R. v. Lavallee, [1990] 1 S.C.R. 852

Angelique Lyn Lavallee Appellant

v.

Her Majesty The Queen Respondent

indexed as: r. v. lavallee

File No.: 21022.

1989: October 31; 1990: May 3.

Present: Dickson C.J. and Lamer, Wilson, L'Heureux-Dubé, Sopinka, Gonthier and McLachlin JJ.

on appeal from the court of appeal for manitoba

Evidence -- Admissibility -- Expert evidence -- Battered woman, fearing attack and possible death, killing spouse -- Defence of self-defence -- Expert witness giving psychiatric assessment of battered woman -- Assessment based in part on inadmissible evidence -- Whether or not expert evidence admissible ‑‑ Whether trial judge's charge to the jury with respect to expert evidence adequate -- Criminal Code, R.S.C., 1985, c. C-46, s. 34(2)(a), (b).

Criminal law -- Battered women -- Battered woman, fearing attack and possible death, killing spouse -- Defence of self-defence -- Expert witness giving psychiatric assessment of battered woman -- Assessment based in part on inadmissible evidence -- Whether or not expert evidence admissible -- Whether trial judge's charge to the jury with respect to expert evidence adequate.

Appellant, a battered woman in a volatile common law relationship, killed her partner late one night by shooting him in the back of the head as he left her room. The shooting occurred after an argument where the appellant had been physically abused and was fearful for her life after being taunted with the threat that either she kill him or he would get her. She had frequently been a victim of his physical abuse and had concocted excuses to explain her injuries to medical staff on those occasions. A psychiatrist with extensive professional experience in the treatment of battered wives prepared a psychiatric assessment of the appellant which was used in support of her defence of self-defence. He explained her ongoing terror, her inability to escape the relationship despite the violence and the continuing pattern of abuse which put her life in danger. He testified that in his opinion the appellant's shooting of the deceased the final desperate act of a woman who sincerely believed that she would be killed that night. In the course of his testimony, he related many things told to him by the appellant for which there was no admissible evidence. She did not testify at the trial. The jury acquitted the appellant but its verdict was overturned by a majority of the Manitoba Court of Appeal.

The issues before this Court were whether the evidence of the psychiatrist should have been before the court at all and whether, if it should, the trial judge's instructions with respect to it were adequate.

Held: The appeal should be allowed.

Per Dickson C.J. and Lamer, Wilson, L'Heureux-Dubé, Gonthier and McLachlin JJ.: Expert testimony is admissible to assist the fact-finder in drawing inferences in areas where the expert has relevant knowledge or experience beyond that of the lay person. It is difficult for the lay person to comprehend the battered wife syndrome. It is commonly thought that battered women are not really beaten as badly as they claim, otherwise they would have left the relationship. Alternatively, some believe that women enjoy being beaten, that they have a masochistic strain in them. Each of these stereotypes may adversely affect consideration of a battered woman's claim to have acted in self-defence in killing her partner. Expert evidence can assist the jury in dispelling these myths.

Expert testimony relating to the ability of an accused to perceive danger from her partner may go to the issue of whether she "reasonably apprehended" death or grievous bodily harm on a particular occasion. Expert testimony pertaining to why an accused remained in the battering relationship may be relevant in assessing the nature and extent of the alleged abuse. By providing an explanation as to why an accused did not flee when she perceived her life to be in danger, expert testimony may also assist the jury in assessing the reasonableness of her belief that killing her batterer was the only way to save her own life.

Expert evidence does not and cannot usurp the jury's function of deciding whether, in fact, the accused's perceptions and actions were reasonable. But fairness and the integrity of the trial process demand that the jury have the opportunity to hear that opinion.

Here, there was ample evidence on which the trial judge could conclude, apart from the psychiatrist's evidence, that the appellant was battered repeatedly and brutally by the deceased over the course of their relationship. The expert testimony was properly admitted in order to assist the jury in determining whether the appellant had a reasonable apprehension of death or grievous bodily harm and believed on reasonable grounds that she had no alternative but to shoot. Each of the specific facts underlying the expert's opinion need not be proven in evidence before any weight could be given to it. As long as there is some admissible evidence to establish the foundation for the expert's opinion, the trial judge cannot subsequently instruct the jury to completely ignore the testimony. The judge must, of course, warn the jury that the more the expert relies on facts not proved in evidence the less weight the jury may attribute to the opinion.

Per Sopinka J.: The very special facts in R. v. Abbey, and the decision required on those facts, have contributed to the development of a principle concerning the admissibility and weight of expert opinion evidence that is self-contradictory: an expert opinion relevant in the abstract to a material issue in a trial but based entirely on unproven hearsay is admissible but entitled to no weight whatsoever. Such an opinion, however, is irrelevant and therefore inadmissible. A practical distinction exists between evidence that an expert obtains and acts upon within the scope of his or her expertise, as in consultation with colleagues, and evidence that an expert obtains from a party to litigation touching a matter directly in issue. Where the information upon which an expert forms his or her opinion comes from a party to the litigation, or from any other source that is inherently suspect, a court ought to require independent proof of that information. The lack of such proof will have a direct effect on the weight to be given to the opinion. Where an expert's opinion is based in part upon suspect information and in part upon either admitted facts or facts sought to be proved, the matter is purely one of weight. That was the situation here, and in the circumstances, the trial judge properly admitted the expert evidence and adequately charged the jury.

Cases Cited

By Wilson J.

Applied: R. v. Abbey, 1982 CanLII 25 (SCC), [1982] 2 S.C.R. 24; considered: State v. Wanrow, 559 P.2d 548 (Wash. 1977); R. v. Whynot (1983), 1972 CanLII 165 (SCC), 9 C.C.C. 449; referred to: Kelliher (Village of) v. Smith, 1931 CanLII 1 (SCC), [1931] S.C.R. 672; R. v. Béland, 1987 CanLII 27 (SCC), [1987] 2 S.C.R. 398; R. v. Lyons, 1987 CanLII 25 (SCC), [1987] 2 S.C.R. 309; State v. Kelly, 478 A.2d 364 (N.J. 1984); Reilly v. The Queen, 1984 CanLII 83 (SCC), [1984] 2 S.C.R. 396; R. v. Baxter (1975), 1975 CanLII 1510 (ON CA), 33 C.R.N.S. 22; R. v. Bogue (1976), 1976 CanLII 871 (ON CA), 30 C.C.C. (2d) 403; State v. Gallegos, 719 P.2d 1268 (N.M. 1986); R. v. Antley (1963), 1963 CanLII 258 (ON CA), 42 C.R. 384.

By Sopinka J.

Considered: R. v. Abbey, 1982 CanLII 25 (SCC), [1982] 2 S.C.R. 24; referred to: City of St. John v. Irving Oil Co., 1966 CanLII 64 (SCC), [1966] S.C.R. 581; Wilband v. The Queen, 1966 CanLII 3 (SCC), [1967] S.C.R. 14; R. v. Lupien, 1969 CanLII 55 (SCC), [1970] S.C.R. 263; Ares v. Venner, 1970 CanLII 5 (SCC), [1970] S.C.R. 608; R. v. Jordan (1984), 1984 CanLII 635 (BC CA), 39 C.R. (3d) 50; R. v. Zundel (1987), 1987 CanLII 121 (ON CA), 56 C.R. (3d) 1.

Statutes and Regulations Cited

Criminal Code, R.S.C., 1985, c. C-46, ss. 34(2)(a), (b), 37.

Authors Cited

Blackman, Julie. "Potential Uses for Expert Testimony: Ideas Toward the Representation of Battered Women Who Kill" (1986), 9 Women's Rights Law Reporter 227.

Crocker, Phyllis. "The Meaning of Equality for Battered Women Who Kill Men in Self-Defense" (1985), 8 Harv. Women's L.J. 121.

Delisle, R. J. Evidence: Principles and Problems, 2nd ed. Toronto: Carswells, 1989.

Ewing, Charles Patrick. Battered Women Who Kill. Lexington, Mass.: Lexington Books, 1987.

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

Walker, Lenore E. The Battered Woman. New York: Harper & Row, 1979.

Walker, Lenore E. The Battered Woman Syndrome. New York: Springer Pub. Co., 1984.

Wardle, Peter. "R. v. Abbey and Psychiatric Opinion Evidence: Requiring the Accused to Testify" (1984), 17 Ottawa L. Rev. 116.

Willoughby, M. J. "Rendering Each Woman Her Due: Can a Battered Woman Claim Self-Defense When She Kills Her Sleeping Batterer" (1989), 38 Kan. L. Rev. 169.

APPEAL from judgment of the Manitoba Court of Appeal (1988), 1988 CanLII 7071 (MB CA), 52 Man. R. (2d) 274, 44 C.C.C. (3d) 113, 65 C.R. (3d) 387, allowing an appeal from acquittal by Scott A.C.J.Q.B. sitting with jury. Appeal allowed.

G. Greg Brodsky, Q.C., and S. Hoeppner, for the appellant.

J. G. B. Dangerfield, Q.C., for the respondent.

//Wilson J.//

The judgment of Dickson C.J. and Lamer, Wilson, L'Heureux-Dubé, Gonthier and McLachlin JJ.^*^ was delivered by

Wilson J. -- The narrow issue raised on this appeal is the adequacy of a trial judge's instructions to the jury regarding expert evidence. The broader issue concerns the utility of expert evidence in assisting a jury confronted by a plea of self-defence to a murder charge by a common law wife who had been battered by the deceased.

1. The Facts

The appellant, who was 22 years old at the time, had been living with Kevin Rust for some three to four years. Their residence was the scene of a boisterous party on August 30, 1986. In the early hours of August 31 after most of the guests had departed the appellant and Rust had an argument in the upstairs bedroom which was used by the appellant. Rust was killed by a single shot in the back of the head from a .303 calibre rifle fired by the appellant as he was leaving the room.

The appellant did not testify but her statement made to police on the night of the shooting was put in evidence. Portions of it read as follows:

Me and Wendy argued as usual and I ran in the house after Kevin pushed me. I was scared, I was really scared. I locked the door. Herb was downstairs with Joanne and I called for Herb but I was crying when I called him. I said, "Herb come up here please." Herb came up to the top of the stairs and I told him that Kevin was going to hit me actually beat on me again. Herb said he knew and that if I was his old lady things would be different, he gave me a hug. OK, we're friends, there's nothing between us. He said "Yeah, I know" and he went outside to talk to Kevin leaving the door unlocked. I went upstairs and hid in my closet from Kevin. I was so scared... . My window was open and I could hear Kevin asking questions about what I was doing and what I was saying. Next thing I know he was coming up the stairs for me. He came into my bedroom and said "Wench, where are you?" And he turned on my light and he said "Your purse is on the floor" and he kicked it. OK then he turned and he saw me in the closet. He wanted me to come out but I didn't want to come out because I was scared. I was so scared. [The officer who took the statement then testified that the appellant started to cry at this point and stopped after a minute or two.] He grabbed me by the arm right there. There's a bruise on my face also where he slapped me. He didn't slap me right then, first he yelled at me then he pushed me and I pushed him back and he hit me twice on the right hand side of my head. I was scared. All I thought about was all the other times he used to beat me, I was scared, I was shaking as usual. The rest is a blank, all I remember is he gave me the gun and a shot was fired through my screen. This is all so fast. And then the guns were in another room and he loaded it the second shot and gave it to me. And I was going to shoot myself. I pointed it to myself, I was so upset. OK and then he went and I was sitting on the bed and he started going like this with his finger [the appellant made a shaking motion with an index finger] and said something like "You're my old lady and you do as you're told" or something like that. He said "wait till everybody leaves, you'll get it then" and he said something to the effect of "either you kill me or I'll get you" that was what it was. He kind of smiled and then he turned around. I shot him but I aimed out. I thought I aimed above him and a piece of his head went that way.

The relationship between the appellant and Rust was volatile and punctuated by frequent arguments and violence. They would apparently fight for two or three days at a time or several times a week. Considerable evidence was led at trial indicating that the appellant was frequently a victim of physical abuse at the hands of Rust. Between 1983 and 1986 the appellant made several trips to hospital for injuries including severe bruises, a fractured nose, multiple contusions and a black eye. One of the attending physicians, Dr. Dirks, testified that he disbelieved the appellant's explanation on one such occasion that she had sustained her injuries by falling from a horse.

A friend of the deceased, Robert Ezako, testified that he had witnessed several fights between the appellant and the deceased and that he had seen the appellant point a gun at the deceased twice and threaten to kill him if he ever touched her again. Under cross-examination Ezako admitted to seeing or hearing the deceased beat up the appellant on several occasions and, during the preliminary inquiry, described her screaming during one such incident like "a pig being butchered". He also saw the appellant with a black eye on one occasion and doubted that it was the result of an accident as she and the deceased stated at the time. Another acquaintance of the couple recalled seeing the appellant with a split lip.

At one point on the night of his death Rust chased the appellant outside the house and a mutual friend, Norman Kolish, testified that the appellant pleaded with Rust to "leave me alone" and sought Kolish's protection by trying to hide behind him. A neighbour overheard Rust and the appellant arguing and described the tone of the former as "argumentative" and the latter as "scared". Later, between the first and second gunshot, he testified that he could hear that "somebody was beating up somebody" and the screams were female. Another neighbour testified to hearing noises like gunshots and then a woman's voice sounding upset saying "Fuck. He punched me in the face. He punched me in the face." He looked out the window and saw a woman matching the description of the appellant.

Three witnesses who attended the party testified to hearing sounds of yelling, pushing, shoving and thumping coming from upstairs prior to the gunshots. It is not disputed that two shots were fired by the appellant. The first one went through a window screen. It is not clear where Rust was at the time. The appellant in her statement says that he was upstairs, while another witness places him in the basement. The second shot was the fatal one. After the second shot was fired the appellant was seen visibly shaken and upset and was heard to say "Rooster [the deceased] was beating me so I shot him," and "You know how he treated me, you've got to help me." The arresting officer testified that en route to the police station the appellant made various comments in the police car, including "He said if I didn't kill him first he would kill me. I hope he lives. I really love him," and "He told me he was gonna kill me when everyone left."

The police officer who took the appellant's statement testified to seeing a red mark on her arm where she said the deceased had grabbed her. When the coroner who performed an autopsy on the deceased was shown pictures of the appellant (who had various bruises), he testified that it was "entirely possible" that bruises on the deceased's left hand were occasioned by an assault on the appellant. Another doctor noted an injury to the appellant's pinkie finger consistent with those sustained by the adoption of a defensive stance.

The expert evidence which forms the subject matter of the appeal came from Dr. Fred Shane, a psychiatrist with extensive professional experience in the treatment of battered wives. At the request of defence counsel Dr. Shane prepared a psychiatric assessment of the appellant. The substance of Dr. Shane's opinion was that the appellant had been terrorized by Rust to the point of feeling trapped, vulnerable, worthless and unable to escape the relationship despite the violence. At the same time, the continuing pattern of abuse put her life in danger. In Dr. Shane's opinion the appellant's shooting of the deceased was a final desperate act by a woman who sincerely believed that she would be killed that night:

... I think she felt, she felt in the final tragic moment that her life was on the line, that unless she defended herself, unless she reacted in a violent way that she would die. I mean he made it very explicit to her, from what she told me and from the information I have from the material that you forwarded to me, that she had, I think, to defend herself against his violence.

Dr. Shane stated that his opinion was based on four hours of formal interviews with the appellant, a police report of the incident (including the appellant's statement), hospital reports documenting eight of her visits to emergency departments between 1983 and 1985, and an interview with the appellant's mother. In the course of his testimony Dr. Shane related many things told to him by the appellant for which there was no admissible evidence. They were not in the appellant's statement to the police and she did not testify at trial. For example, Dr. Shane mentioned several episodes of abuse described by the appellant for which there were no hospital reports. He also related the appellant's disclosure to him that she had lied to doctors about the cause of her injuries. Dr. Shane testified that such fabrication was typical of battered women. The appellant also recounted to Dr. Shane occasions on which Rust would allegedly beat her, then beg her forgiveness and ply her with flowers and temporary displays of kindness. Dr. Shane was aware of the incidents described by Ezako about the appellant's pointing a gun at Rust on two occasions and explained it as "an issue for trying to defend herself. She was afraid that she would be assaulted." The appellant denied to Dr. Shane that she had homicidal fantasies about Rust and mentioned that she had smoked some marijuana on the night in question. These facts were related by Dr. Shane in the course of his testimony.

The appellant was acquitted by a jury but the verdict was overturned by a majority of the Manitoba Court of Appeal and the case sent back for retrial.

2. Lower Court Judgments

Manitoba Queen's Bench (Scott A.C.J.Q.B.)

After Dr. Shane testified and was cross-examined Crown counsel brought an application to have the evidence of Dr. Shane withdrawn from the jury. The first reason he gave was that the jury was perfectly capable of deciding the issue on the admissible evidence and that expert evidence was therefore "unnecessary and superfluous". The second reason was that Dr. Shane's comment that he found the accused credible was "wholly improper" in light of her failure to testify as to the facts upon which Dr. Shane based his opinion. The trial judge denied the application stating that the Crown's concerns could be met through an appropriate charge to the jury:

But I understand fully the concern that the Crown has at this time because a substantial chunk of the factual evidence that Dr. Shane relied on is simply not evidence in these proceedings and is not before the jury and my task, even with a very attentive jury such as this one, is going to be very difficult because of that fact.

But I think, under the circumstances, that the better course of action and the more realistic one to follow is to deal with the fact that it is in evidence and to attempt to explain to the jury as adequately and as fully as I can the difference between what is evidence and what is not in evidence and the impact that that ought to have on the weight that they choose to attach to the opinion of Dr. Shane.

With respect to the appellant's out-of-court statements, the trial judge cautioned the jury that, "[a]s with the verbal testimony, you may accept all, part or none of the statements attributed to Lyn Lavallee and as with all evidence, the real question is whether the things reported to have been said are true." Later he introduced Dr. Shane's testimony as follows:

As counsel put it yesterday, you cannot decide this case on things you didn't hear. You cannot decide this case on things the witnesses didn't see or hear.

A somewhat different, though related, evidentiary caution has to be noted with respect to the expert opinion evidence of Dr. Shane. There were two matters in his evidence, two facts, two sources of information that he had reference to which are not evidence in this case and that is the suggestion that people had been smoking marijuana at the party and the confirmatory evidence, as he called it, received from the mother of Lyn Lavallee. These are not matters in evidence before you.

For example, there is absolutely no evidence that anyone was smoking marijuana at this party and you must not consider that it took place. There is no evidence from the mother of the accused before you.

The extent to which this impacts on the weight of the opinion of Dr. Shane is a matter for you to decide. You must appraise the value of the resulting opinion in light of the fact that there is no evidence about these matters before you. In terms of the matters considered by Dr. Shane he is left, therefore, with the deceased's statement, some supplementary information from the police report and his interpretation of the hospital records.

If the premises upon which the information is substantially based has not been proven in evidence, it is up to you to conclude that it is not safe to attach a great deal of weight to the opinion. An opinion of an expert depends, to a large extent, on the validity of the facts assumed by the evidence by the expert.

If there are some errors and the factual assumptions aren't too important to the eventual opinion, that's one thing. If there are errors or matters not in evidence and those matters are substantial, in you view, in terms of the impact on the expert's opinion, then you will want to look at the value and weight of that opinion very carefully. It depends on how important you think the matters were that Dr. Shane relied on that are not in evidence. [Emphasis added.]

The trial judge then reviewed the evidence given by Dr. Shane regarding the appellant's emotional and mental state at the time of the killing. He reiterated Dr. Shane's opinion that the appellant's act was "a reflection of her catastrophic fear that she had to defend herself". He also drew attention to Dr. Shane's awareness that the appellant would occasionally be the aggressor despite her denial to him that she had homicidal fantasies:

[Dr. Shane] noted that at times Lyn Lavallee would be the aggressor from all of the underlying hostility. She couldn't leave psychologically because there were steel fences in her mind and she was tyrannized psychologically. She said she loved him and he felt that she did.

She denied to him thinking at any time of killing Kevin Rust. That is to say she did not entertain any homicidal fantasies and he felt that what she told him was reasonable.

It is the position of the Crown that Dr. Shane's opinion stands or falls on the veracity of Lyn Lavallee because he relied so heavily and extensively on what she told him and the evidence contained in the statement, Exhibit 16. That's for you to decide.

Undoubtably [sic] she was a very important source, if not the major source, of his information. Dr. Shane agreed that if what she told him was erroneous, he would have to reassess his position.

On cross-examination he reiterated that in his opinion her action was spontaneous to the moment to try to defend herself. The straw that broke the camel's back was the threat, "When the others leave you're going to get it.", even though similar statements had been made to her on other occasions. According to what she told him, the accused felt compelled to shoot.

Based on the information he had in his interview, it was his opinion that the acts of the accused were impulsive and not premeditated. He disagreed with the Crown's suggestion that Lyn Lavallee took the opportunity when it presented itself.

He conceded that patients had, on occasion, lied and misled him in the past.

Manitoba Court of Appeal (Monnin C.J.M., Philp and Huband JJ.) (1988), [1988 CanLII 7429 (MB CA)](https://www.canlii.org/en/mb/mbca/doc/1988/1988canlii7429/1988canlii7429.html), 52 Man. R. (2d) 274

Writing for himself and Monnin C.J.M., Philp J.A. begins by observing, at p. 275, that there was "ample evidence for the jury to conclude that Rust abused the accused." He adds that it "was a reasonable inference for the jury to draw that the injuries resulted from Rust's violent and abusive behaviour, notwithstanding her explanations at the time to the contrary".

Turning to Dr. Shane's evidence, the majority comments that in the course of stating the factual basis of his opinions and conclusions, Dr. Shane referred to many facts, incidents and events which were not before the court in the form of admissible evidence. These included: the smoking of marijuana on the night of the shooting; the deterioration of the intimate relationship between the appellant and Rust (the appellant had told Shane that they were sleeping in separate bedrooms); a reference to an abortion the appellant had obtained, after which Rust allegedly threatened to tell her parents that she was a "baby killer"; incidents where Rust would allegedly beg forgiveness from the appellant after beating her up; the appellant's "incredible remorse" after killing Rust, and the appellant's denial to Dr. Shane that she harboured homicidal fantasies about Rust.

Philp J.A. then refers to the appellant's written statement to the police in which she professed her love for Rust and her hope that he wouldn't die. At p. 277, he pointed out "discrepancies and conflicts in the narrative of events in the accused's statement, and the evidence of witnesses who testified at her trial", particularly with respect to the location of Rust when the first shot was fired. With respect to the accused's unsworn statement he concludes at p. 278:

... in the circumstances of this case, where much of the factual basis for the plea of self-defence lay in the statement of the accused, the jury ought not to have been told to "give this evidence no more nor less weight than any other evidence heard by you"; that the frailties of such assertions should have been pointed out.

The instructions of the trial judge to the jury with respect to the evidence of Dr. Shane are a more troubling matter. The problem presented by the accused's out of court statement and comments, in my view, comes to a head in that context.

Philp J.A. then turns to the judgment of Dickson J. (as he then was) in R. v. Abbey, 1982 CanLII 25 (SCC), [1982] 2 S.C.R. 24, a case from this Court dealing with the admissibility of expert evidence and the use to which it can be put. After quoting from the judgment, Philp J.A. states at p. 279:

Canadian authorities support the view that an expert can state to the court the basis for his opinion, and that it is desirable that he do so. In Abbey, Dickson, J., confirmed this approach and referred to the "obligation" of the party tendering evidence of the factual basis for the opinions of experts, to establish, "through properly admissible evidence, the factual basis on which such opinions are based". He cautioned: "Before any weight can be given to an expert's opinion, the facts upon which the opinion is based must be found to exist."

Referring back to the case at bar, Philp J.A. comments, at p. 279, that the record did not disclose "the full extent of these secondhand facts, or their importance in the formation of Dr. Shane's opinion; nor can one speculate what his opinion might have been had his inquiries been limited to the admissible evidence properly before the court."

In his assessment of the trial judge's charge to the jury, Philp J.A. remarks that the trial judge properly pointed out that there was no evidence about marijuana smoking on the night in question, nor was there any evidence before them from the mother of the appellant. Philp J.A. found this latter warning insufficient. While he considered the trial judge's general instructions regarding the weight that should be placed on expert evidence to be proper, he felt that they "did not go far enough in the circumstances of this case". He gives three reasons (at p. 280):

Firstly, the comments, placed in juxtaposition to the trial judge's reference to the "two facts, two sources of information that (Dr. Shane) had reference to which are not evidence in this case . . .", lose their impact. The jury may well have concluded that the trial judge's warning related only to Dr. Shane's reference to the marijuana, and to the "confirmatory evidence" of the accused's mother.

Secondly, I think the trial judge was in error in telling the jury that the police report (presumably, the document referred to by Dr. Shane as the "police summary of the incident") was a matter left for Dr. Shane to consider. That document was not evidence before the court, nor do we know what facts it contained.

Finally, although the trial judge did not refer to Dr. Shane's interviews with the accused (and her mother) when he told the jury what matters were left for Dr. Shane to consider, the conclusion that the jury was to ignore facts related in these interviews unless they were otherwise established by admissible evidence (and to weigh Dr. Shane's opinion accordingly) is dispelled by the trial judge's later references to these interviews.

Philp J.A. then quotes the passages from the trial judge's charge in which he reviewed Dr. Shane's admission that he would have to reassess his position if what the appellant had told him was not true. Philp J.A. also draws attention to the remark by the trial judge that the Crown emphasized that Shane's opinion would stand or fall on the appellant's veracity. In Philp J.A.'s view, these aspects of the trial judge's instructions were also deficient (at p. 281):

With respect, those comments of the trial judge, so crucial to the plea of self-defence, amounted to a misdirection. The issue was not just the veracity of the accused (and at this point, a careful charge with respect to the accused's unsworn self-serving evidence would have been appropriate). The pivotal questions the jury had to decide were the extent to which Dr. Shane's opinion was based on facts not established by admissible evidence; and the weight to be accorded to his opinion.

Finally, Philp J.A. finds, at p. 281, that the trial judge's charge fell so short of the standard required in Abbey that a new trial was warranted:

This was an unusual case. The accused shot Rust in the back of the head when he was leaving the bedroom. The accused says Rust loaded the rifle and handed it to her. Friends of the accused and Rust, including the couple who had planned to stay overnight, were present in another part of the residence. In these circumstances, absent the evidence of Dr. Shane, it is unlikely that the jury, properly instructed, would have accepted the accused's plea of self-defence. The accused did not testify, and the foundation for her plea of self-defence was, in the main, her unsworn exculpatory evidence and the hearsay evidence related by Dr. Shane. Because Dr. Shane relied upon facts not in evidence, including those related to him in his lengthy interviews with the accused, the factual basis for his opinion should have been detailed in his evidence.

Philp J.A. concludes by suggesting to the Crown that they proceed with a charge of manslaughter rather than second degree murder since a properly instructed jury would, in his opinion, be unlikely to convict the appellant of the latter offence.

Writing in dissent Huband J.A. summarizes the basis of Dr. Shane's opinion that the appellant acted out of a genuine fear for her life. He acknowledges, at p. 282, that "self-defence in this context finds some support in the evidence presented to the jury."

Huband J.A. points out the conflict in the evidence about how the appellant obtained the gun and where Rust was when the first shot was fired. Noting the appellant's statement to the police about how frightened of the deceased she was, he states, at p. 282, that "the significance of the statement is that, if believed, it establishes some foundation for a psychiatric opinion that she acted out of fear for her own safety as a person who had been subjected to continuous abuse." Ezako's evidence confirms that Rust beat the appellant and that, although the appellant may have "often provoked" the arguments, she "invariably got the worst of it".

With respect to Dr. Shane's evidence, Huband J.A. remarks, at p. 273, that in addition to hospital records and the accused's statement to the police, Dr. Shane "had the advantage of speaking at length with the accused herself, and also with the mother of the accused, in formulating his opinion". In Huband J.A.'s view (at p. 283), the "learned trial judge was well aware of the need to give adequate warning to the jury as to the weight to be placed upon the testimony of Dr. Shane. It was quite obvious that Dr. Shane relied upon statements by the accused and her mother which were unsworn hearsay comments and not part of the evidence in the case."

Turning to this Court's judgment in Abbey, Huband J.A. expresses the view, at p. 283, that the "learned trial judge followed the advice of the Supreme Court of Canada to the letter":

The learned trial judge begins by making it clear to the jury that they could not rely on the opinion of Dr. Shane on matters where there was no supporting evidence. He chose a good example. In relating what the accused had told him Dr. Shane said that marijuana had been smoked at the party that evening. None of the witnesses who testified as to what occurred during the evening mentioned any involvement with marijuana. Nor is such an involvement indicated in the accused's statement to the police. Dr. Shane's testimony constitutes no proof that marijuana was smoked, and to the extent that it became a factor in formulating his opinion, then his opinion must be discounted.

The learned trial judge also noted that while Dr. Shane referred to conversations with the accused's mother, no such evidence had been presented to the jury.

But the learned trial judge could not tell the jury to disregard Dr. Shane's report in its entirely. He was required to tell the jury, as he did, that there was some evidentiary support for Dr. Shane's opinion, -- the accused's own statement and the hospital records. Indeed, if anything, I think the learned trial judge was unfair to the accused in not also mentioning the evidence of Mr. Ezako as constituting evidentiary foundation supporting Dr. Shane's opinion.

Huband J.A. finds that the trial judge specifically warned the jury that Dr. Shane's conversations with the accused and her mother extended beyond the evidence before the court when he told them that "it is not safe to attach a great deal of weight" to the opinion of an expert when the information on which it is predicated has not been proven in evidence. After quoting the relevant passage Huband J.A. comments at p. 284:

I suppose one could argue that the learned trial judge should have said it is not safe to attach "any" weight to the opinion rather than it is not safe to attach "a great deal" of weight to the opinion. He could have said that an expert's opinion depends "totally" instead of "to a large extent" on the validity of the factual foundation upon which he has proceeded. In my view, however, the jury would fully comprehend the import of the learned trial judge's remarks. After the jury received its instruction and retired, counsel was invited to comment on the charge ... Crown counsel voiced no complaints.

This accused was acquitted by a jury of her peers on the basis of self-defence, which might strike one as being somewhat fanciful. We should not, however, search out semantic excuses to order a new trial, at high public cost, in the belief that the jury should have been more skeptical and arrived at a different verdict.

3. Relevant Legislation

Criminal Code, R.S.C., 1985, c. C-46:

34. ...

(2) Every one who is unlawfully assaulted and who causes death or grievous bodily harm in repelling the assault is justified if

(a) he causes it under reasonable apprehension of death or grievous bodily harm from the violence with which the assault was originally made or with which the assailant pursues his purposes, and

(b) he believes on reasonable and probable grounds, that he cannot otherwise preserve himself from death or grievous bodily harm.

4. Issues on Appeal

It should be noted that two bases for ordering a new trial are implicit in the reasons of the majority of the Court of Appeal. In finding that "absent the evidence of Dr. Shane, it is unlikely that the jury, properly instructed, would have accepted the accused's plea of self‑defence" the Court of Appeal suggests that the evidence of Dr. Shane ought to have been excluded entirely. The alternative ground for allowing the Crown's appeal was that Dr. Shane's testimony was properly admitted but the trial judge's instructions with respect to it were deficient. Thus, the issues before this Court are as follows:

1. Did the majority of the Manitoba Court of Appeal err in concluding that the jury should have considered the plea of self-defence absent the expert evidence of Dr. Shane?

2. Did the majority of the Manitoba Court of Appeal err in holding that the trial judge's charge to the jury with respect to Dr. Shane's expert evidence did not meet the requirements set out by this Court in Abbey, thus warranting a new trial?

5. Analysis

(i) Admissibility of Expert Evidence

In Kelliher (Village of) v. Smith, 1931 CanLII 1 (SCC), [1931] S.C.R. 672, at p. 684, this Court adopted the principle that in order for expert evidence to be admissible "the subject-matter of the inquiry must be such that ordinary people are unlikely to form a correct judgment about it, if unassisted by persons with special knowledge". More recently, this Court addressed the admissibility of expert psychiatric evidence in criminal cases in R. v. Abbey, supra. At p. 42 of the unanimous judgment Dickson J. stated the rule as follows:

With respect to matters calling for special knowledge, an expert in the field may draw inferences and state his opinion. An expert's function is precisely this: to provide the judge and jury with a ready-made inference which the judge and jury, due to the technical nature of the facts, are unable to formulate. "An expert's opinion is admissible to furnish the Court with scientific information which is likely to be outside the experience and knowledge of a judge or jury. If on the proven facts a judge or jury can form their own conclusions without help, then the opinion of the expert is unnecessary" (Turner (1974), 60 Crim. App. R. 80, at p. 83, per Lawton L.J.)

See also R. v. Béland, 1987 CanLII 27 (SCC), [1987] 2 S.C.R. 398, at p. 415, in which McIntyre J. speaks of an expert witness possessing "special knowledge and experience going beyond that of the trier of fact".

Where expert evidence is tendered in such fields as engineering or pathology, the paucity of the lay person's knowledge is uncontentious. The long-standing recognition that psychiatric or psychological testimony also falls within the realm of expert evidence is predicated on the realization that in some circumstances the average person may not have sufficient knowledge of or experience with human behaviour to draw an appropriate inference from the facts before him or her. An example may be found in R. v. Lyons, 1987 CanLII 25 (SCC), [1987] 2 S.C.R. 309, in which this Court approved the use of psychiatric testimony in dangerous offender applications. At p. 366, La Forest J. remarks that "psychiatric evidence is clearly relevant to the issue whether a person is likely to behave in a certain way and, indeed, is probably relatively superior in this regard to the evidence of other clinicians and lay persons".

The need for expert evidence in these areas can, however, be obfuscated by the belief that judges and juries are thoroughly knowledgeable about "human nature" and that no more is needed. They are, so to speak, their own experts on human behaviour. This, in effect, was the primary submission of the Crown to this Court.

The bare facts of this case, which I think are amply supported by the evidence, are that the appellant was repeatedly abused by the deceased but did not leave him (although she twice pointed a gun at him), and ultimately shot him in the back of the head as he was leaving her room. The Crown submits that these facts disclose all the information a jury needs in order to decide whether or not the appellant acted in self-defence. I have no hesitation in rejecting the Crown's submission.

Expert evidence on the psychological effect of battering on wives and common law partners must, it seems to me, be both relevant and necessary in the context of the present case. How can the mental state of the appellant be appreciated without it? The average member of the public (or of the jury) can be forgiven for asking: Why would a woman put up with this kind of treatment? Why should she continue to live with such a man? How could she love a partner who beat her to the point of requiring hospitalization? We would expect the woman to pack her bags and go. Where is her self-respect? Why does she not cut loose and make a new life for herself? Such is the reaction of the average person confronted with the so-called "battered wife syndrome". We need help to understand it and help is available from trained professionals.

The gravity, indeed, the tragedy of domestic violence can hardly be overstated. Greater media attention to this phenomenon in recent years has revealed both its prevalence and its horrific impact on women from all walks of life. Far from protecting women from it the law historically sanctioned the abuse of women within marriage as an aspect of the husband's ownership of his wife and his "right" to chastise her. One need only recall the centuries old law that a man is entitled to beat his wife with a stick "no thicker than his thumb".

Laws do not spring out of a social vacuum. The notion that a man has a right to "discipline" his wife is deeply rooted in the history of our society. The woman's duty was to serve her husband and to stay in the marriage at all costs "till death do us part" and to accept as her due any "punishment" that was meted out for failing to please her husband. One consequence of this attitude was that "wife battering" was rarely spoken of, rarely reported, rarely prosecuted, and even more rarely punished. Long after society abandoned its formal approval of spousal abuse tolerance of it continued and continues in some circles to this day.

Fortunately, there has been a growing awareness in recent years that no man has a right to abuse any woman under any circumstances. Legislative initiatives designed to educate police, judicial officers and the public, as well as more aggressive investigation and charging policies all signal a concerted effort by the criminal justice system to take spousal abuse seriously. However, a woman who comes before a judge or jury with the claim that she has been battered and suggests that this may be a relevant factor in evaluating her subsequent actions still faces the prospect of being condemned by popular mythology about domestic violence. Either she was not as badly beaten as she claims or she would have left the man long ago. Or, if she was battered that severely, she must have stayed out of some masochistic enjoyment of it.

Expert testimony on the psychological effects of battering have been admitted in American courts in recent years. In State v. Kelly, 478 A.2d 364 (1984), at p. 378, the New Jersey Supreme Court commended the value of expert testimony in these terms:

It is aimed at an area where the purported common knowledge of the jury may be very much mistaken, an area where jurors' logic, drawn from their own experience, may lead to a wholly incorrect conclusion, an area where expert knowledge would enable the jurors to disregard their prior conclusions as being common myths rather than common knowledge.

The Court concludes at p. 379 that the battering relationship is "subject to a large group of myths and stereotypes." As such, it is "beyond the ken of the average juror and thus is suitable for explanation through expert testimony." I share that view.

(ii) The Relevance of Expert Testimony to the Elements of Self-Defence

In my view, there are two elements of the defence under s. 34(2) of the Code which merit scrutiny for present purposes. The first is the temporal connection in s. 34(2)(a) between the apprehension of death or grievous bodily harm and the act allegedly taken in self-defence. Was the appellant "under reasonable apprehension of death or grievous bodily harm" from Rust as he was walking out of the room? The second is the assessment in s. 34(2)(b) of the magnitude of the force used by the accused. Was the accused's belief that she could not "otherwise preserve herself from death or grievous bodily harm" except by shooting the deceased based "on reasonable grounds"?

The feature common to both s. 34(2)(a) and (b) is the imposition of an objective standard of reasonableness on the apprehension of death and the need to repel the assault with deadly force. In Reilly v. The Queen, 1984 CanLII 83 (SCC), [1984] 2 S.C.R. 396, this Court considered the interaction of the objective and subjective components of s. 34(2), at p. 404:

Subsection (2) of s. 34 places in issue the accused's state of mind at the time he caused death. The subsection can only afford protection to the accused if he apprehended death or grievous bodily harm from the assault he was repelling and if he believed he could not preserve himself from death or grievous bodily harm otherwise than by the force he used. Nonetheless, his apprehension must be a reasonable one and his belief must be based upon reasonable and probable grounds. The subsection requires that the jury consider, and be guided by, what they decide on the evidence was the accused's appreciation of the situation and his belief as to the reaction it required, so long as there exists an objectively verifiable basis for his perception.

Since s. 34(2) places in issue the accused's perception of the attack upon him and the response required to meet it, the accused may still be found to have acted in self-defence even if he was mistaken in his perception. Reasonable and probable grounds must still exist for this mistaken perception in the sense that the mistake must have been one which an ordinary man using ordinary care could have made in the same circumstances. [Emphasis in original.]

If it strains credulity to imagine what the "ordinary man" would do in the position of a battered spouse, it is probably because men do not typically find themselves in that situation. Some women do, however. The definition of what is reasonable must be adapted to circumstances which are, by and large, foreign to the world inhabited by the hypothetical "reasonable man".

I find the case of State v. Wanrow, 559 P.2d 548 (1977), helpful in illustrating how the factor of gender can be germane to the assessment of what is reasonable. In Wanrow the Washington Supreme Court addressed the standard by which a jury ought to assess the reasonableness of the female appellant's use of a gun against an unarmed intruder. The Court pointed out that the appellant had reason to believe that the intruder had molested her daughter in the past and was coming back for her son. The appellant was a 5'4" woman with a broken leg. The assailant was 6'2" and intoxicated. The Court first observed, at p. 558, that "in our society women suffer from a conspicuous lack of access to training in and the means of developing those skills necessary to effectively repel a male assailant without resorting to the use of deadly weapons." Later it found that the trial judge erred in his instructions to the jury by creating the impression that the objective standard of reasonableness to be applied to the accused was that of an altercation between two men. At p. 559, the Court makes the following remarks which I find apposite to the case before us:

The respondent was entitled to have the jury consider her actions in the light of her own perceptions of the situation, including those perceptions which were the product of our nation's "long and unfortunate history of sex discrimination." Until such time as the effects of that history are eradicated, care must be taken to assure that our self-defense instructions afford women the right to have their conduct judged in light of the individual physical handicaps which are the product of sex discrimination. To fail to do so is to deny the right of the individual woman involved to trial by the same rules which are applicable to male defendants.

I turn now to a consideration of the specific components of self-defence under s. 34(2) of the Criminal Code.

A. Reasonable Apprehension of Death

Section 34(2)(a) requires that an accused who intentionally causes death or grievous bodily harm in repelling an assault is justified if he or she does so "under reasonable apprehension of death or grievous bodily harm". In the present case, the assault precipitating the appellant's alleged defensive act was Rust's threat to kill her when everyone else had gone.

It will be observed that s. 34(2)(a) does not actually stipulate that the accused apprehend imminent danger when he or she acts. Case law has, however, read that requirement into the defence: see Reilly v. The Queen, supra; R. v. Baxter (1975), 1975 CanLII 1510 (ON CA), 33 C.R.N.S. 22 (Ont. C.A.); R. v. Bogue (1976), 1976 CanLII 871 (ON CA), 30 C.C.C. (2d) 403 (Ont. C.A.) The sense in which "imminent" is used conjures up the image of "an uplifted knife" or a pointed gun. The rationale for the imminence rule seems obvious. The law of self-defence is designed to ensure that the use of defensive force is really necessary. It justifies the act because the defender reasonably believed that he or she had no alternative but to take the attacker's life. If there is a significant time interval between the original unlawful assault and the accused's response, one tends to suspect that the accused was motivated by revenge rather than self-defence. In the paradigmatic case of a one-time barroom brawl between two men of equal size and strength, this inference makes sense. How can one feel endangered to the point of firing a gun at an unarmed man who utters a death threat, then turns his back and walks out of the room? One cannot be certain of the gravity of the threat or his capacity to carry it out. Besides, one can always take the opportunity to flee or to call the police. If he comes back and raises his fist, one can respond in kind if need be. These are the tacit assumptions that underlie the imminence rule.

All of these assumptions were brought to bear on the respondent in R. v. Whynot (1983), 1972 CanLII 165 (SCC), 9 C.C.C. 449 (N.S.C.A.) The respondent, Jane Stafford, shot her sleeping common law husband as he lay passed out in his truck. The evidence at trial indicated that the deceased "dominated the household and exerted his authority by striking and slapping the various members and from time to time administering beatings to Jane Stafford and the others" (at p. 452). The respondent testified that the deceased threatened to kill all of the members of her family, one by one, if she tried to leave him. On the night in question he threatened to kill her son. After he passed out the respondent got one of the many shotguns kept by her husband and shot him. The Nova Scotia Court of Appeal held that the trial judge erred in leaving s. 37 (preventing assault against oneself or anyone under one's protection) with the jury. The Court stated at p. 464:

I do not believe that the trial judge was justified in placing s. 37 of the Code before the jury any more than he would have been justified in giving them s. 34. Under s. 34 the assault must have been underway and unprovoked, and under s. 37 the assault must be such that it is necessary to defend the person assaulted by the use of force. No more force may be used than necessary to prevent the assault or the repetition of it. In my opinion, no person has the right in anticipation of an assault that may or may not happen, to apply force to prevent the imaginary assault.

The implication of the Court's reasoning is that it is inherently unreasonable to apprehend death or grievous bodily harm unless and until the physical assault is actually in progress, at which point the victim can presumably gauge the requisite amount of force needed to repel the attack and act accordingly. In my view, expert testimony can cast doubt on these assumptions as they are applied in the context of a battered wife's efforts to repel an assault.

The situation of the appellant was not unlike that of Jane Stafford in the sense that she too was routinely beaten over the course of her relationship with the man she ultimately killed. According to the testimony of Dr. Shane these assaults were not entirely random in their occurrence. The following exchange during direct examination elicited a discernible pattern to the abuse:

Q. How did they react during the tension that preceded the beatings? How would her...

A. Well, typically before a beating there's usually some verbal interchange and there are threats and typically she would feel, you know, very threatened by him and for various reasons.

He didn't like the way she dressed or if she -- didn't like the way she handled money or she wasn't paying him enough attention or she was looking at other men, all sorts of reasons, and she would be defending herself, trying to placate him, which was typical, saying, you know, trying to calm him down, trying to soothe him, you know, so nothing violent would happen and sometimes it would work. You know, as people's experiences indicated or as people who write about this process, if you will, have indicated.

But often, as reflected by what she has told me, and the information I have from other people, such as her mother, often it would fail and she would end up being beaten and assaulted.

Q. And that would be followed by this forgiveness state?

A. It typically would be followed by, you know, this make-up period.

Earlier in his testimony Dr. Shane explained how this "make-up" period would be characterized by contrite and affectionate behaviour by Rust:

In this particular case she documented many times, after he would beat her, he would send her flowers and he would beg her for forgiveness and he would love her and then the relationship would come back to a sense of equilibrium, if you will. . . . But then, because of the nature of the personalities, it would occur again.

The cycle described by Dr. Shane conforms to the Walker Cycle Theory of Violence named for clinical psychologist Dr. Lenore Walker, the pioneer researcher in the field of the battered wife syndrome. Dr. Shane acknowledged his debt to Dr. Walker in the course of establishing his credentials as an expert at trial. Dr. Walker first describes the cycle in the book The Battered Woman, (1979). In her 1984 book, The Battered Woman Syndrome, Dr. Walker reports the results of a study involving 400 battered women. Her research was designed to test empirically the theories expounded in her earlier book. At pp. 95-96 of The Battered Woman Syndrome she summarizes the Cycle Theory as follows:

A second major theory that was tested in this project is the Walker Cycle Theory of Violence (Walker, 1979). This tension reduction theory states that there are three distinct phases associated in a recurring battering cycle: (1) tension building, (2) the acute battering incident, and (3) loving contrition. During the first phase, there is a gradual escalation of tension displayed by discrete acts causing increased friction such as name-calling, other mean intentional behaviors, and/or physical abuse. The batterer expresses dissatisfaction and hostility but not in an extreme or maximally explosive form. The woman attempts to placate the batterer, doing what she thinks might please him, calm him down, or at least, what will not further aggravate him. She tries not to respond to his hostile actions and uses general anger reduction techniques. Often she succeeds for a little while which reinforces her unrealistic belief that she can control this man ....

The tension continues to escalate and eventually she is unable to continue controlling his angry response pattern. "Exhausted from the constant stress, she usually withdraws from the batterer, fearing she will inadvertently set off an explosion. He begins to move more oppressively toward her as he observes her withdrawal. . . . Tension between the two becomes unbearable" (Walker, 1979, p. 59). The second phase, the acute battering incident, becomes inevitable without intervention. Sometimes, she precipitates the inevitable explosion so as to control where and when it occurs, allowing her to take better precautions to minimize her injuries and pain.

"Phase two is characterized by the uncontrollable discharge of the tensions that have built up during phase one" (p. 59). The batterer typically unleashes a barrage of verbal and physical aggression that can leave the woman severely shaken and injured. In fact, when injuries do occur it usually happens during this second phase. It is also the time police become involved, if they are called at all. The acute battering phase is concluded when the batterer stops, usually bringing with its cessation a sharp physiological reduction in tension. This in itself is naturally reinforcing. Violence often succeeds because it does work.

In phase three which follows, the batterer may apologize profusely, try to assist his victim, show kindness and remorse, and shower her with gifts and/or promises. The batterer himself may believe at this point that he will never allow himself to be violent again. The woman wants to believe the batterer and, early in the relationship at least, may renew her hope in his ability to change. This third phase provides the positive reinforcement for remaining in the relationship, for the woman. In fact, our results showed that phase three could also be characterized by an absence of tension or violence, and no observable loving-contrition behaviour, and still be reinforcing for the woman.

Dr. Walker defines a battered woman as a woman who has gone through the battering cycle at least twice. As she explains in her introduction to The Battered Woman, at p. xv, "Any woman may find herself in an abusive relationship with a man once. If it occurs a second time, and she remains in the situation, she is defined as a battered woman."

Given the relational context in which the violence occurs, the mental state of an accused at the critical moment she pulls the trigger cannot be understood except in terms of the cumulative effect of months or years of brutality. As Dr. Shane explained in his testimony, the deterioration of the relationship between the appellant and Rust in the period immediately preceding the killing led to feelings of escalating terror on the part of the appellant:

But their relationship some weeks to months before was definitely escalating in terms of tension and in terms of the discordant quality about it. They were sleeping in separate bedrooms. Their intimate relationship was lacking and things were building and building and to a point, I think, where it built to that particular point where she couldn't -- she felt so threatened and so overwhelmed that she had to -- that she reacted in a violent way because of her fear of survival and also because, I think because of her, I guess, final sense that she was -- that she had to defend herself and her own sense of violence towards this man who had really desecrated her and damaged her for so long.

Another aspect of the cyclical nature of the abuse is that it begets a degree of predictability to the violence that is absent in an isolated violent encounter between two strangers. This also means that it may in fact be possible for a battered spouse to accurately predict the onset of violence before the first blow is struck, even if an outsider to the relationship cannot. Indeed, it has been suggested that a battered woman's knowledge of her partner's violence is so heightened that she is able to anticipate the nature and extent (though not the onset) of the violence by his conduct beforehand. In her article "Potential Uses for Expert Testimony: Ideas Toward the Representation of Battered Women Who Kill" (1986), 9 Women's Rights Law Reporter 227, psychologist Julie Blackman describes this characteristic, at p. 229:

Repeated instances of violence enable battered women to develop a continuum along which they can "rate" the tolerability or survivability of episodes of their partner's violence. Thus, signs of unusual violence are detected. For battered women, this response to the ongoing violence of their situations is a survival skill. Research shows that battered women who kill experience remarkably severe and frequent violence relative to battered women who do not kill. They know what sorts of danger are familiar and which are novel. They have had myriad opportunities to develop and hone their perceptions of their partner's violence. And, importantly, they can say what made the final episode of violence different from the others: they can name the features of the last battering that enabled them to know that this episode would result in life-threatening action by the abuser.

At p. 236, Dr. Blackman relates the role of expert testimony in cases where a battered woman kills her batterer while he is sleeping (or not actively posing a threat her) and pleads self-defence:

Perhaps the single most important idea conveyed by expert testimony in such a case pertains to the notion that a battered woman, because of her extensive experience with her abuser's violence, can detect changes or signs of novelty in the pattern of normal violence that connote increased danger. Support for this assertion must come from the woman herself, in her spontaneous, self-initiated description of the events that precede her action against the abuser. Only then can testimony from an expert offer scientific support for the idea that such a danger detection process can occur and can be expected to be as accurate as the "reasonable man" standard would imply.

Of course, as Dr. Blackman points out, it is up to the jury to decide whether the distinction drawn between "typical" violence and the particular events the accused perceived as "life threatening" is compelling. According to the appellant's statement to police, Rust actually handed her a shotgun and warned her that if she did not kill him, he would kill her. I note in passing a remarkable observation made by Dr. Walker in her 1984 study The Battered Woman Syndrome. Writing about the fifty battered women she interviewed who had killed their partners, she comments at p. 40:

Most of the time the women killed the men with a gun; usually one of several that belonged to him. Many of the men actually dared or demanded the woman use the gun on him first, or else he said he'd kill her with it. [Emphasis added.]

Where evidence exists that an accused is in a battering relationship, expert testimony can assist the jury in determining whether the accused had a "reasonable" apprehension of death when she acted by explaining the heightened sensitivity of a battered woman to her partner's acts. Without such testimony I am skeptical that the average fact-finder would be capable of appreciating why her subjective fear may have been reasonable in the context of the relationship. After all, the hypothetical "reasonable man" observing only the final incident may have been unlikely to recognize the batterer's threat as potentially lethal. Using the case at bar as an example the "reasonable man" might have thought, as the majority of the Court of Appeal seemed to, that it was unlikely that Rust would make good on his threat to kill the appellant that night because they had guests staying overnight.

The issue is not, however, what an outsider would have reasonably perceived but what the accused reasonably perceived, given her situation and her experience.

Even accepting that a battered woman may be uniquely sensitized to danger from her batterer, it may yet be contended that the law ought to require her to wait until the knife is uplifted, the gun pointed or the fist clenched before her apprehension is deemed reasonable. This would allegedly reduce the risk that the woman is mistaken in her fear, although the law does not require her fear to be correct, only reasonable. In response to this contention, I need only point to the observation made by Huband J.A. that the evidence showed that when the appellant and Rust physically fought the appellant "invariably got the worst of it". I do not think it is an unwarranted generalization to say that due to their size, strength, socialization and lack of training, women are typically no match for men in hand-to-hand combat. The requirement imposed in Whynot that a battered woman wait until the physical assault is "underway" before her apprehensions can be validated in law would, in the words of an American court, be tantamount to sentencing her to `murder by installment': State v. Gallegos, 719 P.2d 1268 (N.M. 1986), at p. 1271. I share the view expressed by Willoughby in "Rendering Each Woman Her Due: Can a Battered Woman Claim Self-Defense When She Kills Her Sleeping Batterer" (1989), 38 Kan. L. Rev. 169, at p. 184, that "society gains nothing, except perhaps the additional risk that the battered woman will herself be killed, because she must wait until her abusive husband instigates another battering episode before she can justifiably act".

B. Lack of Alternatives to Self-Help

Section 34(2) requires an accused who pleads self-defence to believe "on reasonable grounds" that it is not possible to otherwise preserve him or herself from death or grievous bodily harm. The obvious question is if the violence was so intolerable, why did the appellant not leave her abuser long ago? This question does not really go to whether she had an alternative to killing the deceased at the critical moment. Rather, it plays on the popular myth already referred to that a woman who says she was battered yet stayed with her batterer was either not as badly beaten as she claimed or else she liked it. Nevertheless, to the extent that her failure to leave the abusive relationship earlier may be used in support of the proposition that she was free to leave at the final moment, expert testimony can provide useful insights. Dr. Shane attempted to explain in his testimony how and why, in the case at bar, the appellant remained with Rust:

She had stayed in this relationship, I think, because of the strange, almost unbelievable, but yet it happens, relationship that sometimes develops between people who develop this very disturbed, I think, very disturbed quality of a relationship. Trying to understand it, I think, isn't always easy and there's been a lot written about it recently, in the recent years, in psychiatric literature. But basically it involves two people who are involved in what appears to be an attachment which may have sexual or romantic or affectionate overtones.

And the one individual, and it's usually the women in our society, but there have been occasions where it's been reversed, but what happens is the spouse who becomes battered, if you will, stays in the relationship probably because of a number of reasons.

One is that the spouse gets beaten so badly -- so badly -- that he or she loses the motivation to react and becomes helpless and becomes powerless. And it's also been shown sometimes, you know, in -- not that you can compare animals to human beings, but in laboratories, what you do if you shock an animal, after a while it can't respond to a threat of its life. It becomes just helpless and lies there in an amotivational state, if you will, where it feels there's no power and there's no energy to do anything.

So in a sense it happens in human beings as well. It's almost like a concentration camp, if you will. You get paralyzed with fear.

The other thing that happens often in these types of relationships with human beings is that the person who beats or assaults, who batters, often tries -- he makes up and begs for forgiveness. And this individual, who basically has a very disturbed or damaged self-esteem, all of a sudden feels that he or she -- we'll use women in this case because it's so much more common -- the spouse feels that she again can do the spouse a favour and it can make her feel needed and boost her self-esteem for a while and make her feel worthwhile and the spouse says he'll forgive her and whatnot.

Apparently, another manifestation of this victimization is a reluctance to disclose to others the fact or extent of the beatings. For example, the hospital records indicate that on each occasion the appellant attended the emergency department to be treated for various injuries she explained the cause of those injuries as accidental. Both in its address to the jury and in its written submissions before this Court the Crown insisted that the appellant's injuries were as consistent with her explanations as with being battered and, therefore, in the words of Crown counsel at trial, "the myth is, in this particular case, that Miss Lavallee was a battered spouse". In his testimony Dr. Shane testified that the appellant admitted to him that she lied to hospital staff and others about the cause of her injuries. In Dr. Shane's opinion this was consistent with her overall feeling of being trapped and helpless:

... she would never say that she'd been abused by the man with whom she was living and that usually happened because of this whole process. He would beg her. I mean she would tell me that on occasions he would beat her and then the police would be called by, I think, on one occasion a neighbour and he got down on his knees and he begged forgiveness and he loved her and he felt so terrible about it. And so this would be a typical scenario. Whenever she would go to the hospital, that he would attempt to, I think, attempt to have her forgive him and he would love her so much more.

Again she would feel so needed and this would start the whole cycle over again.

And he would also blackmail her on occasions. She had an abortion when she was in the early part of their relationship and he would blackmail her saying, "You know, I will tell your parents that you were a baby killer", et cetera.

But basically the manner in which, I think, she would be prevented from telling the doctors or other people about the beatings was related to the fact that this whole process would repeat itself. He would want forgiveness and tell her he would love her and it would never happen again and she would feel grateful. She would feel a little loved. It would help her self-esteem again and she would feel a little safer for a while too. It would allow her to have a sense, a window of security for a period because she felt so trapped in this relationship.

The account given by Dr. Shane comports with that documented in the literature. Reference is often made to it as a condition of "learned helplessness", a phrase coined by Dr. Charles Seligman, the psychologist who first developed the theory by experimenting on animals in the manner described by Dr. Shane in his testimony. A related theory used to explain the failure of women to leave battering relationships is described by psychologist and lawyer Charles Patrick Ewing in his book Battered Women Who Kill (1987). Ewing describes a phenomenon labelled "traumatic bonding" that has been observed between hostages and captors, battered children and their parents, concentration camp prisoners and guards, and batterers and their spouses. According to the research cited by Ewing there are two features common to the social structure in each of these apparently diverse relationships. At pp. 19-20, he states:

The first of these common features is an imbalance of power "wherein the maltreated person perceives himself or herself to be subjugated or dominated by the other". The less powerful person in the relationship -- whether battered woman, hostage, abused child, cult follower, or prisoner -- becomes extremely dependent upon, and may even come to identify with, the more powerful person. In many cases, the result of such dependency and identification is that the less powerful, subjugated persons become "more negative in their self-appraisal, more incapable of fending for themselves, and thus more in need of the high power person." As this "cycle of dependency and lowered self-esteem" is repeated over time, the less powerful person develops a "strong affective bond" to the more powerful person in the abusive relationship.

The second feature common to the relationships between battered woman and batterer, hostage and captor, battered child and abusive parent, cult follower and leader, and prisoner and guard is the periodic nature of the abuse. In each relationship, the less powerful person is subjected to intermittent periods of abuse, which alternate with periods during which the more powerful, abusive person treats the less powerful person in a "more normal and acceptable fashion."

Given the clear power differential between battered women and their batterers and the intermittent nature of physical and psychological abuse common to battering relationships, it seems fair to conclude...that many battered women are psychologically unable to leave their batterers because they have developed a traumatic bond with them. [Citations omitted.]

This strong "affective bond" may be helpful in explaining not only why some battered women remain with their abusers but why they even profess to love them. Of course, as Dr. Ewing adds, environmental factors may also impair the woman's ability to leave -- lack of job skills, the presence of children to care for, fear of retaliation by the man, etc. may each have a role to play in some cases.

This is not to say that in the course of a battering relationship a woman may never attempt to leave her partner or try to defend herself from assault. In The Battered Woman Syndrome, supra, Dr. Walker notes, at p. 30, that women may sometimes "react to men's violence against them by striking back, but their actions are generally ineffective at hurting or stopping the men. They may be effective in controlling the level of the man's violence against them". In the case at bar Dr. Shane was aware that the appellant had pointed a gun at Rust in the past. In direct examination he stated:

And what would also happen from time to time is that there would be moments where she would attempt to hit back to defend herself or she may take a weapon to defend herself in order to prevent herself from being harmed or even, when the underlying rage may accumulate, if you will, the feeling that she had to do something to him in order to survive, in order to defend herself.

The same psychological factors that account for a woman's inability to leave a battering relationship may also help to explain why she did not attempt to escape at the moment she perceived her life to be in danger. The following extract from Dr. Shane's testimony on direct examination elucidates this point:

Q. Now, we understand from the evidence that on this night she went -- I think you've already described it in your evidence -- and hid in the closet?

A. Yes.

Q. Can you tell the jury why she, for instance, would stay in that house if she had this fear? Why wouldn't she so [sic] someplace else? Why would she have to hide in the closet in the same house?

A. Well, I think this is a reflection of what I've been talking about, this ongoing psychological process, her own psychology and the relationship, that she felt trapped. There was no out for her, this learned helplessness, if you will, the fact that she felt paralyzed, she felt tyrannized. She felt, although there were obviously no steel fences around, keeping her in, there were steel fences in her mind which created for her an incredible barrier psychologically that prevented her from moving out. Although she had attempted on occasion, she came back in a magnetic sort of a way. And she felt also that she couldn't expect anything more. Not only this learned helplessness about being beaten, beaten, where her motivation is taken away, but her whole sense of herself. She felt this victim mentality, this concentration camp mentality if you will, where she could not see herself be in any other situation except being tyrannized, punished and crucified physically and psychologically.

I emphasize at this juncture that it is not for the jury to pass judgment on the fact that an accused battered woman stayed in the relationship. Still less is it entitled to conclude that she forfeited her right to self-defence for having done so. I would also point out that traditional self-defence doctrine does not require a person to retreat from her home instead of defending herself: R. v. Antley (1963), 1963 CanLII 258 (ON CA), 42 C.R. 384 (Ont. C.A.) A man's home may be his castle but it is also the woman's home even if it seems to her more like a prison in the circumstances.

If, after hearing the evidence (including the expert testimony), the jury is satisfied that the accused had a reasonable apprehension of death or grievous bodily harm and felt incapable of escape, it must ask itself what the "reasonable person" would do in such a situation. The situation of the battered woman as described by Dr. Shane strikes me as somewhat analogous to that of a hostage. If the captor tells her that he will kill her in three days time, is it potentially reasonable for her to seize an opportunity presented on the first day to kill the captor or must she wait until he makes the attempt on the third day? I think the question the jury must ask itself is whether, given the history, circumstances and perceptions of the appellant, her belief that she could not preserve herself from being killed by Rust that night except by killing him first was reasonable. To the extent that expert evidence can assist the jury in making that determination, I would find such testimony to be both relevant and necessary.

In light of the foregoing discussion I would summarize as follows the principles upon which expert testimony is properly admitted in cases such as this:

1. Expert testimony is admissible to assist the fact-finder in drawing inferences in areas where the expert has relevant knowledge or experience beyond that of the lay person.

2. It is difficult for the lay person to comprehend the batter wife syndrome. It is commonly thought that battered women are not really beaten as badly as they claim, otherwise they would have left the relationship. Alternatively, some believe that women enjoy being beaten, that they have a masochist strain in them. Each of these stereotypes may adversely affect consideration of a battered woman's claim to have acted in self-defence in killing her mate.

3. Expert evidence can assist the jury in dispelling these myths.

4. Expert testimony relating to the ability of an accused to perceive danger from her mate may go to the issue of whether she "reasonably apprehended" death or grievous bodily harm on a particular occasion.

5. Expert testimony pertaining to why an accused remained in the battering relationship may be relevant in assessing the nature and extent of the alleged abuse.

6. By providing an explanation as to why an accused did not flee when she perceived her life to be in danger, expert testimony may also assist the jury in assessing the reasonableness of her belief that killing her batterer was the only way to save her own life.

Quite apart from Dr. Shane's testimony there was ample evidence on which the trial judge could conclude that the appellant was battered repeatedly and brutally by Kevin Rust over the course of their relationship. The fact that she may have exhibited aggressive behaviour on occasion or tried (unsuccessfully) to leave does not detract from a finding of systematic and relentless abuse. In my view, the trial judge did not err in admitting Dr. Shane's expert testimony in order to assist the jury in determining whether the appellant had a reasonable apprehension of death or grievous bodily harm and believed on reasonable grounds that she had no alternative but to shoot Kevin Rust on the night in question.

Obviously the fact that the appellant was a battered woman does not entitle her to an acquittal. Battered women may well kill their partners other than in self-defence. The focus is not on who the woman is, but on what she did. In "The Meaning of Equality for Battered Women Who Kill Men in Self-Defense" (1985), 8 Harv. Women's L.J. 121, at p. 149, Phyllis Crocker makes the point succinctly:

The issue in a self-defence trial is not whether the defendant is a battered woman, but whether she justifiably killed her husband. The defendant introduces testimony to offer the jury an explanation of reasonableness that is an alternative to the prosecution's stereotypic explanations. It is not intended to earn her the status of a battered woman, as if that would make her not guilty.

The trial judge, to his credit, articulated the same principle when introducing Dr. Shane's testimony in the course of his instructions to the jury. After referring to "the so-called battered spouse syndrome", he cautions:

Let me say at the outset that I think it is better that we try not to attach labels to this. It doesn't matter what we call it. What is important is the evidence itself and how it impacts on the critical areas of the intent of the accused and the issue of self-defence.

Ultimately, it is up to the jury to decide whether, in fact, the accused's perceptions and actions were reasonable. Expert evidence does not and cannot usurp that function of the jury. The jury is not compelled to accept the opinions proffered by the expert about the effects of battering on the mental state of victims generally or on the mental state of the accused in particular. But fairness and the integrity of the trial process demand that the jury have the opportunity to hear them.

(iii) Adequacy of Trial Judge's Charge to the Jury

The second issue raised in this case is the adequacy of the trial judge's charge to the jury with respect to the expert evidence furnished by Dr. Shane. It appears that Dr. Shane relied on various sources in formulating his opinion -- his series of interviews with the appellant, an interview with her mother, a police report of the incident (including information regarding her statement to the police), and hospital records documenting eight of her visits to emergency departments between 1983 and 1986. Neither the appellant nor her mother testified at trial. The contents of their statements to Dr. Shane were hearsay.

In Abbey, supra, this Court addressed the bases upon which expert evidence that relies on hearsay is admissible. The accused in that case was charged with importing cocaine and his defence was insanity. The accused did not testify. A psychiatrist gave his opinion as to the sanity of the accused and, in the course of giving the basis for his conclusions, referred to incidents and hallucinations related to him by the accused for which there was no admissible evidence. The Crown submitted before this Court that the trial judge "accepted and treated as factual much of this hearsay evidence" related to the psychiatrist. Dickson J. found that the point was "well taken". This was the preliminary finding on which the case was based and I think it is fair to say that the trial judge in the case at bar clearly did not make the same mistake as did the trial judge in Abbey. At pp. 44-46 of his judgment, Dickson J. articulated the hazards inherent in admitting expert testimony based on hearsay:

The danger, of course, in admitting such testimony is the ever present possibility, here exemplified, that the judge or jury, without more, will accept the evidence as going to the truth of the facts stated in it. The danger is real and lies at the heart of this case. Once such testimony is admitted, a careful charge to the jury by the judge or direction to himself is essential. The problem, however, as pointed out by Fauteux J. in Wilband resides not in the admissibility of the testimony but rather the weight to be accorded to the opinion. Although admissible in the context of his opinion, to the extent that it is second-hand his testimony is not proof of the facts stated.

It was appropriate for the doctors to state the basis for their opinions and in the course of doing so, to refer to what they were told not only by Abbey but by others, but it was error for the judge to accept as having been proved the facts upon which the doctors had relied in forming their opinions. While it is not questioned that medical experts are entitled to take into consideration all possible information in forming their opinions, this in no way removes from the party tendering such evidence the obligation of establishing, through properly admissible evidence, the factual basis on which such opinions are based. Before any weight can be given to an expert's opinion, the facts upon which the opinion is based must be found to exist.

For present purposes I think the ratio of Abbey can be distilled into the following propositions:

1. An expert opinion is admissible if relevant, even if it is based on second-hand evidence.

2. This second-hand evidence (hearsay) is admissible to show the information on which the expert opinion is based, not as evidence going to the existence of the facts on which the opinion is based.

3. Where the psychiatric evidence is comprised of hearsay evidence, the problem is the weight to be attributed to the opinion.

4. Before any weight can be given to an expert's opinion, the facts upon which the opinion is based must be found to exist.

In the case at bar the trial judge was clearly of the view that Dr. Shane's evidence was relevant. He would not have admitted it otherwise. As I stated above, in light of the evidence of the battering relationship which subsisted between the appellant and the deceased, the trial judge was correct in so doing.

With respect to the second point, the trial judge warned the jury generally that they could not "decide the case on the basis of things the witnesses did not see or hear," which would seem to include those matters which Dr. Shane neither saw nor heard. He then gave the marijuana smoking and the confirmatory evidence of the appellant's mother as two sources of information which were not evidence in the case. In my opinion, it would have been preferable if the trial judge had described the interview with the appellant as a source of inadmissible evidence, the marijuana smoking being an example of inadmissible evidence from that source. Nevertheless, I think the trial judge makes his meaning clear to the jury in the subsequent passage:

In terms of the matters considered by Dr. Shane he is left, therefore, with the deceased's [sic - he means accused's] statement, some supplementary information from the police report and his interpretation of the hospital records.

The trial judge thus eliminates the interview with the appellant and his conversation with her mother as sources of admissible evidence. Elsewhere he reinforces the rule that the jury can only consider the admissible evidence. He refers to the hospital visits made by the appellant:

Another evidentiary caution is necessary here. Mr. Brodsky, in his remarks, said, as he did in calling some of the evidence respecting hospital attendances that this is only a representative sample. He ought not to have said that. It is not evidence and must be completely disregarded by you. The only evidence before you are the eight attendances that you heard about and nothing else -- eight attendances and nothing else.

The trial judge's instructions regarding the weight attributable to Dr. Shane's opinion also emphasize his distinction between admissible evidence and hearsay:

If the premises upon which the information is substantially based has not been proven in evidence, it is up to you to conclude that it is not safe to attach a great deal of weight to the opinion. An opinion of an expert depends, to a large extent, on the validity of the facts assumed by the evidence of the expert.

If there are some errors and the factual assumptions aren't too important to the eventual opinion, that's one thing. If there are errors or matters not in evidence and those matters are substantial, in you view, in terms of the impact on the expert's opinion, then you will want to look at the value and weight of that expert's opinion very carefully. It depends on how important you think the matters were that Dr. Shane relied on that are not in evidence. [Emphasis added.]

I agree with Huband J.A. that these instructions with respect to weight conform to this Court's judgment in Abbey. The only complaint can be with the trial judge's attempt to distinguish admissible from inadmissible evidence. The trial judge was certainly not as clear as he might have been but I have no hesitation in finding that a retrial is not warranted on this account.

Given that Dr. Shane relied extensively on his interview with the appellant, the trial judge drew particular attention to the additional element of credibility that could affect the quality of Dr. Shane's opinion:

It is the position of the Crown that Dr. Shane's opinion stands or falls on the veracity of Lyn Lavallee because he relied so heavily and extensively on what she told him and the evidence contained in the statement, Exhibit 16. That's for you to decide.

Later in the charge, he elaborates:

Undoubtably [sic] she was a very important source, if not the major source, of his information. Dr. Shane agreed that if what she told him was erroneous, he would have to reassess his position.

On cross-examination he reiterated that in his opinion her action was spontaneous to the moment to try to defend herself. The straw that broke the camel's back was the threat, "When the others leave you're going to get it", even though similar statements had been made to her on other occasions. According to what she told him, the accused felt compelled to shoot.

Based on the information he had in the interview, it was his opinion that the acts of the accused were impulsive and not premeditated. He disagreed with the Crown's suggestion that Lyn Lavallee took the opportunity when it presented itself.

He conceded that patients had, on occasion, lied and misled him in the past.

The fourth proposition I have extracted from Abbey is that there must be admissible evidence to support the facts on which the expert relies before any weight can be attributed to the opinion. The majority of the Manitoba Court of Appeal appears to interpret this as a requirement that each and every fact relied upon by the expert must be independently proven and admitted into evidence before the entire opinion can be given any weight.

Dr. Shane referred in his testimony to various facts for which there was no admissible evidence. The information was elicited from his interviews with the appellant. It included the smoking of marijuana prior to the killing, the deterioration of the intimate relationship between the appellant and Rust, past episodes of physical and psychological abuse followed by intervals of contrition, the apparent denial of homicidal fantasies on the appellant's part, and her remorse after killing Rust.

If the majority of the Court of Appeal is suggesting that each of these specific facts must be proven in evidence before any weight could be given to Dr. Shane's opinion about the accused's mental state, I must respectfully disagree. Abbey does not, in my view, provide any authority for that proposition. The Court's conclusion in that case was that the trial judge erred in treating as proven the facts upon which the psychiatrist relied in formulating his opinion. The solution was an appropriate charge to the jury, not an effective withdrawal of the evidence. In my view, as long as there is some admissible evidence to establish the foundation for the expert's opinion, the trial judge cannot subsequently instruct the jury to completely ignore the testimony. The judge must, of course, warn the jury that the more the expert relies on facts not proved in evidence the less weight the jury may attribute to the opinion.

On my reading of the record Dr. Shane had before him admissible evidence about the nature of the relationship between the appellant and Rust in the form of the appellant's statement to the police and the hospital records. In addition, there was substantial corroborative evidence provided at trial by Ezako, the emergency room doctor who testified to doubting the appellant's explanation of her injuries. There was also the evidence of the witnesses on the night of the shooting who testified to the appellant's frightened appearance, tone of voice, and conduct in dealing with Rust. The evidence pointed to the image of a woman who was brutally abused, who lied about the cause of her injuries, and who was incapable of leaving her abuser. As Huband J.A. comments in dissent, if the trial judge erred at all, he was probably remiss in not mentioning the corroborative evidence of Ezako as buttressing the evidentiary foundation on which Dr. Shane premised his opinion.

The majority of the Court of Appeal attached particular significance to the absence of admissible evidence on the question whether the appellant had homicidal fantasies about Rust. As I read the evidence the appellant's alleged denial of homicidal fantasies appeared to add little to Dr. Shane's overall opinion about her mental state on the night in question. Moreover, the evidence given by Ezako about her being an aggressor in the past and even pointing a gun at Rust were far more incriminating in terms of evincing a prior intent to kill than the presence or absence of homicidal fantasies. The gun pointing incidents were explained by Dr. Shane as not inconsistent with her victimized condition and not necessarily indicative of pre-meditation. Clearly, Dr. Shane's explanation was something the jury could evaluate in the context of all the evidence.

Where the factual basis of an expert's opinion is a melange of admissible and inadmissible evidence the duty of the trial judge is to caution the jury that the weight attributable to the expert testimony is directly related to the amount and quality of admissible evidence on which it relies. The trial judge openly acknowledged to counsel the inherent difficulty in discharging such a duty in the case at bar. In my view, the trial judge performed his task adequately in this regard. A new trial is not warranted on the basis of the trial judge's charge to the jury.

I would accordingly allow the appeal, set aside the order of the Court of Appeal, and restore the acquittal.

//Sopinka J.//

The following are the reasons delivered by

Sopinka J. -- I have read the reasons of my colleague Justice Wilson, and I agree in the result that this appeal must be allowed. I find it necessary, however, to add a few words concerning the interpretation of this Court's decision in R. v. Abbey, 1982 CanLII 25 (SCC), [1982] 2 S.C.R. 24.

Abbey has been roundly criticized: see, e.g., Schiff, Evidence in the Litigation Process, vol. 1 (3rd ed. 1988), at pp. 473-76; and Delisle, Evidence: Principles and Problems (2nd ed. 1989), at pp. 477-79. The essence of the criticism is that Abbey sets out more restrictive conditions for the use of expert evidence than did previous decisions of this Court (i.e., City of St. John v. Irving Oil Co., 1966 CanLII 64 (SCC), [1966] S.C.R. 581; Wilband v. The Queen, 1966 CanLII 3 (SCC), [1967] S.C.R. 14; and R. v. Lupien, 1969 CanLII 55 (SCC), [1970] S.C.R. 263). Upon reflection, it seems to me that the very special facts in Abbey, and the decision required on those facts, have contributed to the development of a principle concerning the admissibility and weight of expert opinion evidence that is self-contradictory. The contradiction is apparent in the four principles set out by Wilson J. in the present case, at pp. 000, which I reproduce here for the sake of convenience:

1. An expert opinion is admissible if relevant, even if it is based on second-hand evidence.

2. This second-hand evidence (hearsay) is admissible to show the information on which the expert opinion is based, not as evidence going to the existence of the facts on which the opinion is based.

3. Where the psychiatric evidence is comprised of hearsay evidence, the problem is the weight to be attributed to the opinion.

4. Before any weight can be given to an expert's opinion, the facts upon which the opinion is based must be found to exist.

The combined effect of numbers 1, 3 and 4 is that an expert opinion relevant in the abstract to a material issue in a trial but based entirely on unproven hearsay (e.g., from the mouth of the accused, as in Abbey) is admissible but entitled to no weight whatsoever. The question that arises is how any evidence can be admissible and yet entitled to no weight. As one commentator has pointed out, an expert opinion based entirely on unproven hearsay must, if anything, be inadmissible by reason of irrelevance, since the facts underlying the expert opinion are the only connection between the opinion and the case: see Wardle, "R. v. Abbey and Psychiatric Opinion Evidence: Requiring the Accused to Testify" (1984), 17 Ottawa L. Rev. 116, at pp. 122-23.

The resolution of the contradiction inherent in Abbey, and the answer to the criticism Abbey has drawn, is to be found in the practical distinction between evidence that an expert obtains and acts upon within the scope of his or her expertise (as in City of St. John), and evidence that an expert obtains from a party to litigation touching a matter directly in issue (as in Abbey).

In the former instance, an expert arrives at an opinion on the basis of forms of enquiry and practice that are accepted means of decision within that expertise. A physician, for example, daily determines questions of immense importance on the basis of the observations of colleagues, often in the form of second- or third-hand hearsay. For a court to accord no weight to, or to exclude, this sort of professional judgment, arrived at in accordance with sound medical practices, would be to ignore the strong circumstantial guarantees of trustworthiness that surround it, and would be, in my view, contrary to the approach this Court has taken to the analysis of hearsay evidence in general, exemplified in Ares v. Venner, 1970 CanLII 5 (SCC), [1970] S.C.R. 608. In R. v. Jordan (1984), 1984 CanLII 635 (BC CA), 39 C.R. (3d) 50 (B.C.C.A.), a case concerning an expert's evaluation of the chemical composition of an alleged heroin specimen, Anderson J.A. held, and I respectfully agree, that Abbey does not apply in such circumstances. (See also R. v. Zundel (1987), 1987 CanLII 121 (ON CA), 56 C.R. (3d) 1 (Ont. C.A.), at p. 52, where the court recognized an expert opinion based upon evidence "... of a general nature which is widely used and acknowledged as reliable by experts in that field.")

Where, however, the information upon which an expert forms his or her opinion comes from the mouth of a party to the litigation, or from any other source that is inherently suspect, a court ought to require independent proof of that information. The lack of such proof will, consistent with Abbey, have a direct effect on the weight to be given to the opinion, perhaps to the vanishing point. But it must be recognized that it will only be very rarely that an expert's opinion is entirely based upon such information, with no independent proof of any of it. Where an expert's opinion is based in part upon suspect information and in part upon either admitted facts or facts sought to be proved, the matter is purely one of weight. In this respect, I agree with the statement of Wilson J. at p. 000, as applied to circumstances such as those in the present case:

... as long as there is some admissible evidence to establish the foundation for the expert's opinion, the trial judge cannot subsequently instruct the jury to completely ignore the testimony. The judge must, of course, warn the jury that the more the expert relies on facts not proved in evidence the less weight the jury may attribute to the opinion.

As Wilson J. holds, the trial judge's charge to the jury was adequate, and the appeal ought therefore to be allowed.

Appeal allowed.

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

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

^*^ See Erratum [2009] 1 S.C.R. iv.