By: Erin Sheley

PDF Version: When Should Judicial Discretion Trump Expert Testimony?

Case Commented On: R v Clark, 2016 ABCA 72 (CanLII)

In Regina v Clark the Alberta Court of Appeal reinforced the principle that trial courts should enjoy broad discretion in making evidentiary decisions. On the other side of the scale in this particular case was the great problem of ensuring the accuracy of witness identifications when they are the primary basis for conviction. In the United States at least, 70% of exonerations obtained through DNA evidence occurred in cases involving eyewitness misidentifications (see data collected by the Innocence Project, available here).

Clark involved a trial by judge of a bank robbery case. During the crime the suspect had partially obscured his face with a hood and a hat pulled down over most of his features (at paras 3-4). At trial, the Crown relied on the testimony of three eyewitnesses, and in particular that of one woman who had stood about 5-6 feet away from him at the bank counter and glanced at him several times during the robbery at para 54). Several other witnesses identified a photograph of the robber taken from the security camera as an individual who went by the street name “Lips,” a name by which the accused had identified himself to a police officer prior to the robbery (at para 51).

The defense sought to lead testimony by a Dr. Reid, an expert in the construction and administration of photo lineups (at para 38). Dr. Reid proposed to testify to the general unreliability of photo line-ups (and in particular the lack of correlation between the confidence of a witness in their identification and actual accuracy) (at para 39). He also proposed to testify to various aspects of the photo lineups in question which he believed made them less likely to be accurate. Specifically:

  • The lineups were suspect-based, rather than description-based.
  • The lineups included other people taken from the immediate community, making it more likely that a witness might recognize someone they knew and, thus, that the number of genuine suspects would be reduced.
  • The photos showed more of the suspect’s face than the witnesses had actually seen in real life due to his disguise.
  • The police officer administering the line-up left the witness alone with the pictures for a time and then drew her attention, upon his return, to a particular photo.
  • The lineups occurred several months after the robbery itself (at paras 40-43).

In excluding Dr. Reid’s testimony, the trial court emphasized that it was unnecessary to resort to expert testimony for the purposes of reinforcing the frailties of eyewitness testimony (at para 48). In his reasons for judgment the trial judge noted that he had been a lawyer for 37 years prior to coming to the bench and that he had “conducted many civil and criminal trials as well as complex regulatory hearings, many of which involved the extensive use of technical and scientific experts who gave opinion evidence clearly outside the knowledge base of the decision makers involved” (at para 48). The judge also noted that since becoming a judge he had taken a number of National Judicial Institute courses dealing with the assessment of witnesses and the types of evidence allowed in criminal cases (at para 48).

The accused raised four issues on appeal:

  1. Whether the trial judge erred in failing to admit the evidence of Dr. Reid
  2. Whether the trial judge failed to apply the correct legal test to the issue of eyewitness identification
  3. Whether the trial judge reversed the burden of proof, requiring the defence to demonstrate a lack of link between the appellant and the crime
  4. Whether the trial judge failed to apply the proper legal standard for a circumstantial case, resulting in an unreasonable verdict

The Court of Appeals dismissed the accused’s arguments in fairly abbreviated reasons. Referring to the rule in R v Seaboyer, [1991] 2 SCR 177, that probative evidence should be admitted unless an exclusion can be justified on some ground, the Court concluded that “once the trial judge reasonably concludes that the admissibility threshold is not met, he is entitled to refuse to admit the evidence” (at para 65). In holding that the trial judge had not erred in excluding the evidence of Dr. Reid, the Court repeatedly emphasized the “gatekeeping” function of trial courts, which “deserves deference absent prejudicial error” (at para 67).

The Court addresses the second and third issues at once, holding that nothing in the trial judge’s reasons demonstrate that he had reversed the burden of proof as to the issue of identification, but only that none of the evidence raised a reasonable doubt in his mind as to the appellant’s guilt (at para 72). In coming to this conclusion the Court noted that the trial judge was “alive to the frailties of identification evidence” and had in fact taken these problems into account in evaluating whether the Crown had proven its case beyond a reasonable doubt (at para 71).

On the final point, the Court states that the test for the unreasonableness of a verdict is whether “the verdict is one that a properly instructed jury, acting judicially, could reasonably have rendered” (at para 75, citing R v Biniaris [2000] 1 SCR 381). It holds that in this case the trial court’s conclusion that the Crown had proved its case beyond a reasonable doubt was reasonable because, in fact, it was “the only reasonable inference on the totality of the evidence” (at para 76, emphasis added).

A couple of observations flow from the Court’s reasons in this case. The first is that Clark provides a particularly striking example of the general disinclination of appellate courts to interfere with trial courts’ prerogatives in balancing the probative value of potential evidence against other reasons for exclusion, such as redundancy and waste of time (both factors which appeared to motivate the trial judge in this case). In describing the trial judge as a “gatekeeper” the Court alludes to certain systemic characteristics that drive this discretion. The trial court has the best view of the facts of an individual case, and—to the extent that the justice system would not have the bandwidth to function at all if it attempted to process ALL proffered probative evidence—the trial court must be empowered to make game time decisions without overly invasive appellate review.

The second observation relates to the trial court’s function as factfinder. In discussing the first issue on appeal, the Court noted that “appellate courts have consistently held that this type of evidence is unnecessary because it is clearly within the knowledge of judges and properly instructed jurors” (at para 66). While the reference to properly instructed jurors is obiter, it nonetheless suggests that the outcome of the case would have been the same regardless of whether it had been a jury deciding the case instead of a judge with many years of experience and training. This begs the question: can we in fact rely on jury instructions to drive home the scientific context in which they must understand evidence?

This problem arose in R v D.(D.), [2000] 2 SCR 275, in which the Supreme Court of Canada held that expert evidence is not admissible to counter the argument that a complainant’s delay in reporting sexual assault is evidence of fabrication. Because the trial court must instruct the jury as a matter of law that it cannot infer fabrication from the complainant’s delay, the Court held that expert testimony on the matter is therefore unnecessary. The dissent in that case, authored by Justice McLachlin, argued that such evidence should be considered on a case-by-case basis because the jury might in fact need assistance to understand psychological facts that go beyond immediate experience. The dissent’s concerns in that case—that jury instructions may inadequately substitute for expert explanation of such phenomena—apply equally in cases of eyewitness identifications. When an untrained jury is faced with a witness who expresses certainty as to identification, we cannot be certain what weight they might give it, even under circumstances that might warrant caution. R v Clark, like R v D.(D.) is a vote of confidence not only in trial judges acting appropriately as gatekeepers, but in juries understanding and fully taking to heart the often labyrinthine instructions they are given. Whether such confidence is warranted may be a question that requires more empirical research to answer properly. In the meantime, it seems clear that courts are willing to treat a trial judge’s occupational familiarity with certain specialized areas of knowledge such as the psychology of sexual assault and the risks of misidentification as adequate substitutes for a jury’s. These assumptions should give us pause.

The failed prosecution of Jian Ghomeshi provides a particularly striking example of where expert evidence might have assisted even a judge as trier-of-fact. In his reasons for judgment Justice Horkins noted the understandable limitations on the complainants’ memories of long-ago assaults, but at other moments questioned them about details as precise as the orientation of the accused’s hands around their neck (see discussion of the evidence and critiques of trial court’s reasons in R v Ghomeshi by Joshua Sealy-Harrington). Furthermore he made a number of factually inaccurate statements about the ease with which a sexual assault complainant may navigate the criminal justice system, and the facility she should have in assessing the relevance of particular aspects of a traumatic event in the course of reporting it. Despite the training and experience judges bring to bear on these questions, the Ghomeshi case illustrates how lack of expert testimony can result in at least partially uninformed credibility determinations.

To subscribe to ABlawg by email or RSS feed, please go to https://ablawg.ca

Follow us on Twitter @ABlawg