R. v. Starr, 2000 SCC 40

R. v. Starr, [2000] 2 S.C.R. 144

Robert Dennis Starr Appellant

v.

Her Majesty The Queen Respondent

and

The Attorney General of Canada, the Attorney General for

Ontario and the Attorney General of British Columbia Interveners

Indexed as: R. v. Starr

Neutral citation: 2000 SCC 40.

File No.: 26514.

Hearing: December 3, 1998.

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

Rehearing: February 24, 2000; September 29, 2000.

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

on appeal from the court of appeal for manitoba

Criminal law -- Evidence -- Hearsay -- Admissibility -- Present intentions exception -- Accused convicted of first degree murder -- Trial judge admitting statement of intention made by deceased -- Whether statement admissible under “present intentions” exception to hearsay rule -- Whether statement admissible under principled approach to hearsay admissibility -- Whether exceptions to hearsay rule must conform to principled approach to hearsay admissibility -- Whether hearsay that fits within traditional exception may still be inadmissible if it is not sufficiently reliable and necessary.

Criminal law -- Evidence -- Hearsay -- Admissibility -- Prior identification exception -- Accused convicted of first degree murder -- Trial judge admitting police officers’ testimony regarding out‑of‑court identification by witness -- Witness not testifying at trial that she had made out‑of‑court identification -- Whether police evidence admissible under traditional exception to hearsay rule -- Whether police evidence admissible under principled approach to hearsay admissibility.

Criminal law -- Charge to jury -- Reasonable doubt -- Accused convicted of first degree murder -- Whether trial judge explained concept of reasonable doubt to jury in adequate manner.

The accused was convicted of two counts of first degree murder. He had been accused of shooting C and W by the side of a highway. C and W had been drinking with the accused in a hotel. Outside, C and W offered a couple a ride home in W’s station wagon. W drove, and the group first stopped at an adjacent gas station, where G, a sometime girlfriend of C, approached the station wagon and had a conversation with C. During the conversation, G observed a car beside the gas station, and saw the accused in the car. She became angry with C because he was out with W rather than her, and she walked away from the car. C got out of the car and followed her into a laneway, where they had a further conversation. G asked C why he would not come home with her. According to G, C replied that he had to “go and do an Autopac scam with Robert”. She understood “Robert” to be the accused. A day or two later, G saw a picture in the newspaper of what she believed was the car in which she had seen the accused. The car had been found at the scene of the murder. She phoned the police and told them she had seen the car on the night of the murders at the gas station, with the accused in it. The Crown’s theory was that the killing was a gang‑related execution perpetrated by the accused. W was an unfortunate witness who was killed simply because she was in the wrong place at the wrong time. The theory was that the accused had used an Autopac scam as a pretext to get C out into the countryside. The trial judge found that G’s anticipated testimony regarding the scam was admissible under the “present intentions” or “state of mind” exception to the hearsay rule.

Two police officers visited the couple who had been given a ride. One of the officers testified that the wife, B, had told them that she had seen a man talking to C at the gas station. The officer testified that B indicated that the man in one of the photographs she was shown looked like the man whom she had seen at the gas station talking to C and who was also “probably driving the other car”. The photograph was a photo of the accused. Following a voir dire the trial judge ruled that the officers’ anticipated testimony was admissible pursuant to the prior identification exception to the hearsay rule, notwithstanding the fact that B had not testified at trial as to having seen a man talking to C at the gas station, or as to having identified that man in one of the photographs presented to her by the police.

The Court of Appeal, in a majority decision, upheld the convictions. At issue here is whether the court erred in affirming the trial judge’s decision to admit G’s testimony regarding a statement of intention made by the deceased C, in affirming the trial judge’s decision to admit the testimony of the police officers regarding B’s out‑of‑court identification and in finding that the trial judge had explained the concept of reasonable doubt to the jury in an adequate manner.

Held (McLachlin C.J. and L’Heureux‑Dubé, Gonthier and Bastarache JJ. dissenting): The appeal should be allowed and a new trial ordered.

Per Iacobucci, Major, Binnie, Arbour and LeBel JJ.: Since C’s out‑of‑court statement to G that he had to “go and do an Autopac scam with Robert”, meaning the accused, is sought to be adduced in order to prove the truth of its contents, G’s testimony regarding C’s statement to her is hearsay and would generally be inadmissible as such. The “state of mind” or “present intentions” exception to the hearsay rule as it has developed in Canada permits the admission into evidence of statements of intent or of other mental states for the truth of their contents and also, in the case of statements of intention in particular, to support an inference that the declarant followed through on the intended course of action, provided it is reasonable on the evidence for the trier of fact to infer that the declarant did so. A statement of intention cannot be admitted to prove the intentions of someone other than the declarant, unless a hearsay exception can be established for each level of hearsay. The trial judge erred in admitting C’s statement to G under the present intentions exception and, having admitted it, in not limiting its use by the jury. The statement contained no indicia of reliability since it was made under circumstances of suspicion. C may have had a motive to lie in order to make it seem that he was not romantically involved with W, and could easily point to the accused, who was sitting nearby in a car but out of earshot, as being the person with whom he was going to do an Autopac scam. Moreover, the trial judge failed to instruct the jury that the statement was only admissible as evidence regarding the intentions of C, not the accused. It is well‑established that when a piece of evidence may conceivably be put to both proper and improper uses, the trial judge in a criminal case must give the jury a limiting instruction regarding the permissible inferences that may be drawn from the evidence. In this case the trial judge did not instruct the jury on the proper uses of C’s statement; in fact, he did the opposite by expressly inviting the jury to use the evidence to infer the accused’s intentions. In so doing, he clearly committed an error of law warranting reversal. Finally, when properly limited, the evidence was more prejudicial than probative. The trial judge erred by not considering whether the prejudicial effect of the prohibited use of the evidence overbears its probative value on the permitted use. The impermissible inferences that the jury might well have drawn from C’s statement are that the accused was in the car that followed C, that the accused was alone in the car (since C referred only to the accused), and that the accused went with C as part of a plan to lure C to a secluded area and kill him. The prejudicial effect of the admission of C’s statement outweighed the statement’s probative value. The statement ought to have been excluded on this basis as well.

In Khan, Smith, and subsequent cases, this Court allowed the admission of hearsay not fitting within an established exception where it was sufficiently reliable and necessary to address the traditional hearsay dangers. This concern for reliability and necessity should be no less present when the hearsay is sought to be introduced under an established exception. This is particularly true in the criminal context given the fundamental principle of justice, protected by the Canadian Charter of Rights and Freedoms, that the innocent must not be convicted. It would compromise trial fairness, and raise the spectre of wrongful convictions, if the Crown is allowed to introduce unreliable hearsay against the accused, regardless of whether it happens to fall within an existing exception. In addition to improving trial fairness, bringing the hearsay exceptions into line with the principled approach will also improve the intellectual coherence of the law of hearsay. To the extent that the various exceptions may conflict with the requirements of a principled analysis, it is the principled analysis that should prevail. It is nevertheless important for a court to exercise a certain degree of caution when reconsidering the traditional exceptions, which continue to play an important role under the principled approach. In some rare cases, it may be possible under the particular circumstances of a case for evidence clearly falling within an otherwise valid exception nonetheless not to meet the principled approach’s requirements of necessity and reliability. In such a case, the evidence would have to be excluded. However, these cases will no doubt be unusual, and the party challenging the admissibility of evidence falling within a traditional exception will bear the burden of showing that the evidence should nevertheless be inadmissible. The trial judge will determine the procedure (whether by voir dire or otherwise) to determine admissibility under the principled approach’s requirements of reasonable necessity and reliability.

C’s statement to G was also inadmissible under the principled approach. Since it was made under “circumstances of suspicion”, the statement was not reliable. Nor are there any other circumstantial guarantees of trustworthiness that could render the statement reliable. Having found that the statement is unreliable, it is unnecessary to go on to ask whether it was necessary or not. Since it does not fall under an existing exception either, the courts below erred in admitting this evidence. There was no serious argument that the error was one that could be saved by the curative proviso.

The trial judge erred in admitting the police evidence regarding B’s out‑of‑court identification under a traditional exception to the hearsay rule. Under the “prior identification” exception, prior statements identifying or describing the accused are admissible where the identifying witness identifies the accused at trial, or where the identifying witness is unable to identify the accused at trial, but can testify that he or she previously gave an accurate description or made an accurate identification. These requirements are not satisfied in the circumstances of this case. Since B did not identify the accused in court, only the second branch of the exception could possibly be applicable. However, B did not testify that she could not remember whether the accused was the person whom she identified. She was not asked to compare the accused with her recollections about the person she saw on the night of the murders. Accordingly, the underlying circumstances of necessity required to trigger the second branch of the traditional exception did not exist. Moreover, the police evidence went far beyond the scope of the “prior identification” exception. The officers’ testimony went beyond simply asserting that B had identified the accused, and provided almost the entirety of the narrative underlying the identification. The police testimony was equally inadmissible under the principled approach. The testimony was not necessary, since B was a witness at trial and could have provided first‑hand evidence, had the Crown chosen to question her on the point. Furthermore, there are strong indications that B’s identification was unreliable. The trial judge’s cautionary instruction to the jury was insufficient to remedy the harm caused by the admission of the evidence.

The reasonable doubt instruction given in this case falls prey to many of the same difficulties outlined in Lifchus, and likely misled the jury as to the content of the criminal standard of proof. The key difficulty with this instruction is that it was not made clear to the jury that the Crown was required to do more than prove the accused’s guilt on a balance of probabilities. The trial judge told the jury that they could convict on the basis of something less than absolute certainty of guilt, but did not explain, in essence, how much less. In addition, rather than telling the jury that the words “reasonable doubt” have a specific meaning in the legal context, the trial judge expressly instructed the jury that the words have no “special connotation” and “no magic meaning that is peculiar to the law”. By asserting that absolute certainty was not required, and then linking the standard of proof to the “ordinary everyday” meaning of the words “reasonable doubt”, the trial judge could easily have been understood by the jury as asserting a probability standard as the applicable standard of proof. The trial judge did refer to the Crown’s onus and to the presumption of innocence, and he stated that the accused should receive the benefit of any reasonable doubt. The error in the charge is that the jury was not told how a reasonable doubt is to be defined. As was emphasized repeatedly in Lifchus and again in Bisson, a jury must be instructed that the standard of proof in a criminal trial is higher than the probability standard used in making everyday decisions and in civil trials. In this case the jury was not told that something more than probability was required in order to convict, and nearly all of the instructions they were given weakened the content of the reasonable doubt standard in such a manner as to suggest that probability was indeed the requisite standard of proof. The reasonable likelihood that the jury applied the wrong standard of proof raises a realistic possibility that the accused’s convictions constitute a miscarriage of justice.

Per L’Heureux‑Dubé and Gonthier JJ. (dissenting): The following framework of analysis should be adopted for hearsay statements. First, it must be determined whether the statement is hearsay. Second, the trial judge should determine whether the hearsay statement falls within an established exception to the hearsay rule. Third, if the evidence does not fall within an established exception, the trial judge should determine whether it would still be admissible under the principled approach. Fourth, the trial judge maintains the limited residual discretion to exclude evidence where the risk of undue prejudice substantially exceeds the evidence’s probative value. Finally, once the statements are found admissible, it is for the trier of fact to weigh the evidence and make a determination as to the ultimate reliability of the hearsay evidence at issue. The case‑by‑case application of the principled approach to statements falling within accepted exceptions to the rule against hearsay should not be countenanced. Individual cases may illuminate or illustrate the need to modify a particular traditional exception, but every piece of evidence that falls within a traditional exception should not be subjected to the principled approach and the concomitant voir dire that it may entail.

While it is the duty of the courts to review common law rules, this duty carries with it a corresponding responsibility to proceed prudently. The usurpation of the traditional hearsay exceptions by the principled approach is not warranted under the current state of our law. Our jurisprudence has recognized the need to relax the hearsay rule to keep it in step with our changing society, specifically our greater appreciation of jurors’ abilities. However, the existing common law exceptions should be retained. The principled necessity‑reliability analysis, while appropriate where hearsay evidence does not fall within an established exception to the hearsay rule, has not replaced and should not supplant the traditional exceptions to the hearsay rule.

The traditional exceptions are built upon a determination that a threshold of reliability is met in particular instances of hearsay statements. Reliability under the principled approach is similarly restricted to a threshold inquiry. In re‑evaluating a hearsay exception, the court must ask whether some reason exists to doubt that the rationale underlying the exception applies in certain circumstances. The scope of this inquiry is restricted; the court should not consider every possible permutation of indicators of reliability or unreliability. The only time a court should entertain a challenge to an existing exception is where there are facts, generally applicable to a class of persons, which weaken the theoretical justification for the exception. While in principle the trial judge’s residual discretion to exclude admissible evidence where its prejudicial effect substantially outweighs its probative value could, in appropriate circumstances, apply to evidence falling within an exception to the hearsay rule, this point was not addressed in the Court of Appeal and was not argued before this Court.

Since the Crown sought to use C’s statement to G as proof of the truth of its contents, it is hearsay. The statement falls squarely within the “present intentions” exception to the hearsay rule. An exception to the hearsay rule arises when the declarant’s statement is adduced in order to demonstrate the intentions, or state of mind, of the declarant at the time when the statement was made. A requirement that the statement must appear to have been made in a natural manner and not under circumstances of suspicion should not be added. While statements of intention may be admissible despite the fact that they refer to a joint act, the “present intentions” exception may not be used to infer that a third party acted in accordance with the declarant’s stated intention. The trial judge did not err in leaving C’s statement of intention to the jury. Properly cautioned by the trial judge, juries are perfectly capable of determining what weight ought to be attached to such evidence, and of drawing reasonable inferences therefrom. The Crown did not ask the jury to draw an impermissible inference, and the trial judge did not err in instructing the jury on this issue.

With respect to B’s out‑of‑court identification, while the trial judge may have committed an error by allowing the police officers to testify as to the out‑of‑court identification, any harm that such evidence may have caused was effectively negated by the trial judge’s instructions to the jury. The harm caused by the introduction of the evidence was minimal given other testimony that placed the accused at the gas station before the murders. B’s statement that “he was probably driving the other car” could not have suggested that B identified the accused in the town where the murders took place, rather than at the gas station. Given the trial judge’s admonition to the jury and the accused’s admission of the marginal role that the identification evidence played, there is no reasonable possibility that the verdict would have been different if the alleged error had not been made.

While trial judges are now expected to follow the Lifchus model charge, failure to do so in cases tried before Lifchus does not constitute reversible error if the charge conveys to the jury the special meaning attached to reasonable doubt. Despite not having the benefit of this Court’s decision in Lifchus, the trial judge included most of the suggested elements in his charge to the jury. The charge instructed the jurors that a reasonable doubt must not be an imaginary or frivolous doubt; that it is a doubt based upon reason and common sense; that it must be based on the evidence that the jurors heard in the courtroom; and that the Crown is not required to prove its case to absolute certainty. The trial judge’s charge was not flawless, in that the trial judge erroneously stated that the words “reasonable doubt” are used in their everyday, ordinary sense and have no special legal meaning. The verdict ought not to be disturbed, however, because the charge, when read as a whole, makes it clear that the jury could not have been under any misapprehension as to the correct burden and standard of proof to apply. The charge communicated clearly to the jury that they could not find the accused guilty on a balance of probabilities. Moreover, the charge made it clear to them that the standard of proof beyond a reasonable doubt is inextricably linked to the presumption of innocence and that this burden never shifts to the accused. In light of the trial judge’s compliance with the bulk of the principles enunciated in Lifchus, the charge was not automatically vitiated by the failure to include a specific item mentioned in Lifchus or by the inclusion of an improper item.

Per McLachlin C.J. and Bastarache J. (dissenting): L’Heureux‑Dubé J.’s reasons on the issues of B’s out‑of‑court identification and the charge to the jury on reasonable doubt were agreed with, as were her findings that the victim C’s statement that he intended to do an Autopac scam with the accused later the night of the murder was admissible and that the trial judge’s charge to the jury adequately warned them of the dangers associated with this evidence.

The following principles govern the admissibility of hearsay evidence: (1) Hearsay evidence is admissible if it falls under an exception to the hearsay rule; (2) The exceptions can be interpreted and reviewed as required to conform to the values of necessity and reliability that justify exceptions to the hearsay rule; (3) Where the evidence is admissible under an exception to the hearsay rule, the judge may still refuse to admit the evidence if its prejudicial effect outweighs its probative value; (4) Where evidence is not admissible under an exception to the hearsay rule, the judge may admit it provided that necessity and reliability are established. C’s statement that he intended to do an Autopac scam with the accused later that night is a statement of present intention. Statements of present intention presented for the truth of their contents (i.e., to permit inferences as to what the person in fact did) are admissible, provided they were not made in circumstances of suspicion. There were no circumstances of suspicion here that precluded the trial judge from admitting C’s statement that he was doing an Autopac scam with the accused later that night. The statement should accordingly be admitted as evidence of what the deceased intended to do at the time he made the statement. While a statement of joint intention cannot support an inference as to the state of mind of the third party, in some circumstances it can be fairly considered along with other evidence in deciding what the third party did. C’s statement may thus be viewed as one piece of circumstantial evidence supporting the inference that the accused was with C later that night. It may be that where the only source of inference as to the third party’s conduct is the statement of joint intention, it would be unsafe to permit the jury to rely on it for that purpose. When this occurs, the jury should be so directed. This was not such a case; the statement was merely one of a matrix of circumstances that the jury could consider in determining whether the accused met C later that night as C stated was their common intention. Accordingly, the trial judge was not required to tell the jury that they could not consider the statement on the question of what the accused in fact did. While a more complete warning as to the danger of drawing inferences on actual conduct from statements of joint intention would have been preferable, the trial judge’s instruction to the jury that it was “for [them] to decide whether the evidence of [C’s] statement about the scam goes as far as the Crown would have [them] believe” could have left the jury in no doubt that they must not facilely jump from C’s statement to the conclusion that C and the accused actually met later that evening. The matter was fairly put to the jury and no new trial is required on this ground.

Cases Cited

By Iacobucci J.

Applied: R. v. Lifchus, [1997] 3 S.C.R. 320; referred to: R. v. Khan, [1990] 2 S.C.R. 531; R. v. Smith, [1992] 2 S.C.R. 915; R. v. Bevan, [1993] 2 S.C.R. 599; R. v. P. (R.) (1990), 58 C.C.C. (3d) 334; R. v. Tat (1997), 1997 2234 (ON CA), 117 C.C.C. (3d) 481; R. v. Abbey, [1982] 2 S.C.R. 24; R. v. B. (K.G.), [1993] 1 S.C.R. 740; R. v. Sharp, [1988] 1 W.L.R. 7; Giles v. United States, 432 A.2d 739 (1981); United States v. Brown, 490 F.2d 758 (1973); People v. Madson, 638 P.2d 18 (1981); Shepard v. United States, 290 U.S. 96 (1933); R. v. Carter, [1982] 1 S.C.R. 938; R. v. D. (L.E.), [1989] 2 S.C.R. 111, rev’g (1987), 1987 2536 (BC CA), 20 B.C.L.R. (2d) 384; R. v. Corbett, [1988] 1 S.C.R. 670; R. v. Rockey, [1996] 3 S.C.R. 829; R. v. Seaboyer, [1991] 2 S.C.R. 577; R. v. U. (F.J.), [1995] 3 S.C.R. 764; R. v. Hawkins, [1996] 3 S.C.R. 1043; R. v. Kelly (1999), 1999 13120 (NB CA), 213 N.B.R. (2d) 1; R. v. R. (D.) (1995), 1995 3972 (SK CA), 98 C.C.C. (3d) 353; R. v. Grand‑Pierre (1998), 1998 13202 (QC CA), 124 C.C.C. (3d) 236; R. v. Bisson (1997), 1997 10791 (QC CA), 114 C.C.C. (3d) 154; R. v. Chahley (1992), 1992 2387 (BC CA), 72 C.C.C. (3d) 193; Wepruk (Guardian ad litem of) v. McMillan Estate (1993), 1993 6857 (BC CA), 77 B.C.L.R. (2d) 273; R. v. Crossley (1997), 1997 3280 (BC CA), 117 C.C.C. (3d) 533; R. v. Collins (1997), 1997 12516 (BC CA), 118 C.C.C. (3d) 514; R. v. Warner (1994), 1994 842 (ON CA), 94 C.C.C. (3d) 540; R. v. Leipert, [1997] 1 S.C.R. 281; R. v. Mills, [1999] 3 S.C.R. 668; R. v. L. (J.W.) (1994), 1994 8734 (ON CA), 94 C.C.C. (3d) 263; R. v. Tam (1995), 1995 16084 (BC CA), 100 C.C.C. (3d) 196; R. v. Rose (1998), 1998 4564 (BC CA), 108 B.C.A.C. 221; R. v. C. (B.) (1993), 1993 8564 (ON CA), 12 O.R. (3d) 608; Idaho v. Wright, 497 U.S. 805 (1990); R. v. Bisson, [1998] 1 S.C.R. 306; R. v. W. (D.), [1991] 1 S.C.R. 742.

By L’Heureux‑Dubé J. (dissenting)

R. v. Khan, [1990] 2 S.C.R. 531; R. v. Lifchus, [1997] 3 S.C.R. 320; R. v. Smith, [1992] 2 S.C.R. 915; R. v. Levogiannis, [1993] 4 S.C.R. 475; R. v. Seaboyer, [1991] 2 S.C.R. 577; Ares v. Venner, [1970] S.C.R. 608; Myers v. Director of Public Prosecutions, [1965] A.C. 1001; R. v. B. (K.G.), [1993] 1 S.C.R. 740; R. v. Terry, [1996] 2 S.C.R. 207; Watkins v. Olafson, [1989] 2 S.C.R. 750; R. v. Salituro, [1991] 3 S.C.R. 654; R. v. Hawkins, [1996] 3 S.C.R. 1043; R. v. Chahley (1992), 1992 2387 (BC CA), 72 C.C.C. (3d) 193; R. v. Collins (1997), 1997 12516 (BC CA), 118 C.C.C. (3d) 514; R. v. McKenzie (1986), 1986 1194 (BC CA), 32 C.C.C. (3d) 527; R. v. P. (R.) (1990), 58 C.C.C. (3d) 334; R. v. Moore (1984), 1984 3542 (ON CA), 15 C.C.C. (3d) 541, leave to appeal refused, [1985] 1 S.C.R. x; M. (L.N.) v. Green (1995), 1995 2629 (BC SC), 14 B.C.L.R. (3d) 194; Neufeld v. Child and Family Services of Winnipeg (1993), 1993 14942 (MB QB), 89 Man. R. (2d) 150; R. v. Jack (1992), 1992 2764 (MB CA), 70 C.C.C. (3d) 67; R. v. McLeod (1991), 1991 903 (BC CA), 6 B.C.A.C. 223; R. v. Miller (1991), 1991 2704 (ON CA), 5 O.R. (3d) 678; R. v. Carriere (1986), 1986 4898 (MB CA), 40 Man. R. (2d) 150; R. v. Baron von Lindberg (1977), 1977 237 (BC SC), 66 B.C.L.R. 277; R. v. Flemming (1980), 43 N.S.R. (2d) 249; R. v. Jackson (1980), 1980 2945 (ON CA), 57 C.C.C. (2d) 154; Hildebrand v. Butler (1979), 1979 320 (BC SC), 11 B.C.L.R. 234; R. v. Strongquil (1978), 1978 1815 (SK CA), 43 C.C.C. (2d) 232; R. v. Baltzer (1974), 1974 1668 (NS CA), 10 N.S.R. (2d) 561; R. v. Bencardino (1973), 1973 804 (ON CA), 2 O.R. (2d) 351; Re Grant Estate, 1970 1217 (BC SC), [1971] 1 W.W.R. 555; Great West Uranium Mines Ltd. v. Rock Hill Uranium Ltd., 1955 211 (SK CA), [1955] 4 D.L.R. 307; Gray v. New Augarita Porcupine Mines Ltd., 1952 322 (UK JCPC), [1952] 3 D.L.R. 1; R. v. Wysochan (1930), 1930 483 (SK CA), 54 C.C.C. 172; R. v. Covlin, 1923 620 (AB SCAD), [1923] 3 W.W.R. 555; R. v. Ferry (1992), 1992 15048 (ON SC), 18 C.R. (4th) 221; Re J.B. (1998), 1998 18071 (NL CA), 166 Nfld. & P.E.I.R. 1; Collins v. Collins (1993), 1993 8273 (NL SC), 114 Nfld. & P.E.I.R. 215; R. v. Mafi (1998), 1998 6221 (BC CA), 21 C.R. (5th) 139; Home v. Corbeil, 1955 395 (ON SC), [1955] 4 D.L.R. 750; Mutual Life Insurance Co. v. Hillmon, 145 U.S. 285 (1892); United States v. Pheaster, 544 F.2d 353 (1976); People v. Alcalde, 148 P.2d 627 (1944); R. v. Bisson, [1998] 1 S.C.R. 306; R. v. W. (D.), [1991] 1 S.C.R. 742; R. v. Tombran (2000), 2000 2688 (ON CA), 47 O.R. (3d) 182.

By McLachlin C.J. (dissenting)

Mutual Life Insurance Co. v. Hillmon, 145 U.S. 285 (1892); R. v. Rockey, [1996] 3 S.C.R. 829.

Statutes and Regulations Cited

Criminal Code, R.S.C., 1985, c. C‑46, s. 686(1)(b)(iii).

Evidence Act 1995 (Austl.), 1995, No. 2, ss. 59‑75.

Federal Rules of Evidence, Rules 105, 801 et seq.

Authors Cited

Archibald, Bruce P. “The Canadian Hearsay Revolution: Is Half a Loaf Better Than No Loaf at All?” (1999), 25 Queen’s L.J. 1.

Blok, Murray B. “The Changed Law of Hearsay Evidence (Or, “Hearsay Today, Gone Tomorrow”)” (1993), 51 Advocate 675.

Canada. Law Reform Commission. Report on Evidence. Ottawa: The Commission, 1975.

Carter, P. B. “Hearsay: Whether and Whither?” (1993), 109 L.Q.R. 573.

Choo, Andrew L.‑T. Hearsay and Confrontation in Criminal Trials. Oxford: Clarendon Press, 1996.

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

Cross on Evidence, 7th ed. By the late Sir Rupert Cross and Colin Tapper. London: Butterworths, 1990.

Delisle, R. J. “B. (K.G.) and Its Progeny” (1998), 14 C.R. (5th) 75.

Delisle, R. J. “R. v. Smith: The Relevance of Hearsay” (1991), 2 C.R. (4th) 260.

Ireland. Law Reform Commission. “The Rule Against Hearsay”. In Reports, vol. 2. Dublin: The Commission, 1979, 105.

Kiesel, Diane. “One Person’s Thoughts, Another Person’s Acts: How the Federal Circuit Courts Interpret the Hillmon Doctrine” (1984), 33 Cath. U. L. Rev. 699.

MacCrimmon, Marilyn T. “Developments in the Law of Evidence: The 1990‑91 Term — Social Science, Law Reform and Equality” (1992), 3 Supreme Court L.R. (2d) 269.

Maguire, John MacArthur. “The Hillmon Case — Thirty‑Three Years After” (1925), 38 Harv. L. Rev. 709.

Marin, André. “How to Assess Reliability in Khan and K.G.B. Applications” (1996), 38 Crim. L.Q. 353.

McCormick on Evidence, 5th ed. St. Paul, Minn.: West Publishing Co., 1999.

McCrea, Patricia McNeill. “Judicial Law‑Making: The Development of the Principled Exception to the Hearsay Rule — Implications for Preliminary Hearing Recantations” (1998), 61 Sask. L. Rev. 199.

McWilliams, Peter K. Canadian Criminal Evidence, vol. 1, 3rd ed. Aurora, Ont.: Canada Law Book, 1988 (loose‑leaf updated April 2000, release 24).

Morgan, Edmund M. “Hearsay Dangers and the Application of the Hearsay Concept” (1948), 62 Harv. L. Rev. 177.

Oleskiw, Diane. “Recent Developments in the Law of Hearsay” (1994), 1 Crown’s Newsletter 37.

Ontario. Law Reform Commission. Report on the Law of Evidence. Toronto: Ministry of the Attorney General, 1976.

Paciocco, David M. “The Supreme Court of Canada and Hearsay: The Relevance for Arbitration”. In William Kaplan, Jeffrey Sack and Morley Gunderson, eds., Labour Arbitration Yearbook 1994‑95. Toronto: Lancaster House, 1995, 123.

Phipson on Evidence, 12th ed. By John Huxley Buzzard, Richard May and M. N. Howard. London: Sweet & Maxwell, 1976 (supplemented to 1980).

Phipson on Evidence, 15th ed. London: Sweet & Maxwell, 2000.

Prithipaul, Ravi. “Observations on the Current Status of the Hearsay Rule” (1997), 39 Crim. L.Q. 84.

Rosenberg, Marc. “B. (K.G.) — Necessity and Reliability: The New Pigeon‑holes” (1993), 19 C.R. (4th) 69.

Rosenberg, Marc. “Developments in the Law of Evidence: The 1992‑93 Term — Applying the Rules” (1994), 5 Supreme Court L.R. (2d) 421.

Rowsell, Derek. “Necessity and Reliability: What is the Impact of Khan on the Admissibility of Hearsay in Canada?” (1991), 49 U.T. Fac. L. Rev. 294.

Schiff, Stanley. Evidence in the Litigation Process, vol. 1, 4th ed. Scarborough, Ont.: Carswell, 1993.

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

Stewart, Hamish. “Prior Identifications and Hearsay: A Note on R. v. Tat” (1998), 3 Can. Crim. L. Rev. 61.

Stuesser, Lee. “R. v. B. (K.G.): Prior Inconsistent Statements as Truth” (1991), 5 C.R. (4th) 373.

Then, Edward. “Dying Declarations Following Khan and Smith: Are They Necessarily Reliable?”. In National Criminal Law Program: Criminal Evidence, vol. 1. Vancouver: Federation of Law Societies of Canada, 1994, section 6.5.

Thompson, D. A. Rollie. “The Supreme Court Goes Hunting and Nearly Catches a Hearsay Woozle” (1995), 37 C.R. (4th) 282.

United Kingdom. Law Commission. Consultation Paper No. 138. Criminal Law -- Evidence in Criminal Proceedings: Hearsay and Related Topics. London: HMSO, 1995.

Watt, David. Watt’s Manual of Criminal Evidence. Scarborough, Ont.: Carswell, 1999.

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

Wigmore, John Henry. Evidence in Trials at Common Law, vol. 5. Revised by James H. Chadbourn. Boston: Little, Brown, 1974.

Wigmore, John Henry. Evidence in Trials at Common Law, vol. 6. Revised by James H. Chadbourn. Boston: Little, Brown, 1976.

Wood, Josiah. “Hearsay -‑ Necessity and Reliability” (1997), 20 Prov. Judges J. 5.

Younger, Irving. An Irreverent Introduction to Hearsay. Chicago: American Bar Association, Section of Litigation, 1977.

APPEAL from a judgment of the Manitoba Court of Appeal (1998), 1998 6760 (MB CA), 123 C.C.C. (3d) 145, 123 Man. R. (2d) 292, 159 W.A.C. 292, [1998] M.J. No. 80 (QL), dismissing the accused’s appeal from his conviction on two counts of first degree murder. Appeal allowed and new trial ordered, McLachlin C.J. and L’Heureux‑Dubé, Gonthier and Bastarache JJ. dissenting.

G. Greg Brodsky, Q.C., and Anthony H. Dalmyn, for the appellant.

Gregg Lawlor, for the respondent.

Bernard Laprade and Silvie Kovacevich, for the intervener the Attorney General of Canada.

Written submissions only by Jamie C. Klukach, for the intervener the Attorney General for Ontario.

Alexander Budlovsky and Marian K. Brown, for the intervener the Attorney General of British Columbia.

The reasons of McLachlin C.J. and Bastarache J. were delivered by

1The Chief Justice (dissenting) – I agree with Justice L’Heureux-Dubé on the issues of Ms. Ball’s out-of-court identification and the charge to the jury on reasonable doubt. I also agree that the victim Cook’s statement that he intended to do an Autopac scam with the accused later the night of the murder was admissible and that the trial judge’s charge to the jury adequately warned them of the dangers associated with this evidence. Accordingly, I would dismiss the appeal. However, I would deal with the hearsay issue relating to the victim’s statement of intention somewhat differently than either of my colleagues.

2In my view, the following principles govern the admissibility of hearsay evidence:

1. Hearsay evidence is admissible if it falls under an exception to the hearsay rule;

2. The exceptions can be interpreted and reviewed as required to conform to the values of necessity and reliability that justify exceptions to the hearsay rule;

3. Where the evidence is admissible under an exception to the hearsay rule, the judge may still refuse to admit the evidence if its prejudicial effect outweighs its probative value;

4. Where evidence is not admissible under an exception to the hearsay rule, the judge may admit it provided that necessity and reliability are established.

3In short, the common law exceptions to the hearsay rule remain the law, as interpreted and updated to conform to the twin requirements of necessity and reliability. Additionally, evidence not falling within an exception may be admitted if the requirements of necessity and reliability are established. This retains the certainty and predictability associated with the common law exceptions to the hearsay rule and avoids the need to hold a voir dire when evidence falls within an established exception. At the same time, it permits the exceptions to evolve and evidence outside the exceptions to be admitted where necessity and circumstantial guarantees of reliability exist. As with all evidence, the trial judge has an overriding discretion to exclude the evidence if its prejudicial effect outweighs its probative value.

4Applying these rules to this case, the first question is whether an established exception to the hearsay rule applies to the evidence at issue. The answer in this case is yes. The victim’s statement that he intended to do an Autopac scam with the accused later that night is a statement of present intention. Statements of present intention constitute a long-recognized exception to the rule against admitting hearsay evidence.

5The next question concerns the ambit of the exception of statements of present intention. Here my colleagues differ. L’Heureux-Dubé J. holds that all statements of present intention are admissible. Iacobucci J., following Professor Wigmore’s formulation of the exception, holds that statements of present intention are admissible, unless made in circumstances of suspicion. Canadian law has not yet settled this question.

6Interpreting the exception in light of the underlying requirements of necessity and reliability, I share Iacobucci J.’s view that statements of present intention presented for the truth of their contents (i.e., to permit inferences as to what the person in fact did) are admissible, provided they were not made in circumstances of suspicion. Contemporaneity is cited as providing a guarantee of trustworthiness for statements of present intention. In the normal course, the words are contemporaneous with a present intention to do that act. If a person as she heads out the door says, “I’m going to the store”, there is every reason to believe that is what she intends to do. This flows from the fact that in the great majority of cases, people making such statements actually intend to do the indicated act. The statement of intention is admitted as a statement of mental condition: inferences that may be drawn as to whether the intended act occurred are another matter: Wigmore on Evidence, vol. 6 (Chadbourn rev. 1976), at §1725, p. 139; R. J. Deslisle: “R. v. Smith: The Relevance of Hearsay” (1991), 2 C.R. (4th) 260, at p. 264.

7The reason statements of present intention are generally reliable indicators of the speaker’s “present” or contemporaneous state of mind was captured by Gray J. in Mutual Life Insurance Co. v. Hillmon, 145 U.S. 285 (1892), at p. 299, quoting Beasley C.J. in Hunter v. State, 40 N.J.L.:

In the ordinary course of things, it was the usual information that a man about leaving home would communicate, for the convenience of his family, the information of his friends, or the regulation of his business. At the time it was given, such declarations could, in the nature of things, mean harm to no one; he who uttered them was bent on no expedition of mischief or wrong, and the attitude of affairs at the time entirely explodes the idea that such utterances were intended to serve any purpose but that for which they were obviously designed. . . . At the time the words were uttered. . . the reference to the companion who was to go with him was nothing more, as matters then stood, than an indication of an additional circumstance of his going.

8Sometimes, however, statements of intention may not reflect the actual present intention of the speaker. The circumstances may suggest that the speaker had reason to lie about his or her intentions. In such cases, the circumstantial guarantee of trustworthiness that underlies the exception disappears. This is why Wigmore held that the exception did not apply where the circumstances cast suspicion on whether the statement of intention represented the speaker’s actual intention. Where circumstances of suspicion exist, the presumption of reliability that normally underlies the exception is removed, and it would be inconsistent with the principles underlying the exceptions to the hearsay rule to admit the evidence. (Given the presumption of reliability underlying the exception, it falls to the person opposing the evidence to show circumstances of suspicion.) For this reason, I agree with Iacobucci J. that the statements of present intention are admissible, absent circumstances of suspicion.

9This brings us to the question of whether there were circumstances of suspicion here that precluded the trial judge from admitting the victim’s statement that he was doing an Autopac scam with the accused later that night. The majority of the Court of Appeal held there were not. Iacobucci J., by contrast, finds there were. I find myself in agreement with the Court of Appeal. The victim Cook, who had another woman and two other people in his car, encountered his girlfriend at a service station. They conversed. He told her that he was going to do an Autopac scam with the accused later that night. They parted. The question is whether the circumstances of this encounter cast suspicion on whether the victim’s statement represented his actual intention at the time he spoke. It is argued that the circumstances suggest that he may have made up the Autopac scam story to explain or offset the presence of another woman in his car. It is not clear to me how the statement that he intended to do an Autopac scam with the accused explains why another woman would be in the car with him. Nor is it clear how the statement countered the suggestion that the victim was romantically involved with the woman in the car. It begs the question, “If you are doing an Autopac scam with someone else later tonight, why are you with this woman?” If the victim had said he intended to do the scam with the woman, it might have attracted suspicion as to whether it represented his actual intentions. But that is not what occurred. Accordingly, I would admit the statement as evidence of what the deceased intended to do at the time he made the statement.

10Iacobucci J. argues that the statement is not admissible for a second reason -- because the statement involves the intention to do a joint act with a third person, the accused. In my view, this goes not to admissibility, but to the inferences which may be drawn from the statement. However, my colleague goes on to argue that no inference as to what the accused did can be drawn from the statement, and that the jury should have been so instructed.

11It is common ground that the trial judge must give the jury appropriate instructions on how the jury may use hearsay evidence: R. v. Rockey, [1996] 3 S.C.R. 829, at para. 38. It is also beyond dispute that if failure to give such instructions renders the trial unfair, a new trial must be ordered.

12Two issues arise: (1) Should the trial judge have told the jury they could not draw an inference from Cook’s statement as to the accused’s conduct? and (2) If not, should the trial judge have specifically warned the jury about the danger of making such an inference?

13On the first issue, my colleagues suggest that a jury can never infer what a person other than the speaker did from a statement of joint intention. I would not state the matter so categorically. Certainly, such a statement cannot support an inference as to the state of mind of the third party. However, in some circumstances the statement of joint intention can be fairly considered along with other evidence in deciding what the third party did. The declarant’s state of mind may be one piece of evidence amongst others for the jury to consider in determining what happened. Viewed thus, Cook’s statement may be viewed as one piece of circumstantial evidence supporting the inference that Starr was with the deceased later that night. The cases to which we have been referred, other than some American cases, do not go so far as to say that statements of joint intention can never be used on the issue of what the third party did.

14This said, it may be that where the only source of inference as to the third party’s conduct is the statement of joint intention, it would be unsafe to permit the jury to rely on it for that purpose. When this occurs, the jury should be so directed. In my opinion, this was not such a case; the statement was merely one of a matrix of circumstances that the jury could consider in determining whether the accused met Cook later that night as Cook stated was their common intention. Accordingly, I conclude that the trial judge was not required to tell the jury that they could not consider the statement on the question of what the accused in fact did.

15The second issue is whether the trial judge in any event should have specifically warned the jury about the danger of using the statement of joint intention as evidence on the issue of what the accused did. To answer this question, we must return to what the trial judge said. The Crown had suggested that the statement of joint intention supported an inference that the accused proposed the scam to the victim as a way to get the victim in an isolated place. After this, the trial judge told the jury that it was “for [them] to decide whether the evidence of Cook’s statement about the scam goes as far as the Crown would have [them] believe”.

16An express instruction about the danger involved in drawing inferences from one person’s statement of present intention to what another person did is generally desirable. However, the issue on appeal is whether the trial judge’s instruction in fact conveyed to the jury the need to be careful and not to jump facilely from Cook’s statement that he was doing an Autopac scam with the accused that night to the conclusion that the accused in fact met Cook at the proposed site of the scam. I am satisfied that the trial judge’s direction accomplished this. The trial judge told the jury that they could use the statement for a hearsay purpose, i.e., in deciding what the accused did. He went on to suggest that they might not wish to draw this inference, notwithstanding the Crown’s invitation. The jury thus knew the inference was for them and them alone and that it was one they must carefully consider. While a more complete warning as to the danger of drawing inferences on actual conduct from statements of joint intention would have been preferable, I am satisfied that the trial judge’s instruction in this case could have left the jury in no doubt that they must not facilely jump from Cook’s statement to the conclusion that Cook and the accused actually met later that evening. The matter was fairly put to the jury and no new trial is required on this ground. On this view, either no error was committed, or if it was, it does not constitute reversible error.

17I conclude that the victim Cook’s statement that he intended to do an Autopac scam with the accused was admissible as evidence of the victim’s present intention, and that the trial judge’s instructions sufficed in the circumstances to obviate the danger of impermissible inferences.

18In the result, I agree with L’Heureux-Dubé J. that the appeal should be dismissed.

The reasons of L’Heureux-Dubé and Gonthier JJ. were delivered by

L’Heureux-Dubé J. (dissenting) --

I. Introduction

19The appellant was convicted at trial before a judge and jury on two counts of first degree murder. He was convicted of killing Bernard (Bo) Cook and Darlene Weselowski by shooting them by the side of a provincial highway on the outskirts of Winnipeg in the early morning hours of August 21, 1994. The Manitoba Court of Appeal ((1998), 1998 6760 (MB CA), 123 C.C.C. (3d) 145) upheld the verdict, dismissing the grounds of appeal raised before us. While this case originally came to us as of right, the Court ordered a re-hearing of the appeal and invited submissions on the scope of the present intentions exception to the hearsay rule and on the application of the “principled approach” as set out in R. v. Khan, [1990] 2 S.C.R. 531, and subsequent cases to the traditional exceptions to the hearsay rule. This case also requires us to explicate and apply the principles regarding a charge to the jury on reasonable doubt as set out by this Court in R. v. Lifchus, [1997] 3 S.C.R. 320.

20I have had the benefit of reviewing the reasons of Iacobucci J. and I am unable to agree with his analysis of the issues. While we are both in agreement that the principled approach applies to all hearsay statements, we have very different conceptions as to what this entails.

21Our difference of opinion can be reduced to two crucial points. First, whereas Iacobucci J. would hold that the principled approach must “prevail” over the traditional exceptions to the hearsay rule, I would not endorse such an approach because I believe that it invites challenges to every piece of evidence that falls under a traditional exception. I would hold that the traditional exceptions endure, subject to their categorical re-examination under the principled approach should the issue arise in a proper case. Second, beneath this first disagreement lie two very different conceptions of what the threshold reliability analysis under the principled approach entails. As demonstrated by his analysis of the hearsay statements in this case, Iacobucci J. would require the trial judge to determine the credibility of the hearsay declarant at the threshold stage. I believe that such determinations encroach on the trier of fact’s determination of ultimate reliability. Moreover, I believe that the traditional exceptions already incorporate a finding of threshold reliability and the trial judge need not engage in the type of reliability determination favoured by Iacobucci J. on a “threshold” inquiry of evidence that falls within a traditional exception to the hearsay rule.

22The end result of Iacobucci J.’s approach is an open invitation to challenge previously admissible hearsay evidence. I do not believe that this approach flows from our previous decisions in Khan, supra, and R. v. Smith, [1992] 2 S.C.R. 915. These cases were not about the exceptions to the hearsay rule but about expanding the scope of admissible evidence beyond the boundaries of the traditional categories. They accepted the traditional exceptions as a given but sought other means to admit reliable and necessary evidence through the adoption of a “principled approach” to hearsay. I fear that to adopt the course charted by Iacobucci J. would sacrifice the experience, certainty and predictability of centuries of jurisprudence in the name of a quest for purported intellectual coherence that is untested by the forges of our courtrooms.

23On the issue of the jury charge on reasonable doubt, the Lifchus standard provides guidelines for trial judges rather than an iron-clad roster of proscriptions and prohibitions. I do not read Lifchus as providing that the inclusion or exclusion of certain phrases automatically vitiates a jury charge. Instead, I read Lifchus as mandating that a charge be examined in its entirety to determine whether the jury properly understood the concept of proof beyond a reasonable doubt. I am of the opinion that the jury charge in this case properly communicated this concept to the jury.

24My colleague Iacobucci J. has adequately set forth the factual background and judicial history of this case. In dealing with the individual grounds for appeal, I will elaborate when necessary.

II. Analysis

A. The Admissibility of Cook’s Statement

(1) Introduction and Summary

25Approximately one hour before he was murdered, Bernard (Bo) Cook spoke with Jodie Giesbrecht and told her that he had to “go and do an Autopac scam with Robert” (the appellant). The trial judge admitted Ms. Giesbrecht’s testimony as to what Mr. Cook told her the night he and Darlene Weselowski were murdered. The appellant challenges the admission of this statement on the grounds that it is inadmissible hearsay.

26Iacobucci J. and I disagree regarding the admission of this statement as an exception to the rule against hearsay for declarations of “state of mind” or “present intentions” as recognized by this Court in Smith, supra. For ease of reference, I will refer to this exception as the “present intentions” exception as that term best encapsulates the type of statement at issue before us. Several points of disagreement arise in the course of our analysis of this statement. The first is the scope of this exception; the second is the proper use to which the statement can be put by the jury.

27The third level of disagreement involves the question of how to analyze the hearsay statement. Should it be examined under the traditional exceptions to the hearsay rule or under the new principled approach developed by this Court in Khan and subsequent cases? This question in turn requires us to determine the present status of the traditional exceptions in light of and in relationship to the principled approach.

28I wish to articulate clearly my position on the state of hearsay law after Khan and its progeny. First, I do not believe that Khan “prevails” over the traditional exceptions to the hearsay rule. The effect of Iacobucci J.’s analysis is to reduce these exceptions to mere reference points that provide little more than a convenient launching pad for the application of the principled approach to previously accepted exceptions to the hearsay rule. Second, I would not follow Iacobucci J.’s suggested approach because in my mind it impermissibly shifts responsibility from the trier of fact onto the judge. I believe that this framework directly contradicts the underlying essence of the principled approach and the thrust of hearsay reform in the common law world over the last 30 years. An overly strict view of hearsay is inconsistent with our desire to provide the trier of fact with more, rather than less, evidence. I of course recognize that some restrictions on the admissibility of evidence are nonetheless necessary, for both judges and juries.

29I would adopt the following framework of analysis for hearsay statements. First, it must be determined whether the statement is hearsay. Second, the trial judge should determine whether the hearsay statement falls within an established exception to the hearsay rule. If it does, the evidence is admissible. Third, if the evidence does not fall within an established exception, the trial judge should determine whether it would still be admissible under the principled approach. Fourth, the trial judge maintains the limited residual discretion to exclude evidence where the risk of undue prejudice substantially exceeds the evidence’s probative value. Finally, once the statements are found admissible, it is for the trier of fact to weigh the evidence and make a determination as to the ultimate reliability of the hearsay evidence at issue. The Chief Justice essentially adopts this framework in para. 2 of her reasons.

30In addition, I would allow for the re-evaluation, in a proper case, of a traditional exception under the principled approach. That is to say, I believe that the principled approach may be applied to reconsider the reliability and necessity of the class of statements that are included in a categorical exception to the rule against hearsay. I would not countenance the case-by-case application of the principled approach to statements falling within accepted exceptions to the rule against hearsay. Individual cases may illuminate or illustrate the need to modify a particular traditional exception, but every piece of evidence that falls within a traditional exception should not be subjected to the principled approach and the concomitant voir dire that it may entail. To do so would unnecessarily complicate the trial process and sacrifice experience, certainty and predictability in the name of the vague and uncertain mantra of “principle”.

(2) Hearsay Reform

31The rule against hearsay developed at the same time as the modern form of trial and is associated with a deep-seated distrust of the jury system: Cross on Evidence (7th ed. 1990), at p. 510. It is premised on a belief that the jury will erroneously assess the probative value of evidence and the retention of the rule reflects continued suspicions about jury deliberations: J. Sopinka, S. N. Lederman and A. W. Bryant, The Law of Evidence in Canada (2nd ed. 1999), at p. 176. The rule against hearsay is “founded on a lack of faith in the capacity of the trier of fact properly to evaluate evidence of a statement”: Smith, supra, at p. 935. The modern trend to reform the rules of evidence in general and the hearsay rules in particular constitutes a sharp break from this distrust of the jury. As the Law Reform Commission of Canada explained in its 1975 Report on Evidence, at p. 5:

. . . at least 95% of criminal cases in Canada are now tried by a judge alone, so there is no need for all the precautions about what is to be allowed in evidence now. Judges are experienced in weighing evidence and, in any event, they have to learn of disputed evidence to determine whether it is admissible or not. Even when there is a jury, the situation is profoundly different from the past. Jurors today are far more sophisticated and better educated than in the past. [Emphasis added.]

We have recognized these changes, commenting in R. v. Levogiannis, [1993] 4 S.C.R. 475, at p. 487, that “the recent trend in courts has been to remove barriers to the truth-seeking process”. This Court has taken a flexible approach to the rules of evidence, “reflect[ing] a keen sensibility to the need to receive evidence which has real probative force in the absence of overriding countervailing considerations”: R. v. Seaboyer, [1991] 2 S.C.R. 577, at p. 623. In the specific context of hearsay evidence, Lamer C.J. speaking for a unanimous Court in Smith, supra, at p. 932, explained that “[t]he movement towards a flexible approach was motivated by the realization that, as a general rule, reliable evidence ought not to be excluded simply because it cannot be tested by cross-examination.” Our motivation in reforming the rules of evidence has been “a genuine attempt to bring the relevant and probative evidence before the trier of fact in order to foster the search for truth”: Levogiannis, supra, at p. 487. These principles must guide all of our evidentiary reform endeavours.

32Our reform of the hearsay rules began in Ares v. Venner, [1970] S.C.R. 608, when we refused to follow the House of Lords’ position that the hearsay rule was frozen in time subject to reform only through the initiative of Parliament: see Myers v. Director of Public Prosecutions, [1965] A.C. 1001. In Ares, at p. 624, we adopted the minority position of Lord Donovan in Myers that “[t]he common law is moulded by the judges and it is still their province to adapt it from time to time so as to make it serve the interests of those it binds.” What is at issue in this appeal is whether Khan and its progeny have followed Lord Donovan’s imperative to “adapt” the common law from time to time or whether this Court has embarked upon an entirely different course and revolutionized the law of hearsay in this country.

33In Khan, supra, McLachlin J. (as she then was), writing for a unanimous Court, held that hearsay evidence of a child’s statement on crimes committed against the child should be received, provided that the guarantees of necessity and reliability are met. The effect and reach of Khan were unclear to many with some commentators opining that the decision was analogous to Ares, supra, in creating a new categorical exception to the hearsay rule. See, e.g., D. Rowsell, “Necessity and Reliability: What is the Impact of Khan on the Admissibility of Hearsay in Canada?” (1991), 49 U.T. Fac. L. Rev. 294, at p. 295. However, this interpretation of Khan was expressly disavowed by this Court in Smith, supra.

34In Smith, at p. 932, Lamer C.J., speaking also for a unanimous Court, stated that “Khan should not be understood as turning on its particular facts, but, instead, must be seen as a particular expression of the fundamental principles that underlie the hearsay rule and the exceptions to it”. At p. 930, he had said:

The principled basis of the hearsay rule, and its exceptions, was thus understood by commentators on the common law of evidence early in this century. The decision of this Court in Khan, therefore, should be understood as the triumph of a principled analysis over a set of ossified judicially created categories.

Later at p. 932, Lamer C.J. characterized Khan as signalling a departure “from a view of hearsay characterized by a general prohibition on the reception of such evidence, subject to a limited number of defined categorical exceptions” and moving towards “an approach governed by the principles which underlie the rule and its exceptions alike”. Finally, at p. 933, Lamer C.J. stated that Khan “signalled an end to the old categorical approach to the admission of hearsay evidence. Hearsay evidence is now admissible on a principled basis, the governing principles being the reliability of the evidence, and its necessity.”

35The next year, in R. v. B. (K.G.), [1993] 1 S.C.R. 740, this Court held that prior inconsistent statements may be admitted for the truth of their contents if they are necessary and reliable. The Court established an elaborate voir dire requirement which has spawned much commentary on how to show reliability and necessity. See R. J. Delisle, “B. (K.G.) and Its Progeny” (1998), 14 C.R. (5th) 75, at p. 75; P. M. McCrea, “Judicial Law-Making: The Development of the Principled Exception to the Hearsay Rule -- Implications for Preliminary Hearing Recantations” (1998), 61 Sask. L. Rev. 199, at p. 208; D. A. R. Thompson, “The Supreme Court Goes Hunting and Nearly Catches a Hearsay Woozle” (1995), 37 C.R. (4th) 282, at p. 284; J. Wood, “Hearsay -- Necessity and Reliability” (1997), 20 Prov. Judges J. 5; A. Marin, “How to Assess Reliability in Khan and K.G.B. Applications” (1996), 38 Crim. L.Q. 353.

36The attempts at explanation made in Smith and B. (K.G.) left open the question of where the traditional exceptions stood in the aftermath of Khan and its progeny. Courts and commentators remained uncertain over the application of the principled approach to the traditional categories.

(3) Relationship Between the Principled Approach and the Hearsay Exceptions

37Iacobucci J. takes the position that the traditional hearsay exceptions should not automatically be applied without consideration of the principled approach. He states that in the event of a conflict between the principled approach and the existing exceptions, “it is the principled approach that must prevail” (para. 155). I do not quarrel with this idea but I do take issue with what Iacobucci J. asserts that this entails. At bottom, I read Iacobucci J.’s approach as diminishing the role of the traditional exceptions in the hearsay analysis. I do not believe that such an approach is warranted either by our jurisprudence or by the underlying rationale of the principled approach.

38It is possible to identify five different interpretations of Khan and its progeny: (1) Khan created a particular new exception to the rule against hearsay for child testimony; (2) Khan provides the basis for creating new categorical exceptions for necessary and reliable evidence; (3) Khan created a residual exception for necessary and reliable evidence, leaving the traditional exceptions intact; (4) Khan not only permits the admission of evidence which does not fall within the existing exceptions but allows courts to revisit a traditional exception to determine whether it can still be justified on the basis of necessity and reliability; (5) Khan abolishes the traditional exceptions and supplants them with the principled approach of necessity and reliability which are applied on a case-by-case basis. I will examine each possibility in turn.

39The first interpretation of Khan can be dealt with quickly. The reading of Khan that argued that it created a narrow exception for children’s testimony was expressly disavowed by this Court in Smith, supra, at p. 932, and B. (K.G.), supra, at p. 798.

40The second interpretation of Khan views “necessity” and “reliability” as the basis for creating new categorical exceptions. See Thompson, supra, at p. 284: “Despite the language of ‘principled flexibility’, two of the three cases so far have led to the creation of new categorical exceptions: child abuse hearsay in Khan and some prior inconsistent statements in K.G.B.” (emphasis in original). In B. (K.G.), at p. 798, we similarly expressly repudiated this possibility: “the decisions in Khan and Smith established that Canadian courts will no longer carve out categorical ‘exceptions’”.

41The third interpretation of Khan sees it as creating a new residual exception for “necessary” and “reliable” evidence that does not fall within the traditional categories. This was first expressed by M. T. MacCrimmon in “Developments in the Law of Evidence: The 1990-91 Term -- Social Science, Law Reform and Equality” (1992), 3 Supreme Court L.R. (2d) 269, at p. 322, where she stated that Khan “created a residual exception to the hearsay rule similar to that in the Federal Rules of Evidence”. This interpretation is also adopted by Sopinka, Lederman and Bryant, supra, at p. 191. By far, this is the most widely accepted interpretation of Khan. See McCrea, supra, at p. 208; R. Prithipaul, “Observations on the Current Status of the Hearsay Rule” (1997), 39 Crim. L.Q. 84, at p. 92; Thompson, supra, at p. 284; D. Oleskiw, “Recent Developments in the Law of Hearsay” (1994), 1 Crown’s Newsletter 37, at p. 38; M. B. Blok, “The Changed Law of Hearsay Evidence (Or, “Hearsay Today, Gone Tomorrow”)” (1993), 51 Advocate 675, at p. 685; H. Stewart, “Prior Identifications and Hearsay: A Note on R. v. Tat” (1998), 3 Can. Crim. L. Rev. 61, at p. 64; L. Stuesser, “R. v. B. (K.G.): Prior Inconsistent Statements as Truth” (1991), 5 C.R. (5th) 373, at p. 378.

42The fourth possibility is that Khan permits the re-evaluation of the traditional exceptions on a categorical basis. This was suggested by Marc Rosenberg, before his appointment to the bench, in the article “B. (K.G.) — Necessity and Reliability: The New Pigeon-holes” (1993), 19 C.R. (4th) 69, at p. 71, where he said, “[m]oreover, I think that not only is it possible to admit evidence which does not fall within the existing exceptions but it must also be open to the courts to revisit any established exception to determine whether that exception can still be justified on the basis of necessity and reliability.” For an application of this approach see Rowsell, supra, at pp. 304-8 (re-examining dying declarations and spontaneous declarations in light of Khan). Rosenberg explains the basis for his assertion as follows (at pp. 80-81):

. . . the reason which justifies an existing exception may in time be found to be wanting. The exceptions were created as the courts from time to time applied common sense and experience, but ideas which seemed reasonable in the 19th century may appear questionable in the late 20th century.

I would accept this approach as dictated by the tenets embodied in the principled approach. Yet, as I will explain below, I have a radically different conception than Iacobucci J. of what this entails.

43The final possibility is that Khan and its progeny abolish the traditional exceptions and replace them with a case-by-case determination of reliability and necessity. That the passages previously quoted from Smith, supra, may provide this impression is explained by Professor Schiff in Evidence in the Litigation Process (4th ed. 1993), vol. 1, at p. 660:

Upon even a careful reading of Smith, the court might appear to abolish the existing hearsay exceptions, leaving in their place the judge’s authority to admit any item of hearsay evidence shown to be “necessary” and “reliable” in the particular circumstances. The existing exceptions are nonetheless alive and well. . . .

In Sopinka, Lederman and Bryant, supra, at p. 193, the authors state that “[t]here is no suggestion in the cases that the court was countenancing the exclusion of hearsay which would be otherwise admitted through one of the recognized exceptions.” Furthermore, in B. (K.G.), supra, Lamer C.J. speaking for a majority of the Court, stated (at p. 798): “While the decisions in Khan and Smith established that Canadian courts will no longer carve out categorical ‘exceptions’, the new approach shares the same principled basis as the existing exceptions.” I take this statement as affirming the continued existence of the traditional exceptions while laying the groundwork for the application of the principled approach to new developments. See McCrea, supra, at p. 208. Moreover, we have admitted evidence based solely on the traditional exceptions in cases subsequent to Khan: see R. v. Terry, [1996] 2 S.C.R. 207, at para. 28. In that case, with no discussion of the principled approach, we stated: “An admission against interest made by the accused is admissible as a recognized exception to the hearsay rule, provided that its probative value outweighs its prejudicial effect.”

44I am able to find only a single author who proposes sweeping aside the existing exceptions in the name of the principled approach. However, that author is hardly complimentary in his assessment of this approach. In “The Supreme Court of Canada and Hearsay: The Relevance for Arbitration”, in W. Kaplan, J. Sack and M. Gunderson, eds., Labour Arbitration Yearbook 1994-95, 123, Professor D. M. Paciocco states (at p. 139):

In summary, the Supreme Court of Canada has rejected the traditional approach to identifying admissible hearsay. It has replaced it with an ad hoc or case-by-case assessment in which the admission of hearsay evidence is determined as a matter of judicial discretion, according to the general principles of “necessity” and “reliability.”

Most of the other commentators assert that the traditional categories have not been replaced by the principled approach. See McCrea, supra, at p. 208 (stating that “in B. (K.G.), decided shortly after Smith, the Court clarified that the established hearsay exceptions remained very much intact”); Thompson, supra, at p. 284 (“Smith does not sweep away the existing exceptions”) and at p. 291 (“It is hard to believe that the real intent of Smith was to wipe out the whole structure of hearsay exceptions, constructed over the years by the common law.”); P. B. Carter, “Hearsay: Whether and Whither?” (1993), 109 L.Q.R. 573, at p. 579 (“R. v. Smith should be seen as supplementing and developing, rather than as totally replacing, the traditional framework of the common law”); Stewart, supra, at p. 64 (“Before reaching the principled approach, the Court must be satisfied that the statement is not admissible under any existing exception to the rule against hearsay.”); A. L.-T. Choo, Hearsay and Confrontation in Criminal Trials (1996), at p. 169.

45It is the duty of the courts to review common law rules and as recognized by the Law Reform Commission of Canada in its Report on Evidence, supra, at p. 69, the hearsay rule is the most characteristic rule of our system of judge-made rules of evidence. However, the duty to review the common law carries with it a corresponding responsibility to proceed prudently. This Court has limited changes in the common law to those which are “slow and incremental” rather than “major and far-reaching”: Watkins v. Olafson, [1989] 2 S.C.R. 750, at p. 760, per McLachlin J. This Court must restrict reforms to only “those incremental changes which are necessary to keep the common law in step with the dynamic and evolving fabric of our society”: R. v. Salituro, [1991] 3 S.C.R. 654, at p. 670, per Iacobucci J. The adoption of the principled approach in Khan was based on these calculations and, similarly, any extension of it must be evaluated in terms of the above considerations.

46I do not see the usurpation of the traditional exceptions by the principled approach as warranted under the current state of our law. After studying the issue for several years, the Law Reform Commission of Canada in its 1975 Report on Evidence concluded (at p. 5): “Many of the existing rules are sound and well-tested, and have the advantage of being familiar to the profession. We would retain these.” Similarly, the Ontario Law Reform Commission in its 1976 Report on the Law of Evidence concluded (at p. 15): “In our view, the general exclusionary rule now in effect should be retained, together with the existing common law exceptions. The hearsay rule should, however, be relaxed by way of further, codified, exceptions.” Our jurisprudence has recognized the need to relax the hearsay rule to keep it in step with our changing society, specifically our greater appreciation of jurors’ abilities. However, I am not convinced that changes have occurred that are so substantial as to vitiate the general conclusions of these commissions which of course studied these issues in far greater detail than we can.

47I find it persuasive that other common law jurisdictions that have studied the question have reached similar conclusions, retaining the traditional exceptions while relaxing the standards for admissibility under the hearsay rule. See, e.g., United Kingdom Law Commission, Consultation Paper No. 138, Criminal Law -- Evidence in Criminal Proceedings: Hearsay and Related Topics (1995), at pp. 145-59; Law Reform Commission of Ireland, “The Rule Against Hearsay”, in Reports (1979), vol. 2, 105, at pp. 201-9; Evidence Act 1995, 1995, No. 2, ss. 59-75 (Australia); Federal Rules of Evidence, Rules 801 et seq. (United States).

48The reforms that Iacobucci J. proposes will have “major and far-reaching” effects on the common law and the trial process and therefore should be avoided: see Watkins, supra, at p. 760. Within Iacobucci J.’s approach, evidence that falls under a traditional exception would be doubly challenged. First, while according to Iacobucci J. evidence falling within a traditional exception is presumptively admissible (para. 212), it is open to revisit the exceptions themselves to re-interpret them in a manner consistent with the requirements of necessity and reliability (para. 213). Second, Iacobucci J. asserts that “[i]n some rare cases, it may also be possible under the particular circumstances of a case for evidence clearly falling within an otherwise valid exception nonetheless not to meet the principled approach’s requirement of necessity and reliability” (para. 214). I fear that this invitation will create further uncertainty in the law as lawyers being lawyers will be quick to claim that their case is the “rare” one of which Iacobucci J. speaks. Iacobucci J. does not specify how trial judges are to deal with these rare cases, stating that they will determine the procedure “whether by voir dire or otherwise” (para. 214), but I think it is safe to assume that trial judges will turn to the procedure familiar to them and mandated by our previous jurisprudence: the voir dire (see B. (K.G.), supra).

49In any event, voir dire or not, this approach would require an in-depth re-evaluation of the admissibility of evidence already found to fall within an exception to the hearsay rule. The voir dire is merely the most formal mode of conducting this inquiry. With or without a voir dire, the double analysis mandated in every case of hearsay evidence on this approach would inevitably and unnecessarily complicate and lengthen trials.

50It is generally acknowledged that the principled approach will result in lengthening trials as litigation over the admissibility of evidence becomes more common. See Schiff, supra, at p. 661; Thompson, supra, at p. 295; Prithipaul, supra, at p. 92; Blok, supra, at p. 685. See also Rosenberg, “B. (K.G.) -- Necessity and Reliability: The New Pigeon-holes”, supra, at p. 75. In adopting the principled approach to hearsay evidence not falling within an established exception, we have chosen this course in the name of the search for truth and because of confidence that juries will be able to accurately weigh the quality of the evidence before them. However,

[t]aking away all of the rules and replacing them with necessity and reliability makes the role of the trial judge that much more difficult. . . . Trial judges simply do not have the luxury to spend hours, let alone days, pondering the reliability of a particular piece of evidence, an issue normally reserved to the fact-finding phase of the trial. What is required is a set of rules which can be applied to most situations. [Emphasis added.]

(M. Rosenberg, “Developments in the Law of Evidence: The 1992-93 Term -- Applying the Rules” (1994), 5 Supreme Court L.R. (2d) 421, at p. 487.)

51I would conclude that the principled necessity-reliability analysis, while appropriate where hearsay evidence does not fall within an established exception to the hearsay rule, has not replaced and should not supplant the traditional exceptions to the hearsay rule. I am not persuaded by Iacobucci J.’s attempt to split the difference and hold that the principled approach applies to the evidence that falls within a traditional exception but that a voir dire may only sometimes be necessary. Either the principled approach applies to the traditional exceptions or it does not. If it does, then a voir dire is required to consider the admissibility of such evidence. See B. (K.G.), supra. See also Thompson, supra, at p. 295 (“A voir dire is required, sometimes a complicated one under the residual exception.”). If the traditional exceptions endure, as I believe they should, then no voir dire is required.

52The source of the disagreement between my colleague Iacobucci J. and me can be traced to variant understandings of threshold reliability inquiry under the principled approach. The traditional exceptions are built upon a determination that a threshold of reliability is met in particular instances of hearsay statements. Reliability under the principled approach is similarly restricted to a threshold inquiry: B. (K.G.), supra, at p. 787. In R. v. Hawkins, [1996] 3 S.C.R. 1043, this Court explained the nature of the reliability analysis (at para. 75):

The criterion of reliability is concerned with threshold reliability, not ultimate reliability. The function of the trial judge is limited to determining whether the particular hearsay statement exhibits sufficient indicia of reliability so as to afford the trier of fact a satisfactory basis for evaluating the truth of the statement. More specifically, the judge must identify the specific hearsay dangers raised by the statement, and then determine whether the facts surrounding the utterance of the statement offer sufficient circumstantial guarantees of trustworthiness to compensate for those dangers. The ultimate reliability of the statement, and the weight to be attached to it, remain determinations for the trier of fact.

53Although it may appear that the above pronouncement conflicts with certain statements made in Khan, supra, and Smith, supra, it is possible to reconcile the approaches taken in all of these cases. In Khan, although McLachlin J. spoke in terms of the complainant not having a reason to lie, in essence, the recognition there was that children of such a tender age do not have a reason to lie about sexual subjects. Moreover, it is almost impossible for a child of such a tender age to have knowledge about the functioning of male anatomy in that way, rendering statements made by people in that class inherently reliable. Similarly, in Smith, one can see how people who are in need of assistance generally, as a class of people, have no reason to lie when they say that they need help. It is certainly true that these facts were present in the individual cases in Khan and Smith, but the rule would be applicable to a larger class of persons.

54Threshold reliability exists where there is a circumstantial guarantee of trustworthiness. It does not mean that the hearsay is true or even likely to be true, but rather it asks whether the circumstances are such that there is sufficient reliability for the hearsay to be properly assessed by the jury. The traditional hearsay exceptions are based on a determination of threshold reliability: Smith, supra, at pp. 929-30. These exceptions have historically been founded on truisms common to classes of people or common to circumstances applicable to all people. There is no reason why that should not continue to be the case.

55I acknowledge that some of the existing exceptions may require fine tuning. It may be that society has changed in such a manner that the rationale for the exception no longer applies. And, there may be circumstances which could arise in an individual case which challenge the inherent reliability that underpins a hearsay exception. However, the court must differentiate between individual circumstances that go to the weight of the evidence (such as motive to lie or other extraneous circumstances unique to the individual) and circumstances that are properly considered at the threshold stage.

56In re-evaluating a hearsay exception, the court must ask whether some reason exists to doubt that the rationale underlying the exception applies in certain circumstances. The scope of this inquiry is restricted; the court should not consider every possible permutation of indicators of reliability or unreliability. To do so would result in the excision of the categories and abolition of their utility. Rather, the questions should always be: Is there something wrong with the exception itself? Should the exception be inapplicable in some circumstances?

57The only time when a court should entertain a challenge to an existing exception is where there are facts, generally applicable to a class of persons, which weaken the theoretical justification for the exception. If facts arise which demonstrate that within an exception, we no longer think that statements made by a particular class of persons are inherently reliable, then to a certain extent, the rationale for the exception has been displaced and the exception must be modified to exclude people in such circumstances. For example, it may be that a court may wish to reconsider the dying declaration under certain circumstances such as when a dying person is under the influence of powerful drugs. Another exemption that may benefit from re-examination is the business records exception where the records are written by persons who are not disinterested parties. Both of these examples reflect particular circumstances under which reliability may be questioned for all people in such a situation. I believe that this approach will force the courts to reconsider the exceptions only where a problem exists with the theoretical foundation of the exception. Furthermore, having the courts remember the reasons for the rule will prove very beneficial. The problem with the exceptions was that they were thought to be “ossified judicially created categories” (Smith, supra, at p. 930), but this approach will foster a renewed understanding of the rationales underlying these categories.

58In sum, if one wishes to displace an exception, which is based on concepts of necessity and reliability, one will have to displace either necessity or the reliability upon which the exception is based. This can only be done where the circumstances would be generally applicable to a class of persons or in a particular circumstance faced by all persons.

59In principle, the trial judge’s residual discretion to exclude admissible evidence where its prejudicial effect substantially outweighs its probative value (Seaboyer, supra, at p. 611) could, in appropriate circumstances, apply to evidence falling within an exception to the hearsay rule: see, e.g., Hawkins, supra, at para. 85. This point was not addressed in the Court of Appeal and was not argued before this Court and therefore I see no need to undertake an analysis of the issue.

60I would summarize my position in the following terms which I believe are largely consistent with the approach taken by the British Columbia Court of Appeal in R. v. Chahley (1992), 1992 2387 (BC CA), 72 C.C.C. (3d) 193, and R. v. Collins (1997), 1997 12516 (BC CA), 118 C.C.C. (3d) 514. First, Khan and its progeny permit hearsay evidence to be admitted in new situations where necessary and where indicia of reliability are present, but did not abolish the traditional exceptions. Second, the traditional exceptions are largely consistent with the necessity-reliability criteria and so do not generally require revision. Finally, the traditional exceptions may be modified or supplemented as appropriate to conform to the principled approach (on this last point, see Rosenberg, “B. (K.G.) – Necessity and Reliability: The New Pigeon-holes”, supra, at p. 71). Trial judges should continue to apply the traditional exceptions to the hearsay rule.

(4) Admissibility of Cook’s Statement of Intention

(a) Is the Statement Hearsay?

61No disagreement exists that the Crown sought to use Cook’s statement to Giesbrecht that he had to “go and do an Autopac scam with Robert” as proof of the truth of its contents. Accordingly, as Iacobucci J. explains in para. 165, it is hearsay.

(b) The “Present Intentions” Exception to the Hearsay Rule

62The next question is whether the statement falls within a recognized exception to the rule against hearsay. All three judges in the Court of Appeal recognized that the statement fell squarely within the “present intentions” exception recognized by this Court in Smith, supra, at p. 925. Twaddle J.A., dissenting, would have excluded the statement not because it exceeded the confines of the “present intentions” exception but because he believed that the principled approach should be applied in this case. I disagree both with Iacobucci J.’s description of the “present intentions” exception and with his modification of the exception to comply with the principled approach. Furthermore, I do not agree with his characterization of the so-called “joint act” rule under this exception.

63This Court recently examined the “present intentions” exception in detail in Smith, supra, where we recognized that an “exception to the hearsay rule arises when the declarant’s statement is adduced in order to demonstrate the intentions, or state of mind, of the declarant at the time when the statement was made” (p. 925). Drawing on Wigmore, Iacobucci J. would add the requirement that the statement “must appear to have been made in a natural manner and not under circumstances of suspicion” (para. 168). Let me be quite clear that this second requirement is a new addition to the common law exception. In the course of reviewing nearly eighty years of jurisprudence of the present intentions exception in this country, I have been able to unearth but a single reference to the Wigmore requirement of “circumstantial guarantee of trusthworthiness”: see R. v. McKenzie (1986), 1986 1194 (BC CA), 32 C.C.C. (3d) 527 (B.C.C.A.), at p. 534. In R. v. P. (R.) (1990), 58 C.C.C. (3d) 334 (Ont. H.C.), at p. 338, Doherty J. (as he then was), while not mentioning Wigmore did however note in passing that “[i]t also appears that there is some circumstantial guarantee of trustworthiness attaching to these utterances as there is no suggestion that the deceased had reason to misrepresent her state of mind when she made these various statements to her sister and other close associates.” Neither of these cases provides any explanation, elaboration or application of the “circumstantial guarantee of trustworthiness” component. The clear weight of authority in this country examining the “present intentions” exception lacks any mention of this component: see, e.g., Smith, supra, R. v. Moore (1984), 1984 3542 (ON CA), 15 C.C.C. (3d) 541 (Ont. C.A.) (per Martin J.A.), leave to appeal to the Supreme Court of Canada refused, [1985] 1 S.C.R. x; M. (L.N.) v. Green (1995), 1995 2629 (BC SC), 14 B.C.L.R. (3d) 194 (S.C.); Neufeld v. Child and Family Services of Winnipeg (1993), 1993 14942 (MB QB), 89 Man. R. (2d) 150 (Q.B.); R. v. Jack (1992), 1992 2764 (MB CA), 70 C.C.C. (3d) 67 (Man. C.A.); R. v. McLeod (1991), 1991 903 (BC CA), 6 B.C.A.C. 223; R. v. Miller (1991), 1991 2704 (ON CA), 5 O.R. (3d) 678 (C.A.); R. v. Carriere (1986), 1986 4898 (MB CA), 40 Man. R. (2d) 150 (C.A.); R. v. Baron von Lindberg (1977), 1977 237 (BC SC), 66 B.C.L.R. 277 (S.C.); R. v. Flemming (1980), 43 N.S.R. (2d) 249 (Co. Ct.); R. v. Jackson (1980), 1980 2945 (ON CA), 57 C.C.C. (2d) 154 (Ont. C.A.); Hildebrand v. Butler (1979), 1979 320 (BC SC), 11 B.C.L.R. 234 (S.C.); R. v. Strongquil (1978), 1978 1815 (SK CA), 43 C.C.C. (2d) 232 (Sask. C.A.); R. v. Baltzer (1974), 1974 1668 (NS CA), 10 N.S.R. (2d) 561 (S.C.A.D.); R. v. Bencardino (1973), 1973 804 (ON CA), 2 O.R. (2d) 351 (C.A.); Re Grant Estate, 1970 1217 (BC SC), [1971] 1 W.W.R. 555 (B.C.S.C.); Great West Uranium Mines Ltd. v. Rock Hill Uranium Ltd., 1955 211 (SK CA), [1955] 4 D.L.R. 307 (Sask. C.A.); Gray v. New Augarita Porcupine Mines Ltd., 1952 322 (UK JCPC), [1952] 3 D.L.R. 1 (P.C.); R. v. Wysochan (1930), 1930 483 (SK CA), 54 C.C.C. 172 (Sask. C.A.); R. v. Covlin, 1923 620 (AB SCAD), [1923] 3 W.W.R. 555 (Alta. S.C.A.D.); R. v. Ferry (1992), 1992 15048 (ON SC), 18 C.R. (4th) 221 (Ont. Ct. (Gen. Div.)); Re J.B. (1998), 1998 18071 (NL CA), 166 Nfld. & P.E.I.R. 1 (Nfld. C.A.); Collins v. Collins (1993), 1993 8273 (NL SC), 114 Nfld. & P.E.I.R. 215 (Nfld. U.F.C.); R. v. Mafi (1998), 1998 6221 (BC CA), 21 C.R. (5th) 139 (B.C.C.A.); Home v. Corbeil, 1955 395 (ON SC), [1955] 4 D.L.R. 750 (Ont. H.C.), at pp. 754-55. Nor does the leading treatise in this country on evidence discuss the addition proposed by Iacobucci J.: see Sopinka, Lederman and Bryant, supra, at §§ 6.225-6.242. I do not see this addition as either necessary or desirable.

64By interpreting “circumstances of suspicion” in a broad fashion (see paras. 177-79), Iacobucci J.’s approach neutralizes the utility of the exceptions which provide predictability and a certain degree of swiftness in our trial process. By requiring the negation of the possibility that the declarant was untruthful, Iacobucci J. creates an extremely high threshold for the introduction of previously acceptable hearsay evidence. The possibility of untruthfulness is inherent in every statement. Ascertaining truth is the fundamental task for the trier of fact, not for the trier of law determining threshold reliability.

65In addition, by adding the requirement that a statement of intention not be made under circumstances of suspicion so as to conform with the principled approach, Iacobucci J. reveals the essential problem with his proposed mode of hearsay analysis. For if each exception is to incorporate a necessity and reliability component in examining each individual statement that comes under the exception, then the end result is the eradication of the exceptions themselves. I appreciate that under the approach proposed by Iacobucci J., the exceptions are supposed to continue to serve a useful role as, inter alia, “practical manifestations of the principled approach in concrete and meaningful form” (para. 205). However, once a trial judge is required to examine every hearsay statement in terms of reliability and necessity, I do not think that there is much left of the exceptions.

66Earlier, I explained my preferred mode of rationalizing the traditional exceptions with the principled approach. I will not repeat it except to say that I believe that the categories as a whole may be revisited and refined in light of the principled approach, but they are not supplanted by it. I see no reason to depart from the existing definition of this exception as outlined by Lamer C.J. in Smith, supra, and Martin J.A. in Moore, supra.

(c) The “Joint Act” Rule and the “Present Intentions” Exception

67It is common cause that the “present intentions” exception may not be used to infer that a third party acted in accordance with the declarant’s stated intention. Doherty J. explained the permissible chain of inferences in P. (R.), supra, at p. 343:

Evidence of the deceased’s state of mind may, in turn, be relevant as circumstantial evidence that the deceased subsequently acted in accordance with that avowed state of mind. When a deceased says, “I will go to Ottawa tomorrow”, the statement affords direct evidence of the state of mind –– an intention to go to Ottawa tomorrow –– and circumstantial evidence that the deceased in fact went to Ottawa on that day. If either the state of mind, or the fact to be inferred from the existence of the state of mind is relevant, the evidence is receivable subject to objections based on undue prejudice.

Later at p. 344, Doherty J. wrote: “The evidence is not, however, admissible . . . to show that persons other than the deceased acted in accordance with the deceased’s stated intentions, save perhaps cases where the act was a joint one involving the deceased and another person.” Lamer C.J. adopted this statement of the joint act exception in Smith, supra, at pp. 926-27.

68This exception simply recognizes basic principles of logic. As Professor R. J. Delisle explains in “R. v. Smith: The Relevance of Hearsay” (1991), 2 C.R. (4th) 260, at p. 264:

The statement of the declarant’s then state of mind should be received as evidence of that state of mind. . . . The next step in the chain –– whether an inference should be drawn that the deceased acted in accordance with that state of mind, by himself, or jointly with another –– presents a problem, not of hearsay, but of relevance and prejudicial value.

Doherty J. properly recognized this in P. (R.), supra, at pp. 346-49.

69Statements of intention may be admissible despite the fact that they refer to a joint act: see the leading cases of Mutual Life Insurance Co. v. Hillmon, 145 U.S. 285 (1892), at p. 296; United States v. Pheaster, 544 F.2d 353 (9th Cir. 1976), at pp. 374-80; People v. Alcalde, 148 P.2d 627 (Cal. 1944), at pp. 631-33. Alcalde presented a situation similar to the one before us: the deceased declarant had said that she “was going out with Frank” that evening. The California Supreme Court, over the dissent of Traynor J., held that the statement was admissible to show the decedent’s intent that evening. Generally, this sort of evidence is admissible as an indication of the declarant’s intention at the time he or she spoke. The inferences to be drawn from the evidence are for the jury, properly cautioned.

(d) Use of Cook’s Statement

70I do not believe that the trial judge erred in leaving Cook’s statement of intention to the jury. “Properly cautioned by the trial judge, juries are perfectly capable of determining what weight ought to be attached to such evidence, and of drawing reasonable inferences therefrom” (Smith, supra, at p. 937). The trial judge told the jury that “[t]he Crown says that Ms. Giesbrecht’s evidence is the second thread of evidence to link the accused, Robert Starr, with the deaths of Cook and Weselowski. Again, it is for you to decide whether the evidence of Cook’s statement about the scam goes as far as the Crown would have you believe” (emphasis added). I do not believe that this instruction invited the jury to draw an impermissible inference against the appellant. In examining what the Crown “would have [the jury] believe”, the Crown, referring to another statement not at issue here, explained “evidence of intention” as “What was the deceased thinking about just before his death?”. The Crown continued:

The difference between the statement of intention by Mr. Cook to Young and the statement of intention by Mr. Cook to Giesbrecht is that Cook carried the Giesbrecht intention out, didn’t he? He ended up in a car, say the prosecution, driving down a ditch, smashing it up outside of the Turski home. He travelled all that way. He had to have; he was found dead there, and the car that was smashed against the pole was the car that Starr had been in, in the Mohawk garage. So that expression of intention appears to have been carried out. . . .

I believe that the Crown was entitled to ask the jury to draw an inference that Cook in fact acted in accordance with his stated intention, even if that stated intention involved a joint act: see P. (R.), supra. In this situation, circumstances of joint intention may be considered along with the other evidence in determining what the third party did. Thus, I do not believe that the Crown asked the jury to draw an impermissible inference nor do I believe that the trial judge erred in instructing the jury on this issue.

71Accordingly, I would dismiss this ground of the appeal.

B. The Admissibility of Cheryl Ball’s Out-of-Court Identification

72On this issue, I agree with Monnin J.A.’s decision in the Court of Appeal that the trial judge may have committed an error by allowing the police officers to testify as to Ms. Ball’s out-of-court identification of the appellant, but that any harm that such evidence may have caused was effectively negated by the trial judge’s instructions to the jury. Therefore, I do not find it necessary to analyze this evidence under the “prior identification” exception or under the principled approach as my colleague Iacobucci J. does. I am convinced that the harm caused by the introduction of the evidence was minimal given other testimony that placed the appellant at the Mohawk gas station before the murders. Any harm that may have been caused by the impugned evidence was nullified by the trial judge’s instruction to the jury.

73In the days immediately following the murders, RCMP Constables Madden and MacLeod visited the home of Cheryl and Daniel Ball, interviewing them on four occasions on three different days. Ms. Ball testified at trial that on the second visit, two days after the murder, the police showed her some photographs. She testified that her memory of seeing the photos was hazy and that she did not think she had recognized anyone in the photos. Regarding the RCMP’s third visit on the same day as the second, Ms. Ball testified that she thought she had pointed to one of the three photographs shown to her and said that the person in the photo “looks kind of familiar”. Ms. Ball met with the police a week later, but at trial she did not recall having said anything to them about any of the photos on that occasion. Ms. Ball did not testify at trial as to why the person in the photo shown to her on the second visit looked familiar or where she had seen the person. She was not asked whether she had seen the appellant on the night of the murder and she did not testify that she had done so. In fact, she did not mention the appellant in her testimony or identify him in court.

74Constable Madden testified that on the second visit to her home, Ms. Ball told him and Constable MacLeod that she had seen a man talking to Cook at the Mohawk gas station. Constable Madden testified that Ms. Ball described the man she had seen as white, with glasses and short hair in a ponytail. She mentioned nothing about a goatee. Constable Madden testified that, at the time of the murders, the appellant had long hair that was not in a ponytail, was not wearing glasses, and had a goatee. To Constable Madden’s knowledge, the appellant had never worn glasses. Constable Madden testified that Ms. Ball indicated that the man in Photo No. 3 looked like the man whom she had seen at the Mohawk gas station talking to Cook and who was “probably driving the other car”. Photo No. 3 was a black and white photocopy of a photo of the appellant in which he had a moustache, no goatee, long hair, and no glasses.

75Constable Madden further testified that when he and Constable MacLeod returned to the Ball residence one week later, Ms. Ball picked out the same photograph of the appellant that she had identified on the previous occasion, from amongst a set of eight photographs. None of the other photos had been included among the three shown to her the week before. Ms. Ball was not asked whether her previous selection of the same photo influenced her selection of it one week later.

76The dissent in the Court of Appeal and Iacobucci J. in his reasons assert that Constable Madden’s testimony of Ms. Ball’s out-of-court identification was important because it placed the appellant in St. Norbert, whereas without Ms. Ball’s identification the appellant was last seen “halfway across the city from where the crime occurred”. Both Twaddle J.A. and Iacobucci J. are of the opinion that this piece of evidence might have been the “last straw”, allowing the Crown to meet its onus. With respect, I cannot agree.

77Constable Madden testified that Ms. Ball had stated that the appellant “was the one that she saw at the Mohawk Station and that he was probably driving the other car” (emphasis added). According to Twaddle J.A. and Iacobucci J., this second half of the sentence may suggest that Ms. Ball identified the appellant in St. Norbert where the murders took place, rather than at the Mohawk gas station. I do not believe that the evidence supports such an interpretation. After Constable Madden made the above statement in his testimony, the trial judge asked for a clarification:

THE COURT: I’m sorry, Mr. Dangerfield would you have Constable Madden repeat the words that Mrs. Ball said. “He is the one I saw at the Mohawk” is that what you said?

THE WITNESS: She felt that he was the one at the Mohawk Station.

THE COURT: Yes.

THE WITNESS: And she described him.

THE COURT: Yes.

THE WITNESS: And felt he was the one there that had been talking to Bo Cook and that had been driving the car.

A few questions later, on direct examination, Constable Madden again returned to the specifics of Ms. Ball’s identification of the appellant:

Q All right. You put this before Mrs. Ball. What did she say on that occasion?

A Without any hesitation at all, she pointed at photo number 5.

Q Did she say anything?

A And said: That was the one at the Mohawk.

The appellant’s trial counsel clearly understood Constable Madden to be testifying that Ms. Ball identified the appellant as seeing him at the Mohawk gas station. On cross-examination, counsel for the appellant asked the following regarding Ms. Ball’s out-of-court identification:

Q So did you say to Mrs. Ball, you also saw the picture from the first time, are you sure that the fellow you saw at the Mohawk is photograph 5 and you’re not just repeating what you say on August 23rd?

A No, we did not.

The cross-examination on this point continued:

Q And even when she picked somebody out, she wasn’t sure that was the person. She was telling you this person looks like?

A She says: I believe this is the one at the Mohawk.

Q Did you ask her if she was positive?

A She gave a description. I did not ask her if you are absolutely positive. She said: I think that’s the one at the Mohawk.

Q When she gave you the description of the person at the Mohawk, did she tell you that that person had a goatee?

A No, sir.

Q She told you that she thought the person in the photograph was wearing glasses -- pardon me -- the person at the Mohawk was wearing glasses?

A She said she [sic] was white, light coloured hair.

In the course of Constable Madden’s testimony, he did not make a single reference to St. Norbert; neither did Crown counsel nor counsel for the appellant. Furthermore, Constable MacLeod’s testimony similarly lacks any reference to St. Norbert:

Q And what did Mrs. Ball say, if anything, on seeing them this time?

A She pointed to one photograph and she identified that person as being at the Mohawk Station on the night that they were there.

Q I see. Is that all she said about him that you recall?

A She said it was him, it was the same guy.

Q I see. She mentioned the name Cook in this connection?

A Yes, she mentioned that this person had been talking to Bernard Cook.

78In light of the above testimony, I cannot agree with my colleague Iacobucci J.’s assertion that there were two distinct elements to Constable Madden’s testimony: namely, being at the Mohawk station and “probably” being in the other car in St. Norbert. Nor do I accept his assertion in para. 228 that the jury could have “easily” taken the reference to the “other car” to refer “not to . . . the Mohawk station, but to the second car seen in St. Norbert when the Balls last saw Cook and Weselowski alive”. In light of the above testimony, I find it stretches credulity to assert that the jury could have taken Constable Madden’s testimony to refer to Ms. Ball placing the appellant in St. Norbert. Moreover, I do not see how these seven obscure words – “he was probably driving the other car” -- subsequently clarified on no less than eight occasions, could have constituted the “last straw” that allowed the Crown to meet its onus. Additionally, Ms. Ball’s own limited testimony directly contradicts the inference that Constable Madden was referring to Ms. Ball seeing the appellant in St. Norbert. Ms. Ball testified in court that when she was dropped off in St. Norbert she saw a small car next to the car in which she had been riding but that she could not see if anyone was in the car. In a discussion outside the presence of the jury, the court reiterated that Ms. Ball had just testified that she did not see anyone in the car that was alongside the car in which she had been a passenger. Given that Ms. Ball testified clearly that she did not see the driver of the car in St. Norbert, I do not see how the jury could have drawn a contrary inference from Constable Madden’s testimony.

79Furthermore, no harm was done by Constable Madden’s testimony because Jodie Giesbrecht’s testimony also placed the appellant at the Mohawk station and also saw the appellant sitting in a car at the station. Moreover, any conceivable prejudice that might have been suffered by the appellant was cured by the trial judge’s cautionary instruction to the jury. On this issue, he instructed the jury as follows:

I think it is of some significance that Cheryl Ball was not asked if the reason she could identify the individual in photo number 5 of Exhibit 11 was because she had seen a photo, that is, photo number 3 in Exhibit 10, a week before. I think it is significant that photo number 3 in Exhibit 10 is the only one, the only photograph of the three people in Exhibit 10 repeated in Exhibit 11. It is significant, I think, that she was not told that the description she gave of the man she saw talking to Cook did not in fact fit the description of the man in photograph number 3 of Exhibit 10. It is significant, I think, as well that, in court, she did not mention Starr by name and that she was not asked to identify nor did she identify the accused, Starr, in the dock as being the man she saw at the Mohawk station. I would tell you, then, that the identification evidence of Mrs. Ball is, to say the least, extremely frail, and I would express the opinion that it does not link the accused, Starr, to the deaths of either Weselowski or Cook. However, it is for you to decide the value of her evidence.

Even the appellant’s own counsel characterized the introduction of the constables’ testimony as much ado about not very much. Rather than characterize this evidence as the “last straw” as Twaddle J.A. did in the Court of Appeal, the appellant’s counsel conceded that the case did not rise and fall with the constables’ testimony.

80In sum, given the trial judge’s admonition to the jury and the appellant’s admission of the marginal role that the identification evidence played, I cannot say that there is a reasonable possibility that the verdict would have been different if the alleged error had not been made. I would dismiss this ground of the appellant’s appeal.

C. The Adequacy of the Instruction to the Jury on Reasonable Doubt

(1) The Decisions in Lifchus and Bisson

81As Iacobucci J. ably describes, our recent decisions in Lifchus, supra, and R. v. Bisson, [1998] 1 S.C.R. 306, provide the framework for a review of a trial judge’s charge to the jury on reasonable doubt. I wish to emphasize the following points regarding these decisions.

82In Lifchus, at para. 22, this Court held that the expression “beyond a reasonable doubt” must be explained to a jury. In four different manners, Cory J. articulated guidelines for how trial judges should go about explaining this concept to a jury. First, he discussed what should be avoided (at paras. 23-26). Second, Cory J. explained what should be included in the definition of reasonable doubt (at paras. 27-35). Third, Cory J. provided a summary of what the definition “should and should not contain” (at paras. 36 and 37). Iacobucci J. reproduces this list of principles in paras. 231 and 232 of his reasons. Finally, Cory J. offered a suggested charge for trial judges to use in explaining the concept of reasonable doubt to a jury (at para. 39).

83Several matters are noteworthy about Cory J.’s reasons in Lifchus as set out above. First, in grappling with the issue of whether reasonable doubt must be explained to a jury, Cory J. acknowledged the confusing nature of this term and implicitly recognized the challenge facing trial judges in articulating it to a jury.

84Second, I believe it is notable that in articulating how the expression “reasonable doubt” should be explained to the jury, Cory J. generally employed the words “should” and “should not” rather than the imperative “must” and “must not”: Lifchus, supra, at paras. 23, 26, 28, 29, 31, 33, 35, 36 and 37. I cannot believe that Cory J. intended to use “should” and “must” interchangeably so as to make his list of items that “should be avoided” equivalent to a list of items that are prohibited and the inclusion of which automatically constitutes reversible error. I find support for this position in Cory J.’s statement that the equation of the term “moral certainty” with “reasonable doubt” in a trial judge’s charge to the jury “although not necessarily fatal to a charge on reasonable doubt, should be avoided” (para. 25 (emphasis added)).

85Third, Cory J. stated expressly that “[a]ny form of instruction that complied with the applicable principles and avoided the pitfalls referred to would be satisfactory” (para. 40). He did not view his suggested charge as a “magic incantation that needs to be repeated word for word” (para. 40).

86Fourth, and most importantly, Cory J. expressly stated that “an error in the instructions as to the standard of proof may not constitute a reversible error” (para. 41). He reiterated this Court’s statement in R. v. W. (D.), [1991] 1 S.C.R. 742, at p. 758, that the verdict ought not be disturbed “if the charge, when read as a whole, makes it clear that the jury could not have been under any misapprehension as to the correct burden and standard of proof to apply”. Cory J. explained that “[o]n the other hand, if the charge as a whole gives rise to the reasonable likelihood that the jury misapprehended the standard of proof, then as a general rule the verdict will have to be set aside and a new trial directed” (para. 41).

87Two further annotations to Lifchus are necessary. Lifchus, at para. 27, emphasized the need for the trial judge to link the standard of proof beyond a reasonable doubt to

that basic premise which is fundamental to all criminal trials: the presumption of innocence. The two concepts are forever as closely linked as Romeo with Juliet or Oberon with Titania and they must be presented together as a unit . . . Jurors must be reminded that the burden of proving beyond a reasonable doubt that the accused committed the crime rests with the prosecution throughout the trial and never shifts to the accused.

In addition, Cory J. stated that jurors “must be advised that the Crown is not required to prove its case to an absolute certainty since such an unrealistically high standard could seldom be achieved” (para. 31).

88Bisson, supra, applied the Lifchus principles to the problems that arise when examples are given to illustrate the phrase “beyond a reasonable doubt”. In Bisson, the trial judge’s charge ran afoul of Lifchus because it contained numerous misstatements. First, the charge stated that a reasonable doubt consists of [translation] “a degree of certainty which you use every day, in your important activities, whether at work, at home or in your leisure activities. It is a degree of certainty which you employ, which you have used since you reached the age of reason, without perhaps having dissected it as we are doing at present” (para. 3). Under Lifchus, at paras. 23-24, such a statement was clearly wrong. Second, this misstatement was compounded rather than cured by looking at the charge as a whole: Bisson, at para. 9. The charge proceeded to provide an example of the steps taken in order to determine the level of oil in an automobile (at paras. 3 and 4). The charge then reiterated its earlier error, stating, [translation] “the degree of certainty that is required before concluding that someone is guilty of something, is exactly the same degree of certainty, certainty beyond a reasonable doubt, as the degree of certainty you use in important activities in your life. Reasonable doubt is no more than that, it is no less than that” (para. 4).

89Cory J. discussed the problems that examples may create, noting that the example used seemed to suggest to jurors that they can apply the same standard they would use in everyday, routine decisions to the determination of guilt beyond a reasonable doubt (at para. 7). The trial judge in Bisson erred in providing an example because it invited the jurors to apply a standard of proof significantly lower than the standard to which the prosecution must be held in a criminal case (at para. 8). Cory J. then noted that an error in the instructions as to the standard of proof may not constitute a reversible error if the charge when read as a whole does not give rise to the reasonable likelihood that the jury misapprehended the correct standard of proof (at para. 9). Looking at the instructions in their entirety, the Court concluded that a reasonable likelihood existed that the jury misapprehended the correct standard of proof (at para. 10).

90I believe that these are the relevant principles to apply in considering the propriety of a charge to the jury on proof beyond a reasonable doubt.

(2) The Adequacy of the Reasonable Doubt Charge in This Case

91In determining the adequacy of the reasonable doubt charge in this case, I find the reasons of Sharpe J.A. in the recent case of R. v. Tombran (2000), 2000 2688 (ON CA), 47 O.R. (3d) 182 (C.A.), particularly helpful because the charge in that case was strikingly similar to the one at issue before us. I would adopt Sharpe J.A.’s holding, at p. 194, that “[w]hile trial judges are now expected to follow the Lifchus model charge, failure to do so in cases tried before Lifchus, supra, does not constitute reversible error if the charge conveys to the jury the special meaning attached to reasonable doubt”.

92Despite not having the benefit of this Court’s decision in Lifchus, the trial judge included most of the suggested elements in his charge to the jury. I agree with the conclusion of the Court of Appeal that the charge met the key requirements of Lifchus and that the flaws in the charge do not vitiate it. For reference, I will set out the impugned portion of the charge below:

It is rarely possible to prove anything with absolute certainty and so the burden of proof on the Crown is only to prove the guilt of the accused beyond reasonable doubt. What, then, is proof beyond a reasonable doubt?

The words "reasonable doubt" are used in their everyday, ordinary sense and not as a legal term having some special connotation. The words have no magic meaning that is peculiar to the law. A reasonable doubt is an honest, fair doubt, based upon reason and common sense. It is a real doubt, not an imaginary or frivolous one resting on speculation or guess rather than upon the evidence you heard in this courtroom.

So you can see, the words "reasonable doubt" are ordinary words we use in our everyday language. So if you can say, I am satisfied beyond a reasonable doubt, the Crown has met the onus upon it. If you cannot say those words ‑‑ if you cannot say, I am satisfied beyond a reasonable doubt, the Crown has not met the onus on it, and the accused is entitled to have your doubt resolved in his favour.

93It is clear from the above selection that, like in Tombran, supra, the trial judge succeeded, without the benefit of our decision in Lifchus, in complying with most of its guidelines. This portion of the charge instructed the jurors that a reasonable doubt must not be an imaginary or frivolous doubt (Lifchus, at para. 31); that it is a doubt based upon reason and common sense (Lifchus, at para. 30); that it must be based on the evidence that the jurors heard in the courtroom (Lifchus, at para. 30); and that the Crown is not required to prove its case to absolute certainty (Lifchus, at para. 31). This portion of the charge did not equate proof beyond a reasonable doubt to proof to a moral certainty (Lifchus, at paras. 25 and 37); nor did it instruct jurors that they might convict if they were “sure” that the accused was guilty (Lifchus, at para. 37); nor did it invite jurors to apply the same standard of proof that they apply to important decisions in their own lives (Lifchus, at para. 37; Bisson, supra). Elsewhere in his charge, the trial judge exhorted the jury to put any feelings of prejudice or sympathy out of their minds (Lifchus, at paras. 31 and 36).

94The trial judge’s charge, however, was not flawless. Specifically, we must examine the effect of the trial judge’s misstatement that the words “reasonable doubt” are used in their everyday, ordinary sense and have no special legal meaning. It is asserted that this flaw in the charge, together with the failure of the trial judge to state expressly that the Crown was required to do more than prove the appellant’s guilt on a balance of probabilities, constitutes reversible error. With respect, I cannot agree.

95The verdict ought not to be disturbed because the charge, “when read as a whole, makes it clear that the jury could not have been under any misapprehension as to the correct burden and standard of proof to apply”: W. (D.), supra, at p. 758; Lifchus, at para. 41. While the judge told the jury that the words “reasonable doubt” had no special meaning, this was a harmless error because he proceeded to give them all the legal information they required. The jury knew the accused was presumed innocent, they knew the standard of proof was very high — a standard just below absolute certainty — and they knew that they could not convict on a balance of probabilities. Cory J.’s objection to the “no special meaning” instruction in Lifchus was founded on the danger that the jury might judge on the basis of their everyday affairs: see Bisson, supra. This danger was obviated in this case by the trial judge’s instruction. Therefore, the charge as a whole does not give rise to the reasonable likelihood that the jury misapprehended the standard of proof: Lifchus, supra.

96What is essential is that the charge communicate clearly to the jury that they cannot find the accused guilty on a balance of probabilities. I believe that the charge did this. The opening sentence of the charge clearly implies that proof beyond a reasonable doubt is just one notch lower than absolute certainty; it is the highest level of proof that can be humanly achieved. This negates any suggestion that proof on a balance of probabilities might suffice. “[R]eading the charge as a whole, this jury could not have been left with any doubt that the burden was on the Crown to prove every element of the offence beyond a reasonable doubt rather than on a mere balance of probabilities”: Tombran, supra, at p. 194.

97Moreover, the charge complied with Lifchus’s first principle that it must be made clear to the jury that the standard of proof beyond a reasonable doubt is inextricably linked to the presumption of innocence and that this burden never shifts to the accused: Lifchus, at para. 27. The relevant portion of the trial judge’s charge on this point is:

Let me first emphasize the presumption of innocence. Simply put, the accused is presumed to be innocent and he continues to be and remain innocent unless and until the Crown has satisfied you beyond reasonable doubt of his guilt. This presumption remains with the accused and for his benefit from the beginning of the case until the end of the case. The onus of proving guilt rests upon the Crown from the beginning to the end of the case and it never shifts. There is no burden whatsoever on the accused to prove his innocence. The Crown must prove beyond a reasonable doubt that an accused person is guilty of the offence or offences with which he is charged before he can be convicted.

98In light of the trial judge’s compliance with the bulk of the principles enunciated in Lifchus, I am loath to find that the charge was automatically vitiated by the failure to include a specific item mentioned in Lifchus or by the inclusion of an improper item. To do so would contravene our holding that “an error in the instructions as to the standard of proof may not constitute a reversible error”: Lifchus, at para. 41.

99Jurors are sophisticated persons who are instructed to listen to and follow the entirety of the judge’s charge to them. We must assume that they do so. It would ill behoove a reviewing court to do exactly what the jury is commanded not to do and isolate a phrase or a small section of the charge, ignoring the whole of it. I do not believe that is what Cory J. intended in Lifchus. Rather, he was attempting to provide guidance to trial judges in the extremely difficult task of articulating what is proof beyond a reasonable doubt and communicating this standard to the jury.

100Lifchus should be viewed as a broad template for trial judges to assist them in their difficult task. Cory J.’s suggestions provide a touchstone for comparison for courts in reviewing jury charges. However, reviewing courts must resist the temptation to use the Lifchus suggestions as a mandatory checklist. The tendency to do so is natural and understandable as it would make the task of reviewing a jury charge far easier. However, a jury charge is not a multiple choice exam that can be marked by a computer. Rather, it is akin to a work of literature that must be studied in its entirety in order to evaluate it as a whole.

101Examining the charge in its entirety, I do not find that a reasonable likelihood existed that the jury misapprehended the correct standard of proof. Accordingly, I would dismiss this ground of appeal.

III. Disposition

102I would dismiss the appeal.

The judgment of Iacobucci, Major, Binnie, Arbour and Lebel JJ. was delivered by

IACOBUCCI J.--

I. Introduction and Summary

103The appellant, Robert Starr, was convicted at trial before a judge and jury of two counts of first degree murder. He had been accused of killing Bernard (Bo) Cook and Darlene Weselowski by shooting them by the side of a provincial highway on the outskirts of Winnipeg in the early morning hours of August 21, 1994.

104This appeal was originally taken as of right from the decision of the Manitoba Court of Appeal, dismissing the appellant’s appeal from the two convictions. Twaddle J.A. dissented in the Court of Appeal on two distinct hearsay issues and on the sufficiency of the trial judge’s charge to the jury on the issue of reasonable doubt, which are the issues that form the basis for this appeal. Twaddle J.A. would have allowed the appeal and ordered a new trial. On November 9, 1999, the Court ordered a re-hearing of the appeal, and, by letter dated November 18, 1999, invited submissions on the application of the “principled approach” to the traditional hearsay exceptions, and on the scope of the common law exception for statements of present intention.

105As a general matter, I agree with the lucid reasons of Twaddle J.A., and with the remedy he proposed. However, because this case raises important issues regarding the admissibility of hearsay evidence, including, in particular, the question of whether exceptions to the hearsay rule must conform to the principled approach to hearsay admissibility adopted by this Court in R. v. Khan, [1990] 2 S.C.R. 531, and subsequent cases, I will set out my reasons in some detail. The appeal also provides a useful opportunity to comment upon the proper application of the principles set out by this Court in R. v. Lifchus, [1997] 3 S.C.R. 320, regarding the appropriate manner of instructing a jury on reasonable doubt. This is particularly important in light of the frequency with which this issue has arisen of late, particularly in R. v. Russell, S.C.C., No. 26699 (heard November 5, 1999), R. v. Beauchamp, S.C.C., No. 27075 (heard December 8, 1999), and R. v. Avetysan, S.C.C., No. 27279 (heard January 28, 2000)^*.

106In the result, I conclude that the Court of Appeal erred in admitting the statement in question under the “present intentions” exception to the hearsay rule. However, Khan, supra, and subsequent cases have established that hearsay that does not fit within a traditional exception may nonetheless be admissible if it meets the twin criteria of reliability and necessity. This case therefore requires that we determine the admissibility of evidence under the principled approach, and more particularly, the interaction between the principled approach and the traditional exceptions. In so doing, I conclude that hearsay that does fit within a traditional hearsay exception, as currently understood, may still be inadmissible if it is not sufficiently reliable and necessary. The traditional exception must therefore yield to comply with the principled approach.

107With respect to the jury instructions on the phrase “beyond a reasonable doubt”, as set out in Lifchus, supra, an instruction like the one in this case, which fails to explain that the beyond a reasonable doubt standard has special legal significance and requires a significantly higher quantum of proof than the balance of probabilities, will not satisfy the Lifchus standard.

II. Factual Background

108Bernard Cook and Darlene Weselowski were drinking with the appellant at the Westbrook Hotel in the north end of Winnipeg during the late evening and early morning of August 20-21, 1994. Cook had been released from prison the previous day. A witness, Janet Daly, testified that at around 1:50 a.m., the appellant stood up and told Cook: “If we are going to get this done, we better get this done now”. Cook, Weselowski, and the appellant then left the hotel.

109Cook and Weselowski parted ways with the appellant. Outside, Cook and Weselowski offered a couple named Cheryl and Daniel Ball a ride home in Weselowski’s station wagon to the Balls’ residence in St. Norbert, a 20 to 30 minute drive away. Cook and Weselowski were intoxicated. Cook’s blood-alcohol content at the time of his death, around 3:00 a.m., was .250. Weselowski’s was .140. By Cheryl Ball’s account at trial, the Balls were also “really, really drunk”.

110Weselowski drove, and the group first stopped at an adjacent Mohawk gas station, where Jodie Giesbrecht, a sometime girlfriend of Cook, approached the station wagon and had a conversation with Cook. During the conversation, Giesbrecht observed a car beside the Mohawk gas station, and saw the appellant in the car. She could not determine whether anyone else was in the car with the appellant. A day or two later, Giesbrecht saw a picture in the newspaper of what she believed was the car in which she had seen the appellant. The car had been found at the scene of the murder. After seeing the picture, Giesbrecht phoned the police and told them she had seen the car on the night of August 20-21, 1994 at the Mohawk gas station, with the appellant in it.

111Giesbrecht had spent part of the previous evening with Cook, just after his release from prison. She had been sitting with Cook and the appellant at the Westbrook Hotel about 45 minutes before Cook left the hotel with Weselowski. Giesbrecht testified that she attempted to avoid being seen by Cook as she approached the car, because she did not want Cook “to take off on me while he was with Darlene [Weselowski]”. In their discussion beside the station wagon, Cook told Giesbrecht that he was driving the Balls home with Weselowski. Giesbrecht became angry with Cook because he was out with Weselowski rather than her, and she walked away from the car. Cook got out of the car and followed Giesbrecht into a laneway, where they had a further conversation. Giesbrecht asked Cook why he would not come home with her. According to Giesbrecht, Cook replied that he had to “go and do an Autopac scam with Robert”. Giesbrecht understood “Robert” to be the appellant. Cook said he was to receive $500 for his involvement in wrecking a car for insurance purposes. Giesbrecht testified that it was strange for Cook to discuss business matters like this with her.

112At trial, the defence unsuccessfully sought to exclude as hearsay Giesbrecht’s testimony regarding Cook’s stated intent to participate in an Autopac scam with the appellant.

113Weselowski, Cook, and the Balls drove to Le Maire Street in St. Norbert, where the Balls got out of the car. Shortly after exiting the station wagon, Cheryl Ball looked back and observed a smaller car alongside the station wagon. Cheryl Ball had seen the smaller car follow them as they turned onto Le Maire Street. She could not recall the colour or make of the smaller car. She could not see how many people were in the smaller car. She could not hear any voices coming from either of the two cars. Cheryl Ball then turned away and continued to walk home with her husband.

114Around 3:00 a.m., Albert and Darlene Turski were awakened by their dog barking and by the sound of spinning tires outside their bedroom window. The Turskis lived on a farm bordering on provincial highway 247, several kilometres from where the Balls had been dropped off.

115Albert Turski testified that he looked out the window and saw on the road the lights of two cars. One car seemed to be larger than the other, judging by the sound of its engine. He could not see the vehicles clearly, other than their lights. The road was 50 to 75 yards away. The smaller car was driving along a ditch and its tires sounded like they were spinning in the grass. The larger car approached the smaller car but did not come right up to it. The smaller car got stuck but then emerged from the ditch. The lights on the larger car turned off. Albert Turski heard mumbling from people talking for about 10 to 15 seconds. He could not determine whether it was two people talking, or more than two. He then heard two “pops”, and about five seconds later the smaller car drove across the road into the ditch again and back out, weaving, while the larger car turned its lights on and drove off down the highway at a normal speed. Albert Turski went back to bed.

116Darlene Turski’s testimony was largely similar to that of her husband. However, she heard the mumbling of people talking for what she considered to be one or two minutes, and she testified that the talking sounded as though it involved “more than two” people. Right after the conversation ended, she heard two “pops”. About fifteen seconds later, she said, the two cars pulled away and she and her husband lay back down in bed. Albert fell asleep immediately, but Darlene Turski stayed awake for a period of time. She then heard more “pops” in a fast series, which sounded as if they were further away from the house.

117At 4:26 a.m., a young couple drove by the Turskis’ house on provincial highway 247 and discovered the body of Darlene Weselowski on the side of road outside the house, and the body of Bernard Cook lying 400 metres up the road. Weselowski had been shot twice in the head. Cook had been shot three times in the head and three times in the chest. Both victims had been shot with the same 9mm Glock semi-automatic pistol. A damaged car was found driven into a telephone pole in the ditch near Cook’s body.

118Four days later, on August 25, 1994, the Weselowski station wagon was found parked on Buchanan Street, a block and a half from the home of the appellant’s brother William on Risbey Street. There were blood spatters all over the front passenger side of the car, and several shell casings were found in the car. The blood was determined to be Cook’s. Some of the appellant’s scalp hairs were found on the floor of the driver’s side of the vehicle.

119On August 22, 1994, the day after the murders, the appellant and his common law wife Shelley Letexier checked into a room at the Downs Hotel. After they left the hotel at 3:28 a.m. on August 25, 1994, they drove to the home of the appellant’s brother, William, on Risbey Street. After their departure, a leather jacket in good condition was found in a dumpster at the back of the hotel. A search of the appellant’s home resulted in the seizure of a purchase receipt from the Sidney I. Robinson store for 9mm ammunition capable of being loaded into a 9mm Glock semi-automatic pistol such as that which killed Weselowski and Cook. Store records from the purchase bore the appellant’s name.

120RCMP Constables Patrick Madden and Harvey MacLeod paid three visits to the home of Cheryl and Daniel Ball after the murders. Cheryl Ball testified that on the second visit, on August 23, 1994, the police showed her some photographs. At trial, she had a hazy memory of seeing the photos, and testified that she did not think she had recognized anyone in the photos. The police left, but then Daniel Ball called the police back into the house, and Cheryl Ball met with the police again. She testified at trial that she thought she pointed to one of the three photographs shown to her and said that the person in the photo “looks kind of familiar”. She met with the police again on August 30, 1994, but did not recall at trial having said anything to police about any of the photos at that time. Cheryl Ball did not testify at trial as to why the person in the photo shown to her on August 23, 1994 looked familiar, or where she had seen the person. She was not asked whether she had seen the appellant on the night of the murders, and she did not testify that she had done so. She did not mention the appellant in her testimony, or identify the appellant in court. She did testify, however, that nothing she said in her conversations with police was untrue.

121Daniel Ball had less recollection of his meetings with police in August 1994. He could not remember at trial whether he was shown any of the photos that police later testified he was shown when they met with him. He did not recall having identified anyone in the pictures as being familiar.

122Constable Madden testified that on August 23, 1994, Cheryl Ball told him and Constable MacLeod that she had seen a man talking to Cook at the Mohawk gas station. The man she had seen was white, had short hair in a ponytail, and wore glasses. She did not mention him having a goatee. Madden testified that, at the time of the murders, the appellant had long hair that was not in a ponytail, was not wearing glasses, and wore a goatee. To Madden’s knowledge, the appellant had never worn glasses. Madden testified that after Daniel Ball beckoned him and MacLeod back into the house to speak to Cheryl Ball on August 23, Cheryl Ball indicated that the man in Photo No. 3 looked like the man whom she had seen at the Mohawk gas station talking to Cook and who was also “probably driving the other car”. Photo No. 3 was a photo of the appellant, in which he wore a moustache, no goatee, long hair, and no glasses. The photograph was a black and white photocopy.

123Madden testified that he and MacLeod returned to the Ball residence on August 30, 1994 and, in separate interviews, showed Cheryl and Daniel Ball a set of eight photographs. Photo No. 5 was the same photo of the appellant that had been included in photocopied form as Photo No. 3 on August 23, 1994. None of the other photos had been included among the three photos shown to the Balls on August 23rd. According to Madden, Cheryl and Daniel Ball separately picked Photo No. 5 as being the person from the gas station, although Daniel Ball told them that he may have picked that photo because he had seen the same photo on August 23. Cheryl Ball was not asked whether her previous selection of the same photo on August 23 had influenced her selection on August 30.

124Madden’s testimony to the effect that Cheryl Ball identified the appellant as having spoken to Cook at the Mohawk gas station, as well as MacLeod’s testimony to the same effect, constitute a second group of statements that the defence unsuccessfully sought to have excluded as hearsay at trial.

125Police intercepted private communications involving the appellant and his common law wife, Shelley Letexier, who were recorded making several statements that were relied upon to some degree at trial by both the Crown and the defence. On the tapes, the appellant is heard making statements to Letexier which the Crown argued were consistent with his having been in a stolen car that followed Cook, Weselowski and the Balls up to the point that the Balls were dropped off in St. Norbert, and with his having thrown a leather jacket into the dumpster at the Downs Hotel. However, the appellant was also recorded stating to Letexier that he “never killed anyone”, and Letexier was recorded stating that the appellant and unspecified others were stupid to have followed a car: “Oh well you guys are so fuckin’ stupid. You got fuckin’ witnesses seen you guys, seen you behind that car. It should ah, been called off right away when there were four people.”

126Several other witnesses testified at trial, including Cecil Starr, a cousin of the appellant and a friend of Cook’s, who testified that, about ten days after the murders, the appellant confessed to having killed Cook and Weselowski. Cecil Starr said that the confession occurred at the apartment of Dwight Bamlett, and that Cecil Starr had subsequently attended a party on a reserve where several people, including Cecil’s brother, Mervin Starr, as well as Dwight Bamlett, were discussing the confession.

127Several defence witnesses, including Cecil Starr’s sister Audrey Malcolm, testified that Cecil Starr was a habitual liar, and that his story of the appellant’s alleged confession was false. Malcolm stated that Cecil Starr had since told her that he had not seen the appellant for over two years, and that the appellant had never confessed. Mervin Starr testified that he had never been at a party of the type described by Cecil. Dwight Bamlett testified that Cecil Starr and the appellant had never been present together in his apartment, and that he had never been at a party on a reserve of the type described by Cecil. Former Staff Sergeant Savage of the RCMP testified that he had interviewed Cecil Starr shortly before trial, and that Cecil had denied ever having said that the appellant confessed to him, and had said he could not remember any party on the reserve.

128Finally, the trial judge called Constable Robert Young of the Winnipeg police to testify regarding his interactions with Cook on August 20, 1994, several hours before the murders. Young testified that Cook was the “sergeant at arms” of the Manitoba Warriors gang, meaning he was the person responsible for all of the gang’s criminal matters. Young stated that he and Cook had had meetings previous to August 20, 1994, at which Cook had expressed his desire to dissociate himself from the Manitoba Warriors. Young stated that he met with Cook on August 20, 1994 at about 6:00 p.m. Cook told Young about a break-in that had occurred earlier that day at the Sidney I. Robinson store, in which guns had been stolen. Cook said that he was to meet with members of the Los Bravos gang later that night at the Westbrook Hotel to purchase some of these guns, and that he had been told by Joseph Flett, the President of the Manitoba Warriors, to wear his gang colours. Cook expressed a concern as to why he was being asked to wear gang colours. Cook told Young that if he purchased guns he would turn one over to Young, and that he would alert Young of success in the purchase by calling Young at around 1:00 a.m. Young never received the anticipated telephone call from Cook. When Cook’s body was found, he was wearing the gang colours of the Manitoba Warriors.

129The theory of the Crown at trial was that the killing of Cook was a gang-related execution perpetrated by the appellant. Weselowski was an unfortunate witness who was killed simply because she was in the wrong place at the wrong time. The theory was that the appellant had used an Autopac scam as a pretext to get Cook out into the countryside. Outside the Turskis’ home, Cook got into the smaller car and drove it into the ditch, hitting telephone poles in an effort to damage the car. The appellant shot Weselowski twice in the head, then drove Weselowski’s station wagon up the road to where Cook had stopped the smaller car in the ditch. When Cook entered the station wagon on the passenger side, the appellant shot him from the driver’s seat three times in the head and three times in the chest. He then pushed Cook’s body out of the vehicle and drove away, parking near his brother’s house, where the appellant abandoned the station wagon.

130The defence theory focussed on the issue of identity. The defence argued that the circumstantial evidence adduced by the Crown had failed to prove that the appellant actually shot the victims, and had failed to dispel the real possibility that other gang-related individuals were the killers.

III. Judicial History

A. Manitoba Court of Queen’s Bench

(1) Ruling and jury instructions on Cook’s statement of intention

131The Crown brought an application at trial to permit Jodie Giesbrecht to testify that Cook had told her, inter alia, that he had to “go and do an Autopac scam with Robert”. The Crown sought to have this testimony admitted as evidence of Cook’s state of mind on the evening of the murder and of his intention to go with the appellant to perpetrate a car insurance scam. The defence objected on the basis that the evidence was hearsay if used to prove that the appellant acted in accordance with Cook’s stated intention, and that its admission would be more prejudicial than probative because the jury would likely draw such an impermissible inference.

132The trial judge did not hold a voir dire as part of the Crown motion. Hirschfield J. found that Giesbrecht’s anticipated testimony regarding the Autopac scam was admissible under the “present intentions” or “state of mind” exception to the hearsay rule. Hirschfield J. found that the prejudicial effect of the admission of the evidence was that Cook’s alleged statement implicated the appellant in the commission of insurance fraud. The trial judge noted that, in this regard, the words said to have been spoken by Cook “would be as prejudicial as against the deceased as they are against the [appellant]”. Hirschfield J. concluded that, in light of the other evidence that the Crown anticipated calling in order to link the appellant to the murders, the probative value of Giesbrecht’s testimony outweighed any prejudicial effect.

133In his subsequent instructions to the jury regarding the use to be made of Giesbrecht’s testimony regarding Cook’s statements to her on the night of the murders, Hirschfield J. told the jury that it was for them to decide whether Cook’s statement about the scam linked the appellant with the deaths of Cook and Weselowski.

(2) Ruling and jury instructions on Cheryl Ball’s out-of-court identification

134A voir dire was held to determine the admissibility of the testimony of Constables Madden and MacLeod regarding Cheryl Ball’s alleged identification of the appellant as being a person she had seen talking to Cook at the Mohawk gas station. Madden and MacLeod both testified. The Crown argued it was entitled to adduce the statements of the two officers, notwithstanding the fact that Cheryl Ball had not testified at trial as to having seen a man talking to Cook at the gas station, or to having identified that man in one of the photographs presented to her by police. The Crown relied upon the prior identification exception to the hearsay rule. The defence argued that the use of hearsay evidence to establish a prior out-of-court identification of the accused by a witness is permissible only where the witness acknowledges in court having made the prior identification. Although the record is not entirely clear on this point, Hirschfield J. appears to have accepted the Crown’s submissions. He ruled Madden and MacLeod’s anticipated evidence admissible.

135In his instructions to the jury, Hirschfield J. reviewed the out-of-court identification evidence said to have been provided by Cheryl Ball to Constables Madden and MacLeod. The trial judge concluded his instructions with a caution that Cheryl Ball’s identification evidence was “extremely frail” and, in his view, did not link the appellant to the deaths of Cook and Weselowski. However, Hirschfield J. left it to the jury to decide the value of Ball’s evidence on this point.

(3) Charge to the jury on reasonable doubt

136In his instructions to the jury, Hirschfield J. explained that the appellant was presumed to be innocent and that the onus was on the Crown to prove the appellant’s guilt beyond a reasonable doubt. He then defined the reasonable doubt standard in the following terms:

It is rarely possible to prove anything with absolute certainty and so the burden of proof on the Crown is only to prove the guilt of the accused beyond reasonable doubt. What, then, is proof beyond a reasonable doubt?

The words “reasonable doubt” are used in their everyday, ordinary sense and not as a legal term having some special connotation. The words have no magic meaning that is peculiar to the law. A reasonable doubt is an honest, fair doubt, based upon reason and common sense. It is a real doubt, not an imaginary or frivolous one resting on speculation or guess rather than upon the evidence you heard in this courtroom.

So you can see, the words “reasonable doubt” are ordinary words we use in our everyday language. So if you can say, I am satisfied beyond a reasonable doubt, the Crown has met the onus upon it. If you cannot say those words -- if you cannot say, I am satisfied beyond a reasonable doubt, the Crown has not met the onus on it, and the accused is entitled to have your doubt resolved in his favour.

No further definition of the reasonable doubt standard was provided. The trial judge went on to explain that the jury might be left with a reasonable doubt unless it rejected as untrue some or all of the evidence, if any, that favoured the appellant.

B. Manitoba Court of Appeal (1998), 1998 6760 (MB CA), 123 C.C.C. (3d) 145

(1) Monnin J.A., Kroft J.A. concurring

(a) Admissibility of Cook’s Statement of Intention

137Monnin J.A. found that Giesbrecht’s testimony that Cook told her he was going to do an Autopac scam with the appellant was hearsay, but was nonetheless admissible. While he did cite the decision of this Court in R. v. Smith, [1992] 2 S.C.R. 915, Monnin J.A. primarily relied upon the “present intentions” or “state of mind” exception to the hearsay rule, to the effect that, where the intention or state of mind of an out-of-court declarant is relevant to a fact in issue, evidence that would otherwise be hearsay is admissible to prove the declarant’s intention or state of mind. Monnin J.A. endorsed the trial judge’s ruling, stating that Cook’s alleged statement to Giesbrecht was admissible to demonstrate Cook’s state of mind at the time, and to demonstrate that Cook acted in accordance with his “stated intention or plan” (p. 158). Monnin J.A. also found that the probative value of Cook’s statement outweighed any prejudicial effect.

(b) Admissibility of Cheryl Ball’s Out-of-Court Identification

138Monnin J.A. found that the trial judge may have committed an error in permitting the Crown to adduce evidence through Constables Madden and MacLeod that Cheryl Ball had identified the appellant as a person she had seen talking to Cook at the Mohawk gas station. However, Monnin J.A. held that the trial judge’s cautionary instruction regarding the frailty of the identification evidence “effectively negated any harm that such evidence might have caused” (p. 159). He stated that the appellant’s case was not one that “turned on the issue of identification” (p. 159), and that there was ample other evidence that the jury could have relied upon to arrive at a guilty verdict. Citing R. v. Bevan, [1993] 2 S.C.R. 599, as establishing the applicable test for confirming a guilty verdict in the face of an error in law, Monnin J.A. stated that he was satisfied that, even if the evidence of Madden and MacLeod had been excluded, a properly instructed jury would have reached the same verdict.

(c) The Instruction on Reasonable Doubt

139Monnin J.A. stated that the trial judge’s charge to the jury, considered as a whole, was more than fair and balanced, and if anything favoured the appellant. On the issue of reasonable doubt, Monnin J.A. held that the trial judge’s charge to the jury met the requirements set out by this Court in Lifchus, supra, both in its actual language and in overall tone. In particular, Monnin J.A. found that the trial judge had explained the link between the reasonable doubt standard and the presumption of innocence, the requirement that the Crown must prove more than mere probability of guilt, and the principle that a reasonable doubt is based on reason and common sense, logically founded upon the evidence or lack of evidence. Monnin J.A. noted that, while the trial judge might have erred in stating that the words “reasonable doubt” should be understood in their everyday ordinary sense, any such error was remedied by the fact that the trial judge went significantly further than had the trial judge in Lifchus, offering “interpretive assistance” and linking the concept of reasonable doubt to the evidence.

(2) Twaddle J.A., dissenting

(a) Admissibility of Cook’s Statement of Intention

140Twaddle J.A. stated that four questions arose from the trial judge’s admission of Giesbrecht’s evidence for the purpose of showing Cook’s intention or state of mind on the night of the murders, namely: (1) whether a statement that is adduced to prove the intention or state of mind of a deceased person should be considered for admission under a categorized exception to the hearsay rule, or only under the principled approach to the admission of hearsay evidence enunciated in Khan, supra, and subsequent cases; (2) whether Cook’s statement was in fact properly admitted for the purpose of illustrating Cook’s state of mind or intention; (3) whether the statement would have been admissible on the basis of necessity and reliability alone, pursuant to the principles established in Khan, supra, and subsequent cases; and (4) whether the jury was properly instructed as to the limits of the statement’s use.

141With respect to the issue of whether the principled approach to the admissibility of hearsay should be understood as having made the exceptions to the hearsay rule irrelevant, Twaddle J.A. stated that, in Smith, supra, this Court made it “very clear that the new approach was to be preferred” (p. 166). Nonetheless, Twaddle J.A. noted that, even in Smith, this Court analysed first whether statements of the deceased were admissible under the “present intentions” or “state of mind” exception to the hearsay rule, and then analysed whether they were admissible using the necessity and reliability approach. Twaddle J.A. stated that this bifurcated approach was likely taken because the trial judgment in Smith was rendered prior to Khan, supra. He stated that the appropriate test for the trial judge to have applied in the appellant’s case was likely the necessity and reliability test alone. However, in light of Smith and the trial judge’s focus on the “state of mind” exception, Twaddle J.A. dealt with both approaches to the admissibility of Giesbrecht’s testimony.

142Twaddle J.A. stated that it was generally accepted that evidence of a statement made by a person who has since died is admissible to prove the state of mind or present intentions of the deceased at the time of making the statement. The controversy, Twaddle J.A. noted, is over the impermissible uses to which such ante mortem statements might be put by the trier of fact. In particular, there is a concern that evidence regarding the state of mind of the deceased might be misused by the trier of fact to infer the state of mind, character, or subsequent actions of the accused rather than the declarant. Twaddle J.A. stated that this concern over impermissible uses of the evidence was relevant in determining whether the probative value of the hearsay statement outweighs its prejudicial effect upon the accused, relying in this regard upon the judgment of Doherty J. (as he then was) in R. v. P. (R.) (1990), 58 C.C.C. (3d) 334 (Ont. H.C.), which was cited with approval by this Court in Smith, supra.

143Twaddle J.A. stated that the trial judge had erred in finding that the probative value of Cook’s statement to Giesbrecht outweighed its prejudicial effect. Part of the error, Twaddle J.A. explained, stemmed from the trial judge’s finding that the relevant prejudice that would be suffered by the appellant if the statement was admitted was only that the jury might be inflamed by information suggesting he was inclined to commit a criminal act such as insurance fraud. Rather, Twaddle J.A. found, the prejudice was the more serious problem that the jury may have used evidence of Cook’s state of mind to infer the appellant’s state of mind and the appellant’s subsequent course of conduct, i.e., that the appellant did in fact meet with Cook later that night on the side of provincial highway 247.

144Twaddle J.A. held further that the trial judge had erred, after finding that Giesbrecht’s testimony was admissible pursuant to the “state of mind” exception, in not then holding a voir dire to determine the reliability of Cook’s statement, in accordance with the principled approach to hearsay admissibility set out in Khan and Smith. Twaddle J.A. found that Giesbrecht’s evidence was equally inadmissible on this basis. He reviewed several factors indicating that Cook’s statement was unreliable, including especially Cook’s strong motive to lie to Giesbrecht about his intentions on the night of the murders.

145Finally, Twaddle J.A. held that the trial judge had erred in his instruction to the jury regarding the permissible use of Cook’s statement. Twaddle J.A. noted that, rather than caution the jury against using Cook’s statement in order to prove the appellant’s state of mind and subsequent actions, as he ought to have done, the trial judge in fact invited the jury to draw this impermissible inference if they felt the evidence warranted it. Twaddle J.A. held that Cook’s statement to Giesbrecht was inadmissible on any basis, and that the statement’s admission, particularly with an erroneous instruction to the jury regarding its permissible use, constituted reversible error.

(b) Admissibility of Cheryl Ball’s Out-of-Court Identification

146Twaddle J.A. characterized this issue as whether a police officer’s evidence of an out-of-court identification of the accused by a witness is admissible where the witness does not testify as to whether she made the identification in question. Twaddle J.A. felt that it was important to consider this question, notwithstanding the Crown’s concession on appeal that the identification evidence might have been inadmissible, because he did not agree that the trial judge’s cautionary instruction to the jury regarding the frailty of the identification evidence was sufficient to dispel the possibility of a miscarriage of justice.

147Twaddle J.A. noted that the more common fact scenario is where a witness testifies that he or she recalls having made a previous identification of the accused, but is unable to repeat the identification in court or to recall who was identified. He stated that the law is fairly clear that, in such circumstances, a police officer may be permitted to testify simply as to the specific photograph or person pinpointed by the witness on that previous occasion: R. v. Tat (1997), 1997 2234 (ON CA), 117 C.C.C. (3d) 481 (Ont. C.A.). After reviewing some of the relevant jurisprudence, Twaddle J.A. distinguished the appellant’s case from cases such as Tat. He stated that, where a witness does not acknowledge having made a prior identification, there is a problem because it is no longer the eyewitness implicating the accused, but the police. There is no way to test whether, in fact, the eyewitness ever implicated the accused, because there is no testimony on point.

148Twaddle J.A. stated that, given the inherent frailty of identification evidence in any event, it would take exceptional circumstances for identification evidence introduced as hearsay to be sufficiently reliable to be admissible under the principled approach to hearsay admissibility. He did not find such exceptional circumstances to exist in the appellant’s case.

149Moving on to consider whether this ground of appeal could nonetheless be dismissed on the basis that the trial judge’s cautionary instruction to the jury negated any miscarriage of justice, Twaddle J.A. held that the jury verdict might reasonably have been different if the two officers had not been permitted to testify on the alleged identification by Cheryl Ball. He stated that Ball’s identification of the appellant was important because it placed the appellant in St. Norbert, whereas without Ball’s identification the appellant was last seen “halfway across the city from where the crime occurred” (p. 176). Although the trial judge cautioned the jury as to the frailty of the evidence, he also left it to the jury to decide whether to accept the identification evidence. Twaddle J.A. found that there was a reasonable possibility that the jury would not have convicted the appellant if Ball had not identified the appellant as the driver of the car that followed Cook and Weselowski to St. Norbert. This piece of evidence might have been the “last straw”, he said, allowing the Crown to meet its onus.

(c) The Instruction on Reasonable Doubt

150Twaddle J.A. stated that there is a problem in instructing a jury, as the trial judge did in this case, that the word “reasonable” in the criminal standard of proof beyond a reasonable doubt is to be understood in its everyday, ordinary meaning. By this understanding, he said, the term “reasonable doubt” would mean that even a legitimate doubt would not stand in the way of conviction if the doubt was only moderate, or only made guilt probable. Twaddle J.A. referred on this point to the reasons of this Court in Lifchus, supra, where Cory J. stated that, while a jury should be told that absolute certainty is not required in order to convict, the jury should also be told that something more than mere probability of guilt is required for a conviction.

151Twaddle J.A. stated that an omission to instruct the jury that a probability of guilt is not a sufficient basis upon which to convict, when the jury is instructed that absolute certainty of guilt is not required, will ordinarily be a fatal error justifying a new trial. He found that the requirement to instruct the jury that “beyond a reasonable doubt” means something more than “proof on the balance of probabilities” is particularly acute where, as in the appellant’s case, the trial judge tells the jury that the words “reasonable doubt” are to be understood in their ordinary, everyday meaning and not as having a special connotation in the criminal law context. Following Lifchus, supra, Twaddle J.A. held that a review of the trial judge’s charge to the jury as a whole in this case gave rise to a reasonable likelihood that the jury misapprehended the standard of proof, such that a new trial was required.

IV. Issues

152Three issues of law are presented by this appeal:

1. Did the majority of the Manitoba Court of Appeal err in law in affirming the trial judge’s decision to admit Jodie Giesbrecht’s testimony regarding a statement of intention made by the deceased Bernard Cook?

2. Did the majority of the Manitoba Court of Appeal err in law in affirming the trial judge’s decision to admit the testimony of Constables Madden and MacLeod regarding an out-of-court identification made by Cheryl Ball?

3. Did the majority of the Manitoba Court of Appeal err in law in finding that the trial judge had explained the concept of reasonable doubt to the jury in an adequate manner?

V. Analysis

A. The Hearsay Issues

(1) Introduction

153The law of hearsay in Canada and throughout the common law world has long been governed by a strict exclusionary rule relaxed by a complex array of exceptions. Recently, as noted in Smith, supra, at p. 932, this Court has moved in a new direction by adopting a principled approach to hearsay “governed by the principles which underlie the rule and its exceptions alike”. According to this approach, hearsay evidence may be admissible, notwithstanding the inapplicability of the categorical exceptions on the facts of the case, provided the criteria of necessity and reliability set out in Khan are met.

154The majority in the Court of Appeal below was of the view that hearsay evidence should be admissible either where the evidence falls within a traditional hearsay exception, or where the evidence is admissible using the principled approach developed in Khan and Smith. Twaddle J.A., dissenting, held that the principled approach must prevail in determining hearsay admissibility, and that the fact that the evidence in the particular case fell within a traditional hearsay exception was insufficient standing alone to determine its admissibility.

155My approach in these reasons is as follows. First of all, and with respect for the contrary view, I am of the opinion that the Court of Appeal erred in finding that Cook’s statement to Giesbrecht fit within the present intentions exception to the hearsay rule. I reach this conclusion for two reasons: the statement was made under circumstances of suspicion, and it was used to prove the intentions of someone other than the declarant. Having so concluded, it is necessary to ask whether the statement was admissible under the principled approach, as enunciated in Khan, supra, and Smith, supra. I conclude that it was not, much for the same reasons that it does not fall within the present intentions exception. Answering this question also raises issues respecting the interaction between the principled approach and the existing exceptions. I conclude that in the event of a conflict between the two, it is the principled approach that must prevail. The governing principles for hearsay admissibility must be reliability and necessity.

156I begin my analysis with a brief discussion of the nature and purpose of hearsay evidence, which I hope will be helpful in defining the precise hearsay issues in the appeal, and the problem posed for the trial process by hearsay evidence more broadly. Next, I consider in turn the admissibility of Jodie Giesbrecht’s testimony regarding Bernard Cook’s statement of intention, and the admissibility of Constables Madden’s and MacLeod’s testimony regarding Cheryl Ball’s out-of-court identification. As I will explain below, both statements were unreliable and should have been excluded under both the traditional exception and the principled approach.

(2) The Nature and Purpose of Hearsay Evidence

157The hearsay rule is a notable exception to the general evidentiary principle that all relevant evidence is admissible. To repeat, hearsay evidence was traditionally excluded unless the hearsay evidence came within one of the so-called exceptions to the rule. While it is very difficult to express an entirely complete definition of “hearsay evidence”, it will be useful to elaborate briefly on the principal defining features of hearsay before beginning to analyse the hearsay issues in this appeal.

158Hearsay issues most typically arise when a witness at trial testifies regarding the contents of an earlier, out-of-court statement. Bernard Cook’s alleged statement to Jodie Giesbrecht fits within this most simple definition. A statement can be hearsay even though the out-of-court statement was made by the witness at trial, or even in some cases where the only “statement” was a non-verbal assertion. The breadth of the hearsay rule is illustrated in this case by the hearsay evidence regarding Cheryl Ball’s out-of-court identification, discussed more fully below. Ball was a witness in the trial proceedings in this case yet has also been posited as an out-of-court declarant for the purpose of the hearsay rule, because Constables Madden and MacLeod testified about her out-of-court statements identifying the appellant.

159The difficulty encountered in defining hearsay has been acknowledged many times by courts and by learned authors on the law of evidence: see, e.g., R. v. Abbey, [1982] 2 S.C.R. 24, at pp. 40-41, per Dickson J. (as he then was), citing Phipson on Evidence (12th ed. 1976 (supplemented to 1980)), para. 625, at pp. 263-64. More recent definitions of hearsay have focussed upon the precise evidentiary concerns underlying the exclusionary rule, namely the absence of an opportunity for meaningful, contemporaneous cross-examination of the out-of-court declarant in court under oath or solemn affirmation, regarding the truth of the specific statement or expressive conduct that is sought to be admitted as proof of its contents. The central concern revolves around the inability to test the reliability of the declarant’s assertion. As stated by the Law Reform Commission of Canada in its 1975 Report on Evidence, at pp. 68-69:

Hearsay statements are excluded from evidence in trials because of the difficulty of testing their reliability. If a person who actually observed a fact is not in court, but a statement he made to someone about it is introduced in evidence, there is no way of inquiring into that person’s perception, memory, narration or sincerity. His statement about the fact might be false because he misperceived it or did not remember it correctly, or he may have misled the person to whom it was made because he used words not commonly used, or he may simply have lied about it. These factors, which determine the reliability of his statement, can only be tested if he is in the courtroom and subject to cross-examination.

160It was this fundamental concern with reliability to which Lamer C.J. was adverting when, speaking for the majority of the Court in R. v. B. (K.G.), [1993] 1 S.C.R. 740, at p. 764, he articulated the central “hearsay dangers” as “the absence of an oath or solemn affirmation when the statement was made, the inability of the trier of fact to assess the demeanour and therefore the credibility of the declarant when the statement was made (as well as the trier’s inability to ensure that the witness actually said what is claimed), and the lack of contemporaneous cross-examination by the opponent”. See similarly E. M. Morgan, “Hearsay Dangers and the Application of the Hearsay Concept” (1948), 62 Harv. L. Rev. 177; S. Schiff, Evidence in the Litigation Process (4th ed. 1993), vol. 1, at pp. 239-41.

161Cross on Evidence (7th ed. 1990) describes the hearsay rule in the following terms, at p. 42:

. . . an assertion other than one made by a person while giving oral evidence in the proceedings is inadmissible as evidence of any fact asserted. [Emphasis in original.]

It is of interest to note that the above formulation was adopted as correct by the House of Lords in R. v. Sharp, [1988] 1 W.L.R. 7, at p. 11. Though formulated differently, the definition of hearsay promulgated by the United States Federal Rule of Evidence 801 is similar: “‘Hearsay’ is a statement, other than one made by the declarant while testifying at the trial or hearing, offered in evidence to prove the truth of the matter asserted”. “Statement” is defined as “(1) an oral or written assertion or (2) nonverbal conduct of a person, if it is intended by the person as an assertion”. The most recent edition of Professor McCormick’s treatise on evidence (McCormick on Evidence (5th ed. 1999), vol. 2, at § 246) has adopted this definition.

162These articulations of the hearsay rule make clear that hearsay evidence is defined not by the nature of the evidence per se, but by the use to which the evidence is sought to be put: namely, to prove that what is asserted is true. When the out-of-court statement is offered for its truth, the inability to cross-examine or “test” the source of the evidence in court under oath or solemn affirmation as to the truth of the assertion undermines its reliability: see I. Younger, An Irreverent Introduction to Hearsay (1977), an address to the American Bar Association Annual Meeting, Atlanta, Georgia, August 11, 1976. In short, the essential defining features of hearsay are the purpose for which the evidence is adduced, and the absence of a meaningful opportunity to cross-examine the declarant in court under oath or solemn affirmation as to the truth of its contents.

163With this understanding of the nature and purpose of the hearsay rule in mind, I now turn to consider the hearsay evidence and exceptions at issue in this appeal.

(3) Admissibility of Cook’s Statement of Intention

(a) The Proposed Use of Cook’s Statement

164Assuming the veracity of Jodie Giesbrecht’s testimony at trial, approximately one hour before Cook was killed, he told Giesbrecht that he had to “go and do an Autopac scam with Robert”, meaning the appellant.

165The first stage of the hearsay analysis is to ask whether Cook’s out-of-court statement to Giesbrecht is sought to be adduced in order to prove the truth of its contents. The Crown acknowledged in the courts below that the purpose of adducing Cook’s statement was to illustrate Cook’s immediate intention, shortly prior to his death, to go with the appellant to wreck a car for insurance purposes. Cook’s intended course of action on the night of his murder was relevant, the Crown argued, because the jury could infer from this evidence of intention that Cook followed through on his intention, thus shedding light upon the location of the murders and the presence of a wrecked car at the scene. The Crown submitted that the statement, if true, also linked the appellant to Cook, the car, and the scene. In short, the intention to “go and do an Autopac scam with Robert” is the content of Cook’s statement, and the Crown sought to use the statement as proof of its contents.

166It is very important to note that the Crown’s proposed use of Cook’s statement to Giesbrecht went beyond proving just Cook’s intentions. In his closing address to the jury, Crown counsel argued that Cook’s statement proved that the appellant “proposed that [Autopac] scheme to Cook because it would isolate Cook if he followed along with it, without alarming him”. Given the trial judge’s instruction to the jury that it was “for [them] to decide whether the evidence of Cook’s statement about the scam goes as far as the Crown would have [them] believe”, it is apparent that the evidence was also admitted in order to prove the intentions, and subsequent actions in conformity therewith, of the appellant. By permitting the Crown’s argument to go to the jury, the trial judge expressly permitted the jury to infer the appellant’s intentions and subsequent conduct based on Cook’s statement of intention to Giesbrecht.

167In light of these proposed uses of the statement, Giesbrecht’s testimony regarding Cook’s statement to her is hearsay and would generally be inadmissible as such. It was an out-of-court statement, and it was offered by the Crown to prove the truth of the matter asserted; namely, that Cook intended to do an Autopac scam with Starr. The next stage of the hearsay analysis is to examine whether Giesbrecht’s evidence is nonetheless admissible under the appropriate exception.

(b) The “State of Mind” or “Present Intentions” Exception to the Hearsay Rule

(i) Scope of the Rule

168The Crown argued that the “state of mind” or “present intentions” exception to the hearsay rule applied to render Cook’s statement to Giesbrecht admissible. This exception was most recently discussed in detail by this Court in Smith, supra, where it was recognized that an “exception to the hearsay rule arises when the declarant’s statement is adduced in order to demonstrate the intentions, or state of mind, of the declarant at the time when the statement was made” (p. 925). Wigmore has argued that the present intentions exception also includes a requirement that a statement “be of a present existing state of mind, and must appear to have been made in a natural manner and not under circumstances of suspicion”: Wigmore on Evidence, vol. 6 (Chadbourn rev. 1976), at § 1725, p. 129 (emphasis in original). L’Heureux-Dubé J., at para. 63 of her reasons, denies that Wigmore’s suggestion has ever been adopted in our jurisprudence. As I will discuss below, regardless of whether the present intentions requirement ever had such a requirement, the principled approach demands that it must have it now. I will therefore examine the admissibility of Cook’s statement under the present intentions exception in light of that understanding.

169In Smith, Lamer C.J. explained that the exception as it has developed in Canada permits the admission into evidence of statements of intent or of other mental states for the truth of their contents and also, in the case of statements of intention in particular, to support an inference that the declarant followed through on the intended course of action, provided it is reasonable on the evidence for the trier of fact to infer that the declarant did so. At the same time, there are certain inferences that may not permissibly be drawn from hearsay evidence of the out-of-court declarant’s intentions. On this point, Lamer C.J. cited with approval (at p. 927) from the judgment of Doherty J. in P. (R.), supra, at pp. 343-44, where the case law was summarized as follows:

The evidence is not, however, admissible to show the state of mind of persons other than the deceased (unless they were aware of the statements), or to show that persons other than the deceased acted in accordance with the deceased’s stated intentions, save perhaps cases where the act was a joint one involving the deceased and another person. The evidence is also not admissible to establish that past acts or events referred to in the utterances occurred.

170As noted by J. Sopinka, S. N. Lederman and A. W. Bryant, in The Law of Evidence in Canada (2nd ed. 1999), at § 6.236, in Smith the Court adopted “the proposition that the admissibility of statements of intention were to be limited to the declarant’s state of mind and could not be used to prove the act or intention of any other person”. It is important to emphasize that even in “cases where the act was a joint one involving the deceased and another person”, the hearsay is not generally admissible to show the intentions of a third party. I draw this conclusion for two reasons.

171First, I can find no support in Canadian jurisprudence for the proposition that statements of intention are admissible against someone other than the declarant, apart from the one comment by Doherty J. noted above. Any other interpretation focusses on the exception and ignores the rule. In support of the proposition quoted above, Doherty J. cited three U.S. cases. All three had held that statements about joint acts were only admissible to prove the state of mind of the declarant: see Giles v. United States, 432 A.2d 739 (D.C. App. 1981), at pp. 745-46; United States v. Brown, 490 F.2d 758 (D.C. Cir. 1973); People v. Madson, 638 P.2d 18 (Colo. 1981). See also Shepard v. United States, 290 U.S. 96 (1933), at pp. 105-6; D. Kiesel, “One Person’s Thoughts, Another Person’s Acts: How the Federal Circuit Courts Interpret the Hillmon Doctrine” (1984), 33 Cath. U. L. Rev. 699, at pp. 738-39; and J. M. Maguire, “The Hillmon Case -- Thirty-Three Years After” (1925), 38 Harv. L. Rev. 709, at p. 721.

172Second, there are very good reasons behind the rule against allowing statements of present intention to be used to prove the state of mind of someone other than the declarant. As noted above, the central concern with hearsay is the inability of the trier of fact to test the reliability of the declarant’s assertion. When the statement is tendered to prove the intentions of a third party, this danger is multiplied. If a declarant makes a statement about the intentions of a third party, there are three possible bases for this statement: first, it could be based on a prior conversation with the accused; second, it could be based on a prior conversation with a fourth party, who indicated the third party’s intentions to the declarant; or third, it could be based on pure speculation on the part of the declarant. Under the first scenario, the statement is double hearsay. Since each level of double hearsay must fall within an exception, or be admissible under the principled approach, the mere fact that the declarant is making a statement of present intention is insufficient to render it admissible. The second level of hearsay must also be admissible.

173The other two scenarios also clearly require exclusion. If the statement about joint acts is based on a conversation with a fourth party, then the statement is triple hearsay, or worse. If, on the other hand, it is based on pure speculation, then it clearly is unreliable and does not fit within the rationale underlying the present intentions exception.

174In conclusion then, a statement of intention cannot be admitted to prove the intentions of someone other than the declarant, unless a hearsay exception can be established for each level of hearsay. One way to establish this would obviously be the co-conspirator exception: see R. v. Carter, [1982] 1 S.C.R. 938; Sopinka, Lederman and Bryant, supra, at pp. 303-7. This is no doubt what Doherty J. was referring to in P. (R.), supra, when he spoke of “cases where the act was a joint one involving the deceased and another person” (p. 344). Barring the applicability of this or some other exception to each level of hearsay involved, statements of joint intention are only admissible to prove the declarant’s intentions.

(ii) Application to this Appeal

175As noted above, the trial judge below admitted Cook’s statement to Jodie Giesbrecht as admissible hearsay evidence of Cook’s “state of mind to go with the accused”, and as evidence from which the jury could infer that Cook acted in accordance with his stated intentions. He also admitted it as evidence from which the jury could conclude that the appellant “proposed that [Autopac] scheme to Cook because it would isolate Cook if he followed along with it, without alarming him”. Therefore the evidence was admitted not only to prove the intentions of the declarant Cook, but also of a third party – the appellant-accused.

176All three judges in the Court of Appeal below accepted as a preliminary proposition that Cook’s statement to Giesbrecht fell within the traditional “present intentions” or “state of mind” exception to the hearsay rule. Although Twaddle J.A., dissenting, considered it inappropriate to admit Cook’s statement, especially to show the intentions and subsequent course of conduct of the appellant, he would have excluded the statement through an application of the principled approach to hearsay admissibility, or alternatively pursuant to the court’s residual discretion to exclude evidence where its prejudicial effect outweighs its probative force. Ordinarily, given our limited jurisdiction on appeals as of right, our scope of review would be limited to those issues raised by Twaddle J.A.’s dissent. However, by ordering a re-hearing we expanded the scope of our review to include the question, answered unanimously in the affirmative below, of whether Cook’s statement to Giesbrecht falls within the present intentions exception.

177With great respect to the Court of Appeal, I conclude that the trial judge erred in admitting Cook’s statement to Giesbrecht under the present intentions exception and, having admitted it, in not limiting its use by the jury, for three reasons. First, the statement contained no indicia of reliability since it was made under circumstances of suspicion; second, the trial judge failed to instruct the jury that the statement was only admissible as evidence regarding the intentions of Cook, not the appellant; and third, even if it had been properly limited, the evidence was more prejudicial than probative.

178Turning first to the circumstances of suspicion, I agree with Twaddle J.A. that the statement lacked circumstantial guarantees of trustworthiness. As Twaddle J.A. noted, Cook and Giesbrecht had been romantically involved for almost two years. Cook had lived with Giesbrecht and her mother for a time, and had spent the night before his murder with Giesbrecht, after getting out of jail. Then, in the early morning hours of August 21, 1994, Giesbrecht observed Cook in the car of another woman, Darlene Weselowski. Giesbrecht testified that she thought Cook might try to “take off on her” if he saw Giesbrecht approaching the car, and she endeavoured not to be seen by Cook until she was close enough to talk to him. After an initial confrontation, Giesbrecht walked away into an alley behind the gas station, where Cook followed her. Their conversation ended in an argument because Cook was with Weselowski. She was angry at Cook for being with another woman, and asked him expressly why Cook would not come home with her rather than remain with Weselowski. It was at this point, and in this heated context, that Cook said he was going to engage in an Autopac scam with the appellant, who was sitting in a car nearby. Giesbrecht testified that it was unusual for Cook to discuss such business matters with her.

179Twaddle J.A. found that the circumstances surrounding the making of the statement cast serious doubt upon the reliability of the statement. The possibility that Cook was untruthful could not be said to have been substantially negated. Twaddle J.A. relied, in particular, upon the fact that Cook may have had a motive to lie in order to make it seem that he was not romantically involved with Weselowski, and upon the ease with which Cook could point to the appellant, who was sitting nearby in a car but out of earshot, as being the person with whom he was going to do a scam. In my view, Twaddle J.A. was correct in finding that these circumstances bring the reliability of Cook’s statement into doubt. The statement was made under “circumstances of suspicion”, and therefore does not fall within the present intentions exception. The statement should have been excluded.

180The statement was also inadmissible for the purpose tendered because it was a statement of joint intention. Even assuming, contrary to the foregoing, that the Court of Appeal was correct in concluding that Cook’s statement was admissible with respect to his own intentions under the present intentions exception, we must remember that Cook’s statement was at least double hearsay, if not worse. The Crown did not establish how Cook became qualified to comment on the appellant’s intentions. The only hearsay exception that could conceivably apply is the co-conspirator exception. However, this exception was never raised at trial, and therefore the trial judge did not even attempt to comply with the strict requirements for this exception set out in Carter, supra.

181I should emphasize that statements of intention are not automatically inadmissible simply because they refer to joint acts. As Twaddle J.A. noted at p. 167 of his dissent, “[t]he controversy is not so much over whether such a statement can be admitted in evidence, but rather over the use to which it can be put”. Therefore statements of intention, which refer to intentions of persons other than the declarant, may be admissible if the trial judge clearly restricts their use to proving the declarant’s intentions, and if it is more probative than prejudicial.

182McLachlin C.J. disagrees with the foregoing, particularly with the statement of L’Heureux-Dubé J. at para. 67 (with which I agree) that “[i]t is common cause that the ‘present intentions’ exception may not be used to infer that a third party acted in accordance with the declarant’s stated intention”. McLachlin C.J., at para. 13, argues that she would “not state the matter so categorically” as “in some circumstances the statement of joint intention can be fairly considered along with other evidence in deciding what the third party did”. In fact, no such blanket rule is being laid down in this case. Doherty J.’s statement on this point in P. (R.), supra, at p. 344 (as mentioned by L’Heureux-Dubé J. at para. 67 of her reasons) correctly states the law:

The rules of evidence as developed to this point do not exclude evidence of utterances by a deceased which reveal her state of mind, but rather appear to provide specifically for their admission where relevant. The evidence is not, however, admissible to show the state of mind of persons other than the deceased (unless they were aware of the statements), or to show that persons other than the deceased acted in accordance with the deceased's stated intentions, save perhaps cases where the act was a joint one involving the deceased and another person. The evidence is also not admissible to establish that past acts or events referred to in the utterances occurred.

This statement of the present intention exception was adopted by this Court in Smith, supra, at p. 927.

183Where we disagree, however, is over the application of these principles in this case. McLachlin C.J., at para. 13, concludes that “Cook’s statement may be viewed as one piece of circumstantial evidence supporting the inference that Starr was with the deceased later that night”. In my opinion, this is an impermissible use of Cook’s statement for the reasons given above. The jury should have been explicitly warned that they could not rely on Cook’s statement to infer the appellant’s intentions.

184McLachlin C.J. further concludes that the trial judge’s direction to the jury in this case provided a sufficient warning as to how Cook’s statement about the Autopac scam could legitimately be used, despite the fact that an express instruction on this point was not given. With respect, I also disagree. It is well-established that when a piece of evidence may conceivably be put to both proper and improper uses, the trial judge in a criminal case must give the jury a limiting instruction regarding the permissible inferences that may be drawn from the evidence. See R. v. D. (L.E.), [1989] 2 S.C.R. 111, at p. 128; R. v. Corbett, [1988] 1 S.C.R. 670, at p. 695; R. v. D. (L.E.) (1987), 1987 2536 (BC CA), 20 B.C.L.R. (2d) 384 (C.A.), at pp. 398-400, per McLachlin J.A. (as she then was), dissenting; McCormick on Evidence, supra, vol. 1, at § 59; P. K. McWilliams, Canadian Criminal Evidence (3rd ed. (looseleaf)), vol. 1, at p. 3-8; Phipson on Evidence (15th ed. 2000), at p. 110; Sopinka, Lederman and Bryant, supra, at § 2.83; Cross and Tapper on Evidence (9th ed. 1999), at p. 58; Wigmore on Evidence, vol. 1 (Tillers rev. 1983), at § 13; United States Federal Rule of Evidence 105. Aside from her cogent dissent in D. (L.E.), supra, which was upheld on appeal, authority in support of my position comes from the reasons of McLachlin J. in R. v. Rockey, [1996] 3 S.C.R. 829, at para. 38:

It has long been accepted that trial judges charging juries on out‑of‑court statements must instruct them on how they may use the statements — whether as evidence of the truth of their contents or for some other purpose, such as credit. In this case the trial judge did not do this. It may be that it was apparent to everyone in the courtroom that the subsequent statements were tendered on the issue of consistency, as the majority of the Court of Appeal suggests. Nevertheless, the usual rule requires this to be stated expressly. [Emphasis added.]

185I can see no reason why hearsay statements about joint acts should be immune from this most sensible rule. Cook’s statement, assuming its admissibility, may be used to show the intentions of the declarant only, not the intentions of the appellant. The jury should have been instructed as such. As McLachlin J. pointed out in Rockey, juries cannot be trusted to determine what are proper and improper uses of evidence; indeed, the entire law of evidence is premised on the notion that relevant, probative evidence should on occasion be kept from juries because of their potential inability to assess its weight properly.

186In this appeal, the trial judge did not instruct the jury on the proper uses of Cook’s statement; in fact, he did the opposite by expressly inviting the jury to use the evidence to infer the appellant’s intentions. In so doing, he clearly committed an error of law warranting reversal.

187Finally, I would exclude Cook’s statement as more prejudicial than probative. The trial judge did not make a finding on the issue of reliability. His focus was upon the impermissible inferences that the jury might draw from otherwise admissible hearsay, and he regarded the primary prejudice to the appellant to be that the jury might infer that he was the type of person likely to commit insurance fraud. However, as noted above, this was not the primary source of prejudice.

188The trial judge erred by not considering whether “the prejudicial effect of the prohibited use of the evidence [i.e., the appellant’s intentions] overbears its probative value on the permitted use [i.e., Cook’s intentions]”: Watt’s Manual of Criminal Evidence (1999), at p. 281 (emphasis in original). The impermissible inferences that the jury might well have drawn from Cook’s statement are that the appellant was in the car that followed Cook, that the appellant was alone in the car (since Cook referred only to the appellant), and that the appellant went with Cook as part of a plan to lure Cook to a secluded area and kill him. These were the specific impermissible inferences that the jury might have drawn in this regard — indeed, they are inferences that the Crown specifically invited the jury to draw — quite apart from the inferences that they might have drawn regarding his general criminality. In my view, Twaddle J.A. was correct in finding that the prejudicial effect of the admission of Cook’s statement accordingly outweighed the statement’s probative value. The statement ought to have been excluded on this basis as well: see R. v. Seaboyer, [1991] 2 S.C.R. 577, at pp. 609-11, per McLachlin J.

189I have concluded that Cook’s statement does not fall within the present intentions exception to the hearsay rule. Earlier, I recognized the conflict regarding the precise scope of this exception, and noted that the principled approach requires adopting Wigmore’s requirement that the statement not be made under circumstances of suspicion. I turn now to this question of the relationship between the principled approach and the traditional hearsay exceptions.

(c) Admissibility of Cook’s Statement to Giesbrecht Under the Principled Approach

(i) Why the Principled Approach Should Prevail

Previous Hearsay Jurisprudence

190Before turning to the question of whether Cook’s statement to Giesbrecht might be admissible under the principled approach, I would like to offer some general observations on the relationship between the principled approach and the hearsay exceptions. Speaking for this Court in Smith, supra, at pp. 932-33, Lamer C.J. emphasized that the rules regarding hearsay admissibility in Canada were changed by the Court’s decision in Khan, supra:

. . . Khan should not be understood as turning on its particular facts, but, instead, must be seen as a particular expression of the fundamental principles that underlie the hearsay rule and the exceptions to it. What is important, in my view, is the departure signalled by Khan from a view of hearsay characterized by a general prohibition on the reception of such evidence, subject to a limited number of defined categorical exceptions, and a movement towards an approach governed by the principles which underlie the rule and its exceptions alike. . . .

This Court’s decision in Khan, therefore, signalled an end to the old categorical approach to the admission of hearsay evidence. Hearsay evidence is now admissible on a principled basis, the governing principles being the reliability of the evidence, and its necessity.

Similarly, at p. 930, Lamer C.J. stated:

The decision of this Court in Khan . . . should be understood as the triumph of a principled analysis over a set of ossified judicially created categories.

191In R. v. U. (F.J.), [1995] 3 S.C.R. 764, the Court suggested the need to reconsider the traditional hearsay exceptions in light of the principled approach established by Khan and Smith. Speaking for the majority of the Court, Lamer C.J. pointed to the illogical nature of some of the exceptions, and to the need for principled reform (at para. 20):

Briefly stated, Khan and Smith announced this Court’s commitment to ensure that the rule against the admission of hearsay as evidence would be sufficiently flexible to adapt to new situations. The hearsay rule and its rigidly formulated exceptions had become a sometimes illogical and frequently confusing series of pigeon-hole categories. Khan and Smith examined the principles underlying the hearsay rule to ensure that new developments in the evidentiary treatment of hearsay would reflect those tenets. [Emphasis added.]

Similarly, in R. v. Hawkins, [1996] 3 S.C.R. 1043, at para. 66, the majority of the Court emphasized:

In Khan and Smith, this Court signalled the beginning of a modern principled framework for defining exceptions to the hearsay rule. The Court rejected the traditional approach of the common law premised on rigid, categorical exceptions to the hearsay rule in favour of a more flexible approach which seeks to give effect to the underlying purposes of the rule. [Emphasis added.]

192Up to the present, this Court’s application of the principled approach to hearsay admissibility in practice has involved only expanding the scope of hearsay admissibility beyond the traditional exceptions. The focus of the Court’s analysis and commentary has been upon the need to increase the flexibility of the existing exceptions, and not specifically upon the need to re-examine the exceptions themselves. However, this case requires that we examine an exception to the hearsay rule and determine its co-existence with the principled approach. As I will discuss further, to the extent that the various exceptions may conflict with the requirements of a principled analysis, it is the principled analysis that should prevail.

193The applicability of the principled approach to the exclusion as well as inclusion of evidence is implicitly confirmed by Khan itself. At p. 543 of her reasons for a unanimous panel, McLachlin J. addressed recent developments in the law with respect to the testimony of child victims of sexual assault:

These developments underline the need for increased flexibility in the interpretation of the hearsay rule to permit the admission in evidence of statements made by children to others about sexual abuse. In so far as they are tied to the exception to the hearsay rule of spontaneous declarations, however, they suffer from certain defects. There is no requirement that resort to the hearsay evidence be necessary. Even where the evidence of the child might easily be obtained without undue trauma, the Crown would be able to use hearsay evidence. Nor is there any requirement that the reliability of the evidence in the particular case be established; hence inherently unreliable evidence might be admitted. [Emphasis added.]

Thus the Court in Khan declined simply to enlarge the traditional exceptions because doing so might have allowed into evidence unnecessary or unreliable evidence, thereby falling afoul of the principled approach. It therefore makes little sense to say that the current exceptions need not meet the same standard.

194In this light, it is noteworthy that several provincial courts of appeal have already begun a reconsideration of the appropriateness of relying upon the traditional hearsay exceptions. Several Courts of Appeal have suggested that “the traditional rules governing the admissibility of hearsay evidence are no longer controlling. Reliability and necessity are now the determining factors”: see, e.g., R. v. Kelly (1999), 1999 13120 (NB CA), 213 N.B.R. (2d) 1 (C.A.), at p. 39; and R. v. R. (D.) (1995), 1995 3972 (SK CA), 98 C.C.C. (3d) 353 (Sask. C.A.), at p. 428, per Vancise J.A. (dissenting, but not on this point). However, even a relatively cursory review of lower court decisions reveals a wide range of approaches to the principled approach and its effect on existing exceptions. See R. v. Grand-Pierre (1998), 1998 13202 (QC CA), 124 C.C.C. (3d) 236 (Que. C.A.), at pp. 242-43; R. v. Bisson (1997), 1997 10791 (QC CA), 114 C.C.C. (3d) 154 (Que. C.A.), at p. 177; R. v. Chahley (1992), 1992 2387 (BC CA), 72 C.C.C. (3d) 193 (B.C.C.A.); Wepruk (Guardian ad litem of) v. McMillan Estate (1993), 1993 6857 (BC CA), 77 B.C.L.R. (2d) 273 (C.A.); R. v. Crossley (1997), 1997 3280 (BC CA), 117 C.C.C. (3d) 533 (B.C.C.A.); R. v. Collins (1997), 1997 12516 (BC CA), 118 C.C.C. (3d) 514 (B.C.C.A.); and R. v. Warner (1994), 1994 842 (ON CA), 94 C.C.C. (3d) 540 (Ont. C.A.), at p. 551.

Commentators

195Questions as to whether and how the traditional hearsay exceptions should be modified in light of the principled approach have also been discussed in the academic literature. Some commentators favour reforming the exceptions to make them an inherent aspect of a logical and coherent approach to hearsay admissibility based on reliability and necessity: see, e.g., M. Rosenberg, “B. (K.G.) -- Necessity and Reliability: The New Pigeon-holes” (1993), 19 C.R. (4th) 69; D. Rowsell, “Necessity and Reliability: What is the Impact of Khan on the Admissibility of Hearsay in Canada?” (1991), 49 U.T. Fac. L. Rev. 294; E. Then, “Dying Declarations Following Khan and Smith: Are They Necessarily Reliable?”, in National Criminal Law Program: Criminal Evidence (1994), vol. 1, section 6.5; A. L.-T. Choo, Hearsay and Confrontation in Criminal Trials (1996), at pp. 166-70.

196In “B. (K.G.) -- Necessity and Reliability: The New Pigeon-holes”, an article published before his appointment to the bench, Marc Rosenberg states that this Court’s decisions in Khan, Smith, and B. (K.G.) signify that “it must . . . be open to the courts to revisit any established exception to determine whether that exception can still be justified on the basis of necessity and reliability” (p. 71). Rosenberg explains, at pp. 80-81:

[T]he reason which justifies an existing exception may in time be found to be wanting. The exceptions were created as the courts from time to time applied common sense and experience, but ideas which seemed reasonable in the 19th century may appear questionable in the late 20th century. As early as 1913 Hamilton L.J. in considering the exception for statements against pecuniary interest where the circumstantial guarantee of trustworthiness is said to lie in the fact that persons will not lie to their pecuniary disadvantage observed that as a reason for admitting hearsay this one was “sordid and unconvincing” noting that “Men lie for so many reasons and some for no reason at all; and some tell the truth without thinking about their pockets”.

197Other writers have been less convinced of the benefits of altering the traditional hearsay exceptions. Their concerns have generally been focussed upon the harm that they suggest would result from the complete abolition of the exceptions, rather than upon the effects of a piecemeal reform of the exceptions to comply with the principled approach. See, e.g., P. B. Carter, “Hearsay: Whether and Whither?” (1993), 109 L.Q.R. 573; D. A. R. Thompson, “The Supreme Court Goes Hunting and Nearly Catches a Hearsay Woozle” (1995), 37 C.R. (4th) 282.

198The concerns expressed in the academic commentary warrant careful consideration, and are of assistance in determining how best to rationalize the traditional exceptions in light of the underlying principles of the rule. In my view, there are two paramount reasons to reconsider the hearsay exceptions, namely: first, trial fairness and the integrity of the justice system; and second, the intellectual coherence of the law of hearsay.

Why the Exceptions Must be Rationalized

199As I have already discussed, a fundamental concern with reliability lies at the heart of the hearsay rule. By excluding evidence that might produce unfair verdicts, and by ensuring that litigants will generally have the opportunity to confront adverse witnesses, the hearsay rule serves as a cornerstone of a fair justice system.

200In Khan, Smith, and subsequent cases, this Court allowed the admission of hearsay not fitting within an established exception where it was sufficiently reliable and necessary to address the traditional hearsay dangers. However, this concern for reliability and necessity should be no less present when the hearsay is sought to be introduced under an established exception. This is particularly true in the criminal context given the “fundamental principle of justice, protected by the Charter, that the innocent must not be convicted”: R. v. Leipert, [1997] 1 S.C.R. 281, at para. 24, quoted in R. v. Mills, [1999] 3 S.C.R. 668, at para. 71. It would compromise trial fairness, and raise the spectre of wrongful convictions, if the Crown is allowed to introduce unreliable hearsay against the accused, regardless of whether it happens to fall within an existing exception.

201In addition to improving trial fairness, bringing the hearsay exceptions into line with the principled approach will also improve the intellectual coherence of the law of hearsay. It would seem anomalous to label an approach “principled” that applies only to the admission of evidence, not its exclusion. Rationalizing the hearsay exceptions into the principled approach shows that the former are simply specific manifestations of general principles, rather than the isolated “pigeon-holes” referred to in U. (F.J.), supra, at para. 20.

The Continuing Importance of the Existing Exceptions

202Having recognized the primacy of the principled approach, it is nevertheless important for a court to exercise a certain degree of caution when reconsidering the traditional exceptions. While the exceptions may need to be reexamined in light of the principled approach, their complete abolition is not the answer. Rather, the exceptions continue to play an important role under the principled approach. Our task therefore is to reconcile the traditional exceptions with the principled approach.

203One important function that the hearsay exceptions have served has been to add predictability and certainty to the law of hearsay. In light of the exceptions, and regardless of how illogical or arbitrary they may be, litigants can be more or less certain when going into court of the types of issues that will be relevant in debating admissibility in a particular context, and of the likelihood that the evidence will indeed be admitted. This certainty has fostered greater efficiency in the use of court time both at trial and on appeal, and has facilitated the task of the too frequently overburdened trial judge who is called upon to rule on hearsay admissibility with speed and considerable regularity. As suggested by Rosenberg, supra, at p. 75, a complete abolition of the exceptions and their replacement by the principled approach standing alone would complicate the judicial task:

[I]t is unfair to simply leave the decision as to the admission of hearsay completely open-ended, leaving the trial judges without any analytical tool for determining what is reasonably necessary. To simply define the test in terms of reliability and necessity is just too vague to be of any practical use. With all its rigidity and anomalies a hearsay rule consisting of a broad rule of exclusion with certain well-defined exceptions was relatively easy to apply. Taking away all of the rules and replacing them with necessity and reliability, while perhaps not inviting chaos, does make the role of the trial judge that much more difficult. [Emphasis in original.]

204Second, in addition to serving the utilitarian goals of providing greater certainty and fostering judicial efficiency, the exceptions have served an explanatory or educative function, instructing litigants and judges about the relevant factors to consider in determining whether to admit a particular type of hearsay evidence, or whether to admit hearsay in a particular factual context. Different hearsay scenarios by their nature raise different reliability concerns, and different issues of necessity. The specific requirements of the individual exceptions have had the useful effect of focussing attention upon the peculiar factors that make it desirable, or undesirable, to admit a particular form of out-of-court statement. This should be no surprise given Lamer C.J.’s statement in Smith, supra, that the principled approach is “governed by the principles which underlie the [hearsay] rule and its exceptions alike” (p. 932). Since the principled approach is implicit in most of the exceptions, they are likely to be strong evidence of necessity and reliability.

205It is true that there is guidance inherent in the principled approach itself, which directs a court to gauge whether a particular hearsay statement is reliable and whether its admission is necessary in the circumstances. However, the exceptions are more fact-specific and contextually sensitive. Properly modified to conform to the principled approach, the exceptions are practical manifestations of the principled approach in concrete and meaningful form. Indeed, it is precisely to illustrate the form of analysis under the principled approach that must occur in a particular factual context that this Court in its recent cases has outlined carefully the type of inquiry that must occur when dealing with a particular type of hearsay, whether it be the testimony of a child witness (Khan, supra), a prior inconsistent statement (B. (K.G.), supra, and U. (F.J.), supra), or prior testimony (Hawkins, supra). Some commentators have suggested that the Court’s recent hearsay jurisprudence may accordingly be seen as creating new hearsay exceptions to supplement the traditional exceptions: see, e.g., Carter, supra, at p. 579. Perhaps a more accurate characterization is to say that all of the hearsay “exceptions” should be seen simply as concrete examples of the practical application of the purpose and principles of the hearsay rule in a particular context.

206A third important function played by the traditional hearsay exceptions is that they teach us about the historical and contemporary rationale for admitting certain forms of hearsay. It has quite properly been noted that some hearsay exceptions allow for the admission of evidence that is unreliable, unnecessary, or both. In the interest of fairness for the litigant against whom it is used, unreliable hearsay evidence should never be admitted. Apart from that, a review of the traditional exceptions reveals that there are reasons beyond “pure” necessity why a court might wish to admit reliable hearsay evidence. This point was addressed by Lamer C.J. in B. (K.G.), at pp. 796-97, where he explained that the need to permit the admission of certain forms of hearsay can stem not only from the unavailability of the out-of-court declarant, but also from the quality of the evidence itself. Lamer C.J. cited Professor Wigmore’s explanation (Wigmore on Evidence, vol. 5 (Chadbourn rev. 1974), at p. 253) that some hearsay evidence “may be such that we cannot expect, again, or at this time, to get evidence of the same value from the same or other sources” (emphasis in original). Such hearsay may be admitted, where appropriate, less on the basis of necessity and more on the basis of “expediency or convenience”. The traditional exceptions are useful, therefore, because they are instructive as to the types of situations that may produce hearsay that is the best evidence in the circumstances.

207There are other important functions served by the traditional hearsay exceptions, but the issues I have referred to are sufficient to illustrate that it is neither desirable nor necessary to abolish these exceptions outright. The more appropriate approach is to seek to derive the benefits of certainty, efficiency, and guidance that the exceptions offer, while adding the benefits of fairness and logic that the principled approach provides. The task is to rid the exceptions of their arbitrary aspects, in order to avoid admitting hearsay evidence that should be excluded.

(ii) Is Cook’s Statement to Giesbrecht Admissible Under the Principled Approach?

208For much the same reasons why the statement did not meet the requirements for admissibility under the present intentions exception, I conclude that the statement is not admissible under the principled approach either. This should not be particularly surprising -- as I have discussed above, the traditional exceptions are based on the concepts of reliability and necessity. While occasionally, as in Khan, supra, a statement not falling within an existing exception will be admissible under the principled approach, this will likely be the exception, not the rule.

209The first requirement for admissibility under the principled approach is reliability. Given my conclusion above that Cook’s statement was made under “circumstances of suspicion”, it follows that the statement was not reliable. Nor are there any other circumstantial guarantees of trustworthiness that could render the statement reliable. Having found that the statement is unreliable, it is unnecessary to go on to ask whether it was necessary or not. I conclude that Cook’s statement to Giesbrecht was inadmissible under the principled approach. Since it does not fall under an existing exception either, for all the reasons given above, the courts below erred in admitting this evidence. There being no serious argument that the error was one that could be saved by the curative provisio, s. 686(1)(b)(iii) of the Criminal Code, R.S.C., 1985, c. C-46, the appeal must be allowed.

(d) Revisiting the Hearsay Exceptions in Future Cases

210While the foregoing is perhaps sufficient to dispose of this appeal, the majority of the arguments both in the court below and before this Court focussed on the relationship between the principled approach and the traditional exceptions. Given this, and the substantial controversy among both lower courts and commentators regarding the appropriate relationship between the principled approach and the traditional hearsay exceptions, I would like to offer some general remarks on this issue. I have no doubt that the lower courts will develop guidelines over time as circumstances warrant.

211I hope from the foregoing that it is clear that the existing exceptions are a long-standing and important aspect of our law of evidence. I am cognizant of their important role, and the need for caution in reforming them. Given their continuing importance, I would expect that in the clear majority of cases, the presence or absence of a traditional exception will be determinative of admissibility.

212While Khan, supra, and its progeny have set out the approach for evidence falling outside a traditional exception, I would note that evidence falling within a traditional exception is presumptively admissible. These exceptions traditionally incorporate an inherent reliability component. For example, testimony in former proceedings is admitted, at least in part, because many of the traditional dangers associated with hearsay are not present. As pointed out in Sopinka, Lederman and Bryant, supra, at pp. 278-79:

. . . a statement which was earlier made under oath, subjected to cross-examination and admitted as testimony at a former proceeding is received in a subsequent trial because the dangers underlying hearsay evidence are absent. [Emphasis added.]

Other exceptions are based not on negating traditional hearsay dangers, but on the fact that the statement provides circumstantial guarantees of reliability. This approach is embodied in recognized exceptions such as dying declarations, spontaneous utterances, and statements against pecuniary interest.

213All this being said, it is also clear that the logic of the principled approach demands that it must prevail in situations where it is in conflict with an existing exception. For example, had there been any doubt in this appeal whether the present intentions exception required that the statement not be made under circumstances of suspicion, the principled approach would require holding that it does now. Hearsay evidence may only be admitted if it is necessary and reliable, and the traditional exceptions should be interpreted in a manner consistent with this requirement.

214In some rare cases, it may also be possible under the particular circumstances of a case for evidence clearly falling within an otherwise valid exception nonetheless not to meet the principled approach’s requirements of necessity and reliability. In such a case, the evidence would have to be excluded. However, I wish to emphasize that these cases will no doubt be unusual, and that the party challenging the admissibility of evidence falling within a traditional exception will bear the burden of showing that the evidence should nevertheless be inadmissible. The trial judge will determine the procedure (whether by voir dire or otherwise) to determine admissibility under the principled approach’s requirements of reasonable necessity and reliability.

215In this connection, it is important when examining the reliability of a statement under the principled approach to distinguish between threshold and ultimate reliability. Only the former is relevant to admissibility: see Hawkins, supra, at p. 1084. Again, it is not appropriate in the circumstances of this appeal to provide an exhaustive catalogue of the factors that may influence threshold reliability. However, our jurisprudence does provide some guidance on this subject. Threshold reliability is concerned not with whether the statement is true or not; that is a question of ultimate reliability. Instead, it is concerned with whether or not the circumstances surrounding the statement itself provide circumstantial guarantees of trustworthiness. This could be because the declarant had no motive to lie (see Khan, supra; Smith, supra), or because there were safeguards in place such that a lie could be discovered (see Hawkins, supra; U. (F.J.), supra; B. (K.G.), supra).

216And indeed, lower courts have recognized that the absence of a motive to lie is a relevant factor in admitting evidence under the principled approach: see R. v. L. (J.W.) (1994), 1994 8734 (ON CA), 94 C.C.C. (3d) 263 (Ont. C.A.); R. v. Tam (1995), 1995 16084 (BC CA), 100 C.C.C. (3d) 196 (B.C.C.A.); R. v. Rose (1998), 1998 4564 (BC CA), 108 B.C.A.C. 221; see also B. P. Archibald, “The Canadian Hearsay Revolution: Is Half a Loaf Better Than No Loaf at All?” (1999), 25 Queen’s L.J. 1, at p. 34. Conversely, the presence of a motive to lie may be grounds for exclusion of evidence under the principled approach. Put another way, it is the role of the trial judge to determine threshold reliability by satisfying him- or herself that notwithstanding the absence of the declarant for cross-examination purposes, the statement possesses sufficient elements of reliability that it should be passed on to be considered by the trier of fact.

217At the stage of hearsay admissibility the trial judge should not consider the declarant’s general reputation for truthfulness, nor any prior or subsequent statements, consistent or not. These factors do not concern the circumstances of the statement itself. Similarly, I would not consider the presence of corroborating or conflicting evidence. On this point, I agree with the Ontario Court of Appeal’s decision in R. v. C. (B.) (1993), 1993 8564 (ON CA), 12 O.R. (3d) 608; see also Idaho v. Wright, 497 U.S. 805 (1990). In summary, under the principled approach a court must not invade the province of the trier of fact and condition admissibility of hearsay on whether the evidence is ultimately reliable. However, it will need to examine whether the circumstances in which the statement was made lend sufficient credibility to allow a finding of threshold reliability.

(4) Admissibility of Cheryl Ball’s Out-of-Court Identification

218The analysis of Cheryl Ball’s identification evidence proceeds in much the same way as the foregoing analysis of Cook’s statement to Giesbrecht. I conclude that the statement is hearsay, and that it does not fall within an existing exception. Since I also conclude that the evidence was inadmissible under the principled approach, it follows that the trial judge erred in admitting the evidence.

219Again I will assume for the purposes of analysis the veracity of the testimony of Constables Madden and MacLeod at trial. They testified that Cheryl Ball identified the appellant as being a man she had seen talking to Cook at the Mohawk gas station on the night of the murders, and, according to Madden, as the man who was also “probably driving the other car”. It is clear that these statements are hearsay and so are inadmissible unless reliable and necessary, because the Crown sought to use the statements to show that the appellant was present at the Mohawk station, thus creating an inference that he had followed them from the station and was driving the smaller car Cheryl Ball saw when she left Cook and Weselowski in St. Norbert. The issue before the trial judge was whether the evidence was admissible pursuant to the “prior identification” exception to the hearsay rule. The trial judge found that the exception did apply.

220It is not necessary in this case to review the entirety of the “prior identification” exception to the hearsay rule. The narrow issue before this Court is whether hearsay evidence of an out-of-court identification by a trial witness is admissible where the witness does not testify at trial that she made the identification. On the particular facts of this appeal, Cheryl Ball testified at trial that she told police that one of the men in the photographs she was shown on August 23, 1994 “look[ed] kind of familiar”, but she did not testify that she had seen the person at the Mohawk gas station or at the wheel of the car that followed the Weselowski station wagon to St. Norbert.

221The scope of the “prior identification” exception to the hearsay rule was recently thoroughly canvassed in the lucid reasons of Doherty J.A. in Tat, supra. As Doherty J.A. sets out, there are two situations in which out-of-court statements of identification may be admitted for the truth of their contents. First, “prior statements identifying or describing the accused are admissible where the identifying witness identifies the accused at trial” (pp. 497-98). Second, such statements are admissible “where the identifying witness is unable to identify the accused at trial, but can testify that he or she previously gave an accurate description or made an accurate identification” (p. 500). In the latter circumstance, Doherty J.A. explained, “the identifying witness may testify to what he or she said or did on those earlier occasions and those who heard the description given by the witness or witnessed the identification made by the witness may give evidence of what the witness said or did” (ibid.).

222In the present case, only the second branch of the “prior identification” exception could possibly be applicable to permit the admission of the police testimony under the exception, because Ball did not identify the appellant in court. However, in my opinion, the requirements of this second branch are not satisfied in the circumstances. Ball did not testify that she could not remember whether the appellant was the person whom she identified. She was not asked to compare the appellant with her recollections about the person she saw on the night of the murders. Accordingly, the underlying circumstances of necessity required to trigger the second branch of the traditional exception did not exist. Even aside from this point, the police evidence went far beyond the scope of the “prior identification” exception. Part of the rationale underlying the second branch of the exception is that the testimony that is being admitted to complement the testimony of the identifying witness does not truly constitute hearsay. If the witness can at least testify that at some point she made an accurate identification, then a police officer’s testimony that he or she observed the identifying witness in the act of identification is original evidence that the identifying witness did indeed select a particular person, and that that person is the accused. However, for this rationale to apply, the identifying witness must confirm that the person he or she identified in the police officer’s presence was the person who committed an act that is relevant in the immediate proceedings. The testimony of the identifying witness may thus have its own hearsay component, but this issue is beyond the scope of this appeal: see H. Stewart, “Prior Identifications and Hearsay: A Note on R. v. Tat” (1998), 3 Can. Crim. L. Rev. 61. The point is that the officer’s testimony should merely state who the witness identified, once the witness has already testified to the identification itself, and why the identification is relevant to the case.

223The evidence provided by Constables Madden and MacLeod went beyond simply asserting that Cheryl Ball identified the accused on August 23, 1994. Their testimony provided almost the entirety of the narrative underlying the identification. Of particular importance, as already mentioned, Cheryl Ball did not testify that the person she pointed to in the photo line-up as being “familiar” was present at the Mohawk gas station or in the car that followed the Weselowski station wagon to St. Norbert. All of the links between the act of identification and the reasons for the identification were provided by Constables Madden and MacLeod. As Doherty J.A. emphasized in Tat, the evidence explaining why the identifying witness identified the accused must come from the identifying witness himself or herself. In his words, at p. 505: “Absent some evidence connecting the person identified at the line-up to the crime, I fail to see how identification at the line-up has any relevance to the issue of identity”. Similar comments were made by Twaddle J.A. below, who stated (at p. 175) that “[w]here the identifying witness is unable to recall the prior identification, the situation is the same as if he or she has not testified. In that scenario, the trier of fact is left with the bald assertion of the recipient as to the person identified without the ability to cross-examine the eyewitness on the actual identification”. It follows that the trial judge erred in admitting the police evidence regarding Ball’s out-of-court identification under a traditional exception to the hearsay rule.

224The next question, following Khan, supra, and Smith, supra, is whether the hearsay evidence of Constables Madden and MacLeod is nonetheless admissible under the principled approach to hearsay admissibility. Parenthetically, I should mention that since the “prior identification” exception is not directly brought into play on these facts, I leave to another day the question of whether the exception requires revisiting in order to conform to the principled approach.

225In my view, the police testimony was equally inadmissible under the principled approach. First of all, quite simply, the police hearsay was not necessary. Cheryl Ball was a witness at trial and could have provided first-hand evidence, had the Crown chosen to question her on point. That the Crown chose not to do so is a question of tactics, not a basis for admitting hearsay evidence to fill the gap. Moreover, there are strong indications that Cheryl Ball’s identification was unreliable, even if one accepts the evidence of Constables Madden and MacLeod that Ball told them she saw a man in a car talking to Cook at the Mohawk gas station. According to Madden’s testimony, Ball described the person she saw speaking to Cook as having had short hair in a ponytail, glasses, and no goatee. The black-and-white photocopied photograph that she pointed to as looking “familiar” showed the appellant with long hair and a goatee, with no ponytail and no glasses. Moreover, Madden testified that the appellant had never worn glasses. This inconsistency in the identification clearly does not meet the threshold of reliability. It is quite clear that the Constables’ testimony should not have been admitted at trial.

226Before this Court as in the court below, the Crown argued that, even if the trial judge erred in admitting hearsay evidence regarding Ball’s out-of-court identification, any prejudice that might otherwise have been suffered by the appellant as a result was cured by the trial judge’s cautionary instruction to the jury. The trial judge’s instruction on this point was as follows:

I think it is of some significance that Cheryl Ball was not asked if the reason she could identify the individual in photo number 5 of Exhibit 11 was because she had seen a photo, that is, photo number 3 in Exhibit 10, a week before. I think it is significant that photo number 3 in Exhibit 10 is the only one, the only photograph of the three people in Exhibit 10 repeated in Exhibit 11. It is significant, I think, that she was not told that the description she gave of the man she saw talking to Cook did not in fact fit the description of the man in photograph number 3 of Exhibit 10. It is significant, I think, as well that, in court, she did not mention Starr by name and that she was not asked to identify nor did she identify the accused, Starr, in the dock as being the man she saw at the Mohawk station. I would tell you, then, that the identification evidence of Mrs. Ball is, to say the least, extremely frail, and I would express the opinion that it does not link the accused, Starr, to the deaths of either Weselowski or Cook. However, it is for you to decide the value of her evidence.

227The majority in the Court of Appeal was of the view that the above instruction effectively negated any harm that the hearsay evidence of Constables Madden and MacLeod might have caused. In particular, the majority found that the evidence was relatively insignificant in the larger context of the case, because, in its words, “[t]his was not a case that turned on the issue of identification” (p. 159).

228With respect, I disagree with the Court of Appeal’s assessment of the harm that may have been caused by the admission of the out-of-court identification evidence. Constable Madden testified that Cheryl Ball had stated that the appellant “was the one that she saw at the Mohawk Station and that he was probably driving the other car” (emphasis added). It is true that Jodie Giesbrecht’s testimony had also put the appellant at the Mohawk station. However, according to Constable Madden, Cheryl Ball identified the appellant not only as being at the Mohawk station, but also as someone who “was probably driving the other car”. The use of a conjunctive “and” indicates that there were two distinct elements to this testimony — namely, being at the Mohawk station, and “probably” being in the other car. The jury could easily have taken the reference to the “other car” to refer not to the appellant’s presence at the Mohawk station, but to the second car seen in St. Norbert when the Balls last saw Cook and Weselowski alive.

229While admittedly the jury could also have taken the “other car” testimony to refer to the appellant’s presence at the Mohawk station, certainly both interpretations are possible. The former interpretation would have placed the appellant at the last place the deceased were seen alive and provided a crucial gap in the prosecution’s case. While Cheryl Ball clearly stated at trial that she did not see who was in the other car she saw at St. Norbert, the entire problem with the Constables’ testimony is that they went far beyond what Cheryl Ball herself said at trial. Having offered the evidence, the respondent is now hard pressed to argue the jury would have undoubtedly disregarded it. Thus, even if, as the trial judge cautioned, the identification evidence had little weight, there remained the problem of the serious prejudice that may have been caused to the appellant by its admission. A crucial aspect of the defence theory in this case is that some unknown other person may have been involved in the murders. By potentially putting the appellant with the victims in St. Norbert, the identification evidence could have gone a long way to assuaging the jury’s concerns about the non-existence of such other persons. The cautionary instruction was insufficient to remedy the harm caused by the admission of the evidence.

B. Adequacy of the Instruction to the Jury on Reasonable Doubt

(1) The Decisions in Lifchus and Bisson

230In its decisions in Lifchus, supra, and R. v. Bisson, [1998] 1 S.C.R. 306, this Court has recently addressed the appropriate manner of instructing a jury on the nature of the criminal standard of proof. Cory J., writing for the full Court in Lifchus, held that it is essential for jurors to be instructed that the term “beyond a reasonable doubt” has a special meaning in a criminal trial. He explained that an appropriate jury instruction on reasonable doubt is a fundamental component of a fair trial, because a jury might otherwise convict the innocent by finding guilt on the basis of mere probability, rather than on the basis of proof to a near certainty as is required in criminal proceedings. In his words, at para. 14:

No matter how exemplary the directions to the jury may be in every other respect if they are wanting in this aspect the trial must be lacking in fairness. It is true the term has come echoing down the centuries in words of deceptive simplicity. Yet jurors must appreciate their meaning and significance. They must be aware that the standard of proof is higher than the standard applied in civil actions of proof based upon a balance of probabilities yet less than proof to an absolute certainty. [Emphasis added.]

231In Lifchus, Cory J. held that there are certain essential elements that should be included in any effective jury charge on reasonable doubt, as well as several instructions that would generally lessen or impair the effectiveness of such a charge. Cory J. summarized the essential elements in the following terms, at para. 36:

It should be explained that:

• the standard of proof beyond a reasonable doubt is inextricably intertwined with that principle fundamental to all criminal trials, the presumption of innocence;

• the burden of proof rests on the prosecution throughout the trial and never shifts to the accused;

• a reasonable doubt is not a doubt based upon sympathy or prejudice;

• rather, it is based upon reason and common sense;

• it is logically connected to the evidence or absence of evidence;

• it does not involve proof to an absolute certainty; it is not proof beyond any doubt nor is it an imaginary or frivolous doubt; and

• more is required than proof that the accused is probably guilty -- a jury which concludes only that the accused is probably guilty must acquit. [Emphasis in original.]

232Cory J. also summarized, at para. 37, certain references to the required standard of proof that should be avoided:

• describing the term “reasonable doubt” as an ordinary expression which has no special meaning in the criminal law context;

• inviting jurors to apply to the task before them the same standard of proof that they apply to important, or even the most important, decisions in their own lives;

• equating proof “beyond a reasonable doubt” to proof “to a moral certainty”;

• qualifying the word “doubt” with adjectives other than “reasonable”, such as “serious”, “substantial” or “haunting”, which may mislead the jury; and

• instructing jurors that they may convict if they are “sure” that the accused is guilty, before providing them with a proper definition as to the meaning of the words “beyond a reasonable doubt”.

233Cory J. stressed that the precise wording of any given jury charge on reasonable doubt will not necessarily be determinative of its effectiveness at transmitting the meaning of the term, and so at securing the fairness of the accused’s trial. A charge must be examined in its entirety to determine whether the essential elements of a fair and accurate instruction on reasonable doubt are present and have been properly explained. The question in every case in which a trial judge’s instructions on reasonable doubt are impugned is whether there is a reasonable likelihood that the jury was under a misapprehension as to the correct standard of proof to apply. If the charge, when read as a whole, could not have placed the jury under a misapprehension as to the correct standard of proof, then the jury verdict should not be disturbed: R. v. W. (D.), [1991] 1 S.C.R. 742, at p. 758. However, in Cory J.’s words in Lifchus, at para. 41: “if the charge as a whole gives rise to the reasonable likelihood that the jury misapprehended the standard of proof, then as a general rule the verdict will have to be set aside and a new trial directed”.

234In both Lifchus, supra, and Bisson, supra, this Court held that the reasonable doubt instruction provided to the jury in the particular cases created a reasonable likelihood of a misapprehension of the standard of proof. In Lifchus, the trial judge instructed the jury on reasonable doubt by referring to the ordinary everyday meaning of the phrase.

235In finding that the trial judge’s instruction constituted an error in law justifying a new trial, Cory J. explained that the instruction contained two central flaws. First, it did not provide a definition of “reasonable doubt”. Second, it described the words “reasonable doubt” as “ordinary, every day words” with no specific meaning in the context of a criminal trial. Cory J. found that these errors were not cured when the jury charge was viewed as a whole, even though the charge was, in all other respects, “a model of clarity and conciseness” (para. 44).

236The trial judge in Bisson, supra, explained to the jury that proof to an absolute certainty was not required, and that a reasonable doubt was something more than simply a frivolous, capricious, or imaginary doubt. He then drew an analogy between the standard of proof beyond a reasonable doubt and the standard of proof used in everyday activities such as checking the oil in one’s car, emphasizing at some length that there was no distinction between the two standards. Writing for the Court, Cory J. explained that a trial judge should avoid providing examples from daily life of what may constitute a reasonable doubt, for two key reasons. First, everyday examples are problematic because, almost invariably, they involve the application of a standard of probability rather than one of proof beyond a reasonable doubt. Second, examples tend to be applied subjectively, depending upon the degree of care taken by different jurors prior to making everyday decisions, whereas the reasonable doubt standard is a single, objective, and exacting standard of proof.

237Before turning to examine the charge to the jury in the present appeal, I would like to emphasize that in Lifchus and Bisson, we set out to improve the existing standard for instructions on reasonable doubt. This was not intended to suggest that a new trial is warranted for all previous convictions obtained following jury charges that were not in strict compliance with every aspect of Lifchus and Bisson. To the contrary, as Cory J. made clear in Lifchus, supra, at para. 40, there is no “magic incantation”. A court reviewing a pre-Lifchus jury charge must examine it to make sure that it was in substantial compliance with the principles set out in that case.

(2) Inadequacy of the Reasonable Doubt Charge in This Case

238It should be pointed out that the trial judge in the present appeal, like the trial judge in Bisson, supra, gave his charge to the jury without the benefit of the decision of this Court in Lifchus. However, I agree with Twaddle J.A. that the reasonable doubt instruction given in this case falls prey to many of the same difficulties outlined in Lifchus, and likely misled the jury as to the content of the criminal standard of proof. For ease of reference, I will set out again the relevant portion of the charge in this case:

It is rarely possible to prove anything with absolute certainty and so the burden of proof on the Crown is only to prove the guilt of the accused beyond reasonable doubt. What, then, is proof beyond a reasonable doubt?

The words “reasonable doubt” are used in their everyday, ordinary sense and not as a legal term having some special connotation. The words have no magic meaning that is peculiar to the law. A reasonable doubt is an honest, fair doubt, based upon reason and common sense. It is a real doubt, not an imaginary or frivolous one resting on speculation or guess rather than upon the evidence you heard in this courtroom.

So you can see, the words “reasonable doubt” are ordinary words we use in our everyday language. So if you can say, I am satisfied beyond a reasonable doubt, the Crown has met the onus upon it. If you cannot say those words -- if you cannot say, I am satisfied beyond a reasonable doubt, the Crown has not met the onus on it, and the accused is entitled to have your doubt resolved in his favour.

239The key difficulty with this instruction is that it was not made clear to the jury that the Crown was required to do more than prove the appellant’s guilt on a balance of probabilities. The trial judge told the jury that they could convict on the basis of something less than absolute certainty of guilt, but did not explain, in essence, how much less. In addition, rather than telling the jury that the words “reasonable doubt” have a specific meaning in the legal context, the trial judge expressly instructed the jury that the words have no “special connotation” and “no magic meaning that is peculiar to the law”. By asserting that absolute certainty was not required, and then linking the standard of proof to the “ordinary everyday” meaning of the words “reasonable doubt”, the trial judge could easily have been understood by the jury as asserting a probability standard as the applicable standard of proof.

240The trial judge did comply with some of the requirements discussed in Lifchus, supra. He explained that a reasonable doubt is not an imaginary or frivolous doubt resting on speculation or guess, and that a reasonable doubt is a real doubt based on reason and common sense upon a review of the evidence. He also explained, as mentioned, that the standard of proof beyond a reasonable doubt does not involve proof to an absolute certainty. However, the trial judge’s adherence to these requirements would have benefited primarily the Crown, not the appellant.

241In the present case, the trial judge did refer to the Crown’s onus and to the presumption of innocence, and he stated that the appellant should receive the benefit of any reasonable doubt. The error in the charge is that the jury was not told how a reasonable doubt is to be defined. As was emphasized repeatedly in Lifchus and again in Bisson, a jury must be instructed that the standard of proof in a criminal trial is higher than the probability standard used in making everyday decisions and in civil trials. Indeed, it is this very requirement to go beyond probability that meshes the standard of proof in criminal cases with the presumption of innocence and the Crown’s onus. However, as Cory J. explained in these earlier decisions, it is generally inappropriate to define the meaning of the term “reasonable doubt” through examples from daily life, through the use of synonyms, or through analogy to moral choices. The criminal standard of proof has a special significance unique to the legal process. It is an exacting standard of proof rarely encountered in everyday life, and there is no universally intelligible illustration of the concept, such as the scales of justice with respect to the balance of probabilities standard. Unlike absolute certainty or the balance of probabilities, reasonable doubt is not an easily quantifiable standard. It cannot be measured or described by analogy. It must be explained. However, precisely because it is not quantifiable, it is difficult to explain.

242In my view, an effective way to define the reasonable doubt standard for a jury is to explain that it falls much closer to absolute certainty than to proof on a balance of probabilities. As stated in Lifchus, a trial judge is required to explain that something less than absolute certainty is required, and that something more than probable guilt is required, in order for the jury to convict. Both of these alternative standards are fairly and easily comprehensible. It will be of great assistance for a jury if the trial judge situates the reasonable doubt standard appropriately between these two standards. The additional instructions to the jury set out in Lifchus as to the meaning and appropriate manner of determining the existence of a reasonable doubt serve to define the space between absolute certainty and proof beyond a reasonable doubt. In this regard, I am in agreement with Twaddle J.A. in the court below, when he said, at p. 177:

If standards of proof were marked on a measure, proof “beyond reasonable doubt” would lie much closer to “absolute certainty” than to “a balance of probabilities”. Just as a judge has a duty to instruct the jury that absolute certainty is not required, he or she has a duty, in my view, to instruct the jury that the criminal standard is more than a probability. The words he or she uses to convey this idea are of no significance, but the idea itself must be conveyed....

243In the appellant’s case, with respect, the trial judge did not give instructions that could be construed as having located the reasonable doubt standard above the probability standard. Not only was the jury not told that something more than probability was required in order to convict, but nearly all of the instructions they were given (i.e., less than absolute certainty required, ordinary everyday words, no special meaning, more than a frivolous doubt required) weakened the content of the reasonable doubt standard in such a manner as to suggest that probability was indeed the requisite standard of proof. In this regard, the reasonable doubt instruction in the appellant’s case cannot be meaningfully distinguished from the impugned instructions in Lifchus and Bisson. The reasonable likelihood that the jury applied the wrong standard of proof raises a realistic possibility that the appellant’s convictions constitute a miscarriage of justice. While obviously a mistake in the charge will not always be fatal, at no point did the instructions in this case cure the mistake. The fact that the trial judge repeatedly stated that the prosecution must prove guilt beyond a reasonable doubt is no cure given his failure to ever define reasonable doubt correctly. To hold otherwise would eviscerate the holdings of Lifchus and Bisson. I reach this finding notwithstanding the fact that, like the jury charge in Lifchus when read as a whole, the jury charge in the present case was largely a model of fairness to the accused. The fact that other elements of the charge were fair to the accused cannot eliminate the prejudice caused by the improper instructions. Thus I conclude that there was not substantial compliance with the Lifchus principles. I would therefore allow the appeal on this ground as well.

244I note that, appropriately, the respondent has not seriously attempted to apply the curative proviso contained in s. 686(1)(b)(iii) of the Criminal Code.

VI. Disposition

245The appeal is allowed, the judgment of the Court of Appeal is set aside, and a new trial is directed.

Appeal allowed and new trial ordered, McLachlin C.J. and L’Heureux‑Dubé, Gonthier and Bastarache JJ. dissenting.

Solicitors for the appellant: Walsh, Micay & Company, Winnipeg.

Solicitor for the respondent: The Attorney General of Manitoba, Winnipeg.

Solicitor for the intervener the Attorney General of Canada: Bernard Laprade, Ottawa.

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

Solicitor for the intervener the Attorney General of British Columbia: The Ministry of the Attorney General, Vancouver.

R. v. Russell, 2000 SCC 55, R. v. Beauchamp, 2000 SCC 54, and R. v. Avetysan, 2000 SCC 56, were released November 10, 2000.