R. v. Barton, 2019 SCC 33

Criminal law — Evidence — Admissibility — Mistaken belief in communicated consent

On appeal from a judgment of the Alberta Court of Appeal (Fraser C.J. and Watson and Martin JJ.A.), 2017 ABCA 216, setting aside the acquittal of the accused and ordering a new trial.

The accused was charged with first degree murder in the death of an Indigenous woman and sex worker, who was found dead in the bathroom of the accused’s hotel room. The cause of death was determined to be loss of blood due to an 11 cm wound in her vaginal wall. The Crown’s theory was that during the course of commercial sexual activities while the deceased was incapacitated by alcohol, the accused cut the inside of her vagina with a sharp object with intent to seriously harm or kill her. Alternatively, the Crown took the position that if the accused did not murder the deceased, he committed the lesser and included offence of unlawful act manslaughter, by causing her death in the course of a sexual assault. The accused, however, maintained his innocence. He testified that he and the deceased engaged in similar consensual sexual activity on both the night leading up to her death and the previous night, and that on both occasions, he penetrated her vagina with his fingers and thrusted repeatedly. He claimed that she started to bleed unexpectedly on the second night, bringing the sexual activity to a halt, and he awoke the next morning to find her dead in the bathtub. He then left the hotel in a panic, returned, called 911, and fabricated different versions of a false story. Although he admitted that he caused her death, he claimed that it was a non-culpable accident. He denied using a sharp object and asserted that the deceased consented to the sexual activities in question — or at least he honestly believed that she did.

In its opening address to the jury, the Crown referred to the deceased as a prostitute and explained that she and the accused struck up a working relationship on the night before her death. In addition, without having submitted an application under ss. 276.1(1) and 276.1(2) of the Criminal Code to adduce evidence of the deceased’s prior sexual activity, the accused testified at length about his previous sexual activity with the deceased. The Crown did not object, nor did the trial judge order a separate hearing to consider the admissibility and permissible uses of this evidence. The jury acquitted the accused. The Court of Appeal allowed the Crown’s appeal, set aside the accused’s acquittal, and ordered a new trial on first degree murder.

Held (Wagner C.J. and Abella and Karakatsanis JJ. dissenting in part): The appeal should be allowed in part and a new trial on unlawful act manslaughter ordered.

Per Moldaver, Côté, Brown and Rowe JJ.:

The trial judge erred in failing to comply with the mandatory requirements set out in s. 276 of the Criminal Code . That error had ripple effects, most acutely in the instructions on the defence of honest but mistaken belief in communicated consent, upon which the accused relied. In particular, non‑compliance with the s. 276 regime translated into a failure to expose and properly address misleading evidence and mistakes of law arising from the accused’s defence. This resulted in a reversible error warranting a new trial. However, the new trial should be restricted to the offence of unlawful act manslaughter, as it has not been demonstrated that the acquittal on murder was tainted by reversible error.

A. Section 276 and Prior Sexual Activity Evidence

Section 276 of the Criminal Code governs the admissibility of evidence about a complainant’s prior sexual activities and the uses to which that evidence may be put. The animating purposes behind the s. 276 regime are to protect the integrity of the trial by excluding irrelevant and misleading evidence, protect the accused’s right to a fair trial, and encourage the reporting of sexual offences by protecting the security and privacy of complainants. Section 276(1) provides that in proceedings in respect of certain listed offences, evidence of a complainant’s prior sexual activity is not admissible to support an inference that, by reason of the sexual nature of that activity, the complainant is more likely to have consented to the sexual activity in question or is less worthy of belief. This section is categorical in nature and applies irrespective of which party led the evidence. Section 276(2) provides that evidence of the complainant’s prior sexual activity adduced by or on behalf of the accused is presumptively inadmissible unless, after certain procedures have been followed, the trial judge rules to the contrary.The s. 276 regime applies to any proceeding in which an offence listed in s. 276(1) has some connection to the offence charged, even if no listed offence was particularized in the charging document. Crown‑led prior sexual activity evidence is subject to the common law principles articulated in R. v. Seaboyer, [1991] 2 S.C.R. 577.

B. Defence of Honest but Mistaken Belief in Communicated Consent

An accused may respond to a charge of sexual assault by relying on the defence of honest but mistaken belief in communicated consent. Consent is defined in s. 273.1 of the Criminal Code as the voluntary agreement of the complainant to engage in the sexual activity in question. For purposes of the actus reus, consent means that the complainant in her mind wanted the sexual touching to take place. At this stage, the focus is placed squarely on the complainant’s state of mind, and the accused’s perception of that state of mind is irrelevant. For purposes of the mens rea, and specifically for the purposes of the defence of honest but mistaken belief in communicated consent, consent means that the complainant had affirmatively communicated by words or conduct her agreement to engage in the sexual activity with the accused. Hence, the focus at this stage shifts to the mental state of the accused, and the question becomes whether the accused honestly believed the complainant effectively said “yes” through her words and actions.

While the jurisprudence has consistently referred to the relevant defence as being premised on an honest but mistaken belief in consent, it is clear that in order to make out this defence, the accused must have an honest but mistaken belief that the complainant actually communicated consent, whether by words or conduct. It is therefore appropriate to refine the judicial lexicon and refer to the defence more accurately as an “honest but mistaken belief in communicated consent”. This refinement is intended to focus all justice system participants on the crucial question of communication of consent and avoid inadvertently straying into the forbidden territory of assumed or implied consent. In seeking to rely on the complainant’s prior sexual activities in support of a defence of honest but mistaken belief in communicated consent, the accused must be able to explain how and why that evidence informed his honest but mistaken belief that the complainant communicated consent to the sexual activity in question at the time it occurred. The accused cannot rest his defence on the false logic that the complainant’s prior sexual activities, by reason of their sexual nature, made her more likely to have consented to the sexual activity in question, and on this basis he believed that she consented.

Honest but mistaken belief in communicated consent is a mistake of fact defence, which operates where the accused mistakenly perceived facts that negate, or raise a reasonable doubt about, the fault element of the offence. By contrast, as a general rule, mistakes of law offer no excuse. Therefore, to the extent an accused’s defence of honest but mistaken belief in communicated consent rests on a mistake of law — including what counts as consent from a legal perspective — rather than a mistake of fact, the defence is of no avail. For example, it is an error of law — not fact — to assume that unless and until a woman says “no”, she has implicitly given her consent to any and all sexual activity. Implied consent, which rests on the assumption that unless a woman protests or resists, she should be deemed to consent, has no place in Canadian law. In addition, a belief that the complainant gave broad advance consent to sexual activity of an undefined scope will afford the accused no defence, as that belief is also premised on a mistake of law, not fact. As a further example, an accused’s belief that the complainant’s prior sexual activities, by reason of their sexual nature, made it more likely that she was consenting to the sexual activity in question is again a mistake of law.

The availability of the defence of honest but mistaken belief in communicated consent is not unlimited. The reasonable steps requirement under s. 273.2 (b) of the Criminal Code imposes a precondition to this defence. This requirement, which rejects the outmoded idea that women can be taken to be consenting unless they say “no”, has both objective and subjective dimensions: the accused must take steps to ascertain consent that are objectively reasonable, and the reasonableness of those steps must be assessed in light of the circumstances known to the accused at the time. The reasonable steps inquiry is highly fact‑specific. Trial judges and juries should take a purposive approach, keeping in mind that the reasonable steps requirement reaffirms that the accused cannot equate silence, passivity, or ambiguity with the communication of consent. Trial judges and juries should also be guided by the need to protect and preserve every person’s bodily integrity, sexual autonomy, and human dignity. Steps based on rape myths or stereotypical assumptions about women and consent cannot constitute reasonable steps.

The concept of reasonable steps to ascertain consent under s. 273.2 (b) of the Criminal Code must be distinguished from the concept of reasonable grounds to support an honest belief in consent under s. 265(4) . Where the accused is charged with some form of assault, the presence or absence of reasonable grounds is simply a factor to be considered in assessing the honesty of the accused’s asserted belief in consent in accordance with s. 265(4) . By contrast, where the accused is charged with a sexual offence under ss. 271 , 272 , or 273 , a failure to take reasonable steps is fatal to the defence of honest but mistaken belief in communicated consent by virtue of s. 273.2 (b). In the context of a charge under ss. 271 , 272 , or 273 where the accused asserts an honest but mistaken belief in communicated consent, if either there is no evidence upon which the trier of fact could find that the accused took reasonable steps to ascertain consent or the Crown proves beyond a reasonable doubt that the accused failed to take reasonable steps to ascertain consent, then there would be no reason to consider the presence or absence of reasonable grounds to support an honest belief in consent under s. 265(4) , since the accused would be legally barred from raising the defence due to the operation of s. 273.2 (b).

An accused who wishes to rely on the defence of honest but mistaken belief in communicated consent must first demonstrate that there is an air of reality to the defence. If there is no evidence upon which the trier of fact could find that the accused took reasonable steps to ascertain consent, then the defence of honest but mistaken belief in communicated consent has no air of reality and must not be left with the jury. By contrast, if there is an air of reality to the defence of honest but mistaken belief in communicated consent, including the reasonable steps requirement, then the defence should be left with the jury. The onus would then shift to the Crown to negative the defence, which could be achieved by proving beyond a reasonable doubt that the accused failed to take reasonable steps. Where the Crown does not prove beyond a reasonable doubt that the accused failed to take reasonable steps, that does not lead automatically to an acquittal. In those circumstances, the trial judge should instruct the jury that they are required, as a matter of law, to go on to consider whether the Crown has nonetheless proven beyond a reasonable doubt that the accused did not have an honest but mistaken belief in communicated consent. This requirement flows from the fact that the defence is ultimately one of an honest but mistaken belief in communicated consent, not one of reasonable steps. Ultimately, if the Crown fails to disprove the defence beyond a reasonable doubt, then the accused would be entitled to an acquittal.

C. Application

(1) Applicability of the Section 276 Regime

In this case, while the Crown did not object to the accused’s testimony about the deceased’s prior sexual activity, its failure to do so was not fatal. The ultimate responsibility for enforcing compliance with the mandatory s. 276 regime lies squarely with the trial judge, not with the Crown. It is also plain that the proceeding implicated an offence listed in s. 276(1) because the offence charged in this case, first degree murder, was premised on sexual assault with a weapon, which is an offence listed in s. 276(1). It follows that the s. 276 regime was engaged. Furthermore, the limited information conveyed in the Crown’s opening address did not exclude the application of s. 276(2) to the accused’s detailed testimony about the deceased’s sexual activity on the night before her death, which went well beyond the basic narrative recounted by the Crown. Therefore, before adducing evidence of the deceased’s sexual activity on the night before her death, the procedural requirements under s. 276 should have been observed, and if any of the evidence was deemed admissible, a careful limiting instruction by the trial judge was essential to instruct the jury on the permissible and impermissible uses of that evidence.

(2) Instructions on the Defence of Honest but Mistaken Belief in Communicated Consent

At trial, the accused relied on the defence of honest but mistaken belief in communicated consent, and his testimony about the deceased’s prior sexual activities featured prominently in his defence. The trial judge erred by failing to inoculate the jury against mistakes of law masquerading as mistakes of fact, as the accused’s defence raised the spectre of several mistakes of law: a belief that the absence of signs of disagreement could be substituted for affirmative communication of consent; a belief that prior similar sexual activities between the accused and the deceased, the deceased’s status as a sex worker, or the accused’s own speculation about what was going through the deceased’s mind could be substituted for communicated consent to the sexual activity in question at the time; a belief that the deceased could give broad advance consent to whatever the accused wanted to do to her; and the inference that the deceased’s past sexual activities, by reason of their sexual nature, may make it more likely that she consented to the sexual activity in question. It was incumbent on the trial judge to caution the jury against acting on these mistakes of law. The absence of such an instruction had an immediate impact on the defence of honest but mistaken belief in communicated consent, as it allowed the defence to proceed while these mistakes of law were left unaddressed, thereby leaving the jurors without the necessary tools to engage in a proper analysis. This error was inextricably linked to the failure to hold a s. 276 hearing, which would have subjected the admissibility and permissible uses of the evidence of the deceased’s prior sexual history to rigorous scrutiny and assisted in filtering out the mistakes of law raised by the accused’s defence.

(3) Instructions on Motive

Where motive does not form an essential element of the offence, the necessity of charging a jury on the subject depends upon the course of the trial and the nature and probative value of the evidence adduced. In this case, motive was a relevant consideration bearing upon whether the accused intended to seriously harm or kill the deceased, which would go to the fault element for murder. However, the Crown led no evidence of motive. Since there was neither a proven motive nor a proven absence of motive, it fell within the trial judge’s substantial discretion to charge on motive. Further, the charge was not so unbalanced as to amount to misdirection.

(4) Instructions on the Objective Fault Element of Unlawful Act Manslaughter

The fault element of unlawful act manslaughter consists of objective foreseeability of the risk of bodily harm which is neither trivial nor transitory, in the context of a dangerous act. At trial, the defence conceded that the act in question was dangerous, and the Crown accepted that it would be appropriate to instruct the jury that if they were satisfied that the accused committed an unlawful act, then it was implicit that the act was dangerous. In addition, the Crown agreed to the request of the defence that the language of objectively foreseeable bodily harm be removed from the charge. On appeal, the Crown reversed its position, arguing that the instructions on dangerousness were deficient and the trial judge’s failure to refer to the objective fault element of unlawful act manslaughter unfairly minimized its expert evidence. The Court of Appeal accepted the Crown’s position on appeal. However, it should not have done so, as the Crown had to live with the decision it made at trial.

(5) Instructions on After‑the‑Fact Conduct

In light of procedural fairness concerns — namely, the Crown’s limited right to appeal an acquittal and the requirements that must be observed by appellate courts when raising new issues — the Court of Appeal should not have ordered a new trial on the issue of after‑the‑fact conduct evidence. The Crown was actively involved in drafting the jury charge, and at no point did it request a specific instruction directing the jury to consider the accused’s after‑the‑fact conduct in assessing his credibility. Further, although the Court of Appeal notified the parties at the outset of the hearing that it would raise new issues, it did not specify the precise nature of those issues or indicate whether one or more could result in the accused’s acquittal being set aside. It also allowed the Crown to advance certain arguments on after‑the‑fact conduct for the first time in reply submissions, and at the end of the hearing the court indicated that there was no need for further written argument. Lastly, though it is not necessary to finally decide the issue, there is reason to be skeptical of the Crown’s argument that the trial judge’s instructions on after-the-fact conduct were so defective as to amount to reversible error. When read fairly and as a whole, the trial judge’s charge on after‑the‑fact conduct adequately, albeit imperfectly, conveyed to the jury that they could consider the accused’s after‑the‑fact conduct in assessing guilt and equipped them to do so.

(6) Instructions on the Defence of Accident

The term “accident” is used to signal one or both of the following: (1) that the act in question was involuntary (i.e., non‑volitional), thereby negating the actus reus of the offence; or (2) that the accused did not have the requisite mens rea. In assessing whether a claim of “accident” may negate mens rea in any particular case, it is obviously essential to consider what the relevant mens rea requirement is in the first place. In carrying out this inquiry, it must be kept in mind that mens rea requirements vary and include, for example: (1) a subjective intention to bring about a prohibited consequence; (2) a subjective awareness of prohibited circumstances; and (3) objective fault. Where the offence charged requires proof of subjective intent to bring about a particular consequence, the claim that the accused did not intend to bring about that consequence, making it a mere accident, is legally relevant, as it could negate the mens rea required for a conviction. By contrast, where the offence only requires a subjective awareness of particular circumstances, an accused’s claim that the consequences of his act were unintentional and unexpected, making those consequences a mere accident, is naturally of no assistance. Finally, if the offence requires proof of objective fault — for instance, that the prohibited consequence was objectively foreseeable — then a claim of accident could negate that fault element if the prohibited consequence was such a chance occurrence that the trier of fact is left in a state of reasonable doubt as to whether, objectively, it was foreseeable. To avoid confusion in future cases, trial judges should focus on the questions of voluntariness and/or negation of mens rea, as appropriate, when instructing jurors on the so‑called “defence” of “accident”.

(7) Instructions Addressing Prejudice Against Indigenous Women and Girls in Sexual Assault Cases

There is no denying that Indigenous people — in particular Indigenous women, girls, and sex workers — have endured serious injustices, including high rates of sexual violence. Furthermore, the Court has acknowledged on several occasions the detrimental effects of widespread racism against Indigenous people within our criminal justice system. With this in mind, our criminal justice system and all participants within it should take reasonable steps to address systemic biases, prejudices, and stereotypes against Indigenous women and sex workers. As an additional safeguard going forward, in sexual assault cases where the complainant is an Indigenous woman or girl, trial judges would be well advised to provide an express instruction aimed at countering prejudice against Indigenous women and girls. However, any such instruction must not privilege the rights of the complainant over those of the accused. The objective would be to identify specific biases, prejudices, and stereotypes that may reasonably be expected to arise in the particular case and attempt to remove them from the jury’s deliberative process in a fair, balanced way, without prejudicing the accused.

(8) New Trial

Applying the test set out in R. v. Graveline, 2006 SCC 16, [2006] 1 S.C.R. 609, a new trial on unlawful act manslaughter is warranted. The failure to implement the s. 276 regime carried a significant risk that the jury would, whether consciously or unconsciously, engage in impermissible forms of reasoning on the central questions of whether the deceased subjectively consented to the sexual activity in question and, if not, whether the accused honestly but mistakenly believed she communicated her consent to that sexual activity at the time it occurred. The trial judge’s failure to implement the s. 276 regime was exacerbated by, and was inseparable from, the failure to caution the jury against mistakes of law masquerading as mistakes of fact when considering the defence of honest but mistaken belief in communicated consent. It can reasonably be thought that the trial judge’s errors had a material bearing on the accused’s acquittal for unlawful act manslaughter, and therefore a new trial should be ordered on that offence.

However, a new trial on first degree murder is not warranted. The Crown’s case on first degree murder turned primarily on its expert evidence that the deceased’s fatal wound was a cut. Evidently, the jury was not persuaded. Moreover, the Crown provided no plausible explanation for how the jury could have used prior sexual activity evidence to improperly reason its way through the first degree murder charge. Furthermore, the Crown acknowledged in the court below that the only ground of appeal implicating the murder charge was the motive issue. However, the motive instructions were not tainted by reversible error. Finally, there was a simple and obvious explanation for why the jury unanimously acquitted the accused of murder that does not require the Court to speculate about the potential influence of conscious or unconscious bias: the Crown’s theory simply did not hold up under scrutiny.

Per Wagner C.J. and Abella and Karakatsanis JJ. (dissenting in part):

Section 276 makes evidence of a complainant’s prior sexual activity inadmissible unless the accused complies with the criteria and procedures set out in ss. 276 , 276.1 and 276.2 . In this case, the trial judge permitted the accused to lead such evidence without following the procedure required by s. 276 , thereby allowing him to make unrestricted reference to the victim’s sexual history. He also failed to give the jurors any kind of limiting instruction to advise them that such evidence could not be used to show that the victim was more likely to have consented. All of this was compounded by the fact that the trial judge permitted, on dozens of occasions, the deceased to be referred to as a Native prostitute without providing any instruction to guard against potential prejudicial reasoning based on these descriptions. There was thus no filter for the victim’s prior sexual history and no specific warning to the jury to avoid drawing prejudicial and stereotypical assumptions about Indigenous women working in the sex trade. This left the jury with an essentially unchallenged version of the accused’s interactions with the victim. The trial judge failed to appreciate that the victim’s prior sexual conduct, occupation and race required the jury to be specifically alerted to the dangers of discriminatory attitudes toward Indigenous women, particularly those working in the sex trade. He provided no specific instructions crafted to confront the operative social and racial bias potentially at work. This rendered the whole trial unfair.

The devastatingly prejudicial effects of this error cannot be said to be confined to the included offence of manslaughter, but may also have had a material bearing on the jury’s reasoning on the charge of first degree murder. The prejudicial impact of the accused’s detailed testimony — without either the screening required by s. 276 or any limiting instructions — necessarily infected the whole trial and the entirety of the jury’s fact‑finding process. Indeed, the jury’s portrait of the victim was painted almost exclusively through the accused’s testimony, which meant that there was a significant possibility that the jury’s entire deliberations would have been based on fundamentally flawed — and prohibited — legal premises.

The potential for prejudicial reasoning was further exacerbated by the repeated description of the victim as a “prostitute”, and as a “Native”, without any limiting instruction from the trial judge. Specific safeguards are required in jury trials to prevent the systemic biases that can affect jury deliberations. Trial judges have an important role to play in instructing juries so that they can recognize and set aside racial and other biases, including those against Indigenous peoples and sex‑trade workers. Acknowledging, as this Court has for the last two decades, that racial prejudice is a social fact not capable of reasonable dispute, is not an insult to the jury system, it is a wake-up call to trial judges to be acutely attentive to the undisputed reality of pervasive prejudice and to provide the jury instructions required by law. Not only did that not happen here, the opposite occurred: inflammatory terminology was frequent, and was gratuitously used without any corrective intervention by the trial judge.

In summary, the trial judge’s failure to apply the requirements in s. 276 created a significant risk that the evidence of the victim’s prior sexual conduct not only tainted the jury’s perception of her character and conduct, but also fundamentally affected the factual foundation upon which their deliberations were based. This error permeated the entire trial and may have had a material bearing on the jury’s deliberations, affecting their verdicts for both murder and manslaughter. Given the prejudicial impact of these references, and the risk that they would affect the jury’s assessment of the victim and the accused’s credibility, it is difficult to see how it is realistically possible to conclude that their effect was confined to the jury’s verdict on manslaughter. The risk of harmful effects on the jury’s deliberations on murder would have been no less profound. When a trial with intimately connected issues, such as this one, is riddled with highly prejudicial testimony, it affects the very foundations of a jury’s fact‑finding function and decision making.

In addition, the trial judge’s error in the instruction on after‑the‑fact‑conduct is significant. In his own testimony, the accused admitted to lying, disposing of evidence and providing contradictory exculpatory explanations to numerous people after the victim’s death. It was open to the jury to conclude that additional incriminating after‑the‑fact conduct evidence came from the hotel video camera footage, physical evidence found by the police, and the testimony of numerous individuals. The accused did not call 911 immediately after finding the victim in the bathtub. Instead there was evidence that he attempted to erase his link to the scene by attempting to clean the bathroom, re‑arranging the bedding, putting his belongings in his van, checking out of the hotel room, and that he attempted to conceal and destroy evidence by throwing the bloody towel he had used to wipe the victim’s blood from his feet and the bathroom floor into a garbage can in the parking lot of the hotel. He also concocted and fabricated multiple stories and excuses. There is a strong possibility that, properly instructed, it would have had a material bearing on the jury’s assessment of the accused’s testimony and, ultimately, its verdict.

Instead, the jury was given contradictory and confusing directions. In effect, the trial judge did not leave it open to the jury to consider the impact of the after‑the-fact conduct evidence, such as the admitted exculpatory lies the accused told after the victim’s death, except when such evidence favoured an acquittal.

Juries, although expected to apply common sense, are above all expected to follow the instructions given by the trial judge. Where those instructions are confusing and contradictory, there is no roadmap for common sense to follow.

Citation: R. v. Barton, 2019 SCC 33

SCC File No. : 37769

Reasons for Judgment: Moldaver J. (Côté, Brown and Rowe JJ. concurring)

Dissenting Reasons: Abella and Karakatsanis JJ. (Wagner C.J. concurring)