R v Bingley: Drugs, Discretion, and Deference

In 1994, Justice Sopinka warned against the dangers of expert evidence distorting the fact-finding process:

“Dressed up in scientific language which the jury does not easily understand and submitted through a witness of impressive antecedents, this evidence is apt to be accepted by the jury as being virtually infallible and as having more weight than it deserves.” (R v Mohan [1994] 2 SCR 9, paras 21-22)

Twenty-three year later, the Supreme Court of Canada (“SCC”) grappled with the issue again in R v Bingley, 2017 SCC 12. The SCC attempted to determine the point at which a court or trial judge conducting an evidentiary analysis must defer to parliament—if at all. The decision, however, does not provide any conclusory answers. Instead, both the majority and the dissenting opinions raise questions that expose the statutory framework governing drug impairment testing in Canada to constitutional challenge.

Some Context

Mr. Bingley was arrested in 2009 and charged with driving under the influence of drugs. For a detailed and helpful review of the facts, please consult the earlier post of my colleague, Nicholas Hay. I will re-iterate two important pieces of information on which the SCC decision was centred. Firstly, Mr. Bingley’s drug impairment was determined by a Drug Recognition Expert (“DRE”). A DRE is a law enforcement officer who is trained to conduct a 12-step drug recognition evaluation, which is certified under the Criminal Code and the Evaluation of Impaired Operation (Drugs and Alcohol) Regulation.

Secondly, Mr. Bingley was compelled by the DRE under section 254(3.1) of the Criminal Code to submit to a drug recognition evaluation. Section 254(3.1) provides law enforcement—for the first time ever—with this power to compel, provided that there are reasonable grounds to believe the person being compelled was impaired by a drug. The statutory scheme is meant to work like so:

  • Step 11/12: After completing ten steps of the evaluation, the DRE forms an opinion as to whether the subject is impaired. If the DRE determines that the subject is impaired, the DRE must indicate the categories of drugs that he or she believes have contributed to the subject’s impairment. The DRE bases these conclusions on his or her training and experience, as well as something called the “Drug Symptomatology Matrix.”
  • Step 12/12: If the DRE determines that the subject is impaired, the subject is compelled to give a toxicological sample, which is then sent to a forensic laboratory for analyses to confirm or refute the findings of the evaluator. The mere presence of a drug in the sample is not sufficient evidence to charge a person with drug impairment. The evaluation must show impairment, as well as signs and symptoms consistent with the drug categories indicated by the DRE. Generally, the DRE’s findings must be supported by the toxicology.

At Mr. Bingley’s trial, the DRE that evaluated him was called by the Crown to explain the results of the 12-step drug evaluation (i.e. how the DRE formed the opinion that Mr. Bingley was driving under the influence of drugs). The DRE’s testimony was then presented as evidence of Mr. Bingley’s impairment.

The Issue of Deference

 The central issue was whether the DRE’s evidence was admissible, which depended on the answers to two[1] questions:

  1. Does section 254(3.1) of the Criminal Code provide for the automatic admissibility of DRE’s opinion as evidence? In other words, does section 254(3.1) supersede and replace the common law rules of evidence and was this Parliament’s intention in drafting 254(3.1)?
  2. If section 254(3.1) does not provide for automatic admissibility, is voir dire (an oral hearing) needed to establish the DRE’s special expertise pursuant to the common law rules of evidence?

Ultimately, I believe, the question came down to deference: at what point in the admissibility analysis must the court or trial judge defer to Parliament—if at all? As a whole, the answers provided by the SCC and the courts below map out a sort of gradient for determining admissibility. I view this gradient as having three levels, and I will explore each in turn.

Level 1: Full Deference and Automatic Admission

The first trial judge, the summary conviction appeal judge, and the Ontario Could of Appeal (“ONCA”) agreed that the DRE evidence was automatically admissible. In so doing, all three levels of court found that Parliament intended for section 254(3.1) to supplant the common law evidence rules entirely. Effectively, the courts ruled that judges have no discretion whether to include or exclude the evidence. ONCA based its conclusion on a “plain reading” of section 254(3.1), focusing especially on the term “to determine” within the provision. As long as the DRE is certified according to the Regulations, the determination that the DRE reaches was intended by Parliament to be admissible expert opinion evidence. This is because, according to ONCA, Parliament created a detailed regulatory regime, which shows that it is “satisfied [with] the science underlying the drug evaluations” (ONCA para 48). Thus, holding a voir dire pursuant to the common law admissibility test would be out be out of harmony with the object of the statutory scheme and would be contrary to Parliament’s intention.

The second trial judge and the SCC, on the other hand, were not satisfied with this analytical shortcut. The SCC found that the words “to determine” do not indicate with any sufficient clarity the intention of Parliament to make DRE evidence automatically admissible at trial. Additionally, the SCC disagreed that the purpose of section 254(3.1) is to provide a shortcut to the evidence admissibility analysis. Rather, the purpose of the provision is to provide law enforcement with “investigative tools” to enforce laws against drug impaired driving. While the provision explicitly allows the DRE to form an opinion on a subject’s impairment, the provision does not speak to the admissibility of that opinion as evidence. Since Parliament did not expressly speak to admissibility, deference to an unclear intention should not be given. Therefore, the SCC found that the common law rules of evidence apply.

Level 2: Partial Deference and Special Expertise

Moving on to the second level of partial deference, it is helpful to review the common law rules of evidence. As “expert opinion evidence,” the DRE’s evaluation-based opinion is subject to two stages of analysis first articulated in R v Mohan. The Mohan test proceeds as follows:

  1. First, the evidence must meet four factors and judges usually determine whether the four factors are met through holding a voir dire. The evidence must be:
    1. Relevant
    2. Necessary
    3. Not subject to an exclusionary rule
    4. Considered to be special expertise
  2. Second, the judge must weigh potential risks (e.g. prejudicial effect) against the benefits (e.g. probative value) of admitting the evidence.

Mr. Bingley conceded that the evidence was relevant, necessary, not subject to an exclusionary rule, and that its risks did not outweigh its benefits. The only issue, therefore, was whether the DRE’s met the fourth requirement of “special expertise.” The majority of the SCC found that it did. Here, the majority employed a “plain reading” approach to statutory interpretation—the same approach that they had earlier rejected from the Court of Appeal’s judgment. The majority narrowed the test for special expertise to “[a] witness [possessing] expertise outside the experience and knowledge of the trier of fact.” Then, the majority recognized that the DRE is “literally” an expert, whose designation as an expert depends on special training and certification under the legislative scheme. Because the DRE’s opinion is based on his or her special training, the majority concluded that the DRE “undoubtedly [possesses] expertise on determining drug impairment that is outside the experience and knowledge of the trier of fact” (para 21). Through deference to Parliament at this later point in the analysis, the majority conveniently resolved the fourth Mohan factor. Since Mr. Bingley conceded to the other three factors, the majority noted that requiring a trial judge to hold a voir dire on the issue of special expertise—as is common practice—would be “absurd” and “a waste of judicial resources” (para 28).

Level 3: No Deference, Full Discretion

The dissenting justices, Justices Karakatsanis and Gascon, nevertheless found the deference to be out of place, and subtly accused the majority of creating yet another problematic shortcut to the admissibility analysis. The dissent focused primarily on the special rule of “novel science”—a sub-requirement of the special expertise factor—that the majority rejected in their narrow test. The novel science rule dictates that when an expert opinion is based on a novel scientific theory or technique, the underlying science of the theory or technique must be established as reliable by precedent, evidence, or statute (para 15). The majority found that in this case, the reliability of the 12-step drug recognition evaluation came from the statutory framework provided by section 254(3.1) of the Criminal Code and the Regulations, and dispensed with the issue.

The dissenting opinion questioned the finding that reliability was prescribed by Parliament. Echoing the majority’s reasoning, the dissent stated that Parliament did not indicate with sufficient clarity that the a 12-step evaluation was reliable. Therefore, the dissent suggested that Parliament did not conclusively determine the reliability of the 12-step evaluation, but rather, merely endorsed the reliability of the evaluation as an “investigative tool.” Absent clear statutory language, the dissent insisted that a basic threshold of reliability must therefore be established through evidence and precedent.

To obtain evidence, the trial judge should be able to hold a voir dire to determine whether the underlying science is reliable and valid. Obliging the trial judge to defer to Parliament, as the majority opinion does, in effect would lead to automatic admissibility of the DRE’s opinion as evidence, despite the majority opinion finding to the contrary. Importantly, this would also rid the trial of the ability to use their discretion to exclude evidence. This is particularly troubling given that the second trial judge excluded the DRE’s evidence upon finding through a voir dire that the DRE was not trained on the reliability of the 12-step evaluation and had insufficient knowledge of its scientific foundation.

To set a precedent, the dissent also noted that the use of novel science should be established in a courtroom even if its use is well established outside. Determining reliability this way would be a burdensome task which would require many different specialists to testify and provide various opinions. The dissent, nevertheless, stated that this is a necessary task. Given the unsettled nature of the case law surrounding DRE opinions and the relatively recent use of DRE evidence in Canada, simply deferring to Parliament would constitute an “unqualified endorsement of the underlying science” (para 46). This would set the wrong precedent, the dissent argued, and rid the common law system of the important safeguard of judicial discretion in novel circumstances. According to the dissent, discretion should be preserved through to the end of the evidentiary analysis, and deference to Parliament on the matter of admissibility must be reserved for explicit statutory language to that effect.


In the end, Bingley does not bear any satisfying conclusions or conclusive answers. On the one hand, inferring the admissibility of potentially unreliable science at best, and “junk science” at worst, as evidence may lead to a distorted fact-finding process, mistrials, and wrongful convictions. On the other hand, extending judicial discretion in such circumstances raises more questions than answers. For example: can reliability ever be prescribed by Parliament? What would that prescription look like? Furthermore, what are the factors that determine the “threshold of reliability”? In the case at bar, the 12-step evaluation has a 98% reliability rate. Is it open to the trial judge to brand any science that the scientific community doesn’t wholly accept as unreliable? While the DRE’s evidence doesn’t presume the ultimate issue of guilt, this inconclusive decision, as my colleague correctly predicted, opens the floor to Charter challenges aimed at the legislative framework itself. Given the grossly undeveloped nature of the drug-testing regime in Canada, these Charter challenges might just be a welcomed course of action for everyone involved.

[1] Note: there were a total of three questions, but I will not engage with the question of whether the DRE evidence was admissible as lay opinion evidence because the SCC didn’t consider it.

Join the conversation

Loading Facebook Comments ...