R. v. MacKenzie, 1993 CanLII 149 (SCC), [1993] 1 SCR 212

R. v. MacKenzie, [1993] 1 S.C.R. 212

John Alexander MacKenzie Appellant

v.

Her Majesty The Queen Respondent

Indexed as: R. v. MacKenzie

File No.: 22423.

1992: April 3; 1993: January 21.

Present: Lamer C.J. and La Forest, Gonthier, McLachlin and Stevenson* JJ.

on appeal from the nova scotia supreme court, appeal division

Criminal law ‑‑ Evidence ‑‑ Charge to jury ‑‑ Accused making incriminating statement to police ‑‑ Accused stating at trial that his "confession" was in fact a recounting of second‑hand knowledge ‑‑ Trial judge instructing jury to reject accused's statement to police if they had any reasonable doubt about it ‑‑ Whether trial judge's charge misleading.

Criminal law -- Appeals to Supreme Court of Canada -- Crown appeals -- Accused abandoning one of his grounds of appeal -- Court refusing to hear Crown's arguments on point since Crown had not cross-appealed -- Crown precluded from cross-appealing in view of its overall success in court below -- Criminal Code, R.S.C., 1985, c. C-46, s. 693(1).

Criminal law -- Appeals -- Appeal from acquittal -- Standard of appellate review -- Court of Appeal not misstating standard by requiring only that Crown show that jury's verdict might have been different had it been properly instructed.

The accused was arrested and charged with murdering three of his neighbours, who had been killed with shots from his revolver. He volunteered an incriminating statement about the shootings and later, after he had requested legal counsel, started in on a more detailed and explicit statement. This second statement was excluded by the trial judge because it was made after the accused had exercised his right to counsel. At trial, the accused said his "confession" to the police was in fact a recounting of his second‑hand knowledge of the shootings gained from other neighbours during the day of drinking following the shootings. The trial judge instructed the jury on how to evaluate the contradiction between the accused's statement to the police and his testimony at trial. As part of this instruction he told them to "reject entirely" the accused's statement to the police if they had any reasonable doubt about it. He also said that if the jury accepted the accused's statement, then "it will be considered by you with the other evidence that you decide to accept". The jury found the accused not guilty on all counts. The Court of Appeal, in a majority decision, overturned the acquittal and ordered a new trial. The majority found that there had been a significant misdirection by the trial judge, in that part of the charge may have suggested to the jury that they should consider the accused's confession separate and apart from the other evidence.

The accused's appeal to this Court raised the issues of (1) whether the trial judge erred in instructing the jury that the accused's out-of-court inculpatory statement must be rejected entirely if his testimony at trial raised any reasonable doubt about it; (2) whether the trial judge erred in excluding from evidence the accused's second statement to the police; and (3) whether the majority of the Court of Appeal erred in holding that the verdict might have been different, thereby usurping the jury's function. At the hearing the accused abandoned the second issue concerning the exclusion of his second statement to the police, and the Crown then urged the Court to adopt a position in favour of a new trial because of the excluded statement, but the Court refused to hear its argument on this point.

Held: The appeal should be allowed and the acquittal restored.

Per La Forest, Gonthier and McLachlin JJ.: On the first issue of the jury instruction, the trial judge's charge, when viewed as a whole, was not misleading to the jury. In R. v. Morin, this Court determined that a jury charge is in error if it leaves the jury with the impression that each item of evidence is to be considered piecemeal against the criminal standard of proof or if the jury is instructed to take a two-stage approach to their deliberations, whereby an initial fact-finding stage would weed out certain items of evidence, leaving the determination of guilt or innocence to be based only on the surviving evidence. In the case at bar, there was no general instruction to apply the criminal standard of proof to items of evidence in isolation. While the charge does focus on one item of evidence, namely the accused's statement to the police, the direction on this evidence is carefully couched in terms of all the other evidence in the case. The charge does not instruct the jury to divide its deliberations into two phases, and there was almost no chance for the jury to form the impression that "the whole of the evidence" was intended to mean only the whole of the piecemeal bits of evidence already proven beyond a reasonable doubt.

The contradiction between the accused's out‑of‑court statement and his testimony at trial was a key issue in the case, and the trial judge was entitled to give the jury some guidance on how to handle this discrepancy in the evidence. As a matter of logic, the two stories could not be reconciled, and the judge's suggestion that one of the statements must be "rejected", couched as it was with the proviso that all of the other evidence must be taken into account, in no way prejudiced the Crown. The jury instruction was an accurate reflection of the principle enunciated by this Court in Nadeau that an accused's version of events is entitled to the benefit of the doubt as against a competing version, providing the comparison is made having due regard to all the evidence. It is both acceptable and desirable for a trial judge to focus the jury's attention on vital issues, and to direct their minds to the proper burden of proof on those discrete questions.

While the statement "it will be considered by you with the other evidence that you decide to accept", read in isolation, may be viewed as an erroneous statement of the law, since it suggests that a determination of guilt should be based solely on certain pre‑selected items of evidence, read in context it would not have misdirected the jury. This statement follows the trial judge's direction to consider all the evidence in making a determination, and therefore could not be interpreted as invoking a two‑stage process. Consequently, the instruction to the jury is in keeping with the rule in Morin and was not in error.

On the second issue, which was abandoned by the accused, the Crown had not cross-appealed nor was it entitled to do so in view of its overall success in the court below. Under s. 693(1) of the Criminal Code, the Crown may appeal to this Court only where the Court of Appeal has set aside a conviction or dismissed a Crown appeal. In this case the Court of Appeal allowed the Crown's appeal, albeit on a different issue.

On the third issue, the Court of Appeal did not misstate the standard of appellate review by requiring only that the Crown show that the jury's verdict might have been different had it been properly instructed. A test of "might have been different" is simply the converse of "would not necessarily have been the same", and is thus consistent with the test established by the majority of this Court in Vézeau, which is the governing law today.

Per Lamer C.J.: La Forest J.'s reasons were substantially agreed with. While it is a misdirection to instruct juries to "reject" evidence, to tell juries to reject factual propositions which the Crown's evidence does not establish beyond a reasonable doubt is to state the law correctly. The jury never rejects evidence, but it can and must decide whether to accept or reject the factual assertions made by that evidence before it uses them to support or infer other factual assertions in reaching its verdict. Such factual assertions can only be accepted and used by the jury to convict the accused if they are established by the evidence beyond a reasonable doubt. Facts which are not so established cannot corroborate or be allowed to "bootstrap" other doubtful facts. In the circumstances of this case, the trial judge was not in error in inviting the jury to "reject" the factual proposition in one of the two statements, depending on which they believed, because both statements emanated from the accused and could not logically co‑exist. There was only a single determination of credibility to be made. If the jury believed the accused's exculpatory evidence at trial, they must of necessity have disbelieved his inculpatory statements to the police. The accused need never establish his version of events beyond a reasonable doubt, but must only raise a reasonable doubt with his evidence, even where he bears an evidentiary burden.

Cases Cited

By La Forest J.

Considered: R. v. Morin, [1988] 2 S.C.R. 345; R. v. Minhas (1986), 29 C.C.C. (3d) 193 (Ont. C.A.), leave to appeal refused, [1987] 2 S.C.R. viii; Vézeau v. The Queen, [1977] 2 S.C.R. 277; referred to: Nadeau v. The Queen, [1984] 2 S.C.R. 570; R. v. Barnes, [1991] 1 S.C.R. 449; Guillemette v. The Queen, [1986] 1 S.C.R. 356; R. v. Challice (1979), 45 C.C.C. (2d) 546; R. v. Thatcher, [1987] 1 S.C.R. 652; R. v. Gauthier, [1977] 1 S.C.R. 441; R. v. Demeter (1975), 25 C.C.C. (2d) 417.

By Lamer C.J.

Referred to: R. v. Morin, [1988] 2 S.C.R. 345; Nadeau v. The Queen, [1984] 2 S.C.R. 570; R. v. Proudlock, [1979] 1 S.C.R. 525.

Statutes and Regulations Cited

Criminal Code, R.S.C., 1985, c. C‑46, ss. 686(4), 693 [am. c. 27 (1st Supp.), s. 146; am. c. 34 (3rd Supp.), s. 12].

Authors Cited

Eggleston, Sir Richard. Evidence, Proof and Probability, 2nd ed. London: Weidenfeld and Nicolson, 1983.

APPEAL from a judgment of the Nova Scotia Supreme Court, Appeal Division (1991), 103 N.S.R. (2d) 91, 282 A.P.R. 91, 64 C.C.C. (3d) 336, allowing the Crown's appeal from the accused's acquittal on charges of first degree murder. Appeal allowed.

Joel E. Pink, Q.C., and Donald C. Murray, for the appellant.

Robert C. Hagell and Denise Smith, for the respondent.

//Lamer C.J.//

The following are the reasons delivered by

Lamer C.J. -- I have had the benefit of reading the reasons of Justice La Forest in this appeal, and I agree substantially with his analysis of the issues before the Court and his disposition of the appeal. However, I feel compelled to add a few comments which follow from my concurrence in Wilson J.'s reasons in R. v. Morin, [1988] 2 S.C.R. 345. In retrospect, it appears to me that confusion exists in two aspects of this area of the law: (i) the essential distinction between facts and evidence of facts in the process of the jury's deliberations, and (ii) the relationship between credibility and the standard of proof in a criminal trial.

The Distinction Between Facts and Evidence

In my view, it is crucial to recognize the difference between the roles of "facts" and "evidence of facts" in a criminal trial; the failure to apply this basic distinction has confused the debate on this issue. As an example of this tendency, see R. Eggleston, Evidence, Proof and Probability (2nd ed. 1983), who at p. 122 begins talking of "facts that are not certainly established" but concludes with comments directed to the effect of "evidence" not being taken into account, and stating that a requirement to prove evidence beyond a reasonable doubt obviates the entire field of corroboration. As I state below, doubtful evidence may be corroborated by other evidence, even other doubtful evidence; doubtful facts, however, cannot be corroborated by, or inferred from, other doubtful facts. Throughout many discussions of this point, the terms "fact" and "evidence" are used interchangeably, and thereby improperly. It is beyond question that the jury must always consider the whole of the evidence in reaching their verdict, and that individual items of evidence are not to be examined in isolation. Evidence which may be doubtful standing alone may nevertheless be corroborated or otherwise supported by other evidence, and the jury is always to evaluate the evidence as a whole with a view to such corroboration or support. Therefore, it is an error to refer to the jury "rejecting" evidence at any stage of their deliberations. All of the evidence must always be considered.

However, the jury can and must decide whether the whole of evidence establishes, beyond a reasonable doubt, the individual facts necessary to support a conviction. This must, of necessity, be done in a sequential manner, and I agree with La Forest J. that it is unrealistic to expect the jury to have an "epiphanic" experience in reaching its verdict. In Morin, Wilson J. stated (at p. 378) that the process of the jury's deliberations with respect to the Crown evidence "requires a fact elicited through the mouth of a witness to be assessed by the jury in the context of all the evidence and to be rejected if it has not been proved beyond a reasonable doubt." Therefore, while the jury never rejects evidence, it can and must decide whether to accept or reject the factual assertions made by that evidence before it uses those factual assertions to support or infer other factual assertions towards reaching its verdict. Such factual assertions can only be accepted and used by the jury to convict the accused if they are established by the evidence beyond a reasonable doubt. Facts which are not so established cannot corroborate or be allowed to "bootstrap" other doubtful facts. Any lower standard would present the possibility that an accused could be convicted on the basis of facts which are established as matters of conjecture only.

An analogy I often used in charging juries, especially in cases where the Crown's case was circumstantial, was that of a fisherman's net. The evidence presented at trial by the Crown seeks to establish factual propositions. Once established, facts may be used to infer other facts. In this way, established factual propositions intertwine to construct a net of such propositions. If a factual proposition is established as a mere probability or likelihood, and not beyond a reasonable doubt, it cannot be used to infer any further facts. The interweaving of facts breaks down and there is a hole in the net. A net with a hole, however small, is no useful net at all, since there remains a critical factual proposition which is not consistent only with the accused's guilt. Thus, a fact which is not established beyond a reasonable doubt can play no part in the jury's decision to convict, either as a fact on which they rely to find an essential element of the offence, or as a fact used to infer such facts.

Therefore, while it is a misdirection to instruct juries to "reject" evidence, to tell juries to reject factual propositions which the Crown's evidence does not establish beyond a reasonable doubt is to state the law correctly.

The Relationship Between Credibility and the Standard of Proof

It is also important to understand the relationship between the jury's assessment of credibility and the criminal standard of proof when the jury is faced with conflicting accounts from Crown and defence witnesses. It is a clear error for a trial judge to instruct the jury to choose the more believable, persuasive or credible of the two accounts in such circumstances. As with all facts on which a conviction is to be supported, the factual propositions in the Crown witnesses' evidence must be established by that and other evidence beyond a reasonable doubt. If the jury does not believe the Crown witnesses' version, or any part of it, beyond a reasonable doubt, it cannot be used to convict the accused, even if they do not believe the contrary version of the defence. It may very well be that both versions are wrong, and the accused is entitled to the benefit of this doubt with respect to the credibility of any witness. This was my holding in Nadeau v. The Queen, [1984] 2 S.C.R. 570, in the passage cited by La Forest J. at pp. 27-28 of his reasons.

In the circumstances of this case, however, the trial judge was not in error in inviting the jury to "reject" the factual proposition in one of the two statements, depending on which they believed, because both statements emanated from the accused, and, as La Forest J. states, could not logically co-exist. There was only a single determination of credibility to be made. If they believed the accused's exculpatory evidence at trial, they must of necessity have disbelieved his inculpatory statements to the police. It must be remembered, though, that the accused need never establish his or her version of events beyond a reasonable doubt; that standard only and always applies to the Crown. The accused must only raise a reasonable doubt with his or her evidence, even where he or she bears an evidentiary burden; see R. v. Proudlock, [1979] 1 S.C.R. 525.

With these observations, I would dispose of the appeal in the manner proposed by La Forest J.

//Le juge La Forest//

The judgment of La Forest, Gonthier and McLachlin JJ. was delivered by

La Forest J. -- The appellant, John Alexander MacKenzie, was accused of murdering three of his neighbours in the early hours of June 24, 1989. At trial, a jury found him not guilty. The acquittal was overturned on appeal, and a new trial was ordered. The appellant appealed to this Court as of right, seeking the restoration of his acquittal.

At the time of his trial, the appellant was 47 years old. He was an alcoholic. In 1985, after living for 25 years in Toronto, he moved back to his home community in rural Antigonish County, Nova Scotia. He purchased a plot of land in Beech Hill, and in 1988 moved a mobile home onto the lot and made it his home. The three victims of the shooting, John Boucher, Joseph Deon and Edmund Deon, were neighbours of the appellant. Each owned a home on the Beech Hill Road within a kilometre of the appellant's property. The Deons were brothers, and were good friends of Boucher.

For some reason, the Deons and Boucher apparently engaged in a concerted campaign of harassment against the appellant that lasted from the time he moved to Beech Hill until their deaths. This harassment verged on terrorism, consisting of repeated property damage, the killing of the appellant's pets, shooting bullet holes through his mailbox, and a series of threats to shoot him or dynamite his trailer.

In the early morning of Saturday, June 24, 1989 the Deons and Boucher were shot to death. Boucher was shot three times in his yard, Edmund Deon was shot seven times in his bedroom, and Joseph Deon was shot twice while sitting at the wheel of his pick-up truck, which was parked in his driveway. All the shots came from the appellant's .22 calibre revolver. The revolver was retrieved by police the next day from the appellant's well, along with a pair of his shoes.

During the six weeks before the shootings the appellant was drinking heavily. On June 12, two weeks before the shootings, he discovered that a calf of his was missing. Suspecting his neighbours, he took his revolver and confronted Boucher and his wife Lorraine at their home. They were able to calm the appellant down, and she drove him home. At trial, the appellant testified that Lorraine Boucher took his revolver from him at that time, and that this was the last he saw it. Mrs. Boucher did not testify on this point. Later that evening, the appellant decided to confront Edmund Deon about the calf. He could not find his revolver, so he took a .303 rifle with him. During the confrontation that ensued the appellant discharged the rifle into the floor of Deon's house. The next day the RCMP arrested the appellant, detained him overnight, and confiscated his rifle.

On June 23 the appellant spent most of the day drinking at his mobile home with his brother-in-law, William Cogger. The appellant consumed about 21 beers during the day. Boucher visited them for about an hour in the evening, drank at least one beer, and left without any dispute or incident. At around 11:00 p.m. the appellant and Cogger went to sleep. Sometime during the night, they were awakened by a knock at the door. Cogger testified that he heard the appellant say "who's there?", that a voice responded, and that the appellant said "I'm sleeping. Go home. I'm going back to sleep". Cogger heard nothing more, and fell back asleep. The appellant testified that the visitor was Joseph Deon.

At 4:17 a.m. Daniel Girrior, another neighbour on the Beech Hill Road, was awakened by three gunshots. It was not unusual to hear gunshots at night in that area, so Girrior went back to sleep. William Cogger awoke at 4:45 a.m. to find the appellant sitting fully dressed at a table. Cogger testified that the appellant said he got the "three sons-of-whores last night." Cogger said "tell me this is not the truth", to which the appellant replied "no brother it's not true." The appellant then said he was going to put his gun down the well, and left the house. On returning, he spoke about getting rid of his shoes. At 4:58 a.m. Daniel Girrior heard three shots. Two minutes later he heard three more shots.

At trial, the appellant testified that he fell back to sleep after Boucher's early morning visit. The next thing he remembered was waking up on the floor, with the idea that he had seen the bodies of the Deons and Boucher on the Beech Hill Road. He felt he must have been dreaming, and told Cogger about the dream. He was surprised to see his revolver on the table, as he had not seen the gun since Lorraine Boucher took it from him two weeks earlier. The appellant had no recollection of shooting the Deons and Boucher, or of throwing the revolver and the shoes down the well.

The appellant and Cogger spent the rest of June 24 drinking. At about 9:00 p.m. the appellant was arrested by the RCMP. Realizing that he was drunk, the police did not attempt to question him. He was given food and at 11:30 p.m. was put in jail for the night.

The next morning the appellant was given breakfast, and then brought before two officers for questioning. He was advised of his right to silence and to counsel, and the officers then told him they had reason to believe he had put his gun and shoes down his well, and that he had previously been seen at the Boucher house. They asked the appellant for his side of the story, and he volunteered an incriminating statement about the shootings. One of the officers took notes for about ten minutes, and then suggested that the statement should be in a formal written form. At this point the appellant requested legal counsel, and the officers told the appellant that there would be no more questions. The officers left the appellant alone to call his sister so she could retain a lawyer. When they returned a pause of several minutes ensued when nothing was said. The accused then started in on a more detailed and explicit statement. This second statement was excluded by the trial judge because it was made after the appellant had exercised his right to counsel.

At trial, the appellant said his "confession" to the police was in fact a recounting of his second-hand knowledge of the shootings gained from other neighbours during the day of drinking following the shootings. He remembers little of his statements to the police, other than that he was suffering at the time from "the worst hangover of my life".

During a four-hour jury charge, MacIntosh J. instructed the jury on the law regarding the defences of provocation and drunkenness, and that there were four possible verdicts on each murder count: not guilty, guilty of first degree murder, second degree murder, or manslaughter. He also instructed the jury on how to evaluate the contradiction between the appellant's statements to the police and his testimony at trial. Part of this instruction was as follows:

Here we recall that the accused is saying he doesn't remember anything of what happened, that these statements he gave is what he had heard from other people. If you decide to accept part or all of it, it will be considered by you with the other evidence that you decide to accept. You, of course, must reach your verdict on the whole of the evidence that you decide is worthy of belief. The accused, as a witness, you can accept his earlier testimony as the statement that the police -- the Crown alleges he made to the police. You can accept that as the truth of what happened ("He" being the accused person) as opposed to what he said in court. This we cannot do with an ordinary witness but with an accused, you are free to accept either, what he said in the statement or what he said in court. If you find that his evidence at trial represented the true facts, or if you have any reasonable doubt about it, you will reject entirely the earlier statement. [Emphasis added.]

A majority of the Appeal Division of the Nova Scotia Supreme Court found the underlined passage a significant misdirection, in that it may have suggested to the jury that they should consider the appellant's confession separate and apart from the other evidence.

After two hours of deliberation, the jury asked to have the evidence of William Cogger, Lorraine Boucher and Daniel Girrior read back. After this, the jury deliberated for one and a half days, and then returned a verdict of not guilty on all counts.

Judgments

The Crown appealed the acquittal on a total of eleven grounds. Only two of these were seriously considered by the Appeal Division: that MacIntosh J. erred in (1) excluding the appellant's second out-of-court statement to the police, and (2) instructing the jury to reject entirely the accused's first out-of-court statement to the police if his testimony at trial raised any reasonable doubt about it. The appeal was allowed on the second ground; see (1991), 103 N.S.R. (2d) 91. Chipman and Hart JJ.A. wrote separate judgments for the majority, Freeman J.A. dissented.

On the first question, Chipman J.A. reviewed the governing case law on an accused's rights to counsel and silence, and concluded that the police did nothing wrong in this instance. He reasoned that the duty of the police under the Canadian Charter of Rights and Freedoms does not go so far as to require them to cease listening to unelicited statements that a detainee may make after requesting counsel. As such, Chipman J.A. found that the trial judge had erred in excluding the statement. However, he also found that the admission of the statement would not necessarily have affected the jury's verdict, because the crucial first statement had been put before the jury. Hart J.A. agreed that the trial judge had erred in excluding the second statement, but in his view the exclusion might have materially affected the jury's deliberations. He would have ordered a new trial on this ground. Freeman J.A. differed from his colleagues, concluding that the trial judge was correct in excluding the statement. In the result, the Crown's appeal on this first ground was dismissed.

On the second question, Chipman J.A. reviewed the jury charge, referred to R. v. Morin, [1988] 2 S.C.R. 345; R. v. Minhas (1986), 29 C.C.C. (3d) 193 (Ont. C.A.), leave to appeal refused, [1987] 2 S.C.R. viii, and Nadeau v. The Queen, [1984] 2 S.C.R. 570, and concluded as follows, at pp. 109-10:

On reading the trial judge's charge in its entirety, I have the same concern as did Sopinka, J., in Morin, that the instruction to regard the evidence as a whole was not sufficient to correct the impression they might well have got, that they were to examine the Crown's evidence about the confession in isolation without taking into account the evidence as a whole in determining if it established guilt beyond a reasonable doubt. It seems to me that basically the same error was made here as was made in Morin.

Moreover, this error was of sufficient gravity that it cannot be said that had it not been made the result would necessarily have been the same. Sopinka, J., said in Morin, supra, the charge with respect to the burden of proof lays down for the jury one of the most fundamental rules of the game. The confession was undoubtedly a crucial part of the Crown's case. Without it, the jury could easily have formed a reasonable doubt. With it, it is difficult to see how a reasonable doubt as to some degree of guilt at least could be entertained. It can therefore be said with a reasonable degree of certainty that the outcome may have been affected by the misdirection relating to the confession. The appeal should be allowed and a new trial ordered.

Hart J.A. shared Chipman J.A.'s view that the instruction was incorrect, finding that it permitted the jury to apply the doctrine of reasonable doubt to parts of the evidence rather than to the evidence as a whole, and that the overall instruction did not remedy this defect.

In dissent, Freeman J.A. conceded that the language used in the trial judge's charge to the jury raised the possibility of a misdirection, but he did not agree that an error warranting a new trial had been made. He reasoned as follows, at pp. 133-35:

At its worst, the part of the charge in question, balanced as it is by instructions that the whole of the evidence must be considered, is not so clear as to cause the jury to override their common sense and prematurely exclude a piece of evidence as important as the statement. Once accepted as possibly true, common sense would dictate that the statement remain part of their deliberations to be rejected as a true description of events only when the whole of the evidence including the statement failed to persuade them beyond a reasonable doubt of the respondent's guilt.

If one were to assume that as a result of Mr. Justice MacIntosh's charge, the jury thought their duty was to apply the standard of proof beyond a reasonable doubt to the statement as a piece of evidence in isolation from the other evidence, how might they have proceeded? They could only reject the statement in isolation if it raised a reasonable doubt within itself, if it contained contradictions or inconsistencies that might realize a reasonable doubt. No such internal flaws are apparent. Therefore it had to be considered in the light of other evidence.

In considering the truth of [the appellant's] evidence about the statement, the jury had been instructed to take into account corroboration, circumstances and the condition of the accused. That is to say, the jury was instructed to consider the statement with the whole of the evidence. The charge would be objectionable only if it could be seen as directing the jury to reject the statement before it had done so. It is difficult to see how the jury might have considered the statement except in the context of all of the other evidence. It did not lend itself to consideration in either total or partial isolation.

Accordingly, Freeman, J.A. would have dismissed this ground of appeal.

Points in Issue

1. Whether the trial judge erred in law in instructing the jury that the appellant's out-of-court inculpatory statement must be rejected entirely if his testimony at trial raised any reasonable doubt about it.
2. Whether the trial judge erred in law in excluding from evidence certain statements made by the appellant to police officers.
3. Whether the majority of the Court of Appeal erred in law in holding that the verdict might have been different, thereby usurping the function of the jury.

Analysis

All three of the points in issue were argued in written submissions to this Court, but at the commencement of the hearing appellant's counsel abandoned the second issue concerning the exclusion of the appellant's second statement to the police. This abandonment created certain procedural difficulties for the respondent, which I will comment on briefly before turning to the remaining issues, which concern the charge to the jury.

The Excluded Statement

In the court below, the Crown won a partial victory on the excluded confession: a majority found that the trial judge had erred in excluding it, but only Hart J.A. would have ordered a new trial because of this error. The appellant appealed only the first point of law to this Court, arguing that the trial judge had not erred. I presume the reason for appealing on this point was to foreclose the admission of the statement at a new trial, in the event that this Court upholds the decision of the court below that such a re-trial is necessary. In any event, as I have noted, the appellant eventually abandoned this ground of appeal. As the hearing of this appeal progressed, the Crown urged this Court to adopt Hart J.A.'s minority position in favour of a new trial because of the excluded statement. The Chief Justice pointed out that the Crown had not cross-appealed on this point, and indeed was not entitled to cross-appeal. Accordingly, the Court refused to hear the Crown's arguments on this issue.

Our refusal to hear the Crown on this point was based on the limited scope for Crown appeals to this Court, as determined by s. 693(1) of the Criminal Code, R.S.C., 1985, c. C-46:

693. (1) Where a judgment of a court of appeal sets aside a conviction pursuant to an appeal taken under section 675 or dismisses an appeal taken pursuant to paragraph 676(1)(a), (b) or (c) or subsection 676(3), the Attorney General may appeal to the Supreme Court of Canada

(a) on any question of law on which a judge of the court of appeal dissents; or

(b) on any question of law, if leave to appeal is granted by the Supreme Court of Canada. [Emphasis added.]

The problem for the Crown in this case is that the Court of Appeal allowed the Crown's appeal, albeit on a different issue than that which the Crown sought to pursue in this Court. The Crown's overall success in the court below precluded any further appeal, or cross-appeal, to this Court.

A similar question was recently considered by this Court in R. v. Barnes, [1991] 1 S.C.R. 449. In that case, a trial judge stayed proceedings in a drug case because of entrapment. The Crown appealed, seeking a reversal of the stay and a substitution of convictions against the accused. The appeal was successful in part: the Court of Appeal lifted the stay, but ordered a new trial rather than entering convictions. The accused appealed, and although the Crown did not cross-appeal it argued not only that the lifting of the stay should be affirmed, but that this Court should impose the convictions that the Court of Appeal had refused. Lamer C.J., writing for the majority, rejected the Crown's arguments on this latter point. He found that s. 693 of the Code has no application when a Crown appeal has been allowed. Moreover, Lamer C.J. relied on Guillemette v. The Queen, [1986] 1 S.C.R. 356, in finding that this Court's jurisdiction to vary a judgment of a court of appeal (pursuant to s. 695 of the Code) applies only when there is a valid Crown appeal before the Court. He concluded with the following passage, at p. 466:

In the absence of an appeal by the Crown, this Court has no jurisdiction to allow the Crown's request that the order below be modified. To hold otherwise would allow the Crown to appeal to this Court where such an opportunity has not been provided by the Criminal Code or the Supreme Court Act. The Crown is not given by statute the ability to appeal to this Court a decision which allowed its appeal from an acquittal or judicial stay of proceedings, but which gave the Crown less than what had been requested. As a result, there is no statutory provision which would allow the Crown to appeal from the Court of Appeal's judgment. Absent a statutory right of appeal, there is no right of appeal. [Emphasis in original.]

In my view, this reasoning is directly applicable to the present situation. As in Barnes, a court of appeal has allowed a Crown appeal, thereby precluding any appeal, or cross-appeal, by the Crown to this Court. The subdivision of a case on appeal into discrete grounds does not assist the Crown in this regard: an unfavourable ruling by a court of appeal on one point of law is overtaken by the Crown's success on other grounds.

In the result, since the second ground of appeal has been abandoned by the appellant, it is unnecessary for this Court to consider the Appeal Division's decision regarding the exclusion of the accused's second statement.

The Jury Charge

I turn now to the first ground of appeal, which concerns the charge to the jury on the accused's first statement to the police. All three judgments at the Appeal Division concluded that the jury instruction was to some degree flawed in that it might have left the jury with the impression that it was required to consider the out-of-court statement separate and apart from the evidence as a whole. The majority concluded that this error warranted a new trial, but in dissent Freeman J.A. found that any error was cured by other elements of the charge. Before considering the charge and these judgments in detail, it is useful to review R. v. Morin, supra, as that decision looms large in the debate in the court below and before this Court.

In Morin, this Court found that a jury charge will be in error if it leaves the jury with the impression that each item of evidence is to be considered piecemeal against the criminal standard of proof. As well, a majority found that it is an error to instruct a jury to take a two-stage approach to their deliberations, whereby an initial fact-finding stage would weed out certain items of evidence, leaving the determination of guilt or innocence to be based only on the surviving evidence. The jury charge at issue in Morin is reproduced in the majority reasons (at p. 355) beginning with the following general instruction:

You are not obliged to accept any part of the evidence of a witness just because there is no denial of it. If you have a reasonable doubt about any of the evidence you will give the benefit of that doubt to the accused with respect to such evidence. Having decided what evidence you consider worthy of belief, you will consider it as a whole, of course, in arriving at your verdict. [Emphasis added by Sopinka J.]

Later in the charge, the trial judge stated that the doctrine of reasonable doubt did not apply to each individual item of evidence, but rather to the whole of the case and to "the total body of evidence". However, he then gave a series of instructions on particular items of evidence, each of which suggested that the criminal standard was to be applied, in isolation, to that item. For example, with regard to certain hair and fibre evidence, the trial judge said:

It seems to me that this evidence does not go beyond proving that Christine could have been in the Honda motor vehicle and that the accused could have been at the scene of the killing and, of course, that is not proof beyond a reasonable doubt.

Sopinka J., writing for the majority, found that the charge as a whole was misleading to the jury. He commented as follows, at p. 356:

In my opinion, based on my reading of the charge as a whole, a jury would likely have concluded that in examining the evidence they were to give the accused the benefit of the doubt in respect of any evidence. This process of examination and elimination would occur during the so-called "fact finding" stage, to use the appellant's phrase. The evidence as a whole to which the jury was to apply itself in order to determine guilt or innocence was the residuum after the "fact finding" stage. There is no other way of reading the first excerpt from the charge. [Emphasis in original.]

Sopinka J. rejected the argument that the trial judge's second instruction cured the erroneous impression left by the first (at p. 356):

The second excerpt refers to the "whole of the case" and the "total body of evidence". Having been told earlier that the "whole" upon which the verdict was to be based consisted of the evidence that had been accepted, I am not satisfied that the jury would have interpreted this passage as a correction. Rather, they might very well have assumed that the earlier definition of the "whole" still applied. At best, from the appellant's standpoint the jury would be confused. Subsequent passages in the charge illustrate what is meant by the first excerpt and would confirm that individual pieces of evidence were to be examined by reference to the criminal standard.

Thus, the reversible error in Morin arose from the combined effect of two misdirections. The first was that the jury should assess the evidence in piecemeal fashion. This was exacerbated by the instruction to break the jury deliberations into two stages. In the end, Sopinka J. concluded, at p. 358:

The effect of the misdirections referred to above may very well have been that the jury examined evidence that was crucial to the Crown's case in bits and pieces. Standing alone or pitted against the evidence of the accused without the support of other evidence, much of this evidence might have been discarded as not measuring up to the test. When the jury came to consider the Crown's case as a whole there may not have been very much left of it. We cannot know for certain, but this scenario is a very likely one and the charge therefore constituted a serious misdirection.

With these findings from Morin in mind, I turn now to the jury charge in the present case.

In argument before this Court, the respondent focused on one particular passage in the charge, which instructed the jury members to "reject entirely" the accused's statement to the police if they had any reasonable doubt about it. It seems that the respondent's real complaint is with this single sentence, although other passages are impugned, and I will discuss them all.

The charge lasted over four hours, with two short breaks. MacIntosh J. organized his address to the jury as follows: general principles of a criminal trial, specific principles regarding murder, the theories of the Crown and defence, review and comment on the evidence, and finally general instructions on the available verdicts. The relevant portions of the charge are set out below, and for ease of reference I have lettered the paragraphs A to G. The impugned passages are underlined.

At the outset the trial judge explained the concept of proof beyond a reasonable doubt. He then said:

[A] The accused is entitled to the benefit of reasonable doubt on the whole of the case with respect to each count, and on each and every issue in the case. If after considering the whole of the case and the submissions made on behalf of the accused you entertain a reasonable doubt of guilt, you must resolve that doubt in his favour and acquit him.

The judge then turned to the question of the accused's testimony. He stated:

[B] In this case the accused himself gave evidence, and he is in the same position as any other witness as far as his credibility is concerned. Shortly I will instruct you how to weigh testimony, but for the present let me tell you that if you believe the accused, that he did not commit the offense, or what he did lacked some essential element of the offense, which I'll describe later, or if the evidence of the accused, either standing alone or taken together with all the other evidence, leaves you in a state of reasonable doubt, then you must acquit him. But if upon consideration of all the evidence, the arguments of counsel, that you are satisfied that the accused has been proven guilty beyond a reasonable doubt, it is your duty to convict.

Counsel for the respondent contended that paragraph A, in directing the jury to give the accused the benefit of reasonable doubt "on each and every issue in the case", exacerbated the putative errors which follow. I do not agree. It is entirely correct to state that the accused has the benefit of the doubt on each "issue", which is a reference to the Crown's burden of proving each element of the offence charged.

MacIntosh J. next gave a summary of the factors to be considered in weighing testimony, and stressed that the jury was not obliged to accept what a witness says. He then reviewed the law relating to the charge of murder and its included offences. Following this he turned to the portion of the charge to which the Crown objects. It deals with how the jury should approach its consideration of the appellant's statements to the police. He stated:

[C] Statements of the accused: You've heard evidence that -- it'd be from the evidence of Sergeant Peebles and Constable Cleary regarding statements to have been made by the accused to them on June the 25th. The fact that evidence of this -- these statements were given does not mean that they were made or that they were true. That's up for you to decide. It's for you to decide whether the statements were made and if you have a reasonable doubt about whether or not a particular statement was made in whole or in part, you must reject it entirely or reject those parts as to which you have a reasonable doubt as to the making. If you find the statements were made, you may believe all the statement, part of it, or you can reject them entirely. You are the sole judges whether an unsworn statement alleged to have been made by the accused in whole or in part is an acknowledgment by the accused of the truth of the matters contained in it.

[D] In determining whether or not the accused's statement is true in whole or in part, you might consider the following, one, has there been any corroboration by independent evidence of important parts of it, the circumstances leading up to the making of it, and the circumstances under which it was made, the condition of the accused at the time he is alleged to have made it? Statements were not given under oath, but you may act on those parts which in your judgment are an acknowledgement by the accused of the truth of the facts in the statement.

[E] Here we recall that the accused is saying he doesn't remember anything of what happened, that these statements he gave is what he had heard from other people. If you decide to accept part or all of it, it will be considered by you with the other evidence that you decide to accept. You, of course, must reach your verdict on the whole of the evidence that you decide is worthy of belief. The accused, as a witness, you can accept his earlier testimony as the statement that the police -- the Crown alleges he made to the police. You can accept that as the truth of what happened ("He" being the accused person.) as opposed to what he said in court. This we cannot do with an ordinary witness but with an accused, you are free to accept either, what he said in the statement or what he said in court. If you find that his evidence at trial represented the true facts, or if you have any reasonable doubt about it, you will reject entirely the earlier statement. [Emphasis added.]

All the instructions to this point were given in the first part of the jury charge. Towards the end of the charge, while reviewing the evidence, the trial judge recalled the accused's statement to the police and commented:

[F] Now you remember what I told you about statements of an accused persons and how you deal with it. You've seen the circumstances under which it was taken and you have the evidence of the condition of the accused at the time and so on. The evidence of Sergeant Peebles and Constable Cleary I found -- (That's up to you.) I found it pretty straight forward.

[G] You will recall that the accused is saying that that statement which I have just read was made up from what he had heard from other people. That's up for you to decide if that's true or not. Look at that statement and see whether all the information contained in it could have been obtained from other people or did it have to come from the accused itself.

That is the last of the relevant portions of the instruction to the jury.

As the underscoring suggests, the respondent points to three passages as reversible errors, although as I have noted counsel focused principally on the last sentence in paragraph E. I will consider each of the passages in detail, but it is important at the outset to make certain general observations about the charge. First, it seems to me that this charge is readily distinguishable from that in Morin. In that case the trial judge directed the jury to consider the evidence in piecemeal bits, and compounded that error by instructing the jury to break their deliberations into two stages. These errors were exacerbated by repeated instructions on specific items of evidence, all of which suggested a piecemeal approach and said nothing about considering that evidence with due regard to all the other evidence. In the present case there is no general instruction to apply the criminal standard of proof to items of evidence in isolation. The charge does focus on one item of evidence -- the appellant's statement to the police -- but the direction on this evidence is carefully couched in terms of all the other evidence in the case. Each time the trial judge addresses the appellant's statement he is careful to return to the "whole of the evidence". In Morin, by contrast, the charge to the jury on individual items of evidence never made this link. Finally, the charge in the present case does not instruct the jury to divide its deliberations into two phases. It is true that the first part of paragraph E refers to reaching a verdict "on the whole of the evidence that you decide is worthy of belief", but this stops short of suggesting a two-stage process. The choice of words is perhaps unfortunate, but I do not believe that this passing reference would have misled the jury.

In this case, unlike Morin, there was almost no chance for the jury to form the impression that "the whole of the evidence" was intended to mean only the whole of the piecemeal bits of evidence already proven beyond a reasonable doubt. Appellant's counsel pointed out that the trial judge gave fourteen distinct warnings to consider the evidence as a whole. These warnings, which occurred before, during and after the excerpts of the charge repeated above, strongly militate against any possibility that the jurors may have misconceived their role. I note also that the Crown took no exception to the jury charge at the time it was delivered. This is of course of no great moment, but it does suggest that any misdirection was not readily apparent to those who heard the charge.

As a final general comment, it is important to keep in mind just what MacIntosh J. was striving to achieve in these impugned passages. The contradiction between the accused's out-of-court statement and his testimony at trial was a key issue in the case, and the trial judge was entitled, in his review of the evidence, to give hints to the jury on how to assess such important issues; see Morin at p. 361. MacIntosh J. suggested that the jury focus on the two statements, and pointed out that as a matter of logic the two could not coexist. In my view his suggestion that one of the statements must be "rejected", couched as it was with the proviso that all of the other evidence must be taken into account, in no way prejudiced the Crown.

I turn now to the first and third of the impugned passages in the charge to the jury. In speaking about the accused's statement, the trial judge instructed the jurors in paragraph C that they must reject the accused's statement if they had a reasonable doubt as to whether the statement was made. In paragraph E he returned to this point with respect to the content of the first statement, and again invited the jury to reject the statement if they had a reasonable doubt as to its veracity. Chipman J.A. found the latter statement to be objectionable, and also stated a concern about the passage in paragraph C.

Chipman J.A. noted that paragraph C was virtually the same as an instruction disapproved by the Ontario Court of Appeal in R. v. Minhas (1986), 29 C.C.C. (3d) 193. The latter charge is reproduced in the Ontario Court of Appeal's decision at p. 209:

However, it is for you to decide in each instance whether the statements alleged to have been made by the accused were in fact made to the police or to Vijay Gupta or to Nishi Gupta. If you have a reasonable doubt as to whether any particular statement relied upon by the Crown was made in whole or in part, then you must reject it entirely or reject those parts about which you have a reasonable doubt in the making. If you are satisfied that the statement was made, then it is for you to determine whether it amounted to an acknowledgement of guilt by the accused.

Commenting on this instruction, Martin J.A. said this, at p. 210:

I observe that the instruction that the jury must be satisfied beyond a reasonable doubt that the statements alleged to have been made by the appellant were, in fact, made by her, and that they constituted an acknowledgement of guilt, was, in my view, more favourable to the appellant than the instruction to which she was by law entitled. The jury must on all the evidence be satisfied beyond a reasonable doubt of the accused's guilt in order to convict. If the jury at the end of the case entertains a reasonable doubt as to the accused's guilt they must acquit. The requirement of proof beyond a reasonable doubt does not, however, apply to each individual item of evidence that makes up the Crown's case.... [Emphasis in original.]

Minhas was decided before Morin, but Martin J.A.'s statement of the law is largely consistent with the approach taken by this Court in Morin, and I have no doubt that Martin J.A. was correct in pointing out that all of the evidence must be considered when determining the guilt of the accused beyond a reasonable doubt. In Minhas it appears that there was no instruction accompanying the "rejection" charge to emphasize that the decision to reject evidence could only be taken in light of all the evidence. As such, Martin J.A. was correct to observe that the charge before him was in error. In the present case, however, the impugned paragraphs C and E must be read with paragraph D, in which MacIntosh J. gave the jury explicit directions on how to assess the conflicting statements of the accused in light of all the evidence: they should consider "corroborating independent evidence", the "circumstances leading up to" the making of the first statement, and the "condition of the accused". So as to further emphasize this point, the trial judge returns to it in paragraphs F and G, which are his last words on the subject of the accused's conflicting evidence. Here again we find another direction to consider the appellant's statement in light of the other evidence from the trial.

In my view it is acceptable for a trial judge to focus the jury's attention on the vital issues of its inquiry. Perhaps the jury should not be directed to compartmentalize their deliberations, but at the same time it is unrealistic to view a jury's decision as some epiphanic pronouncement of guilt or innocence; rather, the jurors engage in a deliberate process of evaluating the evidence presented to them. In this case, the jury deliberated for two days on evidence from a trial that lasted more than three weeks. To restrain jurors from rejecting evidence during this process is to impose an artificial constraint on their mandate. Having said this, a trial judge should proceed with the utmost caution in advising the jury that certain evidence may be "rejected". If the trial judge ventures into this realm, then the advice must be complete -- he or she must go on to stress to the jury that this rejection must not be done in isolation -- the evidence is not to be evaluated in a piecemeal fashion.

I do not view this discretion of a trial judge to instruct on "rejecting" evidence as some exception to, or modification of this Court's disapproval of a two-stage process of evaluating evidence. Rather, it is merely a reflection of the reality of jury deliberations -- on important items of evidence the jury may require guidance on how to approach its task. As in this case, where a statement by an accused at trial is entirely at odds with a previous out-of-court statement by the accused, and the jury believes the statement at trial, or is left in reasonable doubt that it is true, then the jury must reject the out-of-court statement; the accused must be given the benefit of the doubt. In arriving at that conclusion, the jury should, of course, give consideration to the evidence as a whole.

At this juncture I would emphasize that the jury's "rejection" of evidence goes to the veracity of that evidence rather than its admissibility. The latter consideration is, of course, wholly within the province of the trial judge. On this point Freeman J.A. aptly commented in the court below (at p. 133):

The concern with Mr. Justice MacIntosh's charge is his use of the word "reject" while speaking of reasonable doubt as to the statement. If the meaning of "reject" were equated with "exclude", this could be problematical. However the jury must have known they could not exclude any evidence because they were instructed that their verdict had to be on the whole of the evidence. The ordinary meaning of "reject" is to refuse to accept, and on the whole of the charge there is no reason to conclude the jury would have understood it in any other sense. In that light if the jury understood they were to refuse to accept the statement as proof of the events it described until they were satisfied of the guilt of the accused on the whole of the evidence including the statement, I can see no difficulty.

Thus the "rejection" of certain evidence during the jury's deliberations does not engage a two-stage process.

The contradictory statements of the accused were a key element of this trial, and the judge was entitled to give the jury some guidance on how to handle this discrepancy in the evidence. As a matter of logic, the two stories could not be reconciled, and the advice to reject one of them is one way to deal with the dilemma. When a judge ventures into this kind of instruction, a reviewing court will have two concerns: (1) the jury must know that their job is not to choose the most believable of the versions, but that the version most favourable to the accused is entitled to the benefit of the doubt; and (2) the two versions cannot be simply pitted against one another in isolation, rather all of the other evidence must also be considered.

The first of these concerns was considered in this Court's decision in Nadeau v. The Queen, [1984] 2 S.C.R. 570. In that case, two contradictory versions of the events leading up to a murder were presented: one by the accused, the other by a Crown witness. The judge told the jury that they must choose between the two versions, and then said the following, which is reproduced at p. 572 of this Court's decision:

[translation] I will simply say that in deciding how you make your choice, you must have one thing clearly in mind: you must choose the more persuasive, the clearer version ....

You must keep in mind that, as the accused has the benefit of the doubt on all the evidence, if you come to the conclusion that the two (2) versions are equally consistent with the evidence, are equally valid, you must give -- you must accept the version more favourable to the accused. [Emphasis added by Lamer J.]

Lamer J. (as he then was), writing for the Court at pp. 572-73, assessed this instruction in the following terms:

With respect, this direction is in error. The accused benefits from any reasonable doubt at the outset, not merely if "the two (2) versions are equally consistent with the evidence, are equally valid". Moreover, the jury does not have to choose between two versions. It is not because they would not believe the accused that they would then have to agree with Landry's version. The jurors cannot accept his version, or any part of it, unless they are satisfied beyond all reasonable doubt, having regard to all the evidence, that the events took place in this manner; otherwise, the accused is entitled, unless a fact has been established beyond a reasonable doubt, to the finding of fact the most favourable to him, provided of course that it is based on evidence in the record and not mere speculation.

Although disapproving the charge as framed in that case, Lamer J. in Nadeau endorses the idea that a jury should be charged to "reject" inculpatory evidence contrary to an accused's statements if, having regard to all the evidence, the jury has a reasonable doubt about it. The jury instruction in the present case is an accurate reflection of the law as stated in Nadeau: an accused's version of events is entitled to the benefit of the doubt as against a competing version, providing the comparison is made having due regard to all the evidence.

The dual concerns of a reviewing court were considered together by Sopinka J. in Morin, at pp. 361-62:

While the charge may and often does include many helpful tips on the weighing of evidence such as observing demeanour, taking into account the interest of the witness and so forth, the law lays down only one basic requirement: during the process of deliberation the jury or other trier of fact must consider the evidence as a whole and determine whether guilt is established by the prosecution beyond a reasonable doubt....

The reason we have juries is so that lay persons and not lawyers decide the facts. To inject into the process artificial legal rules with respect to the natural human activity of deliberation and decision would tend to detract from the value of the jury system. Accordingly, it is wrong for a trial judge to lay down additional rules for the weighing of the evidence. Indeed, it is unwise to attempt to elaborate on the basic requirement referred to above. I would make two exceptions. The jury should be told that the facts are not to be examined separately and in isolation with reference to the criminal standard....

The second exception is that it is appropriate where issues of credibility arise between the evidence for the prosecution and the defence that the jury be charged as suggested by Morden J.A. in Challice, [infra]. There is a danger in such a situation that a jury might conclude that it is simply a matter as to which side they believe. The suggested charge alerts them to the fact that, if the defence evidence leaves them in a state of doubt after considering it in the context of the whole of the evidence, then they are to acquit.

The charge Sopinka J. was referring to is found in the Ontario Court of Appeal's decision in R. v. Challice (1979), 45 C.C.C. (2d) 546, at p. 557:

Understandably, a jury have to give careful consideration to issues of credibility when deliberating upon their verdict, and with respect to various pieces of evidence they may have differing views: total acceptance, total rejection, or something in between. An effective and desirable way of recognizing this necessary part of the process, and putting it to the jury in a way that accurately comports with their duty respecting the burden and standard of proof, is to instruct the jury that it is not necessary for them to believe the defence evidence on a vital issue -- but that it is sufficient if it, viewed in the context of all the evidence, leaves them in a state of reasonable doubt as to the accused's guilt....

It seems to me that MacIntosh J. instructed the jury in exactly the manner prescribed by Challice, and such a charge was appropriate in this case. This is made clear by Sopinka J.'s use of the Challice charge in Morin. There, certain out-of-court statements made by the accused were tendered in evidence against him, and the accused presented a competing version of the meaning of his statements at trial (exactly the situation in the present case). It was with reference to this problem that Sopinka J. allowed, at p. 357, that a jury should be charged as in Challice -- "where the credibility of defence evidence is at issue". It was precisely this scenario that confronted the jury in the present case, and for the reasons stated by Sopinka J. in Morin the charge by MacIntosh J. was entirely appropriate.

Morin demonstrates that it is both acceptable and desirable for a trial judge to focus the jury's attention on vital issues, and to direct their minds to the proper burden of proof on those discrete questions; see also R. v. Thatcher, [1987] 1 S.C.R. 652, at pp. 700-701. In my view, this is exactly what the trial judge has done in the present case. As Freeman J.A. commented in the court below, at p. 134, "... the words of Mr. Justice MacIntosh referred to above, to say nothing of his instructions as a whole, would have left the jury in a state of mind no different than if he had charged them in the words of Mr. Justice Lamer in Nadeau."

I turn now to the final passage of concern, which is found in the first part of paragraph E. The trial judge commented there that if the jury accepted the accused's statement to the police, then "it will be considered by you with the other evidence that you decide to accept. You, of course, must reach your verdict on the whole of the evidence that you decide is worthy of belief." This passage is arguably reminiscent of the impugned two-stage instruction in Morin, but in my view the statement here had a very different effect. Read in isolation, the passage may be viewed as an erroneous statement of the law: it suggests a determination of guilt should be based solely on certain pre-selected items of evidence. However, when the statement is read in context I do not think that it would have misdirected the jury. It should be recalled that the analogous instruction in Morin was made lockstep with the misdirection on taking a piecemeal approach to the evidence. By contrast, the statement here follows on the heels of the trial judge's direction in paragraph D to consider all of the evidence in making its determination. A juror who had just been instructed to corroborate the accused's statement with other evidence in the case could not have interpreted paragraph E as invoking a two-stage process -- the two concepts are irreconcilable.

I can deal briefly with the remaining arguments advanced by counsel for the respondent. First, he submitted that MacIntosh J. neglected to explicitly instruct the jury against taking a piecemeal approach to the evidence, thereby compounding the effect of his misdirections. On this point I have already cited the remarks of Sopinka J. in Morin, at p. 362, to the effect that "[t]he jury should be told that the facts are not to be examined separately and in isolation with reference to the criminal standard", but I do not regard this as an invariable formulation that must be invoked verbatim in every jury charge. Rather, it is sufficient if the charge conveys to the jury the message that they are to consider the evidence as a whole, and not in piecemeal fashion. Here, the instructions of MacIntosh J. made this point very clearly, and amount to the very direction suggested by Sopinka J.

The respondent also contended that the "compartmentalization" of the jury charge aggravated the putative error of the trial judge. That is, by setting up a separate segment of the charge headed "Statements of the accused", the trial judge reinforced the impression that a piecemeal approach to this evidence should be taken. I cannot accept this argument, and I confess to seeing no way to effectively present a four-hour jury charge other than by compartmentalization. Indeed, I would have thought that proper structure and organization would only enhance the jury's understanding of the charge.

Two other points were raised by the respondent in oral argument, both of which in my view were effectively rebutted by the appellant in reply. First, the respondent contended that even a slight misdirection regarding the accused's statements was fatal to the Crown, because it was the "whole case" for the prosecution. In reply, counsel for the appellant pointed to the array of circumstantial evidence implicating the accused as the killer, and to the accused's admission to William Cogger, which, if believed, tended to identify the accused as the murderer. Given this other evidence, it must surely be an overstatement to say that the Crown's "whole case" rested on the accused's statement to the police.

The respondent also submitted that MacIntosh J. erred in instructing the jury, in paragraph C, that it was for them to determine whether the statement to the police had even been made. The respondent contended that the voir dire, in which the statement was found to be voluntary and thus admissible, settled the issue of whether the statement had been made, and therefore this question should not have been left to the jury. To this the appellant responded by citing R. v. Gauthier, [1977] 1 S.C.R. 441, in which this Court found that it is for the jury, after a voir dire, to decide whether a confession had in fact been made. That seems to me to be a complete answer to the respondent's argument.

To summarize, none of the three impugned passages, taken in context, were misdirections. If I am wrong and one or more of these passages constitutes an error of law, then I believe that any such error was cured by the rest of the charge. As I have noted, the jury was instructed some fourteen times to consider the evidence as a whole. As well, MacIntosh J. explicitly directed the jury to evaluate the accused's statements in light of the other evidence before them. I cannot believe that the solitary instructions in question, placed as they are in a sea of contrary statements, would mislead a jury into a flawed process of considering evidence in a piecemeal fashion. As was stated in R. v. Demeter (1975), 25 C.C.C. (2d) 417, at p. 436, there will "probably never be a perfect charge or one that cloistered appellate counsel cannot find objectionable after minute scrutiny." Simply put, the charge to the jury must be read as a whole, bearing in mind the heavy onus that rests with the Crown to establish that the jury's verdict may have been affected by the impugned passage: Vézeau v. The Queen, [1977] 2 S.C.R. 277. In that light it is clear to me that MacIntosh J.'s charge, when viewed as a whole, was not misleading to the jury.

The Test for Review

The appellant's third ground of appeal is that the court below erred in holding that the jury's verdict "might have been different" had it been properly instructed. As I understand it, counsel for the appellant makes two distinct arguments on this ground. First, he submits that the Appeal Division wrongly applied a relaxed standard of review so as to effectively enable it to correct a verdict it viewed as unreasonable. Secondly, he contends that as a matter of logic the jury's verdict would not have been different even if they had accepted the accused's out-of-court statement to the police. Given my conclusions on the first ground of appeal, it is not strictly necessary for me to entertain the appellant's submissions on these issues, but I will touch upon them because they were fully argued before us.

The test for reviewing a Crown appeal of an acquittal pursuant to s. 686(4) of the Criminal Code was established by this Court in Vézeau v. The Queen, supra. The majority there stated the test in this way, at p. 292:

In the present case, therefore, it was the duty of the Crown, in order to obtain a new trial, to satisfy the Court that the verdict would not necessarily have been the same if the trial judge had properly directed the jury.

In dissent, Dickson J. appears to have stated the test more strictly, at p. 282:

... for a new trial to be ordered the Crown must satisfy the Court with a substantial degree of certainty that the jury would necessarily have convicted the accused in the absence of the offending words.

It is unclear from the judgments in Vézeau whether anything turned on these different formulations, and I would not wish to make too much of the point. However, Dickson J.'s formulation would obviously present a more difficult task for the Crown -- it would have to show that the jury's verdict would have been different, rather than establishing only that it may not have been the same, which is the majority's formula. In any event, the majority's view remains the law today, as was made clear by Sopinka J. in Morin, at p. 374:

The onus resting on the Crown when it appeals an acquittal was settled in Vézeau v. The Queen, [1977] 2 S.C.R. 277. It is the duty of the Crown to satisfy the court that the verdict would not necessarily have been the same if the jury had been properly instructed.

I am prepared to accept that the onus is a heavy one and that the Crown must satisfy the court with a reasonable degree of certainty. An accused who has been acquitted once should not be sent back to be tried again unless it appears that the error at the first trial was such that there is a reasonable degree of certainty that the outcome may well have been affected by it. Any more stringent test would require an appellate court to predict with certainty what happened in the jury room. That it cannot do.

The more stringent test stated by Dickson J. has not been revived with the passage of time.

The appellant, in oral submissions, took exception to the following passage from Chipman J.A., at p. 110, following his conclusion that a misdirection had occurred:

Moreover, this error was of sufficient gravity that it cannot be said that had it not been made the result would necessarily have been the same. Sopinka, J., said in Morin, supra, the charge with respect to the burden of proof lays down for the jury one of the most fundamental rules of the game. The confession was undoubtedly a crucial part of the Crown's case. Without it, the jury could easily have formed a reasonable doubt. With it, it is difficult to see how a reasonable doubt as to some degree of guilt at least could be entertained. It can therefore be said with a reasonable degree of certainty that the outcome may have been affected by the misdirection relating to the confession. The appeal should be allowed and a new trial ordered. [Emphasis ordered.]

The appellant argues that these statements relax the burden placed on the Crown. Counsel paraphrases the underlined passages as requiring only that the Crown show that the verdict "might have been different," suggesting that this is itself an improper standard. I do not agree. A test of "might have been different" seems to me to simply be the converse of "would not necessarily have been the same," and stating the proposition in this way does not, of course, mean that the onus somehow shifts away from the Crown. I would note that the first of Chipman J.A.'s impugned statements is a virtual duplicate of an alternative expression of the majority's formulation in Vézeau, supra, at p. 291. These statements are simply different ways of saying the same thing. As such, I must reject the appellant's submission that the court below in some way misstated the standard of appellate review.

The appellant's second argument is that the jurors had enough evidence before them to acquit the accused even absent a reasonable doubt about the accused's out-of-court statement. In other words, even if the appellant's confession was believed, other evidence at the trial was sufficient to raise a reasonable doubt in the minds of the jurors. In particular, the appellant cites the flaws in the circumstantial case presented by the Crown at trial. The appellant argues that the jury must have acquitted the accused because of a reasonable doubt as to the identification of the murderer, and that this doubt could have been raised solely by the weaknesses in the circumstantial evidence implicating the accused in the murders. Again, I must disagree.

Counsel's argument betrays a certain overconfidence in the strength of his case at trial. Essentially, the defence argued that gunshots heard before and after the time that the accused was with William Cogger raised a sort of alibi for the accused. This defence is premised on the following assumptions: that the shots heard by Daniel Girrior emanated from the killer's gun, that the time of those shots was accurately noted and recollected, and that William Cogger accurately recalled the precise time he spoke with the accused during this sequence of events. In my view these are substantial assumptions, and I do not share the appellant's confidence that the jury's reasonable doubt was likely derived solely from this defence theory. It seems more probable that the acquittal was based on a combination of doubts with respect to both the accused's out-of-court statement and the Crown's circumstantial evidence. If the jury was misled by the judge's instructions, then it is entirely possible that the jury's verdict would have been different. As such, the reasoning of the majority in the court below cannot be impugned in this respect.

The appellant (along with the Crown and the courts below) speculates as to what the jury actually did in response to the impugned instructions from the trial judge. The jury deliberated for two days, but very early in their deliberations they asked to re-hear the evidence of Daniel Girrior, William Cogger and Lorraine Boucher. Of course, it is impossible to know why the jury called for this evidence. However, their actions are at least consistent with the theory that the jurors were doing exactly what the trial judge instructed them to do -- they were attempting to test the veracity of the confession with independent evidence. The common link between the three witnesses whose testimony was replayed is that their evidence is the basis for the defence theory that the accused could not have fired all the shots. Cogger's testimony places the appellant back at his home between the groups of shots heard by Girrior, and Boucher testified that she heard a gunshot between 3:30 and 4:00 a.m. This evidence would have enabled the jurors to corroborate details of the appellant's confession -- the number of shots he fired -- with the evidence of the witnesses. The jury might also have been interested in the appellant's admission to Cogger that he had killed the men.

The Crown's theory of why the jurors asked to re-hear this evidence is that they began by considering the confession in isolation, gave the accused the benefit of the doubt, and accepted his explanations at trial, all within the first two hours of their deliberations. At this point, they would turn to the other evidence, and would be particularly concerned with the theory of the defence. In the absence of the confession (which they had already rejected) the defence theory would seem plausible, resulting in the acquittal.

As is apparent, almost anything can be read into the chronology of the jury's deliberations. These theories resolve very little, and do not assist in the task of determining whether the verdict would necessarily have been the same if the alleged misdirection had not occurred. For the reasons stated above, had the judge's instructions misled the jury (which, in my view, they did not) the resolution of this third ground would be that the verdict would not necessarily have been the same, and I would agree with the majority of the court below on this aspect of its reasons.

Disposition

In the result, because the jury charge was substantially error-free and did not mislead the jurors, I would allow the appeal and restore the acquittal.

Appeal allowed.

Solicitors for the appellant: Stewart McKelvey Stirling Scales, Halifax.

Solicitor for the respondent: The Attorney General of Nova Scotia, Halifax.

*Stevenson J. took no part in the judgment.