R v Zullo [1993] QCA 208

The relevant incident occurred about 3 a.m. in Orchid Avenue, Surfers Paradise on 2 May, 1991. There were a substantial number of people in the vicinity at the time, some of whom had been engaged in a fight or fights shortly before a stabbing with a knife occurred; that caused Gilligan's death. Only two people, occupants of a taxi which was in the vicinity, claimed to have seen the stabbing and both said that the assailant was a man in a red (or similar colour) shirt; the Crown case was that the appellant was that man. The two people in the taxi were Messrs Lankester and Brown. The appellant was not otherwise identified but there was some evidence from police officers, Dunn and Perrett, that on

13 May, 1991 they had conversations with the appellant, in the course of which he made statements tending to incriminate himself.

The appellant gave no evidence before the jury. Although there were other witnesses who described events which took place before the killing, the Crown case depended essentially upon Lankester and Brown, and the critical point which emerged from their evidence was the colour of the shirt of the person who wielded the knife which brought about Gilligan's death. Brown had drunk a great deal of alcohol on the night in question and it appeared that Lankester's evidence was regarded at the trial as the more important. We were told that counsel for the Crown below invited the jury to accept the evidence of Lankester rather than that of Brown, where the two differed. Neither Lankester nor Brown was able to identify the appellant as the killer, when shown photographs by the police, but it was contended that their identification by the colour of the shirt of the person who stabbed Gilligan was sufficiently certain to warrant a conviction. There was evidence, which was not contradicted, that Zullo had a shirt of a reddish colour on and that he was in the vicinity when Gilligan died; there was no suggestion that any person wearing a red shirt, other than the appellant, was involved in the fights which preceded the killing.

There was evidence from an undercover police officer that another person who was present at the scene, one Beard, confessed to having killed Gilligan. That evidence was led by the Crown. There was evidence that Beard used to carry a knife with him, and indeed he discussed with Hassan that not long before the Gilligan stabbing, he (Beard) had stabbed a person with his knife. Beard was one of the group involved in the melee which culminated in the stabbing of Gilligan. The question arose at the trial whether the jury should have attached any weight to Beard's confession. One difficulty about accepting it was that when interviewed by Perrett and Dunn, Beard asserted that the killer was Zullo. When asked about the confession at this trial, Beard claimed privilege on the ground that the answer might tend to incriminate him.

Counsel for the appellant, Mr A Kimmins, advanced a number of criticisms of the judge's directions to the jury, but it appears to be necessary to deal with two of them only.

Mr Kimmins contended that the judge's directions on this point were unsatisfactory, in that his Honour might have left the jury in some doubt about the proposition that, considering all the evidence including Beard's confession, the jury had to be left in a state of having no reasonable doubt about Zullo's guilt, to warrant a conviction. The critical part of the judge's summing up is as follows :

"Before an out of Court confession by some other person would result in an acquittal of an accused, you would have to be satisfied that it had such substance, after scrutinising all the evidence relevant to it carefully, that you had a reasonable doubt as to the accused's guilt. If after scrutinising all the evidence relevant to that confession you reached the conclusion that there is no substance in it, or that it was not truthful, then you may conclude that nothing relating to that evidence causes you to have a reasonable doubt as to the accused's guilt." (emphasis added)

On behalf of the appellant, complaint was particularly made about the first sentence quoted above, on the ground that it might have conveyed the idea that the Beard confession could not warrant an acquittal of the appellant unless the jury were positively satisfied that it had substance, presumably meaning a solid claim to acceptance. In our respectful opinion the sentence is well capable of that meaning, and so construed constitutes, considered in isolation, a misdirection. There was no question of the jury having to be satisfied that Beard's confession had substance. It was evidence which the jury had to consider when coming to their ultimate conclusion on the guilt or innocence of the appellant and of course the onus lay on the Crown throughout. If the jury were inclined to suspect that the Beard confession was spurious, that would not necessarily have been a justification for ignoring it when coming to their final conclusion.

It should be noted that there is authority of some strength in favour of the view that Hassan's evidence that Beard had confessed to him was not admissible at the instance of the defence: Blastland [1986] A.C. 41 at 52, 53, re van Beelen (1974) 9 S.A.S.R. 163, Donnelly v. United States [1912] USSC 171; 225 U.S. 243; but one must respectfully agree with the comment of the learned authors of the Australian edition of Cross on Evidence, para 33045, that there is "much force in a remark made by Holmes J., who dissented in the American case:

`The exception to the hearsay rule in the case of declarations against interest is well known; no other statement is so much against interest as a confession of murder; it is far more calculated to convince than dying declarations, which would be let in to hang a man; ...'"

The decision of the Queensland Court of Criminal Appeal in Condren [1991] 1 Qd.R. 574 illustrates the importance which, as a practical matter, a confession by a person other than the accused can have in determining whether to convict: see especially per Thomas J. at 581 and 588 et seq. The confession made by Beard was, in our opinion, to be considered by the jury for what they thought it was worth, and may very well have inclined the jury towards a "not guilty" verdict, even if they were by no means convinced that it was truthful.

It was argued for the Crown that any damage done by that part of the summing-up which has been quoted was neutralised by other directions concerning the Beard confession. Of these, the principal passages were, in our opinion:

"You can only reach that position in this case if, after scrutinising all the evidence relevant to Beard's confession, that evidence does not create in your mind a reasonable doubt that Zullo did the stabbing. Because of those considerations, it appears necessary for you to consider whether there was substance to Beard's confession, whether it was truthful.

If you are left with a reasonable doubt as to the accused Zullo's guilt because of substance you see in Beard's statement, then you will acquit. If you consider there is no substance in any of Beard's statements such as would cause you to have a reasonable doubt as to Zullo's guilt, then, if you are satisfied beyond reasonable doubt by the evidence that Zullo stabbed Gilligan, a guilty verdict would follow."

It should be added that "that position" in the first of these quotations is, in the context, a reference to a state of satisfaction beyond reasonable doubt of the guilt of the appellant.

In our respectful opinion these passages would not necessarily have eradicated the thought that the jury had to be satisfied that there was substance in Beard's confession in order to base a "not guilty" verdict on it. In their reference to consideration whether the Beard confession had substance, they echo the direction discussed above. Further, the statements are so framed as to encourage the jury to approach the case by determining the truthfulness of Beard's confession, as a separate issue; the Beard confession was simply a piece of evidence to be considered along with the rest, in the jury's task of determining whether the appellant's guilt had been proved.

At the conclusion of his summing up the judge asked whether counsel for the appellant had any application for

re-directions and the reply was in the affirmative; but no complaint was made of the judge's direction concerning the Beard confession. That must, in our view, be taken into account in determining whether there was a miscarriage of justice, but it is convenient to defer discussion of that subject.

It is our view, expressed in our reasons for judgment in the first appeal, that this is a case of a kind to which the remarks made by the High Court in Domican (1992) 66 A.L.J.R. 285 apply, as to the content of the judge's direction on the question of identification. The identification relied on in that case, as here, was directly visual, rather than, for example, based on fingerprints. It is therefore unnecessary to consider whether what was said about identification directions in Domican extends beyond visual identification. That is commonly based, in large part, on recognition of facial features. But such identification is often supplemented by other aspects such as stature, bodily colour, clothing type and so forth. Our understanding of the principle underlying Domican is that, at least in some circumstances, visual identification of an offender at the scene of the crime is likely to be unreliable for reasons some of which are exemplified by the discussion in Domican, p. 288: the quality of the light, opportunities of observation, the distance between the observer and the perpetrator, and the speed with which events happened. Apparent errors in evidence of observations made, other than those directly relevant to identification of the perpetrator, can be relevant. Examples in Domican were that the observer saw only one person in a car in which the gunman was seated, whereas others saw more than one, and that the observer claimed to have seen Domican jogging near her home at a time when, in fact, his leg was in plaster (288).

Here, the principal difficulty in the jury's relying on the accuracy and reliability of Lankester's observation was simply that it was unclear whether he actually saw the stabbing. One tends to approach evidence on that subject with a predisposition to think that if his account of what occurred was at all reliable, it might have been expected to be consistent on that point; it was not.

Lankester gave evidence, accompanied by some detail, to the effect that he saw the stabbing. He "saw the person with the knife draw it back and push it into the person who was against a parked car". He explained that that took place about six feet away from his taxi and that the thrust or thrusts was or were "sort of horizontal and upward". On the other hand, when he discussed the matter with a solicitor, one Box, a few weeks after the event, he said that he saw the man with the knife "move into the group," whereupon the group dispersed; he did not claim to have seen any stabbing. During that conversation he said to the solicitor that people told him (Lankester) of the stabbing.

The version given to the solicitor was, in this respect, not inconsistent with the statement which Lankester had already made to the police, which was put to Lankester in cross-examination. He said the statement was made two weeks after the stabbing. It concluded:

"During the fight I saw no other person produce knives or any other weapons. After seeing the youth with the knife, I saw him walk up to the other youths that were fighting against the parked car and did not see the outcome as I was talking to the security guard.".

This is also reasonably consistent with the evidence of one Harvey, a security guard, who was in the vicinity but did not see the stabbing. Lankester gave evidence to the effect that he spoke to Harvey. Lankester said:

"Not long after we stopped he [Harvey] came over and said, `It's all right, the police are on their way', and while that was going on this is when I said, `Well, they had better hurry because there is a guy in there with a knife'..."

This is more easily reconcilable with the notion that Lankester saw a man with a knife, but not the stabbing, than with his having seen both. Harvey's evidence was that he spoke to the taxi driver (who was, apparently, Lankester) after having heard someone call out "He's been stabbed".

To our mind, the point did not clearly emerge from the summing-up that although in court Lankester claimed to have seen the stabbing and described it, there is no mention of his having done so in the statement to the police, nor in that to the solicitor, Box. Of these omissions the former is the more puzzling; when asked at the trial about his not having told the police that he had seen the stabbing, Lankester was unable to give any explanation. He said he had spoken to Det. Dunn on the telephone and also at the police station and agreed that he had provided Dunn with his "best recollection of what occurred".

The judge made reference to the conversation with the solicitor:

"So far as Lankester is concerned, you have got his account to the police in his statement, the written statement. You have got his account on evidence, his explanation for the differences, and you have also got his account in the taped conversation with the solicitor. That taped conversation, Exhibit 15, is part of the evidence, and you may use anything in it for such purposes as you, the jury, see fit. He has the man in the maroon coloured shirt and dark trousers going down the side of the car, up to the man who was being held. He saw a movement and he thought it was the stabbing movement into the body. It was of that particular nature. In that phone conversation he did say, and I think you will find this at page 12 of that - he was asked :

`Anyone else in the group have reddish shirts or maroon shirts, or -----"?-- Not that I recall.

Do colourful shirts ring a bell?-- Not that I recall.'"

It will be noticed that the sentence, "He saw a movement and he thought it was the stabbing movement into the body," appears in the middle of the judge's discussion of the conversation with the solicitor. But that conversation, the transcript of which is Exhibit 15A, includes no reference to a stabbing movement into the body. Apparently, the sentence to which we refer was intended to convey the effect of Lankester's evidence in court, not anything said in the phone conversation with the solicitor.

The point that emerges from that version is that on no less than three occasions he made it clear that he did not see the "guy in the red shirt" stab anybody; he saw him carrying a knife and running towards the end of a group that was fighting, after which they quickly dispersed. Mr Lankester did not tell the solicitor he saw what happened when the man went into the group; he only saw him run into it. That account was given in response to fairly detailed and quite fair questioning only four weeks after the incident. At that time he had already given a statement to the police which likewise is devoid of any suggestion that he saw the stabbing. That he did not see that part of the incident tends to be confirmed by other details volunteered to Mr Box, including that after the dispersal "a couple of them" ran up past the cab "to the security guy" and "they told me he was stabbed". He added that some time later after dropping off his fares, he revisited the scene, asked what had happened, and was again told that the man was stabbed.

Mr Kimmins argued for the appellant that there were other difficulties with Lankester's version, and there is substance in that submission. It is certainly true that Lankester's evidence varied on the question whether the man in the red shirt ran or walked to Gilligan, and there were other points of that kind which might have raised doubts about the accuracy of Lankester's account of his observations of the incident. But in our respectful opinion, if there was an obligation, in summing-up on the question of identification, to draw to weaknesses in the identification evidence, the judge must at least have covered Lankester's inconsistency on the question whether he saw the stabbing or not. The court said in Domican:

"The jury must have the benefit of a direction which has the authority of the judge's office behind it... It follows that the trial judge should isolate and identify for the benefit of the jury any matter of significance which may reasonably be regarded as undermining the reliability of the identification evidence" (288).

The judge's direction on the general difficulties of identification evidence was not attacked before us. His Honour gave proper directions on that topic, on the need for proof that no-one else in the vicinity was in a red shirt; and of the specific danger of error in associating hand, knife and shirt in an incident of this kind. His Honour's directions included the comment -

"If there are a number of people in a group, say even a group of three, one of them might have a very distinctive shirt on, and one hand in that group has a knife. It still becomes important when you are considering this identification-type evidence to be satisfied that the witness is accurate when he said that the hand in which he saw the knife was the hand that went with the maroon shirt."

That, with respect, was an apt direction in the present matter. It is only in respect of the inconsistencies referred to that the summing up fails to satisfy the requirement of isolating "any matter of significance which may reasonably be regarded as undermining the reliability of identification." (Domican, p. 288).

Counsel for the appellant asked for redirections on the identification question, saying, amongst other things, that the judge had "drawn to the jury's attention that the only identification point here really was in relation to the shirt". Counsel added that the judge was "required to draw specific inconsistencies to the jury". The judge replied:

"I think that's really only in relation to the identification, and I don't know that there are any inconsistencies in relation to the shirt identification."

In the course of further discussion, the judge expressed "grave doubts" as to whether there was any necessity to give a "Domican-type warning".

In our opinion, although the judge was not bound to follow the majority's expression of view on this point in the first appeal, since it was obiter, that expression of view was correct. In this Court, the suggestion that no Domican warning was necessary was, although made in a written submission, not pressed before us. It should be noted that in its explanation of the warning necessary in identification cases, the High Court did not expressly confine itself to identification by facial appearance, or indeed even to visual identification. If one reads what was said in Domican literally, it applies to all kinds of identification, including, for example, identification by voice; see R v. Harris (No. 3) [1990] VicRp 28; [1990] V.R. 310. In King (1975) 12 S.A.S.R. 404, it was held in the South Australian Supreme Court that a witness' description of observable characteristics of the offender, whom the witness does not claim to identify, does not call for any direction of a kind appropriate to identification cases. If that is correct, after Domican, then it would seem to follow that the Crown case may depend substantially upon observed characteristics of the offender put forward to identify the accused, without calling for any special direction. In our respectful opinion, that is taking too narrow a view of the problems associated with visual identification. Just as Lankester had apparent difficulty in recalling and recounting whether or not he had seen the man who ran up with the knife actually stab Gilligan, close to his taxi, which one would think to be a most memorable aspect, he might have had trouble in being accurate about whether it was the man with the knife, rather than another, who was wearing a red shirt; one would think that could well be less memorable.

Identification may be wholly or partly indirect, depending, for example, on clothing as in Ramsden referred to in volume 1 of Archbold at p.1625 or upon identification of a person accompanying the offender: see Bath [1990] Crim. L.R. 716, 717. Domican should not be applied as if what the High Court said were a statute, but there appears to be no reason for declining to apply it where the identification, although visual, is two-stage: where the observer of events ascribes certain characteristics to the offender, and other evidence identifies the accused as having those characteristics.

Lankester's evidence did not stand alone. There was evidence from Brown that he saw a man in a red shirt "beating into" the other person and that he then noticed blood coming from the shirt of the other person. He also in the course of his evidence made references to the man in the red shirt as the "guy who stabbed him", but these statements may be considered inferential. However, Brown's evidence is markedly different from Lankester's in relation to practically every surrounding detail of the story. Brown had consumed 15 pots of beer that night, and the account is not particularly clear or coherent. It differed substantially from the accounts of the witnesses Lankester, Greenfield and Harvey.

There was, in addition, support for the Crown case in two unrecorded conversations which the appellant had with police officers, Dunn and Perrett. Those conversations did not constitute a direct admission of the commission of an offence; there was reason to be doubtful about the accuracy of the police account of the conversations. Accuracy was particularly important, because what the appellant is reported to have said is only incriminating by implication and the jury should not have been prepared to make implications unless confident of what was said. There was no attempt to record the conversations in any fashion at the time when they occurred, although they could presumably have been recorded by use of a small tape-recorder, had the police bothered to take one with them. Some hours after the alleged conversations, Perrett wrote an account of them in his note-book. The procedure followed was, as has been said in the reasons in the previous appeal, unsatisfactory. The written account was said by Perrett to be the product of the joint recollection of the two police officers. The failure of the officers concerned to make a proper record of these important conversations has contributed to the outcome of this case. One can understand a failure to record admissions which are made unexpectedly, but here there were, on the evidence, two premeditated interviews, designed to extract admissions, with a person suspected of a most serious offence.

In our respectful opinion the direction with respect to the use of Beard's confession may not, had the Crown case otherwise been overwhelmingly strong, have necessitated a setting aside of the conviction. But the Crown case, and particularly the police evidence, had substantial weaknesses. That must often be so where the question is the identification of a particular person as having performed a certain action, in the midst of a complicated melee at night. When one adds, to the difficulty with the summing up in relation to the Beard confession, the fact that the learned primary judge did not give directions of the kind Domican requires, with respect to Lankester's evidence identifying the offender as a man in a red shirt, it becomes impossible to uphold the conviction.

There have been two trials with different results: the first conviction being of murder, and the second one of manslaughter. We allow the appeal and set aside the conviction, but make no order for a new trial; the accumulation of contradictions in the evidence of Lankester and Brown is an important factor in our determining that there should be no new trial.

The Attorney-General's appeal against sentence must be dismissed.