SUPREME COURT OF CANADA

Citation: R. v. Sinclair, 2011 SCC 40, [2011] 3 S.C.R. 3

Date: 20110728

Docket: 33359

Between:

Her Majesty The Queen

Appellant

and

Terrence Sinclair

Respondent

• and -

Director of Public Prosecutions of Canada

and Attorney General of Ontario

Interveners

Coram: McLachlin C.J. and Binnie, LeBel, Deschamps, Fish, Abella, Charron, Rothstein and Cromwell JJ.

Reasons Dissenting in Result: (paras. 1 to 42)

Reasons for Judgment: (paras. 43 to 63)

Reasons Concurring in Result With Those of LeBel J.: (paras. 64 to 86)

Reasons Concurring in Result With Those of LeBel J.: (para. 87)

Fish J. (McLachlin C.J. and Binnie and Cromwell JJ. concurring)

LeBel J. (Deschamps and Rothstein JJ. concurring)

Charron J.

Abella J.

R. v. Sinclair, 2011 SCC 40, [2011] 3 S.C.R. 3

Her Majesty The Queen Appellant

v.

Terrence Sinclair Respondent

and

Director of Public Prosecutions of Canada

and Attorney General of Ontario Interveners

Indexed as: R. v. Sinclair

2011 SCC 40

File No.: 33359.

2010: December 14; 2011: July 28.

Present: McLachlin C.J. and Binnie, LeBel, Deschamps, Fish, Abella, Charron, Rothstein and Cromwell JJ.

on appeal from the court of appeal for manitoba

Criminal law — Appeals — Powers of Court of Appeal — Unreasonable verdict — Misapprehension of evidence — Accused convicted of manslaughter — Court of Appeal setting aside conviction on basis of trial judge’s misapprehension of evidence — Whether Court of Appeal applied proper test in setting aside conviction — Whether trial judge misapprehended evidence — Criminal Code, R.S.C. 1985, c. C-46, s. 686(1)(a)(i).

A driver observed three males kicking and stomping Adam Lecours on a roadway. They left him lying on the street bleeding. Another driver failed to see Mr. Lecours lying in the road in time to avoid hitting him. Mr. Lecours died from his injuries. The trial judge convicted accused and a co‑accused of manslaughter. The Court of Appeal set aside the conviction on the basis that the trial judge had misapprehended evidence by accepting that there was evidence that the accused had planned to go out to commit a robbery together with his co‑accused and a youth, and by inferring from this evidence that the accused was with his co‑accused at the scene of the crime. In setting aside the conviction, the Court of Appeal partially based its decision on this Court’s decision in R. v. Beaudry, 2007 SCC 5, [2007] 1 S.C.R. 190.

Held (McLachlin C.J and Binnie, Fish and Cromwell JJ. dissenting): The appeal should be allowed. The order for a new trial should be set aside and the conviction restored.

I. The Beaudry Issue

Per McLachlin C.J. and Binnie, Fish and Cromwell JJ.: R. v. Beaudry has no application to the error committed by the trial judge in this case. This error falls squarely within R. v. Lohrer, 2004 SCC 80, [2004] 3 S.C.R. 732. Nothing in Beaudry modifies the principles set out in R. v. Yebes, 1987 CanLII 17 (SCC), [1987] 2 S.C.R. 168, and R. v. Biniaris, 2000 SCC 15, [2000] 1 S.C.R. 381. Yebes and Biniaris continue to apply where the issue is whether the verdict could have been reached reasonably by a properly instructed jury or a judge sitting alone. Beaudry serves an important function of limited scope on an appellate review for unreasonableness under s. 686(1)(a)(i) of the Criminal Code. It addresses the reasonableness of the judge’s verdict, notably by scrutinizing the logic of the judge’s findings of fact or inferences drawn from the evidence admitted at trial. A trial judge who is not mistaken as to the evidence but reaches a verdict by an illogical or irrational reasoning process commits an error under Beaudry. A misapprehension of evidence invites appellate scrutiny under Lohrer. To conflate these conceptually distinct errors is to disregard the rationale of Beaudry and the jurisprudential gap it fills.

A verdict reached illogically or irrationally is “unreasonable” because it is not reached judicially, or in accordance with the rule of law. Further, as stated in Beaudry, a verdict that was reached illogically or irrationally is hardly made reasonable by the fact that another judge (who never did and never will try the case) could reasonably have convicted or acquitted the accused. A court of appeal may intervene pursuant to Beaudry where a trial judge draws an inference or makes a finding of fact that is plainly contradicted by the evidence relied on for that purpose by the judge or that is demonstrably incompatible with evidence that is not otherwise contradicted or rejected by the trial judge. Unreasonable verdicts of this sort are rare. But when they do occur, appellate courts are authorized — indeed bound — to intervene.

Per LeBel, Deschamps and Rothstein JJ.: The Court of Appeal erred in law to the extent that it based its judgment on R. v. Beaudry. Beaudry has no application to a mistake as to the substance of the evidence. Rather, errors of this sort are governed by the standard set out in R. v. Lohrer. Beaudry expanded the traditional scope of s. 686(1)(a)(i) of the Criminal Code. Fish J. carried a majority in Beaudry on the question whether a verdict reached “illogically or irrationally” was “unreasonable” within the meaning of s. 686(1)(a)(i) and his reasons in Beaudry represent the state of the law on this point. Furthermore, the reasons of Fish J. in this case correctly reflect the position that was taken by the majority in Beaudry, and there is concurrence with his reasons to that extent.

Per Abella and Charron JJ.: The Court of Appeal should not have rested its judgment on R. v. Beaudry. Beaudry did not change any of the existing tests for appellate review under s. 686(1)(a). Beaudry did, however, effect an expansion of the test for unreasonableness under s. 686(1)(a)(i). The expanded test for unreasonableness is better understood if the focus remains on the verdict reached at trial, rather than on the kind of errors that may have occurred in the appreciation of the evidence or in the fact‑finding process. While the reviewing court may now consider flaws in the reasoning process in the assessment for reasonableness, the focus must remain on the conclusion reached at trial. A verdict‑focussed approach avoids reviewing courts having to make rather fine distinctions between different types of mistakes in order to identify the applicable test. It is also more consistent with the wording, legislative context and history of s. 686(1)(a)(i) of the Criminal Code. Any error in reasoning identified under the Beaudry test must be carefully assessed in context to determine whether it vitiates the verdict. It would be wrong to conclude that whenever there is an error in reasoning, even if of the kind “demonstrably incompatible” with the uncontradicted evidence at trial, the verdict is unreasonable.

Under a s. 686(1)(a)(i) inquiry in judge‑alone trials, the reviewing court should first apply the Yebes/Biniaris test. If the verdict is not one that a properly instructed jury, acting judicially, could reasonably have rendered on the evidence as a whole, the inquiry is over, and an acquittal is entered. If the verdict is available on the evidence, the reviewing court may go on to assess the reasonableness of the verdict under the Beaudry test by scrutinizing the actual findings of fact and inferences made by the trial judge. It is not necessary to conduct this assessment in every case. The Beaudry test may apply in exceptional cases where the reasoning process is so irrational, or so at odds with the evidence, that it vitiates the verdict — even though that verdict is available on the evidence as a whole. In these rare cases where Beaudry applies, a new trial is ordered, as the court will already have determined that the verdict is otherwise available on the evidence.

II. The Application of R. v. Lohrer

Per LeBel, Deschamps, Abella and Rothstein JJ.: The Court of Appeal erred in its application of Lohrer. It failed to identify any error that justified appellate intervention. The trial judge did not find that the accused was involved in a plan to commit a robbery and she did not rely on such evidence to infer that the accused was with his co‑accused throughout the night and was therefore one of the assailants. The trial judge based her finding that the accused was present at the crime scene on other evidence: he was out of the residence at the same time as his co‑accused at time of the assault; he knew that his co‑accused had been stabbed and stormed into the house minutes after the incident; he led efforts to locate his co‑accused; DNA linked the co‑accused to the crime; and the co-accused’s blood was found on the accused’s jeans and hoodie. Nothing supports a finding that she accepted the Crown’s theory of a planned robbery, and her reasoning discloses no mischaracterization of the evidence. Furthermore, the Court of Appeal referred to the trial judge’s “apparent acceptance” of evidence that the accused had planned to commit robbery with his co‑accused. Ordering a new trial based on a misapprehension of evidence requires more than an apparent mistake. The reasons must disclose an actual mistake.

Even if the trial judge did err, the misapprehension was not materially linked to her verdict. The accused was charged with manslaughter, not conspiracy to commit robbery. In light of the circumstantial evidence the trial judge accepted, it cannot seriously be doubted that she would nonetheless have found the accused was at the crime scene at the time of the assault.

Per Charron J.: The trial judge did not misapprehend the evidence. For the reasons set out by LeBel J., the Court of Appeal erred in finding that the trial judge made an error that met the standard set out in R. v. Lohrer.

Per McLachlin C.J. and Binnie, Fish and Cromwell JJ. (dissenting): The trial judge misapprehended the substance of the evidence on a material matter, and the Court of Appeal explicitly and correctly applied the Lohrer test. The Court of Appeal’s reference to the trial judge’s apparent acceptance that there was evidence of the accused and his two co‑accused planning to go and commit another robbery aptly describes an actual mistake. The trial judge’s mistaken view of the evidence coloured her evaluation of the other circumstantial evidence upon which the Crown relied. The supposed plan of the accused and his co‑accused to stay together throughout the night underpinned her inference that the accused was at the scene of the crime. The evidence does not establish the accused’s participation in the assault or his presence at the assault. No DNA evidence links him to the scene. Most importantly, even if his presence at the scene had been established, that would not have proven that he participated in the assault.

Cases Cited

By Fish J. (dissenting)

Applied: R. v. Lohrer, 2004 SCC 80, [2004] 3 S.C.R. 732; explained: R. v. Beaudry, 2007 SCC 5, [2007] 1 S.C.R. 190; referred to: R. v. Sheppard, 2002 SCC 26, [2002] 1 S.C.R. 869; R. v. Yebes, 1987 CanLII 17 (SCC), [1987] 2 S.C.R. 168; R. v. Biniaris, 2000 SCC 15, [2000] 1 S.C.R. 381; R. v. Morrissey (1995), 1995 CanLII 3498 (ON CA), 97 C.C.C. (3d) 193; R. v. Pittiman, 2006 SCC 9, [2006] 1 S.C.R. 381.

By LeBel J.

Applied: R. v. Lohrer, 2004 SCC 80, [2004] 3 S.C.R. 732; explained: R. v. Beaudry, 2007 SCC 5, [2007] 1 S.C.R. 190; referred to: R. v. Yebes, 1987 CanLII 17 (SCC), [1987] 2 S.C.R. 168; R. v. Biniaris, 2000 SCC 15, [2000] 1 S.C.R. 381; R. v. O’Brien, 2011 SCC 29, [2011] 2 S.C.R. 485; R. v. C.L.Y., 2008 SCC 2, [2008] 1 S.C.R. 5; R. v. Morrissey (1995), 1995 CanLII 3498 (ON CA), 97 C.C.C. (3d) 193.

By Charron J. (concurring with LeBel J. in the result)

Applied: R. v. Lohrer, 2004 SCC 80, [2004] 3 S.C.R. 732; explained: R. v. Beaudry, 2007 SCC 5, [2007] 1 S.C.R. 190; referred to: R. v. Yebes, 1987 CanLII 17 (SCC), [1987] 2 S.C.R. 168; R. v. Biniaris, 2000 SCC 15, [2000] 1 S.C.R. 381; R. v. Morrissey (1995), 1995 CanLII 3498 (ON CA), 97 C.C.C. (3d) 193; R. v. Pittiman, 2006 SCC 9, [2006] 1 S.C.R. 381; R. v. Jaw, 2009 SCC 42, [2009] 3 S.C.R. 26.

By Abella J. (concurring with LeBel J. in the result)

Referred to: R. v. Beaudry, 2007 SCC 5, [2007] 1 S.C.R. 190.

Statutes and Regulations Cited

Criminal Code, R.S.C. 1985, c. C‑46, ss. 686(1)(a), (b).

Authors Cited

Canadian Oxford Dictionary, 2nd ed. Don Mills, Ont.: Oxford University Press, 2004, “apparent”.

APPEAL from a judgment of the Manitoba Court of Appeal (Monnin, Hamilton and Freedman JJ.A.), 2009 MBCA 71, 240 Man. R. (2d) 135, 456 W.A.C. 135, 245 C.C.C. (3d) 331, 69 C.R. (6th) 163, [2009] 8 W.W.R. 581, [2009] M.J. No. 252 (QL), 2009 CarswellMan 342, setting aside the conviction entered by Simonsen J., 2007 MBQB 219, 219 Man. R. (2d) 63, [2007] M.J. No. 324 (QL), 2007 CarswellMan 352, and ordering a new trial. Appeal allowed, McLachlin C.J and Binnie, Fish and Cromwell JJ. dissenting.

Elizabeth A. Thomson and Ami Kotler, for the appellant.

Richard J. Wolson, Q.C., and Evan J. Roitenberg, for the respondent.

James D. Sutton and Carole Sheppard, for the intervener the Director of Public Prosecutions of Canada.

Joan Barrett, for the intervener the Attorney General of Ontario.

The reasons of McLachlin C.J. and Binnie, Fish and Cromwell JJ. were delivered by

Fish J. (dissenting) —

I

[1] Terrence Sinclair was tried and convicted of manslaughter in the Manitoba Court of Queen’s Bench (2007 MBQB 219, 219 Man. R. (2d) 63). The Court of Appeal set aside Mr. Sinclair’s conviction and ordered a new trial (2009 MBCA 71, 240 Man. R. (2d) 135). This is an appeal by the Crown against that judgment, with leave of the Court, and the issue raised concerns the governing test for unreasonable verdicts under s. 686(1)(a)(i) of the Criminal Code, R.S.C. 1985, c. C-46, pursuant to R. v. Beaudry, 2007 SCC 5, [2007] 1 S.C.R. 190.

[2] I agree with the Court of Appeal that the appellant’s conviction must be set aside and that a new trial must be had. More particularly, and with respect, I agree that the trial judge misapprehended the substance of the evidence on a material matter and that this error was essential to the trial judge’s reasoning. Finally, I agree that the trial judge’s misapprehension of the evidence is therefore fatal to her verdict.

[3] While a verdict that rests on a mistake as to the substance of the evidence may well be “unreasonable” in the broad sense of that term, Beaudry has no application to errors of this sort. Rather, they are governed by R. v. Lohrer, 2004 SCC 80, [2004] 3 S.C.R. 732. Nor does Beaudry govern trial court decisions that are rendered inscrutable by an absence of sufficient reasons, as in R. v. Sheppard, 2002 SCC 26, [2002] 1 S.C.R. 869.

[4] Nothing in Beaudry should be taken as a departure from the well-established principles of appellate review set out in R. v. Yebes, 1987 CanLII 17 (SCC), [1987] 2 S.C.R. 168, and R. v. Biniaris, 2000 SCC 15, [2000] 1 S.C.R. 381. Yebes and Biniaris continue to apply where the issue is whether the verdict could have been reached reasonably by a properly instructed jury or a judge sitting alone. Beaudry, as we shall see, involves a narrower inquiry. Its concern is whether it can be seen from the reasons for judgment that the trial judge’s conclusion ― that is to say, the judge’s verdict ― was reached illogically or irrationally. As Justice Charron puts it, Beaudry is concerned with “fundamental flaws in the reasoning process that led to [the trial judge’s verdict]” (para. 77).

[5] In short, Beaudry serves an important function of limited scope on an appellate review for unreasonableness under s. 686(1)(a)(i) of the Code. And I agree with the Crown that the Court of Appeal, to the extent that it appears in some measure to have done so, should not have rested its judgment in this case on Beaudry. The decisive error it identified does, however, fall squarely within Lohrer, and the Court of Appeal, notably at paras. 94-97 of its judgment, applied the Lohrer test explicitly and correctly. I would affirm the judgment of the Court of Appeal for that reason, and for the reasons that follow.

II

[6] Mr. Sinclair was jointly tried with Dallas Pruden-Wilson on charges of manslaughter and aggravated assault. The trial judge convicted them both of manslaughter. Having concluded that a new trial should be had, I will say no more about the facts than is necessary to explain my conclusion.

[7] Essentially, the trial judge rested her finding of guilt on the circumstantial evidence adduced by the Crown. In her view, that evidence supported the inferences that Mr. Sinclair and the co-accused were together throughout the night and that Mr. Sinclair participated in the assault of the victim, Adam Lecours. It is undisputed that Mr. Lecours was left lying in the street by his assailants, where he was later run over and killed by a passing automobile.

[8] In her reasons for judgment, the trial judge began by noting what she understood as a reference by the Crown to “evidence that, prior to the assault, [Mr.] Pruden-Wilson, [Mr.] Sinclair and [a] youth [separately charged] planned to go out together to commit another robbery” (para. 21). This, along with evidence that the three accused were out of the house at the same time and other apparently incriminating circumstantial evidence, satisfied the trial judge that Mr. Sinclair was guilty of manslaughter (paras. 21-22).

[9] The Court of Appeal set aside the verdict unanimously. In the court’s view, the trial judge’s reasons were generally “clear, comprehensive, cogent and consistent with the evidence” (para. 72), but contained a fatal error. The trial judge, said the court, had misapprehended the evidence on an essential point: She mistook for evidence the Crown’s theory concerning a planned second robbery and Mr. Sinclair’s presence when it was committed (para. 94).

[10] After a thorough review of the entire trial record, the Court of Appeal concluded that no such evidence had been adduced, and that the inference drawn by the trial judge was thus based in part on non-existent testimony. Because there may nonetheless have been “evidence capable of supporting a conviction”, it ordered a new trial (paras. 99-100).

III

[11] The Court of Appeal framed the issue before it this way:

We must, therefore, ask whether in reaching her verdict, the judge relied (and, we would add, on a material matter) on evidence not in the record. If she did, then according to Beaudry, her reasons cannot serve as a basis for the verdict. [Emphasis added; para. 93.]

[12] I agree, of course, that a verdict based on an erroneous understanding of the evidence with regard to a material matter cannot support a finding of guilt. As mentioned earlier, I agree as well that this sort of misapprehension of the evidence may well render the verdict at trial “unreasonable” within the meaning of s. 686(1)(a)(i) of the Criminal Code.

[13] In R. v. Morrissey (1995), 1995 CanLII 3498 (ON CA), 97 C.C.C. (3d) 193 (Ont. C.A.), Doherty J.A. explained that misapprehensions of the evidence include not only a mistake as to the substance of the evidence, but also “a failure to consider evidence relevant to a material issue” and “a failure to give proper effect to evidence” (p. 218). I agree with Justice Doherty that “[i]n cases tried without juries, a finding that the trial judge did misapprehend the evidence can . . . figure prominently in an argument that the resulting verdict was unreasonable” (p. 220).

[14] While a misapprehension of the evidence may help to establish that a verdict is unreasonable within the meaning of Yebes and Biniaris, Beaudry does not govern an error of this sort.

[15] Rather, the Beaudry test addresses the reasonableness of the judge’s verdict, notably by scrutinizing the logic of the judge’s findings of fact or inferences drawn from the evidence admitted at trial. A trial judge who is mistaken as to the evidence admitted at trial misapprehends the evidence, inviting appellate scrutiny under Lohrer, not Beaudry; a trial judge who is not mistaken as to the evidence but reaches a verdict by an illogical or irrational reasoning process commits an error under Beaudry, not Lohrer. These are conceptually distinct errors. To conflate them is to disregard the rationale of Beaudry and the jurisprudential gap it has filled.

[16] An appellate court will thus be justified in intervening, pursuant to Beaudry, where a trial judge draws an inference or makes a finding of fact that is (1) plainly contradicted by the evidence relied on for that purpose by the judge, or (2) demonstrably incompatible with evidence that is not otherwise contradicted or rejected by the trial judge.

[17] In Beaudry, all nine justices agreed that a verdict may be unreasonable even if supported by the evidence. While four members of the Court pointed in this regard only to inconsistent verdicts (R. v. Pittiman, 2006 SCC 9, [2006] 1 S.C.R. 381, at para. 14, cited in Beaudry, at para. 57), five justices characterized as unreasonable, within the meaning of s. 686(1)(a)(i) of the Criminal Code, decisions by a trial judge “reached illogically or irrationally” (my reasons, at para. 97; Binnie J., concurring in substance, at paras. 77-80).

[18] I think it helpful, for the sake of context, to reproduce here this passage from my reasons in Beaudry:

It is important to remember that the unreasonable verdict test has more often than not been described and explained in cases involving jury trials, where particular considerations govern: See, for example, R. v. Yebes, 1987 CanLII 17 (SCC), [1987] 2 S.C.R. 168.

Unlike judges, juries are neither required nor even permitted to give reasons. Their reasons, in this country at least, are forever shrouded in the compelled secrecy of their deliberations: Section 649 of the Criminal Code prohibits the disclosure, subject only to narrow exceptions, of any information relating to the proceedings of a jury. Why the jury concluded as it did is thus beyond the ken of the court, both at trial and on appeal.

Appellate courts are no more entitled to speculate about than to know how or why the jury reached its verdict. The jury is presumed to have been composed of reasonable people acting reasonably. It follows that its verdict must be deemed to be reasonable unless no properly instructed jury could reasonably have reached that verdict or, in the words of s. 686(1)(a)(i), it “cannot be supported by the evidence”.

The same, however, cannot be said for the verdict of a judge. The delivery of reasoned decisions is inherent in the judge’s role: R. v. Sheppard, [2002] 1 S.C.R. 869, 2002 SCC 26. And evaluating the reasonableness of those reasons is not a matter of speculation. Moreover, while the necessary degree of detail will vary according to the circumstances of the case, the reasons must be sufficient to permit meaningful review on appeal. There would be no need for this requirement if the sole test of unreasonableness under s. 686(1)(a)(i) is whether there is any evidence capable of supporting the verdict. [Emphasis in original; paras. 90-93.]

[19] Illogical or irrational reasoning can render verdicts unreasonable under s. 686(1)(a)(i) of the Code, in various ways. Beaudry identifies two. First, a verdict is unreasonable where the judge draws an inference or makes a finding of fact essential to the verdict that is “plainly contradicted by the very evidence from which it was drawn” or upon which it has been made to rest (para. 105). In that case, the essential finding is illogical or unreasonable in light of the evidence relied upon in making the finding. Here, the rule of law speaks the language of logic: From accepted evidence “X”, a court cannot lawfully infer “not X”.

[20] I found in Beaudry that the trial judge had committed an error of this sort. The issue in that case was whether the accused had the specific intent to obstruct, pervert or defeat the course of justice when he decided not to take the breath samples needed to charge a fellow officer. In concluding that he did, the trial judge relied on evidence which in my view demonstrated that Mr. Beaudry did not seek to conceal the offence and did not wish to foreclose the possibility of prosecution (para. 111). Accordingly, I found that the judge’s inference that Mr. Beaudry intended to obstruct justice was plainly contradicted by the very evidence from which it was drawn and that his verdict was therefore “unreasonable” within the meaning of s. 686(1)(a)(i).

[21] A verdict is likewise unreasonable where the judge draws an inference or makes a finding of fact essential to the verdict if that inference or finding of fact is “‘demonstrably incompatible’ with evidence that is neither contradicted by other evidence nor rejected by the trial judge” (Beaudry, at para. 79, per Binnie J.). Here, an essential inference that appears compatible with the evidence from which it is drawn implodes on contextual scrutiny. This is so where a trial judge infers from some of the available evidence that an accused had the requisite intent to obstruct justice and this inference is supportable on the evidence relied upon but is not supportable on consideration of other evidence that has been neither contradicted by other evidence nor rejected by the trial judge. Here, too, a verdict of guilt “would lack legitimacy and would properly . . . be treated as ‘unreasonable’” (ibid.).

[22] Fortunately, unreasonable verdicts of either sort are exceedingly rare. But when they do occur, appellate courts are authorized ― indeed, bound ― to intervene.

[23] The remedy will depend on the circumstances of the case. Where the verdict is found to be unreasonable under Beaudry and, in any event, unavailable on the record, an acquittal will ensue pursuant to Biniaris. But where the verdict is found to be unreasonable under Beaudry and the record discloses “evidence capable of supporting a conviction”, a new trial will be ordered (Beaudry, at para. 97).

[24] The rationale for ordering a new trial in the latter case was explained this way in Beaudry:

No one should stand convicted on the strength of manifestly bad reasons . . . on the ground that another judge (who never did and never will try the case) could but might not necessarily have reached the same conclusion for other reasons. A verdict that was reached illogically or irrationally is hardly made reasonable by the fact that another judge could reasonably have convicted or acquitted the accused. [Emphasis in original; para. 97.]

[25] Here, “manifestly bad reasons” refers to the fatal flaw in the reasoning process that underpins the trial judge’s conclusion, not to the insufficiency or inadequacy of the reasons in other respects. As mentioned earlier, Beaudry complements but neither duplicates nor displaces Sheppard, which continues to govern the requirement of adequate reasons to ensure meaningful appellate review. Indeed, in the absence of reasons that are adequate under Sheppard, a reviewing court could hardly assess their reasonableness under Beaudry.

[26] A verdict reached illogically or irrationally is “unreasonable” because it is not reached judicially, or in accordance with the rule of law.

IV

[27] I agree with the Court of Appeal that the record in this case contains “evidence capable of supporting a conviction” (para. 100). Since a properly instructed jury, acting judicially, could reasonably have reached the same conclusion as the trial judge, her verdict is not unreasonable within the meaning of Biniaris. Nor can it be characterized as unreasonable within the more limited scope of Beaudry.

[28] Accordingly, as mentioned at the outset, I agree with the Crown that the Court of Appeal, to the extent that it appears in some measure to have done so, could not rest its judgment in this case on Beaudry. But I think it right to remember that the Court of Appeal did not have the benefit of these reasons, which more fully explain the conceptual distinction between Beaudry and Lohrer.

[29] As I mentioned as well, I believe the Court of Appeal nonetheless reached the right conclusion for substantially the right reason.

[30] Speaking for a unanimous Court in Lohrer, Binnie J. stated:

This is an appeal as of right from convictions of the appellant for aggravated assault and uttering a threat. A majority of the B.C. Court of Appeal affirmed the convictions. Hollinrake J.A. dissented. He found applicable to this case what was said by Doherty J.A. of the Ontario Court of Appeal in R. v. Morrissey (1995), 1995 CanLII 3498 (ON CA), 97 C.C.C. (3d) 193, as follows at p. 221:

Where a trial judge is mistaken as to the substance of material parts of the evidence and those errors play an essential part in the reasoning process resulting in a conviction, then, in my view, the accused’s conviction is not based exclusively on the evidence and is not a “true” verdict.

Later in the same paragraph, Doherty J.A. stated:

If an appellant can demonstrate that the conviction depends on a misapprehension of the evidence then, in my view, it must follow that the appellant has not received a fair trial, and was the victim of a miscarriage of justice. This is so even if the evidence, as actually adduced at trial, was capable of supporting a conviction.

We agree with these observations. Where a miscarriage of justice within the meaning of s. 686(1)(a)(iii) of the Criminal Code, R.S.C. 1985, c. C-46, has been demonstrated an accused appellant is not bound to show in addition that the verdict cannot “be supported by the evidence” within the meaning of s. 686(1)(a)(i). [para. 1]

[31] The decisive error identified by the Court of Appeal in this case falls squarely within Lohrer, and the court applied the Lohrer test explicitly and correctly, notably at paras. 94-97 of its extensive and unanimous reasons:

The judge appears to have accepted that there was evidence that Sinclair and the other two planned to go out together to commit another robbery. She said that there was evidence that they were out together. She described [these] matters as evidentiary, when they were not. As has been explained above, this was more than factual findings based on inference; this amounted to a mischaracterization or misapprehension of the evidence.

Since the judge treated as evidentiary that which was not, she did not consider inferences, other than that the three men were out together, that were available to be drawn on the evidence. For example, there was evidence, particularly of Audy relating to the potential use of the phone at the store, that could have led the judge to a different inference about Sinclair’s whereabouts and involvement.

Moreover, these mischaracterizations [in essence, misapprehensions of the evidence] were not on peripheral points. As stated by Binnie, J., for the Supreme Court in R. v. Lohrer . . . (at para. 2): “The misapprehension of the evidence must go to the substance rather than to the detail. It must be material rather than peripheral to the reasoning of the trial judge.” It was obviously a fundamental question whether Sinclair was at the scene of the crime. The two “evidentiary” matters, among other evidence, on which the judge relied to establish that he was indeed at the scene were essential to her ultimate verdict. It cannot be stated with confidence that, had she not mischaracterized the evidence, she would have found as she did.

[32] I believe it important to add that the Court of Appeal, in ordering a new trial, took care to recall expressly that “[a]ppellate courts should not parse and dissect reasons of trial judges as if under a microscope” (para. 72). I agree, of course. In my respectful view, care must likewise be taken not to attribute to a single word in the reasons of the Court of Appeal a fatal significance divorced from its context and contrary to its literal and ordinary meaning.

[33] The Court of Appeal was troubled, as am I, by the trial judge’s “apparent acceptance . . . that there was evidence of Sinclair, Pruden-Wilson and the young person planning to go out together to commit another robbery” (para. 61). According to the Canadian Oxford Dictionary (2nd ed. 2004), at p. 61, the primary meaning of “apparent” is “readily visible or perceivable”. And, in this sense, the judge’s error is indeed apparent ― not speculative, or even imputed by inference.

[34] After noting the Crown’s submission that circumstantial evidence established Mr. Sinclair’s participation in the assault of the victim, the trial judge states: “The Crown refers to the evidence that, prior to the assault, Pruden-Wilson, Sinclair and the youth planned to go out together to commit another robbery” (para. 21 (emphasis added)). With respect for those who read this sentence differently, it seems plain to me that the Court of Appeal’s reference to the trial judge’s “apparent acceptance” of this “evidence” is entirely grounded in the trial judge’s own words. It is, I repeat, hardly “speculative”: It aptly describes an actual mistake.

[35] Moreover, this mistaken view of the evidence appears from the trial judge’s ensuing reasons to have coloured her evaluation of the other circumstantial evidence upon which the Crown relied. Citing an example, the Court of Appeal explained: “Since the judge treated as evidentiary that which was not, she did not consider inferences, other than that the three men were out together, that were available to be drawn on the evidence” (reproduced above, at para. 31).

[36] The supposed plan of the three accused to commit a robbery ― and to stay together for this purpose throughout the night ― underpinned the judge’s inference that Mr. Sinclair was at the scene of the crime. It was the first step of the reasoning that led her to that conclusion. At no point did the judge state explicitly, or even suggest indirectly, that the “evidence” referred to by the Crown simply did not exist.

[37] For the sake of clarity, I reproduce the judge’s words once more: “The Crown refers to the evidence that, prior to the assault, Pruden-Wilson, Sinclair and the youth planned to go out together to commit another robbery” (emphasis added). Bearing in mind the plain and ordinary meaning of these words and the context in which they appear, I see no basis for assuming that the trial judge meant “the Crown submits” when she said the “Crown refers to the evidence”.

[38] Finally, and with respect, I am unable to share Justice LeBel’s view (at para. 60) that the evidence established Mr. Sinclair’s participation in the assault or even his presence at the scene when the crime was committed.

[39] First, there was “no DNA evidence linking [Mr. Sinclair] to the scene” (para. 21 of the trial judge’s reasons). Second, Mr. Sinclair bore no burden to prove that he was not the third “mysterious individual whose identity remains unknown to this day” (para. 60 of Justice LeBel’s reasons). It was for the Crown to prove that he was.

[40] Third, Mr. Sinclair was not bound to explain the undated cuts in his jeans and lacerations of his legs. The evidence in this regard is inconclusive on its own, and unsupported by forensic evidence of any sort. It is no more probative of Mr. Sinclair’s presence at the scene of the crime than the fact that blood of another accused was on the victim’s knife.

[41] Fourth, and perhaps most important, mere presence at the scene of the crime ― even if it had been established ― would not suffice to prove, as a matter of fact or matter of law, that Mr. Sinclair had participated in the assault. Indeed, as Crown counsel conceded on the hearing of this appeal, there was evidence that “earlier in the evening, [Mr. Sinclair] had been present but had not participated in another robbery” (transcript, at p. 32).

V

[42] For all of these reasons, I would dismiss this appeal, affirm the judgment of the Court of Appeal, set aside the respondent’s conviction, and order a new trial.

The judgment of LeBel, Deschamps and Rothstein JJ. was delivered by

[43] LeBel J. — I have had the benefit of reading the reasons of my colleague Fish J. With respect, I cannot agree with him that the Manitoba Court of Appeal was right to order a new trial on the ground that the trial judge had misapprehended the evidence in her reasons and had thus made an error that met the standard set out in R. v. Lohrer, 2004 SCC 80, [2004] 3 S.C.R. 732. For the reasons that follow, I would allow the appeal and restore the respondent’s conviction.

[44] Before turning to the core of my disagreement with my colleague, I wish to briefly address the legal issues raised by this case. First and foremost, the Court must clarify in this appeal whether Fish J., dissenting in the result in R. v. Beaudry, 2007 SCC 5, [2007] 1 S.C.R. 190, expanded the “traditional scope” of s. 686(1)(a)(i) of the Criminal Code, R.S.C. 1985, c. C-46 (at para. 77), as delineated in R. v. Yebes, 1987 CanLII 17 (SCC), [1987] 2 S.C.R. 168, and R. v. Biniaris, 2000 SCC 15, [2000] 1 S.C.R. 381. On this point, I agree that, because Binnie J. concurred in substance in Fish J.’s reasons in Beaudry, Fish J. carried a majority on the issue of whether a verdict reached “illogically or irrationally” (para. 97) is “unreasonable” within the meaning of s. 686(1)(a)(i) of the Criminal Code (reasons of Fish J., at para. 17). I also acknowledge that Fish J.’s reasons in Beaudry represent the state of the law on this point.

[45] In light of this conclusion, the second — and more subtle — issue in the case at bar concerns the nature of the error requiring appellate intervention under s. 686(1)(a)(i) that Beaudry purported to identify. Fish J. sets out clearly, in comments with which I fully agree, what Beaudry does not stand for (at paras. 3-4):

While a verdict that rests on a mistake as to the substance of the evidence may well be “unreasonable” in the broad sense of that term, Beaudry has no application to errors of this sort. Rather, they are governed by R. v. Lohrer, 2004 SCC 80, [2004] 3 S.C.R. 732. Nor does Beaudry govern trial court decisions that are rendered inscrutable by an absence of sufficient reasons, as in R. v. Sheppard, 2002 SCC 26, [2002] 1 S.C.R. 869.

Nothing in Beaudry should be taken as a departure from the well-established principles of appellate review set out in R. v. Yebes, 1987 CanLII 17 (SCC), [1987] 2 S.C.R. 168, and R. v. Biniaris, 2000 SCC 15, [2000] 1 S.C.R. 381.

Fish J. adds that “the Beaudry test addresses the reasonableness of the judge’s verdict, notably by scrutinizing the logic of the judge’s findings of fact or inferences drawn from the evidence admitted at trial” (para. 15 (emphasis in original)). Despite the fact that I concurred with the opinion of Charron J. in Beaudry, I am convinced that Fish J.’s reasons in the instant case correctly reflect the position that was taken by the majority in that case. Accordingly, I concur with his reasons to that extent.

[46] My disagreement with Fish J. does not therefore pertain to his interpretation of how Beaudry fits into the expanded scope of s. 686(1)(a)(i). Nor does it relate to his opinion that, to the extent that the Manitoba Court of Appeal appears to have based its judgment in the instant case on Beaudry, it erred in law. Rather, the only point on which I disagree with Fish J. relates to the application of Lohrer to the facts of this case. I do not agree that the Court of Appeal identified an error in the trial judge’s reasons that “fall[s] squarely within Lohrer” (reasons of Fish J., at para. 5). In my respectful opinion, the trial judge did not commit an error that justified appellate intervention.

[47] Hamilton and Freedman JJ.A., writing for a unanimous Court of Appeal, held that the trial judge had misapprehended the evidence by asserting that Mr. Sinclair had been involved in a plan to commit a robbery (2009 MBCA 71, 240 Man. R. (2d) 135). In their view, as a result of this misapprehension, the trial judge’s reasoning process was flawed in two respects. They wrote:

The judge drew an inference that Sinclair and Pruden-Wilson were together throughout, but she based this inference partially on that “evidence” referred to by the Crown, when there was no such evidence. Moreover, she then found that Sinclair was one of the participants in the assault, basing this in part on the “evidence” that he and Pruden-Wilson were out together. [para. 71]

Fish J. agrees with this reading of the trial judge’s reasons. He writes that “[s]he mistook for evidence the Crown’s theory concerning a planned second robbery and Mr. Sinclair’s presence when it was committed” (para. 9 (emphasis in original)).

[48] In the view of Hamilton and Freedman JJ.A., this two-fold misapprehension of the evidence thus amounted to a miscarriage of justice within the meaning of s. 686(1)(a)(iii) of the Criminal Code. They saw it as a basis, along with a number of other factors, for the inferences drawn by the trial judge that Mr. Sinclair was at the crime scene at the time Mr. Lecours was brutally assaulted and that he participated in the assault. They added the following comment:

Moreover, these mischaracterizations were not on peripheral points. . . . It was obviously a fundamental question whether Sinclair was at the scene of the crime. The two “evidentiary” matters, among other evidence, on which the judge relied to establish that he was indeed at the scene were essential to her ultimate verdict. It cannot be stated with confidence that, had she not mischaracterized the evidence, she would have found as she did. [Emphasis added; para. 97.]

[49] While I agree that the record does not disclose any evidence of a planned robbery involving Mr. Sinclair, I do not agree that the trial judge relied on such “evidence” to infer that Mr. Sinclair was at the crime scene. In my view, the trial judge did not actually commit the error the Court of Appeal said she had committed. She drew two key inferences to find Mr. Sinclair guilty: (1) that he was at the crime scene; and (2) that he was one of Mr. Lecours’s assailants (2007 MBQB 219, 219 Man. R. (2d) 63). It is important not to be misled about the evidence that led the trial judge to draw these two inferences. Regarding Mr. Sinclair’s presence at the crime scene, she considered the circumstantial evidence adduced by the Crown and found

• that Mr. Sinclair, Mr. Pruden-Wilson and the youth were all out of the residence at the time of the assault on the victim (para. 21);

• that Mr. Sinclair knew Mr. Pruden-Wilson had been stabbed, because (1) he stormed into the house only minutes after the incident and (2) he led the subsequent efforts to locate Mr. Pruden-Wilson while the latter was still bleeding in the snow (paras. 12, 19 and 21); and

• that DNA evidence linked Mr. Pruden-Wilson to the crime, and Mr. Pruden-Wilson’s blood was found on Mr. Sinclair’s jeans and hoodie (para. 21).

[50] From this evidence, which she assessed correctly, the trial judge drew the inference “that Sinclair and Pruden-Wilson were together throughout and that Sinclair was at the scene” (para. 21). The record supported this inference even though there was no evidence of a plan to commit a robbery, and the trial judge’s reasoning discloses no mischaracterization of the evidence. With respect, it is inaccurate to state that the trial judge relied on the existence of a plan to commit a robbery to infer that Mr. Sinclair was at the crime scene. Indeed, nowhere in her reasons can support be found for the view that she accepted the Crown’s theory of a planned robbery involving Mr. Sinclair. All she did was to refer to such a plan as part of the Crown’s theory (at para. 21):

As for Sinclair, there is no DNA evidence linking him to the scene. Rather, the evidence is circumstantial. The Crown submits that the circumstantial evidence proves beyond a reasonable doubt Sinclair’s presence at the scene and his involvement in the unlawful act of assaulting Lecours. The Crown refers to the evidence that, prior to the assault, Pruden-Wilson, Sinclair and the youth planned to go out together to commit another robbery. Although nobody saw the three leave 110 Kushner together, there was clearly a period of time, in the early morning hours, when all three were out of the residence. Further, when Sinclair returned to 110 Kushner, he knew that Pruden-Wilson had been stabbed. The Forensic Reports show that Pruden-Wilson’s blood was on Sinclair’s jeans and hoodie. As well, Audy’s evidence was that Sinclair led their subsequent efforts to locate Pruden-Wilson, and Constable Tully testified that Audy and Sinclair were running (and arrested) in the area where Pruden-Wilson was in the snow, bleeding. I am satisfied that the only rational inference to be drawn from all of this evidence is that Sinclair and Pruden-Wilson were together throughout and that Sinclair was at the scene. [Emphasis added.]

[51] Although I concede that this passage from the trial judge’s reasons could have been better written, it is not ambiguous. Read in context, it does not support a conclusion that the trial judge mistook the Crown’s theory about a planned robbery involving Mr. Sinclair for evidence. Rather, she stated clearly that she inferred that Mr. Sinclair was at the crime scene despite the fact that there was no evidence, such as the simultaneous departure of the three young men, confirming the existence of a plan. In my opinion, the trial judge’s reasons thus indicate that she was satisfied beyond a reasonable doubt — regardless of whether the Crown had made out its case with respect to the planned robbery — that Mr. Sinclair was at the crime scene. In the first part of the paragraph quoted above, the trial judge presented some of the Crown’s submissions, while in the second part, she explained how she reached the conclusion that Mr. Sinclair was at the crime scene. The Court of Appeal was not justified in turning the trial judge’s description of a part of the Crown’s theory into an acceptance of its evidentiary foundation.

[52] I wish to emphasize the fact that the Court of Appeal did not find that the impugned error was explicit in the trial judge’s reasons. Indeed, Hamilton and Freedman JJ.A. referred to her “apparent acceptance . . . that there was evidence of Sinclair, Pruden-Wilson and the young person planning to go . . . commit another robbery” (para. 61 (emphasis added)). In the absence of an explicit acceptance by the trial judge, the Court of Appeal appears to have “inferred” that her explanation of the basis for her inference was flawed. As a result, the Court of Appeal’s decision to order a new trial in this case rests solely on this implicit, but “apparent”, mistake.

[53] When reasons are reviewed on appeal, silence should not be taken to mean acceptance (R. v. O’Brien, 2011 SCC 29, [2011] 2 S.C.R. 485, at para. 16). In my opinion, for an appellate court to decide to order a new trial on the basis of a miscarriage of justice resulting from a misapprehension of the evidence, more is needed than an “apparent” mistake (e.g., an error that the trial judge may have committed) in the reasons. A court of appeal should not, in applying the Lohrer test, order a new trial unless the trial judge has made a real error; its decision cannot be speculative. The plain language or the thrust of the reasons must disclose an actual mistake. When such errors are in fact committed, appellate courts have no difficulty in explaining why they caused the trial judge’s reasoning process to be fatally flawed and where they may be found in the reasons. In such situations, the errors are readily obvious.

[54] In R. v. C.L.Y., 2008 SCC 2, [2008] 1 S.C.R. 5, Abella J. observed that in the oft-cited decision in R. v. Morrissey (1995), 1995 CanLII 3498 (ON CA), 97 C.C.C. (3d) 193 (Ont. C.A.), on which the Lohrer test is based, Doherty J.A. had rightly “caution[ed] appellate judges not to dissect, parse, or microscopically examine the reasons of a trial judge” (para. 11). With respect, the Court of Appeal’s reasoning in the case at bar, which my colleague Fish J. endorses, disregards this cautionary rule. The order for a new trial results from an overly critical reading of the trial judge’s reasons and is not faithful to the thrust of her reasoning. This is illustrated by the fact that neither the Court of Appeal nor Fish J. in this Court has been able to identify a point in the trial judge’s reasons at which she accepted the Crown’s theory that Mr. Sinclair was involved in a plan to commit a robbery, or to satisfactorily explain why this “misapprehension” was central to her reasoning process.

[55] It is therefore my view that the trial judge did not infer that Mr. Sinclair was at the crime scene on the basis of the “evidence” referred to by the Crown concerning a planned robbery. It is clear from the plain language of her reasons that, after considering all the relevant evidence, including evidence that the individuals were all out of the residence at the time of the assault, that Mr. Sinclair reported that Mr. Pruden-Wilson had been stabbed shortly after it happened, and that Mr. Pruden-Wilson’s blood was found on Mr. Sinclair’s jeans and hoodie, she inferred from this that the three young men had been “together throughout” and that Mr. Sinclair had been at the crime scene.

[56] Even if I were prepared to accept, for the sake of the argument, that the trial judge agreed with the Crown’s theory that Mr. Sinclair had been involved in a plan to commit a robbery, it would nonetheless be my view that the misapprehension was not materially linked to the inference she drew with respect to Mr. Sinclair’s presence at the crime scene. For a misapprehension of evidence to be material within the meaning of the Lohrer test, it must go to a central element of the trial judge’s reasoning on which the conviction is based (Lohrer, at para. 2, per Binnie J.). As Binnie J. correctly stated in Lohrer, the standard described by Doherty J.A. in Morrissey is a stringent one (para. 2). In other words, an error in the assessment of the evidence will amount to a miscarriage of justice only if striking it from the judgment would leave the trial judge’s reasoning on which the conviction is based on unsteady ground.

[57] In the case at bar, striking the “error” would not have this effect. In light of the circumstantial evidence the trial judge accepted, it cannot seriously be doubted that she would nonetheless have found that Mr. Sinclair was at the crime scene at the time of the assault on Mr. Lecours. In my opinion, if the trial judge could in fact be assumed to have accepted the “evidence” of a plan to commit a robbery, that acceptance would clearly have been incidental to her finding that Mr. Sinclair was at the crime scene. How little importance she attached to the alleged plan to commit a robbery should also come as no surprise: Mr. Sinclair was charged with manslaughter, not with conspiracy to commit robbery. Although evidence of such a plan would likely have strengthened the inference that Mr. Sinclair and Mr. Pruden-Wilson had been “together throughout”, it did not need to be central to the trial judge’s reasoning, because it was not an element of the Crown’s burden.

[58] In his reasons, Fish J. downplays the persuasiveness of the evidence on which the trial judge based the verdict by referring to it as “apparently incriminating circumstantial evidence” (para. 8). In this respect, the fact that Mr. Sinclair reported that Mr. Pruden-Wilson had been stabbed only minutes after it happened and the fact that his clothes were stained with Mr. Pruden-Wilson’s blood are undisputed pieces of evidence that are more than “apparently incriminating”. Furthermore, my colleague’s interpretation of the weight of this circumstantial evidence does not accord with the trial judge’s reasoning. It is important to reflect upon the meaning of the position he has adopted.

[59] In essence, Fish J. suggests that the reasoning process that led the trial judge to convict Mr. Sinclair was dependent upon the “evidence” of a planned robbery. He thus endorses the Court of Appeal’s position that “[i]t cannot be stated with confidence that, had she not mischaracterized the evidence, she would have found as she did” (para. 97). In my respectful view, not only does this position exaggerate the importance of the trial judge’s reference to this planned robbery, but it suggests, erroneously, that had the trial judge not misapprehended the evidence in this regard, she could plausibly have had a reasonable doubt about Mr. Sinclair’s presence at the crime scene.

[60] This implies that the trial judge might have been inclined to conclude that Mr. Lecours was brutally attacked by three men, including a mysterious individual whose identity remains unknown to this day, and that, for his part, Mr. Sinclair simply happened to fortuitously come across his wounded friend in the minutes following the attack. I have difficulty understanding how she could logically have been inclined to reach such a conclusion. It is even more difficult to understand how that might be if the evidence of Mr. Sinclair’s presence at the crime scene is considered in conjunction with the evidence on which the trial judge relied to infer that he had participated in the assault. In this respect, the evidence accepted by the trial judge shows

• that Emilia Rzedzian, who was found to be a credible witness, observed three young males kicking the victim (paras. 15 and 22);

• that a knife was found at the scene and Mr. Pruden-Wilson’s blood was found on its blade (para. 19);

• that Mr. Sinclair’s jeans had cuts in them, as did Mr. Pruden-Wilson’s (paras. 20-22); and

• that the police photos show lacerations on Mr. Sinclair’s legs in the approximate areas of the cuts (para. 22).

[61] Ultimately, I agree with the Crown that the conclusion that “some other unknown third male, also wearing light coloured clothing, took part in the attack on the deceased, only to then disappear just as [Mr. Sinclair] suddenly appeared in an isolated and out-of-the-way field to assist Pruden-Wilson while also somehow coincidentally acquiring cuts on his clothes and legs similar to those suffered by the other assailants, is utterly speculative and improbable” (A.F., at para. 67). Nothing in the trial judge’s reasons suggests that she might have been inclined to come to such an illogical conclusion if she had expressly rejected the Crown’s theory about the evidence of a planned robbery. The contrary is much more consistent with her actual words. She wrote that the only “rational inference” that could be drawn was that Mr. Sinclair was at the crime scene, which implies that it would be irrational to conclude that he was not there.

[62] From a factual standpoint, there were only two possible conclusions in this case: either Mr. Sinclair was present at the crime scene and, on the basis of the evidence accepted by the trial judge, he was one of the assailants, or he was somewhere else at the time of the assault. In light of both the language and the structure of her reasons for judgment, I am persuaded that the trial judge found the circumstantial evidence to be seriously incriminating for Mr. Sinclair. In my view, her reasoning process was not disturbed by the question of the planned robbery to the extent my colleague Fish J. suggests. Therefore, even assuming that she did misapprehend the evidence concerning the existence of a plan to commit a robbery, the error would not have been material within the meaning of the Lohrer test. The contention that an explicit acknowledgement of the absence of evidence in this regard could have raised a reasonable doubt in the trial judge’s mind about Mr. Sinclair’s presence at the crime scene is inconsistent with the reasoning that led her to convict him.

[63] For all these reasons, I would allow the appeal and restore the respondent’s conviction.

The following are the reasons delivered by

Charron J. —

1. Introduction

[64] Following his trial by a judge sitting without a jury, the respondent, Terrence Sinclair, was convicted of manslaughter (2007 MBQB 219, 219 Man. R. (2d) 63). The conviction was based on the trial judge’s finding that Mr. Sinclair had participated, together with his co-accused and a youth, in the assault that resulted in the victim’s death.

[65] On appeal to the Manitoba Court of Appeal, Mr. Sinclair challenged his conviction as unreasonable. The Court of Appeal held that the trial judge mischaracterized or misapprehended the evidence by inferring both that Mr. Sinclair had planned to go out with his co-accused and the youth to commit a robbery, and that they were out together at the time the fateful assault was committed. In its view, these inferences were based, at least in part, on “evidence” of those facts when there was no such evidence. The court further held that these were not peripheral matters. In its view, “the substratum of findings on which to a material extent the verdict rests is not present” (2009 MBCA 71, 240 Man. R. (2d) 135, at para. 98). Thus, purportedly on the authority of this Court’s decision in R. v. Beaudry, 2007 SCC 5, [2007] 1 S.C.R. 190, the court concluded that the verdict was unreasonable and, as there was evidence capable of supporting a conviction, ordered a new trial.

[66] The Crown appeals with leave of this Court and raises two issues. First, the Crown asks this Court to clarify the test for unreasonable verdicts under s. 686(1)(a)(i) of the Criminal Code, R.S.C. 1985, c. C-46, set out in Beaudry. In Beaudry, the Court was divided five-four in favour of adopting an expansion of the traditional test for unreasonable verdicts, but then split differently on the application of the proposed test. This division led to some uncertainty about the governing test. Second, the Crown argues that none of the subparagraphs of s. 686(1)(a) provides a basis for overturning the verdict in this case. As there was no error of law, miscarriage of justice or unreasonable verdict, the Crown submits that the Manitoba Court of Appeal erred in ordering a new trial and asks that the conviction be restored.

[67] I have had the benefit of reading the reasons of LeBel J. and Fish J. On the first issue, I agree with much of Fish J.’s reasons on the import of this Court’s decision in Beaudry regarding the test for unreasonable verdicts under s. 686(1)(a)(i). In particular, I agree that Beaudry does not change any of the existing tests for appellate review under s. 686(1)(a). As Fish J. states:

Nothing in Beaudry should be taken as a departure from the well-established principles of appellate review set out in R. v. Yebes, 1987 CanLII 17 (SCC), [1987] 2 S.C.R. 168, and R. v. Biniaris, 2000 SCC 15, [2000] 1 S.C.R. 381. [para. 4]

. . . Nor does Beaudry govern trial court decisions that are rendered inscrutable by an absence of sufficient reasons, as in R. v. Sheppard, 2002 SCC 26, [2002] 1 S.C.R. 869. [para. 3]

[68] Further, nothing in Beaudry should be taken as a departure from the test established in R. v. Lohrer, 2004 SCC 80, [2004] 3 S.C.R. 732. Thus, in this appeal, I agree with Fish J. that “the Court of Appeal, to the extent that it appears in some measure to have done so, should not have rested its judgment in this case on Beaudry” (para. 5). As he explains, “[w]hile a misapprehension of the evidence may help to establish that a verdict is unreasonable within the meaning of Yebes and Biniaris, Beaudry does not govern an error of this sort” (para. 14). If the trial judge had indeed misapprehended the evidence as contended by the Court of Appeal, Fish J. rightly states that this kind of error would fall squarely within Lohrer. However, I agree with LeBel J. that the Court of Appeal erred in finding that the trial judge made an error that met the standard set out in Lohrer. I would therefore allow the appeal and reinstate Mr. Sinclair’s conviction for the reasons of LeBel J.

[69] While none of the existing tests have been changed, the approach adopted by the majority in Beaudry, and reiterated in this appeal, did effect an expansion of the test for unreasonableness under s. 686(1)(a)(i). Under the traditional Yebes/Biniaris test, the reasonableness of the verdict is to be measured against the totality of the evidence adduced at trial (R. v. Yebes, 1987 CanLII 17 (SCC), [1987] 2 S.C.R. 168; R. v. Biniaris, 2000 SCC 15, [2000] 1 S.C.R. 381). This test continues to apply. However, where the verdict under review is that of a trial judge without a jury, the reasonableness of the verdict may also be assessed, as Fish J. puts it, “by scrutinizing the logic of the judge’s findings of fact or inferences drawn from the evidence admitted at trial” (para. 15 (emphasis in original)). Explaining how this added dimension of the reasonableness test should be applied on appellate review is somewhat more difficult. This is where I depart from Fish J.’s analysis.

[70] As I read his reasons, Fish J. explains how this expanded approach may play out by describing the kind of errors in the appreciation of the evidence or in the fact-finding process that may result in a verdict which would properly be treated as unreasonable. See in particular paras. 16 and 19-21. There may well be a logical connection between certain mistakes with respect to the substance of the evidence and the reasonableness of the verdict. However, with great respect, I find my colleague’s “reasoning-based approach”, if I may describe it that way, unhelpful. Reviewing courts would have to make rather fine distinctions between different types of mistakes in order to identify the applicable test. This distinction is especially hard to make because the fact-finding process is difficult to articulate with precision. For example, at what point is the trial judge making a finding of fact “that is . . . plainly contradicted by the evidence relied on for that purpose by the judge” (reasons of Fish J., at para. 16), rather than a “mistak[e] as to the substance of material parts of the evidence” (Lohrer, at para. 1, citing R. v. Morrissey (1995), 1995 CanLII 3498 (ON CA), 97 C.C.C. (3d) 193 (Ont. C.A.), at p. 221)? What, really, is the difference between a “mistak[e] as to the substance of . . . the evidence” (Lohrer) and a finding that is “demonstrably incompatible with evidence that is not otherwise contradicted or rejected by the trial judge” (reasons of Fish J., at para. 16)?

[71] This appeal exemplifies the difficulty. As Fish J. himself concludes, the impact on the verdict of errors in the trial judge’s assessment of the evidence and reasoning process of the kind identified by the Court of Appeal was to be assessed, not by asking whether the verdict was unreasonable, but by asking whether it resulted in a miscarriage of justice within the meaning of s. 686(1)(a)(iii).

[72] In my respectful view, the scope of the test for unreasonableness under s. 686(1)(a)(i), as expanded in Beaudry, is better understood if the focus remains on the conclusion reached at trial, rather than on the trial judge’s appreciation of the evidence or the reasoning process that led to the verdict. As I will explain, a verdict-focussed approach is more consistent with the wording of the provision, the legislative context and history, and the established jurisprudence. It also avoids any unnecessary blurring of the line between unreasonable verdicts under s. 686(1)(a)(i) and misapprehensions of evidence resulting in a miscarriage of justice under s. 686(1)(a)(iii).

2. Analysis

[73] The question that occupies us is one of statutory interpretation. A proper understanding of the scope of s. 686(1)(a)(i) requires that this power of appellate intervention be considered in context, according to its grammatical and ordinary sense, harmoniously with the scheme and object of the Act, and with the intention of Parliament. I start with the statutory context.

[74] In s. 686 of the Criminal Code, Parliament has provided a complete set of appellate powers to address both procedural and substantive errors that may take place during a trial. The provision at issue here is s. 686(1)(a)(i), which I refer to at times in brief form as the “unreasonable verdict” ground. The unreasonable verdict ground is one of the three grounds of appellate intervention on an appeal against conviction which are set out in s. 686(1)(a). The three grounds are the following:

686. (1) On the hearing of an appeal against a conviction or against a verdict that the appellant is unfit to stand trial or not criminally responsible on account of mental disorder, the court of appeal

(a) may allow the appeal where it is of the opinion that

(i) the verdict should be set aside on the ground that it is unreasonable or cannot be supported by the evidence,

(ii) the judgment of the trial court should be set aside on the ground of a wrong decision on a question of law, or

(iii) on any ground there was a miscarriage of justice;

[75] The power to intervene with the judgment of the trial court on the ground of a wrong decision on a question of law under s. 686(1)(a)(ii) is qualified by the terms of s. 686(1)(b)(iii):

(b) [the Court of Appeal] may dismiss the appeal where

(iii) notwithstanding that the court is of the opinion that on any ground mentioned in subparagraph (a)(ii) the appeal might be decided in favour of the appellant, it is of the opinion that no substantial wrong or miscarriage of justice has occurred . . .

Notably, this provision, commonly referred to as the curative proviso, does not apply to unreasonable verdicts under s. 686(1)(a)(i).

[76] Read together, it is apparent that the three grounds allowing an appellate court to overturn a conviction share a common purpose. Doherty J.A. usefully described the underlying rationale in Morrissey, as follows (at p. 219):

While s. 686(1)(a) provides three distinct bases upon which [an appellate court] may quash a conviction, each shares the same underlying rationale. A conviction which is the product of a miscarriage of justice cannot stand. Section 686(1)(a)(i) is concerned with the most obvious example of a miscarriage of justice, a conviction which no reasonable trier of fact properly instructed could have returned on the evidence adduced at trial. Section 686(1)(a)(ii) read along with s. 686(1)(b)(iii) presumes that an error in law produces a miscarriage of justice unless the Crown can demonstrate the contrary with the requisite degree of certainty. Section 686(1)(a)(iii) addresses all other miscarriages of justice not caught by the two preceding subsections. [Emphasis added.]

[77] Of course, in describing the scope of the reasonableness test, Doherty J.A. is referring to the traditional Yebes/Biniaris test. I will return to this test later. However, as the wording makes plain, the focus under s. 686(1)(a)(i) is on the conclusion reached at trial. It provides that the verdict is set aside when it is unreasonable or cannot be supported by the evidence. In Beaudry, Fish J. expressed the view that the use of the disjunctive “or” in s. 686(1)(a)(i) was indicative of a “clear Parliamentary intention to differentiate between verdicts that cannot be supported by the evidence and verdicts that may properly be characterized as unreasonable on some other ground” (para. 89 (emphasis in original)). I agree that a verdict can be unreasonable even if it can be supported by the evidence. As I noted in Beaudry, the case of inconsistent verdicts is one obvious example (R. v. Pittiman, 2006 SCC 9, [2006] 1 S.C.R. 381, at para. 14, cited in Beaudry, at para. 57). The majority in Beaudry held further that a verdict may be unreasonable because of fundamental flaws in the reasoning process that led to it.

[78] However, in my view, the disjunctive “or” should not be read as indicative of Parliament’s intention to change the focus of the reasonableness analysis in judge-alone trials from the conclusion reached at trial to the reasoning process that led to it. As the intervener the Attorney General of Ontario rightly argues, such an interpretation does not accord with the legislative history of s. 686(1)(a) of the Criminal Code. The legislative history is succinctly summarized in its factum as follows (at para. 7):

In Canada, the power to overturn a conviction on the ground it is “unreasonable” or “cannot be supported by the evidence” has existed since 1923, and is modeled on s. 4 of England’s Criminal Appeal Act. Originally, the power was expressly limited in its application to jury verdicts. The 1955 [sic] amendments to the Code altered the phrase “verdict of the jury” to “the verdict”, thereby availing this ground to verdicts rendered in judge alone trials. Nothing suggests however that this amendment was intended to alter the purpose of this provision beyond ensuring the reasonableness of the conclusion reached. The “unreasonable verdict” provision has remained unchanged since 1955. [Emphasis deleted.]

Given this legislative history, it is clear that the phrases “unreasonable” and “cannot be supported by the evidence” both apply to a bare jury verdict as much as they do to a trial judge’s verdict. As such, Parliament’s use of the word “or” in s. 686(1)(a)(i) does not mandate a focus on the reasoning process that led to the verdict on appeals from a trial judge’s verdict. Indeed, when that provision is considered in its statutory context, and according to its grammatical and ordinary sense, it indicates that the appellate court’s focus should be on the verdict.

[79] Until this Court’s decision in Beaudry, the test for the application of the unreasonable verdict ground under s. 686(1)(a)(i) was settled. It involved an assessment of the reasonableness of the verdict when measured against the totality of the evidence adduced at trial. In Yebes, a case involving a jury, the Court stated the test as follows (at p. 186):

The function of the Court of Appeal, under s. 613(1)(a)(i) of the Criminal Code, goes beyond merely finding that there is evidence to support a conviction. The Court [of Appeal] must determine on the whole of the evidence whether the verdict is one that a properly instructed jury, acting judicially, could reasonably have rendered. While the Court of Appeal must not merely substitute its view for that of the jury, in order to apply the test the Court [of Appeal] must re-examine and to some extent reweigh and consider the effect of the evidence. This process will be the same whether the case is based on circumstantial or direct evidence. [Emphasis added.]

[80] The Yebes test is expressed in terms of a verdict reached by a jury. The question that occupied this Court in Beaudry and in this appeal is whether the test for an unreasonable verdict is the same when the verdict is reached by a judge sitting without a jury. The Court was divided on this issue in Beaudry. As I have indicated at the outset, this division led to some uncertainty about the governing test. In this appeal, the Court confirms that the Yebes/Biniaris test continues to govern, but that Beaudry has effectively added a new dimension to the test for unreasonableness in judge-alone trials, albeit one of limited scope. In exceptional cases, flaws in the trial judge’s reasoning process may result in an unreasonable verdict even though the verdict is otherwise available on the evidence. As Fish J. notes, unreasonable verdicts of the sort targeted by the Beaudry test “are exceedingly rare” (para. 22).

[81] In my view, it is important to stress that, while the reviewing court may now look beyond the evidence adduced at trial and consider flaws in the reasoning process in the assessment for reasonableness, the focus must remain on the conclusion reached at trial. For s. 686(1)(a)(i) to be properly invoked, it is the verdict that must be in error. As such, it would be wrong to conclude that whenever there is an error in reasoning, even if of the kind “demonstrably incompatible” with the uncontradicted evidence at trial, the verdict is unreasonable. As stated by LeBel J. in R. v. Jaw, 2009 SCC 42, [2009] 3 S.C.R. 26, “[a]n error in proceedings that does not lead to an unreasonable verdict, a wrong decision on a question of law or a miscarriage of justice is indeed an error, but one without legal effect” (para. 29).

[82] Under each ground of appeal set out in s. 686(1)(a), any error must be assessed in context. For example, before an appellate court concludes that a misapprehension of evidence has resulted in a miscarriage of justice under s. 686(1)(a)(iii), it must consider whether the misapprehension played “an essential part not just in the narrative of the judgment but ‘in the reasoning process resulting in a conviction’” (Lohrer, at para. 2). Likewise, not every error of law will require a new hearing under s. 686(1)(a)(ii). Section 686(1)(b)(iii) provides that the conviction may be affirmed where “no substantial wrong or miscarriage of justice has occurred”. Similarly, any error in reasoning identified under the Beaudry test must be carefully assessed to determine whether it vitiates the verdict.

[83] It is useful in the assessment of reasonableness to recall that the curative proviso does not apply to an unreasonable verdict under s. 686(1)(a)(i). Thus, if the reviewing court focusses on the type of error committed in the reasoning process, rather than on the verdict itself, it may wrongfully leap to the conclusion that a simple error in logic in the trial judge’s reasons necessitates a new trial. For example, as the Crown states in its factum, “[i]t would be incongruous if a conviction could be affirmed under [the curative proviso] . . . in a case where the trial judge had incorrectly admitted a piece of evidence (e.g., on the basis that the other evidence in the case was overwhelming) but not in a case where the trial judge had correctly excluded the evidence but then mistakenly referred to it in convicting the accused” (para. 53).

[84] In my view, the s. 686(1)(a)(i) inquiry in judge-alone trials should proceed as follows:

(a) The reviewing court should first apply the test in Yebes/Biniaris. The question is whether the verdict is one that a properly instructed jury, acting judicially, could reasonably have rendered on the evidence as a whole. If not, then the test is met, the inquiry is over, and an acquittal is entered.

(b) If the verdict is available on the evidence, the reviewing court may go on to assess the reasonableness of the verdict under the Beaudry test by scrutinizing the actual findings of fact and inferences made by the trial judge. As stated earlier, an unreasonable verdict of the kind identified in Beaudry will be exceedingly rare. It is therefore not necessary to conduct this assessment in every case. Unless the issue is specifically raised, or the circumstances clearly call for this further inquiry, the impact of errors in the reasoning process that led to the verdict falls to be assessed under either s. 686(1)(a)(ii) (errors of law) or s. 686(1)(a)(iii) (miscarriage of justice).

(c) The Beaudry test may apply in exceptional cases where the reasoning process of the trial judge is so irrational, or so at odds with the evidence, that it vitiates the verdict — even though that verdict is available on the evidence as a whole. In these rare cases, an appellate court will be justified in concluding that the verdict itself is unreasonable. What precise kinds of errors would justify this conclusion is a question better left to be decided in cases where the issue will arise. If the reviewing court so concludes, a new trial is ordered, as the court will have already determined under the Yebes/Biniaris test that the verdict is otherwise available on the evidence.

[85] In the present case, the Court of Appeal rightly concluded that the evidence adduced at trial was capable of supporting a conviction. Thus, Mr. Sinclair cannot meet the Yebes/Biniaris test. Nor can he properly invoke the Beaudry test as there is no fundamental flaw in the reasoning process that led to the trial judge’s verdict. Finally, for the reasons of LeBel J., I conclude that the trial judge did not misapprehend the evidence in finding that Mr. Sinclair had participated in the assault that resulted in the victim’s death.

3. Disposition

[86] For these reasons, I would allow the appeal, set aside the order for a new trial and restore the conviction.

The following are the reasons delivered by

[87] Abella J. — I agree with Charron J.’s analysis on the application of the Beaudry test (R. v. Beaudry, 2007 SCC 5, [2007] 1 S.C.R. 190). Like her, I also agree with LeBel J. that the trial judge did not misapprehend the evidence. Therefore, I would allow the appeal, set aside the order for a new trial, and restore the conviction.

Appeal allowed, McLachlin C.J. and Binnie, Fish and Cromwell JJ. dissenting.

Solicitor for the appellant: Manitoba Prosecution Service, Winnipeg.

Solicitors for the respondent: Gindin, Wolson, Simmonds, Winnipeg.

Solicitor for the intervener the Director of Public Prosecutions of Canada: Public Prosecution Service of Canada, Gatineau.

Solicitor for the intervener the Attorney General of Ontario: Attorney General of Ontario, Toronto.