R. v. Clarke

R. v. Clarke Collection Decisions of the Court of Appeal Date 1998-09-09 Docket numbers C25825 Judges McMurtry, Roy; Labrosse, Jean-Marc; Rosenberg, Marc Subject Criminal Decision Content DATE: 19980909 DOCKET: C25825 COURT OF APPEAL FOR ONTARIO McMURTRY C.J.O., LABROSSE and ROSENBERG JJ.A. BETWEEN: ) ) Trevor Shaw, HER MAJESTY THE QUEEN ) for the appellant ) (Appellant) ) ) and ) James McGrath, ) for the respondent HUBERT EDWARD CLARKE ) ) (Respondent) ) ) Heard: May 4, 1998 ) ROSENBERG J.A.: [1] The rules of evidence respecting the admissibility of character evidence are not always logical or founded in good policy. In this Crown appeal, the court is invited to reconsider the rationality of one of the more anomalous rules. [2] In R. v. Gonzague (1983), 4 C.C.C. (3d) 505, this court held that established authority permits the following line of inquiry of a witness called by the defence to give evidence about the character of a Crown witness: 1. Do you know the reputation of the witness as to truth and veracity in the community in which the witness resides? If the answer is “yes” the questioning proceeds. 2. Is that reputation good or bad? If the answer is “bad” a final question is permitted. 3. From that reputation, would you believe the witness on oath? [3] On behalf of the Crown, Mr. Shaw submits that the rule permitting the third question should be abrogated. He also urges this court to reassess whether the defence should automatically be entitled to ask the first two questions without an assessment of the probative value of the proffered testimony. Finally, Mr. Shaw argues that in any event the charge to the jury in this case as to the use of the evidence was inadequate. [4] I agree with Mr. Shaw that it is time to reassess this rule. I would hold that a judge has a limited discretion to prevent counsel from asking the first two questions of a witness called to attack the credibility of another witness. I would also hold that only rarely should the third question be permitted. I also agree that the charge to the jury in this case was, to some extent, inadequate. However, in the circumstances I am not satisfied that either the admission of this evidence or the impugned jury instructions could have affected the verdict. Accordingly, I would dismiss the appeal. THE FACTS [5] As a result of an incident on August 26, 1995, the respondent was charged with three offences: forcible seizure, assault with a weapon and possession of a weapon for the purpose of committing forcible seizure. He was tried by Sedgewick J. and a jury and acquitted on all three counts. [6] The respondent was a 60-year-old widower at the time of the alleged offences. He met the complainant in the spring of 1995 and the two dated for several months. The complainant broke off the relationship when she learned the respondent was seeing another woman. The respondent made an unsuccessful attempt at reconciliation. The day before the incident, the complainant wrote a letter to the respondent. In the letter, she asked him not to call or come to see her any more. She left the letter in his mailbox. When he returned from work the respondent’s son handed him the letter. The respondent was not pleased with having received the letter and intended to confront the complainant and give it back to her. Later that night and into the early morning hours of the following day there were a number of incidents in which the respondent attempted to speak to the complainant. [7] The respondent was able to speak to the complainant when they met in her apartment sometime after 1:00 a.m. The two then left the apartment. The charges against the respondent arise out of the confrontation between the complainant and the respondent at the complainant’s van, which was parked outside her apartment building. [8] The complainant testified that she was able to get into her van and lock it without interference by the respondent. She then agreed to talk to the respondent and lowered her window for that purpose. Once the window was lowered, the respondent was able to reach in and unlock the van. He then opened the door and pulled the complainant out. The respondent took her keys and purse and attempted to force her in to his car, which was parked nearby. In resisting these attempts the complainant bumped her head. At one point, the respondent had his arm around her throat. He then reached down into his pocket and removed an object, which he then opened and put under her chin. The complainant believed the object to be a knife or nail file. The respondent then stopped and the complainant began to cry. [9] The respondent gave a very different version. He testified that the two began to talk outside the complainant’s van. She told the respondent that she did not love him anymore. She began to cry and he attempted to comfort her. She then started to leave, apparently forgetting that she had left her purse on the hood of the van. The respondent retrieved the purse and again tried to calm her down. The respondent denied that the complainant asked him to leave. He denied striking the complainant or putting a knife to her throat. [10] At about 4:00 a.m., two police officers arrived to investigate another matter. The officers noticed that the complainant was upset and one of them took her aside. The complainant told the officer about the respondent forcing her into his car and his use of the knife. The respondent was arrested and searched. No knife was found at that time, but a small knife was later found in the back of the police cruiser used to transport the respondent to the police station. Since the cruiser had not been searched before the respondent had been placed in it, the officers could not be sure that the respondent put it there. According to the police, when they showed the respondent the knife, he confirmed that it was his. At trial, the respondent admitted telling the police that he had a knife that looks the one they found. However, he testified that the knife was not his. THE TRIAL JUDGE’S RULING [11] The events in this case took place in the City of Trenton. It appears that the complainant and the respondent were members of the small Caribbean community in that city and the surrounding towns. Defence counsel proposed to call five witnesses to testify about the reputation of the respondent and the complainant. Crown counsel objected to the admissibility of the evidence. She argued that the “community” was small and transient and so the witnesses would not know the reputation of the complainant and the respondent in that community. She argued that the prejudicial effect of the proposed evidence substantially outweighed its probative value. Finally, she argued that witnesses should not be permitted to give their opinion as to whether another witness is telling the truth. [12] The trial judge held that the witnesses would be permitted to give their opinions as to the reputation of the respondent and the complainant in the “community”, which he described as the City of Trenton. Further, based on established authority, defence counsel could impeach the credibility of the complainant by asking these questions: Do you know the reputation of the witness as to truth and veracity in the community in which he or she resides? Is that reputation good or bad? From that reputation, would you believe him or her on oath? [13] In addition, the character witnesses could be asked whether they would believe the witnesses based on their personal opinion. During examination-in-chief, the witnesses would not be permitted to give the basis for their belief. [14] The evidence of the five witnesses can be briefly summarized. Dorothy McKillup, a friend of the respondent, testified that the respondent had a good reputation for truth and veracity and from that reputation she would believe the respondent under oath. Ms. McKillup testified that the complainant had a bad reputation and she would not believe her under oath. [15] Kathy Taylor, the complainant’s sister, blamed the complainant for the break-up of her marriage. She testified that the complainant had a bad reputation for truthfulness and based on that reputation she could not be believed under oath. [16] Milton Valentine had known the respondent for many years as an active member of the Caribbean community. He testified that the respondent had a good reputation for truthfulness and would believe him under oath. Mr. Valentine also knew the complainant and in fact had been sued by her and convicted of assaulting her. He testified that the complainant had a bad reputation for truthfulness and should not be trusted. However, he testified that he did not know if she could be believed when placed under oath. [17] Two other witnesses testified about the respondent’s reputation for truthfulness. They did not give any evidence about the complainant. [18] In the charge to the jury, the trial judge gave instructions as to the use of the character evidence, as will be seen below. THE ISSUES [19] Mr. Shaw used the labels “oath-helping” and “oath-attacking” to describe the testimony given by the five character witnesses called by the defence. Using his terminology, this appeal raises these questions for determination with respect to such testimony. I. Is an accused entitled to lead “oath-helping” evidence before his own character has been attacked? II. Is an accused entitled to lead “oath-attacking” evidence and, if so, in what circumstances? III. Was the charge to the jury adequate? IV. Has the Crown met the burden of showing that the verdict would not necessarily have been the same if any errors were made? ANALYSIS I. Oath-helping evidence [20] As I read his ruling, the trial judge held that defence counsel was entitled to ask the character witnesses whether the complainant’s reputation for truth or veracity was good or bad. Only if the character witness answered “bad” was counsel entitled to proceed further. However, defence counsel asked four of the character witnesses about the respondent’s reputation for truth and veracity. Having answered that the reputation was good, the witnesses then gave the opinion that they would believe the respondent under oath. [21] The general rule of evidence is that a party may not bolster the character of his or her witness until the opposite party has attempted to impeach the witness’ character. So-called oath-helping evidence is inadmissible. See R. v. Béland, [1987] 2 S.C.R. 398 at 405-408 and R. v. B. (F.F.), [1993] 1 S.C.R. 697 at 729-30. There are some exceptions and, for example, within limits the parties are permitted to present their witnesses in the best allowable light. Moreover, the law has always permitted the defence to call character witnesses to testify to the good character of the accused, whether or not the Crown has expressly attempted to impeach the character of the accused. Such evidence is relevant to whether the accused is likely to have committed the offence charged and, if the accused has testified, whether he or she is telling the truth. See: R. v. Tarrant (1981), 34 O.R. (2d) 747 (C.A.). Ordinarily, non-expert witnesses may only provide evidence of the good character of the accused by testifying as to the good reputation of the accused in the community. [1] The rules, as I understand them, do not usually permit the witnesses to provide their personal opinion about the accused’s character or testify in chief to particular incidents. [22] We were not referred to any authority that permits the accused to lead evidence from character witnesses that they would believe the accused under oath. To the contrary, the authorities, which I will review later that permit the accused to ask character witnesses whether they would believe another witness under oath, depend upon the character witnesses first having stated that the witness has a bad character for veracity. As I understand it, it is only in reply, after the witness’ credibility has been attacked by other witnesses testifying that they would not believe the witness under oath, that the party calling the witness whose credit has been impeached may call other witnesses to vouch for the oath of the impeached witness. See Schiff, Evidence in the Litigation Process, 4th ed., Master Edition vol. 2 (1993), at 897 – 98. [23] In Béland, at p. 414, McIntyre J. adopted the following summary of the law from McWilliams, Canadian Criminal Evidence, 2d ed. (1984), at 282: When the defence seeks to put the character of the accused in issue by cross-examination of prosecution witnesses, or by calling defence witnesses other than the accused it is submitted that the rule should be strictly enforced, that is, it is confined to evidence of general reputation. However, when the accused himself puts his character in issue he is not so confined. [Emphasis added.] [24] Accordingly, while the defence may lead evidence of the accused’s reputation in the community, including the accused’s reputation for truthfulness and veracity, it may not ask those witnesses for their opinion whether they would believe the accused under oath. Defence counsel should not have asked the character witnesses whether they would believe the respondent under oath. I will consider the effect of this error at the conclusion of these reasons. II. Oath-attacking evidence a. The admissibility of reputation evidence [25] Authority binding on this court establishes that certain evidence of character traits is admissible to show that a witness lacks credibility. In R. v. Corbett, [1988] 1 S.C.R. 670, the Supreme Court of Canada held that the criminal record of a witness, including the accused, was probative of the witness’ lack of credibility. Although the admissibility of criminal records is grounded in statute, s. 12 of the Canada Evidence Act, R.S.C. 1985, c. C-5, the principle that justifies admission of that evidence can apply to admission of other evidence capable of establishing the witness’s character. At p. 686, Dickson C.J.C. referred with approval to the following excerpt from State v. Duke, 123 A.2d 745 at 746 (1956 N.H.): What a person is often determines whether he should be believed. When a defendant voluntarily testifies in a criminal case, he asks the jury to accept his word. No sufficient reason appears why the jury should not be informed what sort of person is asking them to take his word. In transactions of everyday life this is probably the first thing that they would wish to know. So it seems to us in a real sense that when a defendant goes onto a stand, “he takes his character with him ...”. Lack of trustworthiness may be evinced by his abiding and repeated contempt for laws which he is legally and morally bound to obey, as in the case at bar, though the violations are not concerned solely with crimes involving “dishonesty and false statement.” [Emphasis added.] [26] The theory upon which the admissibility of this evidence is based, the “trait or generality theory” has been criticized. It has been argued that there is no such thing as stable personality traits from which one could reasonably predict how a person would act in a given situation. See Schiff at pp. 1154-55. To the contrary, it is argued that a theory of “situationism” provides a more reasonable basis for predicting behaviour. According to this theory, behaviour is determined almost exclusively by environmental factors, by the situation in which the actors find themselves. However, even this theory has been found to be flawed and trait theory has come back into its own. Susan M. Davies describes the understanding of social scientists, S.M. Davies, “Evidence of Character to Prove Conduct: A Reassessment of Relevancy” (1991), 27 Crim.L.B. 504 at 516-17: Using improved methodology, trait theorists are now able to demonstrate the existence in individuals of consistent behavioral tendencies over a sample of situations, and to predict average behavior accurately. In fact, the usefulness of trait information in predicting behavior is no longer controverted by members of the psychology community. The most outspoken critic of trait theory has conceded that traits exist and that trait theorists “can predict many things about people at levels of confidence that are reasonable for various goals and purposes.” Even more significant for the forensic consideration of character is the fact that most psychologists now recognize that, as a general matter, a lay person, given information about a subject’s past behavior, can predict the subject’s future behavior with a significant degree of accuracy. [Italics in original, underlining added; footnotes omitted.] [27] Moreover, the test of relevancy in the law of evidence is not an exacting one and is not dependent upon scientific proof. In R. v. B. (L.); R. v. G. (M.A.) (1997), 116 C.C.C. (3d) 481 at 492 (Ont. C.A.), Charron J.A. quoted Paciocco & Stuesser, Essentials of Canadian Law: The Law of Evidence (1996), at 19. The authors explained that evidence will be deemed relevant “where it has some tendency as a matter of logic and human experience to make the proposition for which it is advanced more likely than that proposition would appear to be in the absence of that evidence”. The issue in this case really concerns the type of evidence that is admissible to prove the character trait of truthfulness or veracity and whether that trait is relevant to prove the likelihood of perjury. [28] At least three different modes of proof of character are recognized at common law and by statute. Character may in some circumstances be proved by specific acts. See for example, R. v. Scopelliti (1981), 63 C.C.C. (2d) 481 at 494 (Ont. C.A.) or s. 12 of the Canada Evidence Act, R.S.C. 1985, c. C-5. It can in some circumstances be proved by expert evidence. See for example, McMillan v. The Queen, [1977] 2 S.C.R. 824 at 827. Finally, character may be proved by the person’s reputation in the community. See Scopelliti at pp. 492-3. [29] Evidence of reputation may not be as probative as other forms of proof and its use at trials in this day and age has been criticized. Reputation is based on hearsay and may be based on nothing more than rumour and unfair and unreliable information. It has rightly been suggested that the personal opinion of someone who has had frequent opportunities to observe and interact with the subject is likely to be more reliable and accurate than reputation. It is also suggested that the whole notion of reputation is outmoded because people no longer live in “communities” in which they have reputations. However, I do not think it is open to doubt that in many circumstances evidence of reputation can have some value. See Michelson v. United States, 335 U.S. 469 (1948) at 477-78 and Sopinka, Lederman & Bryant, The Law of Evidence in Canada (1992) at 444-47. [30] I recognize that reputation for veracity as a means of predicting whether a witness will lie under oath is subject to particular frailties. First, there is the concern with reputation evidence generally and whether it is possible to accurately state a person’s reputation for veracity. See Sopinka at pp. 888-89. Second, there is a particular concern with reputation as a foundation for prediction that a witness will not tell the truth under oath. Testifying under oath is so rare and a trial is such an unusual event that a person’s reputation for honesty and veracity in everyday affairs is not likely to be of substantial assistance in predicting behaviour on the witness stand. Specific safeguards surround testimony in a court of law that are simply not part of the environment of everyday life. In court, the witness is sworn to tell the truth and is subject to prosecution for lying under oath. The testimony takes place before a judge and sometimes a jury and the witnesses know that they are subject to cross-examination. This is not to say that witnesses do not lie under oath. However, the circumstances under which a person may be willing to lie under oath are probably considerably different from the circumstances in which he or she would lie to an acquaintance or even acquire a reputation for lack of veracity. [31] Nevertheless, I am not prepared to hold that reputation evidence is irrelevant. In my view, the problematic probative value of reputation evidence for veracity is no basis for excluding defence evidence to impeach the credit of a Crown witness. b. Discretion to exclude relevant defence evidence [32] Although evidence has some probative value, it is not necessarily admissible. Even if the evidence is not subject to some exclusionary rule, such as the hearsay rule, it may still be held to be inadmissible having regard to the particular circumstances of the case. Relevance cannot be determined in a vacuum but only on the basis of the particular issues raised in the trial. Thus, evidence that may be relevant to one issue may be irrelevant to another or may actually mislead the trier of fact on the second issue. See R. v. Seaboyer; R. v. Gayme, [1991] 2 S.C.R. 577 at 609. The law of evidence therefore gives to the trial judge the task of balancing the value of the evidence against its potential prejudice. In a proper case, the judge has the power to exclude otherwise relevant evidence because the prejudicial effect of admitting the evidence outweighs its value. [33] In Seaboyer, McLachlin J. clarified the test for exclusion of otherwise relevant evidence. She held, at p. 611, that the judge has the power to exclude relevant evidence tendered by the Crown on the basis simply that its prejudicial effect outweighs its probative value. The power to exclude relevant defence evidence, however, is narrower and constrained by the fundamental tenet that an innocent person not be convicted, a tenet which now has constitutional protection. As she said at p. 608, “The right of the innocent not to be convicted is dependent on the right to present full answer and defence. This, in turn, depends on being able to call evidence necessary to establish a defence and to challenge the evidence called by the prosecution.” McLachlin J. held, at pp. 611-12, that defence evidence may be excluded where the prejudice substantially outweighs the value of the evidence. In the result, “the circumstances where truly relevant and reliable evidence is excluded are few, particularly where the evidence goes to the defence” (at p. 611). Exclusion of defence evidence can be justified only where the “potential prejudice to the trial process of admitting the evidence clearly outweighs its value" (at pp. 611-12). [34] In Seaboyer, McLachlin J. also discussed the factors a trial judge should consider in performing this balancing test. She referred in particular, at p. 610, to the principle of legal relevancy as explained in McCormick’s Handbook of the Law of Evidence, 2nd ed. (1972) at 438-40. Professor McCormick lists four factors that a judge should take into account in determining whether evidence that has some probative value meets the test for legal relevancy. Those four factors can be summarized as follows: 1. The danger that the evidence will arouse the jury’s emotions of prejudice, hostility or sympathy. 2. The danger that the proposed evidence and any evidence in response will create a side issue that will unduly distract the jury from the main issue in the case. 3. The likelihood that the evidence will consume an undue amount of time. 4. The danger of unfair surprise to the opponent who had no reasonable ground to anticipate the issue and was unprepared to meet it. [35] I would add a fifth counter-balancing factor to this list: 5. The danger that the evidence will be presented in such a form as to usurp the function of the jury. [36] These then are the general principles that ought to guide a court in determining the admissibility of defence evidence of reputation and opinion as to veracity. Before considering the application of these principles to this case it is necessary to review the state of the law in this province concerning the admissibility of oath-attacking evidence. c. The authorities respecting "oath-attacking" evidence [37] The present understanding of the rule respecting the admissibility of oath-attacking evidence in this province is founded in the decision of Martin J.A. in R. v. Gonzague, supra. In that case, the trial judge had refused to permit a witness to give his personal opinion that a Crown witness could not be believed on oath and had held that the impeaching witness was confined to expressing an opinion based upon the Crown witness’s general reputation. Martin J.A. held at p. 512 that the trial judge’s ruling was in accord with “long-established practice in his province” as described in an article by Dr. D.A. MacRae at 7 C.E.D. (Ont. 2nd) at 364: The rule in Ontario has long been that the witness called to impeach the credit of another is first to be asked “Do you know the reputation of the witness for truth and veracity in the community in which he resides?” Unless the answer is in the affirmative the witness cannot proceed further. If he answers “Yes”, he is then to be asked “Is that reputation good or bad?” Unless he answers “Bad”, he cannot proceed. If he answers “Bad”, he may be asked “From that reputation, would you believe him on oath?” These and only these are the questions permitted to be asked in chief. [38] Martin J.A. then noted that the rule had been broadened by the English Court of Appeal in R. v. Gunewardene (1951), 35 Cr. App. R. 80 to permit the impeaching witnesses to state that from their own knowledge they believed the Crown witness to be unworthy of credit under oath. This holding by the Court of Appeal was later approved by the House of Lords in Toohey v. Metropolitan Police Commissioner, [1965] A.C. 595. At p. 512, Martin J.A. made this observation with respect to Toohey: Although the matter is not free from doubt, because of the views expressed by the House of Lords in Toohey v. Metropolitan Police Com'r, supra, and in view of the opinion expressed by Professor Cross in his well-known work Cross on Evidence, 2nd ed. (1963), which was also quoted with approval in the Toohey case, we think it would have been preferable if the defence had been allowed to elicit the evidence which it desired. The evidence, in our view, however, had little weight and we would not, on this ground, have interfered with the verdict of the jury. [Emphasis added.] [39] The reservations expressed by Martin J.A. were, as noted, based on certain comments in Toohey itself and by Professor Cross. In Toohey, Lord Pearce had the following to say about the rule, at pp. 605-606: From olden times it has been the practice to allow evidence of bad reputation to discredit a witness’s testimony. It is perhaps not very logical and not very useful to allow such evidence founded on hearsay. None of your Lordships and none of the counsel before you could remember being concerned in a case where such evidence was called. But the rule has been sanctified through the centuries in legal examinations and textbooks and in some rare cases, and it does not create injustice. [Emphasis added.] [40] The authors of Cross and Tapper on Evidence, 8th ed. (1995) at 338 make these comments concerning the rule: From the very beginning of the modern law of evidence it has been possible to call a witness to swear that the opponent’s witness cannot be believed upon his oath. It was established that the impugning witness must speak from his personal knowledge, but that he could not refer to particular events to justify his belief. The general form of the question went from the general to the more particular, by asking first ‘Have you the means of knowing what the general character of the witness is?’, and then ‘From such knowledge of his general character would you believe him on oath?’ It was not necessary that the witness need have personal knowledge of false testimony by the witness, and even if he did, he would not be allowed to refer to it under the ban on collateral matters. The rule was reconsidered in R. v. Brown and Hedley [(1867) 10 Cox CC 453] in the light of the decision in R. v. Rowton [(1865) Le & Ca 520] that a witness could not express his personal opinion of the accused’s character in rebuttal of a character witness called by the accused. The reasoning of the court was not wholly satisfactory, but the outcome was to endorse the existing practice, and to condense the questioning to elicit a simple asseveration that the impugning witness would not believe the opponent’s witness on his oath. It became very rare in practice for this cumbersome, anomalous and unconvincing exercise to be conducted. Its continuing availability was however endorsed by a unanimous House of Lords in Toohey v. Metropolitan Police Comr. [Emphasis added; footnotes omitted.] [41] Finally, I should mention one further Ontario development. In R. v. Taylor (1986), 55 C.R. (3d) 321 (Ont. C.A.), the defence had been permitted to lead the evidence of twenty character witnesses that they would not believe the two sexual assault complainants under oath. The witnesses were also permitted to testify as to the basis for their opinion, namely the tendency of the complainants to recount strange and apparently provably false tales of sexual assaults upon them. In reply, the Crown called two expert witnesses to explain that sexually abused children tend to have a rich fantasy life. The accused was convicted and the grounds of appeal concerned the admissibility of some of this expert evidence and the trial judge’s charge to the jury with respect to that evidence. However, in obiter dicta, Cory J.A. made this observation, at p. 326: It was certainly appropriate for the defence to call witnesses who could testify that the complainants should not be believed under oath. These same witnesses could testify as to the basis of their opinion, specifically, the incredible stories of sexual assaults related to them by the complainants. These witnesses were not recounting the bizarre tales in order to contradict the evidence given by the complainants on collateral issues pertaining to other real or imagined sexual assaults. Rather, they were putting forward the fantastic accounts related by the complainants as the basis for their belief of their testimony. [Emphasis added.] [42] Cory J.A. gave Cross on Evidence, 6th ed. 293-94, and Phipson on Evidence, 13th ed. at 820-1 as authority for this proposition. In his annotation to Taylor at pp. 321-23, Professor Delisle points out that these texts do not appear to support this expansion of the rule. While experts may testify as to the basis for their opinion, the law as set out in Cross and Phipson is that reputation witnesses may not give the basis for their opinion during examination-in-chief, although they may be cross-examined on the basis for their opinion. It may well be that there is some other basis for the admissibility of the particular evidence in Taylor. The issue was not directly before the court in that case and is not an issue in this appeal. I would leave to another day whether the law of evidence would be better served by permitting evidence of character for lack of veracity to be proved by evidence of specific acts as is allowed, for example, in a self-defence case where the accused seeks to adduce the victim’s character for violence. See Scopelliti at p. 494, Sopinka, at pp. 465-66, and Davies, at p. 533. [43] It remains to consider whether the general principles that I have outlined above entitle this court to reconsider the holding in Gonzague. d. Application of the principles to oath-attacking evidence [44] In my view, the decisions of the Supreme Court of Canada in Seaboyer and Béland have to some extent overtaken the decision of this court in R. v. Gonzague and it is appropriate to reconsider the oath-attacking rule. I am comforted in undertaking this exercise in view of Martin J.A.’s own reservations about the rule as expressed in the excerpt from Gonzague set out earlier. [45] The most important development in the law of evidence since the decisions of this court is the recognition by McLachlin J. in Seaboyer that relevant defence evidence is subject to the balancing of probative value against prejudicial effect. The test for exclusion of relevant defence evidence is strict, but it exists. [46] I also consider the Supreme Court of Canada’s strong disapproval of oath-helping evidence in Béland to be an important development. The concerns expressed in that case about oath-helping evidence apply equally to oath-attacking evidence. Although McIntyre J. was dealing with polygraph evidence in that case, his comments, at pp. 418‑19, about oath-helpers are of assistance in considering other related types of evidence: … I would not wish to see a return to the method of pre-Norman trials where parties relied heavily upon oath-helpers who swore to their veracity. … I would seek to preserve the principle that in the resolution of disputes in litigation, issues of credibility will be decided by human triers of fact, using their experience of human affairs and basing judgment upon their assessment of the witness and on consideration of how an individual's evidence fits into the general picture revealed on a consideration of the whole of the case. [Emphasis added.] [47] In my view, these comments although directed to oath-helping are equally applicable to the kind of oath-attacking evidence implied in the answer to the third question, which invites one witness to express a personal opinion as to the lack of veracity of another witness. [2] [48] In view of these developments, it is my view that it is open to this court to reconsider the holding in Gonzague. For convenience, I will repeat the three questions authorized by that case. 1. Do you know the reputation of the witness as to truth and veracity in the community? [3] If the answer is “yes” the questioning proceeds. 2. Is that reputation good or bad? If the answer is “bad” a final question is permitted. 3. From that reputation [or from your own knowledge], would you believe the witness on oath? [49] In my view, the prejudicial effect of the answer to the third question will almost invariably substantially outweigh its probative value. The form in which the evidence is presented tends to usurp the function of the jury and engages what I have referred to as the fifth counter-balancing factor. This factor is most often mentioned where it is proposed to lead expert opinion evidence. See R. v. Mohan, [1994] 2 S.C.R. 9 at 24-25 and R. v. Millar (1989), 49 C.C.C. (3d) 193 at 220 (Ont. C.A.). Thus a judge has a discretion to exclude a conclusory statement [4] by an expert where the opinion can be framed in less conclusory terms. However, I believe that this concern can apply where lay witnesses are permitted to express an opinion on a critical issue in the case. This can occur in circumstances where the qualifications of the lay witnesses to provide that opinion may be overvalued by the jury. [50] The Supreme Court of Canada addressed this issue in Graat v. The Queen, [1982] 2 S.C.R. 819, at pp. 839‑40. Dickson J. rejected the argument that on a trial for impaired driving, the opinion of the arresting police officer that the accused was impaired was inadmissible because it went to the ultimate issue. However, he registered “two caveats”. First, the trial judge must exercise a “large measure of discretion” in determining whether an opinion is admissible. Second, there may be a tendency for the trier of fact to let the opinion of police witnesses “overwhelm the opinion evidence of other witnesses.” He pointed out that there is no special reason to prefer the opinion of the non-expert police witness and that as in all cases the trier of fact must decide in each case what weight to give what evidence. I have similar concerns with reputation evidence as to veracity. While the defence witnesses may know the Crown witness' reputation for telling the truth in everyday affairs, for the reasons expressed earlier, their ability to predict the witness' behaviour in court is limited and entitled to no special deference. [51] Where the outcome of the case depends upon the evidence of a single witness, an expression of opinion as to that witness' veracity is a comment on the ultimate issue. There is the risk that in some cases the jury will simply defer to the opinion of the character witness rather than embarking on the difficult task of examining the evidence and measuring it against the standard of proof beyond a reasonable doubt, on the theory that the character witness obviously knows the Crown witness and is in a much better position to determine the outcome of the case. Put another way, the ability of a character witness, who has not heard the evidence in the case, to predict whether another witness has told the truth under oath is very limited. The jury may, however, overvalue that opinion because the character witness knows the witness. [52] Accordingly, I would hold that the accused does not have the absolute right to ask the third question and that in most cases the trial judge would be justified in refusing to permit that question to be asked. I would adopt the holding by Cumming J.A. in Masztalar v. Wiens (1992), 10 B.C.A.C. 19 at 23 to the effect that while the rule need not be absolutely abolished it should be retained "to be sparingly applied only in the rare case where the interests of justice require it". [5] To the extent that the decisions of this court in Gonzague and Taylor hold to the contrary, they should no longer be followed. [53] With respect to the first two questions, in my view, the trial judge also has a discretion to exclude witnesses that would provide that evidence where the prejudicial effect would substantially outweigh its probative value. I should say, however, that in my view it would be an extremely rare case where a trial judge would be warranted in excluding the evidence. Unlike the third question, the first two questions do not invite answers that would have the tendency to usurp the jury's function. The reputation evidence is simply another piece of circumstantial evidence that the jury can use to assess the credibility of the Crown witness' story. I also do not think that this evidence has the same tendency to distract the jury from the main issue in the case, nor is it likely to create side issues. [54] The only serious problem about this defence evidence is whether the evidence will consume an undue amount of time. It seems to me that a trial judge has the discretion to limit the number of reputation witnesses called by the defence where the judge is satisfied that the prejudicial effect of calling an endless series of witnesses testifying to essentially the same fact substantially outweighs the probative value of this additional evidence. The judge will, however, wish to keep in mind that the Supreme Court in Seaboyer has held that the discretion to exclude relevant evidence must be exercised with extreme caution and that in most cases the better course would be to permit the defence to call all of those witnesses. e. The Charge to the Jury [55] Where reputation evidence as to the credibility of a witness is admitted, in addition to the normal factors affecting the weight of evidence of any witness such as their interest in the case, bias and means of knowledge, it would be preferable for the trial judge to cover at least two points peculiar to this kind of evidence. First, whatever the witness' reputation for veracity in the community, testifying in court under oath is a very different circumstance and the jury will want to bear this in mind. Second, the character witnesses have not heard all the evidence, and are not sworn to the heavy duty of the juror to render a true verdict. The jury may find the reputation evidence helpful in determining the credibility of the witnesses, but they should not automatically defer to that evidence. There may be other important factors for the jury to consider which will have been reviewed by the trial judge in the standard part of the charge to the jury. III APPLICATION TO THIS CASE a. Admission of the oath-helping and oath-attacking evidence [56] I have already indicated that in my view the trial judge erred in permitting the defence witnesses to give their opinion on whether the respondent would lie under oath. I have also held that the trial judge has the power to limit the number of reputation witnesses and to prevent the defence from asking the third question concerning the complainant. This would not have been a proper case to prevent the accused from calling defence witnesses to testify about the complainant’s reputation in the community for veracity. The witnesses appeared to have had a sufficient opportunity to discover the complainant’s reputation and the evident bias of two of the witnesses was not a ground upon which to disqualify them. However, the trial judge should not have permitted defence counsel to ask the third question. For the reasons expressed above, the prejudicial effect of the evidence obtained in response to that third question substantially outweighed its probative value. [57] Nevertheless, I am not satisfied that the Crown has met the heavy burden of demonstrating with a reasonable degree of certainty that the verdict would not necessarily have been the same had the impugned evidence not been admitted. The reputation evidence in favour of the respondent was given by his friends who would not be perceived by the jury as giving neutral or impartial evidence of such weight that it could be said to have unfairly prejudiced the Crown’s case. It was only the opinion of these witnesses that the respondent could be believed when testifying under oath that was inadmissible and this additional testimony could not in my view have affected the verdict. [58] The oath-attacking evidence was for the most part substantially undermined by an effective cross-examination by Crown counsel. The cross-examination revealed that two of the three witnesses had very personal reasons for testifying in the manner that they did and were very much biased against the complainant. Mr. Valentine had been charged and convicted of assaulting the complainant. Although he claimed that the complainant could not be believed under oath, he in fact had pleaded guilty in response to the allegation made by her. Ms. Taylor blamed the complainant for the break-up of her marriage. It cannot reasonably be said that this evidence would have improperly affected the jury. b. The Charge to the Jury [59] On appeal, the Crown also complains about the charge to the jury with respect to this evidence. After giving the standard instruction to the jury concerning criminal records, the trial judge gave the following instruction concerning the reputation witnesses. The second evidential matter is that in this case we have heard evidence of good and bad reputation for truthfulness called by the defence to establish that Mr. Clarke has a good reputation for telling the truth in the community in which he lives: Trenton; and to establish that Ms. Reid has a bad reputation for telling the truth in the community or communities in which she lives or has lived: Trenton, Belleville, Picton and, in the personal opinion of these witnesses, is not a person whose word they as individuals would believe under oath. The accused is permitted under our law to lead this evidence by witnesses as to the reputation for truth and veracity of other witnesses (in this case, the other witnesses being the accused and Ms. Reid) in the community in which they live. In assessing the evidence of these five witnesses as to reputation you must consider – you must decide, first, whether you believe them, that is, whether their evidence is credible; and, second, if you believe some or all of their evidence, what weight to attach to it. You must use the same factors as you use in assessing the credibility of other witnesses, their ability and opportunity to observe, their appearance and manner when testifying before you, their powers of recollection, any interest, bias or prejudice they may have, their sincerity when testifying, inconsistencies in their testimony and its reasonableness when considered in the light of all the evidence in this case. As to the first of these tests, the ability and opportunity of the witnesses to observe, you must consider in relation to their evidence how well the witnesses know the accused or Marlene Reid and in what context the witnesses have come to know them. You must also consider that their testimony as to the reputation for truthfulness of those witnesses must be directed towards the general reputation of the person in the community in which he or she lives. How well do the witnesses know the accused and Marlene Reid whose reputations for truthfulness they are testifying to? You will note that all of the witnesses described themselves as good friends of the accused. None of them described themselves as good friends of Marlene Reid. All expressed favourable opinions as to the credibility of Mr. Clarke. All but two of them, Grace Boyce and Cindy McKavanagh, expressed unfavourable opinions about Marlene Reid. You will also note that one of the witnesses, Kathy Lee Taylor is the sister of Marlene Reid who testifies that she holds Ms. Reid responsible for the breakup of her previous marriage. Another, Milton Ballentine who formerly had a relationship with Marlene Reid, was charged with committing an assault on her, a charge to which he pleaded guilty. A third, Dorothy McKillop described Ms. Reid as a former friend. I would remind you that you are entitled to weigh in the factor of interest and bias or prejudice in considering the evidence of these three witnesses who were the only witnesses who testified unfavourably to the reputation of Marlene Reid for truthfulness. [Emphasis added.] [60] It would have been preferable for the trial judge to have included the instructions I have suggested above; namely, that (1) whatever the witness’s reputation for veracity in the community, testifying in court under oath is a very different circumstance and this should be borne in mind and (2) the character witnesses have not heard all the evidence, and are not sworn to the heavy duty of the juror to render a true verdict. [61] On the other hand, this second factor was largely covered in the earlier general instructions where the trial judge gave these directions: You are the final judges as to the weight to be given to the evidence which has been tendered at this trial. … It is not only your right but your duty to make your own decisions on the evidence heard at this trial. You determine the true facts and the weight to be given to the evidence tendered. … I would like to address the subject of the credibility of witnesses. You are the sole judges of the truthfulness of the witnesses and of the weight to be given to the testimony of each of them. … Furthermore, you are not obliged to accept any part of the evidence of a witness just because there is no denial of it. Keep in mind that in determining guilt or innocence you must assess the evidence cumulatively and as a whole in arriving at your verdicts. Moreover, you are not bound to decide an issue in favour of a party who calls the largest number of witnesses on that issue. On the other hand, one witness may suffice to satisfy you on the issue if that witness’s evidence is credible and compelling. [62] Moreover, in this case, the instructions that were given focused particularly on the personal bias that the character witnesses had as friends of the respondent and their motives for testifying against the complainant. These were the factors that in this case were the most telling against the value of the defence evidence and they were clearly and forcefully brought to the attention of the jury. The trial judge’s charge to the jury was thorough and he dealt at length with issues of credibility. I also note that Crown counsel at trial raised no objection to the charge to the jury. I am not persuaded that these minor deficiencies in the charge to the jury could have affected the verdict. DISPOSITION [63] Accordingly, I would dismiss the appeal. RELEASED: September 9, 1998 [1] Although not raised by counsel for the respondent, I should point out that, in my view, the trial judge was in error in holding that the character witnesses could only give evidence about reputation if they were aware of the accused’s or complainant’s reputation in a particular city or town where they lived, such as Trenton. With the increasing urbanization of society, a person’s community will not necessarily coincide with a particular geographic location. Thus, in this case, it would seem to me that it was open to the defence to lead evidence of the respondent’s reputation in the Caribbean community in the area. In R. v. Levasseur (1987), 35 C.C.C. (3d) 136 (Alta. C.A.), the court held that the accused should have been permitted to call a witness to testify as to the accused’s reputation at her place of work even though the witness knew nothing of the accused’s reputation in the community where she resided. At p. 141, Harradence J.A. referred to Wigmore on Evidence Chadbourn Revision, vol. V (1974) at 591-92, including this excerpt: There may be distinct circles of persons, each circle having no relation to the other, and yet each having a reputation based on constant and intimate personal observation of the man. There is every reason why the law should recognize this. Time has produced new conditions for reputations. The traditional requirement about "neighbourhood" reputation was appropriate to the conditions of the time; but it should not be taken as imposing arbitrary limitations not appropriate in other times. Alia tempora, alii mores. What the law, then as now, desired was a trustworthy reputation; if that is to be found among a circle of persons other than the circle of dwellers about a sleeping-place, it should be received. [Emphasis added by Harradence J.A.] [2] In R. v. Moran (1992), 77 C.C.C. (3d) 300 at 307-8 (Ont. C.A.), Finlayson J.A. (dissenting in the result as the majority found that any improper questioning could not have affected the result) cited Béland as authority that one witness’s opinion as to whether he or she believed another witness was telling the truth was inadmissible. [3] For the reasons set out in footnote 1, I have removed the words "in which the witness resides" from this question. [4] Such as, that the deceased infant was the victim of “child abuse”. [5] There may be unusual circumstances where the reputation of the Crown witness is such that it is possible to more accurately predict the likelihood that the witness would lie under oath. In those cases, the answer to the third question may provide the jury with useful information. With a careful jury instruction the danger that this testimony will usurp the jury's function can be minimized.