R. v. Handy, [2002] 2 S.C.R. 908, 2002 SCC 56

Her Majesty The Queen Appellant

v.

James Handy Respondent

Indexed as: R. v. Handy

Neutral citation: 2002 SCC 56.

File No.: 27996.

2001: October 9; 2002: June 21.

Present: McLachlin C.J. and L’Heureux‑Dubé, Gonthier, Iacobucci, Major, Bastarache, Binnie, Arbour and LeBel JJ.

on appeal from the court of appeal for ontario

Evidence -- Admissibility -- Similar fact evidence -- Accused convicted of sexual assault -- Whether evidence by accused’s former wife of alleged past assaults against her should have been admitted as similar fact evidence.

Evidence -- Collusion -- Admissibility of testimony of witness -- Whether determination that complainant and witness had not colluded should have been condition precedent to admitting similar fact evidence of witness.

The accused was charged with sexual assault causing bodily harm. His defence was that the sex was consensual. The complainant’s position was that she had consented to vaginal sex but not hurtful or anal sex. The Crown sought to introduce similar fact evidence from the accused’s former wife to the effect that the accused has a propensity to inflict painful sex, including anal sex, and when aroused will not take no for an answer. The similar fact evidence concerned seven alleged prior incidents. The accused denied assaulting the complainant and committing any of the alleged assaults on his ex-wife. He argued that his ex-wife and the complainant had colluded. His ex-wife acknowledged that she had met the complainant a few months before the alleged sexual assault took place and that she had told the complainant about the accused’s criminal record, her allegations of abuse, that she had received $16,500 from the Criminal Injuries Compensation Board, and that all she had to do to collect the money was say that she had been abused. The trial judge admitted the similar fact evidence and ruled that it was not for him to resolve the possibility of collusion. The jury convicted the accused of sexual assault. The Court of Appeal held that the former wife’s testimony had been wrongly admitted and ordered a new trial.

Held: The appeal should be dismissed.

The similar fact evidence was wrongly admitted. The former wife’s testimony related to incidents removed in time, place and circumstances from the charge. It was thus only circumstantial evidence of the matters the jury was called on to decide and, as with any circumstantial evidence, its usefulness rested entirely on the validity of the inferences it was said to support with respect to the matters in issue. The argument for admitting the circumstantial evidence was that the jury might infer firstly that the accused is an individual who derives pleasure from sex that is painful to his partner, and will not take no for an answer, and secondly, that his character or propensity thus established gave rise to the further inference that he proceeded wilfully in this case knowing the complainant did not consent.

The prejudicial effect of this evidence outweighed its probative value and the trial judge had no discretion to admit it. Furthermore, his refusal to resolve the issue of collusion as a condition precedent to the admissibility of the evidence was an error of law. A new trial is required.

The general exclusionary rule that similar fact evidence is presumptively inadmissible has been affirmed repeatedly and recognizes that the potential for prejudice, distraction and time consumption associated with the evidence generally outweighs its probative value. Issues may arise, however, for which its probative value outweighs the potential for misuse. Similar circumstances may defy coincidence or other innocent explanation. As the evidence becomes more focussed and specific to the charge, its probative value becomes more cogent. The onus is on the prosecution to show on a balance of probabilities that the probative value of the similar fact evidence outweighs its potential for prejudice.

Similar fact evidence does not cease to be propensity evidence because it relates to an issue other than general disposition.

The principal driver of probative value is the connectedness of the evidence to the alleged offences. Factors that may support admission of such evidence include the proximity in time of the similar acts, similarity in detail, the number of occurrences of similar acts, similarities of circumstances, and any distinctive features. Exclusionary factors include the inflammatory nature of the similar acts, whether the Crown can prove its point with less prejudicial evidence, the potential for distraction, and whether admitting the evidence will consume undue time.

If the evidence of collusion amounts to no more than opportunity to collude, the issue usually is best left to the jury. In this case, however, there was some evidence of actual collusion, or at least an “air of reality” to the allegations. The Crown was thus required to satisfy the trial judge, on a balance of probabilities, that the evidence of similar facts was not tainted with collusion. That much would gain admission. It would then be for the jury to make the ultimate determination of its worth.

It was not sufficient for the Crown simply to proffer dicey evidence that if believed would have probative value. It was not incumbent on the defence to prove collusion. It was a condition precedent to admissibility that the probative value of the proffered evidence outweigh its prejudicial effect and the onus was on the Crown to satisfy that condition. The trial judge erred in law in deferring the whole issue of collusion to the jury.

The issue at trial to which the similar fact evidence related was the consent component of the actus reus and, in relation to that issue, the accused’s alleged propensity to refuse to take no for an answer. Identifying the issue merely as credibility risked admitting evidence of nothing more than general disposition. The similar fact evidence of the former wife was capable of raising an inference that the accused derived pleasure from sex that was painful to her and would not take no for an answer. The second inference, i.e., that he proceeded in this case, knowing the complainant did not consent, is more problematic. The trial judge paid insufficient attention to dissimilarities between the alleged similar acts and the offence charged. At least one allegedly similar incident was largely irrelevant and there were important dissimilarities in the other incidents. None of the allegedly similar incidents began as consensual, the dynamics of the situations differed, and all occurred in the very different context of a long-term, dysfunctional marriage.

The former wife’s evidence described incidents more reprehensible than the actual charge before the court and had a serious potential for moral prejudice. It also had the potential to create significant reasoning prejudice by distracting the jury from their proper focus and by consuming undue time. Prejudice does not necessarily recede as probative value advances. As the Crown did not discharge its onus of establishing on a balance of probabilities that the probative value of the similar fact evidence outweighed its potential for prejudice, it ought to have been excluded.

Cases Cited

Discussed: R. v. Robertson, 1987 CanLII 61 (SCC), [1987] 1 S.C.R. 918; Makin v. Attorney-General for New South Wales, [1894] A.C. 57; Morris v. The Queen, 1983 CanLII 28 (SCC), [1983] 2 S.C.R. 190; Sweitzer v. The Queen, 1982 CanLII 23 (SCC), [1982] 1 S.C.R. 949; R. v. D. (L.E.), 1989 CanLII 74 (SCC), [1989] 2 S.C.R. 111, rev’g (1987), 1987 CanLII 2536 (BC CA), 20 B.C.L.R. (2d) 384; explained: R. v. B. (C.R.), 1990 CanLII 142 (SCC), [1990] 1 S.C.R. 717; R. v. Arp, 1998 CanLII 769 (SCC), [1998] 3 S.C.R. 339; R. v. B. (F.F.), 1993 CanLII 167 (SCC), [1993] 1 S.C.R. 697; R. v. Morin, 1988 CanLII 8 (SCC), [1988] 2 S.C.R. 345; not followed: Pfennig v. R. (1995), 127 A.L.R. 99; referred to: R. v. Watson (1996), 1996 CanLII 4008 (ON CA), 50 C.R. (4th) 245; R. v. B. (L.) (1997), 1997 CanLII 3187 (ON CA), 35 O.R. (3d) 35; Director of Public Prosecutions v. Boardman, [1975] A.C. 421; Harrison’s Trial (1862), 12 How. St. Tr. 833; R. v. Seaboyer, 1991 CanLII 76 (SCC), [1991] 2 S.C.R. 577; Leblanc v. The Queen, 1975 CanLII 190 (SCC), [1977] 1 S.C.R. 339; United States v. York, 933 F.2d 1343 (1991); R. v. C. (M.H.), 1991 CanLII 94 (SCC), [1991] 1 S.C.R. 763; R. v. Batte (2000), 2000 CanLII 5751 (ON CA), 34 C.R. (5th) 197; R. v. Litchfield, 1993 CanLII 44 (SCC), [1993] 4 S.C.R. 333; R. v. Lepage, 1995 CanLII 123 (SCC), [1995] 1 S.C.R. 654; R. v. Sims, [1946] 1 All E.R. 697; R. v. Clermont, 1986 CanLII 26 (SCC), [1986] 2 S.C.R. 131; R. v. Bosley (1992), 1992 CanLII 2838 (ON CA), 18 C.R. (4th) 347; R. v. Proctor (1992), 1992 CanLII 2763 (MB CA), 69 C.C.C. (3d) 436; R. v. Hanna (1990), 1990 CanLII 11027 (BC CA), 57 C.C.C. (3d) 392; R. v. Scopelliti (1981), 1981 CanLII 1787 (ON CA), 63 C.C.C. (2d) 481; R. v. Straffen, [1952] 2 Q.B. 911; R. v. Carpenter (1982), 1982 CanLII 3308 (ON CA), 142 D.L.R. (3d) 237; R. v. Smith, 1992 CanLII 79 (SCC), [1992] 2 S.C.R. 915; Cloutier v. The Queen, 1979 CanLII 25 (SCC), [1979] 2 S.C.R. 709; R. v. Simpson (1977), 1977 CanLII 1142 (ON CA), 35 C.C.C. (2d) 337; R. v. Huot (1993), 1993 CanLII 8652 (ON CA), 16 O.R. (3d) 214; R. v. Rulli (1999), 1999 CanLII 3712 (ON CA), 134 C.C.C. (3d) 465; R. v. Fleming (1999), 1999 CanLII 18921 (NL CA), 171 Nfld. & P.E.I.R. 183; R. v. Dupras, 2000 BCSC 1128, [2000] B.C.J. No. 1513 (QL); Director of Public Prosecutions v. Kilbourne, [1973] A.C. 729; United States v. Enjady, 134 F.3d 1427 (1998), certiorari denied, 525 U.S. 887 (1998); Hodge’s Case (1838), 1838 CanLII 1 (FOREP), 2 Lewin 227, 168 E.R. 1136; R. v. H., [1995] 2 A.C. 596; R. v. Kenny (1996), 1996 CanLII 6633 (NL CA), 108 C.C.C. (3d) 349; R. v. McDonald (2000), 2000 CanLII 16871 (ON CA), 148 C.C.C. (3d) 273; R. v. Ewanchuk, 1999 CanLII 711 (SCC), [1999] 1 S.C.R. 330; R. v. Smith (1915), 84 L.J.K.B. 2153; R. v. Minhas (1986), 1986 CanLII 144 (ON CA), 29 C.C.C. (3d) 193; R. v. Corbett, 1988 CanLII 80 (SCC), [1988] 1 S.C.R. 670; Director of Public Prosecutions v. P., [1991] 2 A.C. 447; R. v. Marquard, 1993 CanLII 37 (SCC), [1993] 4 S.C.R. 223.

Statutes and Regulations Cited

Canadian Charter of Rights and Freedoms, ss. 7, 11(d).

Criminal Code, R.S.C. 1985, c. C-46, s. 718.2.

Authors Cited

Acorn, A. E. “Similar Fact Evidence and the Principle of Inductive Reasoning: Makin Sense” (1991), 11 Oxford J. Legal Stud. 63.

Cross and Tapper on Evidence, 9th ed. By Colin Tapper. London: Butterworths, 1999.

Delisle, R. J. “The Direct Approach to Similar Fact Evidence” (1996), 50 C.R. (4th) 286.

Durston, Gregory. “Similar Fact Evidence: A Guide for the Perplexed in the Light of Recent Cases” (1996), 160 Justice of the Peace & Local Government Law 359.

Great Britain. Law Commission. Consultation Paper No. 141. Evidence in Criminal Proceedings: Previous Misconduct of a Defendant. London: HMSO, 1996.

Lloyd-Bostock, S. “The Effects on Juries of Hearing About the Defendant’s Previous Criminal Record: A Simulation Study”, [2000] Crim. L.R. 734.

Martin, G. Arthur. “Similar Fact Evidence”, [1984] Spec. Lect. L.S.U.C. 1.

McCormick on Evidence, vol. 1, 5th ed. By John W. Strong, General Editor. St. Paul, Minn.: West Group, 1999.

McWilliams, Peter K. Canadian Criminal Evidence, vol. 1, 3rd ed. Aurora, Ont.: Canada Law Book, 1988 (loose-leaf updated October 2001, release 27).

Mueller, Christopher B., and Laird C. Kirkpatrick. Federal Evidence, Cumulative Supplement to vol. 2, 2nd ed. Rochester, N.Y.: Lawyers Cooperative Publishing, 1994 & Supp. 2001.

Paciocco, David M., and Lee Stuesser. The Law of Evidence, 2nd ed. Toronto: Irwin Law, 1999.

Pickel, Kerri L. “Inducing Jurors to Disregard Inadmissible Evidence: A Legal Explanation Does Not Help” (1995), 19 Law & Hum. Behav. 407.

Rosenberg, Marc. “Evidence of Similar Acts and Other Extrinsic Misconduct”. In National Criminal Law Program, Criminal Evidence. Vancouver: Federation of Law Societies of Canada, 1994, section 8.1.

Sopinka, John, Sidney N. Lederman and Alan W. Bryant. The Law of Evidence in Canada, 2nd ed. Toronto: Butterworths, 1999.

Stuesser, Lee. “Similar Fact Evidence in Sexual Offence Cases” (1997), 39 Crim. L.Q. 160.

Wigmore, John Henry. Evidence in Trials at Common Law, vol. 1A. Revised by Peter Tillers. Boston: Little, Brown & Co., 1983.

Wigmore, John Henry. Evidence in Trials at Common Law, vol. 2. Revised by James H. Chadbourn. Boston: Little, Brown & Co., 1979.

Wissler, Roselle L., and Michael J. Saks. “On the Inefficacy of Limiting Instructions: When Jurors Use Prior Conviction Evidence to Decide on Guilt” (1985), 9 Law & Hum. Behav. 37.

APPEAL from a judgment of the Ontario Court of Appeal (2000), 2000 CanLII 5707 (ON CA), 48 O.R. (3d) 257, 131 O.A.C. 297, 32 C.R. (5th) 316, 145 C.C.C. (3d) 177, [2000] O.J. No. 1373 (QL), allowing the accused’s appeal from a decision of the Ontario Court (General Division). Appeal dismissed.

Christopher Webb, for the appellant.

Richard N. Stern and David E. Harris, for the respondent.

The judgment of the Court was delivered by

[1] Binnie J. -- The principal issues in this case are (i) the test for the admissibility of discreditable similar fact evidence where the credibility of the complainant (as distinguished from the identification of the accused) is the issue, and (ii) the impact of potential collusion on the admissibility of such evidence.

[2] The respondent was charged with sexual assault causing bodily harm. The complainant, a casual acquaintance, says that consensual vaginal sex following a drinking session at a bar turned into hurtful non-consensual vaginal and subsequently anal sex accompanied by physical abuse. At the respondent’s trial, the Crown sought to introduce the evidence of his ex-wife about seven allegedly “similar fact” incidents (or “similar acts”) that occurred during their seven-year, abusive and sometimes violent cohabitation (interrupted by his incarceration for unrelated sexual assaults) which produced three children. The trial judge admitted the evidence and the jury convicted the respondent of the lesser offence of sexual assault.

[3] The respondent says that the jury ought not to have considered evidence of alleged misconduct which was outside the subject matter of the charge in the indictment, and that evidence of his allegedly brutal disposition, or alleged propensity for hurtful sex, was highly prejudicial to a fair trial. Moreover, he says, the “similar” facts are not similar and in any event the complainant and his ex-wife colluded. The Ontario Court of Appeal ruled that the similar fact evidence was wrongly admitted and ordered a new trial. I agree and would dismiss the appeal.

I. Facts

[4] The complainant’s evidence was that on the evening of December 6, 1996, she went out drinking with some friends. The respondent, whom she had met six months earlier, was also at the bar. The two spent the evening drinking and flirting with one another. After leaving the bar, they went to the home of one of the complainant’s friends to smoke marijuana. The respondent and the complainant left the house together and drove to a nearby motel intending to have sex. In the course of vaginal intercourse, she became upset because the respondent was hurting her, forcing himself into her. She told him that it was painful but he continued. He then brusquely switched to anal intercourse. She said, “Stop that, it hurts”. She tried to get him off her or to make him stop but he would not. She slapped his face. She says he hit her on the chest, he grabbed her arms, squeezed her stomach and choked her, and he punched her. She says she was pleading and crying. She had consented to vaginal sex but she did not consent to and did not want anal sex. After the incident, she told the respondent that he had made her bleed. He allegedly responded to her by saying, “What the hell am I doing here? Why does this kee[p] happening to me?”

[5] A number of witnesses testified that they had seen bruises on her throat, chest and arms in the days following the incident. The complainant was diagnosed with post-traumatic stress.

A. The Similar Fact Evidence

[6] The respondent’s defence was that the sex was consensual. The issue thus came down to credibility on the consent issue. The Crown sought to introduce similar fact evidence from the respondent’s former wife to the effect that the respondent has a propensity to inflict painful sex and when aroused will not take no for an answer. It was thus tendered to explain why the complainant should be believed when she testified that the assault proceeded despite her protest.

(1) Incident One

[7] In March 1990, a few weeks after their first child was born, the ex-wife says the respondent wanted to have sexual intercourse with her to “see what it would feel like”. She did not want to do so because she thought that it would be painful. The respondent insisted that they have vaginal intercourse. Once they started she told the respondent that she was in pain but he did not stop.

(2) Incident Two

[8] Five or six months later she and the respondent visited her sister and brother-in-law in their mobile trailer. After everyone went to bed, the respondent wanted to have sexual intercourse. She told the respondent that she did not want to have sex because her sister and her husband were at the other end of the trailer. She tried to move away from him. The respondent told her to shut up and had vaginal intercourse with her anyway.

(3) Incident Three

[9] She returned home one day to find that the respondent had invited a number of people to their apartment for a party. After seeing the respondent tickle two women on the couch, she got angry and told everyone to leave. After most of the guests departed, she went into the bedroom. The respondent followed her. He was upset that she had broken up his party. He tried to have intercourse with her. She tried to get away but he blocked the door with a dresser. She then attempted to flee through the second floor bedroom window, but he pulled her back in. He then forced her to have vaginal intercourse and passed out.

(4) Incident Four

[10] Sometime early in 1992, the respondent came home drunk and wanted to have anal intercourse. She told him that she did not want to do so because it had hurt her on previous occasions. The respondent initiated anal intercourse nonetheless. She kept moving and tried to get away. Eventually, he grabbed a bottle of baby oil from underneath the bed and applied the oil to his penis and her anus. He initiated anal intercourse. They were interrupted by a crying baby, and she used the distraction to escape to the basement but the respondent followed her. He told her that if she did not stop running, he would tie her up with a rope. She ran naked from the house and over to the neighbour’s house. The police were called but she did not lay charges.

(5) Incident Five

[11] The respondent was imprisoned from 1992 until 1995 for sexual assaults on two other women (although the fact they were “other” women was withheld from the jury by agreement of counsel). In that period he placed a threatening phone call to his then wife, which precipitated their divorce. They resumed living together soon after he was released. Shortly thereafter, she became upset because the respondent had gone out with a woman he had once dated. The respondent became angry, grabbed her by the throat, threw her around, pinned her against the wall and broke their glass coffee table. He did not, however, sexually assault her on that occasion.

(6) Incident Six

[12] One night during the summer of 1996, she and the respondent were returning home after dropping off their friends. The respondent told her that instead of going home, they were going to a gravel pit where she “was going to get it up the ass”. She testified that he had forced her to have sex with him at the gravel pit in the past. She told him that she was willing to do anything other than anal intercourse because it hurt too much. The respondent, however, insisted on anal intercourse. Once at the gravel pit he attempted anal intercourse, but was unsuccessful because there was insufficient room in the back seat of the car. The respondent took her out of his car and put her face down on the hood. He attempted anal intercourse again. He eventually turned her over onto her back and had vaginal intercourse.

(7) Incident Seven

[13] In October 1996, her grandfather passed away. She and the respondent were alone in her mother’s home. She was crying and upset. She testified that her crying “turned [the respondent] on” and that he wanted to have sexual intercourse on her mother’s new couch. She told him that she did not want to. The respondent put her on the couch and commenced vaginal intercourse. She cried. While they were having intercourse, he punched her a number of times in the stomach to make her cry louder.

B. The Respondent’s Testimony

[14] The respondent denied committing any of the alleged assaults on his ex-wife. With respect to the complainant’s allegations, he testified that he met her at the bar, that they were both intoxicated and that they left the bar together. Eventually they went to a motel room. He testified that once inside the room, the complainant straddled him while he lay on his back and they engaged in approximately 15 to 20 minutes of vaginal intercourse. He denied that she had complained or told him to stop. He also denied hitting her and choking her. He testified that she drove him home at approximately 6:40 a.m. He did not see her again.

C. The Evidence of Collusion

[15] The ex-wife testified that she had met the complainant a few months before the alleged sexual assault took place. She had told the complainant at that time about the respondent’s criminal record and her allegations of his abuse of her during their marriage. The ex-wife told the complainant that she had received $16,500 from the Criminal Injuries Compensation Board and agreed when it was put to her in cross-examination that “[a]ll you had to do [to get the money] was say that you were abused”. The ex-wife’s cross-examination was, in part, as follows:

Q. You knew [the complainant]?

A. Yes, I did.

Q. You had met her in the summer of ‘96?

A. That’s correct.

Q. She had come over and visited with you, right?

A. That’s correct.

Q. At one point, she actually said to you that she thought that [the respondent] loved you very much?

A. Yes, she did.

Q. And you straightened her out?

A. That’s correct.

Q. And you told her that he had been to jail?

A. Yes, I did.

Q. You told her that he abused you?

A. Yes, I did.

Q. And you told her that you collected $16,500 from the government. All you had to do was say that you were abused.

A. Yes.

Q. So she knew all of that before December of 1996?

A. Yes. [Emphasis added.]

[16] Subsequently, on December 6, 1996, the complainant met up with the respondent at the bar and, after sharing some marijuana, agreed to accompany him to a motel for sex.

II. Judicial History

A. Ontario Court (General Division) -- Jennings J.

[17] At the conclusion of the voir dire, the trial judge admitted the “similar fact” evidence on the basis that:

(i) the ex-wife’s proposed evidence might assist the jury in determining how he had acted with the complainant;

(ii) the evidence was discreditable to the respondent and could only be admitted if its probative value outweighed its prejudicial effect;

(iii) the issue for the jury was the credibility of the complainant’s allegation that sex continued in a violent manner in the face of attempts to refuse sex and not simply whether a withdrawal of consent had been communicated; and

(iv) the similar fact evidence, if believed, “establishe[d] a pattern of using an initially consensual situation to escalate into violent, painful sexual connection, with both vaginal and anal penetration”. It would show a pattern of behaviour and confirm the credibility of the complainant, both of which the trial judge described as legitimate purposes for the reception of the evidence. This showed more than a mere propensity to commit the acts based upon bad character. The cogency was derived from

the overriding similarity of the conversion of an occasion when consensual sex may be anticipated, into one of continuing vaginal sex after complaint, pain and request to stop, accompanied by physical attack, and of initiating and continuing anal sex without consent, persuades me the proposed evidence has the substantial probative value required, and as was the case in R. v. B. (L.) [1997 CanLII 3187 (ON CA), [1997] 35 O.R. (3d) 35] the proposed evidence is relevant to an important issue, the credibility of [the complainant].

[18] The trial judge ruled that it was not for him to resolve the possibility of collusion between the former wife and the complainant.

There is no direct evidence of collusion between [the ex-wife and the complainant], although the former told the latter of the assaults upon her. Regardless, this is a decision for the trier of fact to make. [Emphasis added.]

B. Ontario Court of Appeal (2000), [2000 CanLII 5707 (ON CA)](https://www.canlii.org/en/on/onca/doc/2000/2000canlii5707/2000canlii5707.html), 48 O.R. (3d) 257

[19] Charron J.A. for the court held that the trial judge had identified the correct test for admitting similar fact evidence but he had erred in its application. In her opinion, the evidence should not have been admitted at trial since the probative value of the evidence was outweighed by its potential prejudicial effect.

[20] The strength of the evidence was weakened by the fact that the respondent had denied the incidents and that they formed the subject matter of other proceedings in which they were as yet unproven.

[21] The alleged similar acts were quite disparate in nature and, despite sharing certain characteristics, it was difficult to fit them into any pattern specific enough to bolster the complainant’s credibility. Charron J.A. further held that there were non-superficial dissimilarities. While the acts alleged by the ex-wife took place during a conjugal, long-term relationship, the acts alleged by the complainant took place during a short, casual affair that had began with her consent.

[22] Charron J.A. also held that there had been a potential for collusion that further weakened any probative value that could be derived from the former wife’s testimony. She held that potential for collusion “is always a serious consideration in the assessment of the strength of this kind of evidence” (para. 41) since collusion between witnesses may deprive similar fact evidence of most of its probative value. The prospect of collusion is “not a matter that can simply be left for the jury to determine without giving it due consideration in the assessment of the probative value of the evidence” (para. 41).

[23] The credibility of the ex-wife was problematic. She had considerably delayed reporting any of the incidents. The eventual timing of her complaints raised issues with respect to her motives. The complaint with respect to four incidents had first been made in support of an uncontested application for compensation before the Criminal Injuries Compensation Board when the respondent was in prison. The rest of the complaints had been made after her final separation from the respondent and shortly after she had learned of the charges laid in this case.

III. Analysis

[24] The trial judge admitted the similar fact evidence in this case because he thought the way the respondent “acted on previous occasions with [his ex-wife], may very well assist the jury in determining how he acted with [the complainant]”. The common thread, according to the Crown, is that the respondent derives pleasure from inflicting pain on a sexual partner and insists on sex “his way” irrespective of consent.

[25] From the respondent’s point of view, introduction of the similar fact evidence radically changed the trial. He was on trial for one incident, to which he pleaded not guilty, but was instead confronted with eight different incidents, of which seven were not the subject matter of any charge. The jury might conclude that a man with a track record of discreditable treatment of his ex-wife in their sexual relations would be acting in character by forcing himself on the resisting complainant, but this he says was unfair because it bolstered the complainant’s credibility by exogenous evidence that related neither to the complainant nor to the charge. At the least the jury might conclude that the respondent was a repugnant individual deserving of punishment and a conviction would, as a matter of rough equivalence, give him his just desserts.

A. The Disputed Inferences

[26] The ex-wife’s testimony relates to incidents removed in time, place and circumstances from the charge. It is thus only circumstantial evidence of the matters the jury was called on to decide and, as with any circumstantial evidence, its usefulness rests entirely on the validity of the inferences it is said to support with respect to the matters in issue. The argument for admitting this circumstantial evidence is that the jury may infer firstly that the respondent is an individual who derives pleasure from sex that is painful to his partner, and will not take no for an answer, and secondly, that his character or propensity thus established gives rise to the further inference that he proceeded wilfully in this case knowing the complainant did not consent. As stated by Wilson J. in R. v. Robertson, 1987 CanLII 61 (SCC), [1987] 1 S.C.R. 918, at p. 943:

In discussing the probative value we must consider the degree of relevance to the facts in issue and the strength of the inference that can be drawn.

See also R. v. Watson (1996), 1996 CanLII 4008 (ON CA), 50 C.R. (4th) 245 (Ont. C.A.), at p. 258, per Doherty J.A.; R. v. B. (L.) (1997), 1997 CanLII 3187 (ON CA), 35 O.R. (3d) 35 (C.A.), at p. 45, per Charron J.A. See also: J. Sopinka, S. N. Lederman and A. W. Bryant, The Law of Evidence in Canada (2nd ed. 1999), at § 11.113; D. M. Paciocco and L. Stuesser, The Law of Evidence (2nd ed. 1999), at pp. 39-41; Cross and Tapper on Evidence (9th ed. 1999), at pp. 350 et seq.; Wigmore on Evidence, vol. 1A (Tillers rev. 1983), at pp. 1152-53.

[27] The contest over the admissibility of similar fact evidence is all about inferences, i.e., when do they arise? What are they intended to prove? By what process of reasoning do they prove it? How strong is the proof they provide? When are they so unfair as to be excluded on the grounds of judicial policy and the presumption of innocence? The answers to these questions have proven so controversial as to create what Lord Hailsham described as a “pitted battlefield”: Director of Public Prosecutions v. Boardman, [1975] A.C. 421 (H.L.), at p. 445.

[28] There is more consensus on the nature of the problems than there is on the correctness of the solutions: compare the differing approaches in R. J. Delisle, “The Direct Approach to Similar Fact Evidence” (1996), 50 C.R. (4th) 286; A. E. Acorn, “Similar Fact Evidence and the Principle of Inductive Reasoning: Makin Sense” (1991), 11 Oxford J. Legal Stud. 63; P. K. McWilliams, Canadian Criminal Evidence (3rd ed. (loose-leaf)), vol. 1, at p. 11-4; M. Rosenberg (now J.A.), “Evidence of Similar Acts and Other Extrinsic Misconduct”, in National Criminal Law Program, Criminal Evidence (1994), section 8.1, at p. 3; and L. Stuesser, “Similar Fact Evidence in Sexual Offence Cases” (1997), 39 Crim. L.Q. 160.

[29] The immediate questions are whether the double inferences are properly raised on the facts of this case, and if so, whether they add such probative strength in the circumstances here that the ex-wife’s evidence ought to be admitted despite the potential prejudice. The respondent says that the so-called similar facts are insufficiently connected in time and circumstances to the offence charged, i.e., there is an insufficient nexus to make the conduct with his ex-wife a reliable guide to his alleged conduct with the complainant. Moreover, even if they are, he says he should not as a matter of policy be put at risk of conviction by confusing the jury about what he allegedly did in other times and at other places.

[30] I should note that the Crown did not attempt to call any expert evidence in relation to the validity of the inferences respecting the respondent’s psychological make-up that it sought to have the jury draw from the ex-wife’s evidence and their applicability to the facts in issue.

B. The General Exclusionary Rule

[31] The respondent is clearly correct in saying that evidence of misconduct beyond what is alleged in the indictment which does no more than blacken his character is inadmissible. Nobody is charged with having a “general” disposition or propensity for theft or violence or whatever. The exclusion thus generally prohibits character evidence to be used as circumstantial proof of conduct, i.e., to allow an inference from the “similar facts” that the accused has the propensity or disposition to do the type of acts charged and is therefore guilty of the offence. The danger is that the jury might be confused by the multiplicity of incidents and put more weight than is logically justified on the ex-wife’s testimony (“reasoning prejudice”) or by convicting based on bad personhood (“moral prejudice”): Great Britain Law Commission, Consultation Paper No. 141, Evidence in Criminal Proceedings: Previous Misconduct of a Defendant (1996), at § 7.2.

[32] This is a very old rule of the common law. Reference may be made to seventeenth-century trials in which the prosecution was scolded for raising prior felonious conduct, as for example to Lord Holt C.J. in Harrison’s Trial (1692), 12 How. St. Tr. 833 (Old Bailey (London)), at p. 864: “Are you going to arraign his whole life? Away, Away, that ought not to be; that is nothing to the matter.”

[33] Subsequently, and most famously, the general exclusionary rule was laid down by Lord Herschell L.C. Makin v. Attorney-General for New South Wales, [1894] A.C. 57 (P.C.), in these terms, at p. 65:

It is undoubtedly not competent for the prosecution to adduce evidence tending to shew that the accused has been guilty of criminal acts other than those covered by the indictment, for the purpose of leading to the conclusion that the accused is a person likely from his criminal conduct or character to have committed the offence for which he is being tried.

[34] The court spoke there of “criminal acts”, but this has been broadened to include any proffered “similar facts” of a discreditable nature (Robertson, supra, at p. 941; B. (L.), supra, at pp. 45-46), a category which includes the conduct alleged by the ex-wife in this case.

[35] The dangers of propensity reasoning are well recognized. Not only can people change their ways but they are not robotic. While juries in fourteenth-century England were expected to determine facts based on their personal knowledge of the character of the participants, it is now said that to infer guilt from a knowledge of the mere character of the accused is a “forbidden type of reasoning”: Boardman, supra, at p. 453, per Lord Hailsham.

[36] The exclusion of evidence of general propensity or disposition has been repeatedly affirmed in this Court and is not controversial. See Morris v. The Queen, 1983 CanLII 28 (SCC), [1983] 2 S.C.R. 190; R. v. Morin, 1988 CanLII 8 (SCC), [1988] 2 S.C.R. 345; R. v. B. (C.R.), 1990 CanLII 142 (SCC), [1990] 1 S.C.R. 717; R. v. Arp, 1998 CanLII 769 (SCC), [1998] 3 S.C.R. 339.

Policy Basis for the Exclusion

[37] The policy basis for the exclusion is that while in some cases propensity inferred from similar facts may be relevant, it may also capture the attention of the trier of fact to an unwarranted degree. Its potential for prejudice, distraction and time consumption is very great and these disadvantages will almost always outweigh its probative value. It ought, in general, to form no part of the case which the accused is called on to answer. It is excluded notwithstanding the general rule that all relevant evidence is admissible: Arp, supra, at para. 38; Robertson, supra, at p. 941; Morris, supra, at pp. 201-2; R. v. Seaboyer, 1991 CanLII 76 (SCC), [1991] 2 S.C.R. 577, at p. 613.

[38] If propensity evidence were routinely admitted, it might encourage the police simply to “round up the usual suspects” instead of making a proper unblinkered investigation of each particular case. One of the objectives of the criminal justice system is the rehabilitation of offenders. Achievement of this objective is undermined to the extent the law doubts the “usual suspects” are capable of turning the page and starting a new life.

[39] It is, of course, common human experience that people generally act consistently with their known character. We make everyday judgments about the reliability or honesty of particular individuals based on what we know of their track record. If the jurors in this case had been the respondent’s inquisitive neighbours, instead of sitting in judgment in a court of law, they would undoubtedly have wanted to know everything about his character and related activities. His ex-wife’s anecdotal evidence would have been of great interest. Perhaps too great, as pointed out by Sopinka J. in B. (C.R.), supra, at p. 744:

The principal reason for the exclusionary rule relating to propensity is that there is a natural human tendency to judge a person’s action on the basis of character. Particularly with juries there would be a strong inclination to conclude that a thief has stolen, a violent man has assaulted and a pedophile has engaged in pedophilic acts. Yet the policy of the law is wholly against this process of reasoning.

[40] The policy of the law recognizes the difficulty of containing the effects of such information which, once dropped like poison in the juror’s ear, “swift as quicksilver it courses through the natural gates and alleys of the body”: Hamlet, Act I, Scene v, ll. 66-67.

C. The Narrow Exception of Admissibility

[41] While emphasizing the general rule of exclusion, courts have recognized that an issue may arise in the trial of the offence charged to which evidence of previous misconduct may be so highly relevant and cogent that its probative value in the search for truth outweighs any potential for misuse, per Sopinka J., dissenting, in B. (C.R.), supra, at p. 751:

The fact that the alleged similar facts had common characteristics with the acts charged, could render them admissible, and, therefore, supportive of the evidence of the complainant. In order to be admissible, however, it would be necessary to conclude that the similarities were such that absent collaboration, it would be an affront to common sense to suggest that the similarities were due to coincidence . . . . [Emphasis added.]

[42] The “common sense” condemnation of exclusion of what may be seen as highly relevant evidence has prompted much judicial agonizing, particularly in cases of alleged sexual abuse of children and adolescents, whose word was sometimes unfairly discounted when opposed to that of ostensibly upstanding adults. The denial of the adult, misleadingly persuasive on first impression, would melt under the history of so many prior incidents as to defy innocent explanation. That said, there is no special rule for sexual abuse cases. In any case, the strength of the similar fact evidence must be such as to outweigh “reasoning prejudice” and “moral prejudice”. The inferences sought to be drawn must accord with common sense, intuitive notions of probability and the unlikelihood of coincidence. Although an element of “moral prejudice” may be introduced, it must be concluded by the trial judge on a balance of probabilities that the probative value of the sound inferences exceeds any prejudice likely to be created.

[43] As set out in the dissenting reasons of Dickson J., concurred in by Laskin C.J., in Leblanc v. The Queen, 1975 CanLII 190 (SCC), [1977] 1 S.C.R. 339, at p. 345: “There is, however, a limited but legitimate area of exception to the general exclusionary rule that the accused has only to answer the specific charge contained in the indictment.”

[44] The criminal trial is, after all, about the search for truth as well as fairness to an accused. Thus Lord Herschell L.C., in what is called the second “branch” of Makin, supra, said at p. 65:

On the other hand, the mere fact that the evidence adduced tends to shew the commission of other crimes does not render it inadmissible if it be relevant to an issue before the jury, and it may be so relevant if it bears upon the question whether the acts alleged to constitute the crime charged in the indictment were designed or accidental, or to rebut a defence which would otherwise be open to the accused.

[45] In Makin itself, the death of one small child found buried in the back garden of the accused could conceivably be thought to be from natural or perhaps accidental causes, but when numerous other bodies were later dug up in the same backyard and at previous residences of the accused, the coincidence of so many premature deaths by innocent causes of babies of recent good health defied belief. Coincidence, as an explanation, has its limitations. As it was put in one American case: “The man who wins the lottery once is envied; the one who wins it twice is investigated” (United States v. York, 933 F.2d 1343 (7th Cir. 1991), at p. 1350).

[46] It was thus held in Makin that the accumulation of babies found dead in similar circumstances permitted, in relation to the accused, the double inferences of propensity mentioned above. The improbability of an innocent explanation was manifest.

Policy Basis for the Exception

[47] The policy basis for the exception is that the deficit of probative value weighed against prejudice on which the original exclusionary rule is predicated is reversed. Probative value exceeds prejudice, because the force of similar circumstances defies coincidence or other innocent explanation.

[48] Canadian case law recognizes that as the “similar facts” become more focussed and specific to circumstances similar to the charge (i.e., more situation specific), the probative value of propensity, thus circumscribed, becomes more cogent. As the differences and variables that distinguish the earlier “similar facts” from the subject matter of the charge in this type of case are reduced, the cogency of the desired inferences is thought to increase. Ultimately the policy premise of the general exclusionary rule (prejudice exceeds probative value) ceases to be true.

D. The Test of Admissibility

[49] The present rule was succinctly formulated by McIntyre J. in Sweitzer v. The Queen, 1982 CanLII 23 (SCC), [1982] 1 S.C.R. 949, at p. 953:

The general principle described by Lord Herschell may and should be applied in all cases where similar fact evidence is tendered and its admissibility will depend upon the probative effect of the evidence balanced against the prejudice caused to the accused by its admission whatever the purpose of its admission.

[50] The rule received more extended and comprehensive treatment by McLachlin J. in B. (C.R.), supra, itself. The accused was charged with sexual offences against his daughter. The daughter testified that the alleged acts began when she was 11 years old and continued for two years. The accused denied the allegations. In support of the child’s testimony, the Crown sought to adduce evidence showing that 10 years earlier, the accused had had sexual relations with his common-law wife’s 15-year-old daughter, with whom he had enjoyed a father-daughter relationship. Writing for a five-judge majority, McLachlin J. held that the evidence, though “borderline”, was admissible. Although evidence relating solely to the accused’s disposition will generally be excluded, exceptions to this rule will arise when the probative value of the evidence outweighs its prejudicial effect (at pp. 734-35):

This review of the jurisprudence leads me to the following conclusions as to the law of similar fact evidence as it now stands in Canada. The analysis of whether the evidence in question is admissible must begin with the recognition of the general exclusionary rule against evidence going merely to disposition. . . . [E]vidence which is adduced solely to show that the accused is the sort of person likely to have committed an offence is, as a rule, inadmissible. Whether the evidence in question constitutes an exception to this general rule depends on whether the probative value of the proposed evidence outweighs its prejudicial effect.

[51] The Court thus affirmed that evidence classified as “disposition” or “propensity” evidence is, exceptionally, admissible. McLachlin J. continued at p. 735:

In a case such as the present, where the similar fact evidence sought to be adduced is prosecution evidence of a morally repugnant act committed by the accused, the potential prejudice is great and the probative value of the evidence must be high indeed to permit its reception. The judge must consider such factors as the degree of distinctiveness or uniqueness between the similar fact evidence and the offences alleged against the accused, as well as the connection, if any, of the evidence to issues other than propensity, to the end of determining whether, in the context of the case before him, the probative value of the evidence outweighs its potential prejudice and justifies its reception.

[52] McLachlin J. formulated the test for admissibility of disposition or propensity evidence, at p. 732:

. . . evidence of propensity, while generally inadmissible, may exceptionally be admitted where the probative value of the evidence in relation to an issue in question is so high that it displaces the heavy prejudice which will inevitably inure to the accused where evidence of prior immoral or illegal acts is presented to the jury.

[53] Subsequently, in R. v. C. (M.H.), 1991 CanLII 94 (SCC), [1991] 1 S.C.R. 763, McLachlin J. referred to B. (C.R.) as the governing authority (at pp. 771-72), as did a unanimous Court almost a decade later in its most recent pronouncement on the topic, in Arp, supra, per Cory J., at para. 41:

. . . evidence which is adduced solely to show that the accused is the sort of person likely to have committed an offence is, as a rule, inadmissible. Whether the evidence in question constitutes an exception to this general rule depends on whether the probative value of the proposed evidence outweighs its prejudicial effect. [Emphasis added.]

[54] Cory J. added some observations in Arp at para. 80 about the trial judge’s instructions to the jury about the use to be made of propensity evidence. These observations should not be taken out of context. Contrary to some commentary, Arp did not qualify the Court’s endorsement of the general test set out in B. (C.R.) as is evident from Cory J.’s repeated references thereto (at paras. 42, 50 and 65):

It can be seen that in considering whether similar fact evidence should be admitted the basic and fundamental question that must be determined is whether the probative value of the evidence outweighs its prejudicial effect.

In summary, in considering the admissibility of similar fact evidence, the basic rule is that the trial judge must first determine whether the probative value of the evidence outweighs its prejudicial effect.

The issue in every case is whether the probative value of the evidence outweighs its prejudicial effect. [Emphasis added.]

The B. (C.R.) test can thus be taken as stating the law in Canada.

[55] Similar fact evidence is thus presumptively inadmissible. The onus is on the prosecution to satisfy the trial judge on a balance of probabilities that in the context of the particular case the probative value of the evidence in relation to a particular issue outweighs its potential prejudice and thereby justifies its reception.

E. Difficulties in the Application of the Test

[56] It has been recognized since Lord Herschell L.C.’s time that it is one thing to talk about so general a test as balancing probative value against prejudice, and a different and much more difficult thing to apply the test in a practical way (Makin, supra, at p. 65). What are the manageable criteria? How “probative” must the evidence be to get over the admissibility hurdle? How much prejudice is too much? How do we calibrate the scales that balance probative value against prejudice?

[57] In an attempt to provide more precise guidance, Canadian appellate courts have from time to time advocated, amongst others, a “categories” approach, a multi-step “purpose” approach and a “conclusiveness” approach. Each of these attempts, helpful as they were in practice, were ultimately thought to obfuscate and detract from the principled approach eventually adopted in Sweitzer, B. (C.R.) and Arp. It was found for example that squeezing propensity evidence into a pre-authorized pigeon hole or recognized “category” sometimes unfairly gained its admission even though, in context, the prejudice was overwhelming: Sopinka, Lederman and Bryant, supra, at § 11.30.

[58] Nevertheless, Sweitzer, B. (C.R.) and Arp did not advocate a free-wheeling approach. They fully recognized the potentially poisonous nature of propensity evidence, and sharply circumscribed the circumstances in which it can be introduced.

(1) Propensity Evidence by Any Other Name Is Still Propensity Evidence

[59] It is occasionally suggested that once the similar fact evidence is related to an issue other than “mere” propensity or “general” disposition, it somehow ceases to be propensity evidence. I do not think this is true.

[60] One of the virtues of B. (C.R.) is its candid acknowledgment that “evidence of propensity, while generally inadmissible, may exceptionally be admitted” (p. 732) to help establish that the accused did or did not do the act in question (at pp. 731-32):

While the language of some of the assertions of the exclusionary rule admittedly might be taken to suggest that mere disposition evidence can never be admissible, the preponderant view prevailing in Canada is the view taken by the majority in Boardman -- evidence of propensity, while generally inadmissible, may exceptionally be admitted where the probative value of the evidence in relation to an issue in question is so high that it displaces the heavy prejudice which will inevitably inure to the accused where evidence of prior immoral or illegal acts is presented to the jury. [Emphasis in original.]

[61] In other words, while identification of the issue defines the precise purpose for which the evidence is proffered, it does not (and cannot) change the inherent nature of the propensity evidence, which must be recognized for what it is. By affirming its true character, in my view, the Court keeps front and centre its dangerous potential.

[62] I refer again to Arp, supra, where Cory J., for the Court, reaffirmed the proposition that in exceptional circumstances propensity evidence is admissible, at para. 40:

Thus evidence of propensity or disposition may be relevant to the crime charged, but it is usually inadmissible because its slight probative value is ultimately outweighed by its highly prejudicial effect. [Emphasis added.]

[63] While Cory J. rested admissibility on the improbability of coincidence (paras. 43 and 45), this does not in my view detract from his recognition that the underlying reasoning was through propensity. When similar facts are attributed to an accused acting “in character”, it is the inferred continuity of character and nothing else that displaces what might otherwise be explained innocently as mere “coincidence”.

[64] I emphasize the reference in Arp to “usually inadmissible”. Cory J. recognized, as did McLachlin J. in B. (C.R.), supra, that disposition evidence could unusually and exceptionally be admitted if it survives the rigours of balancing probative value against prejudice.

[65] While some of Sopinka J.’s judgments are occasionally cited for the proposition that propensity evidence, as such, is never admissible, I think this overstates his position. He, like McLachlin J. in C. (M.H.), at p. 771, occasionally uses the word “disposition” as shorthand for “general” disposition, or bad character, which both agree to be inadmissible. Sopinka J.’s agreement with the use of situation specific evidence of propensity is confirmed, I think, in his majority opinion in Morin, supra, a case in which propensity evidence was sought to be established by expert testimony rather than similar fact evidence. He stated at p. 370:

It seems to me that the policy against the admission of such evidence is satisfied if its probative value exceeds its prejudicial effect. On the other hand, the mere fact that the evidence has some relevance does not secure its admissibility if it does not meet this test.

[66] In R. v. D. (L.E.), 1989 CanLII 74 (SCC), [1989] 2 S.C.R. 111, Sopinka J. further confirmed his approach to propensity evidence at p. 120:

Cross on Evidence (6th ed. 1985) contains a concise statement of the “similar facts rule” at p. 311 with which I agree:

. . . evidence of the character or of the misconduct of the accused on other occasions . . . tendered to show his bad disposition, is inadmissible unless it is so highly probative of the issues in the case as to outweigh the prejudice it may cause.

[67] The acknowledgement that similar fact evidence uses propensity as its mode of reasoning is also supported by Wigmore on Evidence, supra, at pp. 1152-53, and Sopinka, Lederman and Bryant, supra, at § 11.141.

[68] It follows, as stated by Charron J.A. in B. (L.), supra, at p. 57:

. . . propensity reasoning in and of itself is not prohibited. Indeed, it is usually inevitable, given the nature of the evidence and the reason for its admission. . . .

It is propensity reasoning that is based solely on the general bad character of the accused, as revealed through this evidence of discreditable conduct, which is prohibited.

See also R. v. Batte (2000), 2000 CanLII 5751 (ON CA), 34 C.R. (5th) 197 (Ont. C.A.), at p. 226, per Doherty J.A.

(2) Identification of the “Issue in Question” is an Important Control

[69] McLachlin J. speaks in B. (C.R.), supra, of the “value of the evidence in relation to an issue in question” (p. 732 (emphasis added)). McIntyre J., in Sweitzer, supra, emphasized that whether or not probative value exceeds prejudicial effect can only be determined in light of the purpose for which the evidence is proffered (p. 953). The importance of issue identification was also emphasized in D. (L.E.), supra, at p. 121; C. (M.H.), supra, at p. 771; R. v. Litchfield, 1993 CanLII 44 (SCC), [1993] 4 S.C.R. 333, at p. 358; R. v. B. (F.F.), 1993 CanLII 167 (SCC), [1993] 1 S.C.R. 697, at p. 731; R. v. Lepage, 1995 CanLII 123 (SCC), [1995] 1 S.C.R. 654, at para. 35; and Arp, supra, at para. 48.

[70] An indication of the importance of identifying “the issue in question” is that the trial judge is required to instruct the jury that they may use the evidence in relation to that issue and not otherwise.

[71] This Court has frequently gone out of its way to emphasize that the general disposition of the accused does not qualify as “an issue in question”. As stated, the similar fact evidence may be admissible if, but only if, it goes beyond showing general propensity (moral prejudice) and is more probative than prejudicial in relation to an issue in the crime now charged. I accept as correct the dictum of Lord Goddard C.J. in R. v. Sims, [1946] 1 All E.R. 697 (C.C.A.), at p. 700, that “[e]vidence is not to be excluded merely because it tends to show the accused to be of a bad disposition, but only if it shows nothing more”, provided the “something more” is taken to refer to an excess of probative value over prejudice. Thus, for example, in B. (F.F.), supra, the accused was charged with the sexual abuse of a young girl in his care. Similar fact evidence was led from the complainant’s brothers about physical abuse and the violent domination by the accused of the household. Iacobucci J., for the majority, stated, at p. 731:

. . . evidence which tends to show bad character or a criminal disposition on the part of the accused is admissible if (1) relevant to some other issue beyond disposition or character, and (2) the probative value outweighs the prejudicial effect. [Emphasis added.]

[72] Proof of general disposition is a prohibited purpose. Bad character is not an offence known to the law. Discreditable disposition or character evidence, at large, creates nothing but “moral prejudice” and the Crown is not entitled to ease its burden by stigmatizing the accused as a bad person. The defence of “innocent association” in B. (F.F.) was simply another way of expressing the denial by an accused of an element of the offence. The evidence of his prior discreditable conduct of a distinctive and particular nature, was considered to be strongly probative of specific issues in the case. Thus read, B. (F.F.) is quite consistent with B. (C.R.), and should not be interpreted as a rival “two-step” variant of the test.

[73] The requirement to identify the material issue “in question” (i.e., the purpose for which the similar fact evidence is proffered) does not detract from the probative value/prejudice balance, but is in fact essential to it. Probative value cannot be assessed in the abstract. The utility of the evidence lies precisely in its ability to advance or refute a live issue pending before the trier of fact.

[74] The issues in question derive from the facts alleged in the charge and the defences advanced or reasonably anticipated. It is therefore incumbent on the Crown to identify the live issue in the trial to which the evidence of disposition is said to relate. If the issue has ceased to be in dispute, as for example when the fact is admitted by the accused, then the evidence is irrelevant and it must be excluded: R. v. Clermont, 1986 CanLII 26 (SCC), [1986] 2 S.C.R. 131, at p. 136; R. v. Bosley (1992), 1992 CanLII 2838 (ON CA), 18 C.R. (4th) 347 (Ont. C.A.), at p. 360; R. v. Proctor (1992), 1992 CanLII 2763 (MB CA), 69 C.C.C. (3d) 436 (Man. C.A.), at p. 447; R. v. Hanna (1990), 1990 CanLII 11027 (BC CA), 57 C.C.C. (3d) 392 (B.C.C.A.); and B. (L.), supra, at p. 50. The relative importance of the issue in the particular trial may also have a bearing on the weighing up of factors for and against admissibility. Similar fact evidence that is virtually conclusive of a minor issue may still be excluded for reasons of overall prejudice.

[75] The “issues in question” are not, it should be emphasized, categories of admissibility. Their identification is simply an element of the admissibility analysis which, as stated, turns on weighing probative value against prejudice.

(3) Identification of the Required Degree of Similarity

[76] The principal driver of probative value in a case such as this is the connectedness (or nexus) that is established between the similar fact evidence and the offences alleged, particularly where the connections reveal a “degree of distinctiveness or uniqueness” (B. (C.R.), supra, at p. 735). As stated by Cory J. in Arp, supra, at para. 48:

. . . where similar fact evidence is adduced to prove a fact in issue, in order to be admissible, the trial judge should evaluate the degree of similarity of the alleged acts and decide whether the objective improbability of coincidence has been established. Only then will the evidence have sufficient probative value to be admitted.

[77] Thus in Arp, where the issue was identification, Cory J. cited at para. 43 R. v. Scopelliti (1981), 1981 CanLII 1787 (ON CA), 63 C.C.C. (2d) 481 (Ont. C.A.), where Martin J.A. observed that evidence of propensity on the issue of identification is not admissible “unless the propensity is so highly distinctive or unique as to constitute a signature” (p. 496). Martin J.A. made the propensity point again in his lecture on “Similar Fact Evidence” published in [1984] Spec. Lect. L.S.U.C. 1, at pp. 9-10, in speaking of the Moors Murderer case (R. v. Straffen, [1952] 2 Q.B. 911):

Although evidence is not admissible to show a propensity to commit crimes, or even crimes of a particular class, evidence of a propensity to commit a particular crime in a peculiar and distinctive way was admissible and sufficient to identify [Straffen] as the killer of the deceased. [Emphasis in original.]

[78] The issue in the present case is not identification but the actus reus of the offence. The point is not that the degree of similarity in such a case must be higher or lower than in an identification case. The point is that the issue is different, and the drivers of cogency in relation to the desired inferences will therefore not be the same. As Grange J.A. correctly pointed out 20 years ago in R. v. Carpenter (1982), 1982 CanLII 3308 (ON CA), 142 D.L.R. (3d) 237 (Ont. C.A.), at p. 244:

The degree of similarity required will depend upon the issues in the particular case, the purpose for which the evidence is sought to be introduced and the other evidence.

[79] If, for example, the complainant in this case had not been able to identify the respondent as the perpetrator of the alleged offence, the conduct described by the ex-wife was not so “peculiar and distinctive” as to amount to a “signature” or “fingerprints at the scene of the crime” that would safely differentiate him from other possible assailants.

[80] On the other hand, in a case where the issue is the animus of the accused towards the deceased, a prior incident of the accused stabbing the victim may be admissible even though the victim was ultimately shot – the accused says accidentally (Rosenberg, supra, at p. 8). The acts could be said to be dissimilar but the inference on the “issue in question” would nonetheless be compelling.

(4) Identification of Connecting Factors – Is the Similar Fact Evidence Appropriately Connected to the Facts Alleged in the Charge?

[81] The decided cases suggest the need to pay close attention to similarities in character, proximity in time and frequency of occurrence. Wigmore put it this way:

Since it is the improbability of a like result being repeated by mere chance that carried probative weight, the essence of this probative effect is the likeness of the instance. . . .

It is just this requirement of similarity which leaves so much room for difference of opinion, and accounts for the bewildering variances of rulings in the different jurisdictions and even in the same jurisdiction and in cases of the same offense.

(Wigmore on Evidence, vol. 2 (Chadbourn rev. 1979), at pp. 245-46)

See also: Arp, supra, at para. 44; R. v. Smith, 1992 CanLII 79 (SCC), [1992] 2 S.C.R. 915, at p. 941; and Cloutier v. The Queen, 1979 CanLII 25 (SCC), [1979] 2 S.C.R. 709, at pp. 730-31. Thus it was the required degree of similarity, or the lack of it, that divided the Court in B. (C.R.), supra, at pp. 739 and 753. Similarity in this respect does not necessarily require a strong peculiarity or unusual distinctiveness underlying the events being compared, although similar facts manifesting a singular trait (such as necrophilia) would likely be a powerful tool in the hands of the prosecution.

[82] The trial judge was called on to consider the cogency of the proffered similar fact evidence in relation to the inferences sought to be drawn, as well as the strength of the proof of the similar facts themselves. Factors connecting the similar facts to the circumstances set out in the charge include:

(1) proximity in time of the similar acts: D. (L.E.), supra, at p. 125; R. v. Simpson (1977), 1977 CanLII 1142 (ON CA), 35 C.C.C. (2d) 337 (Ont. C.A.), at p. 345; R. v. Huot (1993), 1993 CanLII 8652 (ON CA), 16 O.R. (3d) 214 (C.A.), at p. 220;

(2) extent to which the other acts are similar in detail to the charged conduct: Huot, supra, at p. 218; R. v. Rulli (1999), 1999 CanLII 3712 (ON CA), 134 C.C.C. (3d) 465 (Ont. C.A.), at p. 471; C. (M.H.), supra, at p. 772;

(3) number of occurrences of the similar acts: Batte, supra, at pp. 227-28;

(4) circumstances surrounding or relating to the similar acts (Litchfield, supra, at p. 358);

(5) any distinctive feature(s) unifying the incidents: Arp, supra, at paras. 43-45; R. v. Fleming (1999), 1999 CanLII 18921 (NL CA), 171 Nfld. & P.E.I.R. 183 (Nfld. C.A.), at paras. 104-5; Rulli, supra, at p. 472;

(6) intervening events: R. v. Dupras, 2000 BCSC 1128, [2000] B.C.J. No. 1513 (QL) (S.C.), at para. 12;

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

[83] On the other hand, countervailing factors which have been found helpful in assessing prejudice include the inflammatory nature of the similar acts (D. (L.E.), at p. 124) and whether the Crown can prove its point with less prejudicial evidence. In addition, as stated, the court was required to take into account the potential distraction of the trier of fact from its proper focus on the facts charged, and the potential for undue time consumption. These were collectively described earlier as moral prejudice and reasoning prejudice.

[84] This list is intended to be helpful rather than exhaustive. Not all factors will exist (or be necessary) in every case. A comparable approach is utilized in other common law jurisdictions, including England (see Director of Public Prosecutions v. Kilbourne, [1973] A.C. 729 (H.L.), at p. 758), and in the United States (see C. B. Mueller and L. C. Kirkpatrick, Federal Evidence (2nd ed. 1994 & Supp. 2001), vol. 2, at § 161; United States v. Enjady, 134 F.3d 1427 (10th Cir. 1998), certiorari denied, 525 U.S. 887 (1998)).

(5) Differentiating Admissible from Inadmissible Propensity Evidence

[85] Part of the conceptual problem with similar fact evidence is that words like “disposition” or “propensity” are apt to describe a whole spectrum of human character and behaviour of varying degrees of potential relevance. At the vague end of the spectrum, it might be said that the respondent has a general disposition or propensity “for violence”. This, by itself, proved nothing of value in this trial. The respondent was not charged with having a brutal personality, and his general character was, in that sense, irrelevant.

[86] At a more specific level, it is alleged here that the propensity to violence emerges in this respondent in a desire for hurtful sex. This formulation provides more context, but the definition of so general a propensity is still of little real use, particularly when it is sought to use “propensity” not to predict future conduct in a general way, but to conclude that the respondent is guilty of acting in the specific way under the specific circumstances on December 6, 1996 alleged by this complainant.

[87] Cogency increases as the fact situation moves further to the specific end of the spectrum. In Lepage, supra, the accused was charged with possession of LSD for the purpose of trafficking in narcotics. The trial judge admitted evidence of a roommate that the accused was a “major [drug] dealer in the house” and on that basis expressed the opinion that the drugs belonged to the accused. On appeal, Sopinka J., writing for the majority, ruled that the evidence was admissible on the issue of possession, at paras. 36-37:

In the present case, the testimony of Thelland is not merely relevant to the character of the respondent, but is also relevant to possession which is a key issue in the case. In the circumstances of this case, there were three people living in the house and it was clear that the drugs belonged to one of the three. Surely, it is relevant to the issue of possession to have one of the three testify that the drugs were not his and furthermore, indicate that the respondent is in the business and therefore it is more likely that he was the owner of the drugs.

The evidence is not being adduced solely for the purpose of showing that the respondent is likely to have committed the crime because he is the type of person who would be likely to possess drugs.

[88] The issue to which the evidence was relevant was possession (not just character or propensity), but the housemate’s evidence derived its cogency from what it said about the character or disposition of the accused, as the dissenters, per Major J., pointed out, at para. 55:

In Justice Sopinka’s view, the evidence of Thelland was relevant not solely to character, but also to possession, in that someone in the business of dealing narcotics had more opportunity and was more likely to be in possession of narcotics. With respect, this is evidence of propensity to deal in drugs, and nothing more.

[89] The difference between the majority and the minority, it seems, was that the majority considered the connections or correspondence between the act charged and the prior acts of possession (what McCormick on Evidence (5th ed. 1999), vol. 1, at p. 687, calls “situation-specific behavior”) sufficiently compelling to draw safely the inference of possession on the facts charged, whereas the minority considered any linkage to be so general as to have no probative value with respect to the particular facts of the offence on the particular date charged.

[90] On the facts of B. (C.R.), the majority concluded that the accused was shown to have a situation specific propensity to abuse sexually children to whom he stood in parental relationship, and there was a close match between the “distinct and particular” propensity demonstrated in the similar fact evidence and the misconduct alleged in the charge, although even the majority considered the admissibility to be “borderline” (p. 739). Similar fact evidence is sometimes said to demonstrate a “system” or “modus operandi”, but in essence the idea of “modus operandi” or “system” is simply the observed pattern of propensity operating in a closely defined and circumscribed context.

[91] References to “calling cards” or “signatures” or “hallmarks” or “fingerprints” similarly describe propensity at the admissible end of the spectrum precisely because the pattern of circumstances in which an accused is disposed to act in a certain way are so clearly linked to the offence charged that the possibility of mere coincidence, or mistaken identity or a mistake in the character of the act, is so slight as to justify consideration of the similar fact evidence by the trier of fact. The issue at that stage is no longer “pure” propensity or “general disposition” but repeated conduct in a particular and highly specific type of situation. At that point, the evidence of similar facts provides a compelling inference that may fill a remaining gap in the jigsaw puzzle of proof, depending on the view ultimately taken (in this case) by the jury.

[92] This view also seems to have taken hold in Australia where the High Court, relying in part on McLachlin J.’s judgment in B. (C.R.), made the following observations in Pfennig v. R. (1995), 127 A.L.R. 99, at p. 115:

Thus, evidence of mere propensity, like evidence of a general criminal disposition having no identifiable hallmark, lacks cogency yet is prejudicial. On the other hand, evidence of a particular distinctive propensity demonstrated by acts constituting particular manifestations or exemplifications of it will have greater cogency, so long as it has some specific connection with or relation to the issues for decision in the subject case. That evidence, as has been said, will be admissible only if its probative value exceeds its prejudicial effect. [Emphasis added.]

[93] I note parenthetically that the court then added a “conclusiveness” criteria which in my view ought to be rejected as too great an intrusion by the trial judge in the fact finding mandate of the jury. This is considered in greater detail below.

(6) Similar Fact Evidence Need Not Be Conclusive

[94] Some authorities urge adoption of a further refinement that has been accepted in some common law jurisdictions in the balancing of prejudice against probative value, namely that similar fact evidence should only be admitted if its probative value is so great as to be virtually conclusive of guilt. As Lord Cross put it in Boardman, supra, at p. 457:

The question must always be whether the similar fact evidence taken together with the other evidence would do no more than raise or strengthen a suspicion that the accused committed the offence with which he is charged or would point so strongly to his guilt that only an ultra-cautious jury . . . would acquit in face of it.

[95] The High Court of Australia has ruled that unless the similar fact evidence, taken together with the other evidence in the case, would be consistent with guilt and with no other conclusion, it ought to be rejected. In Pfennig, the lead judgment authored by Mason C.J., Deane and Dawson JJ., stated at p. 116:

. . . that propensity evidence is circumstantial evidence and that, as such, it should not be used to draw an inference adverse to the accused unless it is the only reasonable inference in the circumstances. More than that, the evidence ought not to be admitted if the trial judge concludes that, viewed in the context of the prosecution case, there is a reasonable view of it which is consistent with innocence.

[96] The test is a variant of the rule generally applicable to circumstantial evidence laid down in Hodge’s Case (1838), 1838 CanLII 1 (FOREP), 2 Lewin 227, 168 E.R. 1136, i.e., that the circumstances must be consistent with the conclusion of guilt and inconsistent with any other rational conclusion. The difference, of course, is that we are dealing here with admissibility, not adjudication. The conclusiveness test does not sit well with the balancing model set out in B. (C.R.). If the evidence were truly “conclusive”, its probative value would ex hypothesi outweigh its prejudice: Pfennig, supra, at p. 138.

[97] In my view, the “conclusiveness” test takes the trial judge’s “gatekeeper” function too far into the domain of the trier of fact.

F. Application of the Test to the Facts of this Case

[98] I proceed to apply the test for similar fact evidence to the facts of this case under the following headings:

(1) The Probative Value of the Evidence

[99] Under this heading it is necessary first to determine the precise “issue in question” for which the Crown seeks to adduce the similar fact evidence. I will then address the cogency of the similar fact evidence in relation to that particular question. This will require consideration of the various connecting factors which the Crown considers persuasive, together with those factors which the defence regards as fatally weakening the inferences desired by the prosecution. An important element of the probative weight analysis is the issue of potential collusion between the complainant and the ex-wife. I agree with the respondent that it was part of the trial judge’s “gatekeeper” function to consider this issue because collusion, if established to the satisfaction of the trial judge on a balance of probabilities, would be destructive of the very basis on which the similar fact evidence was sought to be admitted, namely the improbability that two women would independently concoct stories with so many (as the Crown contends) similar features.

(2) Assessment of the Prejudice

[100] Under this heading, it is necessary to evaluate both moral prejudice (i.e., the potential stigma of “bad personhood”) and reasoning prejudice (including potential confusion and distraction of the jury from the actual charge against the respondent). Of importance in this respect is the inflammatory nature of the sexual and domestic abuse alleged by the ex-wife, and the need for the jury to keep separate consideration of the seven “similar fact” incidents from the only charge they were asked to decide, the sexual assault alleged by the complainant.

(3) Weighing up Probative Value Versus Prejudice

[101] The starting point, of course, is that the similar fact evidence is presumptively inadmissible. It is for the Crown to establish on a balance of probabilities that the likely probative value will outweigh the potential prejudice.

(1) The Probative Value of the Evidence

[102] The issue at this stage is to determine whether the similar fact evidence is indeed strong enough to be capable of properly raising in the eyes of the jury the double inferences contended for by the Crown.

[103] The respondent disputes the probative value of the similar fact evidence on three principal grounds: firstly, the opportunity for collusion between his ex-wife and the complainant; secondly, the dissimilarities of the so-called similar facts; and thirdly, the frailties of the ex-wife’s evidence.

(a) The Potential for Collusion

[104] I mention this issue at the outset because if collusion is present, it destroys the foundation on which admissibility is sought, namely that the events described by the ex-wife and the complainant, testifying independently of one another, are too similar to be credibly explained by coincidence. 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 added.]

[105] The gatekeeper function was similarly dealt with by Cory J. in Arp, supra, at paras. 47-48:

. . . in determining the admissibility of similar fact evidence the trial judge must, to a certain extent, invade this province [of the jury]. As Professor Smith stated in Case and Comment on R. v. Hurren, [1962] Crim. L. Rev. 770, at p. 771:

It should be noted that judges commonly distinguish facts as going to weight rather than admissibility (see, e.g., R. v. Wyatt); but it is submitted that, as regards similar fact evidence, no sharp line can be drawn and that admissibility depends on weight.

Thus, where similar face evidence is adduced to prove a fact in issue, in order to be admissible, the trial judge should evaluate the degree of similarity of the alleged acts and decide whether the objective improbability of coincidence has been established. Only then will the evidence have sufficient probative value to be admitted. [Emphasis added.]

[106] In my view, the trial judge cannot assess “the objective improbability of coincidence” without addressing the issue of whether the apparent “coincidence” is in fact the product of collusion. Admissibility is a question of law for the judge alone. I agree with the respondent that “[i]t is only through an accurate and full preliminary assessment of probative value that prejudice can be kept within manageable bounds”.

[107] The trial judge held that he ought not to reach even a preliminary view of the likelihood of collusion. In this, he may have been influenced by the decision of the House of Lords in R. v. H., [1995] 2 A.C. 596, where it was held that for purposes of admissibility the proffered similar fact evidence ought to be accepted “as true” (p. 611) in all but very exceptional cases. The question of weight should, on this view, be left wholly to the jury.

[108] A different position was advocated by Sopinka J. in dissent in B. (C.R.), supra, at p. 752, namely that the Crown “must negative conspiracy or collaboration in accordance with the criminal standard” (i.e., beyond a reasonable doubt). See also R. v. Kenny (1996), 1996 CanLII 6633 (NL CA), 108 C.C.C. (3d) 349 (Nfld. C.A.), at p. 359.

[109] An intermediate position was proposed by Charron J.A. in this case, namely that the possibility of collusion is merely “a factor” to be considered on the issue of admissibility. In doing so, she followed her own ruling in B. (L.), which in turn was followed in R. v. McDonald (2000), 2000 CanLII 16871 (ON CA), 148 C.C.C. (3d) 273 (Ont. C.A.), by another panel of that court differently constituted. To the extent that treating collusion as a “factor” is intended merely to emphasize the overall framework of the probative value versus prejudice balance, I agree with it.

[110] I would not agree, however, that suspected collusion would play less strongly against otherwise powerful evidence than in a borderline case. In that sense, suspected collusion is more than just another “factor”. Cogency is derived from the improbability of coincidence. Collusion is a factor, yes, but more than that it is a crucial factor because the existence of collusion rebuts the premise on which admissibility depends.

[111] Charron J.A. found, and I agree, that there was an issue of potential collusion between the complainant and the ex-wife. The evidence went beyond mere “opportunity”, which will be a feature in many cases alleging sexual abuse with multiple complainants. The issue is concoction or collaboration, not contact. If the evidence amounts to no more than opportunity, it will usually best be left to the jury. Here there is something more. It is the whiff of profit. The ex-wife acknowledged that she had told the complainant of the $16,500 she received from the Criminal Injuries Compensation Board on the basis, she agreed, that “[a]ll you had to do was say that you were abused”. A few days later the complainant, armed with this information, meets the respondent and goes off with him to have sex in a motel room.

[112] The Court in Arp, supra, concluded that the test for the admission of similar fact evidence is based on probability rather than reasonable doubt (paras. 65, 66 and 72). Accordingly where, as here, there is some evidence of actual collusion, or at least an “air of reality” to the allegations, the Crown is required to satisfy the trial judge, on a balance of probabilities, that the evidence of similar facts is not tainted with collusion. That much would gain admission. It would then be for the jury to make the ultimate determination of its worth.

[113] Here it was not sufficient for the Crown simply to proffer dicey evidence that if believed would have probative value. It was not incumbent on the defence to prove collusion. It was a condition precedent to admissibility that the probative value of the proffered evidence outweigh its prejudicial effect and the onus was on the Crown to satisfy that condition. The trial judge erred in law in deferring the whole issue of collusion to the jury.

[114] While that error of law is sufficient to affirm the need for a new trial as ordered by the Court of Appeal, I proceed to examine the other elements of the test previously described.

(b) Identification of “the Issue in Question”

[115] The Crown says the issue generally is “the credibility of the complainant” and more specifically “that the accused has a strong disposition to do the very act alleged in the charges against him”, but this requires some refinement. Care must be taken not to allow too broad a gateway for the admission of propensity evidence or, as it is sometimes put, to allow it to bear too much of the burden of the Crown’s case (Sopinka, Lederman and Bryant, supra, at § 11.26). Credibility is an issue that pervades most trials, and at its broadest may amount to a decision on guilt or innocence.

[116] Anything that blackens the character of an accused may, as a by-product, enhance the credibility of a complainant. Identification of credibility as the “issue in question” may, unless circumscribed, risk the admission of evidence of nothing more than general disposition (“bad personhood”).

[117] Moreover, broadly speaking, the non-consent of the ex-wife on the different occasions described in her evidence is of no relevance to whether the complainant here consented or not: Clermont, supra, at p. 135. Because complainant A refused consent in 1992 scarcely establishes that complainant B refused consent in 1996.

[118] A conviction for sexual assault requires proof beyond reasonable doubt of two basic elements, that the accused committed the actus reus and that he had the necessary mens rea. The actus reus of assault is unwanted sexual touching. The mens rea is the intention to touch, knowing of, or being reckless of, or wilfully blind to, a lack of consent: R. v. Ewanchuk, 1999 CanLII 711 (SCC), [1999] 1 S.C.R. 330, at para. 23.

[119] The respondent admits that sexual touching took place and that he intended it. He denies that it was unwanted. He therefore puts in issue the consent element of the actus reus: Ewanchuk, supra, at para. 27. Is he to be believed when he says consent was never withdrawn, or is the prosecution correct that he has a demonstrated situation-specific propensity to proceed regardless, indeed to derive heightened pleasure from being rejected and forcing sex on his sex partner? If so, was it manifested in this case?

[120] If the jury could legitimately infer sexual intransigence in closely comparable circumstances from the respondent’s past behaviour and refusal to take his wife’s no for an answer, the present complainant’s testimony that intercourse occurred despite her lack of consent gains in credibility. The issue broadly framed is credibility, but more accurately and precisely framed, the “issue in question” in this trial was the consent component of the actus reus and in relation to that issue the respondent’s alleged propensity to refuse to take no for an answer.

(c) Similarities and Dissimilarities Between the Facts Charged and the Similar Fact Evidence

[121] I propose to assess the evidence in light of the relevant “connecting factors” listed above at para. 82. I repeat that not every factor is useful in every case, and that cogency also depends on the other evidence.

(i) Proximity in Time of the Similar Acts

[122] Lapse of time opens up a greater possibility of character reform or “maturing out” personality change, and would tend to undermine the premise of continuity of character or disposition. Remoteness in time may also affect relevance and reliability. The charge against the respondent relates to December 6, 1996. The ex-wife’s seven alleged incidents occurred between March 1990 and October 1996, interrupted by the respondent’s incarceration from 1992 to 1995. The evidence of the respondent’s inability to take no for an answer gains cogency both from its repetition over many years and its most recent manifestation a couple of months before the offence charged.

(ii) Extent to Which the Other Acts Are Similar in Detail to the Charged Conduct

[123] In this case, in my view, with respect, the learned trial judge paid insufficient attention to the dissimilarities.

[124] At least one of the incidents is largely irrelevant. Incident five involved choking, did not demonstrate sexual misconduct and was not remotely connected to the factual allegations in the charge. While the Crown can legitimately argue for the cumulative effect of a string of “similar” facts, I think an incident so remote from the charge could do nothing but blacken the respondent’s character in a general way. Conduct that is so dissimilar or equivocal does not raise an inference capable of overcoming the prejudice.

[125] There are other important dissimilarities. None of the incidents described by the ex-wife began as consensual, then allegedly became non-consensual. Each of the incidents recounted by the ex-wife were bound up with the intimacy of a long-term relationship. Incident one relates to premature sex after birth of their child. Incident five arose out of expressions of jealousy by one conjointe to another. Incident seven followed a death in the family. The dynamic of these situations is not the same as the motel scene, although it is true that they all did lead (apart from incident five) to the respondent’s refusal to accept his ex-wife’s rejection of his sexual demands.

[126] Incident two (where the ex-wife’s initial concern was based on being in close proximity to her sister and brother-in-law in a trailer) bears no obvious similarities, although again, the respondent’s aggression seemed to be heightened by his ex-wife’s resistance.

[127] It should be repeated that the search for similarities is a question of degree (Boardman, supra, at p. 442, per Lord Wilberforce). Sexual activity may not show much diversity or distinctiveness. Not every dissimilarity is fatal, but for the reasons already mentioned, substantial dissimilarities may dilute probative strength and, by compounding the confusion and distraction, aggravate the prejudice.

(iii) Number of Occurrences of the Similar Acts

[128] An alleged pattern of conduct may gain strength in the number of instances that compose it. The cogency of the similar act evidence in the “brides in the bathtub” case undoubtedly gathered strength from the fact the charge related to the third victim who had died under identical circumstances to her two predecessors: R. v. Smith (1915), 84 L.J.K.B. 2153 (C.C.A.). The ex-wife’s evidence here, if believed, established a pattern over many years that the jury might think showed that the respondent’s pleasure in not taking no for an answer in sexual encounters was a predictable characteristic of general application.

(iv) Circumstances Surrounding or Relating to the Similar Acts

[129] Perhaps the most important dissimilarity, as Charron J.A. points out, lies not in the acts themselves but in the broader context. The “similar fact” evidence occurred in the course of a long-term dysfunctional marriage whereas the charge relates to a one-night stand following a chance meeting of casual acquaintances in a bar.

[130] The ex-wife admitted in her testimony that, as one would expect, there were numerous periods of consensual sex during their relationship. They produced three children. She testified that the alleged abuse did not begin until after she and the respondent were married, at which time their relationship demonstrated many complexities that have no parallel with the situation in which the complainant found herself. To what extent was the respondent’s behaviour with his ex-wife an incident of a particular conjugal relationship and to what extent did it reflect a propensity to deal in a certain way with casual sex partners, including the complainant? To what extent can “common sense” be safely relied upon to answer this question? With what confidence can the necessary inferences be drawn? There is no satisfactory answer to these basic questions in this record.

(v) Any Distinctive Feature(s) Unifying the Incidents

[131] It is not alleged that the sex acts themselves or the surrounding circumstances were highly distinctive. Cogency was said to derive from repetition rather than distinctiveness.

(vi) Intervening Events

[132] If the similar facts were sufficient to raise the inferences suggested by the Crown, there were no “intervening events” as such to undermine their probative value. An example (not applicable here) might be evidence of supervening physical incapacity: R. v. Minhas (1986), 1986 CanLII 144 (ON CA), 29 C.C.C. (3d) 193 (Ont. C.A.), at p. 219.

(d) Strength of the Evidence that the Similar Acts Actually Occurred

[133] The respondent did not admit the prior misconduct, and (quite apart from the issue of collusion) a vigorous attack was made in cross-examination on the ex-wife’s credibility. The evidence relating to incident six, for example, was said to be confused and contradictory. The ex-wife initially told the police that the alleged assault only involved vaginal sex. Her evidence on this incident subsequently varied. At the preliminary hearing, she testified that there was anal intercourse but did not mention vaginal intercourse. (An incomplete trial with respect to her allegations was held in April 1998 where she repeated her allegations of anal sex but, contrary to the initial testimony, did not mention vaginal sex.) At both the voir dire and the trial in this case, she testified that the respondent assaulted her vaginally and anally.

[134] 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.

[135] I conclude that the similar fact evidence, if admitted, is certainly capable of raising the first inference, namely that the respondent derived pleasure from sex that was painful to his ex-wife and would not take no for an answer. The second inference (that he proceeded wilfully in this case knowing the complainant did not consent) is a good deal more problematic, for the reasons mentioned.

[136] 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. In this case, the issues were fully argued and I therefore go on to the next stage.

(2) Assessment of the Prejudice

[137] The principal wellsprings of prejudice flowing from propensity evidence were described above in outlining its presumptive exclusion, and there is no need to repeat those worries here.

[138] The poisonous potential of similar fact evidence cannot be doubted. Sopinka, Lederman and Bryant, supra, at § 11.173, refer to the observations of an English barrister who has written of that jurisdiction:

Similar fact evidence poses enormous problems for Judges, jurors and magistrates alike. The reason for this is the headlong conflict between probative force and prejudicial effect. Often, in the Crown Court, it is as close as a Judge comes to singlehandedly deciding the outcome of a case. [Emphasis added.]

(G. Durston, “Similar Fact Evidence: A Guide for the Perplexed in the Light of Recent Cases” (1996), 160 Justice of the Peace & Local Government Law 359, at p. 359)

Canadian trial lawyers take the same view.

(a) Moral Prejudice

[139] It is frequently mentioned that “prejudice” in this context is not the risk of conviction. It is, more properly, the risk of an unfocussed trial and a wrongful conviction. The forbidden chain of reasoning is to infer guilt from general disposition or propensity. The evidence, if believed, shows that an accused has discreditable tendencies. In the end, the verdict may be based on prejudice rather than proof, thereby undermining the presumption of innocence enshrined in ss. 7 and 11(d) of the Canadian Charter of Rights and Freedoms.

[140] The inflammatory nature of the ex-wife’s evidence in this case cannot be doubted. It is, to the extent these things can be ranked, more reprehensible than the actual charge before the court. The jury would likely be more appalled by the pattern of domestic sexual abuse than by the alleged misconduct of an inebriated lout in a motel room on an isolated occasion. It may be noted that s. 718.2 of the Criminal Code, R.S.C. 1985, c. C-46, reflects society’s denunciation of spousal abuse by making such abuse an aggravating factor for the purposes of sentencing.

[141] Some model studies of jury behaviour have put into question the effectiveness of the trial judge’s instruction as to the limited use that may be made of propensity evidence: R. L. Wissler and M. J. Saks, “On the Inefficacy of Limiting Instructions: When Jurors Use Prior Conviction Evidence to Decide on Guilt” (1985), 9 Law & Hum. Behav. 37, at p. 43; S. Lloyd-Bostock, “The Effects on Juries of Hearing About the Defendant’s Previous Criminal Record: A Simulation Study”, [2000] Crim. L.R. 734, at p. 742; and K. L. Pickel, “Inducing Jurors to Disregard Inadmissible Evidence: A Legal Explanation Does Not Help” (1995), 19 Law & Hum. Behav. 407. This is not to undermine our belief in the ability of the jury to do its job, but it underlines the poisonous nature of propensity evidence, and the need to maintain a high awareness of its potentially prejudicial effect.

[142] To some extent, the prejudice could be contained by limiting the extent and nature of the ex-wife’s evidence, even if some of it were admitted, by a process analogous to that followed in R. v. Corbett, 1988 CanLII 80 (SCC), [1988] 1 S.C.R. 670, with respect to criminal convictions. That approach was adopted here only to the limited extent that the fact of the respondent’s jail time for two sexual assaults on other parties was suppressed by agreement of counsel.

[143] I conclude that this evidence has a serious potential for moral prejudice.

(b) Reasoning Prejudice

[144] The major issue here is the distraction of members of the jury from their proper focus on the charge itself aggravated by the consumption of time in dealing with allegations of multiple incidents involving two victims in divergent circumstances rather than the single offence charged.

[145] Distraction can take different forms. In R. v. D. (L.E.) (1987), 1987 CanLII 2536 (BC CA), 20 B.C.L.R. (2d) 384 (C.A.), McLachlin J.A. (as she then was) observed at p. 399 that the similar facts may induce

in the minds of the jury sentiments of revulsion and condemnation which might well deflect them from the rational, dispassionate analysis upon which the criminal process should rest.

[146] Further, there is a risk, evident in this case, that where the “similar facts” are denied by the accused, the court will be caught in a conflict between seeking to admit what appears to be cogent evidence bearing on a material issue and the need to avoid unfairness to the right of the accused to respond. The accused has a limited opportunity to respond. Logistical problems may be compounded by the lapse of time, surprise, and the collateral issue rule, which will prevent (in the interest of effective use of court resources) trials within trials on the similar facts. Nor is the accused allowed to counter evidence of discreditable conduct with similar fact evidence in support of his or her credibility (as discussed in Sopinka, Lederman and Bryant, supra, at § 11.74). Thus the practical realities of the trial process reinforce the prejudice inherent in the poisonous nature of the propensity evidence itself.

[147] In my view, the evidence of the ex-wife had the potential to create, in addition to moral prejudice, significant reasoning prejudice at the respondent’s trial.

(3) Weighing Up Probative Value Versus Prejudice

[148] One of the difficulties, as McHugh J. pointed out in Pfennig, supra, at p. 147, is the absence of a common basis of measurement: “The probative value of the evidence goes to proof of an issue, the prejudicial effect to the fairness of the trial.” The two variables do not operate on the same plane.

[149] As probative value advances, prejudice does not necessarily recede. On the contrary, the two weighing pans on the scales of justice may rise and fall together. Nevertheless, probative value and prejudice pull in opposite directions on the admissibility issue and their conflicting demands must be resolved.

[150] In Director of Public Prosecutions v. P., [1991] 2 A.C. 447 (H.L.), at p. 460, Lord Mackay suggested that similar fact evidence should be admitted when its probative value is “sufficiently great to make it just to admit the evidence”, notwithstanding its prejudicial value. Lord Wilberforce in Boardman, at p. 442, also referred to “the interests of justice”. See also Pfennig, supra, at pp. 147-48. Justice is achieved when relevant evidence whose prejudice outweighs any probative value is excluded (R. v. Marquard, 1993 CanLII 37 (SCC), [1993] 4 S.C.R. 223, at p. 246) and where evidence whose probative value exceeds its prejudice (albeit an exceptional circumstance) is admitted. Justice includes society’s interest in getting to the truth of the charges as well as the interest of both society and the accused in a fair process. A criminal justice system that has suffered some serious wrongful convictions in part because of misconceived notions of character and propensity should not (and does not) take lightly the dangers of misapplied propensity evidence.

[151] In this case, the similar fact evidence was prima facie inadmissible and I agree with Charron J.A. that the Crown did not discharge the onus of establishing on a balance of probabilities that its probative value outweighed its undoubted prejudice. The probative value of the evidence, especially with respect to potential collusion, was not properly evaluated. The potential of such evidence for distraction and prejudice was understated. The threshold for admission of this sort of evidence was set too low.

[152] Consent, or the lack of it, and the complainant’s credibility in relation thereto, was the crucial issue at the trial. It can hardly be doubted that the jury, listening to the ex-wife’s evidence, would form a very low opinion of the respondent as an individual who behaved abominably towards his wife, and be readier on that account to believe the worst of him in his conduct towards the complainant. This is precisely the sort of general disposition reasoning (moral prejudice) that the similar fact exclusion rule was designed to prevent.

G. Review of the Trial Judge’s Decision

[153] A trial judge has no discretion to admit similar fact evidence whose prejudicial effect outweighs its probative value. Nevertheless, a trial judge’s decision to admit similar fact evidence is entitled to substantial deference: B. (C.R.), supra, at p. 739; and Arp, supra, at para. 42. In this case, however, quite apart from the other frailties of the similar fact evidence previously discussed, the trial judge’s refusal to resolve the issue of collusion as a condition precedent to admissibility was an error of law. A new trial is required.

IV. Conclusion

[154] The Crown’s appeal is dismissed.

Appeal dismissed.

Solicitor for the appellant: The Ministry of the Attorney General for Ontario, Toronto.

Solicitors for the respondent: Richard N. Stern and David E. Harris, Toronto.