Davies v R [1937] HCA 27; (1937) 57 CLR 170 (4 June 1937)

This is an application by two prisoners, found guilty of murder, for special leave to appeal from two orders of the Supreme Court of Victoria upholding their conviction. One order refused applications by the prisoners for leave to appeal from their conviction. The second order was made on a reference by the Attorney-General under sec. 610 (a) of the Crimes Act 1928, a provision which enables him to refer the whole case of a person convicted on indictment to the Full Court, and provides that the case shall then be heard and determined by that court as in the case of an appeal by a person convicted.

The crime with which the prisoners were charged was committed in the course of a robbery under arms by three men. The victim was one of two officers who were taking a large sum of money from the Stamp Duty Office to a taxi-cab waiting in the street outside the building. On the opposite side of the street a motor car stood waiting with a man in the driving seat. Just as the bag of money was placed in the taxi-cab two men stepped forward, one of whom seized the bag. One of the officers made some resistance, and was immediately shot dead. The robbers escaped to the waiting car which was rapidly driven away. The car was driven into the yard of the city council destructor close at hand at the corner of Spencer and Lonsdale Streets, where the three men were seen to leave it by the foreman and another council employee. The foreman spoke to two of them. Several people witnessed the crime. After the robber had shot his victim, he dropped his weapon. It proved to be an automatic pistol which had undergone some repair. A young man was called as a witness who said that some days before the crime he had done the repairs for a stranger whom he identified as one of the prisoners.

The question upon the trial of the prisoners was whether they were two of the three men carrying out the robbery in the course of which the murder was committed.

The evidence to establish their identity fell under three heads. There was the direct testimony of persons who saw the men who committed the crime either at the scene or afterwards in the city council yard, and the testimony of one other witness who said that two days before he had seen the car by which the robbers escaped, and that it then contained four men, two of whom he recognized as the prisoners. There was the youth who said that he repaired the pistol, and the fact that according to the evidence it was fired, not by the prisoner he said that he recognized, but by the other of the two assailants. In the third place, there was the evidence of a man named Stevens, who said that he was an associate of the prisoners, and that they had each admitted to him that they were concerned in the robbery.

The prisoners were arrested about nine months after the date of the crime. None of the witnesses who identified them or either of them claimed any previous knowledge of the prisoners. But when, after the prisoners were taken into custody, it was desired to ascertain whether the witnesses could identify either of them, none of the precautions was taken upon which the English Court of Criminal Appeal has so long insisted in such circumstances. The prisoners were not placed in company with other men, but each was shown singly to the witness, who was asked to say whether he was one of the men in question. In some cases this was done at the detective office, in other cases, when the prisoners were in the dock at the police court charged with the crime for which the identification was sought.

In one case, although the prisoners were in custody in the Metropolitan Gaol, the witness was shown some photographs from which he picked out one of the prisoners, and some days afterwards, when that prisoner was brought up to the police station, he was shown him singly for the purpose of identifying him.

It is said that in Victoria it has not been the practice of the police to ask potential witnesses to say whether the person whom they are prepared to identify is among a number of persons presented before him together, although the gaol authorities have adopted this practice.

Notwithstanding the strong views upon this matter expressed by the English Court of Criminal Appeal and the manner in which that court has set aside convictions if the proof of identity depends on the belief of a few witnesses whose recognition of the prisoner has taken place when he has been shown to the witness singly as a suspect, the Supreme Court of Victoria, as a court of criminal appeal, has not hitherto regarded such a method of identification as open to so much danger as to warrant the setting aside of a verdict of guilty founded upon evidence of identity so obtained, or as to require the trial judge to include in his charge to the jury any special warning.

Owing, no doubt, to the position which the Supreme Court has taken in reference to the English rule, the learned judge who tried the prisoners left the matter to the jury without any definite warning of the dangers which such a method of identification is in other jurisdictions considered to involve. Among other things he said:— "There has been a most acrimonious debate about what is the best way for persons to be identified by the police, and it is neither your business nor mine to say which is the best way. The Crown has suggested one way is the best, counsel for the defence, with great enthusiasm, has suggested the other is the only fair and proper one. I think the only important thing at all in this case is that it is one of the circumstances you want to look at." "It is said a more proper way to identify him would be to put him amongst others of about his own size and kind and let him be picked out. Whether that be so or not, the only important things here are that first of all, if you take the view, as probably you will, that to pick out a man in that way in the dock or under the lights by himself does not add anything to the identification, then the identification loses something it might have or would have if he picked out the accused man from among a number of others—and that is one thing. The second thing is that if a man is pointed out to a witness by himself under a light, or still more in the dock, that that in effect is an effort by the police to force him into saying That is the man. That it is the use of suggestion—Of course he must be the man, I see him in the dock accused of murder and he must be the man. I do not know if you were brought up there to identify a man, that you would say he was the man if you did not believe it, just because he was in the dock or presented to you alone under the lights. However, you will use your own knowledge of human affairs and see how far it is likely to influence a man, but you will remember that in this case there are a large number of witnesses who were asked to identify, and the results are very different."

In England, where the Court of Criminal Appeal has no power to order a new trial, it has been unnecessary to draw a distinction between, on the one hand, the question whether a conviction ought to be allowed to stand when, having regard to the whole treatment of the case, including the manner in which the identification was conducted, the number of identifying witnesses, their want of familiarity with the features and characteristics of the prisoner and the nature of the charge, the court is dissatisfied with the result, and, on the other hand, the question whether, subject to a proper direction to the jury, sufficient material to justify a verdict of guilty cannot be found in the whole evidence, including the testimony of witnesses whose preliminary identification of the prisoners was carried out in a mode which the English Court of Criminal Appeal discountenances. From the beginning, that court has acted upon no narrow view of the cases covered by its duty to quash a conviction when it thinks that on any ground there was a miscarriage of justice, a duty also imposed upon the Supreme Court of Victoria (Crimes Act 1928, sec. 594 (1)). It has consistently regarded that duty as covering not only cases where there is affirmative reason to suppose that the appellant is innocent, but also cases of quite another description. For it will set aside a conviction whenever it appears unjust or unsafe to allow the verdict to stand because some failure has occurred in observing the conditions which, in the court's view, are essential to a satisfactory trial, or because there is some feature of the case raising a substantial possibility that, either in the conclusion itself, or in the manner in which it has been reached, the jury may have been mistaken or misled. This is the basis upon which the English court has set aside convictions resting upon identification conducted in an unfair or unsatisfactory manner. As is not unnatural, the judges of that court have more often than not combined with the statement of their view that in a given case a conviction cannot stand some expression of their general condemnation of the method of identification which has led to that result. Such observations, no doubt, have been made as having a salutary effect in future cases where identification might be sought. But, in strictness, they go beyond the decision of the particular case, because in each case the question must be, not whether the identification has been conducted with propriety and fairness, but whether upon the whole evidence as it in fact existed when it came to be laid before the jury, and having full regard to the treatment of the matter at the trial, the actual verdict ought not to stand because a miscarriage of the kind described occurred.

It is almost unnecessary to say that the amount of care and the nature of the precautions which should be taken when a potential witness is brought to identify an accused or suspected person must vary according to the familiarity of the witness with that person. It would be ridiculous, because the prisoner has been shown alone to a potential witness, to deny the value or reliability of the identification if the witness' knowledge of the prisoner arose from long and close association or from every day intercourse in business affairs. But where, before the occasion with which it is sought to connect the person accused or suspected, the witness has seldom or never seen him, experience has led the English court to look for the greatest care to avoid a mistake or prejudice. They treat it as indisputable that a witness, if shown the person to be identified singly and as the person whom the police have reason to suspect, will be much more likely, however fair and careful he may be, to assent to the view that the man he is shown corresponds to his recollection.

If, on the other hand, he were called upon to say whether anyone of a number of persons were the man, his entire mental attitude would be different. A witness who is taken by the police for the purpose of seeing whether he can identify a person who is in custody in relation to a particular crime has in his mind a recollection or impression of the person whom he saw, or, it may be, heard, at the scene of the crime or in relation to some matter which is connected with the crime. The recollection probably relates to the appearance of the person, and possibly to his mode of standing, moving, or speaking or some other characteristic. It is important that this recollection should not be overlaid or in any way affected by suggestions that a particular person in custody is either the person previously seen by the witness or is the person suspected of or charged with the crime. Moreover, inspection of a photograph of the person in custody before viewing him naturally tends to impress on the mind the characteristics shown in the photograph, so that the witness, however honest he may be, tends to identify the person in custody with the person shown in the photograph rather than with the person whom he himself saw previously.

Similarly, if a witness is shown a single person and he knows that that person is suspected of or charged with the crime, his natural inclination to think that there is probably some reason for the arrest will tend to prevent an independent reliance upon his own recollection when he is asked whether he can identify him. This tendency will be greatly increased if he is shown the person actually in the dock charged with the very crime in question.

We think the view accepted in England and, as far as we know, elsewhere in the Dominions where the provisions of the Criminal Appeal Act have been adopted, should be applied in Victoria. That view, as we understand it, is that, if a witness whose previous knowledge of the accused man has not made him familiar with his appearance has been shown the accused alone as a suspect and has on that occasion first identified him, the liability to mistake is so increased as to make it unsafe to convict the accused unless his identity is further proved by other evidence direct or circumstantial. Where that further evidence consists in or includes other witnesses whose identification has been of the same kind, the number of witnesses, their opportunities of obtaining an impression or knowledge of the prisoner and other circumstances in the case must be taken into account by the court of criminal appeal for the purpose of deciding whether on the whole case the possibility of error is so substantial as to make the conviction unsafe.

As the responsibility of convicting must rest with the jury their appreciation of the question is an important consideration, and in a case where the method of identification is open to the objections we have discussed, they should be clearly warned of the dangers, which according to the accepted view, do exist.

In the present case, we think that the observations of the learned judge do not amount to a fulfilment of this requirement. Following the view apparently prevailing in the Supreme Court of Victoria, he treated the matter as one depending upon a choice between rival systems, between different "schools of thought," and did not give the weight of his judicial authority to a statement of the dangers which beset the method in fact adopted, and he did not fully explain those dangers. His direction cannot be described as a warning.

The present case does not, in our opinion, call for any discussion of the difficulties which arise from the use of police photographs as a means of identification. The embarrassments arising from the use of such photographs are discussed by Ferguson J. in R. v. Fannon and Walsh[1], and in R. v. Bagley[2]. We refer particularly to the judgment of Macdonald C.J.A. (dissenting)[3], because it contains observations of wider application in relation to identification and the effect of a proper direction thereon, a matter also mentioned by Lord Alverstone C.J. in R. v. Chapman[4].

If the only ground were the manner in which the learned trial judge dealt with the question we have discussed, we might have hesitated in intervening and granting special leave. But, as we have attempted to show, the whole question of identification is necessarily bound up with the nature of the other evidence in the case. The evidence of Stevens, if believed, would of course have carried the case against the prisoners the whole distance. Stevens was a man of very bad character, and the learned judge warned the jury of the danger of acting on his evidence, but his observations were of such a nature, we think, as to make it not at all improbable that the jury might add his evidence to the rest of the testimony in the case as fitting in with it and making, as a whole, a case establishing the identity of the prisoners with the culprits.

After the appeal to the Supreme Court, Stevens swore a declaration stating that his evidence was false in every material particular. He then swore another declaration stating that his evidence was all true, and that his earlier declaration was false. A declaration by a witness that he has committed perjury cannot possibly be accepted as a ground in itself for setting aside the result of a trial in which the witness has given evidence. If the contrary were held, the whole administration of both civil and criminal justice would be undermined. The subsequent discovery that some evidence (as in this case) is said by the witness who gave it to be false, or is actually proved to be false, cannot, as a general rule, be allowed as a ground in itself for setting aside a verdict or judgment. But if the verdict is open to objection upon a ground affected by such evidence, the case is different. It would not be wise to attempt to frame a universal rule even for such cases. As the Full Court indicates in its judgment, the subsequent statement that the original evidence is false may be explainable by pressure brought to bear upon a witness or by the operation of any one of an indefinite number of motives. Each case should be treated in relation to its own facts. In this case the evidence of Stevens, if believed by the jury, was conclusive of the guilt of the accused persons. The Supreme Court took the view that Stevens' recantation and his subsequent withdrawal of his recantation threw no further light on the credit to be attached to his evidence. Mann C.J. said:—"In our opinion the material submitted to us throws no new light of any importance upon that problem. It was relevant for the jury to consider the temptations, the hopes and fears affecting such a man in the circumstances existing up to the time he gave his evidence. It would be only indirectly relevant and might well be misleading to consider his subsequent words and acts spoken and done in altered circumstances and in the presence of other influences"[5].

We respectfully think that this does not sufficiently take into account the not remote possibility of the jury's having given some definite weight to the fact that Stevens, however criminal in instincts, was prepared to give evidence against the prisoners, with whom, he swore, he had associated. We know that his Honour the Chief Justice, when he presided at an earlier trial, expressed the view that no effect at all should be given to Stevens' evidence, and, if this view had again been strongly commended to the jury which convicted the prisoners, there might be much to be said for the view that Stevens' recantation could not have much importance. But it must be remembered that the Crown chose to rely upon the man's evidence and press its probative value, and the judge's charge does not advise the jury to reject his testimony. It is now known that it is completely untrustworthy, and ought not to be allowed to enter into the reasons for any verdict of guilty. Whether the jury believed his evidence or gave any weight to it in fact cannot be known, but all the other evidence implicating the accused depended upon evidence of identity, and, in this case, the jury was not, as we have already said, adequately instructed with respect to the matters which they should consider in determining the value of that evidence. In these particular circumstances, the facts relating to Stevens' evidence are sufficient, in our view, to entitle the accused to a new trial. We are clearly of opinion that, notwithstanding the mode of identification adopted, the evidence, without the testimony of Stevens, is enough to support a conviction if there were a proper warning to the jury.

We do not think the grounds of appeal not related to these particular matters are of sufficient substance to support the appeal, but for the reasons stated, the order of the Full Court of the Supreme Court upon the hearing of the case in pursuance of sec. 610 of the Crimes Act 1928 is set aside, the verdict of the jury and the sentence of the court are set aside, and a new trial of both accused persons is ordered.

Special leave to appeal granted to both applicants. Appeals allowed. Set aside order of Supreme Court dismissing appeal constituted by the reference of the Attorney-General dated 5th April 1937. Set aside verdict of jury and sentences of Supreme Court. Order new trial of both accused.

[1] (1922) 22 S.R. (N.S.W.) 427; 39 W.N. (N.S.W.) 130.

[2] (1926) 3 D.L.R. 717.

[3] (1926) 3 D.L.R., at pp. 718 et seq.

[4] (1911) 7 Cr. App. R. 53, at p. 56.

[5] (1937) V.L.R., at p. 158.

High Court of Australia

Davies v R [1937] HCA 27; (1937) 57 CLR 170 (4 June 1937)

HIGH COURT OF AUSTRALIA