“Don’t believe your ears!” - R v Prasad Reaffirms the Role of Expert Opinion Evidence

Author:  Derek Zaporzan

“The point is to preserve trial by judge and jury, not devolve to trial by expert.” – Justice Cromwell[1]

In criminal trials, the courts have stressed that the triåal process must be respected, and that the role of the trier of fact must never be usurped by another.[2] However, there are often circumstances where, in order for a trier of fact to draw the desired inferences from the facts presented at trial, the court will allow a witness with specialized knowledge or expertise to provide opinion evidence to assist the trier of fact in their analysis.

The Ontario Court of Appeal (the “ONCA”) recently had to rule on whether expert opinion evidence on psychoacoustics and cognitive neuroscience should be entered in a case focused on wiretap evidence. Had the opinion evidence been entered and accepted, it would have fundamentally altered the role of the trier of fact and the process in which wiretap evidence would be accepted (if it were to be accepted at all). Fortunately, the court refused to enter the opinion evidence, and in doing so reaffirmed its commitment to the traditional roles of the trier of fact and expert opinion evidence.[3] This blog will examine the traditional role of expert opinion evidence and analyze why its admission was refused by the court in Prasad.

Background – Expert Opinion Evidence

In a trial setting, expert witnesses “take information accumulated from their own work and experience, combine it with evidence offered by other witnesses, and present an opinion as to a factual inference that should be drawn from that material.”[4]

Expert testimony is admissible to assist the trier of fact in drawing inferences in areas where the expert has relevant knowledge or experience beyond that of the lay person.[5] In trials, such experts may include police gang and drug experts, although experts are not limited to being police officers. For example, in R v Lavallee, the expert called was a psychiatrist with extensive professional experience in treating women who suffered from domestic abuse.[6]

Expert evidence may be admitted where it meets a basic threshold of reliability and is necessary to assist a trier of fact in coming to a satisfactory conclusion.[7] For expert evidence to be admitted by a trier of law, it must be (a) relevant; (b) necessary to assist the trier of fact; (c) there must be an absence of any exclusionary rule; and (d) the expert must be properly qualified.[8] If the expert evidence meets the above criteria, the trier of law must then undertake the cost-benefit analysis of admitting the evidence. In this “gatekeeping stage,” the trier of law must assess whether the probative value of the proposed expert evidence is outweighed by its prejudicial effect.[9]

R v Prasad, 2024 ONCA 601

In 2018, Akshay Prasad (Prasad) was convicted in the Ontario Superior Court of two counts of trafficking cocaine in a judge-alone trial. The trial judge relied heavily on intercepted communications between Prasad and a co-conspirator. Tendered as evidence at the trial were the original audio recordings of the intercepted communications, which were of poor quality, as well as enhanced recordings and transcripts based on those enhanced recordings.[10]

On appeal, Prasad asked to have the conviction quashed based on fresh expert evidence. The proposed evidence consisted of three opinions, one by a professor of brain and cognitive sciences, another by a cognitive neuroscientist and psychologist, and the third by a professor of psychoacoustics.[11] Prasad asserted that “[t]he trial judge’s decision to use the transcripts as an aid, and his own attempt to listen to the tapes are tainted due to listener bias.”[12]

Although the experts could not point to a single word they thought was wrongly transcribed, they all concurred that the entirety of the transcription was unreliable. The experts argued that the transcript cannot be accurate because the audio’s poor quality brings into play the psychological phenomenon of “priming.”[13] “Priming” describes how prior knowledge of the circumstances from which the audio was obtained can “prime” the transcriber to “supplement the extraneous information that comports with the circumstances, thus biasing the transcriptions.”[14] According to the experts, this phenomenon would apply to the Police monitor who transcribed the audio communication based on the assumption that the parties were discussing a drug deal, and would extend to anyone reading the transcript, including the trial judge and assumedly, the appellate justices.[15]

The experts further opined that “in order to prevent the biasing effect of the transcript, the trier of fact should listen to the audio without assistance; if the transcript is read beforehand, then the danger is that the text of the transcript will bias the listener to hear what the transcript says.”[16] One expert even went so far as to conclude that any unbiased evaluation would be unachievable, as any listener might infer that the recordings were the result of surveillance and contained discussions of drug terminology, and that such inferences would inevitably influence the perception of what was being said.[17] It is not surprising that the court rejected this assertion, as it is akin to arguing that judicial objectivity in weighing wiretap evidence was

impossible.

ANALYSIS

The court in Prasad ultimately refused to enter the proposed expert evidence as it failed on the criterion of necessity. Citing Justice Cromwell in White Burgess, the court stated that it is “for the jury to form opinions, and draw inferences and conclusions, not for the witness.”[18]

The ONCA found that the concept of “priming” was not beyond the common experience of lay people, holding that “[w]ith all due respect to the experts… [priming] has always been a staple of human communication and persuasion.”[19] The court went on to say that one who has been primed is not thereafter “invincibly convinced by the truth of the position that the priming seeks to entrench.”[20]The ONCA also disagreed with the experts’ assertion that the poor quality of the audio combined with the “priming” nature of the transcripts created an “intolerably high” likelihood that a wrongful conviction would result.[21] The court stated that there was plenty of circumstantial evidence to guarantee the reliability of the transcripts. It was clear that Prasad and his co-conspirator were not talking about the weather, sports, or music, but rather that they were discussing drugs.[22]

In Lavallee, expert opinion evidence about “battered woman syndrome” – the psychological experiences of women who have suffered violence at the hands of their intimate partners – was admitted to provide the trier of fact with the necessary background to determine whether the accused, who had shot her partner in the back of the head, did so in self-defence.[23] As was the case in Prasad, the proposed expert evidence in Lavallee was put forward by the defence. The similarities, however, end there. The expert opinion evidence in Lavallee was entered to assist the trier of fact by expanding their knowledge base, assisting with the truth-seeking function of the court. The proposed expert evidence in Prasad would have had the opposite effect, unjustifiably limiting the evidence available to the court to be considered and undermining the objectivity of the trial judge.  

The purpose of the proposed expert evidence in this case was to undermine confidence in wiretap evidence in general. Audio probes, for obvious reasons, must be hidden from the targets of the investigation. As a result, the audio intercepts are often of poor quality, with certain words and sometimes even entire conversations becoming inaudible and incapable of being transcribed. However, to exclude such evidence as being unreliable in all circumstances would be unduly detrimental to police investigations and the administration of justice. Such a determination should be left to the trial judge. Fortunately, the court here agreed, and even went so far as to characterize the expert evidence in this case as “an unwarranted wholesale assault on the criminal trial process.”[24] A logical extension of the appellant’s argument in Prasad would be that, due to the effects of priming, an objective interpretation of poor-quality video surveillance showing a crime − for example, a commercial robbery − would be equally unachievable. If the footage is too grainy, or of such poor resolution so as to leave any question as to the precise sequence of events taking place, then it should not even be made available to the trier of fact. The argument that because someone is charged with a robbery, the video footage must therefore show a robber would be absurd, but no more so than the argument presented in Prasad.

A process already exists for cases where an accused disputes the accuracy of the transcription adduced. When the tape is played in court, the wording of the transcript, usually presented by a police officer, can be challenged by way of cross-examination. If cross-examination cannot settle the issue, the defence can tender a witness who has listened to the tapes and prepared their own transcript, setting out an alternative interpretation of the words spoken on the tape. Both transcripts are then provided to the jury for consideration.[25] It again must be stressed that, in Prasad, none of the experts could point to a single word that they believed was wrongly transcribed.

The ONCA in Prasad did not, however, completely disregard the role of priming and its possible deleterious effects in criminal trials. The court suggested that the issue of priming can, and sometime must be, the subject of a jury instruction.[26] This ruling therefore leaves the door open for the issue of priming to be considered in future cases.

wiretap

[1] White Burgess Langille Inman v. Abbott and Haliburton Co., 2015 SCC 23 at para 18 [White Burgess]

[2] R v Prasad, 2024 ONCA 601 at para 37 [Prasad], citing White Burgess at para 14.

[3] Ibid.

[4] R v Abbey, 2009 ONCA 624 at para 71 [Abbey].

[5] 1990 CanLII 95 (SCC) at page 889 [Lavallee].

[6] Ibid.

[7] R v Mohan, 1994 CanLII 80 (SCC) at page 5 [Mohan].

[8] Ibid at page 4.

[9] White Burgess, supra note 1 at para 24.

[10] Prasad, supra note 2 at paras 1-2.

[11] Ibid at paras 4-5.

[12] Ibid at para 7.

[13] Ibid at para 13.

[14] Ibid at para 14.

[15] Ibid.

[16] Ibid at para 20.

[17] Ibid at para 24.

[18] Ibid at 37, citing White Burgess at para 14.

[19] Ibid at para 55.

[20] Ibid at para 56

[21] Ibid at para 60

[22] Ibid at para 62

[23] Lavallee, supra note 5

[24] Prasad, supra note 2 at para 79

[25] R v Rowbotham, 1988 CanLII 147 (ONCA) at para 107

[26] Prasad, supra note 2 at para 77