The criminal justice system can be daunting, but you don’t need to go through it alone. Our Criminal lawyers are here to guide you every step of the way.
Contact Our Firm
Joseph Neuberger, Michael Bury and Diana Davison, Neuberger & Partners LLP, Criminal Lawyers Toronto
In R. v. E.O., 2026 ONCA 575, the Court of Appeal for Ontario returned to a question that keeps surfacing in sexual assault appeals: where is the line between a trial judge simply believing a complainant and a trial judge believing her for the wrong reasons? The distinction matters, because the first is an unreviewable finding of fact entitled to deference, while the second is an error of law that can send a case back for a new trial.
The court’s answer, delivered by Rouleau J.A. (Wilson and Madsen JJ.A. concurring), is a useful reminder that a single awkwardly worded paragraph rarely undoes an otherwise sound judgment.
(The case carries a publication ban under s. 486.4 of the Criminal Code, so the parties are identified only by initials.)
01 • THE BACKGROUND
The appellant was convicted of two counts of sex assault arising from two incidents of non-consensual slapping during otherwise consensual sexual activity. He received a nine-month conditional sentence followed by two years’ probation. His summary conviction appeal to the Superior Court was dismissed, and he reached the Court of Appeal with leave.
The appeal had nothing to do with the facts of the slapping. It was about how the trial judge reasoned his way to believing the complainant.
02 • TWO FLAVOURS OF FORBIDDEN REASONING
Ontario appellate law recognizes a couple of credibility shortcuts that judges are not supposed to take.
The first is treating a lack of embellishment as something that adds to a witness’s credibility — the intuition that, because a complainant “could have said something worse” but didn’t, she must be telling the truth. Cases like R. v. Alisaleh and R. v. Kiss caution against this. Restraint in an account is not affirmative proof of honesty.
The second is crediting a complainant because she was willing to endure the ordeal of a trial — the reasoning that no one would put themselves through the discomfort and scrutiny of a criminal trial unless their allegations were true. R. v. J.C. identifies this as an impermissible line of thought, because it effectively rewards the act of complaining.
The appellant argued that the trial judge had fallen into both traps.
03 • THE PARAGRAPH AT THE CENTRE OF IT ALL
The offending reasoning appeared late in the trial judge’s decision, where he was rejecting any suggestion that the complainant had a motive to fabricate. In substance, he asked: if she were making this up, why invent this story? Why place the events inside consensual sex and foreplay, and why expose herself to being questioned in public about intimate moments, rather than simply claim she had been slapped during an argument? A fabricator, he suggested, would have come up with a “better story.”
To the appellant, that passage did two forbidden things at once: it leaned on the absence of embellishment, and it leaned on the complainant’s willingness to testify.
The summary conviction appeal judge acknowledged both concerns — and then dismissed the appeal anyway, reasoning that these observations did not overcome the deference owed to the trial judge’s credibility findings, and that the lack of embellishment had not been used as a “makeweight.” The appellant seized on this: if the summary conviction appeal judge had accepted that legal errors occurred, he could only have upheld the convictions by applying the curative proviso — and he never did.
04 • WHAT THE COURT OF APPEAL SAW
Rouleau J.A. dismissed the appeal, and the reasoning is worth unpacking because it resolves the case on a more fundamental footing than the curative proviso.
The court found that the trial judge never actually invoked embellishment or exaggeration. Properly understood, he was assessing plausibility — responding to the implicit suggestion (built into the appellant’s denial) that the complainant had invented the incidents. His point was that a fabricator would likely have fabricated different allegations, ones not anchored in consensual activity. That is an assessment of how believable the narrative is, not a credibility bonus for holding back.
R. v. Kiss permits a judge to note that a witness did not embellish, so long as the observation is not improperly converted into a reason the witness is more credible. Read in context, the trial judge engaged with the absence of embellishment only on the narrow question of whether the complainant had a motive to lie — a purpose the Supreme Court expressly blessed in R. v. Gerrard.
The trial judge was not commenting on the fact that the complainant was prepared to testify. He was commenting on the content of what she testified about — consensual intimate moments — as part of gauging whether her account rang true. The distinction is subtle but real: querying why a fabricator would choose an embarrassing story is not the same as crediting a witness for her bravery in coming forward.
Drawing on R. v. G.F., Rouleau J.A. reaffirmed that trial judges are presumed to know the law. Where reasons are open to more than one interpretation, the reading consistent with that presumption prevails over the one that assumes error. Credibility findings, in particular, deserve deference, and reasons must be read generously and as a whole rather than parsed for isolated missteps.
On that footing, the summary conviction appeal judge — despite a passage that seemed, at one point, to concede error — ultimately reached the right conclusion: the trial judge had not erred, so there was simply nothing for the curative proviso to cure.
05 • WHY THE DECISION MATTERS
E.O. does not break new doctrinal ground, and that is rather the point. It is a clean application of principles that already govern how appellate courts read trial reasons in sexual assault cases.
A few practical takeaways stand out. Context beats extraction: a single paragraph will not be read in isolation when the surrounding reasons show credibility was established on other grounds. The line between permitted and forbidden embellishment reasoning turns on purpose — using it to weigh a motive to lie is fine, using it to inflate credibility is not. Commenting on the plausibility of what a complainant alleges is different from crediting her for alleging it. And the G.F. presumption remains a powerful headwind for appellants: ambiguity in a trial judge’s reasons is far more likely to be read charitably than fatally.
For counsel, the lesson cuts both ways. If you are drafting reasons, precision around credibility saves appeals. If you are challenging them, an inelegant sentence is not the same as an error of law — and after E.O., you will need to show the reasoning genuinely went astray, not merely that it could be read that way.
READ THE FULL DECISION HERE: R. v. E.O., 2026 ONCA 575
This post discusses R. v. E.O., 2026 ONCA 575. The case carries a publication ban under s. 486.4 of the Criminal Code. Nothing herein constitutes legal advice.
