Criminal Law: Sexual Offences; Sentencing; Appellate Review Standard
R. v. Sheppard, 2023 ABCA 381, 2025 SCC 29 (41126)
“S was convicted by a jury of one count each of sexual interference and invitation to sexual touching in respect of the complainant for crimes that occurred in the early 1990s, when the complainant was in grade 7 and S was a teacher at the complainant’s boarding school. In her sentencing decision, the sentencing judge set out the facts that she found were proven beyond a reasonable doubt at trial, including that S sexually violated the complainant approximately 10 times. After identifying aggravating factors and finding no mitigating factors and after considering the case law and relevant starting points for the offences in issue through the lens of the Court’s decision in R. v. Friesen, 2020 SCC 9, [2020] 1 S.C.R. 424, she concluded that a fit sentence was six years’ incarceration on each count, to be served concurrently.
A majority in the Court of Appeal allowed S’s appeal and varied the sentence to a global sentence of 3 years and 11 months’ incarceration. The two main bases for the majority’s intervention were that the sentencing judge erred in principle by providing insufficient reasons for her factual findings and by declining to consider pre‑Friesen sentencing jurisprudence when determining a fit sentence for the historical offences. The Crown appealed to the Court and S brought a motion to adduce new evidence in support of his arguments against reincarceration in the event the Crown appeal was successful.”
The SCC (9:0) held: the motion to adduce new evidence allowed; appeal allowed; trial sentence upheld.
The Chief Justice wrote as follows (at paras. 1-4, 45-50, 57-65, 77, 80-81, 83, 89-90, 98, 117):
“Sexual offences against children are among the most profoundly immoral acts an individual can commit. Historical sexual offences against children are no less grave, and demand no less accountability, than ones committed today. This appeal provides an opportunity to clarify the applicability of the principles set out in R. v. Friesen, 2020 SCC 9, [2020] 1 S.C.R. 424, to historical sexual offences against children and to reaffirm the standard of appellate review set out in R. v. Lacasse, 2015 SCC 64, [2015] 3 S.C.R. 1089.
The respondent, Paul Sheppard, was convicted by a jury of sexual interference and invitation to sexual touching in respect of the complainant. The crimes occurred in 1993 and 1994, when the respondent was a teacher at a boarding school where the complainant was enrolled. The sentencing judge sentenced the respondent to six years’ incarceration on each count, to be served concurrently.
The respondent appealed his sentence. The majority in the court below allowed the appeal and varied his sentence from 6 years’ incarceration to 3 years and 11 months’ incarceration. The court intervened on two main bases: first, the sentencing judge’s reasons for her factual findings were insufficient, and second, she erred by declining to consider pre-Friesen sentencing jurisprudence.
The Crown appealed the sentence reduction to this Court. Following the hearing, this Court delivered its judgment orally from the bench. We unanimously allowed the appeal, restored the original sentence of six years’ incarceration, and ordered the respondent to surrender himself to the proper authorities to serve the remainder of his custodial sentence. We also unanimously allowed the respondent’s motion to adduce new evidence.
…
Reviewing reasons functionally means assessing whether they are sufficient not in the abstract, but in relation to the fundamental purposes that reasons serve (Sheppard 2002, at para. 42). Those purposes are to hold judges accountable to the public, to provide an intelligible result to the parties, and to enable effective appellate review (paras. 15, 42 and 55; G.F., at para. 68). What is necessary to achieve those purposes will depend on the circumstances of the case and, in particular, the issues that were in dispute (G.F., at para. 68; R. v. Dinardo, 2008 SCC 24, [2008] 1 S.C.R. 788, at para. 31).
Under a contextual approach, “[t]he foundations of the judge’s decision must be discernable, when looked at in the context of the evidence, the submissions of counsel and the history of how the trial unfolded” (R.E.M., at para. 17). As this Court explained in R. v. Parranto, 2021 SCC 46, [2021] 3 S.C.R. 366, “[i]n sentencing, as with any decision, the reasons must, when read in conjunction with the record, show why the judge reached a particular result” (para. 39; see generally R.E.M., at para. 37). Sentencing reasons must also be read and understood in conjunction with the reasons for conviction or, in the case of a jury trial, the verdict and the essential facts it implies.
On this approach, an appellate court may intervene only if the reasons are legally or factually insufficient or if they are so muddled that meaningful appellate review is impossible (G.F., at paras. 71 and 108; Sheppard 2002, at paras. 28 and 46; R.E.M., at para. 17). Legal sufficiency requires that the aggrieved party be able to understand the legal basis of the decision in order to meaningfully exercise any right of appeal (G.F., at para. 74; Sheppard 2002, at paras. 64-66). However, it does not demand that the sentencing judge expound on features of the criminal law that are uncontroversial in the context of the case. Sentencing judges are presumed to know the basic principles of criminal law at issue in the case (R.E.M., at para. 45). Factual sufficiency requires that the factual basis for the decision be discernible, even if articulated unartfully (G.F., at para. 71; Sheppard 2002, at para. 55).
The functional and contextual approach rejects a fine parsing of reasons in search of error (R. v. Chung, 2020 SCC 8, [2020] 1 S.C.R. 405, at paras. 13 and 33). In G.F., Karakatsanis J. expressed concern that some appellate courts continue to parse reasons, often in a manner that attacks the trial judge’s credibility findings:
- Despite this Court’s clear guidance in the 19 years since Sheppard to review reasons functionally and contextually, we continue to encounter appellate court decisions that scrutinize the text of trial reasons in a search for error, particularly in sexual assault cases, where safe convictions after fair trials are being overturned not on the basis of legal error but on the basis of parsing imperfect or summary expression on the part of the trial judge. Frequently, it is the findings of credibility that are challenged. [para. 76]
I echo my colleague’s concern, which is no less salient in the context of sentence appeals. On a sentence appeal, the factual findings and basic credibility determinations that are implied by the verdict are not up for debate. The sentencing judge may be called upon to make further findings of fact or credibility determinations that relate only to issues on sentencing (Criminal Code, s. 724). While the basis for those further findings must be discernible upon a contextual review of the record, the credibility findings themselves are owed “particular” deference on appeal (G.F., at para. 81; see also Dinardo, at para. 26).
The foregoing principles guide appellate courts in determining whether reasons are insufficient, which is an error in principle (Nahanee, at para. 59(ii)). Under the Lacasse framework, however, the appellant must not only identify an error in principle but must also show that the error impacted the sentence. In my view, the “impact” criterion dovetails with the functional approach to reviewing the sufficiency of reasons: appellate intervention is permitted only when a deficiency renders the sentence incomprehensible or appellate review impossible. When this is the case, the appellate court is justified in assuming that the deficiency impacted the sentence, since the inherent nature of insufficient reasons is that they make the presence of faulty reasoning impossible to prove.
…
Wakeling J.A. objected that the sentencing judge should have addressed various reliability concerns, which he identified in his reasons. They included the fact that the events took place 28 years earlier; that the complainant’s recall was limited and skeletal in nature; that the complainant had communicated with the respondent years later and never accused him of any wrongdoing that the complainant’s testimony at trial was more detailed than the account they gave to the police; that the complainant made two implausible claims, namely, that they remembered being spanked 34 times and that most sexual acts took place in a room that was not private; that the complainant was a troubled child at the time of the alleged offences; and that the complainant had been to therapy since the events, which Wakeling J.A. said could have distorted their memory.
After canvassing these concerns, Wakeling J.A. concluded that “it would be very dangerous to rely on any other evidence than that the jury must have accepted to find the accused guilty on the three counts” (para. 165). In essence, Wakeling J.A. determined that the complainant’s testimony was not reliable, full stop, and that the sentencing judge should have declined to make any factual findings over and above those that she was bound to accept because they were essential to the jury’s verdict.
I reject this line of reasoning. The reliability concerns that Wakeling J.A. identified served only to challenge the complainant’s reliability generally and could not logically form a basis for distinguishing between various aspects of their testimony. In my view, Wakeling J.A.’s conclusion amounts to an indirect rejection of the jury’s guilty verdict, which was not up for debate on sentencing.
Furthermore, I agree with Crighton J.A. and the intervener, the Raoul Wallenberg Centre for Human Rights, that many of the issues Wakeling J.A. identified engage discredited myths and stereotypes. This Court has repeatedly and unequivocally held that “myths and stereotypes have no place in a rational and just system of law” (R. v. Kruk, 2024 SCC 7, at para. 43, quoting R. v. A.G., 2000 SCC 17, [2000] 1 S.C.R. 439, at para. 2). In particular, the law no longer tolerates any suggestion that the complainant is less credible because they did not confront their abuser. As the Alberta Court of Appeal explained in R. v. A.R.D., 2017 ABCA 237, 422 D.L.R. (4th) 471, aff’d 2018 SCC 6, [2018] 1 S.C.R. 218, there is no single or expected way for victims of sexual abuse to behave, and their failure to distance themselves from their abuser cannot be taken to undermine their credibility (para. 42, citing R. v. D.D., 2000 SCC 43, [2000] 2 S.C.R. 275, at paras. 63 and 65, per Major J., dissenting, but not on this point; see also Kruk, at para. 41).
Similarly, the suggestion that the complainant’s history of undergoing therapy per se impugned their credibility or reliability is not tenable. This Court stated in R. v. Mills, [1999] 3 S.C.R. 668, that “[t]he notion that consultation with a psychiatrist is, by itself, an indication of untrustworthiness is a more recent, but equally invidious, example of such a myth” (para. 119; see also R. v. J.J., 2022 SCC 28, [2022] 2 S.C.R. 3, at para. 132; Kruk, at para. 41). Absent some reason to believe that the complainant’s recollection was distorted by therapy, the conclusion that the sentencing judge ought to have counted this as a reliability concern is wrong in law (A.F., at para. 47; I.F., at para. 15).
Finally, I reject the idea that the complainant was less credible or reliable because they were already troubled and isolated at the time of the assaults or because they described their time at the boarding school as traumatic. Far from undermining their reliability, the complainant’s testimony suggested that the respondent targeted them because they were troubled and isolated (A.R., vol. II, at p. 73). Certainly the complainant’s account of life at the boarding school as traumatic was not inconsistent with their allegations of sexual violence.
Building upon Wakeling J.A.’s criticisms, Feehan J.A. held that the sentencing judge’s reasons were insufficient because they did not clarify which factual findings were essential to the jury’s verdict and which were made through her own independent fact-finding process. He interpreted Ferguson as imposing a requirement on sentencing judges to categorize factual determinations in that manner. In his view, without an explanation of which facts were essential and which were not, it was impossible to understand the basis of the sentencing judge’s decision or to conduct a meaningful appellate review (paras. 286 and 289).
Respectfully, I fail to see the problem. Nowhere in Ferguson did McLachlin C.J. assert that sentencing judges are required to follow a two-step fact-finding process, as Feehan J.A. suggested. The question of which facts are essential to a verdict is a question of law, not one of discretion. On the basis of the essential elements of the offence, it is possible to deduce what the essential facts were. Indeed, in the very same paragraph in which Feehan J.A. faulted the sentencing judge for failing to specify which facts were essential, he himself identified which facts were essential (para. 288). The sentencing judge was not required to explain legal matters that were uncontroversial before her, such as the essential elements of the offences of which the respondent was convicted (R. v. Burns, [1994] 1 S.C.R. 656, at p. 664), nor was she required to spell out her reasoning process in a “watch me think” fashion (R.E.M., at para. 17).
The court below erred by intervening on the basis that the sentencing judge’s reasons were insufficient. When the sentencing judge’s reasons are viewed in context, it is clear how she arrived at her factual findings, and she was not obligated to spell out which findings were essential to the jury’s verdict and which she made through her own fact-finding process.
…
There is no constitutional reason or compelling logical justification for continuing to enforce artificially lenient common law sentencing ranges for historical offences (R.O., at para. 52; Stuckless, at para. 61). So long as the sentence imposed does not exceed the maximum penalty at the time of the offence, sentencing judges should base their reasoning on the most accurate and up-to-date understanding of the gravity of the offence (Friesen, at para. 87; R.O., at para. 50; Stuckless, at para. 110; R. v. X, 2022 QCCA 266, at para. 22).
…
While contemporary case law should guide the analysis, sentencing judges are entitled to consider dated case law insofar as it aligns with contemporary principles. I agree with the Alberta Court of Appeal’s direction in M.A.C. that pre-Friesen sentencing decisions are not automatically irrelevant, particularly where the facts are highly analogous; rather, “the sentencing court must carefully consider those precedents in light of Friesen and whether those sentences appropriately reflect the harm to the children” (para. 50).
Generally speaking, the usefulness of a historical decision will depend on its factual comparability to the case at hand, as well as on the degree to which relevant legal principles, societal attitudes, and legislative provisions have evolved since the decision was rendered. For example, it is safe to assume that historical jurisprudence will not be useful, at least not on its face, when considering offences that are now viewed through a markedly different lens.
…
To reiterate, sentencing judges are entitled to consider historical precedents, though only insofar as they align with contemporary sentencing principles. This should not be taken to suggest that sentencing judges are required to consider such precedents or that declining to do so is an error in principle.
…
The court below erred in intervening on the basis that the sentencing judge failed to consider historical jurisprudence. She was not required to do so and committed no error in her approach to sentencing for historical offences. Finally, the court below found additional errors that formed, in part, the basis for its intervention. With respect, these errors did not justify appellate intervention.
…
I have concluded that there was no valid basis for the court below to disturb the sentence imposed by the sentencing judge and that the original sentence should be restored. While it is not strictly necessary to address the manner in which the majority in the court below conducted the resentencing exercise, I would take this opportunity to reiterate this Court’s guidance on starting points as set out in Parranto and Friesen.
…
In the circumstances of this case, I am satisfied that denunciation and deterrence cannot be achieved without reincarceration given the gravity of the respondent’s offences, and in particular his pattern of abusing children while in a position of power. These objectives are especially pressing when considering reincarceration for sexual offences against children, as Parliament has legislated the primacy of denunciation and deterrence in relation to these offences (Criminal Code, s. 718.01). I also note that the respondent still has the majority of his sentence left to serve, and, unlike in other cases, the Crown has not conceded that reincarceration should not be ordered (see, e.g., Bertrand Marchand, at para. 102; Proulx, at para. 132). On the whole, the interests of justice compel reincarceration in the circumstances of this case. The respondent is therefore to be reincarcerated to serve the remainder of his custodial sentence.”