For many years, eyewitness identification evidence has played a crucial role in the formulation of cases in the criminal and civil justice systems in Canada. It forms “an important component of evidence, prosecution and plea negotiation”, and often proves to be accurate and valuable evidence that leads to just convictions and acquittals. Despite its potential usefulness, however, “identification evidence requires careful and cautious consideration... [as] it can [also] easily lead to wrongful convictions.”[1]

I.     INTRODUCTION

Perjury is a crime, because lying under oath can subvert the integrity of a trial and the legitimacy of the judicial system. However, perjury is defined as knowingly making a false statement—merely misremembering is not a crime. Moreover, the jury makes its determinations of witness credibility and veracity in secret, without revealing the reason for its final judgement. Recognizing the fallibility of witness memories, then, is especially important to participants in the judicial process, since many trials revolve around factual determinations of whom to believe.[2]

The advent of DNA evidence in the courtroom, and the resultant exonerations of people convicted based mainly on eyewitness testimony, have again brought the fallibility of eyewitness evidence to the forefront.  A series of commissions looking at several such cases have already taken place in Canada, including The Commission on Proceedings Involving Guy Paul Morin,[3] The Royal Commission on the Donald Marshall Junior Prosecution,[4] The Commission of Inquiry into the Wrongful Conviction of David Milgaard,[5] and The Inquiry Regarding Thomas Sophonow.[6] In addition, the issue of wrongful conviction was examined generally by the FPT Heads of Prosecutions Committee Working Group[7].  All of these commissions made recommendations as to how the criminal justice system could best avoid the pitfalls of the past. 

II.     LINE-UP PROCEDURES

Thomas Sophonow was tried and convicted for the murder of 16 year-old Barbara Stoppel,[8] a crime of which he was later exonerated.  One of the concerns raised by Justice Cory during the subsequent inquiry into Sophonow’s conviction was the way in which the line-ups were conducted. It appeared that in the initial line-up, which was comprised of photographs, Sophonow's photograph was a standout: it alone had a border around it and was the only photograph taken outside.  Moreover, in the line-up photos, he was the only person wearing a cowboy hat and standing in front of a pickup truck.  The manner in which the line-up was conducted prompted Justice Cory to issue a series of recommendations concerning the conduct of a photograph line-up.  Ten years after the release of the report, it appears to have had some positive impact.  Two of the recommendations, the use of a double-blind procedure and the use of a sequential line-up procedure, have gained currency.  In a double-blind procedure, the investigator and the witness are equally unapprised as to whether or not the suspect is actually present in the line-up; in the sequential line-up, the suspects, or their photographs, are presented one at a time.  Research has demonstrated that both of these procedures help to minimize false identifications.[9] A survey of police officers conducted by Ron Lindsay of Queen’s University in 2011 suggests that upwards of 72 percent of the officers who responded to the survey use a double-blind technique, while 81 percent use a sequential line-up process.[10]  Other recommendations, however, have not been as readily embraced. 

III.      INTERVIEW TECHNIQUES

Almost all the reports mentioned made recommendations concerning interviews. More specifically, most recommended that interviews be video or audio-taped. The Report of the Kaufman Commission on Proceedings Involving Guy Paul Morinmade specific recommendations regarding interviews:

Recommendation 101 - Police protocols for interviewing to enhance reliability

The Ministry of the Solicitor General should establish province-wide written protocols for the interviewing of suspects and witnesses by police officers. These protocols should be designed to enhance the reliability of the product of the interview process and to accurately preserve the contents of the interview.

Recommendation 102 - Training respecting interviewing protocols

All Ontario investigators should be fully trained as to the techniques which enhance the reliability of witness statements and as to the techniques which detract from their reliability. This training should draw upon the lessons learned at this Inquiry.

Financial and other resources must be provided to ensure that such training takes place.

Recommendation 103 - Prevention of contamination of witnesses through information conveyed

Police officers should be specifically instructed on the dangers of unnecessarily communicating information (known to them) to a witness, where such information may colour that witness’ account of events.

Recommendation 104 - Prevention of contamination of witnesses through commentary on case or accused

Police officers should be specifically instructed on the dangers of communicating their assessment of the strength of the case against a suspect or accused, their opinion of the accused’s character, or analogous comments to a witness, which may colour that witness’ account of events.[11]

Recommendations regarding the training of police officers in the proper use of interviewing techniques were also made by the FTP Heads of Prosecution Committee Working Group (the Committee).  The Committee suggested the development of a protocol to ensure the most accurate testimony possible by the witness.[12]  In order to obtain the most complete and accurate recollection from a witness, Kevin Colwell, Cheryl Hiscock, and Amina Memon suggest that the interviewing officer should use open-ended questions and allow the witness to answer these fully without interruptions.[13] Elizabeth Loftus and John Palmer have demonstrated that the type of questions asked by an investigating officer can affect a witness’ memory of an event.[14]  In their classic study, participants were shown a film of an automobile accident and were then asked, “About have fast were the cars going when they _________ in to each other?”  Each of the participants received one of the following verbs completing the question: “collided,” “bumped,” “hit,” “contacted,” or “smashed”.  Those participants who received the word “smashed” reported that the cars were travelling at a higher rate of speed then those who received the other possibilities.[15]  More importantly, when asked one week later to recollect what they had seen, these participants claimed to have seen glass at the scene of the accident.[16] 

Recently, two of the authors of this article were witness to a car accident and were impressed by the way in which the police officer handled the interview.  The officer simply asked them to write down their contact information and then to write down, in as much detail as possible, what they had witnessed.  Unfortunately, this may be the exception rather than the rule.  A review of the interviewing practices of police officers in Canada suggests that officers tend to use closed-ended questions and will often interrupt the witness being interviewed.[17]  Indeed, police officers in Canada are still trained in the use of an accusatory interrogation technique, which condones the use of deception by the officer when interviewing suspects, even though there is a much more ethical, non-accusatory alternative that has been adopted in some jurisdictions in the United Kingdom.  This alternative technique, called P.E.A.C.E. (Preparation and Planning, Engage and Explain, Account, Closure and Evaluate) has proven to be as effective at obtaining confessions, while reducing the possibility of false confessions brought about by coercion.[18]

IV.     PROCEDURAL OR "SYSTEM" VARIABLES

The slow and inconsistent adoption of some of the recommendations that were intended to improve the system results largely from the very fact that they are only recommendations and, therefore, not binding.  The decision as to whether or not to adopt the recommendations is left up to the different police forces and jurisdictions across Canada.  In fact, in a recent article published in the Criminal Law Quarterly, Narissa Somji argues that it is time to turn the recommendations into legislation.[19] Though Somji suggests that strides are being made towards changes in the judicial process and that legislation may, indeed, help the law to catch up with science, legislation will deal only with procedural, or “system,” variables, and this is only half of the battle. There remains a series of “estimator” variables that can impact a person's memory for an event, over which the investigative process has no control.

A.     Estimator Variables

In response to the comments made by those involved in the judicial system suggesting that the research evidence on eyewitness memory was not readily applicable to the judicial process, psychologists began to gear their testing towards real-world examples.  Gary Wells, a prominent researcher in the area, went as far as to classify all variables that can influence a witness’s memory into either “system” or “estimator” variables.[20]  System variables are those procedural variables within the direct control of “the system.”[21] These variables are directly related to the investigative process.  Estimator variables, on the other hand, are those variables that occur during the crime itself, and are, therefore, beyond the control of the system.[22]  The effect of these variables can only be estimated.  Such variables would include factors such as the time of day, the position of the witness, the line-of-sight, and the age of the witness.  Thus, in the case of estimator variables, direct legislation would not be an option, and yet the impact of estimator variables is an important aspect of the investigative process.

Two of the authors of this article had a first-hand experience which underscored the importance of estimator variables.  Having witnessed the car accident mentioned earlier, they gave a written statement to the police.  The accident involved a mini-van which was travelling directly in front of them.  It was t-boned by a car attempting to cross traffic to reach a side-road opposite.  The authors had a front-row seat to the impact which rolled the mini-van.  In their report, they stressed their position in relation to the accident.  There were many witnesses to the accident, as the road was busy and lined with fast-food restaurants.  Following the impact, scores of people came to help the victims. Concerned for the well-being of those involved in the accident, the authors bought the local newspaper the following day, hoping to hear that those involved were doing well.  However, had it not been for the date, place, and time of the accident reported, the authors would not have recognised this accident as the one they had witnessed.  Obviously, the angles and positions of the witnesses interviewed by the newspaper reporter had a profound effect on what they thought they had seen.

B.     The Impact of Estimator Variables

The importance of introducing the impact of estimator variables on an eyewitness’s memory for an event is obvious, and a recent US case provides a case-in-point.[23]  In September 2011, Troy Davis was executed for the murder of Mark McPhail, an off-duty police officer.  While moonlighting as a security guard, McPhail witnessed a vagrant being beaten and intervened on his behalf.  In his attempt to protect the vagrant, McPhail was shot, and later died in hospital.  The gun was never recovered.  Tory Davis, who was at the scene, was identified as the shooter by a number of the witnesses.  Davis maintained his innocence until the end and sought a stay of execution so that further investigation into new evidence could be pursued.  A number of the witnesses had recanted their prior statements saying that, during their interviews, the police investigators coerced them into choosing Troy Davis.  One of these witnesses was Dorothy Ferrell, who was approximately 200 feet away from the incident when it happened.  Research evidence has shown that the ability to recognize an individual, even a famous face, drops to zero at 110 feet.[24]  This type of evidence, had it been considered at the time of the trial, could have cast doubt on the accuracy of Ferrell’s evidence. 

C.   Introducing the Impact of System and Estimator Variables in the Courtroom

Most of the literature concerning the impact of system and estimator variables in the courtroom focuses on criminal cases, perhaps because of the high-profile nature of the crimes and punishments that can be involved.  However, it is important to note that system and estimator variables can also play an important part in civil proceedings, such as litigation regarding automobile accidents. These variables can obviously impact any criminal or civil case where eyewitness testimony is accepted as evidence. Information concerning the impact of both system and estimator variables on eyewitness testimony can be introduced during the court proceedings in a number of ways.  It can be introduced, for example, in a voir dire, during examination of an eyewitness, in the judge's instruction to the jury, or by the testimony of an expert witness.  However, it is the last of these possibilities that has been met with the most resistance. 

In The Inquiry Regarding Thomas Sophonow, Justice Cory recommended that expert witnesses in the area of eyewitness fallibility should be allowed to give testimony when requested:

I would recommend that judges consider favourably and readily admit properly qualified expert evidence pertaining to eyewitness identification. This is certainly not junk science. Careful studies have been made with regard to memory and its effect upon eyewitness identification. Jurors would benefit from the studies and learning of experts in this field. Meticulous studies of human memory and eyewitness identification have been conducted. The empirical evidence has been compiled. The tragic consequences of mistaken eyewitness identification in cases have been chronicled and jurors and Trial Judges should have the benefit of expert evidence on this important subject. The expert witness can explain the process of memory and its frailties and dispel myths, such as that which assesses the accuracy of identification by the certainty of a witness. The testimony of an expert in this field would be helpful to the triers of fact and assist in providing a fair trial.[25]

However, Justice Cory’s view has not been popular.  The prevailing view is that the information regarding the possible fallibility of eyewitness testimony is more-or-less common knowledge.  Any concerns, it has been suggested, could be raised during the voir dire, the examination of the witness, or during the judge's instructions.  This, the opponents argue, would save both time and money. Professor Lee Stuesser, is one such opponent.  In response to Justice Cory’s recommendation to accept expert witnesses in the area of eyewitness fallibility, Stuesser wrote:

I accept the inherent frailties of eyewitness identification and, at the same time, the persuasive impact of such testimony. I accept that mistaken eyewitness identifications have contributed to wrongful convictions. I also accept that there exists an established legitimate body of studies on memory and the process of eyewitness identification; in order words, this is not junk science. What I do not accept is the need for such expert evidence. In my view, the admitting of this type of expert evidence, with its associated costs, is not necessary to ensure a fair trial. Simply put, leave the educating of the jury on eyewitness identification to the trial process and not to the experts. To be sure there may be exceptional cases where expert evidence may well assist, but they would be the rare exception.[26]

The Ontario Court of Appeal appears to be in agreement with Stuesser.  The Court has ruled that the empirical information concerning the fallibility of eyewitness testimony, which it regards as “opinion,” is common knowledge and that an expert witness in the area is, therefore, unnecessary.  According to the Court’s ruling, the court process can remind the jury of the fallibility of memory when necessary:

This opinion evidence is noteworthy in that, unlike most expert psychological or psychiatric testimony, it is not directed to making the testimony of a particular witness more understandable to the trier of fact and therefore more believable (e.g. an explanation of repressed memory syndrome or battered spouse syndrome). This opinion evidence is directed to instructing the jury that all witnesses have problems in perception and recall with respect to what occurred during any given circumstance that is brief and stressful. Accordingly, Dr. Yarmey is not testifying to matters that are outside the normal experience of the trier of fact: he is reminding the jury of the normal experience.[27]

Just recently, however, the New York State Supreme Court ordered a new trial for a man who was convicted of a crime based on the evidence of three eyewitnesses.[28]  The judge in the initial case had refused the defence's request to hear testimony from an expert witness concerning the fallibility of memory.  The Supreme Court concluded that the judge had erred because the whole case hinged on the witnesses’ identification.

If courts continue to be reluctant to allow expert witnesses to testify in similar cases, then a fall-back position is assumed. As mentioned, there are still other ways of addressing the issue, but they assume two things: 1) that the fallibility of eyewitness memory is common knowledge, and 2) that those in the system are indeed aware of the current scientific evidence and can draw upon it when necessary.  Preliminary evidence over the last twenty years has cast doubt on these assumptions.

V.     COMMON KNOWLEDGE 

One of the basic assumptions of many in the judicial system is that everyone is familiar with the fallibility of eyewitness testimony and that an expert witness could redundantly reiterate only what is already common knowledge.  Studies testing this assumption, however, have shown that the information is far from being common knowledge.  In 1983, for example, Daniel Yarmey and Hazel Tressillian Jones conducted such a study to determine the knowledge of both psychology professors and potential jurors on the subject of eyewitness testimony.[29]  The potential jurors were comprised of non-psychology students and regular citizens. They found that the potential jurors answered correctly 41 percent of the time, whereas professors fared significantly better at 77 percent.[30]  However, these numbers do not support the idea that this information is common knowledge. More recently, in 2011, the authors of this article conducted a similar study in which they tested psychology students and psychology professors, and noted that the average score of the psychology students was higher than that of the potential jurors tested by Yarmey and Jones.[31] The psychology students scored an average of 66 percent, but the scores of the psychology professors remained much the same as in Yarmey and Jones’ study at 78.9 percent.[32] These results again underscore the fact that the information is not common knowledge and that there is significant room for necessary improvement. 

VI.     JUDICIAL EXPERTISE

There are a number of ways in which the information concerning the fallibility of a witness’ testimony can be introduced in the courtroom. Since courts are currently reluctant to hear from expert witnesses on the subject, the responsibility rests with the players within the judicial system, and requires that these individuals be well versed in the scientific literature.  Again, the scientific evidence does not bear this out.  In their 1983 study, Yarmey and Jones also tested lawyers, judges, and law students.  It has been noted that these groups did not differ significantly from the potential jurors.[33]  More recently, in 2003, Richard Wise and Martin Safer tested the knowledge of US judges concerning the fallibility of eyewitness testimony.[34]  The judges scored 55 percent or, at most universities, just above a pass. 

The study by Wise and Safer was conducted more than eight years ago, and much has changed since then.  The number of commissions that have reported in the interim, the recommendations they have made, and the universal accessibility of these reports might well have had an effect.  Certainly, these reports have had some impact on the way in which police investigators conduct their investigations.  The present study was conducted to ascertain if the information concerning the research on fallibility of eyewitness testimony is being actively taught within Canadian law schools, and whether or not these soon-to-be lawyers have a firm grasp on the scientific evidence.  A preliminary survey of the courses within the Anglophone law schools in Canada suggests that at least three schools have a course directly related to the subject.  The following is a description of a course entitled “Studies in Criminal Law: Wrongful Convictions,” which is offered at the University of Ottawa:

This seminar course will examine the phenomenon of wrongful convictions - in Canada and elsewhere. It will explore why such miscarriages of justice occur and what can be done to prevent them in the first place, investigate them when allegations are made and remedy them once they are proven. Among the issues discussed will be false confessions, mistaken eyewitness identification, jailhouse informants and "tunnel vision" by criminal justice participants. Teaching will involve lectures, case studies, films and guest speakers.[35]

Though only three law schools appear to offer a course directly on point, this does not mean that the other schools are not teaching the information as part of another course.  The current study used a modified version of the Wise and Safer questionnaire, which was first used by the present authors when testing the knowledge and beliefs of psychology professors and first year psychology students in early 2010.  The questionnaire consists of a couple of demographic questions, as well as 12 knowledge and two belief questions (see Table 1 for the students’ scores for the twelve knowledge questions).  The Deans of all of the Anglophone law schools in Canada were invited to ask their students to participate in the web-based survey.  In total, 136 students responded to the survey: 46 first years, 48 second years, 36 third years, four fourth years, and two students with more than four years at law school.   The average score on the knowledge part of the questionnaire was 73.7 percent.  We suggest that this is inadequate knowledge for future professionals directly responsible for imparting this type of information in a courtroom setting.

Since the general belief amongst the members of the judicial system is that this type of information is common knowledge, it is surprising to note that only 20, or 14.7 percent, of the law students surveyed believed that lawyers knew enough about eyewitness fallibility research.  Ninety-seven students, or 71.3 percent, believed that judges should have more training in the area.   This suggests an understanding of the need to stay current and to follow the developments in the area. Furthermore, only 23 students, or 16.9 percent, said they had had a course related to the material, while 50 of the 136 students, or 36.8 percent, responded that they had had no exposure to this type of information.  The majority of the students who claimed to have had exposure, 50.7 percent, had read an article on the subject.

VII.     CONCLUSION AND RECOMMENDATIONS

It appears from this evidence that knowledge concerning the fallibility of eyewitness testimony is not, as has been believed by some judges, common.  In fact, the supposition that this information is common knowledge is diametrically opposed to the evidence (see Table 1, Question 2).  This is particularly concerting in light of the opposition to allowing expert testimony on this subject in the courtroom setting.  This would not be a problem if the players in the justice system were up-to-date on the current research.  However, evidence suggests that this is not the case.  There are, however, available remedies which we suggest:

1. Expert witnesses should be allowed to give testimony, when necessary, as was the case in the New York ruling.
2. Refresher courses on eyewitness testimony should be made available to all practicing lawyers and judges.
3. All law schools should implement a mandatory course concerning the impact that system and estimator variables can have on eyewitness testimony.

Table 1.

* A statistical test (one-way chi square test) was conducted to identify which questions yielded results significantly better or worse than would be expected by chance (50 percent). The questions marked with an asterisk indicate statistically significant results.

Disclaimer

The contents and the opinions expressed by the authors of Alberta Law Review Online Supplement articles are theirs alone and do not necessarily reflect the views of the Alberta Law Review. The purpose of the Alberta Law Review Supplement is to support the dissemination of legal information, but no representation is made concerning the accuracy or currency of the information.

Links to other websites may be provided as a means of convenient access to the information contained therein. The Alberta Law Review is not responsible for the information accessed by the use of the links or the accuracy of such information contained therein.

Authors that are published on the Alberta Law Review Supplement are not precluded from submitting posted material to other publications.