[Home] [Databases] [World Law] [Multidatabase Search] [Help] [Feedback]
	England and Wales Court of Appeal (Criminal Division) Decisions
You are here: BAILII >> Databases >> England and Wales Court of Appeal (Criminal Division) Decisions >> R E G I N A v Ian Hersey [1997] EWCA Crim 3106 (01 December 1997) URL: http://www.bailii.org/ew/cases/EWCA/Crim/1997/3106.html Cite as: [1997] EWCA Crim 3106, [1998] Crim LR 281

[New search] [Printable RTF version] [Help]

IN THE COURT OF APPEAL
CRIMINAL DIVISION

		Royal Courts of Justice The Strand London WC2A 2LL
		1 December 1997

LORD JUSTICE SWINTON THOMAS: On 6 December 1996, in the Crown Court at Portsmouth, before His Honour Judge Lewisohn QC, the appellant was convicted of robbery and sentenced to serve seven years' imprisonment. He appeals against his conviction with the leave of the single judge.

The primary evidence against the appellant was that of the owner of the shop which was attacked, who identified the appellant as one of the robbers by means of a recognition of his voice. The issues that have arisen in this appeal relate entirely to voice recognition or voice identification.

The robbery took place on 28 October 1995 in a small shop in Hillsley Road, Paulsgrove in Hampshire, just as the shop was about to close a little after 9.00pm. At that time Mr Whiting and two female assistants were in the shop. Two men entered the shop, wearing balaclava masks and each holding a gun. One of the men held a gun to Mr Whiting's head and ordered him to lie down on the floor. Both of the men made demands for money. The first of the two men put a knife to Mr Whiting's throat and demanded to know where the money was. Mr Whiting told him that the day's takings were in a cupboard in the stockroom. The man went into the stockroom and took a blue cash bag. When he returned, that man demanded to know where the safe was and Mr Whiting told him that there was no safe in the shop. They then demanded to have the video recorder. There was further conversation whilst the men removed the video, and as they left the first of the two men (as he was described) said, "Don't do anything for five minutes or I'll come back and blow you away."

It will be appreciated from that very short recitation of the relevant facts that there was a considerable amount of conversation during this period of time, which Mr Whiting estimated lasted overall for about a quarter of an hour. Mr Whiting said that as time passed on so he recognised the voice of the first man as that of the appellant. He said that the appellant had been a customer of his for several years; he had come into the shop on many occasions in order to make purchases. There was one particular instance in relation to the purchase of trailer, and another when he tried to persuade Mr Whiting to take back into his employment a girl whom he (Mr Whiting) had sacked. He said in the course of his evidence that during the attack he gradually appreciated that the voice he was hearing was that of his customer and that the customer was the appellant Mr Ian Hersey. In his evidence Mr Whiting said that the voice that he heard was a distinctive one, a deep, rough voice. He said, "There was a point when I was totally convinced I was right".

The two female assistants did not recognise the appellant's voice and were unable to pick out his voice at the subsequent voice identification parade to which we will refer in a moment. One of those witnesses picked out a different voice.

The appellant was arrested on 18 November 1995 and he declined to answer the questions put to him by the police. He did not give evidence at the trial.

A voice identification parade was held on 7 January 1996. It was conducted by Police Inspector Jarman. There were 17 volunteers for the parade and 11 of the voices were actually used. The men who volunteered read a piece of text which had been extracted from an interview with the appellant in respect of which no charges had arisen.

An application was made to the trial judge by counsel acting for the appellant to exclude the evidence of the voice identification parade pursuant to the provisions of section 78 of the Police and Criminal Evidence Act 1984. The judge ruled that the evidence should be admitted and the first submission made by Mr Donne QC on behalf of the appellant was that the judge was wrong to admit that evidence. A voir dire was held in relation to this issue. The judge heard evidence from Inspector Jarman and from a Mr Duckworth, an expert in voice identification instructed by the defence.

Mr Donne has summarised his submissions in his full and helpful skeleton argument and has elaborated on them in his oral argument. Mr Donne submits that the judge was wrong to allow the Crown to adduce the evidence of the identification parade because:

(1) Mr Whiting already knew the appellant's voice as a regular customer of three-and-a-half years' standing.

(2) Mr Whiting had alleged immediately after the robbery that he recognised the appellant's voice as that of one of the robbers.

(3) The purported recognition should have been allowed to stand on its own. How, it is asked, could it fairly be tested/strengthened by a later voice identification parade?

(4) This was not an appropriate case for a voice identification parade; alternatively, it was not one in the form which it took. Mr Donne in support of that relies upon the statement made by Mr Duckworth and also a statement taken from Dr Logan, to which again we will refer a little later.

(5) Mr Whiting was virtually bound to pick out the appellant's voice in the parade; it was the only one which he knew.

(6) There were two potential dangers inherent in the holding of a voice identification parade held in the way it was:

(a) that Mr Whiting's inevitable recognition of the appellant's voice might wrongly and/or unfairly influence his evidence of his purported recognition of the appellant's voice at the time of the robbery; and

(b) the jury might attach too much weight to Mr Whiting's recognition of the appellant's voice on the parade when the crucial issue was not did Mr Whiting recognise the appellant's voice from twelve others, but did Mr Whiting correctly recognise the appellant's voice at the time of the robbery?

That submission is central to the totality of Mr Donne's submissions. He says that insofar as the identification parade is concerned, it was, in truth, inevitable that Mr Whiting would recognise and so identify the appellant's voice because he thought (wrongly, submits Mr Donne) that that was the voice that he had heard in the course of the robbery. Accordingly, it is submitted that the voice identification was not only valueless, but dangerous from the point of view of the jury's appreciation of this case.

(7) The evidence of Mr Duckworth in the voir dire demonstrated the potential dangers and unfairness of the appellant in this particular voice identification parade because:

(a) twelve voices were too many;

(b) the pitch of the appellant's voice was markedly different to all but one of the other voices;

(c) the other speakers all read the text differently to the appellant and in such a way that only his words made sense;

(d) the appellant's pause at this point in the text was not repeated by the other speakers;

(e) the police had not taken advice from an expert so as to seek to ensure that the other voices were of similar pitch to the appellant's; and

(f) the evidence which Mr Duckworth gave that problems will arise over the pitch of a man's voice when the situation is one which causes stress.

Mr Donne accordingly submits that the admission of this evidence was highly prejudicial. He submits that there is a very real danger that a person who is asked to attend as a witness on a parade of this nature will pick out the voice that he knows as opposed to the voice that he heard at the scene of the crime. There is, Mr Donne submits, the added danger that the jury will place too much emphasis upon the identification parade as opposed to the evidence of what the witness heard at the scene. Consequently Mr Donne submits that all that prejudice greatly outweighs any probative value.

The judge gave a very full and, in our judgment, admirable ruling on this issue. He set out fully the relevant considerations as he saw them. He then went on to say that there were no decided cases which lay down hard and fast rules or guidance in relation to visual identification parade cases as to the appropriateness of an identification parade in recognition cases, and still less were there cases on the appropriateness of an identification parade in a case of voice recognition. He then continued:

"Is it fair that the prosecution should be allowed to adduce the voice identification evidence in the light of the evidence given by the defence expert Mr Duckworth?"

He then referred in a little detail to the evidence that had been given. He referred to the differences in the voices that were used on the parade. He continued:

"It is in my view unrealistic to expect that the police aided or unaided to assemble a collection of voices where there are not differences in pitch, timbre, flow, forcefulness. Obviously there may be cases where the accused's voice is so markedly different from the other voices that one would be driven to say that it was not a fair test that in such cases it would be right to exclude the evidence under section 78. In my view this is not such a case and Mr McCarraher was justified in making the point that two of the three witnesses failed to identify the accused's voice. Therefore I am not prepared to exclude the voice identification evidence under section 78."

The first point to be made in relation to the judge's ruling is that he was exercising a discretion. The judge had to consider whether the admission of the evidence would have such an adverse effect on the fairness of the proceedings that the court ought not to admit it. In our judgment it is, in truth, difficult to see any valid reason why in the circumstances of this case the judge should not have admitted this evidence. It is said that Mr Whiting knew the appellant's voice and that the Crown's case would not in those circumstances be strengthened by any subsequent further identification. That may or may not be true. However, one of the purposes of an identification parade is to give the witness an opportunity to test whether his original identification was or was not correct and, even more important, as has been recognised in cases concerning a visual recognition, in appropriate circumstances out of fairness to the accused so that he can be excluded if the original identification was erroneous. It is often overlooked that identification parades may be as valuable to an accused as they are to the prosecution. It is then said that Mr Whiting was virtually bound to pick out the appellant's voice on the parade. That is by no means necessarily correct. It is then said that the jury is likely to place undue weight on the subsequent recognition on the voice, whereas the true evidence is the initial identification. As we indicated, that is central to Mr Donne's case. We understand the point that he makes in relation to that, but as against that it is a point that will be perfectly obvious to a jury and one which can be readily dealt with by the judge in his summing-up and by counsel in their speeches to the jury.

Finally, it is said that the particular voice identification in this case was flawed. It was said, as we have indicated, that the twelve voices were too many; that the pitch of the appellant's voice was different to all but one of the other voices; and that the others spoke the text differently. Identification parades of their nature, whether visual or by voice, present difficulties for the police in obtaining suitable volunteers. They must do their best. If, for example, the suspect has an unusual voice, appropriate consideration must be given to that. If the suspect has an accent, then that must be taken into account: it would almost certainly not be valuable to hold a voice identification when the suspect speaks with a strong Glaswegian accent and all the other persons on the parade are Londoners. In this particular case the pitch of the appellant's voice was low; so was that of one of the other volunteers, albeit not so low as that of the appellant. Having said all of that, the police must do their best, but it is obviously quite impossible to obtain volunteers who all have similar voices. A judge, when called upon to do so, will undoubtedly rule out the evidence of an identification parade, whether visual or by voice, if he considers overall that it was unfair. That is the test which he must pose for himself. Once the evidence is before the jury they will have little difficulty in our view, particularly when assisted by counsel and the judge, in appreciating any differences in the voices on the parade and make due allowance for that fact.

In our judgment the judge's reasoning for admitting this evidence cannot be faulted and in our view he was correct to admit it. As we have already indicated, the defence had available the evidence of an expert, Mr Duckworth. The judge ruled that Mr Duckworth's evidence was not admissible. This was a re-trial and Mr Duckworth had given evidence without objection at the first trial. The point as to admissibility was taken by the judge himself. He said at page 6 of the transcript:

"The final point, not raised by Mr McCarraher but myself, was whether it is appropriate that Mr Duckworth should be permitted to repeat his evidence in front of the jury. Mr McCarraher said that he was minded to argue the point at the first hearing of the case before Judge MacKean, but Judge MacKean in the course of the voir dire, which he held, expressed the view that no doubt the jury would find Mr Duckworth's evidence helpful, so Mr McCarraher desisted from arguing the point.

My view is that it does not need an expert to draw attention to the differences in pitch and to explain the potential significance thereof. It is something that the jury can notice for themselves. Mr Stopa is as well able to emphasise the differences and dangers. Mr Stopa relied on the case of Robb (1993) Cr App R 161, and in particular to the passage in the judgment of Bingham LJ, as he then was, at page 165, where he was dealing with the expert's experiences and expertise and his auditory techniques.

The expert had said that he paid close attention to voice quality, voice pitch, pronunciation of vowels and consonants, but the expert's function in the case of Robb was entirely different to the function of Mr Duckworth in this case. In Robb the expert had been set the task of comparing the undisputed voice of the defendant with the voice that made ransom demands and giving his expert opinion whether it was the same voice. Here, Mr Duckworth is pointing out to a jury and explaining a difference which they can hear for themselves."

The judge then referred to a well-known passage in the judgment of Lawton LJ in R v Turner 60 Cr App R 80 where he said:

"An expert's opinion is admissible to furnish the court with scientific information which is likely to be outside the experience and knowledge of a judge or jury. If on the proven facts a judge or jury can form their own conclusions without help, then the opinion of an expert is unnecessary. In such a case, if it is just out of the scientific jargon, it may make judgment more difficult."

The trial judge continued finally:

"It is for all these reasons that I have come to the decision that Mr Duckworth's evidence is not admissible."

The passage we have just cited from Turner has been approved in a number of cases, for example, R v Stockwell (1993) 97 Cr App R 260 and R v Clarke [1995] 2 Cr App R 420. In Stockwell the Lord Chief Justice quoted the passage from Turner which we have just cited, and he added:

"It is to be noted that Lawton LJ there referred to a jury forming their own conclusions 'without help'. Where, for example, there is a clear photograph and no suggestion that the subject has changed his appearance, a jury could usually reach a conclusion without help. Where, as here, however, it is admitted that the appellant had grown a beard shortly before his arrest, and it is suggested further that the robber may have been wearing clear spectacles and a wig for disguise, a comparison of photograph and defendant may not be straightforward. In such circumstances we can see no reason why expert evidence, if it can provide the jury with information and assistance which they would otherwise lack, should not be given. In each case it must be for the judge to decide whether the issue is one on which the jury could be assisted by expert evidence, and whether the expert tendered has the expertise to provide such evidence."

We stress the passage in which the Lord Chief Justice said:

"In each case it must be for the jury to decide whether the issue is one on which the jury could be assisted by expert evidence."

Mr Donne submits that the judge was wrong not to admit this evidence. He has referred us to a number of authorities, including R v Robb (1991) 93 Cr App R 161, and to the report of Mr Duckworth. In Robb it was held that the evidence of an expert to prove identification by voice was admissible, but the facts of that case were very different from the facts of this case because the expert had been called not, as it was said he should have been in this case, in order to criticise the identification parade but to carry out the voice identification himself.

Mr Duckworth in his report sets out his qualifications and the work that he had carried out in this case. On page 2 he says:

"There are a number of similarities between Mr Hersey and the remaining 11 speakers on [the video] in respect of the accent used. It is reasonable to suggest that all the speakers come from the same broad geographical region.

There are a number of similarities in respect of the loudness of the speech and the rate of speech (excluding pauses) between Mr Hersey and the remaining 11 speakers.

There are a number of differences in the content, pausing and pitch used by Mr Hersey and the remaining 11 speakers."

Mr Duckworth then spelled out those differences and came to his conclusions. He said:

"There are three reasons why Mr Hersey's speech is likely to stand out intrinsically from the remaining 11 speakers on [the video].

1. his utterance makes sense, the others sound rather odd;

2. he has moments of non-fluency in his speech which are characteristic of unrehearsed conversational speech. The only other speaker who achieves this effect is No. 5 who has a long pause and an unintended repetition;

3. his pitch is significantly lower than any of the other speakers. This alone is likely to make his speech stand out from the others in the line up.

On this basis it would be fair to suggest that there are marked difference in aspects of speech and of delivery between Mr Hersey and the 11 other speakers ...."

In those circumstances Mr Donne submits that the evidence of Mr Duckworth was admissible and ought not to have been excluded by the judge. Mr Duckworth, he submits, was giving evidence as to fact, in particular his criticisms of the procedure used which he (Mr Donne) says was intrinsically flawed. The defence, it is said, wanted to call evidence in relation to the techniques that had been used. As a matter of generality in the field of voice identification, there will undoubtedly be cases in which the jury will need the assistance of an expert and there will be other cases when the matters in issue are matters which are well within a jury's capacity to form their own conclusions. Mr Malcolm QC for the Crown in his submissions to us said that all the matters raised in Mr Duckworth's report are in reality matters which would be well within the knowledge and experience of 12 jurors. There would obviously be little difficulty for a jury, particularly assisted by counsel and the judge, in forming the conclusion that the appellant had a deep voice, whereas most of the others on the parade did not. Certainly the jury did not need expert evidence for that opinion. It is important in all fields of the law, the criminal field included, that there should not be a proliferation of expert evidence and expert witnesses. In our judgment Mr Malcolm's submission is correct. There is nothing in our judgment in Mr Duckworth's report which would not be within the experience of the jury, given the help as they were given. In our judgment the judge was right in this case to hold that the evidence of Mr Duckworth was inadmissible.

Mr Donne then makes certain criticisms of the judge's summing-up. His main criticism relates to the judge's summing-up in relation to the manner in which the jury should approach the evidence that was given in relation to the identification parade. There are two issues that the judge had to deal with in relation to the jury's overall approach: first, their approach to the evidence as to recognition overall; and second, their approach to the identification parade. Mr Donne makes no criticism of the judge's direction in relation to the first limb, but he does criticise the approach to the second limb.

There is not a great deal of authority as to the manner in which a judge should direct a jury in respect of voice identification, particularly where the witness is indicating that he recognises a voice which is known to him. In our judgment, a judge should tailor his directions to the jury on the lines indicated by this court and in the specimen directions issued by the Judicial Studies Board in respect of visual recognition or identification, but tailored for the purposes of voice identification or recognition. The judge should follow, suitably adapted, the guidelines laid down in R v Turnbull (1977) 63 Cr App R 132 and the cases which have followed. Above all, it is vital that a judge should spell out to the jury the risk of a mistaken identification, the reason why a witness may be mistaken, pointing out that a truthful witness may yet be a mistaken witness, and dealing with the particular strengths and weaknesses of the identification in the instant case.

As we indicated in the course of submissions, it is not possible wholly to divorce the judge's direction to the jury in relation to their overall approach to the issue as to recognition from his direction in relation to the identification parade. The judge gave the jury a very full direction in relation to the general issue. He started on page 10 of the transcript and he then said on page 11:

"In a case of visual identification the courts have laid down that juries shall be told that there is a special need for caution before convicting an accused person in reliance solely on the correctness of an identification. They have to be warned that mistakes can happen in the case of a visual identification and that mistakes have happened. We all know that. They are broadcast, talked about, written about and rightly so.

Juries are reminded and have to be reminded in visual identification cases that a conscientious, responsible and fair minded person can make a mistake when it comes to identification as well as an impulsive, irresponsible and not very bright person. Juries are told that they must take on board the reluctance of even a perfectly fair minded person once he has made up his mind about the matter to admit that he may be wrong in his identification of the person. This is a fact of life that people who had made an identification are reluctant to show that there is a possibility of their having made a mistake. It may not apply to everybody. Juries have to be told that they should be on their guard against accepting and acting upon the witness's identification simply because they were impressed by him as a witness. All these observations which apply to a visual identification parade apply equally to a voice identification. You should have regard to all those points that I have mentioned."

The judge then set out the particular aspects of this case on which he suggested that the jury should concentrate and he then said:

"To sum it up, this case rests upon Mr Whiting's identification of the defendant's voice. The law does not require that there shall be any other evidence apart from a witness's identification. You may convict solely upon Mr Whiting's identification but please bear in the forefront of your minds all those matters of warning that I have mentioned to you. That is all I want to say about that and I am going to remind you now of the evidence."

In relation to the identification parade it is of particular importance that in the passage which I have just read the judge told the jury that a perfectly fair-minded person who has made up his mind will be reluctant to admit subsequently that they may be wrong and people who have made an identification are reluctant to admit that there is a possibility of their having made a mistake. That direction carries through to the direction that the judge gave in relation to the identification parade itself, on page 17 where the judge said:

"I have now almost finished my summing-up. I must say a word about the voice identification parade of which criticism has been made by Mr Stopa. Let me say a word or two. There is some substance, you may thing, in Mr Stopa's point that Mr Whiting in identifying the accused on the tape when the tapes were played to him was doing no more than identifying the voice of the accused which was a voice that was familiar to him as being the voice of a customer in his shop. Mr Stopa says that the crucial point in this case is whether he was correct in his identification of the voice on the evening of the robbery. He says that there is no inconsistency between his having been wrong in that identification but right in his identification of the accused as the speaker on the tape. That is a point that you should have regard to. It is entirely up to you whether you find that the recognition by Mr Whiting of the accused's voice when he listened to the tape adds to the strength of the prosecution case against the accused that he was the robber."

Mr Donne criticises that part of the summing-up as being inadequate. He submits that the judge should have given additional guidance to the jury in relation to the identification parade along the lines of Mr Duckworth's report and the evidence that he had heard on the voir dire. Mr Donne submits that the judge should have told the jury that the real danger in a case such as this is that the witness recognised the voice not because it was the voice he had heard at the scene, but because it was a voice that he knew well. In the passage that we have read the judge told the jury of Mr Stopa's submission, which they had heard shortly before, to the effect that the witness was doing no more than identifying a voice which was a voice that was already familiar to him. In our judgment the judge dealt fully and adequately with this aspect of the case and no realistic criticism can be made of his directions.

Certain other criticisms have been made of the summing-up, but we do not think that there is any substance in them.

Finally, Mr Donne made an application to the court for leave to put in the further evidence of Dr Nolan who is, like Mr Duckworth, an expert in voice recognition and identification but with greater qualifications. The application is made pursuant to section 23 of the Criminal Appeal Act 1968, which reads:

"(1) For the purposes of this Part of this Act the Court of Appeal may, if think it necessary or expedient in the interests of justice --

....

(c) receive any evidence which was not adduced in the proceedings from which the appeal lies.

(2) The Court of Appeal shall, in considering whether to receive any evidence, have regard in particular to --

(a) whether the evidence appears to the Court to be capable of belief;

(b) whether it appears to the Court that the evidence may afford any ground for allowing the appeal;

(c) whether the evidence would have been admissible in the proceedings from which the appeal lies on an issue which is the subject of the appeal; and

(d) whether there is a reasonable explanation for the failure to adduce the evidence in those proceedings."

Quite clearly the evidence of Dr Nolan is capable of belief; his evidence fulfils that criterion. It is also true that the defence were unaware of the existence of Dr Nolan at the time of this hearing. However, the defence did have available expert evidence and they were well aware of the availability of such evidence. In our judgment this evidence does not fulfil the requirement of section 23(2)(d). However, we would not refuse to admit the evidence on that ground alone. We have read Dr Nolan's report with great care and there is no need for us to go through it in this judgment. In our view Dr Nolan's evidence would not have been admissible under section 23(2)(c) for the same reasons that we consider that Mr Duckworth's evidence would not be admissible. However, by far the most important aspect in our judgment is that this evidence would not afford any ground for allowing this appeal.

For all those reasons we do not consider that there is anything unsafe about this conviction and the appeal against it is accordingly dismissed.

_____________________________________