TRAN DUC CUONG (VBP No.23461) and Another v. R

1993, No.424
(Criminal)

Headnote

Criminal law - Prosecution case based solely on identification evidence - Need for judge to give full Turnbull warning to jury on danger of misidentification : R. v. Beckford [1993]97 Cr.App.R. 409 referred to - Principles in applying the proviso to s83(1) Criminal Procedure Ordinance: (by a majority) the test formulated by Goff L.J. in R. v. Edwards [1983]77 Cr.App.R. 5 at 8 followed.

1993, No.424
(Criminal)

IN THE COURT OF APPEAL

_______________

_______________

J U D G M E N T

_______________

Litton, J.A.:

Introduction

1. Dang Cuong-anh, a young Vietnamese man aged about 18, was stabbed to death at Section 1, Whitehead Detention Centre, on the evening of 19 June 1991. A total of 19 wounds were inflicted, all over his body.

2. The applicants in this case (D1 and D2) are also young Vietnamese men and were, at the material time, also inmates of Whitehead Detention Centre, Section 1. They were known in the detention centre by their respective names of Mou and Keung (sometimes spelt as Cuong). They are cousins.

3. The prosecution evidence at the trial came from five eye-witnesses, whose accounts as to what had occurred, leading to the death of the victim, was not in any way challenged in cross-examination nor contradicted by other evidence. The issue at the trial was not as to how the victim came to be killed. The sole issue was whether D1 and D2 were the murderers.

4. Although the murder occurred at night, the areas in Section 1 where the attack first took place and was then pressed to its ultimate fatal conclusion were well-lit. The assailants were armed with home-made lethal weapons. There was no suggestion from any of the witnesses that the assailants had attempted to disguise their identities in any way. This was a cold-blooded attack on a defenceless victim, in the presence of a substantial number of persons. There were only two views the jury could have taken at the trial: (i) the assailants were wholly indifferent as to the consequences of their crime, or (ii) they were confident that the inmates would not reveal their identities to the authorities; perhaps because they felt sure the witnesses would be terrorised into silence. The latter is not an unreasonable inference to draw, because when the staff of the Correctional Services Department made their initial investigation, no one had gone forward to identify the assailants. The murder took place shortly after 9 pm. It was not until 12.45 am the following morning that D1 was detained and handed over to the senior officer at the detention centre for further investigation. This meant, of course, that if D1 was indeed one of the assailants, he had ample time to get rid of the items of clothing which might have contained traces of the victim's blood, and to wash himself. D2 was not detained until much later. This might explain how it was that, despite the brutality of the attack on the victim, the prosecution was unable to adduce any forensic evidence linking the applicants to the crime and had to rely solely upon the testimony of the identifying witnesses.

5. The atmosphere of fear prevailing at the detention centre might be gauged by the fact that, at the trial, which took place over two years later, some of the witnesses still expressed concern for their lives and those of their families. For instance PW4, Nguyen Van-noi (p125) said to the judge:

"I am coming here to testify the persons who were involved in the murder case and after testifying I want the court to protect my life".

6. Also PW7 Luu Cong-minh, after he had concluded his evidence, said (p259):

"At the moment I have received threats from the Mou's group of people who threatened my life and my family's life. After this trial I hope that the court can protect my life and my family's life".

The event

7. The account of the attack, leading to the victim's death, as given by the eye-witnesses was as follows. At about 9 pm on 19 June 1991 the victim was on his bed, No. 15A, in Hut 6A in Section 1. His cousin Nguyen Van Noi (PW4) was on the neighbouring bed, No. 14A. Both beds were on the top tier, separated by a gap. Two young men rushed into Hut 6A. One of them said "Here he is" and the other said "Stab him to death". PW4 knew one of the two young men as Mou, because he had worked with Mou in the cafeteria distributing meals. He did not know the other young man: but understood his name was Keung because of what he was told afterwards by the inmates of the hut.

8. PW4 said that Mou pulled two weapons out from a black shirt he was holding: a knife and a metal rod. Keung pulled similar weapons from his waist. Mou was wearing a dark coffee-coloured shirt and shorts, and a dark green cap. Keung was wearing a white shirt and dark shorts. Keung started climbing onto bed 15A by the ladder, and the victim went over to PW4's bed, walking across by means of a blackboard placed between the two beds. Mou then climbed onto PW4's-bed with his weapons. PW4 pulled the victim behind him to protect him from Mou but Keung had come over from bed No. 15A and stabbed the victim in his right thigh. Mou then stabbed the victim in his right ribs. The victim shouted and jumped down to the ground, fell, got up and ran to the lavatory pursued by the two assailants. Keung dropped his knife whilst jumping to the ground but he still had his sharpened metal rod. Mou had his two weapons. They pursued the victim into the lavatory and then about one minute later the victim emerged, covered with blood, pursued by the two assailants.

9. At an identification parade held on 25 June 1991 PW4 picked out D1 at the line out as the assailant Mou, and identified him again in giving evidence at the trial.

10. PW4's account of what occurred in Hut 6A was supported, to some extent, by the testimony of PW2 Madam Ha Thu Hang who, however, only saw the assailants climbing onto the top tier of the bed. She was so frightened that she climbed into her own bed on the lower tier and sat there. Later she identified the two applicants in an identification parade held at the police station as the two men who climbed up onto the victim's bed.

11. It appears from PW2's testimony that the two assailants were not total strangers to her. In her cross-examination, the following exchange took place:

"Q. .....before the incident, I wish you will describe to the court and his Lordship, you had seen the two men, had you, walking around in the camp, they were occupants of three camps?

A. They were living in the same section. We once in a while will see faces, but we don't see each other often."

12. As to what happened outside Hut 6A, the story was picked up by Nguyen Van Loi (PW5), an older man, a hut representative and member of the peace committee. He was an uncle of the victim and of PW4. He knew both D1 and D2 and knew them as respectively Mou and Keung. They all lived in Section 1 at Whitehead Detention Centre. As regards D1, PW5 knew him because they played soccer together in the playground and as regards D2, he too played soccer with him and PW5 said he had often seen him as they watched TV. Before the actual attack on the victim, there had been another incident involving D1 and D2. PW5 had noticed them during that incident, looking angry. PW5 had tried to get the inmates to return to their own huts. Then, a little later, when PW5 was standing between Huts 1A and 2B, he saw his nephew the victim running down the alley-way between the huts, pursued by Mou and Keung holding weapons. Mou was wearing a dark shirt and shorts, and Keung was wearing a white shirt and shorts. PW5 ran after them and shouted out: "Mou, don't hit my nephew". He tripped and lost one of his slippers and momentarily lost sight of them. When PW5 was at the junction of Huts 1B and 4A, he saw the victim kneeling down. Mou and Keung were attacking the victim whilst other people were watching. He saw one of the two applicants stab the victim once on the back, but he could not tell whether it was Mou or Keung who did the stabbing. There was a girl who tried to pull Mou and Keung away.

13. Later, at an identification parade held at the police station, he identified the applicants as the assailants and, in giving evidence, he identified them again in court: D1 as Mou and D2 as Keung.

14. PW5's version of events was supported by the testimony of Le Van- guy PW6 who had lived in Hut 6A in Section 1 Whitehead Detention Centre since late 1989. He was not related to the victim nor to the two previous witnesses. He had known D1 Mou for about two-and-a-half years prior to the incident, from the time when they were at the Argyle Detention Centre together, before transfer to Whitehead Detention Centre. He came to know D2 Keung after he was transferred to Whitehead. His testimony was to the effect that he saw the victim being pursued by Mou and Keung down the alley-way between the huts, with PW5 and a girl running after the assailants. He heard PW5 say: "Mou, this is my nephew". Mou and Keung had weapons in their hands. Mou was carrying a knife and a metal rod. Keung just had a metal rod. Mou was wearing a dark shirt and shorts; Keung was wearing a white shirt and shorts. He joined in the chase, but lost sight of them as they rounded the corner. The next thing he saw was the victim kneeling down, at the side of Hut 1A. Mou said to the victim: "You must kneel down to me" then stabbed him in the lower abdomen, followed by a thrust delivered by Keung. PW6 and others carried the victim to the clinic.

15. On 25 June 1991 he picked out D1 and D2 at a line-up of 14 persons at an identification parade as the two assailants. In giving evidence in court he identified D1 as Mou and D2 as Keung.

16. The last eye-witness to testify in court was PW7 Luu Cong-minh who had been at the Whitehead Detention Centre since February 1989 and lived in Hut 1A in Section 1. He was a barber at Section 1 and knew D1 Mou and D2 Keung because they had gone to him for hair cuts and had, on one occasion, gone together.

17. PW7 witnessed the last episode in the alleyway between Hut 1A and the United Nations Office. D1 Mou was standing facing the victim who was covered with blood. The victim was falling to the ground. Keung and Mou were pulled away by Mou's elder sister. PW7 helped in carrying the victim to the clinic.

18. At the identification parade on 25th June PW 7 also identified both D1 and D2 and repeated that identification in court.

19. As can be seen from the summary of the evidence above, the prosecution case against D1 and D2 was cumulatively overwhelming. As to the truthfulness or otherwise of the testimony of the four male eye-witnesses, this is what the judge said:

"Now you have heard evidence that NGUYEN Van-noi, NGUYEN Van-loi, LE Van-guy and LUU Cong-minh were all interviewed by officers of Correctional Services Department shortly after the incident. I think after the body of the deceased was carried to the clinic and they gave these officers statements. Now in those statements they all said that they did not witness the attack on the deceased and, of course, what they said in those statements was completely different from what they later told the police and that was a few days later and what they said subsequently in court both here and on the other occasion. Now this issue has been a heated point of contention by the defence. The explanation given by these witnesses was that at the time they were interviewed by Correctional Services Department Officers, they were in fear of revenge because a group of people whom they believed to be associated with the accused were present in the clinic and they also feared that the Vietnamese interpreter might disclose what they told Correctional Services Department Officers to the accused or their friends. It is, of course, entirely up to you whether you will accept their explanations. You have seen them in the witness box and observed their demeanour, and that is the way how they gave their evidence. It is for you to assess and evaluate their reliability. As I have said earlier, the credibility and the veracity of these witnesses are of the most crucial importance when you come to decide whether the accused are guilty or not guilty."

20. There is no complaint made of this passage in the summing-up.

The defence case

21. D1 Mou gave evidence at the trial and called one witness. The defence was an alibi. In essence, D1's case was that he and his girlfriend were in Hut 3A attending a celebration party. When after about 10 minutes they heard a disturbance going on outside, they decided to leave and went to the peace committee hut and remained there for about an hour or so. D1, his girlfriend and the alibi witness then left and D1 went back to his own Hut 10A. He never saw his cousin D2 that evening until he returned to Hut 10A. When he got to Hut 10A the door was closed but he called out to his cousin D2 who opened the door for him and let him in.

22. He denied that he had the nick-name Mou and when he was asked by his counsel: "Did anyone in the camp or in the section refer to you as Ah Mou?" his answer was: "I have no idea".

23. As regards the eye-witnesses, D1 said that he knew PW4 Noi who, he said, was related to PW2 Madam Ha. They were, he said, cousins. He also knew PW5 Loi and knew that all three Madam Ha, Noi and Loi came from Hanoi. When he was asked:

"Q. Well, how did you come to know that?

A. Because we have been living in the detention for quite some time, so we know a lot of things".

24. As regards PW6 Le Van-guy, D1 knew him as well. When asked: "In what connection or what relationship did you have with him?" D1's answer was:

"A. Because anyone who had lived in the same section for quite some time with me, I know exactly about that person and also their families. But for the other inmates who were just transferred in, I have no idea about these people.

Q. Did you know or hear about whether Le Van-guy had any kind of connection or relationship with the two men, Noi and Loi?

A. They are close friends, that is why they share the same hut."

25. D1 was then questioned about PW7 Luu Cong-minh who came from a different hut, No. 1A. D1 knew him as well. When asked as to what dealings he had with PW7, his answer was:

"A. We are not close friends. Once he suspected me of taking away his fan because he had a fan in his barber shop. He is a barber. The peace committee staff had once fled (?) him. As a result of that, we were fighting, we had a quarrel."

D1 was questioned further about this episode, and said that the matter was referred to the peace committee. He knew that PW7 Luu came from a "small town" which was part of Hanoi city. He also knew that PW7 Luu was friends with Noi and Loi.

26. D1 had in his own testimony denied that he had the nick-name Mou, and said that his nick-name was Luan. The following exchange in cross- examination took place:

"Q. But obviously you did not tell the people in the camp because that is not what they knew you as, isn't it?

A. Personally, no one ever called my nick-name, Mou. No one ever called myself Ah Mou. I never replied to that name.

Q. In fact, even your older sister referred to you as Ah Mou, didn't she?

A. That is her personal matter. I do not know whether she called me that nick name but many times I had told the Immigration Department my nick name is Luan and also my name, Kuong.

Q. You said it is a personal matter, but do you know why she would call you Mou, your sister?

........................

Q. Do you know, Mr. Tran, that your sister has referred to you as Mou?

A. She never referred to me as Ah Mou.

Q. Do you say that all the other witnesses we have heard from who called you Ah Mou are mistaken about your nick name, Mr. Tran?

A. That is from their personal point of view. I have no idea."

However, when his alibi witness Dinh Van Thang came to give evidence, he said that D1 had the nick-name Mou (p342-N). Moreover, he the witness normally addressed D1 as Mou (p347-1).

27. The testimony of the alibi witness, plainly contradicting D1 in this respect, must have had a most damaging effect on his case in the eyes of the jury.

28. As regards D2 Keung, he gave no evidence and called no witnesses.

The summing-up

29. As regards the identification parades held on 25 June 1991, when all five of the eye-witnesses except PW4 had identified D1 and D2 as the two assailants (PW4 had identified D1 only), the defence sought to minimize its impact by pointing out that out of the 14 persons at the line-out only D1 and D2 had come from section 1 at Whitehead Detention Centre and, as they were almost without exception persons known to the witnesses, or in the case of PW2 were persons whose faces were reasonably familiar to her, it was understandable that they would have been picked out at the parades. As to this, the judge said to the Jury:

"Now you have heard criticism from defence counsel about the identifications made in identification parades. I think both Mr MULLICK and Mr ADAMS very fairly and readily agreed that there was nothing wrong with the identification parades. They were properly conducted and the officers and especially the Superintendent who conducted these parades were following normal procedure to carry out the parades.

But what it comes down to, members of the jury, is this. At the end of the day it is for you to decide whether you believe these witnesses. If you do, if you are sure, if you are satisfied beyond reasonable doubt that they were telling the truth and they were not mistaken that these two men, the two accused, attacked the deceased and wounded Mr HOANG, then that is enough evidence to sustain convictions. The identification parades would be just formalities. It is a red herring. But that, of course, is a matter for you. You may think there is some importance. If you do think so, of course, please do take that into account."

In summing-up the main issue to the jury the judge said:

"It has been suggested by defence counsel that the evidence given by these witnesses was completely false. It was made up by them based on what they might have heard from other inmates and the allegations of these witnesses were completely without any foundation. It was further suggested that these witnesses had every reason and motive to tell lies, to tell false stories out of their desire to obtain refugee status and also revenge because they were either relatives or good friends of the deceased and they had a grudge against the accused. All these allegations have been denied by these witnesses on oath and I don't think there is any evidence to the contrary. Of course, they would like to have refugee status and they admitted it. But do you think that would be a reason that four or five people would conspire together and to give false evidence against two persons with whom they had very little to do? Mr. MADONNA has described this as a smoke screen put up by the defence to cloud the true issue and whether it is a smoke screen or not it is a -or if there is any substance, it is a matter entirely for you to decide. It is for you to decide whether these suggestions have any merits or substance in the light of the evidence that has been given before you."

Fabrication or mistake

30. The judge's statement: "At the end of the day it is for you to decide whether you believe these witnesses" (emphasis added) was too broad. It was suitable as regards PW4's identification of D1 as one of the two assailants, and PW5's and PW6's identifications of D1 and D2 as the assailants Mou and Keung. It was not wholly suitable as regards PW2's identifications of the two assailants, nor was it so as regards PW7's identification of D2 (though accurate regarding his identification of D1). As regards PW2 there was room for the possibility of her being truthful but mistaken, though less so as regards PW4's identification of D2: the assailant he thought was Keung had climbed over to his bed from the victim's bed and had stabbed the victim in the thigh whilst PW4 was shielding him; this was the person identified as having dropped the knife when he jumped from the bed in pursuit of the victim; and he was seen chasing the victim into the lavatory and out again. At the time all this was taking place Hut 6A was well-lit.

31. It is the law that when a case turns on identification the judge should (1) warn the jury of the special need for caution, (2) direct the jury to examine closely the circumstances of the identification and (3) remind the jury of any specific weaknesses in the identification evidence: see Scarman LJ in R. v. Keene [1977] Cr.App.R. 247 at 248, referred to in R. v. Beckford [1993] Cr.App.R. 409 at 415. As regards PW2 the judge did to some extent touch upon points (2) and (3) above in the course of his summary of PW2's evidence, but had given the jury no warning as such, beyond saying that "the witnesses could be mistaken in identifying these two persons". As regards PW4, the judge had a difficult task: the chances of PW4 having been honestly mistaken concerning his identification of D1 were slim: to have reminded the jury of the circumstances of his identification of D1 as one of the assailants who came onto his bed in Hut 6A and then suggested he might have been mistaken would simply have rubbed it in for the jury as to how favourable the circumstances were for a truthful and accurate identification.

32. The courts have laid down no universal rule as to how trial judges should warn the jury about the dangers of mis-identification. The Privy Council's example in R. v. Beckford [1993]97 Cr.App.R. 409 at 415 of "the workmate whom the [appellant] has known for 20 years" was not meant to be taken literally, and trial judges must exercise judgment as to when, for the purposes of a particular trial, the jury should be told that essentially the issue before them is credibility and not the possibility of an honest mistake. Above all, clarity in summing-up is important. To put alternatives to the jury - of an identification being willfully false, alternatively mistaken - could be highly confusing to the jury. Assuming that the judge's direction "the witnesses could be mistaken in identifying these two persons" was not a sufficient general warning regarding the danger of misidentification of the applicants by PW2 Madam Ha, and of D2 by PW7, the question then arises: Was this a serious misdirection? If the essential issue before the jury was nevertheless the truthfulness of PW4, PW5 and PW6 and as regards D1's identification, of PW7, how far was the judge's failure to underline the danger with examples a material irregularity at the trial? Or rendered the verdicts unsafe and unsatisfactory?

PW4

33. As regards PW4, he had not pointed the finger at D2, but in Mr. Mullick's cross-examination on D1's behalf, the case put was that PW4 bore a grudge against D1, and deliberately lied in purporting to identify D1: not that PW4 was mistaken, or could be mistaken. For the judge to have suggested mistake when the whole case for the defence was deliberate fabrication would have been to cut the ground to some extent from the defence's feet.

PW5

34. As regards PW5, it was put on D1's behalf that he bore "animosity, enmity towards Mou" (p169-Q) not that he was mistaken. On behalf of D2, it was suggested to PW5 that he had "named" D2 because he "wanted him out of the camp" and that he "told a pack of lies against D2" (p190-P): not that PW5 was mistaken.

PW6

35. As regards PW6, Mr. Mullick's first question on behalf of D1 was:

"Q. Witness, you had a very good view of the stabbing of the deceased by Ah Mou, is that right?

A. Yes"

(p202-D)

36. And later it was put by Mr. Mullick to the witness that he was "prepared to go along" with whatever PW4 and PW5 told him concerning the incident. The defence was that PW6's testimony was pure fabrication; not mistake, nor any possibility of mistake.

37. The same case was put on behalf of D2 by Mr. Adams. At p232-H, it was put as a fact that PW6 "and the people in Hut 6A knew quite well" that the assailants had climbed over the fence from another section of the camp. At p232-R there was the following:

"Q. Well, I suggest you never saw this attack at all. Yes or no?

A. Using which reason can you tell me that I never saw the attack?

Q. Based on your statement to C.S.D. Do you agree or not?

A. No, I disagree."

PW7

38. As regards PW7, it was put to this witness by Mr. Mullick for D1 that he bore a grudge against D1 because he PW7 had accused D1 of stealing two electric fans from him. D1's case was that PW7's testimony was a fabrication. To have suggested to the jury in the alternative that PW7 might have been mistaken would have been absurd.

39. But as regards D2, it was the case as put by his counsel that PW7 was mistaken. The passage in the cross-examination at p258 A-D proceeded thus:

"Q. I put it to you you weren't sure it was this Keung who was involved in the alley. You were not sure that it was this Keung in the dock now who was involved in the alley.

A. Because I did not see clearly whether he was carrying any weapons in his hands, so I cannot be sure.

Q. I suggest to you you couldn't see him clearly, couldn't see his face clearly.

A. No, I saw his face clearly."

Conclusion

40. As can be seen from the analysis of the evidence above, the judge's failure to give a warning of the dangers of wrongful identification could not have affected the jury's evaluation of the testimony of PW5 and PW6. Their evidence was crucial to the case, and if there was any doubt as to whether they had given truthful accounts - that is, if there was a possibility that they might have pointed the finger at D1 and D2 when they knew that the assailants were some other persons - then inevitably the jury must have acquitted. Any other verdict would have been perverse. Seen in this light the failure to give a detailed warning concerning the identification of the two applicants by PW2, and PW7's identification of D2, had in my judgment no impact on the outcome of this case.

41. The judge's paramount duty was to put the issues fairly and simply to the jury, avoiding if at all possible over-sophistication. The need for simplicity is of particular importance in this jurisdiction where many jurors are operating in an unfamiliar language. I am not satisfied that in the circumstances of this case the judge's failure to enlarge upon what he meant by saying "the witnesses could be mistaken in identifying these two persons" amounted to a serious misdirection. But assuming it were, this is plainly a case where the proviso to s83(1) of the Criminal Procedure Ordinance should apply. The jury could not have convicted if they had any doubts about the truthfulness of the witnesses. In these circumstances, and with a more elaborate warning concerning the dangers of misidentification, the jury must nevertheless have convicted. I should add that I have had the advantage of reading in draft Mr Justice Penlington's judgment and I agree with his adoption of the test formulated by Goff L.J. in R v. Edwards [1983]77 Cr.App.R. 5 at 8 on applying the proviso.

42. None of the other grounds of appeal put forward by counsel cause me any anxiety concerning the jury's verdict and it is not necessary to refer to them here.

43. I would allow the applications for leave and treating the hearing of the applications as the hearing of the appeal dismiss the appeal.

Penlington, J.A.:

44. There can be no doubt that following the decision in Turnbull there is a duty on a trial judge to give a general warning that the jury should regard identification evidence with caution and they should also be given the reasons why such a warning is required. That requirement exists even when the witnesses already know the person being identified was stated in Turnbull by Lord Widgery C.J. at 137:

"Recognition may be more reliable than identification of a stranger; but even when the witness is purporting to recognise someone who he knows, the jury should be reminded that mistakes in recognition of close relatives and friends are sometimes made".

That has now been reiterated in Beckford and Scott (1989) 89 Cr. App. Rep. 153. Such a warning and the reasons for it were not given here in clear terms, probably because the main thrust of the defence was an attack on the credibility of most of the prosecution witnesses, not their reliability, but Scott and Beckford

make it clear that even when a witness is well acquainted with the accused the warning should be given save in exceptional cases. That the evidence was for the main part recognition not identification is however relevant to the consideration of whether this court should nevertheless uphold the conviction pursuant to the proviso in Section 83(1) of the Criminal Procedure Ordinance and I adopt what Litton J.A. has said regarding the evidence given here by Nguyen Van-noi, Luu Long-minh, Nguyen Van-loi and Le Van-guy. All knew the accused quite well and their evidence was attacked, not on the basis that they could have been mistaken but because they had made statements after the incident in which they denied seeing the attack on the deceased, statements they went back on when later interviewed by the Police and when they gave evidence. The jury were given a full and clear direction on that which cannot be faulted in any way. Quite rightly they were told that "the credibility and veracity of these witnesses are of the most crucial importance".

45. There can however be no doubt that a Turnbull warning should have been given, particularly in respect of the evidence of Madame Ha Thu Hang who was the only witness against whom no attack was made as to her impartiality although even she was not identifying total strangers. That failure must constitute a material irregularity in the course of the trial and means that, pursuant to Section 83 of the Criminal Procedure Ordinance, the appeal must be allowed unless this court considers that no miscarriage of justice has actually occurred; is the conviction safe and satisfactory in the absence of the warning.

46. There is some conflict in the authorities on how we should consider the application of the proviso but in my opinion the correct principles are as set out in the 1994 edition of Archbold, Criminal Pleading, Evidence and Practice Vol. 1 at 7-113 to 117 and in the judgment of Robert Goff L.J. in R. v. Edwards 1983.77.Crim. App. Rep. 5. That was an appeal where there was a failure to direct the jury on a vital matter, the standard of proof. After referring to Stirland v. D.P.P. (1944) 30 Crim. App. Rep. 40, Robert Goff L.J. said at 8 "we shall ask ourselves whether, on the evidence, a reasonable jury properly directed on the standard of proof, would without doubt have convicted the appellant". Here that evidence was from eye witnesses, mostly challenged only as to their motives, not their powers of observation, who saw at close range, in excellent lighting conditions, events which constituted an execution and must have created a vivid impression. That evidence was supported by Madame Ha, who was not alleged to be biased, and who had seen the applicants before even though she did not know them well and is set out in the judgment of Litton J.A. The only gap in the Crown's evidence was the failure of PW4 Nguyen Van Noi, to identify D2. I have no doubt that a properly directed jury would have convicted both applicants. Bokhary J.A. in his judgment, which I have had the opportunity to consider in draft, is of the view that we should not speculate as to how the jury reached its decision - it may be that they only accepted the evidence of the other witnesses because it was supported by Madame Ha and we should disregard her evidence as it clearly required aTurnbull warning. I do not consider that is the correct approach. To apply the proviso the court must simply be satisfied, bearing in mind the defect in the summing-up, that on the evidence before them the verdict delivered by the jury would have been the same if a proper direction had been given.

47. We should look at all the evidence including that of Madame Ha and ask ourselves the question posed by Robert Goff L.J. in Edwards - would the jury inevitably have convicted if properly directed? Having studied that evidence in the light of the attacks made on it by the very experienced counsel who appeared for each applicant at the trial, I have no doubt whatever that the jury would inevitably have convicted both applicants. Reliance was placed by Mr. Mullick on R. v. Domican and the relevant passages of that decision have been cited by Bokhary J.A. There the question was whether there was sufficient evidence, apart from that of identification, on which the jury must have convicted. If the court there was saying that if a proper warning is not given the identification evidence must be totally disregarded in considering the proviso I would not accept that proposition. If the lack of such a warning means that such evidence must be totally disregarded, no matter how strong, that would of course be fatal to the convictions here but it would be contrary to what was said in Stirland and, more particularly, in Edwards. It would also, in my view, be contrary to common sense. If the identification evidence is not good the lack of a warning would preclude the application of the proviso but if, as here, it is very good indeed. I do not accept that the High Court of Australia was saying that, no matter how strong the identification evidence, in the absence of a Turnbull warning it must be disregarded when the application of the proviso is being considered. In Edwards there was a failure to give any direction on the standard of proof - a very serious defect which must of course affect all the evidence. Nevertheless the court there, having considered the facts before them, felt they could apply the proviso. Edwards was not referred to in Beckford but if that decision is to be taken as meaning that if an adequate direction is not given in relation to any part of the Crown's evidence such evidence must be totally disregarded when considering the proviso it would mean that Edwards was in effect overruled. I do not think such was the case.

48. Lord Widgery said in Turnbull at 139 in reference to the Devlin Report of April 1976:

"We have not followed that Report in using the phrase 'exceptional circumstances' to describe situations in which the risk of mistaken identification is reduced. In our judgment the use of such a phrase is likely to result in the build up of case law as to what circumstances can properly be described as exceptional and what cannot. Case law of this kind is likely to be a fetter on the administration of justice when so much depends on the quality of the evidence in each case. Quality is what matters in the end (my emphasis)."

Here the quality of the identification evidence was very good indeed and the absence of a full warning cannot mean that it must be totally discarded, which a strict reading of Domican may suggest. I do not believe that was what was intended.

49. I do not consider that the other grounds of appeal argued by Mr. Mullick have been made out. The only matter which did give cause for concern as to the safety of the conviction was the inadequate direction on identification.

Bokhary J.A. :

50. These two applicants, cousins who happen to bear exactly the same full name, were tried before Wong J. and a jury on an indictment containing two counts.

51. Under the 1st count, they were charged with wounding with intent. And under the 2nd count, they were charged with murder.

52. The case concerns events at Whitehead Detention Centre on the night of June 19, 1991:

53. Both applicants, the person whom they were charged with having wounded, the person whom they were charged with having murdered, all the eye-witnesses called by the prosecution and the alibi witness called by the 1st applicant were Vietnamese boat persons detained at that centre at that time.

54. Shortly stated, the case for the prosecution was that on the night in question, the applicants, both of them then in their late teens, stabbed one man wounding him and then, in a separate incident, stabbed another man killing him.

55. On August 8 last year, the jury acquitted them on the wounding count, but convicted them on the murder count.

56. They now seek leave to appeal against their convictions of murder.

57. The evidence against them on that count consisted solely of visual identification evidence unsupported by any other evidence. The view that such evidence was "cumulatively overwhelming" is possible only on the assumption that the eye-witnesses were reliable. And to make that assumption reverses the burden of proof.

58. It has been established by a long line of cases starting with the decision of the Court of Appeal in England in R. v. Turnbull [1977] QB 224 (which decision our courts have consistently followed) and culminating in the recent decision of the Privy Council in R. v. Beckford (1993)97 Cr.App.R.409 (which our courts are bound to follow) that :-

(1) The general rule is that in all cases of visual identification, the judge must give the jury a general warning of the dangers of visual identification.

(2) That general rule was a strong one subject only to very rare exceptions of the type illustrated by this example given by the Privy Council in R. v. Beckford (supra) at p.415 : "If, for example, the witness's identification evidence is that the accused was his workmate whom he has known for 20 years and that he was conversing with him for half an hour face to face in the same room and the witness is sane and sober, then, if credibility is in issue, it will be the only issue."

(3) In giving the general warning, the judge had to elaborate and illustrate the reasons for such a warning.

(4) No particular form of words had to be used. But this had to be got across to the jury. They must always approach identification evidence with particular care. Experience has shown that identification can be wrong even when it comes from a number of honest witnesses and even if some or all of them were familiar with the accused beforehand.

(5) The general warning had to be given even in "recognition" cases of such familiarity beforehand.

(6) It had to be given even if the defence does not rely on the possibility of mistake.

(7) Where the judge has failed in his duty to give a general warning, such a failure will nearly always by itself be enough to invalidate a conviction based substantially on identification evidence.

Those propositions do not constitute the whole of the law relating to visual identification. They go only to the general warning. But in respect thereof, they constitute the irreducible, simple and straightforward minimum.

59. Turning to the facts of the present case, the applicants having pleaded not guilty, the prosecution had the burden of proving its case beyond reasonable doubt. For that purpose, as indicated, it relied solely on visual identification evidence unsupported by any other evidence. The accuracy of that identification evidence not being accepted by the defence, the questions were those which the Privy Council identified in this passage in R. v. Beckford (supra) at p.415 :-

"The need to give the general warning even in recognition cases where the main challenge is to the truthfulness of the witness should be obvious. The first question for the jury is whether the witness is honest. If the answer is yes, the next question is the same as that which must be asked concerning every honest witness who purports to make an identification, namely, is he right or could he be mistaken?"

60. Of the five eye-witnesses called by the prosecution, the defence challenged the truthfulness of four. As to the remaining one, Madam Ha Thu-hang, the defence suggested that she was mistaken.

61. Rightly, the judge left to the jury the possibility of mistake on the part of all the eye-witnesses.

62. But that is all he did. If he had done less, the complaint would be - and would rightly be - that he had wrongly withdrawn from the jury an issue which was for them to resolve.

63. What the judge said to the jury on the possibility of mistake was this :-

"... you will of course consider whether these witnesses could have been mistaken in identifying these two persons, bearing in mind all the evidence."

64. That left the issue to the jury. But neither there nor elsewhere were they given the general warning which, for good reason, the law requires. And the object of the warning, let it be said lest it be forgotten, is the avoidance of any repetition of miscarriages of justice to which errors of identification have led in the past. The courts have not formed a morbid fear of identification evidence. But we have learned a bitter lesson. And we have devised a measured response to the danger of which we have learned. The warning is part of that response. And it is not lightly to be discarded - certainly not just because we see what we think is a strong case of a dreadful crime. The most notorious cases of miscarriages of justice involved dreadful crimes and what were perceived as strong cases.

65. Whether for places like Whitehead Detention Centre or anywhere else, the duty of the courts to protect people from crime is always accompanied by the duty to protect people from being wrongly convicted of crime.

66. Appearing for the Crown before us, Mr Bruce (who did not appear below) readily concedes that the general warning had not been given in the present case. All that he does is to invite us to consider applying - not urge us to apply - the proviso. And he draws attention to factors on each side of the question.

67. In support of applying the proviso there is, Mr Bruce suggests, this. As far as the eye-witnesses whose truthfulness was challenged are concerned - and especially as far as one of them, Mr Nguyen Van-loi, is concerned - the possibility of mistake is so slight that the absence of a warning of the danger of mistake is immaterial.

68. But even assuming that to be correct, there is nevertheless another side to the coin. As Mr Bruce recognizes, there is, on any view, this difficulty in the way of applying the proviso. The warning was certainly vital in respect of Madam Ha. So the absence of the warning would be fatal unless it can be said that the jury would inevitably have convicted even without any evidence from her. That they might have done so, or even that they would probably have done so, is not enough. It is necessary that they would inevitably have done so. Although he has invited us to consider whether we can go as far as to say that, Mr Bruce, commendably level-headed even in the face of a killing so horrifying as this one was, has not felt able actively to argue that we can go that far.

69. And he is right. There is nothing which enables us to go to that length.

70. The defence challenged the truthfulness of the eye-witnesses other than Madam Ha. As to why any inmate of the centre might wish to implicate these applicants, something came from none other than counsel for the prosecution in the court below (not Mr Bruce, it is to be repeated). In cross-examining the 1st applicant, the prosecutor put it to him that he had been a "trouble-maker" in the centre. And the 2nd applicant was the 1st applicant's cousin. That was something to which the prosecutor made express reference in his very next question, by which he suggested that the applicants went to attack the deceased.

71. Then in his closing speech, the prosecutor referred to what he called "a smoke-screen of the squabbles between the witnesses and the accused". And the view which he urged upon the jury was that "these squabbles are not important". The jury may or may not have taken that view. We simply do not know.

72. Both counsel before us have referred to the decision of the High Court of Australia in R. v. Domican (1992)66 ALJR 285. In R. v. Beckford (supra) at p.415, their Lordships cited that case for this proposition laid down by their Honours at p.298 :-

"The trial judge is not absolved from his or her duty to give general and specific warnings concerning the danger of convicting on identification evidence because there is other evidence, which, if accepted, is sufficient to convict the accused ... unless the Court of Criminal Appeal concludes that the jury must inevitably have convicted the accused independently of the identification evidence, the inadequacy or lack of a warning concerning that evidence constitutes a miscarriage of justice even though the other evidence made a strong case against the accused."

73. Although that High Court of Australia's decision in that case was by a majority, the dissenting judge, Brennan J., dissented only because he thought that the warning there was adequate. Certainly, as one sees from the majority judgment (at p.287) the jury in that case received "a very full direction on the dangers of convicting on identification evidence" so that the only complaint was that the jury's attention had not been drawn to a number of alleged weaknesses in the evidence of the eye-witness.

74. As to the application - or rather the non-application - of the proviso, Brennan J. did not dissent. Indeed, his contribution added to the strength of what the majority said on the point.

75. This is how Brennan J. put it (at pp.291-292) :-

" The general principle is this: where, on the evidence and consistently with the directions of the trial judge, it is open to a jury to convict on any of two or more independent bases, a misdirection or an inadequate direction which would vitiate a conviction on one of those bases necessarily results in the setting aside of a guilty verdict despite the availability of another sound basis for conviction. That is because it is not possible to conclude that a guilty verdict has been founded on a sound basis when it was open to the jury to convict on a basis affected by the misdirection or inadequate direction. A Court of Criminal Appeal cannot apply the proviso by speculating either that the jury acted on a body of evidence which was unaffected by the misdirection or inadequate direction; nor can the Court speculate that, if the jury had acted on such evidence, they would have convicted. If a misdirection or inadequate direction would vitiate a conviction based on identification evidence and that basis of conviction was open to the jury, it is impossible to be satisfied that, by reason of the misdirection or inadequate direction, the accused did not lose a chance of acquittal.

The problem in the present case is not whether the case depends to some degree on Mrs Flannery's evidence. (Cf R. v. Evans (1985) 38 SASR 344 at 351; R. v. Bartels (1986) 44 SASR 260 at 270-271.) Clearly, the verdict might have been returned though the jury rejected her evidence. The problem arises because the verdict might have been based solely on her evidence. As it is possible that the jury's verdict was based solely on Mrs Flannery's identification evidence, a defect in the trial judge's warning of the danger of acting on that evidence results in the appeal being allowed and the making of an order for a new trial."

76. Simply by substituting "Madam Ha" for "Mrs Flannery", all of that could be said in this case. Or one could put it on the footing that Madam Ha's evidence might have made the difference between convictions and acquittals. "Took the Crown's case over the hump" is the expression which Mr Bruce used.

77. There is no choice in this case. For the reasons given, these convictions cannot stand. Quite simply, the general warning had to be given. It was easy to give. But it was not given. And, the appellate court not having seen or heard the witnesses and not knowing what the jury thought of each witness, the omission cannot be seen not to matter.

78. As to what happened to the deceased, the prosecution's case was as follows. The deceased was on the top tier of a bunk bed in the hut which he occupied. Two men rushed into the hut. They were each armed with a knife and a metal rod. Climbing up the bunk bed to get at the deceased, one of them stabbed him in the thigh and the other stabbed him in the ribs. Jumping to the ground, falling and getting up, the deceased ran to the lavatory. His assailants pursued him into the lavatory. And then they pursued him as he emerged from the lavatory covered with blood. The chase continued outdoors, where the two armed men, each delivering a thrust, finally despatched their kneeling victim.

79. There was no dispute that the deceased was brutally done to death in that horrifying way.

80. The issue was whether the applicants were the two men who did that to him.

81. In support of its case that they were, the prosecution relied solely on visual identification.

82. Madam Ha identified the applicants as the two men who climbed up the bunk bed to get at the deceased. If her identification was accepted, it alone would be enough to convict the applicants of murder.

83. She was the only eye-witness whose truthfulness was not challenged by the defence.

84. The jury may have acted on her identification alone. Or her identification may have influenced them to accept the evidence of those eye-witnesses whose truthfulness was challenged : we do not know if the jury believed any of those witnesses or, if they did, why. Or the jury may have ignored Madam Ha's identification altogether. We do not know. And we cannot speculate.

85. In my judgment, the only course open to us is to : treat the hearing of these applications for leave to appeal as the hearing of the appeals; allow them; quash the convictions; set aside the sentences; entertain an application for a retrial if such an application is made; direct acquittals if a retrial does not emerge as the just thing to order; or order a retrial if it does. That is the course which I would follow.

86. Witnesses may have been slow to come forward because they were afraid of the killers. But, unless you assume that the applicants were the killers, that fear cannot tell against them.

87. It will, I trust, be appreciated that I do not lightly dissent from my brethren in this, or any other, case. But this is a murder case. Moreover, an extremely important and far-reaching principle of law is involved. My Lord, Penlington JA says that there is "some conflict in the authorities on how we should apply the proviso", and that my approach, as he understands it, is not what he considers to be the correct one. In my judgment, there is no conflict in the authorities which bind this Court. The Privy Council has spoken recently and, as always, clearly. And it is the approach laid down by their Lordships which I follow.

Penlington, J.A.

88. Accordingly, by a majority, the applications are allowed but treating the hearing of the applications as the hearing of the appeal, we dismiss the appeal.