R. v. B. (G.), 1990 CanLII 113 (SCC), [1990] 2 SCR 3

G.B. Appellant

v.

Her Majesty The Queen Respondent

indexed as: r. v. b. (g.)

File No.: 20905.

1989: November 29; 1990: June 7.

Present: Wilson, L'Heureux-Dubé, Gonthier, Cory and McLachlin JJ.

on appeal from the court of appeal for saskatchewan

Criminal law -- Evidence -- Witnesses -- Corroboration -- Evidence of children -- Criminal Code requiring that unsworn evidence of a child be corroborated in a material particular by evidence that implicates the accused --Whether there is corroboration in law if the evidence is corroborated with respect to the occurrence of the event but not with respect to the identity of the accused -- Criminal Code, R.S.C. 1970, c. C-34, s. 586.

Appellant, a young offender, was charged with committing an aggravated sexual assault on the complainant, a kindergarten student who was five years old at the time of the alleged offence. The trial judge found no evidence of wounding and therefore considered only the included offences of sexual assault and sexual assault causing bodily harm. The complainant gave unsworn testimony at the trial. There was corroborating evidence that the assault took place at the time and place the complainant alleged, but there was no evidence, apart from the complainant's identification of appellant, that pointed to appellant's having been the perpetrator of the assault. The Criminal Code provided at the time that no person could be convicted upon the unsworn evidence of a child unless the evidence was "corroborated in a material particular by evidence that implicates the accused." The trial judge found that although there was evidence corroborating the fact that the event described had taken place, there was no independent evidence connecting appellant to the event. He accordingly acquitted him. The Court of Appeal found that the trial judge erred in insisting that the corroborating evidence connect the accused to the commission of the offence, and in failing to find that the complainant's evidence was confirmed in a material particular. It set aside the acquittal and ordered a new trial.

Held: The appeal should be dismissed.

In its decision in Vetrovec this Court effectively put an end to the strict rule in Baskerville, which required that in order for there to be corroboration of unsworn evidence there must be independent evidence that implicates the accused. Although Vetrovec dealt with corroboration of the evidence of accomplices in situations where corroboration is required as a matter of common law, it is clear that this Court has rejected an overly technical approach to corroboration and has returned to a common sense approach which reflects the original rationale for the rule. The requirement in s. 586 of the Criminal Code that the corroborating evidence implicate the accused requires only that the evidence confirm the witness's story in some material particular.

This conclusion is consistent with the language and intent of s. 586. While it is clear that the section requires that there be evidence that implicates the accused, it is not clear whether it is the evidence of the unsworn child or the corroborating evidence that must implicate the accused. Given the semantic difficulties in the section, the legislative intention cannot be determined by looking at the wording alone. The section's purpose, which is to allow the evidence of a witness otherwise feared to be untrustworthy to be given weight, must also be looked at. In order to achieve that purpose, what is required is additional evidence that renders it probable that the complainant's story is true and may be safely acted upon. Provided that the complainant's evidence is corroborated in a material particular, with or without implicating the accused, the veracity of the witness will be strengthened. Since the only evidence implicating the accused in many sexual offences against children will be the evidence of the child, imposing too restrictive a standard on their testimony may permit serious offences to go unpunished and perhaps to continue.

Cases Cited

Considered: Murphy and Butt v. The Queen, 1976 CanLII 198 (SCC), [1977] 2 S.C.R. 603; Vetrovec v. The Queen, 1982 CanLII 20 (SCC), [1982] 1 S.C.R. 811; Warkentin v. The Queen, 1976 CanLII 190 (SCC), [1977] 2 S.C.R. 355; referred to: R. v. Baskerville, [1916] 2 K.B. 658; Paige v. The King, 1948 CanLII 46 (SCC), [1948] S.C.R. 349; R. v. Chayko (1984), 1984 ABCA 80, 12 C.C.C. (3d) 157; R. v. Brasier (1779), 1 Leach 199, 168 E.R. 202; R. v. Silverstone, 1934 CanLII 152 (ON CA), [1934] 1 D.L.R. 726; Hubin v. The King, 1927 CanLII 79 (SCC), [1927] S.C.R. 442; R. v. Parish, 1968 CanLII 120 (SCC), [1968] S.C.R. 466; R. v. McNamara (No. 1) (1981), 1981 CanLII 3120 (ON CA), 56 C.C.C. (2d) 193.

Statutes and Regulations Cited

Act further to amend the Criminal Law, S.C. 1890, c. 37, s. 13.

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

Canada Evidence Act, 1893, S.C. 1893, c. 31, s. 25.

Canada Evidence Act, R.S.C. 1970, c. E-10, s. 16(2).

Criminal Code, 1892, S.C. 1892, c. 29, s. 685.

Criminal Code, R.S.C. 1970, c. C-34, ss. 586, 618(2)(a) [am. 1974-75-76, c. 105, s. 18].

Criminal Code, S.C. 1953-54, c. 51.

Criminal Law Amendment Act, 1885 (U.K.), 48 & 49 Vict., c. 69, s. 4.

Criminal Law Amendment Act, 1975, S.C. 1974-75-76, c. 93, s. 8.

Young Offenders Act, S.C. 1980-81-82-83, c. 110, ss. 27 [am. 1986, c. 32, s. 20], 61(2) [rep. 1986, c. 32, s. 40].

Authors Cited

Canada. Committee on Sexual Offences Against Children and Youths (Badgley Committee). Sexual Offences Against Children in Canada: Summary of the Report of the Committee on Sexual Offences Against Children and Youths. Ottawa: Minister of Supply and Services Canada, 1984.

Canada. Law Reform Commission. Report on Evidence. Ottawa: Information Canada, 1975. Reprinted in (1976), 34 C.R.N.S. 26.

Clarke, Andrew B. "Corroboration in Sexual Cases", [1980] Crim. L.R. 362.

McWilliams, Peter K. Canadian Criminal Evidence, 2nd ed. Aurora, Ontario: Canada Law Book, 1984.

Schiff, Stanley. Evidence in the Litigation Process, vol. 1, 3rd ed. Toronto: Carswells, 1988.

APPEAL from a judgment of the Saskatchewan Court of Appeal (1988), 1988 CanLII 208 (SK CA), 65 Sask. R. 134, allowing the Crown's appeal from appellant's acquittal on charges of sexual assault and sexual assault causing bodily harm. Appeal dismissed.

Donna Taylor, Mervin Ozirny and Wayne Rusnak, for the appellant.

Kenneth W. MacKay, Q.C., for the respondent.

//Wilson J.//

The judgment of the Court was delivered by

Wilson J. -- The appellant is a young offender who was charged with a number of sexual assaults on younger children at Sheho Elementary School in Saskatchewan between September 1985 and May 1986. One of these assaults is the subject of this appeal. A.B., H.H., C.S. and S.S. are other young offenders who were also charged with one or more of the alleged sexual assaults and the circumstances of their involvement will be dealt with in the two related judgments of R. v. B. (G.), [1990] 2 S.C.R. 000 (hereinafter R. v. G.B., A.B. and C.S.) and R. v. B. (G.), [1990] 2 S.C.R. 000 (hereinafter R. v. G.B., C.S., H.H., S.S. and A.B.)

In the first instance trials were conducted in the following three matters: R. v. G.B.; R. v. G.B., A.B. and C.S.; and R. v. G.B., C.S., H.H., S.S., A.B. and P.H. All of the youths were acquitted of all charges by the trial judge, who wrote three separate judgments. Appeals by the Crown to the Saskatchewan Court of Appeal were allowed and the court delivered one judgment encompassing all three appeals. The young offenders now appeal to this Court as of right pursuant to s. 27 of the Young Offenders Act, S.C. 1980-81-82-83, c. 110, as amended, and s. 618(2)(a) of the Criminal Code, R.S.C. 1970, c. C-34 (now R.S.C., 1985, c. C-46, s. 691(2)(a)). Five separate appeals were initially filed but by order of this Court the appeals were consolidated and the appellants were granted leave to file a joint factum.

I find that it is more convenient to deal with the appeals in three separate judgments because of the distinct facts pertaining to each and the different issues raised by the appellants on each appeal. The main issue which arises for determination in this appeal is the issue of corroboration and in particular whether there is corroboration in law if the evidence of an unsworn child is corroborated with respect to the occurrence of the event but not with respect to the identity of the accused. In R. v. G.B., A.B. and C.S., on the other hand, the main issue is with what degree of specificity the time of the offence must be established by the Crown in order to proceed to a determination on the merits. And in R. v. G.B., C.S., H.H., S.S. and A.B., the Court is called upon to clarify the rule applicable to appellate review of an acquittal and to determine whether the Court of Appeal in the circumstances of this case was entitled to interfere with the decision of the trial judge. Each appeal thus raises discrete issues some of which are fairly narrow and technical. The broader question, however, raised by each of them is how the courts should deal with the evidence of child witnesses and, although each judgment will touch upon this question, it will be dealt with most extensively in the related judgment of R. v. G.B., A.B. and C.S.

1. The Facts

The appellant G.B. was charged that:

On or about the 9th day of May A.D. 1986 at Sheho in the Province of Saskatchewan being a young person within the meaning of the Young Offenders Act did in committing a sexual assault on D.M. wound the said D.M. thereby committing an aggravated sexual assault contrary to Section 246.3 of the Criminal Code.

The trial judge found no evidence of wounding and therefore considered only the included offences of sexual assault and sexual assault causing bodily harm.

The complainant, D.M., a kindergarten student at Sheho Elementary School, was five years old at the time of the alleged offence. He gave unsworn testimony at the trial. He testified that during the afternoon recess on May 9, 1986, when he left his classroom to get a drink of water, the appellant grabbed him and carried him to the janitor's room located next to the senior boys' washroom. According to the complainant the appellant then pulled down the complainant's pants, pulled and twisted his penis, injuring him, and leaving him in the janitor's room. There were some discrepancies at trial with respect to the complainant's evidence as to the perpetrator(s) involved but the complainant consistently included the appellant and claimed that he was the one who assaulted him in the janitor's room.

P.M., the older brother of the victim, also gave unsworn evidence at trial. He testified that he had seen someone he could not identify take his brother into the janitor's room. After that person left, P.M. said that he let his brother out of the janitor's room and that D.M. then returned to his classroom.

Mrs. Seidlick, the complainant's kindergarten teacher, testified that on the day in question she noticed the complainant lying down on the carpet in the classroom away from the other students. She also noticed that he was upset and when she tried to find out what was bothering him he complained of soreness. She further testified that the complainant had informed her that the big boys had hurt him and that he had named the appellant as one of the students involved.

D.M's parents testified as to his physical and emotional condition on the evening of the alleged offence. D.M. had complained to his father of a stomachache. The father testified that when he asked his son where it hurt the child indicated the area of his penis. His father examined him and noticed that the penis was swollen and distended. The child was immediately taken to Dr. Raju, who examined him. The doctor testified at trial and stated that when he examined the complainant he found that the child's penis was diffusely swollen and had a three millimetre abrasion inside the foreskin where it joins the glans. The doctor further testified that the injury was fresh, not self-inflicted, and estimated it to have been inflicted within 24 hours of the examination.

There was other Crown evidence including the testimony of Constable Lehman and Mrs. Bolingbrooke, a teacher at the school. Each testified that some time after the event the complainant was taken to the school and when the janitor's room was opened he exhibited signs of terror and fear. The complainant's parents also testified that they had noticed bed-wetting and other behavioural changes in the child.

2. The Issue

The sole issue raised on this appeal is whether there was evidence that the trial judge should have accepted as being capable of corroborating the evidence of the unsworn complainant. The determination of this question turns on the interpretation of s. 586 of the Criminal Code which was in force at the time of trial. It provides:

586. No person shall be convicted of an offence upon the unsworn evidence of a child unless the evidence of the child is corroborated in a material particular by evidence that implicates the accused.

This provision 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. There is currently no statutory requirement for corroboration of an unsworn child's evidence in the Criminal Code. The related provisions of the Canada Evidence Act (s. 16(2), R.S.C. 1970, c. E-10) and the Young Offenders Act (s. 61(2)), which required an unsworn child's evidence to be corroborated by some other material evidence, have also been repealed.

3. The Courts Below

Saskatchewan Provincial Court (Chorneyko Prov. Ct. J., unreported)

The trial judge first stated that the evidence of the kindergarten teacher, parents and doctor satisfied him that the event described by the complainant took place on the date alleged in the information. Having decided that the event occurred Chorneyko Prov. Ct. J. did not rule on the probative value of the evidence of Constable Lehman and Mrs. Bolingbrooke.

The trial judge then turned to the question whether the complainant's unsworn evidence was corroborated in law, stating that this was the critical issue he had to deal with. He found first that the complainant's evidence was the only evidence which implicated the appellant since the evidence of P.M. could not be used for corroborative purposes. He referred to R. v. Baskerville, [1916] 2 K.B. 658 (C.A.), for the principle applicable to the corroboration of unsworn evidence and noted that this principle was adopted in Canada in this Court's decision in Paige v. The King, 1948 CanLII 46 (SCC), [1948] S.C.R. 349. In his view, these cases clearly indicated that in order to constitute corroboration there must not only be independent evidence on some material particular but that such evidence must implicate the accused. The trial judge then considered some later cases such as R. v. Chayko (1984), 1984 ABCA 80, 12 C.C.C. (3d) 157, in which the majority of the Alberta Court of Appeal suggested that the Baskerville test was too narrow. He reviewed the facts of the cases that were alleged to have expanded the test and stated:

Whether one considers the narrow view contained in the Baskerville principle or the wider view in Chayko and Fiddler, in my view, the result is the same. Although Chayko and Fiddler state that independent evidence does not have to show that the accused committed the crime, it is clear that the independent evidence must somehow connect the accused to the event complained of. This was the case in Chayko where the sworn evidence showed the accused buttoning up his trousers, coming from the area where the offences were alleged to have happened and messy hair. All this connects the accused to the event described by the unsworn testimony.

Likewise in Bear, the father of Joey comes to the house where the accused is holding the child in a battered condition. The brother slaps him, and the accused said nothing. Again, this is sworn evidence tending to connect the accused to the event.

In Fiddler, there is the sworn evidence of the accused being near the scene and coupled with his attempt to deny he was near the scene during the critical time. This evidence also connects the accused to the event.

Chorneyko Prov. Ct. J. concluded that in order to have corroboration in law there must be sworn evidence which somehow connected the accused to the event described. Applying this test to the facts he decided that even although there was evidence corroborating the fact that the event had taken place, there was no independent evidence connecting the appellant to the event. He therefore found the appellant not guilty.

Saskatchewan Court of Appeal ((1988), 1988 CanLII 208 (SK CA), 65 Sask. R. 134)

The Court of Appeal (Vancise, Wakeling and Gerwing JJ.A.) unanimously allowed the Crown's appeal and ordered a new trial. Writing for the court Vancise J.A. agreed that the origin of the English rule concerning corroboration is to be found in Baskerville, supra. He went on to state, however, that the narrow interpretation of the Baskerville test was considered and rejected by this Court in Murphy and Butt v. The Queen, 1976 CanLII 198 (SCC), [1977] 2 S.C.R. 603. Murphy and Butt dealt with corroboration of the evidence of a complainant alleging sexual assault under s. 142 of the Criminal Code. After reviewing Murphy and Butt in depth the Court of Appeal then moved to this Court's decision in Vetrovec v. The Queen, 1982 CanLII 20 (SCC), [1982] 1 S.C.R. 811, and noted that Dickson J. in Vetrovec had expressed the view that the inadequacy of the Baskerville test of corroboration was recognized by this Court in Murphy and Butt.

Vancise J.A., relying on the above dicta in Vetrovec and the approach taken to s. 142 of the Criminal Code in Murphy and Butt, concluded that this Court has not accepted the proposition that s. 586 of the Criminal Code embodies the Baskerville rule. The Court of Appeal found therefore that the trial judge was in error in insisting that the corroborating evidence connect the accused to the commission of the offence. It was the view of the court that the appropriate approach was to ask (1) has the child given evidence which implicates the accused? and, if so, (2) is that evidence corroborated in some material particular? A material particular, the court found, is one which tends to confirm that the witness's testimony is true and could be, but need not be, the "particular" of identification. Vancise J.A., applying this approach, found as follows at p. 142:

In this case the child, D.M., gave evidence as to what transpired. That evidence implicated the accused as a perpetrator. The evidence as to what transpired was confirmed in every material respect (except in respect of the identity of the perpetrator) by independent evidence. As the trial judge noted, the evidence of the teacher, the parents and the doctor, confirmed that the story as told by D.M. took place. It was material evidence which elevated the testimony of D.M. and confirmed the story, even though there was no evidence, save that of D.M., pointing to the accused as the perpetrator.

Vancise J.A. concluded that the trial judge erred in law in failing to find that the evidence of D.M. was confirmed in a material particular. Because of this conclusion it was necessary to order a new trial. He went on to state that it did not follow that a conviction must automatically result, only that there was evidence which was capable of constituting corroboration and which must be considered on the question of guilt or innocence.

I note at this point that Wakeling J.A., with Gerwing J.A. concurring, made some additional comments with respect to the helpfulness of expert testimony on the issue of corroboration and the treatment that should be accorded such evidence. However, since this concurrence is raised directly in the matter of R. v. G.B., A.B. and C.S. in the appellants' written argument, I defer my comments and a review of Wakeling J.A.'s reasons until I deal specifically with that appeal.

4. Analysis

The appellant submits that in order to corroborate the complainant's testimony and allow a decision to be reached by the trier of fact on the merits of the case, there must be evidence which in some way connects the accused to the commission of the crime. The appellant does not argue that the corroboration must be direct evidence of the accused's involvement but contends rather that it is not enough for the corroborating evidence to enhance the credibility of the complainant if it does not relate to a material particular implicating the accused. Accordingly, the Court of Appeal erred in holding that the evidence of the complainant's teacher, his parents, and a doctor who examined him, corroborated his evidence as required by s. 586 of the Criminal Code.

The Crown, on the other hand, supports the Court of Appeal's interpretation of the section and argues that the evidence need only corroborate in some material particular the statements of the complainant which implicate the accused. The issue joined on the appeal then is which of these two positions is correct.

I note at the outset that the facts of this case are clear. As found by both the trial judge and the Court of Appeal there is corroborating evidence that the assault took place at the time and place the complainant alleged. However, there is no evidence, apart from the complainant's identification of the appellant, that points to the appellant as having been the perpetrator of the assault. Accordingly, if the appellant's position is correct, the appeal must be allowed; if the Crown's position is correct, the appeal must be dismissed.

I begin with a brief historical overview of corroboration in order to set the stage for the review of the case law and discussion that follows.

(a) Historical Overview

In the early history of the common law children under the age of fourteen were automatically disqualified from giving testimony regardless of their intelligence or religious instruction. This was due to a perceived natural lack of intelligence of young children which would lessen their capacity to understand either what they were testifying about or the nature of the obligations that testifying under oath entailed. However, in the late eighteenth century the law concentrated on the young witness's ability to understand the nature and consequences of an oath and the age limit was dropped as a strict requirement. In R. v. Brasier (1779), 1 Leach 199, 168 E.R. 202, for example, the court held that there was no precise or fixed rule as to the time within which infants must be precluded from giving evidence.

In the late nineteenth century England enacted legislation to permit the receipt of unsworn testimony from children provided "other material evidence" was available to corroborate such testimony (Criminal Law Amendment Act, 1885 (U.K.), 48 & 49 Vict., c. 69, s. 4). Canada adopted similar legislation in 1890 (An Act further to amend the Criminal Law, S.C. 1890, c. 37, s. 13) and in 1893 enacted the forerunner of the recently repealed provision of the Canada Evidence Act (S.C. 1893, c. 31, s. 25) which allowed the unsworn evidence of children to be admitted provided the child possessed sufficient intelligence to justify the reception of the evidence and understood the duty to speak the truth. Under the original Criminal Code of 1892, S.C. 1892, c. 29, s. 685, the unsworn evidence of children could not be admitted unless it was corroborated by "some other evidence in support thereof implicating the accused" and was only allowed for the offences of carnal knowledge and indecent assault. The unsworn evidence of a child in any other case had to be received under the Canada Evidence Act until the Criminal Code of 1954, S.C. 1953-54, c. 51, was enacted. Despite the difference in wording between the Criminal Code and the Canada Evidence Act, the degree of corroboration required was thought to be the same: see R. v. Silverstone, 1934 CanLII 152 (ON CA), [1934] 1 D.L.R. 726 (Ont. C.A.) Thus, for over 100 years the law in Canada has required the unsworn evidence of children to be corroborated.

In recent years there has been considerable criticism of the rule regarding corroboration both in judicial decisions and academic commentaries. In 1976 the Law Reform Commission of Canada expressed the view that the corroboration rule was unnecessary: see Report on Evidence (1975), reprinted in (1976), 34 C.R.N.S. 26, at p. 33. The Badgley Report recommended that the statutory requirement of corroboration of an unsworn child's evidence be abolished and that the weight to be given their evidence should be a matter for the trier of fact: see Sexual Offences Against Children in Canada: Summary of the Report of the Committee on Sexual Offences Against Children and Youths (1984).

The criticism of the rules was not without effect. This Court's decision in Vetrovec, supra, eliminates the common law rule relating to accomplices and the Criminal Code provision dealing with corroboration of a complainant's evidence in rape cases was repealed by the Criminal Law Amendment Act, 1975, S.C. 1974-75-76, c. 93, s. 8. As mentioned, the provision requiring corroboration of proof based on the unsworn evidence of a child has also been repealed and the provisions allowing unsworn children to give evidence have become more liberal. These changes are evidence of the decline in importance of the need for corroboration due to the recognition that the trier of fact is competent to weigh the evidence and credibility of all witnesses. It also reflects a desire to overcome technical impediments in the prosecution of offences. Nevertheless, this appeal must be decided on the basis of the legislative limitations upon an unsworn child's testimony that were in place at the time of trial.

(b) Case Law

Any review of the case law dealing with corroboration must begin with a discussion of Baskerville. At issue in the case was the scope of the rule of practice requiring corroboration of the evidence of an accomplice but Lord Reading C.J. for the court stated that the rule was the same for statutory requirements. Lord Reading pointed out that it was clearly not necessary for the evidence of an accomplice to be confirmed in every detail but that a difference of opinion had arisen as to whether it was essential that the corroborative evidence connect the accused with the crime. After a review of the authorities the court concluded that the better opinion was that the evidence must be confirmed not only as to the circumstances of the crime but also as to the identity of the accused. Quoting from p. 667:

We hold that evidence in corroboration must be independent testimony which affects the accused by connecting or tending to connect him with the crime. In other words, it must be evidence which implicates him, that is, which confirms in some material particular not only the evidence that the crime has been committed, but also that the prisoner committed it. The test applicable to determine the nature and extent of the corroboration is thus the same whether the case falls within the rule of practice at common law or within that class of offences for which corroboration is required by statute. The language of the statute, "implicates the accused," compendiously incorporates the test applicable at common law in the rule of practice. The nature of the corroboration will necessarily vary according to the particular circumstances of the offence charged. It would be in high degree dangerous to attempt to formulate the kind of evidence which would be regarded as corroboration, except to say that corroborative evidence is evidence which shows or tends to show that the story of the accomplice that the accused committed the crime is true, not merely that the crime has been committed, but that it was committed by the accused. [Emphasis added.]

It is the interpretation of this passage which seems to have caused confusion in recent years. The majority of the Alberta Court of Appeal in Chayko found the judgment in Baskerville hard to interpret and noted that different courts have interpreted it in different ways. Kerans J.A., for the majority, relying on the latter portion of the above excerpt, concluded that Lord Reading intended the words "implicates the accused" to refer only to an implication of guilt arising from the evidence that requires corroboration.

The Crown relies heavily on Kerans J.A.'s judgment. It submits that the Baskerville rule is open to two interpretations. The first, or narrow rule, sees corroborative evidence as independent evidence that itself implicates the accused. The second, and considerably broader interpretation, is that if the witness identifies the accused and the evidence of the witness is confirmed in some material particular, then there is corroboration in law of that witness's evidence. The Crown advocates the broader interpretation. However, in my view, support for the broader interpretation is not to be found in Lord Reading's judgment. He made it abundantly clear throughout his reasons that there had to be corroborative evidence as to a material circumstance of the crime and as to the identity of the accused in relation to that crime. In Vetrovec, Dickson J. shared this view, stating at p. 826:

Prior to the judgment of Lord Reading, there had been controversy over whether corroborative evidence must implicate the accused, or whether it was sufficient if it simply strengthened the credibility of the accomplice. Lord Reading settled the controversy in favour of the former view.

In the years following Baskerville the narrow interpretation of the rule was approved in numerous decisions of this Court. One text writer has commented that this Court acted upon the narrow interpretation of the rule on at least fifteen occasions over a period of sixty years: see Schiff, Evidence in the Litigation Process (1988), vol. 1, at p. 613. For example, in Hubin v. The King, 1927 CanLII 79 (SCC), [1927] S.C.R. 442, Anglin C.J., for the Court, stated at p. 444:

Since the decision of the Court of Criminal Appeal in R. v. Baskerville, the requirements of the provision now found in s. 1002 admit of no doubt. The corroboration must be by evidence independent of the complainant; and it "must tend to show that the accused committed the crime charged".

In R. v. Parish, 1968 CanLII 120 (SCC), [1968] S.C.R. 466, this Court, per Ritchie J., again relied on Baskerville when applying the rule for corroboration as set out in the Criminal Code (then s. 134) for evidence given by the complainant in a rape case, the same provision at issue in Hubin. Ritchie J. held that even although the corroborating evidence need not be direct evidence that the accused committed the crime, circumstantial evidence being sufficient, it must nonetheless touch upon the accused's connection with the crime. On the facts of that case the corroborating evidence implicated the accused without touching upon the question whether or not there had been actual intercourse between the accused and the complainant. This was held sufficient to constitute corroboration in law.

There was little controversy as to what the common law rule was or what the statute required until 1976 when this Court delivered two contemporaneous judgments addressing statutory corroboration in the context of rape. As in this case, the provision in question was in effect when the events took place but was repealed prior to this Court's decisions. Several commentators and courts have viewed these cases as leading to the demise of Baskerville.

In Warkentin v. The Queen, 1976 CanLII 190 (SCC), [1977] 2 S.C.R. 355, this Court considered the requirements for statutory corroboration when one of the live issues for determination was the identity of the alleged rapists. Section 142 of the Criminal Code was still in force at the relevant time and provided that there must be a warning to the jury if the evidence of the complainant "is not corroborated in a material particular by evidence that implicates the accused".

In Warkentin, the complainant testified that she had been seized by the three accused and a fourth man, forced into a red Ford Mustang, and taken to an isolated place where she was raped by one of the men while two others held her arms and the fourth stood by laughing. After managing to escape she was picked up by two friends who testified to her distraught condition at the time. The trial judge left the following items with the jury as evidence of corroboration and the accused were convicted: evidence of the association of the accused with the complainant earlier in the evening, the complainant's distraught condition, the finding of sperm on the complainant, the finding of scalp hair that could have come from one of the accused, and the finding of pine needles in the complainant's underclothing that could have come from the scene of the alleged rape.

In this Court the appellants argued that there were three separate issues, namely intercourse, consent, and identity, in respect of each accused, and that the trial judge had erred in failing to distinguish which pieces of evidence were capable of corroborating the evidence of the complainant on each issue. The majority of this Court (Laskin C.J. and Spence, Pigeon and Dickson JJ. dissenting) affirmed the trial judge's decision and concluded that, while none of the items on their own could have provided corroboration, taken together the evidence was capable of establishing each of the elements of the offence and greatly enhanced the credibility of the complainant's evidence. De Grandpré J., for the majority, rejected the view that the identity of each accused had to be corroborated separately in the case of a gang rape. He wrote at p. 379:

I am satisfied that the corroborative evidence of which s. 142 speaks need not identify each accused separately when the evidence to be corroborated is that a gang rape has been committed. It is sufficient to establish that intercourse without consent has taken place and that the group was a party to it. In the same fashion, I cannot accept the submission that the corroborative evidence of s. 142 must be pigeonholed in three different slots, namely intercourse, non-consent and identity. The wording of the section goes against that interpretation. On the text of the article, there is corroboration when the story of the complainant is "corroborated in a material particular by evidence that implicates the accused". It is the entire picture that must be looked at, not a portion thereof.

During the course of his reasons, de Grandpré J. made several statements suggesting that there should be a relaxation of the requirements of corroboration and that corroboration should be treated as a matter of common sense without strict adherence to formalities. He stated at p. 374:

Corroboration is not a word of art. It is a matter of common sense. In recent years, this Court has repeatedly refused to give a narrow legalistic reading of that word and to impose upon trial judges artificial restraints in their instructions to juries or to themselves.

In writing the dissenting opinion, Dickson J. stated that the corroboration required by the section was not corroboration in the sense of evidence which tends merely to confirm or support the story of the complainant. Rather, the section specifies that in order to be corroborative the evidence must be on a material point in the case and, above all, must implicate the accused by connecting or tending to connect him with the alleged offence. Dickson J. also noted that the corroborating evidence must be independent of the acts or words of the complainant.

The Court's judgment in Murphy and Butt, supra, was delivered on the same day, Spence J. writing the majority opinion and Laskin C.J. and Dickson J. dissenting in part. The Crown relies on this case for the proposition that the narrow interpretation of Baskerville has been rejected by this Court. The two appellants had been convicted of rape at trial. Murphy, one of the appellants, admitted having sexual relations with the complainant but claimed it was consensual. Butt, the other appellant, denied that any sexual relations had taken place. The corroborating evidence put forward by the Crown was the police evidence of the complainant's distraught condition shortly after the alleged offence. It was clear that this evidence could corroborate the complainant's absence of consent with respect to Murphy. The difficulty arose as to whether the same evidence could corroborate the complainant's testimony against Butt. Spence J., writing for the majority, found that there was sufficient corroborating evidence as to the intercourse between the complainant and Butt, a conclusion which suggests a departure from the previous strict interpretation of the statutory provision. He stated at pp. 615-16:

It [s. 142 of the Code] is firstly called into force when "the only evidence that implicates the accused is the evidence, given under oath, of the female person in respect of whom the offence is alleged to have been committed". Those conditions do apply to the present case. Then what is required is "that evidence is not corroborated in a material particular by evidence that implicates the accused". So that what is required to be corroborated is a material particular of the evidence of the complainant. The evidence of the complainant was, as I have already recited, that she was taken by these two appellants acting under a pretence of friendship to the apartment which they occupied and then raped first by the appellant Murphy and then by the appellant Butt. It is a material particular of that evidence which must be corroborated. There is no requirement that the whole of her evidence be corroborated. Were that the requirement, there would be no need for even the evidence of the complainant. The so-called corroborative evidence would be sufficient for a conviction.

It is all of that evidence plus the complainant's distraught condition upon which the Crown relies as corroboration of not only Murphy's but Butt's rape of the complainant. The jury were entitled to consider all of that evidence and to come to the conclusion that that evidence with its rather unusual outline of events does corroborate the evidence of the complainant. It was that evidence which the learned trial judge left to the jury as evidence which they might find corroborative of the complainant's testimony.

In my view, the learned trial judge was correct in his conclusion that that evidence was capable of corroborating the complainant's story implicating each of the accused. [Emphasis added.]

Chief Justice Laskin, however, was of a different view. He applied the stricter requirements of s. 142 as had Dickson J. in Warkentin. He concluded that the complainant's distraught condition was not corroborative because it did not implicate Butt. He would have allowed Butt's appeal, stating at p. 606:

I do not quarrel with the proposition that a complainant's hysterical condition or emotional distress, evident after the commission of an alleged rape, may be adduced in evidence as corroboration provided, however, that, in the words of s. 142, it is "evidence that implicates the accused". It cannot be said that evidence of a complainant's hysterical condition or emotional distress is capable of being corroborative at large and that it can be used accordingly against a particular accused simply because the complainant has alleged that he raped her. It may be evidence that is corroborative against an accused who admits sexual intercourse, as Murphy did, but who at the same time alleges consent. In such a case it may properly be regarded as supporting the complaint in a material particular, that is want of consent, and to implicate the accused because of his admission of intercourse. I am unable to understand, however, how a complainant's hysterical condition can implicate an accused in any material particular when he has denied intercourse and there is no other evidence (apart from the complainant's) that can support a finding of intercourse.

Both McWilliams (Canadian Criminal Evidence (2nd ed. 1984)) and Schiff, op. cit., view these two cases as considerably weakening the requirements for statutory corroborative evidence. McWilliams states that the evidence in Warkentin was barely sufficient to be corroborative of identity and that Spence J. in Murphy and Butt construed s. 142 so that it would be sufficient if corroborative evidence went to one of the issues whether that issue was a live one or not. Schiff comments that Murphy and Butt can be taken as the case in which the majority rejected "implicating the accused" as a necessary element of corroboration. A. B. Clarke, an Associate of King's College, London, has also reviewed the two cases and says the Court rejected the Baskerville approach in favour of a search for evidence to support the victim's credibility with the jury: see "Corroboration in Sexual Cases", [1980] Crim. L.R. 362. Clarke writes at p. 363:

The rigid Baskerville formulation has given way to a search for circumstances and other testimony, more consistent with the truth of the victim's statements than their falsity, particularly in the areas of dispute -- a concentration on the victim's story and the support for it.

On the other hand, both cases were analyzed by the Ontario Court of Appeal in R. v. McNamara (No. 1) (1981), 1981 CanLII 3120 (ON CA), 56 C.C.C. (2d) 193, and the court came to the conclusion that in both cases the requirement that the corroborating evidence link the accused in some way to the crime was maintained. Regarding Warkentin the court wrote at pp. 274-75:

In our view, the basis of the majority judgment is that the hair on the complainant's jeans, which was similar to that found on Warkentin, was capable of linking Warkentin to the offence. The admission by all the accused that they had been together earlier in the evening, coupled with the fact that they were arrested together, was capable of linking the other accused to Warkentin at the relevant time and, thus, the evidence was capable of implicating the entire group and each member of it in the commission of the offence. Whether those items, in fact, constituted corroboration was, of course, for the jury to determine.

With respect to Murphy and Butt, the court was of the view that the evidence of Butt and Murphy looking for a prostitute and taking the complainant back to the apartment provided the implicating link to Butt that might not have been provided if the evidence of the complainant's distraught condition had been considered on its own. The court stated at p. 277:

Again, we consider that the difference of opinion reflected in the majority and minority judgments was merely a difference as to whether it was open to the jury to find that the evidence left with them as capable of being corroborative implicated the accused. Both the majority and minority judgments accept the proposition that evidence to be corroborative must implicate the accused in the commission of the offence.

It may be that at first glance Warkentin tends to support the Crown's submission that the element of identity by itself need not be corroborated and that any evidence that makes the complainant's story more credible on any issue is sufficient. However, it is my view that while this may be the correct interpretation of s. 586 of the Criminal Code, the majority decision in Warkentin provides only limited support for it. De Grandpré J. stated that while the circumstantial evidence taken piece by piece did not tend to show that intercourse had taken place without the consent of the complainant and with one or other of the accused, taken in its totality it was capable of establishing the three elements of the offence, including identity. This, I believe, tends to show that the majority believed there was independent evidence corroborative of identity. However, the judgment does suggest a weakening of the strict requirement of evidence corroborative of the accused's connection with the offence.

A broad reading of the judgment of Spence J. in Murphy and Butt leads me to conclude that he thought that any evidence that tended to corroborate in any respect the testimony of the witness whose evidence required corroboration would suffice. Thus, it would be sufficient if the corroborative evidence confirmed the complainant's story implicating the accused. Spence J. did not, however, make it clear whether the complainant's distraught condition was corroborative because it could be taken as implicating Butt or because it raised the credibility of the complainant's story which had implicated Butt. Instead, he referred to all the evidence, i.e. Murphy's admission of intercourse, the fact that both accused had picked up the complainant and taken her to their apartment thinking she was a prostitute, and the complainant's distraught condition, as providing the corroboration.

In my opinion, it is only when the Chief Justice's dissent in Murphy and Butt is compared to the majority judgment that the difference in approaches becomes clear. Laskin C.J. carefully reviewed the Canadian cases in which the complainant's distraught condition was admissible as corroboration implicating the accused and found that in each case the issue was one of consent, not whether the accused had intercourse with the complainant. He thus concluded that there was no evidence capable of corroborating the complainant's evidence with respect to Butt. Applying the strict interpretation of the statutory requirement this conclusion would appear to be correct. It is difficult to see how the evidence of the complainant's distraught condition, even together with the admitted fact that both the accused had picked her up and taken her to her apartment, is of any probative value, when viewed independently of the complainant's testimony, on the issue of the alleged intercourse with Butt. For the majority to reach this result it had to accept that the corroborative evidence implicated Butt because of the evidence of the complainant. Unlike Warkentin, it was not a gang rape situation in which proof against one would be sufficient for the other. In Murphy and Butt there had to be corroboration against each of the accused.

This interpretation of the majority decision in Murphy and Butt was accepted by this Court in its unanimous decision in Vetrovec. At issue in Vetrovec was corroboration of the evidence of accomplices and this Court effectively put an end to the strict rule in Baskerville at least in so far as it applied to the common law, thereby affirming its willingness to depart from its own prior decisions. Dickson J. reviewed the rationale for the rule and concluded that there was no logical need for it. He pointed out that in all categories of witnesses some will be untrustworthy and if the trial judge thinks that the jury should be cautioned as to the evidence of such a witness, he or she may instruct accordingly.

The Court expressed a preference for a common sense approach rather than the overly technical approach in Baskerville. It found at least three problems with Baskerville. The first was that it confuses the reason behind the accomplice warning and prompts the courts to determine whether the corroborative evidence fits the definition rather than deciding whether there is evidence that bolsters the credibility of the accomplice. Secondly, because corroboration became a legal term of art the law in the area became increasingly complex and technical. Thirdly, and most importantly, the Court was of the view that the definition was unsound in principle. Dickson J. stated at p. 826:

With great respect, on principle Lord Reading's approach seems perhaps over-cautious. The reason for requiring corroboration is that we believe the witness has good reason to lie. We therefore want some other piece of evidence which tends to convince us that he is telling the truth. Evidence which implicates the accused does indeed serve to accomplish that purpose but it cannot be said that this is the only sort of evidence which will accredit the accomplice.

In conclusion, the Court held in Vetrovec that there should be no special category for accomplices. An accomplice should be treated like any other witness. While careful to point out that his comments were limited to situations in which corroboration is required as a matter of common law, and even although the focus of the Court's analysis was on the branch of the corroboration doctrine involving accomplices, Dickson J. also wove into the discussion a comment on Murphy and Butt. After quoting the decision of Spence J., he stated at p. 828:

Mr. Justice Spence's approach was to look for evidence which confirmed the story of the complainant. Once the story was confirmed, the complainant could be believed and the accused convicted. Implicit in this approach, it seems to me, is a recognition of the inadequacy of the Baskerville definition of corroboration. Evidence implicating the accused is a possible but not a necessary element for corroboration. Here, even though there was no evidence implicating Butt, there was evidence confirming the story of the complainant and thus it was safe to convict. The important question, as Wigmore pointed out, is not how our trust is restored, but whether it is restored at all. [Emphasis added.]

It seems to me, therefore, that this Court has clearly rejected an ultra technical approach to corroboration and has returned to a common sense approach which reflects the original rationale for the rule and allows cases to be determined on their merits. This was also the view of the majority of the Alberta Court of Appeal in Chayko. Kerans J.A. cited Dickson J.'s comments on Murphy and Butt in Vetrovec and held that the requirement in s. 586 that the corroborating evidence implicate the accused requires only that the evidence confirm in some material particular the story of the witness giving the evidence which required corroboration.

I am, accordingly, in agreement with the Crown's position. In my view, the Court of Appeal was correct to dispose of this appeal on the strength of this Court's statements in Murphy and Butt and Vetrovec. However, I would be remiss if I did not address the interpretation of s. 586 independently of authority in order to confirm that this conclusion is consistent with the language and intent of the legislative provision.

(c) Statutory Interpretation

I reproduce s. 586 in full for ease of reference:

586. No person shall be convicted upon the unsworn evidence of a child unless the evidence of the child is corroborated in a material particular by evidence that implicates the accused.

Turning first to a literal interpretation, the section speaks of corroboration (a) in a material particular and refers to (b) evidence which implicates the accused. The appellant strongly contends that the language of the section compels one to conclude that for the corroboration requirement to be met there must be some link between the corroborating evidence and the accused. I do not agree. In my view, while it is clear that the section requires that there be evidence that implicates the accused, it is not clear whether it is the evidence of the unsworn child in need of corroboration that must implicate the accused or whether it is the corroborating evidence that must implicate the accused. In Chayko, Kerans J.A., when dealing with the Baskerville definition of corroboration which contains substantially the same wording as s. 586, also found ambiguity. He stated at p. 168:

This is compounded by a semantic problem with the words "which implicate the accused". These can refer to an implication of guilt which rests only on the corroborating evidence independently of the evidence needing corroboration. Alternatively, they can refer to an implication of guilt which arises from the evidence needing corroboration and which is now reliable because corroborated.

The semantic difficulties in the section are highlighted if one rewords the section in line with the two possible interpretations. If the wording of the section were "corroborated . . . by evidence that implicates the accused" without the intervening words "in a material particular", a literal interpretation would support the appellant's position. But if the appellant's interpretation is correct, one is forced to conclude that the intervening words "in a material particular" are redundant because, if the corroborative evidence implicates the accused, it will inevitably be corroboration in a material particular. The identity of the accused is always material.

On the other hand, to support the Crown's position, the only wording that is necessary is "corroborated in a material particular". If the section stopped there, its interpretation would not be contentious. However, the section continues and adds "by evidence that implicates the accused". Nonetheless, the addition of these words does not create the same redundancy as does interpreting the provision in accordance with the appellant's position. The extra words are necessary if what is required is that the evidence of the unsworn child implicate the accused. However it does create a redundancy of a different kind since there can be no conviction unless the accused person is implicated by some evidence: indeed, implicated beyond a reasonable doubt.

Given the interpretive problems in the language of the section, it is difficult to discern the intention of the legislature by looking at the wording alone. I think we have to apply a purpose approach. In my view, the purpose of s. 586 is to allow the evidence of a witness, otherwise feared to be untrustworthy, to be given weight. Another way of formulating the purpose is to state that the section is designed to ensure that no accused will be convicted on the basis of testimonial evidence that is by its very nature unreliable. In order to achieve the section's purpose, therefore, what is required is additional evidence that renders it probable that the complainant's story is true and may safely be acted upon. Provided that the complainant's evidence is corroborated in a material particular, with or without implicating the accused, the veracity of the witness will be strengthened. A good example in this case would be the evidence of the doctor that the child had indeed been sexually assaulted.

Also in favour of the liberal interpretation are the presumptions that the law does not require the impossible and the legislator intends only what is just and reasonable. Since the only evidence implicating the accused in many sexual offences against children will be the evidence of the child, imposing too restrictive a standard on their testimony may permit serious offences to go unpunished and perhaps to continue. Moreover, it is reasonable to assume that the legislator did not intend an accused to benefit from the youthful age of his victim by placing unnecessary impediments in the way of prosecuting offences against small children.

Nevertheless, I am fully conscious of the fact that a strict interpretation of s. 586 provides a strong protection against false convictions since establishing the identity of the perpetrator is a necessary pre-condition to any criminal prosecution. However, after weighing the considerations on either side of the issue I have concluded that of the two possible interpretations, each of which can be supported on the wording of the section, the preferable interpretation is that s. 586 of the Criminal Code requires the evidence of the complainant which implicates the accused to be corroborated in a material particular by other independent testimony.

5. Disposition

I would, accordingly, dismiss the appeal.

Appeal dismissed.

Solicitors for the appellant: Rusnak, Balacko, Kachur & Rusnak, Yorkton, Saskatchewan.

Solicitor for the respondent: The Attorney General for Saskatchewan, Regina.