Wilson v R: Guarding Against Standard of Proof Errors in Judge-Alone Trials
In Wilson v R, 2013 NBCA 38 [Wilson], the New Brunswick Court of Appeal provides a thoughtful judgment that develops the law of evidence surrounding the criminal standard of proof in judge-alone cases. In a unanimous decision, Richard JA of the Court of Appeal examines the trial judge’s reasons, locates a reasoning error previously recognized in judge-and-jury contexts, and proposes a modernized formulation of an existing rule that guards against that same error in judge-alone decisions. That error, first recognized and guarded against in R v W(D),  1 SCR 742, is to fail to recognize that an accused testimony, even if not believed, can ground a reasonable doubt in the mind of the trier of fact.
The Decision of the Trial Judge
Following an incident at a bar in St. John, New Brunswick, Bradley Scott Wilson was charged with aggravated sexual assault, assault causing bodily harm, and unlawful confinement (respectively sections 273(2)(b), 267(b), and 279(2)(a) of the Criminal Code, RSC 1985, c C-46). At trial, Mr. Wilson and the complainant offered very divergent accounts of what occurred. With no other testimony speaking to the events in question, the trial judge had little else to consider. After hearing the arguments and taking a twenty minute adjournment, the trial judge filed written reasons and orally rendered his decision to convict the accused of sexual assault. The basic structure of the written reasons are as follows:
- Recognition of inconsistencies: After summarizing the facts, the trial judge noted that there were “a number of inconsistencies and weaknesses” in Mr. Wilson’s version of events. He specifically mentions only one “inconsistency or weakness,” that of Mr. Wilson’s testimony that he had left a one-hundred dollar bill in his jacket. The trial judge wrote that “as I understand human nature, people are much more careful with hundred dollar bills than to leave them with their coat when they could put them in their pants pocket or somewhere safe” (R v Wilson, 2012 NBQB 286 at para 81 [“Trial Judgment”]).
- Finding of disbelief: The trial judge then explicitly states that he does not believe this portion of the accused testimony: “I do not believe his story about the hundred dollar bill” (Trial Judgment at para 81).
- Statement of judicial role: He then states what he takes to be his role in this matter: “Although I do not believe that, I have to consider whether the Crown has proven guilt beyond a reasonable doubt. I have to attempt to resolve the inconsistencies and understand the whole context in this case” (para 81).
- Conclusion of guilt: Finally, the trial judge writes that “after considering all the circumstances and the the legal principles for the cases that apply and the definition of the Criminal Code, I am satisfied beyond a reasonable doubt that the oral sexual activity and related sexual touching was not consensual between the parties (para 86).
Mr. Wilson was sentenced to 24 months, less time spent while remanded, and three years of probation with conditions. He appealed his conviction on a single ground: “The judge did not apply the law regarding reasonable doubt.” (Wilson, para 25).
The Issue and Arguments before New Brunswick Court of Appeal
The issue before the Court of Appeal was whether the trial judge gave sufficiently clear reasons for his decision that the prosecution’s case satisfied him beyond a reasonable doubt? Mr. Wilson, for his part, argued that because the sole issue at trial related to credibility, the judge was required to give clear reasons for his decision to convict. In particular, Mr. Wilson argued that those “clear reasons” ought to have demonstrated that the trial judge followed the reasoning framework set out in R v W(D).
In reply, the Attorney General pointed out that trial judges are not required to recite the W(D) formula in each case, that they are presumed to know the law, and that the key issue was actually whether the judge had applied the proper standard of proof. In the Attorney General’s view, the trial judge had done so (Wilson, paras 25-26).
The Framework of R v W(D) – Avoiding Standard of Proof Errors in Jury Trials
R v W(D), a case involving two charges of sexual assault, essentially pitted the testimony of the complainant against the testimony of the accused. Credibility was the central issue, and the accused denied guilt. In a recharge to the jury, the trial judge “characterized the core issue to be determined by the jury as whether they believed the complainant or whether they believed the accused” (R v W(D), para 21).
All Supreme Court judges present found this directive in error, as the charge manifested the danger that the trial could become a ‘credibility contest’ in the minds of the jurors and that as a result, they might improperly apply the standard of proof. That is, without proper instruction, the jury might erroneously convict an accused simply because they do not believe his testimony. The error recognized here is reasoning from disbelief of an accused’s testimony to a finding of guilt, without considering whether the testimony, although not believed, nevertheless raises a reasonable doubt.
To avoid this error in cases with juror as trier of fact, Cory J proposed three part formula for a trial judge to follow when instructing the jury on the standard of proof where the accused denies guilt and credibility is in issue at para 28 of R v W(D):
First, if you believe the accused, obviously you must acquit.
Second, if you do not believe the testimony of the accused but you are left in reasonable doubt by it, you must acquit.
Third, even if you not left in doubt by the evidence of the accused, you must ask yourself whether, on the basis of the evidence you do accept, you are convinced beyond a reasonable doubt by that evidence of the guilt of the accused.
In addition to this three-part instruction formula, Cory noted that a trial judge need not use this language explicitly, and that “the failure to use such language is not fatal if the charge, when read as a whole, makes it clear that the jury could not have been under any misapprehension as to the correct burden and standard of proof to apply” (para 29). We can call this the “holistic assessment” of a charge on reasonable doubt.
Again, the goal of R v W(D) was to guard against a particular danger: that in instances where credibility was in issue, a jury would fail to apply the proper standard of proof and infer guilt from disbelief of an accused’s testimony. The jury must be aware that an accused testimony, although not believed, may ground a reasonable doubt. It is precisely this danger that is found the Richard JA in Wilson during his consideration of the trial judge’s application of the standard of proof.
Decision of the Court of Appeal
Richard JA begins by agreeing with the Attorney General that the trial judge was not required to recite the R v W(D) formula in his reasons, and that the trial judge is presumed to know the basic principles of criminal law. However, he does find that the judge erred in his application of the criminal law standard of proof.
The Trial Judge Misunderstood his Role
Richard JA’s first objection to the trial judge’s reasons is that the trial judge improperly described the role of the trial court in cases surrounding credibility. Recall from point “3.” above that the trial judge wrote that he had to “attempt to resolve the inconsistencies and understand the whole context in this case.” Richard JA calls this an incorrect statement of law: the trail judge did not have to attempt to resolve any inconsistencies – rather, his role is to consider unexplained inconsistencies while determining what testimony to believe. Further, these inconsistencies should also be considered to determine whether reasonable doubt lingers notwithstanding disbelief (Wilson at para 49).
The Trial Judge Fell Prey to the W(D) Danger
Richard JA’s second observation is that the trial judge’s reasons demonstrate that he ignored the second step of the W(D) formula (para 56). Recall from above that the trial judge found that he did not believe Mr. Wilson’s testimony. The reasons, without more, suggest that he then reasoned from this finding of disbelief to a decision that the evidence he did accept convinced him beyond a reasonable doubt. In doing so, the trail judge did not consider whether the evidence of Mr. Wilson that he did not accept might have raised a reasonable doubt. At paragraph 56, Richard JA writes:
The flaw in the impugned reasoning in [this] case is that it wholly ignores the second step in the W(D) reasoning process. The judge went from step 1 – deciding whether to believe the accused — directly to step 3, in which he decided whether the evidence he accepted convinced him beyond a reasonable doubt. In the process, he ignored whether the evidence not accepted might have raised that doubt. This was exactly the flaw in reasoning meant to be corrected by the W(D) formula.
Further the Court of Appeal remarks that the trial judge’s decision makes it clear that Mr. Wilson’s testimony “was never subjected to the reasonable doubt standard because the judge did not consider it to be true beyond a reasonable doubt” (para 58). Put this way, the error is to misapply the standard of reasonable doubt. The standard of reasonable doubt functions as a way of testing the overall strength of the prosecution’s case, and not as a standard for admitting or excluding testimony from consideration.
For these reasons, the Court of Appeal finds that the trial judge erred in his application of the standard of proof.
Modernizing W(D) for Judge Alone Trials
Again, Richard JA recognizes that the W(D) formula need not be strictly followed as long as the proper standard of proof is applied. Richard JA also recognizes that generally, trial judge’s guilty verdict can ground the inference that the accused’s denial of a charge failed to raise a reasonable doubt in that judge’s mind. For this proposition, he cites R v Boucher,  3 SCR 499, a recent Supreme Court case in which the court recognized that if the accused is not believed, trail judge is implicitly addressing the first two steps in W(D) (Wilson at para 57).
However, Richard JA holds that this W(D)-satisfying inference is not available in cases where the decision of the trial judge demonstrates a flawed reasoning process. Recall that in what we called the “holistic assessment” of a reasonable doubt charge W(D) Cory J noted that guiding question was whether the tier of fact (here, the jury) could have been under a misapprehension of the correct standard of proof to apply.
Here, in Wilson, Richard JA takes this same sentiment and applies it to the judge-alone context. In a judge-alone case, where there is no jury charge to be scrutinized, the judge’s reasons should be examined to determine whether the decision was rendered using the correct standard of proof. Richard JA proposes the following statement of law which incorporates the Boucher judge-alone inference and the overall holistic assessment from R v W(D):
If, in a judge alone case, the accused is not believed, the judge is implicitly addressing the first two steps in W(D), unless the reasons, when read as a whole, make it clear that the judge was under some misapprehension as to the correct burden and standard of proof.” (para 57)
Conclusion and Comments
Wilson represents an important development in the law surrounding the criminal standard of proof. In particular, Richard JA’s proposal of a modernized statement of the W(D) safeguard protects against bad reasoning in an area of law that trades in very high stakes. The burden on the Crown to satisfy the trier of fact beyond a reasonable doubt, easily stated but explained with difficulty, is rests at the core of our criminal law. In fact, in W(D), Sopinka J called it “the most fundamental rule of the game” (at page 751). It is vitally important that our triers of fact, judges and juries alike, have a firm understanding of the concept and of its proper application. Wilson helps ensure this.