo Oath. But not only that – they actually constitute the core substantive requirement of article 104 of the Basic Law: “ When assuming office, the Chief Executive, principal officials, members of the Executive Council and of the Legislative Council, judges of the courts at all levels and other members of the judiciary in the Hong Kong Special Administrative Region must, in accordance with law, swear to uphold the Basic Law of the Hong Kong Special Administrative Region of the People's Republic of China and swear allegiance to the Hong Kong Special Administrative Region of the People's Republic of China.” 7. On 18 October 2016, the President of the LegCo (elected to his office after the oath taking incident described above) gave a written “ruling”, after obtaining senior counsel’s advice, that “[Leung and Yau] could not be serious about their oath and were unwilling to be bound by it” (paragraph 6), and their oaths were invalid. Nonetheless, he said he was “prepared to allow Mr LEUNG and Ms YAU to take their oath afresh at a Council meeting if they put forward their requests in writing” (paragraph 7), which requests Leung and Yau immediately made on the same day. 8. As is only too well-known, these events triggered the urgent commencement by the Chief Executive and the Secretary of Justice of two sets of proceedings below on 18 October 2016, which were described in some detail in paragraphs 9 to 12 of the judgment below, as well as an interpretation by the Standing Committee of the National People’s Congress (“NPCSC”), pursuant to article 67(4) of the Constitution of the People’s Republic of China and article 158(1) of the Basic Law, of the true meaning of article 104 on 7 November 2016 (“the Interpretation”)[1]. The judgment below 9. After an expedited hearing, the judge concluded in his judgment that what Leung and Yau did amounted to their respectively declining or “wilfully” omitting to take the LegCo Oath when duly requested to do so. The consequence of what they did, the judge decided, is governed by section 21 of the Ordinance: “ Any person who declines or neglects to take an oath duly requested which he is required to take by this Part, shall- (a) if he has already entered on his office, vacate it, and (b) if he has not entered on his office, be disqualified from entering on it.” 10. Rejecting an argument to the contrary, the judge decided that once a person declines or neglects to take the relevant oath when duly requested to do so, he is automatically regarded as having vacated his office in question under section 21. In other words, what Leung and Yau did on 12 October 2016 led automatically, by operation of law under section 21, to the vacation of their respective offices as Legislative Councillor. 11. In those circumstances, the judge agreed with the Chief Executive and the Secretary for Justice that contrary to his ruling dated 18 October 2016, the President of the LegCo had no power to give Leung and Yau a second chance to take the LegCo Oath again. The judge rejected various arguments raised by Leung and Yau, including their main argument based on the non-intervention principle, and decided the proceedings in favour of the Chief Executive and the Secretary for Justice by granting declaratory and other relief against the President, “on the basis that Mr Leung and Ms Yau have already vacated the office as a member of the LegCo, and are not entitled to act as a member of the LegCo” (paragraph 130(2)(a)). In other words, Leung and Yau have lost their seats in the LegCo, their offices are vacant and there will be by-elections. The arguments on appeal 12. Aggrieved by the judge’s decision, Leung and Yau appealed. Mr Hectar Pun SC (Mr Anson YY Wong with him), for Leung, essentially argued that the principle of non-intervention applied to the present case so that the judge was wrong to interfere with the internal workings or business of the LegCo, that is, the taking or the retaking of the LegCo Oath by Leung. Mr Pun accepted that the principle of non‑intervention as applied in Hong Kong “is subject to the constitutional requirements of the [Basic Law]” (paragraph 12(1) of his skeleton arguments), but he emphasised that in the present case, the constitutional requirement under article 104 of the Basic Law was simply “not engaged” because the dispute here was not about whether his client had taken the LegCo Oath – he had not. Rather, the issue was about the consequence of his failure to do so, which turned on whether Leung had declined or neglected to take the LegCo Oath – in which case section 21 of the Ordinance would be triggered and his client would be obliged to vacate his seat. Mr Pun submitted that section 21 does not form part of the constitutional requirement under article 104. He argued that the present case was not distinguishable from the English case of Bradlaugh v Gossett (1884) 12 QBD 271, where the English court refused to interfere with an oath taking dispute between Parliament and one of its members. 13. Mr Pun further argued neither the Clerk nor the President had determined that Leung had declined or neglected to take the LegCo Oath on 12 October 2016, whereas Leung had indicated in evidence that he had been “ready, prepared and willing” to take the Oath afresh at the next meeting on 19 October 2016 and had not “declined” or “neglected” to do so. Furthermore, the President had indicated in his capacity as the oath administrator that he was prepared to allow Leung to take the Oath afresh on 19 October 2016. All these matters were internal business of the LegCo in which the courts cannot intervene. 14. Mr Pun took the further point, relying on Makucha v Sydney Water Corporation [2013] NSWCA 177, that even if a LegCo member had declined or neglected to take the LegCo Oath, there was to be no automatic vacation of office. Rather, section 21 merely requires the member to vacate it. Mr Pun submitted that the member had to resign or the President could make a declaration, pursuant to article 79(1) of the Basic Law,[2] that the member was “no longer qualified for the office” as he or she had “[lost] the ability to discharge his or her duties as a result of ... [section 21]”. 15. Mr Pun also submitted that section 73 of the Legislative Council Ordinance (Cap 542)[3], based on which one set of proceedings below was commenced, does not confer any jurisdiction on the court to determine matters such as whether a LegCo member has been disqualified from being a member or has ceased to be one, and does not therefore provide a bypass to the non-intervention principle. He argued that as Leung had not been “disqualified” under section 21 of the Ordinance, section 73(1) simply had no application. Section 73(1) is only applicable after a declaration under article 79(1) of the Basic Law has been made by the President as described above. In any event, section 73(1) does not give the Chief Executive any locus to sue. 16. Mr Pun further argued that the judge was wrong in usurping the fact-finding function of the President regarding whether Leung had declined or neglected to take the LegCo Oath. 17. Insofar as may be necessary but not otherwise, Mr Pun relied on the Interpretation, particularly paragraph 2(3) and (4) to say that whether an oath taker has fallen foul of section 21 by declining to take the LegCo Oath is a matter for the Clerk or President, but not the court to decide. However, insofar as the Chief Executive and the Secretary for Justice sought to make a case against his appeals based on the Interpretation, Mr Pun raised the issues of whether the Interpretation has retrospective effect in the sense that it covers the present case now before the court, and whether it is truly an interpretation falling within the meaning of article 158 of the Basic Law (or merely an interpretation in name but an amendment of the Basic Law in substance). 18. Finally, in written submissions, Mr Pun also prayed in aid article 77 of the Basic Law regarding members of the LegCo’s immunity from suit, and argued that the immunity covered the oath taking event in the present case. 19. Mr Philip Dykes SC (Mr Jeffrey Tam with him), for Yau, also relied on the principle of non-intervention. Essentially, Mr Dykes submitted that oath taking is an internal business of the LegCo. Primarily, it is for the Clerk or the President to determine whether an oath has been validly taken, and in particular, whether the member has declined or neglected to take the oath. Mr Dykes accepted that once a decision has been made by the Clerk or the President, the court has the jurisdiction to decide whether article 104 of the Basic Law has been complied with. However, Mr Dykes submitted, neither the Clerk nor the President has made any such decision. It is simply premature for the court to intervene in the matter. 20. Likewise, counsel submitted, an application made under section 73 of the Legislative Council Ordinance is, except in the case of a member’s resignation, predicated on a declaration by the President under article 79 of the Basic Law. In the present case, the President has made no such declaration. 21. As Mr Benjamin Yu SC (Mr Johnny Mok SC, Mr Jimmy Ma and Mr Jenkin Suen with him) for the Chief Executive and the Secretary for Justice submitted, the applicants’ arguments essentially raised several subject matters, that is, the non-intervention principle; the role of the oath administrator; whether vacation of the office is automatic under section 21 of the Ordinance; the scope of application of section 73 of the Legislative Council Ordinance; and the immunity from suit under article 77 of the Basic Law. Counsel submitted that the appellants’ arguments were unsustainable. The principle of non-intervention 22. The principle of non‑intervention has been dealt with by the courts in recent years: Leung Kwok Hung v President of Legislative Council [2007] 1 HKLRD 387; Cheng Kar Shun v Li Fung Ying [2011] 2 HKLRD 555; Leung Kwok Hung v President of the Legislative Council of the Hong Kong Special Administrative Region, CACV 123/2012, 1 February 2013 (CA); affirmed on appeal: (2014) 17 HKCFAR 689 (CFA). It is an established principle of common law which is of seminal importance and high constitutional significance. Historically, it was derived from or justified by historical development, functional necessity, the constitutional doctrine of separation of powers and (in the United Kingdom) the sovereignty of Parliament. The preferred view in a jurisdiction like Hong Kong now is to justify it on the common law principle of separation of powers. The principle makes good constitutional as well as practical sense. Under it, the court respects and recognises the exclusive authority of the legislature in managing its own internal processes in the conduct of its business. The court will not intervene to rule on the regularity or irregularity of the internal processes of the legislature but will leave it to determine exclusively for itself matters of this kind. On a practical level, this principle allows the legislature to be left freely to manage and to resolve its internal affairs, free from intervention by the courts and from the possible disruption, delays and uncertainties which could result from such intervention. Freedom from these problems is both desirable and necessary in the interest of the orderly, efficient and fair disposition of the legislature’s business. Leung Kwok Hung (CFA), paragraphs 27 to 30. 23. In the United Kingdom where Parliament (or more precisely, the Queen in Parliament), replacing the absolute monarchy in old times, is supreme and sovereign, a further explanation for this principle of non-intervention is that it gives effect to Parliament’s supremacy and sovereignty. In the context of Parliament’s law making function, the court’s role there is confined to interpreting and applying what Parliament has enacted. When an enactment is passed there is finality unless and until it is amended or repealed by Parliament: The Bahamas District of the Methodist Church v Symonette [2000] 5 LRC 196, 207h‑208a. 24. However, in a jurisdiction like Hong Kong where a written constitution (that is, the Basic Law), rather than the legislature, is supreme, where the rule of law reigns and where the courts are given under the constitution the independent power of adjudication, this principle of non-intervention has its own inherent limit. 25. First and foremost, the supremacy of the Basic Law means that no one – the legislature included – is above the Basic Law. In other words, where a constitutional requirement under the Basic Law is in issue, even the legislature cannot act contrary to that requirement under the Basic Law. Secondly, given that the courts are given under the constitution the independent power of adjudication of the Special Administrative Region, the question of whether that constitutional requirement has been complied with or breached is a matter which it is both the power and responsibility of the courts to decide. As the Court of Final Appeal importantly pointed out in Leung Kwok Hung, paragraph 32: “In this respect it is important to recognise that the principle of non-intervention is necessarily subject to constitutional requirements.” Article 104 is a constitutional requirement 26. In the present case, there cannot be any doubt that article 104 of the Basic Law lays down a constitutional requirement. Article 104 is found in Chapter IV of the Basic Law setting out the political structure of the Hong Kong Special Administrative Region. It contains six sections, dealing respectively with the Chief Executive, the Executive Authorities, the Legislature, the Judiciary, District Organizations and lastly, Public Servants. In essence, it covers everyone who is empowered to and charged with the responsibility for running the Special Administrative Region. At the very end of this long chapter, one finds article 104, which expressly requires that when assuming office, the Chief Executive, principal officials, members of the Executive Council and of the LegCo, judges of the courts at all levels and other members of the judiciary to, in accordance with law, swear to uphold the Basic Law and swear allegiance to the Special Administrative Region. 27. As the judge correctly explained (paragraphs 31 to 33), the taking of an oath and pledging of allegiance are serious matters. When taking an oath, no less a promissory oath such as the LegCo Oath, both the form and the substance matter greatly. The requirement under article 104 is plainly designed to secure the genuine, solemn and sincere declaration and pledge by the holders of the important offices mentioned in that article to do their utmost, in accordance with the Basic Law, to discharge the high responsibilities entrusted to them in running the Special Administrative Region in their respective roles assigned under the Basic Law. Article 104 clearly lays down a constitutional requirement that an oath must be taken in accordance with what is required under that article. Moreover, it says “when assuming office”, the oath must be taken. It must mean that taking the oath is a prerequisite and precondition to the assumption of office. 28. All this is now put beyond doubt by the Interpretation. Consequence of non‑compliance is part of the constitutional requirement 29. The Interpretation gives the true meaning of article 104. Paragraph 2(3) of the Interpretation specifically sets out the consequence of an oath taker’s declining to take the relevant oath – automatic disqualification, as part of the true meaning of article 104. It conclusively defeats Mr Pun’s argument that the consequence of a failure to take the relevant oath as required by article 104 does not form part of the constitutional requirement, so that the principle of non‑intervention applies. 30. Furthermore, article 104 says the oath must be taken “in accordance with law”. The relevant provisions in the Ordinance actually predated the drafting of the Basic Law. When article 104 refers to law, the drafters must have in mind the provisions in the Ordinance. Section 21(a) of the Ordinance says that if an office holder declines or neglects to take the relevant oath, he shall vacate his office. That is perfectly consistent with article 104. Since article 104 specifically refers to the implementing law, it provides another reason for rejecting Mr Pun’s argument that article 104 is not engaged but only section 21(a) of the Ordinance is – and therefore the principle of non‑intervention still applies. It is neither right nor realistic to look at article 104 without looking at its implementing law (in the present case, the relevant provisions in the Ordinance) together, or to look at the statutory provisions without looking at article 104 at the same time, in deciding whether the constitutional requirement under article 104 has been satisfied. 31. Mr Pun argued that “law” in the phrase “in accordance with law” only includes sections 16 and 19 of the Ordinance, but not section 21. With respect, this is taking far too narrow a reading of article 104. “Law” in article 104 must be a reference to the whole of the implementing law, including in particular that part of the law which prescribes the consequence of a failure to take the oath in question (that is, section 21). As Mr Yu submitted, when article 104 lays down a requirement, the consequence of failing to meet that requirement is necessarily part and parcel of the requirement itself. The principle of non‑intervention cannot prevent the court from adjudicating on the consequence of a failure to meet the constitutional requirement. A matter for the court to decide 32. Since articles 19(1) and (2) and 80 of the Basic Law vest the independent judicial power of the Special Administrative Region in the courts, giving the courts “jurisdiction over all cases in the Region”, and establishing them as the judiciary of the Region, “exercising the judicial power of the Region”, it is for the courts, not anyone else, to determine whether the constitutional requirement described above has been satisfied. 33. This disposes of three arguments mounted on behalf of Leung and Yau. First, as mentioned, the courts cannot shrink from their constitutional duty to adjudicate on the question of whether the constitutional requirement under article 104 has been satisfied by not intervening in the present dispute in the name of the principle of non-intervention or separation of powers. The Basic Law, not the legislature, is supreme. Secondly, it also disposes of the further argument that it is for the oath administrator (that is, the Clerk or the President) to determine whether Leung and Yau have respectively taken a valid oath, and if not, whether they have respectively declined or neglected to take the oath. Thirdly, it also disposes of the more limited argument raised by Mr Dykes that before the oath administrator makes a decision on the issues just described, it is premature to invite the court to intervene.[4] Plainly, section 21 says none of these. Nor does the principle of non‑intervention require any such interpretation be put on section 21, as the principle simply has no application given that a constitutional requirement is involved. 34. All these arguments must be rejected. Paragraph 2(4) of the Interpretation 35. Mr Pun relied on paragraph 2(4) of the Interpretation to back his argument that it is for the oath administrator, rather than the court, to decide whether the LegCo Oath has been validly taken, or whether the (purported) oath taker has declined or neglected to take the oath. I reject the argument. Paragraph 2(4) of the Interpretation reads in Chinese: “宣誓必須在法律規定的監誓人面前進行。監誓人負有確保宣誓合法進行的責任，對符合本解釋和香港特別行政區法律規定的宣誓，應確定為有效宣誓；對不符合本解釋和香港特別行政區法律規定的宣誓，應確定為無效宣誓，並不得重新安排宣誓。” 36. It is clear from the Chinese version, particularly the use of the words “應確定” (which is better rendered as “should confirm/affirm”), that paragraph 2(4) seeks to emphasise the important administrative duty of the oath administrator to ensure that the oath taker has taken the relevant oath properly and validly in full accordance with the Interpretation and the law, and that when the office holder declines to take the oath (paragraph 2(3)), the oath administrator must resolutely say so and refuse to make any administrative arrangement for the retaking of the oath. What it plainly does not say is it gives the oath administrator any judicial power of the Special Administrative Region to determine whether the oath taken is in accordance with the requirements of the Basic Law and the Ordinance. Still less does it take away the courts’ judicial power of the Special Administrative Region, granted under the Basic Law, to adjudicate on a dispute. 37. Neither does the Interpretation give the oath administrator any fact-finding role in any judicial sense. In other words, it does not give the oath administrator a judicial power of the Special Administrative Region to make any finding of fact. Nor does it constitute the oath administrator as a sort of administrative tribunal of fact (subjecting him thereby to all that standard administrative law requires of such a tribunal of fact to observe by way of procedural fairness etc). Still less does it exclude the courts’ judicial power, conferred under the Basic Law, to make the relevant findings of fact. 38. If anything, paragraph 4 highlights the absolute importance of full compliance with the oath taking requirements under article 104 and the implementing law. Indeed, one may ask rhetorically: if the oath administrator is required, as indeed he is under paragraph 2(4), to use his utmost to ensure compliance, how much more are the courts of the Special Administrative Region expected and required to do so? 39. In the final analysis, what is at stake is the compliance of a constitutional requirement of great significance. In any given set of facts, this can admit of one correct answer only. There is no room for a court to simply sit back without correcting an answer given by the oath administrator which the court considers to be wrong, at the expense of the constitutional requirement. What is in issue is squarely a judicial matter which the courts alone are given the judicial power of the Special Administrative Region under the Basic Law to determine. What is involved is not an ordinary judicial review type of situation where the court only conducts a Wednesbury unreasonableness review. Rather, there can be only one right answer when the issue of compliance with the constitutional requirement is raised and nothing short of a full merit review will suffice. The court, according to the Basic Law, is the ordained organ to determine the question. 40. Of course, what I have said above does not prevent at all a court from, when hearing a dispute on the validity of an oath taken or one regarding whether the oath taker has declined or refused to take the oath when duly requested to do so, receiving evidence from the oath administrator on what his views are and the reasons for those views, insofar as they are relevant and admissible, and according them weight accordingly. Both Leung and Yau have declined to take the Oath 41. On the facts, there can be no dispute that both Leung and Yau have declined respectively to take the LegCo Oath. They have put forward no argument to dispute this. Nor can they. There can be no innocent explanation for what they uttered and did on 12 October 2016. What has been done was done deliberately and intentionally. This conclusion, reached by the judge after careful consideration, is unassailable. Disqualification forthwith and automatic vacation of office 42. As a matter of law and fact, Leung and Yau have failed the constitutional requirement. They are caught by paragraph 2(3) of the Interpretation as well as section 21 of the Ordinance which gives effect to the constitutional requirement. Under the former, they were automatically disqualified forthwith from assuming their offices. Under the latter, they “shall ... vacate [their respective offices].” There is therefore no question of allowing them to retake the LegCo Oath. 43. That leaves only the question of whether vacation of office under section 21 is automatic. This is a necessary plank to Mr Pun’s and Mr Dykes’ procedural argument, based on section 15(1)(e) of the Legislative Council Ordinance[5], that a declaration by the President under article 79(1) of the Basic Law is a prerequisite to any proceedings under section 73 of the Legislative Council Ordinance. 44. I agree with the judge that vacation of office is automatic. It simply serves no useful purpose to require a further step to be taken by the person in question to vacate his office. By definition, this is someone who has declined or wilfully neglected to do the very first and basic thing required of him or her when assuming office, that is, to take the oath when duly requested to do so. There is simply no point in requiring him or her to take the further step of vacating the office. To do so would simply invite further dispute as well as create confusion and uncertainties. The Australian case of Makucha is not binding on this court. It was dealing with a totally different situation, and the court there was simply expressing an obiter view without any reasoning (paragraph 18). It is fair to say that the court there was mainly concerned with upholding the judicial acts done by the judge below the validity of whose judicial oath was called into question. Article 79(1) of the Basic Law 45. I reject the further argument based on article 79(1) of the Basic Law. With respect, one only needs to read the Chinese text of article 79(1)[6] to see that it plainly does not apply to the present type of situation. Disqualification under section 21 is miles away from “無力履行職務” (“loses the ability to discharge his or her duties”) in article 79(1). Section 73 of the Legislative Council Ordinance 46. In any event, I reject the argument that the scope of application of section 73(1) is limited to those situations set out in section 15 of the Legislative Council Ordinance. It does not say so. The fallacy of the argument is to treat section 15 as being exhaustive. It is not. 47. Given my views above, the arguments over the scope of application of section 73 of the Legislative Council Ordinance are of little significance so far as the principle of non-intervention is concerned. It is unnecessary to resort to section 73 to get round, as it were, the principle of non‑intervention. When a constitutional requirement is at stake, this court’s jurisdiction is not founded on the legislature’s voluntary concession of jurisdiction. This court’s jurisdiction is conferred by the Basic Law, the exercise of which is a matter of constitutional duty, which is not restrained by the common law principle of non‑intervention. 48. Section 73 is of significance in the arguments raised in these appeals also for a procedural reason. It is said that section 73 only gives the Secretary for Justice, but not the Chief Executive, the locus to sue. Further, it is said that section 73(7) excludes all other forms of proceedings. 49. All I wish to say is this. I accept that proceedings under section 73 may only be brought by the Secretary for Justice or an elector, but not the Chief Executive in his official capacity as such. However, I do not agree that apart from section 73, no proceedings can be brought by the Chief Executive. Given the Chief Executive’s constitutional role under article 48(2) of the Basic Law (to be responsible for the implementation of the Basic Law and other laws), any attempted restriction on the Chief Executive’s right to take steps, including the commencement of court proceedings, to implement the Basic Law must be incompatible with article 48(2) and therefore invalid. Plainly, section 21J of the High Court Ordinance (Cap 4)[7] entitles the Chief Executive to sue. On its proper construction, section 73(7) does not prevent the Chief Executive from doing so. 50. In any dispute, one must look at the facts concerned and the relief one seeks to determine whether proceedings should be commenced under section 73 of the Legislative Council Ordinance, or under section 21J of the High Court Ordinance by means of an application for judicial review (section 21K(1)(b) of the High Court Ordinance), or both. Immunity from suit under article 77 51. Immunity from suit enjoyed by legislators under article 77 of the Basic Law has no relevance in the present dispute. The article gives members of the LegCo immunity from legal action in respect of their statements at meetings of the Council. It is neither necessary nor desirable to define the scope of immunity conferred by that article in these appeals. One thing is clear. The Basic Law must be read as a whole. Article 77 must be read together with article 104. Given the importance of article 104 as explained above, it is simply unarguable that the drafters of the Basic Law intended to permit members of the LegCo – yet not anyone else as article 77 only applies to them – to get round the constitutional requirement on oath taking laid down in article 104 via the backdoor of article 77. It would make a mockery of article 104 and serve no discernable, meaningful purpose. Thus analysed, the old English case of Bradlaugh v Gossett, a case heavily relied on by Mr Pun and Mr Dykes, loses much if not all of its relevance in the present debate. First, the English court there was not faced with a constitutional requirement. Secondly, Parliament is supreme and sovereign in the United Kingdom. The English court was simply not in a position to adjudicate on the dispute between Parliament and its member in question. In any event, the English case is not binding on this court, and insofar as may be necessary, I would, with respect, decline to follow it. The Interpretation 52. Finally, I turn to Mr Pun’s arguments on the Interpretation. 53. First, the Interpretation, by definition, sets out the true and proper meaning of article 104 from day one. The question of whether it applies “retrospectively” to any given set of facts whether pending before the court or not therefore cannot and does not arise, save for the situation specifically provided in article 158(3) of the Basic Law, which is not the case here. As the Court of Final Appeal explained in Lau Kong Yung v Director of Immigration (1999) 2 HKCFAR 300, 326D: “The Interpretation, being an interpretation of the relevant provisions, dates from 1 July 1997 when the Basic Law came into effect. It declared what the law has always been.” 54. Secondly, regarding Mr Pun’s argument that an interpretation only applies retrospectively if it merely clarifies the law in question but not when it supplements it, this submission has no support from the authorities. The Court of Final Appeal said in Director of Immigration v Chong Fung Yuen (2001) 4 HKCFAR 211, 222J‑223C: “ The Standing Committee’s power to interpret the Basic Law is derived from the Chinese Constitution and the Basic Law. In interpreting the Basic Law, the Standing Committee functions under a system which is different from the system in Hong Kong. As has been pointed out, under the Mainland system, legislative interpretation by the Standing Committee can clarify or supplement laws. Where the Standing Committee makes an interpretation of a provision of the Basic Law, whether under art. 158(1) which relates to any provision, or under art. 158(3) which relates to the excluded provisions, the courts in Hong Kong are bound to follow it. Thus, the authority of the Standing Committee to interpret the Basic Law is fully acknowledged and respected in the Region. This is the effect of the Basic Law implementing the ‘one country, two systems’ principle as was held by the Court in Lau Kong Yung. Both systems being within one country, the Standing Committee’s interpretation made in conformity with art. 158 under a different system is binding in and part of the system in the Region.” 55. No distinction is drawn between the two types of situations under discussion. The suggested distinction by Mr Pun must be rejected. 56. Thirdly, as to the submission that the Interpretation is only an interpretation in name but an amendment of the Basic Law in substance which can only be done by the National People’s Congress in accordance with article 159 of the Basic Law, this argument does not even get off the evidential ground and must be rejected. Here, as explained in Chong Fung Yuen in the passage cited above, one is in the interface of “one country two systems”, and, in particular, one is concerned with the other system – that is, the civil law system practised on the Mainland, when it comes to an interpretation by the NPCSC. When the NPCSC interprets the Basic Law, it is operating under the Mainland’s civil law system. 57. In the absence of any evidence to show what, under a civil law system, particularly the civil law system practised on the Mainland, is regarded as the proper scope of an interpretation of the present type, one simply has no material to argue, let alone to conclude, that what has been done has gone outside the permissible scope of an interpretation. The view of a common law lawyer, untrained in the civil law system, particularly the civil law system practised on the Mainland, is, with respect, simply quite irrelevant. 58. But more importantly, this present argument raises an a priori question of whether under the Basic Law, the courts of the Hong Kong Special Administrative Region have ever been vested with the jurisdiction to determine whether an interpretation officially promulgated as such by the NPCSC in accordance with article 67(4) of the Constitution and article 158 of the Basic Law and the procedure therein is invalid on the ground under discussion. Apart from citing to the court a passage in Ng Ka Ling v Director of Immigration (1999) 2 HKCFAR 4, 26A-B which must be read together with Ng Ka Ling (No 2) (1999) 2 HKCFAR 141, where the Court of Final Appeal clarified in no uncertain terms that the courts in Hong Kong cannot question “the authority of the National People’s Congress or the Standing Committee to do any act which is in accordance with the provisions of the Basic Law and the procedure therein” (p 142E), Mr Pun has simply made no submission on this fundamental question of jurisdiction. 59. In my view, the court has no jurisdiction to deal with the issue raised. Disposal 60. For all these reasons, I would dismiss these appeals with costs. Hon Lam VP: 61. I respectfully agree with the judgment of the Chief Judge and for the reasons he gave the appeals must be dismissed. In view of the constitutional importance of these appeals and the apparent misapprehension on the scope of the principle of non-intervention, I would add some observations of my own. 62. As accepted by Mr Dykes (but disputed by Mr Pun), in Hong Kong the principle of non-intervention is only a self-restraint imposed by the courts in the exercise of jurisdiction rather than a matter that goes to jurisdiction. 63. The Basic Law is the foundation of all executive, legislative and judicial authorities in Hong Kong. As my Lord the Chief Judge pointed out, the judicial power of the courts in Hong Kong and the correspondent adjudicative authority are derived from article 80 of the Basic Law. Further, under article 85, the courts shall exercise judicial power independently and free from any interference. In the exercise of judicial power, judges must adjudicate issues strictly in accordance with laws, and nothing else. The laws, as set out in article 8 are, subject to a very important rider, the common law and statutes. The rider is that these laws must not contravene the Basic Law. 64. Thus, as explained by the Chief Judge, the LegCo, whilst designated as the legislative authority under the Basic Law, must also be subject to the Basic Law. In this respect, our LegCo is different from Parliament of the United Kingdom. 65. The Basic Law does not confer any judicial authority on the LegCo. Hence, in respect of issues arising from disputes concerning compliance with the Basic Law, the LegCo does not have any judicial authority. Whilst article 79 of the Basic Law does confer authority on the President of the LegCo to declare a member no longer qualified for office in certain specified circumstances, it is not the same as saying that the President has judicial authority to determine all questions relating to membership of the LegCo. A clear example where membership of LegCo is to be determined by courts through a judicial process is election petition concerning LegCo elections. 66. Since the source of the courts’ judicial power is derived from the Basic Law, a judge has a constitutional duty to consider the constitutionality of legislation passed by the LegCo if it is an issue arising from the facts of a case before him. Likewise, where there is a complaint as to the non-compliance with a constitutional prerequisite before assumption of office, our courts are duty-bound to examine the same according to law. As explained by the Chief Judge, article 104 is an important constitutional requirement and the courts have a constitutional duty to uphold the same. 67. The exercise of such judicial power should not be perceived as the courts being embroiled in political disputes. In the judgment of the U.S. Supreme Court in Powell v Mc Cormack (1969) 395 US 486cited by Mr Dykes, Warren CJ held at pp.548-9 that the determination of a Congressman’s right to sit required no more than an interpretation of the Constitution which fell within the traditional role accorded to courts to interpret the law. Even if the exercise of such power conflicted with the resolution of the Congress, Warren CJ said these: “The alleged conflict that such an adjudication may cause cannot justify the courts’ avoiding their constitutional responsibility.... For, as we noted in Baker v Carr ... it is the responsibility of this Court to act as the ultimate interpreter of the Constitution.” 68. As I have said, in adjudicating on these issues the courts only concern themselves with legal questions, and the issues before us in these appeals are whether the requirement in article 104 has been complied with and if not, what are the consequences. In our deliberations, we address legal arguments advanced before us and apply the law (including the Interpretation, which upon its pronouncement becomes part of our laws) strictly as we find them. It is for this very reason that the Chief Judge had to stop Mr Yu when counsel at one stage unwittingly treaded beyond the proper scope of legal arguments by quoting from Socrates on abuse of democracy. It is important that we keep politics out of the judicial process. 69. Nor should the exercise of judicial power in these instances be regarded as the court’s intervention into the internal affairs of the LegCo. The principle of non-intervention is a principle governing the exercise of the court’s jurisdiction, placing importance on the integrity and efficiency of the legislative process. Thus, in cases where the intervention of the court would disrupt the legislative process as in Cheng Kar Shun v Li Fung Ying [2011] 1 HKLRD 555 and Leung Kwok Hung v President of the Legislative Council (2014) 17 HKCFAR 689, the court would refrain from intervention. But the court would not shy away from examining the products of such process as in cases where the courts examined the constitutionality of a piece of legislation. In Leung Kwok Hung, the Court of Final Appeal highlighted at [39] to [43] that in cases where the legislative authority is subject to a written constitution (as in the case of our LegCo), the judicial authority will determine whether the legislature has a particular power, privilege or immunity though it would not exercise jurisdiction to determine the occasion or the manner of the exercise of the same. 70. The Court of Final Appeal expressed the principle of non-intervention in these terms at [28] by reference to the relationship between the legislature and the courts: “....This relationship includes the principle that the courts will recognise the exclusive authority of the legislature in managing its own internal processes in the conduct of its business ...The corollary is the proposition that the courts will not intervene to rule on the regularity or irregularity of the internal processes of the legislature but will leave it to determine exclusively for itself matters of this kind...” 71. The Court of Final Appeal also stressed that the principle of non-intervention is necessarily subject to constitutional requirements, see [32]. This has been consistently regarded by our courts as the limits of that principle, see the decision at first instance in Leung Kwok Hung v President of the Legislative Council [2012] 3 HKLRD 483, at [36]; at the Court of Appeal, in CACV 123 of 2012, 1 Feb 2013, at [24]; Cheng Kar Shun v Li Fung Ying, supra at [220]. 72. For the reasons given by the Chief Judge, the non-compliance with the oath-taking prerequisite in article 104 and its consequence are matters of constitutional requirement which the courts have a constitutional duty to examine when it is an issue in a case coming before us. Compliance with such constitutional requirement is not merely a matter of internal process of the LegCo. Oath taking is not simply a requirement of LegCo’s internal rules. Though it took place within the walls of the LegCo Building, it is a matter of immense importance to the whole Hong Kong Special Administrative Region including the Government as well as its general public because the requirement underpins the allegiance owed to the Hong Kong Special Administrative Region of the People’s Republic of China and the sincerity and integrity of the oath taker in upholding the Basic Law. Without the oath properly taken, the member concerned could not assume office and no authority is conferred upon him or her to act as our lawmaker. 73. As illustrated by election petition cases, membership of LegCo is not an internal matter for the LegCo. The courts must intervene when non-compliance with article 104 is brought to our attention. 74. The soundness of this analysis is easily borne out by examining a hypothetical case where a member is wrongly ruled by the Clerk or President to have failed to comply with article 104 and disqualified under Section 21 accordingly. It is hard to see why in such circumstances such a member cannot come to court to seek relief notwithstanding that the oath taking took place within the walls of the LegCo Building. Mr Dykes and Mr Pun readily agreed that the court must intervene in such circumstances. 75. In the converse situation, where a member should have been disqualified but was not so disqualified, electors in the electoral constituency of that member should, as a matter of law, be entitled to have a by-election for the seat and other candidates should also be entitled to run for the same. Hence, it is impossible to characterise the oath taking as a mere internal affair or a matter in the internal process of the LegCo. There is no reason why the courts should refrain from intervening in such circumstances. 76. Properly understood, Bradlaugh v Gossett (1884) 12 QBD 271 on which Mr Pun placed great reliance cannot stand in the way of the above analysis. What was in issue in that case was the resolution of the House of Commons excluding a member from the House until he shall engage not to attempt to take the oath in disregard of the resolution of the House then in force. What happened was that Mr Bradlaugh was prohibited from taking the oath because he did not believe in the existence of a Supreme Being, see AG v Bradlaugh (1885) 14 QBD 667, also the judgment of Stephen J in Bradlaugh v Gossett, supra at p.279. In that case, Mr Bradlaugh sued the Serjeant-at-Arms who excluded him from the House pursuant to the resolution of the House. It was in such context that Lord Coleridge CJ said at p.275: “What is said or done within the walls of Parliament cannot be inquired into in a court of law... The jurisdiction of the Houses over their own members, their right to impose discipline within their walls, is absolute and exclusive.” Coleridge CJ obviously regarded it as a matter of internal discipline as he said at pp.276-277: “Consistently with all the statements in the claim, it may be that the plaintiff insisted on taking the oath in a manner and under circumstances which the House had a clear right to object to or prevent.” At p.277 he also agreed with Stephen J’s broader analysis as to why the court should not examine the basis on which the House passed the resolution. Stephen J proceeded on the assumption that there was inconsistency between the resolution and the Parliamentary Oaths Act and he ruled that the court should not intervene because (a) The court could not do justice without infringing the dignity and independence of the House as it would necessitate the examination of the reasons for the House’s resolution, see pp.280-281; (b) In respect of matters of parliamentary privilege in respect of affairs within the House of Commons, even if the House decided the same not in accordance with law, it would be akin to an error by a judge whose decision was not subject to any appeal, see p.284-286. 77. It should be noted that the courts did not hold that they had no jurisdiction to examine if the affirmation adopted by a member of the House was valid. In earlier proceedings in Clarke v Bradlaugh (1881) 7 QBD 38, the court held that the affirmation by Mr Bradlaugh purportedly under section 4 of the Parliamentary Oaths Act 1870 was invalid as he was not a person falling within the scope of that section. Though the actual result was overturned on appeal in the House of Lords reported in (1883) 8 App Cas 354 as the court held that the person who brought the proceedings had no locus to do so, the finding on the invalidity of the affirmation was not disturbed. Stephen J accepted that the courts had the power to make such decision in Bradlaugh v Gossett, supra, at pp.281 to 282. Subsequently in AG v Bradlaugh (1885) 14 QBD 667, in proceedings brought by the Attorney General penalty was imposed by the Divisional Court on account of Mr Bradlaugh’s sitting and voting in the parliament despite his invalid affirmation and it was upheld on appeal to the Court of Appeal. Thus, the reference to the court’s inability to investigate into what happened within the walls of the Parliament should not be taken too literally. 78. Whatever the position may be in respect of the relationship between the courts and the House of Commons in 19th century England, for the reasons already given, the respective roles of the courts in Hong Kong and the LegCo as prescribed by the Basic Law are different. As the Chief Judge observed, LegCo is subject to the Basic Law. By reason of article 104, there is no right on the part of any elected member to assume office in LegCo without fulfilling the constitutional requirement under that article and the refusal or neglect to take the oath when duly requested would lead to vacation from office or disqualification. Hence, it is not a matter of internal discipline. Nor is it a matter of parliamentary privilege. Instead, it is a constitutional requirement which our courts have the constitutional responsibility to adjudicate on in the same way as the U.S. Supreme Court did in Powell v Mc Cormack, supra. 79. The above analysis also resolves the question as to identity of the decision maker under Section 21. Though the oath administrator would have the power to administer the oath under Section 19, he is not given any judicial role in the determination if an oath taker has declined or neglected to take the oath. In this connection, Section 21 does not say that an oath taker would be disqualified if he or she had, in the opinion of the oath administrator, declined or neglected to take the oath. I agree with the Chief Judge that paragraph 2(4) of the Interpretation does not confer any judicial role upon the oath administrator and any dispute on the application of Section 21 has to be determined by the courts. It also follows that in the determination of such dispute, the courts must conduct a full merit review with full opportunity for the parties to adduce relevant and admissible evidence on the issues instead of the limited review by way of traditional approach in a judicial review. 80. My analysis also reinforces the conclusion of the Chief Judge that Section 15 of the Legislative Council Ordinance is not exhaustive as to the circumstances in which a member is disqualified. Though the scope of Section 73 is of little significance in the present appeals (as 2 sets of proceedings have been commenced in respect of each Interested Party), it would be helpful if we can give some guidance on the interface between Section 73 and Section 21J. 81. In my view, these statutory jurisdictions have a common origin in the prerogative writ of quo warranto by which the authority of a person holding a public office could be challenged in courts. In Hong Kong, the prerogative writ has become obsolete and been largely (if not wholly) overtaken by statutory jurisdictions. Section 21J is a general provision, applicable to any public office created by any enactment whilst Section 73 specifically applies to a person acting or claiming to be entitled to act as a member of LegCo. Proceedings under Section 21J must be brought by way of judicial review, see Section 21K(1) of the High Court Ordinance and they are subject to the rules governing applications for judicial review. Leave is therefore required and it is subject to the usual time constraint in applications for judicial review. On the other hand, Section 73 is to be brought by way of some other form of civil proceedings and leave is not required. However, the time limit is 6 months: section 73(2) must be construed purposively as referring to 6 months from the date on which the person first acted or claimed to be entitled to act whilst disqualified, otherwise the time limit will be meaningless. Further, Section 73 proceedings can only be brought by an elector or the Secretary for Justice, see Section 73(1). In the case of proceedings by an elector, such proceedings would be stayed pending the payment of security for costs, a sum to be ordered by the court which shall not exceed $20,000 in respect of each member challenged, see Section 73(5) and (6). 82. Section 73(7) provides that proceedings within the scope of that section can only be brought in accordance with that section. In other words, it would not be open to persons coming within the scope of this section to bring proceedings by other means like an application for judicial review under Section 21J. 83. However, the Court of Final Appeal in Albert Ho v Leung Chun Ying v Ho Chun Yan, Albert (2013) 16 HKCFAR 735 held that election petition is not the only means to challenge an election and judicial review is an available alternative for those who are not eligible to bring proceedings under section 33 of the Chief Executive Election Ordinance. By the same token, it may be arguable that Section 73(7) does not preclude someone who is not an elector to commence proceedings under Section 21J in respect of a disqualified legislator. As the Chief Judge observed, the Chief Executive is not a person who is entitled to proceed under Section 73. He has to proceed under Section 21J to seek the remedy of declaration and injunction against the Interested Parties in HCAL 185/2016. 84. In any event, given that Section 73 was enacted to protect members of the LegCo against unlimited challenges to their offices (by way of security for costs and time limit for application), I believe even in cases where an applicant is outside the scope of that section and an application is brought by way of judicial review, the court must bear such protection in mind in assessing whether leave should be granted. In particular, given that judicial review should not be permitted when an alternative remedy by way of Section 73 is available, the court should consider whether an applicant who is not an elector in the constituency of the member concerned should be allowed to make the challenge and enquire whether the Secretary for Justice or any elector is willing to bring proceedings under Section 73. The court must also examine the locus standi of such an applicant carefully if neither the Secretary for Justice nor any elector in that constituency is willing to bring proceedings under Section 73. Hon Poon JA: 85. I agree with the judgment of the Chief Judge and the judgment of Lam VP. I would like to add a few words of my own on the principle of non-intervention in the Hong Kong context. 86. As the Court of Final Appeal pointed out in Leung Kwok Hung, supra, at [32], the common law principle of non-intervention is necessarily subject to the constitutional requirements in the Basic Law. As our written constitution, the Basic law is supreme. It binds all organs and branches of the government, including LegCo. The court is the guardian of the Basic Law. The court is constitutionally tasked to protect the provisions of the Basic Law and to ensure that all governmental organs and branches, including LegCo, stay within its bounds. 87. Being subject to the Basic Law, LegCo must uphold and abide by its provisions, always acting within its framework and fulfilling its constitutional requirements. So must all LegCo Members, whether they are elected from functional constituencies or geographical constituencies through direct elections. When disputes arise as to whether LegCo has acted in breach of the constitutional requirements in the Basic Law, such as passing a law which is contrary to certain provisions of the Basic Law, or whether individual LegCo Members have breached the constitutional requirements as mandated in the Basic Law, such as the present, the court has a constitutional duty to adjudicate and rule on the matters. In so doing, the court does not seek to undermine the authority or function of LegCo as the legislature or diminish the mandate that the electors gave to the LegCo Members concerned. Rather the court ensures that LegCo or the Members concerned exercise their powers lawfully in accordance with the constitutional requirements of the Basic Law. Only thus can the public confidence in LegCo be maintained. Only thus can the integrity of the very many important acts performed by LegCo be preserved. Hon Cheung CJHC: 88. Accordingly, the appeals are dismissed with costs to all other parties (including the LegCo President for his costs for the hearing on 17 November 2016) together with a certificate for 3 counsel. The costs order is made on a nisi basis and any application to vary it shall be dealt with by written submissions only. 89. As indicated at the conclusion of the hearing, we will hear any application for leave to appeal to the Court of Final Appeal tomorrow at 9:30 am. All time requirements are abridged accordingly, and we will accept undertakings to file and serve the formal documents in lieu of actual filing or service before the hearing so long as draft documents are made available to the court and the opposite parties in advance. Mr Benjamin Yu SC, Mr Johnny Mok SC, Mr Jimmy Ma and Mr Jenkin Suen, instructed by the Department of Justice, for the 1st and 2nd applicants in CACV 224 & 225/2016 and the 1st and 2nd plaintiffs in CACV 226 & 227/2016 Mr Hectar Pun SC and Mr Anson Wong Yu Yat, instructed by Ho Tse Wai and Partners, for the 1st interested party in CACV 224 & 225/2016 and the 2nd defendant in CACV 226 & 227/2016 Mr Philip Dykes SC and Mr Jeffrey Tam, instructed by Khoo & Co, for the 2nd interested party in CACV 224 & 225/2016 and the 1st defendant in CACV 226 & 227/2016 Lo & Lo, for the respondent in CACV 224 & 225/2016 and the 3rd defendant in CACV 226 & 227/2016, did not appear [1] The Interpretation reads: “1. ‘To uphold the Basic Law of the Hong Kong Special Administrative Region of the People’s Republic of China’ and to bear ‘allegiance to the Hong Kong Special Administrative Region of the People’s Republic of China’ as stipulated in Article 104 of the Basic Law of the Hong Kong Special Administrative Region of the People’s Republic of China, are not only the legal content which must be included in the oath prescribed by the Article, but also the legal requirements and preconditions for standing for election in respect of or taking up the public office specified in the Article. 2. The provisions in Article 104 of the Basic Law of the Hong Kong Special Administrative Region of the People’s Republic of China that ‘When assuming office’, the relevant public officers ‘must, in accordance with law, swear’ bear the following meaning: (1) Oath taking is the legal prerequisite and required procedure for public officers specified in the Article to assume office. No public office shall be assumed, no corresponding powers and functions shall be exercised, and no corresponding entitlements shall be enjoyed by anyone who fails to lawfully and validly take the oath or who declines to take the oath. (2) Oath taking must comply with the legal requirements in respect of its form and content. An oath taker must take the oath sincerely and solemnly, and must accurately, completely and solemnly read out the oath prescribed by law, the content of which includes ‘will uphold the Basic Law of the Hong Kong Special Administrative Region of the People’s Republic of China, bear allegiance to the Hong Kong Special Administrative Region of the People’s Republic of China’. (3) An oath taker is disqualified forthwith from assuming the public office specified in the Article if he or she declines to take the oath. An oath taker who intentionally reads out words which do not accord with the wording of the oath prescribed by law, or takes the oath in a manner which is not sincere or not solemn, shall be treated as declining to take the oath. The oath so taken is invalid and the oath taker is disqualified forthwith from assuming the public office specified in the Article. (4) The oath must be taken before the person authorized by law to administer the oath. The person administering the oath has the duty to ensure that the oath is taken in a lawful manner. He or she shall determine that an oath taken in compliance with this Interpretation and the requirements under the laws of the Hong Kong Special Administrative Region is valid, and that an oath which is not taken in compliance with this Interpretation and the requirements under the laws of the Hong Kong Special Administrative Region is invalid. If the oath taken is determined as invalid, no arrangement shall be made for retaking the oath. 3. The taking of the oath stipulated by Article 104 of the Basic Law of the Hong Kong Special Administrative Region of the People’s Republic of China is a legal pledge made by the public officers specified in the Article to the People’s Republic of China and its Hong Kong Special Administrative Region, and is legally binding. The oath taker must sincerely believe in and strictly abide by the relevant oath prescribed by law. An oath taker who makes a false oath, or, who, after taking the oath, engages in conduct in breach of the oath, shall bear legal responsibility in accordance with law.” [2] Article 79(1) of the Basic Law reads: “ The President of the Legislative Council of the Hong Kong Special Administrative Region shall declare that a member of the Council is no longer qualified for the office under any of the following circumstances: (1) When he or she loses the ability to discharge his or her duties as a result of serious illness or other reasons;” [3] Section 73 reads: “ (1) An elector, or the Secretary for Justice, may bring proceedings in the Court against any person who is acting, [or] claims to be entitled to act, as a Member on the ground that the person is disqualified from acting as such. ... (7) Proceedings against a person on the ground that the person has, while disqualified from acting ..., [acted,] or claimed to have been entitled to act, as a Member may be brought only in accordance with this section. (8) For the purposes of this section, a person is disqualified from acting as a Member if the person- (a) is not qualified to be, or is disqualified from being, a Member; or (b) has ceased to hold office as a Member.” [4] In any event, I take the view that the President has already in his ruling on 18 October 2016 decided, in substance (although not in so many words), that both Leung and Yau have declined or wilfully neglected to take the LegCo Oath. With respect, what he did not do (upon strong legal advice) was to follow that decision to its logical conclusion, and apply and act on section 21 accordingly. [5] Section 15(1) of the Legislative Council Ordinance reads: “(1) A Member’s office becomes vacant if the Member - (a) resigns in accordance with section 14 or is taken to have resigned from that office in accordance with section 13; or (b) dies; or (c) subject to subsection (2), alters either the Member's nationality or the fact as to whether the Member has a right of abode in a country other than the People's Republic of China as declared under section 40(1)(b)(ii); or (d) is the President and has been found under the Mental Health Ordinance (Cap. 136) to be incapable, by reason of mental incapacity, of managing and administering his or her property and affairs; or (e) is declared in accordance with Article 79 of the Basic Law to be no longer qualified to hold that office.” [6] Article 79(1) reads in Chinese: “香港特別行政區立法會議員如有下列情況之一，由立法會主席宣告其喪失立法會議員的資格： (一) 因嚴重疾病或其他情況無力履行職務；” [7] Section 21J reads: “(1) Where a person not entitled to do so acts in an office to which this section applies, the Court of First Instance may - (a) grant an injunction restraining him from so acting; and (b) if the case so requires, declare the office to be vacant. (2) This section applies to any public office or office which has been created by any enactment.” 
Chief Justice Ma (giving the Reasons of the Court): 1. The present appeal was brought by the Defendant against the Judgment of the Court of Appeal[1] dismissing her appeal from the First Instance judgment of Recorder Horace Wong SC.[2] Leave to appeal to this Court was granted by the Court of Appeal[3] under the former s 22(1)(a) of the Hong Kong Court of Final Appeal Ordinance Cap 484 on the basis that the subject matter of the appeal concerned property over the value of $1 million; in other words, an appeal under the previous “as of right” provision in that Ordinance.[4] 2. After hearing counsel for the Defendant appellant,[5] we dismissed the appeal with costs,[6] stating that the Reasons for Judgment would be handed down in due course. The appeal was wholly without merit. 3. The Plaintiff respondent is the mother of the Defendant, her daughter. At the time of the trial before the learned Recorder, she was aged 85. The case concerned the transfer of some properties by the Plaintiff to the Defendant in 1999. As we shall presently see, the determination of the factual disputes between the parties by the Recorder on the evidence before him was critical to the outcome of the litigation. The Court of Appeal upheld the trial judge’s findings of fact, so that there were concurrent findings of fact before this Court. As the parties were reminded, the Court does not, save in rare and exceptional cases, review concurrent findings of fact. It is no part of the Court of Final Appeal’s function to debate “yet again the factual findings made at first instance and previously reviewed by the Court of Appeal” in the hope on the part of an appellant of finally obtaining favourable findings.[7] This was decisive in our dismissing the Defendant’s appeal. We shall briefly explain why the Court arrived at this conclusion. 4. The dispute between the parties concerned two properties (“the properties”): a shop located in Che Cheung Building in Kowloon (“the shop”) and a flat located in Ha Heung Road, also in Kowloon (“the flat”). These properties were at one stage owned by the Plaintiff’s husband, Tam Yuk Lam (“Tam”), together with another property in Manning Theatre Building (“the Manning Theatre Building property”). The Plaintiff and Tam had four children (three sons and their daughter, the Defendant), all born in the Mainland where the Plaintiff once lived. In 1982, the Plaintiff and the Defendant settled in Hong Kong, the three sons remaining in the Mainland. At the time, Tam was already living in Hong Kong and had taken a concubine, a Madam Lou Nui[8], with whom he had three children. 5. After the flat was purchased by Tam in 1986, the Plaintiff began to live there, with Tam staying there every other day. The Defendant, who by this time had married, did not live in the flat. She married in 1984 and had three children of her own. In 1995, she divorced her husband. From July 1984, she began to live on Comprehensive Social Security Assistance (“CSSA”). 6. In August 1997, having suffered a stroke in 1994, Tam divided his properties between the Plaintiff and Lou Nui. The shop and the flat were transferred to the Plaintiff while the Manning Theatre Building property was transferred to Lou Nui. 7. In relation to the transfer of the properties to the Defendant in 1999, according to land search records, the shop was assigned by the Plaintiff to the Defendant on 22 September 1999 and the flat was assigned to the Defendant on 5 October 1999. The Assignments were prepared by an executive in the offices of Messrs George YC Mok and Co. (“GYCM”). The parties advanced at trial different versions of the circumstances surrounding these Assignments :- (1) According to the Plaintiff, one day in early 1999, the Defendant went to the flat to show her a letter which indicated that Tam intended to reclaim both properties and to transfer them to Lou Nui. This letter was not produced at trial. The Plaintiff was told by the Defendant that Tam had given her ten days to move out. When the Plaintiff became worried at this, the Defendant then suggested that if the properties were to be transferred into her (the Defendant’s) name, Tam would not then be able to claim the properties for himself. The Plaintiff agreed. There was, however, no intention on her part to transfer the beneficial title in the properties to the Defendant; it was merely a device to avoid Tam’s threatened claims. (2) The Defendant’s version was that the Plaintiff intended to make a gift of the properties to her. According to the Defendant, it was the Plaintiff who told her about the letter and Tam’s intentions regarding the properties. The Plaintiff said she wanted to make a gift of them to her for essentially two reasons: first, Tam had already made provision for the Defendant’s brothers but not her and therefore the Plaintiff wanted to ensure that the Defendant was also looked after; secondly, in any event the Plaintiff would rather her daughter benefited from having the properties than Lou Nui. The Defendant claimed she was at first reluctant to agree to this transfer (albeit the beneficial interest in the properties would pass to her) since she would no longer be able to receive CSSA. It was only when the Plaintiff agreed to let her have the rental incomes for the properties as well[9] (less $1,000 for the Plaintiff’s own use) that the Defendant was persuaded to agree to the transfer. 8. The Assignments of the properties showed on their face a purchase by the Defendant of the Plaintiff’s interest in them. The consideration for the shop was said to be $300,000, the consideration for the flat, $500,000. The Plaintiff also signed receipts for these amounts. It was common ground that the Plaintiff did not receive these (or any) amounts from the Defendant, evidently so because on both sides’ cases, the Assignments did not intend and were never intended to reflect the true arrangements between them. This is a point of some importance as it effectively disposed of one of the Defendant’s arguments sought to be raised in this appeal, namely, that the Plaintiff was estopped by deed from denying the contents of the two Assignments. 9. Subsequent to these transfers, the Plaintiff continued to live in the flat, paying the various expenses associated with the flat such as rates and telephone bills. She also (as the Recorder eventually found) paid the Defendant the rental income of the two properties less $1,000 for herself. From 2003 to 2009, the Plaintiff lived intermittently in the flat, dividing her time between Hong Kong and the Mainland (where her sons lived). She lived in Huidong from June 2007 to June 2009. When she returned to Hong Kong, she found that the locks to the door in the flat had been changed. In August 2010, the Defendant sold the flat to a third party, Land Crown International Limited, for $2,988,000, with an additional $1,328,000 as “decoration and removal expenses”. 10. The trial judge heard evidence[10] from 4 witnesses: the Plaintiff, the Defendant, and a conveyancing executive and solicitor from GYCM who had handled the assignment of the properties. In a 48-page judgment, the Recorder analyzed the evidence, assessed the credibility of the witnesses and concluded that the Plaintiff’s intention in making the Assignments was merely to avoid Tam’s threatened actions (accepting the Plaintiff’s version) but not to make a gift of the properties to the Defendant (rejecting the Defendant’s version). The legal effect was that the properties were held by the Defendant on resulting trust for the Plaintiff. Apart from making declarations to this effect, the Recorder also ordered the Defendant to provide an account of her dealings with the properties and that there should be the liberty to apply in relation to both the declarations and the order for an account.[11] 11. In his judgment, Recorder Wong SC gave detailed reasons for his conclusions. It is not necessary to go through them in detail; they are amply set out in the written judgment. He found the Plaintiff to be a credible witness but doubted the credibility of the Defendant and found her to be evasive. Wewould only by way of example highlight some of the reasoning:- (1) There was no sensible reason for the Plaintiff to make an outright gift of the properties to the Defendant. She had four children and the Recorder saw no justification for the Plaintiff to favour the Defendant above her brothers. There was, moreover, no evidence to suggest that Tam had made any provision for the sons as the Defendant had alleged. (2) While the Recorder found that the Plaintiff allowed the Defendant to keep the rental income from the properties subsequent to the Assignments, it did not however make sense to divest herself of the beneficial interest in the properties. They were, after all, her only source of wealth, she having no other income. She also continued to pay the rates and other expenses of the flat. (3) The Defendant’s version was found to be unconvincing. In particular, the Recorder found it difficult to believe the Defendant when she claimed to hesitate to receive a gift of the properties on the ground she would then lose her entitlement to CSSA: as the Judge remarked, the value of the properties far outweighed any CSSA to which she may have been receiving.[12] Further, on the Defendant’s version, her hesitation was eventually overcome only when the Plaintiff agreed to let her have the rental income for the properties (less $1,000). As stated earlier, the Judge did find as a fact that the Defendant was permitted to keep the rental income, but this was clearly consistent with the Plaintiff’s version of events. It was however inconsistent with the Defendant’s version: if she was indeed to take the beneficial interest in the properties, it could not have been any additional incentive to have the rental income as well. She would surely be entitled to such income as beneficial owner. 12. The Court of Appeal could not fault the Recorder’s analysis of the evidence. This is conveniently summarized in one paragraph in Cheung JA’s judgment (with whom the other Judges agreed) :- “8.3 The defendant is challenging the Recorder’s finding of fact. The principle is clear that the appellate court will not disturb a finding of fact by the trial Judge who had the benefit of hearing and observing the evidence at first hand, unless the finding is plainly wrong. In our view the defendant has failed to overcome this high threshold. As the Recorder had made it plain, he decided the issue without reliance on presumptions. His express finding is that the plaintiff did not intend to benefit the defendant when she transferred the properties to her and it was solely done to avoid the properties coming into the hands of Lou Nui. He observed that there really was no reason why the plaintiff should suddenly give the properties to the defendant in 1999 when they were her only valuable assets and she was dependent on them. He also found that the defendant was aware of the plaintiff’s intention at that time.” 13. There being concurrent findings of fact along these lines, the Judge and the Court of Appeal were fully entitled to reach the legal conclusion they did, namely, that despite the Assignments, the Defendant held the properties on resulting trust for the Plaintiff. 14. Despite recognizing this difficulty, Mr Chu sought first to maintain before us that the analysis and assessment of the evidence by the Recorder was faulty and that his erroneous approach had been perpetuated by the Court of Appeal. This was hopeless and amounted to no more than a third attempt by the Defendant to seek to persuade a court to try the case again. 15. Various attempts were made in the Appellant’s written Case and in the oral submissions before us to raise legal issues to undermine the decisions of the lower courts. They were all, in ourview, equally hopeless (with some points all but abandoned when Mr Chu made his submissions before us). Weneed only deal with them briefly:- (1) First, it was said that a resulting trust had not been properly pleaded by the Plaintiff. Both the trial Judge and the Court of Appeal rejected this argument and were right to do so. The Statement of Claim specifically pleaded that the properties were assigned to the Defendant at her suggestion to avoid Tam’s claims and that accordingly, the Defendant held them as trustee for her.[13] The Defendant was under no misapprehension as to what claim she was facing. In her Defence and Counterclaim, in denying the Plaintiff’s claim, she pleaded facts to make out her case that the assignment of the properties represented a gift. The object of pleadings is “fairly and precisely to inform the other party or parties in the litigation of the stance of the pleading party (in other words, that party’s case) so that proper preparation is made possible, and to ensure that time and effort are not expended unnecessarily on other issues.”[14] The Plaintiff’s pleadings in the present case did exactly that. (2) Next, in the Appellant’s Case, Mr Chu argued that the Recorder and the Court of Appeal failed to consider the impact of Chinese customary law on the issue whether the Plaintiff ought to have been believed when she said she only agreed to transfer the properties to the Defendant so as to put them out of Tam’s reach. The point sought to be derived from this reliance on Chinese customary law was that the threat on the part of Tam to reclaim the properties was an empty and legally unsustainable one. This was an entirely new point raised for the first time in this Court (and unpleaded as well) without the benefit of the views (not to mention, assessment) by the Recorder or the Court of Appeal.[15] It was also pointless: it will be remembered that as part of the defendant’s own case, there was a threat by Tam to reclaim the properties.[16] It was immaterial whether this claim had merit or not. It was enough that the threat was regarded by both Plaintiff and Defendant as a real one. (3) Thirdly, the Defendant relied on estoppel by deed, meaning that the Plaintiff was bound by the terms of the two Assignments of the properties as well as the receipt clauses in which the Plaintiff confirmed receipt of the purchase monies. Quite apart from the principle that the law will not permit instruments such as an assignment to be used as an instrument of fraud,[17] the Defendant’s own case at trial was that the Assignments did not represent what they purported to say. (4) Fourthly, before us, Mr Chu sought to argue that the Plaintiff had not come to court with clean hands. This was again a point never raised before and therefore not developed or dealt with by the courts below. 16. Finally, weshould just briefly mention the presumption of advancement. Both the Judge and the Court of Appeal discussed this doctrine at some length before coming to the view that in any event the presumption, even if it applied, was displaced by the evidence before the court. Interesting though the question may be, in particular whether a modern view should be taken of the presumption to include transfers from mother to daughter (this presumption, if it applies, is of course usually relevant only for evidential purposes), a detailed consideration was and is simply not called for in the present case. We would reserve a detailed consideration of this doctrine for another occasion. 17. For these reasons, the appeal was dismissed. Mr George Chu, instructed by Joseph Li & Co, assigned by Director of Legal Aid, for the Defendant / Appellant Ms Josephine Tjia, instructed by Tsangs, assigned by Director of Legal Aid, for the Plaintiff / Respondent [1] Dated 4 July 2014 (Cheung & Chu JJA and Mimmie Chan J). [2] Dated 15 August 2013. [3] On 29 December 2014. [4] Section 22(1)(a) of the Ordinance was repealed by s 8(1) of the Administration of Justice (Miscellaneous Provisions) Ordinance 2014. Although the repeal of the provision took effect as from 24 December 2014, it only applied to final judgments of the Court of Appeal after that date: see s 7 of the 2014 Ordinance. [5] Mr George Chu. Ms Josephine Tjia appeared for the Plaintiff respondent. Though not called upon to make oral submissions, we have taken into account the Respondent’s Case drafted by her. [6] As both parties were legally aided, we also made an order that the costs of the plaintiff and the defendant should be taxed in accordance with the Legal Aid Regulations. [7] See Sky Heart Ltd v Lee Hysan Co Ltd (1997-98) 1 HKCFAR 318, 333I-338D; Chinachem Charitable Foundation Ltd v Chan Chun Chuen (2011) 14 HKCFAR 798, at para 57. [8] Lou Nui has been referred to throughout the proceedingsas a concubine. If this was indeed her status, this must have been attained before 7 October 1971 : see ss 2 and 5(1) of the Marriage Reform Ordinance Cap 178. [9] The evidence showed that the shop was leased out. The Plaintiff lived in only a part of the flat. The other parts were rented out. [10] In a trial lasting 4 days. [11] The reason was that, as we have seen, subsequent to the Assignments, in August 2010 the Defendant sold the flat to a third party. The proceeds of sale are held by GYCM as stakeholders pending the outcome of the present case. [12] Adopting the figure of $8,000 (which Defendant said she was receiving by way of CSSA every month) and taking into account the figure of $4,316,000 alone (the consideration for the sale of the flat in 2010), this purchase price represented about 45 years in CSSA payments. [13] The Statement of Claim pleads a constructive trust but it is clear that what was meant was a resulting trust. [14] Kwok Chin Wing v 21 Holdings Ltd (2013) 16 HKCFAR 663, at para 21 referring to Wing Hang Bank Ltd v Crystal Jet International Limited [2005] 2 HKLRD 795. [15] The principles of Flywin Co Ltd v Strong and Associates Ltd (2002) 5 HKCFAR 356 therefore apply. [16] See para 7(2) above. [17] See, among many authorities, Booth v Turle (1873) LR 16 Eq 182. 
The Court: 1. These two appeals, heard together, raise overlapping issues concerning the offences of unlawful assembly and riot under sections 18 and 19 of the Public Order Ordinance (“POO”).[1] 2. The appellant Lo Kin Man (“Lo”) was convicted after trial before A Pang J and a jury[2] of the offence of riot contrary to section 19 of the POO involving an occurrence of public disorder in Portland Street, Mongkok, on 8 and 9 February 2016. His appeal to the Court of Appeal was dismissed.[3] Leave to appeal to this Court was granted by the Appeal Committee[4] in respect of the questions of law and the “substantial and grave injustice” ground set out in Annex 1 to this Judgment. We shall refer to the proceedings leading to the present appeal as “the Lo Case”. 3. In relation to the second appeal, Tong Wai Hung (“Tong”) was charged alongside two other defendants with unlawful assembly and riot under POO sections 18 and 19 in connection with a riot which occurred some three years later amidst the widespread and serious social unrest which has been described in Kwok Wing Hang v Chief Executive in Council.[5] The riot in question took place on 28 July 2019 at Des Voeux Road West. He was acquitted after trial before HH Judge Anthony Kwok in the District Court.[6] This prompted the Secretary for Justice to refer two questions of law to the Court of Appeal for its opinion pursuant to section 81D of the Criminal Procedure Ordinance (“CPO”).[7] Having been provided with its opinion,[8] Tong applied for certification of the two abovementioned questions for leave to appeal to this Court. The Court of Appeal was aware of the grant of leave in the Lo Case and, noting the overlapping issues, granted certification.[9] By consent, the Appeal Committee granted leave to appeal in respect of the said questions which are set out in Annex 2 to this judgment. The proceedings leading to the present appeal are referred to as “the Tong Case”. 4. The questions raised in the two Annexes concern a proper understanding of the structure and elements of the offences created by sections 18 and 19 and their relationship with certain common law doctrines. In particular, issues arise as to (i) the existence and nature of a requirement (if any) for proof of a “common purpose” shared by the defendant and other persons assembled;[10] (ii) the applicability of the doctrine of joint enterprise to the two statutory offences and whether that doctrine enables liability to be established without the defendant being present at the scene;[11] and (iii) whether a defendant can be found guilty under sections 18 and 19 on the basis of “encouragement through [the defendant’s] presence”, without committing acts specifically prohibited by those sections.[12] Under the “substantial and grave injustice” heading, a question is raised as to the consequences of the indictment having made no mention of any participants in the riot other than the named co-defendants who were not convicted after trial together with the defendant.[13] A. The structure and elements of the offences under sections 18 and 19 A.1 The sections 5. A textual examination of the statutory offences is the appropriate starting-point. Indeed, as the relevant provisions form part of a codifying Ordinance, the primary emphasis should be on giving effect to their terms, properly construed, rather than delving too extensively into pre-existing common law authority. As Lord Herschell explained in The Governor and Company of the Bank of England v Vagliano Brothers:[14] “I think the proper course is in the first instance to examine the language of the statute and to ask what is its natural meaning, uninfluenced by any considerations derived from the previous state of the law, and not to start with inquiring how the law previously stood, and then, assuming that it was probably intended to leave it unaltered, to see if the words of the enactment will bear an interpretation in conformity with this view. If a statute, intended to embody in a code a particular branch of the law, is to be treated in this fashion, it appears to me that its utility will be almost entirely destroyed, and the very object with which it was enacted will be frustrated. The purpose of such a statute surely was that on any point specifically dealt with by it, the law should be ascertained by interpreting the language used instead of, as before, by roaming over a vast number of authorities in order to discover what the law was, extracting it by a minute critical examination of the prior decisions, dependent upon a knowledge of the exact effect even of an obsolete proceeding such as a demurrer to evidence.” 6. POO section 18 (unlawful assembly) provides as follows: “(1) When 3 or more persons, assembled together, conduct themselves in a disorderly, intimidating, insulting or provocative manner intended or likely to cause any person reasonably to fear that the persons so assembled will commit a breach of the peace, or will by such conduct provoke other persons to commit a breach of the peace, they are an unlawful assembly. (2) It is immaterial that the original assembly was lawful if being assembled, they conduct themselves in such a manner as aforesaid. (3) Any person who takes part in an assembly which is an unlawful assembly by virtue of subsection (1) shall be guilty of the offence of unlawful assembly and shall be liable – (a) on conviction on indictment, to imprisonment for 5 years; and (b) on summary conviction, to a fine at level 2 and to imprisonment for 3 years.” 7. And section 19 (riot) states: “(1) When any person taking part in an assembly which is an unlawful assembly by virtue of section 18(1) commits a breach of the peace, the assembly is a riot and the persons assembled are riotously assembled. (2) Any person who takes part in a riot shall be guilty of the offence of riot and shall be liable- (a) on conviction on indictment, to imprisonment for 10 years; and (b) on summary conviction, to a fine at level 2 and to imprisonment for 5 years.” 8. As was done in HKSAR v Leung Chung Hang Sixtus,[15] it is convenient for the purposes of exposition to divide up and number (in square brackets) the elements making up the two offences as follows: Unlawful assembly (section 18) [1] When 3 or more persons, [2] assembled together, [3] conduct themselves in a disorderly, intimidating, insulting or provocative manner [4] intended or likely to cause any person reasonably to fear that the persons so assembled will commit a breach of the peace, or will by such conduct provoke other persons to commit a breach of the peace, [5] they are an unlawful assembly. [section 18(1)] [6] It is immaterial that the original assembly was lawful if being assembled, they conduct themselves in such a manner as aforesaid. [section 18(2)] [7] Any person who takes part in an assembly which is an unlawful assembly by virtue of subsection (1) shall be guilty of the offence of unlawful assembly ... [section 18(3)] Riot (section 19) [8] When any person taking part in an assembly which is an unlawful assembly by virtue of section 18(1) [9] commits a breach of the peace, [10] the assembly is a riot and the persons assembled are riotously assembled. [section 19(1)] [11] Any person who takes part in a riot shall be guilty of the offence of riot ... [section 19(2)] A.2 Unlawful assembly 9. Elements [1] to [4] are the constituent elements of an unlawful assembly. Where they are established, the assembly is (as [5] declares) an unlawful assembly. Element [1] requires there to be 3 or more persons [2] assembled together, who conduct themselves in the disorderly manner specified in [3], with the intended or likely consequences stated in [4]. Those persons will be referred to as “the constituent offenders” and their conduct specified in [3] and [4] as “the prohibited conduct”. The true construction of element [4] is dealt with in detail in HKSAR v Leung Chung Hang Sixtus.[16] Notably, the “likely” limb is an aspect of the actus reus requiring the objectionable conduct, objectively assessed, to be likely to produce the reasonable fear of a breach of the peace specified.[17] 10. Element [6] reflects the scheme of graduated liability. An assembly might start off as lawful, then turn into an unlawful assembly and then become a riot. The time such a transformation may take will vary according to the circumstances. An unlawful assembly could develop almost immediately into a riot. 11. Element [7] is the offence-creating provision. Any person who “takes part” in an assembly which is an unlawful assembly commits the section 18 offence. The actus reus is “taking part”. This may or may not involve the same acts as, and should not be confused with, the prohibited conduct of the constituent offenders referred to in [3] and [4]. 12. If the defendant was one of the constituent offenders, by engaging in the prohibited conduct, he or she will (along with other constituent offenders) have taken part in the unlawful assembly that they will together have brought into being. But the defendant need not be one of the constituent offenders and can “take part” by joining in later. This is clear from element [7] which provides that the offence is committed by any person who takes part in the unlawful assembly. The submissions of Ms Gladys Li SC[18] to the contrary are rejected. 13. What then constitutes “taking part” in the unlawful assembly? What acts must the defendant perform? The sections do not spell out the meaning of those words. As a matter of textual analysis, it is implicit that if the defendant is one of the constituent offenders whose conduct falls within [3] and [4] he or she would “take part” since it is by such acts that the unlawful assembly is established, being of the essence of the offence. And if the defendant was not among the constituent offenders, but joined in and similarly did acts prohibited by [3] and [4], he or she would also be found to have “taken part”. 14. However, the offence is not confined to such conduct. As a matter of language, “taking part” is a broad expression. In our view, those words also embrace conduct which does not itself fall within [3] and [4] but which involves the defendant facilitating, assisting or encouraging the performance of such conduct by others participating in the assembly. Such conduct would traditionally give rise to accessorial liability but, by offering such facilitation, assistance or encouragement, the defendant acts in furtherance of the prohibited conduct and may thus also be regarded as “taking part” in the unlawful assembly. In so doing, he or she may attract liability either as a principal offender under section 18 or an aider and abettor. 15. An important feature which emerges from the statutory language is that unlawful assembly is what might be called a “participatory offence”. Thus, the offence requires the constituent offenders who are “assembled together” to “conduct themselves” in the prohibited manner so that the intended or likely fear of a breach of the peace is fear of what “the persons so assembled” will do. Element [6] draws the line between lawful and unlawful assemblies by reference to persons who “being assembled” engage in the prohibited conduct. The offence is committed by someone “taking part” in the unlawful assembly. 16. The defendant must therefore be shown not merely to have been engaging in disorderly conduct alone, but to have acted as part of an assembly with others who were also participants. The offence is “participatory” in that sense. Such participation is a requirement recognised by Lam JA (as Lam PJ then was) in SJ v Leung Kwok Wah,[19] holding that the defendant’s conduct has to be assessed to see if a sufficient nexus with other participants exists to justify regarding them as acting together.[20] His Lordship held that their conduct has to justify the inference that they had what he called “a common purpose in acting in the statutorily prescribed manner;”[21] ie, a shared objective of engaging in the ‘prohibited conduct’ forming elements [3] and [4]. 17. The defendant must accordingly intend to take part in, that is, become part of, the unlawful assembly, being aware of the related conduct of other participants and intending, while assembled together with them, to engage in or act in furtherance of the prohibited conduct. The defendant must, in other words, have what we will call a “participatory intent”. 18. It follows (leaving aside for now accessory and inchoate liability (especially regarding liability for conspiracy and incitement) that may be incurred by a person who is absent[22]), that to be guilty under section 18 as a principal offender, the defendant has to be present as part of the assembly, together with other participants. A.3 Riot 19. The offence of riot builds on that of unlawful assembly. Its starting-point is that an unlawful assembly exists, ie, that elements [1] to [5] are established. Element [8] specifies that when any person taking part in the unlawful assembly [9] commits a breach of the peace, the assembly [10] becomes a riot and the people assembled are riotously assembled. The actus reus of the offence of riot under section 19 is committed [11] when any person “takes part” in the riot. 20. Thus, as with unlawful assembly, the offence of riot has its initial constituent elements which are distinct from the actus reus of “taking part in a riot”. Any person taking part in an unlawful assembly may turn that assembly into a riot by committing a breach of the peace. That person does not have to be one of the constituent offenders who initially constituted the unlawful assembly, but he or she has to be a participant in the unlawful assembly. Any person who “takes part” in a riot which has come into existence commits the offence of riot [11]. Such a person does not have to be the person responsible for the initial, constituent breach of the peace. Neither does he or she have to have taken part in the unlawful assembly prior to it turning into a riot. Any person may take part by joining in after the unlawful assembly has become a riot. 21. Mirroring the analysis regarding unlawful assembly, the defendant’s conduct amounting to “taking part” in the riot must involve acts in furtherance of the riot. It must involve committing breaches of the peace or doing acts facilitating, assisting or encouraging breaches of the peace by others, making the defendant guilty as a principal or as an aider and abettor. We will return below[23] to consider what committing a breach of the peace entails. 22. The offence of riot is also participatory in nature. The constituent act involves a breach of the peace by a person who was taking part in an unlawful assembly. When that occurs, the “persons assembled” are declared by element [10] to be “riotously assembled”. The person who first committed a breach of the peace would thus be taking part in an assembly which has become a riotous assembly. Any other person who commits the offence by “taking part” in the riot, does so as part of the “riotous assembly”. Such person must have a participatory intent, intending to take part in the riot by committing, or acting in furtherance of, breaches of the peace together with other participants engaged in riotous activities. 23. As with unlawful assembly, it follows that liability for the offence of riot as a principal implicitly requires the defendant to be present and acting with the others riotously assembled. Again, this is subject to what is said about accessorial and inchoate liability incurred by persons who are absent, discussed below.[24] 24. It should be noted that the Court of Appeal in the Lo Case[25] and Poon CJHC writing for the Court in the Tong Case[26] omitted reference to the offence-creating, “taking part” provisions when setting out sections 18 and 19. This failure to focus on “taking part” as the actus reus of the two offences appears to have had an important influence on their approach to joint enterprise to which we shall return.[27] B. “Common purpose” 25. In the Lo Case, the appellant’s propositions on “common purpose”, reflected in Annex 1, Questions 1a-1d, may be summarised as follows: (a) The offences of unlawful assembly and riot were common law offences. At common law, it was necessary for the prosecution to prove against the defendant all the elements of the unlawful assembly which then became a riot. These included the prohibited disorderly conduct and a shared intent to engage in such conduct (what we have called a “participatory intent”). (b) Additionally, so the argument runs, it was necessary to prove that the assembled persons had a “common purpose” which they jointly intended to pursue, such as to wreck a dinner or cause a work stoppage.[28] Such a purpose may be referred to as an “extraneous common purpose” which is distinct from and goes beyond the participatory intent to engage in the prohibited conduct or to commit breaches of the peace. It is “extraneous” in that it is not a purpose which relates to any element of the offences but involves an external objective motivating the offenders. Lo’s submission is that the common law required such an extraneous purpose, which might or might not be lawful in itself, to be shared in common by those taking part in the unlawful assembly or riot. (c) It is argued that an extraneous common purpose continues to be required by the POO as a necessary element of both unlawful assembly and riot. The Judge, it is said, erred in failing to recognise, and give directions on, this additional element. 26. The prosecution’s position, adopted by the Judge in her directions to the jury and upheld by the Court of Appeal,[29] is that the prosecution has to prove that the defendant, assembled together with other persons, acted with a participatory intent, intending together with others assembled to behave in the prohibited manner or to commit or further the commission of the prohibited acts or breaches of the peace; but that there is no extraneous common purpose to be proved. 27. The competing arguments raise two questions: (i) What was the position regarding “common purpose” at common law? (ii) To what extent is the common law position still applicable after passage of the POO in 1967? B.1 Common purpose at common law 28. The common law authorities were in some disarray. Some, for instance, referred to the persons assembled having an extraneous common purpose but without necessarily suggesting that such a purpose was an essential ingredient of the offences. Their emphasis was rather on the use of unlawful means to achieve such purpose. 29. Thus, in R v Graham and Burns,[30] many persons assembled with the common purpose of holding a public meeting in Trafalgar Square to demand the release of certain Irish politicians, but the emphasis in Charles J’s direction to the jury (which has often been cited) was on the unlawfulness of the means adopted to pursue that objective: “It has not been suggested that the object of that meeting was not perfectly innocent. There is no doubt whatever that a meeting of British citizens can be held to protest, if they think fit, against the imprisonment of Mr O’Brien, but that is not the point… [The] question is, whether the mode in which it was attempted to assert the right of meeting in Trafalgar-square was or was not unlawful, that is the question you have to consider and decide.”[31] 30. In Field and Others v The Receiver of Metropolitan Police,[32] a number of youths conducted themselves in a disorderly way, pushing against a wall, leading to its collapse. As soon as it fell the caretaker of the premises came out into the street and the youths then dispersed in different directions. Phillimore and Bray JJ conducted a review of the early English authorities, including Lord Coke, Blackstone, Hawkins, Stephen and Lord Holt CJ which revealed markedly different views. Their Lordships concluded from the passages cited: “... that there are five necessary elements of a riot—(1.) number of persons, three at least; (2.) common purpose; (3.) execution or inception of the common purpose; (4.) an intent to help one another by force if necessary against any person who may oppose them in the execution of their common purpose; (5.) force or violence not merely used in demolishing, but displayed in such a manner as to alarm at least one person of reasonable firmness and courage.”[33] 31. Their Lordships therefore held that a riot required proof of a common purpose and its execution “or inception” (elements (2) and (3)). While they considered that there was evidence to justify the judge’s finding that those elements were satisfied, they commented that they “should not have found the same way.” However, the appeal was allowed on the basis that elements (4) and (5) had not been established. The youths had run away when the caretaker emerged so that there was no evidence of any “intent to help one another by force if necessary against any person who may oppose them in the execution of their common purpose”. [34] 32. In R v Caird,[35] the Cambridge riot might be said to have involved the extraneous common purpose of wrecking the restaurant and Greek-themed dinner for political motives. However, Sachs LJ stressed that what mattered was the violence of the protesters’ conduct: “When there is wanton and vicious violence of gross degree the Court is not concerned with whether it originates from gang rivalry or from political motives. It is the degree of mob violence that matters and the extent to which the public peace is being broken.”[36] 33. In R v John McKinsie Jones,[37] a case where picketers who were trying to force a work stoppage visited several different sites sequentially, James LJ stated: “The ingredients of the offence are (i) the actus reus of being or coming together—the assembly, and (ii) the mens rea involved in the intention of fulfilling a common purpose in such a manner as to endanger the public peace.” So again, while a common purpose was mentioned, it was the manner in which it was pursued, endangering the public peace, that was important. 34. McHugh JA (as His Honour then was) when sitting in the New South Wales Court of Appeal,[38] examined the English authorities and stated: “Despite the antiquity of the offence, the precise elements of the offence of riot are not settled. But there is almost unanimous agreement that an essential element is that the rioters must have an intention mutually to assist one another against any person who opposes them.”[39] 35. The mutual assistance element is derived from Hawkins, Pleas of the Crown (1st ed, 1716), Bk 1, p 155. The judgment holds that it was essential for the assembled persons to have a common purpose and an intention to render such mutual assistance: “... the authorities strongly support the proposition that it is an essential element in the offence of riot that each of the participants had the intention to help each other, by force if necessary, against any person who might oppose them in the execution of their common purpose. It necessarily follows that, if a number of persons are charged with the one offence of riot, they must have each had both a common purpose and an intention to help each other, by force if necessary, against any person who might oppose them in execution of that common purpose.”[40] 36. When the English Law Commission reported on Offences Relating to Public Order,[41] it stated, “There is no agreement as to precisely what constitutes an unlawful assembly” and counselled against relying on the authority of the older institutional writers. In relation to “common purpose”, they adopted the alternatives forming part of a working definition of the offence put forward by Smith & Hogan,[42] namely: “(1) An assembly of three or more persons; (2) a common purpose (a) to commit a crime of violence or (b) to achieve some other object, whether lawful or not, in such a way as to cause reasonable men to apprehend a breach of the peace.” 37. Subsequently, the common law offences of unlawful assembly and riot (as well as rout and affray) were abolished and replaced by new statutory offences under the Public Order Act 1986 in England and Wales. Notably, the offence of riot as defined by section 1 is differently structured from the offence under POO section 19 and expressly requires proof of a “common purpose” to be inferred from conduct.[43] B.2 “Common purpose” and sections 18 and 19 38. In our jurisdiction, statutory changes were introduced somewhat earlier by the Public Order Bill 1967 (“POB”).[44] At the Bill’s First Reading, the Attorney-General[45] explained that it was a codifying Bill, stating that it was: “... intended to be a comprehensive piece of legislation dealing with all aspects of public order ... [and] more than a mere consolidation of existing laws ... the opportunity [having] been taken to expand certain provisions and to fill in some gaps ...” He added: “Some of the requirements of the common law in relation to these offences were rather technical and ill-adapted to modern conditions. Although the basic principles of unlawful assembly and riot have been retained therefore certain modifications have been made.” 39. The Objects and Reasons of the Bill[46] make it clear that sections 18 and 19 “replace the common law” as part of the codification exercise. In particular, the intention is to eliminate any requirement for proof of a “common purpose”: “Clauses 18 and 19 introduce provisions dealing with unlawful assemblies and riots and replace the common law in this respect. The proposed statutory provisions are not identical with the common law, which is not adequate in certain respects for local circumstances. Firstly, there will no longer be the necessary element of common purpose before a crowd is an unlawful assembly or a riot. Secondly, an unlawful assembly will become a riot if any person taking part in the assembly commits a breach of the peace.” (Italics supplied) 40. The drafters of the POO were evidently aware of the uncertainties concerning the element of “common purpose” at common law. As we have seen, a common purpose – in the sense of an extraneous common purpose – is not one of the elements of either offence as presently enacted. The Objects and Reasons indicate that this was a deliberate feature of the codification. Thus, it is preferable not to refer to “common purpose” but to recognise instead the requirement of a participatory intent, reflecting the participatory nature of the two offences. It is in any event clear that no requirement for proof of an extraneous common purpose exists. 41. As previously noted, that was the approach to section 18 adopted by Lam JA (as Lam PJ then was) in SJ v Leung Kwok Wah.[47] His Lordship held that the elements of the offence reflected what he called the “corporate” nature of the offence: “... the conduct of the defendants had to be assessed together to see whether this criterion can be satisfied. There must be a sufficient nexus between the conduct of these defendants to justify having them considered together. And the fear required is that such persons so assembled, viz acting together, will commit a breach of the peace.”[48] 42. He added: “... if three persons in a lawful assembly committed acts of the prescribed nature at different parts of the place of assembly for different purposes, sparking off different incidents, involving and affecting an entirely different mix of persons, there would not be a sufficient nexus to turn these independent acts into an unlawful assembly of those three persons.”[49] 43. His Lordship was thus concerned with identifying the participatory requirements of unlawful assembly as a matter of statutory construction and held that the “corporate nature” of the offence entailed proof of a “common purpose” in the sense of what we have called a “participatory intent”. This is how his statement that “the requirement of having a common purpose in acting in the statutorily prescribed manner remains good law in dealing with a charge under s 18”[50] should be understood. He was certainly not suggesting that an additional element involving proof of an extraneous common purpose is required. 44. A Pang J below adopted the same approach, stating in her written directions that a necessary ingredient was that the relevant defendant “assembled together with those people for the common purpose of conducting themselves in a disorderly, intimidating, insulting or provocative manner”.[51] She was thus referring to an intention to take part in the unlawful assembly by doing the prohibited acts, not to an extraneous common purpose. 45. The Court of Appeal endorsed Lam JA’s approach[52] stating (in translation): “According to the true construction of s 18(1), the actus reus that constitutes unlawful assembly must have corporate nature to satisfy the requirement of offenders being jointly responsible. The offenders must have assembled together and conducted themselves in the manner prescribed in the ordinance when assembling. The offenders must have a common purpose so that the court can regard them as assembling to act together. Therefore, the requirement of having a common purpose is still applicable to s 18(1). In other words, the common law requirement of having a common purpose remains an ingredient of the offence of unlawful assembly under s 18(1).”[53] 46. The “common law requirement of having a common purpose” is clearly a reference to a participatory intent. The Court made this clear by adding: “... if the offenders just had the common purpose of conducting themselves in the prescribed manner, it is sufficient to satisfy the requirement of common purpose under s 18(1)”.[54] The Court was not endorsing an “extraneous common purpose”. Indeed, the submission that such a purpose is required was rejected as “completely wrong in law”.[55] 47. The foregoing analysis is equally applicable to the offence of riot. A defendant committing the offence must have a participatory intent. He or she must intend to take part in the riot along with other participants in the riotous assembly. No extraneous common purpose has to be shown. 48. The notion of an extraneous common purpose poses conceptual and practical problems. Rioters may nurture various purposes: to attack the police; to breach a police cordon so as to attack a target on the other side; to disrupt a transportation or other infrastructural system; to oppose a government policy; to cause a government minister to resign; and so forth. They might even be motivated to riot in return for payment by an instigator. Quite apart from the practical difficulties of proving that one or other purpose was shared by the defendants, arguments would arise as to how the extraneous common purpose should be formulated and as to whether motive has to be distinguished from purpose in this context. In contrast, proof of a defendant’s participatory intent may generally be inferred from his or her conduct. 49. Ms Li SC submitted that it is necessary to prove that a defendant took part in a particular unlawful or riotous assembly, sharing a specific extraneous common purpose with other participants in that peculiar assembly, such purpose and such assembly to be identified and spelt out in the indictment and distinguished from other unlawful or riotous assemblies pursuing different extraneous common purposes, forming part of the general mob. Those submissions are rejected. We have already excluded any requirement of proving an extraneous common purpose. Certainly nothing in the language or intent of sections 18 and 19 lends any support for the unworkable and unrealistic approach espoused by Counsel. 50. We would therefore answer Questions 1a and 1b in Annex 1, which postulate the possible requirement of an extraneous common purpose, in the negative. As to Question 1c, as explained above, the defendant must have a participatory intent, intending to participate in the unlawful assembly or riot in concert with other participants, being aware of their related prohibited conduct. We would answer Question 1d, which addresses what we have called a “mutual assistance” element, in the negative as it represents an aspect of the common law that is no longer applicable. C. Joint enterprise and sections 18 and 19 51. Question 2a in Annex 1 asks “Whether the doctrine of joint enterprise applies to the offences under sections 18 and 19 of the POO”. Question 1 in Annex 2 is to like effect. In answering those questions, it is important to note, as explained in HKSAR v Chan Kam Shing,[56] that the common law has developed two forms of the joint enterprise doctrine, referred to as the basic and extended forms. C.1 Basic Joint Enterprise (“BJE”) and Extended Joint Enterprise (“EJE”) 52. The basic form (which Lord Hoffmann called “the plain vanilla version of joint enterprise”[57]) – BJE – involves the co-adventurers simply agreeing to carry out and then executing a planned crime. It is taking part in this criminal joint enterprise that makes all participants guilty as principals, whoever the actual perpetrator(s) of the actus reus might have been. Such liability is independently based on each defendant’s participation in the joint criminal enterprise with the requisite mental state to constitute the offence relevant to the defendant in question.[58] The doctrine is especially efficacious in dealing with situations of evidential uncertainty where, for instance, it is clear that A and B agreed to carry out a murder but it is unclear whether A or B actually struck the fatal blow. Of course, if it is known that both A and B struck the blows which cumulatively killed the victim, they would be jointly guilty of murder as principals. And if it was known that A did the killing while B was present and helped or encouraged A to do it, B would be guilty on the basis of secondary liability as an aider and abettor.[59] And if the agreement to commit the murder is proved, both are liable as conspirators. 53. The extended version of joint enterprise – EJE – addresses the situation where certain defendants (say A, B and C) agree on a certain BJE but, where in the course of carrying out the agreed plan (say, a burglary), one of them (say A) commits a more serious offence (say, murder). The common law (as applied in Hong Kong)[60] holds that B and C are equally guilty of A’s crime on the basis of EJE if it is proved that A’s commission of the further offence (if murder, meaning his acting with intent to kill or at least to cause grievous bodily harm) was foreseen by them as a possible incident of the execution of their planned joint enterprise. 54. As was pointed out in Chan Kam Shing,[61] the joint enterprise doctrine operates alongside the traditional accessorial liability principles. Thus, CPO section 89 makes any person who aids, abets, counsels or procures the commission by another person of any offence guilty of the like offence and so punishable to a like extent. The joint enterprise doctrine also operates alongside the principles relating to inchoate liability for conspiracy, incitement and attempt which are also punishable to the like extent as the principal offences.[62] 55. In examining how far the joint enterprise doctrine applies to unlawful assembly and riot, two questions arise in the present context: (i) is the doctrine invoked in the courts below the BJE or the EJE form of the doctrine? (ii) What consequences does the prosecution allege to flow from application of the doctrine to unlawful assembly and riot? C.2 The Judge’s summing-up and the Court of Appeal’s decision in the Lo Case 56. The trial Judge summed up for the jury on the footing that the defendants could be convicted on the basis of joint enterprise and elaborated as follows (in translation): “If two or more people commit criminal offences together, as mentioned before, but I read out once again, even if each person assumes a different role, if they act in accordance with a common criminal enterprise or agreement, each shall be guilty. Similarly, the words ‘plan’ and ‘agreement’ do not mean that there has to be any formality about it. An agreement commit an offence may arise on the spur of the moment. Nothing needs be said at all. It can be made with a nod and a wink, or a knowing look, or it can be inferred from the behaviour of the parties. The essence of the joint culpability for a criminal offence is that each defendant shares a common intention to commit the offence and played his part in it (however great or small) so as to achieve that common object. In other words, the prosecution alleged in his case that with regard to Counts 3 and 4, they said that the defendant in question committed the offence of riot together with the other defendants listed in the charges.” 57. And in relation to riot, in her written directions, her Ladyship stated: “Where a criminal offence is committed by two or more persons, even if each of them may play a different part, but where they are acting together as part of a joint plan or agreement to commit the offence, each one of them is guilty.”[63] 58. The Judge’s direction therefore concerned a BJE involving participation in an unlawful or riotous assembly. 59. In the Court of Appeal, the focus was on an alleged misdirection on “common purpose”[64] and the Court did not comment on the applicability of the doctrine of joint enterprise while upholding the conviction. It therefore implicitly accepted that the BJE doctrine applies to sections 18 and 19. C.3 The decisions of the Judge and the Court of Appeal in the Tong Case 60. The issue was directly addressed both at first instance and in the Court of Appeal in the Tong Case. The Judge had acquitted the defendants, declining to find by inference that they had been physically present and either taking part in the riot in person or encouraging others who were committing the offence. His Honour also rejected the prosecution’s contention that even if the defendants were not present as part of the said assembly, they should be convicted as participants in a joint enterprise with other rioters. He held that the doctrine was excluded as a matter of statutory construction.[65] As we have seen, this caused the Secretary for Justice to submit questions for the opinion of the Court of Appeal now reproduced in Annex 2 of this judgment. The objective of the Secretary for Justice was to establish that, by operation of the joint enterprise doctrine, it is possible to fix a defendant with liability for riot as principal without that person being present at the scene of the riot. 61. Poon CJHC, with whom the other members of the Court of Appeal agreed, held that Questions 1 and 2 should be answered in the affirmative, taking the view that the joint enterprise doctrine filled what would otherwise be a significant gap in the law of unlawful assembly and riot. He was particularly anxious to hold that persons who played important roles in unlawful assemblies and riots while not physically present at the scene of the criminal assembly should not escape liability: “... unlawful assemblies and riots nowadays are highly fluid in nature. They involve a myriad of participants playing various roles and sometimes with a rather sophisticated division of labour among them. Some physically participate in the unlawful assembly or riot at the scene. Some aid or abet the participants at the scene. Some may not even be present but are clearly participants under the doctrine of joint enterprise. Take the following examples: (1) A mastermind of the unlawful assembly or riot who remotely oversees the situation and gives commands or directions to the participants on the ground. (2) A person who funds or provides materials for the unlawful assembly or riot. (3) A person who encourages or promotes the unlawful assembly or riot by making telephone calls or spreading messages on social media. (4) A person who provides back-up support to the participants in the vicinity of the scene, such as collecting gear, bricks, petrol bombs, other weapons, and other materials to be used by the participants. (5) A lookout stationed in the vicinity who alerts the participants to the advance or deployment of the police. (6) A person who drives a getaway car to allow the participants to leave the scene.”[66] 62. It is evident that Poon CJHC was, like the trial Judge in the Lo Case, referring to the BJE form of the doctrine. The only offences under consideration were the offences of unlawful assembly or riot. No one was seeking to charge any of the participants with a more serious offence committed in the course of their agreed criminal venture so that no question of EJE arose. C.4 On the applicability of the joint enterprise doctrine 63. In our view: (a) On a proper construction of sections 18 and 19 of the POO, the joint enterprise doctrine is not applicable as a basis for fixing a defendant with liability for the offences of riot or unlawful assembly as a principal if he or she was not present at the scene and was not taking part in the criminal assembly. (“The absence point”) (b) On their true construction, POO sections 18 and 19 leave no room for operation of the common law doctrine in its BJE form because the statutory language renders that doctrine otiose and its application would give rise to duplication and possible confusion regarding the central actus reus element of “taking part” in the criminal assembly. (“The taking part point”) (c) The EJE form of the doctrine may be applicable if certain participants in the unlawful assembly or riot can be proved to have foreseen commission by one or more of their number of a more serious offence as a possible incident of the execution of their planned joint participation in the unlawful assembly or riot. (“The EJE point”) (d) The foregoing propositions do not affect potential secondary or inchoate liability for offences under sections 18 or 19 of persons who are not present at the scene. C.5 The absence point 64. The issue here is whether a defendant who is absent and so does not take part in the unlawful or riotous assembly together with others so assembled, can be found guilty of unlawful assembly or riot as a principal offender on the basis of the BJE doctrine. As noted previously, the Court of Appeal in both the Lo Case[67] and the Tong Case[68] omitted reference to the centrally important actus reus element of “taking part” in sections 18 and 19. That resulted in focus on the participatory nature of the offences being lost. 65. As analysed in Sections A.1 to A.3 of this judgment, guilt as a principal offender requires proof that the defendant engaged collectively with other participants in or in furtherance of the prohibited conduct or breaches of the peace. The Ordinance defines participatory offences requiring a defendant not just to “take part”, but to do so as part of an “assembly”, as a person “assembled together” with others taking part. Thus, its language imports a requirement of presence at the scene before one can be held guilty as a principal offender. Such a statutory requirement cannot be displaced by the common law BJE doctrine. The proposition that at common law, a defendant may in certain circumstances be held guilty as a principal according to his own mens rea even though he was not present at the scene where the actus reus was committed only holds good where operation of that doctrine is not excluded by inconsistent statutory provisions. In Chan Kam Shing[69] and Sze Kwan Lung[70] the doctrine was applicable since the substantive offences involved were murder and manslaughter and not statutory offences requiring presence at the scene. C.6 The taking part point 66. This point, which concerns the applicability of the joint enterprise doctrine where the defendant is present at the scene, also flows from recognising the centrality of “taking part” as the actus reus of both offences. Since, as we have just seen, the offences under sections 18 and 19 require proof of the defendant taking part in the disorderly or riotous conduct in common with the other persons assembled, it is otiose and confusing to introduce as a requirement, a prior layer of common intention and collective conduct which merely mirrors the statutory requirement of “taking part”. 67. As noted above, a BJE involves the co-adventurers agreeing to carry out and then executing a planned crime. With the offences of unlawful assembly and riot, proving that the defendants executed their plan would provide proof of their “taking part” in the unlawful assembly or riot and thus establish guilt of the statutory offences, without any need to pray in aid the doctrine of joint enterprise. Their participatory intent would be inferred from their prohibited conduct while assembled together, so from the prosecution’s viewpoint, invoking the BJE doctrine would add the unwarranted burden of showing a prior agreement. From the trial Judge’s point of view, jettisoning BJE in this context would avoid possibly confusing the jury by trying to explain two layers of “taking part” – first in the joint enterprise and then in the unlawful or riotous assembly. 68. This analysis does not leave any gap. As was pointed out in Chan Kam Shing,[71] while a principal carries out the prohibited conduct with the necessary mens rea and an aider and abettor is one who is present and renders assistance or encouragement to the principal in the commission of the offence, liability as a counsellor or procurer does not require the defendant’s presence at the scene: “A person who counsels or procures an offence (referred to also as an accessory before the fact) is not present but provides assistance or encouragement prior to the commission of the offence. Such a person performs the actus reus of ‘procuring’ an offence ‘by setting out to see that it happens and taking the appropriate steps to produce that happening’. A person ‘counsels’ an offence by soliciting or encouraging its commission.”[72] Nor does a defendant who commits the inchoate offence of conspiracy or incitement have to be present at the scene where the principal offence is committed. 69. Thus, to take the examples given by Poon CJHC:[73] The “mastermind” who remotely oversees the situation and gives commands or directions to the participants on the ground would be guilty of incitement or as counsellor and procurer of the criminal assembly. So would the persons who fund or provide materials for the unlawful assembly or riot; or who encourage or promote it on social media. Those who provide back-up support to the participants in the vicinity of the scene, collecting bricks, petrol bombs and other weapons; or who act as lookouts in the vicinity of the riot may either be “taking part” as principals under the POO or liable as aiders and abettors if present at the scene; or, if not present, liable as counsellors or procurers. The culpability of the person who drives a getaway car to help participants leave the scene is likely to include liability for assisting an offender under section 90 of the CPO.[74] 70. There is accordingly no lacuna.[75] Public order can be fully enforced relying on secondary liability and inchoate offences, consistently with the true construction of the statutory offences and without stretching the concept of being “assembled together” beyond its proper limits. As previously pointed out, persons so convicted as accessories or for inchoate crimes are punishable to a like extent as principal offenders.[76] C.7 The EJE point 71. The difficulties associated with applying the BJE doctrine to unlawful assembly or riot do not arise in relation to EJE. There is no duplication or confusion since the liability based on the EJE doctrine relates to the further offence committed in execution of the plan and not just to taking part in the criminal assembly in question. 72. To take a hypothetical example, A, B and C may be among numerous other persons taking part in a riot (satisfying all the statutory elements of the offence) and C then proceeds to commit a further offence – say, of murder, by deliberately stabbing someone to death. If A and B are shown to have participated with C in the riot and foreseen that C might commit murder, meaning his assaulting a victim with intent to kill or with intent to cause grievous bodily harm as a possible incident of the execution of their planned participation in the riot, they could be found guilty of murder on the EJE basis. 73. The point does not arise in the present case since no EJE case is sought to be made against Lo. It is however not an unrealistic consideration. It might, for instance, be possible to prove that a group of persons agreed to take part together in a riot, intending to destroy public property and to erect barriers stopping traffic, while knowing that some amongst them would take along petrol bombs or potentially lethal weapons which they might use. If they proceeded with their plan and the petrol bombs or other weapons were then used to cause serious injury, the EJE doctrine might apply to fix the rioters who foresaw the intentional infliction of such injury as a possible incident of the execution of their agreed plan with liability for the more serious offence. C.8 Presence and the location and scope of the assembly 74. The Court noted in Kwok Wing Hang v Chief Executive in Council,[77] that at the height of the disturbances in 2019: “The frequency of outbreaks of violent protests increased and the locations at which they took place also spread from one or two areas to become a phenomenon described colloquially as ‘blossoming everywhere’in which multiple outbreaks of violence happened simultaneously on Hong Kong Island, in Kowloon and in the New Territories.” 75. And as Poon CJHC observed in the Tong Case:[78] “... unlawful assemblies and riots nowadays are highly fluid in nature.” He pointed to participants assuming different roles and communicating with each other using their phones and on social media. Offenders could not be expected to be assembled as a stationary group with a fixed membership in a single location. Participants would move around in varying groups along main thoroughfares, running into side streets and buildings, spreading out and re-coalescing whether in response to action by the police, in pursuit of different targets or for other reasons. Violence would periodically flare up and die down. Participants would often be in communication with each other, coordinating their activities. 76. It will be necessary in each case for the tribunal to determine where and when an unlawful assembly or riot took place and whether a defendant, if charged as a principal, was present and took part. However, the abovementioned fluidity should be taken into account and an overly rigid view should not be taken of what constitutes the assembly, its location and duration. Evidence regarding the geographical area affected, the conduct of and communications maintained among the participants and the duration of the disturbances should be considered as a whole. The defendant’s role in the assembly, if any, should be considered for the purposes of assessing his or her potential principal, accessorial or inchoate liability. 77. A realistic view should be taken of the duration of the unlawful assembly or riot. So long as three or more participants remain actively engaged in the criminal assembly (not necessarily including the constituent offenders establishing the unlawful assembly nor the person or persons whose breach or breaches of the peace transformed the unlawful assembly into a riot – they may have left), the unlawful assembly or riot remains in being as a matter of law. Such an assembly remains in being as long as the participants remain at the scene even if, in the case of a riot, the violence ebbs and flows. Any person taking part in such a riotous assembly commits the offence. 78. The focus should be on whether the evidence directly proves or supports an irresistible inference that the defendant had taken part in the unlawful assembly or riot. Evidence which might support such an inference could include such matters as the time and place of arrest and items found on the defendant, such as a helmet, body armour, goggles, a respirator, a radio transceiver, plastic ties, laser pointers, weapons and materials to make weapons such as petrol bombs which might have been used by those taking part in the criminal assembly. Such fluidity and the basis of the defendant’s alleged liability should also be borne in mind in the drafting of the charge, catering for alternative possibilities. D. Encouragement 79. Question 2d in Annex 1 asks whether a person can be found guilty of riot “without specific conduct on his part falling under the prescribed conduct provided in sections 18 and 19”, but “merely by virtue of alleged encouragement through his presence”. 80. In her directions to the jury, the Judge stated (in translation): “... mere presence at the scene of crime is not enough to prove guilt. But if you are sure that a particular defendant was at the scene, deliberately by his [mere] presence tried to and did encourage the others to use violent [sic] or threaten to use violence; or if you are sure that his presence at the scene was for playing his role in achieving the purpose of committing riot as jointly agreed by him and other defendants, then he (that is, the defendant) is guilty.” The second occurrence of the word “mere” does not represent a correct translation of the original Chinese and is ignored. 81. It is obviously important to avoid treating innocent passers-by who find themselves caught up in an unlawful assembly or riot as guilty of an offence just because they were present at the scene. Presence at the scene in itself is not enough to constitute “taking part” or aiding and abetting.[79] As the Queensland Court of Appeal held in R v Cook,[80] at common law, mere presence in an unlawful assembly or riot is generally insufficient to found liability. It has traditionally been considered necessary that there be some intentional activity in furtherance of the riot.[81] 82. That is not to say that the bar is set high. It does not take a great deal of activity on the defendant’s part to move the case from the “mere presence” to the “encouragement” category. Thus, in 1810, Mansfield CJ stated in Clifford v Brandon:[82] “The law is, that if any person encourages or promotes, or takes part in riots, whether by words, signs, or gestures, or by wearing the badge or ensign of the rioters, he is himself to be considered a rioter, and he is liable to be arrested for a breach of the peace.” 83. This was echoed more recently in Caird,[83]where Sachs LJ said: “It is the law … that any person who actively encourages or promotes an unlawful assembly or riot, whether by words, by signs or by actions, or who participates in it, is guilty of an offence which derives its great gravity from the simple fact that the persons concerned were acting in numbers and using those numbers to achieve their purpose.” 84. And as Byrne J stated in R v Cook:[84] “Generally, mere presence at the scene of a crime does not involve criminal responsibility. But presence to facilitate the commission of an offence by others has every potential to attract criminal responsibility under s 7 [of the Criminal Code (Qld)]. And so those present to ‘lend the courage of their presence to the rioters, or to assist, if necessary’ may be guilty with the active participants.” 85. Whether a defendant has done enough to constitute “taking part”, especially if by way of encouragement, is a matter of fact and degree, taking all the circumstances into account. 86. Question 2d contains certain rolled-up propositions. Aiders and abettors or counsellors and procurers may obviously perform acts which are not themselves acts of disorderly conduct or breaches of the peace but which offer encouragement or assistance to others in the commission of such acts, thus founding secondary liability (or, as previously explained, possibly also liability as principals for “taking part”). Thus, if the defendant’s presence occurs in circumstances qualifying it as “encouragement” of the prohibited conduct by others, then the answer to Question 2d would be in the affirmative. But mere presence without more is not treated as encouragement, whether for the purpose of “taking part”, or accessorial or inchoate liability. 87. In her summing-up (including in her written directions), the Judge did on a number of occasions correctly direct the jury that their task was to decide whether the appellant had taken part in the riot, coupled with the requirement of a “common purpose” in the sense of a participatory intent discussed above. The case against Lo was that he was a principal offender who had repeatedly performed acts amounting to breaches of the peace. He was not just an innocent by-stander nor even someone who had merely encouraged others. The appellant’s defence was that he had been misidentified, a defence which the jury plainly rejected. E. Breach of the peace 88. As we have seen, breaches of the peace are the distinguishing feature of a riot. In this jurisdiction, R v Howell[85] has often been cited as authority for what constitutes such a breach. Watkins LJ there stated: “... there is a breach of the peace whenever harm is actually done or is likely to be done to a person or in his presence to his property or a person is in fear of being so harmed through an assault, an affray, a riot, unlawful assembly or other disturbance. It is for this breach of the peace when done in his presence or the reasonable apprehension of it taking place that a constable, or anyone else, may arrest an offender without warrant.”[86] 89. With its focus on an act which actually harms a person, or in his presence, harms his property, or is likely to cause such harm, etc, this appears to be a markedly narrow formulation. In recent experience, rioters have often engaged in wanton damage to property which is not privately owned and not done in the presence of the owner – such as tearing down road railings to make barriers, digging up bricks to use as projectiles, smashing traffic lights, damaging CCTV cameras, throwing Molotov cocktails into Mass Transit Railway stations and vandalising shops identified with persons of opposing political views. Such acts often occurred in the middle of the night when the premises were closed and no owner was present. Plainly, such conduct also constitutes breaches of the peace. 90. In R (Laporte) v Chief Constable of Gloucestershire,[87]Lord Bingham of Cornhill put the proposition derived from Howell more widely, stating that “the essence of the concept was to be found in violence or threatened violence”. We would adopt that as a general approach to the phrase “breach of the peace” in the context of section 19. Someone who commits or threatens an act of violence against another person or another’s property or acts so that such violence may reasonably be apprehended, commits a breach of the peace. 91. As is apparent from the passage cited above, Watkins LJ’s statement in Howell was made in affirming the existence of a common law power of arrest without warrant. His reason for referring to the presence of the person apprehending a breach of the peace is indicated as follows (especially in the words italicised): “... even in these days when affrays, riotous behaviour and other disturbances happen all too frequently, we cannot accept that there can be a breach of the peace unless there has been an act done or threatened to be done which either actually harms a person, or in his presence his property, or is likely to cause such harm, or which puts someone in fear of such harm being done. There is nothing more likely to arouse resentment and anger in him, and a desire to take instant revenge, than attacks or threatened attacks upon a person’s body or property.”[88] 92. Thus, the power of arrest, based on an apprehended breach of the peace, was held to arise so that such situations could be prevented from deteriorating into public disorder involving provoked retaliation by persons present and feeling threatened in respect of their persons or property. 93. In POO sections 18 and 19, that policy consideration is covered by element [4]. Conduct prohibited under section 18(1) includes conduct which would “provoke other persons to commit a breach of the peace”. Thus, the concept of a “breach of the peace” in sections 18 and 19 includes, but is not confined to, situations which might give rise to provoked retaliation as envisaged in Howell. In cases involving actual or threatened violence to property, the owner of such property need not be present for there to be a breach of the peace. F. Substantial and grave injustice: non-conviction of the named co-defendants 94. The indictment against Lo alleged that: "LEUNG Tin-kei, LEE Nok-man, LO Kin-man [the appellant] and LAM Ngo-hin, between the 8th day of February, 2016 and the 9th day of February, 2016, at Portland Street, Mong Kok, Kowloon, in Hong Kong, took part in a riot.” 95. Where justified by the evidence, it is common (and good practice) to add the allegation that the defendant’s participation in the unlawful assembly or riot was also “with persons unknown” or “with persons not before the court”. That was not done in Lo’s case and, at the end of the trial, while he was convicted, the jury were unable to reach a verdict in respect of the other three named defendants. Since the offence requires proof of the collective action of at least three persons assembled together as a constituent element, it was a material irregularity to convict the appellant alone without an averment in the indictment regarding the existence of other constituent offenders. Lo argues that this amounted to such a substantial and grave injustice as to require the quashing of his conviction. 96. Such a situation was addressed by the English Court of Appeal in R v Abdul Mahroof,[89] where the appellant was charged with two others (named Biney and Sayers) with violent disorder contrary to section 2 of the Public Order Act 1986, an offence which is constituted “Where three or more persons who are present together use or threaten unlawful violence and the conduct of them (taken together) is such as would cause a person of reasonable firmness present at the scene to fear for his personal safety ...”. 97. The facts were decidedly different from the present case. The evidence was that the appellant and Sayers went to Biney’s home to confront him about a debt he owed the appellant. The threatened violence was alleged only to involve those three men, they being the only persons named in the indictment as participants. A witness testified that there were also two other persons present but they “appeared to be on the fringe of the argument by just adding the odd word during the argument”.[90] The jury acquitted Biney and Sayers, and the question arose whether, given the elements of the offence, the appellant could nonetheless be convicted on his own. 98. On this issue, the Lord Chief Justice stated: “... the question comes down to this: Was it a sufficient allegation in the indictment if no mention was made in the indictment of any other potential participant except the three men, namely, Biney, Mahroof and Sayers? ... It seems to us that the answer to that, and as a consequence the answer to the certified question, is ‘Yes, subject to two very important qualifications’: first of all, that there is evidence before the jury that there were three people involved in the criminal behaviour, though not necessarily those named in the indictment; secondly, that the defence are apprised of what it is they have to meet.”[91] 99. The Court held that it was a material irregularity to convict the appellant in the circumstances. Application of the proviso was considered but, on the facts described above, the Court felt unable to say that there was no miscarriage of justice. The only other persons who might have supplied the needed three participants were the two men at the fringes of the argument, offering little basis on the evidence for being treated as constituent offenders under section 2. The Court decided instead to substitute a conviction for an offence that did not depend on the participation of others. 100. The circumstances of Lo’s case are very different. It is important to note that the guilt of a defendant charged under POO sections 18 or 19 is not dependent on the conviction of anyone else. The prosecution must prove that he or she took part in the criminal assembly which was constituted by at least three persons. If the evidence is incontrovertible that there was such a criminal assembly and that the defendant took part in it with the necessary participatory intent, the elements of the offence are proved and the fact that no one else was charged or convicted, including the only persons named as co-participants, does not exonerate the defendant. In cases where the defendant is convicted but not the only other alleged participants, the failure to include a reference in the indictment to “persons unknown” or “persons not before the court” as co-participants is a material irregularity but by no means necessarily fatal to the prosecution. 101. There was undoubtedly evidence before the jury in the present case that there were at least three persons other than his co-defendants who had taken part in the criminal assembly in which Lo had participated. Thus, the Court of Appeal noted that some 500 people gathered on Portland Street that night behaving in a riotous manner[92] and considered it incontrovertible that Lo (referred to as the 2nd applicant) directly took part in the riot: “The 2nd applicant appeared in Portland Street back at around 9 o’clock in that evening, assisting the hawkers to push hawker trolleys out to the pavement; and subsequently he together with others scolded the FEHD staff and obstructed the police in the investigation of the taxi incident. Police officers testified that they witnessed the 2nd applicant twice throwing objects from the crowd to the police line. The footage produced in court also showed that the 2nd applicant at later stages threw objects to the police line and picked up mud and sand from the ground to attack the police for a total of 11 times.”[93] 102. The evidence of the mob’s activity was described in detail, the Court of Appeal observing[94] that: “The trial judge took into consideration that there were some 500 people gathering on Portland Street, and that the crowd confronted the police for almost four hours, threw objects at the police line, splashed liquid in front of the police line and put up resistance by holding weapons, and also that a police officer was persistently kicked on his head after being pushed to the ground. Moreover, the case happened in a very crowded area of the city on the night of Lunar New Year Day. The violence involved caused extreme danger to the safety of citizens. In the course of it, someone was hit by a concrete slab thrown out, immediately fell onto the ground and lost consciousness. Someone was injured and bled as a result of being struck by a brick. The trial judge found the violence on Portland Street to be of large scale and extremely serious.” 103. The trial Judge’s written directions required the jury to be sure that Lo did take part in an unlawful assembly and a riot involving “3 or more than 3 persons assembled together”, not limited to his named co-defendants. Her Ladyship expanded on this in her oral directions: “The third point is in relation to Count 3 we reviewed yesterday, namely the riot on Portland Street where [Lo] was involved. You will remember what I said. I said that the prosecution relied on the scene, the situation at the scene, the number and state of the crowd and the behaviour of [Lo] etc. to prove the first element of the offence of unlawful assembly ... Of course, the prosecution case, as I said, was to look at the overall situation on Portland Street, the number of people, and the actions of [Lo], etc. that night. He asked us to infer that at Portland Street that night [Lo] did take part in an unlawful assembly.” 104. It was on this basis that the Court of Appeal noted that in sentencing Lo, “the trial Judge held that he “had been actively taking part in the riot of Charge 3”.[95] There was thus ample evidence for the jury to be satisfied of Lo’s participation and so for the first condition referred to in R v Abdul Mahroof to be met. 105. There can also be no doubt that Lo was fully aware that the case against him was of his participation in the riot in Portland Street over some six hours, culminating in his arrest at about 3:00 am. He knew that the prosecution case, based on video footage, was that he had, among other things, on eleven occasions in the course of the riot, thrown objects at the police. He denied having done this, his defence being one of misidentification. But the jury obviously found the case against him proved beyond reasonable doubt. 106. Ms Li SC relied on R v Worton,[96] a case involving a fight in a public house, where the English Court of Appeal held that the two Mahroof conditions had been satisfied but, Lloyd LJ, giving judgment for the Court, went on to say: “In cases of this kind, where there are only three defendants accused of violent disorder, it is not in our view sufficient that the offence should be defined in general terms, as it was here. It is necessary that the Judge should go on and warn the jury specifically that if any one of the three defendants should be acquitted of violent disorder, then they must necessarily acquit the other two, unless satisfied that some other person not charged was taking part in the violent disorder. This the recorder in the present case never did.”[97] 107. While holding that a direction in the aforesaid terms is necessary, it is clear that Lloyd LJ was not ruling out application of the proviso in cases where that direction is not given. Thus he noted that the prosecution had invited the Court to apply the proviso but declined that invitation on the basis that “... we cannot be convinced that the jury would necessarily have reached the same result if they had been given a full and proper direction.”[98] 108. In the present case, in the light of the evidence described above, we are satisfied that the jury plainly would have reached the same conclusion if fully directed so that application of the proviso is appropriate. The case for quashing the conviction on the basis of substantial and grave injustice is not made out. G. Conclusions and disposition of the appeals 109. We summarise our conclusions as follows: (a) POO section 18 defines the elements which constitute an unlawful assembly and makes “taking part” in the unlawful assembly so constituted the actus reus of the offence. The constituent elements are for three or more persons assembled together to conduct themselves in the prohibited disorderly, etc, manner intended or likely to cause a reasonable apprehension that the persons so assembled will commit or provoke a breach of the peace. Any person who takes part in an unlawful assembly commits the offence. (b) POO section 19 builds on section 18, making the existence of an unlawful assembly one of the constituent elements of the offence of riot. A riot comes into being when any person taking part in an unlawful assembly commits a breach of the peace, turning the assembly into a riotous assembly. The offence is committed by any person who takes part in a riot so constituted. (c) Both offences are participatory in nature. The defendant must be shown not just to have been acting alone but to have taken part in the unlawful or riotous assembly, acting together with others so assembled, being aware of their related conduct and with the intention of so taking part, ie, with a participatory intent. There is no requirement for the persons taking part to share some extraneous common purpose. (d) To “take part” in the relevant criminal assembly, the accused must perform the acts prohibited, ie, by behaving in the prohibited disorderly, etc fashion (section 18); or committing a breach of the peace (section 19); or acting in furtherance of such prohibited conduct by facilitating, assisting or encouraging those taking part in the criminal assembly. (e) Mere presence at the scene of an unlawful or riotous assembly does not give rise to criminal liability. However, if the accused, being present, provides encouragement by words, signs or actions, he or she may be held to be “taking part” and guilty as a principal or held to be an aider and abettor. In deciding whether a defendant was present at the scene, the court should take into account the possible fluidity of the criminal assembly and the communications maintained by participants with each other in ascertaining the time, place and scope of the assembly in question. (f) To be guilty as principal offender under section 18 or 19, the defendant must be present at the scene, taking part in the unlawful or riotous assembly together with others so assembled. However, a person who promotes or acts in furtherance of an unlawful assembly or riot while not present at the scene may be guilty as a counsellor and procurer of the relevant offence or guilty of conspiracy or incitement to commit such offence and would be punishable to a like extent as the principal offender. Moreover, a person who assists an offender is liable under section 90 of the CPO. (g) The common law doctrine of joint enterprise in its basic form cannot operate in a manner inconsistent with the statutory language. It cannot be relied on to fix liability as a principal offender on a defendant who was not present and not acting as part of an assembly together with others so assembled, as required by the two sections. Even in cases where the defendant is present at the scene, the basic joint enterprise doctrine is inapplicable, being otiose and liable to cause confusion, since the actus reus of each of the statutory offences already involves “taking part” with others assembled together. (h) The inapplicability of the basic joint enterprise doctrine does not leave any gap since culpable conduct of absent defendants who engage in promoting or acting in furtherance of a criminal assembly is covered by secondary and inchoate liability offences. In given circumstances, the extended form of joint enterprise may operate to fix participants with liability for more serious offences committed in executing their joint plan. (i) “Breach of the peace” in the context of sections 18 and 19 includes, but is not confined to, situations which might give rise to provoked retaliation. It extends to cases involving actual or threatened violence to persons or property, without any need for the owner of such property to be present. (j) In drafting an indictment alleging unlawful assembly or riot against a defendant, it is good practice to allege, where the evidence permits, participation by other persons unknown. Failure to do so may be a material irregularity where the defendant is convicted but not the only other named defendants so that the requirement of 3 or more constituent offenders may not be met. In such cases, the question whether the proviso should be applied depends on whether the evidence plainly establishes that there were at least three persons involved in the criminal assembly in which the defendant took part and on whether the defendant fairly knew the case he or she had to meet. 110. Lo’s appeal must be dismissed. The ground of appeal[99] based on the submission that the Judge had erred in not requiring the jury to find an extraneous common purpose has been rejected.[100] The other grounds, reflected in the other questions in Annex 1 have little bearing on Lo’s case. While the trial Judge erroneously considered joint enterprise applicable, this had no practical effect in Lo’s case since his conviction does not depend on application of that doctrine. Nor was it alleged against him that he was guilty by having encouraged others by his presence.[101] The case against him, amply supported by the evidence and properly dealt with in the summing-up, was that he was present and actively took part in the riot over several hours. We have held that the content of the indictment and non-conviction of the other named defendants constituted a material irregularity but that this is a clear case for application of the proviso. 111. The Tong Case comes to this Court via a reference under CPO section 81D to the Court of Appeal on two questions of law. Since, by virtue of section 81D(4), such a reference does not affect Tong’s acquittal, the issues before this Court are confined to the questions of law set out in Annex 2. It is unnecessary and would be inappropriate to conduct a review of the trial Judge’s findings of fact. In answer to Question 1, we have held that BJE is inapplicable but that, in given circumstances, EJE may be applicable. As to Question 2, we have held that the doctrine attributing liability as a principal offender to a defendant who is absent by application of the BJE doctrine is inapplicable as inconsistent with the statutory language. We have emphasised that a defendant who promotes or acts in furtherance of an unlawful assembly or riot while not present at the scene may be liable as a secondary party or as a conspirator or inciter of the main offence and punishable to the same extent. We have also dealt with the effect of mere presence and what constitutes encouragement in the present context.[102] We have accordingly addressed the questions set out in Annex 2 and so disposed of Tong’s appeal. There is no need for any consequential orders to be made. FACC 6/2021 Ms Gladys Li SC and Mr Lawrence Lau (both assigned by the Director of Legal Aid), Mr Kin Lau and Mr Jason Ko (both on a pro bono basis), instructed by Bond Ng Solicitors, for the Appellant Mr Anthony Chau, DDPP(Ag), Mr Ivan Cheung, SPP, Ms Karen Ng, PP and Mr Wilson Lam, PP, of the Department of Justice, for the Respondent FACC 7/2021 Mr Philip Dykes SC, Mr Hectar Pun SC, Mr Anson Wong Yu Yat and Ms Ferrida Chan, instructed by S. T. Cheng & Co, for the Appellant Mr Anthony Chau, DDPP(Ag), Mr Ivan Cheung, SPP, Ms Karen Ng, PP and Mr Wilson Lam, PP, of the Department of Justice, for the Respondent Annex 1 Question 1a In order to establish the offence of riot under section 19 of the Public Order Ordinance (Cap 245) (“POO”), whether proof of the unlawful assembly required that at least 3 persons be assembled together for a “common purpose” which was distinct from the intention of committing the statutorily prescribed acts, namely, conducting oneself in a disorderly, intimidating, insulting or provocative manner; and therefore whether the Trial Judge had erred in law in directing the jury that the element of “common purpose” is satisfied if they are satisfied that the Applicant assembling together with other defendants had the intention of conducting themselves in a disorderly, intimidating, insulting or provocative manner? Question 1b If the answer to Question 1a is in the affirmative, whether there is a legal requirement that the said prescribed acts and the breach of the peace must be committed with the specific intent to achieve the “common purpose” by such means? Question 1c For the common purpose to be a substantive element in the offence of riot, whether the Prosecution is required legally to prove that such alleged common purpose must be shared, mutually understood or communicated between the accused to a standard that there is a meeting of minds, or the Prosecution is simply required to prove that the same purpose was held individually by the accused without the need to prove further mutual understanding or communication? Question 1d Whether it is a distinct element of riot under section 19 of POO that the defendants must have the mutual intention to assist each other, by force if necessary, against any person who might oppose them in the execution of the common purpose? Question 2a Whether the doctrine of joint enterprise applies to the offences under sections 18 and 19 of the POO? Question 2d Whether a person could be found guilty of riot without specific conduct on his part falling under the prescribed conduct provided in sections 18 and 19, but merely by virtue of alleged encouragement through his presence? On the substantial and grave injustice ground, whether a material irregularity arises in that the indictment does not mention any potential participants other than the co-defendants who were not convicted. Annex 2 Question 1 (1) “For the offences of unlawful assembly and riot respectively under sections 18 and 19 of the Public Order Ordinance, Cap 245, whether the common law doctrine of joint enterprise as elucidated in HKSAR v Chan Kam Shing (2016) 19 HKCFAR 640 is applicable?” (“Question 1”), and Question 2 (2) “If Question 1 is answered in the affirmative, for the offences of unlawful assembly and riot, whether the principle that a defendant’s presence at the scene is not always necessary for criminal liability under the common law doctrine of joint enterprise as enunciated in Sze Kwan Lung & Others v HKSAR (2004) 7 HKCFAR 475 is applicable?” (“Question 2”) [1] Cap 245. [2] HCCC 408/2016 (18 May 2018). [3] Poon CJHC, Chu and Pang JJA, cited as HKSAR v Leung Tin Kei [2020] HKCA 275 (29 April 2020) (“CA-Lo Case”). [4] Ribeiro Ag CJ, Fok PJ and Stock NPJ [2021] HKCFA 17 (17 May 2021). [5] (2020) 23 HKCFAR 518 see §§87-97. [6] [2020] HKDC 588 (24 July 2020). [7] Cap 221. [8] Poon CJHC, Macrae VP and A Pang J, cited as Secretary for Justice v Tong Wai Hung [2021] HKCA 404 (25 March 2021) (“CA-Tong Case”). [9] [2021] HKCA 807 (7 June 2021). [10] Annex 1, Questions 1a to 1d. [11] Ibid Question 2a; Annex 2, Questions 1 and 2. [12] Ibid Question 2d. [13] Annex 1. [14] [1891] AC 107 at 144-145. In a criminal law context, see also R v Fulling [1987] QB 426 at 431-432; R v Smurthwaite (1994) 98 Cr App R 437 at 440; and R v Platt (2016) 1 Cr App R 22 at §27. [15] [2021] HKCFA 24 at §17. [16] [2021] HKCFA 24. [17] Ibid at §§37 and 40. [18] Appearing for Lo with Mr Lawrence Lau, Mr Kin Lau and Mr Jason Ko. [19] Sitting as an additional Judge of the Court of First Instance, on an appeal by way of case stated [2012] 5 HKLRD 556. [20] Ibid at §19. [21] Ibid at §§22 and 83. [22] See §§68-70 below. [23] Section E of this judgment. [24] See §§68-70 below. [25] CA-Lo Case at §24. [26] CA-Tong Case at §§35-36. [27] In Sections C.4 to C.7 below. [28] Examples taken from R v Caird (1970) 54 Cr App R 499 and R v John McKinsie Jones (1974) 59 Cr App R 120 discussed in Section B.1 below. [29] CA-Lo Case at §§51-63. [30] (1888) 16 Cox CC 420. Along the same lines, see Tse Chung v R [1967] HKLR 452 at 455-456, per McMullin J. [31] Ibid at 432. [32] [1907] 2 KB 853. [33] Ibid at 860. [34] Ibid. [35] (1970) 54 Cr App R 499. [36] Ibid at 506. [37] (1974) 59 Cr App R 120 at 127. [38] Anderson v A-G (NSW) (1987) 10 NSWLR 198. [39] Ibid at 113. [40] Ibid at 114-115. [41] Law Com No 123 (24 October 1983) at §5.2. [42] Criminal Law, 4th ed, (1978) p 750. [43] Section 1(1): “Where 12 or more persons who are present together use or threaten unlawful violence for a common purpose and the conduct of them (taken together) is such as would cause a person of reasonable firmness present at the scene to fear for his personal safety, each of the persons using unlawful violence for the common purpose is guilty of riot.” ... (3): “The common purpose may be inferred from conduct.” [44] While such changes were made against the background of the extremely serious riots in 1967, the legislation had apparently been in preparation for some two years previously: Second Reading of the Bill, Official Report of Proceedings of the Legislative Council 15 November 1967, p 474. The POO came into force on 17 November 1967. [45] Sir Denys Roberts, later Mr Justice Roberts NPJ. Official Report of Proceedings of the Legislative Council 1 November 1967, pp 438 and 440. [46] Ibid, p 444at §9. The Explanatory Memorandum of a Bill, including its Objects and Reasons, is admissible for ascertaining the statutory purpose: See HKSAR v Cheung Kwun Yin (2009) 12 HKCFAR 568 at §§12-14; HKSAR v Li Kwok Cheung George (2014) 17 HKCFAR 319 at §37; and, most recently, Secretary for Justice v Leung Kwok Hung [2021] HKCFA 32 at §25. [47] [2012] 5 HKLRD 556. [48] Ibid at §19. [49] Ibid at §21. He added that in such a case, such persons might each be separately guilty of an offence under POO section 17B. [50] Ibid at §22. [51] CA-Lo Case at §26. [52] Ibid at §§57-61. Lam JA’s approach was also cited with approval by the Court of Appeal in CA-Tong Case at §63. [53] Ibid at §58. [54] Ibid at §61. [55] Ibid at §§59 and 60. [56] (2016) 19 HKCFAR 640 at §41. [57] Brown & Anor v The State (Trinidad And Tobago) [2003] UKPC 10 at §13. [58] HKSAR v Chan Kam Shing (2016) 19 HKCFAR 640 at §§33 and 63. [59] Ibid at §§41-44, 63. [60] Ibid at §§45-56, upholding the Privy Council’s decision in Chan Wing Siu [1985] 1 AC 168; and declining to follow R v Jogee [2017] AC 387. The narrow view adopted in the UK in Jogee that the intent must be both shared and specific, is not part of the law of Hong Kong. [61] Ibid at §98. [62] Crimes Ordinance (Cap 200) sections 159A and 159C (conspiracy); sections 159G(1) and 159J(1) (attempts). Incitement is punishable at common law. [63] See CA-Lo Case at §26. [64] CA-Lo Case at §23. [65] CA-Tong Caseat §§4-5, 19-24. [66] Ibid at §56. [67] CA-Lo Case at §24. [68] CA-Tong Case at §§35-36. [69] HKSAR v Chan Kam Shing (2016) 19 HKCFAR 640 at §6. [70] Sze Kwan Lung v HKSAR (2004) 7 HKCFAR 475 at §37 (offences of manslaughter and arson). [71] (2016) 19 HKCFAR 640 at §§8-14. [72] Ibid at §12. [73] Cited at §61 above. [74] Section 90(1): “If a person has committed an arrestable offence, any other person who, knowing or believing him to be guilty of the offence or of some other arrestable offence, does, without lawful authority or reasonable excuse, any act with intent to impede his apprehension or prosecution shall be guilty of an offence.” [75] As was acknowledged by Mr Anthony Chau, DDPP (Ag) appearing for the Respondent with Mr Ivan Cheung, SPP, Ms Karen Ng, PP and Mr Wilson Lam PP. [76] CPO, section 89 and Crimes Ordinance (Cap 200) sections 159A, 159C, 159G(1) and 159J(1). [77] (2020) 23 HKCFAR 518 at §90. [78] CA-Tong Case at §56. [79] HKSAR v Chan Kam Shing (2016) 19 HKCFAR 640 at §10. [80] (1994) 74 A Crim R 1 at 6. [81] See R v Atkinson (1869) 11 Cox CC 330 at 332; Anderson v A-G (NSW) (1987) 10 NSWLR 198 at 212; Boxer v R (1995) 81 A Crim 299 at 310. [82] (1810) 2 Camp 358 at 370; 170 ER 1183 at 1187. [83] (1970) 54 Cr App R 499 at 505. [84] (1994) 74 A Crim R 1 at 8-9. [85] [1982] QB 416. [86] Ibid at 427. [87] [2007] 2 AC 105 at §§27-28. Cited in Chow Nok Hang (2013) 16 HKCFAR 837 at §77. [88] [1982] QB 416 at 426. [89] (1989) 88 Cr App R 317. [90] Ibid at 319-320. [91] Ibid at 321. [92] CA-Lo Case at §13. [93] Ibid at §14. [94] Ibid at §37. [95] Ibid at §39. [96] (1989) 154 JP 201. [97] Ibid at 203-204. [98] Ibid at 204. [99] Reflected in Questions 1a-1d of Annex 1. [100] Section B.2 of this judgment. [101] As reflected in Questions 2a and 2d of Annex 1. [102] Sections C.8 and D of this judgment. 
Chief Justice Ma: 1. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Ribeiro PJ: 2. This appeal raises questions concerning the relationship between offences involving dental companies under the Dentists Registration Ordinance (“DRO”)[1] and disciplinary charges relating to unprofessional conduct brought before the Dental Council of Hong Kong (“the Council”). Issues concerning procedural fairness and the jurisdiction of the Court of Final Appeal and the Court of Appeal to remit issues for re-hearing by the Council also arise. A. Relevant provisions of the DRO and the Disciplinary Code 3. A person who wishes to practise dentistry is required to be registered under the DRO after having passed the qualifying examinations. Practising dentistry without being registered is an offence punishable by a fine and three years’ imprisonment on summary conviction and by five years’ imprisonment on conviction on indictment, and if injury to the “patient” results, by imprisonment for up to seven years.[2] 4. Additionally, the DRO provides that a registered dentist “shall not practise as a dentist in Hong Kong, unless he is the holder of a practising certificate which is then in force”.[3] Practising certificates are issued by the Council’s Secretary and are valid for 12 months. 5. Section 15(3)(ba) provides that the Council may order the removal of a registered dentist’s name from the General Register if he or she has practiced dentistry in Hong Kong for a period exceeding 6 months without having obtained a valid practising certificate. 6. DRO section 12(1) enables a body corporate[4] to carry on the business of dentistry provided (inter alia) that “a majority of the directors and all persons practising dentistry are registered dentists”. 7. Section 12(3) requires every body corporate carrying on the business of dentistry to file an annual return with the Registrar of Dentists (“the Registrar”), stating the names and addresses of all persons who are directors or managers of the company, or who perform dental operations in connection with the business of the company. The subsection goes on to provide that “if any such body corporate fails so to do, it shall be deemed to be carrying on the business of dentistry in contravention of the provisions of this section.” 8. If a company carries on the business of dentistry in contravention of section 12, the company “and every director and manager thereof” commits an offence and is liable on summary conviction to a fine of $2,000 for each offence.[5] However, in relation to a director or manager, section 12(2A) makes it “a defence for him to prove that the offence alleged to be committed by the body corporate was committed without his knowledge.” 9. Section 17 of the Code of Professional Discipline for the Guidance of Dental Practitioners in Hong Kong (“the Code”) issued by the Council[6] refers dentists to DRO section 12 and states the Council’s position regarding directors of dental companies in the following terms: “The Council wishes it to be understood that a dental practitioner ... who becomes a director of a body corporate carrying on the business of dentistry thereby accepts responsibility for the maintenance of a high standard of professional conduct in that practice or business and may be required to answer to the Council for any act or omission in the conduct of that practice or business which appears to the Council to be such as would, if attributed to an individual dental practitioner, constitute unprofessional conduct. Attention is directed to Section 12 of the Dentists Registration Ordinance.” 10. DRO section 27, headed “Covering” provides as follows: “Any registered dentist who practises dentistry in premises in which an unregistered person practises dentistry commits an offence and is liable on summary conviction to a fine at level 3 and to imprisonment for 6 months.”[7] 11. This is also referred to in section 15 of the Code (also headed “Covering”) as follows: “15.1 The Council considers that a dental practitioner should in no way countenance, help, encourage or assist, either wilfully or by neglect, the practice of dentistry by an unregistered person. 15.2 Section 27 of the Dentists Registration Ordinance reads “Any registered dentist who practises dentistry in premises in which an unregistered person practises dentistry shall be guilty of an offence...” 15.3 Section 2(2) of the Ordinance sets out in what circumstances a person practises dentistry...” 12. DRO section 18 empowers the Council to conduct disciplinary proceedings and provides that it may, after due inquiry, if satisfied that any registered dentist has (inter alia) been convicted of any offence punishable with imprisonment or been guilty of unprofessional conduct, impose sanctions ranging from removal of the dentist’s name from the General Register down to a reprimand. 13. It is important for the purposes of this appeal to note that “unprofessional conduct” is defined as: “... an act or omission of a registered dentist which would be reasonably regarded as disgraceful or dishonourable by registered dentists of good repute and competency.”[8] B. The disciplinary proceedings, leave to appeal and the Council’s concession B.1 The disciplinary proceedings 14. The facts are not in dispute. The four appellants are registered dentists who were directors of a dental company called Health & Care Dental Clinic Limited (“the company”) over the relevant period. On 8 August 2005, the company employed Dr Jeremy Tung (“JT”) who had been registered as a dentist three days earlier. JT held valid practising certificates for the years 2005, 2006 and 2007. However, although he was repeatedly asked by company staff for his renewed practising certificate, he failed to renew his practising certificate for the years 2008, 2009 and 2010. His failure to renew led (apparently unknown to himself) to the removal of his name from the General Register on 17 July 2008. However, JT continued thereafter to practise dentistry as an employee of the company. Part of his routine involved practising on Tuesdays and Thursdays at the clinic which the 2nd appellant (“Dr Chu”) attended on Fridays. 15. It was an agreed fact that JT’s failure to apply for renewal of his practising certificate in 2008 was “unknown to the [appellants] and unknown to [the company]”.[9] It was also agreed that: “As soon as it was brought to the attention of the [appellants] on 16th April 2010, that [JT] had failed to maintain his annual practising certificate, they immediately suspended his duties, and subsequently terminated his employment on 6 May 2010.”[10] 16. JT applied to be restored to the General Register, but on 22 March 2011, the Council refused his application and referred his case to the police. The Council’s refusal was quashed for procedural unfairness in judicial review proceedings brought by JT.[11] However, he was prosecuted for practising dentistry as an unregistered person, pleaded guilty and was fined $20,000. The Court was informed that JT has since been restored to the General Register. B.2 The disciplinary charges 17. In 2013, disciplinary charges (as amended) were laid against each of the appellants in the following terms: “That you, being a registered dentist, during the period from 17 July 2008 to 16 April 2010 – (i) and being a director of [the company], employed [JT] to perform dental service when he was no longer a registered dentist after his name had been removed from the General Register on 17 July 2008 (“the 1st charge”); and/or (ii) failed to take adequate steps to ensure that there was no unregistered person practising dentistry in the premises in which you practised dentistry (“the 2nd charge”); and that in relation to the facts alleged you have been guilty of unprofessional conduct.” 18. All the appellants were convicted on the 1st charge and Dr Chu alone was convicted on the 2nd charge since only he practised at the same premises as JT (although on different days of the week). 19. The appellants’ appeal to the Court of Appeal was dismissed.[12] B.3 The grant of leave to appeal 20. The Appeal Committee[13] granted leave to appeal in respect of the following questions of law (which reflect the Court of Appeal’s reasoning, to which I shall return):- "In disciplinary proceedings charging registered dentists who are directors of a dental company for unprofessional conduct by reason of that company having breached its statutory duty under the Dentists' Registration Ordinance (Cap 156) section 12 by employing an unregistered person to perform dental services:- (i) Was the Council entitled to find unprofessional conduct is made out even though that unregistered person had previously been duly registered and his removal from the General Register was unknown to the applicants and unknown to the dental company over the relevant period? ("Question 1") (ii) Does the answer to the foregoing Question depend on whether such absence of knowledge is properly characterised as negligent or as involving negligence rising 'to the level of an elemental and grievous failure'? ("Question 2") (iii) Was the Court of Appeal right to hold that directors of dental companies have a professional duty to ensure that only registered dentists are employed by the company, to verify the registered status of employed dentists in compiling the annual return and to confirm the accuracy of the annual return; and that inserting the relevant person's name in the annual return required to be filed by DRO section 12(3) amounted to a warranty that he was a registered dentist? ("Question 3") 21. In addition, leave was granted on the “or otherwise” limb on the basis that reasonably arguable questions arise relating to want of procedural fairness. B.4 The Council’s concession 22. Mr Stewart Wong SC, appearing for the Council,[14] concedes that the appellants’ convictions cannot stand because of procedural unfairness. The Council acknowledges that the appellants did not know about JT’s lack of registration but submits that the convictions can nonetheless be sustained on the basis that such lack of knowledge was due to “negligence to a level of elemental or grievous failure” on the appellants’ part. The Council accepts, however, that “the Appellants did not have a fair opportunity to deal with this issue at the Inquiry as this had not been expressly and fairly put as part of the case against them (whether in the Charges (originally or as amended) or as presented at the Inquiry).” Thus, while it does not seek to support the convictions, the Council invites the Court, after answering the Questions 1 to 3 in order to “provide authoritative guidance to the Council in this and future cases, to remit the matter back to the Council for fresh consideration (and with further amendment to the Charges (with leave of the Council) if thought fit)”. 23. The premise of this invitation to the Court is that Questions 1 to 3 should be affirmatively answered and that the appellants may properly be found guilty of disciplinary offences as directors of the company on the basis of the high level of negligence contended for despite having no knowledge of JT’s unregistered status. Mr Johnny Mok SC, appearing for the appellants,[15] takes issue with that premise. The appellants also argue that the Court lacks jurisdiction to remit or alternatively, that such remitter would be oppressive and unfair and so should be refused in the exercise of the Court’s discretion. 24. Given that the Council does not seek to uphold the convictions, it would be open to the Court simply to decide whether it can and should remit the case to the Council. However, since the Council’s invitation to remit is premised on affirmative answers to Questions 1 to 3 and, as those questions have been fully argued, it is appropriate for them to be addressed. C. The Council’s decision 25. The Council treated the first charge as based on the offence created by section 12(2) of DRO, interpreting that section as requiring a dental company “to ensure that it employs only registered dentists to perform dental operations”.[16] In support of this interpretation, it relied on section 12(3), stating: “... under section 12 of the Ordinance the dental company is required to make an annual return to the Registrar of Dentists setting out, inter alia, all persons who perform dental operations in connection with the business of the company, stating the status of each of such persons. The company must verify the registered status of each person who performs dental operations before making the annual return in the prescribed form ... This can easily be done by verifying from the General Register which is published on the website of the Dental Council.”[17] 26. The Council regarded the duty of the company so conceived as extending to the appellants, its directors, whom it regarded as the “brain and mind of the company” and “directly responsible for” its affairs.[18] It held that the fact that the staff had repeatedly urged JT to renew his practising certificate meant that: “... they knew that he was practising dentistry without a valid practising certificate.”[19] 27. The Council found that the company had failed to verify, or had turned a blind eye to, JT’s non-registration. It held that: “In either case, the 4 Defendants as directors failed to fulfill their statutory duty to ensure that all persons employed to perform dental operations are registered dentists.”[20] 28. In reaching that conclusion, the Council also referred to section 17 of the Code and decided that failure to ensure that all persons performing dental operations in the dental company were registered dentists was conduct which would be reasonably regarded as disgraceful or dishonourable by registered dentists of good repute and competency. 29. The Council referred to DRO section 27 as the basis for the second disciplinary charge, holding that it “imposes a positive duty on every registered dentist to ensure that there is no unregistered person practising dentistry in the same premises with him”. Pointing out that DRO section 14 requires dentists conspicuously to exhibit their certificates of registration at the premises where they practise, coupled with the fact that Dr Chu was a director of the company, the Council held that it was an “irresistible inference” that Dr Chu knew that JT was practising in the same premises and that it was unprofessional conduct for Dr Chu to have failed to ensure that JT was not an unregistered person so practising. 30. On the 1st charge, the Council ordered the name of the 1st appellant to be removed from the General Register for a period of three months and the names of the other appellants to be removed for a period of two months. On the 2nd charge, Dr Chu’s name was ordered to be removed from the General Register for a period of one month to run concurrently with the order made on the 1st charge. D. The Court of Appeal’s decision 31. The Court of Appeal noted that unprofessional conduct was not confined to conduct constituting an offence under the DRO[21] and also that the Code did not exhaustively list all instances of such conduct.[22] 32. In assessing the appellants’ conduct, it repeatedly stressed the importance of the “regime of registration” laid down by the DRO[23] and held that the Council was entitled to impose professional responsibilities on dentists in order to support the statutory registration regime and to prevent any “dilution” of the protection it affords against members of the public receiving treatment from unregistered persons practising dentistry. McWalters JA, giving the reasons of the Court, put this as follows: “Against this backdrop of the statutory scheme of the DRO, we can see no legal reason why the Dental Council cannot quite properly regard as unprofessional conduct acts or omissions by a dentist which significantly undermine the effectiveness of the registration regime. For the purpose of the disciplinary process it is for the Council to determine what professional responsibilities will be required of dentists and what standards will be set for them in the performance of those responsibilities.”[24] 33. Like the Council, the Court of Appeal found that the appellants were subject to a professional duty as directors of the dental company to ensure that all persons employed by the company to practise dentistry were registered dentists.[25] Although McWalters JA suggested that the Council was not purporting to reach a conclusion as to the legal effect of section 12(2) but merely “expressing its understanding of the spirit and effect of section”[26] in determining whether there had been unprofessional conduct, the Council evidently considered the statutory and professional duties to overlap completely and found that the appellants “as directors failed to fulfil their statutory duty to ensure that all persons employed to perform dental operations are registered dentists”.[27] Indeed, this view of the effect of section 12 was endorsed by the Court of Appeal stating: “In the way the Council approached the statutory provisions of the DRO, and specifically section 12 of that Ordinance, we can find no fault in the Council’s conclusion that the appellants, as directors failed to fulfill their statutory duty to ensure that all persons they employed were registered dentists.”[28] 34. This is reinforced by the Court of Appeal bolstering that conclusion by construing DRO section 12(3) as imposing a “duty to ensure” the registered status of persons employed by the company to practise dentistry before listing them in the annual return. McWalters JA stated: “... it is implicit that by the act of inserting a person’s name in the return the body corporate is warranting that such a person is a registered dentist. Such a representation is crucial to the registration regime and should not be made without verification of the registered status of each person whose name is being included in the return.”[29] 35. Summarising its conclusion, the Court of Appeal held: “We can see nothing wrong in the Council concluding that directors of dental companies have a professional duty to ensure that only registered dentists are employed by the company, have a duty to verify the registered status of employed dentists in compiling the annual return and have a duty to confirm the accuracy of the annual return.”[30] 36. McWalters JA, however, went on to hold that the Council’s finding was “based on negligence”: “It is clear that the Council’s finding was a finding based on negligence. Such a basis for a finding of unprofessional conduct is permitted by the law, but the negligence must be more than ordinary negligence and must rise to the level of ‘an elemental and grievous failure’.”[31] He continued: “Applying this test to the facts of this case as contained in the Statement of Agreed Facts, we are of the view that it was open to the Council to find unprofessional conduct merely on the basis of permitting Mr Tung to practise dentistry through a failure by the appellants to properly discharge their statutory duties. This was a failure, as we have mentioned, that continued over 28 months and encompassed two annual returns. It totally undermined the registration regime and was at the most serious level of non-wilful conduct. The appellants took all the benefits that went with operating a dentistry business through a body corporate but failed to properly discharge their responsibilities that accompanied such a form of practice. In convicting the appellants of non-wilful conduct the Dental Council, in our view, had a correct appreciation of the gravity of the appellants’ conduct. We can see no reason for concluding that it applied the wrong test.”[32] 37. It is not entirely clear how this “test” involving negligence amounting to an “elemental and grievous failure” – which I shall call “gross negligence” – fits in with the postulated “duty to ensure”. It is perhaps best understood as proposing that a failure to ensure the registered status of relevant persons in the absence of knowledge of (or wilful blindness as to) such unregistered status, only constitutes unprofessional conduct if the failure is the result of gross negligence. 38. McWalters JA also pointed to the absence of evidence of what the appellants individually had done to ensure that an adequate administrative system was in place, stating that the evidence “merely described the efforts of various administrative staff of the dental company and did not reveal any specific actions by the appellants”;[33] adding: “The appellants made a deliberate decision not to adduce any evidence to supplement the Statement of Agreed Facts. That left the Council with no evidence from them in respect of what actions, if any, each of them may have taken in the administration of the company and particularly with regard to completing the annual return and ensuring the accuracy of its contents. If the appellants wished to maintain that they had adequate administrative systems in place notwithstanding the obvious failure of those systems in the present case, then it was for them to place such evidence before the Council.”[34] 39. It is unclear whether the Court of Appeal was implicitly convicting the appellants of failing to have in place an “adequate administrative system” and, if so, how that would fit in with the posited duty to ensure or duty based on gross negligence. It is perhaps best understood as translating the postulated duties into a duty placing a burden on defendants to show individual efforts at maintaining an adequate administrative system. 40. In relation to Dr Chu, the Court of Appeal reiterated the legitimacy of imposing professional obligations to bolster the registration regime.[35] It pointed to the “warning” against “covering” set out in section 15 of the Code (which refers dentists to DRO section 27) and stated: “It is hardly surprising in the light of this warning that the Council would view this appellant’s conduct, as a director of the dental company, as constituting unprofessional conduct. Whether the Council would have found the second appellant guilty of unprofessional conduct had he not been a director is another matter. But, that he was a director was clearly a key reason for the Council’s finding for it recognized that there was an overlap in the second appellant’s culpability for the second charge with his culpability for the first charge. Hence, its order that the removal of the second appellant from the General Register for the second charge run concurrently with its removal order in respect of the first charge.”[36] 41. In upholding its decision, the Court of Appeal accorded weight to the Council’s views, stating: “It has long been recognised that bodies, such as the Dental Council, set up to regulate different professions are best placed to decide what standards are required of the professionals subject to their authority and what conduct should constitute professional misconduct.”[37] E. Diffidence of the courts 42. A preliminary matter that arises concerns the degree of diffidence that the courts should show in questioning the disciplinary rulings of the Council. 43. The starting point is section 18(2) which defines unprofessional conduct as “an act or omission of a registered dentist which would be reasonably regarded as disgraceful or dishonourable by registered dentists of good repute and competency”. As the Code explains: “The question of whether any particular course of conduct amounts to unprofessional conduct, and the gravity of such unprofessional conduct and of any conviction, are matters which the Council will determine after considering the evidence in each individual case.”[38] 44. This must of course be understood in the light of the governing statutory definition. What the Code is saying is that the Council decides whether the dentist’s conduct in a particular case should be “reasonably regarded as disgraceful or dishonourable” in accordance with the statutory definition and if so, how grave that misconduct is. It is not saying that the question of what amounts to unprofessional conduct is at large, to be determined by the Council as it sees fit. 45. As noted above the Court of Appeal’s approach was to regard bodies like the Council as best placed to decide whether a dentist’s conduct should be regarded as constituting professional misconduct. In addressing that question within the framework of the statutory definition, that proposition will often be correct. However, much depends on the nature of the issue in question. 46. Thus, in Ghosh v General Medical Council,[39]Lord Millett, delivering the judgment of the Privy Council regarding sentences imposed for professional misconduct, emphasised that the courts’ powers “are not as limited as may be suggested by some of the observations which have been made in the past.”[40] His Lordship held that while the Board would “accord an appropriate measure of respect to the judgment of the committee whether the practitioner's failings amount to serious professional misconduct and on the measures necessary to maintain professional standards and provide adequate protection to the public”, it would “not defer to the committee's judgment more than is warranted by the circumstances.” 47. In Preiss v General Dental Council,[41] Lord Cooke of Thorndon pointed out that with the coming into operation of the Human Rights Act 1998, the tendency to “read down rights of appeal in disciplinary cases is to be resisted”. Echoing Lord Millett in Ghosh v General Medical Council, Lord Cooke added: “An instance, on which some reliance was placed for the General Dental Council in the argument of the present appeal, is the observation in Libman v General Medical Council [1972] AC 217, 221, suggesting that findings of a professional disciplinary committee should not be disturbed unless sufficiently out of tune with the evidence to indicate with reasonable certainty that the evidence was misread. That observation has been applied from time to time in the past, but in their Lordships’ view it can no longer be taken as definitive. This does not mean that respect will not be accorded to the opinion of a professional tribunal on technical matters. But, as indicated in Ghosh, the appropriate degree of deference will depend on the circumstances.” 48. This approach was adopted by the Court of Appeal in Lau Koon Leung v Medical Council of Hong Kong.[42] While acknowledging that the courts should remain cautious in respect of sentences, Yuen JA explained that the earlier reluctance of the courts to interfere was consistent with professional bodies not then being required to give reasons for their disciplinary rulings, the position having since changed. This approach, in which the “appropriate degree of deference” depends on the circumstances, was approved by this Court in Solicitor (24/07) v Law Society of Hong Kong.[43] 49. Clearly, it is sensible for the Court to be diffident about interfering with a disciplinary body’s assessment of matters calling for professional judgment and expertise. This is likely to be the case, for instance, where professional misconduct is alleged in connection with technical or clinical judgments or competence. It may also be the case where the unprofessional conduct raises questions of professional ethics governing, for instance, the relationship between a professional and his or her clients or patients or fellow members of the profession. But diffidence is not called for where the alleged misconduct does not concern such areas of professional expertise. The Court in Mallon v General Medical Council,[44] expressed this as follows: “The spectrum of serious professional misconduct can range from conduct that is entirely non-clinical, such as defrauding the National Health Service or sexually harassing an employee or colleague, to conduct in the course of clinical practice, such as the carrying out of a reckless surgical procedure. In a case of the former kind, a court might conclude that there was little to inhibit it from substituting its own judgment for that of the panel if it should have cause to differ from it. In a case of the latter kind, which involves a technical question of medical malpractice, the court is, we think, at a serious disadvantage to the panel whose decision is impugned.” 50. The present case does not call for reticence on the Court’s part. The alleged unprofessional conduct concerns the responsibility of director-dentists for the fact that their company had employed a dentist who had, without their knowledge or the company’s knowledge, continued to practise after failing to renew his practising certificate and consequently been de-registered. It is not a case involving technical or clinical judgment or professional expertise. It involves attributing responsibility to directors for a corporation’s failure to comply with a regulatory requirement – a matter that the Court is itself well-equipped to deal with albeit in the context of the practice of dentistry. F. The 1st charge 51. The Council was right to concede that the convictions cannot be sustained. The many flaws in the process leading to those convictions began with the way the 1st charge is formulated. 52. Since the case against the appellants was that they were guilty of unprofessional conduct, it was incumbent on the Council, applying the definition in section 18(2), to identify the act or omission which it was alleged would be reasonably regarded as disgraceful or dishonourable by registered dentists of good repute and competency. The 1st charge[45] simply alleges that the appellants “employed JT to perform dental service when he was no longer a registered dentist” and were thereby guilty of unprofessional conduct. The act relied on was therefore the act of employing JT. The transcript shows that, prosecuting the case on the Council’s behalf, its Legal Officer essentially did no more than read out the charges, the Statement of Agreed Facts and official documents relating to registration and annual returns. On that basis, he advised the Council that “The Secretary has no difficulty in proving charge (i)”. The solicitor appearing for the appellants said that she “accepted the facts of charge (i)” and had no comments to make. 53. It was an agreed fact that the act of employing JT was the act of the company with whom he had an employment agreement. It was not the act of the appellants. 54. The charge was thus defective. But as it is framed against each appellant as someone “being a registered dentist ... and being a director [of the company]”, the Council’s intention was evidently to allege unprofessional conduct mirroring the statutory offence created by section 12(2) which applies to director-dentists. The Council’s ruling[46] expressly refers to the appellants as being in breach of the section 12 “statutory duty” and 12(3) is relied on in support. G. Errors in the Council’s decision 55. The Council appears to have drawn no distinction between the appellants as directors and the company itself, stating that they were the company’s “brain and mind” and were “directly responsible for its affairs”. In so far as they were approaching the disciplinary offence as involving unprofessional conduct which reflects the section 12(2) offence, a distinction plainly had to be drawn: the section 12(2A) defence only avails directors and not the company. Directors are not liable if they did not know of the company’s contravention of section 12 whereas the company has no such defence. Directors therefore cannot simply be equated with the company for the purposes of a disciplinary charge based on section 12. 56. The Council attributed to the company knowledge that JT was practising dentistry “without a valid practising certificate”[47] because its staff had repeatedly chased JT for his certificate. It was unnecessary to make such an attribution since the lack of knowledge defence was not in any event available to the company. It may however be that, having equated the directors with the company, the Council wished to treat such knowledge as attributable also to the appellants. If so, that approach was flawed in several respects. 57. First, the administrative staff’s knowledge cannot in law be attributed to the directors, even if it might be attributed to the company. And it cannot somehow be attributed to the directors via an attribution of knowledge to the company. 58. Secondly, the 1st charge concerned employment of JT “when he was no longer a registered dentist after his name had been removed from the General Register”. Practising when de-registered is a very different thing from practising dentistry “without a valid practising certificate”. Section 15(3)(ba) only permits the Council to order the removal of a dentist’s name from the General Register if he or she has practiced dentistry in Hong Kong for a period exceeding six months without having obtained a valid practising certificate. JT’s name was only removed some seven months after his practising certificate had expired at the beginning of 2008. A registered dentist may thus not have a valid practising certificate while remaining on the General Register for a period of six months or more. To infer that the staff knew that JT was practising without a valid practising certificate does not justify a finding that they knew he was no longer registered. 59. Thirdly, and most importantly, it was an agreed fact (read out at the hearing) that neither the appellants nor the company knew that JT was unregistered. It was therefore not open to the Council to find the 1st charge proved on the basis that the appellants is to be treated as having knowledge of JT’s unregistered status. Moreover, the 1st charge as formulated was not based on any attribution of knowledge and it was not suggested at the hearing that the alleged knowledge of the staff should be ascribed to the appellants because they were “directly responsible for” the company’s affairs. Factual questions concerning the management role played by the appellants were not explored and no evidence was adduced to justify a finding that they were “directly responsible for its affairs”. 60. It may be that in order to overcome the difficulties posed by the agreed fact that the appellants had no knowledge, the Council also found that the appellants had failed to verify or had “turned a blind eye” to JT’s non-registration and found them guilty for “failing to fulfil their statutory duty to ensure that all persons employed to perform dental operations are registered dentists”. This too is flawed. “Turning a blind eye” is generally equated in law with knowledge so that this finding remains contrary to the agreed facts. Secondly, the posited statutory duty on the appellants as director-dentists to ensure that all persons employed to perform dental operations were registered dentists cannot be found. Section 12 is of course the prime candidate provision. It penalises the company for carrying on business in contravention of that section but exempts directors who can show that the company’s contravention occurred without their knowledge. Provisions which exclude directors from liability where they are unaware of the company’s default (an agreed fact in the present case) cannot be read as placing on them a duty under section 12 “to ensure that all persons employed to perform dental operations are registered dentists” regardless of the state of their knowledge. 61. Regarding the 2nd charge (which the Council took to be based on DRO section 27), the Council – in my view correctly – considered it necessary to establish knowledge on the part of the dentist charged that the unregistered person was practising dentistry in the same premises. On its face, section 27 penalises the common use of premises for practising dentistry even where the two persons concerned are unaware of each other and have no mutual interaction of any kind. Such a penalty would serve no useful purpose and the legislative intention cannot have been to lay down such a strict, Draconian, liability.[48] Section 27 therefore requires to be construed employing the presumption of mens rea[49] and the Council was right to ask whether Dr Chu knew that JT was practising in the same premises when no longer registered. 62. The Council concluded that such knowledge was irresistibly to be inferred in the present case because (i) DRO section 14 requires dentists conspicuously to exhibit their certificates of registration at the premises where they practise; and (ii) Dr Chu was a director of the company. That inference was, however, contrary to the agreed fact that the appellants, including Dr Chu, did not know of JT’s unregistered status. 63. In my view, the decision of the Council was fatally flawed and the appellants should not have been convicted on either charge. H. Errors in the Court of Appeal’s judgment H.1 Endorsement of the Council’s decision based on section 12 64. Two main strands are detectable in the Court of Appeal’s judgment. First, the Court endorsed the Council’s appellants’ conviction of unprofessional conduct based on breach of their postulated statutory duty under section 12. It supported the finding that the appellants, as directors, had failed to fulfil their statutory “duty to ensure that all persons employed to perform dental operations are registered dentists” under section 12.[50] When construing section 12(3), it held that by filing the annual returns, the company had warranted that JT was a registered dentist.”[51] 65. The Court of Appeal then proceeded from implying this warranty by the company to holding that each of the appellants as directors were individually in breach of both a statutory and a professional duty “to ensure that only registered dentists are employed by the company, ... to verify the registered status of employed dentists in compiling the annual return and ... to confirm the accuracy of the annual return.”[52] 66. McWalters JA found the Council’s inference of knowledge “problematic”,[53] but sought to explain it as a finding “not based on a factual finding that each of the appellants knew the unregistered status” of JT but merely reflecting the Council’s finding of a failure to verify as required by section 12(3).[54] His Lordship held that it was “clear that the Council’s finding was a finding based on negligence”, going on to add that in the present case, this provided a good basis for the conviction since the Council was entitled to view the appellants’ negligence as “more than ordinary negligence” and negligence rising “to the level of an ‘elemental and grievous failure’.”[55] 67. For the reasons given in Section G above, the reasoning of the Council is seriously flawed and the Court of Appeal’s endorsement of that reasoning cannot be supported. I would add that I can see no basis for implying the asserted warranty from the words of section 12(3). As further discussed in Section H.2b below, that subsection merely requires a dental company to make annual returns setting out the names and addresses of the directors or managers of the company and those who perform dental operations in connection with the business of the company. Failure to do so results in the company being deemed to be carrying on the business of dentistry in contravention of section 12, which would then expose the directors to liability under section 12(2) but always subject to the section 12(2A) defence of lack of knowledge – which is available on the facts agreed. The assertion that it was “clear that the Council’s finding was a finding based on negligence” is surprising since there appears to be no such suggestion in the Council’s decision. H.2 A disciplinary offence to support the registration regime 68. The second strand of the Court of Appeal’s decision is based on breach of a professional duty fashioned as a necessary adjunct to the statutory registration regime to prevent dilution of the protection that regime affords to the public. McWalters JA stated that: “It has long been recognised that bodies, such as the Dental Council, set up to regulate different professions are best placed to decide what standards are required of the professionals subject to their authority and what conduct should constitute professional misconduct.”[56] 69. His Lordship held that the Court of Appeal could see no legal reason why the Dental Council cannot “quite properly regard as unprofessional conduct acts or omissions by a dentist which significantly undermine the effectiveness of the registration regime”[57]. Emphasising that the Council is not confined by any mens rea or other requirements of statutory offences in the DRO,[58] McWalters JA added: “If, within the context of the DRO and the registration regime it creates, the Council can reasonably adjudge that imposition on dentists of certain responsibilities will assist in ensuring the effectiveness of the registration regime and contribute to protecting the public from having dental services provided to them by unregistered dentists, then it is entitled to determine that a failure to comply with those responsibilities may, depending on the circumstances of the individual case, constitute unprofessional conduct.”[59] 70. On this strand of its reasoning, it appears that the Court of Appeal considered professional misconduct sufficiently established if gross negligence is proved.[60] 71. So far as penalising individual directors of a dental company was concerned, McWalters JA held that: “Regarding these directors as ‘directly responsible for the affairs of the company’ is quite uncontroversial and consistent with both the law and common sense.”[61] 72. As we have seen, the Court of Appeal concluded that a professional duty could be deduced as follows: “We can see nothing wrong in the Council concluding that directors of dental companies have a professional duty to ensure that only registered dentists are employed by the company, have a duty to verify the registered status of employed dentists in compiling the annual return and have a duty to confirm the accuracy of the annual return.”[62] 73. Moreover, the Court of Appeal evidently considered it a facet of this duty to ensure or verify, that each director was required to make individual efforts to ensure that an adequate system existed to verify the registered status of dentists in the position of JT. 74. With respect, in my view, this strand of the Court of Appeal’s reasoning cannot be supported for three main reasons. H.2a “Unprofessional conduct” and section 18(2) 75. In the first place, it is not correct to hold that the Council is free to create a head of unprofessional conduct simply because the Council “can reasonably adjudge that imposition on dentists of certain responsibilities will assist in ensuring the effectiveness of the registration regime”. 76. What constitutes “unprofessional conduct” is not at large. It is not dependent simply on the Council’s view as to what might reasonably be required to support a desirable statutory objective. The Council’s disciplinary powers are conferred by DRO section 18(1) which provides that if, after due inquiry the Council is satisfied (inter alia) that a registered dentist has been guilty of unprofessional conduct, it may impose specified penalties on him ranging from de-registration down to a reprimand. Crucially, section 18(2) defines “unprofessional conduct” for these purposes as “an act or omission of a registered dentist which would be reasonably regarded as disgraceful or dishonourable by registered dentists of good repute and competency.” 77. This is not to suggest that unprofessional conduct is confined to matters prohibited by specific DRO offences on proof of the elements necessary to sustain criminal liability. On the contrary, the Council is plainly entitled to identify classes of acts or omissions as constituting unprofessional conduct and to give guidance to that effect in the disciplinary Code as well as to adjudicate charges relating to such conduct in individual cases, provided always that the relevant acts or omissions would be reasonably regarded as disgraceful or dishonourable by registered dentists of good repute and competency.[63] While, as noted above,[64] the Court is likely to be reluctant to interfere with the Council’s determination of what constitutes such misconduct where the decision involves clinical expertise or a professional ethical judgment and less reluctant to do so where it involves no such specialist expertise, the question whether the acts or omissions relied on are properly classified as unprofessional conduct is ultimately a question of whether section 18(2) has been properly applied – a matter to be decided by the Court. 78. The first difficulty with the Court of Appeal’s second strand of reasoning is that it seeks to justify the convictions on the basis of a professional duty founded solely on a perceived need to reinforce the statutory registration regime without reference to the section 18(2) definition. The Court of Appeal held that such professional misconduct was established if the appellants’ “failure to ensure” was grossly negligent and that “[applying] this test to the facts ... contained in the Statement of Agreed Facts, ... it was open to the Council to find unprofessional conduct merely on the basis of permitting Mr Tung to practise dentistry through a failure by the appellants to properly discharge their statutory duties”.[65] That was evidently a reference to a failure to verify JT’s registered status when filing annual returns “that continued over 28 months and encompassed two annual returns”.[66] Leaving aside the difficulty of construing section 12 as imposing such a duty discussed above, the key question – whether and if so, how, their alleged failure in relation to the company’s duty to make such annual returns can be reasonably regarded as disgraceful or dishonourable by registered dentists of good repute and competency – was not addressed. H.2b The statutory registration regime and the appellants’ conduct 79. On a proper appreciation of the nature of the statutory registration regime and of any default that may be attributable to the appellants, I am unable to see how their conduct could reasonably be said to fall within the section 18(2) definition. This is the second main reason for my inability to accept this strand of the Court of Appeal’s reasoning. 80. The Court of Appeal evidently considered the conduct of the appellants such as to “significantly undermine the effectiveness of the registration regime”.[67] McWalters JA went so far as to hold that their failure to verify JT’s registered status “totally undermined the registration regime and was at the most serious level of non-wilful conduct”.[68] I am unable to accept that this represents a realistic or fair assessment. 81. It is no doubt correct to say that the registration regime is important in that it lays down the threshold conditions for being permitted to practise dentistry and provides the basic framework for the Council’s regulation of the profession. However, the DRO’s provisions show that the scheme is intended to involve a system for monitoring the registration of dentists operated by the Council and the Registrar. The primary obligation to secure and maintain one’s registered status is placed on the individual dentist. 82. Thus, it is the Council that sets the licensing exams which (alongside university degrees) qualify a dentist for registration.[69] It is the Registrar who keeps and maintains the General and Specialist Registers and it is to the Registrar that a dentist applies for registration.[70] It is the Council that issues the certificate of registration and (through its Secretary) the annual practising certificate.[71] And it is the Council which has power to order the name of a registered dentist to be removed from the Register.[72] 83. The fact that the Council issues both the registration and practising certificates and orders de-registration means that it knows at any moment who are the persons holding such certificates and also who has been de-registered or failed to renew their practising certificates. As the Court of Appeal noted, the Council publishes the names of dentists in the General Register on its website.[73] 84. By requiring a dental company to file an annual return stating the names and addresses of all persons who are directors or managers of the company, or who perform dental operations in connection with the business of the company, section 12(3) is designed to give the Council all the information it needs to enforce the company’s duty not to carry on business in contravention of section 12. Once a company files its annual return, the Council is able to tell from its own records whether any of the persons listed as dentists have not been issued with a registration or practising certificate or has had his or her name removed from the General Register. There is no need to imply a warranty that the company – much less its directors – has independently verified the registered status of all its dentists. The Council, doing its job, would in any event naturally be expected to check the annual return against its own authoritative records. 85. As DRO section 3 makes clear, the system places the primary duty to maintain one’s registered status on the individual dentist. It makes it a criminal offence for any person to practise dentistry, not being a registered dentist, the penalty on summary conviction being a fine of up to $100,000 and imprisonment for three years; and upon conviction on indictment being imprisonment for 5 years or, in cases where injury results, imprisonment for 7 years. 86. In contrast, the maximum penalty prescribed by section 12(2) where a dental company carries on business in contravention of section 12 is a fine of $2,000. This would apply where the company fails to file a proper section 12(3) annual return or where it carries on business without every person practising dentistry as part of that business being a registered dentist. The regime obviously casts dental companies in a merely incidental role, limited to providing the information required in the annual return. By giving directors a defence if they can show that the company’s default occurred without their knowledge, the regime places an even smaller responsibility on them. 87. Significantly, under DRO section 18(1)(a), one of the conditions for the Council to exercise its disciplinary powers is a registered dentist’s conviction “of any offence punishable with imprisonment”. It follows that even if the appellants had been convicted of an offence under section 12(2), such conviction would not have been sufficient to ground disciplinary proceedings since the offence is not punishable with imprisonment. 88. There is therefore no basis in the present case for thinking that the appellants “undermined the effectiveness of the registration regime” significantly or at all. The system might be undermined if something had been done to mislead the Council or prevent it from knowing the true registered status of a person practising dentistry. The appellants did nothing of the sort. In filing the annual returns, the company duly provided the information required. The Council was well aware from its own records that JT had not renewed his practising certificate and, seven months later, it took his name off the General Register. They knew from the annual returns that he had been practising dentistry with the company during that period and brought disciplinary proceedings against the appellants. I am therefore quite unable to see how the appellants’ acts or omission could reasonably be regarded as disgraceful or dishonourable by registered dentists of good repute and competency. There was no justification here for creating a disciplinary offence going beyond the existing statutory sanctions with the purported aim of buttressing the system of registration which has no discernible need for such buttressing. 89. Concern was expressed by the Court of Appeal as to how undermining the registration regime would dilute protection of the public from unregistered dentists. Certainly, in a case involving an unqualified, unlicensed backstreet practitioner, the registration requirement and the severe penalties under DRO section 3 would play an important role to protect the public against the unregistered practitioner. But no such concerns apply in cases like the present. 90. The third reason for my inability to accept the second strand of the Court of Appeal’s reasoning relates to procedural unfairness. The disciplinary offence elaborated by the Council and the Court of Appeal is far removed from the charges faced by the appellants and their convictions are based on factual and legal bases which were not part of the Council’s case and were contrary to the agreed facts, namely: that they were directly responsible for the company’s affairs and could be treated as knowing of JT’s unregistered status; that they were guilty of gross negligence in failing to ensure or to verify the registered status of JT; that they had failed to make individual efforts to ensure that an adequate administrative system existed in the company to verify his registered status; and that by their conduct, they had undermined the statutory registration regime. As Mr Wong SC rightly concedes, the appellants were not given the opportunity to address these issues. I. The Questions on which leave to appeal was granted[74] 91. The Questions of law have substantially been answered in the course of the foregoing discussion. I.1 Questions 1 and 2 92. Questions 1 and 2 can be taken together. The Questions are premised on the unprofessional conduct being based on the company’s breach of section 12. On that footing the answer to Question 1 is plainly “No”. Since Question 1 accepts that the appellants did not know of JT’s removal from the General Register, the appellants are exonerated under section 12(2A). On this footing, the answer to Question 2 would also be “No”. The absence of knowledge would be decisive and unprofessional conduct mirroring statutory liability under section 12 could not be established on the basis of either ordinary or gross negligence. 93. However, I have examined the Court of Appeal’s second strand of reasoning which seeks to justify the convictions on the basis of a free-standing head of unprofessional conduct that is not a mere reflection of the statutory offence. It is fashioned as described in Section H.2 above, and conceives of misconduct consisting of a grossly negligent breach of a duty to ensure JT’s registered status penalised in order to bolster the registration regime. For the reasons set out above, I have concluded that there is no justification for convicting the appellants of such unprofessional conduct and so would equally give negative answers to Questions 1 and 2 on this alternative basis. I.2 Question 3 94. For the reasons given in Sections H.1 and H.2b above, the answer to Question 3 is “No”. A professional duty on individual director-dentists to verify the registered status of employed dentists and to confirm the accuracy of the section 12(3) annual return cannot be imposed on the basis that filing such an annual return amounts to warranting such registered status. J. Remitter and disposal of the appeal J.1 Jurisdiction to remit 95. The Council has invited the Court to “remit the matter back to the Council for fresh consideration (and with further amendment to the Charges (with leave of the Council) if thought fit)”. The first question is therefore whether the Court has jurisdiction to remit. 96. Section 17 of the Court’s statute[75] relevantly provides: (1) The Court may confirm, reverse or vary the decision of the court from which the appeal lies or may remit the matter with its opinion thereon to that court, or may make such other order in the matter, including any order as to costs, as it thinks fit. (2) For the purpose of disposing of an appeal, the Court may exercise any powers of the court from which the appeal lies (including the power to order a retrial), or may remit the case to that court. 97. Mr Mok SC submits that section 17(1) has no application since the Council is not suggesting a remitter back to the Court of Appeal but to the Council which is obviously not the court from which the appeal lay. 98. He also argues that Section 17(2) does not assist because the Court of Appeal itself does not have jurisdiction to order a remitter since DRO section 23 only authorises that Court to “affirm, reverse or vary” the order or decision of the Council appealed from, making no mention of a power to remit. He points out that in contrast, the Medical Registration Ordinance (“MRO”)[76] expressly empowers the Court of Appeal to “remit the case to the [Medical] Council for an, or another, inquiry”. 99. In support of his argument, Counsel traces the legislative history of DRO section 23 which shows that the “affirm, reverse or vary” formula has remained unchanged since 1959, leading the Court of Appeal in Dr Roger Gordon Bishop v The Dental Council of Hong Kong,[77] and Chow Wai Fu v Medical Council of Hong Kong,[78] to hold that there was no jurisdiction to order a re-hearing and to Fuad JA (who delivered both judgments) suggesting that the DRO be amended to add such a power. 100. Mr Mok SC furthermore points out that although the MRO was amended in 1996[79] to insert the express power to remit, when the DRO, including section 23, was amended in 2006,[80] the legislature did not do likewise. 101. Although those submissions were attractively put, I do not accept that jurisdiction to remit is lacking. 102. It is true that DRO section 23, unlike certain other Ordinances, does not expressly confer a power to remit. However, I agree with Mr Wong SC’s submission that an express power to remit is unnecessary in the light of the provisions of Order 59 r 10 of the Rules of the High Court[81] which, by Order 59 r 1,[82] are made applicable to every appeal to the Court of Appeal (subject to immaterial exceptions). Order 59 r 10 relevantly provides as follows: “(2) The Court of Appeal shall have power to receive further evidence on questions of fact, either by oral examination in court, by affidavit, or by deposition taken before an examiner, but no such further evidence (other than evidence as to matters which have occurred after the date of the trial or hearing) shall be admitted except on special grounds. (3) The Court of Appeal shall have power to draw inferences of fact and to give any judgment and make any order which ought to have been given or made, and to make such further or other order as the case may require. (4) The powers of the Court of Appeal under the foregoing provisions of this rule may be exercised notwithstanding that no notice of appeal or respondent's notice has been given in respect of any particular part of the decision of the court below or by any particular party to the proceedings in that court, or that any ground for allowing the appeal or for affirming or varying the decision of that court is not specified in such a notice; and the Court of Appeal may make any order, on such terms as the Court thinks just, to ensure the determination on the merits of the real question in controversy between the parties.” 103. These provisions are intended to supplement enactments which create a right of appeal to the Court of Appeal, making it unnecessary for them to spell out these general powers on each occasion. Order 59 r 10(4) is important. First, it indicates that the powers conferred by the earlier sub-rules are intended to apply in tandem with the Court of Appeal’s powers “for allowing the appeal or ... affirming or varying” the decision appealed from. And secondly, it empowers the Court of Appeal to “make any order, on such terms as the Court thinks just, to ensure the determination on the merits of the real question in controversy between the parties.” 104. Those latter words are apt to cover a situation where the Court of Appeal considers a remitter appropriate consequential upon its disposal of the appeal. Thus, it may for instance decide to reverse the Council because of a procedural flaw, but think that, on the merits, there may have been serious professional misconduct which ought to be investigated. The Court may thus order a remitter so that the merits of the real question in controversy – the charge of unprofessional conduct – can properly be determined. 105. Even where, as applies in this Court, no similar rule has been expressly enacted, it is well-established that there is implied power to make orders reasonably required for the effective exercise of the statutory jurisdiction conferred upon it,[83] mirroring the power provided for in Order 59 r 10(4). 106. In Dr Bishop’s case,[84] the absence of a power to remit was conceded by counsel without reference to the general powers conferred by Order 59 r 10. Similarly, in Chow Wai Fu’s case,[85] those powers were not referred to and the point was not discussed in any detail. 107. Little can be made of the fact that section 23 was not amended to insert an express power to remit when the DRO was amended in 2006. As the Hansard and legislative briefing materials show,[86] the purpose of the amendment was to create a Specialist Register putting practitioners of specialist dentistry on a statutory footing and, in relation to section 23, to lay down a time limit for appealing to the Court of Appeal. The scope of the powers of the Court of Appeal was not touched upon. In the light of Order 59 r 10, an amendment expressly to provide for a power to remit was in any event unnecessary. 108. I therefore conclude that this Court has jurisdiction, in exercising a power available to the Court of Appeal, to remit a disciplinary inquiry to the Council for re-hearing. J.2 Whether there should be a remitter as a matter of discretion 109. In my view, there are powerful discretionary reasons for refusing to order a remitter in the present case. 110. The Council’s purpose in seeking a remitter is to bring disciplinary proceedings on a modified basis, alleging a grossly negligent failure on the appellants’ part to ensure or verify JT’s registered status. That purpose is premised on affirmative answers to Questions 1 to 3. Those questions have, however, been answered in the negative and I have concluded that the proposed alternative basis for alleging unprofessional conduct rests on an erroneous understanding of the registration regime and a flawed assessment of the culpability of the appellants. No viable basis has been demonstrated for describing the conduct of the appellants – who would be exonerated for lack of knowledge of the company’s default if charged under section 12 – as involving acts or omissions which would be reasonably regarded as disgraceful or dishonourable by their peers. The Council’s intended purpose would therefore not be achieved by a remitter. 111. The appellants have had these proceedings hanging over their heads since January 2010. They have had to cope with findings of professional misconduct and severe sentences, upheld by the Court of Appeal, which I have concluded were seriously deficient from the outset. If a remitter were ordered, as Mr Wong SC acknowledged, the Council would not simply be asked to retry the existing disciplinary charges since they are plainly defective. The appellants would have to face quite different charges tailored, if possible, to this Court’s judgment. As a matter of discretion, I can see no justification for permitting the Council such a second bite of the cherry. J.3 Disposal of the appeal 112. For the foregoing reasons, I would allow the appeal and refuse an order remitting the matter back to the Council. In the light of the view I have taken that the proceedings were flawed from the outset, I would make an order nisi that the Council pay the appellants their costs here and below, including the costs of the Inquiry. I would direct that any submissions which the parties may wish to make as to costs should be lodged in writing within 21 days from the date of the judgment and that, in default of such submissions, the order as to costs should stand as an order absolute without further direction. Mr Justice Tang PJ: 113. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Fok PJ: 114. I agree with the judgment of Mr Justice Ribeiro PJ. Lord Phillips of Worth Matravers: 115. I agree with the judgment of Mr Justice Ribeiro PJ. Chief Justice Ma: 116. The Court unanimously allows the appeal and makes the orders referred to in the final paragraph of Mr Justice Ribeiro PJ’s judgment. Mr Johnny Mok SC and Ms Catrina Lam, instructed by Ching & Co., for the Appellants Mr Stewart Wong SC, Mr Anthony Ismail and Ms Bonnie Y.K. Cheng, instructed by the Department of Justice, for the Respondent [1] Cap 156. [2] DRO section 3. The fine on summary conviction is at level 6 (up to $100,000: Criminal Procedure Ordinance, Cap 221, section 113B and Sch 8). The qualifying examinations are prescribed by DRO section 8. [3] DRO section 11A(1). [4] Section 12 deals specifically with dental companies. [5] DRO section 12(2). [6] Promulgated on 1 September 2000 and revised in July 2008. [7] A level three fine is a fine of up to $10,000: Criminal Procedure Ordinance, Cap 221, section 113B and Sch 8). [8] DRO section 18(2). [9] Statement of Agreed Facts §4. [10] Ibid §11. [11] Jeremy Tung v The Dental Council of Hong Kong, HCAL 35/2011 (14 August 2012: Au J) which contains an account of JT’s evidence as to the circumstances of his failure to renew his practising certificate. [12] Hon Lam VP, Kwan and McWalters JJA, CACV 149/2013 (3 July 2014). [13] Ribeiro, Tang and Fok PJJ, FAMV 2/2015 (22 March 2016). [14] With Mr Anthony Ismail and Ms Bonnie Y K Cheng. [15] With Ms Catrina Lam. [16] Inquiry §17. [17] Inquiry §19. [18] Inquiry §15. [19] Inquiry §18. [20] Inquiry §19. [21] Court of Appeal §§46-47. [22] Court of Appeal §39. [23] Court of Appeal §§32-34, 44-45, 47-50, 54, 67-68. [24] Court of Appeal §44. See also §45. [25] Court of Appeal §51. [26] Court of Appeal §49. [27] Inquiry§19 and §21, italics supplied. [28] Court of Appeal §51. This was reiterated at §65. [29] Court of Appeal §50. [30] Court of Appeal §51. [31] Court of Appeal §66, citing Dr Young Yau Yau, Cecilia v The Dental Council of Hong Kong, (Unreported) CACV 150/2007 (30 July 2008). [32] Court of Appeal §67. [33] Court of Appeal §59. [34] Court of Appeal §60. [35] Court of Appeal §68. [36] Court of Appeal §69. [37] Court of Appeal §65, citing O Ting Yuen v The Medical Council of Hong Kong (Unreported) CACV 820/2000 (21 March 2001). [38] Code, Introduction. [39] [2001] 1 WLR 1915 (PC) at §34. [40] Citing Evans v General Medical Council (unreported) 19 November 1984 where the Board said: “The principles upon which this Board acts in reviewing sentences passed by the Professional Conduct Committee are well settled. It has been said time and again that a disciplinary committee are the best possible people for weighing the seriousness of professional misconduct, and that the Board will be very slow to interfere with the exercise of the discretion of such a committee… The committee are familiar with the whole gradation of seriousness of the cases of various types which come before them, and are peculiarly well qualified to say at what point on that gradation erasure becomes the appropriate sentence. This Board does not have that advantage nor can it have the same capacity for judging what measures are from time to time required for the purpose of maintaining professional standards.” [41] [2001] 1 WLR 1926 (PC) at §27. [42] [2006] 3 HKLRD 225 at §§64-72. [43] (2008) 11 HKCFAR 117 at §§53 and 119-120. [44] 2007 SC 426 at §20. [45] Set out in Section B.2 above. [46] Discussed in Section C above. [47] Inquiry §18. [48] Section 15 of the Code indicates that the mischief of “covering” as a form of unprofessional conduct involves a registered dentist countenancing, helping, encouraging or assisting the practice of dentistry by an unregistered person. These are actions involving a form of knowing interaction (although the Code says such conduct may be performed “wilfully or by neglect”). [49] See Hin Lin Yee v HKSAR (2010) 13 HKCFAR 142. [50] Court of Appeal §§51, 62, 63, 65. [51] Court of Appeal §§49 and 50. [52] Court of Appeal §51. [53] Court of Appeal §57. [54] Court of Appeal §63. [55] Court of Appeal §66, citing Dr Young Yau Yau, Cecilia v The Dental Council of Hong Kong, (Unreported) CACV 150/2007 (30 July 2008). [56] Court of Appeal §65, citing O Ting Yuen v The Medical Council of Hong Kong (Unreported) CACV 820/2000 (21 March 2001). [57] Court of Appeal §44. [58] Court of Appeal §46. [59] Court of Appeal §45. [60] Court of Appeal §§66 and 67. [61] Court of Appeal §48. [62] Court of Appeal §51. [63] An example of such a situation is Ng Kin Wai v Dental Council of Hong Kong [2011] 6 HKC 378 which concerned the use by a dentist the title of specialist and the dentist’s duty to familiarise himself with the relevant legislation concerning professional registration and the use of titles. [64] Section E of this judgment. [65] Court of Appeal §67. [66] Ibid. [67] Court of Appeal §44. [68] Court of Appeal §67. [69] DRO sections 4A and 8. [70] DRO sections 6, 7, 9 and 15. [71] DRO sections 10(1) and 11A. [72] DRO section 15(3). [73] Court of Appeal §9. [74] Set out in Section B.3 above. [75] Hong Kong Court of Final Appeal Ordinance (Cap 484). [76] Cap 161, section 26(1). [77] (Unreported) Leonard VP, Cons and Fuad JJA, CACV 52/1983 (12 July 1983) at p 18. [78] (Unreported) Fuad VP, Macdougall and Nazareth JJA, CACV 205/1991 (4 March 1992) at p 18. [79] Medical Registration (Amendment) Ordinance (No 7 of 1996). [80] Dentists Registration (Amendment) Ordinance (No 11 of 2006). [81] Cap 4A. [82] Order 59 r 1 also provides that “references to ‘the court below’ apply to any Court, tribunal or person from which such appeal lies”. [83] PCCW-HKT Telephone Ltd v Telecommunications Authority (2005) 8 HKCFAR 337 at §§40 and 60; Joint & Several Liquidators of Kong Wah Holdings Ltd v Grande Holdings Ltd (2006) 9 HKCFAR 795 at §13; Hall v HKSAR (2014) 17 HKCFAR 17 at §11. [84] (Unreported) Leonard VP, Cons and Fuad JJA, CACV 52/1983 (12 July 1983) at p 18. [85] (Unreported) Fuad VP, Macdougall and Nazareth JJA, CACV 205/1991 (4 March 1992) at p 18. [86] Second Reading of the Dentists Registration (Amendment) Bill, Hansard (6 July 2005); Legislative Council Brief for Bills Committee, LC Paper No CB(2)400/05-06(01) 18 November 2005; Answer to Bills Committee queries, LC Paper No CB(2)1767/05-06(02) 18 April 2006; Legislative Council Report of the Bills Committee for the House Committee meeting on 28 April 2006, LC Paper No CB(2)1831/05-06 28 April 2006. 
The Court: A. INTRODUCTION 1. Since June 2019, Hong Kong has experienced serious social unrests and public disorders marked by protests, escalating violence, vandalisms and arsons across the territory. It is a dire situation that has not been seen in the last 50 years. 2. The increasing violence has resulted in widespread property damage, assaults on persons, serious damage and interruptions to major public transport facilities and highways. The violence and damage are mostly caused by protestors wearing masks and dressed in black outfits. At the same time, it is a common phenomenon that many other protestors participating in public assemblies and processions who are not involved in violence are also wearing masks and dressed in black outfits. 3. To tackle the dire situation, on 4 October 2019, the Chief Executive‑in‑Council (“the CEIC”), in the exercise of the power under section 2 of the Emergency Regulations Ordinance (Cap 241) (“the ERO”), announced the enactment of the Prohibition on Face Covering Regulation (Cap 241K) (“the PFCR”) on the basis that there was an occasion of “public danger” in Hong Kong. 4. In gist, the PFCR makes it an offence for anyone without reasonable excuse wearing a mask that will prevent identification at public assemblies or processions, and empowers the police to demand any person to remove the mask and, if that person refuses to do so, to remove it with force if necessary. The PFCR took effect at midnight on 5 October 2019. 5. Soon thereafter, the applicants under HCAL 2945/2019 (collectively, “KWH”) and the applicant under HCAL 2949/2019 (“LKH”) brought these judicial review proceedings to challenge the PFCR on the basis that it is unlawful and invalid as (a) the ERO is itself (i) unconstitutional (Ground 1); (ii) was impliedly repealed (Ground 2); (iii) falls foul of the “prescribed by law” requirement (Ground 3); and (b) in any event, the PFCR (i) is ultra vires by reason of the principle of legality (Ground 4); and (ii) amounts to disproportionate infringement on the protected fundamental rights to freedom of expression, assembly, movement and the right to privacy (Ground 5A on section 3 and Ground 5B on section 5). 6. The rolled-up hearing of these judicial reviews were heard before G Lam and Chow JJ (collectively, “the Judges”) on 31 October and 1 November 2019. By the judgment (“the Judgment”) and decision (“the Decision”) handed down respectively on 18 and 22 November 2019, the Judges allowed the applicants’ judicial reviews under Grounds 1, 5A and 5B, and declared that: (1) The ERO insofar as it empowers the CEIC to make regulations on “any occasion of public danger” is unconstitutional as it is incompatible with the Basic Law (“BL”). (2) The PFCR is therefore also invalid and of no effect. (3) In any event: (a) sections 3(1)(b), (c) and (d) of the PFCR are incompatible with BL27 and Articles 14, 16 and 17 of the Hong Kong Bill of Rights (“BOR”) and are therefore null, void and of no effect; (b) section 5 of the PFCR is also incompatible with BL28 and BOR5, and is therefore null, void and of no effect. 7. By way of the Notices of Appeals filed on 25 November 2019[1], the respondents appeal against the Judgment and the Decision. 8. In response, KWH filed a Notice of Cross Appeal (“KWH‑NoCA”) under CACV 583/2019 seeking to cross-appeal against the Judges’ decision of rejecting Grounds 2 and 3 of their judicial review. They also filed a Respondent’s Notice (“KWH-RN”) in CACV 542/2019 seeking to affirm the Judges’ decision on additional grounds, in particular on Ground 4. 9. On 16 December 2019, LKH also filed a Notice of Cross Appeal and Respondent’s Notice (“LKH-NoCA”) under CACV 541/2019 against the Judges’ decision seeking to challenge: (1) the Judges’ holding on “legitimate aims” and “rational connection” in the proportionality analysis of section 5 of the PFCR under Ground 5B: LKH-NoCA, [1] - [3]; (2) the Judges’ rejection of Ground 3: LKH-NoCA, [4]; and (3) the Judges’ purported refusal to grant him leave to apply for judicial review on Ground 1 and Ground 5A: LKH-NoCA, [5] - [8]. 10. On 17 December 2019, this Court directed that [5] - [8] of LKH‑NoCA would not be entertained in these appeals. In other words, LKH is only permitted to make submissions in relation to Ground 3 and Ground 5B as set out in the LKH-NoCA. B. BACKGROUND 11. The recent months of protests and social unrest leading to the unprecedented scenes of escalating violence and danger on the streets of Hong Kong were triggered in February 2019 when the Government introduced the Fugitive Offenders and Mutual Legal Assistance in Criminal Matters Legislation (Amendment) Bill 2019 (“the Bill”) to the Legislative Council (“the LegCo”).[2] The Bill was very controversial. Many members of the public, different organisations and professional bodies had expressed grave concerns about it. Many different public assemblies and processions had since been held to urge the Government to amend, delay or withdraw the Bill. 12. After proposing further amendments to the Bill, the Government indicated that it would seek to have the Bill debated and passed by the LegCo before the end of the LegCo session in July 2019. This only led to even more massive protests and public processions. 13. At [2] ‑ [4] of the Judgment, the Judges pointed out that: (1) Although the protests began in opposition to the Bill, the unrest had continued notwithstanding the Government’s decision in June 2019 to suspend the legislative process for the Bill, the acknowledgement since 9 July that the Bill was “dead”, the announcement on 4 September that the Bill would be formally withdrawn and the actual withdrawal of the Bill in the LegCo on 16 October. (2) Since 9 June to 4 October 2019, according to the Government’s records, over 400 “public order events” (referring to assemblies and processions) arising out of the Bill and other matters had been held, with a significant number of them ending up in outbreaks of violence. (3) The degree of violence had escalated and, in particular, on 29 September and 1 October, gatherings took place in many districts in which certain protestors blocked major thoroughfares, vandalised Mass Transit Railway (“MTR”) stations and facilities, government offices and selected shops, and hurled petrol bombs at police officers, vehicles and police stations. 14. Most pertinently, the uncontested evidence shows that the outbreaks of violence are escalated by the more radical and violent protestors employing the “black-bloc” tactics[3] to avoid identification and arrests, who are to a certain extent supported or at least condoned (and hence emboldended) by protestors who participate in initially peaceful public assemblies or processions. Such tactics is facilitated in light of the following[4]: (1) Public order events are highly fluid in nature. A public meeting or procession that is initially lawful and peaceful can quickly turn into an unauthorized or unlawful assembly. (2) The most violent and radical protestors are those who have their faces covered and therefore their identities concealed. They are also often equipped with gas masks and refuse to disperse even when the police deploy tear gas. (3) These radical and violent protestors with their faces covered often mix themselves into a larger group of protestors (taking part in a largely peaceful public meeting or procession) who are also wearing masks. This has rendered identification most difficult when the violent protestors with their faces concealed can easily slip away amidst the chaos they have aroused. (4) The more radical and violent protestors are often supported by many less violent protestors (also wearing facial covers to avoid recognition) by, for example, the provision of resources (such as food and water), tools and even weapons, as well as free rides when the police are taking actions of dispersal and arrests. (5) Moreover, the more radical and violent protestors committing criminal conducts are encouraged by and find it essential to regard that they have the support or back-up of those peaceful protestors who themselves are also wearing face covering masks and present at the public assemblies or processions. (6) In this respect, as the expert evidence shows[5], mask functions as a facilitator of anonymity. When anonymity joins with group function, participants’ responsibilities become easily diffused or shared. Individuals tend to feel they are being supported by a lot of people. This is an emboldenment effect. If the dominant group value or purpose in the situation is anti‑social, the individual will conform to that and more likely to act antisocially[6]. 15. Large scale incidents of breach of peace in Hong Kong ensued. In blatant defiance of the law, the more radical and violent protestors have perpetrated widespread criminal conducts ranging from unlawful assembly to serious property damage, assault on persons, arsons and use of lethal weapons. 16. The unchallenged evidence shows that from 9 June to 4 October 2019, violent and radical protestors had[7]: (1) forcefully and repeatedly charged police cordon lines with the use of weapons whilst protected by body armour, and blocked roads and tunnels (including main thoroughfares) with various objects which had resulted in persons trapped with or in their vehicles and, on some occasions, attacked on drivers who voiced displeasure at such blockages; (2) vandalised and wrecked serious damage to public facilities (such as pavements, roadside fences and barriers, signages, dust bins, lamp posts, traffic lights, street lights and CCTV cameras, etc) and Government buildings (such as the Legislative Council Complex, Police Headquarters, Cheung Sha Wan Government Offices, etc), set fire by burning public properties outside or at police stations and on the streets in multiple districts, and hurled inflammable liquid bombs at police vehicles, police stations and even police officers and at and within the MTR stations; (3) damaged private shopping malls, shops and restaurants, etc; there were also reports of looting and theft in some of the shops that were damaged; (4) damaged residential quarters and harassed residents of the same; (5) crippled the operations of critical transport infrastructure including the Hong Kong International Airport, a large number of the MTR stations and tracks (notwithstanding and in breach of relevant injunctive orders in place, and at the same time causing massive, serious and repeated destruction of facilities in the MTR stations resulting in the closure of multiple MTR stations due to safety concerns and the need for urgent repair as well as taking deliberate acts to stop or delay MTR trains from operating during morning rush), and the Cross-Harbour Tunnel, etc; (6) harassed and attacked ordinary citizens and police officers holding different political views by a range of objects and lethal weapons, such as high-powered laser pointers (which were sometimes shone directly into the eyes of the victim from a short distance), slingshots and catapults to launch a variety of projectiles, sharp or sharpened objects (including box cutters and sharpened bamboo poles), bricks and inflammable liquid bombs etc, causing numerous injuries of various degree; and (7) stopped vehicles and threatened drivers with damage to their vehicles to force them to yield up their mobile telephones for examination or to pay the protestors a sum of money. 17. The evidence therefore has shown that by the beginning of October 2019, the above acts of radical and violent protesters had seriously breached public peace, and posed a grave and genuine danger to the police and other members of the public. Normal functions of the Hong Kong community had been severely disrupted. More importantly, there were signs of and even declared intent by violent protestors to procure further escalation in the degree of violence and vandalism in unlawful assemblies which might as a result turn into riots, pushing Hong Kong to a most perilous situation[8]. 18. As acknowledged by the Judges[9], it was under this “dire situation” of Hong Kong, that the CEIC on 4 October 2019 enacted the PFCR under the ERO. 19. In announcing the reasons and basis for enacting of the PFCR at the press conference on 4 October 2019, the Chief Executive (“the CE”) emphasized that the decision to invoke the ERO as Hong Kong was in “an occasion of serious danger” and was a necessary one in the public interest[10]. In particular: (1) In relation to the current situation, the CE mentioned that: “Protests arising from [the Bill] have continued for nearly four months now. Over this period, protesters’ violence has been escalating and has reached a very alarming level in the past few days, causing numerous injuries and leading Hong Kong to a chaotic and panic situation. We are particularly concerned that many students are participating in these violent protests or even riots, jeopardising their safety and even their future. As a responsible government, we have the duty to use all available means in order to stop the escalating violence and restore calmness in society.”[11] (2) In relation to the PFCR, the CE mentioned that: “… [the PFCR] will create a deterrent effect against masked violent protesters and rioters, and will assist the Police in its law enforcement… … … the objective of [the PFCR] is to end violence and restore order, and I believe this is now the broad consensus of Hong Kong people. … [the PFCR] targets rioters or those who resort to violence. That’s why [the PFCR] contains defence and exemptions to cater for legitimate needs to wear a mask, and we believe that by so doing we have struck the necessary balance.”[12] C. THE ERO AND THE PFCR 20. The ERO is a short piece of legislation with only three substantive sections. They provide as follows: “2. Power to make regulations (1) On any occasion which the Chief Executive in Council may consider to be an occasion of emergency or public danger he may make any regulations whatsoever which he may consider desirable in the public interest. (2) Without prejudice to the generality of the provisions of subsection (1), such regulations may provide for— (a) censorship, and the control and suppression of publications, writings, maps, plans, photographs, communications and means of communication; (b) arrest, detention, exclusion and deportation; (c) control of the harbours, ports and waters of Hong Kong, and the movements of vessels; (d) transportation by land, air or water, and the control of the transport of persons and things; (e) trading, exportation, importation, production and manufacture; (f) appropriation, control, forfeiture and disposition of property, and of the use thereof; (g) amending any enactment, suspending the operation of any enactment and applying any enactment with or without modification; (h) authorizing the entry and search of premises; (i) empowering such authorities or persons as may be specified in the regulations to make orders and rules and to make or issue notices, licences, permits, certificates or other documents for the purposes of the regulations; (j) charging, in respect of the grant or issue of any licence, permit, certificate or other document for the purposes of the regulations, such fees as may be prescribed by the regulations; (k) the taking of possession or control on behalf of the Chief Executive of any property or undertaking; (l) requiring persons to do work or render services; (m) payment of compensation and remuneration to persons affected by the regulations and the determination of such compensation; and (n) the apprehension, trial and punishment of persons offending against the regulations or against any law in force in Hong Kong, and may contain such incidental and supplementary provisions as appear to the Chief Executive to be necessary or expedient for the purposes of the regulations. (3) Any regulations made under the provisions of this section shall continue in force until repealed by order of the Chief Executive in Council. (4) A regulation or any order or rule made in pursuance of such a regulation shall have effect notwithstanding anything inconsistent therewith contained in any enactment; and any provision of an enactment which may be inconsistent with any regulation or any such order or rule shall, whether that provision shall or shall not have been amended, suspended or modified in its operation under subsection (2), to the extent of such inconsistency have no effect so long as such regulation, order or rule shall remain in force. (5) Every document purporting to be an instrument made or issued by the Chief Executive or other authority or person in pursuance of this Ordinance or of any regulation made hereunder and to be signed by or on behalf of the Chief Executive or such other authority or person, shall be received in evidence, and shall, until the contrary is proved, be deemed to be an instrument made or issued by the Chief Executive or that authority or person. 3. Penalties (1) Without prejudice to the powers conferred by section 2, regulations made hereunder may provide for the punishment of any offence (whether such offence is a contravention of the regulations or an offence under any law applicable to Hong Kong) with such penalties and sanctions (including a maximum penalty of mandatory life imprisonment but excluding the penalty of death), and may contain such provisions in relation to forfeiture, disposal and retention of any article connected in any way with such offence and as to revocation or cancellation of any licence, permit, pass or authority issued under the regulations or under any other enactment as to the Chief Executive in Council may appear to be necessary or expedient to secure the enforcement of any regulation or law or to be otherwise in the public interest. (2) Any person who contravenes any regulation made under this Ordinance shall, where no other penalty or punishment is provided by such regulations, be liable on summary conviction to a fine of $5,000 and to imprisonment for 2 years. 4. Declaratory provision as to effect of an amending Ordinance For the purpose of removing doubts it is hereby declared that the words in subsection (1) of section 2 ‘he may make any regulations whatsoever which he may consider desirable in the public interest’ shall be deemed always to have included power to make such regulations as are mentioned in paragraph (g) of subsection (2) of section 2 and it is further declared that the provisions of subsection (4) of section 2 shall be deemed always to have been incorporated herein.” 21. The history of the enactment of the ERO and its use since its enactment in 1922 has been summarized at [15] ‑ [20] of the Judgment. We will elaborate on it when we deal with the submissions pertaining to the ERO below. 22. The PFCR is also a short piece of regulation with six sections. The substantive provisions relevant for the present purposes are as follows[13]. 23. Section 3 of the PFCR imposes a prohibition on the use of and makes it an offence to use facial covering in certain circumstances by providing: “(1) A person must not use any facial covering that is likely to prevent identification while the person is at— (a) an unlawful assembly (whether or not the assembly is a riot within the meaning of section 19 of Cap. 245[14]); (b) unauthorised assembly; (c) a public meeting that— (i) takes place under section 7(1) of Cap. 245; and (ii) does not fall within paragraph (a) or (b); or (d) a public procession that— (i) takes place under section 13(1) of Cap. 245; and (ii) does not fall within paragraph (a) or (b). (2) A person who contravenes subsection (1) commits an offence and is liable on conviction to a fine at level 4[15] and to imprisonment for 1 year.”[16] 24. Section 3 therefore prohibits the use of facial covering likely to prevent identification in four specific situations: (1) Under section 3(1)(a) – at an “unlawful assembly” (非法集結)[17]. (2) Under section 3(1)(b) – at an “unauthorised assembly” (未經批准集結)[18]. (3) Under section 3(1)(c) – at a public meeting (公眾集會) which has been notified to and not prohibited by the Commissioner of Police (“the Commissioner”) and which is not an unlawful or unauthorised assembly[19]. (4) Under section 3(1)(d) – at a public procession (公眾遊行) which has been notified and is not objected to by the Commissioner and which is not an unlawful or unauthorised assembly[20]. 25. In short, the prohibition in section 3(1) applies to persons at unlawful assemblies, unauthorised assemblies, public meetings notified and not prohibited, and public processions notified and not objected to, and does not prima facie apply to public meetings or processions that do not need to be notified, although such meetings or processions may turn into unauthorised assemblies or unlawful assemblies. 26. Section 4 of the PFCR sets out a defence to the offence under section 3(2) of lawful authority or reasonable excuse. The accused thus bears the evidential burden to raise the defence but it is the prosecution’s legal burden to disprove the defence. The scope of reasonable excuse is not exhaustively defined, but three grounds are specifically included, namely, professional or employment reasons, religious reasons and pre‑existing medical or health reasons. 27. Section 5 concerns police powers in relation to facial covering: “(1) This section applies in relation to a person in a public place who is using a facial covering that a police officer reasonably believes is likely to prevent identification. (2) The police officer may— (a) stop the person and require the person to remove the facial covering to enable the officer to verify the identity of the person; and (b) if the person fails to comply with a requirement under paragraph (a) – remove the facial covering. (3) A person who fails to comply with a requirement under subsection (2)(a) commits an offence and is liable on conviction to a fine at level 3[21] and to imprisonment for 6 months.” 28. “Public place” (公眾地方) has the meaning given by section 2(1) of the POO: “any place to which for the time being the public or any section of the public are entitled or permitted to have access, whether on payment or otherwise, and, in relation to any meeting, includes any place which is or will be, on the occasion and for the purposes of such meeting, a public place.” D. PROCEEDINGS BELOW D1. Grounds of judicial review 29. The Judges at [11] of the Judgment summarized the grounds of judicial review that were allowed to be advanced by KWH and LKH and considered by them at the rolled-up hearing. There is no suggestion that the summary is incorrect. We will respectfully adopt them for the purpose of this judgment: (1) Ground 1: the ERO is unconstitutional because it amounts to an impermissible grant or delegation of general legislative power by the legislature to the CEIC and contravenes the constitutional framework under the BL. (2) Ground 2: the ERO was impliedly repealed in 1991 by section 3(2) of the Hong Kong Bill of Rights Ordinance (Cap 383) (“the HKBORO”) either entirely or to the extent it is inconsistent with section 5 of that latter Ordinance; alternatively, it was impliedly repealed in 1997 by Article 4 of the ICCPR[22] (“ICCPR4”) as applied through BL39. (3) Ground 3: the ERO, to the extent that it empowers the CEIC to make regulations restricting fundamental rights protected by the BL and BOR, falls foul of the “prescribed by law” requirement in BL39. (4) Ground 4: by reason of the principle of legality, the general words in section 2(1) of the ERO are not to be read as allowing the CEIC to adopt measures that infringe fundamental rights of the individual in circumstances far removed from emergency situations. The PFCR is therefore ultra vires — beyond the power conferred on the CEIC by the ERO. (5) Ground 5A: section 3 of the PFCR amounts to a disproportionate restriction of a person’s liberty and privacy, freedom of expression and right of peaceful assembly under BOR5, 14, 15, 16, 17 and BL27. (6) Ground 5B: section 5 of the PFCR constitutes a disproportionate interference with the rights and freedoms protected by BL27, 28 and 31 and BOR5(1), 8, 14 and 16. 30. When considering these grounds, it is important to note that [23] : (1) The challenge of the constitutionality of the ERO is confined to the “public danger” ground in the ERO and to the powers it confers when the CEIC considers there to be an occasion of public danger (which is the ground on which the PFCR has been made). (2) All the applicants do not seek to impugn the good faith of the CEIC in invoking the ERO to enact the PFCR. (3) All the applicants do not seek to challenge the CEIC’s decision to enact the PFCR as Wednesbury unreasonable. D2. The Judgment 31. By their comprehensive and detailed Judgment, the Judges allowed the judicial reviews on Grounds 1, 5A and 5B, but rejected all the other grounds. We will look at their reasons in greater details later. For the present purposes, those reasons can be summarised as follows. 32. In allowing Ground 1, the Judges found essentially that: (1) On a proper construction of the BL as a whole, and in particular BL2, 8, 17(2), 18, 48, 56, 62(5), 66 and 73(1), the constitutional order established by the BL after 1997 vests only in the LegCo the general legislative power of the HKSAR. The BL has expressly provided the LegCo to be the legislature of the HKSAR. The CEIC is therefore not given any general legislative power under the BL even though she would have the power to make subordinate legislation. See [38] - [52] of the Judgment. (2) However, on a proper reading of the ERO, and given the widest possible scope and nature of “regulations” that the CEIC can enact under it when the CEIC is “satisfied” that there is any occasion of “public danger” (which by itself is an undefined but very wide term covering many plausible situations), the CEIC is in substance granted or delegated with a general legislative power by the ERO to enact what are in nature primary legislations. This is impermissible under the BL. See [53] - [81] of the Judgment. (3) The theme of continuity and all the relevant authorities decided before 1997 which concluded that the ERO was constitutional and valid do not assist the respondents to uphold the constitutionality of the ERO. This is so as the authorities were premised on the specific constitutional order of Hong Kong before the enactment of the BL and the view that the Hong Kong legislature at that time could be regarded as “supreme” in that respect. These reasons and bases are no longer valid after the enactment of the BL. Under the BL, the LegCo is no longer sovereign and supreme within its province and the constitutional order is that the LegCo is the legislature of Hong Kong. These authorities are therefore distinguishable and no longer applicable. For the same reasons, the ERO also did not become the law of Hong Kong after 1997 through BL8 because the BL does not permit the LegCo to grant to the CEIC “what is essentially the LegCo’s own constitutional power to enact, amend or repeal laws”. See [52] and [82] - [93] of the Judgment. (4) In the premises, the ERO, insofar as it empowers the CEIC to make regulations on any occasion of public danger, is incompatible with the BL. See [37] and [97] of the Judgment. (5) The court leaves open the question of the constitutionality of the ERO insofar as it relates to any occasion of emergency. See [96] of the Judgment. 33. In rejecting Ground 2, the Judges explained at [98] ‑ [109] of the Judgment that: (1) The applicants’ case in support of the implied repeal ground is premised on the argument that the ERO by its terms applies not only in emergencies that fall within the meaning of section 5 of the HKBORO. It therefore allows derogation from the fundamental rights protected by the BOR or ICCPR other than in the exceptional case of “public emergency” specifically defined and permitted under section 5 of the HKBORO or ICCPR4. The ERO is therefore incompatible with section 5 of the HKBORO and, by section 3(2) of the HKBORO, is repealed before 1997. Alternatively, it is inconsistent with ICCPR4 and thus implicitly repealed when the BL commenced operation on 1 July 1997 under BL39[24]. (2) Although accepting that the ERO by its terms is intended to apply not only in emergencies that fall within the meaning of section 5 of the HKBORO[25], the Judges find that the applicants’ argument of compatibility is flawed as it has conflated the concepts of derogation from the BOR itself and restriction of non-absolute rights under and in compliance with the BOR. (3) As rightly accepted by the respondents, the powers of the ERO may not be exercised with the effect of derogation from the BOR except in times of emergency officially proclaimed in accordance with section 5 of the HKBORO. In other occasion of emergencies, measures may be taken under the ERO which have an effect of restricting the rights protected by the BOR, provided the restriction is prescribed by law and compliant with the proportionality test. These are two different concepts. (4) Hence, the proper approach to the ERO is that: (a) Only in times of emergency officially proclaimed and in accordance with the other requirements of section 5 of the HKBORO, measures may be adopted under the ERO which derogate from the BOR. (b) In other situations, measures adopted under the ERO may not derogate from the BOR, and any measure that has the effect of restricting fundamental rights, it has to satisfy the dual requirements that the restriction is prescribed by law and meets the proportionality test. 34. In relation to Ground 3, the Judges rejected the argument that the ERO falls foul of the prescribed by law requirement due to the wide scope of power and the undefined meaning of “public danger”. They concluded that the ERO by itself does not and does not seek to limit any fundamental rights. In the premises, the requirement of legal certainty should be applied to the actual regulations that are enacted under the ERO but not generally to the enabling ERO. See [110] ‑ [120] of the Judgment. 35. Ground 4 seeks to impugn the PFCR itself. Under it, the applicants argued that, properly construed in accordance with the principle of legality, the ERO does not, expressly or by necessary implication, empower the CEIC to make regulations that impose restrictions on fundamental rights of the kind and to the extent found in the PFCR. The Judges found that this ground is inconsistent with Ground 1. Since they had allowed Ground 1, they therefore did not find it necessary to deal with this ground. See [121] ‑ [125] of the Judgment. 36. In allowing Ground 5A, in adopting the 4‑step proportionality test laid down in Hysan Development Co Ltd v Town Planning Board (2016) 19 HKCFAR 372 at [134] - [135][26]: (1) The Judges found that the restrictions imposed by section 3 of the PFCR engage the fundamental rights of the freedom of assembly[27], the freedom of speech or expression[28], and the right to privacy[29]. These rights are however not absolute and may be subject to lawful restrictions satisfying the proportionality test. See [127] - [128] of the Judgment. (2) They concluded that sections 3(1)(a) ‑ (d) of the PFCR satisfy the legitimate aims test[30] and the rational connection test. See: [130] - [146] of the Judgment. (3) Applying the reasonable necessity standard, the Judges also found that section 3(1)(a) (ie, in prohibiting the wearing of facial cover at an unlawful assembly) satisfies steps 3 and 4, in that the restriction is no more than reasonably necessary and strikes a reasonable balance in restricting the relevant fundamental rights and achieving the legitimate aims. They further explained that prohibiting the use of facial cover at unlawful assembly falls within the wide margin of discretion afforded to the Government to devise and implement measures to restrict unlawful and violent conduct. This would not result in an unacceptably harsh burden on the individual. They hold the view that no person should take part in an unlawful assembly in the first place, it being a criminal offence under section 18(1) of the POO. The Judges, therefore concluded that section 3(1)(a) meets the proportionality test and is lawful. See [149] ‑ [151] of the Judgment. (4) However, they did not find that the restrictions under sections 3(1)(b) ‑ (d) satisfy step 3 (and hence not necessary to deal with step 4). They came to that view after considering the following five distinguishing features relating to these provisions[31]: (a) Sections 3(1)(c) and (d) relate to public meetings and processions which may remain authorised and peaceful from beginning to end. The prohibition therefore directly interferes with these participants’ right of privacy or freedom of expression while taking part in perfectly lawful activities in the exercise of their right of peaceful assembly. (b) Although section 3(1)(b) relates to unauthorised assemblies, such assemblies may yet be entirely peaceful, without any violence being used or threatened by anyone participating in them. (c) The prohibition applies to any assembly or procession for whatever causes, but is not restricted to them arising from the Bill. Such assemblies or processions may be taking place for many and different causes, such as LGBT, labour or migrant rights, and traditionally, these gatherings have been orderly and peaceful. Participants in these assemblies or processions may have perfectly legitimate reasons for not wishing to be identified or seen to be supporting such causes. (d) The prohibition applies to any person while he or she is “at” any unauthorised assembly, public meeting or public processions. It is uncertain whether the restrictions imposed by these provisions cover not only participants of the types of gathering referred to in those provisions, but also any person who is physically present at the gathering in question. (e) The prohibition applies to facial covering of any type and used for whatever reasons, even if they are worn for religious, cultural, aesthetic or other legitimate reasons. It catches anyone wearing a facial covering which is “likely” to prevent identification. (5) In the premises, having regard to (a) the reach of the impugned restrictions to perfectly lawful and peaceful public gatherings, (b) the width of the restrictions affecting public gatherings for whatever causes, (c) the lack of clarity as regards the application of the restrictions to persons present at the public gathering other than as participants, (d) the breath of the prohibition against the use of facial covering of any type and worn for whatever reasons, (e) the absence of any mechanism for a case-by-case evaluation or assessment of the risk of violence or crimes such as would justify the application of restrictions, (f) the lack of robust evidence on the effectiveness of the measures, and (g) the importance that the law attaches to the freedom of expression, freedom of assembly, procession and demonstration, and the right to privacy, the Judges did not find the restrictions imposed under sections 3(1)(b) - (d) to be proportionate to the legitimate aims sought to be achieved by imposing those restrictions[32]. See [166] - [167] of the Judgment. 37. In allowing Ground 5B: (1) The Judges found that section 5 engages the non-absolute right of the freedom of the person or right to liberty protected by BL28 and BOR5, and that the measures adopted under it are rationally connected to the legitimate aims to assist in law enforcement, investigation and prosecution by enabling police officers to verify the identity of all masked individuals not during assemblies or processions but also in public places in the prevailing circumstances of public danger in Hong Kong. See [169] ‑ [184] of the Judgment. (2) They however considered that measures provided under section 5 are of a remarkable width: (a) There is practically no limit on the circumstances in which the power under that section can be exercised by a police officer, save that (i) the person is in a public place, and (ii) the facial covering used is reasonably believed by the police officer to be likely to prevent identification; (b) The power can be exercised irrespective of whether there is any public meeting or procession taking place in the vicinity and regardless of whether there is any risk of outbreak of violence or other criminal acts, at the place where the person is found, or in the neighbourhood or anywhere else in Hong Kong; (c) The power may on the face of it be used by a police officer for random stoppage of anyone found wearing a facial covering in any public place. See [185] ‑ [188] of the Judgment. (3) Given this remarkable width, the Judges were of the view that the measures adopted under section 5 exceed what is reasonably necessary to achieve the legitimate aims even in the prevailing turbulent circumstances in Hong Kong, and that they fail to strike a reasonable balance between societal benefits promoted and the inroads made into the protected rights. See [189] ‑ [190] of the Judgment. 38. For all the above reasons, after hearing further submissions on the relief that should be made, the Judges by the Decision, allowed the judicial reviews and declared, among others, that: (1) The ERO, insofar as it empowers the CEIC to make regulations on any occasion of public danger, is incompatible with the BL, and consequently, the PFCR made pursuant to it on an occasion of public danger is also invalid and of no effect. (2) Sections 3(1)(b), (c) and (d) of the PFCR is inconsistent with BL27 and BOR14, 16 and 17, and is therefore null, void and of no effect. (3) Section 5 of the PFCR is inconsistent with BL 28 and BOR5 and is therefore null, void and of no effect. 39. They further ordered that the above declarations be suspended so as to postpone its coming into operation until the end of 29 November 2019 or further order of the court. Given there is no further order later, the declarations eventually took effect on 30 November 2019. 40. The main appeals are brought by the respondents challenging the Judges’ rulings in relation to Grounds 1, 5A and 5B. It is therefore convenient for us to consider these grounds first. E. GROUND 1 E1. Jurisdiction to examine constitutionality 41. Ground 1 involves the court’s scrutiny of the constitutionality of the ERO, a pre‑1 July 1997 law adopted as laws of the HKSAR. The fact that the court has jurisdiction to do so is and should not be controversial. However, given its immense importance, we consider it appropriate to elaborate the point against the relevant framework of the BL as follows. 42. Under BL8, the laws previously in force in Hong Kong, including ordinances, shall be maintained except for any that contravenes the BL, and subject to any amendment by the legislature of the HKSAR. BL18 further provides that the laws in force in the HKSAR shall be the BL, and laws previously in force in Hong Kong as provided for in BL8 and, the laws enacted by the legislature of the Region. BL8 and BL18 are supplemented by BL160(1), which provides that: “Upon the establishment of the [HKSAR], the laws previously in force in Hong Kong shall be adopted as laws of the Region except for those which the Standing Committee of the National People’s Congress (‘the NPCSC’) declares to be in contravention of this Law. If any laws are later discovered to be in contravention of this Law, they shall be amended or cease to have force in accordance with the procedure as prescribed by this Law.” 43. In its Decision dated 23 February 1997 (“the NPCSC’s Decision”), the NPCSC pursuant to BL160 declared that those laws previously in force in Hong Kong listed in Appendixes I and II were in contravention of the BL and were not adopted as the laws of the HKSAR. Other than those laws, all laws previously in force were adopted as laws of the HKSAR. Paragraph 6 of the NPCSC’s Decision repeats the last sentence in BL160(1) with respect to the laws previously in force adopted as laws of Hong Kong. 44. Both BL160 and the NPCSC’s Decision ensure the preeminence of the BL over all pre‑1 July 1997 laws, even after being adopted as laws of the HKSAR upon its establishment. 45. The ERO was enacted in February 1922. It was an ordinance previously in force in Hong Kong prior to 1 July 1997. As it was not included in either Appendix I or II to the Decision, it has been adopted as laws of the HKSAR. It is however not the respondents’ case that the NPCSC’s Decision is final or conclusive on the constitutionality of the ERO. The respondents accept that the constitutionality of the ERO is still subject to judicial scrutiny. The respondents’ position must be correct. For the law on this point is well settled. 46. In HKSAR v Hung Chan Wa & Another (2006) 9 HKCFAR 614, the Court of Final Appeal refused to make a “prospective overruling” in respect of a certain provision which was previously in force but was, after judicial scrutiny on its constitutionality, remedially interpreted to make it constitutionally compliant. Li CJ at [9] said: “[BL 160(1)] supplements articles such as [BL8 and BL18] in making it clear that laws previously in force shall be adopted except for those which the [NPCSC] declares to be in contravention of the Basic Law. Apart from the laws so declared to be in contravention, [BL160] recognizes that there may be laws which are discovered after 1 July 1997 to be in contravention. In relation to them, [BL160(1)] provides that ‘they shall be amended or cease to have force in accordance with the procedure as prescribed by this Law.’…” 47. Li CJ then went on to hold at [11] - [15] that BL160 does not apply to judicial procedure and that the reference of “shall be amended or cease to have force” in BL160(1) connotes a legislative procedure. At [11], Li CJ specifically referred to a judicial declaration of contravention of the BL in relation to a pre‑1 July 1997 law. Evidently, the judicial power to make such a determination derives from BL19(1) and BL80: see Ng Ka Ling & Others v Director of Immigration (1999) 2 HKCFAR 4, per Li CJ at p25G-J. 48. Another ready example of the court scrutinizing the constitutionality of a pre‑1 July 1997 law is W v Registrar of Marriage (2013) 16 HKCFAR 112. There, the Court of Final Appeal, by majority, held that sections 20(1)(d) and 40 of the Marriage Ordinance, underpinned by the common law criteria in Corbett v Corbett (otherwise Ashley) [1971] P 83, for determining who was a “woman” for the purpose of marriage, were unconstitutional because, having regard to the contemporaneous jurisprudence, they failed to give proper effect to the constitutional right to marry under BL37 and article 19(2) of the BOR and impaired the very essence of the right to marry. 49. As BL160(1) and the NPCSC’s Decision both envisage, there may be laws which were previously in force and adopted as laws of the HKSAR but are later found to be in contravention of the BL. Under the framework of the BL, the constitutionality of such a law can be dealt with either by legislative procedure or judicial process when a challenge is mounted in court whereupon the court has the jurisdiction and indeed the constitutional duty to deal with the constitutionality issue and, if found to be the case, to declare such a law to be in contravention of the BL. 50. In so doing, the court does not act contrary to BL160 or in any way diminish the authority of the NPCSC’s Decision. As said, both BL160 and the NPCSC’s Decision maintain the preeminence of the BL over all pre‑1 July 1997 laws, even after adoption as laws of the HKSAR upon its establishment. When law develops, as it must, to cater for the changing needs of the society, it is not inconceivable that a pre‑1 July 1997 law which was formerly constitutional compliant might be found in contravention of the BL later when examined against the context of the current societal circumstances and by reference to the contemporaneous jurisprudence. If and when it happens, the court will declare that law, which is no longer constitutional compliant, to be in contravention of the BL, thereby safeguarding the preeminence of the BL as a living constitutional instrument over all pre‑1 July 1997 laws even after adoption as laws of the HKSAR. This is entirely consistent with the purpose of BL160 and the NPCSC’s Decision. E2. Core issue and the parties’ main submissions 51. For Ground 1, the core issue is whether the emergency regulations made under the ERO are subordinate legislation or primary legislation. If the former, it is acceptable. If the latter, it contravenes the BL, rendering the ERO an unconstitutional delegation of general legislative powers by the LegCo to the CEIC. 52. Mr Yu SC, together with Mr Suen SC, Mr Ma and Mr Lui, for the respondents, complained that in reaching the ruling that the ERO is unconstitutional, the Judges disregarded the theme of continuity which remains despite the change in constitutional order. He submitted that central to the Judges’ reasoning is the erroneous finding that because the constitutional order in Hong Kong has changed, the functions and powers of the LegCo and the Executive have changed so fundamentally under the BL framework that the ERO should be held to be inconsistent with the BL. 53. Further and irrespective of the theme of continuity, Mr Yu argued that the Judges erred in its construction and application of the constitutional order under the BL including BL56(2) which recognizes and reinforces the power of the LegCo to delegate (a fortiori in emergency regulations). He complained that the Judges failed to appreciate the true nature of emergency regulations and the ground of pubic danger, and to examine the same in its analysis and erroneously treated the ERO merely as a piece of legislation authorizing the making of subordinate legislation, having no regard to the true nature of emergency regulations under the ERO. He contended that the Judges erred in ruling that emergency regulations made by the CEIC on the public danger ground are not subordinate legislation referred to in BL56(2). 54. Mr Yu submitted that the real question raised by Ground 1 is whether the ERO contravenes the BL. He argued that there is no such contravention and the Judges erroneously adopted a narrow or rigid approach in construing the BL, and therefore failed to appreciate that the delegation of powers to the CEIC to make subordinate legislation on occasion of public danger is reasonably required and implied under the BL, quite apart from and irrespective of express provisions such as BL56(2) and 62(5). 55. Ms Li SC, with Mr Chan SC, Mr Deng, Mr Tam, Mr Yeung and Ms Wong, for KWH, accepted that the CE has the constitutional power to take emergency measures which can be provided for by suitable legislation catering for the limited circumstances in which derogation from the HKBORO/ICCPR rights are permitted except in the case of non‑derogable rights and/or other circumstances of emergency where no such derogation is permitted and the HKBORO/ICCPR rights as applied to the HKSAR through BL39 must not be infringed. 56. On the theme of continuity, Ms Li emphasized that the Judges did not hold the concept of emergency regulations as no longer constitutional after 1997. The problem is the impermissible and general terms on which this conferment of legislative power was done in the ERO. She argued that the theme of continuity does not operate as a presumption of constitutionality for any particular piece of pre-1997 legislation. Continuity depends on subject matter. The overriding principle is that nothing shall contravene the BL and the principle of continuity must be subject to this overriding principle. There exists core powers and functions that are exclusively reserved to the LegCo as prescribed by the BL which may not be conferred to the CEIC in the manner as the ERO has. She also argued that BL160 does not assist the respondents. 57. Next, Ms Li argued that vague notions of emergency cannot impugn the constitutional scheme. The ERO is exceptionally wide even as an emergency power. The legislative power conferred by the ERO is on the whole general legislative power in substance for the reasons given by the Judges. E3. Theme of continuity 58. In our view, the analysis of the constitutionality of the ERO is informed by the theme of continuity. 59. The theme of continuity is an essential and indeed indispensable policy underlying the BL. In HKSAR v Ma Wai Kwan David & Others [1997] HKLRD 761, Chan CJHC (as he then was) at p774E-F emphasized its utmost importance in these terms: “In my view, the intention of the Basic Law is clear. There is to be no change in our laws and legal system (except those which contravene the Basic Law). These are the very fabric of our society. Continuity is the key to stability. Any disruption will be disastrous. Even one moment of legal vacuum may lead to chaos. Everything relating to the laws and the legal system, except those provisions which contravene the Basic Law, has to continue to be in force. The existing system must already be in place on 1 July 1997. That must be the intention of the Basic Law.” 60. Continuity of course does not mean stagnation of the systems then in place on 1 July 1997: see Chan Yu Nam v Secretary for Justice [2012] 3 HKC 38, per Stock VP (as he then was) at [44]. The systems can further develop to suit the contemporaneous needs and circumstances of our society as Hong Kong progresses as long as they operate within the confines of the BL. 61. That said, Ms Li is correct in submitting that the theme of continuity is not a presumption of interpretation in favour of constitutional compliance when a pre‑1 July 1997 law is challenged. That is not Mr Yu’s argument any way. The significance of the theme for interpreting the BL in this regard is this. 62. As Lord Hoffmann observed in Matadeen v Minister of Education and Science [1999] 1 AC 98 at p108, cited with approval by Stock VP in Chan Yu Nam, at [32]: “… The background of a constitution is an attempt, at a particular moment in history, to lay down an enduring scheme of government in accordance with certain moral and political values. Interpretation must take these purposes into account. ...” Thus, as one of the main policies underlying the BL, the theme of continuity may, depending on the circumstances, relevantly and powerfully inform the interpretation of the provisions of the BL in a challenge against the constitutionality a pre‑1 July 1997 law which has been adopted as part of the laws of the HKSAR after 1 July 1997, such as the ERO. E3.1 History of the ERO in the colonial era 63. In the colonial years, the ERO had been invoked on numerous occasions to cover a host of varying emergencies and public danger. There was a debate in the colonial Legislative Council (“the Pre-97 LegCo”) in 1936 to abrogate one of the emergency regulations made under the ERO. It later survived two challenges before the Full Court in the 1950s. (a) Invocations 64. In The Use and Abuse of Emergency Powers by the Hong Kong Government (1996) HKLJ 47, Norman Miners at pp51 - 55 briefly described the ERO’s legislative history and its invocations since enactment. According to this article, the ERO was not modelled on the then English Emergency Powers Act 1920. Unlike other laws enacted by the Pre-97 LegCo, the emergency regulations made under the ERO were not required to be submitted to London, which meant that they were not subject to the check or censure from London. 65. Over the years, the then Governor in Council (“the GIC”) had made various emergency regulations under the ERO for, inter alia: (1) the general strike and boycott of the colony nearly ruining its economy between June 1925 and October 1926; (2) the fear of subversion by prescribed organizations in 1927; (3) the severe drought in 1929; (4) the prevention and mitigation of cholera in 1932; (5) the outbreak of World War II in 1939; (6) the communist advance to the border in 1949; (7) the banking crisis in 1965; (8) the outbreak of the cultural revolution and riots in 1967; and (9) the oil crisis in 1974. 66. In June 1995, all emergency regulations which were still in force were revoked by the GIC. (b) A failed attempt to abrogate the regulations made under the ERO 67. On 26 August 1936, the Pre‑1997 LegCo debated a motion moved by a member seeking to abrogate an emergency regulation made under the ERO to provide censorship of the Chinese press[33]. The regulation had been made for some 11 years and had not been repealed by the GIC. One of the arguments advanced in support of the motion was that there could not be an existence of emergency or public danger as envisaged under the ERO for so many years. The Government’s position was that the public danger still existed and would continue until a definitely stable government existed in China and in particular, the danger must be admitted to remain while civil war in China was threatening in one of the neighbouring provinces. In the end, the motion was defeated by a majority of 14 to 2. (c) Two challenges in the court 68. The ERO and the emergency regulations made thereunder were challenged in the court in R v To Lam Sin (1952) 36 HKLR 1 and R v Li Bun & Others [1957] HKLR 89. 69. In To Lam Sin, the accused was convicted of being in possession of hand grenades, contrary to regulation 116A of the Emergency (Principal) Regulations made under the ERO. The offence was punishable by death. One of the grounds that the accused sought to quash the conviction was that the ERO was ultra vires the Pre‑97 LegCo. While counsel for the accused conceded that the Pre‑97 LegCo had the power to delegate, he contended that in the case of the ERO, the delegation to the Governor went far beyond the ordinary power of making by‑laws. Counsel argued that it had by the delegation effaced itself, citing Ping Shek and another v The Canossian Institute (1949) 33 HKLR 66 where the Chief Justice at p71 referred to the effacement test. 70. The Full Court (Howe CJ, Gould and Scholes JJ, Howe CJ giving the judgment) apparently did not prefer the effacement test, noting at p12 its limitations: “The word ‘effaced’ may have different shades of meaning. If in the context it is meant to indicate a complete and final parting with legislative authority, we agree that that would be ultra vires, as it would involve an alteration of the Colony’s constitution. That is clearly beyond the powers conferred by the Letters Patent and is clear law. Again, if ‘partial effacement’ means a final parting with legislative authority in some limited sphere the same reasoning applies. But beyond that, the word appears to be of little assistance.” 71. Howe CJ then went on to state that according to the modern judicial view of the colonial legislature as seen from some recent authorities, a colonial legislature was supreme within their own limits and within the powers conferred by the instruments creating it. Based on that modern view, Howe CJ at p14 said: “… the power of a colonial legislature to delegate is a full one, limited only by the necessity not to go outside the powers conferred by or contravene the rights reserved by the Letters Patent or other constitutional document. As is well known, delegation of powers almost parallel with those given by the [ERO] has been resorted to frequently in England under the various Emergency Powers Acts. If the legislature of Hong Kong is supreme (subject to its constitution) in its own area there can be no reason why it should not act similarly – it is not and cannot be suggested that the law is not one for the ‘peace, order and good government’ of the Colony.” 72. Returning to the effacement test, Howe CJ at p14 held that it would not hold that the delegation of the powers was ultra vires because: “Wide though the powers may be, the Legislative Council retains a very firm and close control by virtue of Section 14 of the Interpretation Ordinance (Cap.1). No regulation involving the imposition of the death penalty can become of force or effect without the prior approval of the Legislative Council – this is provided specifically by the [ERO] as well. All other regulations must be laid on the table at the first meeting of the Legislative Council after their publication in the Gazette and the Council may repeal or amend any of them. There is in addition the overriding power to repeal or amend the Ordinance itself. We see nothing there which can be called effacement as we understand it.” 73. In Li Bun, an appeal by way of case stated was brought against convictions for attempting to export motor vehicles without a licence contrary to the Importation and Exportation Ordinance, as amended by the Emergency (Importation and Exportation) (Amendment) Regulations 1953 and 1954 made under the ERO. Two questions of law were raised. The first was whether the ERO was ultra vires the Governor. The second was whether the Emergency (Importation and Exportation) (Amendment) Regulations 1953 and 1954 were ultra vires the ERO. 74. The Full Court (Hogan CJ, Gould and Reece JJ) treated the first question as involving the decision if the ERO was ultra vires the Pre‑97 LegCo. Hogan CJ acknowledged the modern view of a colonial legislature as mentioned in To Lam Sin but at p96 observed that the supremacy of a colonial legislature, even within its own limits, could not be taken to be altogether unqualified. 75. Hogan CJ referred to the effacement test and noted the ERO survived the test in To Lam Sin. He agreed with To Lam Sin but would not place the same reliance on section 14 of the Interpretation Ordinance, as ensuring the control of the legislature over regulations made under the ERO since that section itself was liable to be repealed by the GIC under section 2(3) of the ERO. 76. Hogan CJ next derived from R v Burah 3 AC 889 and In re The Initiative and Referendum Act (1919) AC 935 the proposition that a colonial legislature could not appoint a coordinate or alternative legislative body. Applying it to the ERO, he said at pp101 - 104: “That it may be desirable for the sake of ‘peace, order and good government’ to have, on occasions of emergency or public danger, a delegated power to legislate speedily and effectively in order to meet any and every kind of problem is, I think, obvious. That such power should, as the Attorney General has argued, extend to all existing legislation seems equally apparent, since otherwise its capacity to make adequate provision for some unexpected danger or emergency might be hampered or limited by its inability to alter an existing Ordinance and that, possibly, at a time when the ordinary legislature could not, as a result of the emergency of state of public danger, be brought into session or meet. But however desirable these powers can they be conferred in this way by the Legislature of Hong Kong or should they have been sought by an amendment of the Letters Patent or a similar alteration in the constitution? … … the fact that the powers conferred by the [ERO] are limited to those occasions which in the opinion of the Governor in Council are occasions of public danger or emergency. This, it seems to me, is the one factor or at any rate the principal factor which prevents them from being regarded as that arming, by a dependent Legislature, of another authority with general legislative authority similar in capacity to its own, on which the Privy Council has frowned. It may be argued that to hold so ample a power as falling below the line which cannot be crossed by the Legislative Council, is to push that line so high as to make it almost meaningless; nevertheless it is a real limitation and since it is the principal reason for not treating the ordinance as ultra vires ab initio, a limitation which must, I think, be strictly observed and strictly enforced, since any tendency to regard it as a mere formality would tend to diminish the importance of the principal, if not indeed the only, factor which saves this ordinance from being ultra vires. The decision as to whether an emergency exists is conferred on the Governor in Council and it is of course contemplated that any such decision will be reached bond fide. But it is a decision which must be taken on each occasion that these powers are exercised and to that extent the power thus conferred may be more closely linked with the occasion that gave rise to it than those expressed to be dependent, like so many ‘emergency’ powers, on some formal proclamation and to continue so long as such proclamation remains in force. Indeed, in pursuance of the principle enunciated by Romer J in Land Realisation Co v Postmaster General (1950) Ch 439 when he said:- ‘The legislature in conferring powers … is conferring them on a person who will presumably use those powers bona fide … in furtherance of the objects for the attainment of which the powers were conferred.’ such a power might well be thought to be more closely confined to the purpose of abating the emergency which gave rise to its use than would those powers which are expressed to be exercisable so long as a particular proclamation remains operative.” Hogan CJ therefore agreed that the Full Court in To Lam Sin was right in concluding that the ERO was not ultra vires. 77. Hogan CJ then examined the second question if the Emergency (Importation & Exportation) (Amendment) Regulations were ultra vires the ERO. For the reasons that he set out, he held that they were not. E3.2 Important points to draw from the history 78. From the above history a number of important points arise. They concern (a) the then constitutional arrangements for legislative powers; (b) the Pre-97 LegCo’s role over the emergency regulations made under the ERO; (c) the judicial reasoning in upholding the validity of the ERO; and (d) the varying circumstances of emergency and public danger in which the ERO was invoked with one common feature. They are all highly relevant to the analysis at hand. (a) Constitutional arrangements for general legislative powers 79. We derive from To Lam Sin and Li Bun the following propositions concerning the constitutional arrangements for the exercise of general legislative powers in the colonial era. 80. First, the Pre-97 LegCo was vested with general legislative powers exclusively. The Governor did not have such powers. 81. Article VII of the Letters Patent appeared to have designated the Governor as the law making authority by providing that: “The Governor, by and with the advice and consent of the Legislative Council, may make laws for the peace, order, and good governance of the Colony.” However, the correct understanding of Article VII was that it conferred the general legislative powers on the Pre-97 LegCo. Thus in Li Bun, Hogan CJ at p92 observed that the powers of the Pre-97 LegCo to make laws have their source in Article VII. (Indeed, had the Governor been vested with general legislative powers under Article VII, the question if the ERO was an impermissible delegation of legislative powers by the LegCo to the GIC would not have arisen at all.) Simply put, pursuant to the constitutional arrangement in the Letters Patent on legislative powers, however wide and extensive the powers the Governor otherwise enjoyed under the Letters Patent, and despite the fact the colonial government was very much executive-led, he did not have general legislative powers as the Pre-97 LegCo did. The Governor only had limited legislative power to make subordinate legislation as delegated by the Pre-97 LegCo: see the third proposition below. 82. Second, the Pre-97 LegCo as a colonial legislature was subject to the Letters Patent and other constitutional requirements which established it as the legislature of Hong Kong. Within its constitutional confines, the Pre-97 LegCo was autonomous or “supreme” in exercising is powers and performing its functions and had the authority as plenary and as ample as the Imperial Parliament in the plenitude of its powers possessed and could bestow. 83. Third, under the common law, the Pre-97 LegCo must not act in a manner which effectively abdicated its exclusive general legislative powers, whether completely or partially, by effacement; or divesting itself of such exclusive powers, again whether completely or partially, by creating a co-ordinate legislative authority of a concurrent or alternative character. Subject to this restriction, it could delegate its legislative powers to the Governor to make subordinate legislation. The corollary is that the Governor only had limited legislative powers as properly delegated to him by the Pre-97 LegCo to make subordinate legislation. 84. We pause to add that as a matter of fact, the Pre-97 LegCo had by numerous enabling provisions in various principal ordinances delegated to the GIC and other executive authorities to make regulations in respect of a wide range of matters[34] and even to the Chief Justice to make rules for practice and procedure of court and tribunal[35]. 85. Fourth, an ordinance by which the Pre-97 LegCo delegated to the GIC and other branches of the Government the power to make regulations was subject to judicial scrutiny to see if it was ultra vires the Pre-97 LegCo. The regulations made, like those made under the ERO, were likewise subject to judicial scrutiny to see if it was ultra vires the enabling ordinance. 86. In sum, under the pre-1 July 1997 constitutional arrangement, the Pre-97 LegCo was vested with general legislative powers exclusively by the Letters Patent. The Governor was not. The Pre-97 LegCo was bound by the Letters Patent and other constitutional requirements in exercising its general legislative powers. Under the common law, subject to the restriction that it could not abdicate its general legislative powers, the Pre‑97 LegCo could delegate to the Governor limited legislative power to make subordinate legislation by virtue of a specific enabling ordinance. That was the extent of the Governor’s legislative power. When called upon, the courts would examine the enabling ordinance and the regulations made thereunder to see if they were ultra vires the Pre-97 LegCo or the enabling ordinance, as the case might be. (b) Pre-97 LegCo’s control over emergency regulations 87. Although the attempt to abrogate the emergency regulations in 1936 failed, the whole process of involving a member putting forward of the motion to abrogate the emergency regulations concerned, the ensuing debate and vote by the Pre-97 LegCo as a whole against the motion, clearly showed that despite section 2(3) of the ERO, the Pre-97 LegCo still considered that it had control over the emergency regulations made thereunder and could, if necessary, abrogate it. (c) Judicial reasoning 88. The reasoning of the Full Court in To Lam Sin and Li Bun in holding that the ERO was intra vires the Pre-97 LegCo merits closer attention. Relevantly, the Full Court held: (1) Delegating by the Pre-97 LegCo to the GIC the power to legislate speedily and effectively in cases of emergency and public danger was obviously desirable for “the peace, order and good government of the Colony”, thereby falling within the mandate given to the Pre-97 LegCo under Article VII of the Letters Patent. (2) The GIC’s delegated power to legislate to cater for emergency and public danger should extend to all existing legislation so as not to hamper his capacity to make adequate provision for some unexpected danger or emergency or limit his inability to alter an existing ordinance, possibly, at a time when the Pre‑97 LegCo could not, as a result of the emergency of state of public danger, be brought into session or meet. (3) However, the ERO could only be invoked in the limited circumstances where the GIC considered to be occasions of emergency or public danger. The ERO was not a blanket delegation by the Pre-97 LegCo to the GIC to make any laws under any circumstances. That limitation was not a formality but must be strictly observed and strictly enforced. (4) The GIC must act bona fide in invoking the ERO. He must exercise the powers conferred on him under the ERO for the furtherance of its objects and purposes. (d) Emergency and public danger varied but with one common feature 89. There is no definition of emergency or public danger in the ERO. Based on the past invocations, what constituted, in the opinion of the GIC, emergency or public danger within the meaning of the ERO, warranting the making of the corresponding emergency regulations covered a host of varying circumstances. They ranged from a state of war, which threatened the very existence of Hong Kong; to widespread and serious breaches of public order and security, which threatened the law and order of the entire community; to outbreak of epidemic disease and natural disaster, which threatened life and health of the public; and to financial and other crisis fundamentally rocking Hong Kong’s economy and jeopardizing its citizens’ livelihood. They reflected a broad concept of emergency and public danger with the commonality of serious and immediate threats to Hong Kong and its citizens as a whole subsisting for a period of time. Correspondingly, the provisions of the ERO were wide and flexible enough for the GIC to legislate speedily and effectively to meet all and every kind of emergency and public danger that he had to specifically grapple with at the time. E3.3 After 1 July 1997 90. We first look at the constitutional arrangement for general legislative powers under the BL. (a) Constitutional arrangement for general legislative powers 91. Under the BL, the constitutional arrangement for general legislative powers on the whole mirrors the past subject to the necessary modifications to reflect the change in the constitutional order. 92. First, under the design of the BL, the Government is very much an executive-led government (BL59 - 65): see the observation of Sir Anthony Mason in The Place of Comparative Law in Developing the Jurisprudence on the Rule of Law and Human Rights in Hong Kong (2007) 37 HKLJ 299, at p304. The CE is constituted the head of the HKSAR (BL43) and enjoys wide and extensive powers: see in particular BL48. However, the CE simply does not have general legislative powers. 93. Second, BL56(2) acknowledges that CE may, in consultation with the Executive Council introduce bills and make subordinate legislation. However, as Mr Yu has fairly accepted, BL56(2) is not an enabling provision conferring on the CE the legislative power to make subordinate legislation. The source of such power can only come from the delegation by the LegCo to the CE. 94. Third, BL66 establishes the LegCo as the legislature of the Region to exercise the general legislative powers vested in the HKSAR under BL17(1). BL73 further mandates the LegCo to exercise the powers and functions as prescribed, including, under sub-paragraph (1), to enact, amend, repeal laws in accordance with the provisions of the BL and legal procedures. Importantly, BL11(2) provides that no law enacted by the LegCo shall contravene the BL. 95. Further, as held by this Court in Chief Executive of HKSAR v President of the Legislative Council [2017] 1 HKLRD 460, per Cheung CJHC (as he then was) at [24] - [25] and per Poon JA (as he then was) at [87], the LegCo is subject to the BL and all other constitutional requirements. Within those confines, the LegCo is otherwise a sovereign body under the BL: Leung Kwok Hung v President of Legislative Council [2007] 1 HKLRD 387, per Hartmann J (as he then was) at [10]; Cheung Kar Shun v Li Fung Ying [2011] 2 HKLRD 555, per A Cheung J (as he then was) at [217]. 96. Fourth, the BL does not create a new legislature out of a vacuum: Cheng Kar Shun v Li Fung Ying, supra, per A Cheung J at [121]. In our view, subject to the BL and other constitutional requirements, the common law principles governing the Pre-97 LegCo continue to apply under the theme of continuity. Thus, the principle of delegation in the colonial era under which the Pre-97 LegCo could delegate to the executive branch of government or any person or body to make subordinate legislation continues to apply after 1997, as suitably modified to reflect the change in the constitutional order. Insofar as the CE is concerned, the power to make subordinate legislation pursuant to such delegation is expressly acknowledged in BL56(2). 97. As a matter of fact, the various ordinances by which the Pre-97 LegCo authorized the executive branch of the Government and the Chief Justice to make regulations referred to above remain in force. After 1 July 1997, the LegCo has enacted further ordinances to empower the executive branch and the Chief Justice to make regulations[36]. It demonstrates that the common law principle of delegation continues to thrive in the post-1 July 1997 era. 98. Fifth, it is settled law since the establishment of the HKSAR that our courts, in exercising their independent judicial power vested with them by BL19(1) and BL80, have a constitutional duty to enforce and interpret the BL and to examine whether legislation enacted by the legislature and executive acts are consistent with the BL and, if found to be inconsistent, to hold them to be invalid: Ng Ka Ling & Others v Director of Immigration, supra, per Li CJ at p25G-J. 99. In sum, the constitutional arrangement under the BL for the exercise of legislative power broadly resembles the past before 1 July 1997. The LegCo is the only institution vested with general legislative powers exclusively. Neither the CE nor, for that matter, other branches of the Government has the general power to legislate. In exercising its functions and powers, the LegCo is bound by the BL and other constitutional requirements. Within the constitutional confines and under the common law principle of separation of powers, the LegCo is autonomous. Like the Pre-97 LegCo, the LegCo can under the common law delegate its legislative powers to the CE and other branches of the Government to make subordinate legislation by enacting specific enabling ordinances. Correspondingly, that is the limited extent to which the CE and other branches of the Government can legislate. When a challenge is made, the court can examine the constitutionality of the enabling ordinance, and the constitutionality and vires of the subordinate legislation made thereunder, as the case may be. 100. It follows that under the constitutional framework of the BL, the LegCo can delegate to the CEIC legislative powers by way of an enabling ordinance to make emergency regulations provided that they are subordinate legislation. As we understand the parties’ submissions, they do not differ on this point. 101. This brings us to the ERO. At this juncture, we look at it in the overall context of continuity. (b) The ERO as an integral option for emergency measures 102. The BL contemplates a wide range of options available to the NPCSC and the Government, as the case may be, allowing different and specific measures to be made to address the varying exigencies of emergency or public danger generally. 103. The first option is vested with the NPCSC alone. In the most extreme case of a state of war as declared by the NPCSC or turmoil endangering national unity or security beyond the control of the Government, BL18 enables the Central People’s Government (the “CPG”) to issue an order applying the relevant national laws in Hong Kong. Other than this dire scenario which necessitates the NPCSC’s action, how to tackle emergency and public danger is basically left with the Government. 104. It is worth noting that where there is an official proclamation of public emergency under section 5 of the HKBORO, measures may be taken derogating from the BOR to the extent strictly required by the exigencies of the situation save and except non-derogable rights: section 5(2)(c)[37]. Those measures must be taken in accordance with law: section 5(1). 105. The second option involves the CPG and garrison upon the request of the Government. Under BL14(3), for maintenance of public order or disaster relief, the Government may ask the CPG for assistance from the garrison. 106. Plainly, other occasions of maintenance of public order or disaster relief where the Government does not need to ask the CPG for assistance from the garrison under BL14(3), and many other cases of emergency and public danger, not limited to maintenance of public order of disaster relief, may happen. To cope with those situations, the BL provides the CE with three further options. 107. The CE may adopt executive and administrative measures in emergencies under BL56(2)[38]. The power to do so, it would appear, comes from the general executive powers of the Government[39], of which the CE is the head[40]. 108. If emergency legislative measures are required, the CE may make a request to the President of the LegCo under BL72(5) to convene an emergency session of the legislature to pass the necessary laws. 109. There is another option which the CE can resort to if emergency legislative measures are required, that is, invoking the ERO to make emergency regulations. As explained above, BL56(2) recognizes that the CE can make subordinate legislation under an enabling ordinance pursuant to the common law principle of delegation by the LegCo, including emergency regulations. In respect of emergency and public danger, as at 1 July 1997, the only ordinance, upheld twice by the court, which delegated to the CEIC the power to make emergency regulations was the ERO. So both the ERO and the CEIC’s power to invoke it to make subordinate legislation to deal with emergency and public danger were clearly within the contemplation of the drafters of the BL. They must have regarded the ERO to be compatible with the BL so that it would remain in our statute book after 1 July 1997. 110. Based on the past invocations, the ERO is powerful and versatile enough to enable the CEIC to legislate speedily and effectively to meet all and every kind of emergency and public danger that imposes serious and subsisting threats to Hong Kong and its citizens[41]. 111. In sum, under the theme of continuity, the BL contemplates that the ERO under which the CE has the power to make emergency regulations is one of the integral options for tackling emergency and public danger. The theme of continuity strongly suggests that the ERO is constitutionally compliant and should remain as part of the laws of Hong Kong. If the ERO were held to be unconstitutional, it would leave a significant lacuna in the law for dealing with emergency and public danger generally. The CE would be deprived of the power to respond swiftly, flexibly and sufficiently by making the necessary emergency regulations even though the circumstances clearly warrant it and it is in the public interest to do so. E3.4 The Judges’ views 112. At [82] ‑ [93] of the Judgment, the Judges dealt with the theme of continuity and To Lam Sin and Li Bun in some details. They took the view that the theme of continuity did not assist the respondents for a number of reasons. 113. First, the Judges noted that there was a fundamental change in constitutional order before and after 1997. Before 1997, under Article VII of the Letters Patent, it was the Governor who, by and with the advice and consent of the Pre-97 LegCo, might make laws for the peace, order, and good government of colonial Hong Kong. Further under Article X, the Governor had a discretion to assent or refuse assent to a bill passed by the Pre-97 LegCo or reserved it for the Crown’s signification. And if the Governor refused to give assent, there was nothing the Pre-97 LegCo could do about it. Under the BL, the LegCo is the legislature of Hong Kong (BL66). While a bill passed by the LegCo may take effect only after it is signed and promulgated by the CE (BL76), there are carefully calibrated provisions requiring the CE either to sign bills or refuse to do so in which case a chain of events would be set in motion that could result in the dissolution of the LegCo or resignation of the CE (BL49, 50 and 52). The Judges concluded that the law-making power of the LegCo under the BL is in substance different from that enjoyed by the Pre-97 LegCo whose constitutional role was to provide advice on, and give consent to, bills which the Governor then enacted into law in exercise of the law-making powers conferred on him under the Letters Patent[42]. 114. However, as we have explained above, properly understood, Article VII vested the general legislative powers not with the Governor but with the Pre-97 LegCo. The Pre-97 LegCo was not a mere advisory body assisting the Governor in making laws. It was the legislature of colonial Hong Kong. Had it not been the case, the judicial discussions in To Lam Sin and Li Bun based on the notion that a colonial legislature was sovereign and supreme within its province and how it could permissibly delegate its law making powers to the GIC in the context of the ERO were quite meaningless. And the Full Court would have decided the challenge that the ERO was ultra vires the Pre-97 LegCo on a totally different basis. The references to the difference before and after 1997 regarding the process after passage of a bill do not detract from the fact that the Pre-97 LegCo was the legislature of colonial Hong Kong under the Letters Patent before 1 July 1997, just is the case of the LegCo for HKSAR under the BL after 1 July 1997. 115. Second, the Judges held that the notion that the colonial legislature was sovereign and supreme within its province, which underpinned the reasoning of the Full Court in To Lam Sin and Li Bun, is no longer an apt description of the LegCo. For the BL is now supreme and even the LegCo cannot act contrary to a requirement under the BL. The LegCo no longer has the plenary power enjoyed by the Imperial Parliament but that which is conferred expressly or by implication on it under the BL[43]. They opined that there is a difference between a constitutional order which prescribes the legislature’s authority to make certain laws and binds the legislature to legislate according to certain procedures, and one which treats the legislature as supreme, citing The Executive Council of the Western Cape Legislature & Others v The President of the Republic of South Africa & Others 1995 (4) SA 877, at [59][44]. 116. As we have explained, the Pre-97 LegCo was always subject to the Letters Patent and other constitutional requirements. It was “supreme” or, as what we would prefer, autonomous, within the constitutional confines. That is the same for the LegCo. It is subject to the BL and all other constitutional requirements. Under the constitutional arrangement in the BL and further based on the common law doctrine of separation of powers, it is autonomous within the constitutional confines. The reference to the South African case does not take the analysis any further. 117. Third, the Judges noted that the ERO was enacted at a time when the Pre-97 LegCo consisted of the Governor, the Official Members and not more than six Unofficial Members which were appointed by the Governor, and was presided over by the Governor. They seemed to have endorsed a submission made in Ping Shek and another v The Canossian Institute, supra, at p72 that the GIC was a body that had actually, through its members, a controlling voice in the Pre-97 LegCo itself[45]. They then pointed out under the BL, the transfer of general legislative power by the ERO has to be examined in the context of a constitutional framework that seeks to ensure that laws are enacted, amended or repealed by a legislature constituted by election and whose composition is carefully prescribed. The LegCo is quite separate from the Government. There may be overlap in membership between the LegCo and the Executive Council, but under BL56, the CE need only consult the Executive Council and is not obliged to accept its majority opinion[46]. 118. In our view, the fact that politically the Governor was in control of the Pre-97 LegCo when the ERO was enacted or indeed at any other time does not detract from the fact that constitutionally and legally the Pre-97 LegCo remained the legislature of colonial Hong Kong. The Governor could not make laws on his own without proper delegation by the Pre-97 LegCo. How he was able to secure such delegation politically is irrelevant in terms of legal analysis. 119. Finally, the Judges considered that the reasoning of Hogan CJ in Li Bun, at pp100 and 102 supports the conclusion that inasmuch as the ERO confers “general legislative powers” on the CEIC (subject only to a limitation as to the occasions of public danger or emergency), the LegCo is deprived of any effective role to play in making of regulations which may range over virtually the whole field of legislation. Hogan CJ’s ultimate conclusion that the ERO did not cross the fine line under the then constitutional set-up is not applicable to the constitutional order under the BL[47]. 120. For reasons which we will develop below, we disagree that under the existing constitutional framework the LegCo cannot exercise any real or meaningful control over the emergency regulations made under the ERO. That Pre-97 LegCo, according to Hogan CJ, could not do so because of the restraints under the then constitutional order, is quite beside the point. Further, as will be seen, under the existing constitutional framework the court can also exercise effective control over the CEIC in exercising the powers under the ERO and the emergency regulations made. This is a crucial point in considering the constitutionality of the ERO in the overall constitutional context but was not considered by the Judges. 121. In consequence, the Judges erred in failing to give sufficient weight to the theme of continuity and to To Lam Sin and Li Bun in their analysis. E4. Principal features of the ERO 122. The ERO being constitutionally compliant is not only supported by the theme of continuity. Upon a proper construction by a closer and updated analysis of its principal features with reference to the relevant contemporaneous jurisprudence, the ERO only confers limited legislative power on CEIC to make subordinate emergency regulations, and not primary legislation, on an occasion of emergency or public danger. E4.1 True nature of the ERO and emergency regulations 123. First and foremost, it is imperative to bear in mind the true nature of the ERO and the emergency regulations made thereunder as a matter of substance. 124. There are many reasons for a legislature to delegate to the executive branch of the Government, or any other person or body, by way of an enabling primary legislation, legislative power to make subordinate legislation, as the editors in Bennion on Statutory Interpretation (7th edn), at section 3.1, enumerate: “(a) Modern legislation requires far more detail than Parliament itself has time or inclination for. For example, Parliament may not wish to concern itself with minor procedural matters. (b) To bring a complex legislative scheme into full working operation, consultation with affected interests is required. This can best be done after Parliament has passed the outline legislation, since it is then known that the new law is indeed to take effect and what its main features are. (c) Some details of the overall legislative scheme may need to be tentative or experimental. Delegated legislation provides an easy way of adjusting the scheme without the need for further recourse to Parliament. (d) Within the field of a regulatory Act new developments will from time to time arise. By the use of delegated legislation the scheme can be easily altered to allow for these. (e) If a sudden emergency arises it may be essential to give the executive wide and flexible powers to deal with it whether or not Parliament is sitting.” (emphasis supplied) 125. Scenarios (a) to (d) involve essentially the same legislative technique. The enabling primary legislation legislates on a subject matter in principle usually with a board framework, leaving the subordinate legislation to elaborate by filling in the details as and when required. 126. For scenario (e), the legislative approach can be different for a number of reasons. By nature, emergency or public danger is not capable of exhaustive definition, which means that usually a general or board definition is used. It ordinarily requires an urgent and effective response to avoid an imminent threat, prevent a worsening of the situation or mitigate the effects of the emergency. The executive needs wide and flexible powers to deal with every and all exigencies expeditiously and effectually. It follows that emergency regulations which the primary legislation delegated to the executive to make are necessarily wide and extensive in scope. They may even by virtue of the so-called “Henry VIII Clauses” dis-apply or amend a primary legislation. A ready example of adoption of such legislative approach is the English Civil Contingencies Act 2004. The editors of Wade and Forsyth on Administrative Law, (11th edn), at pp730 - 731 observe: “… [The] definition of an emergency in the 2004 Act is very wide. It comprises ‘serious threats’ to the welfare of any part of the population, the environment, the political, administrative or economic stability or, the security of the United Kingdom. There is no requirement that an emergency be declared, but Her Majesty may by Order in Council make emergency regulations for the purpose of preventing, controlling or mitigating an aspect or effect of the emergency if satisfied that an emergency is occurring or about to occur, that the regulations are necessary and the need is urgent. Practically anything may be required to be done, or prohibited, by the regulations. … The full plenary powers of Parliament have been given to the maker of the regulations for they ‘may make any provision of any kind that could be made by Act or Parliament’ including disapplying or modifying an Act.” 127. As dictated by the considerations peculiar to emergency and public danger, the nature of emergency legislation, both primary and subordinate, differs quite considerably from ordinary primary and subordinate legislations. One must constantly bear this in mind when approaching the constitutionality of the ERO. That is, with respect, what the Judges had failed to do. 128. At [56] of the Judgment, in giving the first reason why, in their opinion, the ERO conferred primary legislative powers on the CEIC, they highlighted the fact that the ERO was not an ordinary piece of primary legislation that legislated on a subject matter in principle leaving another body to devise the detailed legal norms that elaborated or put flesh on the broad matters laid down in the primary legislation. But such a difference always exists and is exactly the reason why the ERO and emergency regulations should be treated differently from ordinary primary and subordinate legislations. In citing that as a reason against the ERO’s constitutionality, the Judges had failed to sufficiently appreciate the true nature of the ERO and emergency regulations. E4.2 Section 2(1) 129. As the main power conferring provision, section 2(1) provides: “On any occasion which the Chief Executive in Council may consider to be an occasion of emergency or public danger he may make any regulations whatsoever which he may consider desirable in the public interest.” 130. Two main points arising from section 2(1) need to be addressed. 131. First, there is no definition for public danger[48]. It is a matter for the CEIC to consider if an occasion of public danger exists. The Judges were understandably concerned. At [60] of the Judgment, they observed that the meaning of public danger could potentially be very broad. They held that since the ERO did not state a “reasonably grounds” test, the discretion conferred by the ERO could be virtually unreviewable. They at [61] went on to query the utility of any review because of the strict confidentiality and pubic interest immunity attaching to information placed before the CEIC. Relying on certain passages in Li Bun, they at [62] noted that the width of the powers left precious little room for the doctrine of ultra vires to operate. 132. In our view, there is nothing objectionable in itself to task the CEIC to determine if an occasion of public danger exists. After all, it is a grave decision with immense consequences which affect the entire community. As the head of the HKSAR, the CE, with the advice of the Executive Council, is evidently the only suitable person to make the call. 133. That said, it does not follow that the CE could act freely without any rein as she wishes. She is subject to close judicial scrutiny. She can only invoke the ERO when there is a public danger. That condition must be strictly adhered to. Although there is no statutory definition for public danger, whether such a state exists at any given time can be objectively gauged by the prevailing circumstances. The past invocations in the colonial era are good examples. As said, they all involved serious and immediate threats to Hong Kong and its citizens as a whole which subsisted for a period of time. In the present instance, the parties do not dispute that a state of public danger arising from the recent ongoing and large-scale social unrest often associated with serious and prevalent violence did exist at the time when the PFCR was made. The present state of affairs shares the same commonality with the past instances of emergency and public danger. In any event, if necessary, the court may provide an interpretation to the meaning of public danger to fill the lacuna, if there is really one, as appropriate[49]. The CE’s decision to invoke the ERO can then be subject to judicial scrutiny by reference to such interpretation. Further, the CE’s decision that an occasion of public danger existed is reviewable under the Wednesbury principle too. She must also act bona fide in furtherance of the statutory purpose of the ERO according to the Padfield principle: Padfield v Minister of Agriculture Fisheries and Food [1968] AC 997. 134. The Judges’ concern that only very limited information is available making a legal challenge against the CE’s decision to invoke the ERO difficult, if not impossible, is misplaced. As will be seen in a while, emergency regulations made under the ERO are subject to the negative vetting procedure. For that procedure, the Government will ordinarily present a LegCo Brief, a public document, to explain the background, the underlying policy, the justification and various matters pertaining to the emergency regulations. The LegCo can also ask the Government for further relevant information. So even assuming that the papers placed before the CEIC are confidential or subject to immunity (which presumably may be waived), the LegCo Brief and any such further information will provide the necessary materials to facilitate the mounting of the legal challenge[50]. 135. The Full Court might have said something in Li Bun suggesting that there was little room for the doctrine of ultra vires to operate. But Li Bun was decided almost 60 years ago. The law has moved on since then. We do not see how Li Bun could have inhibited the application of the doctrine of ultra vires nowadays in any significant way. 136. In any event, all emergency regulations made under the ERO must not contravene the BL. As illustrated by our discussion on the PFCR below, this already provides adequate basis for the court to scrutinize their validity. 137. The second point arising from section 2(1) is this. The powers conferred by section 2(1) on the CEIC are undoubtedly wide and extensive. As explained above, given the nature of the ERO and emergency regulations, such powers are necessary. That the Judges failed to appreciate when they at [57] of the Judgment cited it as a reason for ruling against the constitutionality of the ERO. E4.3 Section 2(2)(g) 138. This is the so-called “Henry VIII Clause”: “Without prejudice to the generality of the provisions of subsection (1), such regulations may provide for– … (g) amending any enactment, suspending the operation of any enactment and applying any enactment with or without modification;” 139. As Hogan CJ observed in Li Bun, at p101, it is necessary to confer on the CE such power, otherwise her capability to make adequate provisions for some unexpected danger or emergency might be hampered or limited by the inability to alter any existing ordinance, especially, when the LegCo could not, for one reason or another, be brought into session or meet. 140. The Judges disagreed for two reasons. 141. At [64] of the Judgment, they first said that there is no provision in the BL that authorizes the CEIC by herself to amend or repeal primary legislation. That must be true. But the CEIC does not derive the power to amend or repeal primary legislation from the BL. She does so from the delegation by the LegCo under the common law. This leads to their second point. 142. At [65] - [66] of the Judgment, the Judges observed that validity of Henry VIII Clauses is based on parliamentary sovereignty which the LegCo does not enjoy and that such Clauses are antithetical to the norm of subsidiary legislation as understood in Hong Kong under section 28(1)(b) of the Interpretation and General Clauses Ordinance (Cap 1) (“the IGCO”) that “no subsidiary legislation shall be inconsistent with the provisions of any Ordinance”. 143. While it is true that the LegCo does not enjoy supremacy in the Diceyan sense, it is wrong to say that it cannot as a matter of Hong Kong common law delegate to the CE the power prescribed by section 2(2)(g) of the ERO in cases of emergency or public danger. There is nothing in the BL or anywhere else to indicate that such common law principle contravenes the BL and has not migrated to the post‑1997 era. 144. The Judges’ reliance on section 28(1)(b) of IGCO is misplaced. Pursuant to section 2(1) of the same Ordinance, the application of the provisions of the IGCO is subject to the contrary intention that appears from the context of any ordinance. Given the context of the ERO, there is definitely a contrary intention to dis-apply section 28(1)(b) to emergency regulations made thereunder. E4.4 Penalties 145. As part of the powers conferred on her, section 3(1) of the ERO enables the CEIC to make regulations to provide for a contravention as a criminal offence with associated penalties up to a maximum of mandatory life imprisonment. This is evidently part and parcel of the wide and extensive powers that the CEIC needs to effectively tackle emergency or public danger, especially when it involves a serious, prevalent and subsisting breach of law and order. 146. The Judges’ only criticism is that it is contrary to the norm for subsidiary legislation under section 28(1)(e) of the IGCO, which specifies the maximum sentence for an offence based on contravention of a subsidiary provision to be a fine of HK$5,000 or six months’ imprisonment. Again, the Judges erred in failing to see that because of the context of the ERO, section 28(1)(e) must have been dis-applied. E4.5 Negative vetting and repeal 147. According to section 2(3) of the ERO, the emergency regulations made shall continue in force until repealed by order of the CEIC. 148. As the Judges rightly held at [69] - [72] of the Judgment, section 2(3), properly construed, does not itself prevent such regulations from repeal by resolution of the LegCo during negative vetting under section 34 of the IGCO or from repeal by a subsequent ordinance. The latter appears to be consistent with what the Pre-97 LegCo did in 1936 in the failed attempt to abrogate the emergency regulations in question. 149. At [74] of the Judgment, the Judges said that it is open to the CEIC, as part of the regulations made under the ERO, to amend or suspend the operation of section 34 of the IGCO. In our view, when the ERO is properly construed, it does not allow the CEIC to do so. Pursuant to section 2A(1) of the IGCO, all laws previously in force in Hong Kong shall be construed with such modifications, adaptions, limitations and exceptions as may be necessary so as not to contravene the BL. Thus the ERO must now be construed to make it compatible with the BL. And under BL73(1), the LegCo’s legislative power to amend or repeal laws in accordance with the provisions of the BL and legal procedures must include the power to scrutinize, and if found necessary, to amend or even repeal, subordinate legislation made by other branches of the Government. So on a proper interpretation to make it in line with BL73(1), the ERO does not empower the CEIC to amend or suspend the operation of section 34, thereby depriving the LegCo the control over the emergency regulations by the negative vetting procedure. E4.6 Check on duration of regulations 150. Despite the point on negative vetting, the Judges at [68] of the Judgment said that there is no time limit on the validity and force of the regulations, nor any mechanism for constant review. However, in accordance with the negative vetting procedure under section 34 of the IGCO, there is a first phase of scrutiny whereby the regulations may take effect pending expiry of the time limit of 28 days, which may be extended for 21 days. If repealed during negative vetting, the regulations will not last beyond such time limit. Further, even if the regulations continue to take effect despite negative vetting, there is a second phase of scrutiny whereby the LegCo can further review them. Nothing can stop the LegCo from stepping in to put an end to the emergency regulations by way of primary legislation. Further, as Mr Yu has rightly accepted, the duration of the emergency regulations is open to challenge by way of judicial review if the CE fails to repeal them when the state of public danger no longer exists. 151. The Judges at [75] - [79] of the Judgment were concerned that once promulgated, the emergency regulations immediately become part of the general law. Even if repealed later, it would not “un-do” the regulation’s effects and revive the original legislation if it was repealed by the regulations. We need not dwell on the point. For such scenario happens whenever a piece of legislation is repealed. By itself it does not operate against the ERO. 152. In our view, when properly analyzed, the emergency regulations which the ERO authorizes the CEIC to make is subordinate legislation. The ERO does not confer on the CEIC general legislative power to make primary legislation. E5. Conclusion 153. In our discussion above, we have not dealt with each and every single point taken by counsel. We consider what we have set out above is sufficient for the present purpose. And for the foregoing reasons, we hold that the ERO does not confer on the CEIC general legislative power to make primary legislation. It does not contravene the BL. 154. In KWH-RN, KWH seek to further support the Judges’ conclusion under Ground 1 on the basis of the principle of legality[51]. In gist, it is submitted that the principle of legality averts a construction of BL17(2), 66 and 73(1) that the drafters intended that the LegCo could enact laws that would have the effect of conferring unlimited and general law‑making power on the CEIC, or conversely that the CEIC could constitutionally rely on BL56(2) and 62(5) to usurp the role of the LegCo. In our views, the principle of legality contention does not add anything further to the applicants’ other contentions that we have considered above. Thus, for the same reasons, we would also reject such contention. 155. In the premises, we would respectfully differ from the Judges and reject Ground 1. We would allow the respondents’ appeal in this regard. F. GROUNDS 5A AND 5B F1. Standard of scrutiny 156. It is not disputed that the PFCR imposed restrictions on fundamental rights protected by the BL and the BOR and must, in addition to being prescribed by law, satisfy the proportionality test in order to be valid. The rights engaged were identified at [127] (BL27 and BOR14, 16 and 17 rights in respect of section 3 of the PFCR) and [169] ‑ [173] (BL27, 28 and 31 and BOR5, 8, 14 and 16 rights in respect of section 5) in the Judgment. 157. The proportionality test involves a four-step examination: (1) whether the restrictions pursue a legitimate aim, and in respect of rights under BOR16 and 17, the legitimate aim must come within the scope of specified purposes set out in those articles; (2) whether the restrictions are rationally connected with such legitimate aim(s); (3) whether the restrictions are no more than necessary for achieving the legitimate aim(s); (4) whether a fair balance has been struck between the societal benefits pursued by the restrictions and the inroads made to the rights of the individuals. 158. In our judgment, notwithstanding that the PFCR touches upon matter of public order and measures which are necessary for the restoration of law and order in Hong Kong and to meet the challenges to the stability of our society presented by the recent turmoil, the appropriate standard that the court should adopt in carrying out the proportionality analysis in this instance should be the stricter standard of “no more than necessary” as submitted by Mr Chan instead of the “manifestly without reasonable foundation” standard advocated by Mr Yu. 159. In Kwok Cheuk Kin v Secretary for Constitutional and Mainland Affairs (2017) 20 HKCFAR 353, Ma CJ explained at [37] ‑ [46] the considerations which guide the Court in deciding the standard to be adopted in the proportionality analysis. Three facets were mentioned in particular: the nature of the right in question and the degree to which it has been encroached upon; the identification of the relevant decision-maker; and the margin of appreciation. 160. The Chief Justice also highlighted the separate constitutional and institutional responsibilities of the judiciary and other organs of the Government. In assessing the extent to which margin of appreciation should be accorded to the decision maker, the court should consider if the decision maker is institutionally likely to be better placed than the court to make an assessment in relation to the particular issue in question. 161. Though we agree with Mr Yu that the provisions in the PFCR do not prohibit the exercise of the rights engaged, viz freedom of expression, right to assembly and demonstration and right to privacy, we also bear in mind that section 3 imposes restrictions which curtail the exercise of these fundamental rights in a substantial manner. In some circumstances (and we recognize that they are real possibilities) the measures may discourage some people from exercising their right of demonstration and even bar them from a particular form or manner of demonstration. 162. Whilst we would not underestimate the onerous and difficult responsibility on the CEIC in deciding whether the prevalent public danger called for measures in the PFCR, as illustrated by previous cases like Leung Kwok Hung & Others v HKSAR (2005) 8 HKCFAR 229, the court is constitutionally and institutionally well-placed to make proportionality assessment in respect of legislation (including subsidiary legislation) relating to the maintenance of law and order in our society. The court is also institutionally tasked with the constitutional duty to strike a balance between conflicting societal interest and fundamental rights of individuals in public order matters. 163. Hence, we respectfully agree with the Judges that the appropriate standard should be the “no more than necessary” standard. 164. Like the Judges, we shall address the proportionality of section 3 and that of section 5 separately. F2. Section 3 F2.1 Legitimate aims and rationality 165. At [130] of the Judgment, the legitimate aims of the PFCR were identified as follows: (1) deterrence and elimination of the emboldening effect for those who may otherwise, with the advantage of facial covering, break the law; (2) facilitation of law enforcement, investigation and prosecution. 166. Mr Yu submitted that these legitimate aims are to be considered against the background of the crisis of public disorder that Hong Kong had experienced in the past few months and that the exercise of the power under the ERO to make the PFCR was deemed necessary by the CEIC in order to restore the stability and to maintain the law and order in the society. 167. Counsel referred to the speech of the CE on 4 October 2019 when she announced the promulgation of the PFCR at a press conference. To recap, the CE recounted the frequent and widespread violent behaviours of some rioters in the course of demonstrations and processions since June 2019 and the escalation in such violence resulting in serious damage (including damage to private properties) and injuries to citizens and law enforcement officers. She also highlighted that lethal weapons were being used by rioters to attack police officers. A high proportion of these violent protesters were students and many of them wore masks and other protective gears to hide their identities in order to escape from criminal responsibilities. 168. In the Judgment, the Judges also alluded to the dire situation facing Hong Kong at [132], reiterating the disorder and violence experienced and witnessed by all those in society incessantly and the development of such violence on an escalating scale. At [137] and [138], the Judges agreed with Mr Yu that many public assemblies or processions which initially took place lawfully and peacefully were turned into unauthorized or unlawful ones with some radical protesters resorting to violence. 169. The Judges found that there could not be a simple dichotomy between peaceful and violent protesters as people’s behaviour may change depending on the circumstances and the influence from others around them. They further accepted that facial covering makes law enforcement, investigation and prosecution more difficult. 170. Against such background, the Judges held that the PFCR served legitimate aims and they are rationally connected with the same. 171. Before us, Mr Chan did not challenge these conclusions of the Judges. He also accepted that the restriction under section 3(1)(a) (prohibition in respect of unlawful assembly) is no more than necessary and strikes a fair balance between individuals’ right and the societal benefit served by the restriction. Mr Pun SC on behalf of LKH likewise did not challenge these conclusions in relation to section 3 (though he challenged the same in respect of section 5, which we shall deal with below). Counsel also accepted section 3(1)(a) is proportionate. 172. Hence, in these appeals, we shall focus on the proportionality of sections 3(1)(b), (c) and (d), in particular, for each of those restrictions whether it is no more than necessary to achieve the legitimate aims and whether a fair balance has been struck. 173. Before we address that key issue, we should first examine the precise scope of operation of section 3(1)(b) (which involves the examination of the concept of unauthorized assembly under the POO and its significance in terms of public order management) and the extent to which it interferes with the fundamental rights of demonstration and freedom of expression. F2.2 Permissible restrictions and scope of section 3(1)(b) 174. We start the analysis by reminding ourselves some relevant authorities on the permissible restrictions on the fundamental rights of demonstration and freedom of expression. 175. The right of demonstration and freedom of assembly are not absolute. In HKSAR v Chow Nok Hang (2013) 16 HKCFAR 837, the Court of Final Appeal highlighted the limits in section B.3 of the judgment. In particular, Ribeiro PJ said at [38] ‑ [40]: “38. Article 17 allows a line to be drawn between peaceful demonstrations (where, as noted above, full rein is given to freedom of expression) and conduct which disrupts or threatens to disrupt public order, as well as conduct which infringes the rights and freedoms of others. In Leung Kwok Hung v HKSAR, having recognized that the interests of ‘public order (ordre public)’ are listed by Article 17 as a legitimate purpose, the Court held that there is no doubt that such concept ‘includes public order in the law and order sense, that is, the maintenance of public order and the prevention of public disorder’. It concluded that a statutory scheme giving the Commissioner of Police discretion to regulate public processions with a view to maintaining public order was constitutionally valid after severance of certain objectionably vague words. 39. Once a demonstrator becomes involved in violence or the threat of violence – somewhat archaically referred to as a ‘breach of the peace’ – that demonstrator crosses the line separating constitutionally protected peaceful demonstration from unlawful activity which is subject to legal sanctions and constraints. The same applies where the demonstrator crosses the line by unlawfully interfering with the rights and freedoms of others. 40. The law therefore imposes bounds on the constitutionally protected activity of peaceful assembly. The need for such limits is sometimes dramatically illustrated in situations involving demonstrations and counter-demonstrations. It is not uncommon for one group, demonstrating in favour of a particular cause, to find itself confronted by another group demonstrating against that cause. The situation may be potentially explosive and the police will generally try to keep them apart. Obviously, if both remain within their lawful bounds, all will be well. But often, conflict and public disorder may result. Sometimes, both sides will have broken the law. But in some cases, the disruption of public order is caused only by one side. The task of the law enforcement agencies and the courts is then to identify the source of such disruption by identifying the demonstrators who have crossed the line into unlawful activity. They thereby avoid curtailing or punishing the constitutionally protected activities of the innocent group.” (emphasis supplied) And at [42]: “42. Lines also have to be drawn where a demonstrator’s conduct impinges unacceptably upon the rights of others (which may or may not be constitutionally protected rights). Such a line had to be drawn, for instance, in Yeung May Wan v HKSAR, where the Court had to decide whether the offence of obstructing a public place was properly applied so as to curtail a static, peaceful demonstration by a small group of Falun Gong protesters which obstructed only part of the pavement, on the basis that they were interfering with the rights of other users of the public highway. To take another example, the Court of First Instance recently had to decide whether the right to demonstrate entitled protesters to take their demonstration into a private residential development without the consent of the owners, or whether that right was constrained by the need to respect the private property rights of the residents.” 176. In the more recent case of Kudrevicius v Lithuania (2016) 62 EHRR 34, the Strasbourg Court considered if the criminal convictions of the applicants were compatible with the right to freedom of assembly. The applicants did not engage in any acts of violence. The regional court found that they had organised a gathering with the aim of seriously breaching public order, viz organising the blockade of a highway. They were convicted of criminal offences and were sentenced to 60 days’ imprisonment which were suspended. The Strasbourg Court held that the criminal conviction was justified as being necessary for pursuing the legitimate aims of prevention of disorder and protection of the rights and freedoms of others and found no violation of the freedom of assembly. 177. In respect of the breach of public order occasioned by blockage of road, the Court observed at [97] (omitting the footnotes): “However, the applicants’ conviction was not based on any involvement in or incitement to violence, but on the breach of public order resulting from the roadblocks. The Court further observes that, in the present case, the disruption of traffic cannot be described as a side-effect of a meeting held in a public place, but rather as the result of intentional action by the farmers, who wished to attract attention to the problems in the agricultural sector and to push the government to accept their demands. In the Court’s view, although not an uncommon occurrence in the context of the exercise of freedom of assembly in modern societies, physical conduct purposely obstructing traffic and the ordinary course of life in order to seriously disrupt the activities carried out by others is not at the core of that freedom as protected by Article 11 of the Convention …” 178. The Court characterized the conduct of the applicants as “reprehensible” (see [101]). Such finding was important because on the facts of the case the demonstration was initially authorized and under Strasbourg case law, a person cannot be subject to a sanction for participation in a demonstration which has not been prohibited so long as that person does not himself commit any reprehensible act on such an occasion, see [149]. The Court further explained the concept of “reprehensible act” at [173] and [174] (omitting the footnotes): “173. As can be seen from the above case-law, the intentional serious disruption, by demonstrators, to ordinary life and to the activities lawfully carried out by others, to a more significant extent than that caused by the normal exercise of the right of peaceful assembly in a public place, might be considered a ‘reprehensible act’ within the meaning of the Court’s case-law. Such behavior might therefore justify the imposition of penalties, even of a criminal nature. 174. The Court considers that, even though the applicants had neither performed acts of violence nor incited others to engage in such acts, the almost complete obstruction of three major highways in blatant disregard of police orders and of the needs and rights of the road users constituted conduct which, even though less serious than recourse to physical violence, can be described as ‘reprehensible.’” 179. In this respect, the concurring speech of Judge Wojtyczek at O1-3 pinpointed the rationale for finding such conducts reprehensible: “… The effects of such roadblocks go far beyond the usual disruptions caused by demonstrations in public places. They also go well beyond the idea of a sit-in protest organised around certain specific places for the purpose of blocking access to them. The applicants took these actions to promote their opinions, not by strength of argument, but by directly undermining the legitimate individual economic interests of a significant number of third parties, and by disrupting the economic life of their country – and thus, more by argument of strength. The demonstrators’ message was meant not only to be heard but also to directly affect their fellow-citizens. In that context, the reasoning of the judgment rightly describes the applicants’ acts as reprehensible …” 180. The Court also reiterated that any measures interfering with freedom of assembly other than in cases of incitement to violence must be subject to careful scrutiny. In particular, at [94] and [146], the Court had these to say: “94. In this connection, it is not without interests to note that an individual does not cease to enjoy the right to freedom of peaceful assembly as a result of sporadic violence or other punishable acts committed by others in the course of the demonstration if the individual in question remains peaceful in his or her own intentions or behaviour. The possibility of persons with violent intentions, not members of the organising association, joining the demonstration cannot as such take away that right. Even if there is a real risk that a public demonstration might result in disorder as a result of developments outside the control of those organising it, such a demonstration does not as such fall outside the scope of paragraph 1 of Article 11, and any restriction placed thereon must be in conformity with the terms of paragraph 2 of that provision. ... 146. … Where the sanctions imposed on the demonstrators are criminal in nature, they require particular justification. A peaceful demonstration should not, in principle, be rendered subject to the threat of a criminal sanction, and notably to deprivation of liberty. Thus, the Court must examine with particular scrutiny the cases where sanctions imposed by the national authorities for non-violent conduct involve a prison sentence.” 181. For demonstration which has not been authorized, the Court made some general observations at [150] ‑ [153]. Whilst acknowledging that it is essential to have a system of prior notification, the absence of prior authorization does not give carte blanche to the authorities in taking enforcement actions. Such actions would still be subject to proportionality requirement. The Court highlighted that though there could be special circumstances which justify the holding of spontaneous demonstrations without prior notification, such exception must not be extended to the point where the absence of prior notification of a spontaneous demonstration can never be a legitimate basis for crowd dispersal. 182. Thus, the proportionality analysis has to be applied on two different levels: (1) examining the systemic proportionality by reference to the legislation or rules in question; (2) examining the operational proportionality by reference to the actual implementation or enforcement of the relevant rule on the facts and specific circumstances of a case at the operational level. 183. In these appeals, it should be emphasized that we are only concerned with the first level of challenges. It remains for the court to assess the proportionality on the second level on the facts and circumstances in a particular case if a charge is brought against person. 184. The Strasbourg Court continued at [155] ‑ [157] in Kudrevicius v Lithuania, supra, to identify the need to have measures to restrict conducts causing disruption to ordinary life to a degree exceeding that which is inevitable for peaceful demonstration and assembly. At [155], the Court alluded to two important mindsets for striking the balance: (1) on the one hand, the public authorities have to show a degree of tolerance; (2) on the other hand, demonstrators should comply with the regulations in force. 185. At [156], the Court said: “The intentional failure by the organisers to abide by these rules and the structuring of a demonstration, or of part of it, in such a way as to cause disruption to ordinary life and other activities to a degree exceeding that which is inevitable in the circumstances constitutes conduct which cannot enjoy the same privileged protection under the Convention as political speech or debate on questions of public interest or the peaceful manifestation of opinions on such matters. On the contrary, the Court considers that the Contracting States enjoy a wide margin of appreciation in their assessment of the necessity in taking measures to restrict such conduct.” 186. The rationale for imposing necessary restrictions was explained at [157]: “Restrictions on freedom of peaceful assembly in public places may serve to protect the rights of others with a view to preventing disorder and maintaining an orderly flow of traffic. Since overcrowding during a public event is fraught with danger, it is not uncommon for State authorities in various countries to impose restrictions on the location, date, time, form or manner of conduct of a planned public gathering.” 187. We shall now turn to consider the effective scope of section 3(1)(b). 188. Section 3(1)(b) prohibits the use of any facial covering that is likely to prevent identification while a person is at an unauthorized assembly. Unauthorized assembly has the meaning ascribed to it as set out in section 17A(2) of the POO. There are three different scenarios: (1) Any public meeting or public procession taking place in contravention of section 7 or 13 of the POO, viz either those without requisite notification being given to the Commissioner or those with notification and the Commissioner does not issue a notice of no objection, or processions not complying with the requirements under section 15. For public meeting, section 7 only applies in respect of a meeting of more than 50 persons. For public procession, section 13 only applies to public procession on public highway or thoroughfare or in a public park which consists of more than 30 persons. (2) Three or more persons taking part in or forming part of a public gathering refuse or wilfully neglect to obey an order given by Commissioner of Police under section 6 of the POO. Under section 6, the Commissioner can control and direct the conduct of all public gatherings and specify the route by which and the time at which any public procession may pass. He may also regulate the playing of music or amplification of human speech or other sound in public places. (3) Three or more persons taking part in or forming part of a public meeting, public procession or public gathering or other meeting, procession or gathering refuse or wilfully neglect to obey an order given under section 17(3) of the POO. Under section 17(3), a police officer may give an order for the purpose of section 17(1) to prevent the holding of, stop or disperse any public meeting in contravention of section 7 or any public procession in contravention of section 13. Also under section 17(3), a police officer of or above the rank of inspector may issue an order for the purpose of section 17(2) to prevent the holding of, stop, disperse or vary the place or route of any public gathering if he reasonably believes that the same is likely to cause or lead to a breach of the peace. Similar orders can be made in respect of public gathering for religious purposes. 189. Public gathering is defined in section 2 of the POO to mean a public meeting, a public procession and any other meeting, gathering or assembly of 10 or more persons in any public place. 190. It should also be noted that: (1) under section 17A(1), any person who refuses or wilfully neglects to obey an order given under section 6 or 17(3) commits an offence; and (2) under section 17A(3), every person who, without lawful authority or reasonable excuse, knowingly takes or continues to take part in or forms or continues to form part of any such unauthorized assembly shall be guilty of an offence. 191. Further, in respect of the conditions imposed for a public meeting which is subject to the requirement of notification under section 7, the conditions would be imposed under section 11(2) instead of section 6. For public procession which is subject to the requirement of notification under section 13, conditions imposed for the same would be issued under section 15(2) instead of section 6. These conditions (unlike conditions imposed under section 6) are subject to the appeal procedure under section 16. 192. Since section 7 applies to public meeting of more than 50 persons and section 13 applies to public procession of more than 30 persons, they cover most public gatherings of a large scale in Hong Kong. 193. The provisions in the POO have to be construed (insofar as it is possible to do so) in a manner which is compatible with the fundamental rights of demonstration and procession: see R v Home Secretary Ex p Simms [2000] 2 AC 115 at p131. They must also be applied in a manner compatible to such fundamental rights. Though section 17A(2)(a) provides that a public meeting or procession taken place in contravention of section 7 or 13 would be an unauthorized assembly, in light of the principle of tolerance and proportionality on the operational level as discussed in Kudrevicius v Lithuania, supra (see [180] ‑ [184] above) as well as a matter of practicality in view of the number of persons present at such gatherings, unless there are violent or other reprehensible conducts on the part of some demonstrators posing serious and imminent risk to public order and safety which requires immediate actions on the part of police, there should be prior warnings and the issuance and announcement of an order under section 17(3) before more drastic actions like arrests and dispersals are taken. 194. For gatherings which are not subject to notification requirement, the Commissioner would not have any advance notice and it is rather unusual for conditions to be imposed by the Commissioner under section 6 for such gatherings. Given that the Commissioner would be unlikely to be present at the scene of gathering, it is also unlikely that a condition will be imposed under section 6 in the course of such gathering. Further, in the context of an unauthorized assembly under section 17A(2)(b), the principle of tolerance and requirement of proportionality on operational level are also engaged. Like the case under section 17A(2)(a), unless there are violent or other reprehensible conducts on the part of some demonstrators posing serious and imminent risk to public order and safety which requires immediate actions on the part of police, there should be prior warnings and the issuance and announcement of an order under section 17(3) before more drastic actions. 195. In cases where conditions are imposed under section 11(2) or 15(2), when there are persons taking part in or forming part of the public meeting or public procession acting in breach of a condition (eg, the routes prescribed under conditions imposed), if the circumstances so required, the proper procedure under the statutory scheme of the POO is for the police to exercise the discretionary power under section 17(1)(a) and (b) with the giving of an order under section 17(3) to prevent the holding of, stop or disperse the gathering. 196. The power to order the stopping and dispersal of a gathering under section 17(3) must be exercised for the purpose of maintenance of public safety, public order and the protection of the rights and freedoms of others since the conditions under sections 11(2) and 15(2) can only be legitimately imposed for such purposes. The purpose of section 17 has to be ascertained by reading all relevant provisions together and in the context of the whole statute: see Medical Council of Hong Kong v Chow Siu Shek (2000) 3 HKCFAR 144. Thus, with the application of the Padfield principle as discussed at great length in Chee Fei Ming v Director of Food and Enviromental Hygiene [2019] HKCA 1425, the power under section 17 cannot be exercised arbitrarily. 197. Further, in terms of unauthorized assembly by virtue of non‑compliance with the notification requirements or the no objection requirement under section 7 or 13, before more direct enforcement actions like actual dispersal or arrest are taken by the police, the proportionality requirement on the operational level (see the discussion at [181] ‑ [184] above by reference to Kudrevicius v Lithuania, supra, prior warnings should have been given about the unauthorized nature of the gathering and an order for stop and dispersal should have been made under sections 17(1) and (3). 198. After such order has been given, the gathering should be discontinued. Otherwise, those without lawful authority or reasonable excuse, knowingly takes or continues to take part in the unauthorized assembly would commit an offence under section 17A(3). 199. Another possibility is that during the course of a public procession or public meeting, the conducts or behaviours of some persons taking part in the gathering lead a police officer of or above the rank of inspector to reasonably believe that it is likely to cause or lead to a breach of the peace. In that case, the police officer may exercise the power under sections 17(2)(a) and (b) in conjunction with the giving of an order under section 17(3) to prevent the holding of, stop or disperse the gathering. 200. Thus, under all these scenarios referred to at [188] above, when it is necessary for the police to take drastic actions like arrests and dispersal in respect of an unauthorized assembly, unless there are violent or other reprehensible conducts on the part of some demonstrators posing serious and imminent risk to public order and safety which requires immediate actions on the part of the police, an order should have been made under section 17(3) and there are at least three or more persons refuse or wilfully neglect to obey the order. 201. There cannot be a refusal or wilful neglect to obey an order unless the persons concerned were aware of the order. 202. Also a person cannot be regarded as knowingly taking part in an unauthorized assembly unless he was aware of the unauthorized nature of the gathering or the order under section 17(3) and that there are three or more persons refusing or wilfully neglecting to comply with such order. 203. The compatibility of the statutory scheme under the POO (in respect of public procession) was examined by the Court of Final Appeal in Leung Kwok Hung & Others v HKSAR, supra. The defendants in that case were convicted of offences under section 17A(3)(b)(i) for holding and assisting in the holding of unauthorized assembly. The assembly in question was a peaceful procession of between 40 and 96 people and it was unauthorized because the defendants refused to give prior notification. 204. At [47], [50] and [54] ‑ [57], the majority of the Court of Final Appeal highlighted that the discretionary powers of the Commissioner must be exercised for the statutory legitimate purposes, viz, he reasonably considers that the restriction is necessary in the interests of national security or public safety, public order or the protection of the rights and freedom of others. The Commissioner must himself apply the proportionality test in that consideration, see [57]. 205. Though the Court of Final Appeal held that the statutory scheme under the then POO (regulating the discretion of the Commissioner by reference, inter alias, to the concept of order public) did not satisfy the “prescribed by law” requirement, the majority held that such reference to order public could be severed and the remaining concept of public order (in the law and order sense, viz, the maintenance of public order and prevention of public disorder, see [82] and [83]) is sufficiently certain. The majority of the Court of Final Appeal also held that the Commissioner’s statutory discretion to restrict the right of peaceful assembly for the purpose of public order is no more than is necessary to accomplish the legitimate purposes: see [92] ‑ [94]. The convictions were upheld accordingly. 206. In this connection, though on the facts of the case the conviction was due to the failure to give notification, the Court of Final Appeal also examined the full range of discretionary powers that the Commissioner and the police could exercise in restricting the freedom of assembly and procession: see the analysis of the statutory scheme and the statutory discretion at [43] ‑ [63]. In light of that, it would be surprising if after the severance of order public from the relevant provisions, the majority of the Court of Final Appeal still regarded some aspects of the statutory discretion concerning unauthorized assembly other than the requirement to give notification to be unconstitutional and made no comment on the same. 207. As explained by the Court of Final Appeal, the public order considerations are not confined to cases where violent confrontations would entail. They also embrace traffic conditions and crowd control. Presence of rival groups and reaction of members of the public are also relevant: see [92(3)]. 208. To sum up, there are valid and serious public order concerns for unauthorized assembly: (1) for those gatherings held without complying with the requisite notification requirement, the justification for imposing criminal sanctions against the same has been explained by the majority in Leung Kwok Hung & Others v HKSAR, supra; (2) for those gatherings involving breach of conditions (including conditions imposed under sections 6, 11(2) and 15(2)), the conditions are imposed in the first place as being necessary in the interests of national security or public safety, public order or the protection of the rights and freedoms of others (and subject to the criteria of proportionality, see Leung Kwok Hung & Others v HKSAR, supra) and in the case of sections 11(2) and 15(2) where a police officer considers that it is necessary to make a further order under section 17(3) to prevent the holding of, stop or disperse the gathering (the making of such order can also be subject to assessment of proportionality on the operational level, see Kudrevicius v Lithuania, supra, as discussed above); (3) for those gatherings involving an order by a police officer of or above the rank of inspector under sections 17(2) and (3), the situations are those where the police officer reasonably believes that the same is likely to cause or lead to a breach of the peace. 209. Unlike the offence of unlawful assembly under section 18 (which is the situation addressed under section 3(1)(a) of the PFCR), there is no specific requirement of an unauthorized assembly under section 17A(2) that the persons assembled conduct themselves in a disorderly, intimidating, insulting or provocative manner intended or likely to cause any person reasonably to fear that they will commit a breach of the peace. 210. Such distinction was regarded by the Judges as significant in assessing if the restriction under section 3(1)(b) is no more than necessary in the proportionality analysis. Thus, at [154] of the Judgment, they observed: “Second, s 3(1)(b) relates to ‘unauthorized’ assemblies as explained in §§26 and 27 above. An assembly which is ‘unauthorized’ may yet be entirely peaceful, without any violence being used or threatened by anyone participating in that assembly, eg a large scale public procession may become an unauthorized procession as a result of the failure by some participants to comply with a condition as regards the route of the procession imposed by the Commissioner under s 15(2) of the POO but the participants may continue to proceed with the procession in an entirely peaceful and orderly manner.” 211. With respect, in so holding, the Judges failed to have regard to the pre-emptive nature of the provisions in imposing sanctions against unauthorized assemblies under the POO and their significance in the context of the maintenance of public order, particularly in the context of the current public danger which, as explained above, the PFCR as a piece of regulations promulgated under the ERO is meant to address. 212. Further, the Judges failed to take account of the actual operation of the statutory scheme as explained above, in particular the circumstances under which an order under section 17(3) of the POO would be made. If it is only a small number of participants deviating from the approved route of procession without any conducts or behaviour causing serious disruption to public order, an order for the stop and dispersal of the gathering will not be compatible with the principle of proportionality on an operational level. 213. As explained above, under the scheme of the POO, in respect of the designation of the route as a condition imposed for a procession under section 15(2), such condition is set because the Commissioner regards it as being necessary and proportionate in the interests of national security or public safety, public order or the protection of the rights and freedoms of others. It follows that deviation from the designated route would have ramifications on the maintenance of public order even though such deviation does not entail violence or threat of violence. 214. Even so, such deviation would not per se turn the procession into an unauthorized assembly. Police officers could give warnings against deviations or take other steps to restore order or to prevent disorder. It is a matter of discretion and judgment on the part of the police officers present at the scene, guided by the principles of law set out in this judgment (in particular the importance of proportionality on the operational level discussed at [180] ‑ [184] and [193] ‑ [197] above), to assess what steps are necessary to redress the disruption to public order occasioned by a deviation. Under section 17(2), a police officer of or above the rank of inspector may vary the route of any public gathering if he reasonably believes that the same is likely to cause or lead to a breach of the peace. There are also other measures that can be taken under sections 17(4) and (6). 215. It is only when a police officer forms a judgment that due to the conducts and behaviours of some participants in breach of conditions imposed for the gathering, such gathering has to be stopped and dispersed and the officer issues an order to that effect under section 17(3) the implications for unauthorized assembly under section 17A(2)(c) (including the application of section 3(1)(b) of the PFCR) kick in. As explained at [201] and [202] above, the persons concerned must have the requisite knowledge before he could be found guilty of the offence under section 17A(3). 216. In this connection, the analysis of the statutory scheme under the POO (and thus the scope of section 3(1)(b) of the PFCR) discussed at [188] ‑ [202] and [212] ‑ [215] above must be borne in mind. In particular, the prerequisite of an order under section 17(3) in cases other than those involving violent or other reprehensible conducts on the part of some demonstrators posing serious and imminent risk to public order and safety which requires immediate actions on the part of the police (see [193] above) and the proportionality on the operational level are safeguards against unjustified interference with the rights of peaceful demonstration and assembly. 217. Thus, the scenario postulated by the Judges that a procession which proceeded peacefully and orderly would by virtue of a deviation of a minor scale turn the whole gathering into an unauthorized assembly is, with respect, rather unreal. 218. At this juncture, we should address the submission that because the use of the word “at” in section 3 of the PFCR, a person could be found guilty simply by virtue of his presence at and wearing a mask in an unauthorized assembly even if he does not participate in the acts occasioning an order being made under section 17(3) and has no knowledge of such order. 219. To recap, section 3(1)(b) of the PFCR provides: “A person must not use any facial covering that is likely to prevent identification while the person is at unauthorized assembly.” 220. A contravention of section 3(1), including subparagraph (1)(b), is an offence by virtue of section 3(2). At the same time, section 4 provides a defence: “(1) It is a defence for a person charged with an offence under section 3(2) to establish that, at the time of the alleged offence, the person had lawful authority or reasonable excuse for using a facial covering. (2) A person is taken to have established that the person had lawful authority or reasonable excuse for using a facial covering if‑ (a) there is sufficient evidence to raise an issue that the person had such lawful authority or reasonable excuse; and (b) the contrary is not proved by the prosecution beyond reasonable doubt. (3) Without limiting the scope of the reasonable excuse referred to in subsection (1), a person had a reasonable excuse if, at the assembly, meeting or procession concerned‑ (a) the person was engaged in a profession or employment and was using the facial covering for the physical safety of the person while performing an act or activity connected with the profession or employment; (b) the person was using the facial covering for religious reasons; or (c) the person was using the facial covering for a pre‑existing medical or health reason.” 221. The problem arising from the word “at” was alluded to by the Judges at [156] of the Judgment: “Fourth, the prohibition applies to any person while he or she is ‘at’ any unauthorized assembly, public meeting or public procession referred to in s 3(1)(b), (c) or (d). It is not clearly stated whether, to be caught by the prohibition, the person must be a participant in the relevant gathering, or whether it suffices for that person to be merely present at the gathering, eg a person who goes to the scene for the purpose of taking photographs, or giving first-aid to persons in need of help, or even a mere passer‑by who has stopped to observe the gathering. The wording of s 3(1) may be contrasted with (i) s 17(3) of the POO, which makes it an offence for any person who, without lawful authority or reasonable excuse, ‘knowingly takes or continues to take part in or forms or continues to form part of’ any unauthorized assembly, and (ii) s 18(3) of the POO, which makes it an offence for any person who ‘takes part in’ an unlawful assembly. It is uncertain whether the restrictions imposed by s 3(1)(b), (c) or (d) cover not only participants of the types of gathering referred to in those sub‑paragraphs, but also any person who is physically present (other than perhaps for a fleeting moment) at the gathering in question.” 222. In his submissions, Mr Yu did not shrink from the position that section 3(1)(b) of the PFCR (as indeed sections 3(1)(a), (c) and (d) also) is applicable when a person is present at the place where the gathering takes place. Counsel submitted that the test of taking part is inappropriate in this context because it has an element of knowingly associated or identified with the unlawful or unauthorized assembly, citing R v Wolfgramm [1978] 2 NZLR 184 and R v Cook & Ors (1994) 74 A Crim R 1. Further, the disorder occurred in the current public danger often arises from gatherings attended by a large number of people. Given the number of people present and the fluidity of the potentially explosive situations, it is more important to require all people taking off facial coverings to minimise the shielding and emboldening effect. This is against the background that one of the legitimate aims of section 3 is to deter the “more peaceful” protestors or bystanders from continuously giving tacit support to the continued challenge of law and order by those radical and violent protestors and shielding the latter from effective police actions and arrests. 223. In our judgment, the taking part in a gathering has to be examined at different stages. It is true that for unlawful assembly, taking part in it requires the proof of knowing association or participation. Apart from the authorities cited by Mr Yu, the same point was made in Secretary for Justice v Leung Kwok Wah [2012] 5 HKLRD 556 at [17] ‑ [22], citing the dicta of Macdougall VP in the earlier case of R v To Kwan Hang [1995] 1 HKCLR 251 at p254. 224. However, in the context of unauthorized assembly, as discussed at [193] above, other than cases involving violent or other reprehensible conducts on the part of some demonstrators posing serious and imminent risk to public order and safety which requires immediate actions on the part of the police, there should be an order issued and announced under section 17(3) of the POO for the stopping or dispersal of the gatherings before further conducts (by way of refusal or wilful neglect to comply) on the part of those at the assembly rendering the assembly unauthorized under section 17A(2)(c). An order of dispersal necessarily requires people at the assembly to leave the scene. If a person remains at the scene notwithstanding the order, he will be acting in defiance of the order. By virtue of section 17A(1)(a), all the persons in the vicinity who are aware of the stop and dispersal order should not remain at the scene. 225. As regards unauthorized assembly held without compliance with section 7 or 13, any person who is at such assembly would be participating in the unauthorized assembly. It is all the more so when such person does not leave the scene after the police announces that an order for stop and dispersal has been made under section 17(3). In this connection, the following observations made by Chantal Masse JCS in the Superior Court of Quebec in Villeneuve v Montreal (City of), 2016 QCCS 2888 at [477] and [478] on demonstration without notification as required by law are pertinent: “[477] Likewise, the fact that police officers may have stated to the representatives that they used the powers conferred on them by Regulation P-6 only when the demonstration involved risks to the safety of citizens does not make it possible to conclude that the provision is not justified. The evaluation of such ‘risks’ necessarily takes into consideration the fact that there is no itinerary. Furthermore, as the organisers and participants have placed themselves in an unlawful situation, they cannot expect that the demonstration cannot be dispersed if it is not peaceful. [478] Moreover, in the context of a demonstration that occurs unlawfully but following a decision to tolerate it, it is not to be expected that the same degree of violence or disturbance is required before it can be dispersed, compared to what would be necessary before ending a demonstration that was taking place legally.” 226. Subject to the requirement of awareness of such person of the unauthorized nature of the assembly and reasonable opportunity being given to those present to disperse and leave the scene, we do not see any fundamental objection to holding him liable for his continued presence at an unauthorized assembly. It is necessary to stress that the maintenance of the rule of law in Hong Kong should be the duty of all citizens who treasure the rule of law as a core value in our society. In the wake of an order for stop and dispersal of a gathering and direction being issued by police to leave the scene, a responsible law-abiding citizen should follow the direction instead of remaining there in defiance of such order and direction. Given the potential for rapid deterioration of the situation and the serious ramifications for the maintenance of law and order and safety of those at the scene in gatherings involving large number of persons, a person who refuses to disperse and leave an assembly after the order is publicly announced perpetuates a state of affairs which disrupts public order and gives rise to a serious threat of breach of the peace. By remaining at the assembly, even without the commission of further act of violence or threat of violence, such person perpetuates the worsening situation which can potentially escalate to serious violent confrontations and frustrates the statutory scheme under the POO for crowd control which is essential to the facilitation of peaceful demonstration and gathering. In short, those choose to remain at the gathering in defiance of an order to stop and dispersal actually participate in the unauthorized assembly. 227. In Secretary for Justice v Leung Kwok Wah, supra, the court observed that section 18 (which deals with unlawful assembly) has a pre-emptive effect and the design of the law is to put a stop to a deteriorating situation. The same observation can be made with regard to the provisions in sections 17 and 17A in relation to unauthorized assembly. The purpose of empowering the police to order a stop and dispersal of a gathering under section 17(3) is to prevent the escalation of confrontation and disorder when a gathering has given rise to imminent threats to the peaceful enjoyment of the right of demonstration and the rights and safety of other citizens. It is essential for effective crowd control and the due performance of the positive duty on the part of the police to facilitate lawful assemblies to take place peacefully (as explained in Leung Kwok Hung & Others v HKSAR, supra) that the police could bring a gathering to a halt when there are signs of such imminent threats. 228. The pre-emptive nature of these provisions is of particular significance in view of the unchallenged evidence on the worrying phenomenon recently witnessed in Hong Kong where the situations were often highly fluid (with peaceful demonstrations rapidly developed into unlawful riots with wanton and reckless violence causing serious damage to properties and even serious injuries to others). The evidence also shows that there were many instances where less radical protestors remaining at the scene to provide moral and actual support (in terms of shielding the identities of those violent protestors). Instead of condemnation of violent acts committed or the public disorder occasioned by the radical protestors, some other protestors provide assistance to the perpetuators of violent and destructive acts. 229. In this connection, it is noteworthy that in the case of Austin v United Kingdom (2012) 55 EHRR 14, the Grand Chamber of the Strasbourg Court had made these observations at [55] and [56] (omitting the footnotes): “55. … In connection with Article 11 of the Convention, the Court has held that interferences with the right of freedom of assembly are in principle justified for the prevention of disorder or crime and for the protection of the rights and freedoms of others where demonstrators engage in acts of violence. It has also held that, in certain well-defined circumstances, Articles 2 and 3 may imply positive obligations on the authorities to take preventive operational measures to protect individuals at risk of serious harm from the criminal acts of other individuals. When considering whether the domestic authorities have complied with such positive obligations, the Court has held that account must be taken of the difficulties involved in policing modern societies, the unpredictability of human conduct and the operational choices which must be made in terms of priorities and resources. 56. As the Court has previously stated, the police must be afforded a degree of discretion in taking operational decisions. Such decisions are almost always complicated and the police, who have access to information and intelligence not available to the general public, will usually be in the best position to make them. Moreover, even by 2001, advances in communications technology had made it possible to mobilise protesters rapidly and covertly on a hitherto unknown scale. Police forces in the contracting states face new challenges, perhaps unforeseen when the Convention was drafted, and have developed new policing techniques to deal with them, including containment or ‘kettling’. Article 5 cannot be interpreted in such a way as to make it impracticable for the police to fulfil their duties of maintaining order and protecting the public, provided that they comply with the underlying principle of Article 5, which is to protect the individual from arbitrariness.” 230. Austin v United Kingdom, supra, was a case about the “kettling” or containment of a group of people carried out by the police on public order grounds. The Strasbourg Court upheld the validity of such measure in the circumstances of that case and rejected submissions that such containment constituted deprivation of liberty. The Court also examined the implications on Article 2 of Protocol No 4 to the European Convention for Protection of Human Rights and Fundamental Freedoms which guarantees the right to liberty of movement. Article 2 §3 of the said Protocol permits restrictions to be placed on the right to liberty of movement where necessary, inter alia, for the maintenance of public order, the prevention of crime or the protection of the rights and freedoms of others. The above observations on maintenance of public order and protection of the rights of others were made in such context. 231. As discussed earlier, there are legitimate public order concerns for unauthorized assembly. In light of the public danger facing Hong Kong with peaceful assemblies or processions by large number of persons wearing masks having a propensity of being hijacked by radical protestors and the rapidity of the worsening of situations, these public order concerns should be accorded with greater weight. Against such background, a deliberate defiance of the regulations against unauthorized assembly which are designed for the maintenance of public order and safety and protection of the rights of others is no less reprehensible than conducts which can be regarded as unlawful assembly. 232. On the other hand, we do not regard bystanders or passer-by as people “at the assembly” notwithstanding that they happened to be at the scene when the assembly becomes an unlawful assembly. Adopting a purposive construction and giving due regard to the principle of legality discussed at R v Home Secretary Ex p Simms [2000] 2 AC 115 at p131 as applied to the basic right of freedom of movement, we are of the view that such bystanders or passer-by cannot be regarded as being “at the assembly” when they are not part of the assembly in the first place nor do they take any steps to join the assembly at some stage. For such persons, a section 17(3) order cannot be directed against them. A person must have joined an assembly at some point in time before he could properly be regarded as “at the assembly” for the purpose of section 3(1) of the PFCR. The Chinese version of “at” in the relevant part of section 3(1) is “身處以下活動時”, not “身處以下活動發生的地點或場所”. In other words, there is a distinction between a person present at the scene or place where the assembly takes place and a person actually present at the assembly. 233. As we shall explain below, we are not minded to uphold the prohibition under section 3(1)(c) and (d) of the PFCR. Thus, there will be cases where a person lawfully wears facial coverings in a lawful public meeting or procession which is somehow hijacked by some people participating in the same gathering. For those persons, we are of the view that the defence of reasonable excuse under section 4 should be applicable when they are not aware of any order made under section 17(3) of the POO; or when they have not had a reasonable opportunity to leave the scene after becoming so aware. 234. As a matter of construction, the scenarios specified under section 4(3) are not exhaustive of the defence of reasonable excuse. By virtue of the potential adverse impact on the right of peaceful demonstration stemming from section 3 of the PFCR, the defence of reasonable excuse should be given a liberal interpretation. As held in R v Home Secretary Ex p Simms, supra, at p131, the principle of legality requires statutes to be interpreted insofar as possible in a manner compatible with fundamental rights. Thus, in cases where there is sufficient evidence to raise an issue of lawful use of facial covering (eg, in the participation of lawful assembly or gathering) and the court cannot be satisfied beyond reasonable doubt that a person remains at the scene (after reasonable opportunity to leave has been given) and persists in wearing such facial covering notwithstanding having made been aware of a section 17(3) order and the breach of the same by others leading to the unauthorized continuation of the assembly or gathering, a defence under section 4 can be made out. 235. In considering sections 3(1)(b), (c) and (d) together (the effect of which is to ban facial coverings from all public demonstrations, processions and gatherings), the Judges found the reasoning in SAS v France (2015) 60 EHRR 11, Yaker v France, Communication No 2747/2016 (17 July 2018) and Villeneuve v Montreal (City of), supra, to be applicable. Those authorities addressed legislations or regulations banning facial coverings in public or public gatherings without distinction as to the nature of the gatherings and their public order significance. For reasons further canvassed below, we agree with them that a total ban of facial coverings from all public demonstrations, processions and gatherings cannot pass the test of reasonable necessity even in the context of the public danger in which the PFCR is meant to address. 236. However, the imposition of such restriction in some defined circumstances can be justified provided that law enforcement officers are not given unduly broad discretionary power to give rise to potential for arbitrariness. In the Guidelines on Freedom of Peaceful Assembly published by a panel of human rights expert, cited by the court in Villeneuve v Montreal (City of), supra, at [496], there is the following commentary: “Restrictions imposed during an assembly: The role of the police or other law-enforcement personnel during an assembly will often be to enforce any prior restrictions imposed in writing by the regulatory body. No additional restrictions should be imposed by law-enforcement personnel unless absolutely necessary in light of demonstrably changed circumstances. On occasion, however, the situation on the ground may deteriorate (participants, for example might begin using of inciting violence), and the authorities may have to impose further measures to ensure that other relevant interests are adequately safeguarded. In the same way that reasons must be adduced to demonstrate the need for prior restrictions, any restrictions imposed in the course of an assembly must be just as rigorously justified. Mere suspicions will not suffice, and the reason must be both relevant and sufficient. In such circumstances, it will be appropriate for other civil authorities (such as an ombudsman’s office) to have an oversight role in relation to the policing operation, and law-enforcement personnel should be accountable to an independent body. Furthermore … unduly broad discretionary powers afforded to law-enforcement officials may breach the principle of legality, given the potential for arbitrariness…” 237. There is no reason why each limb of section 3(1) should not be considered separately. Considering the ban under section 3(1)(b) on its own, given the public order concerns discussed above in respect of unauthorized assembly amidst the fluidity in terms of the rapid deterioration of many large scale processions and demonstrations into violent riots with most of the rioters adopting the black bloc strategy to escape from responsibility for very serious criminal acts, we are of the view that such ban (as interpreted above) is no more than necessary to achieve the legitimate aims. As we have seen, most people who choose to remain at the assembly despite their awareness of the unauthorized nature of the gathering (either because of the issue and notice of an order to stop and disperse made under section 17(3) or warnings given by police that the gathering is unauthorized on account of other reasons) should also be aware that they are at risk of being prosecuted for offences under sections 17A(1)(a) and (3)(a). The need to deter people from wearing facial coverings to frustrate the legitimate aims of the statutory scheme under the POO applies with equal force in the context of unlawful assembly as well as unauthorized assembly. 238. In coming to the above conclusion, we take account of the above analysis as to the scope of section 3(1)(b) of the PFCR (including our construction of the defence of reasonable excuse under section 4) and its potential effects on the right of demonstration and freedom of expression and the background leading to the promulgation of the PFCR and the legitimate aims served by the same. We summarize below the main considerations leading to our conclusion on necessity: (1) The PFCR is intended to operate as an emergency measure when the public danger arising from the events since June 2019 continues to affect Hong Kong. Once the public danger subsides and the overall threat to law and order disappears, it is expected that CEIC would make an order under section 2(3) of the ERO to repeal the same. (2) A significant contributing factor to the current public danger is the impediment to effective law enforcement and efforts to restore law and order in the wake of large scale riots employing the black bloc tactics by the radical rioters backed up by less radical protestors who also wore facial coverings to shield the violent rioters from arrests and criminal sanction. (3) Insofar as section 3(1)(b) is concerned, it only bars the use of facial coverings at unauthorized assemblies. This measure should be assessed against the public danger experienced in Hong Kong from June 2019 onwards. It does not bar the exercise of the right of demonstration or freedom of expression generally. In other words, peaceful demonstration and assembly could still be held and the participants could adopt the wearing of facial coverings as a form of protest provided that the gatherings are authorized and held in accordance with the conditions imposed. (4) The liberty to wear facial coverings is only curtailed when a person remains at an unauthorized assembly and an order to stop and disperse had been made by a police officer under section 17(3) of the POO or in circumstances where violent or other reprehensible conducts on the part of some demonstrators posing serious and imminent risk to public order and safety which requires immediate actions on the part of police. As analysed above, a section 17(3) order could only be made when there is serious public order and safety concerns or immediate need to protect the rights and freedom of other citizens. The enforcement action itself would be subject to proportionality test. (5) Further, the defence of reasonable excuse under section 4 of the PFCR as construed above provides adequate protection to innocent protestors who are unaware of the section 17(3) order or have no reasonable opportunity to leave the scene. (6) As explained above, the refusal to disperse from an unauthorized assembly after a person at the assembly has been notified of the assembly being unauthorized or an order to stop and disperse under section 17(3) being made by itself is an offence. In other words, the persons participating in the unauthorized assemblies are taking part in an unlawful act, liable to be arrested and subject to potential penalty of imprisonment. (7) Even assuming such persons can still be regarded as exercising the right of peaceful demonstration, the wearing of facial coverings for demonstration under such circumstances is not at the core of such right. (8) Given the circumstances under which a section 17(3) order would be issued and the underlying public order significance for the POO regime to regulate and facilitate public processions and gatherings effectively, any non-compliance with such regulations by way of participating in an unauthorized assembly gives rise to a substantial risk of breach of the peace. This is particularly so in light of the propensity of such gatherings being hijacked by some radical protestors turning them into violent confrontations and the adoption of the black bloc tactics by such protestors in the current climate of public danger. (9) The wearing of facial coverings at unauthorized assemblies substantially impedes the effective law enforcement in terms of the identification and arrest of the radical and violent protestors who have crossed the line, which was the task of the police as highlighted in the judgment of Ribeiro PJ in HKSAR v Chow Nok Hang, supra. The inability of the police to pinpoint in a timely manner such radical and violent protestors and to segregate them from other demonstrators by arrests seriously impairs the power of the police to defuse potentially explosive situations and to exercise effective crowd control. The police became seriously handicapped in the performance of their duty to facilitate peaceful demonstration. (10) Such handicap also put the right of peaceful demonstration at serious risk because the Commissioner must take account of the practical difficulties stemming from mass facial coverings in crowd control and prevention of hijacking of lawful assembly or procession in deciding if objection should be made and the conditions to be imposed under sections 11 and 15 of the POO. At the same time, the timely and effective identification and segregation of violent protestors from the other peaceful demonstrators can avoid unnecessary police actions against the latter by virtue of mistaken identifications, thus facilitating the continuation of peaceful demonstrations or processions. (11) In this connection, the observations made by Poon JA (as the Chief Judge then was) in Secretary for Justice v Wong Chi Fung [2018] 2 HKLRD 699 at [118] are pertinent, in particular: “… For the general public, preserving public order helps create a safe and stable social environment to enable individuals to exercise their rights (including human rights of which the freedom of assembly and expression is one), express their views and pursue their goals. In fact, the above‑mentioned rights themselves will be lost in situation of anarchy if public order is not preserved. That is exactly the rationale underlying Article 17 of the Hong Kong Bill of Rights in only safeguarding peaceful assembly: the legal protection of the right of assembly is effective only in a society where public order is preserved …” (12) In light of the above analysis on the operation of the POO regime in respect of unauthorized assembly in conjunction with the common law principles as discussed at [180] ‑ [184], [188] ‑ [202] and [212] ‑ [215] above, there are sufficient safeguards to prevent arbitrariness in the exercise of discretion by the police and the courts in Hong Kong can provide adequate redress in the event of abuses. 239. In such context, SAS v France, supra, and Yaker v France, supra, (concerning legislation which prohibits the wearing of clothing that is designed to conceal the face in public places all the time) are of little relevance. As evidenced by the reasoning set out in those cases, the finding of disproportionality in the cases stems from the absolute ban instead of limiting it to situations where there is a general threat to public safety (see [139] in SAS v France and [8.7] in Yaker v France). These cases did not examine the matter in the context of public order and safety where there had been riots emerging repeatedly from mass demonstrations leading to the state of public danger which the PFCR is meant to tackle. 240. We are also satisfied that on a systemic level the prohibition under section 3(1)(b) of the PFCR strikes a fair balance between the societal benefits pursued by the restriction and the inroads made to the rights of the individual subject to the same. 241. We would therefore reverse the Judges’ holding with regard to the proportionality of section 3(1)(b). Accordingly, we shall set aside the declaration of unconstitutionality in respect of this subsection. F2.3 Sections 3(1)(c) and (d) 242. The position is different in relation to sections 3(1)(c) and (d) of the PFCR. These provisions prohibit the wearing of facial coverings at either a lawful public meeting that takes place under section 7(1) of the POO or a lawful public procession that takes place under section 13(1) of the POO. The holding of such meeting or procession would have been notified to the Commissioner and also not prohibited by the Commissioner. For public procession, the requirements of section 15 were complied with. One of the requirements is that good order and public safety of the procession is maintained. 243. Thus, so long as those meeting and procession proceed in accordance with sections 7(1) and 13(1), there cannot be any serious public order or safety issues which warrant additional restrictions being placed on the same by way of prohibition to wear facial coverings. As we have seen, when such meeting or procession were hijacked by protestors with violent or disorderly disposition, there is ample power on the part of the police under the POO regime to issue an order under section 17(3) including an order to stop and disperse. Disobedience to such an order would turn the meeting or procession into an unauthorized assembly. 244. On the other hand, if the meeting and procession remain peaceful and orderly, it is difficult to see the justification for imposing a restriction on the freedom of demonstration by way of prohibition of wearing facial coverings. 245. Mr Yu submitted that such prohibition is necessary because demonstrators wearing facial coverings would provide a shield to the radical violent protestors and facilitate the black bloc tactics which frustrate the effective policing when violent or criminal acts were committed. 246. However, as we have analysed above, a peaceful demonstration would have already degenerated into an unauthorized assembly or unlawful assembly before actual violence begins. For those fluid situations where such degeneration occurs rapidly, there is still sufficient powers under the POO regime to regulate the same in a proportionate manner. Thus, we have highlighted that there are cases where a lawful assembly can become an unauthorized one without a section 17(3) order when violent or other reprehensible conducts on the part of some demonstrators pose serious and imminent risk to public order and safety which requires immediate actions on the part of the police. For those scenarios, an offence under section 17A(2)(a) and (3) can be committed without an order made under section 17(3). 247. Even taking account of the propensity of mass demonstrations turned violent in the recent turmoil, we are not persuaded that it is necessary to extend the prohibitions under section 3(1) beyond the situations stipulated under subsections (a) and (b). 248. Thus, in respect of the prohibitions in sections 3(1)(c) and (d), we respectfully agree with the Judges that they cannot satisfy the proportionality test. F3. Section 5 249. To recap, section 5 of the PFCR confers power on police officer to require a person using facial covering in a public place to remove such covering to enable the officer to verify the identity of the person. If that person fails to comply with the request, the police officer can remove the covering. It further provides that the failure to comply with the request is an offence and liable to a maximum sentence of imprisonment for six months. 250. The Judges held that the provisions in section 5 engage freedom of the person and the right to liberty under BL28 and BOR5. Freedom of expression, freedom of movement and right to privacy were also relied upon and the Judges regarded that the proportionality analysis would be the same. Though they accepted that section 5 serves legitimate aims of law enforcement, investigation and prosecution and it is rationally connected with such aims, the Judges held that in light of the width of the measure (as explained at [185] of the Judgment), it exceeds what is reasonably necessary to achieve the aims even in the prevailing turbulent circumstances in Hong Kong. They also held that section 5 fails to strike a reasonable balance between the societal benefits derived from the measure and the inroads made to the protected rights. 251. By LKH-NoCA, LKH challenged the finding of the Judges that law enforcement, investigation and prosecution are legitimate aims. He also challenged the conclusion that section 5 is rationally connected with such aims. 252. The main argument of Mr Pun regarding the first challenge is that those aims were too broadly stated so that it could not afford a framework for evaluating the importance of such objectives. He also submitted that for an aim to be legitimate, a genuine need must be established. 253. With respect, we cannot accept these submissions. As demonstrated by the Judgment, the Judges were able to apply the proportionality analysis by reference to these legitimate aims. We do not see any difficulties in the evaluation on proportionality of the measures against these objectives. On the evidence, there is a genuine and pressing need to facilitate effective policing to counter the black bloc tactics. We also agree with the Judges that section 5 is rationally connected with these objectives for the reasons given at [183] and [184] of the Judgment. 254. The other points raised by Mr Pun had sufficiently been addressed by the Judges at [174] ‑ [179] of the Judgment. 255. We would dismiss the cross-appeal in these respects. 256. Coming back to the holding on reasonable necessity, Mr Yu submitted that the Judges erred in holding that section 5 is disproportionate when it involves minimal intrusion to an individual’s right but serves important functions of removing practical difficulties in identification and early intervention in times of public danger. 257. Counsel also submitted that section 5 of the PFCR should be construed together with section 54(1)(a) of the Police Force Ordinance (Cap 232) (“the PFO”) (which empowers a police officer to stop and demand proof of identity of a person acting in a suspicious manner in any street or other public place), section 49(1) of the POO (which empowers a police officer to require any person to produce proof of identity for inspection if he reasonably believes that it is necessary for prevention, detection or investigation of any offence) and section 17C(2) of the Immigration Ordinance (Cap 115) (“the IO”) (which empowers police officer to demand a person to produce proof of identity for inspection). 258. Applying the principle discussed in Medical Council of Hong Kong v Chow Siu Shek, supra, Mr Yu submitted that we should take all these statutory provisions together in construing the limits of the power under section 5. Alternatively, as a fallback position, Mr Yu submitted that the Court can declare that the power under section 5 can only be exercised in conjunction with these other statutory powers. 259. Mr Yu also relied on the decision of the European Commission in Reyntjens v Belgium Application No 16810/90 to support his argument that the restriction on personal liberty under section 5 is minimal. In that case, the European Commission held that a routine identity check did not violate the liberty of the person protected under article 5 and the right to respect for private life under article 8 of the European Convention for the Protection of Human Rights and Fundamental Freedoms. 260. The decision in Reyntjens v Belgium, supra, was made in 1992. The Commission did not regard a check of identity card by the police as an interference in a person’s private life. The identity card carried information as to the bearer’s name, sex, date and place of birth, main address, and his spouse’s name. The Commission took the view (somewhat surprising, if we may say so, in light of the more up-dated jurisprudence) that such identity card did not contain information relating to private life. 261. In DPP v Avery [2002] 1 Cr App R 31, the Divisional Court held that the exercise of the powers conferred by section 60 of the Criminal Justice and Public Order Act 1994 (which provided a police constable upon authorisation being given by an inspector under certain conditions could require any person to remove any item which were worn for the purpose of concealing his identity and to seize such item) was an interference with the liberty of the subject (see [17] and [18] of the judgment). The court also held that the exercise of those powers did not involve the power of search, see [16] and [24]. 262. The interference was found to be justified for the reasons set out at [18]. An important consideration in that case was that the power under section 60 arose only in anticipation of violence and after a decision of a senior police officer was made to authorize the conferment of such power on the specific occasion. At the same time, Newman J also acknowledged at [17] that wearing of apparel to the head and face can be required by custom or religion and wearing of mask can be a potent means of demonstrating in a lawful manner. 263. The Judges in the present case contrasted that provision with section 5 under [185] ‑ [187] and footnote 60 of the Judgment. They, in our view quite rightly, took account of the width of the power under section 5 in assessing the proportionality of the same. 264. The modern European approach was discussed by the European Court of Human Rights in Gillan and Quinton v United Kingdom Application No 4158/05 (which were reiterated in the more recent case of Beghal v United Kingdom Application No 4755/16). Gillan and Quinton v United Kingdom, supra,concerned the stop and search power under section 44 of the Terrorism Act 2000. On the interference with the right to liberty of a person, the Strasbourg Court had these observations at [56] and [57] (omitting the footnotes): “56. The Court recalls that Article 5 § 1 is not concerned with mere restrictions on liberty of movement; such restrictions are governed by Article 2 of Protocol No. 4, which has not been ratified by the United Kingdom. In order to determine whether someone has been ‘deprived of his liberty’ within the meaning of Article 5, the starting point must be his concrete situation and account must be taken of a whole range of criteria such as the type, duration, effects and manner of implementation of the measure in question. The difference between deprivation of and restriction upon liberty is nonetheless merely one of degree or intensity, and not one of nature or substance. Although the process of classification into one or other of these categories sometimes proves to be no easy task in that some borderline cases are a matter of pure opinion, the Court cannot avoid making the selection upon which the applicability or inapplicability of Article 5 depends. 57. The Court observes that although the length of time during which each applicant was stopped and search did not in either case exceed 30 minutes, during this period the applicants were entirely deprived of any freedom of movement. They were obliged to remain where they were and submit to the search and if they had refused they would have been liable to arrest, detention at a police station and criminal charges. This element of coercion is indicative of a deprivation of liberty within the meaning of Article 5 § 1. In the event, however, the Court is not required finally to determine this question in the light of its findings below in connection with Article 8 of the Convention.” 265. On the right to respect for private life, the court discussed the relevant principles and approach at [61] ‑ [63] (omitting the footnotes): “61. As the Court has had previous occasion to remark, the concept of ‘private life’ is a broad term not susceptible to exhaustive definition. It covers the physical and psychological integrity of a person. The notion of personal autonomy is an important principle underlying the interpretation of its guarantees. The Article also protects a right to identity and personal development, and the right to establish relationships with other human beings and the outside world. It may include activities of a professional or business nature. There is, therefore, a zone of interaction of a person with others, even in a public context, which may fall within the scope of ‘private life’. There are a number of elements relevant to a consideration of whether a person’s private life is concerned in measures effected outside a person’s home or private premises. In this connection, a person’s reasonable expectations as to privacy may be a significant, though not necessarily conclusive, factor. In Foka, at [85], where the applicant was subjected to a forced search of her bag by border guards, the Court held that ‘any search effected by the authorities on a person interferes with his or her private life.’ 62. Turning to the facts of the present case, the Court notes that sections 44-47 of the 2000 Act permit a uniformed police officer to stop any person within the geographical area covered by the authorisation and physically search the person and anything carried by him or her. The police officer may request the individual to remove headgear, footwear, outer clothing and gloves. Paragraph 3.5 of the related Code of Practice further clarifies that the police officer may place his or her hand inside the searched person’s pockets, feel around and inside his or her collars, socks and shoes and search the person’s hair. The search takes place in public and failure to submit to it amounts to an offence punishable by imprisonment or a fine or both. In the domestic courts, although the House of Lords doubted whether Article 8 was applicable, since the intrusion did not reach a sufficient level of seriousness, the Metropolitan Police Commissioner conceded that the exercise of the power under section 44 amounted to an interference with the individual’s Article 8 rights and the Court of Appeal described it as, ‘an extremely wide power to intrude on the privacy of the members of the public’. 63. The Government argues that in certain circumstances a particularly intrusive search may amount to an interference with an individual’s Article 8 rights, as may a search which involves perusing an address book or diary or correspondence, but that a superficial search which does not involve the discovery of such items does not do so. The Court is unable to accept this view. Irrespective of whether in any particular case correspondence or diaries or other private documents are discovered and read or other intimate items are revealed in the search, the Court considers that the use of the coercive powers conferred by the legislation to require an individual to submit to a detailed search of his person, his clothing and his personal belongings amounts to a clear interference with the right to respect for private life. Although the search is undertaken in a public place, this does not mean that Article 8 is inapplicable. Indeed, in the Court’s view, the public nature of the search may, in certain cases, compound the seriousness of the interference because of an element of humiliation and embarrassment. Items such as bags, wallets, notebooks and diaries may, moreover, contain personal information which the owner may feel uncomfortable about having exposed to the view of his companions or the wider public.” 266. Coming back to section 5 of the PFCR, even though it does not involve a power of search, the exercise of the power would inevitably impose a temporary restriction on the liberty of the person who is subject to the request. Further, the section imposes a coercive sanction to back up a request for removal of facial coverings. Another facet of section 5 is the empowerment of a police officer to remove a facial covering worn by a person if that person fails to comply with a request for such removal. Though the extent of bodily contact for removal of facial coverings may not be as substantial as a detailed search of a person, it nonetheless authorizes bodily contacts which, from the point of view of a person subject to such treatment, is unwelcome and may even be offensive. Acts and conducts on the part of a person to deflect or resist such removal could be taken as obstruction to due execution of duties and thus bring serious consequences upon such person. 267. We are therefore unable to agree that the interference occasioned by an exercise of section 5 power on the liberty of the person and the right to privacy is minimal. 268. Subject to the situations addressed by sections 3(1)(a) and (b), it is not an offence to wear facial coverings in public place. As we have already held, wearing facial coverings during a peaceful and lawful demonstrations would not be an offence. Yet section 5 confers power to the police to make a request to remove facial coverings in an entirely uneventful situation. 269. As noted by the Judges at [181] of the Judgment, there are other statutory provisions in our statute books empowering police officers to demand proof of identity for inspection under specified circumstances. We alluded to these provisions at [257] above. As explained at [182], there are good public order and immigration control justifications for such powers. 270. In our view, these statutory powers are already sufficient to address the law enforcement objectives, particularly when sections 3(1)(a) and (b) of the PFCR are in place and they could be relied upon in conjunction with the exercise of the power under section 54(1)(a) of the PFO or section 49(1) of the POO. Insofar as necessary, these provisions should be read with section 40(1) of the Interpretation and General Clauses Ordinance (Cap 1) which states: “Where any Ordinance confers upon any person power to do or enforce the doing of any act or thing, all such powers shall be deemed to be also conferred as are reasonably necessary to enable the person to do or enforce the doing of the act or thing.” 271. In order to have meaningful inspection of a proof of identity under these provisions, the police officer must be able to compare the photograph on the identity document with the face of the subject person. Thus, if the person wears facial covering, he must remove the same to facilitate the inspection by the police officer. Failure to remove the facial covering under such circumstances is a failure comply with the demand and it would be an offence under section 49(1) of the POO. 272. Any challenges to such implied power should be laid to rest after this judgment. 273. In any event, there is no justification for such wide power as laid down in section 5 of the PFCR. In this connection we are in full agreement with the Judges’ observations at [189] of the Judgment. We cannot accept the submissions of Mr Yu that section 5 could be construed in a manner to confine its application to circumstances set out in section 54(1)(a) of the PFO or section 49(1) of the POO or section 17C(2) of the IO. There is simply nothing in section 5 itself to suggest that these provisions should be read together. The mere references to “prevent identification” under section 5(1) and verification of identity under section 5(2)(a) do not provide the basis for holding that section 5 should only be applicable in circumstances where section 54(1)(a) of the PFO or section 49(1) of the POO or section 17C(2) of the IO are applicable. 274. In Medical Council of Hong Kong v Chow Siu Shek, supra, Bokhary PJ referred to the following principle of construction at p156B: “It is well-established that the context in which a statute is to be interpreted includes other statutes in pari materia i.e. other statutes dealing with comparable matters.” 275. Even if we were to accept that section 54(1)(a) of the PFO or section 49(1) of the POO can be regarded as dealing with comparable matters, those statutory provisions have different scopes of operation. We cannot distil any common limitations on the exercise of the powers under these statutory provisions and transpose the same to confine the scope of section 5. 276. Bearing in mind the objective for section 17C(2) of the IO as explained by the Judges at [181(3)] of the Judgment, it is plain that it cannot be regarded as dealing with a comparable matter as the PFCR the genesis of which is the public danger brought about by the recent turmoil. In this connection, we are aware that in R v Fung Chi Wood [1991] 1 HKLR 654 Bewley J held that the power under section 17C could be used for purposes unrelated to immigration control. That case was apparently not cited to the Judges. However, that was a decision before the HKBORO came into effect on 8 June 1991. Further, in light of the modern approach to the construction of the scope of statutory power by reference to its legislative purpose as discussed recently in Chee Fei Ming v Director of Food and Environmental Hygiene, supra, it is doubtful if that decision could remain good law today. As there is no appeal against the holding of the Judges in this respect and no submission has been advanced in that light, we are not going to resolve finally the applicable scope of section 17C(2). Suffice to say that we would proceed on the basis that there is no challenge before us to the view of the Judges. 277. Another general principle of construction discussed by Bokhary PJ in Medical Council of Hong Kong v Chow Siu Shek, supra, can be found at p154B: “When the true position under a statute is to be ascertained by interpretation, it is necessary to read all of the relevant provisions together and in the context of the whole statute as a purposive unity in its appropriate legal and social setting.” 278. The obvious contextual background for section 5 is the other provisions in the PFCR itself. However, it is plain that the CEIC did not intend to confine the operation of section 5 to cases where an offence has been committed under section 3. Otherwise, section 5(1) would be drafted differently. And there is actually no need to have section 5 to cover situations where the person is arrested or stopped for enquiry in respect of suspected commission of an offence under section 3. Such need has already been provided for under section 54(2)(a) of the PFO which provides: “If a police officer finds any person in any street or other public place, or on board any vessel, or in any conveyance, at any hour of the day or night, whom he reasonably suspects of having committed or of being about to commit or of intending to commit any offence, it shall be lawful for the police officer to stop the person for the purpose of demanding that he produce proof of his identity for inspection by the police officer.” 279. As explained above, the power under section 54(2)(a) impliedly confers the power to demand the subject person to remove facial coverings for verification of identity. Thus, there is no need for a further power to be conferred by way of section 5 to cater for such situation. We do not see any need for this Court to apply a remedial interpretation to section 5 by reading down its scope of operation. 280. As regard the submission of Mr Yu that section 5 can facilitate early intervention against the background of public danger, for the reasons given at [261] ‑ [268] above, we do not find it proportionate to confer a wide power to the police to demand removal of facial coverings to verify the identity of a person in the absence of any ground for public order concern against the person in question. 281. Further, the conferment of such wide and unqualified power is against the concept of law providing proper safeguard against arbitrary interference of fundamental rights. The relevant principle was reiterated by the European Court of Human Rights in Beghal v United Kingdom Application No 4755/16 at [88]: “For domestic law to meet these requirements it must afford a measure of legal protection against arbitrary interferences by public authorities with the rights safeguarded by the Convention. In matters affecting fundamental rights it would be contrary to the rule of law, one of the basic principles of a democratic society enshrined in the Convention, for a legal discretion granted to the executive to be expressed in terms of an unfettered power. Consequently, the law must indicate with sufficient clarity the scope of any such discretion conferred on the competent authorities and the manner of its exercise (Rotaru v. Romania [GC], no. 28341/95, § 55, ECHR 2000-V; Hasan and Chaush v. Bulgaria [GC], no. 30985/96, § 4, ECHR 2000-XI; Maestri v. Italy [GC], no. 39748/98, § 30, ECHR 2004‑I; see also, amongst other examples, Gillan and Quinton v. the United Kingdom, no. 4158/05, § 77, ECHR 2010 (extracts)). The level of precision required of domestic legislation – which cannot in any case provide for every eventuality – depends to a considerable degree on the content of the instrument in question, the field it is designed to cover and the number and status of those to whom it is addressed (see, for example, Hashman and Harrup v. the United Kingdom [GC], no. 25594/94, § 31, ECHR 1999‑VIII; S. and Marper, cited above, § 96; Gillan and Quinton, cited above, § 77; and Ivashchenko v. Russia, no. 61064/10, § 73, 13 February 2018).” 282. For the above reasons, we would dismiss the appeal by the CEIC in respect of section 5 of the PFCR. G. OTHER GROUNDS 283. In the KWH‑NoCA and KWH-RN, they contend that the Judges were wrong in rejecting Grounds 2, 3 and 4. They are effectively seeking to advance the arguments made under these grounds again in these appeals. 284. In the LKH‑NoCA, he contends that the Judges were wrong to reject Ground 3 and to hold under Ground 5B that the PFCR serves legitimate aims and the measures thereunder are rationally connected to those aims. Effectively, he is seeking to reargue those submissions that have been rejected in the court below. 285. We will deal with them in turn as follows[52]. G1. Ground 2 286. Mr Chan’s submissions in support of this ground are premised fundamentally on the provisions under sections 3 and 5 of the HKBORO. It is therefore useful to set them out first. 287. Section 5 of the HKBORO (which incorporates and is identical to ICCPR4) provides: “(1) In time of public emergency which threatens the life of the nation and the existence of which is officially proclaimed, measures may be taken derogating from the Bill of Rights to the extent strictly required by the exigencies of the situation, but these measures shall be taken in accordance with law. (2) No measure shall be taken under subsection (1) that— (a) is inconsistent with any obligation under international law that applies to Hong Kong (other than an obligation under the International Covenant on Civil and Political Rights); (b) involves discrimination solely on the ground of race, colour, sex, language, religion or social origin; or (c) derogates from articles 2, 3, 4(1) and (2), 7, 12, 13 and 15.” (emphasis supplied) 288. Section 3 provides: “(1) All pre‑existing legislation that admits of a construction consistent with this Ordinance shall be given such a construction. (2) All pre‑existing legislation that does not admit of a construction consistent with this Ordinance is, to the extent of the inconsistency, repealed.” 289. Given these provisions, Mr Chan essentially submitted: (1) Section 5 of the HKBORO prohibits any derogation from the fundamental rights provided in the ICCPR save in the circumstances specified therein and save except for non‑derogable rights. The circumstances specified are detailed and elaborate[53]. (2) On the other hand, the CEIC can make any regulations of the widest scope and nature under the ERO under any occasion of emergency or public danger. The regulations so made can be of such measures that suspend or limit the protected rights under the BOR (and ICCPR). In other words, the regulations made by the CEIC under the ERO can derogate from the protected fundamental rights. (3) However, the occasion of emergency or public danger provided under the ERO are not defined and undisputedly could cover situations other than “public emergency” as provided under section 5 of the HKBORO. In the circumstances, the ERO permits or allows derogation from the protected fundamental rights other than in the circumstances as specified and in accordance with section 5 of the HKBORO (and ICCPR4). (4) In this respect, Mr Chan has further pointed out that section 2(4) of the ERO allows regulations to suspend or repeal existing laws[54], which must include the entire HKBORO or its section 5 specifically. In the premises, the ERO also permits or allows derogation from even non‑derogable rights. (5) The ERO is therefore inconsistent with the provisions in the HKBORO, and thus was already impliedly repealed under section 3(2) of the HKBORO before 1 July 1997. (6) Alternatively, even if the ERO was not repealed before 1 July 1997, it is in any event impliedly repealed or has become unlawful after 1 July 1997 as it is inconsistent with BL39 (which constitutionalised ICCPR as it is incorporated into the domestic law under the HKBORO). 290. Mr Chan further submitted that the Judges were therefore wrong to reject this ground thinking that KWH has somehow conflated the concepts of derogation from the BOR itself and restriction of non-absolute rights under and in compliance with the BOR. He emphasised that his above arguments do not involve any conflation of concepts. His fundamental premise is that the ERO itself permits or allows such derogation to be implemented. 291. With the greatest respect, we think the Judges were clearly right in rejecting this ground. We hope we could be forgiven for not dealing with every aspect of Mr Chan’s very comprehensive and detailed submissions, as we think the proper analysis and approach to this question is this. 292. First, as pointed out by Lam VP at the hearing, it is provided expressly at section 3(1) of the HKBORO that “[a]ll pre‑existing legislation that admits of a construction consistent with this Ordinance shall be given such a construction”. Hence, one must first seek to construe the ERO in a way to be consistent with the HKBORO. It is only when this is not possible that one should resort to section 3(2) to find that the ERO is repealed. 293. However, as rightly pointed out by the Judges at [108] of the Judgment, such a construction is not only permissible but a proper one (given section 3(1) of the HKBORO): the ERO should simply be read subject to section 5 of the HKBORO (which has been fairly accepted by Mr Yu to be the case both before the Judges as well as before this Court). 294. In other words, as explained by the Judges, the ERO should be construed as: (1) In times of a public emergency officially proclaimed and in accordance with the other requirements of section 5 of the HKBORO, measures may be adopted under the ERO which derogate from the BOR (even so, excepting the specified non‑derogable provisions and discrimination on the prohibited grounds). Subject to the conditions of section 5 of the HKBORO (including that the derogations are limited to those strictly required by the exigencies of the situation), this may have the effect of temporarily suspending the relevant human rights norms. (2) In other situations, measures adopted under the ERO may not derogate from the BOR, which means that if any such measure has the effect of restricting fundamental rights, then like any other restriction in normal times, it has to satisfy the dual requirements that the restriction is prescribed by law and meets the proportionality test: Leung Kwok Hung & Others v HKSAR, supra, at [16] - [17]. 295. Mr Chan however argued that such a construction is inappropriate and unworkable, since the court will have to read too much into the provisions of the ERO to meet the elaborate and detailed scheme and safeguards provided under section 5 of the HKBORO (and thus also ICCPR4) to permit derogation from the protected rights. For example, Mr Chan said in relation to the requirement of a proclamation of a “public emergency”, such a construction would further require to be read into the ERO details such as to how the proclamation is to be made, what procedures are to be followed, and how many days of notice should be given to the public, etc. He therefore said the proposed construction is not only unworkable, but would also create uncertainties. 296. We are unable to accept this submission. All that section 5 of the HKBORO (and ICCPR4) has provided for are those criteria that have been set out therein before a derogation from the protected rights is permitted. The ERO is only required to be read to be subject to those broad criteria to make it consistent with section 5. All the so-called details that Mr Chan has suggested are simply not provided for by section 5 (or ICCPR4). They are therefore not specific requirements or criteria prescribed by section 5 of the HKBORO that need to be read into the ERO to make it compatible. 297. This conveniently takes us to the next reason why we find that the Judges were correct in rejecting this ground. 298. As observed by the Judges, although the ERO gives the CEIC the potential power to enact regulations that may result in derogation from the protected rights if she finds it necessary to do so, it is plain from the provisions of the ERO that they do not by themselves provide for any derogation from the protected rights as submitted by Mr Chan. 299. In the premises, the fundamental basis for the systemic challenge made against the ERO to say it is incompatible with BOR5 is incorrect in the first place. 300. If in the circumstances where a regulation enacted by the CEIC under the ERO does have the effect of derogating from any of the protected fundamental rights, that regulation will have to be subject to the dual tests of “prescribed by law” and “proportionality” as provided under section 5 of the HKBOR[55]. This however does not support a systemic challenge against the ERO to say that it is incompatible with section 5 of the HKBOR. 301. This is where and why the Judges were rightly of the view that KWH’s contentions under Ground 2 conflate with the concepts of derogation from right and the restriction of non-absolute rights under and in compliance with the BOR. 302. The ERO is therefore not impliedly repealed before 1 July 1997 for being incompatible with section 5 of the HKBORO. 303. Mr Chan’s alternative contention that the ERO should in any event be repealed after 1 July 1997 (as it is inconsistent with BL39 which incorporates ICCPR4 through the HKBORO) is in all material aspects based on the same arguments considered above. For the same reasons, we will reject it. 304. The Judges were correct in rejecting Ground 2. We would therefore dismiss KWH’s ground of cross-appeal in this respect. G2. Ground 3 305. Under this ground, both Mr Chan and Mr Pun (for LKH) [56] submitted that section 2(1) of the ERO is unconstitutional as it confers practically unlimited power on the CEIC to make regulations that are capable of severely restricting fundamental rights, but the scope of the power and the manner of its exercise is so wide and undefined that it falls foul of the legal certainty test under the prescribed by law requirement. 306. Mr Chan’s submissions can be summarised as follows: (1) The section 2(1) power can be invoked on a most vague ground of a self proclaimed “public danger” (which is not defined in the ERO). In other words, whenever and however. (2) There is also no restriction on the scope of regulations to be made. It is contemplated that these regulations could impose restrictions on liberty of the persons, the right to a fair hearing, freedom of communication, right to private property, and conscription for services. (3) In the premises, it is not possible to reasonably foresee, even with proper legal advice, the limits or the possible consequences of the existence of these powers. There are therefore no independent safeguards against abuses or excesses. 307. Other than relying on Mr Chan’s submissions, Mr Pun further emphasised that the ERO falls foul of the prescribed by law requirement as: (1) Given the widest possible scope of regulations that the CEIC is empowered to make under it, section 2(1) of the ERO confers an “unfettered power” on the CEIC and constitutes “arbitrary interference” in Hong Kong people’s rights and freedom. (2) The unfettered nature of the power (and hence the arbitraries of the interference that would result from it) is underlined by the fact that the ERO does not set out any “guidance” on the way in which the CEIC exercises the power, and only provides for a “purely subjective and inherently vague and uncertain criterion” (ie, the undefined “public danger” ground consideration as a basis for the CEIC to invoke the power). 308. It can be seen from the above that KWH and LKH’s fundamental complaint under this ground is that section 2(1) of the ERO does not meet the prescribed by law requirement as it is uncertain in law as to the scope of the power conferred on the CEIC and the manner in which she should exercise that power. 309. The Judges rejected this ground for the main reasons set out at [112] - [115] of the Judgment as follows: “112. The argument of the applicants in both HCAL 2945 and HCAL 2949 is this. The ERO contemplates and enables the making of regulations which can severely curtail fundamental rights. Section 2(1) is, however, couched in wide terms and lacks any express limit or guidance on the exercise of the power by the CEIC in making regulations that may affect fundamental rights. As such, s 2(1) of the ERO violates the principle of legal certainty mandated by Art 39 of the Basic Law. What the argument focuses on is the certainty (or the lack of it) in the regulation-making power in the ERO, rather than that in respect of any regulations made under the ERO. 113. We do not accept this argument. Legal certainty is not a notion existing in the abstract and in a vacuum. Art 39 of the Basic Law provides that the provisions of the ICCPR as applied to Hong Kong shall remain in force, and states that those rights and freedoms shall not be restricted unless as prescribed by law. As such, the ‘prescribed by law’ requirement applies to the restraints on the rights and freedoms of the individual. It is the ‘norms’ which purport directly to restrict the citizen’s freedom that must be sufficiently precise to enable the citizen to conduct himself accordingly. This is how the requirement has been applied in the decisions of the Court of Final Appeal. Thus in Shum Kwok Sher v HKSAR (2002) 5 HKCFAR 381 and Mo Yuk Ping v HKSAR (2007) 10 HKCFAR 386, the requirement was applied to the common law offence of misconduct in public office and conspiracy to defraud respectively. In Leung Kwok Hung & Others v HKSAR (2005) 8 HKCFAR 229, it is the statutory scheme in the POO restricting the right of assembly and procession that was called into question by reference to the requirement of ‘prescribed by law’. In Hysan Development Co Ltd v Town Planning Board (2016) 19 HKCFAR 372, the point made was that property rights are to be protected by clear and accessible laws. In HKSAR v Fong Kwok Shan Christine (2017) 20 HKCFAR 425, the question was whether a provision in the administrative instructions issued by the President of the LegCo to regulate behaviour within the precincts of the LegCo’s chamber was sufficiently certain. See also Hong Kong Television Network Ltd v Chief Executive in Council [2016] 2 HKLRD 1005 (CA), §90. 114. In contrast, the ERO does not itself purport to limit any fundamental rights. It does not lay down any norms that curtail any right of an individual. Although regulations enacted under it may purport to do so, it seems to us that it is those regulations, if and when enacted, that have to meet the principle of legal certainty, not the enabling Ordinance in itself which has no direct effect on any individual right or freedom. If such regulations are themselves laws of general application, accessible to all residents, and sufficiently well defined, then they cannot be said to fall foul of the requirement of accessibility and foreseeability which is the essence of the principle. 115. The ERO, as the source of power for making regulations, cannot be attacked on its own under the ‘prescribed by law’ requirement. This is not to say that it can never be a matter of concern that executive authorities are given ill-defined powers to make laws that may restrict fundamental rights, but this seems to us to raise the analytically separate and different point in relation to delegation of legislative power, which we have dealt with under Ground 1 above. In addition, the laws thus made will themselves have to possess the quality of accessibility and to afford sufficient safeguards against arbitrary application by indicating with sufficient clarity the scope of any discretionary power conferred.” (emphasis supplied) 310. Mr Chan and Mr Pun both submitted that the Judges had erred in their reasons. 311. With respect to Mr Chan and Mr Pun, we see no substance in this ground. 312. Most recently in Chee Fei Ming v Director of Food and Environmental Hygiene, supra, after reviewing a long line of authorities, this Court explained at [38] - [50] and [52] - [58] the applicable principles under the prescribed by law requirement. 313. Relevant for the present purpose concerning a complaint of lack of sufficient clarity as to the scope of the authority’s statutory power and the manner of its exercise, the principles can be summarised as thus: (1) The requirement stems from BL39(2)[57] and BOR16 and 17. It entails two limbs: (a) the requirement as to the accessibility of the law, and (b) the requirement of foreseeability. (2) The second limb requires that “a norm must be formulated with sufficient precision to enable a citizen to regulate his conduct so that he is able, with legal advice if necessary, to foresee the consequences which a course of action will entail”. This requirement is to be assessed by examining if there is sufficient clarity to protect an individual against arbitrary interference of his rights and freedoms. (3) The crucial question in such context is whether there is sufficient clarity as to the scope of the authority’s power and the manner of which its exercise and whether the law provides adequate safeguards against abuse. If these tests are met, there would be legal protection against arbitrary interference by the authority with the subject freedoms and rights in question. (4) At the same time, under the prescribed by law requirement, absolute precision or certainty is not achieved or required. It only requires that the “core” of the norm should be sufficiently formulated, and “the mere existence of debatable issues surrounding the settled core does not make the norm legally uncertain”. (5) For the purpose of determining what the “law” is for the prescribed by law doctrine, the court should adopt a “holistic approach”. In assessing in a particular case whether foreseeability requirement is satisfied, the court can take into account not only the statutory provision in question, but also unwritten law, including the common law. Under the holistic approach, the court will also examine how the law is actually administered, including the effectiveness of judicial supervision through judicial review. 314. Having regard to these principles, the Judges were clearly correct in rejecting this ground. 315. As explained by the Judges, the ERO itself does not provide for any restriction or limitation of the protected fundamental rights. This is implicitly (if not expressly) acknowledged by Mr Chan in his above submissions[58] (see the italicised words at [306] above). In any event, this must be plain from the fact that the ERO itself cannot be enforced or does not impose any penalty. 316. In the premises, given that it does not by itself restrict or limit any fundamental rights, the Judges were right to conclude at [113] - [115] of the Judgment that the ERO is not subject to the prescribed by law scrutiny as a basis of a systemic challenge. Rather, it is the actual regulation that has been enacted, and if it has the effect of restricting or limiting fundamental rights, that has to meet the prescribed by law requirement. As observed by the Judges, legal certainty “is not a notion existing in the abstract and in a vacuum” and “it is the ‘norms’ which purport directly to restrict the citizen’s freedom that must be sufficiently precise to enable the citizen to conduct himself accordingly”. 317. In this respect, Mr Yu was also correct to point out that KWH is wrong to say that “The differences between [the Judges] and [KWH] is whether a source empowering statute itself is susceptible to such a challenge [for lack of legal certainty]”. The Judges have not held as a general proposition that “a source empowering statute” can never been so challenged. Reading [115] of the Judgment in its proper context, the Judges’ ruling is directed specifically at the ERO where they rightly held that it does not purport to limit fundamental rights or restrict any freedom. 318. Mr Pun however argued that the Judges’ reasoning is wrong as the ERO even as a power conferring statute engages the prescribed by law requirement under BL39(2). 319. First, counsel contended that there is no principle of law that a discretionary power itself is not open to challenge as being in violation of fundamental rights merely because such power has not been actually exercised whether against a particular applicant or the public in general. 320. This submission is misconceived. BL39(2) provides expressly that Hong Kong residents’ rights and freedoms cannot be restricted unless prescribed by law. Hence, it is only when the individuals’ rights and freedoms are or are sought to be restricted that the prescribed by law requirement needs to be met. But as explained above, the CEIC’s exercise of discretionary power under the ERO to make a regulation is not an exercise of power that will necessarily violate any fundamental right. It is only when there is a regulation that has indeed been enacted under the ERO which provides for the restriction of fundamental rights that an individual’s relevant fundamental rights or freedoms are or are sought to be restricted. In such a case, the regulation itself has to meet the prescribed by law requirement and that would not be dependent upon whether or not the power under the regulation has in fact been exercised or not. 321. Second, Mr Pun cited three authorities to support his above contention. They are: Charles v Philips and Sealey (1967) 10 WIR 423, Herbert v Phillips and Sealey (1967) 10 WIR 435 and AG v Reynolds [1980] AC 637. 322. These authorities are distinguishable and do not help him. 323. In Charles, the Governor of St Christopher, Nevis and Anguilla made the Emergency Powers Regulations 1967 (“the Regulations”) pursuant to the Leeward Islands (Emergency Powers) Order in Council 1959 (“the Order in Council”). Charles was detained pursuant to the Regulation 3. He challenged the lawfulness of the Regulation 3. 324. Glasgow J refused the application. The Court of Appeal of West Indies Associated States allowed his appeal. However, the issue raised in the challenge that the court was concerned with is whether Regulation 3 was consistent with sections 3 and 14 of the Constitution itself. This is made clear by Lewis CJ (with whom other members of the court agreed) at p430A-B as follows: “The crucial question in this case is whether, having regard to ss 3 and 14 of the Constitution, the Governor was empowered to make reg 3 of the Emergency Powers Regulations 1967 and to order the detention of [Charles] thereunder. This calls for a close comparison of [the Order in Council] with the relevant provisions of the Constitution.” 325. The Court of Appeal held that the Regulations were invalid as (a) they were not in conformity with the Constitution, and (b) it was impossible to construe them to be in conformity with the Constitution even by way of modification, adaptation, qualification or exception. Thus, the case was not concerned with the doctrine of prescribed by law, but rather with a conflict between the empowering Order in Council and the Constitution. This cannot be an authority to support Mr Pun’s proposition that in relation to a prescribed by law analysis, the court should examine the ERO which by itself does not seek to limit or restrict any rights and freedoms. 326. Herbert was delivered immediately after Charles and was decided in the same way. As acknowledged by Mr Pun, its factual and legal background are in substance the same as in Charles. For the same reason above, this authority does not further Mr Pun’s submission. 327. Reynolds concerns another challenge by the plaintiff there (who was also arrested under Regulation 3) against the Attorney-General seeking damages for false imprisonment and compensation for unlawful detention under section 3 of the Constitution. The High Court gave judgment in favour of the plaintiff. On appeal by the Attorney-General and cross-appeal by the plaintiff as to the amount of damages, the Court of Appeal held that (as in Charles and Herbert) Regulation 3 was not in conformity with the Constitution, and therefore the Regulations were void and the plaintiff’s detention was unlawful. It dismissed the Attorney‑General’s appeal and increased the damages awarded to the plaintiff. The Attorney-General appealed to the Privy Council. 328. The Privy Council dismissed the appeal on the primary basis that on the facts of the case, there was an irresistible presumption that no reasonable grounds existed to detain the plaintiff pursuant to Regulation 3. The detention was therefore unlawful. However, relevant for the present purpose, the Privy Council indeed overruled Charles and Herbert in part and held that the Regulations could be interpreted (by way of modification, adaptation, qualification or exception) to be in conformity with the Constitution and thus the Governor did have the power to enact the Regulations under the Constitution. 329. Despite this, Mr Pun relied on this authority only to purportedly show that the Privy Council did agree in part with Charles and Herbert that, save with the remedial construction, the regulation-making power under section 3 of the Order in Council was on the face of it “not in conformity with” the Constitution[59]. Even for this limited purpose, for the same reason we have explained above in relation to Charles and Herbert, this authority again does not take Mr Pun’s submission any further. 330. Given that the ERO itself does not seek to limit fundamental rights, the question of whether it meets the prescribed by law requirement simply does not arise. Mr Chan and Mr Pun’s principal submissions about the width or unlimited scope of the ERO or lack of safeguards are therefore irrelevant under this context. 331. In the premises, KWH and LKH’s challenge against the ERO under this ground is misconceived. 332. But in any event, for the reasons we have given above under Grounds 1, 5A and B and the principles on the prescribed by law requirement we have highlighted above, the ERO cannot be said to have conferred “unfettered and unlimited” power on the CEIC and there are “no independent safeguards against abuses or excesses”. We will explain why. 333. As explained in our discussion under Ground 1 above, given the true nature of emergency regulations and the ground of public danger, very wide powers have to be given to the CEIC to address occasions of emergency or public danger which cannot be exhaustively defined in advance, and the situations which the ERO is intended to tackle with must be varied and unpredictable. It is therefore necessary for section 2(1) of the ERO to have a wide scope. 334. In this respect, a general enactment does not necessarily mean that it falls foul of the legal certainty test under the prescribed by law requirement. It has been repeatedly stated that the prescribed by law requirement does not require absolute precision or certainty. It only requires that the “core” of the norm should be sufficiently clearly formulated. 335. For this, it is illuminating to remind ourselves again the judgment of Justice Gonthier in the Supreme Court of Canada in R v Nova Scotia Pharmaceutical Society [1992] 2 SCR 606 to highlight the relationship between the scope of precision required and the need for general enactment in the modern world[60]: (1) At pp638 - 640, the learned judge said: “Legal rules only provide a framework, a guide as to how one may behave, but certainty is only reached in instant cases, where law is actualized by a competent authority. In the meanwhile, conduct is guided by approximation. The process of approximation sometimes results in quite a narrow set of options, sometimes in a broader one. Legal dispositions therefore delineate a risk zone, and cannot hope to do more, unless they are directed at individual instances. By setting out the boundaries of permissible and non-permissible conduct, these norms give rise to legal debate. They bear substance, and they allow for a discussion as to their actualization. They therefore limit enforcement discretion by introducing boundaries, and they also sufficiently delineate an area of risk to allow for substantive notice to citizens. Indeed no higher requirement as to certainty can be imposed on law in our modern State. Sematic arguments, based on perception of language as an unequivocal medium, are unrealistic. Language is not the exact tool some may think it is. It cannot be argued that an enactment can and must provide enough guidance to predict the legal consequences of any given course of conduct in advance. All it can do is enunciate some boundaries, which create an area of risk. But it is inherent to our legal system that some conduct will fall along the boundaries of the area of risk; no definite prediction can then be made. Guidance, not direction, of conduct is a more realistic objective… A vague provision does not provide an adequate basis for legal debate, that is for reaching a conclusion as to its meaning by reasoned analysis applying legal criteria. It does not sufficiently delineate any area of risk, and thus can provide neither fair notice to the citizen nor a limitation of enforcement discretion. Such a provision is not intelligible … and therefore it fails to give sufficient indications that could fuel a legal debate. It offers no grasp to the judiciary…” (2) At pp641 - 643, he further observed: “… Often the State attempts to realize a series of social objectives, some of which must be balanced against one another, and which sometimes conflict with the interests of individuals. The modern State, while still acting as an enforcer, assumes more and more of an arbitration role. This arbitration must be done according to law, but often it reaches such a level of complexity that the corresponding enactment will be framed in relatively general terms. In my opinion the generality of these terms may entail a greater role for the judiciary … … One must be wary of using the doctrine of vagueness to prevent or impede State action in furtherance of valid social objectives, by requiring the law to achieve a degree of precision to which the subject-matter does not lend itself. A delicate balance must be maintained between societal interests and individual rights… What becomes more problematic is not so much general terms conferring broad discretion, but terms failing to give direction as to how to exercise this discretion, so that this exercise may be controlled. Once more, an unpermissibly vague law will not provide a sufficient basis for legal debate; it will not give a sufficient indication as to how decisions must be reached, such as factors to be considered or determinative elements. In giving unfettered discretion, it will deprive the judiciary of means of controlling the exercise of this discretion…” 336. Further, and importantly, the exercise of power under the ERO and the emergency regulations is subject to negative vetting by the legislature and judicial supervision by way of judicial review. Any eventual restriction on fundamental rights must also meet the proportionality test and the prescribed by law requirement. 337. Given the above, under a holistic analysis, the exercise of the power granted under section 2(1) of the ERO is sufficiently formulated under a “core” of the norm which is governed and guided by the “law” prescribed by legislative negative vetting, the common law and judicial scrutiny, considered against the above background and context of the ERO. 338. In other words, when the CEIC seeks to exercise her power pursuant to section 2(1) of the ERO to enact a regulation under which certain rights and freedoms would be restricted, she must exercise it in a manner that conforms with what the “law” has prescribed for to render such a restriction to be lawful, and that proper exercise of power will be supervised and checked by judicial supervision and legislative scrutiny under negative vetting as we have explained under Ground 1 above. This constitutes sufficient safeguards in the law against arbitrary interference of freedoms and rights. 339. For all the above reasons, the Judges were correct in rejecting Ground 3, and KWH and LKH fail under this ground of their cross-appeal. G3. Ground 4 340. Under paragraphs 1(2) and 3 of KWH-RN, KWH asks the court to uphold the Judgment also on the principle of legality ground. This is raised in support of both Ground 1 (to challenge the ERO) and Ground 4 (to challenge the validity of the PFCR). 341. Insofar as the principle of legality is relied on in additional support of Ground 1, we have already dealt with it above. 342. Insofar as the principle is relied on in support of Ground 4, Mr Chan’s submissions are in essence these[61]. 343. Under the principle of legality as a principle of construction, if a statute does not expressly or by necessary implication override or restrict fundamental rights, it will not be construed as doing so[62]. In this respect, necessary implication is said to be “a matter of express language and logic not interpretation”[63]. 344. Applying this principle, Mr Chan contended that: (1) Section 2(1) of the ERO (even read together with the opening sentence of section 2(2)[64]) is only a generally worded provision. It therefore only confers a general power but is not to be taken to authorise the CEIC to enact regulations which will interfere with or adversely affect the legal rights of citizens in the absence of clear words. In the premises, on a proper construction, the examples listed under sections 2(2)(a) - (n) should and could only be construed as general words and not words (whether expressly or by necessary implication) permitting abrogation from or interfering with fundamental rights. (2) Alternatively, other than in those areas analogous to those specifically referred to under sections 2(2)(a) - (n), in the absence of clear words to the contrary, section 2(1) of the ERO does not confer an unfettered power on the CEIC to make regulations that interfere with fundamental human rights other than those expressed or necessarily implied from the examples in sections 2(2)(a) - (n). In this respect, it can only be necessarily implied from these specified examples[65] that restriction can be imposed only on these fundamental rights: (i) freedom of expression restricted to reception and broadcast of communication (under section 2(2)(a)); (ii) freedom of liberty, fair trial, and freedom of movement (under sections 2(2)(b), (c), (d) and (n)); (iii) property rights and privacy of the home (in the sense as provided for by BL29 but does not include private life as provided for by BOR14) (under sections 2(2)(f), (h) and (k)); and (iv) freedom of occupation (under sections 2(2)(e) and (l)). (3) As a result, the PFCR are ultra vires the ERO because, either: (a) the PFCR adversely affects or interferes with fundamental rights by section 3 (in restricting freedom of expression and freedom of assembly), and section 5 (in restricting right to privacy); or alternatively, (b) the fundamental rights that the PFCR adversely affects or interferes with are not covered by section 2(2)(a) - (n) of the ERO. 345. We are not persuaded by these submissions. 346. Under the principle of legality, whether or not the general wording of section 2(1) of the ERO (read with the opening sentence of section 2(2)) expressly or by necessary implication gives the CEIC a wide power to enact regulations that may impose restrictions on fundamental rights and freedoms is a matter of proper construction. The authorities do not suggest as a general principle that general wording can never be so interpreted. 347. It is now well established that the court adopts purposive construction in interpreting a statutory provision, and in construing statutory provisions, the court does not merely look at the relevant words. It construes the relevant words having regard to their context and purpose. The context of the relevant statutory provision should be taken in its widest sense and will of course include the other provisions of the statute. It may also be relevant in any given case to look at the history of the relevant provision. See: Town Planning Board v Town Planning Appeal Board (2017) 20 HKCFAR 196 at [29] per Ma CJ. 348. In our views, the general wording of section 2(1) and the opening sentence of section 2(2), when interpreted in its proper context, expressly or by necessary implication provides for the making of regulations imposing restrictions on rights and freedoms, including but not limited to the specific instances set out in sections 2(1)(a) - (n). 349. As examined and analysed under Ground 1 above, the ERO was enacted by the legislature to provide the CEIC with a wide power to enact regulations to implement measures to deal with occasions of emergency and public danger. These occasions by their own nature are wide ranging, varied and could not be exhaustively envisaged or defined at the time of the enactment of the ERO. In this context, the wide and general wording used by the legislature is therefore intended and catered by the legislature for the purpose of vesting in the CEIC all the necessary and flexible power to enact all necessary regulations to deal with those wide ranging, unforeseen and unpredictable occasions as and when they occur. 350. At the same time, as submitted by Mr Yu, it is also clear from the ERO when read as a whole that the legislature did have the imposition of restrictions on rights and freedoms in mind when enacting the ERO. Section 2(2) provides that, without prejudice to the generality of section 2(1), regulations may provide for the specific matters in sections 2(2)(a) - (n), which plainly allow the imposition of restrictions on rights and freedoms. This is reinforced by sections 3(1) and (2), which provide that (without prejudice to the powers conferred by section 2) regulations made may provide for the punishment of any offence with such penalties and sanctions. 351. In the premises, when these are properly construed against the above background and purpose, it is plain as a matter of logic that the use of the general wording in sections 2(1) and (2) was objectively intended by the legislature to give the CEIC a wide power to make regulations, which the legislature also envisaged that might impose restrictions on rights and freedoms, including but not limited to the specific instances in sections 2(2)(a) - (n). 352. For these reasons, we reject KWH’s submissions and dismiss Ground 4. H. DISPOSITIONS 353. We allow the respondents’ appeal under Ground 1 and partially allow their appeal under Ground 5A, but dismiss the appeal under Ground 5B. In the premises, we set aside: (1) The Judges’ declarations that (a) the ERO, insofar as it empowers the CEIC to make regulations on any occasion of public danger, is incompatible with the BL, and (b) the PFCR made pursuant to the ERO on an occasion of public danger is accordingly invalid and of no effect. (2) The Judges’ declaration that section 3(1)(b) of the PFCR is inconsistent with BL27 and BOR14, 16, 17 and is therefore null, void and of no effect. 354. We dismiss the KWH-NoCA and KWH-RN, as well as the LKH-NoCA. 355. In sum, we uphold the constitutionality of the ERO insofar as it empowers the CEIC to make emergency regulations on any occasion of public danger. In respect of the PFCR, there is no challenge against section 3(1)(a) relating to unlawful assembly. We uphold the constitutionality of section 3(1)(b) relating to unauthorized assembly. We however hold that sections 3(1)(c) and (d), relating to public meeting and public procession respectively, and section 5 on police powers in relation to facial covering, are all unconstitutional. 356. We will invite parties’ written submissions on costs in relation to these appeals and the hearing below. The respondents shall lodge and serve their submissions (not more than five pages) within seven days from today and the applicants shall lodge and serve their submissions (not more than five pages) seven days thereafter. Unless otherwise directed, we will determine the costs on paper. 357. Lastly, we thank counsel for their assistance. Mr Benjamin Yu SC, Mr Jenkin Suen SC, Mr Jimmy Ma and Mr Mike Lui, instructed by the Department of Justice, for the respondents in CACV 541/2019, CACV 542/2019 and CACV 583/2019 Mr Hectar Pun SC, Mr Lee Siu Him and Mr Anson Wong Yu Yat, instructed by JCC Cheung & Co, assigned by the Director of Legal Aid, and Mr Jasper Wong (on a pro bono basis), instructed by JCC Cheung & Co., for the applicant in CACV 541/2019 Ms Gladys Li SC, Mr Johannes Chan SC (Hon), Mr Earl Deng, Mr Jeffrey Tam, Mr Geoffrey Yeung and Ms Allison Wong, instructed by Ho Tse Wai & Partners, for the applicants in CACV 542/2019 and CACV 583/2019 [1] Respectively under CACV 541/2019 (on appeal from HCAL 2949/2019) and under CACV 542/2019 (on appeal from HCAL 2945/2019). [2] The undisputed background leading to the enactment of the PFCR is set out in detail in the unchallenged evidence filed by the Government. See the Affirmation of Cheung Tin Lok at [3] - [33], the Affidavit of Chui Shih Yen, Joceline at [6] ‑ [15] and [26] ‑ [30] and the Affidavit of Dr Tsui, Pui Wan Ephraem at [8] - [17]. It has been succinctly set out by the Judges at [3] ‑ [9] of the Judgment. [3] “Black-bloc” tactics is where the protestors would often appear in groups and wear black clothing with little or not distinguishable feature, and cover the whole or a substantial part of their faces with sunglasses, goggles, masks, or respirators etc. See the Affirmation of Cheung Tin Lok at [8] ‑ [9]. [4] See, in particular, the Affirmation of Cheung Tin Lok at [20] - [24] and [31]. [5] See the Affidavit of Dr Tsui, Pui Wan Ephraem at [17]. [6] The word “anti-social” refers to acting against larger social norms which, within the group, can be their own emerging group norm. It implies that the person is not really losing his self in the group, but is just following the group norm. [7] Moreover: (a) the use of inflammable liquid bombs has become more frequent and extensive, and even more so since late September 2019. On 29 September and 1 October 2019, the number of inflammable liquid bombs hurled by the protestors were respectively around 100 and over 100; (b) on 1 October 2019, a police officer was injured by protestors throwing corrosive liquid at him, causing a third degree burn; (c) on the same date, some police officers were seen viciously attacked by large groups of protestors by a range of objects and lethal weapons, and a few of them whose lives were under imminent threat had to defend themselves with firearms; (d) as at 4 October 2019, a total of 2,135 individuals have been arrested for taking part in public order events of an unlawful or criminal nature, or being involved in other unlawful or criminal activities. See the unchallenged evidence in the Affirmation of Cheung Tin Lok at [4] - [6], and the Affidavit of Chui Shih Yen, Joceline at [26] - [28]. [8] See the Affirmation of Cheung Tin Lok at [5] - [6] on the statistics and descriptions showing the significant increase and extensive use of inflammable liquid bombs by the protestors, attacks on police officers and individuals who held different views with the protestors, and the growing degree of violence and vandalism. [9] See [18] of the Decision. [10] See the relevant Press Release. [11] The current situation was described in more detail in the CE’s speech given in Chinese. See the Chinese version of the Press Release. [12] The Chinese speech explained more about the aim of making the PFCR. See the Chinese version of the Press Release. [13] Section 2 contains the definitions while section 6 provides for the time in which prosecution may be brought. [14] That is, the Public Order Ordinance (“the POO”). [15] This means $25,000: see section 113B of and Schedule 8 to the Criminal Procedure Ordinance (Cap 221) (“the CPO”). [16] “Facial covering” is defined in section 2 to mean a mask or any other article of any kind (including paint) that covers all or part of a person’s face. [17] This has the same meaning as in section 18 of the POO. [18] This has the same meaning as in section 17A(2) of the POO. [19] As meeting of no more than 50 persons, meeting in private premises with no more than 500 persons, and certain meetings in school do not have to be notified for the purpose of the POO, they therefore fall outside section 3(1)(c). [20] Public processions of no more than 30 persons and public processions not on a public highway or thoroughfare or in a public park, which do not have to be notified for the purpose of the POO, fall outside section 3(1)(d). [21] This means $10,000: see section 113B of and Schedule 8 to the CPO. [22] International Covenant on Civil and Political Rights. [23] See [13] and [37] of the Judgment. [24] As the ICCPR through the BOR is constitutionally enshrined under BL39. [25] Since section 5 of the HKBORO and ICCPR4 are of identical terms, the Judges therefore deals with this ground by reference to section 5 of the HKBORO only, which they say the analysis and reasons apply equally to the argument based on ICCPR4. See [104] of the Judgment. [26] The 4 steps are: (1) does the measure pursue a legitimate aim; (2) if so, is it rationally connected with advancing that aim; (3) whether the measure is no more than reasonably necessary for that purpose; and (4) whether a reasonable balance has been struck between the societal benefits promoted and the inroads made into the protected rights, asking in particular whether pursuit of societal interest results in an unacceptably harsh burden on the individual. [27] BOR17 and BL27. [28] BOR16 and BL27. [29] BOR14. [30] The legitimate aims being (a) deterrence and elimination of the emboldening effect for those who may otherwise, with the advantage of facial covering, break the law, and (b) facilitation of law enforcement, investigation and prosecution. See [130] of the Judgment. [31] See [152] - [157] of the Judgment. [32] Having concluded that these provisions do not meet step 3, the Judges do not find it necessary to consider step 4. However, they have expressed the view that if it is necessary to do so, they would conclude, for the same reasons, that the provisions have failed to satisfy step 4. [33] See the transcript of the proceedings of the LegCo dated 26 August 1936. [34] See, for example, (a) in respect of the GIC: section 62 of the Employment Ordinance (Cap 57), section 31 of the Import and Export Ordinance (Cap 60), section 25 of the Fire Services Ordinance (Cap 95), section 42 of the New Territories Ordinance (Cap 97), section 59 of the Immigration Ordinance (Cap 115), section 14(1) of the Town Planning Ordinance (Cap 131), section 33(1) of the Medical Registration Ordinance (Cap 161), section 7 of the Registration of Persons Ordinance (Cap 177), section 45 of the Police Force Ordinance (Cap 232), section 22 of the Customs and Excise Service Ordinance (Cap 342), section 41 of the Building Management Ordinance (Cap 344); and (b) in respect of the various Secretaries: section 14(2) of the Town Planning Ordinance (Cap 131) (Secretary for Development), section 72 of the Mental Health Ordinance (Cap 136) and section 33(3A) of the Medical Registration Ordinance (Cap 161) (the Secretary for Food and Health). [35] See for example, section 5 of the Official Languages Ordinance (Cap 5), section 69 of the Evidence Ordinance (Cap 8), section 10 of the Lands Tribunal Ordinance (Cap 17), section 45 of the Labour Tribunal Ordinance (Cap 25), section 117 of the Insurance Ordinance (Cap 41), section 15 of the Telecommunications Ordinance (Cap 106), section 101I of the Banking Ordinance (Cap 155), section 72 of the Legal Practitioners Ordinance (Cap 159), section 8 of the Domestic and Cohabitation Relationships Violence Ordinance (Cap 189), section 32 of the Matrimonial Proceedings and Property Ordinance (Cap 192), section 10 of the Partition Ordinance (Cap 352), section 44 of the Control of Obscene and Indecent Articles Ordinance (Cap 390). [36] See for example, in respect of the CEIC, section 8 of the Prevention and Control of Disease Ordinance (Cap 599) which came into force on 14 July 2008; sections 208 and 273 of the Companies Ordinance (Cap 622) which came into force on 3 March 2014; in respect of the Financial Secretary, section 909 of the Companies Ordinance; in respect of the Secretary for Food and Health, section 7 of the Prevention and Control of Disease Ordinance; in respect of the Chief Justice, sections 233 and 269 of the Securities and Futures Ordinance (Cap 571), which came into force on 1 April 2003, section 37ZJ of the Financial Reporting Council Ordinance (Cap 588), which came into force on 1 December 2006, section 76 of the Anti-Money Laundering and Counter-Terrorist Financing Ordinance (Cap 615) which came into force on 8 July 2011. [37] Those non-derogable rights are set out in articles 2, 3, 4(1) and (2), 7, 12, 13, and 15 of the BOR. [38] It is not the respondents’ case that such measures include legislative measures such as emergency regulation. In other words, the measures envisaged under BL56(2) are executive or administrative measures only. [39] See BL62(2). [40] See BL60(1). [41] We observe by way of judicial notice that in tackling the current Covid-19 pandemic, the Government has invoked sections 7 and 8 of the Prevention and Control of Disease Ordinance (Cap 599) to make emergency regulations. If that Ordinance did not exist, it would appear that the Government would need to invoke the ERO: see [65(4)] above. [42] [88] of the Judgment. [43] [89] of the Judgment. [44] [90] of the Judgment. [45] [91] of the Judgment. [46] [92] of the Judgment. [47] [93] of the Judgment. [48] Nor is there any for emergency. [49] In the course of oral submissions, we invited the parties to assist us on the meaning of “public danger” for the purpose of the ERO. On reflection, we need not dwell on this aspect as it is not necessary to interpret the phrase for present purpose. The only point that we wish to make is that a definition is certainly achievable by means of statutory interpretation. [50] In the present case, apart from the LegCo Brief relating to the PFCR dated October 2019, the CE, the Secretary for Justice and the Secretary for Security explained the Government’s decision and the operation of the PFCR at a press conference on 4 October 2019; the responsible Government officials also attended and answered questions at open meetings of the LegCo Subcommittee on Prohibition on Face Covering Regulation held on 22 and 28 October, 5, 9, 12 and 18 November 2019. They have all been adduced as evidence before the court. [51] For the meaning of the principle of legality as advanced by KWH, see the court’s discussion at Section G3 below for Ground 4. [52] LKH’s contentions under Ground 5B have been addressed above at [251] - [255]. [53] That is, derogation from the rights protected under the ICCPR can only be made when there is a “public emergency”, which has to be one that “threatens the life of nation” and its “existence”, and that the public emergency has to be “official proclaimed”. [54] It may not be entirely correct to say that the existing legislations which are regarded to be inconsistent with a regulation enacted to be “repealed”. Under section 2(4), any such inconsistent provision of an enactment will only be rendered to have “no effect” when the concerned regulation made under the ERO remains in force. [55] As provided in section 5(1) of the HKBORO, the measures to derogate from the rights “shall be taken in accordance with the law”. [56] LKH also seeks to cross appeal the Judges’ rejection of Ground 3. See LKH-NoCA, at [4]. [57] BL39(2) provides: “The rights and freedoms enjoyed by Hong Kong residents shall not be restricted unless prescribed by law ...” [58] See KWH’s skeleton submissions on the cross appeal dated 16 December 2019 at [33]. [59] AG v Reynolds, supra at 655A-B. [60] These are cited in Chee Fei Ming v Director of Food and Environmental Hygiene, supra, at [56] - [57]. [61] These are in substance a rerun of the arguments made before the Judges. As mentioned above, the Judges do not find it necessary to deal with these arguments as they find them to be inconsistent with the arguments raised under Ground 1, which they have found in favour of KWH. [62] See: R v Home Secretary Ex p Simms, supra, at p131 per Lord Hoffmann; A v Commissioner of ICAC (2012) 15 HKCFAR 362 at [24] - [29] perRibeiro PJ; HM Treasury v Ahmed [2010] 2 AC 534 at [45] - [46] per Lord Hope, at [112] per Lord Walker. [63] See: R (Morgan Grenfell Ltd) v Special Commissioner [2003] 1 AC 563, 616 at [45] perLord Hobhouse. [64] Which provides “Without prejudice to the generality of the provisions of [section 2(1)]”. [65] With the exception of sections 2(2)(g) and (i) which are wide empowering provisions and thus could not be construed to necessarily imply enabling the enacting of regulations that restrict or interfere with fundamental rights. 
Mr Justice Ribeiro PJ: 1. I agree with the judgment of Mr Justice Bokhary NPJ. Mr Justice Fok PJ: 2. I agree with the judgment of Mr Justice Bokhary NPJ. Mr Justice Cheung PJ: 3. I agree with the judgment of Mr Justice Bokhary NPJ. Mr Justice Bokhary NPJ: 4. This is an appeal by the Commissioner of Inland Revenue (“the Commissioner”) from a decision of the Court of Appeal (Macrae VP, Yuen and Kwan JJA). They granted him leave to appeal so that he may seek from us an unqualified affirmative answer to a question of law, he having failed to obtain such an answer from them. Question on leave to appeal was granted 5. In the order granting such leave, the question is formulated thus. It begins by postulating the situation “[w]here a contract of employment is terminated by the employer, and the employer agrees at termination to pay to or confer on the employee (1) payment or benefit to eliminate or settle any threatened claim by the employee for, and the payment or benefit is paid or conferred in lieu of, a payment or benefit which if made had the contract of employment not been terminated would be chargeable to salaries tax; [and] (2) a benefit being the entitlement to exercise a right to acquire shares contingently conferred on the taxpayer as the holder of an office in or an employee of the employer”. And it then asks: “is the payment or benefit, or any gain from the exercise of any such benefit, so paid or conferred at termination chargeable to salaries tax under the Inland Revenue Ordinance (Cap 112)?” 6. But are the facts really as postulated in that question? Let us see. Sum D received and Share Option Gain made 7. Immediately prior to 20 July 2008 the respondent Mr John Poon (“the Taxpayer”) was in employment in Hong Kong as the Group Chief Financial Officer and an executive director of a company (“the Employer”) incorporated in Bermuda and listed on the Hong Kong Stock Exchange. On that date the Taxpayer’s employment by the Employer was terminated pursuant to a “Separation Agreement” under which the Taxpayer received certain sums and benefits from the Employer in respect of all claims and rights of actions. 8. One of those sums was a sum of €500,000 paid under clause 4.1.4 of the Separation Agreement “in lieu of a discretionary bonus” for the financial year ended 30 June 2008. This sum of €500,000 has hitherto been - and will in this judgment continue to be - called “Sum D”. 9. As to benefits, clause 5 of the Separation Agreement provided that, despite the cessation of his employment, the Taxpayer was entitled to exercise his stock options set out in Annexure 2 of that agreement. He exercised those options by a number of tranches (which have hitherto been - and will in this judgment continue to be - identified by reference to letters of the alphabet). On 19 August 2008 the Taxpayer subscribed for 360,000 shares in the Employer at $24.20 per share (under “Tranche A”) and 720,000 shares in the Employer at $42.58 per share (under “Tranches B and C”). Those 1,080,000 shares were allotted to him on the following day. Their closing prices on 19th and 20th of that month were $73.15 and $76.50 respectively. The notional gain thus derived has been - and will in this judgment continue to be - referred to as “the Share Option Gain”. Challenge to assessment to salaries tax of that sum and that gain 10. In the 2008/2009 year of assessment, the total on which the Taxpayer was assessed to salaries tax for 2008/2009 included (i) the €500,000 which forms Sum D and (ii) the Share Option Gain which gain was calculated at $43,250,400. He appealed to the Board of Review against their inclusion. The issues in that appeal were these. (1) Is Sum D taxable? (2) Is the Share Option Gain taxable? (3) If the Share Option Gain is taxable, is the date for its computation 19 August 2008 when the share options were exercised (“the Exercise Date”) or 20 August 2008 when the shares were allotted (“the Allotment Date”)? 11. The Commissioner contended that both Sum D and the Share Option Gain are taxable and that computation should be as at the Allotment Date. Disagreeing, the Taxpayer contended that neither Sum D nor the Share Option Gain is taxable but that if the latter is taxable then computation should be as at the Exercise Date. Succeeded on appeal to the Court of Appeal 12. Dismissing the Taxpayer’s appeal, the Board of Review, agreeing with the Commissioner, held that both Sum D and the Share Option Gain are taxable and that computation should be as at the Allotment Date. The Taxpayer then appealed against the Board of Review’s decision. Agreeing with that decision, the Court of First Instance (Anthony Chan J) dismissed the Taxpayer’s appeal, whereupon the Taxpayer appealed against the Court of First Instance’s judgment. The Court of Appeal (by Yuen JA’s judgment with which the other members of the panel agreed) allowed the Taxpayer’s appeal, holding that neither Sum D nor the Share Option Gain is taxable. Relevant statutory provisions 13. Under section 8(1) of the Inland Revenue Ordinance (“the Ordinance”), “income arising in or derived from Hong Kong” from various specified sources is chargeable to salaries tax. One of those sources, being the one specified in item (a) of that subsection, is “any office or employment of profit”. The expression “income from any office or employment” is defined by section 9 of the Ordinance to include various things including those specified in items (a) and (d) of that subsection. Item (a) specifies “any wages, salary, leave pay, fee, commission, bonus, gratuity, perquisite, or allowance, whether derived from the employer or others”. And item (d) specifies “any gain realized by the exercise of… a right to acquire shares or stock in a corporation obtained by a person as the holder of an office in or an employee of that… corporation”. Fuchs analysis 14. As relevant to what we have to decide in the present case, what we held in Fuchs v Commissioner of Inland Revenue (2011) 14 HKCFAR 74 is as follows. Income chargeable to salaries tax under section 8(1) of the Ordinance is not confined to income earned in the course of employment. It includes payments made in return for acting as or being an employee. In other words, it includes rewards for past services. It also includes payments made by way of inducement to enter into employment and provide future services. If a payment, viewed as a matter of substance and not merely of form and without being blinded by some formulae which the parties may have used, is found to be derived from a taxpayer’s employment in the foregoing sense, it is chargeable to salaries tax. That analysis provides guidance on the operation of the relevant statutory words without supplanting or even modifying those words. Payments which are for something else do not come within the analysis, and are not chargeable to salaries tax. All of that appears at paragraphs 17 and 18 of the judgment which was given by Mr Justice Ribeiro PJ and agreed with by all the other members of the panel. Fuchs’s case itself 15. Before going further into the facts of the present case and applying the Fuchs analysis to them, let us examine how that analysis operated in the factual circumstances of Fuchs’s case itself. Mr Fuchs entered the employ of a German bank in 1976. After working for it in Germany and then in Singapore, he was seconded to its Hong Kong branch in July 2003. On 18 November 2003 he entered into a contract of employment to work for the bank at that branch for a three-year period commencing on 1 January 2004 as its Managing Director and CEO Asia. Within that three-year period, the following things happened. In November 2005 the bank was taken over by an Italian banking group. As part of the re-organization resulting from the takeover, it was decided that Mr Fuchs’s employment with the bank would be terminated. There were negotiations over the terms of such termination. The terms agreed upon were set out in an agreement dated 17 October 2005. It was provided in this agreement that Mr Fuchs’s employment was to end by 31 December 2005. That date, it will be observed, would be the last day of the second year of the three-year period for which he was employed. It was also agreed that the bank would pay him “a one-time compensation for the loss of his position due to the termination of the employment relationship for operational reasons” in the total sum of $18,276,667. 16. A breakdown of this total sum into three component sums was provided in the agreement. The first component sum was of $3,120,000. It was equivalent to Mr Fuchs’s salary for the remaining year of the three-year period for which he was employed. The second component sum was of $6,240,000 expressed to represent “two annual salaries for the duration of service with” the bank. The third component sum was of $8,916,667. It was expressed to represent “the average amount of the bonuses paid in the 3 previous years”. 17. Mr Fuchs contended that none of the total sum was taxable. The revenue accepted that the first component sum was not taxable. But it maintained that the second and third component sums were chargeable to salaries tax. Mr Fuchs’s employment contract provided for termination by either party without cause upon expiry of the initial term or termination at any time by the bank for any of a number of specified causes. And it provided that if the bank terminated or purported to terminate Mr Fuchs’s employment on any other grounds, it would pay him as agreed compensation or liquidated damages two annual salaries and an average amount of the bonuses paid in the three previous years of his employment with the bank. Those two annual salaries and that average sum are, it will be noticed, precisely what the second and third component sums respectively represent. So the second and third component sums were paid in satisfaction of rights which had accrued to Mr Fuchs under his contract of employment. That meant that they were derived from his employment and were chargeable to salaries tax. We so held. Both held by the Court of Appeal to be for something else 18. In the present case, on the issue of whether Sum D and the Taxpayer’s stock options entitlement under clause 5 of the Separation Agreement were for something within the Fuchs analysis or were for something else, the Court of Appeal held that both were for something else, so that neither Sum D nor the Share Option Gain was chargeable to salaries tax. That conclusion is now attacked by the Commissioner while the Taxpayer defends it. A closer look at Sum D 19. Sum D consists of, it will be remembered, €500,000 said to be in lieu of a discretionary bonus for the financial year ended 30 June 2008. Under the heading “The true nature of Sum D”, the Board of Review, found the following facts. The Taxpayer’s employment with the Employer was under a service agreement (“the Service Agreement”). Clause 4.3 of the Service Agreement provided that in addition to his salary the Taxpayer “will be eligible to participate in [an annual bonus scheme] on such terms and at such level as [the Employer’s board of directors] may from time to time determine”. The financial year of the Employer was from 1 July to 30 June. After the Employer’s auditors had prepared its audited accounts, the Employer’s executives would look at the audited results and make suggestions to a remuneration committee, probably in August each year. This committee would then make a recommendation further up to the Employer’s board of directors. 20. In his witness statement, the Taxpayer, whom the Board of Review found to be a “truthful witness”, said that the figure of €500,000 which Sum D represented was “arbitrarily arrived at by negotiations”. What negotiations were these? The Board of Review said that “[o]ne may say” that what the Taxpayer received under the Separation Agreement was “consideration to make [him] go away quietly.” The Taxpayer also described Sum D as “a figure, arbitrarily arrived at by negotiations, intended to eliminate any possible claim and lawsuit I might advance against the company for depriving me of the opportunity to be considered for discretionary bonus.” 21. Like the Taxpayer, his solicitor was found by the Board of Review to be a “truthful witness”. From the evidence of these two witnesses the following picture emerges. The Taxpayer had expected to succeed the then Chairman of the Employer’s board of directors. In the afternoon on Friday 18 July 2008, the Chairman informed him that the Employer was preparing to terminate his employment immediately and remove him from the offices which he was holding. He was taken aback. His mood was combative. And he refused to go quietly. Instead he proposed to challenge the plan to remove him from the board. This was to have been done by bringing the matter before the shareholders, he believing that he would have had their support. His remaining on the board, indeed even any delay in his departure from it, was contrary to the wishes of the Chairman and the majority of the other directors. He also disputed the validity of the restraint of trade provisions in the Service Agreement. And he was prepared to take his claims to court. This would have attracted media interest and consequential market reaction. The relationship between the Taxpayer and the Employer was acrimonious. 22. There was a weekend of negotiations involving lawyers on both sides. These negotiations resulted in the Separation Agreement which was entered into on Sunday 20 July 2008. 23. The Employer told the revenue: (i) that the Taxpayer was not awarded any bonus for the financial year ended 30 June 2008; (ii) that Sum D was an “entirely arbitrary amount mutually agreed by [the Taxpayer] and [the Chairman]”; and (iii) that it was paid “to eliminate any claim for unpaid bonus”. And a closer look at the stock options entitlement 24. That closer look at Sum D having been taken, it is now time to take a closer look at the other thing which the Taxpayer received under the Separation Agreement, namely the stock options entitlement thereunder. 25. In 2001 the Employer adopted a written share option scheme (“the Share Option Scheme”). By three letters dated 26 November 2003, 27 November 2004 and 7 February 2007 respectively (which have hitherto been referred to as - and will in this judgment continue to be referred to as - “the Grant Letters”), the Employer offered the Taxpayer options to subscribe for its shares subject to the terms of that scheme. He accepted those offers by signing those letters. 26. Under the 2003 Grant Letter, the subscription price was $24.20 per share. A total of 1,800,000 shares was provided for. And the vesting dates, each in respect of 360,000 shares, were 26 November of 2004, 2005, 2006, 2007 and 2008 respectively. The 360,000 shares at $24.20 per share which the Taxpayer subscribed for and was allotted under Tranche A were those for which the vesting date under this Grant Letter was 26 November 2008. 27. We come now to the 2004 Grant Letter. The subscription price thereunder was $42.58 per share. A total of 1,800,000 shares was provided for. And the vesting dates, each in respect of 360,000 shares, were 27 November of 2005, 2006, 2007, 2008 and 2009 respectively. Of the total of 720,000 shares at $42.58 per share which the Taxpayer subscribed for and was allotted under Tranches B and C, 360,000 shares were those for which the vesting date under this Grant Letter was 27 November 2008 and 360,000 shares were those for which the vesting date thereunder was 27 November 2009. 28. Finally we come to the 2007 Grant Letter. The subscription price thereunder was $83.00 per share. A total of 800,000 shares was provided for. And the vesting dates, each in respect of 160,000 shares, were 7 February of 2008, 2009, 2010, 2011 and 2012 respectively. By a tranche (which has hitherto been identified by - and will in this judgment continue to be identified by - the letter “D”), the Taxpayer subscribed for and was allotted 160,000 shares at $83.00 per share, they being the shares for which the vesting date under this Grant Letter was 7 February 2008. 29. As at the date of the Separation Agreement, the stock option entitlements in respect of the shares subscribed for and allotted under Tranches A, B and C had not yet vested, but the stock options entitlement to which Tranche D relates had vested. Nothing under Tranche D forms any part of the Stock Option Gain. 30. The 2003 and 2004 Grant Letters each contained a term and condition by which this was said to the Taxpayer: “Unless otherwise agreed by the Board in its absolute discretion (and approved by independent non-executive directors of the Company) the Option will only be granted to you in your capacity as Group Chief Financial Officer in the Group (‘the Position’) and may lapse if you cease to be in the Position”. As to the use there of the words “may” in the expression “may lapse”, Yuen JA made two observations. The first is that the expression was used to “cater for the possibility of acceleration”. And the second is that “[i]t was only as a result of the Employer’s decision to accelerate the vesting dates to the Separation Date that the Taxpayer was able to take the benefit of [Tranches A, B and C]”. Although nothing under Tranche D forms any part of the Stock Option Gain, it should be said for the sake of completeness that the 2007 Grant Letter contained a term and condition similar to the one in the 2003 and 2004 Grant Letters set out above. 31. There is a provision in the Grant Letters under which, in the event of the Taxpayer’s employment being terminated and salary being paid in lieu of notice, the Employer’s board may in its absolute discretion accelerate the vesting period by allowing the Taxpayer to exercise part or all of any unvested option that would have vested during the notice period. 32. In a written question posed on 21 January 2011 by the revenue to the Employer, two sets of fact were referred to. The first was that “the vesting dates of the 1,080,000 share options set out in Annexure 2 of the Separation Agreement were accelerated from 26 November 2008, 27 November 2008 or 27 November 2009”. And the second was that “[t]he first two dates were within 6 months from 20 July 2008 whilst the latter date was beyond 6 months from 20 July 2008 (6 months being the notice period prescribed in the Service Agreement)”. The question then continued thus: “In this connection, advise with documentary support (if any): (a) Why the vesting dates of the share options were allowed to be accelerated. (b) The basis on which the 1,080,000 share options were determined. (c) The reasons and justifications for allowing [the Taxpayer] to exercise the share options within the accelerated vesting period.” 33. The Employer’s answers were dated 18 March 2011. On point (a), the Employer’s answer was: “[The Employer] agreed to allow [the Taxpayer] to exercise the share options immediately on the signing of the Separation Agreement as part of the terms of the cessation of [his] employment.” The Employer’s answer on point (b) was “The number of share options with accelerated vesting was an entirely arbitrary number. No specific basis was adopted in determining such number.” And the Employer’s answer on point (c) was: “[The Employer] allowed an acceleration of vesting of the share options so with a view to settling all outstanding matters upon the cessation of [the Taxpayer’s] employment.” 34. From the Employer’s answers, it will be noticed that the settlement did not differentiate between, on the one hand, the dates which were within the notice period and, on the other hand, the date which was beyond that period. Nub of the Court of Appeal’s reasoning 35. The nub of the reasoning by which Yuen JA concluded that Sum D was not chargeable to salaries tax is to be found in paragraph 34.3 of her judgment. At the end of that paragraph, she said that Sum D “was not income ‘from’ the Taxpayer’s employment, but a payment he obtained from the challenges he posed to the Employer which led to negotiations culminating in the Separation Agreement. It was the antithesis to a reward for his services under the contract of employment.” 36. As for the nub of the reasoning by which Yuen JA concluded that the Stock Option Gain was not chargeable to salaries tax, it is to be found in paragraph 36 of her judgment. In that paragraph she said this: “[T]he acceleration of vesting leading to the Stock Option Gain was also not a benefit given for the purpose of rewarding the Taxpayer for services past[,] present or future, but for another reason, viz. it was consideration for him to drop his proposed two-pronged course of action, and to agree to present a united front with the Employer (both internally and to the public) on the reasons for his departure (as set out in the annexes to the Separation Agreement), amongst other additional covenants set out in that Agreement.” Reason why the Court of Appeal granted leave to appeal to us 37. The Court of Appeal’s formal order granting the Commissioner leave to appeal to us says that such leave was granted on the ground that the question which he put forward is of great general or public importance. But the reason which the Court of Appeal gave for granting such leave was only that it may be helpful for us to follow up our decision in Fuchs’s case (where the payment was made under the provisions of the contract of employment) with a decision on the facts of the present case (where the payment was not so made). Commissioner’s arguments 38. Let us now turn to the Commissioner’s arguments in the present appeal. He begins by saying that one must focus on the words of the relevant statutory provisions and that judicial pronouncements do not replace, although they may clarify or explain, such words. That is so. But it does not provide any basis for criticizing the approach taken by the Court of Appeal. They merely applied the Fuchs analysis which clarifies and explains, but in no way operates to replace, the relevant statutory words. 39. It will be remembered that clause 4.1.4 of the Separation Agreement speaks of Sum D as being paid “in lieu of a discretionary bonus”. The Commissioner stresses the words “in lieu”. But as has already been pointed out, questions of taxability or otherwise turn on substance rather than mere form, and in answering them the courts will not be blinded by some formulae which the parties may have used. The Commissioner contends that Sum D was a reward for past services. But was it? 40. Under paragraph 4.1 of the Separation Agreement, the Employer agreed, without admission of liability, to pay the Taxpayer five sums as compensation in respect of possible claims of the type referred to as “Settlement and Waiver” in clause 6 of that agreement. Those five sums are: (i) €500,000 in lieu of notice; (ii) $129,533 statutory long service pay; (iii) €30,137 in lieu of 11 days’ accrued but unused annual leave; (iv) Sum D; and (v) €1,500,000 in consideration of covenants given by the Taxpayer under clause 10 of that agreement, including a covenant not to challenge the restraint of trade clauses in the Service Agreement. Clearly the first three sums were rewards for past services while the fifth sum was not a reward for past services. 41. As has already been mentioned, the Board of Review found as a fact that Sum D was part of what was paid to the Taxpayer to make him “go away quietly”. That is clearly so. On the evidence which the Board of Review found to be truthful, the Chairman and the other directors wanted the Taxpayer to leave the board without delay. Unless he could be persuaded to leave, he might have been able not merely to delay his departure from the board but even to stave it off altogether. And in this connection, even his efforts, let alone his success, would have generated publicity and market reaction of the kind which the Employer wished to avoid. So would have the claims which he was prepared to take to court. 42. The Commissioner contends that in paying the Taxpayer Sum D the Employer was, rather than denying his right to be considered for a discretionary bonus and the efforts that earned him that right, recognising them. This contention runs counter to the established facts which were, as has just been noted, that Sum D was part of what the Employer paid to make the Taxpayer go quietly. And, having been so paid, he did go quietly. There is simply no evidence, let alone any finding, that Sum D was paid to reward the Taxpayer for past services. On the established facts, Sum D was indeed, as Yuen JA said, “the antithesis to a reward”. 43. Now for the taxability or otherwise of the Share Option Gain. 44. The Commissioner contends that the Share Option Gain is “a gain realized by the exercise of… a right to acquire shares or stock in a corporation obtained by a person as the holder of an office in or as an employee of that… corporation” within the meaning of item (d) of section 9(1) of the Inland Revenue Ordinance. 45. With a view to making good that contention, the Commissioner has put forward the following arguments. The answer to the question of whether or not the Share Option Gain is taxable cannot be affected by the fact that it was only because of the acceleration provided for by clause 5 of the Separation Agreement that the Taxpayer was able to exercise the right to acquire the shares. It is clear that the right to acquire shares was obtained by the Taxpayer as an employee. The contingency i.e. the vesting only concerns the exercise, but not the obtaining, of the right to acquire shares as an employee. The acceleration provided by clause 5 of the Separation Agreement did not grant the Taxpayer any right to acquire shares. It only provided that he might exercise the right to acquire shares which he had already obtained as an employee. The only benefit it granted to him was acceleration. It did not grant him any right to acquire shares. 46. It is apparent from clause 1.1 of the Share Option Scheme, the Commissioner says, that the analysis contained in this argument of his is how the parties objectively understood the options exercised under Tranches A, B and C (“the Relevant Options”) i.e. that the vesting only defined when the Relevant Options could be exercised. 47. The Commissioner points to these definitions contained in clause 1.1 of the Share Option Scheme. “Option - a right to subscribe for Shares granted pursuant to this Scheme, including both Vested Option and Unvested Option”. “Unvested Option - an Option that is not exerceisable [sic] pursuant to the terms of the 2001 Share Option Scheme and the terms on which the Option is granted”. “Vested Option - an Option that is exerciseable [sic] pursuant to the terms of the 2001 Share Option Scheme and the terms on which the Option is granted”. “Vesting Period - such period of time, as may be determined by the Board in its absolute discretion and set out in the terms of the grant of the Option, during which the right to exercise the Option in respect of all or some of the Shares to which the Option relates will vest subject to and in accordance with the terms and conditions of the grant of the Option.” 48. Reverting to the Commissioner’s arguments, they continue as follows. It is immaterial that the Relevant Options were contingent at the time when they were granted in the sense that the Taxpayer’s right to exercise them was contingent on vesting. The language of section 9(1)(d) of the Inland Revenue Ordinance does not preclude it from applying to contingent rights to acquire shares. There is no reason why it should be so precluded, since its purpose is to identify whether a gain has the requisite connection to the employee’s employment. 49. The Commissioner’s arguments then proceed to pose the example of an option to purchase shares say five years later (i.e. contingent on the satisfaction of the five-year period), and continues thus. There can be no doubt that when that period elapses, the employee does not receive a new option. He merely exercises the option which he had obtained five years before as an employee. Likewise, on a purposive interpretation of section 9(1)(d), if one were to ask what should be the relevant event to analyse in order to decide whether or not the option has the requisite connection to the employee’s employment, the clear answer is that one should look at the grant of the option (with the contingency), and not at the satisfaction of the contingency. The only difference between the present case (at least in regard to Tranches A and B) and that example is the complexities of the contingencies, but the nature of the contingencies should not make a difference to the legislation’s application. Further, no difference arises from a variation of the contingencies (as in regard to Tranche C). 50. Finally, the Commissioner’s arguments proceed to deal with what he says the position would nevertheless be if, contrary to what he has been arguing up to this point, the Court of Appeal were correct to conclude that the Taxpayer obtained the right to acquire shares only upon acceleration of the vesting of the Relevant Options. Even then, the Commissioner argues, the Share Option Gain would still have been realised from the exercise of a right to acquire shares obtained by the Taxpayer as an employee. The Commissioner argues that given that the Taxpayer obtained a right to acquire shares, from the exercise of which he made the Share Option Gain, the only question would be whether he obtained that right as an employee even though he only obtained the same under the Separation Agreement. And the Commissioner argues that the answer would be in the affirmative for the following reasons. 51. That the stated objective for which the Employer accelerated the Relevant Options was the settlement of all outstanding matters upon the cessation of the Taxpayer’s employment is not determinative. It is necessary to look further to see why the Taxpayer obtained such a right under the Separation Agreement, and to decide whether he obtained it as an employee or otherwise. 52. It was to address the various actions which the Taxpayer threatened that the Employer agreed to, among other things, let him keep and benefit from the Unvested Option which the Taxpayer obtained as an employee. Even on the assumption that the right to acquire shares was only obtained under the Separation Agreement, the Employer did not grant the right from nowhere. All it did was to complete the final step (namely vesting) of a process which the Taxpayer was allowed to access as an employee. Thus the obtaining of the right to acquire shares was as an employee, and the Share Option Gain falls within section 9(1)(d). 53. The present case cannot be treated in the same way as a situation in which the Taxpayer never had even a contingent entitlement to any share options, and had managed to negotiate the share options purely upon termination, out of nothing. It was only because he had already been granted something, namely the Unvested Option, as an employee that the Taxpayer was in a position to demand the acceleration of the Relevant Options as part of the Separation Agreement. Such acceleration could therefore only have been, and certainly was sufficiently substantially, “from” the Taxpayer’s employment and received by him as an employee. He was able to demand and be given something vis-à-vis the Relevant Options because he was granted the Relevant Options as an employee for the employment services which he had given or was to give. This is confirmed by the evidence from the Taxpayer’s solicitor who said that in exchange for the Taxpayer going quietly and agreeing to various ongoing obligations, such as non-disparagement and not challenging the restraint of trade clauses, the Employer was prepared to “make concessions on matters such as compensating [the Taxpayer] for ... the loss of share options that had not vested”. So the right to acquire shares, even if obtained only by the Taxpayer under the Separation Agreement, was obtained by him as an employee. 54. Pausing in this recital of the Commissioner’s arguments, it is to be noted that the Taxpayer’s solicitor did indeed say what he says that she said. But her use of the expression “such as” is to be noticed. So is the fact that clause 6 of the Separation Agreement referred to full and final settlement of “all claims and rights of action (whether under statute, common law or otherwise)”. As to claims and rights of action under statute, there was the possibility of a remedy available to the Taxpayer under the Employment Ordinance (Cap 57) if there was no valid reason for his termination. 55. Pointing to what Yuen JA said in paragraph 32.2 of her judgment, the Commissioner says that the reliance which the Court of Appeal placed on the decision of the House of Lords in Hunter v Dewhurst (1932) 16 TC 605 was “misplaced”. But what Yuen JA said in paragraph 32.2 of her judgment is no more or less than this: “Indeed the present case is similar to the payment in Hunter v Dewhurst where the sum in question was paid for the [employer] ‘to obtain a release from its contingent liability’. That sum was held by the House of Lords to be not chargeable to salaries tax.” There is no inaccuracy in that statement of what happened in Hunter v Dewhurst. And Yuen JA’s view that what the Taxpayer received is similar to the payment to Commander Dewhurst and likewise not taxable was reached by applying the Fuchs analysis to her appreciation of the facts of the present case. It was not reached on any misapplication of Hunter v Dewhurst. 56. From what Lord Woolf said in Mairs v Haughey [1994] 1 AC 303, the Commissioner seeks to extract what he calls a “substitution test”. What does the Commissioner say is that “test”? He says (taking it from how he puts it in paragraph 56 of his written case) that it is that “[t]he payment of a sum in true substitution of another is an acknowledgment and takes the place of, and has the same nature as, the latter.” The use of the word “true” to qualify the word “substitution” is appropriate and telling. It alerts one to the necessity of asking in each case what is and what is not truly a substitution. In any event, it is necessary to look at this statement by Lord Woolf (at p 319D): “It is inevitable that if a payment is made in substitution for a payment which might, subject to a contingency, have been payable that the nature of the payment which is made in lieu will be affected by the nature of the payment which might otherwise have been made. There will usually be no legitimate reason for treating the two payments in a different way.” So even if there is such a substitution, there may be a legitimate reason for treating the two payments differently. Both were considered in Fuchs’s case 57. The Commissioner points to Chadwick LJ’s remark in EMI Group Electronics Ltd v Coldicott [2000] 1 WLR 540 at p 556B that the decision in Hunter v Dewhurst was, as it seemed to him, difficult to reconcile with the substitution test. As to that, it is to be mentioned that the EMI case was looked at by us in Fuchs’s case. More importantly, it is to be stressed that both Hunter v Dewhurst and Mairs v Haughey were considered in Fuchs’s case, and we did not regard them to be inconsistent with each other. As was noted in Fuchs’s case at paragraph 16(d), Lord Woolf stressed in Mairs v Haughey that each case ultimately involves applying the statutory language to the facts. What actually happened in Mairs v Haughey was this. Upon the privatisation of a shipyard, the taxpayer received a payment which had two elements. One element was compensation for abrogation of his rights under a pre-existing redundancy scheme. Such compensation was held to be not taxable. The other element was compensation for the taxpayer accepting a new contract with the privatised owners. Such compensation was held to be taxable. Sum D was not from the Taxpayer’s employment 58. Was Sum D from the Taxpayer’s employment? The question is to be answered by looking at substance. As a matter of form, it was described in clause 4.1.4 of the Separation Agreement as being “in lieu of a discretionary bonus” for the financial year ended 30 June 2008. But was it, as a matter of substance, made in substitution, in the Mairs v Haughey sense, for a discretionary bonus which the Taxpayer might have received under the Service Agreement? The Commissioner says that Sum D could not be called a “discretionary bonus” because the bonus decision process had not been completed. Actually that process had not even begun. 59. And of course the process was directed to matters of substance. In that regard, Yuen JA said this (in paragraph 30 of her judgment): “There was no evidence that the Group’s results for that financial year had been considered for the purpose of deciding whether a bonus should be awarded to any of its staff. Nor was there evidence that the Taxpayer’s performance during that financial year had been considered for the purpose of deciding whether a bonus should be awarded to him. There was no evidence that the quantum of Sum D was decided even on a ‘guesstimate’ of what he might have received if a bonus were to be awarded to him.” (Emphasis in the original.) 60. Where bonuses are concerned, an employer’s results and an employee’s performance are both matters of substance. As a matter of substance, Sum D, which was in an arbitrary amount, was of a wholly different nature from any discretionary bonus under the Service Agreement. There is a legitimate reason, indeed a compelling reason, for treating the two differently. 61. There is no - nor could there possibly be any - suggestion that Sum D was paid to induce the Taxpayer to provide future services. Was it paid to reward him for past services? Plainly not. It was paid for something else, namely to make him go away quietly. It was, as Yuen JA said, the “antithesis” to a reward for past services. 62. All of the Commissioner’s arguments on Sum D were presented with skill, but none of them are capable of prevailing. As the Court of Appeal correctly held, Sum D is not taxable. Nor was the Share Option Gain 63. What about the Share Option Gain? There is no - nor could there possibly be any - suggestion that the acceleration of vesting leading to the Share Option Gain was given to induce the Taxpayer to provide future services. The issue is whether that acceleration was given to reward him for past services or was given for something else. 64. Lord Atkin said in Hunter v Dewhurst at p 645 that “a sum of money paid to obtain a release from a contingent liability under a contract of employment cannot be said to be received ‘under’ the contract of employment, is not remuneration for services rendered or to be rendered under the contract of employment, and is not received ‘from’ the contract of employment.” The Commissioner is not unaware of the difficulty in which that statement places him in the present case, and has questioned its correctness. But in doing so he is confronted by an obstacle consisting of that statement having been cited without question in Fuchs’s case at paragraph 21(a). 65. In any event, the Commissioner would still be in difficulty even if one were to say the opposite of what Lord Atkin said, saying instead that a sum of money paid to obtain a release from a contingent liability under a contract of employment can be said to be received “under” the contract of employment and is received “from” the contract of employment. This is because the Employer was not under any liability to accelerate any vesting and the Taxpayer had no right to any acceleration of vesting. 66. Here it is worth saying a word about Commissioner of Inland Revenue v Elliott [2007] 1 HKLRD 297. That case involved an incentive compensation plan (“the ICP”) consisting of units which would yield an income stream. Mr Elliott’s remuneration package included the immediate allotment to him of 5 million units with a promise that further units would be credited to him depending on the progress of certain projects. Less than five months into his employment, Mr Elliott was asked to resign. A termination agreement was entered into. Under that agreement, Mr Elloitt was paid US$ 11 million in consideration of the cancellation of his participation in the ICP scheme. The revenue accepted that such part of this sum as was attributable to the abrogation of Mr Elliott’s contingent right to be credited with ICP units in the future was not taxable. But the revenue maintained that the existing units had been allotted to Mr Elliott as an inducement to take up the employment so that such part of the US$ 11 million as was paid in substitution for the income that he was entitled to under the existing units was income from his employment. That was accepted by the Board of Review and the Court of First Instance. But it was rejected by the Court of Appeal on the basis that on a correct understanding of the ICP scheme, the rights under the existing units were contingent on Mr Elliott remaining in the employment for at least five years and were not enforceable after cessation of the employment. 67. In Fuchs’s case we did not question the correctness of the Court of Appeal’s decision in Elliott’s case. We distinguished it on the basis that while Mr Elliott could not have sued for those rights under his contract of service, what Mr Fuchs received were sums for which he could have sued under his contract of service. In that regard, the Taxpayer’s position was like Mr Elliott’s rather than Mr Fuchs. 68. The following approach was adopted in some of the cases. Sometimes what an employee received was in satisfaction of his rights under his contract of service and is therefore taxable. At other times what an employee received was in abrogation of his rights under his contract of service and is therefore not taxable. Elliott’s case provides an example of payments falling on the non-taxability side of the line. Fuchs’s case, on the other hand, provides an example of payments falling on the taxability side of the line. The view which I take of Elliott’s case now is, I would underline, exactly the same as the view taken of it in Fuchs’s case. 69. Hunter v Dewhurst involved situations on either side of the line. The payment to Commander Dewhurst was made to obtain a release from a contingent liability to him under his contract of employment. So it was not “from” that contract, and was not taxable. The payments to two other directors of the same company were in accordance with its articles of compensation for loss of office. They were not made to abrogate any rights under their contracts of employment. So they were from their contracts of employment, and were taxable. 70. Mairs v Haughey, too, included, as we have already seen, sums on either side of the line. 71. Henley v Murray (1950) 31 TC 351 and Dale v de Soissons [1950] 2 All ER 460 are cases decided by the Court of Appeal in England at about the same time. They illustrate payments falling on different sides of the line. The payment in Henley v Murray fell on the non-taxability side of the line. That was because it was not provided for in the contract of service, and was consideration paid to the employee for the total abrogation of that contract. On the other hand, the payment in Dale v de Soissons fell on the taxability side of the line. That was because the employee surrendered no rights, and got exactly what he was entitled to under his contract of service. 72. In Comptroller-General of Inland Revenue v Knight [1973] AC 428, which was decided by the Privy Council, the payment to the employee who was made redundant fell on the non-taxability side of the line because it was made to compensate him for the abrogation of his contract of service. 73. In addition to the statements of principle cited above, it is of course possible to cite other statements to similar effect. But the statements already cited provide sufficient understanding of the principles to be applied. Likewise, it is of course possible to give further examples of payments falling on either side of the taxability/non-taxability line. But the examples already given provide sufficient illustration of the correct approach. 74. Coming back now to the circumstances of the present case, it is to be borne in mind that the acceleration in question was not acceleration of the time when the Taxpayer would receive something which he would receive, albeit later, even without any acceleration. It was acceleration without which he would never have received that thing at all. On the evidence and the facts found, it is plain that the acceleration of vesting leading to the Share Option Gain was not given to reward the Taxpayer for past services. It was, on the evidence and the facts found, plainly given for something else, being part of what was given to make him go away quietly. In so far as it may be appropriate to approach the present case by reference to abrogation, it is to be observed that the Separation Agreement abrogated whatever rights the Taxpayer may have had under his contract of employment and that he acquiesced in such abrogation in return for what was given to him to make him go away quietly. That applies equally to Sum D and to what led to the Stock Option Gain. 75. Counsel for the Commissioner said that making the Taxpayer go away quietly was merely the Employer’s motive for giving the Taxpayer Sum D and for giving him what led to the Stock Option Gain. “Purpose” rather than “motive” is the appropriate word here. In any event, whatever word one uses, it would go to the question under the Fuchs analysis of whether those things were given to the Taxpayer to reward him for past services or for something else. The Court of Appeal were entitled to conclude that they were given for something else. That conclusion - as we have seen and is worth repeating - does not rest only on that matter of purpose but rests also on the nature of each of those two things. 76. Like his arguments on Sum D, the Commissioner’s arguments on the Stock Option Gain were presented with skill but are incapable of prevailing. Just as the Court of Appeal were correct to hold that Sum D is not taxable, so were they correct to hold that the Stock Option Gain is not taxable. A last minute suggestion 77. At the very end of the hearing before us, when counsel for the Commissioner was addressing us in reply, it was suggested on the Commissioner’s behalf that the Court of Appeal had not identified any error of law on the part of the Board of Review or the Court of First Instance such as would justify intervention by the Court of Appeal. That was the first time that any such suggestion was raised. No such suggestion is covered by the basis on which the Commissioner obtained leave to appeal to us. Nor does any suggestion appear in the Commissioner’s written case. 78. In any event, quite simply, the Court of Appeal’s decision proceeds on an approach to the law different from that applied by the Board of Review and the Court of First Instance. The Court of Appeal was entitled to entertain the appeal. They themselves proceeded on a correct appreciation of the law. And there is nothing to warrant interference by us with the result which they reached by applying the law to the facts. Answer to question on which leave to appeal was granted 79. As for the question on which leave to appeal to us was granted to the Commissioner by the Court of Appeal, the answer to it cannot be more than as follows. Whether such a payment or the gain derived from such a benefit is chargeable to salaries tax depends on whether it is caught by the words of section 8(1) of the Inland Revenue Ordinance. Guidance on how to decide that is to be found in Fuchs’s case. That case illustrates an instance of chargeability. The present case illustrates an instance of non-chargeability. 80. That answer satisfies the reason which the Court of Appeal gave for granting the Commissioner leave to appeal to us. As noted above, the Court of Appeal said that they granted such leave because they took the view that it may be helpful for us to follow up our decision in Fuchs’s case (where the payment was paid under the provisions of the contract) with a decision on the facts of the present case (where the payment was not so paid). Result 81. For the foregoing reasons, with an expression of thanks to counsel on both sides, I would dismiss this appeal with an order nisi awarding costs, to be taxed if not agreed, here and below to the Taxpayer, such order to become absolute unless within 21 days of today the Commissioner lodges with the Court and serves on the Taxpayer written submissions seeking some other order as to costs, the Taxpayer to file written submissions in opposition within 21 days of such service (if any). Lord Neuberger of Abbotsbury NPJ: 82. I agree with the judgment of Mr Justice Bokhary NPJ. Mr Justice Ribeiro PJ: 83. The Court unanimously dismisses the appeal and makes orders referred to by Mr Justice Bokhary NPJ at paragraph 81 above. Mr Stewart Wong SC and Mr Julian Lam, instructed by the Department of Justice, for the Appellant Mr Adrian Huggins SC, instructed by King & Wood Mallesons, for the Respondent 
Chief Justice Ma: 1. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Ribeiro PJ: 2. This appeal raises issues concerning the principles applicable where a party seeks leave to resist enforcement of a New York Convention arbitration award out of time. 3. The eight respondent companies, members of a Malaysian media group conveniently referred to as “Astro”, were the claimants in the arbitration. The 1st to 5th, 7th and 8th respondents are subsidiaries of the 6th respondent, a substantial company listed on the Kuala Lumpur Stock Exchange. 4. The appellant (“First Media”) is a substantial company listed on the Indonesian Stock Exchange and part of an Indonesian conglomerate referred to as “Lippo”. It was a respondent in the arbitration and Astro seeks to enforce an award against it in a sum exceeding US$130 million. First Media seeks leave to resist enforcement out of time on the basis that the award was made without jurisdiction. A. The underlying dispute and the arbitration 5. By a Subscription and Shareholders’ Agreement (“the SSA”) dated 11 March 2005, Lippo companies including First Media entered into a joint venture with companies in Astro (originally consisting of the 3rd to 5th respondents and then, by novation, the 1st and 2nd respondents) for the provision of multimedia and television services in Indonesia. The joint venture vehicle was to be “Direct Vision” which was the 3rd respondent in the arbitration. 6. The arbitration agreement is contained in the SSA and provides for arbitration through the Singapore International Arbitration Centre (“SIAC”), applying Singapore law. However, the 6th, 7th and 8th respondents were never parties to the SSA. They have been referred to throughout as “the Additional Parties”. 7. The joint venture failed because certain conditions precedent were not fulfilled. However, in the meantime, Direct Vision had pressed ahead with the commercial launch of its pay satellite television service in Indonesia and between about December 2005 and October 2008, the Additional Parties had been providing Direct Vision with substantial funds and services. The breakdown of the joint venture led Lippo to commence court proceedings against Astro in Indonesia, alleging that Astro was obliged to continue the funding and support services under an oral joint venture contract. Astro’s riposte was to commence the arbitration against Lippo, including First Media, at SIAC by notice dated 6 October 2008, seeking an anti-suit injunction to restrain the Indonesian proceedings and advancing monetary claims, inter alia, by way of restitution and quantum meruit. 8. Astro applied to join the Additional Parties (who had the main monetary claims) to the arbitration relying on rule 24(b) of the 2007 SIAC Rules.[1] Such joinder was unsuccessfully resisted by Lippo before the arbitral tribunal.[2] By its Award on Preliminary Issues dated 7 May 2009, the tribunal ruled that on the true construction of rule 24(b), it had power to join persons, such as the Additional Parties, who were not already parties to the agreement to refer the dispute to arbitration.[3] Lippo could have, but did not, challenge that award in the Singapore Court which had supervisory jurisdiction. 9. The arbitration then proceeded on the merits, with the tribunal rendering four additional awards, including an Interim Final Award dated 16 February 2010 in favour of Astro in a sum exceeding US$130 million. Lippo again did not seek to challenge the validity of those awards in the Singapore Court. B. Astro’s enforcement of the awards 10. Astro then proceeded to seek enforcement of the awards, principally against First Media, in various jurisdictions including Singapore and Hong Kong. 11. In Singapore, Astro was initially granted leave to enforce the awards but First Media succeeded on its appeal to the Singapore Court of Appeal which held, by a judgment dated 31 October 2013 (“the SCA Judgment”),[4] that rule 24(b) did not empower the tribunal to order joinder of the Additional Parties since they were not parties to the SSA. The tribunal therefore lacked jurisdiction to make the awards in favour of the Additional Parties and the Singapore enforcement orders in their favour were set aside. 12. On 3 August and 9 September 2010, at about the same time as the enforcement proceedings were commenced in Singapore, Astro obtained orders from Saunders J granting them leave to enforce the tribunal’s awards in Hong Kong against the Lippo parties pursuant to section 2GG[5] of the Arbitration Ordinance.[6] 13. In accordance with Order 73 r 10(6),[7] Lippo had 14 days after service of those orders to apply to set them aside. Believing that they did not have any assets in Hong Kong, Lippo made no such application and, on 9 December 2010, Saunders J entered judgment against them in terms of the awards. 14. However, on 22 July 2011, Astro obtained a garnishee order nisi attaching a debt of US$44 million due from AcrossAsia Limited (“AAL”) to First Media. AAL is a Cayman Islands company listed on the Growth Enterprise Market of the Stock Exchange of Hong Kong and holds 55.1% of the issued shares, and thus a controlling interest, in First Media. The debt arose out of an agreement dated 30 June 2011 whereby First Media granted a loan facility of US$44 million to AAL. When, on 5 August 2011, the garnishee order nisi was served on First Media, AAL filed an affirmation opposing the grant of an order absolute and, on 18 January 2012, First Media took out the summons applying for an extension of time to apply to set aside the Hong Kong enforcement orders and judgment. That summons is at the centre of the present appeal. 15. The SCA Judgment was then pending and, at Astro’s instigation, First Media’s application was stayed to await the Singapore Court of Appeal’s decision. Astro also obtained judgment, dated 31 October 2013,[8] from Deputy High Court Judge Mayo (who took a dim view of what he found to be collusive conduct between First Media and AAL) directing that the garnishee order be made absolute. 16. The SCA Judgment was coincidentally also published on 31 October 2013 and appeals were lodged against Deputy High Court Judge Mayo’s decision by both AAL and First Media. In the light of the SCA Judgment, Mimmie Chan J[9] unconditionally stayed execution of the garnishee order absolute pending determination of First Media’s setting aside application. The Court of Appeal refused Astro leave to appeal against Her Ladyship’s unconditional order[10] and Chow J therefore proceeded to deal substantively with First Media’s summons for leave to make a setting aside application. C. Chow J’s judgment[11] 17. It is common ground that the awards in question are Convention awards.[12] The law’s policy is to aid enforcement of such awards, section 42 making them enforceable, with leave, in the same way as a judgment of the Court. Grounds for refusing (and hence for setting aside) enforcement[13] are strictly limited. Section 44(1) provides that enforcement of a Convention award “shall not be refused except in the cases mentioned in this section”. Accordingly, to succeed in its setting aside application, First Media has to bring itself within one of the cases listed in section 44(2)[14] or section 44(3)[15] (and also persuade the Court that its application should be allowed to proceed although well out of time). 18. Since First Media contends that the awards were made without jurisdiction, the exception which is principally relevant is contained in section 44(2)(b): “Enforcement of a Convention award may be refused if the person against whom it is invoked proves ... (b) that the arbitration agreement was not valid under the law to which the parties subjected it or, failing any indication thereon, under the law of the country where the award was made ...”[16] 19. Chow J upheld the enforcement orders and judgment, rejecting First Media’s application for leave to apply out of time to set them aside on the section 44(2)(b) ground. C.1 The effect of the SCA Judgment 20. His Lordship so decided even though he held that the SCA Judgment had conclusively established that the arbitral tribunal did not have power to join the Additional Parties to the arbitration, that being a matter governed by Singapore law.[17] It follows, so the Singapore Court of Appeal held,[18] that there was no arbitration agreement in existence and thus no valid agreement on which to found the awards. That conclusion, as Lord Collins of Mapesbury JSC pointed out, is in accordance with consistent international practice: “Although article V(1)(a) [of the New York Convention] (and section 103(2)(b) [of the Arbitration Act 1996]) deals expressly only with the case where the arbitration agreement is not valid, the consistent international practice shows that there is no doubt that it also covers the case where a party claims that the agreement is not binding on it because that party was never a party to the arbitration agreement.”[19] 21. Moreover, Chow J held that since the Singapore Court of Appeal was undoubtedly a court of competent jurisdiction which had determined an identical issue between the parties on the merits in a final and conclusive judgment, the SCA Judgment raised an issue estoppel per rem judicatam preventing Astro from denying the absence of a valid arbitration agreement.[20] 22. It follows that if leave had been granted, First Media would have brought itself within section 44(2)(b) as a ground for refusing enforcement. Chow J’s refusal of leave is therefore, as the Judge himself acknowledged, somewhat surprising. He noted that, when refusing Astro leave to appeal against Mimmie Chan J’s grant of an unconditional stay of execution of the garnishee order absolute, the Court of Appeal had commented obiter that it would: “... indeed be remarkable if, despite the Singapore Court of Appeal judgment on the invalidity of arbitration awards, Astro will still be able to enforce a judgment here based on the same arbitration awards that were made without jurisdiction.”[21] 23. Chow J, however, did not consider the SCA Judgment definitive since, he reasoned, questions of enforcement are governed by Hong Kong law, being the law of the forum in which enforcement is sought;[22] our law makes enforcement mandatory unless the case falls within section 44(2) or (3);[23] and refusal of enforcement abroad is not a ground for resisting enforcement of the award in Hong Kong.[24] His Lordship proceeded to refuse leave on two grounds. C.2 The good faith principle 24. First, his Lordship decided that First Media was precluded from relying on section 44(2)(b) on the ground that it was in breach of a duty of good faith. 25. Chow J’s starting-point was that the word “may” in section 44(2) – “Enforcement of a Convention award may be refused” – makes it clear that the Court has a residual discretion to order or uphold an order for enforcement even though the person resisting that order comes within one of the section 44(2) exceptions. 26. The existence of such a residual discretion is supported by the authority of this Court in Hebei Import & Export Corp v Polytek Engineering Co Ltd,[25] where Sir Anthony Mason NPJ endorsed the decision of Kaplan J in China Nanhai Oil Joint Service Corp Shenzhen Branch v Gee Tai Holdings Co Ltd,[26] to that effect. 27. Chow J also relied on those two authorities as establishing that the discretion may be exercised in favour of enforcement where the resisting party (although within a section 44(2) exception) has acted in breach of his duty of good faith, drawing parallels between those two cases and the present. 28. In China Nanhai, having participated in the arbitration and having lost on the merits, the defendant sought to oppose enforcement on the section 44(2)(e) ground, ie, that “the composition of the arbitral authority ... was not in accordance with the agreement of the parties”. It complained that whereas the arbitration agreement provided for arbitration before a CIETAC[27] panel in Beijing, at the plaintiff’s behest, a Shenzhen CIETAC panel had assumed jurisdiction. Kaplan J held that while technically the Shenzhen tribunal lacked jurisdiction, the residual discretion in favour of enforcement ought to be exercised because there was: “... a duty of good faith which in the circumstances of this case required the Defendant to bring to the notice of the full tribunal or the CIETAC Commission in Beijing its objections to the formation of this particular arbitral tribunal. Its failure to do so and its obvious policy of keeping this point up its sleeve to be pulled out only if the arbitration was lost, is not one that I find consistent with the obligation of good faith nor with any notions of justice and fair play.”[28] 29. However, his Lordship regarded the arbitration as one which the parties had in substance agreed to: “The parties agreed on a CIETAC Arbitration under CIETAC Rules. They got it. CIETAC, Shenzhen, is a sub-commission of CIETAC in Beijing. The Defendants participated in the arbitration and have raised no other grounds whatsoever which go to the procedure of the arbitration or the substance of the award. Had they won, they would not have complained. ... I am quite satisfied that the Defendants got what they agreed in their contract in the sense that they got an arbitration conducted by 3 Chinese arbitrators under CIETAC Rules. To exercise my discretion against enforcement on the facts of this case would be a travesty of justice. Had I thought that the Defendants' rights had been violated in any material way, I would, of course, have taken a different view.”[29] 30. The Hebei case involved a CIETAC arbitration regarding the allegedly sub-standard performance of certain equipment supplied under a contract. The complaint was that, without notice to the respondent, an inspection had been conducted by experts appointed by the arbitral tribunal accompanied by the chief arbitrator at the end user’s factory, where technicians who had installed the equipment communicated with the chief arbitrator in the respondent’s absence. Enforcement of the award was resisted on the section 44(2)(c) ground, namely, that the resisting party had been “unable to present his case” and on the section 44(3) ground that enforcement in such circumstances would be contrary to public policy. However, when the respondent discovered that the inspection had occurred it did not raise the matter with the tribunal but continued to participate in the arbitration. It was held that: “The respondent's conduct amounted to a breach of the principle that a party to an arbitration who wishes to rely on a non-compliance with the rules governing an arbitration shall do so promptly and shall not proceed with the arbitration as if there had been compliance, keeping the point up his sleeve for later use.”[30] 31. Chow J considered that the principles established by China Nanhai and Hebei were applicable on the facts of the present case: “In my view, what was considered to be so objectionable in China Nanhai Oil Joint Service Corporation Shenzhen Branch and Hebei Import & Export Corp v Polyteck Engineering Co Ltd was the idea that a party to an arbitration, while being fully aware of an objection (whether in relation to the jurisdiction of the tribunal or the procedure or conduct in the course of the arbitration), should be permitted to keep the objection in reserve, participate fully in the arbitration and raise the objection in the enforcing court only after an award had been made against him by the tribunal. This is effectively what happened in the present case. First Media was fully aware of its right to challenge the Tribunal’s ruling on jurisdiction before the Singapore High Court under article 16(3) of the Model Law, but chose not to do so. It seems clear that what First Media decided to do was to defend the claim on the merits in the hope that it would succeed before the Tribunal, and keep the jurisdictional point in reserve to be deployed in the enforcement court only when it suited its interests to do so.”[31] 32. His Lordship arrived at this conclusion notwithstanding that he: (a) accepted that a party is not obliged to exhaust its remedies by challenging the validity of an award in the courts of the arbitral seat, having a choice between the “active remedy” of making such a challenge and a “passive remedy” of resisting enforcement in the jurisdiction where enforcement is sought;[32] (b) acknowledged that First Media had objected to and expressly reserved its position concerning the tribunal’s jurisdiction both before and after the Award on Preliminary Issues[33] so that it could not be said to have kept the objection “up its sleeve”; and, (c) recognised that the SCA Judgment had conclusively established that the tribunal lacked jurisdiction in respect of the Additional Parties, bringing First Media within section 44(2)(b) and that “it would take a very strong case to permit enforcement of an arbitral award in circumstances where it was made by an arbitral tribunal without jurisdiction”.[34] C.3 Refusal of an extension of time 33. Secondly, Chow J held that even if he had decided not to exercise his residual discretion under section 44(2) in favour of enforcement, he would have refused to extend time for First Media to make its application for leave.[35] 34. The time limit is the limit of 14 days after service of the orders in question, laid down by Order 73 r 10(6).[36] It expired on 1 November 2010 so that First Media’s summons for a time extension, issued on 18 January 2012, was 14 months out of time. The Court’s power to grant an extension is contained in Order 3 r 5 which materially states: “(1) The Court may, on such terms as it thinks just, by order extend or abridge the period within which a person is required or authorized by these rules, or by any judgment, order or direction, to do any act in any proceedings. (2) The Court may extend any such period as is referred to in paragraph (1) although the application for extension is not made until after the expiration of that period.” 35. After having considered authorities on how the discretion should be approached, in particular The Decurion[37] and Terna Bahrain Holding Company WLL v Al Shamsi[38](to which I shall return), Chow J based his refusal of an extension on three factors[39] namely: (a) the fact that the delay of 14 months, viewed in the light of the 14 day limit prescribed by Order 73 r 10(6) was very substantial;[40] (b) the fact that First Media’s delay was a deliberate decision not to take action within the time limit because it thought that it had no assets here, thus taking a “calculated risk”;[41] and, (c) the fact that the awards had not been set aside in Singapore and thus were “still valid and create legally binding obligations on First Media to satisfy them”, so that refusing an extension of time “merely means that Astro is permitted to obtain satisfaction of a legally binding debt due and owing by First Media to Astro”.[42] 36. His Lordship refused an extension even though he accepted “that Astro has not suffered any substantial prejudice (other than costs which can be compensated) as a result of First Media’s delay of 14 months”.[43] D. The Court of Appeal’s judgment[44] D.1 Good faith and the residual discretion 37. The Court of Appeal overturned the Judge’s decision based on the good faith principle. 38. Kwan JA considered the conclusive judgment of the Singapore Court of Appeal (which she referred to as the supervisory court) as to the invalidity of the arbitration agreement “a very strong policy consideration” in the Hong Kong courts for section 44(2) purposes.[45] 39. The Judge had found it objectionable (relying on China Nanhai and Hebei) that First Media, while “fully aware of its right to challenge the Tribunal’s ruling on jurisdiction before the Singapore High Court” chose not to do so, but “decided ... to defend the claim on the merits in the hope that it would succeed before the Tribunal … [keeping] the jurisdictional point in reserve to be deployed in the enforcement court”,[46] but her Ladyship pointed out that First Media was fully entitled under Singapore law to follow that course. The SCA Judgment had held that a challenge to the tribunal’s preliminary ruling was not a “one-shot remedy” and did not affect the availability of defences at the enforcement stage.[47] It had also found that First Media had not waived its rights, but had disputed the tribunal’s jurisdiction and then proceeded with the arbitration reserving its position.[48] Chow J was thus held by the Court of Appeal to have fallen into error “in not giving proper recognition to the findings in the SCA Judgment.”[49] 40. The SCA Judgment is undoubtedly of central importance because it conclusively determined, as a matter governed by Singapore law, that there was no valid arbitration agreement between First Media and the Additional Parties and raised an issue estoppel to that effect. The Singapore Court of Appeal was, however, acting as an enforcement court and not invalidating the awards in its supervisory capacity. The proposition that First Media’s adoption of a passive remedy was in accordance with Singapore law is, with respect, not to the point since the availability of enforcement in Hong Kong is governed by the law of this forum. The true issue was whether, as a matter of Hong Kong law, Chow J was right to find that First Media was in breach of its duty of good faith so as to justify precluding it from relying on section 44(2)(b). Many of the points made by the Court of Appeal as to the consequences of the SCA Judgment are, however, equally relevant to the position under Hong Kong law. 41. In any event, Kwan JA’s rejection of the Judge’s application of the good faith principle is soundly based on her Ladyship’s finding that, when exercising his discretion, Chow J failed to take sufficient account of “the fundamental defect that the Awards were sought to be enforced against the Additional Parties who were wrongly joined by the Tribunal into the Arbitration and the Awards were made without jurisdiction”.[50] 42. It is clear that the absence of a valid arbitration agreement between the parties is a fundamentally important factor militating against discretionary enforcement. Thus, referring to equivalent provisions, Rix LJ stated in the English Court of Appeal in Dallah Real Estate and Tourism Holding Co v Ministry of Religious Affairs of the Government of Pakistan as follows:[51] “... There is no express provision however as to what is to happen if a defence is proven, but the strong inference is that a proven defence is a defence. It is possible to see that a defence allowed under Convention or statute may nevertheless no longer be open because of an estoppel (Professor van den Berg's view, see The New York Arbitration Convention 1958, at p 265), or that a minor and prejudicially irrelevant error, albeit within the Convention or statutory language, might not succeed as a defence (as in the China Agribusiness case [1998] 2 Lloyd's Rep 76). But it is difficult to think that anything as fundamental as the absence of consent or some substantial and material unfairness in the arbitral proceedings could leave it open to a court to ignore the proven defence and instead decide in favour of enforcement.” 43. And in the UK Supreme Court in the same case, Lord Collins of Mapesbury JSC pointed out that “... article V[52] safeguards fundamental rights including the right of a party which has not agreed to arbitration to object to the jurisdiction of the tribunal”.[53] He acknowledged that because of the word “may”: “The court before which recognition or enforcement is sought has a discretion to recognise or enforce even if the party resisting recognition or enforcement has proved that there was no valid arbitration agreement.”[54] However, his Lordship emphasised the limits of that discretion, observing that “it is not easy to see how that could apply to a case where a party had not acceded to an arbitration agreement.”[55] 44. In the light of such authority, at paragraph 59 of its judgment the Court of Appeal, in my view correctly, held as follows: “The judge had misdirected himself and failed to take into account the fundamental defect that the Awards were sought to be enforced against the Additional Parties who were wrongly joined by the Tribunal into the Arbitration and the Awards were made without jurisdiction when he exercised his discretion under section 44(2) whether to refuse enforcement. Had he taken this into account, he could only have exercised his discretion to refuse enforcement.” D.2 Extension of time 45. Given the forceful statement in paragraph 59, one might have expected the Court of Appeal to allow the appeal and grant First Media leave to make its setting aside application. However, instead, it upheld Chow J’s refusal of a time extension and dismissed the appeal. 46. The Court of Appeal declined to interfere with Chow J’s exercise of discretion and endorsed his reliance on the three factors mentioned above, namely, (i) the length of the delay, (ii) the fact that a deliberate decision was taken not to apply to set aside within the time prescribed and (iii) the fact that the awards had not been set aside at the seat of the arbitration. 47. The Court rejected First Media’s argument that factor (iii) was an irrelevant matter erroneously taken into account and also disagreed with the submission that the Judge’s exercise of discretion, guided largely by the decision of Popplewell J in Terna Bahrain Holding Company WLL v Al Shamsi,[56] was inappropriate and plainly wrong. 48. The Appeal Committee granted leave to appeal to this Court on both the question of law and the “or otherwise” bases.[57] The questions of law are formulated as follows: (1) What is the proper test for determining whether an extension of time should be granted for the purposes of an application to resist enforcement of an arbitral award under the 1958 New York Convention on the Recognition and Enforcement of Foreign Arbitral Awards (“the New York Convention”)? (“Question 1”) (2) In determining whether to extend time for the purposes of an application to resist enforcement of an arbitral award under the New York Convention, is the fact that the award has not been set aside by the courts of the seat of arbitration a relevant factor? (“Question 2”) E. This appeal 49. As those questions indicate, the issue before this Court concerns the refusal of a time extension. Although the good faith ground is no longer advanced by Astro, aspects of why that ground was rejected provide a context relevant to considering the refusal to extend time in the Courts below. E.1 Appellate review of discretion 50. The grant or refusal of a time extension is of course discretionary and the principles as to when an appellate court may interfere with a discretionary decision are conveniently summarised by Yuen JA in the Court of Appeal in Tai Fook Futures v Cheung Moon Hoi Jeff[58] as follows: “It is well-established law that an appellate court should not interfere with the exercise of a judge’s discretion unless it is satisfied that the judge has erred in law or in principle, or if she has taken into account some matter which she should not have taken into account, or has left out of account some matter which she should have taken into account, or if the decision was so plainly wrong that it must have been reached by a faulty assessment of the weights of the different factors which have to be taken into account ...” 51. Mr Toby Landau QC[59] submits that (i) as presaged in Question 1, the Judge and the Court of Appeal erred in principle in applying the wrong test when exercising the discretion; (ii) that, as anticipated in Question 2, they erroneously took into account an irrelevant factor, namely, the fact that the award has not been set aside by the Court at the seat; and (iii) that looked at overall, the refusal was plainly wrong, being perverse and disproportionate in its consequences. 52. Mr David Joseph QC[60] seeks to uphold the Court of Appeal’s decision as one made within the proper bounds of its discretion and seeks to rely on a line of cases concerning applications for relief from sanctions under CPR r 3.9(1) of the English Civil Procedure Rules, not previously explored. E.2 The appropriate test 53. It has often been emphasised that the discretion to extend time conferred by Order 3 r 5 is broad and unrestricted, designed to enable justice to be done between the parties. Thus, in Kwan Lee Construction Co Ltd v Elevator Parts Engineering Co Ltd,[61] Litton VP in the Court of Appeal, stated: “The court's jurisdiction to extend time, as conferred by O 3 r 5, is as broad as it can come and, in the exercise of that discretion, the court would, generally speaking, have some regard to what might ultimately be in issue.” 54. In Costellow v Somerset County Council,[62]dealing with the equivalent provision in England and Wales, Sir Thomas Bingham MR noted that the discretion involves the intersection of two principles. The first promotes the enforcement of time limits for the expeditious dispatch of litigation in the public interest and the second recognises that a plaintiff should not ordinarily be denied adjudication of his claim on the merits because of a procedural default “unless the default causes prejudice to his opponent for which an award of costs cannot compensate”. His Lordship noted that the second principle “is reflected in the general discretion to extend time conferred by Ord 3, r 5, a discretion to be exercised in accordance with the requirements of justice in the particular case.” 55. The approach advocated on First Media’s behalf as formulated by the Court of Appeal in The Decurion,[63] is in line with the foregoing authorities. Citing Costellow, Cheung JA acknowledged the intersecting principles and stated: “It is clear that the applicable principle in deciding whether time should be extended is to look at all relevant matters and consider the overall justice of the case. A rigid mechanistic approach is not appropriate...”[64] 56. Although Chow J and the Court of Appeal cited The Decurion, it is clear that they laid primary emphasis on Terna Bahrain Holding Company WLL v Al Shamsi,[65] quoting extensively from the judgment of Popplewell J.[66] In Terna Bahrain,the respondent in a London arbitration sought an extension of time to challenge the validity of the award under sections 67 and 68 of the Arbitration Act 1996 on the footing that the award had been made on a basis not advanced by the claimant so that the award was vitiated by serious irregularity and lack of jurisdiction. 57. The focus of Mr Landau QC’s criticism is on the passages in Popplewell J’s judgment in which, having cited certain authorities for the applicable principles, his Lordship stated: “(1) Section 70(3) of the Act requires challenges to an award under sections 67 and 68 to be brought within 28 days. This relatively short period of time reflects the principle of speedy finality which underpins the Act, and which is enshrined in section 1(a). The party seeking an extension must therefore show that the interests of justice require an exceptional departure from the timetable laid down by the Act. Any significant delay beyond 28 days is to be regarded as inimical to the policy of the Act. (2) The relevant factors are: (i) the length of the delay; (ii) whether the party who permitted the time limit to expire and subsequently delayed was acting reasonably in the circumstances in doing so; (iii) whether the respondent to the application or the arbitrator caused or contributed to the delay; (iv) whether the respondent to the application would by reason of the delay suffer irremediable prejudice in addition to the mere loss of time if the application were permitted to proceed; (v) whether the arbitration has continued during the period of delay and, if so, what impact on the progress of the arbitration, or the costs incurred in respect of the arbitration, the determination of the application by the court might now have; (vi) the strength of the application; (vii) whether in the broadest sense it would be unfair to the applicant for him to be denied the opportunity of having the application determined. (3) Factors (i), (ii), and (iii) are the primary factors.”[67] 58. Popplewell J added four observations: (i) that “the length of delay must be judged against the yardstick of 28 days provided for in the Act. Therefore a delay measured even in days is significant; a delay measured in many weeks or in months is substantial;”[68] (ii) that reasons given for the delay had to be supported by evidence;[69] (iii) that in the light of CPR r 3.9(1) which identifies an intentional failure as a separate factor, an intentional failure to comply militates against a finding that the applicant acted reasonably;[70] and (iv) that since the Court will not normally conduct a substantial investigation into the merits, they are of secondary significance: “Unless the challenge can be seen to be either strong or intrinsically weak on a brief perusal of the grounds, this will not be a factor which is treated as of weight in either direction on the application for an extension of time. If it can readily be seen to be either strong or weak, that is a relevant factor; but it is not a primary factor, because the court is only able to form a provisional view of the merits, a view which might not be confirmed by a full investigation of the challenge, with the benefit of the argument which would take place at the hearing of the application itself if an extension of time were granted.”[71] 59. The elaborately structured approach to discretion in Terna Bahrain is qualitatively different from the broad, unrestricted approach espoused in cases like The Decurion. 60. Popplewell J’s reference to CPR r 3.9(1)[72] is instructive since that is a provision addressing applications for relief from a sanction imposed for failure to comply with any rule, practice direction or court order. Those are situations in which, as one might expect, a substantial burden is placed on an applicant seeking relief from an applicable sanction, and where features such as intentional non-compliance with the rule or order weigh heavily against relief. Given this perspective, it is unsurprising that Popplewell J’s focus was not so much on finding a balance between intersecting policy principles but on the enforcement of procedural orders which have triggered sanctions against the applicant. Thus, his Lordship requires the party seeking an extension to “show that the interests of justice require an exceptional departure from the timetable laid down by the Act”, treating the grant of an extension as “exceptional”. 61. His Lordship’s treatment of the merits is also important. In paragraph (3) cited above, he states that items (i), (ii) and (iii) “are the primary factors”. These comprise the length of delay; the reasonableness of allowing the time limit to expire and whether the other side or the arbitrator contributed to the delay. The strength of the application and fairness to the applicant are only mentioned at the end of the list as items (vi) and (vii) and are thus treated as secondary considerations. As indicated in Popplewell J’s fourth observation, this is because his Lordship’s view was that an investigation of the merits is not generally undertaken so that the court is only able to form a provisional rather than a concluded view of the merits. 62. The Terna Bahrain approach of promoting the importance of certain factors and according to others (including the merits) a secondary status is plainly inconsistent with the The Decurion approach of looking at all relevant matters and considering the overall justice of the case, eschewing a rigid, mechanistic methodology. The downgrading of the merits as a factor is particularly inapt in a case like the present, where the tribunal’s lack of jurisdiction has been conclusively shown. 63. Mr Landau QC convincingly submits that the Judge faithfully applied the approach of Popplewell J set out in the passages cited above, focussing especially on items (i) and (ii) of the listed principles.[73] While it is true, as Mr Joseph QC points out, that Popplewell J did refer to cases where the merits “may be a powerful factor in favour of the grant of an extension” where “the court can determine that the challenge will succeed”,[74] Chow J did not give any weight to, or even mention in this context, the conclusively established merits. 64. It follows, in my judgment, that the test in Terna Bahrain was inappropriately applied in the present case. It led to a failure to accord proper weight to the established lack of a valid arbitration agreement which, if recognised, would have wholly undermined the central arguments made on Astro’s behalf. Thus, much emphasis was laid on section 2AA(1) which states that the object of the Arbitration Ordinance is “to facilitate the fair and speedy resolution of disputes by arbitration without unnecessary expense”. This led the Court of Appeal to state as follows: “As rightly submitted by Mr Joseph, a more disciplined approach is called for in the arbitration context, with its emphasis on speedy finality and the short statutory time limits. It is accordingly in an ‘entirely different territory’ from applications for extensions of time for compliance with interlocutory orders or rules applying during the currency of a case (Soinco Saci v Novokuznetsk Aluminium Plant [1998] 2 Lloyd’s Rep 337 at 338, per Waller LJ).”[75] 65. The policy favouring speedy finality in resolving an arbitration is necessarily premised on a valid arbitration agreement between the parties which is absent in the present case. The trite proposition that binding arbitrations must rest on a consensual basis is reflected in section 2AA itself which, in subsection (2)(a) states: “(2) This Ordinance is based on the principles that- (a) subject to the observance of such safeguards as are necessary in the public interest, the parties to a dispute should be free to agree how the dispute should be resolved...” 66. Soinco Saci v Novokuznetsk Aluminium Plant,[76] cited by the Court of Appeal, is not a case where a valid arbitration agreement was lacking. The challenge to enforcement was there based on the argument (held to be unsustainable) that it would be contrary to public policy to enforce the award. In any event, Waller LJ’s approach differed significantly from that of Terna Bahrain in that the strength of the applicant’s case was given equal prominence with other discretionary factors comprising the extent of and excuse for the delay, and the degree of prejudice to the respondent. 67. Terna Bahrain involved an active remedy challenge to the validity of the award, launched before the Court with supervisory jurisdiction in London. There is force in Mr Landau QC’s submission that, unlike challenges in an enforcement forum other than the seat, such a challenge involves invoking the procedural rules contractually agreed upon by the parties, and may justify a stricter approach to procedural time limits. 68. For the foregoing reasons, I conclude that in adopting the Terna Bahrain approach, the Courts below erred in principle, leading them to downgrade the fundamentally important absence of a valid arbitration agreement between First Media and the Additional Parties. They thereby failed to take proper account of a relevant matter, justifying this Court’s interference with their exercise of discretion. 69. Before leaving this part of the discussion, I note that Mr Joseph QC sought to rely on a line of English authority on the application of CPR r 3.9(1) culminating in Denton v TH White Ltd.[77] That rule provides: “On an application for relief from any sanction imposed for a failure to comply with any rule, practice direction or court order, the court will consider all the circumstances of the case, so as to enable it to deal justly with the application, including the need— (a) for litigation to be conducted efficiently and at proportionate cost; and (b) to enforce compliance with rules, practice directions and orders.” 70. It is a procedural rule which makes no mention of the substantive merits, setting out an approach to applications for relief from sanctions imposed for non-compliance with rules or orders. I do not accept Mr Joseph QC’s submission that First Media’s application “is as a matter of substance analogous to” such applications for relief. Thus, Denton was a consolidated appeal before the English Court of Appeal involving three cases which, respectively, concerned (i) service of witness statements outside time limits set at a case management conference, carrying the automatic sanction of prohibiting the proposed witness from being called; (ii) failing to comply in time with a pre-trial checklist setting a deadline for payment of court fees with an unless order for automatically striking out the claim in default; and (iii) filing a costs budget form 45 minutes late, carrying the automatic sanction of restricting the applicable cost budget. It is obvious that these situations are miles away from the circumstances under discussion. There is no question of the discretion exercised in such cases being overshadowed by a lack of jurisdiction to make the awards upon which the challenged orders are based. 71. Even at a purely procedural level, cases in the Denton line (in which the parties’ substantive rights do not feature) sit uncomfortably with the procedural regime in this jurisdiction. Order 1A r 2(2) provides that: “In giving effect to the underlying objectives of these rules, the Court shall always recognize that the primary aim in exercising the powers of the Court is to secure the just resolution of disputes in accordance with the substantive rights of the parties.” This Court in Wing Fai Construction Co Ltd v Yip Kwong Robert, reiterated the importance of the primary aim of securing the just resolution of disputes in accordance with the parties’ substantive rights, stressing that compliance with the rules is not an end in itself[78] and that a mechanistic approach is to be eschewed.[79] The overriding objective in the English CPR r 1.1 emphasises procedural fairness and economy rather than the parties’ substantive rights. E.3 Deliberate failure to set aside the awards at the arbitral seat 72. The third discretionary factor relied on for refusing a time extension by Chow J and upheld by the Court of Appeal involved the fact that the awards had not been set aside in Singapore and thus were “still valid and create legally binding obligations on First Media to satisfy them”.[80] This is closely linked to Chow J’s second factor, which is that First Media’s delay involved a deliberate decision not to take action to set aside the orders and judgment within time. 73. First Media’s submission, based principally on the “choice of remedies” doctrine, is that, in giving weight to those two factors, the Courts below erred in principle and took account of irrelevant factors. 74. In Paklito Investment Ltd v Klockner East Asia Ltd,[81] Kaplan J referred to the choice of remedies doctrine as follows: “It is clear to me that a party faced with a Convention award against him has two options. Firstly, he can apply to the courts of the country where the award was made to seek the setting aside of the award. If the award is set aside then this becomes a ground in itself for opposing enforcement under the Convention. Secondly, the unsuccessful party can decide to take no steps to set aside the award but wait until enforcement is sought and attempt to establish a Convention ground of opposition.” 75. Those options are mirrored in section 44(2) itself. Thus, where a party opts to set aside the award in the courts of the seat and succeeds in doing so, it acquires a defence against enforcement under section 44(2)(f) which covers cases where the award has been set aside by a competent authority of the country in which, or under the law of which, it was made. The other option is to resist enforcement on other grounds, including section 44(2)(b), without having taken steps to set aside the awards in the supervisory court. They are options which are independently available. 76. As Lord Collins of Mapesbury pointed out in Dallah Real Estate and Tourism Holding Co v Ministry of Religious Affairs of the Government of Pakistan:[82] “There is nothing in the Convention which imposes an obligation on a party seeking to resist an award on the ground of the non-existence of an arbitration agreement to challenge the award before the courts of the seat.” 77. In the Hebei case,[83]Sir Anthony Mason NPJ points out that even where the supervisory court has held that the awards are valid, it would be open to the Hong Kong court in the enforcement forum to refuse enforcement, for example, on the ground of a differing Hong Kong public policy: “The Convention distinguishes between proceedings to set aside an award in the court of supervisory jurisdiction (arts VI and VI(e)) and proceedings in the court of enforcement (art VI). Proceedings to set aside are governed by the law under which the award was made or the law of the place where it was made, while proceedings in the court of enforcement are governed by the law of that forum. The Convention, in providing that enforcement of an award may be resisted on certain specified grounds, recognises that, although an award may be valid by the law of the place where it is made, its making may be attended by such a grave departure from basic concepts of justice as applied by the court of enforcement that the award should not be enforced. It follows, in my view, that it would be inconsistent with the principles on which the Convention is based to hold that the refusal by a court of supervisory jurisdiction to set aside an award debars an unsuccessful applicant from resisting enforcement of the award in the court of enforcement.” 78. Section 44(2) is therefore consonant with the choice of remedies principle and enables the party concerned to resist enforcement in Hong Kong without having challenged the awards in the supervisory court. It follows that the decisions of the Courts below to treat the fact that the awards have not been set aside in Singapore as a major factor in refusing a time extension come into conflict with the choice of remedies principle. 79. Moreover, since the doctrine admits of (and indeed presupposes) a choice being made between an active or passive remedy, to hold it against First Media that it made a deliberate choice in favour of a passive remedy also conflicts with the choice of remedies principle. 80. Respecting that principle, Chow J’s second and third factors should not have been taken into account and, in embracing them, the discretionary exercise in the Courts below miscarried. 81. The decision not to embark upon a setting aside application within the 14 day time limit (which Mr Joseph QC submitted was incumbent on First Media) when there were then no assets in Hong Kong was entirely reasonable, particularly where the tribunal’s jurisdiction had been challenged and the right to bring further challenges was expressly reserved. 82. Mr Joseph QC sought to submit that the continued existence of the Singapore awards is a relevant discretionary factor to be taken into account because, as he puts it, they constitute “documents of title” creating legally binding debts which Astro is entitled to enforce.[84] The Court of Appeal sought to buttress that proposition by reference to section 42(2), stating: “... unless an award is set aside, it is treated as binding for all purposes between the parties as between whom it is made, see section 42(2) of the Ordinance.”[85] 83. In my view, section 42(2) does not assist. It relevantly provides: “Any Convention award which would be enforceable under this Part shall be treated as binding for all purposes on the persons as between whom it was made, and may accordingly be relied on by any of those persons by way of defence, set off or otherwise in any legal proceedings in Hong Kong...” 84. The binding quality of the award therefore depends on whether it would be enforceable under the Ordinance. Such enforceability is of course the very issue between the parties and section 42(2) does not help to resolve it. To argue, as Mr Joseph QC does, that the awards are relevant because they create enforceable debts begs the very question of their enforceability. F. Proper exercise of the discretion 85. For the foregoing reasons, this Court is entitled and bound to set aside the decisions below and to exercise the discretion under Order 3 r 5 afresh, looking at all relevant matters and considering the overall justice of the case. 86. Having discarded two of the three factors given weight by Chow J and the Court of Appeal, one is left only with the factor of the 14 month delay as a possible basis for refusing an extension. Viewed against the 14 day limit prescribed by Order 73 r 10(6), that period is obviously substantial. However, as Chow J accepted, “Astro has not suffered any substantial prejudice (other than costs which can be compensated) as a result of First Media’s delay of 14 months”.[86] Moreover, it is clear that Astro did not feel the need to press on urgently with the litigation since it was on Astro’s application that the hearing of First Media’s summons was stayed to await the Singapore Court of Appeal’s decision, involving a delay of some 20 months. 87. There must be balanced against the 14 month delay, the fundamentally important absence of a valid arbitration agreement between First Media and the Additional Parties, so that those parties were wrongly joined and the tribunal’s awards were made in their favour without jurisdiction. Thus if an extension of time is granted, the section 44(2)(b) defence against enforcement will clearly be available, there being no basis for precluding its operation. To refuse an extension would be to deny First Media a hearing where its application has decisively strong merits and would involve penalising it for a delay which caused Astro no uncompensable prejudice to the extent of permitting enforcement of an award for US$130 million. That would self-evidently be wholly disproportionate. 88. In my view, the Court’s discretion can only properly be exercised by setting aside the decisions below and granting the appellant an extension of time. Mr Joseph QC submitted that if the Court were minded to grant such an extension, it should do so only on terms that First Media makes full payment in satisfaction of all outstanding costs orders. That submission cannot be accepted. It was advanced without prior notice to the other side; without identifying what, if any, costs orders have been left unpaid; without considering whether any such orders may be subject to appeal; and without taking into account any available set-offs. G. Disposal of this appeal 89. I would accordingly allow the appeal and grant the appellant an extension of time to apply to set aside the orders granting leave to enforce the awards and the judgment entered on the strength of those orders. 90. My answer to Question 1 is that the proper test involves looking at all relevant matters and considering the overall justice of the case, eschewing a rigid mechanistic approach, as indicated in The Decurion. And my answer to Question 2 is “No” except in cases where section 44(2)(f) is relied on. It is unnecessary to consider the “or otherwise” ground. 91. I would make an order nisi that the respondents pay to the appellant the costs of the appeals in this Court and in the Court of Appeal. Since it was necessary for the appellant to apply to Chow J for an extension of time, it must bear to some degree the costs of seeking the Court’s indulgence. But since it has succeeded in showing that his Lordship’s exercise of discretion miscarried on both grounds relied on when refusing them an extension, I would set aside his order as to costs and make an order nisi that the appellant pay half of the respondents’ costs before Chow J. 92. If a different order on costs is sought, written submissions should be lodged with the Registrar and served on the other parties within four weeks of the handing down of this judgment with liberty to the other parties to serve written submissions in reply within four weeks thereafter. If no submissions are lodged or served within the initial four-week period, the orders nisi shall stand as orders absolute without further direction. Mr Justice Tang PJ: 93. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Fok PJ: 94. I agree with the judgment of Mr Justice Ribeiro PJ. Lord Reed NPJ: 95. I agree with the judgment of Mr Justice Ribeiro PJ. Chief Justice Ma: 96. The Court unanimously allows the appeal, sets aside the orders of the Courts below and directs that time for the appellant to file its application for leave to set aside the orders granting the respondents leave to enforce the awards and to enter judgment thereon be extended for three months from the date of the handing down of this judgment. The Court also makes the orders as to costs set out in paragraphs 91 and 92 above. Mr Toby Landau, QC, Mr Mark Strachan SC and Mr Jeffery Chau, instructed by Cordells Rompotis, for the Appellant Mr David Joseph QC, Mr Bernard Man SC and Mr Justin Ho, instructed by Clifford Chance, for the Respondents [1] Rule 24(b): “In addition and not in derogation of the powers conferred by any applicable law of the arbitration, the Tribunal shall have the power to:... (b) allow other parties to be joined in the arbitration with their express consent, and make a single final award determining all disputes among the parties to the arbitration”. [2] Comprising Sir Gordon Langley, Sir Simon Tuckey and Stewart C Boyd CBE QC. [3] Award on Preliminary Issues, §97. [4] PT First Media TBK (formerly known as PT Broadband Multimedia TBK) v Astro Nusantara International BV and others and another appeal [2013] SGCA 57, Sundaresh Menon CJ giving the judgment of the Court. [5] Section 2GG: “(1) An award, order or direction made or given in or in relation to arbitration proceedings by an arbitral tribunal is enforceable in the same way as a judgment, order or direction of the Court that has the same effect, but only with the leave of the Court or a judge of the Court. If that leave is given, the Court or judge may enter judgment in terms of the award, order or direction. (2) Notwithstanding anything in this Ordinance, this section applies to an award, order or direction made or given whether in or outside Hong Kong.” The awards being Convention awards, under section 42, they are enforceable in the same manner as under section 2GG. [6] References to sections in this judgment are (unless otherwise indicated) to provisions of the Arbitration Ordinance (Cap 341) applicable by virtue of transitional provisions in Schedule 3 para 1 of the Arbitration Ordinance (Cap 609) which has since replaced Cap 341. [7] Rules of the High Court (Cap 4A). O 73 r 10(6): “Within 14 days after service of the order ... the debtor may apply to set aside the order and the settlement agreement or award shall not be enforced until after the expiration of that period or, if the debtor applies within that period to set aside the order, until after the application is finally disposed of.” [8] HCCT 45/2010 (31 October 2013). [9] HCCT 45/2010 (24 January 2014). [10] Cheung and Barma JJA, HCMP 835/2014 (25 June 2014). [11] HCCT 45/2010 (17 February 2015). [12] Awards “made in pursuance of an arbitration agreement in a State or territory, other than China or any part thereof, which is a party to the New York Convention”, ie, the Convention on the Recognition and Enforcement of Foreign Arbitral Awards adopted by the United Nations Conference on International Commercial Arbitration on 10 June 1958 the text of which is set out in the Third Schedule of the Ordinance: see section 2. Singapore acceded to the Convention on 21 August 1986. [13] Reflecting Article V of the New York Convention. [14] Section 44(2): “Enforcement of a Convention award may be refused if the person against whom it is invoked proves – (a) that a party to the arbitration agreement was (under the law applicable to him) under some incapacity; or (b) that the arbitration agreement was not valid under the law to which the parties subjected it or, failing any indication thereon, under the law of the country where the award was made; or (c) that he was not given proper notice of the appointment of the arbitrator or of the arbitration proceedings or was otherwise unable to present his case; or (d) subject to subsection (4), that the award deals with a difference not contemplated by or not falling within the terms of the submission to arbitration or contains decisions on matters beyond the scope of the submission to arbitration; or (e) that the composition of the arbitral authority or the arbitral procedure was not in accordance with the agreement of the parties or, failing such agreement, with the law of the country where the arbitration took place; or (f) that the award has not yet become binding on the parties, or has been set aside or suspended by a competent authority of the country in which, or under the law of which, it was made.” [15] Section 44(3): “Enforcement of a Convention award may also be refused if the award is in respect of a matter which is not capable of settlement by arbitration, or if it would be contrary to public policy to enforce the award.” [16] It is also arguable that the exception set out in section 44(2)(d) is relevant. The focus of the argument has, however, been on section 44(2)(b). [17] Judgment §95, citing Guangzhou Green‑Enhan Bio‑Engineering Co Ltd v Green Power Health Products International Co Ltd [2004] 4 HKC 163. [18] SCA Judgment §§152-158, §162-164. The applicable Singapore provisions, namely, Article 36(1)(a)(i) and (iii) of the Model Law given statutory effect by section 3(1) of the International Arbitration Act (Cap 143A, 2002 Rev Ed) are directly comparable to those governing the present appeal. [19] Dallah Real Estate and Tourism Holding Co v Ministry of Religious Affairs of the Government of Pakistan [2011] 1 AC 763 at 828, §77. [20] Judgment §96, citing First Laser Ltd v Fujian Enterprises (Holdings) Co Ltd (2012) 15 HKCFAR 569, at paragraphs 43 to 49 per Lord Collins of Mapesbury NPJ; and The Sennar (No 2) [1985] 1 WLR 490, at 493F‑494A per Lord Diplock and at 499A‑C per Lord Brandon. [21] Cheung and Barma JJA, HCMP 835/2014 (25 June 2014), at §13. [22] Judgment §73(5), citing Hebei Import & Export Corp v Polytek Engineering Co Ltd (1999) 2 HKCFAR 111 at 136D-E. [23] Judgment §73(3). [24] Judgment §73(4), citing Société Nationale D’Operations Petrolieres de la Côte d’Ivoire Holding v Keen Lloyd Resources Ltd [2004] 3 HKC 452, at §14 per Burrell J. [25] (1999) 2 HKCFAR 111 at 138. [26] [1995] 2 HKLR 215 at 217. [27] China International Economic and Trade Arbitration Commission. [28] [1995] 2 HKLR 215 at 225. [29] Ibid at 226. [30] (1999) 2 HKCFAR 111 at 137, citing China Nanhai Oil Joint Service Corp Shenzhen Branch v Gee Tai Holdings Co Ltd [1995] 2 HKLR 215. [31] Judgment §91. [32] Judgment §§82-86. [33] Judgment §67 and §89. [34] Judgment §92. [35] Judgment §131. [36] Set out in Section B of this judgment. [37] [2012] 1 HKLRD 1063. [38] [2013] 1 Lloyd’s Rep 86. [39] Judgment §131. [40] Judgment §129(1). [41] Judgment §129(2). [42] Judgment §129. [43] Judgment §130. [44] Kwan JA and Lok J, CACV 272/2015 (5 December 2016), Kwan JA writing for the Court. [45] Court of Appeal §§44 and 47, citing Gao Haiyan v Keeneye Holdings [2012] 1 HKLRD 627 at §§65 to 69 and Minmetals Germany GmbH v Ferco Steel Ltd [1999] CLC 647 at 661. [46] Judgment §91. [47] SCA Judgment §§100-123, 125-132. [48] SCA Judgment §§205-219, 222-224. [49] Court of Appeal §47. [50] Court of Appeal §59. [51] [2011] 1 AC 763 at §89. [52] Reflected in our section 44(2). [53] Ibid at 836, §102. [54] Ibid at 843, §126. [55] Ibid at 844, §127. [56] [2013] 1 Lloyd’s Rep 86. [57] Ribeiro, Tang and Fok PJJ, FAMV 20/2017 (18 August 2017). [58] Yuen JA and Kwan J, CACV 103/2005, at §15 (23 October 2006). [59] Appearing for the appellant with Mr Mark Strachan SC and Mr Jeffrey Chau. [60] Appearing for the respondents with Mr Bernard Man SC and Mr Justin Ho. [61] [1997] HKLRD 965 at 973. [62] [1993] 1 WLR 256 at 263-264. [63] [2012] 1 HKLRD 1063. [64] Ibid at §11. [65] [2013] 1 Lloyd’s Rep 86. [66] Judgment §§127-128 and Court of Appeal §76. [67] At §27. [68] At §28. [69] At §29. [70] At §30. [71] At §31. [72] Set out in Section E.2 of this judgment. [73] Judgment §129. [74] Terna Bahrain at §33, an approach reiterated by Popplewell J in Michael Jonathan Christopher Oldham v QBE Insurance (Europe) Ltd [2017] EWHC 3045 (Comm). [75] Court of Appeal §81. [76] [1998] 2 Lloyd’s Rep 337. [77] [2014] 1 WLR 3926. [78] (2011) 14 HKCFAR 935 at §34. [79] Ibid at §80. [80] Judgment §129. [81] [1993] 2 HKLR 39 at 48-49. [82] [2011] 1 AC 763 at 837, §103. [83] (1999) 2 HKCFAR 111 at 136. [84] Respondents’ written submissions §§76-77. [85] Court of Appeal §86. [86] Judgment §130. 
Chief Justice Ma : 1. I agree with the judgment of Lord Millett NPJ. Mr Justice Ribeiro PJ : 2. I agree with the judgment of Lord Millett NPJ. Mr Justice Tang PJ : 3. I am in full and respectful agreement with the judgment of Lord Millett NPJ which I have read in draft. There is nothing I can usefully add. Mr Justice Litton NPJ : 4. I agree with the judgment of Lord Millett NPJ. Lord Millett NPJ : 5. The question for decision in this appeal is whether the introduction of new accounting standards in Hong Kong in 1998 had the effect of making unrealised increases in the value of the Respondent’s trading stock held at the end of its accounting period chargeable for the first time as taxable profits. 6. The appeal is brought by the Commissioner of Inland Revenue (“the Commissioner”) from a judgment dated 19 June 2012 of the Court of Appeal (Cheung, Hartmann and Fok JJA) dismissing its appeal from a judgment of To J dated 28 June 2011. By his judgment To J had held that the introduction of the new accounting standards did not have the effect for which the Commissioner contended and allowed the Respondent’s appeal from assessments to profits tax in each of the three years from 2003/04 to 2005/06. The facts 7. The facts which are agreed are set out at length in the judgments below and it is not necessary to repeat them. They may be shortly stated as follows. 8. The Respondent is a private company incorporated in Hong Kong. Its principal business consists of trading in marketable securities quoted in Hong Kong. Prior to the introduction of new accounting standards for 1999 and subsequent years, its trading stock like that of other traders and in accordance with the conventional practice was shown in its financial statements at the lower of cost and net realisable value. This had the effect that unrealised increases in the value of its trading stock during the accounting period (in the Respondent’s case marketable securities held for sale) were not reflected in its profit and loss accounts or tax computations. Following the introduction of new accounting standards in 1998, however, the Respondent duly recorded in its profit and loss accounts not only profits and losses which it had realised by the sale or disposal of trading stock during the accounting period but also changes in the value of unrealised trading stock held at the end of the period. 9. There is no dispute that the Respondent’s financial statements for the relevant accounting periods were prepared in accordance with the prevailing albeit new accounting practice in Hong Kong. The Respondent accepts (and its auditors reported) that its financial statements were prepared in accordance with accounting principles generally accepted in Hong Kong and showed a true and fair view of its affairs and of its profits and losses for the relevant accounting periods. But it contends that its profit and loss accounts need to be adjusted for tax purposes by excluding unrealised profits from its tax computations since they are not assessable to profits tax. Accordingly in computing its assessable profits and allowable losses for tax purposes for each of the years from 2003/2004 to 2005/2006, the Respondent excluded unrealised profits (ie increases in value of its unrealised trading stock during the accounting period) but continued to claim to deduct unrealised losses which it described in its profit and loss accounts as provision for the diminution in value of listed investments held at the end of the accounting period. 10. The Commissioner considered that the unrealised gains and losses arising from revaluing the trading stock held at the end of the year should be included in the profits tax assessment for the year of assessment in which the unrealised gains were credited and unrealised losses were debited in the Respondent’s financial statements. Accordingly, he assessed the Respondent to profits tax on the basis of the realised and unrealised losses for the years of assessment 1999/2000 to 2002/03 and profits for the years of assessment 2003/04 to 2005/06. The difference between the amount of the profits tax assessed on this basis over the period and that calculated by the Respondent is of the order of $250 million. 11. The new accounting standards in accordance with which the Respondent’s financial statements were prepared are applicable only to persons who carry on the business of trading in marketable securities, whether listed or not, and while mandatory in other cases are optional in the case of small and medium sized businesses. As the Court of Appeal observed, the surprising effect of the Commissioner’s contentions is that, without any statutory support in the taxing statute, taxpayers who carry on the business of trading in securities are taxable on their unrealised profits while those who carry on other businesses are not; and unlike larger businesses carrying on the same trade small and medium sized businesses may choose whether or not to be taxed on their unrealised profits. In my judgment, as will shortly appear, this is not merely surprising; it is also contrary to the express charging provisions for profits tax in Hong Kong. The statutory provisions 12. Section 14(1) of the Inland Revenue Ordinance, Cap 112 (“the Ordinance”), imposes a charge to profits tax “… for each year of assessment … on every person carrying on a trade, profession or business in Hong Kong in respect of his assessable profits arising in or derived from Hong Kong for that year from such trade, profession or business … as ascertained in accordance with this Part.” (emphasis added) Section 2 defines “assessable profits” to mean: “… the profits in respect of which a person is chargeable to tax for the basis period for any year of assessment, calculated in accordance with the provisions of Part 4” Section 18B(1) provides that, for the relevant years of assessment “the assessable profits for any year of assessment … from any trade, profession or business carried on in Hong Kong shall be computed on the full amount of the profits therefrom arising in or derived from Hong Kong during the year of assessment.” (my emphasis) 13. The question for decision, therefore, is whether for the purpose of profits tax unrealised increases in the value of trading stock held at the end of the accounting period as a result of the revaluation should be included in the computation of “the full amount of the profits … arising in or derived from Hong Kong during the year[s] of assessment”. As the Courts below observed, this raises a matter of statutory construction[1], not accounting practice. The question is one of law: what does the statute mean by the words “the full amount of the profits therefrom during the year of assessment”? Whatever these words mean, the fact that they apply to “every person” means that in the absence of some statutory provision to the contrary they mean the same for every taxpayer to which the Ordinance applies whatever the nature or size of his business. Moreover, the word “therefrom” (meaning from any trade, profession or business) suggests that the profit must derive from some trade, professional or business activity and not merely be the result of a revaluation of assets held for the purpose of the trade, profession or business. The Commissioner’s case 14. At the heart of the Commissioner’s case lay three propositions. First, the word “profits” is not defined in the Ordinance, and in the natural and ordinary meaning of the word unrealised profits are nonetheless profits. Secondly, the amount of the profits during the year of assessment is primarily a question of fact. And thirdly, the amount of any profits or losses during the year of assessment must be ascertained by reference to ordinary principles of commercial accounting unless these are contrary to an express statutory provision in the Ordinance. These principles are not static but so long as they remain current and generally accepted they provide the surest guide to the question that the legislation requires to be answered[2]. Profits 15. While it is true, as the Commissioner submitted, that the amount of any profits is a question of fact, what constitutes “profits” within the meaning of the Ordinance and whether any disputed amount represents an assessable profit are questions of law. 16. The word “profits” is an ordinary English word and as such is capable of a broad variety of meanings. In these circumstances its meaning in any particular case depends on the context in which it is used. It has been consistently held in a series of cases dealing with the prohibition against the payment of dividends by companies except out of profits that the concept of profits in the context of company law is sufficiently broad to embrace unrealised profits[3]. The question in the present case is raised in a very different context; whether for the purpose of profits tax the word “profits” in s 14(1) of the Ordinance includes unrealised profits. 17. In seeking to persuade us to answer this question in his favour the Commissioner has relied heavily on the judgment of Fletcher Moulton LJ in Re Spanish Prospecting Co Ltd[4] where he said[5]: “The word ‘profits’ has in my opinion a well-defined legal meaning, and this meaning coincides with the fundamental conception of profits in general parlance …. ‘Profits’ implies a comparison between the state of a business at two specific dates usually separated by an interval of a year. The fundamental meaning is the amount of gain made by the business during the year. This can only be ascertained by a comparison of the assets of the business at the two dates. For practical purposes these assets in calculating profits must be valued …. Even if the assets were identical at the two periods it would by no means follow that there had been neither gain nor loss, because the market value – the value in exchange – of these assets might have altered greatly in the meanwhile.”(emphasis added) 18. In my judgment the Commissioner’s reliance on that case is misplaced, since the context in which the word “profits” was used was completely different. What was in issue was the meaning of the word “profits” in a contract of employment where the employee’s salary was payable only out of the company’s profits. The decision has rightly been described as “altogether independent of any taxing statute”[6]. In The Naval Colliery Co Ltd v CIR[7] Lord Warrington of Clyffe distinguished it as “not in point” because “The learned Lord Justice was dealing not with a profit and loss account for the purpose of Income Tax but with a balance sheet intended to show the actual financial condition of a business at the end of a business year.” It has been repeatedly recognised in many different jurisdictions that when considering the meaning of the word “profits” in the Spanish Prospecting case Fletcher Moulton LJ was not dealing with its meaning in the context of taxation[8]; and that in that context the word has always been given a more restricted meaning. 19. In Read v The Commonwealth[9] the High Court of Australia was concerned with the meaning of the word “profits” in the context of social security payments. In the majority judgment Mason CJ, Deane and Gaudron JJ said[10] “In our opinion a mere increase in value of an asset does not amount to a capital profit. A profit connotes an actual gain and not a mere potential to achieve a gain. Until a gain is realised it is not ‘earned, derived or received’. A capital gain is realised when an item of capital which has increased in value is ventured, either in whole or in part, in a transaction which returns that increase in value.” The reference to “derived” is not without interest, since under s 14(1) of the Ordinance profits are assessable to profits tax only if they are “arising in or derived from Hong Kong”. While that particular requirement is specific to Hong Kong, the passage may be of wider application, since for more than a century the ordinary principles of commercial accounting have required the adoption of the accruals method of computing profits under which profits from the sale of trading stock are recognised not when the sale proceeds are received but when they are “earned” by the sale of stock. 20. The correctness of that decision in relation to Australian social security payments may be doubted, but not its correctness in relation to capital gains tax. As the High Court of Australia observed in FCT v Sun Alliance Investments[11] the case proceeded on the apparently erroneous assumption that the notion of capital profits in the social security legislation equated with the notion of capital gains for tax purposes. In the latter case, in a judgment delivered by the full court[12], the High Court of Australia rejected Fletcher Moulton LJ’s definition of the word “profits” as of universal application, even as a matter of ordinary usage. It said[13] “… the notion that a profit may be revealed or disclosed by a revaluation even where the composition of the assets held by a business does not change appears at odds with the focus, naturally attendant upon discussions of the ‘ordinary usage’ concept of income, on receipts coming into a taxpayer’s hands.” The cardinal principles 21. There are two cardinal principles of tax law: (i) the word “profits” connotes actual or realised and not potential or anticipated profits; and (ii) neither profits nor losses may be anticipated. The two principles overlap and are often interchangeable, for they both involve questions of timing; but they are not identical. The first is concerned with the subject-matter of the tax, uses the word “anticipated” in its secondary meaning of “expected” or “hoped for”, and excludes profits which have not been and may never be realised. The second is concerned with the allocation of profits to the correct accounting period, uses the word “anticipated” in its primary meaning of “brought forward”, and prevents profits being taxed prematurely. 22. It would be wearisome to cite all the numerous authorities in which these principles have been stated, but it is right to cite some of the more important if only in deference to the industry of counsel[14]. In Duple Motor Bodies Ltd v Inland Revenue Commissioners[15] Viscount Simonds said “… the Crown is not entitled to anticipate a profit which may or may not be made, as it might do if too high a value were put on stock-in-trade. …” In the same case Lord Reid said “… it is a cardinal principle that profit shall not be taxed until realised. If the market value fell before the article was sold the profit might never be realised.” Lord Reid, however, placed a gloss on the principle in relation to losses, for in the passage immediately following he said “But an exception seems to have been recognised for a very long time: if market value has already fallen before the date of valuation so that at that date the market value of the article is less than it cost the taxpayer, then the taxpayer can bring the article in at market value, and in this way anticipate the loss which he will probably incur when he comes to sell it. That is no doubt good conservative accountancy but it is quite illogical.” (emphasis added) 23. Lord Reid’s statement of the cardinal principle has been frequently cited with approval; see B S C Footwear Ltd v Ridgway per Lord Morris of Borth-y-Gest[16] and Willingale v International Commercial Bank per Lord Fraser[17], where Lord Keith[18] simply referred to “the rule that a profit may not be taxed until it is realised.” 24. In Willingale in the Court of Appeal [19] Stamp LJ had referred to “the income tax rule that you may not be taxed on an anticipated profit.” When that case reached the House of Lords[20] Lord Salmon explained the meaning of “realised”[21] “It is well settled by the authorities cited by my noble and learned friends that a profit may not be taxed until it is realised. This does not mean until it has been received in cash but it does mean until it has been ascertained and earned.” Under the prevailing accruals system of accounting, a profit is realised by sale of trading stock, not when the sale price is received, but when the amount of the sale price is credited to cash at bank or debtors. 25. In B S C Footwear Ltd v Ridgway[22] Lord Reid said[23]: “The application of the principles of commercial accounting is, however, subject to one well-established though non-statutory principle. Neither profit nor loss may be anticipated. A trader may have made such a good contract in year one that it is virtually certain to produce a large profit in year two. But he cannot be required to pay tax on that profit until it actually accrues. And conversely he may have made such an improvident contract in year one that he will certainly incur a loss in year two but he cannot use that loss to diminish his liability for tax in year one.” 26. In relation to profits the only doubt is whether the principle is truly non-statutory. In Gallagher v Jones[24] Sir Thomas Bingham MR said that the “general principle of income tax law that neither profit nor loss should be anticipated” which was not in doubt was “squarely based on the statutory provisions”[25]; while Nolan LJ suggested that it might equally be described as a restatement in a particular context of the statutory rule that tax should be charged on “full amount of the profits or gains of the year”[26]. But it does not matter, for whatever the source of the rule it is far too well established as a principle of law to be overturned except by clear and express statutory provision. It certainly cannot be overturned by the adoption of new standards of commercial accounting. But while it is true to say that neither profits nor losses may be anticipated, it is not true to say that the taxpayer may not use an unrealised loss to diminish his liability for tax. Losses 27. In Duple Motor Bodies Ltd v Inland Revenue Commissioners[27] Lord Reid had suggested that in relation to losses there was a long established though illogical exception to the cardinal principle that neither profits nor losses may be anticipated[28], in that the taxpayer could bring into his accounts at market value an article of trading stock which had fallen in value below cost and “in this way anticipate a future loss”. The words “in this way” show that the process which Lord Reid was describing is not strictly an exception to the principle that neither profits nor losses[29] may be anticipated; and for my part I do not consider that it is “quite illogical”. Strictly speaking there is no exception to the rule that losses may not be anticipated. If at the end of an accounting period the value of an item of trading stock is the same as or greater than cost but it is sold in the following accounting period for less than cost, the loss is realised in the later period and cannot be brought forward to the earlier. This is the case even if the loss is realised before the accounts are signed off, for post-balance sheet events are relevant and can be taken into account only if they affect the position as at the balance sheet date. 28. If the market value of an item of trading stock which cost $100 is $120 at the end of year one and the item is sold in year two for $80, the application of ordinary principles of taxation means that for tax purposes in year one there is neither profit (because the profit has not been realised) nor loss (because the loss may not be anticipated); but there will be a loss of $20 in year two. Under the new accounting standards, however, the taxpayer’s financial statements will show a profit of $20 in year one and a loss of $40 in year two. 29. But it does not follow that an unrealised loss cannot be used to reduce liability for profits tax. In a proper case this can be achieved by making provision in the profit and loss account for the diminution in the value of trading stock during the accounting period. At first sight this seems to be merely another way of anticipating unrealised losses; but it is not. The auditors will not normally allow such a provision to be made unless they are satisfied that the diminution in value is material and likely to be permanent. Moreover, if such a provision is made it can be challenged by the Commissioner. The need for such a rule can be seen by considering the case where the trading stock includes shares in a company has become insolvent and the shares worthless. The taxpayer may properly write off the value of the shares by making an appropriate provision when the company is put into liquidation without waiting for the company to be dissolved. 30. The difference between making a provision for a diminution in value, which has always been permitted in a proper case, and substituting market value for cost in accordance with the new accounting standards, which is obligatory in every case, can be seen by considering a very simple example. Suppose the value of an item of trading stock which cost $100 fluctuates between $95 and $105 during the accounting period and is worth (i) $102 or (ii) $98 at the end of the period. The application of the principles of taxation results in neither taxable profit nor allowable loss in either case (because the profit in (i) is unrealised and the loss in (ii) does not justify a provision). Under the new accounting standards, however, the financial statements will be required to show a profit of $2 in (i) and a loss of $2 in (ii). 31. If contrary to expectation a provision later proves to be unjustified it should be wholly or partially written back, thereby producing a so-called “profit” of the amount written back. Thus if the value of an item of trading stock which cost $100 falls to $50 in year one and the reduction in value seems to be permanent, the auditors may agree that a provision of $50 should be made against the value of the stock, producing a loss of $50 in that year. If the item is sold in year two for $80 the provision should be written down by $30, thereby increasing the profits by $30 in year two. If it is sold for $120 the provision should be wholly written off, increasing the profits in year two by $70. 32. It may become clear even without a sale that the market has risen sufficiently to show that the provision was unjustified, in which case the amount of the over-provision should be written off, thereby increasing the profits or reducing the losses appearing in the profit and loss account. This appears to be inconsistent with the principle that profits must be “earned” or derived from a transaction by which they are realised; but the writing back of an unjustified provision is not a “profit” in the ordinary meaning of the term but is merely the reversal of a previous understatement of profits. If the taxpayer fails to write back a provision which later proves to have been unjustified, the Commissioner can challenge the accounts. The role of the principles of commercial accounting 33. The Commissioner submitted that the amount of any profits or losses during the year of assessment must be ascertained by reference to the ordinary principles of commercial accounting unless these are contrary to an express statutory provision in the Ordinance, and relied on the decision of this Court in Commissioner of Inland Revenue v Secan Ltd[30] for this purpose. That is a misreading of my judgment in that case. After citing the celebrated passage in the judgment of Sir John Pennycuick VC in Odeon Associated Theatres Limited v Jones[31], in which he explained the relationship between accountancy evidence and the ascertainment of the taxpayer’s assessable profits, I said[32]: “Both profits and losses therefore must be ascertained in accordance with the ordinary principles of commercial accounting as modified to conform with the Ordinance. Where the taxpayer’s financial statements are correctly drawn in accordance with the ordinary principles of commercial accounting and in conformity with the Ordinance, no further modifications are required or permitted.” It should be noted that I said “in conformity with the Ordinance”, not “in conformity with an express provision of the Ordinance”. 34. It is a fundamental principle of the constitution of Hong Kong, as of England, Australia, the United States and other democratic societies, that the subject is to be taxed by the legislature and not by the courts, and that it is the responsibility of the courts to determine the meaning of legislation. This is not a responsibility which can be delegated to accountants, however eminent. This does not mean that the generally accepted principles of commercial accounting are irrelevant, but their assistance is limited. 35. In the present case the subject matter of the tax is “profit”, and the question what constitutes a taxable profit is a question of law. While the amount of that profit must be computed and ascertained in accordance with the ordinary principles of commercial accounting, these are always subject to the overriding requirement of conformity, not merely with the express words of the statute, but with the way in which they have been judicially interpreted. Even where the question is a question of computation, the court must “always have the last word”[33]. 36. In some of the earlier cases on which the Commissioner relied the courts have used language which might be taken to suggest that the ordinary principles of commercial accounting must prevail unless inconsistent with an express statutory provision in the taxing statute. But the passages in question need to be read in context. In Whimster v The Commissioners of Inland Revenue[34], for example, Lord Clyde said that the taxpayer’s profits must be ascertained in accordance with the ordinary principles of commercial accountancy “and in conformity with the rules of the Income Tax Act”. He observed that the rule that stock should be brought into the account at the lower of cost and net realisable value was derived from the ordinary principles of commercial accounting “although there is nothing about this in the taxing statutes”. In that case, however, the Court of Session rejected an attempt by the taxpayer to deduct future anticipated losses in ascertaining its profits for the accounting period because the taxpayer had failed to show that its accounts were drawn in accordance with ordinary commercial practice. The case, therefore, has no relevance when considering the circumstances which require accounts drawn in accordance with ordinary principles of commercial accounting to be modified for tax purposes. 37. In Sun Insurance Office v Clark[35] Viscount Haldane said that “Questions of law can only arise when (as was not the case here) some express statutory direction applies and excludes ordinary commercial practice, or where, by reason of its being impracticable to ascertain the facts sufficiently, some presumption has to be invoked to fill the gap.” But that was a case in which the amount of the profits had to be estimated and where Lord Loreburn LC observed[36] “There is no rule of law as to the proper way of making an estimate …. A rule of thumb may be very desirable, but cannot be substituted for the only rule of law that I know of, namely, that the true gains are to be ascertained as nearly as it can be done.” 38. In Southern Railway of Peru Ltd v Owen[37] Lord Radcliffe said that he “should view with dismay the assertion of legal theories as to the ascertainment of true annual profits which were in conflict with current accountancy practice and were not required by some special statutory provision of the Income Tax Acts.” In rejecting the Crown’s argument that a provision for contingent future liabilities was contrary to a supposed rule of law, however, Lord Radcliffe did not confine his remarks to the statute. He said “In my opinion, there is no such rule of law governing the ascertainment of annual profits. Where does it come from? Not from anything to be found in the Income Tax Act. … Not from any decided authority which is binding on your Lordships ….” In that case, moreover, the House of Lords overturned the finding of the Special Commissioners in favour of the taxpayer after hearing the evidence of its auditor and another independent accountant of distinction, and despite the auditor having testified that he would not have signed the balance sheet without qualification unless the disputed provision for contingent future liabilities had been made. The provision was disallowed not because it was inconsistent with an express statutory provision but because, in the words of Lord Radcliffe, the provision was “well on the wrong side of what was permissible”. 39. It is clear beyond argument that accounts drawn up in accordance with the ordinary principles of commercial accounting must nevertheless be adjusted for tax purposes if they do not conform to the underlying principles of taxation enunciated by the courts even if these are not expressly stated in the statute[38]. InWillingale v International Commercial Bank[39] Lord Fraser said that “… where ordinary commercial principles run counter to the principles of income tax they must yield to the latter when computing profits or gains for tax purposes.” (my emphasis) There are many other statements in the authorities to the same effect. 40. In particular, the principles of commercial accounting must give way to the core principles that profits are not taxable until they are realised and that profits must not be anticipated. In the passage cited above from his speech in B S C Footwear Ltd v Ridgway[40] Lord Reid stated in terms that “The application of the principles of commercial accountancy is, however, subject to one well-established though non-statutory principle. Neither profit nor loss may be anticipated.” In Willingale v International Commercial Bank[41]it was common ground that the taxpayer’s accounts were drawn up in accordance with the principles of commercial accountancy[42]. Yet they were held to be “not a proper basis for assessing [its] liability for corporation tax” because they contravened the rule that profits may not be anticipated (per Lord Fraser[43]) or that a profit may not be taxed until realised (per Lord Keith[44]). 41. Most of the cases in which the courts have considered the circumstances in which recourse may be had to the evidence of accountants were concerned with the allocation of profits, or more usually expenditure, to the correct period of account. In B S C Footwear Ltd v Ridgway[45] Lord Reid said[46] “In my view, there is a difference between a question whether a sum is or is not taxable at all, and a question as to the proper year to which an admitted profit should be allocated. To the former there can only be one answer. The latter may not be capable of a definite answer: one may say that one answer is preferable but that another is possible. Much will depend on proper accounting practice and that may alter in the course of time.” 42. There is also a difference between determining the amount of any profits, which is a matter of computation, and determining whether a given sum is or is not a taxable profit for the purpose of the Ordinance, which is a question of law. Most of the statements that primacy should be accorded to the ordinary principle of commercial accounting have been made with reference to the first question and not the second. Thus in Revenue and Customs Commissioners v William Grant & Sons Distillers Ltd[47] Lord Hope of Craighead said that “The golden rule is that the profits of a trading company must be computed in accordance with currently accepted accounting principles.” (my emphasis) 43. The question in that case was whether depreciation of unsold stock (which is a cost) could be carried forward and treated as part of the cost of future sales. In his speech Lord Hoffmann indicated that in Secan[48] I gave two contradictory answers to the question, for my statement that the Ordinance did not prohibit the capitalisation of interest suggested one answer and my analysis of the accounting treatment another. With respect to Lord Hoffmann I plainly used the expression “capitalisation of interest” in its normal meaning to describe the accounting treatment which I later explained in detail. As I explained, normally interest is an overhead like rent and deducted as an expense of the business as a whole in the year in which it is paid. When it is capitalised, however, it is treated as a cost of acquisition of an asset. This is similarly deducted in the year in which it is paid but is matched by a corresponding increase in the value of the asset, producing neither profit nor loss in that year but reducing the profit in the year in which the asset is sold by the amount of the increase in its cost. There are not two different systems of computation but a single system which has had the misfortune to be explained in different ways. Financial Statements 44. It must be borne in mind that the new accountancy standards are directed to the preparation of financial statements and not tax computations, and that the two serve different purposes. Financial statements are prepared in order to give investors, potential investors, financial advisers, and the financial markets generally a true and fair view of the state of affairs of the company and in particular its financial position and profitability. Those who read them are concerned not with the past but with the future, and in particular the future profitability of the company. The Ordinance, however, is directed to the past. The Commissioner is not concerned with the likelihood that the taxpayer will make profits in future but whether it made them in the past. 45. The courts have had frequent occasion to comment that while a taxpayer’s financial accounts, drawn in accordance with ordinary principles of commercial accountancy, may be appropriate for the purpose of showing its financial position they may not be appropriate for the assessment of tax[49]. Where they are not appropriate for this purpose, the taxpayer is entitled or may be required to adjust them for tax purposes: the cases show both situations. In Minister of National Revenue v Anaconda American Brass Ltd[50] Viscount Simonds, giving the opinion of the Privy Council, said “Their Lordships do not question that the Lifo method or some variant of it may be appropriate for the corporate purpose of a trading company. Business men and their accountant advisers must have in mind not only the fiscal year with which alone the Minister is concerned. It may well be prudent for them to carry in their books stock valued at a figure which represents neither market value nor its actual cost but the lower cost at which similar stock was bought long ago. A hidden reserve is thus created which may be of use in future years. But the Income Tax Act is not in the year 1947 concerned with the years 1948 or 1949: by that time the company may have gone out of existence and its assets been distributed … the evidence of expert witnesses, that the Lifo method is a generally acceptable, and in this case the most appropriate method of accountancy, is not conclusive of the question that the court has to decide. That may be found as a fact by the Exchequer Court and affirmed by the Supreme Court. The question remains whether it conforms to the prescription of the Income Tax Act. As already indicated, in their Lordships’ opinion it does not.” 46. In Willingale v International Commercial Bank, where the taxpayer’s financial statements were found to be drawn up in a way which anticipated future profits, Lord Fraser said that there were no doubt excellent commercial reasons for preparing the accounts in that way and borrowed the words of Walton J[51] that they “are much better economic indicators than corporation tax accounts would be as to whether a bank is or is not doing what it ought to be doing, that is to say, steadily making an economic profit for its shareholders.” Despite this he held that they were not a proper basis for assessing the bank’s liability to tax. 47. In the present case the taxpayer’s financial statements, prepared in accordance with the new accounting standards, showed a single figure for the profits whether realised or unrealised during the accounting period, followed by a breakdown between realised and unrealised profits. Before 1999 the profit and loss account would not have shown unrealised profits and the balance sheet would have shown stock in trade at the balance sheet date at cost, excluding any increase in its value during the preceding accounting period. But it is difficult to believe that the taxpayer would not have wanted in its own interest, or possibly have been obliged by its auditors, to advertise its success by providing in a note to the accounts the value of the stock still held at the end of the year. If so, then the change is merely a matter of presentation in which the contents of a note to the accounts have been elevated to the balance sheet. The new accounting standards 48. The new accounting standards have been adopted internationally by many different countries including the United Kingdom. Their purpose, eminently laudable, is to harmonise so far as possible the preparation of financial statements so that they may be understood by those who read them and who live in a global world. They are not intended, and cannot sensibly be thought to have been intended, to harmonise the tax liabilities of taxpayers carrying on businesses in countries with greatly different tax regimes. The international nature of the new accounting standards militates against their use for tax purposes. 49. As the Respondent observed in its printed case, the existence of mandatory international accounting standards for the preparation of financial statements provides surer evidence than was available in the past of the ordinary principles of commercial accounting, but they cannot take the place of the Ordinance as interpreted by the courts. 50. It is of some interest to note that the original view of the Inland Revenue in the United Kingdom was that financial statements prepared in accordance with the new accounting standards were not appropriate for the assessment of tax. It is understood that the Inland Revenue have changed their mind, possibly influenced by a government desperate to raise as much revenue as possible to fill a hole in the public finances. It remains to be seen whether their changed approach will be challenged. If it is, then these proceedings will no doubt be considered a dry run. Conclusion 51. In my judgment the taxpayer’s financial statements, prepared in accordance with mandatory international accounting standards, record both profits which the taxpayer has realised during the accounting period and which are assessable to tax and increases in the value of its trading stock during the period which represent unrealised profits and are not assessable to tax. In preparing its tax computations the taxpayer was entitled to remove the amounts of its unrealised profits as not chargeable to tax. 52. I would dismiss the appeal. Chief Justice Ma : 53. For the above reasons, the appeal is unanimously dismissed. As to costs, we made an order nisi that the respondent should have the costs of the appeal, such costs to be taxed if not agreed. If any party wishes to have a different order for costs, written submissions should be served on the other parties and lodged with the Court within fourteen days of the handing down of this judgment, with liberty on the other parties to lodge written submissions within fourteen days thereafter. In the absence of such written submissions, the order nisi will stand absolute at the expiry of the time limited for these submissions. Mr Michael Furness QC, Mr Eugene Fung SC and Mr Julian Lam, instructed by the Department of Justice, for the Appellant Mr David Goldberg QC and Mr Stewart Wong SC, instructed by Woo, Kwan, Lee & Lo, for the Respondent [1] All questions whether tax is payable or not are ultimately questions of statutory construction: see Collector of Stamp Revenue v Arrowtown Assets Ltd (2003) 6 HKCFAR 517 at §§35-36 per Ribeiro PJ and at §105 per Lord Millett NPJ. [2] See Revenue and Customs Commissioners v William Grant & Sons Distillers Ltd[2007] 1 WLR 1448 at p 1458 per Lord Hope of Craighead adopting the observation of Sir Thomas Bingham MR in Gallagher v Jones[1994] Ch 107. [3] See Commissioner of Taxation of the Commonwealth of Australia v Sun Alliance Investments Pty Ltd(In Liquidation) (2005) 225 CLR 448 at p 506 and the cases there cited. [4] [1911] 1 Ch 92 CA. [5] At pp 98-99. [6] See Dalgety v Commissioner of Taxes (1912) 31 NZLR 260 at p 262 per Williams J. [7] (1928) 12 TC 1017 at p 1052 per Lord Warrington of Clyffe. [8] In Re Income Tax Acts (No 2) [1930] VLR 233 at 238 per Irvine CJ; at 245 per Macfarlan J; and at 250 per Lowe J; Jebsen & Co v CIR(1949) HKLR 312 per Gibson CJ at 315-316; Quemont Mining Corp v Canada [1967] 2 Ex CR 169 at §118 per Cattanach J; Marra Developments Ltd v BW Rofe Pty Ltd [1977] 2 NSWLR 616 at 628-629 per Mahoney JA; FCT v Slater Holdings Ltd (1984) 156 CLR 447 at 460 per Gibbs CJ. [9] 1988 167 CLR 57. [10] At p 67. [11] [2005] CLR 288. [12] Gleeson CJ, Gummow, Kirby, Callinan and Heydon JJ. [13] At p 504. [14] Because of the difficulty in doing so no attempt has been made in the citation of authority which follows to allocate the dicta to a particular principle. [15] [1961] 1 WLR 739, 748, 751. [16] [1972] AC 544 at p 560. [17] [1978] AC 834 at p 843. [18] At p 852. [19] [1977] Ch 78 at p 85. [20] Supra note 17. [21] At p 841. [22] Supra note 16. [23] At p 552. [24] [1994] Ch 107 CA. [25] At p 135. [26] The corresponding words in the Ordinance are “the full amount of the profits … during the year of assessment”. [27] Supra note 15. [28] Paragraph 22 supra. [29] Paragraph 25 supra. [30] (2000) 3 HKCFAR 411. [31] 48 TC 257 at p 273. [32] At p 419. [33] Duple Motor Bodies Ltd v Inland Revenue Commissioners(note 15 supra) at p 753 per Lord Reid. [34] (1926) SC 20, 25. [35] [1912] AC 443, 455. [36] At p 454 [37] [1957] AC 334, 360. [38] See B S C Footwear Ltd v Ridgway(note 16 supra) at p 562 per Lord Guest. [39] Note 17 supra. [40] Note 23 supra. [41] Supra note 17. [42] See [1977] Ch At pp 97-8 per Sir John Pennycuick. [43] At p 847. [44] At p 852. [45] Supra note 16. [46] At p 555. [47] [2007] 1 WLR 1448, 1458. The decision has been strongly criticized as leading to the conclusion that the cost of unsold stock, unlike the cost of stock which is sold during the year, is not deductible in the year in which it is incurred, and it may be necessary in future for this court to decide whether it should be applied in this jurisdiction. [48] Note 30 supra. [49] See the observations of Lord Warrington of Clyffe in The Naval Colliery case cited in para 14 supra. [50] [1956] AC 85, 102. [51] [1976] 1 WLR 657,663. 
Mr Justice Ribeiro PJ: 1. I agree with the judgment of Mr Justice Stock NPJ. Mr Justice Tang PJ: 2. I agree with the judgment of Mr Justice Stock NPJ. Mr Justice Chan NPJ: 3. I agree with the judgment of Mr Justice Stock NPJ. Mr Justice Stock NPJ: Introduction 4. These are appeals against concurrent findings of fact. They arise from the determination by the District Court[1] of a preliminary issue in an application by the respondent husband for matrimonial ancillary relief. That issue was whether the beneficial interest in 20 million shares in a privately owned company was held by the petitioner wife or, on the other hand, by her father who was joined as intervener. 5. The shares were registered in the wife’s name but it was her contention and that of her father that she holds them on trust for him. The judge rejected that contention and declared that the shares are part of the wife’s assets and therefore available for distribution upon the ancillary relief application. 6. The judge’s decision was upheld by the Court of Appeal[2], whose judgment predated the repeal of section 22(1)(a) of the Court of Final Appeal Ordinance[3] so that, since the matter in dispute involves a claim to shares clearly quantifiable at a value in excess of $1 million, the appeals come to this Court pursuant to that “as of right” provision[4]. 7. The established practice is that this Court will not interfere with concurrent findings of fact other than in exceptional and rare circumstances, namely, where it is demonstrated that there has been a miscarriage of justice or violation of some principle of law or procedure, and: “That miscarriage of justice means such a departure from the rules which permeate all judicial procedure as to make that which happened not in the proper sense of the word judicial procedure at all. That the violation of some principle of law or procedure must be such an erroneous proposition of law that if that proposition be corrected the finding cannot stand: or it may be the neglect of some principle of law or procedure, whose application will have the same effect. The question whether there is evidence on which the courts could arrive at their finding is such a question of law.”[5] 8. The submission advanced by the petitioner (the wife) and her father is that in this case there are exceptional circumstances which warrant a review of the concurrent findings. Those circumstances are said to be inconsistent findings on the central issue; findings against the credibility of witnesses despite a failure to challenge that credibility; fundamental failures of analysis; and a failure by the first instance judge to provide adequate reasons. We are asked to reverse the declaration, failing which to order a retrial. 9. Most of the complaints can be addressed briefly and do not warrant other than summary disposal, for they do not lend themselves to reasonable argument justifying a full review. There was, however, one aspect of the judgment at first instance which, in my opinion, does warrant review; that is whether the judge at first instance explained what weight he gave to certain events said by the wife and her father to be cogent evidence in support their case and why, despite that evidence, he found against them. The evidence 10. The parties were married in 1999. They have one child. They lived together in properties provided rent-free by the company in question. The wife filed a petition for divorce on 11 October 2011 and a decree nisi was granted on 4 October 2012. 11. The wife was aged 46 years at the time of the trial of the preliminary issue. She has three siblings: her elder sister, Leung Wa Yi, who lives in Canada; Leung Cham Wah (Cham), also in his mid-40s; and Leung Kwun Wa (Kwun), the youngest, aged 40 years at the time of trial. 12. The father, aged 79 years at the time of trial, is a highly successful businessman. In 1977 he established Yee Fung Polyfoam Ltd (Yee Fung) and, subsequently, other companies including HK Sports Helmet Manufacturing Limited, leading manufacturers of crash helmets. 13. The wife commenced work at Yee Fung in about 1990 and continued to work there at all material times. Before he left for Canada in 1997, Cham worked there and Kwun still does. 14. The father invested most of the profits of these businesses in real estate and in March 1995 he acquired a shelf company, Nicegood Properties Limited (NPL), to hold the matrimonial homes of his family members and other properties. 15. In April 1995 the authorised share capital of NPL was increased to $66 million divided into 66 million ordinary shares of $1 each. The father became the registered shareholder of 40 million shares and Cham and Kwun were each allotted 13.2 million shares. 16. The father's case was that he told those of his children who held shares in NPL that they were not to deal in any way with them without his prior consent and that should they leave the family business, the shares were to be returned to him. It was well understood that the shares registered in the names of the two sons belonged to the father beneficially. No dividends were paid to them. He alone provided the working capital for NPL and was the sole authorised signatory of all NPL’s bank accounts. 17. In 1997, Cham went to live in Canada and on 15 December 1997 he transferred his 13.2 million shares to the father. The instrument of transfer and the bought and sold notes state that the transfer was made in consideration of the payment to Cham of $13.2 million. Stamp duty of $22,825 was paid for the transaction. However, the oral evidence was that no consideration was in fact paid. 18. On 11 November 2004, Kwun executed a declaration of trust by which he stated that he held the NPL shares as nominee for his father. The evidence of the father and of the wife was that this occurred because of a falling out between the father and Kwun and that the father wanted thereby to show his lack of trust in Kwun. The effect of the father's testimony was that whereas the trust arrangement had hitherto arisen by reason of oral communications, the relationship between himself and Kwun now required that it be spelt out in writing. In respect of this transaction, stamp duty of $19, 880 was paid. 19. Shortly thereafter, in early December 2004, the father asked the petitioner to replace Kwun as the nominee shareholder of those 13.2 million shares and to hold them on trust for him, the father. There followed, dated 8 December 2004, an instrument of transfer by Kwun to the wife of the 13.2 million shares and the consideration was therein stated to be nil with the words “change of nominee” inserted in manuscript. There is an official stamp on the document stating : “No Ad Valorem Duty payable.” 20. On the same day, the wife executed a declaration of trust stating that she held the shares as nominee for her father, the beneficial owner. The document states that no duty was chargeable. 21. About six weeks later, namely, on 18 January 2005, the wife transferred those shares back to her father, again by an instrument of transfer expressed to be for no consideration followed by the words: “ Being shares transfer back from Nominee to Beneficial Owner”. The document states that no ad valorem duty was payable. 22. According to her testimony, she returned the shares because her involvement in the dispute between her father and her brother had made her uncomfortable. The father’s evidence was that the request to execute a declaration of trust was, he thought, taken by her as a slight, an indication that her word that she held them on trust was not good enough. In those circumstances and since, at the time, the wife was the only one of his children upon whom he felt able to rely in respect of business and family matters, he thought it appropriate for her to hold shares for him but to do so, as had her brothers in the past, without requiring her to sign a declaration of trust. He also thought that transferring shares to her name would add to her authority and bargaining power with third parties such as estate agents and bankers. He told her and she agreed that “if she left the family business or upon my request, she had to return all things (including the shares of Nicegood) to me … .”[6] 23. Consequently, by an instrument of transfer dated 31 March 2005, 13.2 million shares were transferred by the father to the wife but, according to that instrument and the relevant bought and sold notes, that transfer was effected in consideration of the sum of $13.2 million; however, the oral testimony by the father, the wife and the father’s accountant was that no sum was in fact paid. Stamp duty of $15,249 was paid. 24. In December 2006, as a result of tax advice, the authorised share capital of NPL was increased by 60 million shares of $1 each, taking the total authorised share capital to $126 million. The funds were provided by the father, though from what source and whether by director’s loan to the company, the evidence did not disclose. His testimony was that since his relationship with Kwun had by then improved, he decided to restore Kwun to his status as a shareholder and allotted to him 20 million shares in NPL and, to achieve parity between Kwun and the wife, allotted 6.8 million shares to her. Still, according to this evidence, the arrangement remained – although nowhere recorded - that all NPL shares in the names of Kwun and the wife were held on trust for the father; neither was asked by him to sign a declaration of trust because a trust was sufficiently understood. 25. Amongst the matters which the father prayed in aid of his contention that he never intended to confer beneficial ownership of the shares upon Kwun and the wife were that he regarded NPL as his personal “piggybank”; that to donate shares to only some of his children would unfairly discriminate between his children, grandchildren and other family members; that he alone controlled NPL, its working capital and bank accounts; that no dividends were paid to Kwun or to the wife; that no consideration passed from Kwun or the wife for the shares[7]; that in 1997 Cham acceded to the father's request to return the shares in Cham’s name, conduct which sat squarely with the trust understanding; that, similarly, in 2004 Kwun acceded to the father's request to execute a declaration of trust and to assign his shares to the wife, a further instance of the understanding in action; that the share certificates held by the wife and Kwun were never signed; and that in the light of the divorce petition presented in October 2011 and the certain prospect of a claim for ancillary relief, he would never have permitted the subsequent purchase by NPL, in February and April 2012, of two properties for a total price of $208.8 million. 26. The wife’s testimony supported that of her father, as did the evidence of Kwun. There was other evidence that no money had changed hands upon any of the transfers of NPL shares between father and his children; and of the father's complete control of the affairs of NPL and the Yee Fung group. 27. The husband's evidence was that no one ever suggested in his presence that the siblings held NPL shares on trust for the father. His assumption was that, absent declarations of trust, the intention in transferring shares to the siblings who worked for the company was as a reward for having done so and an encouragement to continue. He speculated that the declarations of trust signed in late 2004 were triggered by the unusual tension between Kwun and his father and that the step by which Kwun’s shares were transferred to the wife upon her express declaration of trust was intended to send a message to Kwun that if he maintained his attitude of hostility to his father, his position would be supplanted by her. He asserted that the father was a generous man given to making gifts of blue-chip shares to the wife and providing substantial financial support to her and to him. The absence of pleadings 28. The order for joinder of the father as intervener was made[8] pursuant to a consent summons which sought an order that he be so joined “to seek a declaration in respect of shares in [NPL] held in the name of the petitioner in which he claims beneficial ownership”. [9] In accordance with the terms of that consent summons, a direction was given for the petitioner to serve copies of all pleadings and other relevant documents in the suit on the intervening party, which, in context, can only have meant the pleadings - such as the petition and cross petition - and documents hitherto filed in the matrimonial proceedings. 29. However, no direction was sought and none given for the filing of pleadings or for the discovery of documents in support of and directed to the preliminary issue. That omission was unfortunate and was understandably categorised by the Court of Appeal as contributing to confusion as to “what case was advanced by the husband in opposition to the father's claim, how the husband's case was conducted in the course of the trial, and the findings that should be made by the judge on the competing claims or contentions to resolve the dispute.”[10] In this regard, I endorse the practice commended by the Court of Appeal that in the context of matrimonial proceedings for ancillary relief, disputes between a spouse and a third party as to ownership of property should procedurally be approached as in standard claims for ownership of property: in other words, after a direction for trial of the dispute as a preliminary issue and the joinder of the claimant as intervener, by directions for the filing of pleadings on the preliminary issue and for witness statements and discovery relevant to that issue.[11] It does not however lie at the door of any party to the present proceedings now to complain about the procedural lacuna since the flawed directions were sought on the joint application of the parties and ordered by consent without anyone suggesting that appropriate directions be given for the preliminary issue. The first instance judgment 30. As to the law, the judge said: (1) that “[g]iven that the documents hold out the wife to be the legal and beneficial owner of the shares, shares apparently purchased by her for $13.2 million, it is for her and her father to prove on a balance of probabilities that the wife is the father’s trustee and that he remains the beneficial owner of the shares”[12]; and (2) that it was his function to discover what the father's intention was when he transferred the shares to the wife, in which context he cited several passages from Lavelle v Lavelle[13] “which”, he said, “succinctly [address] what is required of me in the circumstances”[14] : “[13] Where one person A transfers the legal title of property that he owns or purchases to another, B, without receipt of any consideration, the effect will depend on its intention. If he intends to transfer the beneficial interest in the property to B, the transaction will take effect as a gift and A will lose all interest in the property. If he intends to retain the beneficial interest for himself, A will take the legal interest but will hold the property in trust for [B] [15]. “[14] Normally there will be evidence of the intention with which the transfer is made. Where there is not, the law applies presumptions. Where there is no close relationship between A and B , there will be a presumption that A does not intend to part with the beneficial interest in the property and B will take the legal title under the resultant trust for A. Where, however, there is a close relationship between A and B, such as father and child, a presumption of advancement will apply. The implication will be that A intended to give the beneficial interest in the property to B and the transaction will take effect accordingly.” “[19] In these cases equity searches for the subjective intention of the transferor. It seems to me that it is not satisfactory to apply rigid rules of law to the evidence that is admissible to rebut the presumption of advancement.”[16] 31. Under the heading “ Conclusions”, the judge said that there was no doubt but that the father had been very generous towards the wife and her husband; that the father was a traditional patriarchal head of the family, who demanded obedience and loyalty, factors to which he, the judge, had paid great attention. He said that he noted, as well and with care, “the pattern of share transfers both on the basis of trusteeship and of apparent transfers of beneficial interests”[17], and had given careful consideration to “the fact that the father made purchases of valuable property after he knew of the pending divorce.”[18] 32. A key question in the case was what point there was, whether from the perspective of the father or from that of the siblings, in the siblings holding the shares as mere nominees. As to the father's explanation in this regard, most particularly in relation to the transfer of shares to the wife in March 2005, the judge found as follows: “I am bound to say that I have great difficulty in accepting the logic of the father's case that the motive in transferring the shares to the wife in this way was to provide her with ‘a psychological lift’ and in giving her more clout when dealing with third parties. As to the former I cannot really discern any ‘lift’ when the recipient of the shares was a mere nominee. As to added status, the fact of the matter is that status, if any, comes from the fact that she is a director but, even more importantly, that she is her father's daughter who overwhelmingly controls the group. No doubt when she speaks, she speaks for him so much so that she would need to consult with him before entering into any important transaction. This really betrays a certain contradiction; more standing with third parties in representing the company and yet, to the knowledge of the third parties called as witnesses, having to defer and consult her father before any final decision is to be taken. I do not believe this proferred motive for transferring the shares as nominee in this way can hold any water at all. I reject it as a reason.”[19] 33. He concluded as follows: “61. In the final analysis I am afraid to say that the truth of this matter is that this kind and generous father decided to reward his beloved daughter and Kwun, with whom he had happily become reconciled, with this very valuable interest in NPL. As to the point that he would not do so because this would favour some relatives over others, is not a matter that I find at all persuasive. The fact is that when the time comes, he still holds great wealth in his own hands, he will be able to do what is right for the remainder of them. The wife in this case is his right hand and he intended, I use the word advisedly, to transfer the shares to her outright, which is what he did. Had it been otherwise he would have created a specific instrument of trust as he had in the past. This must have been, I find this as a fact, a deliberate choice to bring about a transfer of the beneficial interest in the shares. 62. Now of course this has become very inconvenient because the husband is saying, and rightly so on my finding, that the shares, being the wife's must form part of the matrimonial estate which falls to be divided up between him and the wife in such proportion as the court considers correct and just. In order to avoid such a consequence, I regret to say that these usually honest and decent people have decided that they would give false evidence as to the basis of the transfer of these shares.” 34. He accordingly dismissed the father’s application for a declaration that the beneficial ownership of shares in NPL held in the name of the wife rested with him and ordered that the shareholdings stand as part of her beneficially owned assets available for distribution on the hearing of the ancillary relief application. 35. During the course of the hearing on the preliminary issue, the judge raised of his own motion the question whether the instrument of transfer of shares to the wife and the bought and sold notes constituted an estoppel as against the father and the wife. In his judgment, he said that: “Having found that the father's intention was to and did transfer the beneficial interest in the shares to the daughter is sufficient to dispose of the issue … I am therefore not required to decide whether an estoppel is available [to the husband] as an additional ground for succeeding in this matter.”[20] 36. He nonetheless addressed that issue “in deference to counsel’s researches and in case the matter needs to be considered in another court” and, without providing analysis, held that the father and the wife were indeed estopped “from denying the truth of the contents of the instrument of transfer and the bought and sold notes. So this must stand as a further ground for the dismissal of the Intervener’s summons.”[21] The Court of Appeal 37. In the Court of Appeal, it was contended on behalf of the father and the wife that the judgment at first instance was internally inconsistent in that the judge held that there had been a sale for value of the shares to the wife but, inconsistently, that the father had gifted the shares to her; that the documents purportedly recording a sale were erroneously treated by the judge as precluding the existence of a trust; that the judge failed to distinguish between the different circumstances in which the wife acquired the two parcels of shares (13.2 million in March 2005 and 6.8 million in December 2006); that he had ignored the evidence of a consistent pattern of trust arrangements between the father and three of his children, as demonstrated by the return of shares to the father by Cham in 1997 and Kwun in 2004; that the judge failed to appreciate the significance of the absence of signatures on the share certificates; that he failed to recognise the significance of the father’s investment in properties through NPL after he became aware of the institution of divorce proceedings; and that, contrary to the principle in Browne v Dunn[22], it had never been put to the father or the wife, or to other witnesses, in cross-examination of them on the husband’s behalf, that their accounts of an oral trust agreement were untrue. 38. It was conceded by the respondent before the Court of Appeal that the judge had erred in his finding of an estoppel but, in a careful and thorough judgment[23], that Court rejected the argument that that conclusion had infected the judge’s central findings. The Court was satisfied that the judgment evidenced a single uncontradicted finding by the judge that the father had made a gift of the two parcels of shares to the wife; that there were significant indicia which countered the assertion of a trust; and that the failure to put to the witnesses in terms that their evidence was not true was of no consequence since it must have been obvious to the parties and to the witnesses that the husband's case was a challenge to their assertions of a trust arrangement. 39. The Court was not persuaded that the judge had overlooked relevant evidence. The Court pointed to cogent factors against the wife’s case and that of her father: “78. … the transfer to the wife in March 2005 was made without a declaration of trust, and this came not long after declarations of trust were executed by Kwun in November 2004 and by the wife in December 2004. No plausible explanation was given by the wife why she should have been unwilling to be the express nominee of the father when she returned the shares to him in January 2005 but was happy to take on the status in April 2005 without a declaration of trust. When the father was pressed for an explanation why he executed documents stating that consideration was paid and received when the shares were transferred to the wife, when it was his case that no consideration was paid as it was a trust arrangement, he said the documents were just formalities and claimed he did not remember much and did not know the contents. The father admitted at that time, he understood what a declaration of trust was, that it was a simple way of setting out the true beneficial ownership as opposed to registered ownership. 79. As for transferring the shares to the wife as a nominee to provide motivation for the other children to join the family business and work hard, Mr Sussex asked rhetorically why would Cham come running back from Canada only to be made a mere nominee shareholder as he had already been before, on the father's case. That the father had wanted all along his children to carry on the family business was entirely consistent with the gift of the shares to them.” Further, the Court agreed, “there was no credible explanation why the wife’s shareholding was increased to 20 million in 2006 if it was a merely nominal shareholding for spiritual or psychological purposes.” [24] 40. The Court concluded that the trial judge was entitled to disbelieve the father and the wife, whose evidence he had heard and that there was no basis to impugn the finding of a gift.[25] Accordingly, the appeals were dismissed. Concurrent findings 41. These appeals challenge concurrent findings of fact. The established practice, to which I refer at the outset of this judgment[26], is that this Court will not interfere with concurrent findings of fact other than in exceptional and rare circumstances. The exceptional circumstances advanced on behalf of the father and wife are, in their combined effect, a repetition of the grounds of appeal canvassed before the Court of Appeal albeit, in the case of the father, with some difference of emphasis. I first address those which can be disposed of in short order. Inconsistent findings 42. This complaint is not maintained by Mr Burns SC for the father but is maintained by Mr Coleman SC for the wife. Mr Coleman’s suggestion is, first, that resolution of the preliminary issue required the judge to decide not only whether the wife held the beneficial interest in the shares but, if she did, whether it was by reason of gift or of purchase since the route by which that interest was hers would impact upon the degree to which, if at all, the value of those shares would enure to the benefit of the husband in his claim for ancillary relief. The contention is, with respect, not sound, for the preliminary issue defined by the father's summons and ordered to be tried, was whether he held the beneficial ownership of the shares, no more and no less. In the event, the point matters not since the question whether a gift or a purchase, if not a trust, was in reality destined to be addressed – as indeed it was - and, more particularly, it is clear to me, as it was to the Court of Appeal, that the first instance judgment does not yield any inconsistent finding. The judge decided, as is apparent in that part of his judgment entitled “Conclusions”, that in 2005 and in 2006, the father made a gift to the wife of the NPL shares in contention. The Browne v Dunn point[27] 43. It is true that there was an inconsistency in the husband’s case as canvassed in the District Court,although not, as just pointed out, in the judge's findings. The inconsistency was that in his affirmations, his case was that the shares were transferred to the wife by way of gift whereas at trial his case appeared to be that she had purchased them. Yet what was apparent throughout and must have been clear to the witnesses was that their suggestion of a trust was not accepted. The same complaint, when advanced to the trial judge, was dismissed by him with the comment that the entire curial contest had, to everyone's knowledge, concentrated upon the issue whether the father and the wife were telling the truth when they said that it had been agreed between them that she would hold the shares as trustee for the father. “Everybody knows,” he said “that their evidence is under the severest challenge.” A contrary contention is, in my opinion, not tenable. In the circumstances which prevailed at this trial – and such an issue is always case specific - a requirement to put to the witnesses in terms that they were not telling the truth would have been an insistence for form over substance. Judge’s reasoning: the arguments 44. Mr Burns asserts that the judge erred in respect of the burden of proof. The suggestion is that in the matrimonial proceedings it was for the husband to prove that the shares belonged beneficially to the wife whereas the judge held[28] that since the documents held out the wife to be legal and beneficial owner, it was for her and the father to prove that she was a trustee. 45. He further argues that whereas the judge was required to ascertain the father's subjective intention as at the time of transfers of shares to the wife in March 2005 and December 2006, the judge instead adopted an objective test in that he allowed the documents to distract him from evidence of the father's subjective intention. According to this argument, the documents were neutral in their effect since they purported to evidence sales for value whereas it was accepted that no sale had taken place, yet the documents on their face were inconsistent with a gift as well as with the creation of a trust. There was therefore no inference to be drawn from the documents and all that remained was the testimony of the father. It follows, it is said, that the judge’s approach was fundamentally flawed. 46. It is then contended that since it was the father’s case that his evidence of intention was supported by two key undisputed facts - that shares were twice returned to the father upon his request and that expensive properties were injected into NPL after presentation of the petition - it was incumbent upon the judge to explain what impact those facts had upon his reasoning, whereas all he said in his judgment was that he noted them.[29] This is said to demonstrate either a failure by the judge to understand the significance of the key evidence in support of the father’s case or a failure to provide adequate reasons for his decision; in either event, such a significant neglect as to warrant interference by this Court. 47. The argument concludes with the suggestion that if the father’s testimony was correctly rejected, there remained only the presumption of a resulting trust since the transfers of legal title were made without consideration; a presumption which could be rebutted by the presumption of advancement only if the latter presumption were available in the case of transfer of a legal interest by a parent to an adult independent child, a proposition which counsel categorised as debatable.[30] It was incumbent on the husband to rebut the presumption of a resulting trust but the judge erred in not approaching the issue on that basis. 48. Mr Coleman adopted and further developed these submissions. The burden of proof 49. It is unnecessary to decide upon whom the burden of proof lay and whether the judge erred in this regard. That is because the decision against the wife and the father on the preliminary issue did not rest upon a failure by them to discharge a burden of proof. It flowed from a finding of fact by the judge that the father intended by the transfers of shares to the wife to make a gift of them to her.[31] Subjective or objective approach 50. The judge’s task was to ascertain the subjective intention of the father when in March 2005 he transferred the shares to the wife and when in December 2006 he caused shares to be allotted to her; and the case proceeded on the footing that the intention was the same on both occasions. The judge expressly recognised subjective intention as the key question[32] and I am satisfied that that is the test which he in fact applied. I do not accept the argument that he adopted an objective test; neither do I accept the argument that he should have treated the documents as neutral in their evidential effect. 51. It is plain that what the judge did was to look primarily at the evidence presented to him of words and conduct contemporaneous with and proximate to the transfers in question in order to ascertain the subjective intention of the transferor at the time of the transfers. That evidence was of words allegedly passing between the father and the wife, and of conduct evidenced by the content and pattern of written declarations of trust and instruments of transfer. He concluded that the content of the documents and the stark contrast between them at a time when the father knew the effect of a trust declaration and had the benefit of legal and accounting advice, spoke volumes as to the father's intention in respect of the transfers in issue. At such a time, he chose on some occasions to cause such declarations to be made and on other occasions not to do so. He chose on some occasions instruments of transfer to bear the words "change of nominee" and "transfer back from nominee to beneficial owner" and on other occasions not. Absent a reasonable explanation for the choices he made in March 2005 and December 2006 not to seek written declarations of trust and not to refer to the wife as a nominee, and to pay duty on each such transaction, it was clearly open to the judge to reject the contention that on those occasions he intended the wife to hold the shares on trust. 52. The father and the wife said, however, that there was indeed a reasonable explanation for wanting her and her brothers to hold as mere nominees and, save in the unusual circumstances pertaining at the end of 2004, to do so without the need for documents to evidence that status. But the palpable implausibility of that explanation was soundly revealed by the judge and by the Court of Appeal as not according with the common sense of the matter; and the logic of their analyses is not susceptible to valid criticism. 53. There was no need in the circumstances for the judge to invoke the presumption of a resulting trust or to look to the presumption of advancement and he did not do so. The presumptions of resulting trust and advancement arise as a matter of default; in other words, where there is otherwise insufficient evidence of subjective intention. That is the point made in the passage cited by the judge from Lavelle that : “Normally there will be evidence of the intention with which the transfer is made. Where there is not, the law applies presumptions.”[33] and summarised thus, that: “Even where this [an express declaration of trust] is absent, the court aims to arrive at the parties' real intentions by considering direct evidence of the entire transaction. This requires an objective inference drawn from the parties’ words and conduct. As a result, the presumptions of resulting trust or of advancement are only relied upon as default rules where there is no sufficient evidence to displace them.”[34] Non-proximate conduct 54. In that it is said that the judge’s conclusion evidenced a reliance on conduct proximate to the transfers themselves to the wrongful exclusion of non-proximate conduct (the 1997 and 2004 returns of shares at the father’s request; and the 2012 property purchases), the answer is, first, that he did not ignore the earlier or later events or the suggested significance of them. He mentioned them as matters to which he had close regard[35] and expressly recognised the reliance placed on them[36]. Whether he revealed what he made of those events and why, despite those events, he rejected the contention of a trust arrangement is a matter which merits scrutiny and to which I will shortly turn. 55. A question canvassed in the course of argument before this Court was whether those earlier and later events were admissible as evidence of intention at the time of transfers. It seems to me that they were and that the issue was one of weight. In the context of a question whether evidence of conduct subsequent to the registration of shares in the names of children was admissible to rebut the presumption of advancement, it was held in Shephard v Cartwright[37] that whereas acts and declarations of the parties before or at the time of the transaction or so immediately after it as to render it a part of that transaction are admissible “either for or against the party who did the act or made the declaration.. . subsequent declarations are admissible as evidence only against the party who made them, not in his favour.”[38] The modern approach is less rigid in relation to evidence of subsequent conduct: “… it does not follow that subsequent conduct is necessarily irrelevant. Where the existence of an equitable interest depends upon a rebuttable presumption or inference of the transferor’s intention, evidence may be given of the subsequent conduct in order to rebut the presumption or inference which would otherwise be drawn.”[39] 56. As a matter of common experience, contemporaneous conduct is inherently more likely to be a reliable indicator of intention, to be given greater weight, than are words and conduct after the event, especially in the case of “ self serving statements or conduct of the transferor, who may long after the transaction be regretting earlier generosity.”[40] That rationale is not restricted to evidence in rebuttal of presumptions but embraces any evidence from which an inference of the transferor’s intention may properly be drawn. 57. The 1997 and 2004 transfers to the father by Cham and Kwun, respectively, predated the 2005 transfer and 2006 allocation to the wife; and were admissible in support of the father’s theme that shares in his children’s name were to be returned to him upon his request. So too in my judgment was the father’s 2012 injection of properties, even though subsequent to 2005 and 2006 events and, as to weight, such a substantial investment might justifiably be said not to be in the nature of a self serving afterthought. 58. That the judge treated the 1997 and 2004 transfers to the father and the 2012 purchases of property as admissible evidence as to intention is clear enough and it is also apparent that that evidence did not move him to accept the father’s and the wife’s central contention. What, however, has prevented a summary disposal of this application to review concurrent findings of fact, is the absence of an express explanation by the judge as to why that was so; why, in other words, that evidence did not persuade him to a different result. Adequacy of reasons 59. The question whether reasons provided in a judgment are adequate is always case and issue specific, and is a question which is to be approached with common sense [41]. It must, however, be made apparent to the parties, especially to the losing party, from the reasons which are provided, why the judge has arrived at his decision and : “ … if the appellate process is to work satisfactorily, the judgment must enable the appellate court to understand why the judge reached his decision. This does not mean that every factor which weighed with the judge in his appraisal of the evidence has to be identified and explained. But the issues the resolution of which were vital to the judge’s conclusion should be identified and the manner in which he resolved them explained. It is not possible to provide a template for this process.”[42] 60. The pattern of return of shares to the father, in so far as two occasions might be termed a pattern, as well as the injection of properties in 2012, were key planks of the case for the father and the wife and, in my judgment, it would have been preferable had the judge explained in terms how those events sat comfortably with his finding that, nonetheless, the father’s intention in March 2005 and December 2006 was to gift shares to his daughter. Yet the question is whether, despite the absence of a direct explanation, his reasoning in that regard is sufficiently apparent. 61. The weight of the evidence which constituted these key planks was expressly addressed by the Court of Appeal in the context of submissions that the judge had overlooked them. The Court said that not only had the judge taken them into account but that : “… just as [counsel for the father] had sought to persuade us by looking at the pattern of transfers and allotments in a particular light, equally if not more cogent reasons could be advanced showing that the transfer in dispute was not a trust arrangement.”[43] The Court thereupon tabulated compelling factors which, notwithstanding the pattern relied upon, sat ill with the notion of a transfer to the wife as trustee[44]. 62. It is sufficiently apparent from a sensible reading of the first instance judgment as a whole that the judge’s reasoning as to the significance of the two planks of evidence so much relied upon by the father and the wife, was the same as that subsequently articulated by the Court of Appeal; which is to say, that the account offered by the father and the wife made no sense either in practical terms or in the light of the documentary history preceding and accompanying the transfer in March 2005; and that, since no consideration was in fact paid, the only conclusion to which the evidence and the common sense of the matter drove him was that a gift had been intended. 63. These findings by the judge and the Court of Appeal do not sit at odds with the incidents upon which the appellants rely. They are consistent with a desire by the family’s patriarch to reward by gift those who worked with him but with a willingness on the part of the siblings, as a matter of filial rather than legal obligation, to convey beneficial ownership to him should unusual circumstances so indicate. 64. As for the suggested significance of the injection of property assets into NPL after filing of the petition, the weight of the point is highly doubtful since there was not a scintilla of evidence of the impact of those acquisitions on the value of NPL shares; and it is the shares, not the properties, which were liable to form part of the matrimonial assets. 65. In the event, I am persuaded that the complaint of inadequate reasons must fail. Fresh evidence 66. There was an application to this Court by the wife and father that, for the purpose of these appeals, they be permitted to adduce evidence of the purchase by Yee Fung of Mainland property in the names of several individuals and of a finding by a Mainland court in August 2015 that such of those properties as were conveyed into the name of the husband, were held by him on trust for Yee Fung. That finding was contrary to his contention prior to and in the Mainland proceedings that the four flats in his name were held by him beneficially. The proposed evidence, including the ruling, is said to exemplify the father’s practice of appointing family members and others to hold properties as nominees without formal declarations of trust. 67. We rejected the application. Issues of relevance aside, no adequate reason was advanced for the failure previously to adduce or to apply to adduce the evidence. The properties were purchased in 2006 and it was known by Yee Fung in mid-2014 that the husband disputed Yee Fung’s ownership claim. The hearing of the NPL dispute before the District Court took place in November 2012 and the hearing in the Court of Appeal took place in November 2014 without any reference in either forum to the Mainland property, in respect of which it would have been open to the father and the wife to apply to adduce evidence from persons, other than the husband, in whose names flats in the same district had been purchased at the same time. 68. The father’s explanation was that due to the insignificant value of the flats, he forgot about them during preparation for trial of the preliminary issue and was not reminded of them until May 2014 when, as a result of the termination of the employment by Yee Fung of one of the property holders, he was reminded of them by Yee Fung’s accounts department. Even then, he did not appreciate the relevance of the trust arrangements for those flats to the issue of beneficial ownership of the NPL shares. 69. The first condition for the reception of fresh evidence at the appellate stage is that the evidence could not have been obtained with reasonable diligence for use at the trial[45]. The appellants did not begin to meet that condition. Conclusion 70. For the reasons which I have provided, I would dismiss this appeals and make an order nisi that the costs of the appeals and of the applications to adduce fresh evidence be to the respondent, to be taxed if not agreed. Lord Collins of Mapesbury NPJ: 71. I agree with the judgment of Mr Justice Stock NPJ. Mr Justice Ribeiro PJ: 72. The Court unanimously dismisses the appeals and an order as to costs is made in the terms proposed in paragraph 70 above. Mr Russell Coleman SC and Mr Robin Egerton, instructed by Simon C W Yung & Co, for the Petitioner (Appellant in FACV 10/2015) Mr Charles Sussex SC, Mr Neal Clough and Mr Timothy Parker, instructed by T C Foo & Co, for the Respondent (Respondent in FACV 10/2015 and FACV 11/2015) Mr Ashley Burns SC, Ms Maggie Wong and Ms Tanie Toh, instructed by Li, Wong & Lam & W I Cheung, for the Intervener (Appellant in FACV 11/2015) [1] Deputy District Judge Carlson, FCMC 14018 of 2011, 4 March 2013. [2] Cheung, Yuen and Kwan JJA, CACV 151 of 2013 and CACV 152 of 2013, 4 December 2014. [3] The repeal took effect on 24 December 2014: section 7 Administration of Justice (Miscellaneous Provisions) Ordinance. [4] Leave was granted by the Court of Appeal on 23 April 2015. [5] Srimarti Bibhabati Devi v Kumar Narayan Roy [1946] AC 508 at 521-522, the test applied by this Court in Sky Heart Ltd v Lee Hysan Co td (1997-98) 1 HKCFAR 318; Wu Yee Pak v Un Fong Leung (2004) 7 HKCFAR 498; Chinachem Charitable Foundation Ltd v Chan Chun Chuen (2011) 14 HKCFAR 798; and Wealth Duke Ltd v Bank of China (Hong Kong) Ltd (2011) 14 HKCFAR 863. [6] The father’s affirmation dated 25 September 2012, para 21. [7] A contention described by the judge in the course of proceedings as incontrovertible. [8]Judge Melloy 25 September 2012. [9] Consent summons dated 13 September 2012. [10] Paras 30 and 31, per Kwan JA. [11] Court of Appeal judgment, para 30, citing with approval TL v ML & Ors [2006] 1 FCR 465 at [34] and [37]. [12] District Court judgment para 43. [13] [2004] 2 FCR 418. [14] Ibid para 55. [15] The report of the judgment says “A” but that must be a slip. [16] The emphasis is that supplied by Judge Carlson. [17] Judgment para 58. [18] Ibid para 59. [19] Ibid para 60. [20] Ibid para 66. [21] Ibid para 67. [22] (1894) 6 R 67 HL. See para 43 below, fn. 27. [23] Kwan JA, with whom the other members of the Court agreed. [24] Para 80. [25] Ibid paras 81 and 88. [26] Para 7 above. [27] See para 37 above. Put broadly, the principle in Browne v Dunn is that where a party intends to impeach the credibility of a witness on a material point, that party is bound to challenge the witness in cross-examination so as to give the witness an opportunity of making any explanation open to him; failure to do so may be taken as an acceptance of his evidence. But there are cases “in which [notice of the challenge] has been so distinctly and unmistakably given and the point upon which he is impeached, or is to be impeached, is so manifest, that it is not necessary to waste time in putting questions to him upon it”: Browne v Dunn (1894) 6 R 67 HL at 70-71. See also Markem Corp v Zipher Ltd [2005] R.P.C. 31 at [57] - [61]; and Phipson on Evidence, 18th ed., [12-12] and [12-35]. [28] See para 30(1) above. [29] See para 29 above. [30] The question whether the presumption of advancement applies to the case of a transfer without consideration by a parent to an adult independent child has been left open by the Court of Appeal : Suen Shu Tai v Tam Fung Tai [2014] 4 HKLRD 436 at [10.16] and by this Court upon appeal from that judgment: FACV No 9 of 2015, 5 November 2015. [31] See para 61 judgment; para 30 above: “ … he intended, I use the word advisedly, to transfer the shares to her outright, which is what he did. Had it been otherwise he would have created a specific instrument of trust as he had in the past. This must have been, I find this as a fact, a deliberate choice to bring about a transfer of the beneficial interest in the shares.” (emphasis added). [32] See his emphasis on para [19] of the judgment in Lavell, referred to at para 27(2) above. [33] Lavelle at [14]. See also Au Yuk Lin v Wong Wang Hin [2013] 4 HKLRD 373 at [18] to [19]; and Pettitt v Pettitt [1970] AC 777 at 811 and 813, passages referred to in Lavelle at [17]. [34] Snell’s Equity 33rd ed., at 25-11(a). [35] See para 31 above; paras 58 and 59 of the District Court judgment. [36] Judgment paras 15 and 47. [37] [1955] AC 431 at 445. [38] Citing Snell’s Principles of Equity (24th Ed. p 153). [39] Tribe v Tribe [1996] Ch 107 at 129 per Millett LJ, as he then was. See also Lavelle, above, at [17] to[19] and Snell’s Equity 33rd ed., 25-013. [40] Lavelle para [19]. [41] Oriental Daily Publisher Ltd v Commissioner for Television and Entertainment Licensing Authoriy (1997-98) 1 HKCFAR 279 at 290-291. [42] English v Emery Reimbold & Strick Ltd [2002] 1 WLR 2409 at [19]. See also as to the duty to give adequate reasons: Zhou Cui Hao v Ting Fung Yee 1999 3 HKC 635; Welltus v Fornton Knitting [2013] 5 HKC 105. [43]Court of Appeal judgment, para 77. [44] Ibid paras 78 to 80, repeated at para 39 above. [45] Ladd v Marshall [1954] 3 All ER 745 at 748. 
Chief Justice Ma: 1. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Ribeiro PJ: 2. Under section 122(1) of the Crimes Ordinance,[1] a person who indecently assaults another is guilty of an offence and is liable on conviction on indictment to imprisonment for 10 years. If there is consent to the relevant acts, such conduct, although “indecent”, generally does not constitute an assault so that the offence is not committed. However, section 122(2) provides: “A person under the age of 16 cannot in law give any consent which would prevent an act being an assault for the purposes of this section”. 3. The question which arises in this appeal is whether a person commits the offence if he engages in sexual conduct which is in fact consensual with a girl who is actually aged 13 when he honestly and reasonably believes her to be aged 16 or above.[2] A. The facts 4. In August 2014, the appellant, then a third year university student aged 22, visited an adult website on which the girl (referred to throughout as “PW1”) had posted an advertisement, describing herself as aged 17 and offering sexual services at listed prices. The appellant asked for, and received from her, a half-length photo of herself. They arranged to meet and went to a guest house where they showered together and the appellant ran his hands over her body. PW1 then performed oral sex on him. 5. The appellant testified that he thought that her photo was consistent with her being aged 17 and that he believed that this was borne out when they met, PW1 appearing to him to be relatively tall, with well-developed bodily features, and speaking in a mature manner. He said that he did not suspect that she was under 16. PW1 gave evidence that she would dress more maturely when meeting clients. B. The magistrate’s decision 6. The deputy magistrate, Mr Peter Hui Shiu-keung,[3] rejected the prosecution’s submission that the offence was one of absolute liability and held, following the House of Lords decision in R v K,[4]that where there is actual consent, an honest belief on the defendant’s part that the girl is aged 16 or more results in an acquittal. 7. He found that PW1 “really looked more mature than her actual age, and did not look like a 13-year old girl”[5] in the light “of her appearance, build and speaking tone”.[6] Taking account of PW1 having claimed to be 17 and her “deliberate attempt to appear to be mature when meeting her clients”, the magistrate found that there was “no apparent reason that [the appellant] should have any suspicions about the age PW1 claimed”[7] and concluded that the appellant should be acquitted since “the defence that the defendant honestly and reasonably believed that PW1 was aged 16 or above is not rebutted”[8]. The prosecution appealed by way of case stated. C. The Judge’s decision 8. The magistrate’s ruling was reversed by Deputy High Court Judge Stanley Chan.[9] He relied principally on the decisions of the Court of Appeal[10] and this Court[11] in HKSAR v So Wai Lun (involving unlawful sexual intercourse with a girl under 16)[12] and held that, as a matter of necessary implication, the legislative intent was that indecent assault should be an offence of absolute liability. The appellant therefore could not escape conviction on the basis of an honest and reasonable belief as to PW1’s age. I shall return to consider in greater detail the arguments which led his Lordship to that conclusion. D. Leave to appeal 9. The Appeal Committee[13] granted leave to appeal, certifying the following questions of law as being of the requisite importance, namely: (i) Whether an offence contrary to s 122 (1) & (2) of the Crimes Ordinance, Cap 200, taken together is an offence of absolute liability when the alleged victim is a person under 16 years of age. (ii) Whether an accused charged under s 122 (1) with indecently assaulting a person who was under 16 years of age can legally put forward a defence that the person in fact consented and the accused genuinely believed that he/she was 16 years of age or over. (iii) Whether in a prosecution under s 122 (1) where the alleged victim is a person under 16 years of age the prosecution is required to prove absence of genuine belief on the part of the accused that the person was 16 years of age or over. E. Determining the mental elements of a statutory offence 10. Those questions fall to be answered in the light of significant developments in this jurisdiction within the last decade as to how the mental elements of a statutory offence are to be determined and, in particular, as to when such an offence is to be treated as one of absolute liability. 11. The absolute liability approach to age-related sexual offences illustrated by nineteenth century English cases like R v Prince,[14] had previously held sway. Thus, in Prince, a statute made it an offence unlawfully to take an unmarried girl under the age of sixteen out of the possession and against the will of her father but was silent as to the defendant’s required mental state regarding her age. The defendant was convicted although the jury found that he “bonâ fide, and on reasonable grounds, believed that she was above sixteen, viz, eighteen years old.”[15] This was upheld by a Court consisting of 16 judges[16] with Bramwell B, who wrote one of the main judgments, commenting that to require proof of mens rea in respect of her age would mean “reading the statute with some strange words introduced; as thus: ‘Whosoever shall take any unmarried girl, being under the age of sixteen, and not believing her to be over the age of sixteen, out of the possession’, &c”[17]. Such a construction of the Act was rejected. 12. That approach was applied in Hong Kong, as is evident from cases decided by the Court of Appeal as late as in 1993 and 1997.[18] However, in HKSAR v So Wai Lun,[19] Ma CJHC (as the Chief Justice then was) observed that in the light of more modern decisions,[20] Prince had been “largely discredited and all but overruled by the House of Lords” and held that the abovementioned Court of Appeal decisions had to be regarded as of “extremely dubious” authority. 13. As Ma CJHC noted,[21] instead of demanding a justification for reading in a mens rea requirement, the modern starting-point is that mens rea is presumed to be an essential ingredient where the statute is silent on the mental element unless that presumption is displaced expressly or by necessary implication.[22] 14. The presumption of mens rea flows from recognition that it is a cardinal principle of our criminal law that mens rea, involving the intentional or knowing performance of prohibited conduct, is ordinarily an essential ingredient of guilt of a criminal offence. The law therefore assumes that in creating a statutory offence, the legislature does not intend to dispense with that basic principle unless the enactment does so expressly or by necessary implication. This has been said to reflect the principle of legality[23] or to be a principle of statutory interpretation whereby any ambiguity in a penal statute is resolved in favour of the accused.[24] 15. The major developments in this jurisdiction relate to cases where the presumption of mens rea is held to be dislodged. As this Court concluded in Hin Lin Yee v HKSAR,[25] it becomes necessary in such event to ask: “By what, if any, mental requirement is the supplanted requirement of mens rea to be replaced?” 16. The English approach (previously followed in Hong Kong) presents a stark choice between construing the statute as requiring full mens rea and construing it as imposing absolute liability.[26] This has led eminent judges to lament the absence of an option to establish liability on some intermediate basis, such as a construction requiring the accused to “convince the jury that on balance of probabilities he is innocent of any criminal intention”.[27] Such an option was, however, thought to have been excluded by Woolmington v Director of Public Prosecutions.[28] 17. After considering different approaches in Australia, Canada and New Zealand, this Court decided in 2010 to depart from the confining regime adopted in England and Wales. It was held in Hin Lin Yee v HKSAR,[29] that in appropriate cases, the law permits statutes to be construed as intending to displace the presumption of mens rea in favour of an intermediate basis of liability. Five possible bases of liability were recognised, ranging from full mens rea to absolute liability with three intermediate possibilities. 18. Hin Lin Yee was a case involving the relatively minor offence of selling a drug intended for use by man but unfit for that purpose[30] where the mental element in question related to the circumstance of the drug’s unfitness. In Kulemesin v HKSAR,[31] which involved the more serious offence of endangering the safety of others in a vessel or at sea,[32] the issue concerned the mental element regarding the consequence of endangerment. To accommodate more serious offences and to cater for mens rea as to the consequences of (in addition to the circumstances accompanying or surrounding) the prohibited act, the five possible bases of criminal liability formulated in Hin Lin Yee were slightly amended and re-formulated as follows: “... the five possible alternatives may be stated as follows...: (a) first, that the mens rea presumption persists and the prosecution must prove knowledge, intention or recklessness as to every element of the offence (‘the first alternative’); (b) second, that the prosecution need not set out to prove mens rea, but if there is evidence capable of raising a reasonable doubt that the defendant may have acted or omitted to act in the honest and reasonable belief that the circumstances or likely consequences of his conduct were such that, if true, liability would not attach, he must be acquitted unless the prosecution proves beyond reasonable doubt the absence of such exculpatory belief or that there were no reasonable grounds for such belief (‘the second alternative’); (c) third, that the presumption has been displaced so that the prosecution need not prove mens rea but that the accused has a good defence if he can prove on the balance of probabilities that he acted or omitted to act in the honest and reasonable belief that the circumstances or likely consequences of his conduct were such that, if true, he would not be guilty of the offence (‘the third alternative’); (d) fourth, that the presumption has been displaced and that the accused is confined to relying on the statutory defences expressly provided for, the existence of such defences being inconsistent with the second and third alternatives mentioned above (‘the fourth alternative’); and (e) fifth, that the presumption is displaced and the offence is one of absolute liability so that the prosecution succeeds if the prohibited act or omission is proved against the accused, regardless of his state of mind regarding the relevant elements of the offence in question (‘the fifth alternative’).”[33] 19. I shall refer to these reformulated alternatives as “the Kulemesin alternatives”. The Court held that all five alternatives should be considered as possible conclusions when construing statutory criminal offences – both serious and regulatory – which are silent or ambiguous as to the state of mind relevantly required.[34] 20. The view has been expressed[35] that “reasonableness of belief” should be rejected as a standard for criminal liability, since such liability is properly concerned with a subjective state of mind and ought not to be displaced by an objective standard akin to one involving negligence. 21. However, in incorporating “honest and reasonable belief” as part of the second and third Kulemesin alternatives, a different view was preferred. As Lord Reid pointed out in Sweet v Parsley,[36] the court’s choice is “much more difficult if there were no other way open than either mens rea in the full sense or an absolute offence; for there are many kinds of case where putting on the prosecutor the full burden of proving mens rea creates great difficulties and may lead to many unjust acquittals”; and it would often be much easier to infer that the legislature “must have meant that gross negligence should be the necessary mental element than to infer that [it] intended to create an absolute offence”. 22. I might add that objective standards are in any event applicable to an offence like indecent assault, in that the courts have interpreted “indecency” as “conduct that right-thinking people will consider an affront to the sexual modesty of a woman;”[37] or conduct “so offensive to contemporary standards of modesty and privacy as to be indecent;”[38] or conduct “which right-minded persons would clearly think was indecent”.[39] F. The presumption of mens rea 23. Applying the principles, the analysis should begin by considering how precisely the presumption of mens rea applies to sections 122(1) and 122(2) which provide as follows: (1) Subject to subsection (3),[40] a person who indecently assaults another person shall be guilty of an offence and shall be liable on conviction on indictment to imprisonment for 10 years. (2) A person under the age of 16 cannot in law give any consent which would prevent an act being an assault for the purposes of this section. F.1 The presumption and indecent assaults on persons aged 16 or over 24. As noted above, it is presumed that the prosecution must prove knowledge, intention or recklessness as to every actus reus element of a statutory offence. Where the actus reus ingredient is expressed negatively such as requiring the absence of consent, the presumption is that the prosecution must negative the existence of consent and negative any claimed belief by the defendant in its existence. 25. In indecent assault cases, leaving aside for the moment those involving persons under the age of 16, two main actus reus elements must be proved. First, the prosecution must prove an assault on the other person. For this purpose, “Any intentional touching of another person without the consent of that person and without lawful excuse” constitutes an assault.[41] It follows that where there is genuine consent freely and lawfully given, there is no assault and the offence is not committed.[42] If a potentially believable possibility that there was consent arises on the evidence, the prosecution has the onus of negativing such consent.[43] 26. The second actus reus element which the prosecution must prove is that the assault was “accompanied by circumstances of indecency towards the person alleged to have been assaulted”.[44] This requires proving that the conduct was indecent, applying (as we have seen) objective “right-thinking” standards as to what constitutes indecency. 27. The presumption of mens rea requires the prosecution to prove mens rea corresponding to those two actus reus elements. Thus, the defendant must be shown to have intended to commit an indecent assault on the other person. This means that the prosecution must prove that the defendant intended to lay hands on the victim (or otherwise assault her) without her consent.[45] If there is evidence giving rise to a reasonable doubt as to whether the defendant believed her to be consenting, the prosecution must negative that belief.[46] Insofar as the conduct is equivocal as to whether it is indecent, indecency can be proved by evidence of the defendant’s indecent purpose.[47] F.2 The presumption and indecent assaults on persons under the age of 16 28. Section 122(2) deprives persons under the age of 16 of the capacity to consent to an act which, if not consensual, would amount to an indecent assault. It operates to modify both the actus reus and mens rea requirements of the offence in relation to persons under 16. 29. Thus, as we have seen, for persons aged 16 or over, actual consent negatives the actus reus and an honest belief in the other person’s consent negatives the mens rea. However, by deeming persons under the age of 16 to be incapable of giving consent, section 122(2) makes an indecent act towards such a person as an assault even though the evidence clearly establishes that there was consent in fact. 30. In cases where the accused asserts that there was actual consent, he will naturally also be asserting that he honestly believed that the other person was consenting. If such a belief were capable of negativing the mens rea of the offence, the protective purpose of the age-related restriction on capacity to consent would be defeated. Section 122(2) thus necessarily implies that it does not avail the accused to establish that he honestly (and correctly) believed that the under-aged person was freely consenting to the acts in question. The effect of section 122(2) is, in other words, to eliminate consent as an ingredient from both the actus reus and mens rea of indecent assault. The presumption of mens rea therefore does not concern any element of consent in cases involving persons under the age of 16. 31. However, the defendant’s mental state regarding the other person’s age stands on a different footing. While in cases involving persons aged 16 or over, the victim’s age is not relevant,[48] the fact that the victim is under 16 is an essential ingredient of the offence as modified by section 122(2). Indecent assault becomes an offence which the defendant commits by doing an indecent act towards the victim, being a person under 16, with or without that person’s consent.[49] The age of the victim therefore forms part of the actus reus of the offence as modified by section 122(2) and thus engages the presumption of mens rea. 32. This accords with the analysis by Lord Hobhouse of Woodborough of section 14(2) of the United Kingdom’s Sexual Offences Act 1956 which is replicated by our section 122(2). His Lordship stated: “Section 14(2) provides a fact-based legal rule which, given the stated factual situation, qualifies the requirement that the indecent act be done without the actual consent of the other person. The additional fact is that the other person is under the age of 16 years. The result is that the actus reus becomes an indecent act done either without the consent of the other person or with or without the consent of the other person being a person under the age of 16 years. The prosecution must therefore prove as regards the actus reus either the fact of the absence of consent or the fact of an age of less than 16 years.”[50] 33. Having concluded that the presumption of mens rea is engaged regarding the alleged victim’s age, the questions which fall to be considered are whether that presumption is displaced and, if so, displaced by what. G. Is the presumption displaced? 34. As stated above, a statutory offence is presumed to require proof of mens rea unless the presumption is displaced expressly or by necessary implication.[51] Whether such displacement occurs is a matter of statutory construction requiring examination of the statutory language; the nature and subject-matter of the offence; the legislative purpose and any other matters indicative of the statutory intent.[52] 35. As pointed out in Hin Lin Yee,[53] the availability of intermediate bases of liability inevitably influences the approach to deciding whether a dislodging of the presumption is intended. A court may recoil from imposing absolute liability for a particular offence and so hold against displacement, while it may be prepared to hold that the presumption is supplanted in favour of the second or third Kulemesin alternative. The decision in R v K[54] that the offence of indecent assault requires proof of full mens rea, should be viewed in that context. With no intermediate option, it is not surprising that the House of Lords ruled in favour of requiring mens rea instead of absolute liability. 36. In the present case, sections 122(1) and 122(2) are silent as to the mental ingredient, if any, required regarding the other person’s age. It is a serious offence “in terms of penalty and social obloquy”, carrying a maximum sentence of ten years’ imprisonment on indictment which is a feature tending to favour non-displacement of the presumption of mens rea.[55] 37. However, given the manifest purpose of section 122(2) viewed in the light of the Hong Kong courts’ long-standing policy regarding age-related sexual offences, I am of the view that the presumption of mens rea is clearly displaced in respect of indecent assaults on persons under the age of 16 in this jurisdiction. 38. We have seen that the effect of section 122(2) is to deprive persons under that age of the capacity to consent to an indecent act and thus to transform the offence into one of significantly stricter liability. The manifest statutory intention is to confer special protection on a class of vulnerable persons who are potential objects of indecent acts. The long-standing approach of our courts has been to implement such protection by construing age-related sexual offences as requiring potential defendants to take care “to avoid what may be unlawful and steering well away from the line between legality and illegality”, holding that this “would add materially to the protection for young girls” provided for by the relevant statute.[56] Some of those decisions were arrived at before the development of intermediate bases of liability and led to conclusions in favour of absolute liability, indicating the force of that policy. The existence of intermediate options does not affect its continuing vitality. The statutory purpose would plainly be compromised if the presumption of mens rea is not dislodged in respect of the girl’s age and the prosecution were required to disprove, beyond reasonable doubt, any claim made by the defendant that he honestly believed that she was aged 16 or over. H. Displaced by what? H.1 Absolute liability 39. Absolute liability departs from basic common law principles of criminal responsibility which generally require some degree of knowledge or intention to accompany the prohibited conduct. As this Court recognised: “... it is in principle objectionable, especially where the offence is serious, that a person should be made criminally liable where he did not deliberately or recklessly engage in the prohibited conduct or where he was ignorant of circumstances making his conduct criminal or where he acted harbouring an honest and reasonable belief inconsistent with liability.”[57] 40. It is thus never lightly to be inferred that the legislature intended to create an offence of absolute liability since, as Lord Reid put it: “... someone could be convicted of it who by all reasonable and sensible standards is without fault.”[58] 41. Absolute liability is primarily imposed for what are essentially regulatory offences[59] rather than serious criminal offences, and then only where some useful purpose (such as encouraging preventive safety measures) may be served by the imposition of such liability.[60] As was made clear in Hin Lin Yee[61] and Kulemesin,[62] before concluding that an offence is one of absolute liability, it is necessary to consider whether the statutory purpose can sufficiently be met by construing it as laying down a less Draconian, intermediate form of liability. Absolute liability should only be resorted to if the answer is in the negative. 42. The main reason for the Judge deciding in favour of absolute liability was his application of the Court of Appeal’s judgment in HKSAR v So Wai Lun[63] where it was held that absolute liability was applicable to unlawful sexual intercourse with an under-aged girl contrary to section 124 of the Crimes Ordinance. That section provides: “Subject to subsection (2),[64] a man who has unlawful sexual intercourse with a girl under the age of 16 shall be guilty of an offence and shall be liable on conviction on indictment to imprisonment for 5 years.” 43. That decision is however clearly distinguishable. It was concerned with unlawful sexual intercourse and not indecent assault, and the Court of Appeal determined that absolute liability was imposed principally on the basis of the legislative history of the section 124 offence which differs significantly from that of section 122. 44. The Court of Appeal traced section 124 back to section 6 of the Women and Girls' Protection Ordinance enacted in 1890,[65] which provided as follows: “Any person who carnally knows or attempts to have carnal knowledge of any unmarried girl being of or above the age of twelve years and under the age of sixteen shall be guilty of a misdemeanour and on conviction thereof shall be liable to the punishment hereinafter mentioned. Provided that where both parties are Asiatics a girl shall not be deemed unmarried within the meaning of this ordinance if she is duly married according to the laws and customs of the native country of the girl. Provided also it shall be a sufficient defence to any charge under this section, if it shall be made to appear to the Court or jury before whom the charge shall be brought that the person so charged had reasonable cause to believe that the girl was of or above the age of sixteen years. Provided also that no prosecution shall be commenced for an offence under this section more than three months after the commission of the offence.”[66] (Italics supplied) 45. That section was replaced seven years later in 1897 by section 5 of the Protection of Women and Girls Ordinance[67] which deleted the italicised defence involving “reasonable cause to believe”,[68] leading the Court of Appeal in So Wai Lun to conclude that “the reasonable belief defence has been expressly rejected in Hong Kong”.[69] This was regarded as a decisive indication that the presumption of mens rea was supplanted: “For the above reasons, we are of the view that in relation to a charge under s 124 of the Crimes Ordinance of unlawful sexual intercourse with a girl under the age of 16, the belief of the defendant that the girl was 16 or more does not provide a defence. In other words, the offence is accordingly one of absolute liability in this respect.”[70] (Italics supplied) 46. As indicated above, So Wai Lun may be distinguished in two important respects. First, in the legislative history of indecent assault, there is no equivalent of an express “reasonable belief” defence later being abolished. Indecent assault is traceable back to the Offences Against the Person Ordinance 1865, section 46;[71] which was succeeded by the Protection of Women and Girls Ordinance 1897, section 7;[72] then, in 1978,[73] by what is essentially the present section 122 of the Crimes Ordinance. That provision was then amended in 1991[74] to increase the maximum penalty from five to ten years’ imprisonment. At no point was there any express reasonable belief defence, much less a later abandonment of such a defence. 47. Secondly, the Court of Appeal’s judgment in So Wai Lun was delivered before the recognition of intermediate bases of liability. Thus, as the italicised last sentence of the abovementioned citation indicates, once the Court of Appeal held that the presumption of mens rea was displaced, it concluded that the offence was “accordingly” one of absolute liability. It was a decision reflecting the stark choice then available between full mens rea and absolute liability. 48. If that case had been decided today, the Court would have proceeded to consider whether the statutory purpose of the offence could be sufficiently met by construing section 124 as laying down an intermediate mental requirement in place of full mens rea.[75] It seems entirely possible that a different conclusion would have been reached since the Court was much troubled by its having to decide in favour of absolute liability. This was because its decision logically meant that an offence under section 123,[76] punishable by life imprisonment, ought also to be construed as one of absolute liability: “The aspect that has taxed us most about this conclusion is that there is no doubt that s 124(1) involves a ‘truly criminal’ offence with serious consequences (a maximum of 5 years' imprisonment). Moreover, our conclusion on s 124 must equally apply to the s 123 offence of unlawful sexual intercourse with a girl below the age of 12. For this latter offence, the maximum penalty is life imprisonment. The provisions of ss 123 and 124 are virtually identical save that in respect of the latter, a statutory defence is available (s 124(2)). It seems at first startling that a crime that carries with it such severe penalties (in fact the severest in the case of s 123) should be one of absolute liability (not even strict liability as defined above). Yet, this remains the position in many common law jurisdictions and is a conclusion that is, given the matters already discussed as to the statutory construction and the legislative history, compelling.”[77] 49. It is true that in this Court, it was later acknowledged[78] that So Wai Lun had held a section 124 offence to be one of absolute liability, but the basis of the Court of Appeal’s decision was not subject to argument or scrutiny. It was merely assumed to be correct in its conclusion. 50. The question whether the Court of Appeal’s decision was right, in particular in treating the removal by the 1897 Ordinance of the reasonable belief defence introduced in 1890 as a decisive indication of legislative policy which carries down to the present, should be left open. If it is revisited in the future, it would no doubt be necessary to consider whether the 1890 offence may be said to be a different offence from that created by section 124[79] and also relevant to consider any changes to the general criminal law providing the context in which the two offences operate. 51. The relevance, if any, of R v Brown,[80] would also have to be considered. The United Kingdom Supreme Court there held that the offence of unlawful carnal knowledge of a girl under the age of 14 under the Criminal Law Amendment Acts (Northern Ireland) 1885-1923, section 4 is one of absolute liability in respect of the girl’s age. While its judgment may be thought to contain some echoes of the reasoning of the Court of Appeal in So Wai Lun, R v Brown is a case with markedly special features. Section 4 of the Northern Irish Act was held to be a specific response to the decision in R v Prince[81] having been enacted some 10 years after that decision, deliberately preserving absolute liability in relation to girls under the age of 13 (later raised to 14).[82] It also involved two directly overlapping provisions,[83] and, when a proviso allowing for a reasonable belief defence was eliminated from one of them (section 5), it was expressly enacted that reasonable cause to believe the girl was 17 or above “shall not be a defence” to a charge under that section.[84] Those features had a weighty bearing on the construction of section 4. 52. The Judge[85] also considered the doubling of the maximum sentence for indecent assault from five to ten years’ imprisonment in 1991 an indication that absolute liability was intended. However, the authorities generally hold that the more serious an offence is in terms of penalty and social obloquy, the more likely it is that the presumption of mens rea is sustained.[86] Hence the doubling of the maximum militates against, rather than in favour of, a construction favouring absolute liability. 53. For the aforesaid reasons and because, as indicated below, I do not consider absolute liability necessary to achieve the statutory purposes of section 122, I would reject the prosecution’s submission that the presumption of mens rea is displaced in favour of absolute liability regarding the girl’s age. H.2 The fourth Kulemesin alternative 54. Sections 122(3) and 122(4) state as follows: (3) A person is not, by virtue of subsection (2), guilty of indecently assaulting another person, if that person is, or believes on reasonable grounds that he or she is, married to that other person. (4) A woman who is a mentally incapacitated person cannot in law give any consent which would prevent an act being an assault for the purposes of this section, but a person is only to be treated as guilty of indecently assaulting a mentally incapacitated person by reason of that incapacity to consent, if that person knew or had reason to suspect her to be a mentally incapacitated person. 55. The Judge[87] furthermore held that because reasonable belief defences are provided for by subsections (3) and (4), defendants falling outside those subsections are not intended to have any defence and are subject to absolute liability. In other words, his Lordship held that the presumption of mens rea was displaced by the fourth Kulemesin alternative, subjecting the appellant to absolute liability when he fell outside those defences. 56. A similar argument was rejected by Lord Bingham of Cornhill in R v K[88] when construing section 14 of the Sexual Offences Act 1956 which is replicated in section 122 of the Crimes Ordinance.[89] His Lordship summarised the argument (which corresponds to the fourth Kulemesin alternative) thus: “... subsections (3) and (4) define circumstances in which a defendant's belief, knowledge or suspicion exonerate a defendant from liability for what would otherwise be an indecent assault; if it had been intended to exonerate a defendant who believed a complainant to be 16 or over, this ground of exoneration would have been expressed in subsection (2); the omission of such a provision makes plain that no such ground of exoneration was intended.”[90] 57. He commented that this submission would have had great force “[if] the provisions of section 14 were part of a single, coherent legislative scheme and were read without reference to any overriding presumption of statutory interpretation.” However, his Lordship pointed out that it was plainly “not part of a single, coherent legislative scheme” since the 1956 Act was a consolidation Act with its provisions derived from diverse sources, giving the Act a patchwork or “rag-bag” nature.[91] Lord Bingham therefore concluded (after examining the legislative history and decided cases) that: “... significance cannot be attached to the inclusion of grounds of exoneration in subsections (3) and (4) and the omission of such a ground from subsection (2), although subsections (3) and (4) do reflect parliamentary recognition that a defendant should not be criminally liable if he misapprehends a factual matter on which his criminal liability depends. There is nothing in the language of this statute which justifies, as a matter of necessary implication, the conclusion that Parliament must have intended to exclude this ingredient of mens rea in section 14 any more than in section 1.”[92] 58. His Lordship’s refutation of the argument is equally applicable to the submission made in connection with sections 122(3) and (4) which are an enactment of section 14 of the 1956 Act with its “rag-bag” pedigree. The legislative history of indecent assault and related offences in Hong Kong has no greater coherence. Thus, for instance, the maximum penalty of 10 years’ imprisonment for indecent assault (introduced in 1991) is twice that prescribed for unlawful sexual intercourse with a girl under 16 under section 124, the latter offence potentially being far more serious, involving a risk of unwanted pregnancy. 59. The fourth Kulemesin alternative is only adopted if the availability of the expressly enacted defences is inconsistent with the second and third alternatives also being available. No such inconsistency arises here. Subsections (3) and (4) provide defences in narrow and rare circumstances involving an honest and reasonable belief that one is married to the alleged victim; or the absence of any reason to suspect that the alleged victim is a mentally incapacitated person. The existence of such specialised defences is not inconsistent with having a defence based on belief that the victim was aged 16 or over, nor accordingly with construing section 122(2) as accommodating the second or third Kulemesin alternatives generally in cases falling outside subsections (3) and (4). H.3 The second or third Kulemesin alternatives 60. Having decided that the presumption of mens rea is dislodged and that neither absolute liability nor the fourth Kulemesin alternative are applicable, the choices devolve down to either the second or the third Kulemesin alternative as the appropriate mental requirement for an offence under section 122(2). 61. Both alternatives admit by way of defence, an honest and reasonable belief on the part of the accused that the circumstances or likely consequences of his conduct were such that, if true, he would not be guilty of the offence. The difference between them is that under the second alternative, the defendant bears only an evidential burden whereas the third alternative requires him to discharge a persuasive burden as to his belief. 62. In HKSAR v Lam Kwong Wai,[93] Sir Anthony Mason NPJ explains the difference as follows: “An evidential burden, unlike a persuasive burden, does not expose the defendant to the risk of conviction because he fails to prove some matter on which he bears an evidential onus. An evidential burden: ‘… requires only that the accused must adduce sufficient evidence to raise an issue before it has to be determined as one of the facts in the case. The prosecution does not need to lead any evidence about it, so the accused needs to do this if he wishes to put the point in issue. But if it is put in issue, the burden of proof remains with the prosecution. The accused need only raise a reasonable doubt about his guilt.’...[94] See also R v Lambert [2002] 2 AC 545 at p 588H, where his Lordship said: ‘What the accused must do is put evidence before the court which, if believed, could be taken by a reasonable jury to support his defence.’ A persuasive burden, on the other hand, requires a defendant to prove, on a balance of probabilities, an ultimate fact which is necessary to the determination of his guilt or innocence. The burden relates to an essential element of the offence. It reverses the burden of proof by transferring it from the prosecution to the defendant.[95] It may be either mandatory or discretionary in its operation. With a mandatory persuasive burden, it is possible for a conviction to be returned, even where the tribunal of fact entertains a doubt as to the defendant's guilt.[96]” 63. Under the second alternative, a defendant charged with indecent assault on an under-aged girl is required to put evidence before the court that he honestly and reasonably believed that she was 16 or over, leaving it to the prosecution, if it can, to negative that belief beyond reasonable doubt. And on the third alternative, the defendant must satisfy the court or jury on the balance of probabilities that he did honestly and reasonably believe that she was 16 or over. 64. The Court must determine which of those alternatives most closely reflects the statutory purpose of section 122. It is clear that such purpose is to treat girls (and boys) under the age of 16 as a vulnerable class in need of a high degree of protection against sexual exploitation. Hence, section 122(2) deems them incapable of giving consent to indecent conduct and holds the defendant guilty even though he can prove that consent was, to his knowledge, in fact given. As we have seen, in furtherance of that policy, the Hong Kong courts have construed age-related sexual offences as requiring potential defendants to take care “to avoid what may be unlawful and steering well away from the line between legality and illegality” to the extent of imposing absolute liability (when intermediate bases of liability had not yet been developed).[97] As Lord Kerr JSC put it in R v Brown,[98] it is a policy designed to “promote the objects of the statute by encouraging greater vigilance against sexual intercourse with girls under the age of [16].” 65. In my view, that purpose is achieved by construing section 122(2) in accordance with the third Kulemesin alternative, imposing a persuasive burden on the accused, in preference over the second alternative. 66. The second alternative provides greater protection than a requirement of full mens rea since guilt is established by the prosecution negativing either the accused’s alleged honest belief as to the victim’s age or the reasonableness of his alleged belief. [99] However, such protection does not appear to me to go far enough. Since the prosecution bears the burden of establishing the aforesaid negatives, an accused would still be acquitted if the court or jury thinks that it may well be the case that he did not honestly and reasonably believe she was of sufficient age, but that a reasonable doubt remains as to whether he did harbour such a belief. 67. The third alternative better reflects the statutory purpose. A man who says: “I honestly and reasonably believed the girl was old enough to consent” ought to be required to persuade the court or jury that he probably did in fact so believe, a matter which he is best placed to explain. It is a suitably demanding standard designed to encourage men to steer clear of indecent conduct with young girls who may fall within the protected class, placing them otherwise at peril of being unable to discharge the persuasive burden. 68. A reverse onus derogates from the constitutional right to be presumed innocent.[100] For such a derogation to be justified, it must pass the rationality and proportionality tests: the Court must be satisfied that the reverse onus has a rational connection with the pursuit of a legitimate aim and that it is no more than necessary for the achievement of that aim.[101] If those tests are met, the Court must be satisfied overall that adoption of a reverse onus strikes a reasonable balance between the societal benefits promoted and the inroads made into the constitutionally protected presumption of innocence and that it does not place an unacceptably harsh burden on the individual.[102] 69. In my view, construing section 122(2) as imposing a burden on the accused to prove on the balance of probabilities that he honestly and reasonably believed that the girl in question was aged 16 or more passes those tests. It is rationally connected with the legitimate aim of giving heightened protection to vulnerable under-aged girls and is no more than necessary to achieve such a level of protection. The inroad it makes into the presumption of innocence as the price of promoting necessary protection of a vulnerable class strikes a reasonable balance in the context of a fair trial for the accused. I. Conclusion 70. For the foregoing reasons, I conclude that the Judge was wrong to hold that section 122(2) imposes absolute liability where the victim is in fact under the age of 16. On its proper construction, the presumption has been displaced so that the prosecution does not need to prove mens rea as to the girl’s age, but the accused has a good defence if he can prove on the balance of probabilities that he honestly and reasonably believed that the girl was 16 or over. J. Disposal of the appeal 71. In acquitting the defendant, the magistrate stated: “It is my conclusion that the defence that the defendant honestly and reasonably believed that PW1 was aged 16 or above is not rebutted, and as the benefit of doubt -- reasonable doubt should go to the defendant, I find the defendant not guilty of the charge.”[103] 72. It is unclear whether the magistrate regarded the honest and reasonable belief referred to as a defence established by the accused or as an allegation which the prosecution failed to negative. Although Hin Lin Yee and Kulemesin were not cited, his statement effectively conflates the second and third Kulemesin alternatives. 73. If the magistrate held that the defendant discharged the persuasive burden and established a defence along the lines of the third alternative, his acquittal should be upheld as it would be in line with the conclusion reached in this judgment. If, however, the magistrate determined that the prosecution failed to negative the alleged honest and reasonable belief as to the girl’s age beyond reasonable doubt, the question of a possible re‑trial would arise since the defendant would not have been found to have established a reasonable belief defence on a balance of probabilities. 74. However, Ms Anna Y K Lai SC[104] accepted that the acquittal should be upheld on either reading of the Reasons for Verdict. The Court should therefore resolve the ambiguity by assuming that the accused was held to have successfully established a third alternative defence along the lines indicated in this judgment. Ms Lai SC also stated that in such event, she would not oppose an order for costs in the accused’s favour comprising his costs before the magistrate and his legal aid contributions in the Court of First Instance and in this Court. 75. I would accordingly allow the appeal, set aside the Judge’s decision and restore the defendant’s acquittal. I would also make the order as to costs referred to in the preceding paragraph. 76. My answers to the certified questions are as follows: (i) No. (ii) No. Actual consent is deemed irrelevant. The defence requires proof of an honest and reasonable belief that the girl was aged 16 or over. (iii) No. Mr Justice Tang PJ: 77. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Fok PJ: 78. I agree with the judgment of Mr Justice Ribeiro PJ. Lord Collins of Mapesbury NPJ: 79. I agree with the judgment of Mr Justice Ribeiro PJ. Chief Justice Ma: 80. The Court unanimously allows the appeal, sets aside the Judge’s decision, restores the defendant’s acquittal and awards the appellant costs comprising his costs before the magistrate and his legal aid contributions in the Court of First Instance and in this Court. Mr Gerard McCoy SC and Mr Wong Po Wing, instructed by Eli K.K. Tsui & Co., assigned by the Director of Legal Aid, for the Appellant Ms Anna Y K Lai SC, DDPP, Mr Ned Lai, SADPP and Ms Audrey Parwani, SPP of the Department of Justice, for the Respondent [1] Cap 200. [2] It is clear that indecent assault can be committed by a person and on a person of either gender: R v Hare [1934] 1 KB 354. As the present case concerns a man charged with the offence in respect of a woman or girl, for brevity, I will generally refer to the defendant as a male and the other person as a female. [3] KCCC 388/2015 (12 June 2015). [4] [2002] 1 AC 462. [5] Reasons for Verdict §16. [6] Ibid §18. [7] Ibid. [8] Ibid §19. [9] HCMA 620/2016 (31 July 2017). [10] [2005] 1 HKLRD 443. [11] (2006) 9 HKCFAR 530. [12] Contrary to section 124 of the Crimes Ordinance (Cap 200). [13] Ribeiro, Tang and Fok PJJ, FAMC 28/2017 (1 November 2017). [14] (1872-75) LR 2 CCR 154. [15] At 170. [16] With Brett J dissenting. [17] At 174. [18] R v Poon Ping Kwok [1993] 1 HKCLR 56 at 57, holding that Prince applies to unlawful sexual intercourse with an under-aged girl and to procuring a woman under 21 to have unlawful sexual intercourse; and R v Savage (No 3) [1997] 2 HKC 768 at 772, applying Prince in a case involving gross indecency with or towards a child under the age of 16. [19] [2005] 1 HKLRD 443 at §25. [20] Including the Privy Council’s decision in Gammon (HK) Ltd v Attorney General of Hong Kong [1985] AC 1. [21] HKSAR v So Wai Lun [2005] 1 HKLRD 443 at §12(2). [22] Sweet v Parsley [1970] AC 132 at 148 and 152; Gammon (HK) Ltd v Attorney General of Hong Kong [1985] AC 1 at 14; B (A minor) v DPP [2000] 2 AC 428 at 460; R v K [2002] 1 AC 462 at 471-472, 477. [23] B (A minor) v DPP [2000] 2 AC 428 per Lord Steyn at 470, citing Lord Hoffmann in R v Secretary of State for the Home Department, Ex parte Simms [2000] 2 AC 115 at 131: “Fundamental rights cannot be overridden by general or ambiguous words. This is because there is too great a risk that the full implications of their unqualified meaning may have passed unnoticed in the democratic process. In the absence of express language or necessary implication to the contrary, the courts therefore presume that even the most general words were intended to be subject to the basic rights of the individual.” See also R v K [2002] 1 AC 462 at 477. [24] Sweet v Parsley [1970] AC 132 at 149 and 165. It has however been stressed that the presumption of mens rea is not dependent on there being any ambiguity and operates to supplement the text: B (A minor) v DPP [2000] 2 AC 428 at 470; R v K [2002] 1 AC 462 at 477. [25] (2010) 13 HKCFAR 142 at §45. [26] Discussed in Hin Lin Yee v HKSAR (2010) 13 HKCFAR 142 at §§55-61. [27] Sweet v Parsley [1970] AC 132 per Lord Reid at 150. See also at 157 per Lord Pearce and at 163-164 per Lord Diplock. [28] [1935] AC 462, as discussed in Hin Lin Yee v HKSAR at §§56-59 and 108-111. [29] (2010) 13 HKCFAR 142. [30] In breach of section 54(1) of the Public Health and Municipal Services Ordinance (Cap 132) punishable on summary conviction by a maximum fine of $50,000 and imprisonment for 6 months. [31] (2013) 16 HKCFAR 195. [32] In contravention of s 72 of the Shipping and Port Control Ordinance (Cap 313) punishable on conviction on indictment by a fine of $200,000 and imprisonment for 4 years; and on summary conviction by a fine of $200,000 and imprisonment for 2 years. The Court also noted that conviction was likely to have serious professional ramifications for the mariners involved: Ibid §86. [33] (2013) 16 HKCFAR 195 at §83. [34] Ibid at §90. [35] For example by Lord Nicholls of Birkenhead in B (A minor) v DPP [2000] 2 AC 428 at 462. [36] [1970] AC 132 at 150. [37] R v Court [1989] AC 28 at 33-34. [38] Ibid at 42. [39] Ibid at 43. [40] Set out in Section H.2 below. [41] R v Court [1989] AC 28 at 41-42. [42] Ibid at 44;R v K [2002] 1 AC 462 at 468. [43] R v May [1912] 3 KB 572. [44] R v Court [1989] AC 28 at 42. As Lord Ackner held, it is not necessary to prove that the person assaulted was aware of the circumstances of indecency or apprehended indecency, and an indecent assault can clearly be committed by the touching of someone who is asleep or unconscious. [45] R v Kimber (1983) 77 Cr App R 225 at 229. [46] Ibid. [47] R v Court [1989] AC 28. [48] R v Hodgson [1973] QB 565. [49] A similar view was taken of the offence of buggery with a girl under 21 contrary to section 118D of the Crimes Ordinance in HKSAR v Yee Yiu Sam [2002] 3 HKC 21 at §41 (a sentencing case). [50] R v K [2002] 1 AC 462 at 478-479. His Lordship proceeded to hold that the presumption of mens rea applied and was not displaced, as discussed in Section G below. [51] Sweet v Parsley [1970] AC 132 at 148 and 152; Gammon (HK) Ltd v Attorney General of Hong Kong [1985] AC 1 at 14; B (A minor) v DPP [2000] 2 AC 428 at 460; R v K [2002] 1 AC 462 at 471-472, 477; Hin Lin Yee v HKSAR (2010) 13 HKCFAR 142 at §139. [52] Hin Lin Yee v HKSAR (2010) 13 HKCFAR 142 at §§140-143. In B (A minor) v DPP [2000] 2 AC 428 at 464: Lord Nicholls of Birkenhead explains displacement by necessary implication thus: “‘Necessary implication’ connotes an implication which is compellingly clear. Such an implication may be found in the language used, the nature of the offence, the mischief sought to be prevented and any other circumstances which may assist in determining what intention is properly to be attributed to Parliament when creating the offence.” [53] Hin Lin Yee v HKSAR (2010) 13 HKCFAR 142 at §45. See also Lord Reid in Sweet v Parsley [1970] AC 132 at 150. [54] [2002] 1 AC 462. [55] Hin Lin Yee v HKSAR (2010) 13 HKCFAR 142 at §141. [56] So Wai Lun v HKSAR (2006) 9 HKCFAR 530 at §39; Hin Lin Yee v HKSAR (2010) 13 HKCFAR 142 at §154; Kulemesin v HKSAR (2013) 16 HKCFAR 195 at §93. [57] Kulemesin v HKSAR (2013) 16 HKCFAR 195 at §92. [58] Sweet v Parsley [1970] AC 132 per Lord Reid at 153, and per Lord Diplock at 164. [59] Warner v Metropolitan Police Commissioner [1969] 2 AC 256 at 271-272; Sweet v Parsley [1970] AC 132 at 149 and156; Hin Lin Yee (2010) 13 HKCFAR 142 at §142; Kulemesin v HKSAR (2013) 16 HKCFAR 195 at §93. [60] Lim Chin Aik v The Queen [1963] AC 160 at 174; Sweet v Parsley [1970] AC 132 at 163; Hin Lin Yee v HKSAR (2010) 13 HKCFAR 142 at §158. [61] (2010) 13 HKCFAR 142 at §151. [62] (2013) 16 HKCFAR 195 at §93. [63] [2005] 1 HKLRD 443. [64] Section 124(2) lays down a defence as follows: “Where a marriage is invalid under section 27(2) of the Marriage Ordinance (Cap 181) by reason of the wife being under the age of 16, the invalidity shall not make the husband guilty of an offence under this section because he has sexual intercourse with her, if he believes her to be his wife and has reasonable cause for the belief.” [65] Ord No 11 of 1890. [66] Re-enacted in the Protection of Women and Girls Ordinance 1897 (Ord No 9 of 1897), section 5. [67] Ord No 4 of 1897. [68] HKSAR v So Wai Lun [2005] 1 HKLRD 443 at §31(5). [69] Ibid at §32(2). [70] Ibid at §37. [71] Ord No 4 of 1865. Section 46: “Whosoever shall be convicted of any indecent assault upon any female, or of any attempt to have carnal knowledge of any girl under 12 years of age, shall be liable, at the discretion of the Court, to be imprisoned for any term not exceeding two years, whether without hard labour.” [72] Ord No 4 of 1897. Section 7(1): ““Every person who commits an indecent assault upon any female shall be guilty of a misdemeanor.” 7(2): “It shall be no defence to a charge or indictment for an indecent assault upon a girl under the age of 16 years to prove that she consented to the act of indecency.” This was re-enacted unchanged in section 7 of the Protection of Women and Juveniles Ordinance (Ord 1/ 1951). [73] Ord No 1 of 1978. [74] Ord No 90 of 1991. [75] Hin Lin Yee v HKSAR (2010) 13 HKCFAR 142 at §151; Kulemesin v HKSAR (2013) 16 HKCFAR 195 at §93. [76] Section 123: “A man who has unlawful sexual intercourse with a girl under the age of 13 shall be guilty of an offence and shall be liable on conviction on indictment to imprisonment for life.” [77] HKSAR v So Wai Lun [2005] 1 HKLRD 443 at §38. [78] In So Wai Lun v HKSAR (2006) 9 HKCFAR 530 at §39; Hin Lin Yee v HKSAR (2010) 13 HKCFAR 142 at §102-103; Kulemesin v HKSAR (2013) 16 HKCFAR 195 at §93. [79] Bearing in mind that it applied to carnal knowledge of unmarried girls aged between 12 and 16; contained an exception where customary marriages between “Asiatics” were involved; had a three-month limitation period; and made offences punishable (under section 12) by a maximum of two years imprisonment (one year if tried by a single magistrate). Section 124 is set out in Section C above. [80] [2013] NILR 265. [81] (1872-75) LR 2 CCR 154. Mentioned in Section E above. [82] R v Brown [2013] NILR 265 at §32. [83] Section 4 concerning girls under 14, an offence punishable by life imprisonment, and section 5, a misdemeanour where the girls concerned were under the age of 17. [84] Ibid at §11. [85] Judgment §38. [86] B (A minor) v DPP [2000] 2 AC 428 at 464; HKSAR v So Wai Lun [2005] 1 HKLRD 443 at §16(3). [87] Judgment §38. [88] [2002] 1 AC 462. [89] Section 14 of the 1956 Act provides: “"(1) It is an offence, subject to the exception mentioned in subsection (3) of this section, for a person to make an indecent assault on a woman. (2) A girl under the age of 16 cannot in law give any consent which would prevent an act being an assault for the purposes of this section. (3) Where a marriage is invalid under section two of the Marriage Act 1949, or section one of the Age of Marriage Act 1929 (the wife being a girl under the age of 16), the invalidity does not make the husband guilty of any offence under this section by reason of her incapacity to consent while under that age, if he believes her to be his wife and has reasonable cause for the belief. (4) A woman who is a defective cannot in law give any consent which would prevent an act being an assault for the purposes of this section, but a person is only to be treated as guilty of an indecent assault on a defective by reason of that incapacity to consent, if that person knew or had reason to suspect her to be a defective." [90] [2002] 1 AC 462 at §3. [91] Ibid at §4. [92] Ibid at §21. [93] (2006) 9 HKCFAR 574 at §§26 and 27. [94] Citing R v DPP, ex p Kebilene & Others [2000] 2 AC 326 at pp 378H-379A, per Lord Hope of Craighead. [95] Citing R v DPP, ex p Kebilene [2000] 2 AC 326 at p 378H, per Lord Hope of Craighead. [96] Citing Emmerson and Ashworth, Human Rights and Criminal Justice (2001) para.9‑03. [97] So Wai Lun v HKSAR (2006) 9 HKCFAR 530 at §39; Hin Lin Yee v HKSAR (2010) 13 HKCFAR 142 at §154; Kulemesin v HKSAR (2013) 16 HKCFAR 195 at §93. [98] [2013] NILR 265 at §§19, 38-39. [99] Reflecting the Australian approach: CTM v The Queen (2008) 236 CLR 440 at §§8 and 27; as opposed to the English position requiring only that the mistaken belief be honest, without necessarily being reasonable: R v K [2002] 1 AC 462. [100] Basic Law Art 87; Bill of Rights Art 11(1). [101] HKSAR v Lam Kwong Wai (2006) 9 HKCFAR 574; HKSAR v Ng Po On (2008) 11 HKCFAR 91; Lee To Nei v HKSAR (2012) 15 HKCFAR 162. [102] Hysan Development Co Ltd v Town Planning Board (2016) 19 HKCFAR 372. [103] Reasons for Verdict §19 (in translation). [104] Appearing for the respondent with Mr Ned Lai and Ms Audrey Parwani. 
Chief Justice Ma: 1. This is the judgment of the Court to which each member has contributed. 2. In this appeal, it falls to the Court to consider the constitutionality of legislation which prevents persons who are not of Chinese nationality and who enter Hong Kong for employment as foreign domestic helpers from acquiring Hong Kong permanent resident status. A. The central provisions and the decisions below 3. Article 24(2)(4) of the Basic Law provides for the acquisition of permanent resident status by persons not of Chinese nationality as follows: Article 24(2)(4) “The permanent residents of the Hong Kong Special Administrative Region shall be ... Persons not of Chinese nationality who have entered Hong Kong with valid travel documents, have ordinarily resided in Hong Kong for a continuous period of not less than seven years and have taken Hong Kong as their place of permanent residence before or after the establishment of the Hong Kong Special Administrative Region.”[1] 4. Section 2(4)(a) of the Immigration Ordinance (“the Ordinance”),[2] stipulates that certain categories of persons are not to be treated as “ordinarily resident” in Hong Kong. They are therefore not eligible to acquire Hong Kong permanent resident status and ineligible for the right of abode: Section 2(4) For the purposes of this Ordinance, a person shall not be treated as ordinarily resident in Hong Kong- (a) during any period in which he remains in Hong Kong- (i) with or without the authority of the Director, after landing unlawfully; or (ii) in contravention of any condition of stay; or (iii) as a refugee under section 13A; or (iv) while detained in Hong Kong under section 13D; or (v) while employed as a contract worker, who is from outside Hong Kong, under a Government importation of labour scheme; or (vi) while employed as a domestic helper who is from outside Hong Kong; or (vii) as a member of a consular post within the meaning of the Consular Relations Ordinance; or (viii) as a member of the Hong Kong Garrison; or (ix) as a holder of a prescribed Central People's Government travel document; or (b) during any period, whether before or after the commencement of this Ordinance, of imprisonment or detention pursuant to the sentence or order of any court. 5. Thus, section 2(4)(a)(vi) disqualifies a person from outside who enters Hong Kong for employment as a foreign domestic helper (“FDH”) from building up, while so employed, the continuous seven year period of ordinary residence necessary for acquiring permanent resident status under Article 24(2)(4). This is the case however long such person’s actual period of residence in Hong Kong while so employed may be. It is the constitutionality of section 2(4)(a)(vi) that the appellants seek to challenge. The appellants contend that the restriction it imposes is inconsistent with Article 24(2)(4) and unconstitutional. 6. The appellants’ argument on unconstitutionality succeeded before Mr Justice Lam (as Lam JA then was),[3] although Mr Domingo’s case was remitted to the Registration of Persons Tribunal for an issue concerning an alleged estoppel to be dealt with. However, the Court of Appeal[4] upheld the constitutional validity of section 2(4)(a)(vi) and reversed Lam J’s judgment. B. The FDH scheme 7. Since the mid-1970s, with greater affluence and increasing numbers of households where both spouses go out to work, there has been a growing demand by Hong Kong families for domestic helpers. The local supply of such labour was scarce, so FDHs have been recruited from many countries including the Philippines, Indonesia, Nepal, India, Pakistan, Thailand and Sri Lanka to meet the demand. The number of such FDHs in Hong Kong grew from 881 in 1974, to 28,951 in 1986, to 70,335 in 1990, and stood at 285,681 on 31 December 2010. It is therefore plain that FDHs represent a highly valuable and indeed essential workforce. Many FDHs have had multiple renewals of their contracts with the same employer and have forged close personal ties with the families in which they work. 8. The terms on which FDHs are admitted to work and reside in Hong Kong are and have always been highly restrictive and subject to control by the Director of Immigration. The process of hiring a FDH begins with an application by the prospective employer to the Director for a visa for the prospective FDH. Before granting a visa, the Director satisfies himself of the applicant’s financial capacity and ability to provide suitable accommodation. He also satisfies himself that neither employer nor FDH has previously proved unsuitable by breaking relevant laws, and so forth. 9. The employer and the FDH are required to enter into a standard-form contract which cannot be varied without the prior consent of the Director except in respect of certain matters of detail and except insofar as wages higher than the prescribed minimum may be agreed. The visa issued to the FDH specifies that he or she is admitted for employment as a domestic helper with the named employer under the specified contract. 10. A central feature of the arrangement is that permission to remain in Hong Kong is tied to the duration of each contract. On completion of the standard two-year term, the FDH is required to return to his or her own country of origin. If the contract is terminated prematurely, the FDH is not allowed to change employers, save in exceptional circumstances. A FDH without a current contract is required to leave Hong Kong within his or her approved limit of stay or two weeks after termination of the contract, whichever is the earlier. 11. A FDH wishing to continue in employment must sign a fresh contract and apply for a visa to return to Hong Kong on the strength of that contract. However, he or she is first required to return to the country of origin for a prescribed period of vacation before coming back to Hong Kong. 12. A second important feature of the arrangement is that the FDH is subject to highly restrictive employment conditions. The FDH can only engage in domestic work and has to work and reside in the employer’s residence designated in the contract. He or she cannot work elsewhere or for another employer. The employer must provide the FDH with accommodation and food during the term of the employment and pay for repatriation upon termination of the contract. In the case of death, the FDH’s remains and personal property have to be returned to the place of origin at the employer’s expense. 13. It is made clear in various documents, including the Explanatory Notes attached to the standard-form contract, that FDHs are not admitted to Hong Kong for settlement but permitted to enter and remain for specific employment and for a limited period. FDHs are not allowed to bring their dependents to Hong Kong for residence here. 14. The evidence is that these restrictive arrangements reflect the Government’s policy of allowing FDHs to enter and reside in Hong Kong for the particular purpose of meeting the shortage of domestic helpers in the labour market. That is the basis of their permission to enter and why they are prohibited from taking any other employment while here. They are not admitted for settlement in Hong Kong and are expected to maintain genuine links in their home country which is why, at the end of their contract, they must return home before taking up any fresh contract and why they not allowed to bring their dependents to Hong Kong for residence. 15. In contrast to the restrictive arrangements mentioned above, certain other persons not of Chinese nationality may be granted general employment visas on entering Hong Kong. The differences in the conditions of stay applicable to the two categories of employment were highlighted in the evidence as follows: 16. A person permitted to enter on a general employment visa (and who falls outside the other restrictive categories listed in section 2(4)(a)) may be treated as ordinarily resident and may therefore be eligible for eventual qualification for permanent resident status under Article 24(2)(4). C. The parties 17. Ms Vallejos is a Philippine national and was born in the Philippines in 1952. She married another Philippine national in 1974 and they have five children, the youngest having been born in 1984. Her husband and four of her children remain in the Philippines while her eldest son lives in Ireland. She first came to Hong Kong to work as a domestic helper in August 1986 and has worked for the same employer since 2 February 1987 under a series of contracts. She returned to the Philippines when each contract expired and, on each occasion, her permission to return to Hong Kong as a FDH was renewed. 18. Mr Domingo is also a Philippine national born in the Philippines. He came to Hong Kong in 1985 to work as a domestic helper. Between 1987 and 2007, he worked for the same employer. In 1988, he married another domestic helper, Mrs Domingo, with whom he has had three children, all born in Hong Kong. On 14 November 2007, Mr Domingo was granted permission to remain in Hong Kong without any restriction on his employment, such date stated to be the starting date for calculating the length of his ordinary residence. He was, in other words, taken out of the category of FDH for section 2(4)(a)(vi) purposes and treated as ordinarily resident, eligible for eventual permanent resident status. Mr Domingo, however, claims that his entitlement to such status had already accrued and that he does not need to build up the qualifying period as from 14 November 2007, contending that section 2(4)(a)(vi) is inconsistent with Article 24(2)(4) and unconstitutional. The pending issue on estoppel relates to the effect of his accepting the aforesaid change of status stated to be effective only as from the 2007 date. 19. The Commissioner of Registration and the Registration of Persons Tribunal are named as the respective respondents in each case since the Commissioner refused each of the appellants verification of eligibility for a permanent identity card and since their appeals to the Tribunal were rejected. D. The parties’ respective cases 20. The parties differ as to the meaning of “ordinarily resided” in Article 24(2)(4) and as to the method of arriving at such meaning. The appellants’ argument is simple. Mr Michael Fordham QC, appearing on their behalf,[5] submits that the words “ordinarily resided in Hong Kong” found in Article 24(2)(4) have a well-established “natural and ordinary meaning” which is decisive. The words “ordinarily resided” in Article 24(2)(4) carry that meaning and apply to FDHs residing in Hong Kong so that they must be treated as ordinarily resident within the meaning of that Article. It follows that section 2(4)(a)(vi) is unconstitutional since it purports to exclude FDHs as a class from ever being treated as “ordinarily resident”. 21. Basing himself on Lord Scarman’s formulation in R v Barnet London Borough Council ex parte Shah,[6] (to which we shall return), Mr Fordham submits that a person is “ordinarily resident” in Hong Kong according to the “natural and ordinary meaning” of those words if, quite simply, he or she is in Hong Kong, “living lawfully, voluntarily and for a settled purpose, as part of the regular order of life for the time being”. FDHs, so the argument runs, come within this well-established interpretation just as much as do other persons who come to live in Hong Kong for business, education or employment purposes and who have, unlike FDHs, been treated as ordinarily resident here. The purported exclusion of FDHs by section 2(4)(a)(vi) is thus inconsistent with Article 24(2)(4) and unconstitutional. 22. Lord Pannick QC, appearing for the Commissioner,[7] advances two principal arguments in response. The first concerns the meaning of “ordinarily resided” in Article 24(2)(4). While he accepts Lord Scarman's formulation as a starting-point in construing those words, Lord Pannick does not accept that it is their sole meaning. In factually exceptional cases, he argues, the circumstances of the claimant’s residence in Hong Kong may be so out of the ordinary as to fall outside the concept of “ordinary residence” as generally understood and outside the appellants’ “natural and ordinary meaning”. He contends that the highly restrictive conditions applicable to FDHs place them in such an exceptional category so that excluding them from “ordinary residence” is consonant with Article 24(2)(4). 23. Lord Pannick’s second argument is that the Court should recognize that the Basic Law accords a margin of discretion to the legislature to define excepted or excluded classes of persons around the edges of the concept of “ordinary residence”. This, he says, is what the legislature has done in section 2(4)(a), continuing the role that it has traditionally played in helping to define, by identifying excluded classes, the concept of “ordinary residence” for the purpose of determining who is entitled to be treated as a Hong Kong permanent resident. He submits that the Basic Law should be taken to have adopted this approach as part of its promotion of continuity in the legal system after 1st July 1997. He argues that so long as the legislature respects and preserves a core meaning covered by Lord Scarman’s formulation and confines itself to exclusions around the margins of the “ordinary residence” concept, the Court should regard such enactments as consistent with Article 24(2)(4). 24. If and insofar as the Court may be in doubt as to the meaning of the concept as used in Article 24(2)(4), Lord Pannick submits that the Court should refer to extrinsic materials, including materials which came into existence after April 1990 when the Basic Law was promulgated, in aid of its interpretation of the Article. 25. Finally, if and insofar as the Court considers it necessary to consider the effect of the Interpretation adopted by the Standing Committee of the National People’s Congress on 26 June 1999,[8] Lord Pannick submits that the Court would be bound under Article 158(3) of the Basic Law to refer to the Standing Committee two proposed questions concerning the meaning and scope of an “interpretation” made under Article 158(1).[9] 26. We shall deal in the first place with the meaning of “ordinarily resided” in Article 24(2)(4) with a view to determining the compatibility or otherwise of section 2(4)(a)(vi) with that Article. E. The meaning of “ordinarily resided” in Article 24(2)(4) E.1 The importance of purpose and context 27. In much of the case-law in which the concept of “ordinary residence” is discussed, Lord Scarman's formulation has been adopted on the basis that it lays down “the natural and ordinary meaning” of those words and of cognate expressions such as “ordinarily resident”, “ordinarily resided” and so forth. In many instances, that formulation is likely to supply an appropriate meaning to be attributed to the term. The danger is, however, that describing it as “the natural and ordinary meaning” may be thought to suggest that any other meaning is “unnatural” or “extraordinary” and therefore inappropriate. But it would be an obvious error to think that “ordinary residence” can only have a single meaning whatever the factual and legal context. 28. This was recognized by Lord Scarman himself in Ex parte Shah in the very sentence containing his formulation: “Unless, therefore, it can be shown that the statutory framework or the legal context in which the words are used requires a different meaning, I unhesitatingly subscribe to the view that ‘ordinarily resident’ refers to a man's abode in a particular place or country which he has adopted voluntarily and for settled purposes as part of the regular order of his life for the time being, whether of short or of long duration.”[10] (Italics supplied) 29. And in Fateh Muhammad v Commissioner of Registration,[11] Mr Justice Bokhary PJ, writing for the Court, stated: “No single judicial pronouncement or combination of such pronouncements in regard to the meaning of the expression ‘ordinarily resident’ can be conclusive for the purposes of every context in which that expression appears.” E.2 The meaning of “ordinary residence” in various statutes 30. It should be noted that except for two cases decided by this Court,[12] all the authorities cited to us were cases involving the construction of “ordinary residence” or cognate expressions as they appear in various statutes and not in constitutional instruments. Moreover, many of those statutes did not concern permanent resident status or citizenship claims. Thus, the Court was presented with a survey of cases where statutory provisions concerning “ordinary residence” were relevant to tax liability,[13] eligibility for education grants,[14] eligibility to stand for a District Board election,[15] the court’s bankruptcy jurisdiction[16] and the court’s matrimonial jurisdiction.[17] The citation of such authority must obviously be approached with caution since, as one might expect, the approach to the meaning of the term “ordinary residence” in each case is conditioned by the content and legislative purpose of the statute in question, applied in differing factual contexts. 31. Two obvious illustrations may be found in the concept’s application for the purposes of determining the court’s bankruptcy and matrimonial jurisdictions. In Re Ip Pui Man Nina,[18] the respondent, a Hong Kong permanent resident who claimed to spend two-thirds of her time outside of Hong Kong, sought to challenge the court’s bankruptcy jurisdiction, relying on section 4(1) of the Bankruptcy Ordinance which excludes presentation of a petition unless (among other alternatives) the debtor “at any time in the period of 3 years ending with [the day the petition is presented] has been ordinarily resident, or has had a place of residence, in Hong Kong”. Such a requirement – of ordinary residence of unspecified duration at any time during the preceding three years – obviously has a far lower threshold than Article 24(2)(4)’s requirement of a continuous seven year period of ordinary residence and reflects a statutory purpose wholly different from the purposes underlying the questions facing this Court. Ms Ip’s challenge failed since the Court found that she had, during the earlier part of the three year period, been ordinarily resident in Hong Kong. 32. The court’s matrimonial jurisdiction is founded inter alia on habitual residence in Hong Kong throughout the period of three years[19] immediately preceding the date of the petition or application.[20] In this context, “habitual residence” is treated as interchangeable with “ordinarily resident”.[21] As Baroness Hale explained in Mark v Mark, the statutory purpose is to require the parties and their marriage to have a sufficiently close connection with the country to make it desirable that its courts should have jurisdiction to dissolve the marriage and thus to dissuade mere forum shoppers. The court approaches the concept of “habitual residence” in this connection considering the welfare and financial needs of the spouses and their children, and so forth[22] – self-evidently considerations different from those which arise when examining whether a person is ordinarily resident for the purposes of Article 24(2)(4). 33. Furthermore, even where the statute employs “ordinary residence” in a provision relevant to immigration control and permanent resident status, the precise context in which the issue as to the meaning of the concept arises may vary, so that any attempt to extrapolate from such decisions to the meaning of “ordinarily resided” in Article 24(2)(4) must be treated with caution. 34. This is illustrated by Cheung Cheong v Attorney General[23] and Director of Immigration v Ng Shun Loi.[24] In both of those cases, persons who had been residing lawfully in Hong Kong had their stay here interrupted when they were detained for long periods (five and ten years respectively) on the Mainland by the Mainland authorities. When they eventually returned to Hong Kong (after absences of about 16 and 23 years respectively) they sought to argue that they had acquired the status of “Chinese resident” under Hong Kong law and were therefore exempt from removal orders. Ex parte Shah was relied on, as it were at one remove, the argument being that since their stay on the Mainland was involuntary, they had not become ordinarily resident there and that their ordinary residence in Hong Kong ought therefore to be regarded as having continued throughout the periods of their absence. Their arguments were not accepted although a similar argument, relying on Ex parte Shah in this roundabout fashion, succeeded in Lau San Ching v Apollonia Liu[25]for the purpose of establishing the applicant’s eligibility to stand for election in District Board Elections, such eligibility requiring ten years ordinary residence immediately preceding nomination. 35. It is self-evident that although these are cases where Ex parte Shah was relied on in relation to establishing ordinary residence in Hong Kong, their factual context and the attempt to apply Ex parte Shah at one remove have little to do with the concerns of the present appeal. E.3 The tax cases on which Ex parte Shah was founded 36. An analysis of Ex parte Shah may usefully begin by examining the two 1928 tax cases in the House of Lords on which it was founded, namely Levene v Commissioners of Inland Revenue[26] and Inland Revenue Commissioners v Lysaght[27]. Revisiting those decisions, it has to be said that it is somewhat surprising that Lord Scarman treated them as establishing “the natural and ordinary meaning” of “ordinary residence” for general purposes since they were actually decisions addressing quite particular statutory provisions applied to peculiar factual situations. 37. In both cases, the legal issue was whether the persons concerned were entitled to exemption from income tax on war loan interest under section 46 of the Income Tax Act 1918,[28] as persons not ordinarily resident in the United Kingdom. The House of Lords were not engaged in considering the meaning of “ordinarily resident” in general, but only for the purposes of the Income Tax Acts, and their Lordships found guidance in a particular rule bearing on that expression. Thus, Viscount Cave LC pointed out in Levene[29] that in the case of British subjects (which the persons concerned were): “... regard must be had to r 3 of the General Rules applicable to all the schedules of the Income Tax Act, which provides that every British subject whose ordinary residence has been in the United Kingdom shall be assessed and charged to tax notwithstanding that at the time the assessment or charge is made he may have left the United Kingdom, if he has so left the United Kingdom for the purpose only of occasional residence abroad.” 38. In each of the two cases, the British subjects concerned had initially been ordinarily resident in the United Kingdom; had then left the country; but had thereafter returned to England on a regular basis. In the case of Mr Levene, after leaving the country, he lived in various hotels in Europe but returned to England in the summer, again living in hotels, for about five months each year to obtain medical advice, to visit relatives, to make arrangements for the care of his mentally afflicted brother, to take part in Jewish religious observances and to deal with his income tax affairs. Mr Lysaght had retired as an active director of his company and gone to live in the Irish Free State where he owned a large estate, thereafter returning to attend monthly board meetings, living in hotels and spending a total of between 48 and 101 days in England in any one year. 39. Holding that findings of fact by the Commissioners could not be interfered with unless they were unsupported by evidence or otherwise vitiated by an error of law, their Lordships concluded that the findings of ordinary residence on the abovementioned facts could not be disturbed. Thus, in Levene, Viscount Sumner stated (reflecting the language of rule 3): “...there was ample evidence before the commissioners to show that a man who left England to live abroad as he had been living here, and when warm weather came, returned to his native country and to his permanent associations, and in 1919 ‘left the United Kingdom for the purpose of occasional residence only.’” 40. It was in this context that their Lordships approached “ordinarily resident” as encompassing “the way in which a man’s life is usually ordered”;[30] or as connoting “residence in a place with some degree of continuity and apart from accidental or temporary absences”.[31] It was the ordered regularity of their returning to England that justified a finding of ordinary residence. Thus, Viscount Sumner described Mr Levene as “a bird of passage of almost mechanical regularity”, adding: “...Mr Levene continued to go to and fro during the years in question, leaving at the beginning of winter and coming back in summer, his home, thus remaining as before. He changed his sky but not his home. On this I see no error in law in saying of each year that his purpose in leaving the United Kingdom was occasional residence abroad only. The occasion was the approach of an English winter, and when with the promise of summer here that occasion passed away, back came Mr Levene to attend to the calls of interest, of friendship, and of piety.”[32] 41. Similarly in Lysaght, it was the regularity with which Mr Lysaght returned to England that his Lordship was referring to when Viscount Sumner stated: “I think the converse to ‘ordinarily’ is ‘extraordinarily’ and that part of the regular order of a man's life, adopted voluntarily and for settled purposes, is not ‘extraordinary.’ Having regard to the times and duration, the objects and the obligations of Mr Lysaght's visits to England, there was in my opinion evidence to support, and no rule of law to prevent, a finding that he was ordinarily resident, if he was resident in the United Kingdom at all. No authority was cited, which requires special consideration on this head.”[33] 42. We were also referred to the Canadian Supreme Court’s decision in Thomson v Minister of National Revenue,[34] which proceeded on strikingly similar lines. The legal issue was whether Mr Thomson was liable to tax under the Income War Tax Act, section 9(1) of which provided: “There shall be assessed, levied and paid upon the income during the preceding year of every person (a) residing or ordinarily resident in Canada during such year; or (b) who sojourns in Canada for a period or periods amounting to one hundred and eighty-three days during such year ...” 43. Mr Thomson was born in New Brunswick and had lived there until 1923, when he left and declared himself domiciled in Bermuda and built a house in North Carolina. He returned to New Brunswick for three consecutive summer seasons and then built a house there so his wife could be near her relatives in that Province. Since then he had spent an average of 150 days each year in Canada, leading him to argue that he was exempt from the tax because he was someone who was not residing or ordinarily resident in Canada, but had merely “sojourned” there for less than 183 days a year. 44. The Court was therefore not approaching the concept of “ordinarily resident” with a blank slate, but seeking to decide whether to characterise Mr Thomson as ordinarily resident as opposed to a mere sojourner. Not surprisingly, they held that he was no sojourner, Kerwin J stating:[35] “The family ties of his wife, if not of himself, the erection of a substantial house, the retention of the servants, together with all the surrounding circumstances, make it clear to me that his occupancy of the house and his activities in Canada comprised more than a mere temporary stay therein.” 45. Rand J, having cited the English tax cases mentioned above, stressed the importance of context in approaching the concepts of residence and ordinary residence, stating: “The gradation of degrees of time, object, intention, continuity and other relevant circumstances, shows, I think, that in common parlance ‘residing’ is not a term of invariable elements, all of which must be satisfied in each instance. It is quite impossible to give it a precise and inclusive definition. It is highly flexible, and its many shades of meaning vary not only in the contexts of different matters, but also in different aspects of the same matter. In one case it is satisfied by certain ele­ments, in another by others, some common, some new. The expression ‘ordinarily resident’ carries a restricted signification, and although the first impression seems to be that of preponderance in time, the decisions on the English Act reject that view. It is held to mean residence in the course of the customary mode of life of the person concerned, and it is contrasted with special or occasional or casual residence. The general mode of life is, therefore, relevant to a question of its application.” 46. Their Honours were therefore, like the House of Lords, concerned with a set of facts involving departure from the country followed by regular return visits. The regularity of Mr Thomson’s return, together with his having set up a house for his wife in New Brunswick provided the basis for rejecting the “sojourning” submission. This was the context for Estey J holding that: “... one is ‘ordinarily resident’ in the place where in the settled routine of his life he regularly, normally or customarily lives. One ‘sojourns’ at a place where he unusually, casually or intermittently visits or stays. In the former the element of permanence; in the latter that of the temporary predominates. The difference cannot be stated in precise and definite terms, but each case must be determined after all of the relevant factors are taken into consideration, but the foregoing indicates in a general way the essential difference.” [36] 47. The question of ordinary residence therefore arose and was resolved in these tax cases in rather special circumstances. But Lord Scarman nevertheless drew heavily on those 1928 decisions, generalising from their specifics to the broad proposition that living in a particular country “for a settled purpose, as part of the regular order of life for the time being” was an essential feature of “ordinary residence”. 48. Leaving aside the detailed reasoning in those decisions, it is clear that a court’s orientation in construing a tax statute may be influenced by purposive considerations which are inapplicable in other contexts. Thus, the House of Lords was likely to have been anxious to avoid construing “ordinary residence” in a manner facilitating tax avoidance and for that reason gave the concept a wide meaning. That appears to be the view Rand J took of the English decisions. When commenting on Lysaght, his Honour stated: “...it must, I think, be said that the language of ‘plain men’ was stretched to the breaking point to encompass the facts that had been found by the Commissioners to be residence.”[37] Indeed, Viscount Cave LC had dissented in Lysaght, stressing the fact that Mr Lysaght had a permanent home in the Free Irish State and holding that the evidence did not support the Commissioners’ finding of ordinary residence. E.4 What Ex parte Shah decided 49. Whatever misgivings one may have as to the basis supplied by the tax cases for Lord Scarman’s formulation in Ex parte Shah, that formulation is frequently resorted to when the meaning of “ordinary residence” in a statutory provision is sought: a person is thus often treated as “ordinarily resident” if he is living here “lawfully, voluntarily and for a settled purpose, as part of the regular order of life for the time being”. As previously stated, in many cases that may well be an appropriate reflection of the statutory intent. 50. However, having recognized that the statutory (or constitutional) purpose and the factual context have an important bearing on the meaning of “ordinary residence” in any particular case, the essential question is whether Lord Scarman's formulation is decisive, as the appellants contend; or whether, as the Commissioner argues, it is but a starting-point. How complete is the formulation’s coverage? Are there purposes and contexts which are not adequately addressed by the formulation so that a given set of facts may satisfy all its elements but may nonetheless justify the conclusion that the person or class of persons involved are not “ordinarily resident”? If so, the question which is vital to the outcome of this appeal is whether the circumstances of FDHs as a class present one such factual context when the words “ordinarily resided” in Article 24(2)(4) come to be interpreted. 51. To begin to answer these questions, it is necessary first to examine what precisely Ex parte Shah decided. This is clearly revealed if one compares the approach of the Court of Appeal[38] with that of Lord Scarman for the House of Lords. 52. Ex parte Shah was a test case on whether foreign students studying in the United Kingdom were entitled to local authority education grants. The students in question claimed an entitlement under section 1(1) of the Education Act 1962 which imposed a duty on every local education authority: “... subject to and in accordance with regulations made under this Act, to bestow awards on persons who – (a) are ordinarily resident in the area of the authority, and (b) possess the requisite educational qualifications, in respect of their attendance at courses to which this section applies.” 53. A regulation made under the Act[39] provided that there was no duty to bestow an award: “... upon a person who has not been ordinarily resident, throughout the three years preceding the first year of the course in question, in the United Kingdom ...” 54. In the Court of Appeal, Lord Denning MR accepted that if what was traditionally regarded as the natural and ordinary meaning of the words “ordinarily resident” were to be applied, a person would be held to come within that concept if he or she was “habitually and normally resident here, apart from temporary or occasional absences of long or short duration”. He noted that on that test, all the relevant students would qualify for a mandatory award. However, his Lordship regarded the Immigration Act 1971 as having made a crucial difference, rendering the traditional interpretation of “ordinarily resident” inapplicable.[40] Lord Denning MR stated: “We must say what is the meaning of the words ‘ordinarily resident’ in the context of the situation brought about by the Immigration Act 1971. On this approach, it is my opinion that, whenever a boy comes from overseas on a student's visa, which is renewed every year, he is not to be regarded as ‘ordinarily resident’ here. He is allowed to enter on the terms that he or his parents or friends will pay all his fees and expenses whilst he is here, and that he will leave this country when his leave comes to an end. Such a boy is not ‘ordinarily resident’ here. No matter whether he goes home for holidays or not. No matter whether his parents are dead and he has no home to go to overseas. Suffice it that he has to leave at the end of his time, unless renewed.”[41] 55. In other words, Lord Denning MR considered the immigration status of the students created by the 1971 Act critical to deciding whether they were ordinarily resident and, since that status was not addressed if one simply applied the traditional “natural and ordinary meaning” of those words, his Lordship abandoned “natural and ordinary meaning” as an inadequate test of ordinary residence in the context and gave decisive weight to the claimants’ immigration status. 56. Eveleigh LJ adopted a similar approach. He stated that where a person seeks to claim a benefit from the state: “...it will be necessary to show that he has the right or permission to live here in the manner which has led to the conclusion that he is living as any ordinary member of the community may live. On this basis I approach the cases before the court, and with the immigration legislation in mind.”[42] 57. Referring to one of the students in question, his Lordship concluded: “Hamid Akbarali came here for education. He had no right or permission to live here for all general purposes. Whatever his desire, his permission only extended to living here for the purpose of education. Such a limited purpose could not entitle him to be classed as being ‘ordinarily resident.’”[43] 58. Templeman LJ also gave decisive weight to the immigration status of the relevant students: “An overseas student has no right to remain in the United Kingdom after he has ceased to be a student, and is not ordinarily resident in the United Kingdom because he is resident in this country for educational purposes only, and not for the purpose of making a home indistinguishable from the home of those who are residing in this country for all purposes.”[44] 59. The vital difference in Lord Scarman’s approach involved his Lordship’s decision that (except where presence in the country was unlawful) the immigration status of the students under the 1971 Immigration Act was irrelevant for the purposes of the Education Act. He restricted the test to whether, in the three years immediately preceding the start of the educational course in question, each student’s residence in the United Kingdom came within his Lordship’s formulation of the natural and ordinary meaning of “ordinarily resident” which we have set out above. 60. Lord Scarman framed the issues thus: “Two questions of statutory interpretation, therefore, arise. The first is: what is the natural and ordinary meaning of ‘ordinary residence in the United Kingdom’...? The second is: does the statute in the context of the relevant law against the background of which it was enacted, or in the circumstances of today, including in particular the impact of the Act of 1971, compel one to substitute a special, and, if so, what, meaning to the words ‘ordinarily resident in the United Kingdom’?”[45] 61. His short answer was as follows: “The natural and ordinary meaning of the words has been authoritatively determined in this House in two tax cases reported in 1928. To the second question my answer is ‘No.’ The Act of 1962 and the Regulations are to be construed by giving to the words ‘ordinarily resident in the United Kingdom’ their natural and ordinary meaning.”[46] 62. Lord Scarman answered “No” to the second question because his Lordship and the other members of the House of Lords did not agree with the courts below as to the policy of the Education Act. The lower courts had concluded that enactment of the 1971 Immigration Act had changed the context in which the condition of ordinary residence in the Education Act 1962 was to be applied. The House of Lords disagreed. Lord Scarman traced the legislative policy back to section 81 of the Education Act 1944 which required local education authorities: “...for the purpose of enabling pupils to take advantage without hardship to themselves or their parents of any educational facilities available to them ... (c) to grant scholarships ... and other allowances in respect of pupils over compulsory school age, including pupils undergoing training as teachers.” His Lordship stated: “The section set no limit, imposed no restrictions by way of nationality, origin, or sex. It was an integral part of an enlightened statute, the policy of which has governed educational provision ever since.”[47] 63. He pointed to the fact that immigration control measures had been introduced under the Commonwealth Immigrants Act 1962 and that the Education Act had been enacted in the same year, commenting: “It is inconceivable, and I refuse, therefore, to infer, that Parliament, in enacting the education statute, could have been unaware of the restrictions then being imposed for the first time on Commonwealth immigrants.”[48] 64. The natural and ordinary meaning test on Lord Scarman’s formulation was therefore firmly focussed on whether, in the three years preceding commencement of the relevant educational course, the student in question had lived in the United Kingdom “lawfully, voluntarily and for a settled purpose, as part of the regular order of life for the time being”, ignoring the conditions subject to which he had been admitted into the country. Lord Scarman stated: “Both courts ... agreed in attaching decisive importance to what the Divisional Court called ‘the immigration status’ of the students. ‘Immigration status,’ unless it be that of one who has no right to be here, in which event presence in the United Kingdom is unlawful, means no more than the terms of a person's leave to enter as stamped upon his passport. This may or may not be a guide to a person's intention in establishing a residence in this country: it certainly cannot be the decisive test, as in effect the courts below have treated it. Moreover, in the context with which these appeals are concerned, i.e. past residence, intention or expectations for the future are not critical: what matters is the course of living over the past three years.”[49] (Italics supplied) 65. His Lordship stated the applicable test as follows: “... local education authorities, when considering an application for a mandatory award, must ask themselves the question: has the applicant shown that he has habitually and normally resided in the United Kingdom from choice and for a settled purpose throughout the prescribed period, apart from temporary or occasional absences? ... The relevant period is not the future but one which has largely (or wholly) elapsed, namely that between the date of the commencement of his proposed course and the date of his arrival in the United Kingdom. The terms of an immigrant student's leave to enter and remain here may or may not throw light on the question: it will, however, be of little weight when put into the balance against the fact of continued residence over the prescribed period - unless the residence is itself a breach of the terms of his leave, in which event his residence, being unlawful, could not be ordinary.”[50] E.5 The limits of the decision in Ex parte Shah 66. So analysed, it becomes clear that the decision in Ex parte Shah is limited in at least two major respects. First, its scope or coverage is limited. Because of the view taken by the House of Lords of the legislative policy informing the relevant Education Act, their Lordships discounted as irrelevant any restrictive conditions affecting the students’ immigration status to their entitlement to a grant. On that basis, Lord Scarman fashioned as the “natural and ordinary meaning” of “ordinarily resident” a formulation which focuses solely on the students’ “course of living” over the three years immediately preceding the relevant educational programme, asking whether such course of living was lawful, voluntary and for a settled purpose, as part of their regular order of life for the time being. That may have made good sense in that context.[51] However, in another case, the Court might well consider it highly relevant to the meaning of the words “ordinary residence”, used in a different context and for different statutory or constitutional purposes, to give full or even decisive weight to immigration status. The present case may well be such an instance. 67. The second limitation of Lord Scarman’s formulation is that it leaves open questions regarding the qualitative aspects of a person’s residence in the country. A person may indeed be able to show that he or she had lived lawfully, voluntarily and for a settled purpose, as part of the regular order of life for the time being in the country over a given period of time, but questions are bound to arise in some cases as to whether qualitative aspects of all or part of that period of residence are of such a character as to render it other than “ordinary residence”. The qualitative aspects of FDHs’ residence in Hong Kong are obviously of potential relevance. 68. In the context of Article 24(2)(4), the Court has already had to deal with qualitative aspects of “ordinary residence” to which the “natural and ordinary meaning” approach provides no answer. Thus, after entering Hong Kong lawfully, voluntarily and for a settled purpose, and living here for a substantial period as part of the regular order of his life, a person may be convicted of a crime and sentenced to a period of imprisonment before the seven-year period required by Article 24(2)(4) has been achieved. Section 2(4)(b) of the Ordinance provides that “a person shall not be treated as ordinarily resident in Hong Kong … during any period … of imprisonment or detention pursuant to the sentence or order of any court”. The question arose as to whether that exclusion is consonant with the meaning of “ordinarily resident” in Article 24(2)(4) properly interpreted. 69. In Fateh Muhammed [52]the Court answered that question in the affirmative. Bokhary PJ, writing for the Court, regarded “serving a term of imprisonment, at least when it is not of trivial duration” as “something out of the ordinary”. He took as the object of the Court’s interpretation exercise, the need to determine whether the applicant had satisfied “a qualification prescribed by the Basic Law for attaining a valuable status and right, namely Hong Kong permanent resident status and the right of abode here” and held that “in such a context, there is a very strong case for saying that residence while serving a substantial term of imprisonment or detention in a training or detention centre is not ordinary residence”. The Court so held. As Ribeiro PJ pointed out in Prem Singh, it is the “qualitative aspect of time spent in prison that has led to such periods being excluded from the concept of ‘ordinary residence’ in successive statutory schemes and in the Basic Law.”[53] 70. The Ordinance also supplements Article 24(2)(4) by introducing what has been called the “immediacy condition”. Schedule 1, paragraph 1(4)(b) provides: “For the purposes of calculating the continuous period of 7 years in which a person has ordinarily resided in Hong Kong, the period is reckoned to include a continuous period of 7 years ... for a person under paragraph 2(d), before or after the establishment of the Hong Kong Special Administrative Region but immediately before the date when the person applies to the Director for the status of a permanent resident of the Hong Kong Special Administrative Region.” 71. Was the enactment of such a condition constitutional? No help on that question can be found in Lord Scarman’s “natural and ordinary meaning”. Indeed, he had no need to consider immediacy as a possible implication since regulation 13 expressly required the three year period of ordinary residence to precede the first year of the educational course in question. 72. Adopting a purposive approach to the interpretation of Article 24(2)(4), the Court in Fateh Muhammed affirmed the constitutionality of the immediacy requirement: “...unless its wording simply cannot support such a reading, a purposive construction of art.24(2)(4) drives the Court to say that the seven continuous years required by art.24(2)(4) must come immediately before the time when an application for Hong Kong permanent resident status is made in reliance on those seven continuous years. In my view, the wording of art.24(2)(4) supports such a reading. Such support is to be found generally in the tenor of the provision and particularly in the implicit link between the twin requirements of seven years' ordinary and continuous residence and of having taken Hong Kong as one's place of permanent residence. In my judgment, the seven continuous years required by art.24(2)(4) of the Basic Law must come immediately before the time when an application for Hong Kong permanent resident status is made in reliance on those seven continuous years.”[54] 73. Another supplement to or elaboration of Article 24(2)(4) held by the Court to be valid involves the establishment of a fair and reasonable statutory scheme for the proper verification of a person’s claim to right of abode.[55] 74. But a different legislative initiative was struck down as unconstitutional by the Court in Prem Singh.[56] The Ordinance purported to make it a condition that a person should not be subject to a limit of stay when declaring his intention to make Hong Kong his place of permanent residence in the course of applying for permanent resident status. The Court held that such a condition was incompatible with the language and structure of the permanence requirement as laid down in Article 24(2)(4) and was therefore unconstitutional. The decision does not address the meaning of “ordinary residence”. E.6 A general criticism of Lord Scarman’s approach 75. A more general criticism may, with respect, be made of Lord Scarman’s approach. Although as noted above,[57] his Lordship qualified his formulation by acknowledging the importance of context, it will be recalled that Lord Scarman framed the issues as follows: “Two questions of statutory interpretation, therefore, arise. The first is: what is the natural and ordinary meaning of ‘ordinary residence in the United Kingdom’...? The second is: does the statute in the context of the relevant law against the background of which it was enacted, or in the circumstances of today, including in particular the impact of the Act of 1971, compel one to substitute a special, and, if so, what, meaning to the words ‘ordinarily resident in the United Kingdom’?”[58] 76. Such an approach (adopted by Mr Fordham in the present case) relegates context and purpose into second place, to be consulted after a “natural and ordinary meaning” has been identified. It does not accord with more modern approaches to statutory (and constitutional) interpretation. It involves asserting that the words in question bear a certain “natural and ordinary meaning” and then placing the onus on anyone seeking to advance a different meaning to establish some ground which compels acceptance of that different meaning to be adopted as a “special” meaning in substitution for the “natural and ordinary meaning” previously identified. 77. The Hong Kong courts do not nowadays accept such an approach. As Sir Anthony Mason NPJ pointed out in HKSAR v Lam Kwong Wai:[59] “The modern approach to statutory interpretation insists that context and purpose be considered in the first instance, especially in the case of general words, and not merely at some later stage when ambiguity may be thought to arise (Medical Council of Hong Kong v Chow Siu Shek (2000) 3 HKCFAR 144 at 154B-C; K & S Lake City Freighters Pty Ltd v Gordon & Gotch Ltd (1985) 157 CLR 309 at 315 per Mason J (dissenting, but not on this point); CIC Insurance Ltd v Bankstown Football Club Ltd (1997) 187 CLR 384).” 78. Lord Scarman’s natural and ordinary meaning approach is particularly unhelpful when interpreting words which have some flexibility of meaning. If one approaches the words “ordinarily resident” without trying to assign to them a priori some “natural and ordinary meaning”, but recognizing that they take their meaning from their context, it immediately becomes clear – as illustrated by the various examples discussed in Sections E2 and E3 – that the expression “ordinary residence” is open-textured as a matter of language and inherently capable of assuming different meanings in different contexts. 79. The open-textured and flexible nature of that expression may be contrasted, for instance, with the word “born” used in two of the other categories of entitlement to permanent resident status in Article 24(2): “Chinese citizens born in Hong Kong...”;[60] and “Persons of Chinese nationality born outside Hong Kong of those residents listed in categories (1) and (2)”.[61] The central triggering requirement in each of those categories is the event of birth – a concept which is much more closed and which admits of much less flexibility than the requirement of a continuous seven year period of ordinary residence under Article 24(2)(4). This was indeed the conclusion that the Court reached in Director of Immigration v Chong Fung Yuen.[62] Incidentally, we should point out that we do not necessarily accept all that the Chief Judge stated in relation to Chong Fung Yuen in paragraphs 125 to 131 of the Court of Appeal’s judgment below.[63] E.7 The proper approach to interpreting “ordinarily resided” in Article 24(2)(4) 80. On the basis of the foregoing analysis, it is clear that the “natural and ordinary meaning” attributed by Mr Fordham to the words “ordinarily resident” in Article 24(2)(4) may serve as a starting-point but cannot be decisive. To be able to say of a person or class of persons that they live in Hong Kong “lawfully, voluntarily and for a settled purpose, as part of the regular order of life for the time being” may often justify the conclusion that they are ordinarily resident here, but that will not always be the case. 81. It is always necessary to examine the factual position of the person claiming to be ordinarily resident to see whether there are any special features affecting the nature and quality of his or her residence. If such features exist, one asks whether they result in that person’s residence being qualitatively so far-removed from what would traditionally be recognized as “ordinary residence” as to justify concluding that he or she is not “ordinarily resident”. This is necessarily a question of fact and degree and the outer boundaries of “ordinary residence” are incapable of precise definition. But the exercise is necessary since the meaning of “ordinary residence” in Article 24(2)(4) can only be considered in the factual context presented by the person claiming to come within the meaning of that concept. 82. In the present case, the Court must take as the factual context, the scheme whereby FDHs are allowed to enter and reside in Hong Kong subject to the highly restrictive conditions described in Section B above. Those are the facts that mark out FDHs as a class and characterise the nature and quality of their residence in Hong Kong while employed as such. Those facts must necessarily be at the centre of the Court’s deliberations when it considers whether it is congruent with the concept of “ordinary residence” as employed in Article 24(2)(4) to treat FDHs as a class as not ordinarily resident. 83. It is plainly within the intent of Article 24(2)(4) that the immigration status of persons claiming to come within that category must be taken into account in deciding whether they satisfy the seven-year ordinary residence requirement. This is indicated by the words of the Article itself. They define the category of non-Chinese nationals eligible for Hong Kong permanent resident status, as consisting of: “Persons not of Chinese nationality who have entered Hong Kong with valid travel documents, have ordinarily resided in Hong Kong for a continuous period of not less than seven years and have taken Hong Kong as their place of permanent residence ...” 84. Article 24(2)(4) is therefore addressing the situation of persons who need permission to enter Hong Kong and makes it a requirement that they enter with valid travel documents. The Article therefore commences defining eligibility for permanent resident status in this category by focussing on the point of entry into Hong Kong and laying down the requirement that entry has to be on the basis of a valid travel document. As is to be expected with a constitutional instrument, the Basic Law does not seek to define what does and does not constitute a valid travel document. It leaves it to the Director to determine, as a matter of immigration control, what travel documents to accept as valid for the purposes of entry. 85. Moreover, Article 24(2)(4) implicitly makes immigration control a constant feature in the process of building up eligibility when it stipulates a seven-year qualifying period. The drafters of the Basic Law would plainly not have assumed that a person entering Hong Kong would immediately be given permission to stay for seven or more years. One would naturally expect a person who does not have the right of abode to be allowed to enter subject to a limit in the duration of stay and subject to other conditions, such as whether taking up employment is allowed, etc, while in Hong Kong. It follows that it is implicit in Article 24(2)(4) that the continuous seven year period of ordinary residence can only be achieved subject to continuing immigration control. 86. It has rightly not been suggested that the imposition of such controls is in any way constitutionally objectionable. As was noted in Prem Singh,[64] the Director may undoubtedly exercise his discretions as to whether a non-Chinese person should be allowed to enter Hong Kong and whether permission to remain should be extended, such discretions having a bearing on the entry and seven-year requirements. This is consistent with Article 154(2) which allocates constitutional responsibility for immigration control on the HKSAR Government and provides: “The Government of the Hong Kong Special Administrative Region may apply immigration controls on entry into, stay in and departure from the Region by persons from foreign states and regions.” 87. In the exercise of this power, the Director may impose conditions on the entry of a person which will materially affect the quality of his residence in Hong Kong. F. Are FDHs entitled to be treated as ordinarily resident? 88. Adopting the aforesaid approach, it is clear that prominent distinguishing features have an important bearing on the nature and quality of the residence of FDHs as a class in Hong Kong. Those features have been described in Section B of this judgment. By way of summary, each time a FDH is given permission to enter, such permission is tied to employment solely as a domestic helper with a specific employer (in whose home the FDH is obliged to reside), under a specified contract and for the duration of that contract. The FDH is obliged to return to the country of origin at the end of the contract and is told from the outset that admission is not for the purposes of settlement and that dependents cannot be brought to reside in Hong Kong. 89. It is clear, in our view, that these distinguishing features result in the residence of FDHs in Hong Kong being qualitatively so far-removed from what would traditionally be recognized as “ordinary residence” as to justify concluding that they do not, as a class, come within the meaning of “ordinarily resident” as used in Article 24(2)(4). It follows that in providing that they should not be treated as ordinarily resident, section 2(4)(a)(vi) is consistent with Article 24(2)(4) and constitutionally valid. We accordingly dismiss the appeals. G. Matters which need not be addressed 90. Having arrived at the aforesaid conclusion, it is unnecessary to address Lord Pannick’s second argument which invites the Court to acknowledge a margin of discretion in the legislature to enact exclusionary categories. 91. Moreover, we have concluded that pursuant to the foregoing analysis, the meaning of “ordinarily resided” in Article 24(2)(4) interpreted in the light of its context and purpose is clear, plainly accommodating the exclusionary provisions of section 2(4)(a)(vi). There is accordingly no need and no legal basis for referring to any extrinsic materials in aid of interpreting that Article. H. Reference to the Standing Committee of the National People's Congress under Article 158(3) of the Basic Law H.1 The Commissioner’s request for a reference 92. Article 158 of the Basic Law states:- (1) “The power of interpretation of this Law shall be vested in the Standing Committee of the National People's Congress. (2) The Standing Committee of the National People's Congress shall authorize the courts of the Hong Kong Special Administrative Region to interpret on their own, in adjudicating cases, the provisions of this Law which are within the limits of the autonomy of the Region. (3) The courts of the Hong Kong Special Administrative Region may also interpret other provisions of this Law in adjudicating cases. However, if the courts of the Region, in adjudicating cases, need to interpret the provisions of this Law concerning affairs which are the responsibility of the Central People's Government, or concerning the relationship between the Central Authorities and the Region, and if such interpretation will affect the judgments on the cases, the courts of the Region shall, before making their final judgments which are not appealable, seek an interpretation of the relevant provisions from the Standing Committee of the National People's Congress through the Court of Final Appeal of the Region. When the Standing Committee makes an interpretation of the provisions concerned, the courts of the Region, in applying those provisions, shall follow the interpretation of the Standing Committee. However, judgments previously rendered shall not be affected. (4) The Standing Committee of the National People's Congress shall consult its Committee for the Basic Law of the Hong Kong Special Administrative Region before giving an interpretation of this Law.” We shall, in this judgment, refer to the Standing Committee of the National People’s Congress simply as the Standing Committee. 93. Following the decisions of this Court in Ng Ka Ling v Director of Immigration[65]and Chan Kam Nga v Director of Immigration[66], the State Council submitted a Motion to the Standing Committee requesting an interpretation of Articles 22(4) and 24(2)(3) of the Basic Law. The Motion was submitted after the Chief Executive of the HKSAR had provided a report to the Standing Committee. 94. An interpretation was made and this was adopted by the Standing Committee of the Ninth National People’s Congress at its Tenth Session on 26 June 1999. The interpretation, headed “The Interpretation by the Standing Committee of the National People’s Congress of Articles 22(4) and 24(2)(3) of the Basic Law of the Hong Kong Special Administrative Region of the People’s Republic of China” is annexed to this judgment. We shall refer to this as the 1999 Interpretation. 95. The Commissioner places considerable reliance on the 1999 Interpretation, focusing on the sentence in para 2[67]:- “The legislative intent as stated by this Interpretation, together with the legislative intent of all other categories of Article 24(2) of the Basic Law of the Hong Kong Special Administrative Region of the People’s Republic of China, have been reflected in the “Opinions on the Implementation of Article 24(2) of the Basic Law of the Hong Kong Special Administrative Region of the People’s Republic of China” adopted at the Fourth Plenary Meeting of the Preparatory Committee for the Hong Kong Special Administrative Region of the National People’s Congress on 10 August 1996.” 96. The Commissioner relies on the Statement as part of his arguments on extrinsic materials. It was submitted that the Statement provided at least strong persuasive authority as to the legislative intent behind Article 24(2)(4) of the Basic Law. However, we have already held[68] that there is no need to resort to extrinsic materials for the purposes of construing Article 24(2)(4). 97. As an alternative argument based on the Statement, the Commissioner further submits that the effect of the Statement was that it may well be binding on the Hong Kong courts by reason of Article 158 of the Basic Law. We use the expression “may well be” because although Lord Pannick did not invite this Court actually to hold ourselves that this was the effect of the Statement for the purposes of Article 158, nevertheless it was his submission that this was arguably the position. And for that reason, the meaning of the concept of an “interpretation” under Article 158, and therefore the meaning and effect of the Statement itself, were matters which ought to be referred by this Court to the Standing Committee for an authoritative interpretation. Lord Pannick, however, stressed that a reference would only be necessary if the issue of the true interpretation of Article 24(2)(4) using the common law approach, were determined against him. 98. There were perhaps two reasons why the Commissioner did not at this stage invite the Court, in the absence of an authoritative interpretation by the Standing Committee, to hold that the Statement was actually binding for the purposes of Article 158 (rather than just being persuasive, this being the argument based on extrinsic materials):- (1) A recognition of the point forcefully made by Mr Fordham that on a plain reading of the 1999 Interpretation itself, it was apparent that the only provisions that were interpreted were Articles 22(4) and 24(2)(3) of the Basic Law, as its title indicated. There were also numerous references in the body of that document to an interpretation of only those specific provisions of the Basic Law. Lord Pannick accepted that the Statement was an “incidental” construction of other provisions in the Basic Law, namely, the other categories in Article 24(2). (2) In Director of Immigration v Chong Fung Yuen[69], it was accepted by the Director of Immigration that the Statement was not binding on the Court as an interpretation made under Article 158.[70] 99. Be that as it may, this Court was asked to seek an interpretation from the Standing Committee. The following questions were identified by the Commissioner:- “1. What is the meaning of an “interpretation” which the NPCSC has power to give under Article 158(1) of the Basic Law? 2. Whether the Statement in the 1999 Interpretation is or constitutes part of an “interpretation” within the meaning of Article 158(1), such that it is binding on and shall be applied by the courts of the HKSAR when deciding cases involving any one of the categories under Article 24(2) of the Basic Law (including Article 24(2)(4))?” Before addressing the request to refer these questions, we must first consider the jurisdiction of the Court under Article 158. H.2 The Court’s jurisdiction under Article 158 of the Basic Law 100. Article 158 is a specific provision governing the interpretation of the Basic Law. In this regard, only the Standing Committee and the Hong Kong courts are referred to. The Article sets out in clear terms the power of the Standing Committee to interpret provisions of the Basic Law and the effect of such interpretations. It also sets out the powers of interpretation of the Basic Law vested in the Hong Kong courts and makes provision for mandatory references by the Court of Final Appeal to the Standing Committee for authoritative interpretations of Basic Law provisions. 101. This Court has, on a number of previous occasions, discussed this Article, in particular in Ng Ka Ling v Director of Immigration[71], Lau Kong Yung v Director of Immigration[72], Director of Immigration v Chong Fong Yuen[73], Tam Nga Yin v Director of Immigration[74] and Democratic Republic of the Congo v FG Hemisphere Associates LLC (No. 1)[75]. It is clear from the way this Court has treated it that Article 158 sets out what is a constitutional jurisdiction of the Court in relation to interpretation and reference. 102. As Article 158(1) makes clear, the power of interpretation of the Basic Law rests with the Standing Committee. This is derived from the general power of the Standing Committee to interpret laws under Article 67(4) of the Constitution of the People’s Republic of China. In Lau Kong Yung, this Court stated the power to be in “general and unqualified terms”, unrestricted by the remainder of Article 158[76]. It is plenary in that it covers all articles in the Basic Law and may be exercised even in the absence of any legal proceedings.[77] 103. Only the courts of the HKSAR are also given the power to interpret the Basic Law, but the ambit of this power is precisely defined and is not as wide as the general power vested in the Standing Committee:- (1) Under Article 158(2), the Hong Kong courts are authorized (this is a constitutional authorization[78]) to interpret “on their own, in adjudicating cases, the provisions of [the Basic Law] which are within the limits of the autonomy of [the HKSAR]”. This limitation is, however, not critical because the first sentence of Article 158(3) makes it clear that the Hong Kong courts “may also interpret other provisions of [the Basic Law] in adjudicating cases”. It is therefore open to the Hong Kong courts to interpret all articles in the Basic Law. (2) The real limitations, however, imposed on the Hong Kong courts are two. (3) First, our courts can only become engaged in construing provisions of the Basic Law in the course of adjudicating cases. In other words, the role of our courts is to adjudicate and not give advisory opinions.[79] (4) Secondly, as Article 158(3) states, where, in adjudicating cases, the Hong Kong courts need to interpret Basic Law provisions “concerning affairs which are the responsibility of the Central People’s Government, or concerning the relationship between the Central Authorities and [the HKSAR]”[80], then the Hong Kong courts “shall, before making their final judgments which are not appealable, seek an interpretation of the relevant provisions from [the Standing Committee] through the Court of Final Appeal”. The seeking of an interpretation from the Standing Committee in these circumstances is mandatory[81] and it is the responsibility of the Court of Final Appeal.[82] It is part of the Court’s “constitutional jurisdiction”.[83] (5) Before the Court of Final Appeal can make a reference, it must be satisfied that two conditions (the “classification” and the “necessity” conditions) are fulfilled, both these conditions flowing from the language of Article 158(3) itself. They were described by Chief Justice Li in Ng Ka Ling in this way:- “As far as the Court of Final Appeal is concerned, it has a duty to make a reference to the Standing Committee if two conditions are satisfied: (1) First, the provisions of the Basic Law in question: (a) concern affairs which are the responsibility of the Central People’s Government; or (b) concern the relationship between the Central Authorities and the Region. That is, the excluded provisions. We shall refer to this as “the classification condition”. (2) Secondly, the Court of Final Appeal in adjudicating the case needs to interpret such provisions (that is the excluded provisions) and such interpretation will affect the judgment on the case. We shall refer to this as ‘the necessity condition.’ ”[84] (6) In looking at the classification condition, in determining whether a provision is an excluded provision, the Court will look at the character of the provision itself to see whether it concerns “affairs which are the responsibility of the Central People’s Government, or concerning the relationship between the Central Authorities and [the HKSAR]”. The Court will not look at the effect of the implementation of the provision in any given case in order to determine whether or not the provision is an excluded one: see Chong Fung Yuen[85] and Tam Nga Yin[86]. (7) Underlying any consideration by the Court whether or not to refer under Article 158 is the requirement of arguability. As was said by Chief Justice Li in Ng Ka Ling[87]:- “We are at the stage of considering whether the Court is obliged to make a reference under art.158. At this stage, the Court is concerned with what is arguable and not with deciding the question of interpretation. That is a matter for the Standing Committee if a reference has to be made and a matter for the Court if a reference does not have to be made. Of course, an argument which is plainly and obviously bad would not be arguable. If the Court decides at this stage, that the point is unarguable, that would be an end of the matter as far as a question of reference is concerned. If the Court decides that it is arguable, the Court would then consider whether the classification and necessity conditions are satisfied.” 104. The arguability factor is implicit in Article 158(3) to ensure integrity in the operation of a reference. Otherwise, there will be a risk of potential abuse; all sorts of fanciful arguments could then be made just to seek a reference to the Standing Committee. 105. The seeking of an interpretation by the Court of Final Appeal (or, in common parlance, making a reference to the Standing Committee[88]) is therefore a precisely defined duty under Article 158(3). It is mandatory in character: it is specifically limited in its application to the excluded provisions of the Basic Law and it applies only in the particular circumstances contemplated by that provision, that is, the conditions of classification, necessity and arguability. Then and only then the Court of Final Appeal must make a reference to the Standing Committee.[89] 106. Nor is there any basis for otherwise implying a general power in the Court of Final Appeal to seek an interpretation from the Standing Committee. This conclusion follows from: (a) the fact that that Court of Final Appeal exercises the power of final adjudication as an element in the exercise of judicial power;[90] (b) the responsibility of the courts of the HKSAR under Article 158(3) to adjudicate cases “on their own”, subject to the mandatory reference procedure under that article and to the application of Standing Committee interpretations under that article; and (c) the long established rule that a common law court cannot abdicate any part of its judicial function to any other body. 107. What is the effect of an interpretation made by the Standing Committee? Where an interpretation of a provision of the Basic Law is made by the Standing Committee under Article 158(1) pursuant to its plenary powers, it is of course binding on the courts.[91] Where the Standing Committee has made an interpretation after a judicial reference under Article 158(3), that paragraph states the consequence as follows:- “…When the Standing Committee makes an interpretation of the provisions concerned, the courts of the Region, in applying those provisions, shall follow the interpretation of the Standing Committee. However, judgments previously rendered shall not be affected.” It should be noted that upon an Article 158(3) reference, the interpretation of the relevant provision or provisions by the Standing Committee will bind all courts in Hong Kong.[92] It has the same binding effect as the exercise of the power of interpretation under Article 158(1). In short, the courts of Hong Kong are bound to follow an interpretation made by the Standing Committee whether under Article 158(1) or (3).[93] 108. With these principles in mind, we turn to the question of a reference in the present case. H.3 Determination of the Commissioner’s application for a reference 109. As stated earlier, Article 158 of the Basic Law sets out the constitutional jurisdiction of the Court in relation to both interpretation and reference.[94] In deciding whether or not to refer, the court must be satisfied that both the classification and necessity conditions are fulfilled (as well as the arguability aspect). This is for the Court to decide and for it alone to decide [95]. Lord Pannick confirmed that this was the position. 110. The classification condition is satisfied in the present case. Article 158 of the Basic Law is a provision that does concern the relationship between the Central Authorities and the HKSAR. There was no real dispute about this. 111. The necessity condition, however, is not. In the light of the conclusion this Court has reached on the issue of the true construction of Article 24(2)(4), a reference to the Standing Committee is simply unnecessary. As stated above, Lord Pannick addressed us on this basis[96]. In our view, it is plainly the right approach. 112. The necessity condition not being satisfied, it is not necessary to deal with the arguability factor either. Accordingly, the request for a reference must be rejected. I. Outcome 113. The Appeals are accordingly dismissed. Any submissions as to costs should be lodged in writing within 14 days of the date of this judgment and any replies within 14 days thereafter. Mr Michael Fordham QC, Ms Gladys Li SC, Mr PY Lo and Mr Benson Tsoi instructed by Barnes & Daly and assigned by the Legal Aid Department for the appellants Lord Pannick QC, Mr Anderson Chow SC and Ms Eva Sit instructed by the Department of Justice for the 1st respondent Annex THE INTERPRETATION BY THE STANDING COMMITTEE OF THE NATIONAL PEOPLE'S CONGRESS OF ARTICLES 22(4) AND 24(2)(3) OF THE BASIC LAW OF THE HONG KONG SPECIAL ADMINISTRATIVE REGION OF THE PEOPLE'S REPUBLIC OF CHINA (Adopted by the Standing Committee of the Ninth National		People's Congress at its Tenth Session on 26 June 1999) The Standing Committee of the Ninth National People's Congress examined at its Tenth session the "Motion Regarding the Request for an Interpretation of Articles 22(4) and 24(2)(3) of the Basic Law of the Hong Kong Special Administrative Region of the People's Republic of China" submitted by the State Council. The motion of the State Council was submitted upon the report furnished by the Chief Executive of the Hong Kong Special Administrative Region under the relevant provisions of Articles 43 and 48(2) of the Basic Law of the Hong Kong Special Administrative Region of the People's Republic of China. The issue raised in the Motion concerns the interpretation of the relevant provisions of the Basic Law of the Hong Kong Special Administrative Region of the People's Republic of China by the Court of Final Appeal of the Hong Kong Special Administrative Region in its judgment dated 29 January 1999. Those relevant provisions concern affairs which are the responsibility of the Central People's Government and concern the relationship between the Central Authorities and the Hong Kong Special Administrative Region. Before making its judgment, the Court of Final Appeal had not sought an interpretation of the Standing Committee of the National People's Congress in compliance with the requirement of Article 158(3) of the Basic Law of the Hong Kong Special Administrative Region of the People's Republic of China. Moreover, the interpretation of the Court of Final Appeal is not consistent with the legislative intent. Therefore, having consulted the Committee for the Basic Law of the Hong Kong Special Administrative Region under the Standing Committee of the National People's Congress, the Standing Committee of the National People's Congress has decided to make, under the provisions of Article 67(4) of the Constitution of the People's Republic of China and Article 158(1) of the Basic Law of the Hong Kong Special Administrative Region of the People's Republic of China, an interpretation of the provisions of Articles 22(4) and 24(2)(3) of the Basic Law of the Hong Kong Special Administrative Region of the People's Republic of China as follows: 1. The provisions of Article 22(4) of the Basic Law of the Hong Kong Special Administrative Region of the People's Republic of China regarding "For entry into the Hong Kong Special Administrative Region, people from other parts of China must apply for approval" mean as follows: People from all provinces, autonomous regions, or municipalities directly under the Central Government, including those persons of Chinese nationality born outside Hong Kong of Hong Kong permanent residents, who wish to enter the Hong Kong Special Administrative Region for whatever reason, must apply to the relevant authorities of their residential districts for approval in accordance with the relevant national laws and administrative regulations, and must hold valid documents issued by the relevant authorities before they can enter the Hong Kong Special Administrative Region. It is unlawful for people from all provinces, autonomous regions, or municipalities directly under the Central Government, including persons of Chinese nationality born outside Hong Kong of Hong Kong permanent residents, to enter the Hong Kong Special Administrative Region without complying with the appropriate approval procedure prescribed by the relevant national laws and administrative regulations. 2. It is stipulated in the first three categories of Article 24(2) of the Basic Law of the Hong Kong Special Administrative Region of the People's Republic of China that the "permanent residents of the Hong Kong Special Administrative Region shall be: (1) Chinese citizens born in Hong Kong before or after the establishment of the Hong Kong Special Administrative Region; (2) Chinese citizens who have ordinarily resided in Hong Kong for a continuous period of not less than seven years before or after the establishment of the Hong Kong Special Administrative Region; (3) Persons of Chinese nationality born outside Hong Kong of those residents listed in categories (1) and (2);". The provisions of category (3) regarding the "persons of Chinese nationality born outside Hong Kong of those residents listed in categories (1) and (2)" mean both parents of such persons, whether born before or after the establishment of the Hong Kong Special Administrative Region, or either of such parents must have fulfilled the condition prescribed by category (1) or (2) of Article 24(2) of the Basic Law of the Hong Kong Special Administrative Region of the People's Republic of China at the time of their birth. The legislative intent as stated by this Interpretation, together with the legislative intent of all other categories of Article 24(2) of the Basic Law of the Hong Kong Special Administrative Region of the People's Republic of China, have been reflected in the "Opinions on the Implementation of Article 24(2) of the Basic Law of the Hong Kong Special Administrative Region of the People's Republic of China" adopted at the Fourth Plenary Meeting of the Preparatory Committee for the Hong Kong Special Administrative Region of the National People's Congress on 10 August 1996. As from the promulgation of this Interpretation, the courts of the Hong Kong Special Administrative Region, when referring to the relevant provisions of the Basic Law of the Hong Kong Special Administrative Region of the People's Republic of China, shall adhere to this Interpretation. This Interpretation does not affect the right of abode in the Hong Kong Special Administrative Region which has been acquired under the judgment of the Court of Final Appeal on the relevant cases dated 29 January 1999 by the parties concerned in the relevant legal proceedings. Other than that, the question whether any other person fulfils the conditions prescribed by Article 24(2)(3) of the Basic Law of the Hong Kong Special Administrative Region of the People's Republic of China shall be determined by reference to this Interpretation. [1] Article 24(2)(5) addresses the position of the children of persons coming within Article 24(2)(4). [2] Cap 115. [3] Vallejos Evangeline Banao v Commissioner of Registration HCAL 124/2010 (30 September 2011); Domingo Daniel L v Commissioner of Registration HCAL 128/2010 (3 November 2011). [4] Hon Cheung CJHC, Tang VP and Stock VP, Vallejos Evangeline Banao v Commissioner of Registration CACV 204/2011, (28 March 2012); Domingo Daniel L v Commissioner of Registration CACV 261/2011 (19 July 2012). [5] With Ms Gladys Li SC, Mr P Y Lo and Mr Benson Tsoi. [6] [1983] 2 AC 309. [7] With Mr Anderson Chow SC and Ms Eva Sit. [8] Annexed to this judgment. [9] See Section H. below. [10] R v Barnet London Borough Council ex parte Shah [1983] 2 AC 309 at 343. [11] (2001) 4 HKCFAR 278 at 283-284. [12] Fateh Muhammad v Commissioner of Registration (2001) 4 HKCFAR 278; and Prem Singh v Director of Immigration (2003) 6 HKCFAR 26. [13] Levene v Commissioners of Inland Revenue [1928] AC 217; Inland Revenue Commissioners v Lysaght [1928] AC 234; and Thomson v Minister of National Revenue [1946] SCR 209. [14] R v Barnet London Borough Council ex parte Shah [1983] 2 AC 309. [15] Lau San Ching v Apollonia Liu [1995] 5 HKPLR 23. [16] Re Ip Pui Man Nina [2011] 3 HKLRD 299. [17] Mark v Mark [2006] 1 AC 98. [18] [2011] 3 HKLRD 299. [19] One year in England and Wales under section 5(2) of the Domicile and Matrimonial Proceedings Act 1973. [20] Matrimonial Causes Ordinance, Cap 179, section 3. [21] Mark v Mark [2006] 1 AC 98, §33. [22] Ibid. [23] [1987] HKLR 356. [24] [1987] HKLR 798. [25] [1995] 5 HKPLR 23. [26] [1928] AC 217. [27] [1928] AC 234. [28] Section 46: “Where the Treasury have, before the commencement of this Act, issued or may thereafter issue any securities which they have power to issue for the purpose of raising any money or any loan with a condition that the interest thereon shall not be liable to tax or super tax so long as it is shown, in manner directed by the Treasury, that the securities are in the beneficial ownership of persons who are, not ordinarily resident in the United Kingdom, the interest of securities issued with such a condition shall be exempt accordingly.” [29] At 224. [30] Levene at 232 per Lord Warrington of Clyffe. [31] Levene at 225 per Viscount Cave LC. [32] Levene at 227. [33] Lysaght at 243-244. [34] [1946] SCR 209. [35] At 214. [36] At 231-232. [37] Thomson v Minister of National Revenue [1946] SCR 209 at 224. [38] The approach of the Divisional Court was the same as the Court of Appeal’s on this issue. [39] Local Education Authority Awards Regulations 1979, regulation 13. [40] [1982] QB 688 at 720. [41] At 721. [42] At 722-723. [43] At 723. [44] At 729. [45] [1983] 2 AC 309 at 340. [46] Ibid. [47] At 337. [48] Ibid. [49] At 348. [50] At 349. [51] Although the UK Parliament disagreed and amended the regulation to remove any duty on local authorities to provide an award if any part of the three-year period in question has “been wholly or mainly for the purposes of receiving full-time education”. See R v Hereford and Worcester County Council, ex p Wimbourne The Times, 12 November 1983. [52] (2001) 4 HKCFAR 278 at 283-284. [53] (2003) 6 HKCFAR 26 at §75. [54] (2001) 4 HKCFAR 278 at 285. [55] Ng Ka Ling v Director of Immigration (1999) 2 HKCFAR 4 at 36; Lau Kong Yung v Director of Immigration (1999) 2 HKCFAR 300 at 312; Prem Singh v Director of Immigration (2003) 6 HKCFAR 26 at §§56-57. [56] (2003) 6 HKCFAR 26. [57] Section E1. [58] [1983] 2 AC 309 at 340. [59] (2006) 9 HKCFAR 574 at §63. Reiterated by Li CJ in HKSAR v Cheung Kwun Yin (2009) 12 HKCFAR 568 at§§12-13. [60] Article 24(2)(1). [61] Article 24(2)(3). [62] (2001) 4 HKCFAR 211 at 231-233. [63] CACV 204/2011 (28 March 2012). [64] (2003) 6 HKCFAR 26 at §63. [65] (1999) 2 HKCFAR 4. [66] (1999) 2 HKCFAR 82. [67] This sentence was referred to by Lord Pannick in his submissions as “the Statement” – we are content to use the term in this judgment. [68] See section G above. [69] (2001) 4 HKCFAR 211. [70] This is recorded at 217J-218A. [71] (1999) 2 HKCFAR 4. [72] (1999) 2 HKCFAR 300. [73] (2001) 4 HKCFAR 211. [74] (2001) 4 HKCFAR 251. [75] (2011) 14 HKCFAR 95. [76] At 323C. [77] Lau Kong Yung at 324E-F per Chief Justice Li and 345E-F per Sir Anthony Mason. In his judgment, Chief Justice Li made reference to Hong Kong’s New Constitutional Order (Second Edition) by Professor Yashi Ghai at 198. Two Interpretations, in relation to, respectively, (a) Article 7 of Annex 1 and Article III of Annex II of the Basic Law; and (b) Article 53(2) of the Basic Law (adopted by the Standing Committee of the 10th National People’s Congress on 6 April 2004 and 27 April 2005), were given by the Standing Committee in the absence of any litigation. [78] Ng Ka Ling at 29G. [79] Ng Ka Ling at 30E. [80] These are referred to as the “excluded provisions”: see Ng Ka Ling at 30F-G. [81] This can be referred to as a “duty”: see Lau Kong Yung at 322H. [82] Ng Ka Ling at 30H-I. [83] Ng Ka Ling at 32C. [84] At 30I-31B. It is right to point out that in relation to the classification condition, the Court in Ng Ka Ling added a rider that the Court also had to ask “as a matter of substance” what was “predominantly” the provision in the Basic Law that was sought to be referred for an interpretation: at 33D. Following the 1999 Interpretation, in Lau Kong Yung, the Court said that it may be necessary to revisit the “classification and necessity conditions and the predominant test in an appropriate case”: at 324H-I. These issues however, do not arise in the present appeal. [85] At 229A-H. [86] At 257I-J. [87] At 32F-G. [88] In Lau Kong Yung, the term “judicial reference” is used: at 322H. [89] In saying this we recognize that the Court may arrive at its own interpretation of the relevant provisions in the Basic Law provided it does not arrive at a final view before reference: see Democratic Republic of the Congo v FG Hemisphere Associates LLC (No 1) (2011) 14 HKCFAR 95 at §§183, 223, 415. [90] Basic Law, Articles 2, 19(1) and 82. [91] Lau Kong Yung at 324D. [92] See the remarks of Sir Anthony Mason in Lau Kong Yung at 345H-346B. [93] Chong Fung Yuen at 223B. [94] See para 103 above. [95] Ng Ka Ling at 31C. [96] See para 97 above. 
Chief Justice Cheung: 1. At the conclusion of the hearing, we dismissed this appeal which raised a short point involving the priority of claims to the proceeds of sale of a property. These are my reasons. The facts 2. Tang Shung Ching Sabin[1] was the registered owner of the property known as Flat No. 1F, Ground Floor, Block 2, Lung Cheung Court, No. 29 Broadcast Drive, Kowloon together with a car parking space in the residential development[2]. The development had been constructed under a government housing scheme for civil servants. Tang purchased the Property and became its registered owner in 2004. Under the legal assignment by which he purchased the Property, the Property was assigned to him subject to certain “non-alienation” covenants, which prohibited him from assigning, mortgaging, charging or otherwise parting with possession of the Property or any interest therein, or entering into any agreement to do so, without the previous written consent of the Financial Secretary Incorporated[3]. Such consent would generally be granted only upon payment of a premium to the FSI. 3. To secure the performance of the non-alienation covenants, Tang gave a legal charge over the Property in favour of the FSI. The legal charge, like the assignment containing the non-alienation covenants, was registered in the Land Registry. 4. Over the years, Tang borrowed various loans from no less than five finance companies, including Winland Finance Limited[4] and Gain Hero Finance Limited[5], respectively the appellant and respondent in this appeal. 5. No doubt because of the prohibitions created by the non-alienation covenants, these loans could not be secured by any mortgage or charge over the Property, and means were adopted to obtain some form of security for the loans and have the loan agreements registered in the Land Registry. 6. In the case of Winland, this was sought to be done by means of an assignment of the balance of all future proceeds of sale of the Property receivable by Tang to Winland as security for the repayment of the loan. In clause 9 of the loan agreement dated 23 July 2014[6] whereby the loan of $2.2 million was advanced to Tang, the following was provided : “9.1 In consideration of the Facility advanced by the Lender under this Agreement and as security for repayment of all the indebtedness owed to the Lender by the Borrower, the Borrower as Beneficial Owner hereby assigns and agrees to assign to the Lender all his /her /their right and interest (“Assigned Rights”) to and in the balance of the consideration or purchase money for the sale of the Property more particularly described herein (after deduction and payment of any premium payable to the Government of Hong Kong Special Administrative Region the amount provided for removal of the restriction against alienation of the Property contained in clause 4 of the Assignment registered in the Land Registry by Memorial No.UB9403278 and any redemption money payable to the mortgagee /chargee under the existing Mortgage /Legal Charge of the Property) and any money whatsoever accrued or payable to the Borrower, in the Borrower's capacity as owner, vendor, assignor, transferor or mortgagor, from the purchaser, assignee, transferee or mortgagee of the Property or any other person(s) subject to the provisos contained in clauses 9.4 and 9.3 hereof ... 9.2 Upon full payment to the Lender of all the indebtedness owed by the Borrower to the Lender, the Lender shall at the request and cost of the Borrower reassign the Assigned Rights to the Borrower. 9.3 If, for whatsoever reason, the assignment of the Assigned Rights mentioned in clause 9.1 hereof does not have the effect as such, the agreement to assign made by the Borrower under the said clause 9.1 shall operate to assign the Assigned Rights and there shall deem to be an effective assignment of the Assigned Rights as soon as the Borrower is in the position to make such an assignment and the Borrower hereby covenants to the Lender that the Borrower shall from time to time at the request of the Lender execute all such documents and do all such acts as may be necessary to perfect to give effect to the said assignment. 9.4 The Borrower hereby IRREVOCABLY NOMINATED AND APPOINTS the Lender to be the Borrower's true and lawful attorney for the Borrower and in the Borrower's name or in the name of the Lender as such attorney at any time or times during the continuance of this Agreement to do perform transact and effectuate all or any of the following acts, deeds, matters and things:- (i) to receive the balance of the consideration or purchase money for the sale of the Property (after deduction and payment of any premium payable to the Government of the Hong Kong Special Administrative Region the amount provided for removal of the restriction against alienation of the Property contained in clause 4 of the Assignment registered in the Land Registry by Memorial No.UB9403278 and any redemption money payable to the mortgagee /chargee under the existing Mortgage /Legal Charge of the Property) and any money whatsoever accrued or payable to the Borrower … 9.5 The Borrower hereby covenants to the Lender that except with the written consent of the Lender, the Borrower shall not sell, charge, mortgage, dispose of or otherwise deal with the Property or any interest therein before the Borrower has fully settled all indebtedness under this Agreement. … 9.7 For the purpose of notifying the public of the Lender's right to receive the consideration or purchase money for the sale of the Property and other money as aforesaid, the parties hereto shall register this Agreement against the Property in the relevant Land Registry.” 7. As envisaged in clause 9.7, the parties managed to have their loan agreement registered in the Land Registry on 17 September 2014. 8. Relevantly, on 27 November 2014, Tang borrowed another sum of $1 million from Gain Hero. Under the loan agreement, Tang agreed not to sell or create or enter into any charge, lien or other encumbrance over the Property or otherwise dispose of the Property until full repayment. He also appointed Gain Hero as his attorney to “receive payment of the balance of purchase price” payable on completion of the sale of the Property. The parties also successfully procured the registration of the loan agreement in the Land Registry on 8 December 2014. 9. Tang defaulted in repayment of the loan due to Gain Hero. On 8 June 2015, Gain Hero obtained judgment against Tang for the sum of $1,130,684.93 together with interest and costs[7]. 10. On 29 June 2015, Gain Hero obtained a charging order nisi against the Property in enforcement of the judgment debt. On 4 August 2015, Gain Hero obtained a charging order absolute. Both the charging order nisi and charging order absolute were registered by Gain Hero in the Land Registry against the Property. 11. On 23 June 2016, Gain Hero commenced proceedings[8] for the sale of the Property in enforcement of the charging order. On 13 March 2017, the court made an order for the sale of the Property at a price of not less than $8 million, unless Tang should, within 28 days, redeem the Property by paying the outstanding judgment debt together with interest and costs – which Tang never did. The order directed that the proceeds of sale be distributed in the following order : “(a) for payment of premium (if any) to the Lands Department; (b) for discharge of all Government rent, taxes, rates and other outgoings due and affect the Property; (c) for payment to prior encumbrancer ranking immediately before the Charging Order, if any; (d) for payment of costs and expenses incurred incidental and in relation to the sale and discharge of all auctioneer’s remuneration and/or estate agent’s commission, not exceeding 1% of the selling price; (e) for payment of the legal costs of the sale of the Property, not exceeding HK$12,000.00 excluding disbursements and reimbursement to the Plaintiff for any fees paid to the Lands Department for the purpose of assessment of premium; (f) for payment of such amount due and owing under the Charging order as set out in paragraph 1 above; (g) for payment of the Plaintiff’s costs of the present proceedings as set out in paragraph 7 herein; and (h) any surplus be paid to the subsequent incumbrancer ranking immediately behind the Charging Order and if there is none, be paid to the Defendant.” 12. The order for sale was registered in the Land Registry on 29 March 2017. 13. On 31 August 2017, the Property was agreed to be sold to a purchaser for $10.08 million. 14. On 4 September 2017, Winland was informed of the intended sale of the Property pursuant to the court order. Winland, whose loan to Tang had also remained outstanding, maintained that the price of $10.08 million was below the market value of the Property. In any event, it also contended that it had priority over Gain Hero to the net proceeds of sale. The CFI decision 15. This eventually led to the commencement of two sets of proceedings between Winland and Gain Hero[9], raising a number of issues which were resolved by Deputy High Court Judge Keith Yeung SC (as he then was) in his decision dated 21 March 2019[10]. Of these issues, only one remained contentious in this appeal, namely the question of priority in relation to the net proceeds of sale of the Property after deduction of the amount required to discharge the encumbrance registered in favour of the FSI. However, it is noteworthy that one other issue decided by the deputy judge was that the assignment of proceeds of sale under clause 9 of the Winland loan agreement did not create any interest in or otherwise affect the Property, and was therefore not registrable in the Land Registry under the Land Registration Ordinance[11]. He therefore ordered the registration to be vacated[12]. From that particular decision, there was no appeal. 16. In relation to the question of priority, the deputy judge also held in favour of Gain Hero. He decided that the charging order obtained by Gain Hero was a charge on Tang’s beneficial interest in the Property itself. It was not simply a charge over the proceeds of sale of the Property. He held that Gain Hero’s interest thereunder had priority over any unregistrable interest which Winland had in any future sale proceeds of the Property under the Winland loan agreement[13]. The Court of Appeal judgment 17. On appeal[14], the Court of Appeal agreed with the deputy judge[15]. Giving the lead judgment of the court, Yuen JA referred to the submission on Winland’s behalf that its claim had priority “because [Gain Hero’s] interest under the Charging Order arose after [Winland’s] interest under the Assignment ... (bearing in mind the equitable nature of the interest created by the Charging Order)”[16] and explained : “27.2 I do not accept that submission as it ignores the difference in the class of assets. GH [Gain Hero]’s interest was in a different class (an interest in the Property given by statute) from that of WF [Winland]’s (an interest in future proceeds). 27.3 The enhanced status of a holder of a Charging Order is obvious. He can enforce it by sale, or by the appointment of a receiver to receive rents. And if he applies to court for sale of the land, the debtor can only redeem the land if he makes full repayment of the debt. The debtor cannot get any share of the proceeds for himself before the debt owed to the holder of the Charging Order is entirely satisfied. 28.1 It was argued on behalf of WF that the nature of the Charging Order is equitable. That is so, but that only refers to the nature of the interest. What is crucial is the class of the asset. To compare GH’s interest in the land with WF’s interest in the proceeds is not comparing like with like, for they are not within the same class of assets. Put another way, after GH obtained the Charging Order, WF could not enjoy any “equal equity” because WF’s security interest remained attached to the proceeds only. 28.2 Since Tang must defer to GH in his claim to proceeds, WF (as the assignee of Tang’s rights) cannot enjoy better rights than him. The judge was clearly right when he held that, like Tang, WF can only get its hands on what is left after GH has satisfied its loan.” Leave to appeal 18. By a notice of motion dated 24 May 2021, Winland sought leave to appeal on a question of law of great general or public importance: “What is the priority to the sale proceeds (“Proceeds”) of real property (“Property”) as between (i) an earlier assignee of the Proceeds pursuant to an equitable assignment of the same before the Property was sold and (ii) a judgment creditor who obtained a later charging order against the Property.” 19. By an order dated 9 August 2021, the Court of Appeal gave leave to appeal, and thus this appeal. Counsel’s arguments 20. Ms Audrey Eu SC[17], for Winland, essentially submitted that an assignment of future property takes effect in equity immediately upon the assignment provided the assignee has given value; whereas under section 20B(3) of the High Court Ordinance[18], a charging order has the effect of and is enforceable in the same manner as “an equitable charge created by the debtor by writing under his hand”. As opposed to a mortgage which gives the mortgagee an interest in the property in question, a charging order is simply a statutory security whereby the property is appropriated for the payment of the judgment debt, which is achieved by selling the property and using the proceeds to pay off the judgment debt under a court order for sale. In essence, a charging order, taking effect as an equitable charge, only operates to give the judgment creditor an interest in the proceeds of sale. Since the equitable assignment of the proceeds of sale under clause 9 of the Winland loan agreement was created prior to the charging order obtained by Gain Hero, the applicable priority rule should be that which governs competing priority between equitable interests, namely that the first-in-time prevails[19]. 21. A variation on the same theme was that Tang could not give to Gain Hero what he did not have (nemo dat quod non habet[20]). In other words, Gain Hero could not by its charging order get any more than whatever Tang could honestly give him. Its interest in the Property under the charging order was subject to every liability under which Tang held the Property. Gain Hero was not a subsequent bona fide mortgagee or purchaser for valuable consideration as it gave no fresh consideration in obtaining a charging order[21]. It therefore could not invoke the “equity’s darling” exception to the nemo dat rule. 22. Ms Eu contended that the Court of Appeal erred in effectively holding that, in law, there is a “class” exception to the “nemo dat” or “first-in-time prevails” principle[22]. She submitted that the dispute over priority in this case related to the proceeds, not the Property, and therefore there was no question of any different class of interest involved. 23. Gain Hero basically repeated its arguments below and supported the lower courts’ reasoning. Discussion 24. What was not in issue in this appeal was that an assignment of future property, including, as in the present case, the future proceeds of sale of a property (if and when sold), is more than a contract to assign the proceeds in future but takes effect in equity immediately upon the assignment provided the assignee has given value. When the subject matter comes into existence, the beneficial interest vests immediately in the assignee. Equity, treating that as done which ought to be done, fastens upon the property, and the contract to assign becomes in equity a completed assignment. The equitable ownership in the property passes automatically to the assignee, without any intermediate vesting of the beneficial interest in the assignor. Once the future property comes into existence and is acquired by the assignor, the rights of the assignee are deemed to “relate back” to the date of the assignment[23]. 25. Therefore, there was no dispute in the present appeal that the assignment of the balance of the sale proceeds of the Property (if and when sold) by Tang in favour of Winland was an equitable assignment of the future proceeds and took effect in equity immediately upon the making of the assignment. 26. However, what was also no longer in dispute after the CFI decision was that the assignment of the future proceeds did not create any interest in or otherwise affect the Property, so much so that the registration of the Winland loan agreement in the Land Registry against the Property was rightly vacated[24]. This is because only instruments in writing that affect land may be registered in the Land Registry under the Land Registration Ordinance: section 2(1)[25]. Instruments that do not affect land are simply not registrable. The provisions in sections 3(1) and (2)[26], which govern priority, are only applicable to instruments that are registrable under the Land Registration Ordinance – where the instruments are all so registered, section 3(1) determines priority; where a prior instrument that is registrable is not registered, section 3(2) applies to render it “absolutely null and void to all intents and purposes” as against “any subsequent bona fide purchaser or mortgagee for valuable consideration of the same [land]”. Where, however, an instrument does not affect land and is therefore not registrable, those provisions simply have no application[27]. 27. The significance of all this to the instant appeal lies in the fact that the assignment of the future proceeds of sale simply did not create an interest in or otherwise affect the Property. What Winland got under clause 9 of its loan agreement was merely an equitable chose in action in respect of the future proceeds but, importantly, it obtained no interest whatsoever in the Property itself – that is why the loan agreement was not registrable in the first place. 28. This is important because it means that so far as Tang’s beneficial interest in the Property was concerned, it was unencumbered by the assignment of the future proceeds in favour of Winland. So far as the Property was concerned, immediately before the making of the charging order, the only relevant encumbrance on the Property was the legal charge in favour of the FSI. 29. A charging order, on the other hand, is a creature of statute which affects land. Sections 20A(1) and (2) of the High Court Ordinance provide that “a charge may be imposed by a charging order” on an interest held by the debtor beneficially in land. Section 20B(3) goes on to provide that “a charge imposed by a charging order shall have the like effect and shall be enforceable in the same courts and in the same manner as an equitable charge created by the debtor by writing under his hand”. 30. Sections 20B(2) and (5) specifically provide for the registration and the vacating of the registration of a charging order in the Land Registry upon the making and discharge of a charging order respectively. 31. There can be no dispute that a charging order affects land and is registrable as such in the Land Registry[28]. It has like effect as an equitable charge, and is enforceable by a court order for sale of the land for the purpose of satisfying the outstanding judgment debt[29]. Contrary to what Ms Eu seemed to suggest at one stage during argument, a charging order does not simply affect the proceeds of sale of the land involved but also the land itself. In Yasaki International Co Ltd[30], Godfrey J (as he then was) rejected a similar argument : “Mr Yau then suggested, alternatively, that even if the assignment did not affect an interest in land, it was in no worse case than Yasaki’s charging order, which itself affected only the proceeds of sale of land. But I do not think this is correct. The effect of the charging order is prescribed by s20B(3) of the Supreme Court Ordinance (Cap 4). This provided that a charge imposed by a charging order shall have the like effect and shall be enforceable in the same courts and in the same manner as an equitable charge created by the debtor by writing under his hand. The charge so created would be a charge on the land itself, and not simply a charge over the proceeds of sale of that land.” 32. It is true that at common law, “a mortgage involves a transfer of legal or equitable ownership to the creditor, whereas an equitable charge does not”[31], and an equitable chargee “only gets a right to have the security [ie the charged property] made available by an order of the court”, but “gets no legal right of property, either absolute or special, or any legal right to possession” [32]. It is also true that an equitable charge only confers on the chargee “rights to apply to the court for an order for sale or for the appointment of a receiver [and thus to appropriate the rents and profits of the land], but no right to foreclosure (so as to make the property his own) or [to] take possession”[33]. Nonetheless, the availability of equitable remedies “has the effect of giving the chargee a proprietary interest by way of security in the property charged”[34], albeit that “the interest, though registrable against the chargor, remains inchoate and ineffectual until an order of the court is made”[35]. 33. It does not follow from the fact that the proprietary interest that an equitable chargee (and therefore a judgment creditor under a charging order) has in the charged property is “inchoate and ineffectual” until an order of the court is made for the sale of the property that an equitable charge or a charging order is no or little more than a mere charge over the proceeds of sale. It is one thing to be “inchoate and ineffectual” for the time being; it is another to say that it is not in existence at all as a proprietary interest in the property. Nor should one confuse the subject matter of the charge/security with its means of enforcement. Before the sale, an equitable charge or a charging order remains a proprietary interest in the property and is an encumbrance on it. 34. One consequence of this is that even if the chargee does not apply for a court order for sale to enforce the security, where the chargor/owner for whatever reason wants to sell the property, it can only be sold and a good title passed to the purchaser after or upon the discharge of the equitable charge or charging order as an encumbrance on the property, just like any other encumbrances on it. And where the proceeds of sale from the purchaser are the only source of money to pay off these encumbrances, the proceeds must first be so applied, which means that the chargee must be paid off first before any balance proceeds may be received by the chargor/owner or his assignee of the balance proceeds. 35. Analysed this way, Ms Eu’s attempt to downgrade a charging order to an equitable security over the proceeds of sale in order for her to apply the “first-in-time prevails” priority rule cannot be accepted. 36. In the present case, the charging order that Gain Hero obtained against Tang was directed at Tang’s beneficial interest in the Property. As explained, as at the time when the charging order was made, his interest in the Property was only subject to the legal charge in favour of FSI, but was otherwise unencumbered, as the earlier assignment of the future sale proceeds of the Property was not an assignment which created an interest in or otherwise affected the Property itself. This being the case, the charge created by the charging order, being a charge on the beneficial interest of Tang in the Property, was a charge over the Property subject only to the prior legal charge in favour of FSI but nothing else. It was not subject to the earlier assignment of the future sale proceeds. 37. Pausing here, it should be noted that the nemo dat maxim or the equitable principle of “first-in-time prevails” simply has no relevance to the legal analysis thus far. Tang’s beneficial interest in the Property was not in any way encumbered by the assignment of the future sale proceeds such that he could not pass an unencumbered proprietary interest (subject only to the legal charge) to Gain Hero under the charging order. Likewise, the subject matter of the earlier assignment of the future sale proceeds and the subsequent charging order over the Property being different, the equitable principle of “first-in-time prevails” simply has no application. 38. Moving on, when a charged property is sold pursuant to an order for sale, the property is of course converted into the proceeds of sale. Invariably, the order for sale would give directions for the distribution of the sale proceeds, which was what happened in the present case. Generally speaking, the court order should first provide for the discharge of all rent, taxes and rates and other outgoings due and affecting the property with the proceeds of sale, and then the discharge of all prior encumbrances on the property, that is, encumbrances which have priority over the charging order, as well as the costs and expenses of sale. The remaining proceeds will then be used to pay off the outstanding judgment debt and other monies covered by the charging order. If there is any surplus money, it will be used to repay the indebtedness of any other secured creditors whose securities on the property rank behind the charging order. If there is still any money left, it will go to the chargor/judgment debtor or his assignee of the proceeds[36]. Indeed, the order for sale in the present case gave similar directions for the distribution of the proceeds of sale. 39. As explained, the equitable assignment of the proceeds of sale under the Winland loan agreement did not give Winland any priority over Gain Hero’s charging order on the Property. Indeed, it did not give Winland any proprietary interest in the Property itself at all, and did not constitute an encumbrance on the Property. Winland’s entitlement to the proceeds of sale was derived from the judgment debtor/chargor (Tang)’s entitlement to the surplus proceeds, if any. It must therefore rank last. 40. In this regard, it should be noted that an assignment of the future proceeds of sale by an owner like Tang can only operate to assign the owner’s future entitlement to the proceeds of sale in favour of the assignee. The assignor cannot assign to the assignee what he is not entitled to receive by way of proceeds of sale upon the sale of the property. This is simply an application of the nemo dat maxim. Indeed, clause 9 of the Winland loan agreement in the present case only sought to assign to Winland Tang’s “right and interest … to and in the balance of the consideration or purchase money for the sale of the Property … in [his] capacity as owner, vendor …”, and did not purport to do otherwise. 41. All this means that in the present case, the proceeds of sale from the purchaser must, relevantly, go first towards discharging the outstanding judgment debt and other sums of money covered by the charging order, before any surplus can be released to Tang’s assignee to the proceeds of sale, that is, Winland. Analysed this way, in terms of “priority” to the proceeds, Gain Hero must rank before Winland. Disposition 42. For these reasons, the appeal was dismissed with costs. Mr Justice Ribeiro PJ: 43. I agree with the reasons for judgment of the Chief Justice. Mr Justice Bokhary NPJ: 44. I respectfully agree with the Chief Justice’s judgment, adding only my acknowledgment of the skill with which Ms Audrey Eu SC argued the equitable assignee’s case. No skill could overcome the difficulty that (i) the proceeds equitably assigned were those of a contemplated sale of real property free from encumbrances and that (ii) the charge imposed by the registrable and registered charging order obtained by the other side had become an encumbrance to be removed before arriving at any proceeds to which the equitable assignee could have recourse. Mr Justice Chan NPJ: 45. I agree with the reasons for judgment of the Chief Justice. Madam Justice McLachlin NPJ: 46. I agree with the reasons for judgment of the Chief Justice. Ms Audrey Eu SC, Mr Michael Yin and Mr Isaac Chan, instructed by Ho & Partners, for the plaintiff (appellant) Mr Jeevan Hingorani, Mr Lawrence Cheung and Ms Teresa Leung, instructed by HY Leung & Co LLP, for the defendant (respondent) [1] “Tang”. [2] “Property”. [3] “FSI”. [4] “Winland”. [5] “Gain Hero”. [6] “Winland loan agreement”. [7] HCA 465/2015. [8] HCMP 1608/2016. [9] HCMP 1980/2017 and HCMP 2671/2017. [10] [2019] HKCFI 771 (“CFI decision”). [11] Cap 128. [12] CFI decision, [38]-[55], [71]. [13] CFI decision, [67]. [14] CACV 186/2019. [15] [2021] HKCA 576, Lam VP, Yuen and Chu JJA, 27 April 2021 (“CA judgment”). [16] CA judgment, [27.1]. [17] Mr Michael Yin and Mr Isaac Chan with her. [18] Cap 4. [19] Snell’s Equity (34th ed), §4-002. [20] No one gives who possesses not. See Snell’s Equity, ibid. [21] Si Tou Choi Kam v Wealth Credit Ltd [2018] HKCA 250, [17]; Financial and Investment Services for Asia Ltd v Baik Wha International Trading Co Ltd [1985] HKLR 103, 110J-111G. [22] See CA judgment, [28.1]. [23] Secretary for Justice v Global Merchant Funding Ltd (2016) 19 HKCFAR 192, [37]-[39]; Tailby v Official Receiver (1888) 13 App Cas 523, 533; Hadlee v Commissioner of Inland Revenue [1991] 3 NZLR 517, 527‑528; In re Lind [1915] 2 Ch 345, 357, 363-366. [24] Re Yasaki International Co Ltd [1993] 1 HKC 349. [25] Si Tou Choi Kam, [13]. [26] Section 3 reads: “(1) Subject to this Ordinance, all such deeds, conveyances, and other instruments in writing, and judgments, made, executed, or obtained, and registered in pursuance hereof, shall have priority one over the other according to the priority of their respective dates of registration, which dates shall be determined in accordance with regulations made under this Ordinance. (2) All such deeds, conveyances, and other instruments in writing, and judgments, as last aforesaid, which are not registered shall, as against any subsequent bona fide purchaser or mortgagee for valuable consideration of the same parcels of ground, tenements, or premises, be absolutely null and void to all intents and purposes: Provided that nothing herein contained shall extend to bona fide leases at rack rent for any term not exceeding 3 years …” [27] This is very different from the type of situation, which is not the case here, where an interest in or affecting land is not created or evidenced by any instrument in writing but has arisen by operation of law, in which case there is no instrument to be registered under the Ordinance and the interest itself is often described as “not registrable”. Priority would be determined according to the normal rules under common law. Financial and Investment Services for Asia Ltd (unwritten equity arising from the discharge of a prior mortgage by a subsequent mortgagee) and Si Tou Choi Kam (resulting trust) are examples of this type of situation. [28] Re Yasaki International Co Ltd, 351H-I. [29] Order 50, rule 9A; Order 88, rule 5A, Rules of the High Court (Cap 4A). See generally Cousins on the Law of Mortgages (4th ed), §§19-01 to 19-11. [30] 351H-I. [31] In re Cosslett (Contractors) Ltd [1998] Ch 495, 508. [32] National Provincial and Union Bank of England v Charnley [1924] 1 KB 431, 450. [33] Carreras Rothmans Ltd v Freeman Mathews Treasure Ltd [1985] Ch 207, 227. [34] In re Charge Card Services Ltd [1987] Ch 150, 176. [35] Bland v Ingrams Estates Ltd [2001] Ch 767, [19]. [36] Section 54 of the Conveyancing and Property Ordinance (Cap 219) provides that upon the sale of the mortgaged land, the proceeds of sale shall be applied according to the following priority:- “(a) in discharge of all rent, taxes, rates and other outgoings due and affecting the mortgaged land; (b) unless the mortgaged land is sold subject to a prior incumbrance, in discharge of that prior incumbrance; (c) in payment of the receiver’s lawful remuneration, costs, charges and expenses and all lawful costs and expenses properly incurred in the sale or other dealing; (d) in payment of mortgage money, interest and costs due under the mortgage, and any residue shall be paid to the person who, immediately before any sale or other dealing, was entitled to the mortgaged land or authorized to give a receipt for the proceeds of the sale of that land.” “Mortgage” under the Ordinance is widely defined to mean “a security over land for securing money or money’s worth”: section 2. 
Chief Justice Ma: 1. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Ribeiro PJ: 2. This appeal is brought by the Employees Compensation Assistance Fund Board (“the Board”) raising two questions which are of great practical importance to its operations. The first is whether the courts have jurisdiction to make costs orders against the Board in certain proceedings to which it is a party. The second is whether the Board is empowered to enter into binding settlements regarding claims or potential claims for payments out of the Fund which it administers. A. The Board and the Fund 3. The Board was established in 1991 by the Employees Compensation Assistance Ordinance (“ECAO”)[1] to provide relief for employees and other eligible persons who are entitled to compensation under the Employees’ Compensation Ordinance[2] (“ECO”) or to common law damages in respect of employment-related injuries, but who are unable to make recovery against the employer or any insurer. 4. Payments to those eligible for relief are made out of a Fund established under the ECAO[3] which the Board holds upon trust to administer in accordance with the objects of the Ordinance.[4] The ECO[5] makes it compulsory for employers to take out insurance policies for indemnity against liability to their employees for injury or death by accident arising out of and in the course of their employment. Upon payment of each premium, a statutory levy is imposed[6] and more than half of the levies collected are distributed to the Board[7] for application pursuant to the ECAO.[8] The resources of the Fund derive largely from such insurance levies. 5. An employee who is injured at work may bring proceedings against the employer for compensation under the ECO or for damages at common law, or commonly, for both. The Board has a statutory contingent liability which is likely to become actual in cases where the employer has no substantial means and no insurance cover, whether or not the claim is contested. Upon becoming aware of such a case, the Board may wish to take steps to protect the interests of the Fund. This appeal examines what the Board can properly do to that end and what consequences may follow from the steps which it takes. In particular, where the Board decides to intervene in the proceedings between employee and employer, for instance, to ensure that the claim is not collusive or extravagant and is properly defended, the issue arises as to what the costs consequences of such intervention may be. Or if the Board thinks that seeking an early settlement of a potential claim against the Fund is in its best interests, the question arises as to whether the law permits it to contract a binding settlement. Answers to these questions are matters of statutory construction, as we shall see. B. The present claim and the decisions below 6. The plaintiff was an interior decoration worker who sustained injuries at work. The 2nd defendant had been engaged by the 1st defendant to do the construction work. The plaintiff alleged that the 2nd defendant was his employer but brought proceedings against the defendants for compensation under the ECO and for common law damages. The case was discontinued against the 1st defendant in 2015 when he settled the plaintiff’s claims by payment of $80,000. 7. That left the plaintiff’s claims, which included one for over $4.7 million damages, against the 2nd defendant who had no insurance cover and who acted in person throughout. In these circumstances, the Board obtained leave to be joined as 3rd defendant in the proceedings, stating that it wished to participate in the assessment of damages. It proceeded in 2016 to commission a medical report jointly with the plaintiff and to file an answer to the plaintiff’s statement of damages, putting forward a lower assessment. 8. The plaintiff obtained judgment against the 2nd defendant for $602,380 before Deputy District Judge Chow[9] in the ECO proceedings in September 2016. The common law action then came on for trial in May 2017 and at its commencement, the plaintiff and the Board agreed to settle his potential claim under the ECAO for $1.42 million.[10] The 2nd defendant was not party to this settlement and the trial proceeded against him before Deputy High Court Judge To, resulting in an award of damages in the sum of $2,110,927. Deducting the sums of $80,000 received from the 1st defendant and the ECO award of $602,380, the net amount of damages due to the plaintiff from the 2nd defendant was $1,428,547, which was marginally more than the settlement amount agreed with the Board. The Judge ordered the 2nd defendant to pay the plaintiff’s costs but refused the plaintiff’s application for costs against the Board as from the date of its joinder in the action, directing that there should be no order as to costs as between the plaintiff and the Board.[11] 9. The plaintiff sought and obtained leave from the Court of Appeal to appeal against the direction of no order as to costs. He sought to contend that the Judge should have adopted the approach in Kwan Kam Pui v Fung Man,[12] where Bharwaney J suggested certain “usual orders” that ought to be made against the Board in cases where it had intervened in proceedings which resulted in awards in favour of the plaintiff.[13] By a respondent’s notice, the Board sought to uphold the Judge’s order on the additional ground that the court lacked jurisdiction to make any costs orders against it. The plaintiff’s appeal was dismissed on the ground that there was a wide discretion as to costs and that there was no basis for interfering with the order made. 10. Regarding the challenge to jurisdiction, in Jiang Zhong v Yeung Chun Leung,[14] the Court of Appeal had held that the court does have jurisdiction to make costs orders against the Board where it is joined as a party. Although the Court of Appeal in the present case[15] said that it was unnecessary to determine the issue of jurisdiction, having dismissed the plaintiff’s appeal on the merits, it nevertheless endorsed the decision in Jiang Zhong, rejecting the Board’s submission that that decision was either obiter or plainly wrong. 11. Relying on section 29 of the ECAO,[16] the Court of Appeal also held that it was open to the Board to enter into a settlement agreement in relation to any issue in the proceedings, rejecting the Board’s submission to the contrary.[17] 12. The Board was unsuccessful before the Court of Appeal in its application for leave to appeal to this Court.[18] Leave was, however, granted by the Appeal Committee,[19] confined to the following questions of law: Question 1 Whether, upon proper construction of the provisions of the [ECAO], and in particular s.20B(3) thereof, the court has jurisdiction to order costs against the Board in common law damages claims to which the Board joins in pursuant to s.25A of the ECAO (“Joinder Cases”), and if so, whether there is a “usual order” or “starting point” on costs to be made or ordered in Joinder Cases where the Board (i) disputes liability and quantum, and (ii) disputes quantum only (See [Court of Appeal] Judgment §§14, 20, 32 and 34). Question 2 Whether, in view of the fact that a plaintiff is entitled under s.20B(l) of the ECAO to an amount of relief payment which “shall be the amount of damages for which the employer is liable to pay to the eligible person”, the Board, as a statutory body constituted under ECAO, has power to settle with a plaintiff on the quantum of relief payment payable by the Board before such amount of damages for which the employer is liable is known, i.e. before trial or assessment of damages, and if so: (a) Whether the trial judge was correct that should the quantum of damages assessed against the employer be higher than the agreed amount, the plaintiff cannot apply to the Board for relief payment in respect of the difference, and should the quantum of damages assessed against the employer be lower than the amount agreed between the plaintiff and the Board, the Board may have no authority to pay the plaintiff the higher amount that has been agreed (See §§40-41 of the [Court of Appeal] Judgment and §25 of the judgment of trial judge dated 23 June 2017). (b) Whether the Board has power under the ECAO to pay the agreed relief payment to the plaintiff in the absence of an application under s.20A of the ECAO (Cf. ss.28(1) and 30 of the ECAO in relation to compensation claims). (c) If not, what is the legal effect of such settlement between the plaintiff and the Board? C. Question 1 C.1 Jurisdiction and ECAO section 20B(3) 13. ECAO section 20B(3) states: “Notwithstanding the definitions of ‘compensation’ and ‘damages’, [20] for the purposes of determining the amount of a relief payment— (a) any interest payable on any amount; and (b) any costs, arising from proceedings in respect of any damages or compensation claim shall not be included.” 14. Ms Audrey Eu SC,[21] relies on this provision for the submission that, by excluding any costs from being part of the relief payment, it deprives the court of jurisdiction to award costs against the Board in all damages claims. 15. That submission cannot be accepted. Section 20B(3) must be construed in the context of sections 20A and 20B read together. Section 20A materially provides: (1) An eligible person who is unable to recover from an employer payment of an amount of damages for which the employer is liable may apply for a relief payment of that amount from the Fund. (2) For the purposes of this section, an employer is not to be regarded as liable for the payment of an amount of damages unless the amount is payable pursuant to a judgment or order of a court of competent jurisdiction in Hong Kong. (3) For the purpose of this section, an eligible person is not to be regarded as being unable to recover from an employer payment of an amount of damages for which the employer is liable unless the eligible person has taken such proceedings to recover payment of the amount— (a) from, wherever applicable— (i) the employer; and (ii) an insurer who has issued a policy of insurance— (A) which is in force in relation to the injured employee at the time of the accident concerned; and (B) indemnifying the employer in respect of such liability; and (b) as are reasonable in the circumstances, having regard to the likely cost of such proceedings, the resources available to the eligible person and the amount likely to be recovered from, wherever applicable, the employer and the insurer.” ... 16. Section 20A therefore lays down the conditions for claiming a relief payment arising out of an unsatisfied judgment for common law damages. Those conditions principally involve obtaining a judgment and having unsuccessfully taken reasonable proceedings to secure payment. Section 20B then goes on to prescribe the amount of the relief payment claimable, stating that it shall be the amount of damages which the employer is liable to pay, reduced by any ECO compensation paid or payable and by any amount of the damages already paid.[22] 17. It is in that context that section 20B(3) operates. Like sections 20B(1) and (2), that subsection is concerned with circumscribing the quantum of the relief payment. It specifies that the relief payment is not to include any sum of interest or costs awarded to the employee against the employer. It does not address or seek to limit the court’s power to order costs against the Board. 18. That this is the effect of section 20B(3) appears clearly in cases where the employee applies to the Board for a relief payment arising out of a damages award obtained in proceedings to which the Board was never a party. As we shall see, the ECAO caters both for such cases and for cases in which the Board was joined as a party to the action between employee and employer. Leaving the joinder cases aside for now, when section 20B(3) is applied where the Board played no part at all in the underlying proceedings, there is obviously no question of any costs order being made for or against the Board since it was not a party to any relevant proceedings. It merely receives an application for relief from an employee who has independently taken the case against the employer to judgment and failed to make recovery. In such cases, section 20B(3) clearly functions merely to limit the amount of the relief payment claimable by the employee. It has nothing to do with the court’s jurisdiction to make costs orders against the Board. C.2 Jurisdiction and cases where the Board is joined as a party 19. By section 25A,[23] the ECAO authorises the Board to apply in certain circumstances to be joined as a party to existing proceedings brought against the employer for compensation or damages: “If proceedings have been initiated to claim compensation or damages ..., the Board may apply to the court to be joined as a party to the proceedings as follows— (a) where no policy of insurance is known to be in force at the time of the accident to which the proceedings relate, the Board may apply to the court to join in the proceedings as a party in accordance with Order 15, rule 6 of the Rules of the High Court ... or Order 15, rule 6 of the Rules of the District Court ..., as the case may require, to take over the defence as if it were the employer in the proceedings if— (i) the employer cannot be identified or, if identified, cannot be found; (ii) the employer is insolvent; (iii) the employer is dead or has been dissolved, wound up or struck off the register; (iv) the employer, for any reason, cannot be served with the notice of proceedings; or (v) at any time when the employer fails to attend the hearing, leaving the claim uncontested; (b) where no policy of insurance is known to be in force at the time of the accident to which the proceedings relate and the employer is present at the proceedings, the Board may apply to the court to join in the proceedings as a party in accordance with Order 15, rule 6 of the Rules of the High Court ... or Order 15, rule 6 of the Rules of the District Court ..., as the case may require; (c) where the insurer concerned is insolvent, the Board may apply to the court to join in the proceedings as a party in accordance with Order 15, rule 6 of the Rules of the High Court ... or Order 15, rule 6 of the Rules of the District Court ..., as the case may require.” 20. Where the Board does become a party to the proceedings, the occasion may obviously arise for considering whether costs incurred in the conduct of those proceedings should be awarded whether in favour of or against the Board. 21. The ECAO distinguishes between the Board’s participation in claims for ECO compensation as against participation in actions for damages; and, as appears from section 25A, between situations where the defendant employer is or is not a participant in those proceedings. 22. Thus, ECAO section 25(2) authorises the employee to issue an ECO claim against the Board “as if it were the employer” where the employer is, for one of the specified reasons, unavailable for suit[24] and where there is no applicable insurance policy in force. The Board is thus evidently to be named as the defendant in such cases and section 25(4) provides that the rights and liabilities of the employer in respect of the compensation are vested in and transferred to the Board. 23. But there is no similar provision for an action for damages to be instituted against the Board. Only where proceedings have already been issued by an employee claiming damages (or ECO compensation) against the employer does section 25A enable the Board to apply to the Court in accordance with Order 15 rule 6 of the Rules of the High Court (“RHC”) to be joined as a party to those proceedings. As set out above, section 25A specifies three sets of circumstances in which there might be such joinder which are evidently aimed at permitting the Board’s intervention in two main situations. First, under section 25A(a), the purpose of the Board’s joinder is “to take over the defence as if it were the employer in the proceedings” in circumstances where the employer is effectively not present in, or not defending, the proceedings. Secondly, under paragraphs (b) and (c), while the employer is present, there is no insurance cover so that it is likely that recourse will be had to the Fund. 24. Unlike section 25 (dealing with ECO claims), section 25A(a) is not explicit about the consequences of joinder. It does not, for instance, expressly provide for the rights and liabilities of the employer to be vested in the Board in respect of the action for damages.[25] Nevertheless, in authorising the Board to take over the defence “as if it were the employer”, the legislature must have envisaged the Board being able actively to test, limit or resist the employee’s claim regarding liability and/or quantum to ensure that the Fund’s resources are properly applied. 25. The consequences of the Board joining the proceedings under section 25A(b) and (c) are even less explicitly spelt out. However, it is a compelling inference that the legislative intent was to enable the Board to be joined as a party to keep an eye on the proceedings in which the employer is still participating, given the Fund’s contingent liability to make relief payments tied to the level of judgments obtained by the employee. Whether, in such circumstances, the Board takes an active role defending against the employee’s claim will obviously depend on the circumstances. If the claim is being properly defended by the employer, the Board might be content essentially to maintain a watching brief and no costs consequences are likely to follow. On the other hand, if the claim is extravagant or collusive or if the employer is not mounting a proper defence, the Board may wish actively to participate in the proceedings to mount an effective defence,[26] in which case there may be consequences as to costs arising out of the conduct of the proceedings. 26. In Section C.1 above, I have concluded that ECAO section 20B(3) does not exclude the court’s jurisdiction to order costs for or against the Board. And in the foregoing discussion, the possible relevance of costs orders where the Board takes an active role in proceedings has been pointed out. But does the Court have power to make such orders as to costs? 27. The answer is plainly in the affirmative. Section 52A(1) of the High Court Ordinance[27] confers a general power on the court to make costs orders in all civil proceedings: “Subject to the provisions of rules of court, the costs of and incidental to all proceedings in the Court of Appeal in its civil jurisdiction and in the Court of First Instance, including the administration of estates and trusts, shall be in the discretion of the Court, and the Court shall have full power to determine by whom and to what extent the costs are to be paid.” 28. Since, pursuant to ECAO section 25A, the Board is made a party to the proceedings in accordance with RHC Order 15 rule 6, the power to order costs conferred by section 52A clearly extends to possible orders against the Board. Indeed, section 52A(2) enables costs to be ordered even against a non-party[28] so that, a fortiori, costs can be ordered against a person joined as a party under RHC Order 15 rule 6, whether or not judgment in the action is entered against such person. If the general power conferred by section 52A is to be excluded, one would expect this to be done by some clear provision. There is nothing in the ECAO to such effect. C.3 Jurisdiction and ECAO section 29 29. Moreover, ECAO section 29 offers a clear positive indication that costs may be ordered against the Board. It provides as follows: “(1) Notwithstanding any other law where, in proceedings to which the Board is a party under this Part, the Board makes a written offer to any other party in the proceedings which is expressed to be ‘without prejudice save as to costs’ and which relates to an issue in the proceedings, then in deciding the question of costs the court shall take into account any such offer which has been brought to its attention. (2) An offer made under subsection (1) shall not be communicated to the court until the question of costs falls to be decided.” 30. Thus, where the Board has intervened in an action, section 29 empowers it to make an offer to another party – who will, in the nature of things, generally be the plaintiff employee – placing that party at risk as to costs if the offer is not accepted. It is implicit that acceptance of the offer would save costs on both sides, being an offer to settle the claim or some time-consuming issue in the claim, so that a refusal of the offer, if followed by an outcome in which the plaintiff fails to improve upon what is on offer, may be seen as unreasonable and lead to adverse costs consequences. 31. The underlying premise of section 29 is therefore that costs are in play in the proceedings. In practice, the risk which the plaintiff faces is that the court, when it comes to take the offer into account, may deprive him or her of costs against the Board as from the time of the making of the offer. This necessarily implies potential liability of the Board for costs in connection with the continued conduct of the proceedings which section 29 empowers it to guard against. 32. Whilst it is conceivable that refusal of the offer might lead to costs being ordered against the plaintiff, this may in practice be unlikely, not only because of the injured employee’s probable impecuniosity, but also because the Board’s role in joining as a party is to monitor the proceedings to ensure that relief is appropriately granted, rather than to adopt an excessively adversarial stance against the injured employee. In the normal course, the costs of the Board’s participation in proceedings are treated as part of its operating expenses[29] and it will generally not seek to fund such participation by depriving injured employees of a significant portion of an appropriately quantified relief payment. 33. The court of course retains a wide discretion and, in exceptional cases, if the employee’s conduct viewed overall has been egregious, such as where the claim is fraudulent or seriously extravagant, it may result in an order for costs in the Board’s favour. Section 29 is nevertheless principally about depriving the plaintiff of litigation costs which he or she would otherwise have recovered as against the Board, clearly indicating the existence of jurisdiction to make such costs orders. 34. Ms Eu’s only response to section 29 was to argue that it somehow does not apply to cases in which the Board has applied to join as a party pursuant to ECAO section 25A. I have to say that I find that argument incomprehensible. Section 29 applies “where, in proceedings to which the Board is a party under this Part, the Board makes a written offer (etc)”. Where the Board is joined under section 25A, it is “a party under this Part”, that is, Part IV of the ECAO. It is unarguable that section 29 does not apply to such joinder cases and was rightly “firmly rejected” in the Court of Appeal.[30] 35. That jurisdiction to order such costs exists has been recognised by the Court of Appeal in Jiang Zhong v Yeung Chun Leung,[31] applying an analysis similar to that developed above. C.4 Discretion in ordering costs against the Board 36. In Kwan Kam Pui v Fung Man,[32] Bharwaney J considered how the discretion to order costs against the Board should be approached and decided that a “costs to follow the event” order should be usual, equating the Board with the employer vis-à-vis the employee. His Lordship put it thus: “In my judgment, the usual order should be made against the ECA Fund Board, as is made against an employer defending the proceedings against whom a judgment for damages has been obtained. Such usual order is to pay the entire costs of the action, not just costs from the date of the intervention by the ECA Fund Board, in cases in which the ECA Fund Board has intervened and disputed liability and quantum or has obtained an order to set aside interlocutory judgment for damages to be assessed in order to dispute liability and quantum. The taxation of such costs would ensure that any costs and expenses which were unreasonably incurred would not be allowed on the taxation. The costs of necessary steps taken to prosecute a claim to judgment ought to be recoverable in a case in which the ECA Fund Board intervenes to dispute liability, even though those steps were taken before its intervention. ... Where the ECA Fund Board intervenes in order to dispute the injured employee’s claim on quantum only, then the usual order ought to be the one made by Master Lung, as he then was, in Lau Tuen Ping v Law Wai Kwong & Anor,[33] namely, that the ECA Fund Board ought to pay the costs of the proceedings from the date of its joinder.”[34] 37. At first instance below, Deputy High Court Judge To expressed broad agreement with Bharwaney J although noting that his “usual orders” were merely a starting-point and not invariable.[35] Although he acknowledged that the Board was not like an ordinary litigant and played a “filtering role” to screen out unmeritorious or inappropriate claims,[36] he agreed that where the Board disputes liability it would generally be appropriate that costs should follow the event so that the Board should pay all the costs, including those incurred before the date of joinder, if the employee succeeds in his claim.[37] 38. However, where the Board intervened only on the issue of quantum, his Lordship held that the starting-point should be that the Board pays the employee’s costs from the date of joinder unless the claim was “outside the range of reasonable dispute”. This would prima facie arise “if the amount of damages claimed by the employee exceeds that awarded by the court by one-third” in which case the Board should not have to pay the employee’s costs.[38] He also held that that starting position might be modified in the light of untoward behaviour by either party in the conduct of the proceedings: “Where the Board only plays a filtering role, the proper starting point remains that the Board shall pay the employee’s costs, but the court may, in light of the attitudes of both parties to the proceedings and the difference between the amount of damages awarded by the court and the amount claimed by the employee, make such costs orders as it thinks fit. The factors to be considered by the court include: whether the Board has only assumed a filtering role or has assumed an antagonistic role, whether it has adopted an unnecessarily antagonistic attitude, wasted time, raised unnecessary, frivolous or technical issues, etc; whether the employee has inflated the claim, adduced false evidence, misled the court, misled the Board, made an act or omission which caused the Board to raise unnecessary issues. Where the amount claimed by the employee is outside the range of reasonable dispute, the proper starting point in considering costs is that both parties shall bear their own costs. In appropriate cases, the court may even order the employee to pay part or all of the Board’s costs.”[39] 39. In the present case, the plaintiff claimed $4,871,113.20 but was awarded only $2,110,927 so that the claim went “way beyond the range of reasonable difference”.[40] This resulted in Deputy High Court Judge To making no order as to costs so that each party was to bear its own costs.[41] 40. In the Court of Appeal, the plaintiff sought to argue that Deputy High Court Judge To was wrong not to apply Bharwaney J’s approach and thus not to direct, as a “usual order”, that the Board pay the employee’s costs as from the date of joinder, challenging the Judge’s use instead of the one-third measure.[42] 41. This was rejected by Kwan JA writing for the Court of Appeal. Her Ladyship regarded Bharwaney J’s “usual orders” as merely the starting-point in the exercise of discretion.[43] She held that Deputy High Court Judge To’s one-third measure was avowedly “at random” and not decisive,[44] his Lordship having in fact duly taken into account the conduct of the parties in accordance with RHC Order 62 rule 5(2), including: “(a) whether it was reasonable for a party to raise, pursue or contest a particular allegation or issue; (b) the manner in which a party has pursued or defended his case or a particular allegation or issue; (c) whether a claimant who has succeeded in his claim, in whole or in part, exaggerated his claim; and (d) conduct before, as well as during, the proceedings.”[45] 42. Kwan JA concluded that the Judge’s exercise of discretion had not miscarried and that there was no ground for interfering with it.[46] 43. In my view, the court should not adopt as its starting-point the “costs following the event” criterion and should not equate the Board with an employer conducting adversarial proceedings against the employee. The Board is a statutory authority holding the Fund as a trustee and required to consider applications for payment or relief.[47] The Ordinance plainly requires the Board to be proactive in ensuring that the Fund’s resources are properly applied and not subject to abusive or unjustified claims. Thus section 25B requires potential claimants on the Fund to give the Board notice of proceedings instituted for compensation or damages. Section 25B(7) provides that on receiving notice, the Board “shall carry out such inquiries as it considers necessary to facilitate the proper carrying out of the functions of the Board in relation to the claim to which the notice relates”. As we have seen, section 25A enables the Board to apply for joinder as a defendant in the circumstances discussed in Section C.2 above. Where the employer is not present or not interested in defending, etc, joinder is to allow the Board “to take over the defence as if it were the employer in the proceedings”.[48] And where the employer is present but there is no insurance cover so that the Fund is likely to be drawn on, the Board is by necessary implication, authorised, if necessary, to do likewise.[49] 44. Thus, the Board is authorised and required to scrutinise and, if thought necessary, to intervene in order to test the validity of the plaintiff’s case whether on liability, quantum or both, in carrying out its statutory duties. Where an employee obtains an award of damages after such scrutiny or intervention it does not mean that the Board has “lost” the case so as to constitute “the event” justifying an award of costs against it. The “event” properly viewed, is the outcome of the plaintiff’s action against the employer (whether or not actually present) and not the process involving the Board’s intervention. 45. The Board’s role was properly recognised by the trial judge: “In participating in the proceedings, the Board’s function is neither to contest the employee’s claim in place of the employer nor to help save public funds, but is to ensure that employees in need can obtain such damages as they are entitled to. The Board has a duty to screen out unmeritorious or inappropriate claims to ensure (inter alia) that this assistance mechanism will not be abused, its recipients will not exaggerate their claims, and public funds will not be used for improper purposes. Therefore, unless it is necessary to contest false claims or suspected fraudulent claims, the Board should not adopt an antagonistic attitude in the proceedings. It should act as a filter and adopt an impartial attitude in the proceedings so as to assist the court in coming to a correct ruling.”[50] 46. In my view, such recognition dictates that where the Board has been joined as a party and properly carries out its “filtering” or monitoring functions, the appropriate starting-point should generally be no order as to costs, irrespective of whether the Board sought to test the case as to liability, quantum or both. 47. That is the starting-point and a different costs order might be warranted on particular facts. One naturally expects that, having intervened, the Board will behave in a responsible manner but if, in what will hopefully be a rare case, its conduct is unreasonable or misconceived or unjustifiably antagonistic, unnecessarily prolonging its intervention, or otherwise untoward, the court may, in the exercise of its discretion, consider ordering the Board to pay the plaintiff’s costs. Factors such as those referred to in RHC Order 62 rule 5(2) set out above would be relevant. Where such considerations may arise, ECAO section 29 offers are especially relevant and will be taken into account.[51] It must, however, always be kept in mind that the discretion to order costs is broad and to be flexibly exercised in the light of the specific circumstances. D. Question 2 48. The ECAO does not explicitly deal with the Board settling potential relief payment claims based on actions for damages. In contrast, section 28(1) expressly authorises settlement of potential ECO-based claims under section 16: “Where it appears to the Board that a person is or might be entitled to apply under section 16 for a payment from the Fund, the Board may, in the interests of the protection of the Fund, and on such terms as it sees fit, offer to pay from the Fund to or on behalf of that person an amount in satisfaction of such entitlement.” 49. Section 28 then goes on to prescribe the consequences where a person accepts[52] or rejects[53] the offer and, in a joinder case, authorises the Board to agree the costs without going to taxation. No equivalent section deals with settling cases based on damages claims. 50. Ms Eu relies on the inapplicability of section 28 to damages claims, making the point that prior to the 2002 amendments to the ECAO, the previous section 28 did apply to damages claims, but that power had been expressly removed by the amendment exercise. D.1 The desirability of the Board having power to settle potential claims 51. As a matter of policy, it is obviously desirable that the Board should have power to settle potential relief payment claims just as it has power to settle ECO-based compensation claims. If no such power exists, the misgivings expressed by Bharwaney J in Lau Yuk Hung v Tsang Kwong Ming,[54] and repeated in Kwan Kam Pui v Fung Man,[55] would have even greater force. His Lordship was concerned about: “... the current state of affairs ... under which injured persons, who are unable to recover payment from their employer are required, nevertheless, to prove their claim for common law damages in court, and incur the costs of proving their case in court, before they can claim payment from the [Board] but who, when they claim payment, after obtaining such judgment, are told that the [Board] is not liable to pay interest on the damages that had been awarded and, more surprisingly, that the [Board] is not liable to pay the costs incurred in proving their case in court and obtaining that judgment against the employer. It is a catch 22 situation: the injured employee cannot get payment from the [Board] unless he obtains a judgment from court but he would not be reimbursed for the costs incurred in obtaining that judgment. The net result is that any damages that are awarded are reduced by the amount of the costs incurred to obtain that judgment, which can run into a few hundred thousand dollars.” 52. He pointed out that in the Kwan case, the plaintiff’s taxed costs estimated at $500,000 would have to come out of the net recovery amount of HK$800,000, leaving the plaintiff with only $300,000, which was less than the amount of costs incurred in bringing the proceeding to a conclusion.[56] 53. An important object of the ECAO is obviously to grant financial relief to injured employees without the relief payments being eaten up by legal costs. The problem adverted to by Bharwaney J would be exacerbated if there was no power to settle. It would also mean that where the Board became a party, it would in effect have to remain for the entire proceedings, thereby increasing costs. The need to encourage early settlements to save legal costs was acknowledged at the meeting of the LegCo on 17 May 2001 both by a LegCo member and the government official proposing the legislation.[57] D.2 The issues bearing on the existence of a power to settle 54. Two related issues which principally bear on whether the Board has power to settle potential damages-based relief payment claims appear to me to be as follows: (a) The Board is in the position of a trustee and is not free to apply the Fund save in accordance with the terms of the ECAO. So, given the contrast between the way settlements of ECO-based compensation claims are dealt with by section 28 and the absence of express provision regarding claims based on common law actions, are there any provisions in the ECAO giving the Board power to settle and to disburse amounts from the Fund by way of settlement? (“the first issue”) (b) If there is a power to settle, can the Board contract a settlement without the amount to be paid out being determined by a judgment and assessment of damages against the employer as provided for by ECAO sections 20A and 20B? If so, how is the settlement amount to be determined? (“the second issue”) D.3 Settlement and section 29 55. The resolution of each of those issues depends on ECAO section 29 which has been set out in Section C.3 above. It provides that where the Board has intervened in an action, section 29 empowers it to make an offer to another party, generally the plaintiff employee, to settle any issue in the proceedings (which must include the ability to settle the entire claim), so that a refusal to settle, if proven to have been unreasonable (as by the plaintiff achieving only a lower award and relief payment), may lead to adverse costs consequences. 56. Section 29 is therefore a measure expressly aimed at cutting down legal costs. It is wholly aligned with the policy objective of encouraging settlements so that costs are reduced and erosion of relief payments minimised. In conferring on the Board a broad power to make such offers, section 29 expressly, or at least by necessary implication, empowers it to enter into settlement agreements with other parties to proceedings in which it has intervened. Such settlements are therefore intra vires the Board. So is any payment out to the eligible person of an agreed settlement amount[58] since ECAO section 8(d) materially provides that “[the] Board may pay from the Fund ... any ... sums ... permitted to be paid by the Board under this Ordinance.” The answer to the first issue raised above is therefore in the affirmative. 57. The deletion of a power previously in section 28 to settle is neither here nor there. It is section 29 that provides a legal basis for settlement. It is applicable in the context of section 25A, enacted in the 2002 amendments, expressly permitting the Board to be joined as a party to the proceedings. D.4 The quantum of a settlement offer 58. Given that the Board has power to settle claims, the second issue mentioned above poses the question whether there are any constraints as to the manner in which, and the amount for which, it may settle. This is reflected in Question 2 which asks whether any power to settle may be constrained (i) by the reference in section 20B(1) to the amount of relief payable being “the amount of damages for which the employer is liable to pay to the eligible person [subject to deductions]”; (ii) by such amount having to be determined at trial and on an assessment of damages; (iii) by the need for an application for relief under section 20A which requires the applicant to have secured judgment and to have taken reasonable proceedings to recover payment.[59] 59. It should at once be pointed out that if, although having power to settle, the Board is constrained not to make a settlement agreement unless and until quantum is determined at trial and after an assessment of damages, the important object of facilitating early settlements to avoid running up legal costs would be entirely defeated. The injured employee would be forced to navigate through the statutory machinery, toiling on until the bitter end to secure a judgment and an assessment of damages, incurring costs which would eat significantly into the settlement amount. 60. Section 29 provides the means to avoid such a counter-productive position. Its central premise is that the offer is made, and hence that the claim may be settled, before the proceedings reach the stage of trial and judgment. The possible “prejudice as to costs” only arises if the offer is rejected and the employee presses on with the case after the date of the offer. Section 29 makes no sense otherwise. Accordingly, if an offer to settle is accepted, it is necessarily implicit that a contractually binding agreement comes into being between the Board and the eligible person which disposes of that claim, making it no longer necessary to operate the statutory machinery set out in sections 20A, 20B and so forth. Settlement is, in other words, an alternative route to obtaining the relief provided by the ECAO and disposing of the employee’s claim or potential claim against the Fund. 61. How does the Board decide the amount at which it should settle? As with all settlements arrived at before the court’s award is made, this must depend on the parties’ assessment of the likely range of the award. If the settlement mechanism provided by section 29 is to work, the Board must have power to make an offer based on its assessment of what is a reasonable sum, arrived at in good faith on available information it considers sufficient. It is equipped by the ECAO to secure relevant information. As already noted, section 25B(7) requires the Board to “carry out such inquiries as it considers necessary to facilitate the proper carrying out of the functions of the Board in relation to the claim to which the notice relates”. Section 25B(8) gives it compulsory powers[60] to obtain any required information: “For the purpose of carrying out inquiries under this section, the Board may— (a) require the employer ... the employer’s insurer or any other employee of the employer to furnish such information or particulars relating to the claim; and (b) make such inquiries from any other person connected or associated with the claim, as the Board considers necessary. 62. How much information the Board acquires will differ from case to case and its adequacy may depend on the particular facts and the nature of the defendant’s allegations. If the Board has decided to intervene pursuant to section 25A, it will have ascertained that the defendant employer is uninsured and unlikely to be good for any judgment obtained. The Board may be in a good position, for the purposes of quantum, to assess the employee’s injuries by obtaining medical reports (often jointly with the employee, as happened in the present case). Documentary evidence might be available to throw light on certain liability issues, for instance, where the defendant denies being the employer or where the employee’s earnings are disputed. Evidence of how the accident happened might be available from witnesses bearing on issues of negligence, causation, contributory negligence and the like. In many cases, having made inquiries, one would expect the Board to feel equipped to propose a reasonable sum for settlement. 63. Of course, not every case will be a candidate for settlement. The Board may not feel that it has enough facts and may decide that the case should be left to go to trial and assessment. Or the employee may hold out for a sum which is regarded as extravagant, notwithstanding that a section 29 offer has been made by the Board. 64. The point, however, is that settlement, which is obviously valuable as a possible means of cutting legal costs and making the best use of the Fund’s resources is, on the true construction of the ECAO, an option open to the Board. Its functions, laid down by section 4, include administering the Fund in accordance with the objects of the Ordinance[61] and it has power to “do all such things as are expedient for or conducive to the attainment of the functions referred to in section 4 or which, in the opinion of the Board, are necessary to facilitate the proper carrying out of the functions of the Board”.[62] Once it is accepted that it is empowered by section 29 to settle claims in properly carrying out its functions, it is plainly intra vires the Board to enter into agreements to settle for sums based on its assessment of what is reasonable, arrived at in good faith on available evidence considered sufficient, having made due inquiries. D.5 The legal effect of a settlement agreement 65. Questions 2(a) and (c) address the legal effect of a settlement. In particular, they ask what the position is where the Board and the eligible person settle for a stated amount, where the case proceeds to judgment and where the award made against the employer is either higher or lower than the settlement amount. 66. Deputy High Court Judge To took the following view: “Should the quantum of damages assessed by me be higher than the agreed amount, the Plaintiff cannot apply to the Board for relief in respect of the difference. Should the quantum of damages assessed by me be lower than the agreed amount, the Board may also have no authority to pay the Plaintiff an amount higher than what has been agreed.”[63] [It seems likely that the last word should be “awarded” rather than “agreed”] 67. The Court of Appeal left open the question whether the Board has power to pay a settlement sum in excess of the eventual damages award.[64] But it accepted that there is nothing to prevent an eligible person from agreeing to accept a settlement amount that turns out to be less than the award against the employer.[65] 68. I pause to observe that it seems odd that an action against the employer should be pressed on to judgment, incurring legal costs, after the eligible person has settled the potential relief payment with the Board. Ex hypothesi, the Board has only applied to join in the proceedings under section 25A because there is no insurance cover and the employer is either effectively absent or presumptively not good for the award, exposing the Board to likely liability under the ECAO. So why would the plaintiff choose to pursue the employer after settling with the Board? 69. We were informed by Mr Raymond Leung SC[66] that in the present case, the settlement was apparently as to quantum, without agreeing liability. Why this course was taken was not explained but, as appears from the submissions of Ms Eu, it may have been due to a belief that it is necessary for the plaintiff to secure a judgment and to make a section 20A application to bring himself within section 20B(1) in order to qualify for payment of the settlement sum. That view appears to have been shared by the Court of Appeal who commented that “[the] plaintiff had to continue with the trial to obtain a judgment on liability against the 2nd defendant ...”[67] 70. In my view, such an approach is erroneous and is based on a failure to appreciate that a settlement is an alternative route to disposing of the claim which obviates the necessity to traverse the ECAO’s relief payment claim requirements. A settlement agreement between the Board and the plaintiff is a binding contract. Its terms are not subject to change and its binding quality not vitiated by any subsequent judgment obtained as between two different parties, the employee and the employer. Indeed, one would hope that a wasteful pursuit of the presumptively impecunious employer after settlement with the Board would be avoided. If the parties were not bound by the settlement and if the employee was subject to an asymmetrical rule whereby, depending on the level of the damages award against the employer, he might get less, but could not get more, than the settlement sum bargained for, there would be little incentive for the injured employee to settle. 71. I would accordingly give a negative answer to Question 2(a) and answer Question 2(c) as follows: The settlement amount agreed to by the plaintiff and the Board is contractually binding and not affected by the level of any subsequent judgment obtained against the employer. E. Conclusions 72. In relation to Question 1, my conclusions may be summarised as follows: (a) Section 20B(3) does not deprive the court of jurisdiction to order costs against the Board in relation to proceedings in which it has been joined as a party. (b) Power to make such costs orders is conferred by section 52A of the High Court Ordinance.[68] ECAO section 29 also recognises the existence of this power. (c) In exercising its discretion regarding a possible costs order against the Board, the principle that costs normally follow the event does not apply, given the Board’s statutory role. The starting-point should be that there is no order as to costs. However, the court has a wide discretion and may depart from this if the circumstances warrant a different order. 73. My conclusions regarding Question 2 are summarised as follows: (a) Section 29 authorises the Board to settle potential claims for relief payments in amounts it assesses to be reasonable, arriving at such assessments in good faith on available information considered sufficient after having made due inquiry. (b) Settlement is an alternative means for disposing of a claim and obtaining relief. The settlement agreement constitutes a binding contract and is not affected by any judgment which the employee may subsequently obtain against the employer. It should not generally be necessary to pursue that action after settling with the Board. F. Disposition 74. For the foregoing reasons, I would dismiss this appeal. Although such an outcome would normally entail making an order for costs against the Board, taking into account its special statutory role and accepting that there was a need for clarification of the legal position affecting the Board’s operations, I would, by way of order nisi, direct that there be no order as to costs so that the parties each bear their own costs. I understand that the respondent is on legal aid and I would direct that his costs be taxed in accordance with the Legal Aid Regulations. 75. I would direct that any submissions that the parties may wish to make as to costs be submitted in writing within 14 days of the date of this judgment and that, in default of such submissions, the order nisi stand as an order absolute. Mr Justice Fok PJ: 76. I agree with the judgment of Mr Justice Ribeiro PJ. Mr Justice Stock NPJ: 77. I agree with the judgment of Mr Justice Ribeiro PJ. Madam Justice McLachlin NPJ: 78. I agree with the judgment of Mr Justice Ribeiro PJ. Chief Justice Ma: 79. The Court unanimously dismisses this appeal and makes the orders as to costs referred to in paragraphs 74 and 75 above. Ms Audrey Eu SC and Mr Roger Phang, instructed by P.C. Woo & Co., for the 3rd Defendant (Appellant) Mr Raymond Leung SC and Mr Patrick Szeto, instructed by Kenneth Lam Solicitors, assigned by the Director of Legal Aid, for the Plaintiff (Respondent) [1] Cap 365. [2] Cap 282. [3] Section 7. [4] Section 4(a). [5] Part IV. [6] By the Employees’ Compensation Insurance Levies Ordinance (Cap 411), section 14. [7] Cap 411, Schedule 2. [8] ECAO section 4(a). [9] DCEC 2408/2014 (26 September 2016). [10] As the terms of the parties’ settlement were not placed before the Court, it is unclear what the terms agreed were. [11] Deputy High Court Judge To [2018] 3 HKLRD 897 (23 June 2017). [12] [2014] 6 HKC 361. [13] This decision is considered further in Section C.4 below. [14] [2018] 1 HKLRD 886. [15] Cheung and Kwan JJA, Au J [2019] 1 HKLRD 48 (23 November 2018). [16] Considered in Section C.3 below. [17] Court of Appeal §§41-43. [18] Cheung, Kwan and Au JJA [2019] HKCA 166 (15 February 2019). [19] Ma CJ, Fok PJ and Chan NPJ [2019] HKCFA 27 (25 July 2019). [20] Which are defined by reference to section 3 of the ECO and, unless excluded, would cover the payment of interest. [21] Appearing for the Board with Mr Roger Phang. [22] Section 20B(1): “Subject to section 20A(2) and subsection (3), the amount of a relief payment to an eligible person shall be the amount of damages for which the employer is liable to pay the eligible person after that amount is reduced by— (a) the amount, if any, of compensation which has been paid or is payable under the Employees’ Compensation Ordinance (Cap 282) in respect of the injured employee in relation to the accident concerned; and (b) the amount, if any, of those damages already paid by the employer.” [23] Introduced by the amendments to the ECAO in 2002. [24] Section 25(1)(a) : “…where the employer cannot be identified or, if identified, cannot be found; is insolvent; is dead or in the case of a company, has been dissolved, wound up or struck off the register; or for any reason, cannot be served with proceedings.” [25] Compare section 25(4). It has been held that an award of damages in such cases is made against the employer and not the Board: Tu Zhiqiang and Hap Sang Co (a firm) [2012] 1 HKLRD 1075 at §51; Kwan Kam Pui v Fung Man [2014] 6 HKC 361 at §34; Jiang Zhong v Yeung Chun Leung [2018] 1 HKLRD 886 at §3.11. [26] The aim, as explained in the Legislative Council (“LegCo”) Brief on the Employees Compensation Assistance (Amendment) Bill 2002 (5 February 2002) in support of the amendments introducing the joinder procedure, was mainly to give the Board some control regarding default judgments: “At present, the Ordinance does not explicitly empower the Board to defend claims in legal proceedings. This places the Board in a disadvantaged position because defaulting employers are usually absent in the legal proceedings leaving the claims undefended. It is proposed that where a proceeding has been initiated to make a claim in connection with a work-related accident, the Board may apply to the Court to be joined in the proceedings as a party and defend the claims.” (§17) [27] Cap 4. The equivalent in the District Court being section 53 of the District Court Ordinance (Cap 336). [28] Section 52A(2): “Without prejudice to the generality of subsection (1), the Court of Appeal or the Court of First Instance may, in accordance with rules of court, make an order awarding costs against a person who is not a party to the relevant proceedings, if the Court of Appeal or the Court of First Instance, as the case may be, is satisfied that it is in the interests of justice to do so. [29] As pointed out in the Minutes of the LegCo Meeting on 17 May 2001, §44. [30] Court of Appeal §43. [31] [2018] 1 HKLRD 886. [32] [2014] 6 HKC 361. [33] [2003] HKCU 412; HCPI 1221/2000 (15 April 2003). [34] [2014] 6 HKC 361 at §§66-67. [35] Judgment §§43 and 45. [36] Judgment §47. [37] Judgment §48. [38] Judgment §§51-52. [39] Judgment §52. [40] Judgment §55. [41] Judgment §§55-56. [42] Court of Appeal §§25-28. [43] Court of Appeal §32. [44] Court of Appeal §35. [45] Court of Appeal §33. [46] Court of Appeal §§34, 45 and 46. [47] ECAO section 4. [48] Section 25A(a). [49] Section 25A(b) and (c). [50] Judgment §47. [51] As further discussed in Section D.3 below. [52] Loss of the right to apply under section 16 in respect of that claim. [53] Relieving the Board of paying more than the eventual ECO award and of paying the costs of the person refusing, as from the date of the offer. [54] [2013] 2 HKLRD 614. [55] [2014] 6 HKC 361 at §1. [56] Ibid at §15. [57] The Minutes of the LegCo Meeting, §§45 and 47. [58] Properly quantified, as discussed in Section D.4 below. [59] Question 2 also addresses issues concerning the legal effect of any settlement made, discussed separately below. [60] Section 40(2) makes it an offence for any person, without reasonable excuse, to fail to comply with or respond to an inquiry made under section 25B(8). [61] ECAO section 4(a). The objects include “the protection of the entitlement of employees and others to compensation for employment-related injury [and] for the making of relief payment to eligible persons in relation to damages for employment-related injury” (Long Title). [62] ECAO section 5(1). [63] Judgment §25. [64] Court of Appeal footnote 24. [65] Court of Appeal §41. [66] Appearing for the respondent with Mr Patrick Szeto. [67] Court of Appeal §44. [68] Cap 4. 
Hon Poon CJHC (giving the judgment of the Court): 1. Article 21 of The Law of the People’s Republic of China on Safeguarding National Security in the Hong Kong Special Administrative Region (“NSL”) establishes a two-tier penalty regime for an offence falling thereunder by reference to its severity in these terms: “情節嚴重的，處五年以上十年以下有期徒刑；情節較輕的，處五年以下有期徒刑、拘役或者管制。” Its English translation, which is for information only,[1] reads: “If the circumstances of the offence committed by a person are of a serious nature, the person shall be sentenced to fixed-term imprisonment of not less than five years but not more than ten years; if the circumstances of the offence committed by a person are of a minor nature, the person shall be sentenced to fixed-term imprisonment of not more than five years, short-term detention or restriction.” For present purposes, we will refer to the tier for serious offences as “the Upper Tier” and that for minor offences as “the Lower Tier” respectively. 2. The NSL provisions on penalty in Parts 1 to 4, including NSL 21, are augmented by NSL 31 to NSL 35 in Part 5. NSL 33(1) provides for three disposals to adjust an otherwise appropriate penalty when any of the three conditions specified is established, as follows: “有以下情形的，對有關犯罪行為人、犯罪嫌疑人、被告人可以從輕、減輕處罰；犯罪較輕的，可以免除處罰： （一） 在犯罪過程中，自動放棄犯罪或者自動有效地防止犯罪結果發生的； （二） 自動投案，如實供述自己的罪行的； （三） 揭發他人犯罪行為，查證屬實，或者提供重要線索得以偵破其他案件的。” Its English translation reads: “A lighter penalty may be imposed, or the penalty may be reduced or, in the case of a minor offence, exempted, if an offender, criminal suspect, or defendant: (1) in the process of committing an offence, voluntarily discontinues the commission of the offence or voluntarily and effectively forestalls its consequences; (2) voluntarily surrenders himself or herself and gives a truthful account of the offence; or (3) reports on the offence committed by other person, which is verified to be true, or provides material information which assists in solving other criminal case.” 3. In this application for leave to appeal against sentence, it falls upon this Court to consider two principal questions. The first concerns the categorisation of the circumstances of the NSL 21 offence committed by the applicant as serious by Her Honour Judge Amanda Woodcock below (“the Judge”), which is a question of fact. The second question is one of law on the effect of the Upper Tier and NSL 33(1) in their application to sentencing of a serious NSL 21 offence. A. Proceedings below 4. Before the Judge below, the applicant pleaded guilty to a charge of incitement to secession, contrary to NSL 20 and NSL 21.[2] The particulars of the offence read: “[The applicant], between the 30th day of June, 2020 and the 24th day of September, 2020, both dates inclusive, in Hong Kong, together with other persons, incited other persons to organize, plan, commit or participate in acts, whether or not by force or threat of force, with a view to committing secession or undermining national unification, namely separating the Hong Kong Special Administrative Region from the People’s Republic of China or altering by unlawful means the legal status of the Hong Kong Special Administrative Region.” 5. For the purpose of his guilty plea, the applicant agreed to the Amended Summary of Facts.[3] The admitted facts as agreed may be outlined as follows. A1. Prosecution case as admitted A1.1 The Channel 6. The applicant committed incitement to secession via a Telegram channel (“the Channel”) first established on 14 December 2019 under the name “橙級裝備”.[4] Its name was changed on different occasions: (1) on 21 April 2020, to “得返一個來自香港的黑衣人”;[5] (2) on 30 April 2020, to “變返兩個半退休既冷氣勇武”;[6] (3) on 8 May 2020, to “兩個全力回歸既勇武派”[7] (4) on 22 May 2020, to “兩個勇武派的中共速龍小隊”;[8] (5) on 11 June 2020, to “暫時得返一個會覆機既勇武Admin”;[9] (6) on 3 August 2020, to “2個專叫港獨台獨反共反國安的年輕人”;[10] (7) on 28 August 2020, to “抗共港獨台 H.K.I.A.F.”.[11] 7. The applicant and another male surnamed Tsui were at all material times the administrators of the Channel. Initially, the identity of the administrators of the Channel were concealed but after a change of setting on 26 April 2020, viewers could see which administrator made which posts. Tsui and the applicant used the identity “Equipment for attack” admin and “Equipment for defense” admin respectively. In making posts, Tsui used “Yuri Orlov” with variations; and the applicant used “有為yy”[12]. Tsui and the applicant were also responsible for sale of “equipment for attack” and “equipment for defense” respectively. 8. As at 23 September 2020, a total of 1,040 persons joined and subscribed to the Channel. The link of the Channel was provided publicly under “Channel Info”, an introduction to the Channel on the Telegram platform. The applicant was identified as the commissioner for “defence-model”, whereas three other Telegram accounts were provided as commissioners for “attack-model”, “miscellaneous model” and “chatroom”. A1.2 Secessionist posts and content on the Channel 9. Against the context of widespread violent and unlawful events in Hong Kong since June 2019, available records of the posts made in the Channel began on 31 January 2020 at 1740 hours. The police preserved all the available records of the posts since then up to 23 September 2020, one day before the arrest of the applicant. A total of 1495 posts were so preserved and are set out in Annex B to the Amended Summary of Facts with the English translation in Annex C. The posts made it clear as to which administrator made which post on what day and at what time. Out of the 1495 posts, at least about 338 posts were made by and/or related to the applicant. During the offence period, namely, between 30 June 2020 (starting at 2358 hours, that is, right after the NSL was made applicable to the HKSAR[13]) and 23 September 2020 (the day before the applicant was arrested), there were a total of 357 posts. 10. Before the NSL was applied to the HKSAR, there were a large number of posts in various forms such as texts, photo images, videos and other graphical depictions which the administrators posted or re-posted from other Telegram channels. The posts were generally of a nature designed to incite violence and counsel disobedience to the law by providing and offering for sale weapons and equipment such as pepper sprays, extendable batons, respirators and other body gear to protestors against the police; and providing protestors with information of police operations as well as strategy and tactics against police action. In particular, there were posts which expressly or implicitly suggesting acts to be taken to change the regime or status of the HKSAR unlawfully. For example, there were messages containing the slogans “Liberate Hong Kong, Revolution of our times”, “Hong Kong independence, the only way out” and “Fight against totalitarianism”. 11. The Channel also, for 27 times, organized polls and/or shared polls made by other Telegram channels or groups to seek opinion of viewers of the Channel on matters such as what gear or equipment would be sold or purchased and other issues relating to the protests in Hong Kong since June 2019. 12. After the NSL applies to the HKSAR, during the offence period, the Channel continued to carry posts and re-posts of a seditious nature and inciting hatred against the Central Authorities, the People’s Republic of China, the HKSAR and the police. There were many posts relating to Hong Kong independence, anti-communism, a violent faction that is anti-communist, Hong Kong being a country and appeals to participants to train themselves up to fight for “this country”, and appeals to others to equip themselves with weapons for close quarter combat. The posts for sale of weaponry and equipment also continued. 13. In particular, as set out at [20] of the Amended Summary of Facts, there were many posts to incite other persons by means of the posts in the Channel to organize, plan, commit or participate in acts, whether or not by force or threat of force, with a view to committing secession or undermining national unification, namely separating the HKSAR from the PRC or altering by unlawful means the legal status of the HKSAR, in particular with contents containing “光復香港，時代革命”, advocating for the so-called independence of Hong Kong with the expression “香港獨立”, expressly or otherwise implicitly suggesting acts to be taken to unlawfully change the regime or status of the HKSAR. The applicant admitted that he personally made some of those posts. 14. Moreover, the Channel continued to organize, for 12 times, polls and/or share polls made by other Telegram channels or groups to seek opinion of viewers of the Channel with contents which were secessionist in nature. Responses ranging from 126 to 1,226 voters were received. 15. In sum, the applicant admitted the offence of incitement to secession as charged. A2. Mitigation 16. At the time of sentence, the applicant was aged 25 and had a clear record. He was single and lived with his mother. His father died when he was 15, which left a significant detrimental impact on him and his family. His mother took on the full responsibility of supporting him and his sister both emotionally and financially. He was a first-year student reading civil engineering at the Hong Kong Polytechnic University. 17. The applicant was diagnosed with Attention Deficit Hyperactivity Disorder (ADHD) in 2015 and had been receiving treatment since 2016. The defence stressed that his ADHD affected his DSE results but he did not give up. He motivated himself to study in his own time to repeat the examination and work at the same time to alleviate his mother’s financial burden. He was successful in his retake and was admitted to the HKPU in 2020. 18. According to the mitigation letters written by the applicant’s mother, a family friend, a volunteer who had regularly visited him since remand, his principal, vice principals and supervisor from school: (1) the applicant was a filial son and a kind-hearted young man who helped others and was determined to help himself despite his own difficulties; (2) he was a good student who actively participated in school life, voluntary work and extra-curricular activities as a squadron leader of the Air Cadets and vice president of the Student Association; (3) he was deeply affected by the social unrest in 2019 and concerned about the future of Hong Kong. 19. The defence emphasized that the applicant was remorseful. He himself wrote a letter accepting responsibility and explaining why he committed the offence. He pleaded guilty at the earliest opportunity and had been cooperating with the police. He now realized his wrongdoings and wanted only to better himself to contribute fruitfully to society once he had served his sentence. 20. In terms of seriousness of the offence, the defence submitted that it belonged to the minor category. It was stressed that the circulation of the Channel was likely to have been limited and not the same as appearing on television or publishing similar incitements in newspapers reaching large audiences; that those who logged into the Channel were others who already shared thoughts and sentiments similar to the administrators. They would not be those that would be influenced by the posts to commit acts of secession; and that the applicant lacked celebrity status, was not a well-known figure or public leader so that his messages would not have had much impact. A3. Reasons for sentence 21. The Judge delivered the Reasons for Sentence on 29 April 2022.[14] She first found that the present case was of serious nature for the purpose of NSL 21, warranting a sentence of not less than five years but not more than ten years as prescribed.[15] In so finding, she took into account the context in which the crime was committed. It was at a time when there was social unrest and heightened anti-government sentiment.[16] She emphasized the fact that social media had been used to perpetrate the offence.[17] The content of the posts was of a serious nature, leaving no doubt that the applicant condoned, promoted, advocated and incited others to commit secession or undermine national unification.[18] Amongst the rhetoric, the administrators of the Channel offered for sale weapons or items they described as either of an attacking or defending nature.[19] Finally, in a post on 23 July 2020, the Channel promoted the sale of bottles of hand sanitizer but also provided a means for participants to donate to the administrators for “fighting for Hong Kong independence by this channel”.[20] 22. The Judge adopted a starting point of five years and six months and reduced it by one-third to reflect the applicant’s timely plea.[21] However, before she became functus officio, the prosecution submitted that on a plain reading of NSL 21, the minimum of five years in the Upper Tier was meant to be mandatory so that while the Judge could still give the applicant a discount on account of his guilty plea, the ultimate sentence could not go below five years.[22] The prosecution cited R v Andrew James Jordan and others [2005] 2 Cr App R (S) 44 in support. Accepting that the prosecution appeared to be correct, the defence urged the Judge to re-consider the nature of the offence in light of the mandatory minimum sentence.[23] The Judge agreed with the prosecution’s reading of NSL 21 so that she must impose a minimum of five years’ imprisonment. She maintained her finding that the nature of the offence was serious and having regard to all the circumstances and mitigation, sentenced the applicant to five years’ imprisonment.[24] B. Grounds of Appeal 23. In the Amended Perfected Grounds of Appeal, a total of five grounds of appeal are raised.[25] The first two grounds attack the Judge’s categorization of the offence as serious: (1) The Judge erred in categorizing the circumstances of the offence as serious under NSL 21 by over-emphasizing or attaching too much weight to the factors adverse to the applicant and disregarding or not giving proper and sufficient weight to those favourable to him. (Ground 1) (2) Alternatively, even if the offence was serious, the Judge erred in adopting a manifestly excessive starting point because of her mistaken treatments of the factors as aforesaid. (Ground 2) They call for a closer examination of the facts. 24. The next two grounds complain of the Judge’s failure to give the customary one-third discount in full: (1) The Judge erred in her interpretation of NSL 21 regarding the permissible final sentence by not taking into account NSL 33(1) whereby fulfilment of one of the prescribed circumstances may “reduce” even a minimum sentence; and by not taking into account or giving sufficient weight to the relevant mitigating factors including the guilty plea. (Ground 3) (2) Further, the Judge failed to give sufficient reduction with reference to NSL 33(1)(2) as the circumstances concerning the applicant came very close, if not identical to the scenario provided thereunder; and the relevant mitigating factors including the guilty plea. (Ground 4) They entail a construction exercise to ascertain the legislative intention of the Upper Tier and NSL 33(1) in their application to sentencing of a serious NSL 21 offence. 25. The last ground wraps up the applicant’s case by asserting that because of Grounds 1 to 4, individually or cumulatively, the sentence is wrong in principle or manifestly excessive. It does not add much in terms of substance or analysis and stands or falls with the other Grounds. C. Grounds 1 and 2 26. In light of how Mr Edwin Choy, SC,[26] for the applicant, argued Grounds 1 and 2, it is convenient to deal with them together. C1. The general approach to categorization 27. In HKSAR v Ma Chun Man [2022] HKCA 1151, which was handed down after the time of sentence in the present case, this Court laid down the general approach to the categorization of an offence of incitement to secession under NSL 21 thus: (1) Since the legislative intent of the NSL is to seek convergence, compatibility and complementarity with local laws, in the absence of any NSL provision on how to classify cases of incitement to secession, the local sentencing principles apply. See [67]. (2) According to local sentencing principles, the sentencing of an offence is informed by its gravamen. As a pre-emptive offence, the gravamen of the offence of incitement under NSL 21 is (a) to prevent inciting (which includes enticing or encouraging) another person to commit the crimes of secession; and (b) to allow the law to intervene and prevent at the earliest stage the commission of the crimes of secession by those incited. Its purpose is to fully safeguard the immense public interest in national security, territorial integrity, the constitutional foundation and the legal status of the HKSAR, and to ensure that any crime of secession will be thwarted and punished timely and effectively. See [73]. (3) Whether a case of incitement to secession is serious or minor depends on all the circumstances. Drawing on the common law offence of incitement, which is similar in nature, the court will focus primarily on the offender’s conduct, as well as the consequences, risks and other possible effects that his conduct might entail. The court will take into account, non-exhaustively: (a) the context in which the offence was committed; (b) the modus operandi; (c) the frequency, duration and persistency of the incitement; (d) the scale of the incitement; (e) whether there was premeditation; (f) whether the offence involved violence or threat of violence; (g) whether the offender acted in concert with others; (h) the persons or groups being targeted and the potential influence on them; (i) whether anyone had been incited into committing secession or any other offence; or how imminent or serious such threat or risk was; (j) the actual or potential influence that the offender may generally have on society or a particular sector thereof. See [73] – [75]. (4) Finally, the court has to carefully consider the facts, identify all the relevant factors, and give due weight to them individually and collectively in the overall circumstances in categorizing the case as serious or minor. See [76]. C2. Categorization of the present case 28. Applying the above approach to the present case, we first consider the context of the offence. 29. The applicant committed the offence between 30 June and 24 September 2020. As noted in Ma Chun Man, at [78] and accepted by the applicant, although the scale and severity of unlawful and violent assemblies at that time was less serious than before the NSL was applied to Hong Kong, Hong Kong was still facing considerable threats and risks to national security and public disorder. Viewed against that context, the threats and risks to national security by the applicant’s incitement was more than latent as submitted by Mr Choy. They were real and must be given due weight in the overall assessment of the seriousness of his conduct. 30. Mr Choy compared the facts of the present case with those in Divin and McGinlay v HM Advocate [2013] JC 259 and Attorney General v Tse Chung [1967] HKLR 452 and submitted that the timing of the commission of the offence by the applicant was far from being the worst of its kind. 31. In Divin, the appellants pleaded guilty to an offence of using social networking sites to incite others to riot and were sentenced to 3 years and 3 months’ and 3 years’ imprisonment respectively. The context of the offence was that of extensive and very recent rioting and looting in a number of cities and towns in England which had been widely reported by the media in Scotland. Although such criminal conduct had not been spread to Scotland, concerns existed that it might. Against that background, the High Court of Judiciary at [20] held that the appellants’ culpability was high and despite their young age, a sentence of detention was the only appropriate disposal for each of them. 32. In Tse Chung, the appellant was convicted of unlawful assembly during the 1967 riots in Hong Kong and was sentenced to 12 months’ imprisonment. At the time of the offence, he stood at the forefront with a crowd of people in violent and noisy confrontation with the police. The appellant appealed against sentence on the grounds that the magistrate must have imposed sentence for reasons other than those he actually expressed and that he should not have taken judicial notice of the then disturbed state of Hong Kong. That was roundly rejected by the court at p 458. 33. In our view, the seriousness of a particular offence has to be gauged by its actual circumstances, which by nature, must vary from case to case. With so many variables involved, there are limits to the extent to which comparisons with the detail of other cases can assist the sentencing court in determining the seriousness of the offence and the appropriate sentence before it: see Divin, at [28]. If the facts or contexts are comparable, then those cases may provide some assistance. If not, the utility of such comparison is minimal. Here, we do not find the comparison with Divin or Tse Chung helpful because the facts and contexts were so different. Ma Chun Man is more useful because it also involves an NSL 21 offence and the time of the offence, namely, between August and November 2020, is comparable. Further, it does not assist the assessment to ask if the context of the present offence is or is not the worst of its kind. The more relevant question is whether the context, together with the other features, in the overall assessment of the circumstances justifies the categorization of the offence as a serious one. 34. Next, as is now well-established, the use of social media for committing incitement is an aggravating feature: 律政司司長訴庚家駒[2020] HKCA 1019, at [28]. The aggravation lies in the extreme effectiveness of social media in providing platforms or means to individuals with the ease and ability to communicate or disseminate messages or content to a vast audience instantly without physical contact or geographical constraints, thereby amplifying the effects of the incitement and increases exponentially the threats and risks it poses to national security. As the English Court of Appeal relevantly explained in R v Blackshaw [2012] 1 WLR 1126, in connection with the common law offence of incitement to serious public disorder,[27] at [73]: “We are unimpressed with the suggestion that in each case the defendant did no more than make the appropriate entry in his Facebook. Neither went from door to door looking for friends or like-minded people to join up with him in the riot. All that is true. But modern technology has done away with the need for such direct personal communication. It can all be done through Facebook or other social media. In other words, the abuse of modern technology for criminal purposes extends to and includes incitement of very many people by a single step. Indeed it is a sinister feature of these cases that modern technology almost certainly assisted rioters in other places to organize the rapid movement and congregation of disorderly groups in new and unpoliced areas.”[28] 35. As is agreed, Telegram is a widely used social media platform allowing users to set up private or public channels to publish posts. Public channels are operated by its owners, administrators or authorized persons. Other users can join and subscribe to public channels to view and receive posts as part of the messaging function of Telegram. Users can also view the posts in a public channel without joining it by searching for and accessing the channels with name, link and/or ID of the channel. The Channel used by the applicant was a public channel on Telegram and all Telegram users could access the posts there as aforesaid. And there were 1,040 users who joined and subscribed the Channel, which is not negligible. Using the Channel to commit the offence is clearly an aggravation. 36. Mr Choy took no issue at the judicial abhorrence against the commission of incitement by using social media. However, he argued that there remains a wide spectrum of culpability. Adopting a broad-brush approach would not be able to differentiate different degrees of culpability. The court should consider how the social media was used, to what extent it was used and its actual effectiveness. Thus in Ma Chun Man, the applicant effectively used different social media together with traditional means to achieve the maximum secession effects, including Facebook, Telegram, videos available online, incitements at big shopping malls as well as outside Government Headquarters and police station, interactive journalistic interviews and pre-made propaganda materials for distribution. In stark contrast, the applicant here only administered with another person a single Telegram channel containing posts and photos, the contents of which, he accepted, were objectively far from harmless. 37. By comparison, the modus operandi adopted in Ma Chun Man might be more serious. But the use of the Channel in the manner as it was in the present case remains an aggravation. That must weigh heavily in the court’s assessment of the seriousness of the offence. 38. Moreover, there are other aggravating features: (1) The applicant defiantly denounced the authority of the NSL on the Channel. For example, on 4 July 2020, he published a post in which he said that he would regard the NSL as “a piece of waste paper”.[29] (2) Secessionist posts and videos were published right after the NSL was applied to Hong Kong on 30 June 2020[30] and on 1 July 2020, the establishment day of the HKSAR,[31] both of which were then sensitive dates with clear risks of provoking secessionist and other unlawful acts: see Ma Chun Man, at [83(1)]. Mr Choy submitted that those were the only sensitive dates over a time span of nearly three months, in stark contrast to the much greater number of sensitive dates identified in Ma Chun Man. Strictly in terms of figures, Mr Choy is right. But the risks involved in the present case were still high. (3) There were a total of 357 posts published on the Channel with 8 directly by the applicant himself. Mr Choy submitted that the Judge had failed to consider the limited number of posts (8) made by or related to the applicant. However, as submitted by Mr Anthony Chau, Deputy Director of Public Prosecutions,[32] the applicant pleaded guilty and admitted to have committed the offence with others, that is, jointly as administrators of the Channel. All secessionist and other unlawful content in the Channel were attributable to him, even if they were published by others. The 8 posts published by him in fact only aggravated his criminality by way of direct conduct. (4) The Channel, for 12 times, organized polls and/or shared polls made by other Telegram channels or groups to seek opinion of viewers of the Channel with contents which were secessionist in nature and received responses ranging from 126 to 1,226 voters. Arousing public attention and discussion in the guise of polls by making use of secession materials with a view to advocating secession is clearly an aggravation. (5) The applicant acted in concert with others, including Tsui. According to how the Channel was administered, there was plainly a degree of division of labour between them, although as submitted by Mr Choy it did not appear to involve a high degree of sophistication or scale. (6) As noted by the Judge, during the offence period, the Channel carried posts for sale of weaponry and gear, advocating violence and targeting the sovereignty over the HKSAR by China. Advocating the use of violence is undoubtedly a very serious aggravating factor: Ma Chun Man, at [75(6)]. (7) As further pointed out by the Judge, the Channel was used to appeal to participants for donations to the administrators, including the applicant, “for fighting for Hong Kong independence”. Seeking to raise funds for the secessionist cause is clearly an aggravation. (8) The Channel targeted the general public and its 1,000 plus subscribers. Further, by advertising for sale of weapons and gear, the Channel clearly aimed at people including protestors who were inclined to resorting to violence, thereby posing a greater risk to national security and public disorder. 39. The overall circumstances of the offence well justify the Judge’s categorization. Mr Choy’s complaint that she erred in attaching too much weight to those factors adverse to the applicant cannot stand. 40. Mr Choy further submitted that although it might not be a mitigating factor, the fact that no persons were shown to have been incited adds nothing to aggravate the offence. As the applicant was not a social celebrity, his actual or potential influence on others would be much lesser than otherwise. Even taken at their highest, those factors favourable to the applicant do not in our view detract from the seriousness of his offence. Mr Choy’s complaint that the Judge erred in not giving sufficient weight to them is not substantiated either. 41. Having carefully evaluated the circumstances of the present offence, we find that they were of a serious nature within the meaning of NSL 21. The Judge was correct in her categorization, albeit her reasoning was not as full as this judgment, which is understandable because at the time she did not have the guidance from Ma Chun Man. Ground 1 accordingly fails. C3. Starting point 42. After stating that she stood corrected and agreed that NSL 21 did impose a mandatory minimum of five years and maintaining her categorization of the offence as serious, the Judge did not say anything further about starting points. Understood in context, and agreed by the parties, it means that she stood by the five years and six months starting point. In so doing, she must have found that the applicant’s culpability falls near the lower end of the category. We agree with her finding. The starting point adopted by the Judge is within the reasonable bounds that a sentencing court may impose for a serious offence sitting in that end. There is no basis for this Court to disturb it as being manifestly excessive. Ground 2 therefore also fails. D. Core issues arising from Grounds 3 and 4 43. We now come to Grounds 3 and 4. They give rise to two core issues. 44. The first core issue is: whether it is the legislative intention of the Upper Tier in prescribing the range of penalty for serious NSL 21 offences to lay down a range of starting points between the maximum of ten years and the minimum of five years, as the applicant contended? Or whether it is to set five years as a mandatory minimum, as the respondent contended? We will refer to this as the 1st Core Issue. 45. If the Upper Tier is intended to be a range of starting points, subject to the next core issue on the effect of NSL 33(1), the sentencing court should be able to adjust the penalty to below five years after giving appropriate discount to applicable mitigating circumstances including a guilty plea. If the Upper Tier mandates five years as the minimum, the court must faithfully apply it. Subject to the next core issue, whatever discount the court may give for the mitigating circumstances including a guilty plea, it cannot take the ultimate sentence below it. 46. The second core issue is: whether, as the respondent contended but disputed by the applicant, the legislative intention is that the three conditions specified in NSL 33(1) are exhaustive so that in their absence, the court cannot adjust the penalty of a serious NSL 21 offence lower than the minimum of five years in the Upper Tier on account of other mitigating circumstances including a guilty plea? We will call this the 2nd Core Issue. 47. Once the Core Issues are resolved, Grounds 3 and 4 can be disposed of shortly. E. Construction of NSL 21 and NSL 33(1) 48. NSL 21 and NSL 33(1) and their English translations have already been set out at [1] and [2] above. In construing them, we adopt the contextual and purposive approach as the Court of Final Appeal did in HKSAR v Lai Chee Ying (2021) 24 HKCFAR 33; and HKSAR v Ng Hau Yi Sidney (2021) 24 HKCFAR 417: see also Tong Ying Kit v Secretary for Justice [2021] 3 HKLRD 350 at [31] – [34]. E1. Context of NSL 21 and NSL 33(1) 49. First and foremost, NSL 21 and NSL 33(1) must be examined in the light of the context and purpose of the NSL as a whole, taking into account the constitutional basis upon which the NSL is applied in the HKSAR: Lai Chee Ying, at [8]. Given the special status of the NSL as a national law applied under article 18 of the Basic Law in the HKSAR, and the express reference to the Decision of the National People’s Congress on Establishing and Improving the Legal System and Enforcement Mechanisms for Safeguarding National Security in the Hong Kong Special Administrative Region (28 May 2020) in NSL 1, regard may properly be had to the legislative materials of the NSL to ascertain the context and purpose of the NSL as a whole: Lai Chee Ying, at [11]. 50. In Lai Chee Ying, the Court of Final Appeal in Part B summarised the legislative history of the NSL. In so doing, the Court canvassed the legislative materials of the NSL in some detail against the background of prominent national security risks in the HKSAR, as seen in the serious, prolonged disturbances to public order, the alarming breakdown of law and order and escalating violence, the wholesale challenges to the governments of the PRC and the HKSAR, and the rampant acts and activities threatening national security including those advocating “Hong Kong independence”. As observed by the Court of Final Appeal: (1) Against that background, the Central Authorities found the absence of national security legislation pursuant to article 23 of the Basic Law[33] since 1 July 1997 unacceptable. (2) The Central Authorities therefore decided to enact the NSL at the national level to “establish and improve the legal system and enforcement mechanisms for the HKSAR to safeguard national security, and to change its long-term ‘defenceless’ condition in the field of national security”. The need to improve the HKSAR’s system and to address the institutional deficiencies in relation to safeguarding national security was further stressed as one of the working principles for drafting the NSL. (3) The basic principles underlying the NSL were stated to be: (a) First, “resolutely safeguarding national security”; (b) Secondly, “upholding and enhancing the ‘One Country, Two Systems regime’”; (c) Thirdly, adhering to “administering Hong Kong in accordance with the law” and resolutely upholding “the constitutional order in the HKSAR as established by the Constitution and the Hong Kong Basic Law”; (d) Fourthly, resolutely opposing external interference; and (e) Fifthly, “fully safeguarding the legitimate rights and interests of Hong Kong residents”. 51. These basic principles are contained and reflected in NSL 1 which expressly stipulates the purpose of the NSL in these terms: “This Law is enacted, in accordance with the Constitution of the People’s Republic of China, the Basic Law of the Hong Kong Special Administrative Region of the People’s Republic of China, and the Decision of the National People’s Congress on Establishing and Improving the Legal System and Enforcement Mechanisms for Safeguarding National Security in the Hong Kong Special Administrative Region, for the purpose of: - ensuring the resolute, full and faithful implementation of the policy of One Country, Two Systems under which the people of Hong Kong administer Hong Kong with a high degree of autonomy; - safeguarding national security; - preventing, suppressing and imposing punishment for the offences of secession, subversion, organisation and perpetration of terrorist activities, and collusion with a foreign country or with external elements to endanger national security in relation to the Hong Kong Special Administrative Region; - maintaining prosperity and stability of the Hong Kong Special Administrative Region; and - protecting the lawful rights and interests of the residents of the Hong Kong Special Administrative Region.” The immediate point to note for present purpose is that as an integral component of safeguarding national security, preventing, suppressing and imposing punishment for the NSL offences is distinctly stated as a primary purpose of the NSL (“Primary Purpose”), highlighting its immense importance in the overall scheme of the NSL. 52. Moreover, very relevantly, the NSL lays particular emphasis on the Primary Purpose, requiring strict and full application of the NSL and local laws to prevent, suppress and impose punishment for offences endangering national security when referring to criminal proceedings and penal sanctions. Thus, NSL 3(3) requires the executive authorities, legislature and judiciary of the HKSAR, in discharging their duty to safeguard national security to:[34] “effectively prevent, suppress and impose punishment for any act or activity endangering national security in accordance with this Law and other relevant laws.” Specifically, NSL 8 directs the law enforcement and judicial authorities of the HKSAR, in order to safeguard national security effectively, to fully enforce this Law and the laws in force of the Region:[35] “concerning the prevention of, suppression of, and imposition of punishment for acts and activities endangering national security.” Further, NSL 42(1) requires the law enforcement and judicial authorities, in applying local laws concerning matters such as detention and time limit, to handle cases concerning offence endangering national security in a fair and timely manner so as to:[36] “effectively prevent, suppress and impose punishment for such offence.” 53. It must be emphasised that the imperative in NSL 3(3), NSL 8 and NSL 42(1) on strict and full application of laws to further the Primary Purpose (“the Imperative”) covers not only the NSL but also local laws. As will be elaborated, the Imperative materially impacts on how local sentencing laws are to apply in the framework of NSL 21 and NSL 33(1). 54. While the NSL is tailor-made to cater for the national security risks in the HKSAR, it is by no means intended to be the one and only code of laws to safeguard national security in the Region. For NSL 7 directs the HKSAR to enact BL 23 legislation as soon as possible and to refine relevant laws to protect national security. That article recognizes that the NSL provides an incomplete framework of laws designed to protect national security: Ng Hau Yi Sidney, at [18]. Moreover, after considering NSL 3, NSL 7, NSL 8 and NSL 42, the Court of Final Appeal said at [24]: “Viewed purposively, the intent of the NSL is plainly for national security to be safeguarded by the complementary application of the laws which it creates together with the existing laws of the HKSAR, such as those contained in Part II of the Crimes Ordinance[37].” 55. Arising from the above context of the NSL as a whole are four instructive propositions. 56. First, in the wake of the grave and indeed unprecedented national security risks in the HKSAR and in the absence of BL 23 legislation, the NSL was enacted to partially fill the legal lacuna for safeguarding national security in the Region. The remaining gap after enactment of the NSL is to be filled by (1) the application of relevant existing laws; (2) the refinement of such laws; and (3) the enactment of BL 23 legislation. Pending (2) and (3), the laws available to protect national security are the NSL and relevant existing laws, such as those in Part II of the Crimes Ordinance. Local sentencing laws must therefore operate in tandem with the NSL to achieve the aim of safeguarding national security, giving priority to NSL provisions in case of inconsistency:[38] Ma Chun Man, at [66]. 57. Second, applying the Imperative to the penalty regimes in the NSL, priority should be given to the penological considerations of deterrence, retribution, denunciation and incapacitation, that is, putting out of the power of the offender to commit further offences (“the Penological Considerations”). It follows that the construction of NSL 21 must give full effect to the Penological Considerations. 58. Third, the court may take into account applicable mitigating circumstances to balance against the rigour of the Penological Considerations in arriving at a proportionate sentence. In the context of the NSL, because of the Imperative, not all mitigating circumstances are applicable. Only those which do not compromise the Primary Purpose are permissible. This informs the construction of NSL 33(1). 59. Fourth, the Imperative also governs the application of local sentencing laws to the sentencing of the NSL offences. To achieve convergence, compatibility and complementarity with NSL 21, local sentencing laws on mitigation can apply only if they do not prejudice the effect of the Penological Considerations. To achieve the same result with NSL 33(1), local sentencing laws on mitigation can only apply if they do not compromise the Primary Purpose. E2. Construing NSL 21 and NSL 33(1) 60. When construed with the above considerations in mind, the legislative intention and effect of NSL 21 is clear. 61. For minor offences, the Lower Tier stipulates three penalty options: (1) fixed-term imprisonment; (2) short-term detention; and (3) restriction. For fixed-term imprisonment, the maximum is five years. Insofar as imprisonment is concerned, the Lower Tier sets a range of starting points with five years as the maximum, leaving it to the court to determine what term is appropriate in the particular circumstances of the case. In so doing, the Lower Tier operates in very much the same way as most local statutory penalty regimes do. 62. For serious offences, the Upper Tier stipulates only one penalty option, that is, fixed-term imprisonment. It also sets a maximum of ten years and a minimum of five years. By specifying such a scope of penalty, unwarranted divergence in the assessment of punishment by different courts is avoided. This reflects broadly the notion of fairness in the context of criminal justice that generally there should be a substantial degree of consistency in the punishment of offences with comparable severity. Further, the choice of imprisonment as the only penalty option and the range of penalty in the Upper Tier reflect the drafters’ judgment with respect to the gravity of serious NSL 21 offences and how to give full effect to the Penological Considerations to further the Primary Purpose in sentencing serious NSL 21 offences. 63. Thus viewed purposively, the minimum of five years in the Upper Tier is mandatory. It follows that subject to the effect of NSL 33(1), whatever discount the court may give on account of mitigation, the ultimate sentence imposed cannot go below the minimum of five years. 64. Turning to NSL 33(1), its legislative intention and effect is equally clear. 65. To recap, the three disposals in NSL 33(1) are 從輕處罰, 減輕處罰and免除處罰. 66. The English translations for 從輕處罰 and 減輕處罰, that is, “a lighter penalty may be imposed” and “the penalty may be reduced”, appear to be synonymous. However, in Chinese, which is the official version, their natural and ordinary meanings are quite different. 67. According to 現代汉語規范詞典,第三版 (Xiandai Hanyu Guifan Cidian, the 3rd Edition), at p 222, 從輕 means “在法律或條例規定的範圍內給予較輕的（刑罰或處罰）”, which may be translated as “give a lighter punishment or penalty within the range stipulated by the law or regulations”. And at p 641, 減輕means “數量減少；程度降低”, which may be translated as “quantum reduced; level lowered”. When applied to sentencing, 減輕處罰means reducing or lowering the penalty below the range of penalties stipulated by the law. That is distinct from 從輕處罰and is a more lenient option of adjusting the penalty. Both parties agree that 從輕處罰 and 減輕處罰 bear the above meanings. 68. The three disposals accordingly follow their sliding degree of leniency in their order of appearance in NSL 33(1): (1) 從輕處罰, imposing a lighter penalty, that is, imposing a lighter penalty within the applicable tier as prescribed by the relevant NSL provisions; (2) 減輕處罰, reducing the penalty, that is, reducing the penalty from the applicable tier to a lower tier, which is more lenient; and (3) 免除處罰, that is, exempting the penalty, which is most lenient. As to which disposal the court may adopt, it must depend on the actual circumstances of the case before it. Since 免除處罰 does not arise in the present case, we will just focus on the other two.[39] 69. Whether 從輕處罰 (to impose a lighter penalty) or 減輕處罰 (to reduce a penalty) depends necessarily on the weight to be attached to the available mitigating circumstances. Put differently, it is the weight of the mitigation that determines if the court should or 減輕處罰. Although 從輕處罰 and 減輕處罰define the permissible extent of discount, they essentially involve the same qualitative assessment in which the court evaluates and weighs all the available mitigating factors. The weight so accorded by the court must vary from case to case and there is always a spectrum. In sentencing an NSL 21 offence, if the court decides to從輕處罰, it may impose a lighter penalty within either of the Upper Tier or Lower Tier. For serious offences, whatever discount the court may give, the ultimate sentence cannot go below the mandatory minimum of five years in the Upper Tier. In contrast, if the court decides to減輕處罰, it may reduce the penalty from the Upper Tier to the Lower Tier. 70. In enabling the court to 從輕處罰 (imposing a lighter penalty) or 減輕處罰 (reducing a penalty), not only does NSL 33(1) not compromise the Primary Purpose but it is in fact conducive to achieving it, because each of the specified conditions is broadly consistent with such purpose: (1) NSL 33(1)(1) encourages an offender to desist from completing the NSL offence and to take active advance steps to prevent the offence from taking its toll on the victims and society. They are largely in line with the objective of prevention and suppression of NSL offences. (2) NSL 33(1)(2) encourages an offender to so conduct himself to enable the law enforcement agencies to devote more time and resources on investigating and prosecuting other NSL offences. This is largely consistent with the objective of preventing and suppressing NSL offences. (3) NSL 33(1)(3) encourages the offender to inform on other criminals. It is a powerful weapon in the hands of the law enforcement in the fight against NSL offences, serving the aim of prevention, detection and prosecution of such offences. See Z v HKSAR (2007) 10 HKCFAR 183, per Li CJ at [10]. 71. Moreover, among all forms of mitigation, the specification of the three conditions, without more, is intentional. It reflects the drafters’ judgment that they are the only relevant mitigating factors in the context of national security which may allow the court to not only 從輕處罰 (imposing a lighter penalty) but also減輕處罰 (reducing a penalty), as the case may be, without prejudicing the Penological Considerations or compromising the Primary Purpose. The legislative intention is plainly that, as specified conditions in the NSL, they are exhaustive in that sense. 72. However, it is also the legislative intention that local sentencing laws on mitigation are to operate in tandem with the NSL, provided that they do not prejudice the Penological Considerations or compromise the Primary Purpose. So, as rightly accepted by Mr Chau, other mitigating factors not specified in NSL 33(1) but recognised under the common law, such as guilty pleas, can operate in full for the court to consider 從輕處罰, that is, imposing a lighter sentence within the respective range of the Lower and Upper Tier. It is because whatever the discount those mitigating factors may afford, the ultimate sentence stays within the range of either Tier, which still accords with the drafters’ judgment with respect to the gravity of and the Penological Considerations for NSL 21 offences. As such, they do not compromise the Primary Purpose. 73. Those other mitigating factors, however, cannot apply to 減輕處罰, that is, to reduce the sentence below the minimum of five years in the Upper Tier. It is because if they were to apply to discount the sentence below five years, it would contradict the drafters’ judgment with respect to the gravity of serious NSL 21 offences and prejudice the Penological Considerations for such offences, thereby compromising the Primary Purpose in that regard. As such, they are not compatible with either NSL 21 or NSL 33(1). 74. In contending that the Upper Tier only prescribes a range of starting points, Mr Choy referred to some other legislations which adopt the language similar to “shall be sentenced” used in the English translation of the Upper Tier, such as sections 36(2) and (2A) of the Road Traffic Ordinance,[40] and section 51A of the English Firearms Act 1968. He argued that although mandatory language has been used in those legislations, the legislative intention is to set a range of starting points as opposed to a mandatory minimum. 75. Sections 36(2) and (2A) of the Road Traffic Ordinance provide for disqualification periods for conviction of the offence of dangerous driving causing death. Section 36(2) stipulates: “If a court or magistrate convicts a person of an offence under subsection (1), the court or magistrate shall order that the person be disqualified in accordance with subsection (2A) or (2B) unless the court or magistrate for special reasons orders that the person be disqualified for a shorter period or that the person not be disqualified.” Under subsection (2A), the disqualification is, in the case of a first conviction, for a period of not less than 5 years and in the case of a second or subsequent conviction, for a period of not less than 10 years. 76. In HKSAR v Wong Mun Ming [2016] 1 HKLRD 1057, on which Mr Choy relied, the Court of Appeal at [30]-[32] held that the effect of those provisions is to lay down a starting point and it is one from which the judge may depart upon being satisfied of the existence of “special reasons” as defined in the Ordinance. Although it was not so expressed, it is clear that the Court arrived at that conclusion as a matter of construction. 77. In our view, the context and purpose of the Road Traffic Ordinance is entirely different from the NSL. The construction for sections 36(2) and (2A) is irrelevant to the construction of NSL 21. 78. As will be elaborated at [86] below, section 51A of the Firearms Act 1968 involves an entirely different legislative scheme. Its construction is irrelevant, which means that Mr Choy’s reliance on it is also misplaced. 79. Next, Mr Choy argued that NSL 33(1) fortifies his construction of the Upper Tier. Had the five years been a mandatory minimum, it would be unnecessary to provide in NSL 33(1) that the sentence could be reduced in the specifying circumstances. However, Mr Choy’s argument missed the fundamental point that NSL 33(1) exists, among others, for the very purpose of enabling the rigour of the mandatory minimum to be mitigated without compromising the Penological Considerations in furtherance of the Primary Purpose in sentencing serious NSL 21 offences. 80. Mr Choy then argued that NSL 21 operates very much like guideline judgments set by the Court of Appeal in offences of trafficking in dangerous drugs. 81. Most local statutory regimes set the maximum penalty without specifying a minimum, leaving the court to mete out appropriate sentences in different cases. In exercise of its sentencing jurisdiction, the Court of Appeal has handed down guideline judgments, stating tiered-ranges of sentences which reflect gradations in severity to assist the lower courts in sentencing serious offences, such as child pornography;[41] theft or fraud involving a breach of trust;[42] and trafficking in dangerous drugs.[43] Consistency in punishment and treatment is the primary consideration of issuing guideline judgments, which is a fundamental aspect of any rational and fair system of criminal justice: see HKSAR v Godson Ugochukwu Okoro [2019] 2 HKLRD 451, at [81]-[82]. The ranges of sentences effectively set the ranges of starting points for the court to adopt in individual cases. Depending on the circumstances, the ultimate sentence may stay within the applicable range or fall below the minimum of the range. Thus understood, guideline judgments are premised on and operate within sentencing regimes which set the maximum without a minimum. They do not inform the construction of the Upper Tier, a wholly different penalty regime. 82. In sum, Mr Choy’s submissions failed to take account of the context and purpose of the NSL as a whole sufficiently. His construction of the Upper Tier is unsustainable. 83. Turning to NSL 33(1), Mr Choy argued that the drafters of the NSL were fully aware of local sentencing laws when drafting it. It cannot be accidental that the drafters do not univocally state the three specified conditions to be exhaustive. The plain meaning of the language makes it clear that they are not meant to be exhaustive in allowing the court to 減輕處罰, that is, to reduce the penalty of a serious NSL 21 offence. With respect, Mr Choy also failed to construe NSL 33(1) in the context of the NSL as a whole. He also failed to sufficiently recognize that in their application to sentencing such offences, local sentencing laws on mitigation must operate in tandem with NSL 21 and NSL 33(1) in the manner as explained above. His submissions likewise cannot stand. 84. For completeness, we only need to deal with two points taken by Mr Chau in support of his arguments on the construction of NSL 21 and NSL 33(1). 85. First, as he did before the Judge, Mr Chau relied on Jordan. 86. In Jordan, the appellants pleaded guilty to offences of possession of firearms and prohibited firearms under the Firearms Act 1968. Section 51A of the Act provides that: “The court shall impose an appropriate custodial sentence (or order for detention) for a term of at least the required minimum (with or without fine) unless the court is of the opinion that there are exceptional circumstances relating to the offence or to the offender which justifies not doing so.” One of the issues on appeal was whether the court could give discount on guilty pleas which would lower the sentence beyond the mandatory minimum despite section 51A. The English Court of Appeal approached it by ascertaining the legislative intent of the Parliament in enacting section 51A. Noting that the section was plain and unambiguous on its face; that the Parliament was plainly aware of the statutory provisions on how guilty pleas should be regarded in sentencing; and that the absence from section 51A of any reference to those provisions was plainly deliberate, the Court rejected the contention that the Parliament could not have intended to prevent the reduction of a sentence following a guilty plea. They further held that the rigour of section 51A was mitigated by the possibility of exceptional circumstances being found, but they did not regard a guilty plea as an exceptional circumstance. Thus analysed, Jordan, which involves a wholly different legislative scheme, does not assist the construction exercise at hand. 87. Second, Mr Chau sought to rely on the substantive Mainland sentencing law, making reference to a book entitled刑法条文理解适用与司法实务全书,根据刑法修正案1~10编定,第一卷,第四章,第一節刑罚的具体运用.[44] He submitted that since there is a need for convergence between the NSL and local laws, the Mainland sentencing law on the meanings of terms used in the NSL will help the construction exercise. 88. As to that, there was some very brief discussion between the bench and Mr Chau. Unfortunately, Mr Chau failed to properly articulate the basis for his reliance on the Mainland sentencing law, which required clarification. In the process, certain remarks were made by the Court which we now recast to better express our meaning, as follows. Given the special status of the NSL as a national law applied to the HKSAR, regard may properly be had to the relevant Mainland law in aid of its construction. For as pointed out by the Court of Final Appeal in Lai Chee Ying, at [18], according to the Explanation on the “Draft Law of the People’s Republic of China on Safeguarding National Security in the Hong Kong Special Administrative Region” addressed to the 19th Session of the 13th Standing Committee of the National People’s Congress by the Responsible Official of the Legislative Affairs Commission under the Standing Committee of the National People’s Congress (18 June 2020), one of the working principles for drafting the NSL was: “… accommodating the differences between Mainland China and the HKSAR, and striving to address the convergence, compatibility and complementarity between this Law, and the relevant national laws and local laws of the HKSAR.” (Emphasis supplied) Accordingly, the relevant Mainland law may in principle inform the construction of the NSL or a particular NSL provision. As to which particular Mainland law is relevant for the construction exercise, how and to what extent it is relevant, and how to make reference to it must depend on the actual circumstances of the case before the court. 89. As mentioned at the hearing, for the purpose of the present case, the need to refer to or consult the substantial Mainland sentencing law for the construction of NSL 21 or NSL 33(1) does not arise. So we will just leave the materials relied on by Mr Chau, the status of which has not been properly established by him, as they are. E3. Answers to the Core Issues 90. For these reasons, the answer to the 1st Core Issue is that on a proper construction, the legislative intention of NSL 21 in prescribing the range of penalty for serious offences in the Upper Tier is to lay down a mandatory minimum of five years. 91. The answer to the 2nd Core Issue is that NSL 33(1) is, on a proper construction, exhaustive in enabling the court to 減輕處罰, that is to reduce the sentence for a serious NSL 21 offence below the mandatory minimum of five years in the Upper Tier as appropriate. Other mitigating circumstances recognised under local laws including guilty pleas are not applicable for such purpose. F. Grounds 3 to 5 92. In light of the answers on the Core Issues, both Grounds 3 and 4 fail to the extent as discussed above. As to the remainder, none of the complaints is established either. 93. In respect of Ground 3, Mr Choy submitted that on a proper construction of NSL 21 and NSL 33(1), the minimum term can and should be “reduced” such that a starting point can be adjusted downward for individual justice if one of the prescribed circumstances is satisfied. However, the proper approach to sentencing under NSL 21 is for the court to firstly categorize the offence as either serious or minor; and then to determine the penalty within the applicable tier. The adjusting of an otherwise appropriate sentence, if warranted, would take place at the second step. The practical effect of Mr Choy’s submission is to ask the court to go back to the first step and re-categorize a serious case as a minor one. That is wrong in principle. 94. In respect of Ground 4, Mr Choy submitted that the circumstances of the applicant came very close, if not identical to the scenario specified in NSL 33(1)(2) although he did not voluntarily surrender himself. Thus he should be entitled to avail himself of NSL 33(1)(2). However, the language of NSL 33(1)(2) is plain and unambiguous. Voluntary surrender takes place before any appearance of the offender before the court. It would be too far-fetching to suggest that it covers the making of a plea by him at trial. 95. As said, Ground 5 adds nothing to the applicant’s case. It falls together with Grounds 1 to 4. G. Dispositions 96. For the above reasons, the Judge was correct in her categorization of the offence committed by the applicant as serious. Although she erred in relying on Jordan, her conclusion as to the construction of NSL 21 was correct. Further, on a proper construction of NSL 33(1), the applicant’s timely guilty plea could not apply to adjust the ultimate sentence below the minimum of five years in the Upper Tier. 97. In consequence, we uphold the sentence of five years’ imprisonment that the Judge imposed on the applicant and dismiss his application for leave to appeal against sentence. Mr Edwin Choy SC and Mr Kevin Tang, instructed by Wan Yeung Hau & Co., assigned by the Director of Legal Aid, and Mr Jonathan Lin, instructed by Wan Yeung Hau & Co., for the applicant Mr Anthony Chau, DDPP and Mr Ivan Cheung, ADPP (Ag), of the Department of Justice, for the respondent [1] See G.N. (E.) 72 of 2020. [2] This was Charge 3. The other two charges that the applicant faced and pleaded not guilty to were possession of arms without a licence and possession of offensive weapons with intent, contrary to sections 13(1) and (2) of the Firearms and Ammunition Ordinance, Cap 238 and section 17 of the Summary Offences Ordinance, Cap 228 respectively. Pursuant to the parties’ agreement, those two charges were left on the court file and are not to be dealt with unless with leave of the court. [3] Appeal Bundle, at pp 5-43. [4] Its English translation is: “Orange Gear”. [5] Its English translation is: “The Only Remaining Black Clad from Hong Kong”. [6] Its English translation is: “Changed Back to Two Semi-retired Members of the Air-con Valiant”. [7] Its English translation is: “Two valiant members back in full force”. [8] Its English translation is: “Two Valiants’ CPC Elite Team Raptors”. [9] Its English translation is: “only one valiant warrior giving reply at the moment Admin”. [10] Its English translation is: “2 youths who are anti-communism and anti-national security, and are determined to shout Hong Kong independence and Taiwan independence”. [11] The English translation is “Channel of Anti-Communism and Hong Kong Independence”. [12] Its English translation is: “Promising yy”. [13] The NSL applies to the HKSAR from 11 pm on 30 June 2020: see The Promulgation of National Law 2020 made by the Chief Executive. [14] [2022] HKDC 384. [15] Reasons for Sentence, at [42]. [16] Reasons for Sentence, at [41]. [17] Reasons for Sentence, at [34] – [37]. [18] Reasons for Sentence, at [38]. [19] Reasons for Sentence, at [39]. [20] Reasons for Sentence, at [40]. [21] Reasons for Sentence, at [47]. [22] See the transcripts of the hearing on 29 April 2022, Appeal Bundle, at p 83R-T. [23] Reasons for Sentence, at [50]. [24] Reasons for Sentence, at [51]-[54]. [25] The applicant’s application to amend the Perfected Grounds of Appeal was prompted by the Court’s directions dated 20 September 2022 for further written submissions on the construction of NSL 21 and NSL 33. It was not opposed by the respondent and we allowed it at the hearing on 13 October 2022. The amendments are those italicized in Grounds 3 and 4. [26] Leading Mr Kevin Tang and Mr Jonathan Lin. Both Mr Choy and Mr Tang appeared below. [27] Which, as observed in Ma Chun Man, at [72], is similar in nature to an offence of incitement under NSL 21. [28] For the powerful use of social media, see also HKSAR v Chan Yau Hei (2014) 17 HKCFAR 110, at [89]. [29] Annex B, S/N 1167, published at 1744 hours. [30] Annex B, S/N 1138 and 1139 at 2258 and 2358 hours. [31] Annex B, S/N 1140 to S/N 1143 between 2242 and 2348 hours. [32] Together with Mr Ivan Cheung, Assistant Director of Public Prosecutions (Ag). [33] BL 23 provides: “The [HKSAR] shall enact laws on its own to prohibit any act of treason, secession, sedition, subversion against the Central People’s Government, or theft of state secrets, to prohibit foreign political organizations or bodies from conducting political activities in the Region, and to prohibit political organizations or bodies of the Region from establishing ties with foreign political organizations or bodies.” [34] NSL 3(3) provides: “香港特別行政區行政機關、立法機關、司法機關應當依據本法和其他有關法律規定有效防範、制止和懲治危害國家安全的行為和活動。” [35] NSL 8 provides: “香港特别行政區執法、司法機關應當切實執行本法和香港特別行政區現行法律有關防範、制止和懲治危害國家安全行為和活動的規定，有效維護國家安全。” [36] NSL 42(1) provides: “香港特別行政區執法、司法機關在適用香港特別行政區現行法律有關羈押、審理期限等方面的規定時，應當確保危害國家安全犯罪案件公正、及時辦理，有效防範、制止和懲治危害國家安全犯罪。” [37] Cap 200. [38] Pursuant to NSL 62, which provides: “This Law shall prevail where provisions of the local laws of the [HKSAR] are inconsistent with this Law.” [39] How is 免除處罰 understood and applied when read together with local sentencing laws has to await another occasion. [40] Cap 374. [41] Secretary for Justice v Man Kwong Choi [2008] 5 HKLRD 519. [42] HKSAR v Cheung Mee-kiu [2006] 4 HKLRD 776; HKSAR v Ng Kwok Wing [2008] 4 HKLRD 1017. [43] Attorney General v Tuen Shui-ming & Another [1995] 2 HKC 798 (cannabis); R v Lau Tak-ming & Others [1990] 2 HKLR 370 and HKSAR v Abdallah Anwar Abbas [2009] 2 HKLRD 437 (cocaine and heroin); Secretary for Justice v Hii Siew-cheng [2009] 1 HKLRD 1 (ecstasy and ketamine); HKSAR v Ko Wai-shing [2021] 5 HKLRD 724 (GHB and GB); Attorney General v Ching Kwok-hung [1991] 2 HKLR 125 (ice); Attorney General v Chan Chi-man [1987] HKLR 221 (methaqualone and diazepam); R v Lau Yiu-nam [1986] HKLR 964 (opium). [44] Unofficial translation of the title is “The Complete Book on the Understanding, Application and Judicial Practice of the Criminal Law Provisions, prepared according to Amendments 1~10 of the Criminal Law, Volume 1”, Chapter IV: The Special Application of Penalties, Section 1: Sentencing. 
Mr Justice Fok PJ : 1. After hearing counsel for the appellants, the Court did not call on counsel for the respondent and dismissed the appeal with costs, indicating that it would hand down its reasons for doing so in due course. These are the Court’s reasons. 2. The sole issue in this appeal relates to the ownership of 13 parking spaces in an industrial building in Tsuen Wan called Gold King Industrial Building (“the building”).[1] The appellants are the first owners of the building (“the First Owners”) and their agent, Tai Fat Development (Holding) Co Ltd (“Tai Fat”), and the respondent is the Incorporated Owners of the building (“IOs”). The appellants claimed that the 13 parking spaces are reserved for the separate use and ownership of the First Owners and their successors-in-title whereas the IOs contended that the car parking spaces are common parts of the building. 3. This was one of a number of issues, albeit the main issue, between the parties at trial.[2] The issue relating to the 13 parking spaces turned on the construction of a number of documents but principally the Deed of Mutual Covenant dated 16 April 1987 in respect of the building (“the DMC”). The other documents included the First Assignment, also dated 16 April 1987, relating to a unit in the building[3] (“the First Assignment”) and also the Government Grant in relation to the land on which the building was constructed (and in particular Special Conditions 12 and 13 thereof) and the Approved Building Plans of the building dated 18 March 1987. 4. Suffiad J decided the issue in favour of the IOs.[4] He concluded that, on a construction of the DMC, it was clear that the intention of the parties to it was that the 13 parking spaces “were meant to be part of the loading and unloading areas of the building and as such would be part of the common areas of it”. He also rejected the First Owners’ argument based on tenancy by estoppel.[5] That argument was premised on the fact that the IOs had entered into various tenancy agreements in respect of the 13 parking spaces. The last of those tenancy agreements related to 12 of the 13 spaces and was for the period from 1 December 2004 to 30 November 2006. That agreement was, however, only entered into between the IOs and Tai Fat after the IOs’ solicitors had written to assert that the 13 parking spaces in issue were part of the common areas of the building. 5. The First Owners appealed against Suffiad J’s judgment and essentially advanced the same arguments in support of their case as to ownership of the 13 parking spaces. The Court of Appeal dismissed the appeal at the conclusion of the hearing with the reasons for its decision being given subsequently by Barma JA.[6] 6. In the CA Reasons, Barma JA carefully set out the contentions of the First Owners as to why Suffiad J’s construction of the DMC was erroneous[7] but rejected them and concluded[8] that the judge was “clearly right to conclude that the 13 Car Parking Spaces formed part of the common areas of the building”. The First Owners’ arguments were then each separately addressed by Barma JA in the CA Reasons.[9] 7. With one exception, relating to the way he dealt with the First Owners’ case based on estoppel by convention (addressed below), we agree with the reasons given by Barma JA for rejecting the First Owners’ arguments which in turn were consistent with the reasons given by Suffiad J at first instance. It is unnecessary to rehearse all the reasons leading to the conclusion that the 13 parking spaces in question form part of the common areas of the building. The principal reasons, however, are that: (1) The definition of “common areas” in the DMC includes at sub-paragraph (b) “loading and unloading areas” and the 13 parking spaces were listed in the DMC under the heading “Loading and Unloading Areas” in the First Schedule to the DMC; (2) Only ownership shares were issued in respect of the 13 parking spaces but not management shares, thus suggesting they were intended for common ownership rather than private ownership (since, if the latter, management fees on the basis of management shares would be expected to be payable); (3) The reservation of the 13 parking spaces would be necessary to ensure compliance with Special Conditions 12 and 13 of the Government Grant imposing requirements as to the minimum space for the parking, loading and unloading of vehicles; (4) As shown by the approved building plans, there would be considerable practical difficulties if the 13 parking spaces were not common areas and this fact militates in favour of the IOs’ construction of the DMC. 8. The First Owners’ arguments advanced in its printed case on appeal before this Court, being in substance a repetition of its arguments below, do not affect the conclusion reached by the Court of Appeal affirming that of the judge below. The First Owners’ argument that the reservation of the car parking spaces by the vendors in the First Assignment means that it was intended that the 13 parking spaces would not form common areas but would be reserved for private ownership does not overcome the above points. On the terms of the First Assignment, it is to the DMC that one looks in order to ascertain what parts of the building form its common areas. We are satisfied that the Court of Appeal correctly construed the DMC and did not err by ignoring terms in it as contended by the First Owners. We are also satisfied that, as part of the factual matrix in which to construe the DMC, the Court of Appeal was entitled to have regard to the issue of practical difficulty[10] that would be caused to occupants of the building if the car parking spaces were not common areas. 9. The exception referred to above relates to the First Owners’ reliance on the principle of estoppel by convention. Barma JA said that the estoppel by convention argument was not open to the First Owners because it had not been pleaded or investigated at the trial.[11] This is not correct since the defence was pleaded[12] and it was not dealt with separately by Suffiad J. 10. At trial, as noted above, Suffiad J considered a related but separate pleaded defence of tenancy by estoppel and concluded[13] that this argument did not provide the First Owners with a defence to the IOs’ claim that the 13 parking spaces are common areas. His conclusion was based on the provisional conclusion of the Court of Appeal in an earlier summary judgment appeal in these proceedings (reported at [2007] 4 HKLRD 440) and it is not necessary to consider that particular argument further since it was not pursued by the First Owners on appeal. 11. What the First Owners do maintain is that the principle of estoppel by convention avails them in asserting their title to the 13 parking spaces. However, reliance on this principle does not assist the First Owners. Estoppel by convention is concerned with a common assumption relied upon as a basis upon which the persons sharing such common assumption enter into a transaction or legal relationship: First Laser Ltd v Fujian Enterprises (Holdings) Co Ltd (2012) 15 HKCFAR 569 at [79(6)], citing Unruh v Seeberger (2007) 10 HKCFAR 31 at [142]. It is also necessary for the party relying on the principle to show that it would be unjust for the other party to attempt to depart from the common assumption and that he would suffer detriment arising from his having entered into the relevant transaction on the basis of the common assumption if the other party were permitted to set up rights inconsistent with that assumption: First Laser Ltd v Fujian Enterprises (Holdings) Co Ltd (2012) 15 HKCFAR 569 at [79(7)], citing Unruh v Seeberger (2007) 10 HKCFAR 31 at [150]. 12. Here, the First Owners’ pleaded case relates to (a) the letting of the 13 parking spaces from 1987 to the manager of the building and then from 1998 to the IOs, and (b) the further letting of 12 of those spaces after the IOs had, through their solicitors, disputed the ownership of those spaces. However, although the IOs had claimed the repayment of the rent paid in respect of the 13 parking spaces, Suffiad J dismissed that part of the IOs’ claim and accordingly, the First Owners are unable to show any detriment arising from the transaction or transactions which it is said were entered into on the basis of the common assumption. For this reason alone, the plea of estoppel by convention must fail. 13. On behalf of the First Owners, Mr KM Chong[14] sought to advance an additional argument in his oral submissions, namely that the Court of Appeal should not have given weight to the matter of practical convenience in relation to the layout of the 3rd floor of the building because the designation of the one parking space on the 2nd floor of the building for private ownership gave rise to the same or similar difficulties. This argument is lacking in logic since the 2nd floor layout simply highlights the practical inconvenience of the 3rd floor layout on the First Owners’ case. In any event, being a new point not raised below it also faced the insuperable difficulty that it could not satisfy the conditions in Flywin Co Ltd v Strong & Associates Ltd (2002) 5 HKCFAR 356. 14. For these reasons, we dismissed the appeal with costs to the respondent. Postscript 15. The Court understands that this appeal is the last appeal for which leave to appeal was granted “as of right”,[15] the Court of Appeal’s judgment under appeal being dated 27 August 2014 and so before the coming into effect of the Administration of Justice (Miscellaneous Provisions) Ordinance 2014 on 24 December 2014. If this is indeed the last of the “as of right” appeals, this category of case can now be consigned to the footnotes of legal history since the arguments in favour of the abolition of appeals to this Court “as of right”,[16] fully applicable to the present appeal for which leave to appeal would not otherwise have been granted, have happily now led to the repeal of s.22(1)(a) of the Hong Kong Court of Final Appeal Ordinance.[17] Mr K.M. Chong and Mr Alvin Chong, instructed by Kenneth C.C. Man & Co., for the Plaintiff by original action & the 1st Defendant by counterclaim (1st Appellant) and the 2nd – 4th Defendants by counterclaim (2nd – 4th Appellants) Mr Edward Chan SC, Mr Andrew Mak and Ms Carol Wong, instructed by Chung & Kwan, for the Defendant by original action & the Plaintiff by counterclaim (Respondent) [1] The particular parking spaces in issue are those numbered 1 and 2 on the 1st Floor and those numbered 4, 5, 11, 12, 13, 14, 20, 36, 37, 38 and 39 on the 3rd Floor. [2] In HCA 104/2006 before Suffiad J. [3] Being Unit D on the 19th Floor of the building. [4] HCA 104/2006, Judgment dated 15 January 2014 (“CFI Judgment”), at [85]. [5] CFI Judgment at [90]. [6] CACV 22/2014 (Lam VP, Kwan and Barma JJA), Reasons for Judgment dated 23 December 2014 (“CA Reasons”). [7] At [13]. [8] At [23]. [9] At [29] to [34]. [10] Which Barma JA carefully analysed in the CA Reasons at [22]. [11] CA Reasons at [34]. [12] At paragraph 10 of the Re-Amended Reply and Defence to Counterclaim. [13] CFI Judgment at [90]. [14] Appearing with Mr Alvin Chong. [15] Pursuant to (the now repealed) section 22(1)(a) of the Hong Kong Court of Final Appeal Ordinance (Cap.484). [16] See e.g. China Field Ltd v Appeal Tribunal (Buildings) (No.1) (2009) 12 HKCFAR 68 at [13] to [18], Chinachem Charitable Foundation Ltd v Chan Chun Chuen (2011) 14 HKCFAR 798 at [20] and [107], and Wealth Duke Ltd v Bank of China (Hong Kong) Ltd (2011) 14 HKCFAR 863 at [1]. [17] By section 8(1) of the Administration of Justice (Miscellaneous Provisions) Ordinance 2014. 
Chief Justice Cheung: 1. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Mr Justice Fok PJ. Mr Justice Fok PJ: A. Introduction 3. This appeal arises out of the prosecution of the appellant, then a member of the Legislative Council (“LegCo”), for an alleged offence of contempt during a committee meeting. It raises the question of the extent to which a member of LegCo may be subject to criminal prosecution for disorderly conduct interrupting proceedings. The Court has previously examined the question of the liability of persons acting disruptively in the public gallery[1] and that of persons acting violently against police officers in the precincts of LegCo,[2] but not that raised in this appeal. A.1 The alleged facts 4. The facts alleged to have occurred and on which the prosecution of the appellant was based were simple. On 15 November 2016, LegCo’s Panel on Housing and its Panel on Development held a joint meeting. The meeting was attended by the appellant, then a member of the Panel on Housing, and Mr Ma Siu-cheung, the then Under Secretary for Development, as well as others. Mr Ma had with him a folder containing confidential documents, which he placed on the bench in front of him. In the course of the meeting, after he had asked Mr Ma to produce certain documents relating to an issue under discussion, the appellant rose from his seat, walked over towards Mr Ma, snatched his folder and then passed it to another member of LegCo in attendance, Mr Chu Hoi-dick, for him to read. The Chairperson of the meeting issued repeated verbal demands to the appellant for him to return the folder to Mr Ma and return to his seat. Eventually, the Chairperson ordered the appellant to withdraw from the meeting pursuant to Rule 45(2) of the Rules of Procedure and temporarily suspended the meeting. Mr Chu, having read the contents of Mr Ma’s folder, eventually returned it to a security guard and it was then given back to Mr Ma. A.2 The prosecution and proceedings below 5. Following this incident, with the consent of the Secretary for Justice,[3] the appellant was prosecuted[4] for the offence of contempt under section 17(c) of the Legislative Council (Powers and Privileges) Ordinance (Cap.382) (“LCPPO”), which provides that: “Any person who – … (c) creates or joins in any disturbance which interrupts or is likely to interrupt the proceedings of the Council or a committee while the Council or such committee is sitting, commits an offence and is liable to a fine of $10,000 and to imprisonment for 12 months, and in the case of a continuing offence to a further fine of $2,000 for each day on which the offence continues.” 6. At the request of the defence, the magistrate gave a ruling on preliminary issues regarding the ambit and constitutionality of section 17(c). In particular, she was asked to rule on whether the section applies to a member of LegCo and to proceedings other than those involving a person called to give evidence under oath.[5] She was also asked to rule on whether the provision is unconstitutional in violating “the freedom of speech and debate of the member and/or LegCo’s exclusive authority over its internal affairs enshrined in the Basic Law”.[6] 7. The magistrate ruled that, on its true interpretation, section 17(c) “is applicable to the proceedings of [LegCo] or a committee in general but it is not applicable to the members of [LegCo].”[7] Having so ruled, she did not deal with the issue of constitutionality. 8. The prosecution appealed by way of Case Stated[8] with questions as to the correctness of the magistrate’s ruling and also raising the constitutional validity of section 17(c). By order of a judge of the Court of First Instance, the Case Stated was reserved for the consideration of the Court of Appeal.[9] 9. In a comprehensive judgment, Poon CJHC (with whom Lam VP[10] and D Pang JA agreed) held that the magistrate was: (1) wrong to rule that conduct caught by section 17(c) is protected by the privilege in section 3 of the LCPPO (see below); (2) wrong to hold that section 17(c) is inapplicable to members of LegCo; and (3) right to find that section 17(c) applies to all proceedings of LegCo and its committees and is not confined to proceedings involving the taking of evidence. He also held against the argument that section 17(c) is unconstitutional.[11] B. The issues in this appeal 10. The Appeal Committee granted leave to appeal to the appellant in respect of the following questions of law, namely: “(1) Whether what is said and done during proceedings of the Legislative Council (“LegCo”) and its committees falls within the privilege enjoyed by LegCo provided that it does not amount to an ordinary criminal offence. (‘Question 1’) (2) Upon the true interpretation of the Legislative Council (Powers and Privileges) Ordinance, Cap 382 (“LCPPO”): (a) Whether section 17(c) applies to a member of LegCo. (‘Question 2(a)’) (b) Whether ‘proceedings’ in section 17(c) is confined to proceedings involving the taking of evidence under oath. (‘Question 2(b)’) (3) Whether the Court of Appeal was correct in respect of the following: (a) That the absolute freedom of speech and debate of LegCo under section 3 of the LCPPO does not extend to conduct caught by section 17(c) of the LCPPO. (‘Question 3(a)’) (b) That LegCo had, by enacting section 17(c) of the LCPPO, in fact relinquished to and/or conferred upon the courts a penal jurisdiction over the conduct and discipline of members of LegCo during the proceedings of LegCo and its committees. (‘Question 3(b)’) (c) That section 17(c) of the LCPPO as applicable to a member of LegCo is not unconstitutional. (‘Question 3(c)’).” 11. Despite the breadth of the questions for which leave to appeal was sought and granted, in light of the parties’ submissions in their respective written cases and at the hearing before this Court, the issues on this appeal have been more focused. The appellant contends that, as a matter of law, he is not subject to liability under section 17(c) of the LCPPO because of, first, the constitutional protection of freedom of speech and debate or proceedings in LegCo, and/or, second, the principle of non-intervention by the courts in the conduct of LegCo proceedings. The first contention involves construing section 17(c) of the LCPPO and will require consideration of Article 77 of the Basic Law (“BL77”) and sections 3 and 4 of the LCPPO (set out below). The second contention requires consideration of this Court’s decision in Leung Kwok Hung v President of the Legislative Council (No.1) (2014) 17 HKCFAR 689. 12. In advancing the appeal, Lord Pannick QC[12] draws attention to three particular aspects of the appellant’s case, namely that (i) the appellant was a member of the LegCo Panel on Housing, (ii) his impugned conduct was in relation to the business of the joint meeting of the two LegCo Panels, and (iii) the alleged offence is inextricably linked to the proceedings of the LegCo Panels and is not an “ordinary crime”. C. Whether appellant immune from prosecution by reason of freedom of speech and debate in LegCo C.1 Appellant’s conduct prima facie caught by section 17(c) 13. It is to be noted that not every interruption to proceedings constitutes an offence under section 17(c), which only penalises those interruptions which are the result of a disturbance to the relevant proceedings. The dictionary definition of “disturbance” is relevant and includes: “1. The interruption and breaking up of a … proper functioning; … an instance of … a breach of the public peace … 2. Interference with the continuance of any action or process. … 4. Interference with rights or property.”[13] 14. Accordingly, the offence is committed when a defendant creates a disturbance which interrupts or breaks up the proper functioning of LegCo or its committees and, in particular, occurs when the resulting interruption involves interference with the rights of others. 15. The wording of section 17(c) refers to “any person” and, in context and having regard to the purpose of the LCPPO (see below), there is no reason to exclude a LegCo member from its ambit. A compelling contextual argument for this is the presence, in section 20 of the LCPPO,[14] of the qualification “other than a member or officer of the Council”. Where the LCPPO is not intended to apply to a member of LegCo, it says so. The expression “any person” is also used in section 19 of the LCPPO (creating offences of interfering with members, officers and witnesses),[15] without the section 20 qualification, and, if that were read as not applying to LegCo members, would, like section 17(c), produce anomalous results. Given the statutory purpose of the LCPPO, there is no good reason to exclude a LegCo member from liability for interfering with other members or officers or witnesses. Similarly, as Lord Pannick QC accepted, there is no good reason to exclude a LegCo member, in circumstances where they are not protected by the freedom of speech and debate conferred on them, from liability for contempt constituted by the creation of a disturbance interrupting proceedings. 16. On the facts set out in Section A.1 above, which have yet to be established, the appellant was prima facie caught by section 17(c) of the LCPPO. The joint panel meeting was a proceeding of a committee of LegCo.[16] The appellant was a person who evidently created a disturbance which interrupted the proceedings of that committee while it was sitting. The central question in this appeal, to which I shall now turn, is whether the appellant’s conduct fell within any privilege conferred on members of LegCo. C.2 The argument in reliance on the protection of freedom of speech and debate 17. The first and principal reason advanced by the appellant for not being liable under section 17(c) on the facts as alleged involves reliance on the privileges conferred by law in respect of statements made at meetings of LegCo and its committees. Specifically, the appellant relies on: (1) BL77, which provides: “Members of the Legislative Council of the Hong Kong Special Administrative Region shall be immune from legal action in respect of their statements at meetings of the Council.” (2) LCPPO section 3, which provides: “There shall be freedom of speech and debate in the Council or proceedings before a committee, and such freedom of speech and debate shall not be liable to be questioned in any court or place outside the Council.” (3) LCPPO section 4, which provides: “No civil or criminal proceedings shall be instituted against any member for words spoken before, or written in a report to, the Council or a committee, or by reason of any matter brought by him therein by petition, Bill, resolution, motion or otherwise.” 18. The appellant also contended that, since section 3 of the LCPPO is closely modelled on Article 9 of the Bill of Rights 1689,[17] it is relevant to have regard to English cases concerned with Article 9,[18] including the UK Supreme Court’s interpretation of that provision in R v Chaytor [2011] 1 AC 684. That case concerned the prosecution of members of the UK Parliament for submitting false claims for expenses and allowances. The UK Supreme Court held that the members in question were not immune from prosecution by reason of Parliamentary privilege because the submission of claims was not part of the proceedings in Parliament to which the privilege extended. 19. It was submitted on behalf of the appellant that the ambit of protection conferred by section 3 of the LCPPO (as under Article 9 of the Bill of Rights 1689 on which it is modelled) is “the core or essential business of Parliament [here LegCo], which consists of collective deliberation and decision making”[19] and that the prosecution in the present case would involve the criminal court in addressing “the proceedings in Parliament [that is LegCo]”[20] and “the legislative or deliberative processes of”[21] of LegCo. The appellant’s prosecution in this case would require the court to assess the conduct of the LegCo Panel meeting and what a member of LegCo had done during that meeting in the course of the collective deliberations, and whether this conduct was a disturbance which interrupted the business of LegCo, or was likely to do so. 20. Therefore, it was submitted, so long as the appellant’s alleged conduct occurred during the business of the LegCo Panel meeting, section 3 of the LCPPO (and likewise BL77) confers an absolute privilege and grants him immunity from prosecution even if his conduct was such as to fall within the terms of section 17(c). C.3 The ambit of a LegCo member’s freedom of speech and debate and the correct interpretation of section 17(c) 21. The appellant’s argument requires acceptance of the proposition that the conduct of the appellant constituted an exercise of the protected freedom of speech and debate. 22. Whilst it may be correct to say that section 3 of the LCPPO is modelled on Article 9 of the Bill of Rights 1689 and that the former, like the latter, confers an absolute privilege which cannot be waived,[22] it nevertheless remains for the Court to determine, on its true construction, where the boundaries of section 3 lie. This necessarily also entails a consideration of section 17(c) of the LCPPO, which is part of the context of section 3, and the question of how, as a matter of statutory construction, the two sections intersect. 23. The proposition that it is for the courts to determine the scope of the legislature’s privilege is clearly laid down in Leung Kwok Hung v President of the Legislative Council (No.1) (2014) 17 HKCFAR 689 at [39]-[43]. In particular, that case establishes that “… in the case of a written constitution, which confers law-making powers and functions on the legislature, the courts will determine whether the legislature has a particular power, privilege or immunity” (ibid. at [39]). Similarly, the UK Supreme Court affirmed in R (Miller) v Prime Minister [2019] 3 WLR 589 at [66], by reference to R v Chaytor [2011] 1 AC 684, that “it is for the court and not for Parliament to determine the scope of Parliamentary privilege, whether under article 9 of the Bill of Rights or matters within the ‘exclusive cognisance of Parliament’; …”. 24. Adopting the well-established principles of statutory construction[23] of examining the language of the relevant provisions by reference to their context and purpose, one notes that section 3 of the LCPPO is not in identical terms to Article 9 of the Bill of Rights 1689.[24] The latter extends its protection to “proceedings in Parliament”, as well as freedom of speech and debates, and the English cases emphasise that it protects what is “said or done within the walls of Parliament”.[25] In the LCPPO, the privilege or immunity applies to “speech and debate” (section 3) and “words spoken … or written” (section 4) in LegCo, and is reflected in the immunity for “statements” in BL77. This Court has, of course, previously acknowledged that freedom of expression embraces, as one of its dimensions, the manner in which an individual wishes to express their views and is therefore not limited to spoken or written words.[26] At the same time, however, section 17(c) provides a criminal offence of contempt for interruptive disturbances (see [13] and [14] above) and so the question of whether any particular conduct falls within the protected freedom of speech and debate or not must depend on a proper construction of the relevant provisions of the LCPPO as a whole. 25. The statutory purpose is the starting point of any such construction. The LCPPO was enacted in 1985 in anticipation of the resumption of the exercise of sovereignty over Hong Kong by the People’s Republic of China. It was recognised that the powers and privileges of the former colonial legislature would cease to have effect after 30 June 1997 and so the then Hong Kong Government proposed the Bill which became the LCPPO in order to provide a statutory footing for LegCo’s management of its own affairs, effective investigatory powers and its powers and privileges. That legislative scheme included the creation of offences and penalties, over which jurisdiction was conferred on the courts. This statutory purpose is reflected in the debates in LegCo second reading and committee stage of the Bill, including in particular the speeches of the Chief Secretary moving its second reading.[27] The enactment of the LCPPO in July 1985, included sections 3, 4 and 17(c), as well as other provisions, and preceded the drafting of the Basic Law. There is no suggestion that the Basic Law was intended either to depart from, or extend, the powers and privileges of LegCo in the LCPPO (as noted above, BL77 only immunises “statements at meetings of the Council”). 26. The statutory purpose of the LCPPO, reflected in its Long Title,[28] has more recently been explained in the judgment of Ribeiro PJ in HKSAR v Fong Kwok Shan Christine, as including the provision of: “… a statutory framework aimed at creating a secure and dignified environment in the LegCo complex conducive to the legislature carrying out its constitutional functions at its sittings without disruption or disturbance, while permitting members of the public to observe the proceedings within the Chamber as an open legislative process”.[29] 27. Construing the statutory privilege of free speech and debate in LegCo contextually and purposively, I would reject the appellant’s argument that his impugned conduct fell within the protection of free speech and debate relied upon. The LCPPO is to be construed as a coherent whole with sections 3 and 4 having to be read in context together with other provisions including section 17(c). As the Court of Appeal observed (at [42]): “Protection of the core legislative and deliberative business in terms of free speech and debate in the Council and proceedings in a committee is conferred by sections 3 and 4. Together with other privileges and immunities, they aim at enabling LegCo to carry out its functions independently and without outside interference. The provisions regulating admittance, etc and for offences, including section 17(c) aim at maintaining the secure and dignified environment that LegCo needs to carry out its functions.” 28. The protection of freedom of speech and debate in LegCo is self-evidently an important right. It enables members of LegCo to advocate opinions freely and robustly and without inhibition due to the fear of legal proceedings for such speech and debate. It would be a significant inroad into that freedom if a member of LegCo were subject to legal proceedings for things said by him in the course of sometimes heated political debate. Equally, as the passage quoted in the preceding paragraph demonstrates, the provisions regulating admission and creating offences are designed to achieve the statutory purpose of creating a secure and dignified environment conducive to the legislature carrying out its constitutional functions at its sittings without disruption or disturbance. 29. Accepting the appellant’s broad argument in the present case that, merely because he was present at, and had been participating in, a committee meeting of LegCo, he had absolute immunity for his actions however and whenever occurring and even if they amounted to a disruption caught by section 17(c), would be to extend the privilege of free speech and debate beyond the purpose for which it is granted. 30. In the present case, in my view, his conduct did not fall within the speech and debate protected by sections 3 or 4 of the LCPPO or BL77. At the meeting in question, if the prosecution’s case is established, the appellant created a disturbance by the act of crossing the floor of the chamber during a debate and snatching property belonging to someone else which he passed to a third party over the owner’s objections. He thereby interfered with the rights of Mr Ma, to whom as a public officer the privileges and immunities enjoyed by LegCo members are also extended,[30] including his privacy rights in relation to the confidential documents in his file. The appellant acted in breach of the LegCo rules and ignored the Chairperson’s repeated demands that he resume his seat and return the folder to Mr Ma. This led to the suspension of the meeting whilst the disruptive consequences of the appellant’s actions were addressed. By his actions, the appellant created a disturbance which interfered with the ability of other members of LegCo to carry out their proper functions. In doing so, he was not making a speech, nor was he participating in debating any business that was before the meeting. 31. Concluding that the appellant’s conduct, falling as it did within section 17(c), did not come within the protection of sections 3 and 4 or BL77 is not to say that the freedom of speech and debate in LegCo is qualified, in the same way as, for example, the generally applicable rights of free speech and assembly under Articles 16 and 17 of the Hong Kong Bill of Rights. As noted above, the Court has acknowledged that freedom of expression embraces the manner in which an individual expresses their views. In conveying information and ideas an individual member might well conduct themself in a manner which falls within the freedom of speech and debate conferred. On the other hand, whilst the limits of the freedom are widely drawn and properly described as absolute, conduct which does not form part of any speech or debate in LegCo falls outside the section 3 privilege. Such conduct which creates a disturbance constituting an interruption to proceedings interfering with the proper functioning of LegCo or its committees, and in particular where it interferes with the rights of others, may attract liability under section 17(c). There may be cases where it is more difficult to see the division between conduct which falls within the protection of speech and debate and that which does not. This is not such a case. Here, it is plain that the appellant was not engaged in speech and debate in LegCo when he conducted himself in the manner alleged to have created a disturbance. It follows that his conduct, prima facie contrary to section 17(c), is not protected by the privilege conferred by sections 3 and 4 or BL77. D. Whether appellant immune from prosecution by reason of the non-intervention principle 32. The appellant’s second ground for contending that he is immune from prosecution for the alleged offence involves the proposition that the Court should refrain from exercising criminal jurisdiction over contempts of LegCo under section 17(c) of the LCPPO on the basis that taking jurisdiction would offend the non-intervention principle which is said to derive from the doctrine of separation of powers as affirmed by this Court in Leung Kwok Hung v President of the Legislative Council (No.1).[31] 33. Whilst recognising that the exclusive competence of Parliament (or, in Hong Kong, LegCo) to manage its own affairs without interference from the courts or others can be waived or relinquished,[32] it was submitted on behalf of the appellant that: (1) given the constitutional significance of the principle, clear words would be required to conclude that exclusive competence had been waived; (2) section 17(c) says nothing express about exclusive competence and cannot be read as displacing that principle in the context of an ordinance which includes section 3; and (3) there is scope for section 17(c) to operate even if it does not extend to the conduct of a member of LegCo. 34. The non-intervention principle was stated in Leung Kwok Hung v President of the Legislative Council (No.1) as: “… the principle that the courts will recognise the exclusive authority of the legislature in managing its own internal processes in the conduct of its business, in particular its legislative processes. The corollary is the proposition that the courts will not intervene to rule on the regularity or irregularity of the internal processes of the legislature but will leave it to determine exclusively for itself matters of this kind (the non-intervention principle).”[33] 35. That case concerned the question of whether the court should exercise its powers of judicial review regarding the regularity or otherwise of the President of LegCo’s decision to curtail the time for debate and to bring a long filibuster to an end. This was clearly a matter involving the internal processes of the legislature. The present case is entirely different. In exercising jurisdiction in respect of the appellant’s prosecution under section 17(c), the court is carrying out its judicial function of applying primary legislation enacted by LegCo itself. There is no issue of separation of powers. LegCo has carried out its constitutionally allotted legislative function of enacting the offence provision conferring jurisdiction on the courts and the courts carry out their constitutionally allotted adjudicative function in trying prosecutions for the offence so enacted. Contrary to the appellant’s arguments, the non-intervention principle has no application. 36. Regardless of whether or not conduct caught by section 17(c) would previously have been or is still susceptible to internal LegCo disciplinary proceedings, by enacting section 17(c) as primary legislation, LegCo has deliberately vested the courts with the criminal jurisdiction thereby created. To the extent that it might have been arguable that the conduct was subject to regulation as part of LegCo’s internal processes, LegCo has waived any exclusivity in its disciplinary jurisdiction and conferred penal powers on the courts. LegCo itself has never claimed criminal jurisdiction (as opposed to disciplinary jurisdiction) in respect of misconduct committed inside LegCo. Insofar as there is now overlapping jurisdiction over misconduct of the type covered by section 17(c), the fact that internal LegCo disciplinary proceedings might have been commenced in respect of a particular incident will be a relevant factor in any decision of the Secretary for Justice to grant consent pursuant to section 26 for the institution of a prosecution under the LCPPO in respect of the same incident.[34] 37. None of the appellant’s arguments in support of the contention that LegCo has not waived its exclusive competence over members of LegCo for disturbances falling within section 17(c) are convincing reasons for holding that the non-intervention principle applies here. The wording of section 17(c) and its application to members of LegCo is clear and unambiguous. Part IV of the LCPPO, including section 17(c), is part of the “statutory framework aimed at creating a secure and dignified environment in the LegCo complex conducive to the legislature carrying out its constitutional functions at its sittings without disruption or disturbance, while permitting members of the public to observe the proceedings with the Chamber as an open legislative process” (see [26] above). Excluding members of LegCo from the operation of section 17(c) would be anomalous in providing them with an immunity from prosecution for behaviour which, on any view, is inimical to the legislative and deliberative business of LegCo. E. Conclusions and Disposition 38. For the reasons set out above: (1) Where, by conduct not forming part of any speech or debate, a LegCo member has created a disturbance which interrupts proceedings within the meaning of section 17(c), the freedom of speech and debate conferred on him by sections 3 and 4 of the LCPPO or BL77 does not provide the member with an immunity from prosecution for the offence of contempt under section 17(c). (2) The non-intervention principle does not require that the courts refuse to exercise criminal jurisdiction over a member of LegCo in a prosecution under section 17(c). (3) It is not necessary to address more specifically the separate questions for which leave to appeal was granted. In particular, it was not argued on behalf of the appellant that the Court of Appeal was wrong to conclude that proceedings in section 17(c) were confined to proceedings involving the taking of evidence under oath (Question 2(b)). Nor did the appellant pursue the argument that the Court of Appeal was wrong in holding that section 17(c) is not unconstitutional (Question 3(c)). 39. Accordingly, I would dismiss the appeal. The appellant is not immune from prosecution for the alleged offence and the courts are not precluded from exercising jurisdiction in respect of the charge. Mr Justice Chan NPJ: 40. I agree with the judgment of Mr Justice Fok PJ. Lord Reed of Allermuir NPJ: 41. I agree with the judgment of Mr Justice Fok PJ. Chief Justice Cheung: 42. Accordingly, the appeal is unanimously dismissed. Lord Pannick QC and Ms Margaret Ng, instructed by Ho Tse Wai & Partners, assigned by the Director of Legal Aid, Mr Carter Chim and Mr Jason Lee, instructed by Ho Tse Wai & Partners, on a pro bono basis, for the Appellant Mr Johnny Mok SC, instructed by the Department of Justice, and Ms Vinci Lam SC, DDPP, Mr William Liu, DLO (Civil Law) (Ag.) and Mr Antony Leung, SPP, of the Department of Justice, for the Respondent [1] HKSAR v Fong Kwok Shan Christine (2017) 20 HKCFAR 425. [2] HKSAR v Leung Hiu Yeung (2018) 21 HKCFAR 20. [3] Required by section 26 of the LCPPO. [4] ESS 16969/2017, before Ms Ada Yim Shun-yee, Ag Principal Magistrate. [5] Ruling on Preliminary Issues, 5 March 2018, Amended Question One. [6] Ibid., Amended Question Two. [7] Ibid. at [32]. [8] Pursuant to section 105 of the Magistrates Ordinance (Cap.227). [9] Pursuant to section 118(1)(d) of (Cap.227). [10] Now Lam PJ. [11] [2020] HKCA 424, HCMA 520/2018, Judgment dated 2 June 2020 (“CA Judgment”) at [11] and [83]. [12] Appearing with Ms Margaret Ng, Mr Carter Chim and Mr Jason Lee. [13] Shorter Oxford English Dictionary (6th Ed., 2007) Vol.1, p.722. [14] Section 20 provides: “Any person, other than a member or officer of the Council, who – (a) enters or attempts to enter the Chamber or the precincts of the Chamber in contravention of any of the Rules of Procedure or any resolution under section 8(2); or (b) contravenes any administrative instructions issued under section 8(3), or any direction given thereunder, regulating the admittance of persons to or the conduct of persons within the Chamber or the precincts of the Chamber, commits an offence and is liable to a fine of $2,000 and to imprisonment for 3 months.” [15] Section 19 provides: “Any person who – (a) assaults, obstructs or molests any member going to, being within or going from the precincts of the Chamber, or endeavours to compel any member by force or menace to declare himself in favour of or against any motion or matter pending before the Council or a committee; or (b) assaults, interferes with, molests, resists or obstructs any officer of the Council while in the execution of his duty; or (c) tampers with, deters, threatens, molests or in any way unduly influences any witness in regard to any evidence to be given by him before the Council or a committee; or (d) threatens, molests or in any way punishes or injures or attempts to punish or injure any person for having given evidence before the Council or a committee or on account of any evidence which he has given before the Council or a committee, commits an offence and is liable to a fine of $10,000 and to imprisonment for 12 months.” [16] Section 2 of the LCPPO defines “committee” as “(a) a standing or select committee or any other committee of the Council; (b) a subcommittee of any committee referred to in paragraph (a).” [17] This provides: “That the Freedome of Speech and Debates or Proceedings in Parlyament ought not to be impeached or questioned in any Court or Place out of Parlyament.” [18] Reference was made to Bradlaugh v Gossett (1884) 12 QBD 271, Stockdale v Hansard (1839) 9 Ad & E 1, and Chenard v Arissol [1949] AC 127. [19] [2011] 1 AC 684 per Lord Phillips of Worth Matravers PSC at [62]. [20] Ibid. [21] [2011] 1 AC 684 per Lord Rodger of Earlsferry JSC at [122]. [22] [2011] 1 AC 684 per Lord Phillips of Worth Matravers PSC at [61]. [23] See e.g. Chan Ka Lam v Country and Marine Parks Authority [2020] HKCFA 33, (2020) 23 HKCFAR 414 at [26]-[27]. [24] “That the Freedome of Speech and Debates or Proceedings in Parlyament ought not to be impeached or questioned in any Court or Place out of Parlyament.” [25] Bradlaugh v Gossett (1884) 12 QBD 271 at p.275, R v Chaytor [2011] 1 AC 684 at [29], [52] and [54]. [26] HKSAR v Fong Kwok Shan Christine (2017) 20 HKCFAR 425 at [42], citing Mayor of London v Hall [2011] 1 WLR 504 per Lord Neuberger of Abbotsbury MR (as he then was) at [37]. [27] Such statements being admissible to ascertain statutory purpose: see HKSAR v Cheung Kwun Yin (2009) 12 HKCFAR 568 at [13]-[14], HKSAR v Li Kwok Cheung George (2014) 17 HKCFAR 319 at [37]. [28] “To declare and define certain powers, privileges and immunities of the Legislative Council and of the members and officers thereof and of the Chief Executive and public officers designated by him in relation to attendance at sittings of the Legislative Council and committees thereof; to secure freedom of speech in the Legislative Council; to make provision for regulating admittance to and conduct within the precincts of the Chamber of the Legislative Council; to provide for the giving of evidence in proceedings before the Legislative Council or committees thereof, and for offences in respect of such proceedings and related matters; and for purposes incidental to or connected therewith.” [29] (2017) 20 HKCFAR 425 at [82]. [30] By section 8A of the LCPPO. [31] (2014) 17 HKCFAR 689. [32] R v Chaytor [2011] 1 AC 684 per Lord Phillips of Worth Matravers PSC at [63]. [33] (2014) 17 HKCFAR 689 at [28]. [34] CA Judgment at [72], referring to R v Chaytor [2011] 1 AC 684 at [81]. 
Chief Justice Ma: 1. I agree with the judgment of Madam Justice McLachlin NPJ. Mr Justice Ribeiro PJ: 2. I agree with the judgment of Madam Justice McLachlin NPJ. Mr Justice Fok PJ: 3. I agree with the judgment of Madam Justice McLachlin NPJ. Mr Justice Cheung PJ: 4. I agree with the judgment of Madam Justice McLachlin NPJ. Madam Justice McLachlin NPJ: A. Introduction 5. The appellant was convicted of trafficking in a dangerous drug and sentenced to 8½ years in prison after a trial before a jury. The appellant was absent for a portion of the trial due to a medical condition. The issue on this appeal is whether the trial judge’s refusal to adjourn the trial while he sought medical help for a painful dental problem requires that his conviction be set aside and a new trial ordered. 6. I conclude that while the trial judge should have granted the brief adjournment requested for the appellant to seek medical attention, the appellant’s absence from a portion of the trial did not render his trial unfair, having regard to the entirety of the proceedings. Accordingly, I would dismiss the appeal. B. Facts and Judicial Proceedings B.1. Events Leading to the Arrest 7. In August 2016, PC 4543 (“PW1”) and PC 9687 (“PW2”) were on plain clothes patrol on the street. PW1 spotted the appellant who was accompanied by a female friend, Ms Chan. PW1 thought the appellant was acting suspiciously because of the way he was walking and holding his hands in front of his crotch area as if to conceal something in his clothing. PW1 asked PW2 to join him in intercepting the appellant and Ms Chan. PW1 touched the crotch area of the appellant and felt something hard. The police officers took the appellant to a nearby disabled toilet for a further search, which revealed drugs hidden in the appellant’s underpants. PW1 cautioned the appellant, who made certain admissions. The appellant says that PW2 induced him to admit possession of the drugs by promising that the appellant would only be charged for possession of dangerous drugs and Ms Chan would be released (“inducement”). He was taken to the police station, where he signed a post-record of his verbal admissions given in the disabled toilet and made further admissions in a video-recorded interview (“confession”). He was then charged with trafficking in a dangerous drug, contrary to section 4(1)(a) and (3) of the Dangerous Drugs Ordinance, Cap 134. B.2. The Trial 8. The trial began with a voir dire to determine whether the confession was voluntary and hence admissible. The appellant alleged that the confession was inadmissible because of the inducement. On day 2 of the voir dire, while the prosecution’s first witness, PW1, was still testifying, the appellant indicated that he was unwell with a swollen and infected gum and a headache, for which he had taken painkillers. The trial judge responded that the trial could continue in his absence because he was represented by counsel. The appellant’s counsel replied that this was an option but indicated that after completion of the evidence of PW1, he would seek the trial judge’s further direction. At 11:06am, while PW1 was still being cross-examined, the appellant informed the trial judge that he felt dizzy and requested to leave but the proceedings could continue in his absence. The trial judge agreed, saying he could be brought back when he felt fit. 9. Shortly after the appellant left the courtroom, at 11:29am, the court adjourned for the mid-morning break. After the break, the appellant’s counsel informed the trial judge that the appellant felt a bit better but would go to see the doctor at Lai Chi Kok Reception Centre in the afternoon. The trial judge responded, “we’ll continue this afternoon without him”. The transcript then reveals the following exchange: “MR. LO: Yes. Well, but, your Ladyship, is it I … COURT: That’s perfectly proper. He is not well. There is no reason why he should be here. You said you had full instructions. We can move on. Otherwise, unless he is dead or dying, he stays. So he has a choice. MR. LO: Your Ladyship, but … COURT: I can’t see that he is in any prejudice. It’s a voir dire. You have full instructions. MR. LO: Okay.” After the conclusion of the cross-examination of PW1 at 1:03pm, immediately before the lunch adjournment, the appellant’s counsel renewed his concern: “MR. LO: Your Ladyship, my latest instruction from my client is still – is that he would like to go back to the Lai Chi Kok for … COURT: To be then … MR. LO: Yes. COURT: Well, then we will … MR. LO: But I – your Ladyship, your stance is clear and I know it’s hard to convince your Ladyship otherwise. However, I urge you to – I urge your Ladyship to reconsider whether it is the best interest to the defendant if he is – if … COURT: He is well-represented, Mr Lo. He has given full instructions. Your case has been put. It has been reduced to your grounds of objection. If he is that unwell, then of course he must go back. If he feels he needs to be here, then he can choose but as far as I am concerned, there will be no prejudice to your client whatsoever being represented by you and your instructing solicitor. This isn’t the trial. This is – before a jury at this stage. This is a voir dire. MR. LO: Yes, but such decision would affect to some certain extent his – the trial. COURT: No, Mr, I’m not with you on that. He has a choice. It’s up to him to decide what he wishes to do. You’re more than competent to deal with this. MR. LO: Well, I’m … COURT: He hasn’t with the main witness given you other instructions during your cross-examination. You’re well prepared. MR. LO: All I could say, your Ladyship, is I’m flattered. COURT: And I hope he manages to get some sort of medication which will make him better. MR. LO: Grateful, your Ladyship. COURT: And we’ll see what the doctor has to say. 2:30, gentlemen.” 10. At 2:30pm, the trial resumed in the absence of the appellant. The appellant’s counsel informed the trial judge that the appellant had gone back to Lai Chi Kok Reception Centre to consult a doctor as he had a dental problem with tremendous pain, and it was anticipated that the appellant could be present in court the next morning. He reapplied for a short adjournment to the next day. The trial judge stated the proceedings would continue and told the appellant’s counsel not to try again as she had already made her ruling and decision. The appellant was absent from the voir dire for the whole afternoon and thus not present for the examination in chief and cross-examination of PW2, who had allegedly given the inducement that made the confession involuntary. Having sought medical attention, the appellant returned to the trial the next morning. B.3. The Court of Appeal 11. The appellant applied for leave to appeal against his conviction to the Court of Appeal. Unrepresented, he complained of the refusal of the trial judge to adjourn his case to allow his counsel to prepare for the voir dire, and of the trial judge’s failure to provide adequate jury directions on the possibility of hiding the dangerous drugs inside his underpants. The application was dismissed. 12. In an Addendum to its judgment, the Court of Appeal noted that the clerk’s log revealed that the appellant had claimed he was ill on day 2 of the voir dire but that the trial had continued that afternoon in his absence. However, since it had already issued an order dismissing his application for leave to appeal, the court viewed itself as functus officio and did not deal with this matter or its significance. This is now before us. B.4. The Present Appeal 13. The Appeal Committee of this Court granted leave to appeal on the substantial and grave injustice ground, namely that it was reasonably arguable that the appellant was deprived of a fair trial by the trial judge’s refusal to grant a short adjournment due to the appellant’s stated inability to attend court by reason of illness and instead, continuing with the trial in his absence. C. The Law 14. Article 11(2)(d), section 8 of the Hong Kong Bill of Rights Ordinance, Cap 383,provides that in the determination of any criminal charge against him, a person shall be entitled to be present at his trial: “Article 11 Right of persons charged with or convicted of criminal offence … (2) In the determination of any criminal charge against him, everyone shall be entitled to the following minimum guarantees, in full equality— … (d) to be tried in his presence …” 15. This well-established right allows the accused to see and hear the case against him, confront his accusers, and give prompt and continuous instructions to his legal representatives. The presence of the accused may also place psychological pressure on witnesses to tell the truth, enhancing public confidence that justice has been done. See R v David Jeremy Taylor, Tina Kaisa Paivarinta-Taylor [2008] EWCA Crim 680 at [16]: “...The right to be present at court is vital. It is not only a right so that the defendant can participate by giving instructions and following the trial, but also so that he can see and hear the witnesses who are giving evidence against him. This is not only part of the process by which a defendant participates at trial, but is also a sanction by which the public can be confident that there are pressures on those witnesses who give adverse evidence against a defendant to tell the truth. To give an account of events in a witness statement or within the seclusion of a police station is one thing. It is quite another to have to stand up in public, face-to-face with a defendant, and repeat those assertions. …” 16. The right to be present at one’s trial is part of the broader right of everyone charged with a criminal offence to a fair trial.[1] It is not an absolute right. The trial judge has a discretion to allow the trial to proceed in the accused’s absence in appropriate circumstances. As Lord Bingham said in R v Jones (Anthony), [2003] 1 AC 1 (“Jones”), at [6]: “... The presence of the defendant has been treated as a very important feature of an effective jury trial. But for many years problems have arisen in cases where, although the defendant is present at the beginning of the trial, it cannot (or cannot conveniently or respectably) be continued to the end in his presence. This may be because of genuine but intermittent illness of the defendant … . In all these cases the court has been recognised as having a discretion, to be exercised in all the particular circumstances of the case, whether to continue the trial … . The existence of such a discretion is well established … . But it is of course a discretion to be exercised with great caution and with close regard to the overall fairness of the proceedings; a defendant afflicted by involuntary illness or incapacity will have much stronger grounds for resisting the continuance of the trial than one who has voluntarily chosen to abscond.” 17. The authorities[2] establish a number of factors that should be considered in determining whether a trial should be continued in an accused’s absence. They include the following: - Was the accused’s absence voluntary or involuntary? Where an accused is absent because of illness, the absence is generally treated as involuntary. - Has the accused waived the right to be present at his trial? This is a question of fact to be determined in all the circumstances. - Would an adjournment resolve the problem of the accused’s absence? If so, would the adjournment required be short or long? Would an adjournment impact negatively on the conduct of the trial, for example, the effect of delay on the memory of witnesses? - Is the accused legally represented? If so, to what extent are his legal representatives able to receive and act upon instructions in his absence? - Would the accused be prejudiced by his absence, having regard to the nature of his defence and the evidence against him? - Would there be a risk of the jury reaching an improper conclusion about the accused’s absence? 18. The factors to be considered cannot be reduced to an exhaustive or conclusive checklist. It is clear, however, that a judge in exercising the discretion should carefully consider all the relevant circumstances arising in the case at hand with the “overriding concern … to ensure that the trial, if conducted in the absence of the [accused], will be as fair as circumstances permit and lead to a just outcome.”[3] It is also clear that the judge should proceed with “utmost care and caution”[4]. The decision to proceed with a trial in the absence of an accused should only be in “rare and exceptional cases”[5], and if an accused is absent because of illness “it would very rarely, if ever, be right to exercise the discretion in favour of commencing the trial, at any rate unless the defendant is represented and asks that the trial should begin.”[6] Indeed, it has been said that the discretion “should be sparingly exercised and never if the accused’s defence will be prejudiced by his absence.”[7] 19. Where an appellate court concludes that the discretion has been exercised improperly, the remaining question is whether, viewing the trial as a whole, the absence of the accused rendered the trial unfair. See Jones.[8] 20. As I earlier stated, the right to be present at one’s trial is part of the broader fair trial right. Not every departure from the norm will require a new trial. As the proviso in section 83(1) of the Criminal Procedure Ordinance, Cap 221 states: “…the Court of Appeal may, notwithstanding that it is of opinion that the point raised in the appeal might be decided in favour of the appellant, dismiss the appeal if it considers that no miscarriage of justice has actually occurred.” While this Court is not bound to apply the proviso, the principle enshrined is nevertheless inherent in the Court’s function to do justice. As Bokhary PJ observed in Tang Siu Man v HKSAR [1998] 1 HKLRD 350, at p.379I-J, “To allow an appeal just because something has gone wrong at the trial even though it has not resulted in a miscarriage of justice would not be doing justice.” D. Application of the Law 21. The first question is whether the trial judge in this case exercised her discretion to order the trial to continue in the appellant’s absence correctly. I conclude she did not. 22. The trial judge did not consider the factors referred to in the cases in a careful and systematic way. She did not allude to the authorities that suggest great caution in ordering a trial to proceed where an accused cannot be present because he is genuinely ill. Most importantly, she very quickly concluded that there would be no prejudice to the accused because the matter at that stage was the voir dire, not the trial proper; and because the accused had instructed his counsel. 23. The trial judge’s suggestion that the accused’s presence is less necessary at a voir dire to determine the admissibility of a confession than at the trial proper was, with respect, misplaced. The issue of the admissibility of the confession was crucial to the appellant’s defence. 24. The trial judge’s suggestion that the appellant had a choice to stay or leave is also questionable. The appellant’s counsel advised the trial judge that the appellant was in great pain. To suggest the appellant had a real choice of whether or not to avail himself of his right to be present at the trial when he was in great pain was unrealistic and goes against the weight of authorities, which have consistently treated absence due to genuine illness as involuntary. 25. Similarly, the trial judge’s assumption that the appellant would not be prejudiced because he had instructed his counsel, and his objections to the admissibility of the confession had been reduced to grounds of objections filed with the court, while relevant, does not fully answer the appellant’s concern. One of the purposes of giving the accused the right to be tried in his presence is to allow him to instruct his counsel on the basis of what he hears the witnesses testifying against him say. The trial judge seems to have assumed that, because the appellant did not give his counsel instructions in relation to PW1, he also would not do so for PW2 and accordingly would not be prejudiced by being absent for the testimony of PW2. Yet the two officers played different roles in the arrest, and PW2 was the officer alleged to have made the vitiating inducement. 26. Critically, the trial judge appears not to have taken into account the fact that the adjournment would have been brief: for the afternoon only. When the brevity of the proposed adjournment is weighed against the importance of the appellant’s right to be present at all stages of his trial and his involuntary absence, the cautious approach required by the authorities suggests that the better course here would have been to adjourn the proceedings until the next morning. 27. The remaining question is whether the trial, considered as a whole, was fair, notwithstanding the appellant’s absence during PW2’s evidence. The appellant argues that the admissibility of the confession was important, and suggests that his absence may have prejudiced the possibility of having the confession declared inadmissible. He argues that because of the trial judge’s ruling, he was deprived of the opportunity to confront the most crucial witness of the voir dire in person – a witness whose evidence was consequently given free of pressure from the physical presence of the appellant and resulted in the admission of the confession. He says that this in itself constituted irremediable prejudice, rendering other considerations irrelevant. 28. The respondent’s counsel counters that the court must view the proceedings as a whole and ask whether the appellant’s trial was fair. He submits that the appellant’s absence from the courtroom for PW2’s evidence did not prejudice him and thus did not render his trial unfair. The respondent points to nine considerations. 1. The appellant had experienced counsel throughout the trial and the voir dire. 2. The appellant’s counsel agreed with the trial judge that he had full instructions. 3. The appellant’s counsel had filed detailed written grounds of objection to the confession, containing all conceivable challenges. 4. The alleged inducement by PW2 was made in the presence of PW1, who testified as to how it was not made in the appellant’s presence. 5. The appellant before leaving the courtroom knew that the trial would proceed in his absence and probably instructed his counsel over the lunch break before leaving to seek medical attention. (However, there is no evidence of what was discussed and the appellant was in pain.) 6. The absence of the appellant was for a short period of time. 7. The appellant did not attempt to recall PW2 at the end of the voir dire. 8. The appellant testified in the voir dire and had meaningful participation. 9. The voir dire was recanvassed in the trial proper. The evidence presented on the voir dire was again presented on the trial proper before the jury. (Although it can be countered that at this stage the damage had been done, in that the confession has been admitted and the jury had seen it.) 29. I agree with the respondent’s submission that the proceedings must be viewed as a whole and reject the appellant’s submission that his absence, without more, constituted an irremediable prejudice of his right that rendered the trial unfair. While some of the respondent’s points are stronger than others, I find it impossible, viewing the proceedings as a whole, to conclude that the appellant’s absence from the trial for the afternoon during which PW2 testified on the voir dire rendered his trial unfair. When asked by this Court whether additional or different questions would have been put to PW2 had the appellant been present, the appellant’s counsel did not identify any. He also rightly conceded that the appellant’s presence was only one facet of a fair trial and that other considerations listed by the respondent’s counsel also played a part. I find that any actual or perceived prejudice or unfairness by reason of the appellant’s absence could be, and was in fact, counterbalanced by other safeguards that the appellant enjoyed in his trial. 30. The trial judge should have granted a brief adjournment of the trial for the afternoon while the appellant sought medical assistance. However, in the circumstances of this case, it cannot be said that continuing the trial in his absence prejudiced his defence and thus rendered the trial unfair. Nor is this a case where the trial judge’s improper exercise of discretion would in itself bring the administration of justice into disrepute, requiring a new trial on that ground alone – something the respondent admitted could occur. While the trial judge’s “dead or dying” comment was insensitive and inappropriate, it was isolated, and offset by her stated belief that the appellant would not be prejudiced by continuing the trial that afternoon in his absence. 31. I conclude that the appellant has failed to show prejudice resulting in an unfair trial. In the words of the Hong Kong Court of Final Appeal Ordinance, Cap 484, he has not shown that a substantial and grave injustice has been done to him. 32. I would dismiss the appeal. Mr James Tze, instructed by H.Y. Leung & Co. LLP and assigned by the Director of Legal Aid, and Mr Tom Cheung, instructed by H.Y. Leung & Co. LLP on a pro bono basis, for the Appellant Mr Ira Lui SADPP (Ag.) and Mr Andy Lo SPP, of the Department of Justice, for the Respondent [1] Article 87 of the Basic Law and Article 10 of the Hong Kong Bill of Rights Ordinance. [2] Jones at [13] – [15] approving the Court of Appeal’s list of relevant factors (except “the seriousness of the offence”) in R v Hayward [2001] QB 862, applied in HKSAR v Kwan Wai Keung & Others (CACC 259/2011, unreported) at [6] and HKSAR v Singh Gursewak [2015] 1 HKC 498 at [30]. [3] Jones at [14]. [4] Jones at [13]. [5] R v Hayward at [22(4)]. [6] Jones at [13]. Although this is couched in the context of commencing trial in the accused’s absence, Lord Bingham also specifically recognised that the same discretion exists regardless of whether the question is one of commencing or continuing a trial in the accused’s absence, at [10]. [7] R v Pearson, The Independent25 February 1998 CA, [1998] Lexis Citation 3879 at p.13, citing R v Howson (1982) 74 Cr App R 172 at p.179. [8] Although their Lordships diverged on the question whether or not the lower court was correct in finding that the accused had waived his right to be present at his trial by absconding, the disagreement did not have any bearing on the unanimous decision to dismiss the appeal on the basis that the accused had, in all the circumstances, “enjoyed his Convention right to a fair trial”: [15]. 
Mr Justice Chan, Acting CJ: 1. The freedom of expression may take many forms. As was involved in the present case, they include the freedom of speech, the freedom of assembly and the freedom of demonstration. The right to the freedom of expression is guaranteed by art 27 of the Basic Law and art 17 of the Bill of Rights. This is a fundamental right to enable any person to air his grievances and to express his views on matters of public interest. In a free and democratic society, there are bound to be conflicts of interest and differences in opinion. It is important that those who purport to exercise the right to the freedom of expression must also respect the rights of others and must not abuse such right. Conflicts and differences are to be resolved through dialogue and compromise. Resorting to violence or threat of violence or breach of the peace in the exercise of this right will not advance one’s cause. On the contrary, this will weaken the merits of the cause and result in loss of sympathy and support. The means to achieve a legitimate end must not only be peaceful, it must also be lawful. Violent or unlawful means cannot justify an end however noble. It may also attract criminal liability. 2. The right to freedom of expression, like all fundamental rights and freedoms, must be given a generous interpretation. (See Ng Ka Ling & others v Director of Immigration (1999) 2 HKCFAR 4.) But this right is not absolute and may be subject to restrictions as prescribed by law in the interest of public order and for the protection of the rights and freedoms of others. See art 39(2) of the Basic Law and art 17 of the Bill of Rights. Any law which purports to restrict such right must be narrowly interpreted and the restriction must satisfy the test of necessity and proportionality. See Yeung May Wan & others v HKSAR (2005) 8 HKCFAR 137 and Leung Kwok Hung v HKSAR (2005) 8 HKCFAR 229 (which were cases on the right of assembly and the right of demonstration). In construing the relevant statutory provisions which have the effect of restricting such right, the court must have regard to competing public interests, including the maintenance of public order and the rights and freedoms of others. The right balance has to be struck between the preservation of public order and the exercise of the individual’s rights and freedoms and between the competing rights and freedoms of individuals or groups of individuals. 3. As described in Mr Justice Tang PJ’s judgment and shown on the videos, the conduct of the appellants was disgraceful. They showed absolutely no regard or respect for the rights of others and had abused the right to freedom of expression. The 1st appellant’s conduct was, to say the least, distasteful and the 2nd appellant’s conduct constituted a breach of the peace and an assault. Although they could have been charged with or guilty of some other offences, the question in these appeals is, however, whether their conduct also incurred criminal liability under s.17B of the Public Order Ordinance, Cap 245. 4. There is no challenge on the constitutionality of s.17B. It is clear from the language of this provision that s.17B(2) is aimed at preventing the outbreak of public disorder and s.17B(1) at protecting the rights and freedoms of others in transacting their normal business. 5. Both s.17B(1) and s.17B(2) require proof of disorderly conduct: “acts in a disorderly manner” in s.17B(1) and “behaves … in a disorderly manner” in s.17B(2). Although the 2nd appellant does not dispute his conduct was disorderly, the 1st appellant argues that what he did was not. I must say that this term has caused me some difficulty. I was initially attracted to Mr Justice Tang’s analysis. But having considered the matter further, I would, with respect, prefer to adopt the approach taken by Mr Justice Ribeiro. 6. There is no definition in the statutory provision and no comprehensive definition by any court of this term “acts/behaves in a disorderly manner”. Section 17B had its origin in s.5 of the Public Order Act 1936 in the UK. In Brutus v Cozens [1973] AC 854, which dealt with the meaning of “insulting words” in that section, Lord Reid said (at p.861) that the meaning of an ordinary word of a statute is a question of fact for the trial court. This approach was adopted by the Divisional Court in Chambers v DPP [1995] Crim L R 896 which held that “disorderly behavior” were to be treated as words in ordinary everyday use. This was followed by Beeson J in HKSAR v Cheng Siu Wing [2003] 4 HKC 471 who held that these words “are to be treated as words in everyday use and given their normal meaning”. It was also accepted that the disorderly conduct in s.5 of the 1936 Act (s.17B of our Ordinance) need not involve any element of violence (Chambers v DPP) or amount to a breach of the peace (Campbell v Adair [1945] JC 29). 7. In New Zealand, disorderly behavior in a public place is an offence under s.4(1)(a) of the Summary Offences Act 1981 albeit punishable only with a maximum fine of $1000. Another section, s.3, creates a more serious offence which requires an element similar but not exactly the same as that in our s.17B(2). The Supreme Court of New Zealand in Brooker v Police [2007] 3 NZLR 91 held that disorderly behavior in s.4(1)(a) (the lesser offence) means behavior seriously disruptive of public order. This was followed in Morse v Police [2012] 2 NZLR 1. One would expect that the Supreme Court of New Zealand would give the same meaning to “disorderly behavior” in the more serious s.3 offence (similar to our s.17B(2)). It is however important to note that it was considered necessary by the Supreme Court of New Zealand to set a high threshold for disorderly behavior as an offence. 8. Unlike the position in New Zealand, it is not an offence in Hong Kong to behave in a disorderly manner in public. To constitute an offence under s.17B(1), an accused must have acted in a disorderly manner for the purpose of preventing the transaction of the business of a public gathering and for an offence under s.17B(2), he must have behaved in a disorderly manner either with the intent to provoke a breach of the peace or that a breach of the peace is likely to be caused by his conduct. The reasons in the New Zealand cases for imposing a higher threshold including constitutionality considerations do not necessarily apply with the same force in Hong Kong. 9. Further, not only is disorderly behavior by itself not an offence, s.17B(2) refers also to other types of behavior such as using, distributing, displaying writing containing threatening, abusive or insulting words. These other types of behavior do not necessarily involve a serious disruption of public order. Also, neither s.17B(1) nor 17B(2) requires proof that there is an actual serious disruption of public order, only that the disorderly behavior was done with the intent to provoke a breach of the peace or had the likely effect of causing a breach of the peace. It is unlikely and I do not believe that the legislature intends by “acts/behaves in a disorderly manner” in s.17B to mean conduct causing a serious disruption of public order. 10. Hence, I am more inclined to follow the English authorities and accept that the term “acts/behaves in a disorderly manner” should be given an ordinary and everyday meaning and that whether there is disorderly conduct for the purpose of s.17B is a question of fact for the trial court having regard to the nature and manner of the conduct and the circumstances of the case. As Gleeson CJ said in Coleman v Power (2004) 220 CLR 1, at para. 12: “Concepts of what is disorderly, or indecent, or offensive, vary with time and place, and may be affected by the circumstances in which the relevant conduct occurs.” 11. The s.17B(2) offence also requires the prosecution to prove that the accused in behaving in a disorderly manner has the intent to provoke a breach of the peace or that a breach of the peace is likely to be caused by his conduct. With regard to this element of the offence, I would like to make the following observations. For the sake of brevity, I would use the term disorderly conduct to include also using, distributing or displaying any writing containing threatening, abusive or insulting words. 12. First, as pointed out earlier, this offence is aimed at preventing an outbreak of public disorder. Although in most cases, the accused’s disorderly conduct would probably have constituted a breach of the peace or some offence already, it is not necessary to show that his conduct has actually provoked or caused others to commit a breach of the peace. It is sufficient to show that there is a real or imminent risk that others would breach the peace or resort to violence as a result of the accused’s disorderly conduct. 13. Secondly, the fact that the accused’s disorderly conduct is a breach of the peace is not sufficient to establish the s.17B(2) offence. The language of the section clearly contemplates a breach of the peace by a person or persons other than the accused. (See Secretary for Justice v Chiu Hin Chung [2013] 1 HKLRD 227.) The first limb requires proof of an intent to provoke a breach of the peace and this clearly does not refer to the accused’s own breach of the peace, although one cannot rule out the possibility that his breach of the peace was committed with the requisite intention to provoke others to breach the peace. The second limb is equally clear: it refers to the likely effect of the disorderly conduct on others. As McCulloch J in Marsh v Arscott (1982) 75 Cr App Rep 211 at 216 said of a provision similar to s.17B(2): “This section is describing breaches of the peace which are brought about, or are likely to be brought about, by other words or behavior occurring earlier, although usually not very long before. The phrase “whereby a breach of the peace is likely to be occasioned” indicates that Parliament was concerned with cause and effect i.e. with conduct which is likely to bring about a breach of the peace and not with conduct which is itself a breach of the peace and no more.” (emphasis added) 14. Thirdly, for the first limb of the s.17B(2) offence, it is necessary to consider the subjective intent of the accused and for the second limb, to assess objectively the likely effect of the disorderly conduct. The court has to examine all the circumstances of the case, including the conduct of the accused, the nature and manner of such conduct, the presence of other persons, such as persons holding opposing views, and the likely reaction of those present (see Viscount Dilhorne in Brutus v Cozens at p.865). The accused’s knowledge of such circumstances is also relevant to the issue of his intention. 15. Fourthly, the accused’s disorderly conduct may be aimed at another person or a group of persons or simply those who are present within the sight and hearing of his conduct. There might be persons who would not be affected but there might be others who would be easily provoked into violent retaliation. The accused has to take his “audience” or target as he finds him. (See Jordan v Burgoyne [1963] 2 QB 744.) There were suggestions in some cases that trained police officers are unlikely to be provoked to commit a breach of the peace (see Marsh v Arscott (1982) 75 Cr App R 21, Coleman v Power (2004) 220 CLR 1, and R v Li Wai Kuen (1973-1976) HKC 346) and hence if no other person is present or has been provoked, no offence under s.17B(2) has been committed. While this may be the case generally, I do not think one should rule out the possibility that even a trained officer might in some situations be provoked to react with violence to the disorderly conduct. It is interesting to note that Huggins J in Li Wai Kuen was careful in adding that “I would not suggest that there could never be a case where the abuse was so gross that even a police officer might be likely to be provoked into violence retaliation, particularly if the language used were also threatening …” . 16. With regard to the s.17B(1) offence, I would agree with the analyses in the judgments of Mr Justice Ribeiro and Mr Justice Tang and have nothing useful to add. 17. For the reasons given by Mr Justice Ribeiro and Mr Justice Tang, I would agree that the prosecution in this case had failed to substantiate the two charges against the appellants. I too would allow the appellants’ appeals and dismiss the prosecution’s cross appeal. Mr Justice Ribeiro PJ: 18. The appellants were demonstrating against fare increases on the Mass Transit Railway. They attended at a prize-giving ceremony organized by the MTR in a public square at the conclusion of a sporting event that it had sponsored. The appellants did not content themselves with displaying placards and chanting protest slogans, as other demonstrators were doing. They decided to invade the podium where those officiating at the prize-giving ceremony were gathered. The appellants were charged with offences under sections 17B(1) and 17B(2) of the Public Order Ordinance.[1] I gratefully adopt the detailed account of the facts set out in the judgment of Mr Justice Tang PJ. 19. The public order offences in question can, of course, be committed without engaging any human rights issues. However, in the present case, the charges relate to conduct in the course of a demonstration and the argument has proceeded on the footing that the appellants’ conduct must be judged in the context of the constitutional rights of peaceful assembly and freedom of demonstration. 20. Thus, although the constitutional validity of sections 17B(1) and 17B(2) has not been challenged and the disposal of the appeal turns on whether, in the events which have happened, the requisite elements of those offences have been proved against each appellant, it is necessary to consider the scope and limits of those constitutional rights and how they relate to cases like the present. A. The offences and the decisions below 21. Section 17B is headed “Disorder in Public Places” and provides as follows:- (1) Any person who at any public gathering acts in a disorderly manner for the purpose of preventing the transaction of the business for which the public gathering was called together or incites others so to act shall be guilty of an offence and shall be liable on conviction to a fine at level 2 and to imprisonment for 12 months. (2) Any person who in any public place behaves in a noisy or disorderly manner, or uses, or distributes or displays any writing containing, threatening, abusive or insulting words, with intent to provoke a breach of the peace, or whereby a breach of the peace is likely to be caused, shall be guilty of an offence and shall be liable on conviction to a fine at level 2 and to imprisonment for 12 months. 22. The Magistrate, Mr Marco Li,[2] convicted both appellants under section 17B(2) and sentenced them each to 14 days’ imprisonment, granting them bail pending appeal. He did not consider it necessary to deal with the charges under section 17B(1) but indicated that he would equally have found them guilty under that section. 23. Their appeal to the Court of First Instance was heard by Barnes J,[3] who quashed their convictions under section 17B(2). However, her Ladyship found them guilty under section 17B(1) instead. In place of the sentence of imprisonment imposed by the Magistrate, she fined the 2nd appellant $3,000 and the 1st appellant $2,000. 24. The appellants were granted leave to appeal by the Appeal Committee on 17 November 2012.[4] B. The right of peaceful assembly and freedom of demonstration 25. Hong Kong residents are justly proud of their tradition of peaceful demonstration. Marches involving tens or even hundreds of thousands of demonstrators have frequently been held without a single incident of public disorder or damage to property. B.1 Article 17 of the Bill of Rights and Article 27 of the Basic Law 26. The right to hold such demonstrations is a fundamental constitutional right. Article 17 of the Bill of Rights,[5] which is given constitutional status by Article 39 of the Basic Law, states: “The right of peaceful assembly shall be recognized. No restrictions may be placed on the exercise of this right other than those imposed in conformity with the law and which are necessary in a democratic society in the interests of national security or public safety, public order (ordre public), the protection of public health or morals or the protection of the rights and freedoms of others.” 27. And Article 27 of the Basic Law states: “Hong Kong residents shall have freedom of speech, of the press and of publication; freedom of association, of assembly, of procession and of demonstration; and the right and freedom to form and join trade unions, and to strike.” 28. It must be read together with Article 39 which relevantly provides: “The rights and freedoms enjoyed by Hong Kong residents shall not be restricted unless as prescribed by law. Such restrictions shall not contravene the provisions of the preceding paragraph of this Article.” 29. The “preceding paragraph” referred to in Article 39 is the paragraph which accords constitutional status to the ICCPR as enacted in the Hong Kong Bill of Rights Ordinance.[6] Article 39 therefore endorses the right of peaceful assembly provided for by Article 17 of the Bill of Rights subject to the restrictions mentioned above. 30. As this Court has held,[7] there is no substantive difference between the right of peaceful assembly guaranteed by Article 17 of the Bill of Rights and freedom of demonstration enshrined in Article 27 of the Basic Law. I shall accordingly refer to the rights interchangeably and focus in this judgment on the text of Article 17 as setting out the scope and limits of the right. B.2 The importance of those rights 31. The importance of the right to demonstrate, and of the closely related freedom of expression, is well-recognized. As this Court has acknowledged, they are rights which lie at the heart of Hong Kong’s system. They guarantee freedoms which are of cardinal importance for the stability and progress of society – freedoms which promote the resolution of conflicts, tensions and problems through open dialogue and debate.[8] 32. The Court has emphasised that such fundamental rights must be given a generous interpretation so as to give individuals their full measure, and that restrictions on such rights must be narrowly interpreted.[9] B.2 The limits of those rights 33. Demonstrators are therefore free to assemble and to convey views which may be found to be disagreeable, unpopular, distasteful or even offensive to others and which may be critical of persons in authority. Tolerance of such views and their expression is a hallmark of a pluralistic society. At the same time, it must be recognized that those freedoms are not absolute and demonstrators must ensure that their conduct does not go beyond the constitutional limits of those rights. 34. Such limits are set by the combined effect of the constitutional provisions themselves and compatible laws enacted by the legislature and developed at common law. Basing itself on the text of Article 17, the Court held in Leung Kwok Hung v HKSAR[10] that: “The exercise of the right of peaceful assembly, whether under the Basic Law or under BORO, may be subject to restrictions provided two requirements are satisfied: (1) The restriction must be prescribed by law (the ‘prescribed by law’ requirement). (2) The restriction must be necessary in a democratic society in the interests of national security or public safety, public order (ordre public), the protection of public health or morals or the protection of the rights and freedoms of others (the necessity requirement).” No issue arises in the present case in relation to the “prescribed by law requirement”.[11] 35. It is established that the “necessity requirement” involves the application of the proportionality test[12] which enables a proper balance to be struck between the interests of society on the one hand and the individual’s right of peaceful assembly on the other.[13] Accordingly, limits may be set to the right of demonstration by validly promulgated laws provided that the restrictions they impose are for a legitimate purpose; are rationally connected with the pursuit of such legitimate purpose; and are no more restrictive than necessary to achieve that legitimate purpose.[14] 36. In relation to the right of peaceful assembly, the Court has held that Article 17, reflecting Article 21 of the ICCPR, exhaustively lists the only legitimate purposes for which restrictions on that right may be imposed.[15] Thus, restrictive laws may legitimately have as their purpose the safeguarding of national security, public safety, or public order (ordre public); the protection of public health or morals; or the protection of the rights and freedoms of others. Such laws are valid if they pass the proportionality test. 37. While it is conceivable that each one of those legitimate purposes might, in given circumstances, be relevant to setting limits on the freedom to demonstrate, for the purposes of this appeal, laws aimed at the maintenance of public order and laws aimed at protecting the rights and freedoms of others are of the most immediate relevance. 38. Article 17 allows a line to be drawn between peaceful demonstrations (where, as noted above, full rein is given to freedom of expression) and conduct which disrupts or threatens to disrupt public order, as well as conduct which infringes the rights and freedoms of others. In Leung Kwok Hung v HKSAR,[16] having recognized that the interests of “public order (ordre public)” are listed by Article 17 as a legitimate purpose, the Court held that there is no doubt that such concept “includes public order in the law and order sense, that is, the maintenance of public order and the prevention of public disorder”.[17] It concluded that a statutory scheme giving the Commissioner of Police discretion to regulate public processions with a view to maintaining public order was constitutionally valid after severance of certain objectionably vague words.[18] 39. Once a demonstrator becomes involved in violence or the threat of violence – somewhat archaically referred to as a “breach of the peace” – that demonstrator crosses the line separating constitutionally protected peaceful demonstration from unlawful activity which is subject to legal sanctions and constraints. The same applies where the demonstrator crosses the line by unlawfully interfering with the rights and freedoms of others. 40. The law therefore imposes bounds on the constitutionally protected activity of peaceful assembly. The need for such limits is sometimes dramatically illustrated in situations involving demonstrations and counter-demonstrations. It is not uncommon for one group, demonstrating in favour of a particular cause, to find itself confronted by another group demonstrating against that cause. The situation may be potentially explosive and the police will generally try to keep them apart. Obviously, if both remain within their lawful bounds, all will be well. But often, conflict and public disorder may result. Sometimes, both sides will have broken the law. But in some cases, the disruption of public order is caused only by one side. The task of the law enforcement agencies and the courts is then to identify the source of such disruption by identifying the demonstrators who have crossed the line into unlawful activity. They thereby avoid curtailing or punishing the constitutionally protected activities of the innocent group. 41. Such a situation arose in the classic common law case of Beatty v Gilbanks,[19] where charges of unlawful assembly were brought against the Salvation Army since they had gone ahead with their procession knowing from past experience that an opposing group calling themselves the “Skeleton Army” was likely to cause a breach of the peace. The Court dismissed those charges. It was not the Salvation Army whose conduct posed a threat to public order and their right of peaceful assembly was upheld. In contrast, in Wise v Dunning,[20] a Protestant preacher who used deliberately provocative language and gestures before a hostile Roman Catholic audience was held to have been properly bound over since a breach of the peace by members of his audience was the natural consequence of his acts. Sedley LJ put it this way in a more recent decision:[21] “... a judgment as to the imminence of a breach of the peace does not conclude the constable's task. The next and critical question for the constable, and in turn for the court, is where the threat is coming from, because it is there that the preventive action must be directed.” 42. Lines also have to be drawn where a demonstrator’s conduct impinges unacceptably upon the rights of others (which may or may not be constitutionally protected rights). Such a line had to be drawn, for instance, in Yeung May Wan v HKSAR,[22] where the Court had to decide whether the offence of obstructing a public place[23] was properly applied so as to curtail a static, peaceful demonstration by a small group of Falun Gong protesters which obstructed only part of the pavement, on the basis that they were interfering with the rights of other users of the public highway.[24] To take another example, the Court of First Instance[25] recently had to decide whether the right to demonstrate entitled protesters to take their demonstration into a private residential development without the consent of the owners, or whether that right was constrained by the need to respect the private property rights of the residents. 43. In practice, restrictions on the right to demonstrate are likely to be tested where a demonstrator is subjected to some form of restraining action by a law enforcement agency possibly, but not necessarily, leading to a criminal charge on account of the demonstrator’s conduct. The law relied on to justify the action taken against the demonstrator may then require examination. In some cases, its constitutionality may have to be examined: Does it pursue one of the listed legitimate aims? If so, does it pass the proportionality test? If the law is constitutionally valid, the question becomes whether the demonstrator’s conduct falls within the restriction imposed by the law, bearing it in mind that the restriction is narrowly construed while the constitutional right receives a generous interpretation. C. The appellants’ conduct 44. The appellants were fully entitled to protest against the MTR’s fare increases in the exercise of their rights of peaceful assembly and demonstration. But to stay within their constitutionally protected sphere, they had to avoid committing a breach of the peace – a concept examined below – and they had to avoid infringing the rights and freedoms of others. 45. As McGrath J recently pointed out in the New Zealand Supreme Court: “Freedom of assembly is not limited to gatherings for the purpose of protest. It extends to formal and informal assemblies in participation in community life. Gatherings for purposes that are ostensibly less political are also important to citizens for forming opinions and, ultimately, for participating in the democratic process.”[26] 46. That applies to the persons on the podium who were exercising their rights of peaceful assembly and free expression at the prize-giving ceremony. 47. The Court viewed video recordings showing the conduct of the appellants. The 1st appellant’s conduct was comparatively mild. He climbed over the barriers and rushed onto the stage and, standing some distance away from the speaker, scattered “devil money” in the air as his form of protest. He then turned away and did not resist being marched off the stage by security staff. An employee of the MTR had unsuccessfully attempted to stop him and had suffered a minor injury to his elbow. The interruption to the ceremony was very brief and, although the speaker was obviously surprised by the 1st appellant’s sudden appearance, she took it in her stride, laughing it off. 48. The 2nd appellant’s conduct was significantly more intrusive and threatening. He leapt over the barriers surrounding the stage, dashed at speed onto the podium and lunged towards the speaker (the Secretary for Transport) who was addressing the crowd through a microphone. His conduct undoubtedly caused those on the stage to fear for the speaker’s safety. Persons who had been seated at the back of the stage spontaneously leapt to their feet to try to intercept the 2nd appellant in her defence. The master of ceremonies suffered minor injuries in this process. Security staff swarmed onto the stage to bundle the 2nd appellant off the podium. The reaction of the persons on the stage was quite natural since no one could have known what the 2nd appellant’s intentions were as he lunged towards the speaker. As it turns out, his object was to seize the microphone away from her and to use it to shout his protest slogans. The ceremony resumed about a minute after the 2nd appellant’s intrusion. 49. In my view, the 2nd appellant’s conduct crossed the line and was unacceptable as a form of demonstration. It put the persons on the podium in fear of one of them being harmed, constituting a breach of the peace. Those persons were entitled to exercise their constitutional rights of peaceful assembly and freedom of expression without being subjected to apparent threats of imminent violence. The 2nd appellant’s actions may very well have constituted a common assault since he appears intentionally or recklessly to have caused another to apprehend immediate and unlawful personal violence.[27] And, as discussed below, he could have been arrested and bound over to keep the peace. The police have a duty to prevent such intrusions. The security personnel were entitled to act using reasonable force in the defence of others and in self-defence. If the law were to take a different view, the very constitutional rights now being invoked by the appellants would be at risk of being subverted in counter-demonstrations by thugs and bully boys seeking to suppress the expression of views they do not like. D. The constitutionality of the public order offences charged 50. The appellants were not, however, charged with common assault. Nor was it sought to have them bound over. Instead, the decision was taken to bring charges under section 17B(1) and section 17B(2) of the Public Order Ordinance.[28] I turn now to consider whether those charges were made out. 51. There has rightly been no challenge to the constitutionality of those sections. As noted above, Article 17 recognizes the maintenance of public order and protection of the rights and freedoms of others as legitimate purposes capable of justifying a restriction on the right of peaceful assembly. 52. Section 17B(1) makes it an offence to act in a disorderly manner for the purpose of preventing the transaction of the business for which a public gathering was called together. It is obviously aimed at protecting the right of peaceful assembly exercised by others. 53. Section 17B(2) has two principal elements. It involves in the first place, proof of noisy or disorderly behaviour; or of the use of threatening, abusive or insulting words by the defendant. In the present case, the focus of the charges is on disorderly behaviour. Secondly, the offence requires proof that such conduct was performed with intent to provoke a breach of the peace, or in circumstances where such conduct was likely to cause a breach of the peace. It is therefore plainly an offence aimed at maintaining public order or preventing public disorder. 54. Restricting the right of peaceful assembly by setting the boundaries established by these two sections (and, indeed, by other offences which prohibit violence or the threat of violence) involves in my view, the rational and proportionate furthering of the aforesaid legitimate purposes. They set a proper balance between the interests of demonstrators exercising their right of peaceful assembly on the one hand, and the interests of public order and the rights and freedoms of persons affected by that exercise on the other. It should not ordinarily be necessary in future for a magistrate or judge to dwell on the constitutionality of these offences. E. Was an offence under section 17B(1) proved? 55. The offence under section 17B(1) targets a “person who at any public gathering acts in a disorderly manner for the purpose of preventing the transaction of the business for which the public gathering was called together”. The main issue which arises concerns the meaning of “preventing”. 56. When indicating that he would, if necessary, also have convicted the appellants under section 17B(1), the Magistrate stated (in translation): “... the appellants’ conduct satisfied the meaning of ‘prevent’ as they had disrupted or even brought the ceremony to termination”. While I can see how the ceremony can be said to have been “disrupted”, I am with respect unable to see any basis for suggesting that their conduct “brought the ceremony to termination”. It was not in dispute that it resumed shortly after the interruptions. I can see no basis for inferring that they must have intended to prevent the ceremony from transacting its intended business, the obvious inference being that they merely wanted to interrupt it to dramatise their protest. 57. On appeal before Barnes J, Mr H Y Wong, counsel for the 1st appellant, submitted that the word “preventing” should be held to require proof that the gathering was “completely brought to an end” or “aborted”, something that manifestly had not happened in this case. The Judge did not accept that argument. She pointed to various dictionary meanings and concluded that “preventing” in section 17B(1) covered situations where the gathering had “to stop for a moment due to hindrance”. Her Ladyship added (in translation): “As a result of the 1st appellant’s conduct, the Secretary’s speech stopped for a few seconds. Shortly after she resumed speaking, the 2nd appellant snatched away the microphone, so that she was unable to continue with her speech. It is obvious that the appellants were preventing the prize presentation ceremony of the walking race from going on by behaving in that way in order to stay on the stage and capture everyone’s attention for as long as they could, to fulfil their purpose of escalating the intensity of their protest.” 58. I am with respect unable to agree with that construction of “preventing”. In my view, both as a matter of language and of the evident statutory purpose of section 17B(1), a person only has the “purpose of preventing the transaction of the business for which the public gathering was called together” if his purpose is to make it impossible in practical terms to hold or continue with the gathering; or, at least, to interrupt the gathering for such a duration or by using such means as substantially to impair the intended transaction of business. It certainly does not apply where, as in the present case, the demonstrators’ purpose evidently did not involve more than a brief interruption of the prize-giving. 59. The same result is arrived at on a purposive construction. As stated above, the evident purpose of section 17B(1) is to protect the exercise by others of their right of peaceful assembly. The need for such protection is obviously triggered where the defendant’s purpose is to prevent the public gathering from happening or continuing; or substantially preventing it from transacting the intended business of the gathering. But an intrusion which must have been envisaged to cause only a minor interruption, with the gathering then resuming, cannot be said to have been intended to deny the people attending the gathering that right. It is not the statutory purpose to employ section 17B(1) to punish such minor interruptions. 60. As previously noted, restrictions on fundamental rights are narrowly construed. So if “preventing” can properly be given a range of meanings, the Court inclines towards adopting a meaning which preserves a wider ambit for the relevant rights. 61. For the abovementioned reasons, it is my view that the appellants’ convictions under section 17B(1) must be set aside. I turn then to consider their possible liability under section 17B(2). F. Did the appellants commit offences under section 17B(2)? 62. Section 17B(2) (set out in Section A above) is derived from the United Kingdom’s Public Order Act 1936, section 5.[29] Focussing on the “disorderly behaviour” aspect of the offence,[30] the elements which the prosecution must prove under section 17B(2) are that the defendant: (a) in a public place; (b) behaved in disorderly manner; (c) either (i) with intent to provoke a breach of the peace; or (ii) whereby a breach of the peace was likely to be caused. 63. It is common ground that the relevant incident occurred in a public place. And in the case of the 2nd appellant, Mr Martin Lee SC realistically accepted that his conduct was “disorderly”. However, Mr H Y Wong did not accept that the behaviour of his client, the 1st appellant, could be so described. 64. Both appellants in any event deny that there is any evidential basis for holding that their conduct was either intended to provoke or was likely to cause a breach of the peace when those expressions are properly construed. F.1 The meaning of “disorderly behaviour” 65. The courts have not attempted a definition of “disorderly behaviour”. However, there have been attempts to relate that phrase to other familiar terms. Thus, in the Scottish case of Campbell v Adair,[31] it was held that “disorderly” indicates “less aggressive conduct than would be required to constitute a breach of the peace”. In New Zealand, in Kinney v Police,[32] Woodhouse J said that “disorderly behaviour” meant “something more than unmannerly, or disturbing or annoying” behaviour. And the Divisional Court in England has held that there need not be any element of violence, present or threatened, nor proof of any feeling of insecurity, in an apprehensive sense, on the part of the public to establish disorderly conduct in a harassment case.[33] Moreover, it was pointed out that “disorderly behaviour” did not need to be threatening, abusive or insulting, since the Public Order Act deals separately with the use of words meeting that description. Similarly, in section 17B(2) the use of “threatening, abusive or insulting words” forms a separate limb of the offence. 66. In the same case, the Divisional Court held that the proper approach is that of the House of Lords in Brutus v Cozens[34] where their Lordships decided that the meaning of “insulting behaviour” was a question of fact for the trial court. As Lord Reid held in Brutus v Cozens: “The meaning of an ordinary word of the English language is not a question of law. The proper construction of a statute is a question of law.”[35] 67. In my view, that is the approach which should be adopted in Hong Kong. As Gleeson CJ pointed out in Coleman v Power:[36] “Concepts of what is disorderly, or indecent, or offensive, vary with time and place, and may be affected by the circumstances in which the relevant conduct occurs.” 68. Such concepts are best left to the trial judge to be applied in their ordinary meaning to the time, place and circumstances of the conduct in question. In the present case, the Courts below held that the 1st appellant’s conduct did constitute “disorderly behaviour”. They were entitled to take that view. The 1st appellant climbed over barriers plainly intended to keep uninvited persons off the stage. He evaded interception by an MTR employee who attempted to stop him (and who consequently suffered a minor injury) and ran onto the stage where he momentarily interrupted proceedings with his “devil money” demonstration before being marched off the podium. As a matter of ordinary language, such behaviour is capable of being described as “disorderly”. F.2 “Disorderly behaviour” – the Australian and New Zealand cases 69. There was considerable discussion of the Australian High Court’s decision in Coleman v Power[37] and the decisions of the New Zealand Supreme Court in Brooker v Police[38] and Morse v Police.[39] While there is much of interest to be found in those judgments and while the meaning of “disorderly behaviour” is extensively considered in the New Zealand decisions, those cases must be treated with caution in Hong Kong since their analysis and construction of the relevant statutory provisions (which are materially different from section 17B(2)) were undertaken to address a constitutional problem which we do not face. For the reasons which follow, in my view, they are decisions which have no immediate relevance to the present appeal. 70. As indicated in Section D above, the constitutionality of section 17B(2) is not in issue. It has two main elements, the first involving the defendant’s conduct (relevantly “disorderly behaviour” in the present case) and the second involving a requirement that such conduct be intended to provoke or likely to cause a breach of the peace. Thus, the section 17B(2) offence can be accommodated as a justified restriction in pursuit of a legitimate purpose recognized by Article 17 as previously discussed. 71. In contrast, the three cases mentioned above were all concerned with offences which do not make liability dependent upon proof of an intention to provoke or the likelihood of causing a breach of the peace. Nor is liability dependent upon the defendant denying others the right of peaceful assembly. In Coleman v Power,[40] the offence[41] was constituted simply by any person, in any public place, using “any threatening, abusive or insulting words to any person”. The Court was concerned with a defendant charged with using insulting words addressed to a policeman (asserting that he was corrupt). In Brooker v Police[42]and Morse v Police,[43] the offence[44] makes liable every person who simply “in or within view of any public place, behaves in an offensive or disorderly manner...” The Court in Brooker v Police was concerned with deciding how “behaves in [a] ... disorderly manner” should be interpreted where an individual staged a solo protest against police conduct outside the house of a constable who was trying to get some sleep. And in Morse v Police, it was concerned with “offensive behaviour” involving the burning of a New Zealand flag on ANZAC day. 72. Earlier versions of those offences, like section 17B(2), required proof of a second element involving a breach of the peace. But that requirement had been dropped, as Gleeson CJ explained: “Section 7 of the Vagrants Act replaced s 6 of the Vagrant Act 1851 (Qld). That section prohibited the using of threatening, abusive or insulting words or behaviour in any public street, thoroughfare or place with intent to provoke a breach of the peace or whereby a breach of the peace may be occasioned. The omission of the element relating to a breach of the peace, in the 1931 Act, was plainly deliberate. The legislative changes in Queensland in 1931 were similar to changes in New Zealand in 1927. In New Zealand, the Police Offences Act 1884 (NZ) made it an offence to use any threatening, abusive or insulting words or behaviour in any public place within the hearing or in the view of passers by, with intent to provoke a breach of the peace, or whereby a breach of the peace may be occasioned. By legislation in 1927, the provision was altered by omitting any reference to a breach of the peace, and by expanding the description of the prohibited conduct to cover behaving in a riotous, offensive, threatening, insulting or disorderly manner, or using threatening, abusive or insulting words, or striking or fighting with any other person.”[45] 73. Having eliminated the breach of the peace requirement, the offences acquired an extremely wide reach and potentially came into conflict with constitutionally protected rights. In Australia, constitutional protection was conferred on the implied freedom of communication about government and political matters as laid down in Lange v Australian Broadcasting Corporation.[46] And in New Zealand, the constitutional rights were those protected by section 14 of the New Zealand Bill of Rights Act 1990, guaranteeing freedom of expression, including the freedom to seek, receive, and impart information and opinions of any kind in any form, subject (as provided by section 19(3)) to reasonable restrictions prescribed by law which are necessary to protect other important interests, including public order and the rights and reputations of others. 74. Thus, the Australian High Court and the New Zealand Supreme Court were engaged in construing the concepts of “insulting words” and “disorderly behaviour” and “offensive behaviour” in their legislation to achieve compatibility with constitutionally protected rights.[47] In New Zealand, this was done by interpreting the “disorderly behaviour” offence as one aimed at maintaining public order and confining it to cases where the disorderly behaviour is disruptive of public order, interfering with others’ use of public space;[48] although it was held that this did not mean that it had to lead to violence.[49] “Offensive behaviour” was given a like meaning. The majority in the Australian High Court drew similar lines, holding that by construing the offences as intended to suppress the disturbance of public order, they could be given a meaning compatible with the implied constitutional freedom. Thus, the expression “insulting words” was construed to mean words intended to or reasonably likely to provoke unlawful physical retaliation.[50] 75. In Hong Kong, for the reasons discussed above, there is no need to adopt such a particular construction in relation to the words “behaves in a ... disorderly manner” to make section 17B(2) compatible with Article 17. 76. The 2nd appellant accepts that his conduct was “disorderly” and, for the reasons given above, the Courts below were entitled to find that the same applies to the 1st appellant as a matter of ordinary language. F.3 The requirement that a breach of the peace is likely to be caused F.3a Meaning and consequences of a “breach of the peace” 77. To decide whether the second element of section 17B(2) has been made out against the appellants, it is necessary to consider the meaning of “breach of the peace”. The modern authority is R v Howell,[51] where Watkins LJ explained the concept as follows: “... there is a breach of the peace whenever harm is actually done or is likely to be done to a person or in his presence to his property or a person is in fear of being so harmed through an assault, an affray, a riot, unlawful assembly or other disturbance. It is for this breach of the peace when done in his presence or the reasonable apprehension of it taking place that a constable, or anyone else, may arrest an offender without warrant.”[52] 78. Approving that decision, Lord Bingham of Cornhill noted in R (Laporte) v Chief Constable of Gloucestershire,[53] that “the essence of the concept was to be found in violence or threatened violence” and it was on that basis that the European Court of Human Rights found that the concept possessed sufficient certainty in law.[54] 79. However, a person may provoke a breach of the peace without any violence or threat of violence on his part: “... it suffices that his conduct is such that the natural consequence of it is violence from some third party”.[55] That third party need not be the person provoked or a by-stander, it could, for instance, be a member of the provoker’s group.[56] The actual or feared harm must be unlawful[57] and, where the harm is anticipated, there must be a real risk and not the mere possibility of such harm.[58] Moreover, the anticipated harm must be imminent.[59] 80. As appears in the passage from R v Howell cited above, a breach of the peace or reasonable apprehension of an imminent breach of the peace gives rise to a common law power of arrest without warrant. It also gives rise to a power to take measures short of arrest to prevent such breach.[60] 81. Moreover, while a breach of the peace is not, as such, a criminal offence, it founds an application to bind over.[61] The jurisdiction to bind over dates back at least to the 1361 Statute of Edward III.[62] In Hong Kong, the jurisdiction is now conferred by the Criminal Procedure Ordinance,[63] with procedural provisions contained in the Magistrates Ordinance.[64] It is a jurisdiction aimed at preventing breaches of the peace in the future. A binding-over order may thus be made requiring a person to keep the peace where there are grounds for reasonably apprehending that he may become involved in a breach of the peace in future. That person may, for instance, have previously been involved in violence to a person or property; or may threaten such violence; or be guilty of conduct giving rise to a reasonable apprehension that such violence will take place.[65] 82. It was with these principles in mind that I expressed the view[66] that the 2nd appellant could have been arrested and bound over to keep the peace. His conduct whereby one or more of the persons on the podium were put in fear of being harmed amounted to a breach of the peace giving rise to a power of arrest and provided a basis for binding him over to prevent future breaches. F.3b Section 17B(2)’s second element 83. But section 17B(2) is not designed to penalise persons who simply commit breaches of the peace. That is of central importance to the disposal of this appeal. The appellants would only be guilty of an offence under that section if their disorderly behaviour was either intended or likely to cause a breach of the peace by someone else. It is not enough to show that they were guilty both of disorderly behaviour and of committing a breach of the peace. This is a conclusion dictated by the language of section 17B(2) and reflected in the Hong Kong authorities and the authorities on section 5 of the United Kingdom’s Public Order Act 1936 from which section 17B(2) is derived and which is structured in the same way. 84. There are two forms of the offence and each form involves proof of two elements.[67] For present purposes, the first element in both forms of the offence is disorderly behaviour which as discussed above, is established against both appellants. 85. The second element in the first form of the offence requires such disorderly behaviour to be carried out “with intent to provoke a breach of the peace”. Plainly, the defendant’s intent must be to provoke a breach of the peace (in the sense explained in Howell and discussed above) by another person or other persons as a consequence of his disorderly conduct. It cannot sensibly be applied to a case where the defendant is guilty of disorderly behaviour and where he alone commits a breach of the peace. In such a case, it would make no sense to speak of him acting “with intent to provoke” a breach of the peace. He would have simply have committed a breach of the peace. 86. The second element in the second form of the offence requires the disorderly behaviour to be such that “a breach of the peace is likely to be caused”. Once again, such language is inapt for describing a situation involving the conduct of the defendant and no one else. This form of the offence requires an assessment of the likely reaction to the defendant’s disorderly conduct by the persons who are affected by it. 87. This view has the support of the authorities. Thus, in Jordan v Burgoyne,[68] a case involving section 5 of the Public Order Act 1936, Lord Parker CJ put the focus firmly on the likely reaction of the crowd in Trafalgar Square to the defendant’s pro-Hitler remarks: “...if words are used which threaten, abuse or insult - all very strong words - then that person must take his audience as he finds them, and if those words to that audience or that part of the audience are likely to provoke a breach of the peace, then the speaker is guilty of an offence.” 88. In Marsh v Arscott,[69] McCullough J, sitting with Donaldson LJ in the English Divisional Court stated in relation to the said section 5: “The phrase ‘whereby a breach of the peace is likely to be occasioned’ indicates that Parliament was concerned with cause and effect, ie with conduct which is likely to bring about a breach of the peace and not with conduct which is itself a breach of the peace and no more. Were this the law every common assault occurring in a public place would also be an offence against this section. Many such assaults will in fact be likely to lead very quickly to a breach of the peace, and these will be within the section; but, without more it is not enough that conduct which is threatening, abusive or insulting is of itself a breach of the peace.” 89. In Parkin v Norman,[70] a case involving the “threatening, abusive or insulting words” stream of the offence, McCullough J elaborated as follows: “The purpose of the Act was to promote good order in places to which the public have access ... It is clear, both from the long title and from sections 1, 2 and 4 that Parliament intended to prevent activities liable to lead to public disorder, regardless of whether or not those engaging in them intended that disorder should result. The use of the phrase ‘whereby a breach of the peace is likely to be occasioned’ in section 5 reflects this thinking. ... But not all threats, abuse or insults necessarily have this result. Qualifying words were therefore required, and conduct of this kind was only prohibited if it was likely to lead to a breach of the peace or, as was added, if it was so intended. It was the likely effect of the conduct on those who witnessed it with which Parliament was chiefly concerned. What is likely to cause someone to break the peace is his feeling that he has been threatened or abused or insulted, and this will be so whether or not the words or behaviour were intended to threaten or to abuse or to insult.” 90. In Percy v DPP,[71] Collins J in the Divisional Court emphasised that the provocation had to be directed at others: “The conduct in question does not itself have to be disorderly or a breach of the criminal law. It is sufficient if its natural consequence would, if persisted in, be to provoke othersto violence, and so some actual danger to the peace is established.” 91. The section 17B(2) offence therefore has a somewhat paradoxical feature. Whereas a defendant who acts in a disorderly fashion may commit the offence if he misbehaves in a situation where less than law-abiding people might react by breaching the peace, he escapes liability if he indulges in the same conduct in the presence of law-abiding and disciplined persons who are regarded as unlikely so to react. 92. The purpose of section 17B(2) is in other words, to prevent a person instigating public disorder involving others rather than simply punishing that person for his own misbehaviour. It therefore excludes from its ambit many situations where the defendant’s disorderly behaviour or threatening, abusive or insulting words, and so forth, are not likely to produce such violence. In Coleman v Power,[72] Gleeson CJ described some such situations: “There may be any number of reasons why people who are threatened, abused or insulted do not respond physically. It may be (as with police officers) that they themselves are responsible for keeping the peace. It may be that they are self-disciplined. It may be simply that they are afraid. Depending upon the circumstances, intervention by a third party may also be unlikely. ... And if violence should occur, it is not necessarily unlawful. Depending upon the circumstances, a forceful response to threatening or insulting words or behaviour may be legitimate on the grounds of self-defence or provocation. Furthermore, at common law, in an appropriate case a citizen in whose presence a breach of the peace is about to be committed has a right to use reasonable force to restrain the breach.” F.3c The reaction of police officers and trained security personnel 93. The likelihood of a breach of the peace is assessed as a matter of fact, taking account of the nature of the disorderly behaviour and the circumstances in which such behaviour occurred. 94. As reflected in the passage cited from the judgment of Gleeson CJ, the courts have consistently regarded police officers generally as persons unlikely to react by breaching the peace. Since the police have a legal duty to keep the peace[73] the courts have recoiled from the suggestion that in taking measures to preserve public order they might be regarded as acting unlawfully by themselves committing breaches of the peace. 95. Thus, in Marsh v Arscott,[74] where a defendant directed threatening, abusive or insulting words at the police with no one else present, McCullough J held that: “...no breach of the peace was likely to have been occasioned. No other person was likely to have broken the peace, and all that the police were likely to do was arrest him, as they did.” 96. Donaldson LJ pointed out that there were other offences that could be charged: “...there are a number of other offences which can be charged, where appropriate, and the most obvious of course is common assault or assault on the police in the execution of their duty, and it must never be forgotten in any event that the police, in pursuance of their duty to keep the peace, have a right and a duty to detain those who are threatening it.”[75] 97. Similarly, in Redmond-Bate v DPP,[76] Sedley LJ pointed out that in Percy v DPP:[77] “...a bind-over on a woman who kept climbing over the perimeter fencing into a military base was quashed because there was no sensible likelihood that trained security personnel would be provoked by her conduct to violence.” 98. A similar view has been taken in the Hong Kong courts.[78] In one such case, trained security staff of the Hong Kong Jockey Club were regarded as unlikely to have reacted to disorderly conduct by committing a breach of the peace, especially in full view of members of the racing public.[79] F.3d Was it proved that a breach of the peace was likely to be caused in the present case? 99. Barnes J held that the section 17B(2) offence was concerned with “the effect or influence which the disorderly conduct is likely to produce on the people who witnessed the conduct in question at the scene”.[80] She noted that the Magistrate had not analysed the evidence or made findings in respect of such influence or effect.[81] Her Ladyship held that the evidence showed that the MTR staff had acted in a “very restrained and professional manner” and that they “would not themselves breach the peace when, in the course of the performance of their duty, some people disrupted the order”.[82] She concluded that it had not been established that the appellants’ disorderly conduct was likely to cause other people present to breach the peace and quashed the section 17B(2) convictions.[83] 100. I respectfully agree with Barnes J’s approach and conclusion. The only persons who were in the immediate vicinity of the appellants when they invaded the stage were the officiating guests and MTR staff on the podium, security guards and possibly the police. There was no evidential basis for suggesting that any of them might have reacted unlawfully by committing a breach of the peace in response to the intrusion. Neither was there any evidence that other members of the public present, whether the competing athletes or otherwise, might have been prone to reacting violently to the disorderly behaviour of the two appellants. It follows that Barnes J was right to overturn their convictions under section 17B(2). G. Conclusion 101. For the foregoing reasons, I conclude that convictions both under section 17B(1) and section 17B(2) cannot be sustained against the appellants. I would therefore allow their appeals. I would direct that any submissions as to costs should be lodged in writing within 21 days from the date of this judgment and that in default of such submissions, there be no order as to costs. 102. Before leaving this judgment, I am anxious to reiterate that the 2nd appellant’s acts described in Section C above are unacceptable and exceed the boundaries of the constitutionally protected right of peaceful assembly and demonstration. As it turns out, he was charged with offences that are designed to prevent the instigation of public disorder rather than individual disorderly behaviour. If he had been charged with common assault, he may very well have been convicted. He could in any event have been arrested and bound over to keep the peace. The 1st appellant’s conduct was less objectionable, but he too behaved in a disorderly manner and a binding over order to keep the peace might very well have been justified in his case. Mr Justice Tang PJ: Introduction 103. On 10 April 2011, for the purpose of a prize presentation ceremony following a fund raising event,[84] a stage was erected on a drained pond at Statue Square. The prize presentation ceremony began at about noon. Persons including guests who were involved in the prize presentation ceremony were seated on the stage, among whom the Secretary for Transport and Housing Madam Eva Cheng (“the Secretary”) and the Chairman of the MTR, Mr Raymond Ch’ien (“Mr Ch’ien”). We were told that because protests were expected, the MTR had deployed a total of 29 staff members and 20 contract security guards to maintain order. In addition, a significant number of police constables were strategically positioned. Measures[85] were taken to ensure that only authorised persons could enter certain areas, including the prize presentation stage. 104. The event was recorded on videotape by the MTR Corporation as well as by Cable TV. The police also carried out their own video recording. Four sections of video footage were produced as exhibits at trial. The video recordings show that while Chairman Ch’ien was speaking on the stage, demonstrators were shouting the slogan “Shame on MTR for their fare hike”. The shouting continued even after the Secretary had begun to speak. 105. These appeals are concerned with what happened in the next minute or so, which were captured on video. The summary below is largely taken from the Statement of Findings of the trial magistrate, Mr Marco Li. 106. While the Secretary was speaking, Mr Chow Nok-hang (hereafter called the 1st appellant) rushed onto the stage from the left and scattered hell money. The Secretary was startled and cried out “wow” she then said “This is not some kind of sport. Children, you should not imitate that. Hahaha…”. While the 1st appellant was being taken away, he shouted “Shame on MTR for their fare hike.” Earlier, PW1 the organizer of the event was standing on the drained pond, had tried to stop the 1st appellant from going up onto the stage but he became unbalanced. He got hold of the 1st appellant’s thigh but lost balance again, and his elbow was injured when he hit the kerb. 107. Immediately after that, Mr Wong Hin-wai (hereafter called the 2nd appellant) jumped onto the stage and ran towards the Secretary. Mr Ch’ien sprang up from his seat trying to stop him. The 2nd appellant snatched away the microphone placed in front of the Secretary and shouted the slogan “Shame on MTR for their fare hike” through the microphone. A large number of persons, presumably those involved with security, came forward. They separated the Secretary from the 2nd appellant and took him away together with the microphone. At about that time, a demonstrator wearing a black upper garment went towards the stage, he appeared to be trying to go onto the stage, but he was stopped by the staff. He raised his arm and shouted “Shame on MTR for their fare hike”. The learned magistrate said: “PW2 … the MC … pulled his partner to his side fearing that she would be knocked down. … He tried to stop the second appellant but was not fast enough. He believed that the secondappellant had taken hold of Secretary Cheng’s microphone and said something. Then a group of personnel together with him carried the secondappellant down the stage. He suffered minor injuries during the struggle: … The second appellant put up slight resistance but soon calmed down.” The charges 108. As a result the 1st and 2nd appellants[86] were each charged with one count of “Behaving in a disorderly manner in a public place”, contrary to s17B(2) of the Public Order Ordinance Cap 245 (“POO”) and an alternative count of “Acting in a disorderly manner at a public gathering” under s17B(1) of the Public Order Ordinance. 109. Section 17B provides : “(1) Any person who at any public gathering acts in a disorderly manner for the purpose of preventing the transaction of the business for which the public gathering was called together or incites others so to act shall be guilty of an offence and shall be liable on conviction to a fine of $5000 and to imprisonment for 12 months. (2) Any person who in any public place behaves in a noisy or disorderly manner, or uses, or distributes or displays any writing containing, threatening, abusive or insulting words, with intent to provoke a breach of the peace, or whereby a breach of the peace is likely to be caused, shall be guilty of an offence and shall be liable on conviction to a fine of $5000 and to imprisonment for 12 months.” 110. The particulars of the offence alleged that each of the appellants in respect of the s17B(1) offence “… acted in a disorderly manner for the purpose of preventing the transaction of the business of the ‘MTR Hong Kong Race Walking 2011’ for which the public gathering was called together.” And in respect of the s17B(2) offence that each of them behaved “in a noisy[87] or disorderly manner, with intent to provoke a breach of the peace, or whereby a breach of the peace was likely to be caused.” 111. The trial magistrate convicted them of the s17B(2) offence and sentenced each of them to 14 days imprisonment. He said, had it been necessary to do so, he would have also convicted them on the alternative charges under s17B(1). 112. On appeal Barnes J quashed their conviction under s17B(2) but convicted them on the alternative charges of “Acting in a disorderly manner at a public gathering” contrary to s17B(1). She set aside the prison terms and fined the 1st appellant $2000 and the 2nd appellant $3000. Both the appellants and the prosecution have appealed to us. Public Order Offences 113. Section 17B(1) and (2) are public order offences[88], and may impact upon important freedoms, such as the freedom of speech, of assembly and of demonstration, which are fundamental freedoms protected by the Basic Law and the Hong Kong Bill of Rights Ordinance. The freedom of assembly, demonstration and speech are closely associated.[89] It is clear that at the material time the appellants were engaged in a public demonstration against a MTR fare hike. They were exercising their rights of free speech, assembly and demonstration. 114. In respect of such fundamental freedoms, this court has emphasized: “…It is well established in our jurisprudence that the courts must give such afundamental right a generous interpretation so as to give individuals its full measure. Ng Ka Ling & Others v Director of Immigration (1999) 2 HKCFAR 4 at pp.28-29. On the other hand,restrictions on such a fundamental right must be narrowly interpreted. Gurung Kesh Bahadur v Director of Immigration (2002) 5 HKCFAR 480 at para.24. Plainly, the burden is on the Government to justify any restriction. This approach to constitutional review involving fundamental rights, which has been adopted by the Court, is consistent with that followed in many jurisdictions. Needless to say, in a society governed by the rule of law, the courts must be vigilant in the protection of fundamental rights and must rigorously examine any restriction that may be placed on them.”[90] 115. The reason for such approach was given by the majority earlier in their judgment when they said: “1. The freedom of peaceful assembly is a fundamental right. It is closely associated with the fundamental right of the freedom of speech. The freedom of speech and freedom of peaceful assembly are precious and lie at the foundation of a democratic society. 2. These freedoms are of cardinal importance for the stability and progress of society for a number of inter-connected reasons. The resolution of conflicts, tensions and problems through open dialogue and debate is of the essence of a democratic society. These freedoms enable such dialogue and debate to take place and ensure their vigour. A democratic society is one where the market place of ideas must thrive. These freedoms enable citizens to voice criticisms, air grievances and seek redress. This is relevant not only to institutions exercising powers of government but also to organizations outside the public sector which in modern times have tremendous influence over the lives of citizens. Minority views may be disagreeable, unpopular, distasteful or even offensive to others. But tolerance is a hallmark of a pluralistic society. Through the exercise of these freedoms minority views can be properly ventilated.” 116. Their lordships went on to say that although these freedoms may be restricted as prescribed by law, such restriction must be necessary in a democratic society in the interests of national security or public safety, public order (order public), the protection of public health or morals or the protection of the rights and freedoms of others. Any restriction must be both necessary and proportionate.[91] 117. The possible impact on freedom of assembly and expression in connection with a charge of public obstruction was considered in Yeung May Wan where persons responsible for a largely static and peaceful demonstration and the display of a banner by between 4 and 16 persons outside the Liaison Office of the Central People’s Government were charged with obstruction of a public place contrary to s4A of the Summary Offences Ordinance Cap 228 by setting out the banner (the 1st charge), and doing an act whereby obstruction might accrue to a public place contrary to s4(28) by assembling together and displaying the banner (the 2nd charge), and of wilfully obstructing a police officer contrary to s36(b) of the Offences Against the Person Ordinance Cap 212 (the 3rd charge) and assaulting a police officer contrary to s63 of the Police Force Ordinance Cap 232 (the 4th to 6th charges)(the wilful obstruction and assault charges). The Court of Appeal quashed the convictions on the public place obstruction charges but upheld those on the wilful obstruction and assault charges. The defendants’ appeals to this court were unanimously allowed. In the joint judgment of Li CJ, Chan and Ribeiro PJJ and Sir Anthony Mason NPJ they said: “31. Central to the case is the fact that the arrests were made and the charges of public place obstruction laid against the defendants because of their conduct in the course of a peaceful public demonstration. This was not a simple case of obstruction, for instance, by inconsiderate parking of a vehicle or by dumping waste building materials on the road or by a hawker impeding pedestrians on a pavement. Here, the fact that the defendants were at the time of arrest engaged in a peaceful demonstration meant that the constitutionally protected right to demonstrate was engaged. Indeed, a peaceful demonstration, may also engage the closely related guaranteed freedoms of opinion, expression and assembly. Such fundamental rights, when engaged, have an important bearing on the scope of the offence of obstruction and consequently on the scope of police powers of arrest on suspicion of that offence. … 44. where the obstruction in question results from a peaceful demonstration, a constitutionally protected right is introduced into the equation. In such cases, it is essential that the protection given by the Basic Law to that right is recognised and given substantial weight when assessing the reasonableness of the obstruction. While the interests of those exercising their rights passage along the highway obviously remain important, and where exercise of the right to demonstrate must not cause an obstruction exceeding the bounds of what is reasonable in the circumstances, such bounds must not be so narrowly defined as to devalue, or unduly impair the ability to exercise, the constitutional right.”[92] Disorderly manner 118. The allegations of acts or behaviour in a disorderly manner was common to the offences under s17B(1) and (2). These offences are not unique to Hong Kong and have a long history. There are decisions from other common law jurisdiction on similar offences, which provide helpful illumination. I will refer to some of these decisions below. 119. In New Zealand, s3 and s4 of the Summary Offences Act 1981 dealt with offences involving disorderly manner behaviour. It was an offence under s3 punishable by three months imprisonment or a fine not exceeding $2000, for any person “who, in or within view of any public place, behaves, or incites or encourages any person to behave, in a riotous, offensive, threatening, insulting, or disorderly manner that is likely in the circumstances to cause violence against persons or property to start or continue.” Section 4(1)(a) created a lesser offence, punishable only by a fine not exceeding $1000, which did not require any such conduct to be “likely in the circumstances to cause violence against persons or property to start or continue.” 120. In Brooker v Police [2007] 3 NZLR 91, a decision of the Supreme Court of New Zealand, the defendant, who believed that a constable had behaved unlawfully towards him, went to her home shortly after 9 am, knowing that she had been on duty all night, knocked on her door, and after she opened it and asked him to go away, retreated to the grass verge of the road where he sang songs in a normal voice and played his guitar. When the police, summoned by the constable, told the defendant to stop and go away he refused. The defendant whose conviction under s4(1)(a) was upheld by the Court of Appeal appealed to the Supreme Court on the ground that his behaviour could not be regarded as “disorderly” under s4(1)(a) when read in conformity with s14 of the New Zealand Bill of Rights Act 1990, which guarantee the right to freedom of expression. By a majority, his conviction was set aside. 121. Elias CJ was of the view at [34] that even this lesser offence was capable of significant impact upon important freedoms and that [24] “disorderly behaviour under s4(1)(a) means behaviour seriously disruptive of public order. Simply causing annoyance to someone else, even serious annoyance, is insufficient if public order is not affected.” Thus, the disorderly behaviour must be seriously disruptive of public order. [42] to [47]. Moreover, “the value protected by the Bill of Rights must be specifically considered and weighed against the value of public order … As a result, public order will less readily be seen to have been disturbed by conduct which is intended to convey information or express an opinion than by other forms of behaviour.”[59] 122. Blanchard J [56] said to fall within s4(1)(a), the behaviour must substantially disturb the normal functioning of life in the environs of that place. “It must cause a disturbance of good order which in the particular circumstances of time and place any affected members of the public could not reasonably be expected to endure because of its intensity or its duration or a combination of both those factors” Thus, no one should be convicted “ unless there has been a substantial disruption of public order.” 123. Tipping J said: [90] “Conduct in a qualifying location is disorderly if, as a matter of time, place and circumstance, it causes anxiety or disturbance at a level which is beyond what a reasonable citizen should be expected to bear … Where … the behaviour concerned involves a genuine exercise of the right to freedom of expression, the reasonable member of the public may well be expected to bear a somewhat higher level of anxiety or disturbance than would otherwise be the case.” 124. McGrath J dissented on the facts. He said at [146] “The detriment to the complainant’s privacy interests, because of the time, place and manner of the appellant’s protest which sought to interrupt her from resting at her home, went well beyond what any citizen, public official or not, should have to tolerate in her home environment.” 125. Thomas J also dissented, at [186] he said “Disorderly behaviour may range from behaviour which disrupts public order to behaviour which, because it is an annoyance, impacts on public order. Such behaviour may, or may not, implicate a right or rights. The Court’s decision in any given case will depend on the time, location and circumstances and will essentially be a question of fact and degree.” 126. Morse v Police [2012] 2 NZLR 1, another decision of the New Zealand Supreme Court, was concerned with offensive conduct under s4(1)(a)[93]. There were flag burning, horn blowing by a group of protestors, during a Dawn Parade on ANZAC Day, to attract attention to what they were doing. The head notes in the report stated that the Supreme Court of New Zealand held (unanimously) that “The offence … was concerned with public order. To be offensive under s4(1)(a), a behaviour … had to be productive of disorder”. And by a majority (Elias CJ and Anderson J dissenting) to such an extent that it was beyond what could be expected to be tolerated by other reasonable people in a democratic society. 127. Morse v Police is important because it provides important insight on what persons subject to the impugned conduct could be expected to tolerate. 128. Blanchard J said the test is that of a reasonable person who, “must surely be a person who is sensitive to such values[94] and displays tolerance for the rights of the person whose behaviour is in question. In other words, the hypothetical reasonable person (of the kind affected) is one who takes a balanced, rights-sensitive view, conscious of the requirements of s5 (New Zealand Bill of Rights Act), and therefore is not unreasonably moved to wounded feelings or real anger, resentment, disgust or outrage, particularly when confronted by a protestor.” [64] 129. Tipping J said at [70] “It cannot, however, be right that the unreasonable reactions of those who are affected by the behaviour can be invoked as indicative of a threat to public order. Hence those affected by the behaviour must be prepared to tolerate some degree of offence on account of the rights and freedoms being exercised by those responsible for the behaviour. It is only when the behaviour of those charged under s4(1)(a) causes greater offence than those affected can be expected to tolerate that an offence under s4(1)(a) will have been committed. And it is always necessary for the prosecution to demonstrate a sufficient disturbance of public order. [71] In this context public order is sufficiently disturbed if the behaviour in question causes offence of such a kind or to such an extent that those affected are substantially inhibited in carrying out the purpose of their presence at the place where the impugned behaviour is taking place. Only if the effect of the behaviour reaches that level of interference with the activity in which those affected are engaged is it appropriate for the law to hold that their rights and interests should prevail over the right to freedom of expression of those whose behaviour is in contention. That is the appropriate touchtone. [72] All relevant matters of time, place and circumstance must, however, be brought to account when applying the touchstone to the behaviour in question and thereby deciding whether the defendant's conduct is offensive in law. The application of the touchstone is contextual not abstract; but those affected are required, for the purpose of the necessary assessment, to be appropriately tolerant of the rights of others. Tolerance to the degree thought appropriate by the Court is the pivot on which the law reconciles the competing interests of public order and freedom of expression. A free and democratic society is justified in limiting freedom of expression at the point when public order is sufficiently disturbed.” 130. In Australia, in Coleman v Power [2004] 220 CLR 1, the High Court was concerned with s7(1)(d) of the Vagrants Act 1931 of Queensland, which provided that “any person who, in any public place or so near to any public place ... uses any threatening, abusive, or insulting words to any person … shall be liable to a penalty … ”[95] Section 7(1)(d) was enacted to replace s6 of the Vagrant Act 1851, which prohibited the using of threatening, abusive or insulting words or behaviour in any public street … “with intent to provoke a breach of the peace or whereby a breach of the peace may be occasioned”.[96] 131. The defendant was distributing pamphlets in a mall containing charges of corruption against several police officers. He was approached by a police officer who demanded a pamphlet. The defendant refused and said loudly “this is constable (BP) a corrupt officer”. He then sat down wrapping his arms around a pole and violently resisted attempts to arrest him. His conviction under s7(1)(d) was set aside by a majority, which comprised Gummow, Hayne, Kirby and McHugh JJ.[97] 132. Gleeson CJ said “the removal in 1931 of the requirement concerning a breach of the peace undoubtedly gave rise to a problem of confining the operation of the legislation within reasonable bounds.” He resolved that problem by interpreting as having been built into s7 a requirement related to serious disturbance of public order or affront to standards contemporary behaviour [23]. 133. Gummow and Hayne JJ in a joint judgment said at [183], the insulting words which are proscribed are those which “are intended to provoke unlawful physical retaliation, or they are reasonably likely to provoke unlawful physical retaliation from either the person to whom they are directed or some other who hears the words uttered.” Kirby J agreed. 134. Section 17B(2) has its English equivalent in s5 of the Public Order Act of 1936, which provided “any person who in any public place or at any public meeting uses threatening, abusive or insulting words or behaviour with intent to provoke a breach of the peace, or whereby a breach of the peace is likely to be occasioned, shall be guilty of an offence.” In Redmond-Bates v Director of Public Prosecutions [2000] HRLR 249, a decision of the English Divisional Court (Sedley LJ and Collins J) where the question was whether the conduct of the defendants was likely to provoke a breach of the peace. Although the Human Rights Act 1998 had not yet come into force, Sedley LJ considered the human rights dimension of such an offence. There, 3 women Christian fundamentalists preached from the steps of a cathedral and attracted a crowd some of whom were showing hostility. A policeman fearing a breach of the peace told the women to stop. They refused and were arrested for wilful obstruction of a police officer. The appellant was convicted. The issue on appeal was whether it was reasonable for the police officer, in the light of what he perceived, to believe that the appellant was about to cause a breach of the peace. Her appeal was allowed. At para 18, Sedley LJ[98] said: “…The question for PC Tennant was whether there was a threat of violence and if so, from whom it was coming. If there was no real threat, no question of intervention for breach of the peace arose. If the appellant and her companions were (like the street preacher in Wise v Dunning) being so provocative that someone in the crowd, without behaving wholly unreasonably, might be moved to violence he was entitled to ask them to stop and to arrest them if they would not. If the threat of disorder or violence was coming from passers-by who were taking the opportunity to react so as to cause trouble (like the Salvation Army in Beatty v Gilbanks), then it was they and not the preachers who should be asked to desist and arrested if they would not. … 20. … What Speaker’s Corner (where the law applies as fully as anywhere else) demonstrates is the tolerance which is both extended by the law to opinion of every kind and expected by the law in the conduct of those who disagree, even strongly, with what they hear…” 135. Earlier at para 6, Sedley LJ explained: “… In Beatty v Gilbanks (1882) 9QBD 308, this Court (Field J and Cave J) held that a lawful Salvation Army march which attracted disorderly opposition and was therefore the occasion of a breach of the peace could not found a case of unlawful assembly against the leaders of the Salvation Army. Field J, accepting that a person is liable for the natural consequences of what he does, held nevertheless that the natural consequences of the lawful activity of the Salvation Army did not include the unlawful activities of others, even if the accused knew that others would react unlawfully.” 136. In Jordan v Burgoyne [1963] 2 QB 744, the English Court of Appeal held that, a speaker who used words which threatened, abused or insulted, had to take his audience as he found it and, if the words spoken to that audience were likely to provoke a breach of the peace, the person could be convicted under s5 of the Public Order Act, 1936. Since the offence is concerned with the maintenance of public order, naturally, the likely reaction of the audience is highly relevant. Even so, as the passage from the judgment of Sedley LJ in Redmond-Bate quoted at para 134 above shows, if the threat of disorder or violence was coming from those who were taking the opportunity to react so as to cause trouble, it was they and not the speaker who should be asked to desist and arrested if they would not. In Jordan, Lord Parker CJ said, of the offending speech “I cannot myself, having read the speech, imagine any reasonable citizen, certainly one who was a Jew, not being provoked beyond endurance, and not only a Jew but a coloured man, and quite a number of people of this country who were told that they were merely tools of the Jews, and that they had fought in the war on the wrong side, and matters of that sort.”[99] Jordan recognised that people may be so provoked that they lost control of themselves, and that as a matter of common sense, persons with certain attributes might reasonably be expected to react more strongly than those without. I have no doubt that a person who behaved in a manner which actually provoked persons present beyond reasonable endurance or might do so may be convicted under s17B(2).[100] 137. Although, unlike these appeals, Brooker, Morse and Coleman were concerned with offences which contained no express requirement of any likelihood to cause violence, I do not believe that diminishes the relevance of those decisions. They show that, notwithstanding the minor nature of the relevant offences, they must be judged according to their possible impact on fundamental rights. There is no suggestion in any of these cases that had the relevant offence required a likelihood of violence, a less exacting approach would be adopted in determining whether the impugned conduct was serious enough to satisfy the requirement of the offence. Elias CJ at [31] said that he did not think the “word ‘disorderly’ can have a different meaning in s3 and s4. The additional element of seriousness in s3 arises from the likelihood of violence”. Indeed, the minority in Brooker in part explained their dissent by stressing the difference between s3 and s4. For example, McGrath J said “ …Section 3 … addresses behaviour at an extreme point on the range of what is disorderly.” Thomas J said “the use of the word ‘disorderly’ in s3 cannot have the effect of elevating the seriousness of the behaviour contemplated in s4(1)(a). They are different offences, the one more serious than the other.” I believe, when, as here, the offence requires a likelihood of violence, the impugned conduct must indeed be serious. How then does one decide whether the conduct is sufficiently serious? To that question I now turn. 138. In the present case, the appellants were exercising their freedom of speech, of assembly and of demonstration in a public place and in the presence of a large number of the public. Freedom of expression, assembly or demonstration would be meaningless if they can only take place in private or away from persons who may find the views, ideas or claims that an assembly or demonstration or speech is promoting annoying or offensive.[101] However, as Gleeson CJ noted “it is often the case that one person’s freedom ends where another person’s rights begin.”[102] The rights of those who may be affected by such conduct “obviously remain important” and the exercise of such rights by protesters must not exceed “the bounds of what is reasonable in the circumstances, (but) such bounds must not be so narrowly defined as to devalue, or unduly impair the ability to exercise, the constitutional right.”[103] 139. I would reconcile these competing rights, adopting the language used in the cases cited above, and say that those affected are expected to take a balanced, rights sensitive view, conscious of the requirement of the Hong Kong Bill of Rights, and would not be unreasonably moved to wounded feelings or real anger, resentment, disgust or outrage, particularly, when confronted by a protestor,[104] but the exercise of such rights by protesters must not exceed “the bounds of what is reasonable in the circumstances, (but) such bounds must not be so narrowly defined as to devalue, or unduly impair the ability to exercise, the constitutional right.” It is only when the conduct even when viewed against such a generous standard, “went well beyond what any citizen, public official or not, would have to tolerate”, in the circumstances in which it occurred[105], can such conduct properly be regarded as disorderly conduct within the meaning of s17B(1) or (2). Findings of Disorderly Conduct 140. In para 18 of the Statement of findings the learned magistrate said: “The appellants … rushed onto the stage at a highspeed. They should have anticipated resistance by the personnel and body clashes in the course of it. Eventually PW1 and PW2 indeed sustained injuries as a result. I am of the view that the appellants’ behaviour was unruly and offensive and therefore they did behave in a disorderly manner. I do not accept that the first appellant was genuinely exercising his freedom of expression because the law only protects the freedom to express oneself ‘peacefully’. The first appellant’s behaviour was not peaceful at all by whatever standard.” 141. Article 17 of the Hong Kong Bill of Rights[106] used the expression “right of peaceful demonstration”. A similar expression is used in Article 11 of the European Convention on Human Rights. In this context, I would note the following passages from the guidelines on Freedom of Peaceful Assembly 2nd edition published by OSCE/ODIHR (Organisation for Security and Co-operation in Europe / Office for Democratic Institution and Human Rights) dated 25 October 2010: “25. Peaceful assemblies: Only peaceful assembly is protected by the right to freedom of assembly. The European Court of Human Rights has stated that “[i]n practice, the only type of events that did not qualify as ‘peaceful assemblies’ were those in which the organizers and participants intended to use violence.”[107] Participants must also refrain from using violence (though the use of violence by a small number of participants should not automatically lead to the categorization as non-peaceful of an otherwise peaceful assembly – see para 164). An assembly should, therefore, be deemed peaceful if its organizers have professed peaceful intentions, and this should be presumed unless there is compelling and demonstrable evidence that those organizing or participating in that particular event themselves intend to use, advocate or incite imminent violence. 26. The term “peaceful” should be interpreted to include conduct that may annoy or give offence to persons opposed to the ideas or claims that it is seeking to promote, and even include conduct that temporarily hinders, impedes or obstructs the activities of third parties. Thus, by way of example, assemblies involving purely passive resistance should be characterized as peaceful. Furthermore, in the course of an assembly, ‘an individual does not cease to enjoy the right to peaceful assembly as a result of sporadic violence or other punishable acts committed by others in the course of the demonstration, if the individual in question remains peaceful in his or her own intentions or behaviour’.” 142. It appears that the learned magistrate took the view that because the appellants were not expressing their views peacefully, they were not entitled to the protection which the law extends to freedom of expression. Presumably, the learned magistrate’s finding that they had not expressed their views peacefully proceeded from his view that they had behaved in a disorderly manner. With respect, he ought to have first considered whether their conduct was disorderly because they exceeded the bounds of what is reasonable in the circumstances, such that they could not reasonably be expected to be tolerated. 143. Moreover, he should have considered the conduct of each of the appellants separately. Mr McCoy SC accepted that it was not and never had been the prosecution case that the appellants had acted in concert, or that their conduct should not be considered independently. The Appellants’ conduct 144. In paras 106 and 107 above, I have substantially reproduced the learned magistrate’s description of the disruption to the ceremony. We were shown the video recording at the hearing and I have watched the recording carefully since the hearing. They show that there were 10 persons on stage. They were seated in one single row. The Secretary was seated on the 5th chair from the left, between Sir Chung-kong Chow, at that time, the Chief Executive Officer of the MTR and Dr Ch’ien. Shortly after the Secretary had begun her speech, the 1st appellant stepped onto the stage. He was several feet away from the first person on the left (and I think well over 10 feet from the Secretary) when he scattered the hell money. It is obvious that he stepped onto the stage to scatter hell money, and that was his sole purpose. There was no apparent reaction from those seated on the stage.[108] After that was done he stood there quietly, waiting to be taken away. As he went peacefully, he shouted the slogan. The reaction of the Secretary has been recorded earlier. To complete the picture I should mention the finding that PW1 who was standing on the drained pond had tried to stop the 1st appellant and in doing so became unbalanced and hurt himself. The fact that the 1st appellant's conduct had provoked such a reaction may support a finding that his behaviour was disorderly within the meaning of s17B. I return to para 18 of the Statement of findings where the learned magistrate said: “The appellants being ‘uninvited guests’ rushed onto the stage at a high speed. They should have anticipated resistance by the personnel and body clashes in the course of it.” The learned magistrate had not distinguished between the appellants. It appears from the video recordings that the 1st appellant had not rushed onto the stage at a high speed. Unfortunately, the learned magistrate had not dealt with the 1st appellant’s conduct separately from the 2nd appellant’s conduct. Nor did he have in mind the need to reconcile the 1st appellant’s right of demonstration and, for example, the right of the organizers of the event to stop people from approaching or ascending the stage. In such circumstances, I believe it is unsafe to treat para 18 as a sufficient finding of disorderly conduct on the part of the 1st appellant. 145. Since I have come to the conclusion that there was no disorderly behaviour on the part of the 1st appellant, his appeal must be allowed. In Brutus v Cozens [1973] AC 854, the appellant who went onto the Centre Court in Wimbledon during a match to protest against apartheid, was charged under s5 Public Order Act 1936 with “insulting … behaviour … whereby a breach of the peace was likely to be occasioned”. His acquittal on finding that the behaviour was not insulting was upheld. So too in R v Ambrose (1973) Cr App R 538, where a young man was charged with using insulting words said to be likely to cause a breach of the peace under s5 of the Public Order Act of 1936. The Court held that the words complained of were incapable of being insulting words for the purposes of s5 prosecution and it was unnecessary to decide whether the fact that persons who were told about what was said were very angry and said that they would have felt like assaulting the appellant. The 2nd appellant 146. Mr Martin Lee SC who represented the 2nd appellant rightly accepted that the conduct of the 2nd appellant was of a different character from the 1st appellant.[109] The video tapes showed clearly that the 2nd appellant ran at great speed towards the stage and directly at the Secretary. He grabbed the microphone[110] whilst it was being used by the Secretary. PW2, the MC said he pulled his partner “to his side fearing that she would be knocked down”, as the learned magistrate said the video recording did not show that. The recording showed clearly that both Sir Chung Kong Chow and Dr Ch’ien stood up, and Dr Ch’ien actually came forward, clearly to protect the Secretary. The only reasonable inference is they were apprehensive of the 2nd appellant’s intention. The fact that the 2nd appellant only took away the microphone which the Secretary was using only became apparent later. I believe in rushing onto the stage, directly at the Secretary and grabbing the microphone which she was using, the 2nd appellant’s behaviour, even when viewed most generously, was disorderly. It went well beyond “what any citizen, public official or not, should have to tolerate”, in the circumstances in which it occurred.[111] That being the case I go on to consider whether the charges or any of them has been made out. Section 17B(1) 147. Section 17B(1) is concerned with disorderly behaviour “for the purpose of preventing the transaction of the business for which the public gathering was called together…” The two key words for consideration are “purpose” and “preventing”. 148. The video recordings show that just under one minute elapsed from the scattering of hell money (The 2nd appellant came on the scene shortly thereafter) to the Secretary resuming her speech. The question is whether the 2nd appellant’s conduct was for the purpose of preventing the transaction of the business for which the gathering was called. I will deal with preventing first. 149. Barnes J said at para 94 that “prevent” did not require that the transaction of the business to be brought completely to an end or being aborted. She said “because to stop for a moment due to hindrance also fits the (dictionary) definition. The magistrate’s finding as to the meaning of ‘prevent’ is correct”. 150. Mr McCoy submitted that an interruption of a gathering long enough to remove a heckler would be sufficient for the purpose of s17B (1). 151. Mr Martin Lee SC has traced s17B(1) to s1(1) of the English Public Meetings Act 1908 (“PMA 1908”) which was enacted in response to the suffragettes’ protests against Mr Lloyd-George, the Chancellor of the Exchequer, at Albert Hall, London in December 1908. A contemporary report in the New York Times described the incident in the following terms: “Some of the women were armed with whips, and they repelled vigourously every attempt to eject them. There were fierce tussles every few moments in different parts of the hall, and every time Mr Lloyd-George made an attempt to speak his voice was drowned with mingled groans and cheers. Finally the Chancellor, who for a quarter of an hour had been trying to get in a word, sat down, and the organist present tried to sooth the hysterical sisterhood by playing ‘What Can the Matter Be?’. It was of no use; pandemonium still reigned/…/At the end of half an hour or more the opposition was worn out, and Mr Lloyd-George was able to continue his speech with only occasional interruptions.”[112] 152. Mr Lee showed that there was vigorous debate in both Houses of Parliament on the impact the proposed law would have on the exercise of free speech, in particular when heckling and interjections frequently occurred at public or political meetings, and indeed, in the House of Commons. The debate showed that the proposed law was not directed at such conduct.[113] 153. Mr Lee told us that the predecessor of s17B(1) was first enacted as s5(1) of the Public Order Ordinance 1948 and in introducing the Public Order Bill before the Legislative Council in 1948, the Acting Attorney General said: “The past twenty years have witnessed the growth all over the world of political parties organized more thoroughly than has been the case before and prepared in some cases to enforce their views by forcible methods. Clauses 3-5[114] inclusive of the Bill, which are based on similar provisions in the United Kingdom, are designed to curb the activities of such political organisations while at the same time preserving and strengthening the right of public meeting of ordinary peaceful citizens.”[115] 154. Since “preventing” also impacts on important freedoms, it should be interpreted so that it is confined within reasonable bounds. Here, too, it is necessary to balance the conflicting rights of the protestor and the rights of the persons at the gathering. I believe such rights should be reconciled in the way I suggested in para 137 above. I believe, properly construed “preventing” requires an interruption to an extent beyond what could be expected to be tolerated by other persons in a democratic society,[116] such as breaking up or a substantial interruption of a gathering. 155. Given the cardinal importance of the freedom of expression, demonstration and assembly I have no doubt that the interruption for a minute or so in this case is insufficient. Moreover, s17B(1) provides in terms of preventing the transaction of the business for which the gathering was called. That argues against a brief interruption of a gathering. But, I do not agree with Mr Lee that nothing short of an abortion of the gathering would suffice. I have set out the report from the New York times to show that Mr Lloyd-George was able to continue after a half hour or so. If Mr Lee is right Public Meeting Act 1908 would have been enacted in vain. 156. Section 17B(1) also requires that the disorderly behaviour was for the purpose of preventing the transaction of the business for which the gathering was called. Barnes J was of the view that because the gathering was interrupted that was the purpose of the disorderly conduct. With respect I do not agree. 157. In Sweet v Parsley[1970] AC 132, Lord Morris said in connection with s5 of the Dangerous Drugs Act, under which it was an offence for an occupier to permit premises to be used for the purpose of smoking cannabis, that it “denote a purpose which is other than quite incidental or causal or fortuitous: they denote a purpose which is or has become either a significant one or a recognized one though certainly not necessarily an only one. There is no difficulty in appreciating what is meant if it is said that premises are used for the purposes of a dance hall or a billiard hall or a bowling alley or a hair-dressing saloon or a café.” I believe purpose under s17B(1) is used in a similar sense. 158. I see no basis for the conclusion that the 2nd appellant had as his purpose, the prevention of the transaction of the business for which the gathering was called. 159. For the above reasons, I would allow the 2nd appellant’s appeal against his conviction under s17B(1). Section 17B(2) 160. I turn to the s17B(2) offence. I start with the finding that the 2nd appellant’s conduct was disorderly within the meaning of s17B(2), and go on to consider whether the 2nd appellant conducted himself in that way, “with intent to provoke a breach of the peace, or whereby a breach of the peace is likely to be caused”. 161. Mr McCoy submitted that the 2nd Appellant’s conduct was not only disorderly, it amounted to a breach of the peace. I agree. Indeed, if, as seemed to be the case, he intentionally or recklessly caused another to apprehend immediate and unlawful violence, he could be charged with common assault. Breach of the Peace 162. What amounts to a breach of peace was settled by R v Howell [1982] QB 416. It is clear from Howell that violence is the central element in determining what amounts to a breach of the peace and: “an act done or threatened to be done which either actually harms a person, or in his presence his property, or is likely to cause such harm, or which puts someone in fear of such harm being done. There is nothing more likely to arouse resentment and anger in him, and a desire to take instant revenge, than attacks or threatened attacks upon a person’s body or property.” WatkinsLJ, delivering the judgment of the English Court of Appeal. [426G] 163. In Percy v DPP [1995] 1 WLR 1382, a decision of the English Divisional Court (Balcombe LJ and Collins J), the Divisional Court commented on the last sentence of Watkins LJ’s judgment cited above, and said: “It is clear from the last sentence that harm to property will constitute a breach of the peace only if done or threatened in the owner's presence because the natural consequence of such harm is likely to be a violent retaliation. Thus Reg v Howell makes it clear that there must be violence or threatened violence for there to be a breach of the peace to justify an arrest.” 164. I prefer to take the view that attacks or threatened attacks on property in the owner’s presence are tantamount to threatened violence on the owner and for that reason, it is a breach of the peace. It is commonsense that this kind of conduct would put the owner in fear of violence. 165. The fact that the 2nd appellant’s conduct amounted to a breach of the peace led to several submissions from Mr McCoy to which I will now turn. 166. Mr McCoy submitted that on the basis of the 2nd appellant’s own breach of the peace he could be convicted under s17B(2). With respect, I cannot agree. In Marsh v Arscott (1982) 75 Cr App R 211, a decision of the English Divisional Court (Donaldson LJ and McCullough J), which was concerned with s5 of the Public Order Act 1936, McCulloch J said (with the agreement of Lord Justice Donaldson) at p216 “This section is describing breaches of the peace which are brought about, or are likely to be brought about, by other words or behaviour occurring earlier, although usually not very long before. The phrase ‘whereby a breach of the peace is likely to be occasioned’ indicates that Parliament was concerned with cause and effect, i.e. with conduct which is likely to bring about a breach of the peace and not with conduct which is itself a breach of the peace and no more. Were this the law every common assault occurring in a public place would also be an offence against this section. Many such assaults will in fact be likely to lead very quickly to a breach of the peace, and these will be within the section; but, without more it is not enough that conduct which is threatening, abusive or insulting is of itself a breach of the peace.” 167. With respect, I agree. Indeed, the language of s17B(2) is so clear that I will say no more. 168. Mr McCoy also submitted that because the 2nd appellant’s conduct would inevitably lead to his arrest, the arrest would itself be a breach of the peace which was caused by the 2nd appellant’s conduct. He submitted that an arrest, however gentle, would involve some physical contact with the 2nd appellant by the trained personnel, which would technically be a breach of the peace. So, the 2nd appellant’s conduct was likely to cause a further breach of the peace. 169. The strongest support for this submission comes from the judgment of Lord Denning MR in Reg v Chief Constable of Devon and Cornwall, Ex parte Central Electricity Generating Board (CEGB) [1982] QB 458, 471, where he said : “There is a breach of the peace whenever a person who is lawfully carrying out his work is unlawfully and physically prevented by another from doing it. He is entitled by law peacefully to go on with his work on his lawful occasion. If anyone unlawfully and physically obstructs the workers -- by lying down or chaining himself to a rig or the like – he is guilty of a breach of the peace. Even if this were not enough, I think that their unlawful conduct gives rise to a reasonable apprehension of a breach of the peace. It is at once likely that the lawful worker will resort to self-help by removing the obstructor by force from the vicinity of the work so that he obstruct no longer. He will lift the recumbent obstructor from the ground. This removal would itself be an assault and battery -- unless it was justified as being done by way of self-help. Long years ago Holt CJ declared that ‘the least touching of another in anger is a battery’: see Cole v Turner (1705) 6 Mod. Rep. 149. Salmon on Torts, 17th edition (1977), p 120 adds that even anger is not essential: an ‘unwanted kiss maybe a battery.’ So also the lifting up of a recumbent obstructor would be a battery unless justified as being done in the exercise of self-help. But in deciding whether there is a breach of the peace or the apprehension of it, the law does not go into the rights or wrongs of the matter -- or whether it is justified by self-help or not. Suffice it that the peace is broken or is likely to be broken by one or another of those present. With the result that any citizen -- and certainly any police officer -- can intervene to stop the breach. ” 170. I have quoted a longer passage because in Percy the court regarded the first two sentences quoted above as erroneous. The court said: “Neither Lawton LJ or Templeman LJ agreed with (Lord Denning’s) observations. Indeed, it is in our view implicit in what each said that they took the view that some violence or threat of violence was necessary: see per Lawton LJ at 476 F-G and Templeman LJ at 480 A-C.” 171. CEGE was concerned with protests at a possible site for a nuclear power station. The protestors obstructed entrance to the site. They were well organized, for example, after an injunction was obtained against some, others took their place. So injunction was not an effective remedy. The organizers issued a leaflet giving objectors detailed instructions about peaceful methods of protest and informing them that no attempt could legally be made to manhandle them. The board wrote to the chief constable asking for his assistance in enabling it to perform its statutory duties by preventing further obstruction. The chief constable replied that without “a more definitive legal mandate” the police will not remove the obstructors since there was no actual or apprehended breach of the peace, nor an unlawful assembly. The Board brought judicial review against the police’s refusal to intervene because there was no imminent breach of the peace. 172. I believe the headnote, which did not reflect Lord Denning’s remarks relied on by Mr McCoy, has correctly stated the basis of the decision, it stated that the police had power to remove or arrest obstructors if there was a breach of the peace or the reasonable apprehension of it, or an unlawful assembly; that by wilfully obstructing the operations of the board the obstructors were deliberately breaking the law, so that the board was entitled to use the minimum force reasonably necessary to remove those obstructing the exercise of its powers, that the use of self-help in such circumstances engendered the likelihood of a breach of the peace, and accordingly, the police were entitled to be present in order to intervene as necessary. 173. Moreover, it is implicit in the passages in the judgments of Lawton and Templeman LJJ referred to in Percy that they did not share Lord Denning’s view. For example, Lawton LJ said “as soon as one person starts to, or makes to, lay hands on another, there’s likely to be a breach of the peace.” Lawton LJ did not say that as soon as one person lays hands on another there would be a breach of the peace. Templeman LJ said “The board and the police may instruct the obstructors to leave the site and warn them that if they do not leave the site and remain off the site the obstructors will be liable to be forcibly removed or arrested. If after such a warning the board enter the site with the object of completing the survey, the possibility of a confrontation with the obstructors will at once raise a danger of breaches of the peace when the board’s workmen seek to carry out their work and find the obstructors lying in their path. An obstructor who will not leave the site unless he is forcibly removed presents a threat and danger of a breach of the peace even if he disclaims any intention of causing a breach of the peace.” It is obvious that Templeman LJ did not regard the forcible (but justified) removal of the obstructors to be a breach of the peace. 174. I would add that in Marsh, after the passage quoted in para 166 above, McCullough J went on to say: “In the circumstances here, assuming the defendant to have been acting unlawfully in using threatening words and behaviour, no breach of the peace was likely to have been occasioned. No other person was likely to have broken the peace, and that the police were likely to do was arrest him, as they did. On that basis too an acquittal (of a charge under s5 of the Public Order Act) would, in my judgment, have been inevitable.” 175. If Mr McCoy is right, a breach of the peace which is likely to result in a lawful arrest, though the arrest be unresisted and peaceful, would also constitute an offence under s17B. In other words, a breach of the peace committed in public would also be an offence under this section, and the offender would not only be exposed to being bound over he would liable to imprisonment for 12 month and a fine of $5000 under s17B(2). I do not agree. 176. Mr McCoy further submitted that the 2nd appellant’s conduct was likely to cause a violent or unlawful reaction, from policemen or trained personnel responsible to keep order, which would itself be a breach of the peace. Such risk must be real and not a mere possibility. 177. In R v Li Wai Kuen [1973-1976] HKC 346, Huggins J (as he then was) said at 348 “There is a Canadian case which is in many respects similar to the present, where it was held to be unbelievable that the officer to whom the abuse was addressed could be provoked to commit a breach of the peace: R vZwicker (1938) 1 DLR 461. I myself had occasion to say something similar in this court in an unreported case some years ago. I would not suggest that there could never be a case where the abuse was so gross that even a police officer might be likely to be provoked into violent retaliation, particularly if the language used were also threatening, but it is to the credit of the police forces in those countries where the common law prevails that they conduct themselves with outstanding tolerance and good humour even in the face of provocation more grave than that in the present case.” This has remained the sentiment in Hong Kong. I would also note that in Coleman, Gleeson CJ said in relation to insulting word to a policeman that “It may eliminate, for practical purposes, any likelihood of a breach of the peace.” 178. I believe unlawful reaction from law enforcement officers and other trained personnel can be eliminated for practical purposes because of their training, discipline and professionalism. Moreover, the Community expects and the law requires no less. In the common-law world, I can say with confidence that, if they use excessive violence in effecting arrest, they will be visited with the full force of the law. 179. In para 22 of the Statement of Findings the learned magistrate said: “I am of the view that PW1, PW2 and other staff had shown restraint without over- reaction in stopping and removing the appellants. I find their action under those circumstances reasonable. Before the appellants took the actions, they must have reasonably anticipated obstruction, bodily clashes and even harm caused because of the use of force. I find that their behaviour involved violence breaching public order and the peace of society. In fact their conduct had provoked someone to follow suit. Therefore the only reasonable inference that I draw would be that they had the intent. I am satisfied that the prosecution had proved (with intent to provoke a breach of the peace, or whereby a breach of the peace was likely to be caused) beyond reasonable doubt.” 180. I believe the learned magistrate has insufficiently considered whether there was a real risk of a breach of the peace from the trained personnel on the stage. Since, apparently, no more force than was required was used to subdue the 2nd appellant, there was no breach of the peace by anyone other than the 2nd appellant. 181. That was also Barnes J’s view, she said: “67. I have repeatedly read the magistrate’s Statement of Findings. I do not see he had any analysis or finding as to the inference or effect which the appellants’ disorderly conduct might produce on other people at the scene who witnessed such conduct, except his description about members of the staff and the man dressed in black who, according to the magistrate, ‘followed the appellants’ example’. … 69. Leaving the man dressed in black for a moment, I have said earlier that the facts of this case showed that the staff members who intercepted the appellants exercised great restraint…” 182. As for the person said to have been provoked into following suit, the learned magistrate said at para 21(iv) “P3(d)(one of the tapes) showed that when the appellants were taken away, a protester dressed in black tried to get close to the stage but was stopped. The protester also shouted the same slogan as the 2nd appellant: ‘MTR fare increase, shameful.’ I can draw a reasonable inference that the protester was provoked or encouraged to follow the steps of the appellants.” Earlier, when dealing with the video recordings at para 6, under P3(d) the magistrate noted “12:13:46 — the protestor dressed in black raised his arms and shouted ‘MTR fare increase, shameful!’ ” 183. I have watched the recording most carefully, I see no evidence that the person dressed in black behaved in a disorderly manner, much less committed a breach of the peace. If that was the magistrate’s finding he should state it clearly and explain how he arrived at that conclusion. As it is I find the Magistrate’s finding difficult to follow and am unsure of his meaning. I place no reliance on it. 184. In relation to the man dressed in black Barnes J said: “76. … I do not think that the only reasonable inference is that the man was provoked by the 1st and/or 2nd appellants to try to go onto the stage … there was no solid ground on which the magistrate could find that the prosecution had proven that the 1st and 2nd appellants had the intent to provoke a breach of the peace.” With respect, I agree. 185. I therefore conclude that the learned magistrate had wrongly convicted the 2nd appellant under s17B(2). This is not a case where one could say that the 2nd appellant’s conduct was likely to cause a breach of the peace by persons attending the prize giving ceremony. A protest against MTR fare hike was highly unlikely to arouse strong emotion in those attending the ceremony. Had the 2nd appellant seized a microphone from a speaker at a political rally, to silence the speaker or so that the 2nd appellant could express a contentious view, a finding that he behaved in a disorderly manner, with intent to provoke a breach of the peace, or such that a breach of the peace was likely to be caused, might be justified. Disposition 186. For the above reasons, I would allow the appeals of the 1st and 2nd appellants and dismiss the cross appeals of the Government. Mr Justice Litton NPJ: Introduction 187. The question before this Court, as certified by the Appeal Committee on 27 November 2012, is as follows: “What are the elements of the offences created respectively by section 17B(1) and section 17B(2) of the Public Order Ordinance?” 188. The short answer, a complete answer, is this: The elements of those offences are those found in the statute itself: In the words used in the statute. Nowhere else. 189. The criminal law, here as elsewhere, is maintained by sanctions. Sanctions are imposed by words. Since the criminal justice system is largely codified, those words are found in statutes – Ordinances and subsidiary legislation – which regulate activities within the community, thus ensuring freedom for all under the law. 190. By far the largest number of criminal cases are tried in the magistrate courts. Day in, day out, magistrates apply the words in statutes to the facts of the case: To make a mystery of ordinary words in statutes is to grievously debase the rule of law, and make it an ineffective instrument of civil society. This is particularly so in a bilingual system[117] where obscurities could happen through translation. Nowadays, most cases in the magistrate courts are conducted in Chinese (as it was in the present case). The police force, given the duty of enforcing the law, operates almost exclusively in the Chinese language. When appellate courts give directions to the courts of trial, and to the law enforcement agencies, as to the meaning of words in criminal statutes, through the medium of the English language, they must be sensitive to this inherent disability. 191. The magistrate courts, governed by the Magistrates Ordinance, Cap 227, are courts of summary jurisdiction. That is clear from the long title. Words like “disorderly conduct” and “breach of the peace” have appeared in the statutes of Hong Kong for well over 100 years. Countless magistrates in myriad cases over the past century have applied those words to situations as found in court. 192. In the list of duties imposed on the police force under section 10 of the Police Force Ordinance, Cap 232, the very first is that of “preserving the public peace”. And, as the English Court of Appeal in R v Howell [1982] QB 416 at 427 reminds us, when breach of the peace or the reasonable apprehension of such breach, happens in the ordinary citizen’s presence, that citizen has the right to arrest the offender without warrant. It goes further: At common law, the citizen not only has such right, he has a duty to take reasonable steps “to make the person who is breaking or threatening to break the peace refrain from doing so, and those reasonable steps in appropriate cases will include detaining him against his will”: See Lord Bingham’s judgment in R(Laporte) v Chief Constable of Gloucestershire [2007] 2 AC 105 at 125-6, quoting Lord Diplock in Albert v Lavin [1982] AC 546 at 565. When the common law imposes such a duty on the ordinary citizen, it may seem surprising that there could be doubt as to what “breach of the peace” means in law. And yet, as will be explained later, that is precisely the case. 193. What behaviour amounts to disorderly conduct, or when a situation constitutes a breach of the peace, or a threat thereof, depends on the circumstances of the time and place. The social context in which the events occur forms an important part of the picture. As the author of Public Order: A guide to the 1986 Public Order Act [Format Publishing 1987] at p.93 notes: The conduct of a football crowd would be disorderly if it were repeated in a theatre during a performance. 194. The question is one of degree. It is purely a question of fact, for the fact-finding tribunal to decide. The appellate courts would do the system, and the community, great harm if they interfered too readily in such fact-finding function, or added degrees of sophistication to such function which are unwarranted. 195. As Lord Reid said in Cozens v Brutus [1973] AC 854 at 861C, the meaning of an ordinary word of the English language is not a question of law. It is a matter for the fact-finding tribunal to consider, not as a matter of law, but as fact, whether in the whole of the circumstances the words of the statute do or do not apply to the facts which have been proved. 196. All that having been said, there is in fact a problem with the scope and meaning of the expression “breach of the peace” (“破壞社會安寧”) and this is a case where the expression warrants close scrutiny. “Breach of the peace” 197. There is no offence known to the criminal law as “breach of the peace”; a person cannot be charged with breaching the peace. And yet, under s 61 of the Magistrates Ordinance, he can be bound over to keep the peace[118], and is liable to be imprisoned in default of compliance with the order. The power given to magistrates under s 61 is to administer preventative justice: To prevent a future event happening when public order might be disturbed by a breach of the peace. But this brings us no further to defining the scope of the expression “breach of the peace”, nor to identify with any precision when disorderly behaviour is likely to cause a breach of the peace, which lies at the heart of the offence created under s 17B(2) of the Public Order Ordinance, Cap 245. 198. Take the recent case of a demonstration by a group of people outside the Liaison Office of the Central People’s Government at Connaught Road West, Hong Kong. Secretary for Justice v Chiu Hin Chung [2013] 1 HKLRD 227. Mills barriers had been erected outside the main gate. Police officers were stationed both inside and outside the barriers. There were security guards within the forecourt of the Liaison Office, as well as the manager of the security company. Outside, in the demonstration area, there were reporters present, apart from police officers, security personnel and, of course, the demonstrators themselves. The two defendants (like the appellants here) were charged with disorder in a public place under s 17B(2) of the Public Order Ordinance. Their disorderly conduct was the following: (i) The 1st defendant stood on a barrier and threw a placard and a bottle containing about 1.7 kg of cornstarch into the forecourt of the Liaison Office. A piece of paper was stuck to the bottle with the Chinese characters meaning “toxic melamine”. (ii) The second defendant threw a plastic bag containing cornstarch into the forecourt. The bag broke open and scattered the power, some landing on three people, two of whom complained of discomfort in the eyes as a result, and one of itchiness in the forehead. 199. Shortly after this episode, the demonstration was called off. The entire event, from beginning to end lasted about 18 minutes. Had they been arrested, taken away and brought before a magistrate to be bound over to keep the peace for, say, 18 months, under s 61 of the Magistrates Ordinance, that would most probably have been justified. The magistrate would have examined their past disorderly behaviour and considered whether such behaviour might, on a future occasion, lead to a breach of the peace. But they were charged under s 17B(2)[119], which looks to the actual situation on the ground at the time of the charge. 200. The question at trial then was whether, in light of all the circumstances, their conduct had produced a breach of the peace or was likely to have caused a breach of the peace? 201. The magistrate concluded that those present at the demonstration – the reporters, the security personnel, the police officers, the demonstrators themselves – were unlikely to react to the defendants’ disorderly conduct by resorting to violence or riotous behaviour, and acquitted the defendants. The government’s appeal to the High Court was dismissed. In short, the courts found that the public peace had not been disturbed or was likely to be disturbed: No-one was likely to engage in riotous behaviour as a result of the defendants’ action: Their own disorderly behaviour had no significant impact on the behaviour of others, apart from the fact that they were arrested and taken away. 202. This line of reasoning follows cases of some antiquity: For example Wise & Dunning [1902] 1 KB167 where the question was whether the stipendiary magistrate had authority to bind over the appellant to keep the peace. He, a Protestant lecturer, had held meetings in public places in Liverpool, causing large crowds to gather. He used gestures and language highly insulting to the religion of the Roman Catholic inhabitants. The consequence of his acts at one meeting was to cause some of the listeners to rush towards him; the police intervened and got him away to safety. The appellant had not himself committed any offence. His conduct was not disorderly. But he had provoked a violent reaction from others. His counsel, F.E. Smith (later Lord Birkenhead LC) argued that as his conduct was lawful, and was not disorderly, he could not be bound over simply because the conduct of others might be unlawful. The court disagreed. He could be bound over to keep the peace if the natural consequence of his acts was that others might breach the peace. 203. This bears out the point that counsel makes in this case: For the offence under s 17B(2) to be established, the court looks not only to the acts of the accused but also to the natural reaction of others to his acts. This is always in the context of public order. The aim of the statute is to keep the public peace. In considering whether acts are likely to cause a breach of the peace, the following question is relevant: Are persons in the vicinity likely to engage in an affray, to behave riotously, to act violently or to threaten violence as a result of the behaviour of the accused? Violence and Threat of Violence 204. As can be seen, this view of what constitutes “behaviour likely to cause a breach of the peace” focuses on violence and threats of violence on the part of others. That this seems to be the correct view is borne out by the judgments of Lawton and Templeman LJJ in R v Chief Constable of Devon & Cornwall [1982] 1 QB 458[120] where, at 476G Lawton LJ speaks of violence and tumultuous behaviour being apprehended by the police, justifying arrest for breach of the peace; and at 480C Templeman LJ used the memorable phrase: “Even Mahatma Gandhi discovered to his sorrow that in the conduct of ordinary mortals passive resistance remains passive only so long as the resistance is successful”. The learned Lord Justice was here adverting to the possibility of the Electricity Board’s workmen forcibly removing the objectors and the objectors resisting such removal, causing a breach of the peace: creating a situation where there would be a danger of a breach of the peace, even if the objectors disclaimed any intention of causing such breach. 205. Lord Denning M.R., in the same case at 471, would go further, and would give a wider meaning to the expression breach of the peace. He stated: “There is a breach of the peace whenever a person who is lawfully carrying out his work is unlawfully and physically prevented by another from doing it. He is entitled by law peacefully to go on with his work on his lawful occasions. If anyone unlawfully and physically obstructs the worker – by lying down or chaining himself to a rig or the like – he is guilty of a breach of the peace.” This tends to suggest that breach of the peace occurs simply through act of the objectors unlawfully interfering with the work of the Electricity Board workmen, without considering whether those workmen might take physical steps to have them removed, producing a situation of violent confrontation, or the threat of violent confrontation. This is out of line with the mainstream of authority on the point: see for instance the later case of R (Laporte) v Chief Constable of Gloucestershire (supra) at 134: A case of preventive action to forestall a breach of the peace. At 134B Lord Rodger of Earlsferry speaks of arrangements to contain the demonstrators from reaching their destination (an airbase heavily used by the United States Air Force for hostile operations against Iraq): Should the demonstrators reach their destination, “there would be outbreak of violence. This would constitute a breach of the peace within the definition in R v Howell (Errol) [1982] QB 416, 427”. 206. Lord Denning’s view in R v Chief Constable of Devon and Cornwall was rejected by the Divisional Court in Percy v D.P.P. [1995] 1 WLR 1382 (Balcombe LJ and Collins J) where, at 1394, applying the definition in R v Howell, that court reiterated the proposition that breach of the peace necessarily involved violence or threats of violence. In Percy[121] there was no evidence to suggest that violence was the natural consequence of the defendant’s action. The order made in the magistrate court binding her over to keep the peace was accordingly quashed. Preventative Justice v Punitive Justice 207. For a charge under s 17B(2), the starting point in every case is whether the conduct of the defendant was disorderly. Even if it was not, as in Wise v Dunning then, if the defendant’s action was calculated to provoke others into violence or acts of violence, and such acts (by others) is a real possibility, a binding over order under s 61 of the Magistrates Ordinance can properly be made. But no charge under s 17B(2) in such circumstance can possibly succeed, because disorderly behaviour is an ingredient of the offence. This underlines the difference between preventative justice under s 61 of the Magistrates Ordinance and punitive justice under s 17B(2) of the Public Order Ordinance. 208. Since violence or threat of violence is inherent in the concept of “breach of the peace”, the next question must be: Where is the threat coming from? If it is only from the defendant himself, there can be no breach of the peace, though he would probably be guilty of an offence such as common assault, or worse. 209. Having (hopefully) defined the scope of “breach of the peace” in a charge brought under s 17B(2) of the Public Order Ordinance, I turn to the proceedings in the courts below. The Magistrate’s Findings 210. The actions of the two appellants, leading to the charges brought against them under s 17B(2) were almost simultaneous, but there is no suggestion that they acted in concert. 211. What happened was this. On 10 April 2011, a section of Statue Square in Central Hong Kong had been cordoned off by mills barriers. This was for a prize-giving ceremony organized by the MTR Corporation. A stage had been erected for the purpose. Persons involved in the prize-giving ceremony were seated on the stage including the prize-winners and, at a table in front of the stage, the dignitaries including the chairman of the MTR Corporation Mr Ch’ien and the Secretary for Transport and Housing Madam Eva Cheng. There were a number of MTR Corporation staff members and security personnel inside the barrier. There were also some police officers present, both inside and outside the barrier. Outside the barrier, some demonstrators were shouting slogans, protesting against fare increase by the MTR Corporation. 212. What then happened was this. The two appellants broke through the barrier, and rushed towards the stage where Madam Cheng was giving a speech in front of a microphone, with Mr Ch’ien sitting alongside. The 1st appellant rushed onto the stage and scattered “hell money” in the air and shouted “shame on MTR for their fare hike”[122]. After that, he stood there quietly and made no resistance when he was held. Almost simultaneously, the 2nd appellant dashed onto the stage; Mr Ch’ien pulled Madam Cheng to one side; the 2nd appellant got hold of Madam Cheng’s microphone and shouted “shame on MTR for their fare hike” through the microphone. 213. PW1, the event organizer, had tried to tackle with the 1st appellant, got hold of his thigh but lost balance, falling down and hurting his left elbow. PW2, the master of ceremonies, likewise tried to grapple with the 2nd appellant and sustained minor injuries. As mentioned earlier, a number of guards came forward and held the 1st appellant, who put up no resistance. As regards the 2nd appellant, the security personnel separated Madam Cheng from the 2nd appellant; he put up a slight resistance and the guards took him away, with him holding the microphone. 214. At about this time another person, wearing a black upper garment (“man in black”) managed to penetrate the barrier, went towards the stage, apparently trying to mount it, and was stopped by the staff. This person also shouted “Shame on MTR for their fare hike”. 215. On these facts, the magistrate concluded that each of the appellants had behaved in a disorderly manner with intent to provoke a breach of the peace in terms of s 17B(2) and sentenced each of them to 14 days’ imprisonment. He summed up his findings in this way (§22): “Before the appellants took the actions, they must have reasonably anticipated obstruction, bodily clashes and even harm caused because of the use of force. I find that their behaviour involved violence breaching public order and the peace of society. In fact their conduct had provoked someone to follow suit. Therefore the only reasonable inference that I draw would be that they had the intent.” 216. As can be seen, the finding of intent to provoke a breach of the peace was based to a large degree on the action of the man in black. The magistrate concluded that the man in black had been provoked or at least emboldened by the action of the two appellants into breaching the barrier and attempting to mount the stage. Appeal to Barnes J 217. On appeal to the High Court, the appellate judge Barnes J upheld the magistrate’s finding of disorderly behaviour: a finding which could hardly be in doubt in this case. The behaviour of the appellants penetrating the barrier and rushing onto the stage was momentarily threatening; they brought the prize-giving ceremony to a halt, albeit for a short while. The focus of the appeal before Barnes J was therefore on the finding of intent: Intent to provoke a breach of the peace. 218. The judge held that whilst the 1st appellant’s act of rushing onto the stage and scattering “hell money” was “very insulting”, and that the 2nd appellant had acted in a violent manner in rushing onto the stage and snatching the microphone whilst Madam Cheng was speaking, their intent in both cases was to draw people’s attention to their grievances: Moreover, the judge found that the action of the man in black was equivocal; there was no evidence to sustain the magistrate’s conclusion that his action was the result of anything done by the appellants; accordingly the finding of intent to provoke a breach of the peace could not upheld. 219. The judge’s approach was this: Were the persons within the enclosure – the staff, the security personnel, the persons on the stage – likely to be provoked by the appellant’s disorderly behaviour to act violently and breach the peace? The judge framed the issue in this way: Whilst admittedly the behaviour of the appellants was violent and had caused minor injuries to two staff members, was “their conduct likely to cause the two injured persons or other people who might be injured by their conduct to take revenge and result in a breach of the peace?” (§61 of her judgment). She went on to say that the staff members had acted “in a very restrained and professional manner….They would not themselves breach the peace when, in the course of the performance of their duty, some people disrupted the order”. The result was that Barnes J quashed the conviction on s 17B(2), but substituted for that a conviction on s 17B(1). Section 17B(1) 220. Section 17B(1) provides: “(1) Any person who at any public gathering acts in a disorderly manner for the purpose of preventing the transaction of the business for which the public gathering was called together or incites others so to act shall be guilty of an offence and shall be liable on conviction to a fine of $5,000 and to imprisonment for 12 months.” 221. In my respectful view, the substituted conviction under s 17B(1) cannot be sustained. As mentioned earlier (para 218) the judge had found that the appellants’ intent was to draw people’s attention to their grievances: This does not sit well with her finding that their purpose was to prevent the transaction of the business taking place. I agree with counsel’s submission that for purpose under s 17B(1) to be made out, there has to be something more than a brief interruption to the business of the public gathering: There was no evidence here that either of the two appellants had set out to prevent the prize-giving ceremony taking place. The conviction of the two appellants under s 17B(1) must be quashed. 222. The matter must then be looked at afresh by this Court. Section 17B(2) 223. As can be seen from footnote 119 above, where breach of the peace was likely to be caused by the appellants’ disorderly behaviour, the offence is complete: Though, if the intent to provoke a breach of the peace was made out, a magistrate would obviously treat the offence as one of greater culpability. 224. The point that must be borne in mind in every case is this: Where a person behaves in a disorderly manner in a public place – a place where a large number of people are gathered – the result is often unpredictable. The 1st Appellant 225. Whilst the action of the 1st appellant in rushing onto the stage might, momentarily, have been frightening, there was no finding by the magistrate that his behaviour was seen by any one as threatening violence. The magistrate’s finding was that Secretary Cheng, then in the course of giving her speech, was startled rather than alarmed: Secretary Cheng’s reaction to the 1st appellant’s act of scattering “hell money” was to say: “This is not some kind of sport. Children, you should not imitate that….”. The 1st appellant shouted “shame on MTR for their fare hike” and was almost immediately held by security personnel. He put up no resistance and was quietly led away. 226. The appellate judge, as mentioned earlier (para 218 above), concluded that the finding of intent to provoke a breach of the peace was wrong in law. It was not sustained by the evidence. I agree with that conclusion. The question still remains: Was it established beyond a reasonable doubt, on the facts as found, that the disorderly conduct of the 1st appellant was likely to cause a breach of the peace? 227. In my judgment, the answer is No. The fact that PW1 sustained a mild injury in trying to grapple with the 1st appellant is not enough. The magistrate said (§22) that the appellants must have anticipated “obstruction, bodily clashes and even harm caused because of the use of force” and therefore their behaviour had breached public order and “the peace of society”. In my judgment, the brief episode whereby PW1 failed to physically stop the 1st appellant from mounting the stage is not enough to constitute a breach of the peace. There is no suggestion that it led to any kind of confrontation between them. It follows that his conviction under s 17B(2) cannot be restored. The 2nd Appellant 228. The disorderly behaviour of the 2nd appellant was initially more menacing. When he jumped onto the stage and rushed towards Secretary Cheng, the reaction of those on the stage was one of alarm. Mr Ch’ien sprang from his seat as the 2nd appellant snatched away the microphone in front of Secretary Cheng. The magistrate’s finding was to this effect: “PW2 ... the MC [master of ceremony]... pulled his partner to his side fearing that she would be knocked down. … He tried to stop the 2nd appellant but was not fast enough. He believed that the 2nd appellant had taken hold of Secretary Cheng’s microphone and said something. Then a group of personnel together with him carried the 2nd appellant down the stage. He suffered minor injuries during the struggle: ... The 2nd appellant put up slight resistance but soon calmed down.” 229. Plainly the 2nd appellant had, by his disorderly behaviour, caused a situation where persons on the stage had an immediate and real fear of physical harm. The proceedings on the stage – the prize-giving ceremony organized by the MTR Corporation – were interrupted and brought briefly into chaos. He was undoubtedly guilty of common assault. 230. But, in relation to a charge of public disorder causing (or likely so cause) a breach of the peace, the court looks beyond the misbehaviour of the accused to the reaction of others at the scene. As to this, the appellate judge said this (§67): “I have repeatedly read the magistrate’s Statement of Findings. I do not see he had made any analysis or finding as to the influence or effect which the appellants’ disorderly conduct might produce on other people at the scene who witnessed such conduct, except his description about members of the staff and the man dressed in black who, according to the magistrate, ‘followed the appellants’ example’.” 231. I respectfully agree with her conclusion. It follows that, on the magistrate’s findings of fact, no conviction for disorderly conduct likely to cause a breach of the peace can be sustained. Conclusion 232. In my judgment, had the 2nd appellant been brought before a magistrate to be bound over to keep the peace under s 61 of the Magistrates Ordinance, such order would almost certainly be sustained. The case of the 1st appellant is more equivocal, depending on a magistrate’s view of the matter overall, including any risk of future misconduct. 233. I have had the advantage of reading in draft Ribeiro PJ’s judgment and am in total agreement with that judgment. I concur in the orders he proposed. Lord Millett NPJ: 234. I agree with the Judgments of Mr Justice Chan, Acting CJ, Mr Justice Ribeiro PJ and Mr Justice Litton NPJ. Mr Justice Chan, Acting CJ: 235. The Court unanimously allows the appellants’ appeals and dismisses the HKSAR’s appeal. The question of costs will be dealt with in accordance with para. 101 in Mr Justice Ribeiro’s judgment. Mr Martin Lee SC, Mr Randy Shek and Mr Carter Chim, instructed by Ho Tse Wai, Philip Li & Partners, for Wong Hin Wai Mr H Y Wong, instructed by Ho Tse Wai, Philip Li & Partners, for Chow Nok Hang Mr Gerard McCoy SC, instructed by the Department of Justice and Mr Martin Hui, SADPP and Mr Derek Lau, PP of the Department, for HKSAR [1] Cap 245. [2] ESCC 3256/2011 (11 January 2012). [3] HCMA 193/2012 (20 July 2012). [4] FAMC No 45/2012 (27 November 2012). [5] Reflecting Article 21 of the International Covenant on Civil and Political Rights (“ICCPR”). [6] Cap 383. [7] Leung Kwok Hung v HKSAR (2005) 8 HKCFAR 229 at §20. [8] Ibid, at §§1 and 2; Yeung May Wan v HKSAR (2005) 8 HKCFAR 137 at §1. [9] Ng Ka Ling v Director of Immigration (1999) 2 HKCFAR 4 at 28-29; Gurung Kesh Bahadur v Director of Immigration (2002) 5 HKCFAR 480 at §24; Leung Kwok Hung v HKSAR (2005) 8 HKCFAR 229at §16. [10] (2005) 8 HKCFAR 229 at §17. [11] There is no difference between the Article 17 expression “in conformity with the law” and the Basic Law Article 39(2) expression “as prescribed by law”: Leung Kwok Hung v HKSAR (2005) 8 HKCFAR 229 at §18. The requirement is explained in Shum Kwok Sher v HKSAR (2002) 5 HKCFAR 381 and Leung Kwok Hung v HKSAR (2005) 8 HKCFAR 229 at §§25-29. [12] HKSAR v Ng Kung Siu & Another (1999) 2 HKCFAR 442 at 460; Leung Kwok Hung v HKSAR (2005) 8 HKCFAR 229 at §33. [13] Leung Kwok Hung v HKSAR (2005) 8 HKCFAR 229 at §35. [14] Ibid at §36. [15] Ibid at §35. [16] (2005) 8 HKCFAR 229. [17] At §69, the Court adding that the concept of “public order (ordre public)” went further. [18] Leung Kwok Hung v HKSAR (2005) 8 HKCFAR 229 at §95. [19] (1882) 9 QBD 308. [20] [1902] 1 KB 167. [21] Redmond-Bate v DPP [2000] HRLR 249 at §6. [22] (2005) 8 HKCFAR 137. [23] Contrary to section 4A of the Summary Offences Ordinance (Cap 228). [24] The Court held that since a constitutionally protected right was involved: “While the interests of those exercising their right of passage along the highway obviously remain important, and while exercise of the right to demonstrate must not cause an obstruction exceeding the bounds of what is reasonable in the circumstances, such bounds must not be so narrowly defined as to devalue, or unduly impair the ability to exercise, the constitutional right.” (§44) [25] HKSAR v Au Kwok-kuen [2010] 4 HKC 235. [26] Morse v Police [2012] 2 NZLR 1 at §110. [27] Fagan v Metropolitan Police Commissioner [1969] 1 QB 439 at 444; R v Ireland, R v Burstow [1998] AC 147 at 161-162. [28] Set out in Section A above. [29] “Any person who in any public place or at any public meeting uses threatening, abusive or insulting words or behaviour with intent to provoke a breach of the peace or whereby a breach of the peace is likely to be occasioned, shall be guilty of an offence.” [30] It not having been alleged that they were guilty of using threatening, abusive or insulting words. [31] [1945] JC 29 at 33. The Court was concerned with the Glasgow Police Act, 1866, which by section 135(5), made it an offence if a person was “riotous, disorderly or indecent in his behaviour.” [32] [1971] NZLR 924 at 925. The defendant was charged under section 3D of the Police Offences Act 1927 which stated: “Every person commits an offence ... who, in or within view of any public place ... behaves in a riotous, offensive, threatening, insulting or disorderly manner ...” [33] Under section 5 of the Public Order Act 1986: Chambers v DPP [1995] Crim LR 896. [34] [1973] AC 854. [35] At 861. [36] (2004) 220 CLR 1 at §12. [37] (2004) 220 CLR 1. [38] [2007] 3 NZLR 91. [39] [2012] 2 NZLR 1. [40] (2004) 220 CLR 1 . [41] Under section 7(1)(d) of the Vagrants, Gaming and Other Offences Act 1931 (Q). [42] [2007] 3 NZLR 91. [43] [2012] 2 NZLR 1. [44] Under Section 4(1)(a) of the Summary Offences Act 1981. [45] Coleman v Power (2004) 220 CLR 1 at §§5 and 6. See also Brooker v Police [2007] 3 NZLR 91 at §25. [46] (1997) 189 CLR 520. [47] As required, in New Zealand, by section 6 of the New Zealand Bill of Rights Act. [48] Brooker v Police [2007] 3 NZLR 91 at §§31, 41 and 45 per Elias CJ, at §56 per Blanchard J, and at §90 per Tipping J; Morse v Police [2012] 2 NZLR 1 at §§2-3 and 36 per Elias CJ, at §§62 and 67 per Blanchard J, at §§69-71 per Tipping J and at §§102-§103. [49] Ibid, at §§33 and 45. [50] Coleman v Power (2004) 220 CLR 1 at §§183-184, 193-194 per Gummow and Hayne JJ (Kirby J agreeing). [51] [1982] QB 416. [52] At 427. [53] [2007] 2 AC 105 at §§27-28. [54] Steel v United Kingdom (1998) 28 EHRR 603. [55] Percy v DPP [1995] 1 WLR 1382 at 1391, citing Wise v Dunning [1902] 1 KB 167; R v Morpeth Ward Justices, Ex parte Ward (1992) 95 Cr App R 215. [56] Read v Jones (1983) 77 Cr App R 246 at 251-252. [57] McBean v Parker (1983) 147 JP 205 (Div Ct). [58] Percy v DPP [1995] 1 WLR 1382 at 1394. [59] R (Laporte) v Chief Constable of Gloucestershire [2007] 2 AC 105. [60] Ibid at §§39 and 45. [61] Ibid at §28. However, under section 19 of the Public Order Ordinance (Cap 245), a participant in an unlawful assembly (defined by section 18) who commits a breach of the peace is made guilty of the offence of riot. [62] 34 Edw 3. [63] Cap 221, section 109I: “A judge, a District Judge or a magistrate shall have, as ancillary to his jurisdiction, the power to bind over to keep the peace, and power to bind over to be of good behaviour, a person who or whose case is before the court, by requiring him to enter into his own recognizances or to find sureties or both, and committing him to prison if he does not comply.” [64] Cap 227, sections 36, 61-63. See Lau Wai Wo v HKSAR (2003) 6 HKCFAR 624. [65] Lau Wai Wo v HKSAR (2003) 6 HKCFAR 624 at §47. [66] In Section C above. [67] Leaving aside elements such as “in a public place” which are not presently controversial. [68] [1963] 2 QB 744 at 749. [69] (1982) 75 Cr App R 211 at 216. [70] [1983] QB 92 at 98. [71] [1995] 1 WLR 1382 at 1392. [72] (2004) 220 CLR 1 at §§9-10. [73] In Hong Kong, the first of the duties of the police listed in section 10 of the Police Force Ordinance Cap 232 is the duty “to take lawful measures for ... preserving the public peace ...” See also R (Laporte) v Chief Constable of Gloucestershire (HL(E)) [2007] 2 AC 105 at §66 on the duty of the police “to take whatever steps are reasonably necessary to prevent a breach of the peace”. [74] (1982) 75 Cr App R 211 at 216. [75] Ibid. [76] [2000] HRLR 249 at §9. [77] [1995] 1 WLR 1382. [78] R v Li Wai Kuen [1973-1976] HKC 346 at 348; HKSAR v Morter [2003] 2 HKLRD 510. [79] HKSAR v Pearce [2006] 3 HKC 105 at §29. [80] Judgment §64. [81] Judgment §67. [82] Judgment §62. [83] Judgment §§82-83. [84] Namely, the “MTR Hong Kong Race Walking 2011” which was jointly organised by the MTR and the Hong Kong Amateur Athletic Association. [85] These included several mills barriers. [86] The 1st appellant was a District Councillor Assistant and the 2nd appellant was a Councillor Assistant. [87] At trial, there was no separate reliance on “noisy manner”. [88] As Mr McCoy SC for the prosecution rightly accepted. [89] Leung Kwok Hung and others v HKSAR (2005) 8 HKCFAR 229 at p245 para 1 and Yeung May Wan and Others v HKSAR (2005) 8 HKCFAR 137 at p148 para 1. [90] Leung Kwok Hung, the majority judgment of Li CJ, Chan, Ribeiro PJJ and Sir Anthony Mason NPJ. Bokhary PJ dissented on the result, but his views on the importance of such rights were equally emphatic. [91] Leung Kwok Hung at pp252-254. [92] Bokhary PJ delivered a separate concurrent judgment. He said at 144 “what the public can reasonably be expected to tolerate is a question of fact and degree. But when answering that question, a court must always remember that preservation of the freedom in full measure defines reasonableness and is not merely a factor in deciding what is reasonable.” [93] Which, it will be remembered, does not have to be likely “to cause violence against persons or property to start or continue”. [94] Human right values. [95] Coleman should be noted for the importance which the High Court placed on the impact which such an offence may have on the freedom of expression. [96] As Elias CJ pointed out in Brooker at footnote 41, at p107, Gummow, Kirby and Hayne JJ held that s7(1)(d) required the likelihood of a breach of peace, but not Gleeson CJ, McHugh, Callinan & Heydon JJ who were “influenced in particular by the fact that the Australian Legislation had removed the earlier requirement of breach of the peace (as the New Zealand legislation had also done).” [97] However, it should be noted that whilst McHugh J took the view that s7(1)(d) merely required that they be uttered in a public place and were calculated to hurt the personal feelings of Constable Power and they did so, he agreed that the conviction had to be set aside because [81] “the words used by the appellant were a communication on political or government matters”, and s7(1)(d) which imposed an unqualified prohibition on the use of insulting words in public cannot be justified as compatible with constitutional freedom. McHugh J also expressed the view that [102] “Without seeking to state exhaustively the qualifications needed to prevent a infringement of the freedom of communication, the law would have to make proof of a breach of the peace and the intention to commit the breach elements of the offence. It may well be the case that, in the context of political communications, further qualifications would be required before a law making it an offence to utter insulting words would be valid.” [98] With the agreement of Collins J. [99] At p 748. [100] Of course, Jordan does not absolve those who reacted with violence or threat of violence from criminal liability. [101] Sedley LJ famously said in Redmond-Bate “Free speech includes not only the inoffensive but the irritating, the contentious, the eccentric, the heretical, the unwelcome and the provocative provided it does not tend to provoke violence. Freedom only to speak inoffensively is not worth having.” [102] Coleman at para 32, quoting the submission of the Solicitor General for Queensland. [103] Yeung May Wan at para 44, see para 117 above. [104] Per Blanchard J in Morse at para 64, see para 128 above. [105] Per McGrath J in Brooker at [146], see para 124 above. [106] Which is based on Art 21 of the International Covenant on Civil and Political Rights. [107] Footnote 56 in the text referred to Cisse v France, a decision of the European Court of Human Rights 9.4.2002 at para 37. [108] It seems that neither Sir Chung-kong Chow nor Dr Ch’ien was aware of the 1st appellant’s presence until after he had scattered the hell money. [109] Mr Lee also rightly accepted that the gathering was a public gathering and that it took place in a public place. [110] The microphone was on a stand. [111] Per McGrath J in Brooker at [146], see para 124 above. [112] “Suffragettes Riot in Albert Hall”, New York Times, 6th December 1908. See also House of Lords Hansard,18th December 1908 vol 198 cc2206-11. [113] See House of Lords Hansard, 18th December 1908 vol 198 cc2206-11; and House of Commons Hansard, 19th December 1908 vol 198 cc2328-43. [114] Clause 5 became s5(1) of the POO 1948. [115] Legislative Council Hansard, 20th October 1948. [116] Majority holding in Morse, see para 126 above. [117] Article 9, Basic Law: “In addition to the Chinese language, English may also be used as an official language by the executive authorities, legislative and judiciary of the Hong Kong Special Administrative Region”. [118] 61. Exercise on complaint of power to bind over to keep the peace (1) The power of a magistrate, on complaint of any person, to adjudge a person to enter into a recognizance and find sureties to keep the peace or to be of good behaviour towards such first-mentioned person shall be exercised by an order upon complaint, and the provisions of this Ordinance shall apply accordingly, and the complainant and defendant and witnesses may be called and examined and cross-examined, and the complainant and defendant shall be subject to costs, as in the case of any other complaint. (2) The magistrate may order the defendant, in default of compliance with such last-mentioned order, to be imprisoned for 6 months. [119] “(2) Any person who in any public place behaves in a noisy or disorderly manner, or uses, or distributes or displays any writing containing, threatening, abusive or insulting words, with intent to provoke a breach of the peace, or whereby a breach of the peace is likely to be caused, shall be guilty of an offence and shall be liable on conviction to a fine of $5,000 and to imprisonment for 12 months.” [120] This was yet another case of protestors hindering the work of the Central Electricity Board by so-called passive resistance. The question was whether the Chief Constable should be ordered by mandamus to have the obstructors forcibly removed. [121] Percy involved a woman repeatedly entering a military airbase, with the intention of protesting about the use of the base. She never threatened violence or did damage to property. [122] At §73 Barnes J on appeal commented “In Hong Kong, a Chinese-dominated society, scattering paper money offerings is indeed a very insulting act….” 
Mr Justice Chan PJ : 1. I agree with the judgment of Sir Thomas Gault NPJ. For the reasons given by him, I too would dismiss the appeal. Mr Justice Ribeiro PJ : 2. I agree with the judgment of Sir Thomas Gault NPJ. Mr Justice Tang PJ : 3. I have had the advantage of reading in draft the judgment of Sir Thomas Gault NPJ. For the reasons given by him I agree that the appeals should be dismissed. In deference to counsel’s submissions, I will add a few words. 4. We are concerned with voice identification of an audio recording, by Mr Lau (PW1), who claimed to be familiar with the voice of the appellants. In connection with such evidence, the learned judge reminded himself that he shouldapply: “13. … Turnbull warning with relevant and necessary modifications.” It is clear from the reasons for verdict that throughout the learned judge was mindful of those warnings. 5. Mr Chan submitted that PW1’s voice identification evidence was the result of or influenced by the content of the recorded conversation. If so, I agree, that might undermine the reliability of the voice identification. However, that was not the evidence of PW1. Nor did the reasons for verdict give any hint that the learned judge took that view. 6. Quite the contrary, the learned judge said: “102. Mr. Lau heard the voices of the two Defendants for many times within a long period of time. Evidence of Mr. Lau on the identification of their voices is absolutely credible and reliable.” 7. Nor did it appear that this finding was influenced by the circumstantial evidence in the case[1]. It was after this clear statement (and, an earlier statement to similar effect in para 100), that the learned trial judge went on to say: “103. Besides, there was also other peripheral circumstantial evidence to support the evidence of voice identification of Mr. Lau.” 8. The learned judge then took note of the circumstantial evidence which strongly supported Mr Lau’s evidence that the appellants were the speakers. 9. I believe in deciding what weight to put on the voice recognition evidence of Mr Lau, the learned judge was entitled, indeed, bound to have regard to all relevant and probative evidence in the case. This is not a case where the judge was uncertain about the reliability of the voice identification evidence. There was no undermining of the importance of maintaining the integrity of the modified Turnbullwarnings. Mr Justice Mortimer NPJ : 10. Having had the advantage of reading Sir Thomas Gault NPJ’s judgment in draft, I agree for the reasons he gives, that the appeals must be dismissed. Sir Thomas Gault NPJ : 11. This is an appeal pursuant to leave granted by this Court on 12 November 2012. 12. The issues for consideration are the same in respect of both appellants, and involve evidence of identification said to reside, in part, in a sound tape recording. 13. The appellants were tried in the District Court before HH Judge D Yau on 21 April 2011. Both were convicted on two charges; the first of doing acts tending and intended to pervert the course of public justice, and the second of misconduct in public office. 14. The appellants were police officers. In a criminal trial in 2007 in the District Court (“the earlier proceeding”), the 1st appellant was a prosecution witness having been deployed as an undercover officer investigating allegations of conspiracy to manage a vice establishment and money laundering. The 2nd appellant was an exhibits officer in the same trial. It is claimed that while the 1st appellant was giving evidence, but during the lunch break, the appellants had a discussion in a witness room that was recorded and disclosed the offending for which they were subsequently charged and convicted. 15. When the appellants were tried in the District Court, the judge’s reasons for verdict stated at the outset: “3. The Prosecution and Defence did not dispute on the admissibility of the relevant voice recording pen and the contents therein. In paragraph 4 of the Agreed Facts, it was agreed between all parties that the relevant voice recording pens had been produced by counsel for the 4th and 6th Defendants as exhibits in DCCC 1280/2005. It was also agreed between all parties that those voice recording pens were not handled improperly after they had been produced to the court.” 16. In this Court, Mr Charles J Chan represented both appellants. Mr Gerard McCoy SC and with him, Mr Jonathan Man Tak Ho (“Mr Man”), represented the respondent. 17. Mr Chan contended that the admissions of fact did not preclude an argument that the tape recording is inadmissible as a matter of law. Further, he said the admission, in any event, went no further than accepting the provenance of the taped evidence after the tape was produced to the court in the earlier proceeding. However, the tape, having been received in evidence, the case proceeded essentially as requiring determination of whether the evidence before the court proved that the participants in the recorded discussion were identified as the appellants and that they engaged in the charged criminal conduct. The evidence 18. There is no direct evidence in the present case establishing who made the recording, when it was made or the circumstances in which it was made and kept prior to its production to the court in the earlier proceeding. The prosecution contends that sufficient facts can be inferred from the content of the recordings when taken with other available evidence. 19. The tape records a conversation between two persons from which, the prosecution says, relevant evidence can be inferred. There was no challenge to the finding that whoever it was talking on the recording, there was a discussion concerning the giving of false evidence. 20. In his reasons for verdict, after referring to the admissions and recording the charges, the District Court Judge said that the only real issue in the case was the voice identification evidence of the prosecution witnesses in relation to the interlocutors in the audio recordings. At the outset of his consideration of the evidence, the judge said: “13. When considering the evidence on the voice identification, the court applied Turnbull[2] warning with relevant and necessary modifications, and reminded itself that even an honest witness might make a mistake in voice identification.” 21. Referring to the prosecution evidence, the judge described the witness PW1 (“Mr Lau”) as the most important prosecution witness and the only person who could positively identify the voices of the two appellants in the recorded conversations. He said that Mr Lau is a prosecuting counsel and had been counsel representing two of the defendants in DCCC 1280/2005. It was in the course of the trial in that case that the present appellants are said to have offended. 22. Mr Lau’s evidence in the present case was that towards the end of the prosecution case in DCCC 1280/2005, counsel for other parties produced to the court 45 voice recording pens. One contained the recording with which we are concerned. Mr Lau testified that he could identify the voices on the tape as those of the present appellants. The reliability of that evidence is in contention. 23. Mr Lau’s evidence was that, in the course of the earlier proceeding, in the morning of 1 June 2007, he had complained to the judge in the court that one of the police officers in the prosecution team whom he subsequently identified during an identity parade as the 2nd appellant, had peeped at a document that Mr Lau was reading. He knew that police officer from a previous case in the Kowloon City Magistracy. Mr Lau said that when he listened to the tape, he recognized the voice of the same police officer he knew as “Billy”. 24. The District Court Judge in his reasons for verdict said: “28. The court must handle this part of the evidence of Mr. Lau very carefully. The only useful part of such evidence and the only purpose that it can be admitted as evidence is the identification of voices of the people who spoke in (the audio recordings contained in) the voice recording pen.” 25. The judge summarized the voice identification evidence as follows: “31. First of all, Mr. Lau testified that the 1st Defendant had testified in court in the capacity as prosecution witness in a trial case before Magistrate Mr. Peter Law at Kowloon City Magistracy. Such a case at Kowloon City Magistracy was one of the cases of keeping vice establishments deriving from DCCC 1280/2005. In such a case at Kowloon City Magistracy, the 1st Defendant as an undercover police officer pretended to be a customer and visited the vice establishment concerned. 32. Moreover, in the case (DCCC) 1280/2005, the 1st Defendant testified at the witness box for a long period of time. Mr. Lau was responsible for cross-examining the 1st Defendant so he had sufficient time and chances to familiarize himself with the voice of the 1st Defendant. Although the 1st Defendant had not yet started testifying in court during the first suspended trial, the 1st Defendant had assisted in handling documents. According to Mr. Lau, the 1st Defendant was relatively familiar with documents, and sometimes after Mr. Lau had asked the 2nd Defendant i.e. the exhibits officer something about documents, the 2nd Defendant would seek assistance from the 1st Defendant. 33. Mr. Lau testified that even before commencement of the re-trial of (DCCC) 1280/2005, since Mr. Lau had had contacts with the 1st Defendant in the trial case at Kowloon City Magistracy, he was already able to identify the voice of the 1st Defendant. The trial case at Kowloon City Magistracy lasted for about 12 to 15 days. The 1st Defendant testified for about 1 to 2 days. 34. All in all, in the presence of Mr. Lau, the 1st Defendant testified for about 30 days or 120 hours. The last time Mr. Lau heard the voice of the 1st Defendant was when the 1st Defendant testified in the voir dire of DCCC 1280/2005 i.e. in about late 2007 or early 2008. Regarding the 2nd Defendant 35. Regarding voice identification of the 2nd Defendant, first of all, Mr. Lau testified that the 2nd Defendant never acted as a witness in DCCC 1280/2005 or in other cases handled by Mr. Lau. 36. Mr. Lau said that he talked with the 2nd Defendant for many times because the 2nd Defendant had also assisted in the trial of the case at Kowloon City Magistracy. Besides, the 2nd Defendant also assisted in the process of the trial on many occasions during the suspended trial of DCCC 1280/2005 and the subsequent 130-day retrial. 37. Mr. Lau said that the 2nd Defendant was responsible for passing documents to the defence. Through these contacts and occasional chit-chats with the 2nd Defendant, Mr. Lau became familiar with the voice of the 2nd Defendant. Mr. Lau said that the 2nd Defendant was rather talkative so during court adjournments Mr. Lau talked with the 2nd Defendant for many times. Although contents of those conversations were not something important and also not personal, such kind of chit-chats occurred for about 50 to 60 times, each lasting for about 1 to 2 minutes.” 26. In cross-examination, Mr Lau said that when asked by the police about the matter in April 2010, some 3 years and 2 months later, his initial recall was vague though he remembered having complained to the judge about the person he remembered having peeped at documents he was reading. His evidence was that, after having located and reviewed notes of the earlier proceeding he had retained, he recalled that the person who peeped at his papers was “Billy”. 27. Mr Lau did concede that he had not seen the two appellants alone together in the witness room during court adjournments. 28. The judge heard some expert testimony to the effect that there was no indication that the tape recording had been edited or interfered with and that there was no technical indication of when it had been recorded. 29. PW4 (“Mr Ng”) was the prosecuting counsel in the initial suspended trial and re-trial in DCCC 1280/2005. He confirmed that the 1st appellant was the second prosecution witness. He said that one of the police officers who assisted him was called “Billy”. PW4 had also prosecuted in the trial at Kowloon City Magistracy where the same undercover police witnesses gave evidence. 30. In the reasons for verdict, the District Court Judge noted PW4’s evidence was that he had listened to the relevant audio recordings and was unable to identify the voices recorded. The judge then stated: “57. … After hearing the evidence of Mr. Ng on this aspect, I deliberately remind myself that I must exercise extra care when dealing with the evidence of PW1 Mr. Lau in relation to voice identification. The reason is that Mr. Ng was the prosecutor in that case. He should have more chances and in closer proximity than any other people to have contacts with the Defendants. Theoretically he should be in a better position to identify the voices of the Defendants.” 31. The judge went on to note that PW4 could not recall who was the subject of the complaint on 1 June. 32. In his summary of the evidence of PW4, the judge set out the following findings: “59. …. Mr. Ng testified that as exhibits officer, the 2nd Defendant had been assisting Mr. Ng to pass exhibits to the judge for most of the time. Mr. Ng said that he was sure he had more opportunity than PW1 Mr. Lau to talk to the 2nd Defendant. 60. Another important point in the evidence of Mr. Ng is that when the Defence produced a series of voice recording pens to the court in that case, Mr. Ng and PW1 Mr. Lau were assigned to take part in the preparation of the transcript. According to the evidence of Mr. Ng, during the course of preparing the transcript, Mr. Lau made mistakes for many times and mixed up which voice in the audio recordings belonged to whom. Mr. Ng further said that Mr. Lau had to guess the identity of the persons in the recordings and that Mr. Lau himself had made mistakes for many times and even had to be corrected by his own defence team. 61. However, the Defence did not cross-examine Mr. Ng in relation to the voice identification of the relevant sessions of audio recording in the present case to clarify whether Mr. Lau had made any mistakes concerning such sessions. On the contrary, Mr. Lau gave specific evidence in relation to the relevant session(s) of audio recording in the present case and confirmed accuracy of his evidence on voice identification. 62. Upon invitation by the court and being agreed by the Prosecution and Defence, Mr. Ng listened to the relevant parts of the audio recordings in court again. What Mr. Ng could do was only to speculate that the voice which the Prosecution alleged to belong to the 2nd Defendant could be the voice of the 2nd Defendant. Mr. Ng was completely unable to give evidence in relation to whether the voice of the 1st Defendant was that of the 1st Defendant.” 33. The 5th prosecution witness was the officer-in-charge of case DCCC 1280/2005 over the relevant period. She gave evidence directed to aspects of the proceedings and the roles of participants. In relation to our present task, her evidence does not take matters much further. One point from her evidence noted by the judge in his reasons is: “72. Another important part of the evidence of WDIP Lee is that according to Madam Lee, in the case DCCC 1280/2005 exhibits were all kept in a large box and such large box was in the custody of her subordinates. During lunch breaks or court adjournments, they would normally take the box with them when they left the courtroom. Generally, they would put it in a witness room near the courtroom. Madam Lee said in general the 2nd Defendant was responsible for watching over the said exhibits box.” 34. PW5 was unable to identify the voices on the tapes. 35. Four further police officers gave evidence of having listened to the relevant tape in September 2009 and had been unable to identify the voices recorded. The judge did not find their evidence added significantly. The District Court Judge’s Findings 36. Having set out the evidence, including that of a character witness for the defence (the appellants elected not to give evidence), the judge found all of the witnesses honest and reliable. He found the contents of the transcript of the audio recordings prepared by PW3 are accurate. His findings on the voice identification evidence of PW1 were: “100. I find that evidence of PW1 Mr. Lau on voice identification of the audio recordings is true and accurate. I give full weight to his evidence. I find that regarding the 1st Defendant, Mr. Lau had ample opportunities to fully familiarize himself with the voice of the 1st Defendant at the trial at Kowloon City Magistracy and at the trial of DCCC 1280/2005. 101. Regarding the 2nd Defendant, although Mr. Lau heard the voice of the 2nd Defendant on relatively fewer occasions, he still had sufficient opportunities to hear the voice of the 2nd Defendant in the process of handling documents throughout the whole trial of DCCC 1280/2005. As the officer in charge of exhibits in the case, the 2nd Defendant definitely had many opportunities to talk with Mr. Lau. Furthermore, although the trial period was relatively shorter, the 2nd Defendant was also the exhibits officer in the trial at Kowloon City Magistracy and Mr. Lau was also the defence counsel in that case. This also increased the opportunity of him hearing the voice of the 2nd Defendant. 102. Mr. Lau heard the voices of the two Defendants for many times within a long period of time. Evidence of Mr. Lau on the identification of their voices is absolutely credible and reliable.” 37. The judge then added his findings on the peripheral circumstantial evidence. He accepted that references to names and nicknames on the tape were to police officers and police activities connected with case DCCC 1280/2005 and another case with which the appellants and those referred to on the tape were closely familiar. The District Court Judge found also that the taped conversation revealed that the two males were recorded as talking about the incident in the courtroom on 1 June 2007 when complaint was made about “Billy” peeping at PW1’s documents. 38. The judge also mentioned reference on the tape to another previous case: “110. In counters 205 and 211 of the audio recordings, Male 2 asked Male 1 if he could remember another previous case involving “city”. According to the evidence in our present case, both the 1st and 2nd Defendants had been involved in the trial of a case concerning vice establishments which was heard at Kowloon City Magistracy, and the trial of such case concerning vice establishments also involved the undercover operation that was involved in the trial of DCCC 1280/2005. In both cases, the 1st Defendant was one of the members in the undercover operation, while the 2nd Defendant was an exhibits officer. Notebook of the 1st Defendant confirmed that he had testified in a case called city sauna (literal translation).” 39. With reference to the time when and the place where the recorded conversation occurred, the District Court Judge’s reasons for verdict state: “116. Evidence which is able to assist us in trying to find out on what date and at what time the audio recordings were made are from the contents of the conversations. In the recorded conversations, Male 1 and Male 2 mentioned a floppy disc. Transcript of the trial of DCCC 1280/2005 shows that the 1st Defendant testified before lunch on 1st June that he would bring along with him a floppy disc when he performed the undercover operation. Obviously the two males were discussing the evidence which had been given by one of them. I find that the only irresistible inference is that the two males were the 1st and 2nd Defendants, and that they were discussing the evidence which had been given by the 1st Defendant at the trial of DCCC 1280/2005 at an earlier time on 1 July 2007. Place of making audio recordings 117. Without direct evidence of the person who placed the voice recording pen, the Prosecution could only rely on circumstantial evidence to try to convince the court to make an inference of the place of occurrence of the conversations recorded in the audio recordings. The evidence referred to by the Prosecution included the following. 118. The trial of DCCC 1280/2005 was heard by Mr. Jenkins. Mr. Jenkins’ court was at court no. 24 on 8/F of Wanchai Law Courts, Wanchai Tower at that time. According to the evidence of PW2 the expert witness, such conversation took place in a room. It was heard from the audio recording that the 1st Defendant told someone else over the phone that he needed what kind of food. The 2nd Defendant brought back some food and lemon tea later. They talked with each other while eating the food. According to the evidence of PW5 WDIP Lee, the 2nd Defendant was the officer responsible for watching over the exhibits box in the case during court adjournments. She also testified that in general the exhibits box would be put in the witness room during court adjournments. In counter 232 of the transcript of the audio recordings, someone mentioned quarter past. This seems to indicate that the then time was 2:15pm which was also the finishing time for the lunch, the time for them to prepare to return to the courtroom and continue with the trial. Since they had only a very short time left at that time, they were definitely at a place very near the courtroom. 119. I find that based on the above evidence, I am able to make an irresistible inference that the conversation in the audio recordings took place in the witness room on 8/F of Wanchai Law Courts, Wanchai Tower during lunch time on 1 June 2007.” 40. The judge took into consideration that others were not able to identify the voices on the tape and set out why he nevertheless accepted the evidence of PW1. When doing so, he mentioned again the “Turnbull” warning. He stated his conclusion as: “123. I find that the 1st and 2nd Defendants had conversations in the witness room at 8/F of Wanchai Law Courts, Wanchai Tower, while the 1st Defendant was still giving evidence for the Prosecution in court in the case DCCC 1280/2005 on 1 June 2007. I find that the contents of their conversations are as stated in Exhibit P23 the transcript. I find that their conversations relate to the evidence which had been given by the 1st Defendant at the trial of DCCC 1280/2005. I find that they also discussed how the 1st Defendant should further testify when the trial was resumed, that the 1st Defendant should not mix up the true and false parts of his evidence, and that the 1st Defendant should memorise the false parts so as to avoid confusion in future.” The Case in the Court of Appeal 41. In their appeal to the Court of Appeal, the appellants advanced four grounds all of which were rejected. Of those the first; that the contents of the audio recording did not disclose the charged offending, and the fourth, that the District Court Judge had a conflict of interest, were rejected. They were not pursued in this Court and need not be mentioned further. 42. The remaining grounds were dealt with together in the Court of Appeal judgment. The court considered four matters; they were: the making and custody of the tape recording, the possibility that it had been tampered with, whether the voice identification evidence of Mr Lau should have been admitted by the judge, and the weight that should be accorded in Mr Lau’s evidence. 43. The judgment of the Court of Appeal cited the leading authorities on evidence by audio recordings. By reference to HKSAR v Lee Chi Fai, the court accepted that the content of the tape was obviously relevant to the charges. The court then said: “27. … Secondly, its prima facie authenticity must be considered. In general, the prosecution would have evidence concerning the provenance and history of the audio recording. However, this is not ‘indispensible’. As Roberts CJ said in Choi Kit Kau v The Queen, this is the ‘best’ method of proving. However, the authority case shows that this is not the only way to prove.” 44. The court accepted that the authorities establish that authenticity can be proved circumstantially. The judgment reviewed the expert evidence and accepted the finding of the District Court Judge that the expert evidence established the tape as prima facie authentic. The court then addressed the identification evidence of Mr Lau and concluded that, although he was not an expert, the judge was right to accept his evidence because of the opportunities he had to become familiar with the voices of the appellants he said were recorded on the tape. The court said: “37. … The Applicants emphasized that there were other witnesses in the case who were unable to identify the voices depicted in (Exhibit) P37. However, the judge had the advantage of seeing and hearing testimony of every witness. He could observe the response, wisdom, alertness and manner etc. characteristics of every witness, and then he had the right to accept the evidence of one of the witnesses.” The Case in this Court 45. In this Court, Mr Chan, for the appellants, presented extensive written argument on the difficulties inherent in dealing with evidence in the form of tape recordings. He argued that only with clear evidence of the making and preservation of any recording and expert evidence of its accuracy and reliability with available voice recorded material for comparison, should such evidence be admitted. He contended that warnings adapted from the guidance in Turnbull are not sufficient to avoid the risks of relying on evidence of tape recordings. 46. When he dealt more specifically with the circumstances of the present case, Mr Chan was critical of the inadequacies in the identification evidence. In that regard, he went considerably beyond the evidence and cross-examination of PW1 before the trial judge. That cross-examination focused almost entirely on opportunities PW1 had to become familiar with the voices of the appellants. Counsel’s criticisms of the inadequacy of the evidence explaining how PW1 was able to recognize the voices when others could not, seems to have rested on the one answer that Mr Lau acknowledged having never received any special training on voice identification. 47. The real issue, however, is whether it was open to the judge to conclude that the evidence that was given proved that the voices recorded on the tape were those of the two appellants. Decision 48. The issues surrounding evidence in the form of recordings are broadly similar though the means of recording may be different. Audio tapes, video tapes, films, photographs, even DNA analysis, whatever the technology used, raise issues of authenticity and accuracy that must be addressed with reference to the particular circumstances. When a recording is relied on for identification, not only the accuracy of the record, but also the reliability of the witness identifying what is recorded may arise. These matters are facts which must be proved. 49. Visual identifications are well recognized as requiring care and the courts recognize this as the decision in Turnbull clearly shows. The same care and equivalent warnings are applicable to other forms of identification evidence, including voice identifications. 50. In HKSAR v Lai Wai Cheong[3], the Court of Appeal accepted that the Turnbull guidelines are applicable to evidence of identification, whether of appearance or voice. A tape recording can be admitted and relied upon as evidence of the contents where a proper basis for its reliability is established. In the early case of R v Maqsud Ali[4], Marshall J, for the Court of Criminal Appeal, said: “… We can see no difference in principle between a tape recording and a photograph. In saying this we must not be taken as saying that such recordings are admissible whatever the circumstances, but it does appear to this court wrong to deny to the law of evidence advantages to be gained by new techniques and new devices, provided the accuracy of the recording can be proved and the voices recorded properly identified; provided also that the evidence is relevant and otherwise admissible, we are satisfied that a tape recording is admissible in evidence. Such evidence should always be regarded with some caution and assessed in the light of all the circumstances of each case. There can be no question of laying down any exhaustive set of rules by which the admissibility of such evidence should be judged.” 51. In Choi Kit Kau v The Queen[5], Roberts CJ said: “[T]he best method of proving that a tape recording produced in court is authentic is to show that it has been, since the time it was recorded, continuously in the custody of persons who assert that it was not tampered with. This is a course which should be followed if the authenticity of a tape is challenged and is desirable even if it is not.” 52. Where the “best” method is unavailable, the authenticity of a tape may still be capable of proof. In HKSAR v Lee Chi Fai and Others[6], the Court of Appeal adopted the analysis in R v Murphy and Another[7]: “ Authenticity, in our view, like most facts may be proved circumstantially. In the case of a video film, the direct way is to call the cameraman who took it and the court will normally expect him to be called. But if he is not available, he need not be called; other evidence will suffice if it is logically probative that the video was authentic. That evidence may be adduced in other ways and from other sources.” 53. In R v Chen[8], the Victoria Court of Criminal Appeal received evidence of tape recordings where the makers were dead or unavailable. The court said: “The test is whether there is sufficient material before the court to allow the tribunal of fact acting reasonably to conclude that the recorded sounds reproduce those originally made by the persons identified by the evidence. In other words, there must be evidence, which the tribunal of fact is entitled to accept, that the recording is of a conversation which occurred and which would be admissible if proved by oral testimony. In our opinion, admissibility does not depend on the party tendering the tapes having removed absolutely any chance that they are inaccurate.” 54. There is much of assistance in the lengthy judgment of the Court of Appeal (Criminal Division) of Northern Ireland in R v Murphy and Another. The case involved video-taped evidence forming part of the prosecution case though the makers of the films did not give evidence. 55. The Court of Appeal in England in R v Flynn and St John[9] ruled that evidence of recognition of a voice by a lay person was admissible, but on the facts of that case evidence of police officers of voice recognition should have been ruled inadmissible. The opportunities for familiarization with the voices of the accused were very much less than in this case and there was expert evidence casting doubt on the reliability of the tape recordings. The court’s conclusions were summarized at p 279 as: “ There are other considerations which apply to both appellants. We have already noted the two most important factors, namely the limited opportunity for the officers to acquire familiarity with the appellants’ voices and secondly, the poor quality of the covert recording. To these we add the following. First, the police officers’ purported recognition of the appellants’ voices is in marked contrast to the evidence of the two experts who are unable to recognise their voices; in Dr Holmes’ case she was unable to identify individual voices. Secondly, nothing is known of the ability of any of the police officers to recognise voices. There is no evidence that any of them had any training in auditory analysis. Thirdly, the identification of the voices was carried out by listening to the covert tape on a standard laptop computer, as opposed to the sophisticated equipment used by the experts. Fourthly, in our judgment, it is significant that DC Gittings’ transcript contains words attributed to Flynn which Dr Holmes could not distinguish as words, let alone recognise as attributable to Flynn. Fifthly, the expert evidence shows that lay listeners with considerable familiarity of a voice and listening to a clear recording, can still make mistakes.” 56. To the judgment in that case the court added this postscript: “ As appears from the above we have been dealing in these appeals with issues arising out of voice recognition evidence. Nothing in this judgment should be taken as casting doubt on the admissibility of evidence given by properly qualified experts in this field. On the material before use we think it neither possible nor desirable to go as far as the Northern Ireland Court of Criminal Appeal in O’Doherty which ruled that auditory analysis evidence given by experts in this field was inadmissible unless supported by expert evidence of acoustic analysis. So far as lay listener evidence is concerned, in our opinion, the key to admissibility is the degree of familiarity of the witness with the suspect’s voice. Even then the dangers of a mis-identification remain; the more so where the recording of the voice to be identified is poor.” 57. In Jones & Harris[10], the Court of Criminal Appeal of Victoria rejected submissions for two appellants that evidence of voice identifications should not be received. In the case of the appellant Jones tape-recorded telephone conversations together with circumstantial evidence consistent with factual statements made during the telephone calls was received as probative of identity. Similarly, in the case of Harris, voice identification evidence from witnesses familiar with his voice and supported by circumstantial evidence was held to have been rightly admitted. 58. I see no reason why, in certain circumstances, the identity of participants and, the timing and circumstances of a recorded conversation, cannot be proved by reference to the content of the recording. It must be possible as where the maker is unknown or unavailable and the content when placed in a context of known facts establishes time and place. Similarly, the identity of a speaker may emerge from what is said. Statements revealing the facts or knowledge that only an offender could know frequently can lead to identification. For example, statements made by disguised offenders in the course of a robbery can be relied upon to identify them. Statements made anonymously in writing or in the course of a telephone call demanding ransom after a kipnapping may support identification. 59. Of course other possible contingencies must be excluded – such as voice imitation. Those matters go to the weight to be accorded the evidence of the conversation. 60. The authorities make clear the care with which tape recordings must be considered. But with appropriate safeguards, and careful directions, there is no reason to construct rigid exclusionary rules. As with all evidence, reliability is to be assessed in all the circumstances. 61. In many cases, it would be question-begging to seek to prove the authenticity of a tape recording from the contents of the recording itself. Mr Chan, for the appellants, submitted that the voice identification in this case is flawed because PW1 was not shown to have recognized the speakers solely by their voices and without reliance on the context with which the speakers exhibited familiarity. He argued that: “In substance, the whole exercise would not stand to test the ability of PW1’s ability in the voice identification. It is apparent that both PW1 and essentially the learned trial Judge had been affected by this potential contextual bias, using the content to reinforce their ‘belief’ that the voices belonged to the Appellants. … The Appellants respectfully submit that to assess the evidence of voice identification of PW1, it has to be done in separation of the content of Exh. P37. Otherwise, one would fall into the same flaw or weakness in suggesting that PW1 was capable of identifying something apparently representing the details of a content that he was familiar with and borrow force from this to his ability of voice recognition because it is a ‘content recognition’ which might subconsciously affect the listener and also it is a logical fallacy amplified infinitely because one cannot examine how much the recognition is attributable to voice and content respectively.” 62. I do not accept that the “context”, indeed all the surrounding circumstances, may not be taken into consideration in evaluating evidence of identification. The argument to the contrary is unreal and inconsistent with the authorities already mentioned. 63. Of greater significance is whether PW1’s evidence of his recognition of the identity of the voices on the tape may have been not solely his own, but was contributed to by others who did not give evidence of recognition or identification of the voices. PW1’s evidence was that when the existence of the tapes was exposed at the end of 2007, he was one of the people involved in preparing transcripts. There were 45 tapes produced at the trial of DCCC 1280/2005 though only one contained the conversation with which we are directly concerned. 64. At the trial of this case, when PW1 gave evidence that he could identify the voices on the tape, he was asked when he first disclosed this. His evidence was: “A: Actually, I disclosed (this) at the trial of DCCC 1280 because actually in the process of preparing these transcripts, (the identities of) Male 1 and Male 2 had been mutually agreed by the prosecution and defence in that trial. Q: The prosecution and defence did not agree on who was Male 1 and who was Male 2; otherwise it would have been written at an early stage --- it would have been specified at the beginning that Male 1 was whose voice and Male 2 was whose voice. Do you agree? This was not agreed by both sides. Is it? A: I dare not say (whether it is the case or not) for individual transcript. For example, if you are talking about whether the transcript for 1 June had been mutually agreed at that time, we --- I have to go back and ‘check’ my notes, but regarding the transcripts being produced to the court at that time, most of the Male 1, Male 2 or Female 1, or Male 1, Male 2, Male 3, all were mutually agreed by the prosecution and defence before the transcripts were produced to the court; otherwise the cross-examination could not be continued. Q: I put it to you that, by saying that the prosecution and defence mutually agreed, you mean agreeing the contents of the transcripts but not the identities of the voices appeared in the transcripts. This has never been agreed. Do you agree? Since … A: I have to think about this first. I do not have much recollection about this.” 65. Later in his evidence, PW1 explained why in October 2010 he was prompted to consider the details of the “peeking” incident of 1 June 2007. He said that in connection with other litigation it had become necessary for him to review the relevant documents and his contemporaneous notes. There was also a professionally prepared transcript then available. He said this brought back memory of the incident. It was not put to him that this recollection and identification in 2010 and subsequently were aided by having been involved, with others, in 2007 in preparation of the first transcripts (which in any event did not include the identities of the recorded speakers). Of course this would not exclude the possibility of there having been discussion or speculation at the time, but there is no evidence of that. At the trial of the present proceeding, the tape was played and PW1 identified the voices as those of the appellants. Doubtless by then he would have listened to the tapes a number of times. Standing alone his evidence that he could identify the voices after the lengthy lapse of time would warrant concern. But the circumstantial evidence gives real credibility to his recollections, as they clearly did for the District Court Judge and the Court of Appeal. And the peeking incident gave him a particular reason to remember. 66. I have not been persuaded that the reasons for verdict contain material errors of law. Nor has the judgment of the Court of Appeal been shown to address incorrectly the arguments presented to that court. The case has involved an assessment of the reliability of oral evidence of identification bolstered by circumstantial evidence directed to the authenticity of the tape recording. The courts below were convinced by the evidence tying the recorded conversations to the appellants and to the occasion on which they occurred. Both lower courts reviewed the absence of evidence from others whose exposure to the voices of the appellants was no less than PW1 but nevertheless accepted his evidence. 67. The dangers of misidentification of voices especially by lay listeners were acknowledged. PW1’s familiarity with the voices of the appellants was fully investigated and his evidence was firm. 68. Accordingly, I am satisfied that the verdicts were open on the evidence and there has been demonstrated no “substantial and grave injustice” (So Yiu Fung v HKSAR[11]) as would be required under s 32(2) of the Court of Final Appeal Ordinance to warrant disturbing the convictions. 69. I would dismiss the appeals. Mr Justice Chan PJ : 70. The Court unanimously dismisses the appeal. Mr Charles J Chan, instructed by Munros, for the appellants Mr Gerard McCoy SC, instructed by the Department of Justice, and Mr Jonathan Man, of that Department, for the respondent [1] In the present case, the prosecution also relied on circumstantial evidence to establish the identity of the speakers. In the end, the appellants were convicted on the totality of the evidence. In R v Jones and Harris (1989) 41 A Crim 1, a decision of the Court of Criminal Appeal in Victoria, the prosecution also relied on both circumstantial and voice recognition to identify the speaker. [2] R v Turnbull and Another [1977] QB 224, CA. [3] [1998] 1 HKLRD 665, 668. [4] [1966] 1 QB 688. [5] [1980] HKLR 433, 439. [6] [2003] 3 HKLRD 751, 762. [7] [1990] NI 306, 344. [8] (1993) 66 A Crim R 154. [9] [2008] EWCA Crim 970, [2008] 2 Crim Appeal R 20. [10] (1989) 41 A Crim R 1. [11] (1999) 2 HKCFAR 539.