Decision Content
In the Court of Appeal for the Northwest Territories Citation: R. v. Larsen, 2012 NWTCA 9 Date: 2012 06 04 Docket: A-1-AP-2011-000014 Registry: Yellowknife, N.W.T. Between: Her Majesty the Queen Respondent - and - Gordon Larsen Appellant Restriction on Publication: By Court Order, information that may identify the persons described in this judgment as the complainant may not be published, broadcast, or transmitted in any manner. There is also a ban on publishing the contents of the application for the publication ban or the evidence, information or submissions at the hearing of the application. See the Criminal Code, s. 486.4 _______________________________________________________ The Court: The Honourable Mr. Justice Clifton O’Brien The Honourable Mr. Justice Frans Slatter The Honourable Madam Justice Myra Bielby ______________________________________________________ Memorandum of Judgment of the Honourable Mr. Justice Clifton O’Brien and the Honourable Madam Justice Myra Bielby Memorandum of Judgment of the Honourable Mr. Justice Frans Slatter Concurring in the Result Appeal from the Conviction by The Honourable Mr. Justice J.Z. Vertes Dated the 23rd day of March, 2011 (Docket: 2010115753) _______________________________________________________ Memorandum of Judgment of the Honourable Mr. Justice Clifton O’Brien and the Honourable Madam Justice Myra Bielby _______________________________________________________ OVERVIEW [1] Gordon Larsen appeals his convictions on three counts of sexual interference, contrary to s. 151 of the Criminal Code, RSC 1985, c C 46. The three complainants’ allegations were historic in nature, relating to activities they said took place between 1989 and 1994, when they were young children. The allegations were brought to the attention of the police for the first time in 2010. [2] Two of the complainants are sisters, the other their childhood friend. Mr. Larsen testified at his trial, agreeing that he was a friend of the parents of the complainants. He admitted to living with the sisters and their family for a short time and otherwise visiting their home, staying overnight on two occasions. He also admitted to living with the friend’s family during the general time period in question. [3] The sisters testified that they shared a bedroom. Each testified that Mr. Larsen would come into the bedroom at night, when everyone else in the house was sleeping, take down her underwear and touch her vagina with his hand. The younger sister also testified that, on one occasion, Mr. Larsen used his tongue to touch her vagina and, on another occasion, he touched her vagina over her clothing in his truck. [4] The older sister recalled these events occurring when she was between 8 and 13 years of age. The younger sister testified she was between 4 and 9 years of age when they occurred. The friend testified to a single event, when she was about 9 years of age, which occurred while Mr. Larsen was living with her family; he touched her vagina with his hand while she lay beside him on a couch after having a nightmare. [5] Both sisters testified to telling their parents what had occurred but said they were not believed. Although the older sister revealed these events in a school paper she wrote in Grade 11, with the result that her teacher contacted her mother, the latter did not contact the police until 2010. The sisters testified that they did not discuss the allegations with one another when they were younger and had never discussed them in detail. They also testified they had never discussed the allegations with the friend. The friend testified that she had never discussed the allegations with the sisters and said she spoke to the police after the sisters’ mother contacted her mother. [6] Mr. Larsen testified that none of the incidents occurred. He agreed that he babysat the sisters when he lived with their family but denied entering their bedroom and being alone in his truck with either girl. He said his contact with the sisters’ family ended around 1992. He recalled an event where the friend, aroused at night due to a nightmare, lay down beside him on the couch where he slept. However, he denied assaulting her. [7] Credibility was the main issue at trial. The trial judge made two decisions, both oral, the first to admit similar fact evidence during the course of the trial and the second at the conclusion of the trial proper, in which he convicted Mr. Larsen of all three counts. His credibility conclusions during his conviction decision and circularity of reasoning in his assessment of the complainants’ credibility in the similar fact evidence decision are at issue in this appeal. [8] The appeal is allowed and a new trial ordered. STANDARD OF REVIEW [9] An appeal court must defer to the credibility findings of the trial judge unless a palpable or overriding error, in his appreciation of the law or the evidence, can be shown: R v Gagnon, 2006 SCC 17, [2006] 1 SCR 621 at paras 10 and 24. [10] The admissibility of evidence is usually an extricable question of law reviewable on the correctness standard, although underlying fact findings are reviewed for palpable and overriding error: R v Blea, 2012 ABCA 41, [2012] AJ No 106 (QL) at para 31. As the task of balancing probative value against prejudicial effect involves an exercise of judicial discretion, in the absence of a clear error of law, a trial judge’s decision to admit similar fact evidence is entitled to substantial deference: Blea at para 33; R v Buna, 2009 BCCA 536, 249 CCC (3d) 156 at para 53. ISSUES [11] This appeal raises two main issues: 1. Are the convictions unreasonable because of the treatment the trial judge gave to Mr. Larsen’s evidence? 2. Did the trial judge err in admitting the similar fact evidence because, when deciding that the three complainants’ testimony was sufficiently credible to justify its admission for that purpose, he considered only that very evidence in making the assessment of credibility? ANALYSIS Are the convictions unreasonable ? [12] Several aspects of the trial judge’s decisions raise the concern that his conviction of Mr. Larsen was unsafe. The first of these arise from his principle reason for rejecting his evidence, which was the unevenness or variation in Mr. Larsen’s memory of events. In the conviction decision, although he found that Mr. Larsen was not shaken on cross-examination and there were no inconsistencies or serious contradictions in his evidence, the trial judge nonetheless went on to conclude: Having considered the totality of the evidence, I reject the accused’s evidence. I find it incredible that he would recollect the details of the incident with [the friend] as he did in his testimony. This occurred some 18 or 19 years ago yet he had a clear recollection, a recollection which became more detailed on cross-examination. Yet he has no similar details for any of his interactions with the other complainants. He acknowledged that he sometimes stayed overnight at the [sisters’] home after drinking with the father and [the younger sister] testified that she knew it was the accused who came into her bedroom because she could see him, hear him, and “smell him” - specifically that she recalled smelling alcohol. Yet the accused would not acknowledge that his memory could be affected by alcohol. [13] This is not a rejection of evidence due to the witness’ inability to provide a detailed recollection of the circumstances surrounding the alleged offences. If it were, that would be an error because an innocent person is unlikely to have a detailed memory of distant uneventful occasions: R v Norman (1993), 16 OR (3d) 295, [1993] OJ no 2802 (QL) at paras 43-44 (CA); R v Tucker (1992), 120 AR 393 (CA). The trial judge did not reject Mr. Larsen’s evidence simply because he was unable to reconstruct a scene that was said to have occurred many years ago, or because he failed to recall particular details of his interactions with the sisters. Instead, he rejected Mr. Larsen’s evidence because of a sharp variation in the level of his memory. [14] This is akin to the situation discussed in the recent decision of the Supreme Court of Canada, R. v. R.P. , para. 18, where the majority restored the decision of a trial judge who had rejected the evidence of the accused’s wife, finding it “ unlikely that the respondent’s wife could remember details of daily life where there was no reason for them to have been “fixed” in her memory. While he was unable to recall any circumstance around the events to which the sisters testified, he had a specific memory in relation to an incident involving the friend. Mr. Larsen testified to remembering two occasions when the friend awoke from a nightmare while he was living with her family. On one of those occasions he told her to lay beside him on the couch, in the livingroom where he was sleeping, and calm down. When her father came out and saw her laying there, Mr. Larsen said she had a nightmare and told her to go back to her room. He denies any inappropriate touching. The father was not called to testify. [15] Unlike in R.P., the trial judge in this case did not make a finding that it was either likely or unlikely Mr. Larsen could recall the incident where the friend lay on the couch with him after having a nightmare so many years earlier if nothing untoward happened on that occasion. Had he found it was likely that he would have a particularly detailed memory of the event because it was sufficiently unusual, that would be a reason for his variation in memory between this event and the events involving the sisters. [16] In summary, whether this incident was sufficiently unusual to account for Mr. Larsen’s specific memories was a question for the trier of fact to determine, yet the trial judge did not address the issue. This error is of particular concern because of his findings that Mr. Larsen was not shaken on cross-examination nor inconsistent in his evidence. [17] A verdict must be overturned if it is unreasonable. One of the means in which it becomes unreasonable is if the trial judge has drawn an inference or made a finding of fact essential to the verdict that is shown to be incompatible with evidence that has not otherwise been contradicted by or rejected by the trial judge; see R. P. para. 9. By rejecting the Appellant’s evidence because he had a much more detailed recollection of the incident that other incidents, without first having found that there was no particular reason for that event to have been “fixed” in his mind, the trial judge ran afoul of this prohibition. [18] Another specific reason given by him for rejecting Mr. Larsen’s credibility - that he would not acknowledge that his memory could be affected by alcohol - is not strictly supported by the evidence. What Mr. Larsen stated, in cross-examination, is that when he drank alcohol to the degree that his ability to drive was impaired, his memory would not be affected. That evidence is considerably less conclusive than a denial that his memory could ever be affected by alcohol. [19] Further, the trial judge found that the evidence of each of the complainants was not shaken on cross-examination and there were no inconsistencies or serious contradictions in their evidence. Certain inconsistencies did exist, however. The elder sister recalled being present in her bedroom at night on one occasion when Mr. Larsen assaulted her younger sister who did not recall that incident. The elder sister testified at trial that none of the incidents occurred when her family lived in Saskatoon, but then admitted this was inconsistent with an earlier statement in which she stated she told her mother that an incident or incidents occurred in that city. She testified that she was never alone with the accused during the day, yet four transcript pages earlier she testified to a single incident of oral sex which “might have been during the day”. Two pages earlier she testified to being alone with him in his truck during the day when he touched her over her clothing. The friend, as well, testified to circumstances surrounding her assault in a sleeping bag which were logistically unlikely. [20] While none of these inconsistencies or unlikely details would, of necessity, lead to the conclusion that the complainants’ evidence was not reliable, the trial judge appeared to have overlooked them in making his credibility assessment by concluding that no such inconsistencies, nor successful challenges on cross-examination occurred. [21] We are mindful that the power to overturn a verdict based on credibility findings should be used sparingly, given the advantage a trial judge has over an appellate court in assessing witnesses’ credibility: R v Burke, [1996] 1 SCR 474 at paras 5-6; Buna at para 69. This was not, however, the situation as that in R.P. where the trial judge gave detailed reasons for judgement; rather, his reasons were so sparse that they must be examined with care to find support, or possible support, for the conclusions he reached. [22] We conclude, however, that because the trial judge made a palpable and overriding error through misapprehension of the evidence, which makes these convictions unreasonable , particularly in the context of offences prosecuted almost two decades after they were said to have occurred, the appeal must be allowed and a new trial ordered. Did the trial judge err in admitting the similar fact evidence because, when deciding that the three complainants’ testimony was sufficiently credible to justify its admission for that purpose, he considered only that very evidence in making the assessment of credibility? [23] While it is not necessary to address this issue, given the above conclusion, we do so as it was raised by the Bench in correspondence with counsel in advance of oral argument, then addressed by them during that argument, and is to be avoided on the retrial. It arises from the reasons the trial judge gave for admitting the similar fact evidence. [24] Two stages arise where an assessment of credibility must be made in a trial where similar fact evidence is admitted. The first stage arises when the application to admit is made and the Crown must establish, on a balance of probabilities, that the proposed evidence is sufficiently probative to be considered: R v Handy, 2002 SCC 56, [2002] 2 SCR 908 at para 101; R v MacCormack, 2009 ONCA 72, 241 CCC (3d) 516 at para 54. [25] The credibility or reliability of the proposed similar fact evidence is one aspect of its probative value. At para 104 of Handy, Justice Binnie stated: The trial judge’s gatekeeper role in this respect was addressed in B.(C.R.) by McLachlin J., at pp. 733 34: The difficulty of the trial judge’s task and the amount of discretion entrusted to him or her is great. As Forbes, [Similar Facts (1987)], puts it at pp. 54 55: A judge presented with similar facts for the prosecution has to exercise an extraordinary complex of duties and powers. First he has to assess not only the relevance but also the weight of the disputed evidence, although the latter task is normally one for the jury. Second, he must somehow amalgamate relevance and weight to arrive at “probative value”. [Emphasis in the original] And, at paras 134 and 136, he said: In the usual course, frailties in the evidence would be left to the trier of fact, in this case the jury. However, where admissibility is bound up with, and dependent upon, probative value, the credibility of the similar fact evidence is a factor that the trial judge, exercising his or her gatekeeper function is, in my view, entitled to take into consideration. Where the ultimate assessment of credibility was for the jury and not the judge to make, this evidence was potentially too prejudicial to be admitted unless the judge was of the view that it met the threshold of being reasonably capable of belief. ... If the proffered similar fact evidence is not properly capable of supporting the inferences sought by the Crown, the analysis generally need go no further. ... [Emphasis in the original] [26] While there was no jury in the case under appeal, the Crown remained obliged to establish that the proposed similar fact evidence was capable of belief, on a balance of probabilities, when it applied for its admission. This is to be compared to the Crown’s obligation during the trial itself, during the second stage of credibility assessment, to establish the guilt of Mr. Larsen beyond a reasonable doubt. [27] Here, the trial judge expressly found that the evidence of the three complainants was reasonably capable of belief in his initial reasons which are strikingly sparse. They were issued mid-trial, at the conclusion of a de facto voir dire (although that label was not used by counsel or judge), after the complainants’ evidence had been given and argument heard on the issue, but before the defence led evidence. [28] The trial judge did conclude those reasons by stating that: “[T]hat’s very much a summary of my reasons for the decision. I will likely amplify those reasons in due course.” However, the only place such amplification is found is in his subsequent oral conviction decision, where he appears to give a further six transcript paragraphs of reasons for having admitted the similar fact evidence. [29] In neither decision did he provide reasons for his conclusion that the evidence of the three complainants was reasonably capable of belief unless they can be found in the paragraph immediately preceding this conclusion in the initial decision in which he referred to the “significant degree of similarity between the alleged offences”. It is not clear if he intended this to be a reason for his credibility finding, or rather to relate to a separate prerequisite for admissibility, that there was sufficient “connectedness” among the proposed similar fact evidence. [30] If he intended these remarks to bear on his credibility finding, it would appear that the only factor he may have taken into account in assessing the complainants’ credibility during the application to admit similar fact evidence, was the degree of similarity in their evidence. [31] While the trial judge was entitled to consider similarity as an aid to assessing credibility even at the initial stage, he engaged in circular reasoning here, concluding that the similar fact evidence was admissible only because of the similar fact evidence. [32] Of course, a trial judge is not required to give particular reasons for making findings of credibility in a decision to admit similar fact evidence if his reasons are otherwise intelligible in the context of the evidence and the trial proceedings: Buna at para 54. Had this trial judge clearly given no reasons for his credibility findings at all in similar fact evidence decision, it may be that this Court would have concluded that such reasons existed, as he gave independent reasons for his credibility conclusion in his subsequent conviction decision. However, his reference to the similarity of the evidence as the only potential reason for his initial credibility finding gives rise to the possibility that it was the only reason at that stage. [33] It may also be that the trial judge intended to have his reasons for later finding the complainants to have been credible, set out elsewhere in his conviction decision, to also have acted as reasons for his initial decision to admit the similar fact evidence. However, he does not say so in either decision, and such mixing of reasons between two stages involving differing standards of proof, the former being given before Mr. Larsen testified and the latter being given after he testified, makes such a conclusion dangerous. [34] This is not to suggest that similar fact evidence, once properly admitted, cannot be considered in making credibility findings in the trial proper: R v B(CR), [1990] 1 SCR 717 at 738-739. If it could not, there would be no reason for seeking its admission. However, that addresses its use at the second stage of the credibility assessment, whereas we are concerned here with the first stage. [35] This issue likely arose because of the trial judge’s delay in giving full reasons for his similar fact evidence ruling until after he had heard the balance of the evidence in the trial, and because he failed to expressly delineate which portions of his conviction decision were in fact further reasons for that initial decision. Other issues [36] In light of these conclusions, it is unnecessary to address the balance of the issues raised by counsel. We observe, however, that Mr. Larsen utterly failed to convince us that he was deprived of a fair trial due to the acts or omissions of his trial counsel, notwithstanding the lack of a Corbett application being sought prior to Mr. Larsen’s extensive cross-examination on his dated criminal record, which included some morality-based offences. He also now complains of his counsel’s failure to cross-examine the complainants on inconsistencies between earlier statements and trial evidence and about the possibility of unintentional contamination or collusion. However, in the absence of evidence from trial counsel to the contrary, trial strategy could well explain choices made at that time. CONCLUSION [37] The appeal is allowed and the matter remitted to the Supreme Court of the Northwest Territories to conduct a new trial on all counts. Appeal heard on April 16, 2012 Memorandum filed at Yellowknife, N.W.T. this day of June, 2012 Authorized to sign for: O’Brien J.A. Bielby J.A. Slatter J.A. (concurring in the result): [38] The issues on this appeal are whether the trial judge’s analysis of the evidence discloses palpable and overriding error, and whether similar fact evidence was improperly considered. Facts [39] The appellant was charged with sexually touching three young girls. The acts were alleged to have occurred between 1989 and 1994, but were first brought to the attention of the police in 2010. The trial occurred approximately 20 years after the events in question. [40] Two of the complainants were sisters, and made similar allegations. They both testified that the appellant would enter their bedroom at night when they were asleep, and sexually assault them. The appellant denied any such acts, and testified that he was never in their bedroom at all. [41] The third complainant was a friend of the sisters. In about 1992 the appellant was staying at her family home, and would sleep in a sleeping bag on a couch outside her parents’ bedroom. One night she awoke as a result of a nightmare, and was banging on her parents’ bedroom door. The appellant awoke, and told her to stop. She crawled into the sleeping bag, at which time she alleges that the appellant assaulted her. The appellant testified that he recalled two incidents where this complainant awoke as a result of nightmares, and that on one occasion she came to lie down next to him. He denied, however, that there was any inappropriate touching. [42] The trial judge admitted the evidence of each complainant as being probative similar fact evidence with respect to the charges involving the other two complainants. He was not left with any reasonable doubt as a result of the denials of the appellant, and convicted him of three counts of sexual touching. Standard of Review [43] This appeal is an example of a not uncommon and difficult scenario for the courts. It is a “historic sexual assault”; the events in question took place many years ago. The complainant was often a child at the time. The passage of time has rendered illusory the prospect of corroborative evidence, physical evidence or the proof of an alibi. The memories of the witnesses have faded. The accused denies under oath events that the complainant testifies under oath occurred. Cases like this present many factual, credibility and evidentiary challenges. [44] A number of legal principles are engaged by an appeal of this category of case. Three in particular are at the forefront of this appeal. [45] The first is the principle of justice in the result. It goes without saying that the courts must do all that they can to avoid wrongful convictions. Whatever its full role may be, this Court is still in many respects an error correcting court. While error correction must occur within the parameters of the appropriate standard of review, the correction of errors is still an important function of appellate courts. It is up to this Court to intervene in the face of incorrect, unsafe and dangerous convictions. [46] The second is the principle of deference. Appellate courts are to show respect for the role and conclusions of trial courts. The rationale for this is summarized in Housen v Nikolaisen, 2002 SCC 33 at paras. 15-8, [2002] 2 SCR 235: (a) extending deference to trial courts conserves resources by avoiding needless duplication of judicial proceedings with little, if any improvement in the result; (b) deference supports the presumption underlying the structure of our court system that a trial judge is competent to decide the case before him or her, and that a just and fair outcome will result from the trial process; and (c) the trial judge is better situated to make factual findings owing to his or her extensive exposure to the evidence, the advantage of hearing testimony viva voce, and the judge’s familiarity with the case as a whole. [47] Deference is an important principle, but it does not equate to immunity from review. As was said in a civil context, “. . . appellate courts not only may but must set aside all palpable and overriding errors of fact shown to have been made at trial”: H.L. v Canada (Attorney General), 2005 SCC 25 at para. 75, [2005] 1 SCR 401. The standard of review in the criminal context is set out in R. v Lohrer, 2004 SCC 80 at para. 1, [2004] 3 SCR 732, quoting from R. v Morrissey (1995), 22 OR (3d) 514 at p. 541, 97 CCC (3d) 193 (CA): 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. The test is stringent, and the error “must play an essential part not just in the narrative of the judgment but ‘in the reasoning process resulting in a conviction’” Lohrer at para. 2. This test also confirms that reviewable error may result not only from errors in the weighing of the evidence, but also from a flawed reasoning process. [48] Since Housen was decided in 2002, there has been a universal standard of review of findings of fact in civil proceedings: “palpable and overriding error”. There is, unfortunately, no similar universal standard in criminal law. The decision in R. v Biniaris, 2000 SCC 15, [2000] 1 SCR 381 is often cited for the proposition that the appellant must show that “no reasonable trier of fact, properly instructed, could have reached this verdict”. Because the standard of proof is higher in criminal matters, it seems anomalous that there is a greater burden on the appellant to show reviewable error in a criminal case than there is in a civil case. [49] However, decisions like Lohrer show that there are different formulations of what will amount to a reviewable error, beyond the test in Biniaris. There are arguably others: see R. v Lee, 2010 ABCA 1 at paras. 7-9, 474 AR 203, 23 Alta LR (5th) 76 affm’d other grounds 2010 SCC 52, [2010] 3 SCR 99. The recent decision in R. v R.P., 2012 SCC 22 at paras. 9 and 12 postulated four: (a) rendering a verdict a judge could not reasonably render (Biniaris); (b) drawing an inference or making a finding of fact that was plainly contradicted by the evidence (R. v Sinclair, 2011 SCC 40, [2011] 3 SCR 3); (c) drawing an inference or making a finding of fact that was incompatible with evidence that was not otherwise contradicted or rejected (Sinclair); (d) making an unreasonable assessment of the witnesses’ credibility (R. v Burke, [1996] 1 SCR 474). It is apparent that the standard of review for an unreasonable criminal verdict is more fluid than the specific test set out in Biniaris. [50] The third factor invoked by appeals like this is the principle of transparency. Trial courts have a robust power to make findings and render verdicts, but they are required to explain how they arrived at their conclusions: R. v Sheppard, 2002 SCC 26, [2002] 1 SCR 869. This is to show the parties how the result was arrived at, and to allow meaningful appellate review. The deference owed to trial courts depends in part on their reasons explaining what they did and why. [51] None of these three principles automatically or universally prevails over the other. Which one is operative in a particular appeal depends on the issues raised, and the nature of the alleged error. [52] In this appeal it is the principle of transparency, in support of the principle of avoiding unjust results, that prevails. The reasons rendered by the trial judge demonstrate errors in the path of reasoning used, and therefore the deference this Court must give to the trial process does not prevent interference with the result. [53] This appeal may usefully be contrasted with the recent decision in R. v R.P. That was also a case where the complainant testified about sexual assaults that occurred many years before when she was a child. At trial the respondent denied the events under oath. The trial judge gave detailed reasons why he accepted the evidence of the complainant, rejected the evidence of the respondent, and discounted the exculpatory evidence of the respondent’s wife. The majority of the Court of Appeal concluded that the trial judge had erred in his assessment of credibility, particularly as to the weight to be given to the wife’s evidence, and allowed the appeal. [54] The Supreme Court of Canada restored the conviction. It held that the Court of Appeal should not have asked whether the findings of the trial judge were correct, or were the most likely findings, or were the ones the Court of Appeal would have made. Rather, the issue was whether any properly instructed trier of fact could have convicted. [55] In R.P. the majority (at para. 13) quoted an earlier decision, R. v A.G., 2000 SCC 17 at para. 29, [2000] 1 SCR 439: . . . where a judge gives detailed reasons for judgment and when, as in this case, the reasons reveal that he or she was alive to the recurrent problems in this field of adjudication, the court of appeal brings no special insight to the assessment of the evidence. As this Court’s s. 686(1)(a)(i) jurisprudence makes very clear, the fact that an appeal court judge would have had a doubt when the trial judge did not is insufficient to justify the conclusion that the trial judgment was unreasonable. Both R.P. and A.G. were cases where the trial judge gave “detailed reasons for judgment” and showed that they were “alive to the recurrent problems in this field of adjudication”. Since it was possible to follow the chain of reasoning in the judgments, no overt error had been shown, the verdict was one available on the record, and the appellate courts were required to extend deference. [56] In R.P. and A.G. the principle of deference prevailed. In contrast, in this case this Court cannot be satisfied the trial judge resolved the issue of credibility in an acceptable way, and whether he was alive to the adjudicative challenges he faced. In this case there are logical flaws in the credibility assessment, and insufficient reasons to ascertain that an acceptable path of reasoning followed. Absent that foundation, deference is neither possible nor appropriate. The Reasons [57] All three of the complainants and the appellant testified at the trial. This case turned particularly on the credibility of the witnesses. The complainants’ allegations were flatly denied by the defence. The trial judge’s threshold finding of credibility was: None of the witnesses were shaken on cross-examination. There were no inconsistencies or serious contradictions in the evidence of any witness. There was nothing in the evidence to suggest that the complainants, or any one of them, had a motive to fabricate these allegations. Given these conclusions, the trial judge’s analysis of the appellant’s denial was critical. [58] Having regard to the time that had passed since the events, and the age of the complainants at that time, the trial judge was satisfied with their evidence. However, as my colleagues point out (supra, para. 19), the trial judge overlooked some of the inconsistencies in the complainants’ testimony. [59] The key was, however, the analysis of the defence. In cases like this there are two different situations that can arise: a) The first is where the accused testifies that “nothing happened”. It is not possible to give particulars of “nothing”. That is the situation in this appeal with respect to the two sisters. The appellant denied that he was ever in their bedroom. b) The second situation is where the accused admits that something happened, but offers an innocent explanation. That happened here with respect to the third complainant. The appellant remembered the nightmare event, but testified that there was no sexual activity. In this sort of situation particulars are possible. One issue on credibility is whether the appellant should reasonably have remembered this event, i.e. that is was it sufficiently unusual. If so, he might be expected to remember details, notwithstanding the passage of time. The point is not whether this Court thinks the event was sufficiently unusual that it would be remembered, but whether the trial judge thought so. In this case, the trial judge did not identify this distinction. Further, he discounted the appellant’s evidence because he had no details of the first scenario, but did have the details of the second. This is logically backwards. [60] In his brief oral reasons, the trial judge’s analysis of the appellant’s denial was as follows (AR p. 215, sentences numbered for ease of reference): [1] Having considered the totality of the evidence, I reject the accused’s evidence. [2] I find it incredible that he would recollect the details of the incident with [the third complainant] as he did in his testimony. [3] This occurred some 18 or 19 years ago yet he had a clear recollection, a recollection which became more detailed on cross-examination. [4] Yet he has no similar details for any of his interactions with the other complainants. [5] He acknowledged that he sometimes stayed overnight at the [family] home after drinking with the father and [the second sister] testified that she knew it was the accused who came into her bedroom because she could see him, hear him, and “smell him” - specifically that she recalled smelling alcohol. [6] Yet the accused would not acknowledge that his memory could be affected by alcohol. [61] In the second and third sentences the trial judge finds it “incredible” that the appellant would remember the incident with the third complainant. Remembering details is only remarkable if the event was so routine that a normal person would not remember. The trial judge never asked himself whether the two nightmare events were sufficiently unusual to justify the appellant’s memory of them. If there were only two occasions when he was sleeping on the couch, and the third complainant had a nightmare, and banged on the door, why was it incredible that he would remember it? [62] A larger problem with the analysis is that the trial judge appears to assess the “denials” with the “innocent explanation”, as if the weighing of the evidence for each involves the same process. It is not at all inconsistent to remember details of the latter but not the former. [63] The trial judge noted in the fourth sentence that the appellant had no “details for any of his interactions” with the two sisters. The appellant, however, had denied that anything had happened. There were no “interactions” to detail. It is not possible to give details of “nothing happened”. Further, the sting of the analysis is that his reported remembering of the nightmare event (which he might well have remembered) is contrasted directly with his failure to remember what he says never happened. This flawed analysis undermines the verdict. [64] The problem with criticizing the accused for failing to remember details of events that are denied was discussed in R. v Norman (1993), 16 OR (3d) 295 at p. 313, 68 OAC 22, 87 CCC (3d) 153 (CA), where the accused was charged with (but flatly denied) sexual assaults at a corn roast many years earlier: . . . In fact, the trial judge was quick to excuse variations in the evidence of the complainant and Brenda Goebel while holding the appellant to a standard of accuracy which was unrealistic for a man who, assuming that he is innocent, was attempting to recall an unremarkable corn roast some 18 years earlier. This is a case where the trial judge failed to give effect to the presumption of innocence. Here, the trier of fact was called upon to reconstruct a scene that was said to have occurred many years before. An innocent man is unlikely to have a detailed memory of distant uneventful occasions. Yet, unfortunately, the appellant’s inability to recall the minutiae of the corn roast appears to have been interpreted by the trial judge as “selective” recall. The appellant was ultimately called upon to justify his version of events, while his accusers’ inconsistencies were excused as being “insignificant when related to the charges”. Through this sort of reasoning, the trial judge effectively shifted the onus onto the accused to prove his innocence. . . . The same observation was made in R. v Tucker (1992), 120 AR 393 (CA) and R. v Ward, 2008 NLCA 38 at paras. 29-38, 234 CCC (3d) 159 (Barry J.A.). This is a critical point in this appeal. As the trial judge pointed out, all the witnesses were credible, no one was shaken on cross-examination, and the whole case depended on their testimony. [65] In closing, it is important to emphasize that this Court is not substituting its opinion on credibility for that of the trial judge. The record does not permit this Court to make such a finding. To the extent that the reasons of the trial judge disclose how he made it, the reasoning is flawed. This is not a situation like those postulated in Biniaris or R.P. where this Court can say that no properly instructed trier of fact could have convicted, and so an acquittal must be entered. From the sterile paper record and the insufficient and unsatisfactory reasons of the trial judge, it is impossible to say what a reasonable trier of fact would or would not have done. A new trial is the only solution. Similar Fact Evidence [66] Since a new trial is required in any event, it is not necessary to discuss the admission of similar fact evidence in this case. Given the somewhat unusual procedure followed at trial, and the abbreviated reasons of the trial judge, this case does not establish a good platform for setting out rules of law on the subject. However, since my colleagues have analyzed the issue, a short discussion is called for. [67] Unlike many of the reported similar fact evidence cases, this trial was conducted by a judge sitting without a jury. In a jury trial, the admission of similar fact evidence follows two distinct steps. Firstly there is a voir dire in the absence of the jury, at which the trial judge makes a threshold finding of admissibility. Then all of the evidence is presented to the jury, who must decide if the Crown has proven the case beyond a reasonable doubt. [68] In this case, since there was no jury, both counsel agreed with the trial judge that the two-step process could be collapsed into one. The evidence of each complainant was admissible in any event with respect to the counts involving that complainant. The only point of the similar fact application was whether the evidence of each complainant would be admissible with respect to the counts involving the other two complainants. Since the trial judge was to hear all of the evidence anyway, it was agreed that there would not be a separate voir dire, all three of the complainants would testify, and their evidence would be evidence at the trial. [69] The Crown called its evidence, and all three of the complainants were examined and cross-examined. Then, before the accused testified, the trial judge heard argument and made his preliminary ruling admitting the similar fact evidence. While his reasons could have been more detailed, the trial judge concluded that the complainants were reasonably capable of belief, there were striking similarities in their evidence, and that the probative value of the evidence outweighed its prejudicial effect. [70] The trial continued, the accused testified, and the trial judge gave reasons for conviction. The trial judge relied, in part, on the similar fact evidence in concluding that the Crown had proven the case beyond a reasonable doubt. [71] In the first step in the analysis of similar fact evidence, the trial judge plays a gatekeeper role. As stated in R. v Handy, 2002 SCC 56 at para. 134, [2002] 2 SCR 908: In the usual course, frailties in the evidence would be left to the trier of fact, in this case the jury. However, where admissibility is bound up with, and dependent upon, probative value, the credibility of the similar fact evidence is a factor that the trial judge, exercising his or her gatekeeper function is, in my view, entitled to take into consideration. Where the ultimate assessment of credibility was for the jury and not the judge to make, this evidence was potentially too prejudicial to be admitted unless the judge was of the view that it met the threshold of being reasonably capable of belief. It is therefore clear that, in performing the gatekeeper role, the trial judge must have regard to the basic credibility of the evidence. [72] It is axiomatic that at the second stage of the process, the jury (or other trier of fact) must assess the credibility and weight of all the evidence, including the similar fact evidence. This is part of the ordinary process of determining if the Crown has proven the case beyond a reasonable doubt. [73] Thus, where the trial is by judge alone, the trial judge will consider the credibility of the similar fact evidence at both stages. That is consistent with the approach taken in Handy at para.134. This does not in any respect result in any “circularity”. The credibility of the evidence is relevant in both stages. In the first stage, the credibility of the evidence is used to determine if, on a balance of probabilities, the evidence is admissible. In the second stage, the credibility of the evidence is used to determine if the Crown has proven the case beyond a reasonable doubt. [74] In this case it would have been helpful if the trial judge had given more detailed reasons for admitting the similar fact evidence. However, his admission of that evidence does not disclose any reviewable error. If he found (during the second stage of this blended procedure) that the complainants’ evidence was strong enough to prove the case beyond a reasonable doubt, it must have had sufficient threshold credibility to make it admissible. By the time the trial judge ruled on admissibility, he had already heard the complainants tell their entire story, and be cross-examined. It is a strained reading of the trial judge’s reasons to suggest that he only relied on the similarity of the alleged acts, and not on other aspects of the credibility of this evidence, such as the cross-examination. Prior to hearing the defence’s denial, in the trial judge’s mind the testimony of the three complainants was clearly credible enough to make it admissible. It was “reasonably capable of belief”. At the end of the trial, the defence evidence and the analysis in R. v W. (D.), [1991] 1 SCR 742 must also be brought into play. But that does not undermine the decision on threshold admissibility. [75] The admission of the similar fact evidence therefore does not disclose any reviewable error. Conclusion [76] In conclusion, the appeal must be allowed, and a new trial directed. Appeal heard on April 16, 2012 Memorandum filed at Yellowknife, N.W.T. this day of June, 2012 Slatter J.A. Appearances: A. Godfrey for the Respondent B.A. Beresh, QC for the Appellant A 0001 AP 2011 000014 IN THE COURT OF APPEAL FOR THE NORTHWEST TERRITORIES BETWEEN: Her Majesty the Queen Respondent - and - Gordon Larsen Appellant Restriction on Publication: By Court Order, information that may identify the persons described in this judgment as the complainant may not be published, broadcast, or transmitted in any manner. There is also a ban on publishing the contents of the application for the publication ban or the evidence, information or submissions at the hearing of the application. See the Criminal Code, s. 486.4 MEMORANDUM OF JUDGMENT
You are being directed to the most recent version of the statute which may not be the version considered at the time of the judgment.