Simon's Megalomaniacal Legal Resources

Evidence - Identification (3)

. R. v. G.D.

In R. v. G.D. (Ont CA, 2026) the Ontario Court of Appeal dismissed a criminal appeal, this brought against a conviction for 'sexual abuse'.

Here the court considers the adequacy of a jury charge respecting eyewitness identification:

[33] The appellant argues that the trial judge erred by separating a general caution with respect to eyewitness identification evidence and the witness’ specific evidence, including its frailties. To be clear, the appellant does not argue that the trial judge failed to provide an adequate caution or that he failed to raise frailties with the witness’ evidence for the jury to consider (though the appellant points out the coverage of the frailties was not comprehensive, and did leave out, for example, a problematic in-dock identification of the appellant by the witness as well). Rather, the basis for the error alleged by the appellant is the failure to situate the frailties in the context of the caution.

[34] The appellant relies on R. v. Baltovich (2004), 2004 CanLII 45031 (ON CA), 73 O.R. (3d) 481 (C.A.), at para. 79, where this court, quoting from R. v. Keane (1977), 65 Cr. App. R. 247 (Eng. C.A.) at p. 248, noted:

The principle is the special need for caution when the issue turns on the evidence of visual identification: the practice has to be a careful summing-up, which not only contains a warning but also exposes to the jury the weaknesses and dangers of identification evidence both in general and in the circumstances of the particular case. [Emphasis added in Baltovich.]

[35] The need for the link between the frailties of eyewitness identification evidence generally and the specific frailties of the case before the jury is now well settled: see e.g., R. v. Brown, 2007 ONCA 71, 216 C.C.C. (3d) 299, at paras.17-18.

[36] The Crown does not dispute this principle but argues that a functional approach to the jury charge as a whole makes clear the jury could not be confused by the frailties of the witness’ identification evidence and the risks of eyewitness identification evidence generally. Both were reviewed by the trial judge in his charge, albeit in different sections.

[37] There is no doubt that an ideal charge would include the risks of eyewitness identification evidence together with the frailties of the specific eyewitness evidence before the jury together in the same section of the charge. However, we reject the submission that this proximity is a requirement of a legally valid charge.

[38] As the Supreme Court of Canada held in R. v. Abdullahi, 2023 SCC 19, 428 C.C.C. (3d) 1, at para. 53, the sufficiency of an instruction “must be assessed in the context of the charge as a whole”. There, the Court specifically recognized that an instruction may be insufficiently detailed in one part of the charge but can be supplemented by another part of the charge such that the jury would be equipped with a sufficient understanding of the law to decide the case. In our view, that is what occurred in this case.

[39] The trial judge gave accurate and sufficient instructions on both the general dangers of eyewitness identification, and the specific frailties of the eyewitness identification in this case.

[40] At the outset of the general eyewitness instruction, the trial judge asked the jury to recall the witness’ testimony that she recognized Bobby as the complainant’s uncle when he dropped off the pager at her house. This would have indicated to the jury that her identification evidence should be kept in mind when considering the dangers of eyewitness identification. Later in the charge, after providing detailed summary of the witness’ evidence (which we review next), the trial judge again stated that the witness’ evidence about the delivery of the pager was “not reliable evidence of Bobby as [the complainant’s] uncle because of the inherent dangers of identification evidence.” The jury was therefore alerted to the connection between the dangers of eyewitness identification and this witness’ identification of Bobby both before and after the specific summary of the frailties in her evidence.

[41] The trial judge dealt specifically with the length of the observation, noting that the witness initially testified that “she got a glimpse of him, then she said she got more than a glimpse of him, and then she said she got a solid look at him.” He dealt with the fact that the witness told police she had “buried these memories and friendships many years ago”. The trial judge dealt with the possibility of tainting of this evidence by the complainant or the police as well as the possibility of the complainant’s friend having a motive to assist the complainant. Finally, the trial judge highlighted the disputed evidence about the colour of the car, noting that the witness had testified that Bobby’s car was dark coloured, in contrast with the evidence of other witnesses.

[42] We note that experienced trial counsel reviewed the eyewitness identification sections of the charge in detail and was active in suggesting language for the charge relating to the witness’ evidence. However, counsel raised no concern with the gap between the general eyewitness identification evidence section of the charge and the concerns with the witness’ identification evidence.

[43] For these reasons, we reject this ground of appeal.

. R. v. Fitzpatrick

In R. v. Fitzpatrick (Ont CA, 2026) the Ontario Court of Appeal allowed two criminal appeals, one against a first-degree murder conviction and the other against a manslaughter conviction.

Here the court considers 'resemblance evidence':

1. The trial judge erred by admitting prejudicial “resemblance evidence”

a. The law

[49] Where there is video evidence relating to the commission of an offence and the identity of the perpetrator is at issue, the Crown may seek to prove that the accused is the perpetrator by adducing evidence from a witness who is sufficiently familiar with the accused to recognize them as the person in the video: Leaney, at p. 413; R. v. Hudson, 2020 ONCA 507, 391 C.C.C. (3d) 208, at para. 28. This recognition evidence is a form of identification evidence, which is itself a form of admissible lay opinion evidence: R. v. Olliffe, 2015 ONCA 242, 322 C.C.C. (3d) 501, at para. 39; R. v. Brown (2006), 2006 CanLII 42683 (ON CA), 219 O.A.C. 26 (C.A.), at para. 39.

[50] In R. v. Berhe, 2012 ONCA 716, 113 O.R. (3d) 137, Blair J.A. explained the test for threshold admissibility of recognition evidence. He referred, at para. 14, to this test as the “prior acquaintance/better position” test. The trial judge must determine whether: 1) the witness is sufficiently familiar with the accused to have “some basis” for their identification opinion; and 2) the witness, as a result of their prior acquaintance with the accused, is in a better position than the trier of fact to identify the perpetrator, in the sense that they have “some advantage that can shed light on the evidence in question”: Hudson, at para. 30, citing Berhe, at paras. 14, 21.

....

[64] I cannot accept the Crown’s submission that the stumble evidence constituted resemblance evidence that was admissible because there was other inculpatory evidence linking Fitzpatrick to Person #2: see e.g., R. v. Rybak, 2008 ONCA 354, 90 O.R. (3d) 81, at para. 121. Resemblance evidence is typically used to describe something specific about a suspect that resembles the accused. That resemblance may come from a suspect’s physical features, a distinctive gait, or clothing: see e.g., R. v. Boucher (2000), 2000 CanLII 3270 (ON CA), 146 C.C.C. (3d) 52 (Ont. C.A.). In other words, its probative value comes from its specificity. A non-distinctive stumble or trip over a curb cannot reasonably amount to evidence of a suspect resembling the accused. Indeed, this evidence had no bearing on the issue of Person #2’s identity and was accordingly inadmissible: Cloutier v. The Queen, 1979 CanLII 25 (SCC), [1979] 2 S.C.R. 709, at pp. 730-31.

. R. v. MacLaughlin

In R. v. MacLaughlin (Ont CA, 2025) the Ontario Court of Appeal dismissed a defendant's criminal appeal, here brought against convictions for "child luring, invitation to sexual touching, making child pornography, criminal harassment, and uttering a death threat".

The court considers identification evidence, here in the context of self-instructing by the trial judge:

[20] Turning to the second argument, the appellant correctly notes that “[v]oice identification evidence, like any identification evidence, ought to be treated with extreme caution”: R. v. Clouthier, 2012 ONCA 636, at para. 19; R. v. Dodd, 2015 ONCA 286, 322 C.C.C. (3d) 429, at para. 79; R. v. A.W., 2024 ONCA 564, 439 C.C.C. (3d) 530, at para. 46. However, the need for a specific caution about the frailties of this evidence requires “a case-by-case determination having regard to the balance of the evidence in the case”: A.W., at para. 54.

. R. v. Sturgeon

In R. v. Sturgeon (Ont CA, 2025) the Ontario Court of Appeal dismissed a criminal defendant's appeal, finding that while the trial judge "misapprehended the corroborative evidence", they did so with immaterial effect.

Here the court considers the evidence law of 'eyewitness identification':

[35] The dangers of erroneous convictions based on honest and convincing, but mistaken, eyewitness identification are notorious: see R. v. Layne, 2024 ONCA 435, 439 C.C.C. (3d) 112, at paras. 23-24. Trial judges are required to guard against these dangers by closely scrutinizing the reliability of the witness’s description of the perpetrator and that description’s similarities and dissimilarities with the accused. They must assess the identification evidence against other potentially exculpatory evidence: R. v. Olliffe, 2015 ONCA 242, 322 C.C.C. (3d) 501, at paras. 37, 42-44. Trial judges should also evaluate the evidence in light of the four factors derived from this court’s decision in R. v. Tat (1997), 1997 CanLII 2234 (ON CA), 35 O.R. (3d) 641 (C.A.), at p. 673, namely:

(i) whether the person identified was a stranger or known to the witness since, if the witness recognizes the perpetrator as someone he or she knows, that makes the identification more reliable than trying to remember a stranger;

(ii) whether the circumstances of the identification were conducive to an accurate identification;

(iii) whether the pre-trial identification process was flawed; and

(iv) whether there was independent evidence confirming the identification. This fourth factor, independent confirmatory evidence, “can go a long way to minimizing the dangers inherent in eyewitness identification”: R. v. Quercia (1990), 1990 CanLII 2595 (ON CA), 75 O.R. (2d) 463 (C.A.), at p. 471.

....

[41] Further, this is not a case where there is contrary evidence casting doubt on Mr. Darnell’s identification of the appellant, such as where the witness described physical characteristics that were inconsistent with the perpetrator’s (Olliffe, at paras. 42, 45; Chartier v. Quebec (Attorney General), 1979 CanLII 17 (SCC), [1979] 2 S.C.R. 474, at pp. 489-490), or where there were other witnesses whose evidence contradicted that of the identification witness (Chartier, at p. 490).

. R. v. Smith ['cross-racial identification']

In R. v. Smith (Ont CA, 2025) the Ontario Court of Appeal considered 'cross-racial identification':

[15] We agree that it would have been preferable for the trial judge to self-instruct on this risk and consider it. In oral argument, he suggested that the defence needed to adduce social science evidence before he could do so. This was incorrect because cross-racial misidentification is a well-recognized danger which is the subject of judicial notice and does not require expert evidence: see, e.g., R. v. Bailey, 2016 ONCA 516, 339 C.C.C. (3d) 463, at para. 49; R. v. B.M. (1998), 1998 CanLII 13326 (ON CA), 42 O.R. (3d) 1 (C.A.), at pp. 29-30, 35; R. v. McIntosh (1997), 1997 CanLII 3862 (ON CA), 35 O.R. (3d) 97 (C.A.), at p. 105, leave to appeal refused, [1997] S.C.C.A. No. 610. Failure to grapple with it can be a reversible error if it poses a serious danger in the circumstances of the case: R. v. Bao, 2019 ONCA 458, 146 O.R. (3d) 225, at para. 23.

[16] But this risk was not a serious danger in the circumstances of this case so the trial judge’s failure to consider it was not fatal. This was not a case where the eyewitnesses were purporting to positively identify the appellant as the assailant, rather than simply providing descriptions of what the assailant had looked like. The defence did not articulate any case-specific factors which increased the risk and, as the Crown submits, it did not impact the accuracy of the witnesses’ descriptions of the assailant’s skin colour, height, and clothing on which the trial judge relied. Further, as we explained, the trial judge carefully examined all the other relevant factors, as in R. v. Lewis, 2022 ONCA 282, at paras. 26-32.

. R. v. Vu

In R. v. Vu (Ont CA, 2025) the Ontario Court of Appeal allowed a criminal Crown appeal, here from "acquittals on two counts of sexual assault and one count of forcible confinement."

The court considers 'similar fact' evidence, here in an identification context:

[36] Similar act evidence is presumptively inadmissible. The Crown holds the onus of establishing, on a balance of probabilities, that the evidence should be admitted. That onus is met where the probative value of the evidence outweighs its prejudicial effect: R. v. Handy, 2002 SCC 56, [2002] 2 S.C.R. 908, at para. 55. Although the trial judge did not refer to any case law, he correctly summarized these legal principles.

[37] The probative value of the evidence comes from the objective improbability of coincidence: Handy, at paras. 47-48. The prejudice comes from both moral and reasoning prejudice that may result from the admission of the evidence. Moral prejudice concerns itself with whether the trier of fact will decide the case based upon the perceived bad character of the accused. Reasoning prejudice comes from the injection of delay and complexity into the trial: R. v. Shearing, 2002 SCC 58, [2002] 3 S.C.R. 33, at paras. 65, 68; Handy, at para. 31. Some of these concerns are attenuated in judge-alone trials such as this one: R. v. J.M., 2010 ONCA 117, 251 C.C.C. (3d) 325, at para. 88; R. v. T.B., 2009 ONCA 177, 95 O.R. (3d) 21, at paras. 27, 33; and R. v. MacCormack, 2009 ONCA 72, 95 O.R. (3d) 21, at paras. 56, 68-69.

[38] As the trial judge noted, where the proposed evidence is directed at the issue of identity, the law insists upon a “high degree of similarity” or “striking” similarity”: R. v. Arp, 1998 CanLII 769 (SCC), [1998] 3 S.C.R. 339, at para. 45; R. v. Perrier, 2004 SCC 56, [2004] 3 S.C.R. 228, at para. 20; R. v. Durant, 2019 ONCA 74, 144 O.R. (3d) 465, at para. 98; and R. v. Atwima, 2022 ONCA 268, 161 O.R. (3d) 481, at para. 39. The reason for a heightened bar for similarity relates to the driver of cogency when it comes to the similar act evidence being used to establish identity: “the improbability that two persons would display the same configuration of matching characteristics in committing a crime”: Perrier, at para. 19; see also, Handy, at para. 78; and Atwima, at para. 39.

[39] There exist occasions where acts are so strikingly similar that they will constitute the offender’s signature or trademark: Arp, at para. 45; see e.g., R. v. Jesse, 2012 SCC 21, [2012] 1 S.C.R. 716, at paras. 5, 10, 12, and 24. In those situations where signatures or trademarks do not exist, striking similarities can still arise from an “accumulation of commonalities” that, when considered together, are sufficient to reveal an improbability of coincidence that two or more people would display these matching characteristics: Arp, at para. 45; Handy, at para. 81.

[40] In assessing whether the proffered evidence has this cumulative effect, Binnie J. in Handy, at para. 82, provided a list of helpful considerations:

(a) proximity in time between the acts;

(b) the extent to which they are similar in detail;

(c) the number of acts said to be similar;

(d) the circumstances surrounding or relating to the acts;

(e) the distinctive feature(s) that may be involved in the acts;

(f) whether there exist any intervening events; and

(g) “any other factor which would tend to support or rebut the underlying unity of the similar acts.”

[41] Accordingly, similar act evidence is not considered on a piecemeal basis. Rather, the goal is to consider whether the similarities, viewed collectively, are sufficient to establish on a balance of probabilities that they are the product of the same actor. If so, the trial judge must go on to the second stage and determine whether the evidence is linked in some way to the accused: R. v. Woodcock, (2003), 2003 CanLII 6311 (ON CA), 14 C.R. (6th) 155 (Ont. C.A.), at para. 81; Arp, at paras. 54-56. There need only be “some evidence” linking the accused to the acts: R. v. Sweitzer, 1982 CanLII 23 (SCC), [1982] 1 S.C.R. 949, at p. 954; Perrier, at para. 24; Durant, at para. 91; and Atwima, at para. 42. To be sure, this is a “low evidentiary threshold”: Jesse, at para. 63.