In judge-alone trials, the judge must give reasons for the verdict. This obligation creates difficulties for the judge, especially around adequately explaining reasons for assessments of the credibility and reliability of witnesses. There can be a tendency for judges to refer to their common sense and their experience of the ways of the world. This might lead to a departure from the evidence in the case. How should an appellate court determine whether the judge has reasoned lawfully?
This was the central question in R v Kruk, 2024 SCC 7.
The Court unanimously rejected a rule-based approach called “the rule against ungrounded common-sense assumptions” [1].
There were numerous reasons for that rejection, but of more general interest is the articulation(s) of the correct appellate approach to judicial assumptions not supported (or controverted) by evidence.
Two judgments were delivered, with Rowe J agreeing in the results of these appeals (in two unrelated cases) but setting out the analysis he prefers. How different is this from that of the other judges - Wagner CJ, Coté, Martin, Kasirer Jamal and O’Bonsawin JJ - in their joint judgment?
The joint judgment sets out “the existing and well-established law on assessing a trial judge’s credibility or reliability assessments” at [93]-[99]. As their summary is given “for the utmost clarity” [93], it seems pointless to summarize their summary.
But in asking whether Rowe J’s approach is different, comparison of [93]-[99] with [129]-[132] is necessary. He describes the issue in these terms: “These two appeals ask how appellate courts should review trial judges’ reliance on generalized expectations based on common sense and human experience in the fact-finding process” [128]. Broadly, appellate courts need to be sure that what the judge relied on was indeed a generalized expectation and not an assessment of evidence in the case ([130] - this is what the joint judgment says at [94]), then, if reliance had been placed on a generalized expectation, ask whether that expectation was reasonable ([131] - here the joint judgment at [95] diverts to address unreasonable assumptions and to consider how these should be reviewed on appeal [96]-[97]), and if it was reasonable, ask whether it was used to replace evidence instead of being a benchmark for assessing the evidence ([132] and here the joint judgment follows an identification of error by asking whether it was “palpable” in the sense that it affected the result or went to the very core of the outcome of the case [98]).
One kind of error is an error of law, and the standard for review is simply whether the judge got the law right [96]. Examples of this open-ended category are given at [96]. If the error was not one of law, the standard for review is whether the error was palpable and overriding, and examples are given at [97].
A case that was overruled in Kruk provides a quite amusing (at least, I think so) illustration of how absurd it is to require generalizations that are advanced to support credibility and reliability findings to be grounded in evidence: R v JC, 2018 ONSC 5547. See Kruk per Rowe J at [211]-[213], and the joint judgment at [21]-[23]. Absurd, because “The Crown cannot be expected to elicit evidence on how sexual encounters ordinarily unfold in every sexual assault trial before a trial judge can rely on their common sense or human experience with respect to human sexual behaviour” [211].
