APPEAL COURT, HIGH COURT OF JUSTICIARY
Lord Justice Clerk Lord Reed Lord Brodie	[2011] HCJAC 40 Appeal No: XC408/09 OPINION OF THE COURT delivered by THE LORD JUSTICE CLERK in the REFERRAL BY THE SCOTTISH CRIMINAL CASES REVIEW COMMISSION in causa WILLIAM GAGE Appellant; against HER MAJESTY'S ADVOCATE Respondent: _______

For the appellant: Scott, QC; Mitchell; Aamer Anwar and Co, Glasgow

For the Crown: Shand QC, AD; Crown Agent

15 April 2011

Introduction

[1] On 9 February 2004 at Glasgow High Court the appellant was convicted on the following charge:

"On 7 March 2002 at Acacia Way, Westburn, Cambuslang, Glasgow, you did while acting along with another or others meantime to the prosecutor unknown, assault Justin John McAlroy ... and repeatedly discharge a firearm at him and shoot him on the head and body whereby he was so severely injured that he died on 8 March 2002 ... and you did murder him."

[2] His appeal against conviction was refused (Gage v HM Adv [2006] HCJAC 7).

[3] The Scottish Criminal Cases Review Commission has referred the case to this court. The appellant has taken the opportunity of the referral to table several grounds of appeal against conviction. They relate to the reasonableness of the jury's verdict, dock identification and alleged misdirection. These grounds remain to be considered. The appellant now seeks leave to add a new evidence ground. We have had a hearing on that question.

The evidence

[4] The deceased was an experienced criminal who trafficked in drugs. He was murdered outside his house. The evidence relevant to the proposed ground of appeal was as follows.

The identification of the gunman

[5] Several witnesses saw a man running from the scene. He wore dark clothing. His face was partially obscured by a scarf, hood or something similar. He got into the passenger seat of a white car, which was driven off in the direction of the M73/M74.

[6] Only one witness identified the appellant as that man, namely the deceased's widow Tracy McAlroy. In her police statements and in her Crown and defence precognitions Mrs McAlroy said that she could not identify the man or did not think that she would be able to do so in court. There was no identification parade. The police had arranged one but the appellant refused to take part in it because he thought that the stand-ins were much younger than he.

[7] Mrs McAlroy said that on hearing what were evidently gunshots, she opened her front door and looked out. She saw a man running away at the end of her driveway. He looked at her when she opened the door. He wore a hood. Only part of his face, from the forehead to the nose, was visible. The advocate depute asked her if she had seen anyone who resembled that man. She said yes, and indicated the appellant in the dock. She said that she did so "because I'll never forget the eyes ... They eyes, I think I seen them that night but I'm not 100 percent sure." She said that they were "scary eyes." She confirmed that it was the appellant's eyes that caused her to say that there was a resemblance. There was no objection to this line of evidence. Mrs McAlroy had not previously mentioned the man's eyes.

[8] Stephen Madden was the only other witness to see the gunman's face. He said that he saw the gunman inside the white car removing what he thought was a ski mask. He did not identify him in court.

Identification of clothing

[9] The shooting occurred at about 10pm. At 10.49pm police were alerted to a white Saab that was on fire at Balcurvie Road, Easterhouse. It had been abandoned there at least 10 minutes earlier. In the Saab the police found gloves, a black nylon hooded jacket, a scarf or snood, a drinks bottle and a radio scanner that was switched on and tuned to the police wavelength. The appellant's DNA was on the neck of the bottle and on the hood and cuffs of the jacket. On the gloves and snood there was his DNA and the DNA of at least two other unknown persons. The jacket and snood also bore a small amount of firearms discharge residue of the same type as that found at the locus.

[10] Tracy McAlroy told the police that the man whom she saw running away was wearing a padded jacket with an 'Eskimo-type' hood. Her description did not fit the jacket found in the Saab. In May 2002, following the aborted identification parade, the police took her into a room in the police station and showed her a mannequin dressed in the clothing found in the Saab. She identified the clothing as that worn by the man whom she had seen. There was no objection to this evidence. According to the trial judge's report, the mannequin had very prominent eyes. Other eye-witnesses were shown the jacket but only one, Phyllis Craig, said that it was of a similar type to the jacket that she had seen the man wearing.

The identification of the car

[11] The witness Charles Bowman was working nearby. He heard a car screeching away. He described it to the police as a white car with a spoiler. He thought that it might have been a Volvo. When the police showed him a number of Volvos, he said that he thought that he had seen a Volvo 440. Later, at Paisley police station he was shown the Saab that was found at Easterhouse. He said that it was similar to the car that he had seen. The Saab did not have a spoiler. There was no objection to this evidence. Other witnesses gave varying descriptions of a white car that they said that they had seen near the locus. One of them, James Kearns, said that he was familiar with Saab cars and would have noticed if the car that he saw had been a Saab. He too thought that the car had had a spoiler.

The referral

[12] The Commission summarises its reasons for referring the case as follows:

"The Commission considers that the combination of difficulties with each of these adminicles heightened the importance of the jury being given specific directions on the evidence in question. In relation to the dock identification, the Commission has reached the view that the complete absence of directions on the dangers inherent in the dock identification (such as the increased possibility of mistaken identification), on the need to consider the fairness to the applicant of the process and on the need for particular caution in the circumstances amounts to a misdirection by the trial judge.

That misdirection must be seen in light of the obvious difficulties with, and the absence of similar directions regarding, Mr Bowman's confrontation with the Saab and Mrs McAlroy's confrontation with the mannequin; and the limited challenges put forward by the appellant's solicitor advocate to the three strands of contested evidence in cross-examination and in his speech (as highlighted above); and the potential importance of each of these three strands of evidence to the case as a whole (the Crown expressly relied on them and without them the case, though still sufficient in law, was entirely circumstantial and was undoubtedly much weaker and more tenuous).

The Commission believes that, when the misdirection is considered in light of these factors, a miscarriage of justice may have occurred. The Commission also believes that, when seen in the context of the trial as a whole, the cumulo effect of the Crown having relied on the three strands of contested evidence may have amounted to a violation of the applicant's right to a fair trial under article 6, and that a miscarriage of justice may have occurred as a result (Referral, para 247)."

The proposed new ground of appeal relates directly to these reasons.

The proposed new ground

[13] The appellant seeks to lead the evidence of Professor Tim Valentine, Professor of Psychology at Goldsmiths, University of London. We understand that Professor Valentine is a recognised expert on eye-witness identification, that he has advised the Home Office and police forces in England in relation to identification questions and that his research has informed and influenced the English code of practice governing the conduct of identification parades. According to the proposed ground of appeal, the new evidence will cover:

"The recognised and undisputed factors from psychological science which are relevant to the circumstances of this case and which render the identification evidence made here more or less reliable.

In this case the crucial circumstances relied upon concerned facial identification of the accused (resemblance) and identification of objects. In summary consisting of the evidence that the accused resembled the gunman; that the clothing shown was that worn by the gunman or similar to that worn by the gunman and the motor car shown was the car or similar to the car abandoned in Easterhouse.

The relevant risk factors which arise from this evidence include:-

the opportunity to view the gunman by witnesses

the effect of stress upon witnesses

the delay before the identification procedures were carried out

the potential effect of feedback or post-event information acquired by the witnesses

the potential effect of the suggestible procedures carried out.

The risks to the reliability of such identification evidence which can arise from the way it is obtained and in particular from the identification procedures followed here. Notably the risks arising from the procedures detailed at 3.1-3.3 above, namely

the procedure of showing a single object or motor car to the witness

the procedure whereby the eye-witness was confronted with a mannequin dressed in clothing from the accused

the procedure of dock identification - absent a parade"

[14] Professor Valentine's report relates to the principal psychological properties of human memory; psychological research on facial identification; the psychology of object recognition; and what he calls the most salient factors relevant to eyewitness testimony in this case. He has reviewed the trial judge's report, the Commission's statement of reasons, the report of the aborted identification parade, the statements and transcripts of the evidence of several Crown witnesses, the Crown and defence speeches and the trial judge's charge. His conclusions are:

"Psychological science shows that the following procedures used in this case can result in memory distortion through suggestion.

Showing a single item of property to a witness for identification despite considerable mismatch of the property with the witnesses description. This is leading post-event information and may have affected identification of the car.

Confronting a witness with mannequin dressed in clothing associated with the suspect. By inducing a strong emotional response and by the strongly suggestive nature of the procedure it is likely to result in memory distortion and source misattribution of memory. This factor may have affected the identification of the clothing.

It is inappropriate to attempt an identity parade for witnesses who have not described the perpetrator's face or hair and say they can't identify the man they saw.

Mistaken identification of foils occurs frequently at a line-up. A dock identification does not provide an adequate test of memory because such an error cannot be detected. A request to make a dock identification is a highly suggestive procedure. The outcome will be influenced by the witness' willingness to testify, which is known to be influenced by feedback of prior identification and, in my opinion, is likely to be influenced by knowledge of other evidence against the defendant" (underlining in the original).

Submissions for the appellant

[15] Counsel for the appellant submitted that Professor Valentine's evidence was admissible. The psychological science of eyewitness identification was well-established (cf Lord Devlin, Report of the Committee on Evidence of Identification in Criminal Cases, 1976 Cmnd 338). It was not essential that the proposed new evidence was necessary for the proper resolution of the dispute (cf Wilson and Murray v HM Adv 2009 SCCR 666, para [58]; Grimmond v HM Adv 2001 SCCR 708). In England the test was whether the evidence would assist the jury in understanding a fact in issue (R v Turner [1975] QB 834, R v Raghip The Times, 9 Dec 1991). A jury would know generally of some factors that could affect an identification, but they would have no way of measuring the effect of them. Expert evidence could give guidance on how significant they were. Other factors would be outwith a jury's knowledge; for example, the fact that memory involved an active process of reconstruction that could be affected by suggestion. Research showed that lay people were unaware of some factors affecting identification and that there were common misconceptions, for example about the effect of stress.

[16] Identification evidence had well-recognised dangers. A specific risk arose here because there had been no identification parade and Mrs McAlroy had previously said that she could not identify the man. While the jury could see that the process of dock identification was suggestive, they did not know the extent to which science regarded such identification as unreliable. The same applied to the identification of the clothing and the fire-damaged Saab. The police had used highly suggestive procedures. The trial judge's charge did not address the possibility that the witnesses' memory could have been distorted.

[17] Credibility and reliability were for the trier of fact, but expert evidence was admissible if there were special features that were likely to be unknown to the jury (McBrearty v HM Adv 2004 JC 122; AJE v HM Adv 2002 JC 215; HM Adv v A 2005 SCCR 593); for example, an accused person's susceptibility to pressure when questioned by the police (HM Adv v Gilgannon 1983 SCCR 10; Gilmour v HM Adv 2007 SCCR 417). In the present case the proposed evidence related to conditions that could affect any witness. It was therefore easier to say that it did not usurp the jury's function. The well-known risk that eyewitness identification evidence could cause a miscarriage of justice was itself a reason to admit Professor Valentine's evidence. The Lord Justice General's 1977 Practice Note on the directions to be given to the jury provided only guidance. While the trial judge's directions had gone further than normal, there had been no specific warning on the dangers of dock identification in the absence of an identification parade (Holland v HM Adv 2005 SC(PC) 3, para [58]).

Submissions for the Crown

[18] The advocate depute submitted that expert evidence regarding the credibility or reliability of a witness's evidence, or on aspects of human nature within the limits of normality, was in general inadmissible (Walkers, Evidence, 3rd ed, para 1.6.3). It was admissible only if it furnished scientific information on matters outwith the jury's knowledge. In all of the cases in which expert evidence had been admitted, it had related to matters outwith the jury's knowledge or to a specific fact in issue (eg AJE v HM Adv supra; Campbell v HM Adv 2004 SCCR 220; Gilmour v HM Adv supra). The test was one of necessity, not assistance or usefulness (Dickson, op cit, para 397; Wilson and Murray v HM Adv, supra, at para [58]; R v Raghip, supra; Walkers, op cit, para 16.3.3). It was not met here. Professor Valentine's report added little, if anything, to what the jury had already known. It did not provide any measure by which the court could assess the degree to which identification evidence might be unreliable (cf Campbell v HM Adv, supra). His references to academic studies provided nothing concrete with which to assess the effects of delay or the conditions of observation. The effect of feedback and the lack of a correlation between confidence and accuracy were not in issue. The proposition that memory was liable to suggestion was not outwith the jury's experience. Professor Valentine's opinion that dock identification was a suggestive procedure was irrelevant, as it had been clear to the jury that Mrs McAlroy had made only a tentative identification based on the appellant's eyes. All of the issues to which Professor Valentine referred had been covered by the defence speech and by the trial judge in his charge.

Discussion

Professor Valentine's conclusions

[19] The advocate depute drew to our attention certain remarks reported by the Commission as having been made by Professor Valentine, in an interview for the purposes of a television programme. According to the Commission, Professor Valentine said inter alia that he accepted that Tracy McAlroy's first statement was obtained in stressful conditions on the night of the murder; that he would look to her early statements, taken within a few days of the murder, as the most reliable source of information; that only a small proportion of people would correctly pick out at a line up a person whom they had seen in the circumstances in which Tracy McAlroy had seen the gunman; that her confrontation with the mannequin was a highly suggestive procedure and that there could be no better way of planting misleading information into a witness's memory. The absence of any reference to the man's eyes in her police statements suggested that the confrontation with the mannequin, which had prominent eyes, might have contaminated her memory.

[20] From these and other points Professor Valentine concluded that there was no reliable identification evidence linking the appellant to the scene of the crime; and that the witnesses' statements in regard to the make of the white car seemed to be unreliable, apart from that of James Kearns, whose evidence on the point was most likely to be reliable. We regret to say that in expressing these conclusions, although they are not set out in the report itself, Professor Valentine has, in our view, compromised his position as an expert witness in this case.

The admissibility of expert evidence - the test of necessity

[21] Questions of credibility and reliability are pre-eminently matters for the tribunal of fact. Our system of jury trial proceeds on the basis that jurors, as people of ordinary intelligence and experience, are capable of assessing the credibility and reliability of a witness without expert assistance. For opinion evidence to be admissible its subject-matter must fall outwith the jury's knowledge and experience and be based on a recognised branch of knowledge (Dickson, Evidence, para 397). It was accepted on behalf of the appellant that opinion evidence must not usurp the function of the jury (Davie v Mags of Edinburgh 1953 SC 34).

[22] We do not accept the submission that evidence of the kind proposed is admissible if it might be useful to the jury. That test is not supported by authority and, in our view, would be unworkable in practice. Expert evidence is admissible only if it is necessary for the proper resolution of the dispute. In particular, in a case of this kind it is necessary only if the tribunal of fact would be unable to reach a sound conclusion without it (Wilson and Murray v HM Adv, supra, para [58]; Dickson, op cit, para 397). That will be the case only if there are special features pertaining to a witness or to his evidence that are likely to be outwith the jury's knowledge or experience (HM Adv v A, supra, at para [11]).

[23] For these reasons this court has held that psychiatric or psychological evidence would have been admissible on the question of the reliability of a child witness in light of the interviewing techniques used (AJE v HM Adv, supra) and on the question whether a witness suffered from a severe personality disorder and was a pathological liar (McBrearty v HM Adv, supra). This court has also admitted such evidence on the question of an appellant's susceptibility to pressure when he was being questioned by the police (Gilmour v HM Adv, supra). A trial judge has admitted such evidence on the question whether a complainer who suffered from bi-polar illness (manic depression) exhibited false memory syndrome (HM Adv v A, supra); but it was not admitted in HM Adv v Grimmond (supra) on the question of the reliability of child complainers where there was no suggestion that they were other than ordinary, normal children.

[24] In Campbell v HM Adv (supra), four police officers claimed to have almost identical verbatim recall, without having conferred, of incriminating statements allegedly made by the appellants. The court admitted psycho-linguistic evidence on the likelihood that several people who heard a statement could each recall it in almost identical terms. It refused to admit such evidence in Coubrough v HM Adv (2008 SCCR 317) where the police witnesses did not claim to have verbatim recall and admitted that they had conferred in recording the appellant's incriminating statement.

[25] Counsel for the appellant relied on R v Turner (supra) and R v Raghip (supra); but these decisions support a test of admissibility based on necessity rather than mere usefulness. The issue in R v Turner was whether psychiatric evidence should have been admitted to support the appellant's defence of provocation, there being no indication that he suffered from a mental disorder. Lawton LJ put the point in the following way:

"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 given dressed up in scientific jargon it may make judgment more difficult. The fact that an expert witness has impressive qualifications does not by that fact alone make his opinion on matters of human nature and behaviour within the limits of normality any more helpful than that of the jurors themselves; but there is a danger that they may think it does" (at p 841).

[26] In R v Raghip the Court of Appeal held that psychological evidence was admissible to prove that the appellant was of low intelligence and was abnormally suggestible. It was relevant to the question whether the trial judge should have excluded evidence of his confession on the ground that it was obtained in circumstances likely to make it unreliable. Without evidence from a psychologist the jury would have found it impossible to assess the appellant's abnormally low intellect, still less his abnormal suggestibility, from the manner in which he gave evidence. That was the approach taken in Gilmour v HM Adv (supra). Such evidence was directly relevant to a fact in issue.

Expert evidence on eyewitness identification

[27] Expert evidence on eyewitness identification is admitted in some jurisdictions, but the practice is far from universal. In our opinion, the objections to such evidence are straightforward.

[28] It is inherent in the system of trial by jury that the question of the credibility and reliability of a witness's evidence, in general or on a specific point, is a question for the jury to decide. That question is to be decided on the basis of the jurors' own assessment of the witness and of his evidence by reference to other relevant evidence in the case. Each juror brings to bear on that judgement his experience of life and human affairs.

[29] On the particular issue of identification evidence, the court has its own safeguards. In a prosecution that rests on eye-witness identification the risk of a miscarriage of justice is notorious. The invariable practice in our courts is that the trial judge gives the jury a specific and thorough direction that warns them that in certain circumstances such evidence may be unreliable and refers by way of example to specific considerations that might be thought to affect the reliability of an identification made by an eye-witness. The strength of that safeguard has been recognised in other jurisdictions in which expert evidence on the point is excluded (eg Smith v The Queen (1990) 64 ALJR 588, Deane J (High Court of Australia)).

[30] There is the further safeguard that the defence can highlight the potential unreliability of eye-witness identification in cross-examination of the relevant witnesses and lead evidence of objective physical factors that might affect the reliability of the identification in question; for example, distance, the length of time for which the witness could see the assailant, the visual acuity of an elderly witness (cf R v Toohey ([1965] AC (HL) 595, Lord Pearce at p 608), the lack of street lighting, whether the assailant was masked and so on. These safeguards are consistent with, and stem from, the basic approach of our law which is to recognise that a juror's judgment on the reliability of an identification is a matter of life experience.

[31] For these reasons alone, we consider that our law is right in refusing to admit evidence of the kind proposed. We are opposed to any relaxation of the existing rule.

[32] Moreover, we are opposed to the admission of such evidence on other, more practical, grounds. If evidence of this kind were to be admissible, it is a matter of certainty that the defence would lead psychological evidence in most trials in which identification was an issue, if only for fear that a failure to do so would lead to an Anderson appeal. Inevitably, the Crown would lead psychological evidence in rebuttal. In the result, trials would be considerably prolonged (cf R v Smith, 2000 116 A Crim R 1, Smart AJ at para 69 (New South Wales Court of Criminal Appeal)). The focus of the trial would shift as the jury had to adjudicate on the conflicting expert views and on the cogency of the research evidence on which they were based. Expert evidence of that kind would of course be centred on the weaknesses of identification evidence rather than on the factors that enhance its cogency in any individual case. In our opinion, it would create a climate of disbelief (cf State v Young (2010 WL 1286933 (La. 2010) (Supreme Court of Louisiana)). There would be a serious risk that a juror who was impressed by an expert would be diverted from his own duty to make a proper appraisal of the identification in question.

[33] Moreover, if such evidence were admissible in solemn procedure, it would be admissible too in trials under summary procedure, where the sheriff or justice alone is the trier of fact. These constitute about 95% of trials in Scotland.

Is Professor Valentine's evidence admissible?

[34] In our opinion, Professor Valentine's evidence is inadmissible because it does not meet the test of necessity. It is not suggested that any of the eyewitnesses suffered from some condition that affected his or her vision or power of recall. On the evidence summarised by the Commission, the eye-witnesses appear to have been normal people who witnessed an incident in imperfect conditions. This is not a case in which there is a dispute about the circumstances in which witnesses saw or heard the incident about which they give evidence. The relevant witnesses conceded that they caught only a fleeting glimpse, at night, of the gunman or of the car.

[35] Professor Valentine's report describes problems of perception and recall that affect all witnesses. He highlights weaknesses in the identification evidence that was led, without objection, at the appellant's trial; but these weaknesses were already known to the jury. Counsel for the appellant conceded that they are matters of common sense. They were brought out in cross-examination. The trial judge dealt with them in his charge. In the charge, as counsel for the appellant accepted, the trial judge went further than normal in highlighting the problems inherent in identification evidence. It is not for Professor Valentine now to offer a view on whether a particular witness was reliable. That was for the jury.

[36] Counsel for the appellant argued that without expert assistance, a jury could not measure the extent to which a witness's reliability was compromised; but in our view Professor Valentine's evidence would not supply that deficiency. In all of the cases in which expert evidence has been admitted in our courts, the evidence was specific to the facts of the case, and usually specific to a particular witness. Professor Valentine has conducted no case-specific tests or research. He can only alert the court to some of the factors that might in general affect the reliability of identification evidence. Much of his report has no bearing on the facts of this case. For example, there has been no suggestion that any witness's evidence was distorted by feedback or by any significant delay before identification procedures were carried out. He expresses the opinion that there is no correlation between the confidence with which an identification is asserted and its accuracy; but that is not in point. Tracy McAlroy made only a tentative identification based on resemblance, which was limited to the appearance of the gunman's eyes. That was obvious to the jury. It was for the trial judge to remind the jury of the dangers inherent in such identification, as he clearly did.

Disposal

[37] We refuse to admit the proposed ground of appeal and shall continue the case for consideration of the grounds of referral and the existing grounds of appeal.