FACC7/2012 KULEMESIN YURIY AND ANOTHER v. HKSAR

1. For the reasons which are contained in the judgments of Mr Justice Ribeiro PJ and Lord Clarke of Stone-cum-Ebony NPJ, I agree that the appeal of the 1st appellant be dismissed and that the appeal of the 2nd appellant be allowed. I would also like to express the Court’s gratitude to all counsel for their submissions.

2. I agree with the judgments of Mr Justice Ribeiro PJ and Lord Clarke of Stone-cum-Ebony NPJ.

3. On 22 March 2008, the “YAO HAI” (“YH”), a Chinese flagged bulk carrier, collided with the “NEFTEGAZ 67” (“N67”), a Ukrainian flagged oil rig supply vessel, in a buoyed channel running between the north coast of Lantau Island and the coast near Castle Peak in the New Territories. N67 sank with the tragic loss of 18 of her 25 crew members. YH suffered damage but none of her crew were lost or injured.

4. The collision led to the prosecution of N67’s master, Mr Yuriy Kulemesin (“the 1st appellant”); YH’s master, Mr Liu Bo; the senior pilot who was navigating YH at the time, Mr Tang Dock Wah (“the 2nd appellant”); and the junior pilot assisting him, Mr Bruce Chun Wah Tak. They were named as the 1st to 4th defendants respectively and each was charged with endangering the safety of others in contravention of section 72 of the Shipping and Port Control Ordinance (“SPCO”).

5. The trial took place over some 91 hearing days in the District Court before HH Judge S D’Almada Remedios and resulted in the conviction of all four defendants. However, the Court of Appeal quashed the convictions of YH’s master and the junior pilot. They reduced the 1st appellant’s sentence of imprisonment for 3 years and 2 months to a sentence of imprisonment for 18 months. They also reduced the 2nd appellant’s 3 year sentence so as to permit his immediate release. The present appeal is brought by the 1st and 2nd appellants against their respective convictions.

6. This appeal concerns eight questions of law certified as being questions of great and general importance. The present judgment addresses the first of those questions, certified by the Court of Appeal in the following terms:

7. The other seven questions raise issues which bear on the appellants’ challenges regarding the actus reus elements of the section 72 offence, namely as to whether it ought properly to have been found that by their conduct they caused the endangerment of the safety of others. Those questions are dealt with in the judgment of Lord Clarke of Stone-cum-Ebony NPJ, which I have had the benefit of reading in draft and with which I respectfully agree. I gratefully adopt the account and analysis of the facts it contains.

8. This judgment also deals with the 1st appellant’s application for leave to raise two matters on the “substantial and grave injustice” ground under section 32 of the Court’s statute.[1]

10. The approach of the courts below has been to treat section 72 as creating an offence having two forms: (i) one under the “unlawful act limb” covering cases where the accused performs an unlawful act and thereby endangers the safety of others; and (ii) the other under the second limb where the endangerment is caused by the defendant “in any manner whatsoever without reasonable excuse”. That approach was challenged by Mr Gerard McCoy SC[2] who sought to argue that the Court should relocate the comma after the words “unlawful act” in its construction of the section. For reasons given later in this judgment,[3] I reject that argument.

11. After amending the original charges, the case brought against the 1st and 2nd appellants proceeded under both limbs in the alternative. The Judge eventually convicted the 1st appellant under both limbs, but found the 2nd appellant guilty only under the unlawful act limb.[4]

12. Under the unlawful act limb, the prosecution alleges that the 1st and 2nd appellants were guilty of unlawful acts which consisted of offences under section 10(3) of SPCO which provides as follows:

13. The unlawful act relied on for section 72 purposes was therefore breach of section 10(3) which in turn requires proof of contravention by a vessel of the collision regulations. Those regulations are the International Regulations for Preventing Collisions at Sea 1972 (often referred to as “COLREGS”), the current edition of such regulations issued pursuant to a series of international conferences convened by the International Maritime Organization. The collision regulations are well known to mariners all over the world and by their own terms, apply to all vessels upon the high seas and in all navigable waters connected therewith. They establish internationally recognized rules for the navigation of vessels aimed at preventing maritime collisions.

14. By section 10(1)(b) of SPCO, those rules are made applicable to all vessels within the waters of Hong Kong. And, as we have just seen, if any such vessel contravenes one of the collision regulations, her owner, the master and any person for the time being responsible for her conduct are individually made guilty of an offence under Hong Kong law under section 10(3). That, however, is subject to the defence provided under section 10(4) which states:

15. The contraventions of COLREGS relied on by the prosecution as constituting unlawful acts under section 72 are breaches of rules 5 and 8 in both cases and additionally of rule 9 in the 1st appellant’s case. Those rules relevantly provide as follows:

16. Originally, the charges laid against the appellants proceeded solely under the second limb of section 72. It was in the context of the charges so framed that at an early stage of the trial, the defendants mounted an unsuccessful constitutional challenge to the validity of section 72.[5] The prosecution thereafter obtained leave to amend the charges.

17. In the amended charges[6] brought under the unlawful act limb, the 1st and 2nd appellants were each charged with endangering or causing to be endangered the safety of the crew of N67 in that being master of that vessel and pilot navigating YH respectively, they each were guilty of unlawful acts consisting of contraventions of COLREGS.[7] Those contraventions were particularised as involving breaches of rules 5 and 8 in respect of both of the appellants and additionally breach of rule 9 in the 1st appellant’s case.

18. The rule 5 particulars alleged that each of the appellants had failed to maintain a proper look-out by sight and hearing as well as by radar so as to make a full appraisal of the situation and of the risk of collision between the two vessels.

19. In relation to rule 8, the appellants were each alleged to have failed to take positive action in ample time and with due regard to the observance of good seamanship; failed to make an alteration of course and/or speed large enough to be readily apparent to the other vessel; failed to make a substantial alteration of course in good time to avoid a close-quarters situation; and failed to take action such as to result in passing the other vessel at a safe distance.[8]

20. The rule 8 particulars went on to allege that “if and insofar as” each vessel had a duty not to impede the other, the respective appellants had failed to take early action to allow sufficient sea-room for the safe passage of the other vessel.[9] And “if and insofar as” each vessel was not to be impeded by the other, each appellant had breached rules 5 and 8 for avoiding a risk of collision in approaching the other vessel.[10]

21. The particulars of breach of rule 9 stated that “when proceeding along the course of a narrow channel or fairway in the Western approaches as defined by buoys CP5, CP3 and CP1 on the northern side and by buoys CP4, CP2 and the intervening prohibited area on the southern side, alternatively the 15 metre depth line on each side”, the 1st appellant failed to keep as near to the outer limit of the channel or fairway on the starboard side of N67 as was safe and practicable.

22. The alternative charges under the second limb of section 72 alleged that each appellant had failed to maintain a proper look-out by sight and hearing as well as by radar so as to make a full appraisal of the situation and of the risk of collision with the other vessel; and failed to take adequate positive or preventive action so as to avoid a collision with the other vessel.

23. The Judge recorded that the charges were laid against each appellant under both limbs of section 72 in the alternative and noted that while the unlawful act limb rested on alleged breaches of COLREGS, the charge under the second limb was founded on the same impugned conduct, not described as unlawful acts, but raising similar issues.[11]

25. Section 10(4) is of course a defence to a section 10(3) charge and not to a section 72 charge. However, what the Judge must have meant was that where, as in the present case, a defendant is charged on the basis of having committed an unlawful act consisting of an offence under section 10(3), successful reliance on section 10(4) deprives the prosecution of the unlawful act it needs to secure a conviction under section 72.

26. The centrally important feature of the Judge’s approach to section 72 is that she regarded it as unnecessary for any mental state on the defendant’s part to be established in relation to the consequences of his conduct involving endangerment to the safety of others. In other words, she regarded the offence as one of absolute liability so far as the consequence of endangerment was concerned.

27. The Judge made this clear in her ruling rejecting the defendants’ constitutional challenge to section 72 mentioned above. Her Honour stated that in her interpretation of the section, “no specific mens rea” was required.[13]

28. This is reflected in the Judge’s formulation of the legal issues in the Reasons for Verdict. Under the unlawful act limb, the questions framed begin by asking whether the defendants breached the relevant COLREGS; then whether such breaches caused the endangerment of the N67’s crew’s safety; going on immediately to ask whether the defence of having taken all reasonable precautions to prevent such contravention succeeds or whether it has been negative by the prosecution.[14] There is no mention of any requirement for the accused to have intended or been reckless or to have had a mental state of any description regarding the dangerous consequences of his conduct.

29. Similarly, under the second limb, the issues framed were whether the appellants had failed to keep a proper lookout or to take adequate collision avoidance action; whether such conduct had caused endangerment to the safety of N67’s crew and if so, whether the prosecution had negatived any possible reasonable excuse.[15] Liability was therefore seen as established by proving conduct (which did not have to be unlawful) which actually caused endangerment of safety, coupled with the absence of reasonable excuse without any mental state having to be proved in relation to the consequence of endangerment.

30. In finding the 1st appellant guilty, having earlier rejected his proffered excuses, the Judge simply found beyond all reasonable doubt that he “had breached Rules 5, 8 and 9 and that he endangered the safety of his own crew on N67.”[16] Similarly, she found that the 2nd appellant “had breached rule 5 and 8 and endangered the safety of the crew of N67” adding that he “did not take reasonable precautions to prevent the contravention and did not have a reasonable excuse for his failure in doing so.”[17] She did not consider it necessary to make any findings in either case as to the appellant’s state of mind regarding endangerment of safety.

31. The Judge considered the strictness of the offence as interpreted by her to be mitigated by the existence of the statutory defences, stating:

32. Having considered two amendments to section 72 made in 1979[19] by which the offence was made indictable and made to carry much increased penalties, the Court of Appeal posed the question:

33. After discussion (to which I shall return), the Court of Appeal pronounced itself “satisfied that by the increased penalties available on conviction on indictment the Legislative Council did not intend to imbue section 72 with a general requirement of full mens rea” and that the Judge had correctly identified the ingredients of the offence.[21]

35. Like the Judge, the Court of Appeal took comfort from the existence of the two defences mentioned above:

36. The approach adopted in the courts below therefore meant that they regarded section 72 as an offence of absolute liability with regard to the element of endangerment. They held:

37. It is striking that both judgments below fail to mention the presumption of mens rea when construing section 72. The omission is particularly puzzling in the case of the Court of Appeal since we were told that it was addressed on that presumption in the light of this Court’s decision in Hin Lin Yee v HKSAR, [24] which discusses in detail how the presumption operates and how the mental requirements of statutory offences should be ascertained.

38. As pointed out in that judgment, where the offence-creating provisions are silent or ambiguous as to the mental requirements, the starting-point is that the statute must be construed adopting the presumption that it is incumbent on the prosecution to prove mens rea in relation to each element of the offence.[25] In other words, the court presumes that the prosecution must prove that the accused possessed the state of mind appropriate to each external element of the offence. Thus, if a specified circumstance is required to accompany the defendant’s act or omission as a prescribed ingredient of the offence, the presumption is that the prosecution must prove that the defendant knew of or was reckless as to the existence of that circumstance. And if it is a necessary element of the offence that the impugned conduct causes certain prohibited consequences, it is presumed that the prosecution must prove that when acting or omitting to act, the defendant intended to cause or was reckless as to his causing those consequences.[26]

39. In Hin Lin Yee, the offence charged was selling a drug intended for use by man but unfit for that purpose, so the presumption of mens rea was there directed at the circumstance that the drug sold was unfit for human use. In the present appeal, section 72 requires the unlawful act or other impugned conduct of each appellant to cause danger to the safety of others, and it is to that consequence of endangerment that the presumption of mens rea is directed.

40. If the Hin Lin Yee approach had been adopted, the courts below would have proceeded to consider whether the presumption had been displaced as a matter of statutory construction. They would have engaged in an examination of the language and purpose of section 72 and matters including the nature and seriousness of the offence; its attendant penalties and social or professional stigma; the utility of imposing sanctions; and the prevailing societal conditions, in so far as relevant.[27] They would also have considered what alternative legal regime governing the mental requirements might take the place of mens rea if the presumption were to be supplanted.[28]

41. In Hin Lin Yee, a survey of the case-law of England and Wales, Australia, New Zealand and Canada indicated that in construing statutory offences which are silent or ambiguous as to the mental requirements regarding the circumstances accompanying the defendant’s acts or omissions, five alternative regimes are possible:

42. As the alternatives so formulated refer to the circumstances surrounding the prohibited conduct, they will require a degree of modification to address additionally the mental requirements relating to the consequences of the prohibited conduct, as discussed below.[29]

43. In concluding that section 72 imposes no mens rea requirement regarding the consequence of endangerment, the courts below did not begin with the presumption of mens rea and then identify the basis for its displacement. Nor did they give any consideration to whether an alternative to absolute liability – a “halfway house” requirement – should replace a supplanted mens rea requirement. Since the Judge’s decision pre-dates Hin Lin Yee, it is perfectly understandable that it does not discuss any “halfway house”. But although Hin Lin Yee was available to the Court of Appeal it nevertheless appears to have assumed that rejection of a full mens rea requirement automatically results in absolute liability. On what basis then did the courts below arrive at their conclusion in favour of absolute liability?

44. The Judge appears mainly to have based her conclusion on the Court of Appeal’s decision in HKSAR v Chow Tin Shing,[30] which she thought[31] provided helpful general guidance to section 72. She cited in particular the following passages from the judgment of McMahon J:

45. When making her constitutionality ruling,[32] her Honour also cited paragraphs 35 and 36 in Chow Tin Shing,stating that the Court of Appeal there:

This led her to conclude, as we have seen, that “no specific mens rea” was required.

46. With respect, those passages from Chow Tin Shing were taken out of context and that decision provides no guidance as to the mental requirements of section 72.

47. The defendant in Chow Tin Shing was the master of a motorised oil barge who faced two charges under section 72 relating to two different incidents arising out of his navigation in a fairway leading to Lei Yu Mun. Paragraphs 13 and 14 relied on by the Judge concerned the allegation that defendant had breached rules 9 and 14 of COLREGS in that, when on a nearly reciprocal course with an oncoming tug, he had failed to keep to the starboard side of the fairway and had effected a starboard to starboard passing with the vessels about 30 metres apart.

48. What McMahon J was saying in the paragraphs cited was that it was insufficient merely to allege breaches of the two COLREGS rules and that it was necessary to go on to prove that the defendant’s conduct had actually caused endangerment to the safety of the tug’s crew.[33] He pointed out that passing on the wrong side relatively close to the tug “does not necessarily of itself result in” the necessary endangerment[34]and quashed the conviction on that charge, holding that the evidence did not prove such endangerment to the requisite standard.[35] The paragraphs cited therefore do not address the question of mens rea at all. They point simply to the deficiency in the prosecution’s case on the question of causation.

49. Neither do paragraphs 35 and 36 of Chow Tin Shing throw any light on the question under discussion. The Court of Appeal was there dealing solely with the appropriate sentence in relation to the second incident which involved the defendant’s barge running down a small stationary fishing boat resulting in the death of a person on board. The Court of Appeal was therefore not seeking to explore the ingredients of section 72 since liability was at that stage assumed. It was, for sentencing purposes, comparing the culpability of the defendant to that of human traffickers involved in high speed chases in Hong Kong waters involving police launches and pointing out that in those cases, “the acts of endangerment were deliberately reckless and of great risk to the safety of the passengers in the sampans and to a lesser extent to the crew of the pursuing police launches”. In contrast:

50. Thus, McMahon J assessed the defendant’s culpability with reference to what the defendant knew and the extent to which his conduct had been calculated or deliberate, his sentence being reduced[37]accordingly.

51. Before continuing with this examination of the basis for the decisions below, it is convenient at this point to return to Mr McCoy SC’s challenge to construing section 72 as having two limbs, an argument which I have rejected. That section relevantly provides:

52. It creates an offence with three actus reus elements: (i) prohibited conduct, (ii) causation and (iii) endangerment of safety of persons at sea. The conduct in question may either consist of an unlawful act or an act coming within the words “any manner whatsoever without reasonable excuse” in the second limb. It therefore has two alternative limbs so far as the prohibited conduct is concerned and, on its face, the defence of reasonable excuse only applies where the charge is laid under the second limb.

53. The object of Mr McCoy’s contention was to make that defence equally available where a defendant is charged under the unlawful act limb. To that end, he invited the Court to move the comma which separates the two limbs forward by four words so that section 72 would read: “Any person who(,) by any unlawful act or in any manner whatsoever without reasonable excuse” etc. In support, he points to the suggested punctuation having been adopted in section 72A.[38]

54. I can see no basis for re-arranging section 72’s punctuation. A remedial construction of a statute may be called for where, by some obvious clerical error, the provision is rendered incoherent or absurd, but no such complaint can be made in respect of section 72. On the contrary, it is perfectly plausible to conclude that the legislative intent was to condition liability on two alternative forms of conduct, one involving an unlawful act which, given the unlawfulness, did not admit of a reasonable excuse defence; and the other involving conduct which, while dangerous, was not itself unlawful and thus meriting a reasonable excuse defence. It is nothing to the point that the desired pattern of punctuation can be found in a different section which has different subject-matter and does not presently call for interpretation.[39]

55. It is against this background, namely, that section 72 requires proof of the three actus reus elements of prohibited conduct, causation and endangerment; and that it has two limbs involving alternative forms of prohibited conduct with reasonable excuse as a defence only in respect of the second limb; that the next ground relied on below[40] in support of absolute liability should be approached.

56. That ground concerns the earlier amendments to section 72. The predecessor to the present section first entered our statute book in 1949,[41] stating:

57. The present version of section 72, enacted in January 1979, makes it triable as an indictable offence with a maximum fine of $200,000 and imprisonment for up to 4 years, a ten-fold and eight-fold increase in the respective penalties.

58. While the Court of Appeal acknowledged these changes and evidently thought that they did not stand in the way of its conclusion in favour of absolute liability, it is unclear what effect they attributed to those changes.[42]

59. Certainly, the amendment which made the offence much more serious by making it indictable and punishable by significantly increased penalties militates against treating the offence as one involving absolute liability. As was pointed out in Hin Lin Yee,[43] while exceptions exist, the more serious the offence in terms of penalty and social obloquy, the less likely it is that the presumption of mens rea will be held to have been supplanted.

60. The second aspect of the amendment concerned replacing the limb involving the defendant’s endangerment of safety “by any wilful omission or neglect” with the present second limb which founds liability on endangerment “in any manner whatsoever without reasonable excuse”. That change was given weight by the Judge apparently as a matter in support of absolute liability. In her ruling,[44] she held that deletion of the requirement of wilfulness signifies the absence of any requirement to show an intention to cause danger. In holding that the Judge had correctly identified the ingredients of the offence,[45]the Court of Appeal may be taken to have agreed.

61. With respect, I do not agree that the changes provide any support for absolute liability in relation to the element of endangerment.

62. It is true that by virtue of the amendment, in cases where the prosecution relies on some omission or neglect as the prohibited conduct, it no longer has to show that such omission or neglect was “wilful” – ie, that the defendant knew he was under a duty to act but deliberately failed to do so. As the Court of Appeal pointed out, “conduct, by commission or omission, not amounting to an unlawful act would fall within that phrase, for example, the circumstances of mere negligence”.[46]

63. However, that change does not touch on the requirement of endangerment. As noted above, section 72 has three actus reus requirements: prohibited conduct, causation and endangerment. The amendment affected only the first of these and the statute remains silent as to the state of mind required regarding the consequence of endangerment caused by the prohibited conduct. As noted above,[47] an unrebutted presumption of mens rea makes it incumbent on the prosecution to prove mens rea in relation to each element of the offence. It follows that while the 1979 amendment appears to have removed the need to show wilfulness in respect of the first (conduct) element of the offence in cases under the second limb, it left the presumption regarding the third (endangerment) element untouched.

64. The Court of Appeal sought guidance from English case-law on other offences, notably unlawful act manslaughter and endangering the safety of persons carried on the railway, contrary to section 34 of the Offences Against the Person Act 1861 (“section 34”).[48]

65. It held that, as in manslaughter cases, the requirement of an unlawful act does not of itself import mens rea into section 72.[49] I respectfully agree.

66. More problematical is the Court of Appeal’s extrapolation from the decisions in two English cases on section 34 to hold that under section 72:

67. It relied first on R v Bowray,[51]an 1846 case in which the accused was standing on a bridge over a railway line and dropped or cast a stone on a passing train. He was convicted on a finding that he had deliberately cast the stone, Alderson B directing the jury that it was unnecessary to prove that he had thereby intended to endanger anyone’s safety. That decision was approved in R v Pearce,[52] a decision in 1966, where the defendants disabled railway signals by stealing copper wire from the signal boxes. Widgery J apparently held that it was only necessary to prove that by their intentional unlawful act of larceny, the defendants had created a source of danger to persons travelling on trains, no mention being made of any mens rea requirement in respect of such endangerment.

69. I therefore do not accept that analogies drawn with unlawful act manslaughter and section 34 are capable of providing a sufficient basis for displacing the presumption of mens rea.

70. As we have seen, both courts below considered the existence of the two statutory defences some justification for holding that no mens rea requirement attaches to the consequence of endangerment. To what extent are they capable of displacing the presumption of mens rea?

71. As we have seen in Section B above, the unlawful act relied on by the prosecution against both appellants is breach of section 10(3)[55] based upon alleged breaches of rules 5 and 8, and in the case of the 1st appellant, also of rule 9 of COLREGS, an offence subject to the section 10(4) defence that the accused took all reasonable precautions to prevent such contraventions.[56] In my view, the presumption of mens rea is plainly not dislodged by the existence of that defence.

72. In the first place, liability under the unlawful act limb may be constituted by unlawful acts other than breaches of COLREGS, so that the “reasonable precautions” defence is by no means always available as a qualification to liability under section 72. The existence of that defence in relation to charges under section 10(3) therefore does not justify any broader conclusion as to the mental requirements of the section 72 offence and especially does not warrant the conclusion that it dispenses with mens rea.

73. Secondly, for the same reason as that given in Section G.3 above, the existence of the section 10(4) defence cannot displace the presumption of mens rea in relation to the requirement of endangerment. It is a defence that operates to negative an essential actus reus element in section 72 concerning the conduct of the accused where the unlawful act charged is breach of section 10(3). It does not touch on the third actus reus element of endangerment and so cannot dislodge the presumption in respect thereof.

74. In my view, the reasonable excuse defence which may arise on a charge under the second limb of section 72 also does not address the mental requirements relating to the element of endangerment and therefore does not supplant the presumption of mens rea in respect thereof.

75. Under the second limb, the prosecution has to prove that the defendant’s conduct (which may involve him acting or omitting to act in any manner whatsoever) endangered the safety of persons at sea and that such conduct, resulting in that consequence of endangerment, was without reasonable excuse. It was, in my view correctly, conceded by the prosecution that the reasonable excuse defence imposed merely an evidential burden on the accused, the legal burden of negativing the defence if sufficiently raised, resting throughout on the prosecution.

76. Plainly, the defence of reasonable excuse is separate and distinct from any requirement that intention, recklessness or some other state of mind must accompany the act or omission in question in order to constitute the offence in the first place.

77. This may be illustrated by taking, for example, section 33(1) of the Public Order Ordinance[57] which makes it an offence for any person, without lawful authority or reasonable excuse, to have with him in any public place any offensive weapon. It is plainly necessary for the prosecution to prove that the defendant knew that he had the weapon with him.[58] But even where such intention is duly proved, the statutory defences of lawful authority and reasonable excuse remain available and have spawned much case-law. A person who knowingly has with him an offensive weapon in a public place may, for instance, have the defence of lawful authority since he may be carrying the weapon as part of his lawful duties, being a policeman or a soldier. And a person may have a reasonable excuse in certain cases where he arms himself in self-defence in anticipation of being subjected to imminent unlawful attack[59] or where he has some other “good reason” for carrying the weapon.[60]

78. The reasonable excuse defence in section 72 likewise concerns the existence of good reasons, extrinsic to the defendant’s mental state, which excuse his conduct which has endangered safety. One can imagine, for instance, a case where a vessel sails down the wrong side of a narrow channel endangering the safety of others, but where the master gives convincing evidence that the vessel’s steering gear had failed through no fault of his own and is found to have a reasonable excuse. Unexpected equipment failure might also provide a reasonable excuse for a vessel not showing the masthead and side lights required by rule 17 and not making the sound signals required by rule 35 of COLREGS. The possible availability of a reasonable excuse defence in such cases tells us nothing about whether mens rea must be proved in relation to the consequence of endangerment in the first place.

79. For the foregoing reasons, I conclude that the approach adopted in the decisions below was flawed. They failed to acknowledge the existence of a presumption of mens rea applicable to the element of endangerment in section 72; failed to identify a proper basis, if any, for displacing that presumption; and if displaced, failed to consider what mental requirements should replace the supplanted mens rea.

80. Mr Simon Westbrook SC[61]seeks to support the findings below in favour of absolute liability in relation to the element of endangerment. He submits[62]that section 72 comes within Hin Lin Yee’s fourth alternative,[63] that is, that on its true construction, section 72 dispenses with mens rea and limits the defendants to reliance on the “all reasonable precautions” defence provided by section 10(3) and the “reasonable excuse” defence under section 72 itself.

81. I shall shortly consider the case for absolute liability as a matter of construction, but would say at once that the Hin Lin Yee fourth alternative must be rejected in the present case. It is an alternative which only arises where the presumption is held to have been supplanted and the express statutory defences are inconsistent with the second and third alternatives importing a “halfway house”.[64] Since, as I have endeavoured to show in Section G above, neither of the defences under section 10(3) and section 72 bears on the mental requirements relating to the element of endangerment, they are not inconsistent with either full mens rea or a halfway house mental state being required in respect of that element.

82. Before proceeding to construe section 72, it is necessary to re-visit Hin Lin Yee and to re-formulate the available alternatives in a manner appropriate to a case such as the present.

83. First, as indicated above, the five Hin Lin Yee alternatives need reformulation to address the possible alternative mental requirements relating to the consequences, and not just the circumstances, of the defendant’s conduct. So modified, the five possible alternatives may be stated as follows (with references to the element of consequences here italicised):

84. Secondly, the principle enunciated in Hin Lin Yee should be revisited to address the position regarding a serious criminal offence as opposed to an offence that is merely regulatory.

85. Hin Lin Yee was concerned with the mental element to be proved in relation to a regulatory offence, namely, breach of section 54(1) of the Public Health and Municipal Services Ordinance.[65] The offence involved selling drugs which were intended for use by man but unfit for that purpose (because they had been contaminated by a certain contaminant). It was an offence punishable on summary conviction by a maximum fine of $50,000 and imprisonment for 6 months.[66] The appellants’ case was that they were unaware of the contamination and it was held by this Court that the presumption of mens rea had been dislodged and that the appellants were confined to the statutory defences expressly provided for, which they were unable to make good on the facts.[67] The 1st and 2nd appellants were respectively fined $45,000 and $1,500.[68] As was made clear in Hin Lin Yee, no constitutional issues were raised in that appeal and the discussion proceeded solely on the basis of common law principles.[69]

86. The present case involves an offence which is not merely regulatory but involves serious criminal liability, being triable on indictment and punishable by a fine of up to $200,000 and by imprisonment for up to 4 years. In cases like the present, a conviction is also likely to have serious professional ramifications for the mariners involved. As we have seen, a constitutional challenge was made to the reverse onus apparently imposed by the statutory defences provided for.

87. I concluded in Hin Lin Yee that the halfway house to be adopted in Hong Kong was the third (reverse onus) and not the second (evidential burden) alternative, on the basis that once the Court was prepared to go so far as to accept as a halfway house, the imposition of a reverse onus on the defendant, there was little to be said in favour of the second alternative. I also remarked that the latter alternative was less straightforward, since it requires the prosecution to proceed subject to the contingency that the accused discharges an evidential burden as to his having had reasonable grounds for an exculpatory belief. The prosecution would then have to prove either the absence of an honest belief or the absence of reasonable grounds for such belief.[70]

88. Such reasoning is apt when the Court is dealing with a merely regulatory offence and no constitutional issue is raised. Given the relatively minor gravity of the offence, the Court may be prepared to adopt the reverse onus alternative which falls close to the absolute liability end of the spectrum as best reflecting the presumed legislative intent upon displacement of the presumption of mens rea. Such alternative remains available and may be compelling when construing regulatory offences.

89. However, where, as in the present case, the Court is not dealing with a merely regulatory offence, it may conclude that the presumption of mens rea has been displaced but may not accept that adoption of the third alternative would reflect the legislative intent. Where a serious offence carrying severe penalties is involved, a reverse onus may also be regarded as making excessive inroads into both the presumption of mens rea and the constitutionally protected presumption of innocence. In such cases, the Court may find it more appropriate to infer that the second alternative best reflects the presumed intention of the legislature, doing no violence to the presumption of innocence.

90. It follows that all five of the reformulated alternatives set out above should be considered as possible conclusions when construing statutory criminal offences – both serious and regulatory – which are silent or ambiguous as to the state of mind required in respect of the element of the offence under consideration. I proceed to discuss the proper construction of section 72 on this basis.

91. It is convenient to begin by considering whether the decisions below in favour of absolute liability for the element of endangerment can be justified.

92. A conclusion in favour of absolute liability is of course never lightly reached since it is in principle objectionable, especially where the offence is serious, that a person should be made criminally liable where he did not deliberately or recklessly engage in the prohibited conduct or where he was ignorant of circumstances making his conduct criminal or where he acted harbouring an honest and reasonable belief inconsistent with liability.

93. As pointed out in Hin Lin Yee,[71] absolute liability is usually only entertained as a possibility in relation to “quasi-criminal” or merely regulatory offences where the public interest prohibits and imposes a minor penalty on the conduct in question. But even in such cases, the fact that the offence is merely regulatory is not a sufficient basis for imposing absolute liability, particularly where a middle course exists.[72] Before deciding on absolute liability, the Court will ask whether some less Draconian approach is appropriate. It will have to be satisfied that some useful purpose is served by imposing absolute liability. Illustrations given in Hin Lin Yee include cases where the prohibited conduct is intrinsically unacceptable so that absolute liability encourages individuals to steer well clear of possible contravention. Also instanced are cases where certain tasks are delegated to an agent and where it makes sense to subject the principal to absolute liability to encourage him to take proactive measures to prevent contravention by his agent in carrying out those tasks.

94. Viewed against such considerations, there is plainly no justification for imposing absolute liability in relation to the endangerment of safety under section 72. It may have started off as a relatively minor offence but in 1979, it was made indictable and punishable by a fine of $200,000 and imprisonment for 4 years, a ten-fold and eight-fold increase in the respective penalties.

95. Endangerment of the safety of persons at sea is at the very core of this serious offence and it would in principle be objectionable to find someone guilty of such criminality in the absence of some culpable state of mind (not necessarily full mens rea)with regard to endangerment.

96. Given the breadth of the offence, in many cases covered by the section, little purpose would be served by the imposition of absolute liability. The safety of persons at sea could sensibly be promoted by deterring dangerous conduct which is deliberate, reckless or carried out without an exculpatory state of mind. But no deterrent value or social benefit would be gained by punishing persons who lack such culpable mental states since they could not be expected to behave any differently notwithstanding potential liability.

97. I therefore conclude that the courts below fell into error in holding that liability under section 72 could be established on the basis of absolute liability regarding the element endangerment.

98. I turn then to consider what the proper construction of section 72 is, asking whether the presumption of mens rea is displaced and if so, by what alternative mental element.

99. The question which falls to be examined is whether the legislative intention behind section 72 is to require the prosecution to prove beyond reasonable doubt that the conduct (either under the unlawful act limb or the second limb) of each appellant was intentional or reckless regarding the consequence of endangering the safety of N67’s crew.

100. It is clear on its face that the purpose of section 72 is in general terms to punish and deter conduct which endangers the safety of persons at sea. And as the Court of Appeal pointed out,[73] the 1979 amendments which made the offence much more serious were motivated specifically by a major influx of Vietnamese boat people and targeted owners and masters who conveyed them to Hong Kong in dangerously unseaworthy and overloaded vessels.

101. However, the net cast by section 72 is very much wider. Its language is striking for its inclusivity and breadth. Unlike section 10(3) which applies only to the owner, the master and the person responsible for the conduct of the vessel, section 72 applies to “any person”. The conduct that it prohibits may take virtually any form so long as it endangers safety. Under the first limb, the conduct has to be unlawful, but under the second limb, it is conduct which causes danger “in any manner whatsoever”. Even under the unlawful act limb, there is no restriction on the kind of unlawfulness that suffices: it may or may not consist of an offence which itself requires proof of mens rea. The central criminality targeted is therefore the causing of danger to the safety of others. The legislature appears almost indifferent as to how such endangerment is caused.

102. Since the 1979 amendment was aimed primarily at unscrupulous human traffickers acting in ways which self-evidently endangered the lives and safety of the persons carried in their vessels, it is perhaps unsurprising that the legislature did not expressly address the mental element pertinent to the consequence of endangerment. However, the breadth and lack of specificity of the ingredients of section 72 result in the section covering a wide spectrum of behaviour in very different contexts. Cases against human traffickers sending out overloaded “rust bucket” vessels obviously raise very different policy, evidential and enforcement issues from the issues arising in the context of a maritime collision between ocean-going vessels navigated by certified master mariners or pilots. The section has also been held to apply, for instance, where danger is caused by smugglers deliberately or recklessly running high-powered boats at great speed through congested areas to evade chasing police vessels; and in the different situation involving coastal vessels causing endangerment by failing to keep a proper lookout.

103. Given the range of situations covered by section 72, I do not think that the Court should categorically affirm the presumption and construe the section as invariably requiring proof of full mens rea. There will of course be many cases where the prosecution faces little difficulty proving deliberate or reckless endangerment. But there will equally be cases like the present where significant problems of proof may arise since the collision may be the result of a combination of factors arising out of the interaction between two vessels developing over time, making it difficult for the prosecution to establish the facts upon which firm inferences can be drawn regarding each defendant’s state of mind.

104. The situation just described exemplifies what was envisaged in Hin Lin Yee when it was emphasised that:

105. The question whether the presumption of mens rea should be held to be displaced should therefore be held in abeyance until the possible alternatives are considered.

106. I have already rejected the fourth alternative on the ground that the available statutory “all reasonable precautions” and “reasonable excuse” defences do not bear on the element of endangerment and therefore are not inconsistent with adoption of a halfway house alternative.[75] I have also rejected the fifth alternative which favours absolute liability.[76] That leaves the second and third alternatives as candidates to replace mens rea should the presumption be dislodged.

107. In discussing the need to reformulate the five alternatives,[77] I noted that when dealing with a serious as opposed to regulatory offence, a court considering whether the presumption of mens rea should be displaced – and if so, replaced by what – is likely to find it more appropriate to infer that the second (rather than the third) alternative best reflects the presumed intention of the legislature since that alternative lies closer to the full mens rea end of the spectrum and does no violence to the presumption of innocence, as befits a serious offence. On the other hand, the third alternative is likely to be more compelling in relation to a regulatory offence.

108. In my view, that approach is apposite to the present case and the conclusion to be drawn is that the second alternative applies on the true construction of section 72. Support for imputing to the legislature an intention favouring the second alternative if mens rea is supplanted may be found in the existence of the reasonable excuse defence. Although, as pointed out above,[78] that defence does not bear on the mental requirements relating to the element of endangerment but arises as a subsequent issue, the existence of such defence shows (at least under the second limb) that the legislature’s policy is not to penalise endangerment resulting from conduct for which there is a reasonable excuse. This, in my view, justifies the conclusion that given displacement of the presumption of mens rea, it would likewise have been the legislative intention that mens rea should be replaced by a mental element which focuses on the reasonableness of the grounds of an exculpatory belief. While “reasonable excuse” only arises as a matter of defence under the second limb, the second alternative should be taken to apply as a constituent element of prima facie liability under both limbs.

109. The conclusion reached as to the proper construction of section 72 may therefore be summarised as follows. On its true construction, it is an offence to which the second alternative applies. In other words, it is an offence where:

110. I might add in relation to the last proposition that the prosecution conceded that the reasonable excuse defence operates to impose merely an evidential burden on the accused, with the prosecution bearing the legal burden of negativing such defence if it should be sufficiently raised on the evidence. Mr Westbrook SC indicated that the prosecution saw no practical difficulty in approaching the defence on that basis. In my view, that concession was correctly made and construing the reasonable excuse provision as imposing merely an evidential burden obviates any need to debate the proportionality of a reverse onus in the present context.

111. When the 1st appellant applied to the Appeal Committee for leave to appeal, he sought to advance six grounds on the substantial and grave injustice basis in addition to the legal questions dealt with on this appeal. Leave was only granted on the points of law ground. Mr Peter Duncan SC[79] nevertheless sought leave at the hearing to argue two points on the substantial and grave injustice ground.

112. As was decided in HKSAR v Lee Ming Tee,[80]the Court undoubtedly has jurisdiction to determine all questions which are related to the points which have been certified and on which leave to appeal was granted. However, it will only exercise its discretion to entertain additional questions in exceptional cases, particularly where determination of such questions is necessary or desirable to ensure the effective disposal of the issues arising on the appeal.

113. The Court heard Mr Duncan de bene esse but refused the 1st appellant leave to proceed without calling on the prosecution to reply. No basis was shown for exercising the Court’s discretion in favour of granting leave, especially since the points advanced appeared to the Court to be unarguable.

114. Mr Duncan sought leave to raise two points both based on the complaint that the Director of Marine was not called as a witness by the prosecution. It was suggested that the “official view” of the Marine Department, contrary to that espoused by the prosecution, was that the stretch of water in which the collision occurred was not a narrow channel within rule 9 of COLREGS, a proposition which it was said that the Director could have been expected to confirm, explaining the reasons for his view. The two points sought to be made based on his not being called by the prosecution were:

115. Insofar as the first point involves inviting the Court to rule on whether the prosecution should or should not have been brought, it was not a competent ground of appeal since Article 63 of the Basic Law guarantees the Department of Justice control of criminal prosecutions, free from any interference.

116. Insofar as the proposition was that the absence of the Director’s evidence meant that the case could not be proved beyond reasonable doubt, that is a point which could be and was made in a submission of no case to answer, which was rejected. It was a point that was open as a submission at the end of the trial and indeed, on appeal to the Court of Appeal. It is self-evident that it stood no chance of success since, the 1st appellant was found guilty under section 72 beyond reasonable doubt.

117. If it was genuinely thought that the court’s process was being abused, an application could have been made to stay proceedings, but such application was never made. In any event, there is no viable basis for the purported ground focussing on whether the Secretary for Justice was entitled to bring the prosecution.

118. As to the second point, Mr Duncan accepted that the 1st appellant was himself perfectly entitled to call the Director and to question him about his and the Department’s views on the status of the stretch of water in question. Having chosen not to do so, there can be no basis for complaining about the Director’s absence as a witness, let alone submitting that his absence somehow represents a substantial and grave injustice done to the 1st appellant.

119. The Courts below having fallen into error in treating section 72 as an offence of absolute liability regarding the element of endangerment, the question for this Court is whether it follows that the appeals must be allowed and the convictions quashed or whether either or both convictions ought to be upheld on the basis of the proviso.

120. Section 17(2) of the Court’s statute[81]empowers the Court to exercise any of the powers of the court from which the appeal lies. That includes the Court of Appeal’s powers under section 83 of the Criminal Procedure Ordinance[82]which requires appeals to be allowed where the conviction is unsafe or unsatisfactory; or based on a wrong decision on a point of law; or involved a material irregularity in the course of the trial, but makes this subject to a proviso which states:

121. The relevant test for applying the proviso is well established. As stated in Ewan Quayle Launder v HKSAR[83]and often applied,[84]the test is “whether a reasonable jury, properly instructed, would, on the evidence, without doubt convict or would inevitably come to the same conclusion”. Where, as in the present case, the trial was in the District Court without a jury, the inevitability or otherwise of a conviction can be judged with the benefit, not only of the evidence admitted at the trial, but also of the Judge’s explicit findings. I would add, with respect, that although the Judge was wrong-footed on the mens rea question, her judgment is highly impressive for its comprehensive analysis of the evidence and the meticulous and detailed findings made.

122. The question which this Court must ask arising out of the foregoing discussion of mens rea is whether, notwithstanding the error of law regarding the mental requirements pertinent to the element of endangerment in section 72, this Court can be satisfied that on the evidence and her Honour’s findings, the Judge would inevitably have come to the same conclusion, convicting the appellants, if she had applied the correct legal test.

123. On the particular charges in the present case, that translates into the following questions to be asked of each appellant:

124. So far as the actus reus is concerned, Lord Clarke NPJ demonstrates how the 1st appellant’s conduct caused endangerment to the safety of his vessel and his crew by failing to alter course to starboard at any stage pursuant to his duty to keep to the starboard side of the channel under rule 9.[85]

125. Lord Clarke NPJ also deals in detail with the exculpatory case on mens rea advanced by the1st appellant and the Judge’s findings thereon. I gratefully adopt that analysis and will seek in this judgment only to summarise the conclusions relevant to the proviso question.

126. The 1st appellant seeks to justify his actions leading up to the collision along the following lines:

127. The 1st appellant’s case is that the Court should infer from such evidence that the 1st appellant believed himself to be faithfully following internationally recognized rules of navigation designed to prevent maritime collisions and that he therefore lacked any appreciation or foresight that his conduct might result in endangerment of the safety of his crew.

128. He argues that whether or not he was objectively right in his understanding of the collision regulations, his abovementioned beliefs were honestly and reasonably held and that it cannot be said that the prosecution has shown beyond reasonable doubt either that he did not in fact hold such beliefs or that they were not held on reasonable grounds. The 1st appellant therefore submits that this Court would not be justified in finding that, applying the proper legal test as to section 72’s mental requirements, the Judge would inevitably have reached the same conclusion and convicted him.

L.2 The Judge’s findings regarding the 1st appellant relevant to the proviso

129. The Judge made findings specifically regarding the reasonableness of 1st appellant’s aforesaid beliefs. Those findings were made in the context of deciding whether the “all reasonable precautions” defence under section 10(3) or the “reasonable excuse” defence under section 72 availed the 1st appellant. They were therefore made under a different legal rubric but their substance is directly in point so far as the mental requirements are concerned. Those findings are therefore of direct relevance to the question of whether the proviso should be applied.

130. For the reasons given by her and analysed in detail by Lord Clarke NPJ, the Judge found and was entitled to find:

131. As Lord Clarke NPJ points out, the critical case established against the 1st appellant was his unreasonable failure to appreciate that the buoyed channel was a rule 9 narrow channel; his insistence on maintaining his course until the last and his consequent failure to take starboard helm at any stage. The 1st appellant’s contention that he had honest and reasonable grounds for believing that his conduct did not give rise to any risk of endangerment since he was properly following the crossing rules designed to prevent collisions, was therefore negated. Even assuming that the 1st appellant honestly believed that his conduct was not such as to endanger the safety of others, such a belief was properly found by the Judge not to have been held on reasonable grounds.

132. It follows, in the light of the Judge’s findings, that if she had applied the correct legal test by adopting the second alternative as the mental requirement for the element of endangerment in section 72, she would inevitably have come to the same conclusion and convicted the 1st appellant. The proviso should accordingly be applied in the case of the 1st appellant and his appeal dismissed.

133. On the 2nd appellant’s evidence, his appreciation of the situation was diametrically opposed to that of the 1st appellant:

L.5 The Judge’s findings regarding the 2nd appellant relevant to the proviso

134. The Judge’s assessment of the each appellant’s responsibility was strikingly different. In the case of the 1st appellant, as we have seen, she made a series of specific findings as to the unreasonableness of his beliefs which contradict any submission that he acted in the honest and reasonable belief that his conduct was not such as endanger the safety of others. On the other hand, the Judge’s findings critical of the 2nd appellant focus not on the unreasonableness of his subjective beliefs, but on his conduct falling short of objective standards of good seamanship.

135. The Judge found that the 2nd appellant correctly understood that the buoyed channel was a rule 9 channel;[111] that the crossing rules did not apply;[112]and that it was reasonable for YH to enter the channel between CP1 and CP2 buoys and to make her approach on the starboard side; whereas N67 was hogging the channel and ought to have moved over to her starboard side.[113]

136. Her Honour’s fundamental criticism of the 2nd appellant was that he had left collision avoidance action too late. The collision occurred at 21:13:44h. She found that the order of starboard 20 should have been given at the latest by 21:11:35h and therefore that he had given it about a minute too late.[114] As she puts it: the collision avoidance action by YH was simply “too little too late”.[115]

137. This led the Judge to infer that the 2nd appellant must have failed to keep a proper lookout since he should have determined that there was a collision risk sooner and acted more promptly to avert it.[116]As she puts it:

138. In my view, there is no basis for application of the proviso in the 2nd appellant’s case. I accept the submission of Mr Timothy Brenton QC that on the evidence and the Judge’s findings, the 2nd appellant’s navigation of YH was conditioned throughout by his honest and reasonable belief that N67 would move to starboard and that he has therefore raised at least a reasonable doubt as to whether he honestly and reasonably – albeit mistakenly – believed that his conduct was not such as to endanger the safety of others, in particular the crew of N67. The finding that he had ordered collision avoidance action to be taken a minute too late for it to avert the collision involved criticising an error of judgment on his part. It did not negate his exculpatory honest and reasonable belief. It cannot be said that the Judge would inevitably have convicted, applying the correct legal test.

139. I therefore conclude that the 2nd appellant’s appeal must be allowed and his conviction quashed.

140. I agree with the judgments of Mr Justice Ribeiro PJ and Lord Clarke of Stone-cum-Ebony NPJ.

141. I have read the draft judgment Mr Justice Ribeiro PJ. I respectfully agree with it and will not traverse the same ground in this judgment. I will focus on two issues. In doing so I will use the same abbreviations and terminology as Mr Justice Ribeiro PJ. The first issue involves an analysis of the navigation of the two vessels which led to the collision and to the tragic loss of life that followed. The most important question for consideration under this head is what provisions of the COLREGS applied to the navigation of each vessel. The second is the question what follows with regard to the actus reus and mens rea of the offencefrom the conclusions on the first issue and from the legal principles identified by Mr Justice Ribeiro PJ. In the case of each appellant, the question is whether his appeal against conviction should be allowed or dismissed.

142. The Court of Appeal certified one question of law and this court certified a further seven. The first is concerned with mens rea and has been fully discussed by Mr Justice Ribeiro PJ. The remaining seven are relevant or potentially relevant to the navigation of the vessels. They are these:

143. The N67 was a twin engine and twin screw Ukrainian oil rig supply vessel, of 2,723 gross registered tonnes, 81.37 metres in length and 16.3 metres in beam with a maximum draught of 4.8 metres. She had some containers on board. The YH is a single engine and single screw vessel, of 36,544 gross registered tonnes, 225 metres in length and 32.2 metres in beam with a draught of 11.48 metres forward and 12.01 metres aft. She was laden with a cargo of 57,842 tonnes of maize. Both vessels were exhibiting white masthead lights, red and green side lights and a white stern light. The N67 was considerably more manoeuvrable than the YH.

144. On the day is question (22 March 2008) the weather conditions were good. The wind was west or north-west and was blowing at about 12 kph and visibility was about 1.7 miles or a little more. The tide was flooding in a generally westerly direction at about 1.5 knots and the height of tide was about 1.7 metres.

145. The primary facts are not now in dispute. That is because, as the Judge observed at para 106 of the Reasons for Verdict (references hereafter to paragraph numbers are references to the Reasons for Verdict), the relevant events were recorded and captured on radar at the Vessel Traffic Centre (“VTC”) provided by the Vessel Traffic Services (“VTS”) system maintained by the Marine Department (“MARDEP”) and on the voice data recorder (“VDR”). If the N67 had a VDR it was not available because of course the N67 sank. The Court of Appeal attached to their judgment a plot showing the tracks of the vessels over the ground which had been prepared at trial by one of the expert witnesses, Captain Third. It is not in dispute that the tracks shown are substantially accurate and I annex the same plot, which is drawn on a copy of Admiralty Chart 4122, as Annex 2 to this judgment. The N67 was proceeding at about 10 to 10.5 knots and the YH at about 13.5 knots over the ground I note in passing that, like the courts below, we saw a video of the VTS radar images, which shows clearly how the collision occurred. The Judge based her findings of fact on this material and it is, in my opinion correctly, not suggested that she was wrong to do so. I will return below, so far as necessary, to the navigation of each vessel after considering the principal navigational issue between the parties in this appeal, which is whether the N67 was navigating in a narrow channel within the meaning of rule 9(a) of the COLREGS.

146. For this purpose I attach as Annex 3 a further plot which is also drawn on chart 4122, which is again accepted as broadly accurate. It shows both the geographical area with somewhat more clarity and the tracks of the vessels from 2107 until collision at 2113.44. With one or two false starts the prosecution asserted that the N67 was navigating in a narrow channel delineated by six CP (or Castle Peak) buoys. Castle Peak Bay is some way to the north. There were three CP buoys on the north side of the channel. Each was delineated on the chart by a capital G, meaning green, and, from the west, they were marked as CP5, CP3 and CP1. They were passed to port by the N67, as the incoming vessel.

147. There were only two buoys on the south side, each designated a red buoy. From the west they were marked CP4 and CP2. Buoy CP4 was broadly opposite buoy CP5 and buoy CP2 was broadly opposite buoy CP1. There was no buoy opposite or broadly opposite buoy CP3. The reason for that can be seen from the reference on the chart to Area 6, which is broadly opposite and to the south of CP3. The Note referred to on the chart says that, except with the permission of the Director of MARDEP, Area 6 is prohibited to all vessels with an air draught of more than 15 metres and Area 4 is prohibited to all vessels.

148. The N67 was a vessel with an air draught of more than 15 metres and accordingly could not enter Area 6 without permission. The case for the prosecution was that the channel delineated by the CP buoys was a narrow channel within rule 9(a) of the COLREGS, which provides:

The prosecution submitted that, in the case of any vessel proceeding eastward along the channel which did not have an air draught of 15 metres, her obligation was to keep as near to the southern edge of the channel, which was a line drawn between buoy CP4 to the west and CP2 to the east, as was safe and practicable. That line would pass very close to the pecked line on the chart between the marks to the north-west and the north-east of Siu Mo To Island. Such a vessel could pass close to the northern edge of the prohibited area noted on the chart as Area 4.

149. A vessel like the N67 was in a different position. She could not properly proceed south of the northern edge of Area 6 shown on the chart because, in the light of the notation on the chart, it was not safe and practicable to do so. It was therefore submitted on behalf of the 1st appellant that, assuming that the CP buoyed channel was a narrow channel for the purposes of rule 9(a), in the positions shown on the chart the N67 was complying with the rule between 2107 and 2109. It was accepted on all sides that that was the case. The issue between the parties was what was the position as from about 2109. The case for the prosecution and the 2nd appellant was that the N67 was under an obligation to alter course to starboard in order, in compliance with rule 9, to proceed along the channel as close to the southern edge of the channel as was safe and practicable. It was said that it was safe and practicable to navigate close to the southern edge of the channel, which was on a line approximately between the Siu Mo To NE mark and the CP2 buoy.

150. It was said on behalf of the 1st appellant on the other hand that the CP channel, although a buoyed channel, was not a narrow channel, that the N67 was not under an obligation to alter course to starboard to proceed along the starboard side of the channel and that, as the vessels approached each other, there came a time when they were crossing so as to involve risk of collision, so that the YH was under a duty to keep out of the way of the N67 and to take early and substantial action to do so under rules 15 and 16 of the COLREGS and the N67 was under a duty to maintain her course and speed until a late stage under rule 17. It was thus submitted that the N67 was not to blame for failing to alter course to starboard to proceed close to her starboard side of the channel. The Judge rejected this submission and held that the principal cause of the collision (and therefore the loss of life) was the failure of the second defendant as master of the N67 to alter course to starboard at an early stage. She held that, in colloquial terms, he was hogging the centre of the channel. The Court of Appeal reached the same conclusion. It was submitted that both courts were wrong to reach these conclusions. I will return below, so far as necessary, to the detail of the COLREGS and to other aspects of the navigation of the two vessels but, so far as the navigation is concerned, it seems to me that the most important issue in the appeal is whether the Judge was correct to hold that the buoyed channel was a narrow channel within the meaning of rule 9.

151. I will return below to such assistance as can be derived from the authorities but the question whether a particular stretch of water is a narrow channel within the meaning of rule 9 seems to me to be essentially a question of fact. The question has to be asked by masters and pilots in a myriad of different circumstances all round the world day in and day out. They have their experience and expertise to rely upon. They also have the relevant chart or charts and the relevant Admiralty Pilot or other sailing directions, together with any notices to mariners which may be relevant. They must of course always be aware of the local law and any local regulations or byelaws. Whether a particular area is a narrow channel cannot depend upon the opinion of the harbour master, the director of the port authority or the port authority itself unless that opinion has been disseminated to mariners. It follows that I would not accept the submission made on behalf of the 1st appellant that the opinion of the Director of MARDEP was of any assistance, unless it was disseminated to those navigating Hong Kong waters.

152. What then did the 1st and 2nd appellants have available to them to enable them to decide whether the buoyed channel delineated by the CP buoys was a narrow channel for the purposes of rule 9? The 1st appellant did not have the assistance of a pilot because, as a vessel of less than 3,000 tons gross, the N67 was not obliged to take a pilot and did not do so. The 1st appellant did however have some experience of navigating in the area. He had used the buoyed channel 20 or 30 times before The 2nd appellant was of course the chief pilot on board the YH for which pilotage was compulsory. He was a pilot of very considerable experience.

153. In this judgment I shall use the expression “CP channel” as meaning the channel delineated by the CP buoys and the expression “narrow channel’ as meaning a narrow channel within the meaning of rule 9. The answer to the question posed above is that the principal aids available to the appellants were the Admiralty charts, notably chart 4122, and the Admiralty Pilot Book.

154. The relevant Pilot Book is entitled “Admiralty Sailing Directions China Sea Pilot Volume 1”. Paragraph 9.142 is headed Ma Wan to Urmston Road. Ma Wan is the north-east tip of Lantau Island, which outbound vessels like the YH have to pass to port. It is about 1.75 miles to the east-north-east of the position shown on chart 4122 for the YH at 2107. Urmston Road is a little off chart 4122 to the west and is north of the Hong Kong International Airport. Paragraph 9.142 describes the general passage from the east and identifies Cheung Sok to the south and Brothers Point to the north, both of which are marked on chart 4122. Paragraph 9.142 continues:

155. The Pilot Book thus plainly described the CP channel as a buoyed channel and buoys CP1 and CP2 as starboard and port hand buoys respectively. It is common ground that the buoys are IALA buoys, which means that they conform to and are part of the uniform system of buoyage laid down by the International Association of Lighthouses. They are generally used for well-defined channels and denote the lateral limits of the channel. Under the IALA system vessels should keep the green buoys to starboard when proceeding in the direction of flow. In the CP channel the direction of flow is westward, as indicated by the large arrow and two small rings marked on the chart just up-channel of buoys CP4 and CP5. Thus, if using the channel, the YH would pass the green northern buoys to starboard and the N67 would pass them to port and thus would pass the red southern buoys to starboard.

156. The Pilot Book describes the route through the channel as a deep-water route because the depths of water in the vicinity are such that some vessels are so constrained by their draught that they have to navigate within the buoyed channel. An example of such a vessel was the BERLIOZ, which was a large container vessel of some 300 metres in length which was following the YH. It was apparent from the radar that she was a large vessel and it was not suggested that other vessels could assume that she could safely navigate outside the CP channel. The draught of the YH was not such that she had to navigate in the channel but, given her draught of just over 12 metres, it was readily understandable that she would wish to do so. The Judge found that the buoyed channel was generally used by ocean going vessels and that local vessels use the area to the north of the channel, where traffic is usually heavy. It was common ground that within the area to the north of the channel, outside the limits of the deep water route (ie the channel) vessels did not adhere to the narrow channel rule. This is not surprising because that area is plainly not a narrow channel and the Judge did not find that it was.

157. The width of the CP channel can be seen on the chart. The CP buoys are positioned on the 15 metre contour except for buoys CP1 and CP2 which are located on the 20 metre contour. The width of the channel varies between 3.1 cables at its widest, which is between CP1 and CP2 and 2.2 cables at its narrowest, which is between CP3 and the restricted area north of Siu Mo To.

158. As I read her judgment, the Judge found that the CP channel is a narrow channel on the basis of the following factors: (i) the geography, topology and bathymetry of the area, (ii) the size and manoeuvring characteristics of the vessels that use the channel; (iii) the fact that a buoyed channel had been created within a wider expanse of water; (iv) the object and underlying rationale of rule 9; (v) the manner in which seamen in fact navigate within the channel; and (vi) the expert evidence. I will briefly consider each of these aspects of the case.

159. These are often critical. As I see it, they include both geographical and depth constraints and buoys which are often positioned having regard to such constraints. As the Judge put it at her para 378, the boundaries can be marked by buoys, landmasses or sandbanks. In paras 379 and 380 she accepted the opinion evidence of Captain Third in these terms:

380. The bathymetric contours also define it as a narrow channel. The deep water indicates the deep water route designated for ocean going vessels. CP1 and CP2 are placed on the 20 metres contour lines whilst CP3, CP4 and CP5 are placed on the 15 metres line they are bathymetrically narrow and mark the bathymetrics of the channel.

381. I accept Captain Third’s opinion that even though in this case a vessel can still navigate outside of those buoys as there is sufficient room to do so the channel can still be considered a narrow channel. He said there are numerous places in European port approaches which are of that type where there is the ability to navigate outside the buoys but it was still a narrow channel. He gave the example of The Magellan Straits where there are quite broad stretches of water and the vessels navigate in central zones, but they are quite definitely following the narrow channel rule and ensuring their encounters are red to red.”

That seems to me to be entirely sensible advice which the Judge was entitled to accept.

(ii) The size and manoeuvring characteristics of the vessels that use the channel

160. This is to my mind a relevant consideration on the facts here because it is to be expected that deep water vessels will use the channel and proceed along the channel on the basis that the starboard hand rule applies to vessels in the channel.

161. As I read the Judge’s judgment, her essential reasoning is that the whole point of creating a buoyed channel out of a much wider area where the depths are significantly less is to create a channel in which vessels will be expected to pass safely port to port in accordance with the narrow channel rule. This seems to me to make good sense.

162. It is I think common ground that if a channel is designated as a narrow channel on a chart it will be a narrow channel within rule 9. However it was submitted on behalf of the 1st appellant that otherwise, where there is a dredged channel or, as in this case, a buoyed channel, rule 9 must apply across the entire stretch of water or not at all. It was on this basis that it was submitted that the CP channel could not be a narrow channel. It was submitted that, if it was anything, it was a fairway, and that it could not be a fairway because a fairway cannot exist except within a narrow channel and the water outside the CP channel was not itself a narrow channel. It was submitted that it is or would be impracticable to require vessels navigating within the CP channel to adhere to rule 9 in circumstances in which vessels just outside it were not subject to rule 9.

163. It was further submitted that, where a vessel is in a buoyed channel because she is constrained by her draught, she is expected to display the appropriate signal under rule 28 such that other vessels are expected to avoid impeding her passage under rule 18(d)(i). Moreover in a crossing situation the give-way vessel can slow down or stop. Attention was also drawn to the one way traffic system for vessels of more than 16 metres draft at the western end of the channel and to the restriction at Area 6.

164. Finally it was submitted that the outer limits of the channel were not defined. It was asked rhetorically whether the outer limits of the channel were to be the sounding lines and, if so, which and, if (as the Judge found) the buoys were to be joined up, which buoys and whether they were to be joined up by an imaginary straight line.

165. I would not accept those submissions, essentially for the reason given in para 161 above. I entirely accept that, where the chart marks a fairway or indeed a narrow channel, such a marking would almost certainly be conclusive. So, for example, there are a number of fairways marked on charts which cover Hong Kong, where they are for the most part marked with pecked lines. However, the expert evidence was that it is far from common to have narrow channels designated as rule 9 channels. I would reject the submission that a channel can only be a narrow channel if it is designated as such. There are, for example, at least two narrow channels in the Hong Kong area, apart from the CP channel, which are not so designated. One is a channel in Urmston Road designated by two buoys where a vessel’s position in the channel is ascertained by a light, another is in the south-west corner of chart 4122 and there may be others.

166. It is true that it was at one time suggested in the course of the trial that the channel was bounded, not by the CP buoys but by the 15 metre sounding lines shown on chart 4122. It was also at one time suggested that the CP channel was a fairway within a wider narrow channel encompassed by the 15 metre lines. However, the Judge was in my opinion correct to reject those suggestions. The whole point of the buoys was, and was obviously, to designate a channel, as expressly stated in the Pilot Book. In these circumstances it would make no sense to conclude that a wider and less precise channel is a narrow channel. Moreover the points taken relating to vessels constrained by their draught, to the one way traffic system for vessels of more than 16 metres draft at the western end of the channel and to the restriction at Area 6 do not seem to me to be of any assistance in deciding whether the CP channel is a narrow channel. I see no difficulty in a conclusion which treats the CP channel as a narrow channel and requires vessels proceeding along it to keep as near to the starboard side of the channel as is safe and practicable in accordance with rule 9.

167. This point is closely related to the last. The Judge expressed her conclusions broadly thus at her paras 124 and 125 under the heading “Passing of vessels within this stretch of water”. Vessels passing in the channel should pass port to port, with each vessel proceeding as close to her starboard side as is safe and practicable. Under rule 9 the vessel in the channel should plainly do so because vessels entering the channel will expect her to do so. Vessels navigating in the buoyed channel seeing vessels approaching the buoyed channel from Ma Wan Island in the opposite direction at an early stage are likely to see a number of alterations of course of these vessels as they may be constantly changing their aspect in order to shape a course into the channel. Safety requires a vessel approaching the channel so as to proceed along it to navigate so that if the vessels pass in the channel they will pass port to port. This will be achieved if the narrow channel rule applies. If it does not, there is considerable scope for confusion.

168. There was some discussion in the course of the argument as to whether this was a relevant consideration since it is a factor which might not be known to everyone who navigates in the particular area. However, it does seem to me to be a potentially relevant factor. If, for example, there is evidence that the vast majority of those navigating in the area treat a particular channel as a narrow channel, that seems to me to be a factor to take into account in deciding whether it is a narrow channel.

169. This seems to me to be such a case. The Judge discussed it at her paras 382 to 385. A letter was put in evidence signed by 96 of the 101 members of the Hong Kong Pilots Association (“HKPA”) in which they said that they considered that rule 9 applied to the CP channel. They added that all bulk carriers with a draft of over 11 metres would customarily navigate through the channel. In addition to the 2nd appellant and his co-pilot, who was the fourth defendant, the two pilots on board the BERLIOZ gave evidence to the same effect.

170. The Judge noted at para 383 that there was evidence that in the 24 hour period before the collision, when two vessels were going in opposite directions along the channel, there were no vessels passing starboard to starboard in the channel or at or near the entrance to the channel at CP1 and CP2. The Judge accepted at para 384 that there may be the odd occasion when vessels are going in opposite directions and one vessel may have gone north of the CP1 buoy in an easterly direction while the other vessel was heading west in the channel but she added, in my opinion correctly, that this did not show that seamen did not regard the CP channel as a narrow channel. She concluded at para 385 that seamen navigate the whole area of water from Ma Wan Island to the buoyed channel in accordance with Rule 9 but more particularly when they are approaching the buoyed channel and especially when they are within it.

171. This was an important part of the evidence on many topics which the Judge considered in considerable detail from para 279. In para 365 she identified the witneses who expressed views on either side of the question whether the CP channel was a narrow channel. She identified the following as in favour of the narrow channel view. Captain Cheung and Professor Ng were experts called by the prosecution. The Judge did not place particular emphasis on this part of their evidence. The two BERLIOZ pilots and the pilots on the YH also gave evidence to this effect. Captain Chen, who was the chairman of the HKPA, exhibited the letter to which I have referred together with the views of the 96 pilots.

172. The experts upon whose opinion on the question of narrow channel the Judge placed particular reliance were Captain Third, who gave evidence on behalf of the master of the YH, who was the second defendant, and Captain Simpson, who gave evidence on behalf of the second pilot on the YH, who was the fourth defendant. I have referred to the most important part of Captain Third’s evidence at para 159 above, where I have set out the terms in which the Judge accepted his evidence.

173. It is important to consider the evidence relied upon on behalf of the 1st appellant in support of the conclusion that the CP channel was not a narrow channel which the Judge rejected. This is important because it was submitted on behalf of the 1st appellant that the Judge ought not to have rejected it or that it was in any event evidence on the basis of which the 1st appellant could reasonably think that the CP channel was a narrow channel. The argument is that if independent experts or authorities, as for example the Director of MARDEP, held that opinion, the contrary opinion cannot be unreasonable, even if it is held to be wrong.

174. In para 366 the Judge identified those that took a different view from Captain Third and the others referred to above as being the MARDEP VTC operators, Dr Stephen Li of MARDEP, the 1st appellant and his two experts, Captain Browne and Captain Loynd. It appears to me that it is important to identify the reasons why each of these witnesses expressed the opinions they did and to consider whether they stand up to scrutiny. I will consider each in turn.

175. Three MARDEP VTC operators gave evidence. Two of them said that vessels using the channel normally kept to the starboard side of it. The other did not express a view. It is true that they all expressed the view that the CP channel was not a narrow channel, but the basis of their view was that it was not a channel at all and that, at any rate in some of the evidence, that that was because it was not designated as such. In this Court it is accepted that the channel is a channel, the only issue being whether it is a narrow channel. It is presumably accepted that it is a channel because it is designated as such in the Pilot Book and because the buoyed channel is clearly shown on the chart. In these circumstances the evidence of the VTC operators is of no real assistance.

176. The operators were plainly wrong in so far as they suggested that a channel cannot be a narrow channel unless it is gazetted. For the reasons given above, it does not follow from the proposition that a channel which is gazetted as a narrow channel is a narrow channel that a channel which is not so gazetted is not a narrow channel. All depends upon the circumstances. The Judge was in my view correct to accept Captain Third’s opinion in this regard.

177. There is much the same problem with the evidence of Captain Li of MARDEP. He expressed the view that the CP buoys demarcate the deep water route in the area, which is correct, but he added that as most vessels can navigate outside the buoys, the area is not considered as a narrow channel. In my judgment that view is plainly wrong. As Captain Third said, there are many areas in the world where some vessels can navigate outside a channel and some cannot. This is simply one of them. For the reasons given above, just as there is no reason why an IALA buoyed channel should not be introduced in such an area, so there is no reason why such a channel should not be a narrow channel. It is very likely that it will be but whether it is or not will depend upon a judgment based on all the circumstances of the case, as the Judge in effect held in accepting the opinion of Captain Third.

178. The alternative view is not in my opinion arguable. I accept that that conclusion involves the rejection of the opinions of the 1st appellant’s experts Captain Browne and Captain Loynd. However, I regret to say that the Judge found that both witnesses were biased in favour of the 1st appellant. She said at para 296 that she found that they had no notion of impartiality and that they were not convincing in their attempts to explain his conduct and to suggest that he observed all material rules that night. In the case of Captain Browne she gave detailed reasons at para 297 for her conclusion at the end of that paragraph that he was not giving independent evidence. She reached a similar conclusion in the case of Captain Loynd at para 299. The Judge accordingly placed no reliance upon their evidence in this regard. In my judgment, the Judge was entitled to reach the conclusions which she did and, understandably, no attempt was made to persuade the Court to reach a different conclusion as to their approach.

179. In these circumstances I do not think that it would be appropriate for this Court to hold that the opinions expressed by those witnesses were reasonable or correct, especially in circumstances in which the Judge had contrasted their evidence with that of Captain Third and Captain Simpson. She said at para 295 that the most comprehensive, logical, independent and fair expert was Captain Third. She found him to be an impressive and reliable witness. She concluded that he was clear in his thinking process and his explanations and that his opinions were knowledgeable and fair. She found him to be impartial and unbiased. Also reliable and sensible, she said, was Captain Simpson. She was entitled to reach these conclusions.

180. The other witness who gave evidence that in his opinion the CP channel was not a narrow channel was the 1st appellant. The Judge held that his view to that effect was unreasonable. I agree. I will however return to his position when considering the application of the proviso in his case.

181. Reliance was also placed upon the fact that the prosecution did not initially allege that this was a narrow channel case. That is true. The prosecution’s first case did not refer to the narrow channel rule at all. On the contrary it asserted that it was a head on case, presumably under rule 14, under which vessels meeting on reciprocal or nearly reciprocal courses so as to involve risk of collision must alter course to starboard. No proper thought could have been given to the case before this charge was drafted. Rule 14(b) expressly provides that such a situation shall be deemed to exist when a vessel sees the other vessel ahead or nearly ahead and by night she can see the masthead lights in a line or almost in a line and/or both side lights. In the present case there was no time when the vessels could see both the other’s side lights. As the plot in Annex 2 shows clearly, save perhaps at the last, the position throughout (assuming of course that they were looking) was that the N67 could see the green starboard light of the YH and the YH could see the red port light of the N67. They were crossing vessels not end on vessels. No competent mariner could have thought otherwise.

182. In my opinion the Judge was correct to hold that the CP channel was a narrow channel, essentially for the reasons she gave. I should add that I would reject the submission that the Judge employed a two stage test by first considering the physical criteria and then considering how seamen navigate the channel as a separate consideration only if the decision cannot be made on the basis of the physical criteria. There is some force in this point based on para 361 of the judgment. However, the judge held that the CP channel was a narrow channel based on all these considerations, but primarily on the physical characteristics, including of course the buoys. Moreover the reason why seamen treat it as a narrow channel is also, as I see it, based on those same characteristics. In any event the Judge held at para 387 that the channel was a narrow channel on the grounds, both of it being physically narrow geographically and bathymetrically, and of the manner that seamen in fact navigate in the channel. For the same reasons the Court of Appeal were correct to uphold her decision.

183. Those conclusions are consistent with the authorities as I read them. As stated above, the issue for decision in this case is whether a buoyed channel, which is accepted to be a channel, is a narrow channel. The authorities seem to me to support the proposition that the question whether it is or not is a question of fact. The answer depends upon considering a number of different factors. In particular it involves a detailed examination of the locus in quo. It thus includes a consideration of the nature of the channel, including its buoys (if as here it is a buoyed channel), the geography, topology and bathymetry of the area, the width of the channel, the size and characteristics of the vessels that use it and the manner in which seamen in fact navigate within the channel.

184. In order to be taken into account in what has been called a multi-factorial exercise, as indicated in para 11 above, the relevant information must be available to the pilots and masters so that they are all able to judge the true position based on the same information. Thus the relevant charts, sailing directions and local regulations must be consulted. It must be possible to reach a conclusion as to whether a particular area is a narrow channel from those sources.

185. There is no universal definition of a narrow channel. As para 6-228 of the 13th edition of Marsden on Collisions at Sea says, there is considerable difficulty in defining a narrow channel and a definition was deliberately avoided by IMO in 1973. It does not, however, follow from this that in the vast majority of cases it is not possible to recognize a narrow channel. Paras 6-228 and 6-229 give a large number of examples of areas which have and have not been held to be narrow channels. It is of little, if any, assistance to look at the cases because they turn on their own facts and very few of them contain statements of principle. In any event it is self-evident that no mariner has time to consult Marsden, let alone the cases, in deciding whether a particular area is a narrow channel.

186. Many of the cases were decided in England by Willmer J, later Willmer LJ and even later Sir Gordon Willmer, in the 1940s, 1950s, 1960s and 1970s: see for example The Crackshot (1948-49) 82 Ll L Rep 594, The Jaroslaw Dabrowski [1952] 2 Lloyd’s Rep 20 at 26, The Anna Salen [1954] 1 Lloyd’s Rep 475 at 487, The Sedgepool [1956] 2 Lloyd’s Rep 668 at 678 and The Koningin Juliana [1974] 2 Lloyd’s Rep 353 at 362. He regularly said that it was a question of fact in every case.

187. In The Jaroslaw Dabrowski Willmer J said that he had derived a certain amount of assistance from the decision of Langton J in The Varmdo [1940] P 15, where the judge had dealt with the issue of narrow channel as partly being one of evidence as to how seamen in fact navigate in the locality and partly as one for the advice of the Elder Brethren, which he said really amounted to the same thing. He said almost exactly the same in The Anna Salen and in The Sedgepool, where he said that he based his conclusion that the narrow channel rule was applicable partly on the basis of the evidence he had heard in the case and partly on the advice of the Elder Brethren. He added:

188. Those statements of principle were criticized on the basis that it is unsatisfactory to rely upon the practice of seamen because relative newcomers to a port might not know what that practice was. There is some theoretical force in that criticism but it seems to me to be largely a lawyer’s point because the advice of the Elder Brethren, who advise the judge on nautical matters in collision actions in England, will be based on the kind of assessment to which I have referred, namely the physical and geographical conditions and the information on the chart and in other publications. So, for example, in The Anna Salen Willmer J said at page 487 that the Elder Brethren advised him that it would be impossible to define what was the channel for the purpose of the narrow channel rule. On the facts here there is no such difficulty because of the buoys and there is I think no doubt what advice the Elder Brethren would have given.

189. It was submitted that the approach in the cases and adopted by the Judge and indeed in this judgment is far from satisfactory and should be reformulated or restated. It was submitted that the essence of any test for mariners should be objective certainty and that a mariner who navigates a stretch of water for the first or fiftieth time should be able to determine the application of rule 9 from objective criteria, which must in not only any geographical or bathymetric features but also the official designation of the channel by the relevant regulatory authorities. I agree.

190. However, it was further submitted that those criteria must include the absence of such an official designation (my emphasis). It was correctly submitted that MARDEP had not in fact designated the channel as a narrow channel. In para 372 the Judge referred to an assessment in 2004 of a proposal to declare the area a fairway. Based on evidence from Captain Loynd to the effect that MARDEP’s response to the proposal was that because the number of deep draught vessels using the deep water route was diminishing they did not feel it justified the expense and they intended to continue the system as it was. At para 374 the Judge said:

191. That paragraph was criticized on behalf of the 1st appellant. However, I see no ground for criticism, save that the CP channel was officially declared on the chart to be a channel. It was not declared as a narrow channel. I agree with the Judge that the fact the deep water channel was not gazetted as a narrow channel was of no relevance. As indicated above, there are many narrow channels around the world which are not declared or designated as narrow channels. In cases in which they are so declared or designated, the channel would undoubtedly be a narrow channel, as the Judge recognized, but the absence of gazetting is irrelevant, unless a particular area has been formally noted on the chart or otherwise publicly stated as not being a narrow channel, which was not the case here. Whether the channel is a narrow channel depends upon all the circumstances of the case as explained above.

192. It was submitted that, if in doubt a mariner such as the 1st appellant can contact MARDEP and that, if the 1st appellant had contacted the Director or other appropriate official at MARDEP he would have been told that MARDEP did not treat the CP channel as a narrow channel. It was submitted that in such circumstances the 1st appellant would have been bound, or at least entitled, to proceed on the basis that it was not a narrow channel. Although the judge did not admit it in evidence (on the ground that it was hearsay), we know from a letter dated 29 June 2009 that MARDEP would have replied that the CP buoys demarcate a deep water route and would have referred to the Pilot Book quoted above. They would have added that, as most vessels can navigate outside the above buoys, the area is not considered as a narrow channel. That opinion is irrelevant because it was not disseminated either to mariners generally or to the 1st appellant, but more importantly, as stated above, it is plainly wrong. A moment’s thought would lead to the conclusion, first that, as is common ground, the CP channel is a channel and, secondly, that it is a narrow channel within which vessels must pass port to port. That is so regardless of how vessels navigate outside the channel.

193. Reliance was placed on a number of articles, notably by Professor Craig H Allen, who has written widely and has advanced some criticism of the present COLREGS: see eg Taking Narrow Channel Collision Prevention Seriously to More Effectively Manage Marine Transportation System Risk, Journal of Maritime Law & Commerce, Vol 41, 1 January 2010. His article provides much food for thought and makes recommendations for the future. However, it does not seem to me to be inconsistent with any of the views I have expressed as to the correct approach to what is a narrow channel as things stand at present. It notes at page 7 that the principal factors that distinguish narrow channels and fairways from other waterways are their physical characteristics and usage. The problem is to deal with ships that pass frequently at close quarters. Professor Allen further notes at page 13 that pre-designation of narrow channels or fairways, where rule 9 applies, is the exception not the rule. He stresses the importance of certainty and makes an impressive case for the formal designation of water to which rule 9 applies in order to avoid confusion: see eg pages 34 ff. However, he recognizes that, given their international nature and the importance of obtaining wide international agreement to any changes, the COLREGS are unlikely to be altered in the near future. His plea is for the most part directed to the particular problems which arise in some of the great rivers in the United States. He would like to see the authorities on those rivers introduce both a programme of designation and more guidance. It is of interest to note that he concludes his article with the plea that any such guidance should caution mariners that if there is any doubt whether a waterway is a narrow channel or fairway under rule 9 they should assume that it is.

194. I see the force of the general points made by Professor Allen but I do not see how this Court can for the first time lay down principles of navigation different from those which have been accepted to date. It must approach them as mariners would do. For these reasons I would reject the submission that, in the absence of a published designation or declaration that a channel is a narrow channel, the channel should either be conclusively treated as not being a narrow channel or there should be a strong presumption to that effect.

195. Some reliance was placed upon The Koningin Juliana [1973] 2 Lloyd’s Rep 308 before Brandon J and in the Court of Appeal supra. This case was said to support the proposition that there cannot be a narrow channel within an area where other vessels navigate outside it. It was not, however, considering that question, but a different question, which was a question of fact. It was held that, on the facts, the narrow channel comprised the whole area of navigable water. Within that area was a dredged channel or fairway. The issue was what was meant by “the fair way or mid-channel” under the then COLREGS, rule 25(a) of which provided that a vessel navigating in a narrow channel must keep to the starboard side of “the fairway or mid-channel”. It was held that on the facts the duty was to keep to the starboard side of the dredged fairway and not of the wider narrow channel. See per Brandon J at page 313-4 and Willmer LJ at Page 362, where he said that in the comparatively rare case of a channel which does not include a defined fairway the vessels must keep to the starboard side of mid-channel.

196. As already indicated, the Judge was in my opinion entitled to accept the evidence of Captain Third that there are numerous ports in Europe where vessels navigate in water outside narrow channels. It would to my mind be astonishing if it were otherwise. There are cases in the materials before the Court where that was the position, namely The Gustafsberg [1905] P 10, The La Bretagne (1910) 179 F, 286, The Toluca [1981] 2 Lloyd’s Rep 548 and The Devotion and The Polydinamos [1993] 2 Lloyd’s Rep 464. In the Gustafsberg, for example, it was held that a buoyed channel was a narrow channel notwithstanding that one-third of the traffic to the north of the channel navigated outside it. However, these are no more than examples. I am unaware of a case in which it was argued that a narrow channel can only exist where it takes up the whole of the available water. In my opinion it is unarguable.

197. I note in passing that some reliance was placed upon the decision of Merriman P in The Kirsten Skou (1949-50) 83 Ll L Rep 279 where a gap between two wreck buoys was held not to be a narrow channel. As I read the judgment, the President held that it was not a channel at all, let alone a narrow channel, whereas here it is common ground that the CP channel is a channel and the only question is whether it is a narrow channel. The case contains no statement of principle which might support the 1st appellant’s case.

198. For these reasons I would hold that the Judge and the Court of Appeal were correct to hold that the CP channel is a narrow channel. Indeed, in my opinion there is no arguable case to the contrary. It was not reasonable for any seaman to hold the contrary view.

199. It was submitted on behalf of the 2nd appellant that the answer to the question certified as the second point of law, namely what are the legal principles for determining whether a particular channel is a narrow channel to which rule 9 applies is (a) that it is a question of fact whether a particular channel is a narrow channel and (b) a channel is not precluded from being a narrow channel simply because it lies within a greater expanse of navigable water to which rule 9 does not apply. I would accept that submission but would add two points. The first is that the question of fact involves a consideration of all the various factors identified above as touching upon the issue. The second is that, while an official designation of a narrow channel publicized to mariners would be conclusive, the absence of such a designation is irrelevant unless a particular area is formally noted on the chart or otherwise publicly stated as not being a narrow channel.

200. It is not in dispute that the vessels were aware of each other from about 2108 at the latest. It follows from the conclusion that the N67 was navigating in a narrow channel that it was her duty and that of her master who was navigating her to keep as close to the starboard side of the channel as was safe and practicable. It was thus his duty, if he wished to remain in the channel, to alter course to starboard at about 2109 in order to navigate close to the line between the Siu Mo To NE mark and buoy CP2. That was his duty regardless of the position of the outbound vessels, the YH and the BERLIOZ.

201. He was not bound by the crossing rule but by the narrow channel rule. At her para 401 the Judge quoted this passage from the judgment of Willmer J in The Empire Brent (1948) 81 Ll L Rep 306 at 312:

It was not I think suggested that the crossing rules applied to the N67 if she was navigating in a narrow channel. In any event the principles stated by Willmer J are plainly correct.

202. The Judge so held at paras 124 to 132. She was very critical of the 1st appellant’s navigation. She held at para 125 that vessels meeting each other in opposite directions at or near the entrance to the channel at buoys CP1 and CP2 should expect to pass port to port. As to the N67 she said this at paras 126 and 127:

203. At para 130 the Judge held that the N67 was not navigating on the starboard side of the channel as a narrow channel. She further expressly held that the 1st appellant was unreasonable in not believing that this is a narrow channel. As the plots show, the N67 took no action until less than a minute before the collision. During the whole of the time until then the vessels were closing at about 23 knots, or 2,300 feet a minute. The principal cause of the collision was the failure of the N67 to alter course to starboard at about 2109 or soon thereafter in order to make it clear that she expected to pass port to port at a safe distance when she was close to the southern side of the channel. The Judge held at para 169 that she should have altered about 12º to starboard.

204. There can be no doubt that in failing to comply with the narrow channel rule the 1st appellant caused his vessel and the lives of the crew to be endangered. The purpose of the COLREGS is to ensure, so far as possible, that vessels will pass safely without risk of collision. If the N67 had altered to starboard to proceed on her starboard side of the channel there can be no doubt that the vessels would have passed safely port to port within the channel in the vicinity of buoys CP1 and CP2.

205. The judge made further findings of fault against the 1st appellant. She found that he failed to keep a good lookout by radar, visually and orally. He was under a duty to keep a good lookout under rule 5 of the COLREGS, which provides:

The Judge held that the 1st appellant ought to have appreciated that the YH and the BERLIOZ were shaping to enter the channel. He could not reasonably have thought that the YH would alter course to starboard and leave the CP 1 buoy to port in the ordinary course of navigation. Equally he could not reasonably have thought that the YH would cross ahead and pass his vessel starboard to starboard. He was not keeping a radar lookout at all from about 21.11. The YH sounded one short blast on her whistle at 21.12.32 in order to indicate that she was altering her course to starboard. The 1st appellant did not hear it. She sounded two signals of five short blasts at 21.12.54 and 21.12.57. Such a signal is sounded to mean (under rule 34(b) of the COLREGS) that the YH failed to understand N67’s intentions or that she was in doubt whether sufficient action was being taken to avoid a collision. The 1st appellant did not hear either signal. It is not in dispute that they were sounded because they can be heard on the VDR recording. The Judge held at para 137 that it was shocking and unacceptable that the 1st appellant was, as she put it, unable to hear these audible whistle blasts when the vessels were at such close range. In short, on the findings of the Judge, the lookout kept by the 1st appellant was deplorable. The Judge summarized her conclusions in this respect in para 173 by holding that the 1st appellant incorrectly, erroneously and unreasonably believed that the YH was going to cross the N67’s bow and pass starboard to starboard. As a result he took no action until he went hard to port at the last.

206. Throughout this period the passing distance between the vessels was so small as to be unsafe. The Judge held at para 142 that from 2109 the CPA was always less than 100 metres, which was developing into a grave and real risk as the vessels approached. The Judge described the respective courses of the vessels in her paras 147 and 148. As can be seen on the plot at Annex 2, as they approached each other the N67 could see the green light of the YH and the YH could see the red light of the N67. As shown on Annexes 2 and 3, the N67 was on a steady course of between 090º and 092º to port of mid-channel from 2109 until about 21.12.50, when she went to port. The helm orders given on YH were 257º at 21.07.26, 260º at 21.08.32, 261º at 21.11.12 and 263º at 21.11.37, followed by starboard 10º at 21.12.25, starboard 20º at 21.12.32 and finally hard to starboard at 21.12.43. Until 21.12.25 the course alterations were not collision avoidance measures but were simply small alterations in order to take up a down-channel course by the time she reached the channel. However, they had the effect of reducing the CPA if the N67 took no action. The helm orders of 10º to starboard and 20º to starboard followed by hard to starboard were measures taken in an attempt to avoid collision. Neither vessel reduced speed.

207. I will return below to the faults found against the YH and hence the 2nd appellant. As to the 1st appellant, the Judge recognized that the hard to port order was given too late but blamed him for giving it. However, as I see it, whether or not the 1st appellant was to blame for going to port at the last, the critical case established against him was his unreasonable failure to appreciate that the CP channel was a narrow channel, his insistence on maintaining his course until the last and his consequent failure to take starboard helm action at any stage. I should add that the Judge held that the 1st appellant was also in breach of rules 7 and 8 of COLREGS, which are set out in Annex 1. He failed to use all available means to determine if risk of collision existed and failed to take positive action in ample time with the due regard to the observance of good seamanship. These faults were all bound up with his essential fault, which was that he failed to alter course to starboard at any stage pursuant to his duty to keep to the starboard side of the channel under rule 9.

208. In order to determine the appeal of the 1st appellant it is not necessary to determine any of the other certified issues of law because, apart from issue 3, which affects the navigation of the YH and to which I will return, the other issues, with the possible exception of issues 5 and 6, are relevant only if the CP channel is not a narrow channel. In these circumstances I will mention only issues 4, 5 and 6.

209. Issue 4 asks whether there is a rule of good seamanship which requires a vessel navigating in an IALA buoyed channel which is not a narrow channel to keep to the starboard side of the channel. The Judge answered this question in the affirmative on the basis of the evidence of Captain Third and Captain Simpson. Given my conclusion that the Judge and the Court of Appeal were correct to hold that the CP channel is a narrow channel, it is not strictly necessary to express a view on this point but I will only say that their evidence and the Judge’s conclusion seem to me to make very good sense.

210. Issues 5 and 6 are not relevant. There is no arguable possibility that the N67 had a right of way over vessels entering the channel. It is correctly recognized by the 1st appellant in his case that issue 5 is irrelevant. Equally nobody suggested that the result of these appeals could be affected by the answer to the question posed by issue 6.

211. Mr Justice Ribeiro PJ has carried out a detailed analysis of the application of mens rea to the offence or offences created by section 72 of the Shipping and Port Control Ordinance (Cap 313), with which I entirely agree. On the basis of that analysis he has concluded that in the case of each appellant the appeal must succeed unless the proviso applies. The case of each appellant must be considered separately. I agree with Mr Justice Ribeiro PJ that, under the proviso, the question is whether, if the Judge had applied the correct test as now formulated by this Court, she would inevitably have convicted the appellant. Mr Justice Ribeiro PJ summarises his position at para 122 as follows :

I am content to assume that the 1st appellant can satisfy sub para (a). The question is therefore whether the prosecution has established beyond reasonable doubt that the 1st appellant acted or omitted to act, either without holding the belief that his conduct was not such as to cause danger to the safety of others, or that his belief, although honestly held, was not based on reasonable grounds.

212. I have already expressed the view that there can be no doubt that the 1st appellant’s breach of duty in failing to navigate on the starboard side of the CP channel in breach of rule 9 endangered the N67 and the lives of those on board. It thus appears to me that the question to be asked boils down to this. Have the prosecution established beyond reasonable doubt that the belief of the 1st appellant that the CP channel was not a narrow channel was a belief which was not held on reasonable grounds? The Judge held that he believed that the CP channel was not a narrow channel and that it was for that reason that he did not alter course to starboard as he should have done.

213. It was submitted on behalf of the prosecution that, in the course her judgment, the Judge considered this very question. It was correctly accepted that the Judge was not considering the reasonableness of the 1st appellant’s belief for the purposes of applying the principles of mens rea which Mr Justice Ribeiro PJ has explained apply to a prosecution under section 72. It was submitted on behalf of the 1st appellant that, that being the case, it would be wrong in principle, and in any event wrong on the facts, for this Court to apply the proviso.

214. It appears to me that the Court must consider with care the findings of the Judge in this regard. I do not think that the mere fact that the Judge was considering the subjective belief of the 1st appellant and whether he had reasonable grounds for that belief for one purpose should lead, as a matter of law, to the conclusion that her conclusions are irrelevant for a different purpose. If the question is the same for both purposes and there is no flaw in the Judge’s reasoning, the Court should accept it in both cases.

215. What then were the questions considered by the Judge? Between paras 369 and 386 the Judge considered the question whether the CP channel is a narrow channel. She concluded at para 386 that the overwhelming evidence was that vessels heading in opposite directions would keep to the starboard side of the buoyed channel or just outside the channel north of CP1 going west and that vessels going east would keep to the starboard side of the channel. I have considered that question in some detail between paras 151 and 197 above and have already expressed my agreement with her conclusion at para 387 that the buoyed channel was a narrow channel on the grounds both of it being physically narrow geographically and bathymetrically and of the manner that seamen in fact navigate in the channel. In doing so I have referred to most of the findings made by the Judge: see in particular paras 158 to 182. I have reached the same conclusion as the Judge for essentially the same reasons. Indeed, as stated at para 198, I have expressed my own view that no reasonable mariner could consider that the CP channel is not a narrow channel. I will not of course repeat my reasoning in this part of the judgment.

216. At paras 388 to 398 the Judge considered the 1st appellant’s evidence that the narrow channel does not extend beyond the buoyed channel so that, if the CP channel is a narrow channel, it does not extend to the east of buoys CP1 and CP2. His evidence was that it followed that, on the facts, the N67 was the only vessel in the channel and that it was permissible for her to proceed down the middle of the N67, the vessels were crossing vessels and, again, that his duty was to maintain his course and speed. His evidence was that the vessel outside the channel should act cautiously and that the YH unreasonably entered the channel and hampered the navigation of the N67 which was in the channel. She should have stayed outside the channel until the N67 passed buoys CP1 and CP2. The Judge rejected that evidence.

217. Having referred to The Empire Brent supra, which on its face applies to vessels in the channel, the Judge said at para 403, basing herself on the evidence of Captain Simpson, that vessels approaching the channel and vessels within it must acknowledge that there must be some continuity so that rule 9 applied at the entrance to the channel in order to avoid chaos and ambiguity. She referred to the decision of Hewson J in The Canberra Star [1962] 1 Lloyd’s Rep 24, where he said that, if a vessel is entering a channel to proceed along it in circumstances where there is another vessel already in the channel the crossing rule does not apply. He said this at page 28:

218. That seems good sense to me. At paras 406 and 407 the Judge held that the N67 should not have assumed that the YH (and the BERLIOZ) would leave the channel but should have expected them to enter the channel on their starboard side of the channel close to buoy CP1. She should therefore have altered to starboard on to her starboard side of the channel in order to facilitate that manoeuvre. At paras 408 to 414 the Judge gave her reasons for criticizing the N67 for hogging the centre of the channel. All those conclusions were, in my judgment, justified.

219. The Judge then turned to the case of each individual defendant. At paras 416 and 417 she set out the 1st appellant’s experience, which was considerable. She continued in paras 419 to 423 as follows:

220. The Judge then gave a detailed account of his evidence, including (at para 444) his belief that the YH would go to the north of buoy CP1. In the next section of the judgment the Judge first considered whether it was reasonable for the 1st appellant to think that the CP channel was not a narrow channel. As I see it, that is to ask the same question as to ask whether there were reasonable grounds for believing that it was not a narrow channel, which is of course the critical question for the purposes of the proviso. The Judge expressed her view concisely at para 451:

221. Between paras 452 and 467 the Judge considered whether it was unreasonable for the 1st appellant to believe that the crossing rule applied and held that it was not. As I see it, if it was unreasonable for him to believe that the narrow channel rule did not apply, it was unreasonable of him to think that the crossing rule applied to the N67. As already stated, the narrow channel rule applies to all ships and, where the rule applies, there can be no reasonable basis for concluding that the stand on rule applies.

222. For the reasons I have already given, I have reached the clear view that it was unreasonable for the 1st appellant to believe that the narrow channel rule did not apply. The indicia are all one way. It is true that some evidence was given to the contrary. However I have considered that in detail. It was either given by experts who were biased or was given by those with no rational basis upon which to reach that conclusion. In particular the view expressed by Dr Li and others that there cannot be a narrow channel where vessels can navigate outside the channel is not based on any authority and, to my mind, is irrational. The whole point of a channel marked by buoys in which there are port and starboard hand buoys is that vessels inside the channel proceeding along it will pass port to port. Where, for example, there is a narrow dredged channel to enable deep-draughted vessels to proceed through it, again the whole point of such a channel is to ensure that vessels pass port to port. The fact, as may often be the case, that smaller vessels can navigate outside the channel is surely wholly irrelevant to the question whether the channel is a channel in the first example or a narrow channel in the second example.

223. It is true that the judge does not explain why she asked the question whether the 1st appellant reasonably believed that the CP channel was not a narrow channel. The most likely explanation seems to me to be that she thought it was relevant to the defence of reasonable excuse under section 72 or possibly to the defence of reasonable precautions under section 10(3), which she thought were in essence the same defence. However, whatever the reason, the fact is that the judge considered with great care the reasonableness of the 1st appellant’s belief. Moreover she arrived at what in my opinion were correct conclusions, which were of course arrived at by applying the criminal standard of proof, namely beyond reasonable doubt.

224. Counsel were not able to suggest any distinction between the issue which the Judge considered and the question which is relevant to the application of the proviso. It was not convincingly suggested that the trial might have taken a different course if the correct mens rea had been identified from the outset. I am satisfied, for the reasons given by the Judge, that the prosecution has established beyond reasonable doubt that the 1st appellant had no reasonable grounds for his belief that the CP channel was not a narrow channel and that the crossing rule applied. It follows that he had no reasonable grounds for believing that he was navigating his vessel safely and without danger to her crew. In these circumstances, if the Judge had directed herself correctly as to mens rea she would inevitably have convicted the 1st appellant. For these reasons I would dismiss the 1st appellant’s appeal against conviction.

225. I have already concluded that the CP channel is a narrow channel. I have also expressed the view in para 217 that vessels approaching a narrow channel and intending to proceed along it are not bound by the crossing rule but must enter the channel and, as they do so, keep as near to the starboard side as is safe and practicable in accordance with rule 9. It seems to me to follow that a vessel shaping to enter the channel should, as a matter of good seamanship, navigate in such a manner that, when she reaches the channel, she is on the starboard side of the channel in accordance with rule 9. The Judge in effect so held in paras 400 to 407 and was correct to do so. This approach is consistent with that espoused by the Court of Appeal in England in The Kaiser Wilhelm Der Grosse [1907] P 259 at 264, where Lord Alverstone CJ said:

226. As is plain from the plots in Annexes 1 and 2, the YH in fact navigated in such a way as to shape to enter the channel in a correct position on her starboard side of the channel near buoy CP1. She was under a duty to keep a good lookout and thus to watch the approach of the N67 carefully. Other things being equal, as a vessel drawing 12 metres, she was entitled to navigate in the channel and would be expected to do so. The Judge correctly so held. The 2nd appellant was aware of the N67 from about 2108. I have set out the courses steered by the YH at para 66 above. The Judge analysed the thought processes of the 2nd appellant in some detail between paras 603 and 608. In addition to the other steps taken on board the YH referred to above, she gave two long flashes on her Aldis light at about 2109 in order to alert the N67, although they were not seen by her. The 2nd appellant recognized that, unless one of the vessels altered course, there was going to be a collision or a near miss.

227. The 2nd appellant assumed, he said confidently, that after passing the restricted area, the N67 would move to starboard, although he also thought that it was a possibility that she would alter course to port out of the channel The Judge thought that to be unreasonable: see para 615. However, the Judge did not hold that the YH was at fault for not leaving the channel. The 2nd appellant and those on the bridge of the YH still thought that the N67 would alter to starboard. At 21.11.04 the 2nd appellant is recorded as saying on the bridge: “he knows that I would turn for sure? He knows that I would turn for sure”. The 2nd appellant explained that he was thinking that he would enter the buoyed channel and then turn to starboard to take up an up-channel course. That would be, as he put it, a large turn to starboard and would show the N67 a red light. The Judge held at para 624 that at 21.11.05 the vessels were 1.07 miles apart.

228. Shortly after giving a helm order of 261º at 21.11.12, at 21.11.20 the 2nd appellant is recorded as saying that the N67 “doesn’t need to move right along the fairway. I personally don’t see the need”. He called VTC to tell them to tell the N67 to pass port to port. He gave the helm order of 263º at 21.11.37. There was one communication with MARDEP which was of no effect. He realized that he had a serious situation in front of him and ordered starboard 10º at 21.12.25 and starboard 20º at 21.12.32 giving one short blast on the whistle. At 21.12.43 he ordered hard to starboard and at about the same time the co-pilot sounded five short blasts on the whistle. Just over a minute later he sounded another five short blasts. Finally, at about 21.13.06 he saw both of N67’s side lights and then the red light disappeared. He concluded that N67 intended to cross his bow and ordered port 10º and then amidships at 21.13.26 in an attempt to minimize the damage. The collision occurred almost immediately afterwards and he ordered stop engines.

229. As the Judge held at para 631, the 2nd appellant accepted that he could have turned a little earlier at about 21.11.00 in order to pass between the N67 and the CP1 buoy. He also admitted that he should have ordered at least starboard 20º by 21.11.40 in order to show the N67 a red light. At paras 632 to 636 the Judge set out part of the evidence of Captain Third and Captain Simpson to the effect that by C-2 it was too late to go north of buoy CP1 and why going south of the buoy at C-2 was the preferred option.

230. The Judge analysed the 2nd appellant’s evidence and the action he should have taken in very great detail at paras 637 to 698. The Judge put the 2nd appellant’s failure to taken earlier action down to his failure to keep a proper lookout. As the Judge had put it at para 638, he did not appreciate the close CPA, he did not appreciate the close quarters he was getting himself into and, even though he was using his binoculars, he wholly misjudged what he was seeing. It was as a result that no avoiding action was taken until 21.12.25. If he had been keeping a proper lookout he would have taken action earlier to avoid a collision.

231. The Judge held at para 673 that the YH (and indeed the N67) should have sounded a signal of five short blasts at about C-3, which was 21.10.44. That failure was not however causative of the collision because the N67 did not hear any of her whistle signals. More importantly, the Judge held that the YH should have altered course to starboard earlier than she did. She held at para 685, under the heading “Finding” that, if the N67 was in the 2nd appellant’s view not behaving properly, the only sensible course was to keep well clear. He should have navigated on what he actually saw and not what he expected. The Judge concluded at para 687 that the YH should have turned to starboard after the first short blast, which was sounded at 21.12.32. She held that, once it was clear that the N67 was not responding to that signal, the YH should have made a larger alteration to starboard. The failure of the N67 to alter to starboard at that stage, as the Judge put it, clearly rebutted the assumption made on the YH that the N67 would obey the rules. The Judge held at para 688 that the actions taken to avoid collision by the 2nd appellant were made in a sort of panic. They were not made in ample time and were not alterations large enough to be readily apparent.

234. As is apparent from those paras, the Judge was making those particular findings in the context of the defences to section 72 and section 10(3). The question for determination under the proviso is whether the prosecution can establish beyond reasonable doubt, either that the 2nd appellant acted or omitted to act either without holding the belief that his conduct was not such as to cause danger to the safety of others or that his belief, although honestly held, was not based on reasonable grounds.

235. As explained above the only causative omission to act proved against the 2nd appellant was a failure to keep a good lookout and to take substantial action to avoid collision about a minute earlier than he did. Have the prosecution established beyond reasonable doubt that the belief of the 2nd appellant that it was safe to wait so long was not held on reasonable grounds? I have reached the conclusion that they have not. Although the Judge is critical of the 2nd appellant’s lookout and failure to take action, sometimes in quite strong terms, his position was quite different from that of the 1st appellant. The 1st appellant was responsible for creating the risk of collision. The problem facing the YH and her pilot, the 2nd appellant, was how to react. He naturally expected the N67 to alter course to starboard at any time. The mistake he made was to fail to assess the position with sufficient accuracy and to wait too long before taking sufficient avoidance action. It can fairly be said that, in doing so he contributed to the endangerment of those on board both vessels, including the N67. It does not follow from the fact that the Judge held him to be at fault for breaches of the COLREGS that he did not believe on reasonable grounds that he was not endangering the lives or physical safety of those on board the vessels.

236. I agree with the conclusion expressed by Mr Justice Ribeiro PJ at paras L5 134 and 136, and L6 138 as follows. By contrast with her findings against the 1st appellant, the Judge’s findings critical of the 2nd appellant focus, not on the unreasonableness of his subjective beliefs, but on his conduct falling short of objective standards of good seamanship. The collision avoidance action he took was too little too late. His navigation was conditioned throughout by his honest and reasonable, albeit mistaken, belief that the N67 would alter her course to starboard. In these circumstances, I am far from persuaded that the Judge would inevitably have convicted the 2nd appellant if she had directed herself correctly on the issue of mens rea. It follows that, in agreement with Mr Justice Ribeiro PJ and essentially for the same reasons, I would not apply the proviso in his case. I would allow his appeal and quash his conviction.

237. For these reasons I would dismiss the appeal of the 1st appellant and allow the appeal of the 2nd appellant.

238. By way of postscript I would like to associate myself with Mr Justice Ribeiro PJ’s comments on the judgment of the Judge. Although we have not agreed with all her conclusions and I say nothing one way or the other about the parts of the case upon which I have expressed no view, I agree with him that her judgment is highly impressive for its comprehensive analysis of the evidence and the meticulous and detailed findings made. I can also see that it must have been quite an ordeal to try the case.

(a) These Rules shall apply to all vessels upon the high seas and in all waters connected therewith navigable by seagoing vessels.

(a) Nothing in these Rules shall exonerate any vessel or the owner, master or crew thereof, from the consequences of any neglect to comply with these Rules or of the neglect of any precaution which may be required by the ordinary practice of seamen, or by the special circumstances of the case.

(b) In construing and complying with these Rules due regard shall be had to all dangers of navigation and collision and to any special circumstances, including the limitations of the vessels involved, which may make a departure from these Rules necessary to avoid immediate danger.

Every vessel shall at all times maintain a proper look-out by sight and hearing as well as by all available means appropriate in the prevailing circumstances and conditions so as to make a full appraisal of the situation and of the risk of collision.

(a) Every vessel shall use all available means appropriate to the prevailing circumstances and conditions to determine if risk of collision exists. If there is any doubt such risk shall be deemed to exist.

(b) Proper use shall be made of radar equipment if fitted and operational, including long-range scanning to obtain early warning of risk of collision and radar plotting or equivalent systematic observation of detected objects.

(d) In determining if risk of collision exists the following considerations shall be among those taken into account:

(a) Any action to avoid collision shall, if the circumstances of the case admit, be positive, made in ample time and with due regard to the observance of good seamanship.

(b) Any alteration of course and/or speed to avoid collision shall, if the circumstances of the case admit, be large enough to be readily apparent to another vessel observing visually or by radar; a succession of small alterations of course and/or speed should be avoided.

(c) If there is sufficient sea-room, alteration of course alone may be the most effective action to avoid a close-quarters situation provided that it is made in good time, is substantial and does not result in another close-quarters situation.

(d) Action taken to avoid collision with another vessel shall be such as to result in passing at a safe distance. The effectiveness of the action shall be carefully checked until the other vessel is finally past and clear.

(f) (i) A vessel which, by any of these Rules, is required not to impede the passage or safe passage of another vessel shall, when required by the circumstances of the case, take early action to allow sufficient sea-room for the safe passage of the other vessel.

(a) A vessel proceeding along the course of a narrow channel or fairway shall keep as near to the outer limit of the channel or fairway which lies on her starboard side as is safe and practicable.

(a) When 2 power-driven vessels are meeting on reciprocal or nearly reciprocal courses so as to involve risk of collision each shall alter her course to starboard so that each shall pass on the port side of the other.

(b) Such a situation shall be deemed to exist when a vessel sees the other ahead or nearly ahead and by night she could see the masthead lights of the other in a line or nearly in a line and/or both sidelights and by day she observes the corresponding aspect of the other vessel.

(c) When a vessel is in any doubt as to whether such a situation exists she shall assume that it does exist and act accordingly

When 2 power-driven vessels are crossing so as to involve risk of collision, the vessel which has the other on her own starboard side shall keep out of the way and shall, if the circumstances of the case admit, avoid crossing ahead of the other vessel.

Every vessel which is directed to keep out of the way of another vessel shall, so far as possible, take early and substantial action to keep well clear.

(a) (i) Where one of 2 vessels is to keep out of the way the other shall keep her course and speed.

(ii) The latter vessel may however take action to avoid collision by her manoeuvre alone, as soon as it becomes apparent to her that the vessel required to keep out of the way is not taking appropriate action in compliance with these Rules.

(b) When, from any cause, the vessel required to keep her course and speed finds herself so close that collision cannot be avoided by the action of the give-way vessel alone, she shall take such action as will best aid to avoid collision.

(c) A power-driven vessel which takes action in a crossing situation in accordance with paragraph (a)(ii) of this Rule to avoid collision with another power-driven vessel shall, if the circumstances of the case admit, not alter course to port for a vessel on her own port side

(d) This Rule does not relieve the give-way vessel of her obligation to keep out of the way.

(a) When vessels are in sight of one another, a power driven vessel underway, when manoeuvring as authorized by these Rules, shall indicate that manoeuvre by the following signals of the whistle :

(d) When vessels in sight of one another are approaching each other and from any cause either vessel fails to understand the intentions or actions of the other, or is in doubt whether sufficient action is being taken by the other to avoid collision, the vessel in doubt shall immediately indicate such doubt by giving five short and rapid blasts on the whistle. Such signal may be supplemented by a light signal of at least 5 short and rapid flashes.”

[1] Hong Kong Court of Final Appeal Ordinance (Cap 484).

[2] Appearing with Mr Timothy Brenton QC on the 2nd appellant’s behalf.

[3] In Section G.2 below.

[4] Reasons for Verdict (“RV”)§§786 and 787. The Court of Appeal appears to have approached both convictions on the footing that they arose solely under the unlawful act limb: CA§§5-7, 494.

[5] On the grounds that section 72 was too uncertain and that it offended the presumption of innocence by placing the burden of showing reasonable excuse on the accused, see CA§§82-83.

[6] Set out in full at RV§84.

[7] As made applicable to vessels by the Merchant Shipping (Safety) (Signals of Distress and Prevention of Collisions) Regulations Cap (369N).

[8] Contrary to paragraphs (a), (b), (c) and (d) respectively.

[9] Contrary to rule 8(f)(i).

[10] Contrary to rule 8(f)(iii).

[11] RV§§18-21.

[12] RV§21.

[13] Ruling §70, cited in CA§88.

[14] RV§§24-27. The prosecution had conceded that section 10(4) imposes merely an evidential burden on the defendant and that it had the legal burden of negating the defence if that evidential burden was discharged.

[15] RV§§28-30.

[16] RV§550.

[17] RV§698.

[18] See CA§88.

[19] Shipping and Port Control (Amendment) Ordinance 1979, Ord No 2 of 1979.

[20] CA§109.

[21] CA§§113 and 115.

[22] CA§392.

[23] Court of Appeal at §110.

[24] (2010) 13 HKCFAR 142.

[25] Hin Lin Yee at §§40-41.

[26] Hin Lin Yee at §42.

[27] Hin Lin Yee §44.

[28] Hin Lin Yee §§44-47.

[29] In Section I.

[30] [2007] 4 HKLRD 991 (Stock JA, McMahon and Lunn JJ, with McMahon J giving the judgment of the Court).

[31] RV§80.

[32] See CA§88.

[33] At §15.

[34] At §16.

[35] At §27.

[36] At §36.

[37] From 2 years’ to 17 months’ imprisonment.

[38] Section 72A: “Any person who, by any unlawful act or in any manner whatsoever without reasonable excuse, disables, abandons, scuttles or beaches any vessel within the waters of Hong Kong commits an offence...”

[39] The Court of Appeal took a similar view: CA§§100-101, 108.

[40] See CA§§84-87, 92-115.

[41] As section 26(4) of the Merchant Shipping Ordinance 1899.

[42] See CA§§98-99, 104-105, 111 and 113.

[43] At §141.

[44] Cited by the Court of Appeal at CA§88.

[45] CA§115.

[46] CA§98.

[47] In Section F.

[48] Cited by the Court of Appeal at CA§88.

[49] Court of Appeal §93.

[50] CA§97.

[51] (1846) 10 Jur 211.

[52] [1966] 3 All ER 618.

[53] [1961] AC 290.

[54] Which reversed that view by its section 8.

[55] Section 10(3): “If any of the collision regulations is contravened by a vessel, the owner of the vessel, the master and any person for the time being responsible for the conduct of the vessel shall each be guilty of an offence and liable to a fine of $20,000.”.

[56] Section 10(4): “It shall be a defence to a charge under subsection (3) for the person charged to prove that he took all reasonable precautions to prevent the contravention to which the charge relates.”

[57] Cap 245. Section 33(1): “Any person who, without lawful authority or reasonable excuse, has with him in any public place any offensive weapon shall be guilty of an offence and shall be sentenced, on summary conviction or conviction on indictment, in the manner specified in subsection (2).”

[58] R v Ng Wai Hung [1990] 1 HKC 43.

[59] Evans v Hughes 56 Cr App R 813 at 817.

[60] See Hong Kong Archbold 2013, §§25-124 to 25-126.

[61] Appearing with Mr Robert KY Lee and Ms Jasmine Ching for the respondent.

[62] Respondent’s printed case §44.

[63] See Section F above.

[64] Hin Lin Yee at §§89-91, 96(d), 164, 178 and 198(d).

[65] Cap 132.

[66] PHMSO, section 150, Schd 9 and Criminal Procedure Ordinance (Cap 221), Sch 8.

[67] By sections 70 and 71 of the PHMSO.

[68] Hin Lin Yee §19.

[69] At §28.

[70] Hin Lin Yee §§111-113.

[71] At §146-147.

[72] Hin Lin Yee at §148.

[73] CA§103-105.

[74] At §45.

[75] Section H above.

[76] Section J.1 above.

[77] Section I above.

[78] Section G.5b.

[79] Appearing for the 1st appellant with Mr Nigel Jacobs QC and Mr James McGowan.

[80] (2001) 4 HKCFAR 133 at 187-188.

[81] Cap 484.

[82] Cap 221.

[83] (2001) 4 HKCFAR 457 at §61.

[84] Yuen Kwai Choi v HKSAR (2003) 6 HKCFAR 113 at §54; Leung Fei Wah v HKSAR (2006) 9 HKCFAR 118 at §§17 and 23; Kissel v HKSAR (2010) 12 HKCFAR 27 at §170.

[85] Lord Clarke NPJ §§203, 206

[86] RV§§15, 66, 71, 358.

[87] RV§§13, 67

[88] RV§§205.

[89] RV§206.

[90] RV§§130, 451 and 467.

[91] RV§§411, 412, 466, 492 and 501.

[92] RV§127

[93] RV§517.

[94] RV§516.

[95] RV§§135 and 197.

[96] RV§137.

[97] RV§509.

[98] RV§11.

[99] RV§§158, 194, 358, 596.

[100] RV§§597, 653, 771.

[101] RV§624.

[102] RV§11, 12, 198, 612.

[103] RV§596.

[104] RV§§178-179.

[105] RV§§192, 604.

[106] RV§171.

[107] RV§§136, 172, 182, 203.

[108] RV§161.

[109] RV§651.

[110] RV§648.

[111] RV§386.

[112] RV§400.

[113] RV§§158, 409-411.

[114] RV§668.

[115] RV§§163-164, 652.

[116] RV§§644, 648

[117] RV§638.