R. v. W. (R.), [1992] 2 SCR 122

R. v. W. (R.), [1992] 2 S.C.R. 122

Her Majesty The Queen                                                                   Appellant

v.

R. W.                                                                                                  Respondent

Indexed as:  R. v. W. (R.)

File No.:  21820.

1992:  April 2; 1992:  June 11.

Present:  La Forest, L'Heureux‑Dubé, Gonthier, Cory, McLachlin, Stevenson^* and Iacobucci JJ.

on appeal from the court of appeal for ontario

Criminal law ‑‑ Powers of court of appeal ‑‑ Accused convicted of sexual offences on basis of children's testimony ‑‑ Court of Appeal overturning convictions ‑‑ Whether Court of Appeal exceeded proper limits of appellate review ‑‑ Criminal Code, R.S.C., 1985, c. C‑46, s. 686(1) (a)(i).

Criminal law ‑‑ Evidence ‑‑ Children ‑‑ Accused convicted of sexual offences on basis of children's testimony ‑‑ Proper approach to evidence of young children ‑‑ Whether Court of Appeal erred in overturning convictions.

The accused was charged with indecent assault, gross indecency and sexual assault against three young girls.  The youngest girl, his niece, was between two and four years old when the incidents occurred, seven when they were reported to the authorities, and nine at the time of trial.  The other two girls were his step‑daughters.  The younger one was between nine and ten at the time of the events, eleven when they were reported, and twelve at the time of trial, while the oldest girl was ten at the time of the events, fourteen at the time of reporting and sixteen at the time of trial.  At the trial the girls described the incidents out of which the charges arose, and the accused denied the allegations.  The evidence of the oldest child was uncontradicted, apart from the accused's denial, and internally consistent, but the evidence of the two younger children revealed a number of inconsistencies and was contradicted in some respects.  The accused was convicted on all five counts.  The Court of Appeal set aside the convictions and entered acquittals.  It found that there was "really no confirmatory evidence", that the evidence of the two younger children was "fraught with inaccuracy" and that neither of the older children was "aware or concerned that anything untoward occurred".

Held:  The appeal should be allowed.

In determining whether the trier of fact could reasonably have reached the conclusion that the accused is guilty beyond a reasonable doubt, a court of appeal must re‑examine and to some extent reweigh and consider the effect of the evidence.  This applies to verdicts based on findings of credibility.  The test is whether a jury or judge properly instructed and acting reasonably could have convicted.  In applying this test the appeal court should show great deference to findings of credibility made at trial.  While the Court of Appeal thus did not err in this case in re‑examining and reweighing the evidence, it did err in setting aside the convictions.  The law concerning the evidence of children has undergone two major changes in recent years.  First, the notion, found at common law and codified in legislation, that the evidence of children was inherently unreliable and therefore to be treated with special caution has been eliminated.  Thus various provisions requiring that a child's evidence be corroborated have been repealed.  Second, there is a new appreciation that it may be wrong to apply adult tests for credibility to the evidence of children.  While the evidence of children is still subject to the same standard of proof as the evidence of adult witnesses in criminal cases, it should be approached not from the perspective of rigid stereotypes, but on a common sense basis, taking into account the strengths and weaknesses which characterize the evidence offered in the particular case.  The Court of Appeal went too far in this case in finding lacunae in the evidence which did not exist and in applying a stringent, critical approach to the evidence.  It appears to have been influenced by the old stereotypes relating to the inherent unreliability of children's evidence and the "normal" behaviour of victims of sexual abuse and to have placed insufficient weight on the trial judge's findings of credibility.  The verdicts in this case were ones which a properly instructed jury (or judge), acting judicially, could reasonably have rendered.

Cases Cited

Referred to:  R. v. Yebes, [1987] 2 S.C.R. 168;  Corbett v. The Queen, [1975] 2 S.C.R. 275;  R. v. Howard, [1989] 1 S.C.R. 1337;  R. v. S. (P.L.), [1991] 1 S.C.R. 909;  White v. The King, [1947] S.C.R. 268;  R. v. M. (S.H.), [1989] 2 S.C.R. 446;  R. v. B. (G.), [1990] 2 S.C.R. 30.

Statutes and Regulations Cited

Act to amend the Criminal Code and the Canada Evidence Act, S.C. 1987, c. 24, s. 15.

Canada Evidence Act, R.S.C. 1970, c. E‑10.

Criminal Code, R.S.C. 1970, c. C‑34, s. 586.

Criminal Code , R.S.C., 1985, c. C‑46 , ss. 686(1) (a)(i), 695 .

Young Offenders Act, S.C. 1980‑81‑82‑83, c. 110.

APPEAL from a judgment of the Ontario Court of Appeal setting aside the respondent's convictions on charges of indecent assault, gross indecency and sexual assault.  Appeal allowed.

Catherine A. Cooper, for the appellant.

Robert J. Reynolds, for the respondent.

//McLachlin J.//

The judgment of the Court was delivered by

McLachlin J. -- This case raises the issue of the proper limits of appellate jurisdiction to overturn a conviction under s. 686(1) (a)(i) of the Criminal Code , R.S.C., 1985, c. C-46 .  It also raises the question of the approach which should be taken to the evidence of children.

Facts

The respondent was convicted on three counts of indecent assault, one count of gross indecency and one count of sexual assault against three young girls, B.W., M.W. and S.W.  Without going into the rather complex details of the family relationship, it may be simply stated that the respondent lived with and assumed a quasi-paternal role vis-à-vis the three girls, being the uncle of B.W. and having been married to the mother of S.W. and M.W. 

The first three charges (two of indecent assault and one of gross indecency) concerned B.W.  She was between two and four years old when the incidents occurred in 1981-82, seven years old when the offences were reported to the authorities, and nine years old at the time of trial.

The fourth charge (of sexual assault) concerned M.W.  She was between nine and ten at the time of the events in question, eleven when the offences were reported, and twelve at the time of trial.

The fifth charge (of indecent assault) concerned S.W.  She was ten years old at the time of the events in question, fourteen at the time of reporting, and sixteen at the time of trial.

The substance of the charges is as follows.  B.W. alleged that while she was living with the respondent and his wife (her uncle and aunt) in the red brick house in Napanee, the respondent would come and take her from her bed, which she shared with S.W. and M.W., and lead her upstairs to his bedroom, where he would take down her pants, drop his pyjamas, and rub his soft penis over her vagina.  B.W. stated in evidence that this happened two or three times a week.  On one occasion, the respondent asked B.W. to suck on his "dink", which she did (this incident is the basis of the gross indecency charge).  B.W. also described a further incident which occurred on a camping trip, when the respondent took her into the family's van, laid her on the floor, and rubbed his penis over her vagina.

S.W. described two incidents in the red brick house in Napanee.  S.W. and the respondent were fully clothed throughout both incidents.  The first incident began as play-wrestling, but the respondent began to breathe heavily and rubbed his erect penis against her buttocks.  In the second incident, the respondent lay down beside her in bed, and she could feel his erect penis against her buttocks.

M.W. described three incidents involving the respondent.  Both M.W. and the respondent were fully clothed throughout all three incidents.  In the first incident, she was sitting on the respondent's lap while he combed her hair and rubbed his erect penis against her lower back.  In the second and third incidents, the respondent rolled M.W. onto the floor, forced her legs apart and rubbed his penis against her vagina.

The respondent testified in his own defence. He admitted to wrestling with M.W. and S.W., and to brushing the girls' hair, but denied all the allegations that formed the basis of the charges against him. Aside from the respondent's denial, the evidence of the oldest child, S.W., was uncontradicted and internally consistent. The evidence of the two younger children, however, quite apart from the respondent's denial, revealed a number of inconsistencies and was contradicted in some respects.

Judgments Below

A.  Ontario District Court (Judge O'Flynn, June 27, 1988)

Judge O'Flynn convicted the respondent on all five counts.  The trial judge instructed himself (correctly, in my opinion) on the importance of corroborative evidence in cases of this kind:

I am considering allegations of sexual misconduct where young children are involved and by reason of the fact that such acts were alleged to have occurred in secret, it is important to look at the evidence to see if there is other evidence which tends to confirm the evidence of the particular complainant enhancing the probability that the complainant was telling the truth on the material issues.  However, the law does not require the evidence of the particular complainant to be supported in order to convict the accused and corroboration in law is not necessary.

The trial judge then reviewed the evidence, noting the inconsistencies.

The trial judge's conclusion, in its entirety, was as follows (omitting the re-statement of the five charges against the respondent and the full names of the girls):

I have considered each complainant's intelligence, memory, power of observation, interest in the outcome and demeanor in the witness box.  There does not appear to be collusion between the girls.  I have considered the inconsistencies in the girls' evidence which I have referred to, as minor and attributable to their youth and the passage of time.

I have also considered the fact that B.W. did not promptly disclose the alleged abuse to anyone, however, I have also considered the explanation of why by B.W.  [The trial judge may be referring here to B.W.'s testimony that the respondent had told her not to tell anybody, "or you'll be in trouble" (in-chief, C.O.A. p. 38).]  I have considered the young age of the children and particularly B.W. when the alleged offences were to have taken place and the time that has passed since the alleged incidents.

                                                                    ...

I accept the evidence of the children, B.W., M.W. and S.W. in regard to the alleged offences and I find the accused guilty as charged on Count Nos. 1, 2, 3, 4 and 5 (as amended).

The respondent was sentenced on September 9, 1988.  The total sentence was for fifteen months, broken down as follows:

Count 1 (indecent assault on B.W.):                                   │

Count 2 (gross indecency on B.W.):                                  ├──9 months concurrent

Count 3 (indecent assault on B.W.):                                   │

Count 4 (sexual assault on M.W.):                                     3 months consecutive

Count 5 (indecent assault on S.W.):                                   3 months consecutive

B.                Ontario Court of Appeal (Brook and Krever JJ.A. and Craig J. (ad hoc), December 21, 1989)

The following is the text of the Court of Appeal's endorsement:

This case has caused us very great concern. The case has been carefully argued. We recognize the advantage of the trial judge, but also the responsibility of this court. Yebes v. The Queen (1987), 36 C.C.C. (3d) 417 at p. 430. Giving the matter our best consideration, we are all of the opinion that on this evidence these convictions cannot safely stand. There was really no confirmatory evidence, the evidence of the two younger children was fraught with inaccuracy and in the case of the older children [it was] perfectly clear that neither was aware or concerned that anything untoward occurred which is really the best test of the quality of the acts. The appeal is allowed, the conviction is set aside and an acquittal is entered.

Issues

The central issue is whether the Court of Appeal erred in setting aside the convictions entered by the trial judge.  This raises a number of subsidiary issues:

(1) What is the standard for review under s. 686(1)(a)(i), and did the Court of Appeal depart from it?

(2) Did the Court of Appeal err in the way it treated the evidence of the children?  In particular:

(a) Did the Court of Appeal err in relying on the alleged lack of confirmatory evidence?

(b) Did the Court of Appeal err in its treatment of the inaccuracies in the children's evidence?

(c) Did the Court of Appeal err in relying on the alleged fact that neither  of the older children was aware of or concerned about the nature of the acts?

Analysis

l.                  What is the Standard of Review and Did the Court of Appeal Depart From It?

The appellant contends for a narrow right of review: a court of appeal can set aside a conviction based on findings of credibility, in its submission, only if the trier of fact proceeded on the basis of an error of law or if evidence is accepted on a central issue which cannot logically be true given other undisputed facts.  The respondent, on the other hand, argues that a court of appeal must look at all the evidence, consider its probity and weight, and set aside the conviction only if it concludes that the trier of fact could not reasonably have come to the conclusion that the offence was proved beyond a reasonable doubt.

The ambit of the powers of an appellate court under s. 686(1)(a)(i) has been considered by this Court in a number of recent cases.  The leading case is probably R. v. Yebes, [1987] 2 S.C.R. 168.  Writing for the Court, McIntyre J. reviewed the earlier case of Corbett v. The Queen, [1975] 2 S.C.R. 275.  McIntyre J. acknowledged that there was some ambiguity in Corbett as to the test to be applied.  The ambiguity was between a standard that no jury could possibly have reached the verdict at trial, and that no jury could reasonably have reached the verdict.  McIntyre J. resolved the ambiguity in favour of the "reasonably reached" standard (at p. 185):

I am in agreement with Hutcheon J.A. that the word "possibly" in this context is inappropriate.  In my view, to adopt literally the proposition that the appellate court could only consider whether the impugned verdict could possibly have been reached would render review on appeal under the subsection almost impossible.  "Reasonably could have reached" must be the test, and from a reading of the whole of Pigeon J.'s judgment [in Corbett] I am of the view that it was what was intended.  The concept of reasonableness is clearly expressed in the section which speaks of an unreasonable verdict.  Therefore, curial review is invited whenever a jury goes beyond a reasonable standard.  In my view, then, Corbett is the governing case and the test is "whether the verdict is one that a properly instructed jury acting judicially, could reasonably have rendered".  [Emphasis in original.]

McIntyre J. acknowledged that in conducting a review under s. 686(1)(a)(i), a court of appeal must, to some extent at least, re-examine and reweigh the evidence (at p. 186):

The function of the Court of Appeal, under s. 613(1) (a)(i) of the Criminal Code , goes beyond merely finding that there is evidence to support a conviction.  The Court must determine on the whole of the evidence whether the verdict is one that a properly instructed jury, acting judicially, could reasonably have rendered.  While the Court of Appeal must not merely substitute its view for that of the jury, in order to apply the test the Court must re-examine and to some extent reweigh and consider the effect of the evidence.  This process will be the same whether the case is based on circumstantial or direct evidence.  In the Court of Appeal, the majority clearly found that there was sufficient evidence to justify the verdict and both Macdonald and Craig JJ.A. rejected all rational inferences offering an alternative to the conclusion of guilt.  It is therefore clear that the law was correctly understood and applied.

Yebes has been applied in at least two subsequent decisions of this court: R. v. Howard, [1989] 1 S.C.R. 1337; and R. v. S. (P.L.), [1991] 1 S.C.R. 909.  In Howard, Lamer J. (as he then was), writing for three of five justices on the panel, adopted the test of McIntyre J. in Yebes in upholding the conviction (at p. 1349):

I have read evidence and have come to the conclusion that a verdict of first degree murder "is one that a properly instructed jury acting judicially could reasonably have rendered" (per McIntyre J. for the Court in R. v. Yebes, [1987] 2 S.C.R. 168, at p. 185).

Similarly, in S. (P.L.), Sopinka J., writing for himself, Lamer C.J., and La Forest and McLachlin JJ., adopted the test set out in Yebes.  Sopinka J. emphasized that the court of appeal is properly engaged in a review of the facts pursuant to s. 686(1)(a)(i) (at p. 915):

In an appeal founded on s. 686(1)(a)(i) the court is engaged in a review of the facts.  The role of the Court of Appeal is to determine whether on the facts that were before the trier of fact a jury properly instructed and acting reasonably could convict.  The court reviews the evidence that was before the trier of fact and after re-examining and, to some extent, reweighing the evidence, determines whether it meets the test.  See R. v. Yebes, [1987] 2 S.C.R. 168.                                                                     

It is thus clear that a court of appeal, in determining whether the trier of fact could reasonably have reached the conclusion that the accused is guilty beyond a reasonable doubt, must re-examine, and to some extent at least, reweigh and consider the effect of the evidence.  The only question remaining is whether this rule applies to verdicts based on findings of credibility.  In my opinion, it does.  The test remains the same: could a jury or judge properly instructed and acting reasonably have convicted?  That said, in applying the test the court of appeal should show great deference to findings of credibility made at trial.  This Court has repeatedly affirmed the importance of taking into account the special position of the trier of fact on matters of credibility: White v. The King, [1947] S.C.R. 268, at p. 272; R. v. M. (S.H.), [1989] 2 S.C.R. 446, at pp. 465-66. The trial judge has the advantage, denied to the appellate court, of seeing and hearing the evidence of witnesses.  However, as a matter of law it remains open to an appellate court to overturn a verdict based on findings of credibility where, after considering all the evidence and having due regard to the advantages afforded to the trial judge, it concludes that the verdict is unreasonable.

I therefore conclude that the Court of Appeal did not err in re-examining and reweighing the evidence, as the appellant contends.  That leaves, however, the question of whether, on all the evidence, the Court of Appeal was entitled to conclude that the judge could not reasonably have decided that the accused was guilty beyond a reasonable doubt.

2.                Did the Court of Appeal Err in the Way it Approached the Evidence of the Children?

The appellant contends that the Court of Appeal made errors of fact and of law in the process of re-examining and reweighing the evidence of the children which led it wrongly to conclude that the verdicts were unreasonable.

Before turning to the particular errors alleged, I pause to consider the general question of how courts should approach the evidence of young children.  The law affecting the evidence of children has undergone two major changes in recent years.  The first is removal of the notion, found at common law and codified in legislation, that the evidence of children was inherently unreliable and therefore to be treated with special caution. Thus, for example, the requirement that a child's evidence be corroborated has been removed: s. 586 of the Criminal Code, R.S.C. 1970, c. C-34, which prohibited the conviction of a person on the uncorroborated evidence of a child testifying unsworn, was repealed by An Act to amend the Criminal Code and the Canada Evidence Act, S.C. 1987, c. 24, s. 15, effective January 1, 1988. Similar provisions of the Canada Evidence Act, R.S.C. 1970, c. E-10, and Young Offenders Act, S.C. 1980-81-82-83, c. 110, have also been eliminated. The repeal of provisions creating a legal requirement that children's evidence be corroborated does not prevent the judge or jury from treating a child’s evidence with caution where such caution is merited in the circumstances of the case.  But it does revoke the assumption formerly applied to all evidence of children, often unjustly, that children’s evidence is always less reliable than the evidence of adults.  So if a court proceeds to discount a child’s evidence automatically, without regard to the circumstances of the particular case, it will have fallen into an error.

The second change in the attitude of the law toward the evidence of children in recent years is a new appreciation that it may be wrong to apply adult tests for credibility to the evidence of children.  One finds emerging a new sensitivity to the peculiar perspectives of children.  Since children may experience the world differently from adults, it is hardly surprising that details important to adults, like time and place, may be missing from their recollection.  Wilson J. recognized this in R. v. B. (G.), [1990] 2 S.C.R. 30, at pp. 54-55, when, in referring to submissions regarding the court of appeal judge's treatment of the evidence of the complainant, she said that

... it seems to me that he was simply suggesting that the judiciary should take a common sense approach when dealing with the testimony of young children and not impose the same exacting standard on them as it does on adults. However, this is not to say that the courts should not carefully assess the credibility of child witnesses and I do not read his reasons as suggesting that the standard of proof must be lowered when dealing with children as the appellants submit.  Rather, he was expressing concern that a flaw, such as a contradiction, in a child's testimony should not be given the same effect as a similar flaw in the testimony of an adult. I think his concern is well founded and his comments entirely appropriate. While children may not be able to recount precise details and communicate the when and where of an event with exactitude, this does not mean that they have misconceived what happened to them and who did it. In recent years we have adopted a much more benign attitude to children's evidence, lessening the strict standards of oath taking and corroboration, and I believe that this is a desirable development. The credibility of every witness who testifies before the courts must, of course, be carefully assessed but the standard of the "reasonable adult" is not necessarily appropriate in assessing the credibility of young children.

As Wilson J. emphasized in B. (G.), these changes in the way the courts look at the evidence of children do not mean that the evidence of children should not be subject to the same standard of proof as the evidence of adult witnesses in criminal cases.  Protecting the liberty of the accused and guarding against the injustice of the conviction of an innocent person require a solid foundation for a verdict of guilt, whether the complainant be an adult or a child.  What the changes do mean is that we approach the evidence of children not from the perspective of rigid stereotypes, but on what Wilson J. called a "common sense" basis, taking into account the strengths and weaknesses which characterize the evidence offered in the particular case.

It is neither desirable nor possible to state hard and fast rules as to when a witness's evidence should be assessed by reference to "adult" or "child" standards -- to do so would be to create anew stereotypes potentially as rigid and unjust as those which the recent developments in the law's approach to children's evidence have been designed to dispel. Every person giving testimony in court, of whatever age, is an individual, whose credibility and evidence must be assessed by reference to criteria appropriate to her mental development, understanding and ability to communicate. But I would add this. In general, where an adult is testifying as to events which occurred when she was a child, her credibility should be assessed according to criteria applicable to her as an adult witness. Yet with regard to her evidence pertaining to events which occurred in childhood, the presence of inconsistencies, particularly as to peripheral matters such as time and location, should be considered in the context of the age of the witness at the time of the events to which she is testifying.

Against this background, I turn to a more particular consideration of the Court of Appeal's treatment of the evidence in this case.  First, the Court referred to the fact that "there was really no confirmatory evidence".  This suggests that the Court may have been applying the old rule that the evidence of a child could not found a conviction unless it was confirmed or corroborated by independent evidence.  It may be that in considering the whole of the evidence in accordance with the Yebes test, a court of appeal will take into account, along with other factors, the presence or absence of confirmatory evidence.  So the reference to lack of confirmatory evidence is not in itself an error of law.  But standing as it does as a bald proposition unrelated to a detailed examination of the evidence, it does support the submission that the Court of Appeal was treating the evidence of the children as being inherently less reliable than adult evidence might be.

In any event, the Court of Appeal erred in concluding that there was no supporting evidence.  As the trial judge noted, B.W.'s allegation that the respondent took her into the family's van on a camping trip, and once inside the van, indecently assaulted her, was confirmed to some extent by the evidence of her aunt and S.W., both of whom recalled the respondent taking B.W. to the van on such an outing. In addition, the testimony of M.W. may be considered to support that of S.W. with respect to the wrestling incident S.W. alleged. Although M.W.'s description of the respondent's actions on that occasion differed from S.W.'s, her testimony still confirmed that of S.W. in a number of particulars.

The Court of Appeal next referred to the fact that the evidence of the younger children was fraught with inaccuracy.  This is true, particularly with respect to B.W.'s evidence.  Some of the inconsistencies are minor, for example an error on the distance from a van to a ball game many years ago.  Others are more significant, relating to the sleeping arrangements of the three children, the location of bedrooms in the house and possibly the respondent's nighttime attire.  While it was the proper task of the Court of Appeal to consider such inconsistencies, one finds no mention of the fact that the trial judge was alive to them and resolved them to his satisfaction in his reasons for judgment, nor of the fact that many of the inconsistencies may be explained by reference to the fact that a young child might not be paying particular attention to sleeping arrangements or clothing or that the children had lived in a variety of different arrangements, which might well have given rise to confusion on such details.

Finally, the Court of Appeal relied on the fact that neither of the older children was "aware or concerned that anything untoward occurred which is really the best test of the quality of the acts."  This reference reveals reliance on the stereotypical but suspect view that the victims of sexual aggression are likely to report the acts, a stereotype which found expression in the now discounted doctrine of recent complaint.  In fact, the literature suggests the converse may be true; victims of abuse often in fact do not disclose it, and if they do, it may not be until a substantial length of time has passed.  In any event, the Court of Appeal erred in concluding that there was no complaint. In fact the two older girls did complain to their mother; S.W. about the respondent getting into bed with her, and both M.W. and S.W., it would seem, about the way in which he wrestled with them.

In summary, the Court of Appeal was right to be concerned about the quality of the evidence and correct in entering upon a re-examination and reweighing, to some extent, of the evidence.  It went too far, however, in finding lacunae in the evidence which did not exist and in applying too critical an approach to the evidence, an approach which appears to have placed insufficient weight on the trial judge's findings of credibility, influenced as the Court of Appeal appears to have been by the old stereotypes relating to the inherent unreliability of children's evidence and the "normal" behaviour of victims of sexual abuse.

Placing myself, as I must, in the position of the Court of Appeal (see Criminal Code  s. 695 , and the comments of McIntyre J. in Yebes, supra, at p. 186), I conclude that we are here concerned with verdicts which "a properly instructed jury [or judge], acting judicially, could reasonably have rendered", to repeat the words of Yebes.  I would allow the appeal and restore the convictions.

Appeal allowed.

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

Solicitors for the respondent:  Reynolds Kline Selick, Belleville.