HKSAR v. SHAM YING KIT

CACC 565/1999

IN THE HIGH COURT OF THE

HONG KONG SPECIAL ADMINISTRATIVE REGION

COURT OF APPEAL

CRIMINAL APPEAL NO. 565 OF 1999

(ON APPEAL FROM HCCC NO. 6 OF 1999)

____________

_______________

J U D G M E N T

_______________

Keith JA (giving the judgment of the Court):

Introduction

1. On the evening of 22 March 1998, a fight occurred between two groups of men at the Urban Council Market in Hung Hom. Two men were seriously injured, and a third man died a week later. Four of the men in the group who attacked them were arrested. In due course, they faced trial in the Court of First Instance on an indictment containing three counts. All four of them faced the same counts. Counts 1 and 2 charged them with wounding with intent the two men who had been injured. Count 3 charged them with the murder of the man who had died.

2. After a trial before Nguyen J and a jury which lasted a number of weeks, the four Defendants were all convicted on the two counts of wounding with intent. They were all acquitted on the third count of murder. However, the jury had been directed that if it acquitted any of the Defendants of murder, it was open to it to return verdicts of manslaughter against all or any of the Defendants, either on the basis that they had been provoked into killing the deceased or on the basis that they had lacked the intent to cause the deceased at least really serious bodily injury. The jury acquitted the 2nd, 3rd and 4th Defendants (who we shall refer to as D2, D3 and D4 for convenience) of manslaughter as well, but by a majority of 5:2 it convicted the 1st Defendant of manslaughter on the basis of lack of intent. He was sentenced to 4 years' imprisonment on each of the two counts of wounding with intent, and to 6 years' imprisonment for the offence of manslaughter, all these sentences to be served concurrently with each other, making 6 years' imprisonment in all.

3. The 1st Defendant now applies for leave to appeal against his conviction for the offence of manslaughter. The sole ground on which he relies is that the jury returned inconsistent verdicts. His conviction for the offence of manslaughter on the basis of lack of intent is said to be inconsistent with (i) his convictions for wounding with intent on counts 1 and 2, (ii) the convictions of D2, D3 and D4 for wounding with intent on counts 1 and 2, and (iii) the acquittal of D2, D3 and D4 of both murder and manslaughter. We shall refer to these alleged inconsistencies as (i), (ii) and (iii) respectively. There is no application by the 1st Defendant for leave to appeal against the sentences, nor any application by D2, D3 or D4. In the circumstances, we shall refer to the 1st Defendant from now on as the Applicant.

The evidence at the trial

4. As often happens, the attack which resulted in the death of the deceased started with an incident of the most trivial kind. The victims in the two counts of wounding with intent were PW1 and PW2. Together with the deceased and other friends of theirs, they found themselves in the same lift at the market as the four Defendants and others. There was some pushing and shoving in the lift. An exchange of words took place, and one group tried to stare the other out. No fighting between the two groups took place at that stage, and the two groups parted company.

5. The principal prosecution witness as to what had happened thereafter was PW4. He had been in the group which had included the Applicant and his friends. However the verdicts of the jury are viewed, it is plain that the jury must have accepted his evidence. That was conceded by Mr John McNamara for the Applicant. PW4's evidence was that the group which had included the Applicant and his friends had then gone to the Applicant's shop in the market. They had decided to teach the other group a lesson. The Applicant had distributed various weapons. The Applicant had armed himself with a beef knife which was about 2 feet long, he had provided D3 with a metal pipe, and he had provided D2 and D4 with a wooden pole each. They had gone in search of the group which had included PW1, PW2 and the deceased. They had eventually found the deceased's group. The deceased's group had been unarmed. The Applicant's group had then attacked the deceased's group, injuring PW1, PW2 and the deceased. PW4 had seen the Applicant chop one of the other group with a knife, he had seen the knife being held by the Applicant when the deceased had fallen to the ground, and he had seen D2, D3 and D4 beat members of the other group with the pipe or pole which each had been carrying. After the attack, the Applicant and his group had returned to the Applicant's shop and disposed of their weapons.

6. Following his arrest, the Applicant was interviewed by police officers. He challenged the admissibility of the record of his interview, but following a voir dire the judge overruled the objection. The account which he gave in his interview was essentially in line with the evidence of PW4. In the course of that interview, he admitted having chopped one of the members of the other group. He did not give evidence in the trial, nor did any of the other Defendants.

Joint enterprise

7. The prosecution's case was that the four Defendants had been parties to a joint enterprise to inflict really serious bodily injury to the deceased and the other members of his group. However, there was some evidence that the other members of the Applicant's group might not have known that the weapon with which the Applicant had armed himself was a knife. In the course of his interview, the Applicant had said that no one had been close to him when he had collected the knife before the attack, and that no one had been with him when he had disposed of the knife after the attack. And when the Applicant was asked whether anyone had known that he had armed himself with a knife, he said:

"They were not very sure .... He thought that I went to take other things .... like poles."

The possibility that D2 and D4 might not have known that the Applicant had armed himself with a weapon as lethal as a knife was addressed by their counsel in their closing speeches.

8. This possibility gave rise to the need for the judge to give the jury a direction which reflected the principles referred to in R. v. Uddin [1999] 1 Cr App R 319 at pp. 328F-329F. The relevant principles for present purposes were these:

"(i) Where several persons join to attack a victim in circumstances which show that they intend to inflict serious harm and as a result of the attack the victim sustains fatal injury, they are jointly liable for murder; but if such injury inflicted with that intent is shown to have been caused solely by the actions of one participant of a type entirely different from actions which the others foresaw as part of the attack, only that participant is guilty of murder.

(ii) In deciding whether the actions are of such a different type, the use by that party of a weapon is a significant factor. If the character of the weapon, e.g. its propensity to cause death, is different from any weapon used or contemplated by the others, and if it is used with a specific intent to kill, the others are not responsible for the death unless it is proved that they knew or foresaw the likelihood of the use of such a weapon.

(iii) If some or all of the others are using weapons which could be regarded as equally likely to inflict fatal injury, the mere fact that a different weapon was used is immaterial.

(iv) If the jury conclude that the death of the victim was caused by the actions of one participant which can be said to be of a completely different type to those contemplated by the others, they are not to be regarded as parties to the death whether it amounts to murder or manslaughter. They may nevertheless be guilty of offences of wounding or inflicting grievous bodily harm with intent which they individually commit ....

(vii) The mere fact that by attacking the victim together each of them had the intention to inflict serious harm on the victim is insufficient to make them responsible for the death of the victim caused by the use of a lethal weapon used by one of the participants with the same or shared intention."

9. The judge gave the jury a direction on these lines. He provided them with written directions in which he emphasised the importance of a defendant (a) knowing that another member of his group had a lethal weapon and (b) anticipating that it might be used in the attack. He then said:

".... the prosecution must prove that the defendant knew that the other or others who had lethal weapons had the weapons before they used them, because it is that knowledge which forms the foundation for proving that he participated in the act of the others. If a defendant did not know that the others were armed with lethal weapons and the others used the weapons to kill or to cause really serious bodily harm, the others are acting on their own and the defendant is not responsible for their actions unless there was a specific agreement to kill or cause really serious bodily harm and the defendant was a party to it.

That is why I ask you, members of the jury, to firstly determine was there a specific agreement to kill or cause really serious bodily harm? If there was, was a particular defendant a party to that specific agreement? If there was not a specific agreement to kill or cause really serious bodily harm, or there was an agreement but the defendant was not a party to it, then you really have to look at what he himself did and the state of his knowledge. Because, as I said earlier on, if a person acting well beyond the agreement produced a knife or another lethal weapon and kills or injures somebody very seriously, the other people present at the scene would not be responsible for that person's actions."

A little later, he added:

"Further, since the unforeseen use of the lethal weapon would take the killing outside the scope of a joint enterprise, the defendant is not guilty of manslaughter either. In the case of the 1st and 2nd counts, the defendant will not be guilty either [of] wounding with intent, Section 17, or unlawful and malicious wounding, Section 19."

These directions are not as clear as they might have been, but the attention of the jury was nevertheless alerted to a distinction which could properly be drawn between the Applicant and the other Defendants.

The causation issue

10. The deceased had several chop wounds to his thigh. However, it was common ground at the trial that, despite the savagery of the chop wounds sustained by the deceased, they had not been the cause of his death. The wound from which the deceased had died was a haematoma which had formed as a result of a fracture to his skull. Although that had not been noticed by the doctor who had examined the deceased after he had been referred to the trauma unit at the hospital to which he had been taken, it had been noticed by the doctor who had first examined him in the Accident and Emergency Department and by the doctor who had performed a craniotomy on the deceased when his condition worsened two days after the attack. The fracture of the skull which gave rise to the haematoma could not have been caused by a chop. It could have been caused by a blow with a blunt object, such as a pole or a pipe, but it could also have been caused by a fall to the ground or by a bang on a wall. Moreover, the evidence of a professor of surgery called by the defence was that the fracture could have been caused by a light blow, i.e. one which did not necessarily mean that the person who had inflicted it had intended to cause the deceased really serious bodily injury.

11. In the light of this evidence, one of the issues in the trial related to causation, i.e. whether the death of the deceased had been caused by the act or acts of any of the Defendants. If the death of the deceased had not been caused by the act or acts of any of the Defendants, none of the Defendants could have been convicted of the murder of the deceased (or his manslaughter for that matter). However, it had been open to the jury to find that the wound from which the deceased had died had been caused either by a blow with a blunt object wielded by one of the Defendants or by a fall to the ground which had itself occurred as a result of the attack by the Defendants.

The alternatives left to the jury

12. In his summing-up, the judge directed the jury on various alternative verdicts which were open to it. If the jury was satisfied that any of the Defendants had been a party to the wounding of PW1 or PW2, but had not intended that PW1 or PW2 should suffer really serious bodily injury, the jury was directed that it was open to it to return verdicts of simple wounding as alternatives on counts 1 and 2. Again, if the jury was satisfied that any of the Defendants had been a party to the inflicting of the injuries on the deceased from which the deceased had died, but had not intended that the deceased should sustain really serious bodily injury, the jury was directed that it was open to it to return verdicts of manslaughter as alternatives on count 3. Finally, if the jury was satisfied that any of the Defendants had been a party to the inflicting of the injuries on the deceased from which the deceased had died, but had been provoked into attacking the deceased, the jury was directed that it was open to it to return verdicts of manslaughter as further alternatives on count 3.

13. In view of one of the arguments deployed by Mr McNamara, we should make one or two comments on these alternatives. First, at no stage did the prosecution ever put its case on the basis that the evidence justified leaving these alternatives to the jury. Secondly, the judge's decision to leave these alternatives to the jury had been preceded by a discussion between counsel and the judge of only the most cursory nature. There had been no discussion whatever as to whether the alternatives of simple wounding and manslaughter should be left to the jury on the basis of lack of intent. Indeed, when agreeing with the judge that the failure of the prosecution to satisfy the jury on the issue of causation would have to result in the acquittal of the Defendants on count 3, Mr McNamara pointed out to the judge that there was no alternative to count 3. It was "murder or nothing", said Mr McNamara, and the judge did not disagree with him. As for provocation, that had been raised only by counsel for D2. However, he had not decided yet whether he was going to address the jury on the issue of provocation, and there was no further discussion on the topic. Certainly, the judge did not inform counsel that he intended to leave provocation to the jury. In the event, counsel for D2 did address the jury on the issue of provocation, and the judge directed the jury as to the circumstances in law in which a verdict of manslaughter on the basis of provocation could be returned.

The alleged inconsistencies in the verdicts

14. Inconsistency (i). The Applicant was convicted of manslaughter on the basis that the jury was not sure that he had intended to cause the deceased really serious bodily injury. However, he was convicted on counts 1 and 2 on the basis that the jury was sure that he had intended to cause PW1 and PW2 really serious bodily injury. Mr McNamara contended that there was no rational basis on which the jury could have distinguished between what the Applicant had intended should happen to the deceased and what he had intended should happen to PW1 and PW2. At no time had the prosecution ever advanced such a basis for distinguishing between PW1, PW2 and the deceased, and such a distinction does not make much sense. If the Applicant had intended that really serious bodily injury should have been caused, that intention would have applied to the deceased as much as to PW1 and PW2. If the Applicant had not intended that such injury should be caused, that lack of intent would have applied to PW1 and PW2 as well as to the deceased. If followed that his conviction for the offence of manslaughter was inconsistent with his convictions on counts 1 and 2.

15. Inconsistency (ii). Since there was no evidence as to which of D2, D3 or D4 had caused the particular injuries sustained by PW1 and PW2, D2, D3 and D4 could only have been convicted on counts 1 and 2 on the basis that the jury was sure that they had been parties to a joint enterprise to attack the deceased's group and to cause them really serious bodily injury. Mr McNamara contended that it was difficult to see a rational basis on which the jury could have distinguished in this respect between the Applicant on the one hand and D2, D3 and D4 on the other. The Applicant would have been a party to the same joint enterprise as D2, D3 and D4, since he was convicted on counts 1 and 2, despite the absence of any evidence that he had himself caused the injuries sustained by PW1 and PW2. The jury's finding that the Applicant had lacked the intent to cause the deceased really serious bodily injury was therefore inconsistent with the Applicant's participation in a joint enterprise which had included an intent to cause the deceased and his group really serious bodily injury. It followed that the Applicant's conviction for the offence of manslaughter was inconsistent with the convictions of D2, D3 and D4 on counts 1 and 2.

16. Inconsistency (iii). Since there was no evidence as to which of the four Defendants or the people with them had caused the injuries sustained by PW1 and PW2, the four Defendants could only have been convicted on counts 1 and 2 if the jury was sure that they had been parties to a joint enterprise to attack the group of men which included PW1 and PW2. On the other hand, the fact that D2, D3 and D4 were acquitted of both the murder and the manslaughter of the deceased suggested, said Mr McNamara, that the jury was either not sure that they had been parties to the attack on the deceased or not sure that the injuries from which the deceased had died had been inflicted by anyone from the Applicant's group. That latter possibility had to be rejected in the light of the Applicant's conviction for the offence of manslaughter of the deceased, and that suggested that the jury acquitted D2, D3 and D4 of both the murder and the manslaughter of the deceased on the footing that it had not been sure that they had been parties to the attack on the deceased. However, it was difficult to see a rational basis on which the jury could have distinguished in this respect between PW1 and PW2 on the one hand and the deceased on the other. If the joint enterprise had been to attack the group of men which included PW1 and PW2, that would have included the deceased as well. It followed that the Applicant's conviction for the offence of manslaughter was inconsistent with the acquittals of D2, D3 and D4 of both murder and manslaughter.

The solution proposed on behalf of the Applicant

17. Inconsistencies in verdicts do not necessarily result in convictions based on those verdicts being quashed. That is because (once one leaves aside erroneous rulings of law and material irregularities in the course of the trial) an appeal against conviction can only be allowed, and the conviction can only be quashed, if the conviction is "unsafe or unsatisfactory": see section 83(1)(a) of the Criminal Procedure Ordinance (Cap. 221). Verdicts which are inconsistent with each other may, and usually will, result in one or more of the verdicts being set aside, but in the final analysis inconsistencies in verdicts are no more than a factor to be taken into account in determining whether the convictions based on them are unsafe or unsatisfactory.

18. Mr McNamara realistically recognised that the alleged inconsistencies in the present case should not result in all the convictions being quashed. He took on board the comments made by Auld J (as he then was) in delivering the judgment of the Court of Appeal in England in R. v. McKechnie (1992) 94 Cr App R 51 at pp. 61-62:

"In our view, it does not necessarily follow that inconsistency of verdicts requiring intervention by the court should result in the quashing of all guilty verdicts on one side or the other of the inconsistency. Where one verdict is readily explicable from the way in which the prosecution have put the case, from the evidence and from a proper direction from the judge on the law and facts, and where the second may well be attributable to the jury's attempt to do justice on the facts regardless of the constraints of the charges or to a failure by the judge to give adequate directions as to the consequences for the second verdict of a possible conclusion on the first, the court could properly form the view that only the second verdict should be quashed."

This passage was recently cited with approval and applied by the Court of Appeal in Hong Kong in HKSAR v. Ho Kin Sang (CACV 449/2000). Mr McNamara accepted that, on the footing that the approval of McKechnie in Ho Kin Sang cannot be criticised, this is not a case in which all the Applicant's convictions should be quashed. The critical question, he submitted, is which of the verdicts should be quashed.

19. Ho Kin Sang was a similar case to the present, though without the complication of multiple defendants. The Defendant had been charged with murder and wounding a second man with intent. He was acquitted of murder, but convicted of manslaughter and wounding with intent. The apparent inconsistency in these verdicts was cured by substituting a conviction for simple wounding for the conviction of wounding with intent.

20. Mr McNamara (who had appeared as counsel in Ho Kin Sang) contended that the solution which the court adopted in Ho Kin Sang could not be adopted in the present case. That was because in the present case the inconsistencies could only been cured by quashing the conviction for manslaughter. Substituting convictions for simple wounding for the convictions for wounding with intent would not cure the inconsistency between the conviction of the Applicant for manslaughter and the acquittals of D2, D3 and D4 of manslaughter (i.e. alleged inconsistency (iii)). In any event, the conviction of the Applicant for manslaughter was not, to use Auld J's words, "readily explicable from the way in which the prosecution .... put its case", because the prosecution had never suggested that a conviction for manslaughter was an appropriate alternative to a conviction for murder. The only verdicts which were explicable from the way in which the prosecution had put its case were the verdicts on counts 1 and 2.

21. We do not think that the fact that the prosecution never suggested that a conviction for manslaughter was an appropriate alternative to a conviction for murder helps in this context. The fact of the matter is that the judge left manslaughter to the jury, and the conviction of the Applicant for manslaughter is explicable to that extent. However, we agree with Mr McNamara that substituting convictions for simple wounding for the convictions for wounding with intent would not cure the alleged inconsistency between the conviction of the Applicant for manslaughter and the acquittals of D2, D3 and D4 of manslaughter (i.e. alleged inconsistency (iii)). The critical question is whether that alleged inconsistency, and the others to which we have referred, were in reality inconsistencies, or whether there is a rational and explicable basis for the jury returning the verdicts which it did. It is to that issue that we now turn, though we preface what we have to say by reminding ourselves that it is well settled that a conviction will be quashed on the ground of inconsistency with other verdicts if, and only if, the appellant can establish that the conclusion reached by the jury is one at which no reasonable jury who had applied their minds properly to the facts of the case could have arrived: R. v. Durante (1972) 56 Cr App R 708, approving what Devlin J (as he then was) had said in R. v. Stone (unreported), 13 December 1954, and adopted by the Court of Appeal in Hong Kong in R. v. Cheng Man To [1987] 2 HKC 261.

The possible reconciliation of the verdicts

22. Although Mr Andrew Bruce SC for the Government accepted that the Applicant's conviction for manslaughter was, on the face of it, inconsistent with his convictions on counts 1 and 2, he submitted that there was nevertheless a rational explanation for it. On the evidence, the jury would almost certainly have regarded the Applicant as the most culpable of all the Defendants. It had been to the Applicant's shop that the Applicant's group went to equip themselves for the attack, and it had been the Applicant who had actually distributed the weapons. The beef knife with which the Applicant armed himself was plainly a potentially lethal weapon, whereas the weapons which the others had were less obviously so. Finally, the Applicant was the oldest of the Defendants - indeed, at 22 he was a good deal older than D2 and D3 who were only 16. With his usual realism, Mr McNamara (who had appeared for the Applicant at his trial as well) did not challenge this view of the Applicant's role. He accepted that the evidence pointed "overwhelmingly" to the Applicant having been the ringleader of his group.

23. It is against that background that the judge's direction to the jury on the mens rea required for a conviction for murder has to be considered. The judge directed the jury that either an intent to kill or an intent to cause really serious bodily injury had to be proved on the count of murder. He then gave the jury a direction based on R. v. Nedrick [1986] 1 WLR 1025 in order to assist it in that task. He said:

"When determining whether the defendant had the necessary intent, ..... it may ..... be helpful for you to ask yourselves two questions. One, how probable was the consequence which resulted from the defendant's voluntary act? Two, did he foresee that consequence? If he did not appreciate that death or serious harm was likely to result from his act, he cannot have intended to bring it about. If he did appreciate but thought that the risk to which he was exposing the person killed was only slight, then it may be easy for you to conclude that he did not intend to bring about that result. On the other hand, if you are satisfied that at the material time the defendant recognised that death or serious harm would be virtually certain, barring some unforeseen intervention, to result from his voluntary act, then that is a fact from which you may find it easy to infer that he intended to kill or do serious bodily harm, even though he may not have had any desire to achieve that result."

24. The judge was, of course, required to direct the jury that an intent to cause really serious bodily injury had to be proved on counts 1 and 2. Unlike count 3 on which he had given the jury a Nedrick direction, the judge did not give the jury any assistance as to how it should approach that task on counts 1 and 2. That may have resulted in the jury treating the issue of intent on count 3 differently from the issue of intent on counts 1 and 2. The rationale underlying the Nedrick direction - that an awareness of the possible consequence of one's acts may make an intention to cause that consequence easier to infer - is relatively sophisticated, and its subtlety may be lost even on an intelligent jury. The jury may have thought that, to convict any of the Defendants of murder, they had to be sure, not merely that the Defendant had intended to cause at least really serious bodily injury to the deceased, but also that that Defendant had foreseen that at least really serious bodily injury would almost certainly result from the attack on him. The absence of a Nedrick direction on counts 1 and 2 meant that there was no possibility of the jury thinking that foreseeability (in addition to intent) had to be proved on those counts.

25. It is thus entirely possible that the judge's decisions (a) to give the Nedrick direction on count 3 and (b) not to give a corresponding direction on counts 1 and 2 account for the lack of intent which the jury found when acquitting the Applicant of murder but for the presence of intent when convicting him (and D2, D3 and D4) on counts 1 and 2 (i.e. alleged inconsistencies (i) and (ii)). Suppose that the jury had concluded that the fracture to the skull which had caused the haematoma from which the deceased had eventually died had occurred when he fell to the ground as a result of being chopped on the thigh by the Applicant. In the light of the Nedrick direction which the jury received on count 3, we can well understand the jury concluding that the Applicant would not have regarded (a) his attack on the deceased as virtually certain to result in the deceased banging his head on the ground, and (b) such bang on his head as virtually certain to result in the deceased's death. However, that conclusion would not have been inconsistent with (a) convicting the Applicant of manslaughter on the basis that the deceased had died as a result of the unlawful attack on him by the Applicant, or with (b) convicting the Applicant (and D2, D3 and D4) on counts 1 and 2 on the basis that they had been parties to a joint enterprise to attack the deceased and his group and to cause them really serious bodily injury, and that the injuries which PW1 and PW2 suffered were the result of actions which the Applicant (and D2, D3 and D4) had contemplated.

26. Nor do we see any inconsistency between the Applicant's conviction of manslaughter and the acquittal of D2, D3 and D4 of both murder and manslaughter (i.e. alleged inconsistency (iii)). We have already referred to the fact that in his interview the Applicant had said that the other members of his group might not have known that he had armed himself with a knife. Although there may have been a joint enterprise (to which D2, D3 and D4 were parties as well as the Applicant) to attack the deceased as well as PW1 and PW2, and to cause them really serious bodily injury, the jury might not have been satisfied that D2, D3 and D4 knew that the Applicant had armed himself with a knife, and might therefore not have been satisfied that the deceased's chop wounds (which caused him to fall to the ground and to bang his head with the result that he subsequently died) were the result of actions of the kind which D2, D3 and D4 had contemplated. Indeed, that was the basis on which the judge thought that the jury's verdicts were explicable. When, in the course of mitigation, Mr McNamara argued that the verdicts were inconsistent, the judge said:

"But they convicted your client of manslaughter on the basis that he was the person who wielded the knife and it was the knife which caused the more serious injuries to the deceased .... he was the one wielding the knife and therefore the one who must have known that the knife was used .... the acquittal of the 2nd, 3rd and 4th defendants can be explained, can it not, by the jury finding that because they did not have knives and they each respectively only had a weapon other than a knife, they could be acquitted because on their verdict they did not know that the 1st defendant had a knife and was going to use a knife to inflict some harm?"

The Applicant's explanation for the verdicts

27. Mr McNamara contended that a plausible scenario for the jury returning the verdicts which it did was that, although it was not sure that the attack upon the deceased had caused his death, it did not want the leader of the group which attacked the deceased and his friends to escape responsibility for the deceased's death. The jury therefore convicted the Applicant of manslaughter to give effect to its perception of what the justice of the case required. But that is to assume that there was no other rational explanation for the jury's verdicts, and that there was little or no evidence upon which the jury could reasonably have found that the death of the deceased had been caused by the acts of the Applicant. We have already dealt with the first assumption. As for the second, we have already referred to the fact that causation was an issue in the trial. The fact that the chop wounds were not themselves the cause of the deceased's death is not to the point, so long as it was open to the jury to find on the evidence that the deceased had died either from a blow with a weapon wielded by one of the Defendants, or from a bang to his head when he fell to the ground as a result of an attack on him by the Defendants. The jury was reminded of both of these possibilities by the judge in his summing-up:

".... the prosecution says one of two things happened, members of the jury. Either the fracture of the skull was caused by the assailants during the attack, or that as a result of the injuries sustained by the deceased, either during the attack or shortly after the attack, the deceased had a fall or hit his head against the wall or the ground which caused his skull to be fractured and the haematoma to develop.

If the fall or the hitting of the head against the wall or the ground occurred during or shortly after the attack, the prosecution says this was contributed to significantly by the acts of the defendants in causing the injuries to the deceased, the three chop wounds, and because of those three chop wounds, the deceased therefore had this fall or hit his head against the ground which caused his skull to be fractured."

28. The possibility of the deceased's head having hit the ground was addressed extensively by counsel, especially by Mr McNamara, in their closing speeches to the jury. Although it was conceded that that might have happened when the deceased fell during the fight, the defence suggested that there were other possibilities which had not been excluded. For example, there was evidence that the deceased had been or might have been the worse for drink, and there was a suggestion that he might have been struck by a taxi. In his closing speech, counsel for the prosecution highlighted such evidence as suggested that the deceased had not fallen prior to the assault and that nothing had happened to him on the way to or at the hospital to account for the fracture to his skull. Thus, the issue was fully before the jury, and if it excluded the other possibilities advanced by the defence for the fracture of the deceased's skull, as it was perfectly entitled to do, it was entitled reasonably to conclude that the assault on the deceased had caused him to fall, and that it had been that fall which had caused him to bang his head on the ground thereby sustaining the injury from which he eventually died.

29. We accept that, on more than one occasion, the judge told the jury that if it was not proved that the Defendants' acts had contributed significantly to the death of the deceased, then "the murder count immediately goes", and "the murder charge goes out of the window". He ought to have said, at every point at which he referred to the consequences of such uncertainty, that the Defendants could be guilty of neither murder nor manslaughter in such circumstances. However, we do not accept that the jury would have thought that causation was an issue which related to murder only. The judge gave to the jury a written direction on manslaughter, which informed it that if any of the Defendants were to be convicted of manslaughter, the prosecution had to prove that "the defendant committed an act or acts which caused the deceased's death". Moreover, towards the close of the summing-up, the jury was given a list of questions which it was to answer in order to arrive at the appropriate verdicts on count 3. Question 1 was:

"Are you sure that the defendant committed the act or acts which caused the deceased's death? If the answer is 'No' .... then find him not guilty. If the answer is 'Yes', .... then go to Question 2."

Question 2 dealt with self-defence. Question 3 referred to manslaughter, and question 4 to murder. The progression of questions made it clear that manslaughter only arose if causation had been proved, self-defence had been excluded, and intent had not been proved.

Conclusion

30. It follows that there was, in this case, little room for Mr McNamara's explanation for the verdicts of the jury. Not only did the judge's directions relate the issue of causation both to murder and manslaughter, but it was also open to the jury on the evidence to conclude that the Applicant had by his acts caused the death of the deceased. Since there was a rational explanation for all the jury's verdicts, we have not been persuaded that the verdict of manslaughter was one at which no reasonable jury could have arrived. We think that the questions raised on this appeal justify granting the Applicant leave to appeal against his conviction for manslaughter, but we dismiss the appeal.

The lessons to be learned

31. Before departing from this case, we wish to spell out some of the lessons to be learned from it. First, a Nedrick direction will usually not be necessary. In Nedrick, Lord Lane CJ himself said at p. 1028E that the direction should only be given "in the rare cases where the simple direction [on intent] is not enough". We do not think that this was an appropriate case for a Nedrick direction. Where a man arms himself with a knife with a view to attacking another, and then deliberately stabs or slashes him with it, cutting him in several places to the bone, directions about foreseeability are quite unrealistic. In any event, the Nedrick direction which the judge gave did not take into account the refinements to that direction suggested by Lord Steyn in R. v. Woollin [1999] 1 AC 82 at p. 96G-H.

32. Secondly, not every case of murder and wounding with intent requires the alternatives of manslaughter and simple wounding to be left to the jury. We do not say that these alternatives should not have been left to the jury in the present case, but we have already noted that the prosecution never put its case on the basis that the evidence justified leaving these alternatives to the jury. It is also worth repeating that if the judge proposes to leave these alternatives to the jury, he should discuss the matter with counsel in the absence of the jury beforehand. Not only will such a discussion make his decision to leave these alternatives to the jury a more informed one, but counsel will be able to tailor their closing addresses to the jury in the knowledge of the directions which the judge will be giving.

33. Finally, when a judge gives a direction about joint enterprise, it is sensible to explain to the jury the significance of such a direction. For example, the consequence of a finding by the jury in the present case that the Defendants had participated in a joint enterprise to attack the deceased's group using weapons which they had all known about would have been that it did not matter which of the Defendants had inflicted the injuries individually sustained by PW1, PW2 and the deceased. We think that the jury would have understood that from the tenor of the judge's summing-up, but it may have been better for the judge to have spelled that out.