Her Majesty the Queen v. Bennett [Indexed as: R. v. Bennett]

67 O.R. (3d) 257

[2003] O.J. No. 3810

Docket No. C36225

Court of Appeal for Ontario

McMurtry C.J.O., Armstrong J.A. and Blair R.S.J. (ad hoc)

October 3, 2003

*Application for an extension of time granted and application for leave to appeal dismissed April 1, 2004 (Bastarache, LeBel and Deschamps JJ.).

Criminal law -- Charge to jury -- After-the-fact conduct -- Consciousness of guilt -- Accused giving two statements to police about his activities and whereabouts at time of victim's death -- Minor inconsistencies between statements -- Trial judge instructing jury to consider whether there were material inconsistencies whether they were deliberate or innocent -- Trial judge instructing jury that if they were satisfied that accused concocted alibi they were entitled to draw inference of guilt -- Trial judge erring in not identifying inconsistencies that could warrant finding of concoction -- Slight difference in statements not capable of supporting finding of concoction -- Trial judge erring in leaving evidence of accused's demeanour with jury as evidence capable of supporting inference of guilt -- Trial judge failing to give proper limiting instruction with respect to items of after-the-fact conduct where were admissible -- Crown's case entirely circumstantial -- After-the-fact conduct constituting critical aspect of Crown's case -- Application of curative proviso inappropriate -- New trial ordered.

Criminal law -- Evidence -- Identification evidence -- Crown witness testifying that murder victim appeared to be afraid on day before her death of man standing across street -- Witness identifying man as accused and describing him as being clean- shaven -- Accused having beard at relevant time -- Trial judge erring in leaving it open to jury to conclude that accused was person witness saw across street -- Notable dissimilarity in identification evidence, absent some other inculpatory evidence, rendering identification evidence of resemblance of no probative value -- Trial judge should have instructed jury to that effect and also instructed them that it was open to them to infer from witness' evidence that he had seen another man standing across street on day before victim's death and that victim appeared to be afraid of that man -- Appeal from conviction allowed -- New trial ordered.

Criminal law -- Evidence -- Physical evidence -- Hair evidence -- Trial judge erring in admitting evidence of hairs from murder scene that could not be eliminated as having come from accused -- Prejudicial effect of evidence outweighing its probative value -- Crown expert clearly expressing opinion that evidence of type 2 hair comparison, without confirming DNA evidence, is not probative of identification of donor of hair -- Potential for prejudice dramatically increased by fact that Crown urged jury in closing address to draw inferences unsupported by expert's testimony based on "common sense approach" to hair evidence -- Crown misleadingly stating that type 2 hairs were "all DNA'd" to accused -- Trial judge's instruction to jury that Crown's suggested inference "may have gone beyond" expert's opinion significantly understated unfairness of Crown's treatment of hair evidence.

The accused was charged with first degree murder. The case against him was entirely circumstantial. It included the accused's expressed sexual interest in the victim; scalp and pubic hairs, some of which revealed a DNA match to the accused [page258] and some of which showed microscopic similarities to his hairs, were found on the victim's bedclothes and body; various hairs and hair fragments found on the victim's bedclothes and body which had so-called "Negroid" characteristics (the accused was the only black suspect); a trace of semen found in the victim's vagina; the body was nude; she had suffered multiple, superficial cuts to her neck and significant bruising to her upper body; the accused knew how to access her apartment; and the accused's alleged inculpatory after-the-fact conduct. The hair evidence was complex. Two hundred and ninety two hairs were recovered from the victim's apartment. An expert witness, C, concluded that 13 hairs found in the victim's apartment showed different levels of microscopic similarity to the accused's known hair samples. Four hairs (type 1) were microscopically similar to the accused's hair and could have originated from him. Nine hairs (type 2) which shared some but not all of the microscopic characteristics of the accused's hair could not be eliminated as having originated from him. C also identified six "C" hairs, i.e. short hair fragments or limb and immature hairs that were not useful for identification purposes, but which he testified had "Negroid" characteristics. On the nine hairs submitted for DNA analysis, seven matched the accused. The accused was convicted. He appealed.

Held, the appeal should be allowed.

The trial judge erred in admitting the type 2 hair comparison evidence. The prejudicial effect of that evidence outweighed its probative value. The theory of the defence was that the accused's hairs had either been deposited directly on the victim's bedding when he sat on the bed the week before the death, or were innocently transferred onto the victim's body and bedding when, as the forensic evidence indicated, the body was moved after death. The strength of the defence theory would be reduced as the number of hairs linked to the accused and found on the bedding and body increased. C clearly expressed the opinion that evidence of a type 2 hair comparison, without confirming DNA evidence, was not probative of identification of the donor of the hair. His additional evidence to the effect that taking together the hairs that showed a type 1 and type 2 comparison to the accused's known hair samples reduced by an immeasurable amount the size of the indeterminate class who could be a donor of the type 2 hairs did not elevate the probative value of the type 2 evidence that the defence sought to exclude. Moreover, C testified that he did not compare the type 2 hairs, one with the other, so as to form any opinion as to their similarities. Hairs are not unique, and an assessment of similarities between hairs is highly subjective. The trial judge also erred in admitting the C hair evidence, which had extremely low probative value and significant potential for prejudice given the testimony regarding their racial characteristics.

The potential for prejudice from the admission of the type 2 hair comparison evidence was dramatically increased by the fact that Crown counsel in his closing address asked the jury to draw inferences unsupported by C's testimony based on a "common sense approach" to the hair evidence. Even more problematic was the fact that Crown counsel misleadingly told the jury that the type 2 hairs were, with one exception, "all DNA'd to" the accused. The trial judge instructed the jury that some of the Crown's suggested inferences "may have gone beyond [C's] opinion". This instruction significantly understated the unfairness of Crown counsel's treatment of the hair evidence.

M, a Crown witness with an extensive criminal record and a history of drug abuse, testified at trial that on the day before her death, the victim appeared to be frightened of someone, and that he observed a man leaning against a post across the street. He identified the accused as that man. He also described the man as [page259] clean shaven. At the relevant time, the accused had a beard. A notable dissimilarity in identification evidence, absent some other inculpatory evidence, renders identification evidence of a resemblance of no probative value. The trial judge erred in leaving it open to the jury to conclude that the accused was the person M saw across the street the day before the victim was killed. The trial judge ought to have instructed the jury that M's identification evidence of the accused was worthless, not that it was "worth very, very little". Further, the trial judge ought to have instructed the jury that it was open to them to infer from M's evidence that he had seen a clean-shaven black man, and not the accused, standing across the street on the day before the killing and that the victim appeared to be afraid of that man.

After the victim's body was discovered, the accused gave two statements to the police accounting for his activities and whereabouts in the days surrounding the victim's death. There were some minor inconsistencies between the two statements. The trial judge erred in instructing the jury that based on these inconsistencies the accused had fabricated an alibi and that it was open to them to infer guilt from having concocted an alibi. The trial judge left it for the jury to determine whether there was any inconsistency and, if so, whether it was material. If an instruction permitting a jury to find concoction and to infer guilt based on that finding is to be given, a trial judge is required to identify for the jury what inconsistency in the accused's statements could warrant a finding of concoction. The inconsistency must be compelling in the sense, for example, that there is an indication that the accused was attempting to mislead investigators by fabricating an alibi. The slight differences between the accused's statements were not capable of warranting a finding of concoction. While the trial judge could have told the jury that any differences in the two statements was a factor they could take into account in deciding whether to believe either of them, there was no basis for instructing them that they could go further and use the statements as a separate piece of circumstantial evidence from which guilt could be inferred. Rather, the jury ought to have been told that if they disbelieved the alibi, they should simply discard that evidence, without more.

There were six other items of after-the-fact conduct that the trial judge left with the jury as evidence from which the Crown asked them to infer guilt. Those items of evidence were: the accused's anger at being described as someone capable of killing the victim; his failure to make certain phone calls upon learning of the victim's death; his failure to ask police the identity of the victim; the accused telling a friend of the victim's that the victim was seen going to a mall on the afternoon of her death; his desire to acquire copies of statements made to police by two of his acquaintances; and his reaction to the report that police found a silver letter opener that was missing from the victim's apartment. The first three items of evidence should not have been left with the jury as evidence from which they could infer consciousness of guilt. These forms of conduct were examples of demeanour evidence that is highly suspect and easily misinterpreted. As for the other three items, the trial judge erred in not instructing the jury that the after-the-fact conduct evidence relied on by the Crown had only an indirect bearing upon the issue of guilt, and that the jury should exercise caution in inferring guilt because the conduct might be explained in an alternative manner. In addition, he ought to have instructed the jury that they could not use this conduct to support an inference of guilt unless they rejected any innocent explanation for the conduct. This was not an appropriate case for the application of the curative proviso in s. 686(1)(b)(iii) of the Criminal Code, R.S.C. 1985, c. C-46. The Crown's case was entirely circumstantial, and the after-the-fact conduct evidence was a critical aspect of that case. In light of the trial judge's errors, and also in light of the [page260] errors in the hair comparison evidence and the identification evidence, it could not be said that the result of the trial would necessarily have been the same.

APPEAL by an accused from a conviction for first degree murder.

R. v. Baltrusaitis (2002), 2002 36440 (ON CA), 58 O.R. (3d) 161, 162 C.C.C. (3d) 539 (C.A.); R. v. Boucher (2000), 2000 3270 (ON CA), 146 C.C.C. (3d) 52, [2000] O.J. No. 2373 (QL) (C.A.); R. v. Diu (2000), 2000 4535 (ON CA), 49 O.R. (3d) 40, 144 C.C.C. (3d) 481, 33 C.R. (5th) 203 (C.A.); R. v. Levert (2001), 2001 8606 (ON CA), 159 C.C.C. (3d) 71 (Ont. C.A.), apld Other cases referred to Chartier v. Quebec (Attorney General), 1979 17 (SCC), [1979] 2 S.C.R. 474, 104 D.L.R. (3d) 321, 27 N.R. 1, 48 C.C.C. (2d) 34, 9 C.R. (3d) 97; R. v. Andrade (1985), 1985 3502 (ON CA), 6 O.A.C. 345, 18 C.C.C. (3d) 41 (C.A.); R. v. Blazeiko (2000), 2000 14726 (ON CA), 48 O.R. (3d) 652, 145 C.C.C. (3d) 557 (C.A.); R. v. Campbell (1999), 1999 2688 (ON CA), 139 C.C.C. (3d) 258 (Ont. C.A.); R. v. Charemski, 1998 819 (SCC), [1998] 1 S.C.R. 679, 157 D.L.R. (4th) 603, 224 N.R. 120, 123 C.C.C. (3d) 225, 15 C.R. (5th) 1; R. v. Coutts (1998), 1998 4212 (ON CA), 40 O.R. (3d) 198, 126 C.C.C. (3d) 545, 16 C.R. (5th) 240 (C.A.) [Leave to appeal to S.C.C. refused (1999), 239 N.R. 193n, [1998] S.C.C.A. No. 450]; R. v. Hibbert, [2002] 2 S.C.R. 445, 211 D.L.R. (4th) 223, 287 N.R. 111, 163 C.C.C. (3d) 129, 50 C.R. (5th) 209, 2002 SCC 39, [2002] S.C.J. No. 40 (QL); R. v. Krishantharajah (1999), 1999 3701 (ON CA), 43 O.R. (3d) 663, 133 C.C.C. (3d) 157 (C.A.); R. v. McInnis (1999), 1999 2671 (ON CA), 44 O.R. (3d) 772, 134 C.C.C. (3d) 515 (C.A.); R. v. O'Connor (2002), 2002 3540 (ON CA), 62 O.R. (3d) 263, 100 C.R.R. (2d) 164, 170 C.C.C. (3d) 365, 7 C.R. (6th) 205, [2002] O.J. No. 4410 (C.A.); R. v. Portillo (2003), 2003 5709 (ON CA), 176 C.C.C. (3d) 467, [2003] O.J. No. 3030 (QL) (C.A.); R. v. Price (2000), 2000 5679 (ON CA), 72 C.R.R. (2d) 228, 144 C.C.C. (3d) 343, 33 C.R. (5th) 278 (Ont. C.A.); R. v. Ruddick (1980), 1980 2941 (ON CA), 57 C.C.C. (2d) 421 (Ont. C.A.); R. v. Wristen (1999), 1999 3824 (ON CA), 47 O.R. (3d) 95, 141 C.C.C. (3d) 1 (C.A.) [Leave to appeal to S.C.C. refused [2000] S.C.C.A. No. 419] Statutes referred to Criminal Code, R.S.C. 1985, c. C-46, ss. 231(5), 686(1)(b)(iii) Authorities referred to Report of the Commission on Proceedings Involving Guy Paul Morin (Toronto: Queen's Printer for Ontario, 1998) (Hon. F. Kaufman, C.M., Q.C.)

Delmar Doucette and Catherine Glaister, for appellant. David Finley, for respondent.

The judgment of the court was delivered by

MCMURTRY C.J.O.: --

OVERVIEW

[1] The appellant appeals from his conviction for first degree murder following a trial by judge and jury. The conviction relates to the killing of Jennifer Ueberschlag, who was found dead in her apartment on Sunday, May 10, 1992. The appellant was arrested for first degree murder five years after the killing, on May 13, 1997. He was found guilty of the charge and [page261] sentenced to life in prison with no eligibility for parole for 25 years on December 11, 1999.

[2] The Crown's case against the appellant was entirely circumstantial. It included certain expert hair comparison evidence that, in my view, the trial judge erred in admitting. In addition, there were errors in the trial judge's charge to the jury on the use that they could make of certain identification evidence and after-the-fact conduct evidence, including two statements that the appellant made to police. I am satisfied that the cumulative effect of these errors is that the conviction cannot stand and a new trial on the charge of first degree murder is required.

[3] The theory of the Crown in this case was that the appellant wanted a sexual relationship with the 18-year-old victim, whom he had met a few weeks before her death. This motivation led him to her apartment on the evening of Friday, May 8, 1992, where he asphyxiated her by shoving a cat toy in her mouth to suppress her screams. The killing occurred while he was committing or attempting to commit a sexual assault or a forcible confinement or both. The Crown's position was that the killing took place at around midnight on May 8th.

[4] The circumstantial evidence upon which the Crown's theory rested included the following areas of evidence: the appellant's expressed sexual interest in the victim; scalp and pubic hairs, some of which revealed a DNA match to the appellant and some of which showed microscopic similarities to his hairs, found on the victim's bedclothes and body; various hairs and hair fragments found on the victim's bedclothes and body had so-called "Negroid" characteristics and the appellant was the only black suspect; a trace of semen was found in the vagina (the sample was too small for DNA analysis); the body was nude; the victim had suffered multiple, superficial cuts to her neck and significant bruising to the upper body; the appellant knew how to access the victim's apartment; and the appellant's alleged inculpatory after-the-fact conduct.

[5] The appellant did not testify and the defence did not call any evidence. Defence counsel was unsuccessful on his motion for a directed verdict of acquittal and on his alternative motion to have the first degree murder charge dismissed and the case proceed on a charge of second degree murder.

[6] In his closing address to the jury, defence counsel did not deny that there was evidence that the appellant, who was age 33 at the time of the murder, had expressed sexual attraction for the victim. Nor was it denied that six of his pubic hairs were found on the bedclothes, the body and the body bag in which the victim's body was transported. However, the defence denied a [page262] motive to use violence to satisfy the appellant's sexual interest and took the position that the presence of his hairs could be explained by a theory of innocent transfer.

[7] Defence counsel argued that the jury could not find beyond a reasonable doubt that it was the appellant who killed Ms. Ueberschlag. Counsel further argued that the jury could not convict on the charge of first degree murder because there was not sufficient evidence to establish that the victim was sexually assaulted or unlawfully confined. The defence alternatively argued that death occurred without the intent to kill and that the only possible offence upon which the jury could make a finding of guilt was manslaughter.

[8] Because of my view that a new trial is required, I will limit my discussion of the facts to those necessary to appreciate the six grounds of appeal pursued by appellant's counsel. These grounds are as follows:

(i) the trial judge erred in admitting trace evidence from the scene of the killing that was not probative and was prejudicial;

(ii) the trial judge erred in his charge to the jury by failing to instruct the jury that they must not draw the speculative and prejudicial inference that the Crown invited them to draw with respect to the trace evidence;

(iii) the trial judge erred in his charge in regard to identification evidence and erroneously allowed the jury to consider as evidence against the appellant an identification which excluded the appellant from being the person who was seen following the victim on the day before she was killed;

(iv) the trial judge erred in his charge in regard to after- the-fact conduct by erroneously allowing the jury to use after-the-fact conduct evidence against the appellant that was prejudicial and had little or no probative value;

(v) any verdict of culpable homicide is unreasonable; and

(vi) the verdict of first degree murder is unreasonable.

RELEVANT FACTS

[9] In the spring of 1992, Jennifer Ueberschlag lived alone in an apartment at 152 Homewood Avenue in Kitchener. She socialized with street youth and was introduced to the appellant through her "street dad", John MacDonald (known on the street as "Bulldog"), who was leaving town and who asked the appellant [page263] to look after Jennifer while he was gone. The introduction occurred somewhere between two weeks and two months before her death.

[10] Jennifer's landlord, David Pawlowski, testified that the appellant had visited her building several times driving a "rusty black Cutlass" with a noisy muffler.

[11] Nicole Rowe, a close friend of Jennifer's, testified that on May 1, 1992 (about a week before the killing), Jennifer called her and asked her to come over because she was not comfortable being drunk with the appellant in her apartment. Rowe went to the apartment and while she was there, she saw Jennifer and the appellant sitting beside each other on the bed. He at one point massaged Jennifer's shoulders and tried to push her top off her shoulder. Rowe overheard the appellant ask Jennifer to go out with him and she replied that she loved Jeff [Stadelbauer], her boyfriend, who was then in jail.

[12] When Jennifer was in the bathroom, the appellant asked Rowe how he could win Jennifer over. Rowe replied she would probably not go out with anyone unless Jeff was out of the picture. Rowe testified that the appellant commented, "it could be arranged" and said "his dick ached every time he thought of her".

[13] The Crown introduced a number of witnesses who gave evidence of the appellant's whereabouts during the afternoon and evening of May 8th. None of the witnesses testified to seeing him at or near Jennifer's apartment near the time of the murder. The Crown witnesses' evidence was inconsistent on the timing of the appellant's activities on the evening of May 8th. In addition, the evidence of Charles Russell, who lived in the same house as the appellant, was totally inconsistent with that of the appellant's acquaintances and Crown witnesses, Valerie Dobbin, Michelle Klobucar, Justina Meekison and Dan Stewart, as well as Jennifer's landlord, Pawlowski.

[14] On the evening of May 8th, Jennifer was out with her parents. They dropped her at her apartment at 10 or 10:30 p.m. Rowe telephoned her at about 11:45 p.m., but there was no answer. At 11:55 p.m., James Winters, who lived in the same apartment building, awoke to the sounds of a female screaming and thumping noises. Two or three minutes later he heard another short scream. An estimated five to 30 minutes later, he heard an engine start and a vehicle leave the parking lot. When asked to describe the sound of the engine, he said that it did not sound like his own small Japanese car. He testified that he did not note any unusual sound from the car and also testified that he did not note it having a defective muffler or problems with the exhaust system. [page264]

[15] Another tenant in the building, Michael Sommers, testified that the next night, shortly after 11:45 p.m., he heard a loud car with a bad muffler come into the parking lot and stay there for approximately five or ten minutes. It drove around the back of the building and then exited the building around five or ten minutes later. About 20 minutes after the car had left, he saw a light on in Jennifer's apartment and a large shadow moving in the window. The Crown proposed that the inference to be drawn was that the appellant had returned to clean up the crime scene.

[16] On Sunday, May 10, 1992, at approximately 7:30 p.m., Jennifer Ueberschlag's nude body was found by her father and Rowe lying face down in the living room. Her face was in a bowl of liquid.

PATHOLOGY

[17] Dr. David McAuliffe, a forensic pathologist, testified that he could find no cause of death other than asphyxia. He was unable to determine the cause of the asphyxiation: there was no sign of manual or ligature strangulation. The Crown and defence theory was that a cat toy found near the body had been inserted into the victim's mouth and caused her death. The pathologist agreed that the cat toy could have obstructed the airway and caused asphyxia if it were inserted sufficiently deep.

[18] There was blunt force trauma to the victim's head, neck and arms. There were four significant bruises to the scalp and her right eye was blackened. There was massive bruising with abrasions on the left side of the face and deep bruising on the left side of the neck. There was bruising to the left and right biceps, consistent with grab marks from behind. There was a serrated abrasion on the right wrist, consistent with a defensive wound. There were 18 to 20 cuts on the neck, which were superficial and did not contribute to her death. The cuts had been inflicted before death or possibly a very short time after death. There was a serrated pattern in some of the cuts, but either a serrated or non-serrated knife could have caused them.

[19] The body had been moved postmortem. Lividity patterns indicated that the victim first lay on her back and later was moved and turned onto her front.

[20] There were no signs of trauma to the vagina or anus. Three spermatozoa were found in the vagina, a sample too small to permit DNA analysis. Normal ejaculate contains hundreds of millions of spermatozoa, which can be found for up to seven days in the vagina of a living woman and longer after death. [page265]

THE CRIME SCENE

[21] In the bedroom, the mattress was not sitting squarely on the bed frame and the headboard was at an angle. The bed was not made. There were clothes on the floor near the bed, positioned in a way that suggested the wearer had undressed and dropped them on the floor. The clothes were not damaged.

[22] Police collected hairs and fibres and took swabs and fingerprints from the apartment. The body was examined by laser and luma light in an effort to discover hairs and fibres. Fingerprinting the lower body yielded no fingerprints. Although bruising on the upper body suggested that the victim had been grabbed, the decision was made not to fingerprint the remainder of the body because the body was deteriorating and it was decided that an autopsy needed to be performed immediately.

[23] Twenty identifiable fingerprints were found in the apartment and of these, only five were identified to known individuals. The appellant's fingerprints were found on a bottle of peach schnapps in a cupboard and nowhere else.

FORENSIC EVIDENCE

[24] Kimberley Johnston of the Centre of Forensic Sciences ("CFS") received various items for forensic analysis and DNA testing, including various swabs taken from the victim's body, bedding and other areas of her apartment, the appellant's knife that he customarily carried with him and roots from several hairs.

[25] Ms. Johnston concluded that the trace amount of semen identified on vaginal swabs from the victim was insufficient to permit generation of a DNA profile. Semen was not detected on any of the swabs from the victim's bedding and body. Blood was detected on swabs of a towel from the bathroom and from the kitchen and dining room floors and carpets, but no DNA results could be obtained from these samples. Blood was not detected on the appellant's knife.

[26] Ms. Johnston performed DNA testing on eight hairs found in the apartment. DNA extracted from four of these hairs matched the appellant's DNA profile: one of the hairs was found on the right buttock of the victim, two were found on the bed sheets and one on the duvet. Another hair found on the back of the victim matched the appellant's DNA profile in three of four loci where it was tested. At the fourth locus, the appellant could not be excluded as a donor of the DNA profile.

[27] The carpet underneath where the body was found was not seized until approximately six years later. By that time, it had been cleaned and other tenants had since lived in the apartment. [page266] Ms. Johnston's analysis of a piece of this carpet revealed a semen stain that had not originated from the appellant.

[28] Pamela Newall of the CFS also examined five items that were sent to her for DNA analysis, including a root tissue from a hair found on the carpet below the body and a hair collected from the body bag. She did not find DNA on any of the five items that she examined.

[29] Eric Crocker of the CFS undertook a microscopic examination of some 292 human hairs that were found in the apartment. The complicated evidence surrounding this hair comparison analysis will be discussed in detail below. Suffice it to say at this point that Mr. Crocker concluded that 13 hairs found in the apartment showed different levels of microscopic similarity to the appellant's known hair samples.

[30] There is nothing in the record to indicate the reason for the five-year delay between the killing of Jennifer Ueberschlag and the arrest of the appellant for her murder. It appears that the time lag is at least partially attributable to delay by the CFS which, according to the trial judge, did not perform the DNA testing until approximately five years after the death and did not do much of the microscopic comparison work until 1997 [^1].

[31] Apart from this delay, other aspects of the investigation are troubling. The failure to test the carpet under the body for semen until six years after the murder and the failure to first fingerprint the upper part of the victim's body, where she was beaten and appeared to have been grabbed, gives rise to a concern that evidence going to the killer's identity may not have been collected.

APPELLANT'S STATEMENTS TO POLICE

[32] The appellant gave two statements to police. The first was a verbal statement taken by Detectives Osinga and Close at the appellant's Mill Street residence in Kitchener on May 11, 1992. They advised him of Jennifer's death and told him they would like to rule him out as a suspect and wished him to give a statement. The appellant co-operated, giving a statement and responding to police questions. Detective Close requested the appellant to provide hair, saliva and pubic hair samples and the appellant readily agreed to do so.

[33] On May 13, 1992, Detective Osinga asked the appellant over the phone to attend at police headquarters to answer a few more questions. The appellant asked if he was still a suspect and [page267] [was] told that he was. The appellant agreed to come to the station for nine o'clock the next day to give a statement. Detective Osinga testified as follows about what was said to the appellant when he arrived at the station on May 14, 1992:

I advised Mr. Bennett that I knew he had given a partial alibi for Friday evening in his previous statement. I said we had spoken to Justine [Justina Meekison] and she seemed to back up that part of his previous statement. I said, she speaks highly of you. I advised Milton that we would want to alibi him from Friday morning, the 8th of May, until the Sunday evening. We needed to know what he had been doing during that period of time.

The appellant provided a more detailed written statement regarding his whereabouts for the time period specified by Detective Osinga.

[34] The content of the appellant's two statements are discussed below in connection with the trial judge's charge on the after-the-fact conduct evidence.

ANALYSIS

Issue 1: Did the Trial Judge Err in Admitting Trace Evidence from the Scene of the Killing that was not Probative and was Prejudicial?

[35] The defence brought a pre-trial application challenging the admissibility of certain hair comparison analysis and DNA evidence that the Crown sought to introduce at trial. On appeal, only the admissibility of some of the hair comparison evidence is at issue.

(i) Summary of hair comparison and related DNA evidence

[36] As noted, 292 human hairs were recovered from the victim's apartment and sent for analysis by the CFS. Eric Crocker of the CFS conducted the hair comparison analysis and testified at the voir dire and at trial. His report and testimony included the following terminology and analysis, as quoted verbatim from his report:

Type A-1: Unknown scalp hairs that are microscopically similar to a known hair sample. On the basis of this microscopic similarity, these hairs could have originated from the same source as the known sample.

Type A-2: Unknown scalp hairs that share some, but not all of the microscopic characteristics in common with a known hair sample. On the basis of the shared characteristics, these hairs cannot be eliminated as having originated from the same source as the known sample. Note: This partial similarity, while perhaps useful in an investigative sense, should not be considered to have any probative value in terms of identification. [page268]

Type A-3: Unknown scalp hairs that are dissimilar to the known samples.

Type B-1: Unknown body hairs that are microscopically similar to the known pubic and chest hair sample. On the basis of this microscopic similarity, these hairs could have originated from the same source as the known sample.

Type B-2: Unknown body hairs that share some, but not all of the microscopic characteristics in common with a known pubic and chest hair sample. On the basis of the shared characteristics, these hairs cannot be eliminated as having originated from the same source as the known sample. Note: This partial similarity, while perhaps useful in an investigative sense, should not be considered to have any probative value in terms of identification.

Type B-3: Unknown body hairs that are dissimilar to the known samples.

Type C: Unknown hairs that include short hair fragments and clippings or limb and immature hairs that are not useful for comparison purposes.

(Underlining in original)

[37] Mr. Crocker compared the hairs found at the apartment with comparison samples from the victim, the appellant, and various other suspects and individuals known to have been in contact with the victim, including Joel Coulombe (Jennifer was to be a witness for him at his upcoming sexual assault trial), Chad Tailby (an acquaintance of Jennifer's), Jeff Stadelbauer and others. According to Mr. Crocker, a type B-2 hair similar to Joel Coulombe's body hair sample was found on the victim's pink blanket, although he testified that he had never been in Jennifer's apartment. Mr. Crocker also concluded that two type B-2 hairs similar to Chad Tailby's body hair sample were found on her pink blanket.

[38] A significant number of hairs were found on the victim's body and bedsheets and elsewhere in the apartment that, according to Mr. Crocker, were not similar to any of the known hair samples. The following unknown hairs were found on or under the victim's body or on the body bag:

right hand: 1 unknown scalp hair left and right breast: 1 unknown scalp hair right thigh: 1 unknown scalp hair back: 2 unknown scalp hairs buttocks: 1 unknown scalp hair left buttocks: 2 unknown body hairs body (from laser exam): 1 unknown scalp hair body bag: 1 unknown scalp hair carpet underneath body: 1 unknown body hair [page269]

[39] The following unknown hairs were found on the victim's bedding:

top sheet: 1 unknown scalp hair 1 unknown body hair bottom sheet: 2 unknown scalp hairs pink blanket: 4 unknown scalp hairs 3 unknown body hairs pillow case and pillow: 1 unknown body hair

[40] In sum, Mr. Crocker's hair analysis indicated that there were three body hairs and eight scalp hairs either on or under the body and body bag that did not come from the victim, the appellant or any of the other known donors. The same is true of four body hairs and seven scalp hairs found in the victim's bedding.

[41] Mr. Crocker concluded that four body hairs found in the apartment were microscopically similar to the appellant's pubic hair sample, in other words that they were type B-1 hairs. These four hairs were described in Mr. Crocker's report as "B1MB" hairs, such abbreviation denoting the hair type and the appellant's initials. In addition, he concluded that six body hairs found in the apartment shared some, but not all of the microscopic characteristics in common with the appellant's pubic hair sample, i.e., type B-2 hairs. These hairs were described in Mr. Crocker's report as "B2MB" hairs, again denoting the hair type and appellant's initials. He also concluded that three scalp hairs found in the apartment shared some, but not all of the microscopic characteristics in common with the hair sample provided by the appellant, i.e., type A-2 hairs. These hairs were referred to in Mr. Crocker's report as "A2MB" hairs.

[42] These 13 hairs with some level of microscopic similarity to the appellant's known hair samples were found in the following locations:

B1MB hairs: 1 on the back of the victim 1 on the right buttock of the victim 1 on the body bag 1 on the washroom floor

B2MB hairs: 1 on the back of the victim 1 on the carpet under the victim's body 1 on the duvet 2 on the top sheet 1 on the pink blanket

A2MB hairs: 2 on the duvet 1 on the bottom sheet [page270]

[43] In Mr. Crocker's opinion, the similarities between the B1MB and B2MB hairs found in the apartment and the sample pubic hairs provided by the appellant indicated (quoting Mr. Crocker's rather inscrutable testimony on the voir dire), that: "if these hairs being similar to Mr. Bennett's pubic hairs, if they're not Mr. Bennett's pubic hairs, then they're not from Mr. Bennett." Or as defence counsel put it, and which Mr. Crocker accepted, "that's another way of saying if they're from Mr. Bennett, then they must be from his pubic hair . . . area."

[44] Mr. Crocker identified six C hairs, i.e., short hair fragments or limb and immature hairs that are not useful for comparison purposes, which he testified had "Negroid" characteristics. The location and description of these C hairs are as follows:

LOCATION DESCRIPTION

Victim's perineum Negroid hair tip fragment; natural tapered tip suggests body hair.

Taping of victim's Hair tip fragment of left and right indeterminate origin; some breast Negroid characteristics in terms of pigmentation and flatness.

Negroid limb hair.

Victim's buttocks Negroid hair tip fragment.

Top sheet Negroid limb hair.

Pink blanket Negroid immature hair.

[45] Mr. Crocker testified that C hairs can be readily carried on clothing and transported in and out of a particular area. He also testified that some of the C hairs with Negroid characteristics, including the one found on the victim's perineum (the area between the vagina and anus), exhibited dissimilarities to the appellant's known hair samples. In cross-examination on the voir dire, Mr. Crocker agreed that he was not suggesting that these C hairs came from the appellant and that they could well be from another person.

[46] As noted, Kimberley Johnston and Pamela Newall of the CFS performed DNA analysis on various hairs found by investigators [^2]. Twelve hairs found in the apartment were sent for DNA [page271] testing, nine of which, according to Mr. Crocker's evidence, were either B1MB hairs or B2MB hairs. The results of Ms. Johnston and Ms. Newall's analysis in relation to these nine hairs were as follows:

B1MB Hairs

LOCATION DNA RESULTS

Victim's back DNA match to appellant at 3 of 4 loci

Victim's buttocks DNA match to appellant

Body bag No DNA result attainable

B2MB Hairs

LOCATION DNA RESULTS

Top Sheet DNA match to the appellant

Top Sheet DNA match to the appellant

Duvet DNA match to the appellant

Pink Blanket DNA match to the victim

Victim's back No DNA result attainable

Carpet under No DNA result attainable victim's body

(ii) Voir dire on the admissibility of the hair comparison evidence

[47] In his submissions on the voir dire, defence counsel did not object, in my view, perhaps unwisely, to the admissibility of all of the microscopic comparison evidence of hairs that did not reveal a DNA match to the appellant. The focus of defence counsel's objection was described by the trial judge as follows:

Mr. Ducharme does not challenge the admissibility of the evidence of those hairs from the scene found to have a DNA profile matching that of Mr. Bennett. Nor does he challenge the admissibility of the evidence of those hairs which in the opinion of Mr. Crocker had a type 1 level of comparison to known hairs of Mr. Bennett.

[48] In other words, defence counsel was seeking to exclude evidence of a connection between the appellant and the three B2MB hairs that did not reveal a DNA match to the appellant: one on the pink blanket, one on the victim's back and one from [page272] the carpet under the victim's body. Comparison evidence regarding the three A2MB hairs on the bedding was not the focus of the submissions or the trial judge's analysis, but they are type 2 hairs and thus also subject to defence counsel's objection because there was no DNA evidence linking these hairs to the appellant. Defence counsel also objected in his submissions to the admission of evidence that six of the type C hairs had "Negroid" characteristics, "because the only possible relevance of that would be to suggest that it's [the appellant's] hair".

[49] In a ruling dated July 19, 1999, the trial judge dismissed the defence's application related to the hair comparison evidence and admitted all of the comparison evidence linking the various hairs to the appellant. Although the ruling does not specifically address the admissibility of the evidence regarding the C hairs having "Negroid" characteristics, Mr. Crocker was allowed to give evidence before the jury about the racial characteristics of the C hairs.

[50] In arriving at his decision, the trial judge extensively reviewed the Report of The Commission on Proceedings Involving Guy Paul Morin, 1998 by the Hon. Fred Kaufman, C.M., Q.C. ("Kaufman Report" or "Report"), which discusses issues surrounding the use or misuse of hair and fibre comparison evidence in the wrongful conviction of Guy Paul Morin. As the trial judge noted in his ruling, the Kaufman Report recommends at pp. 311-12 that hair comparison evidence showing only that an accused cannot be excluded as the donor of an unknown hair, or only that an accused may or may not have been the donor, is unlikely to have sufficient probative value to justify its reception at a criminal trial as circumstantial evidence of guilt. This comment would apply to both the type 1 and type 2 hairs referred to in Mr. Crocker's evidence.

[51] Also noted by the trial judge is that the Kaufman Report refers at pp. 312-13 to a possible situation where evidence that an accused cannot be excluded as the donor of a hair left by a perpetrator may have a high degree of probative value. Such a situation would arise, for example, if there are only two likely suspects in a case and hair comparison evidence indicates that hair left by the perpetrator could have come from one suspect and could not have come from the other. The Report goes on to state (at p. 313):

In the vast majority of cases, however, such evidence has extremely limited probative value: it merely permits the trier of fact to infer that the accused is one of a limitless class of persons who cannot be excluded as the perpetrator based upon this analysis. [page273]

[52] The trial judge then referred to the discussion in the Kaufman Report of the prejudicial effect of hair comparison evidence (at pp. 315 and 320):

The added difficulty with hair comparison evidence is that its prejudicial effect may be substantial, since the scientific opinion brings with it an aura of respectability and infallibility. The length and complexity of the testimony which must be examined to produce the minute conclusion that the accused cannot be excluded as the donor of the unknown hair has the potential to mislead the jury and cause the testimony to acquire a prominence and importance out of all proportion to its insignificance. Any trier of fact, hearing an exhaustive detailing of the minutia of hair similarities found, could easily (and understandably) conclude that only some legal or professional restraint prevents the experts from saying that the compared hairs come from a common source.

In the least, paraphrasing Mohan, there is a danger that hair and fibre evidence will be misused and will distort the fact- finding process. Dressed up in scientific language which the jury does not easily understand and submitted through a witness of impressive antecedents, this evidence may be accepted by the jury as being virtually infallible and having more weight than it deserves. Yet its probative value may often be insufficient to justify its reception.

[53] Finally, the trial judge noted that the Kaufman Report recommends against using the language "could have originated from the suspect" in scientific findings by the CFS:

Certain language enhances understanding and more clearly reflects the limitations upon scientific findings. For example, some scientists state that an item "may or may not" have originated from a particular person or object. This language is preferable to a statement that an item "could have" originated from that person or object, not only because the limitations are clearer, but also because the same conclusion is expressed in more neutral terms. (Executive Summary and Recommendations, Recommendation 10 at pp. 47-48)

[54] The trial judge next reviewed in some detail Mr. Crocker's evidence on the voir dire, which fills almost 200 pages of transcript. As noted by the trial judge, Mr. Crocker gave the following evidence with respect to the type 2 hairs, which is reproduced here in summary form:

-- a hair that reveals a type 2 match, whether of a scalp or body hair, from a microscopic point of view is not at all probative in terms of whether or not the hair came from a particular individual;

-- without confirming DNA evidence, the type B2MB hairs do not tell him anything beyond the fact that the unknown hair donor and the appellant can belong to the same class of donors, a class of indeterminate size, and that is why such [page274] classification type is not probative of identification of the donor of the hair;

-- the lack of probative value in regard to one type B-2 hair is not altered if he has five type B-2 hairs, each of which are slightly different from each other but each of which has similarities to one or more hairs in the known sample; the finding of five such hairs would be of use from an investigative point of view;

-- the finding of a body hair having a B-1 level of comparison to the appellant, at a different location in the apartment, does not make it any more likely that hairs with a type B-2 level of comparison to the appellant actually came from him;

-- taking together all of the type B2MB hairs that were found does not make for a stronger case that they are the appellant's hairs; what it does is reduce somewhat the size of the indeterminate class of people from which the hairs could have come, but that size reduction is unquantifiable;

-- the fact that three B2MB hairs generated a DNA profile matching that of the appellant does not change the probative value of the other B2MB hairs;

-- considering the total number of all of the B1MB and B2MB hairs does not increase the probative value of his opinion whether the B-2 hairs are the appellant's; the combined effect of considering all these hairs together is that it would in some way reduce the size of the indeterminate class who could be the donor, but in an immeasurable way so that the class is still of indeterminate size;

-- there was nothing about the type C hairs that increased the probative value of the type B-2 classification of some hairs as having possibly originated from the appellant;

-- if Mr. Crocker had a hundred hairs that he categorized as having a type B-2 level of comparison to a known sample, that would tend to suggest that they came from the same individual, but he was unable to so opine on the strength of five hairs;

-- Mr. Crocker did not compare the various B2MB hairs, the one with the others, so as to form any opinion as to their similarities; [page275]

-- the fact that there are three B1MB hairs, three A2MB hairs and five B2MB hairs has no probative value individually in terms of whether they can be said to be the appellant's; however, viewed collectively, because they have some overlapping characteristics, the highest it can be put is that they may very well have originated from the same individual; the similarities found do not preclude the possibility that these hairs came from more than one donor;

-- considering all of the type B2MB hairs together only suggests that they may have come from a single individual but that they certainly do not prove that and that is as high as his opinion can be put;

-- considering the type B-1 and B-2 hairs, together with the fact that some were found to have a DNA profile matching that of the accused means that the accused may well be the individual who donated all the B-2 hairs, but that opinion would be based on the effect of the DNA identification and not on the basis of his microscopic hair comparison;

-- the submission of one, six or ten type 2 hairs to a jury is prejudicial.

[55] The trial judge commented that if the only hair evidence linking the appellant to the scene consisted of those hairs having a type 2 level of comparison, he would have excluded the evidence on the basis that its low probative value was outweighed by the potential prejudicial effect. However, he concluded that the probative value of the type 2 hairs was "somewhat elevated" in this case:

In this case the probative value is somewhat elevated by the finding of type 1 hairs, by the finding of a DNA profile matching that of Mr. Bennett on some of these B-1 and B-2 hairs, the effect that finding may have in establishing that those hairs came from him, and by virtue of similarities said to exist between the various type 1 and 2 hairs, the available inference that all the B-2 hairs are his. The proximity of various hairs to each other, in terms of location, may also lend some probative value. The finding that some B-2 hairs had a negroid characteristic has some probative value, given other evidence before me that the police investigation revealed no other possibly involved persons of that racial origin, and the deceased's best friend knew of no other such acquaintances. This of course, may well be offset by the finding of negroid characteristics in some type C hairs having microscopic dissimilarities to the accused's sample hairs.

(Emphasis added)

[56] The trial judge further explained his decision to admit the evidence of the type 2 hair linked to the appellant as follows: [page276]

In summary, my view is that the evidence as to type 2 level of comparison has some limited probative value, that in the circumstances of this case the jury is of necessity going to hear hair comparison evidence in any event, that it is not practical to exclude type 2 level of comparison hair evidence without improperly distorting the balance of the hair comparison evidence, and that with appropriate effort by counsel and careful instruction from the court, the potential for prejudice can be satisfactorily addressed.

(iii) Analysis of the ruling on the voir dire

[57] In my view, the trial judge fell into error in deciding to admit the type 2 hair comparison evidence that defence counsel sought to exclude. The prejudicial effect of this evidence outweighed its probative value.

[58] The defence's theory was that the appellant's hairs associated with the body and bedding were found there for an innocent reason: his hairs had either been deposited there directly, as he had been sitting on the bed the week before the death, or through innocent transfer, as lividity patterns showed that the body was moved after death and Mr. Crocker accepted that hairs could be transferred to a body if it was sticky. The strength of the defence's theory would be reduced as the number of hairs linked to the appellant and found on the bedding and the body increased.

[59] The only possible relevance of the type 2 hair comparison evidence was to establish the identity of the killer. The inference going to identity would have to be that the finding of six type 2 hairs (three of which were scalp hairs and three body hairs) on the victim's body and bedding, which were linked by Mr. Crocker's evidence to the appellant, combined with the six pubic hairs found on or near the body and on the bedding which defence counsel conceded belonged to the appellant, supported the inference that the appellant was the killer.

[60] I am of the view that the probative value of the impugned type 2 evidence was not elevated to the extent found by the trial judge. Having said that, in fairness to the trial judge, I would agree entirely with his comment that, "Mr. Crocker's opinion as to the probative value of the type 2 hairs was, with respect, something less than crisp." However, Mr. Crocker clearly expressed the opinion that evidence of a type 2 hair comparison, without confirming DNA evidence, is not probative of identification of the donor of the hair. He also testified that the fact that three B2MB hairs generated a DNA profile matching that of the accused does not change the probative value of the other B2MB hairs. His additional, and rather obscure, testimony to the effect that taking together the hairs that showed a type 1 and type 2 comparison to the appellant's known hair samples reduced by an [page277] immeasurable amount the size of the indeterminate class who could be the donor of the type 2 hairs did not, in my view, significantly elevate the probative value of the type 2 evidence that the defence sought to exclude.

[61] Moreover, to the extent that the trial judge found that the probative value of the type 2 hair comparison evidence was elevated by virtue of similarities said to exist between the various type 1 and 2 hairs, Mr. Crocker's testimony was that he did not compare the type 2 hairs, one with the other, so as to form any opinion as to their similarities. In addition, the Kaufman Report indicates that an assessment of similarities between hairs is highly subjective (at p. 88):

The characteristics of a person's hairs vary from hair to hair, and they may differ even within a single hair on a person's body. Hair comparisons are not akin to fingerprint comparisons. Hairs are not unique, and the assessment of the similarities, differences and importance of hair characteristics is highly subjective. Efforts to quantify, through statistical analysis, the probability that a person was the donor of an unknown hair are not generally accepted in the forensic community -- in my view, with good reason.

(Emphasis in original; underlining added)

[62] A further concern I have with the trial judge's analysis is with his conclusion that the Negroid characteristics of the B2MB hairs increased their probative value. This conclusion is premised on the assumption that there was a finite group of people who could have left these hairs. However, there was not a discrete group of people who could have been the killer in this case. As defence counsel pointed out in his submissions on the voir dire, it had not been established that Jennifer did not have any other black acquaintances. Jennifer's landlord could not say, quoting the trial Crown's language, if there had been any "negro visitors" to the apartment during the time prior to Jennifer's death. Nicole Rowe and Jennifer's parents admitted that they did not know all of Jennifer's acquaintances. And as defence counsel further argued in his pre-charge submissions, it had not been proven that "no other black person could get into the apartment on the night" in question.

[63] Finally, I am of the view that the trial judge erred in concluding that if he excluded the type 2 evidence objected to by defence counsel, he would improperly distort the balance of the hair comparison evidence. The trial judge articulated this concern as follows:

For understandable reasons, Mr. Ducharme does not seek the exclusion of the evidence of the finding of all these hairs. The finding of hair bearing dissimilarity to any of the known samples, and the finding of hair having a similarity to other known persons and suspects has relevance to his defence. If [page278] evidence as to the finding of many other hairs in the apartment is admissible, as is evidence that when microscopically examined some of these hairs were found to have similarity to the hair of other known persons, or to be microscopically dissimilar to any of the known persons, how can such evidence be sensibly placed before the jury without the evidence of a purported B-2 level of comparison of others to Mr. Bennett? Without distorting the evidence?

[64] In my opinion, had defence counsel's position been accepted, Mr. Crocker would not have been put in a difficult position in explaining the evidence to the jury, nor would the evidence have been distorted. Crown counsel could have been permitted to call evidence regarding the type 1 hairs that were microscopically similar to the appellant's, as well as the type 2 body hairs that revealed his DNA profile. For the remaining six A2MB and B2MB hairs, Mr. Crocker could simply have testified that these hairs were found, but that he could not be sure who these hairs came from, instead of being permitted to testify that the appellant could not be excluded as the donor of these hairs. Such an approach would have avoided the strong potential for prejudice associated with this type of evidence, as identified at p. 315 in the Kaufman Report: "Any trier of fact, hearing an exhaustive detailing of the minutia of hair similarities found, could easily (and understandably) conclude that only some legal or professional restraint prevents the experts from saying that the compared hairs come from a common source."

[65] For these reasons, I conclude that the trial judge erred in admitting the type 2 evidence which the defence sought to exclude. Before turning to the next ground of appeal, I also express the opinion that the trial judge further erred in admitting the C hair evidence in this case. Counsel for the appellant did not press this issue on appeal, however, in view of my disposition that a new trial is required, I think it is appropriate to address it.

[66] In his submissions on the voir dire, defence counsel argued that the C hairs should not be admitted. The trial judge said the following about this evidence:

While type C hairs are not suitable for comparison in Mr. Crocker's opinion, he did conduct some microscopic comparison of those hairs and found the majority of them to exhibit negroid characteristics but to be dissimilar to the known hairs of Mr. Bennett. One of these C hairs was found on the perineum of the victim. The fact that he observed dissimilarities in characteristics as between the known Bennett hairs and these unknown C hairs was not mentioned in his report.

[67] In my view, the type C hairs had extremely low probative value and had significant potential for prejudice given the [page279] testimony regarding their racial characteristics. As previously mentioned, Mr. Crocker's evidence was that type C hairs can be readily carried on clothing and transported in and out of a particular area. He also testified that he was not suggesting that these Negroid C hairs came from the appellant and said that they could well have come from another person. He expressed the view in cross-examination that the C hairs in this case do not have any value because they were isolated hair fragments, as opposed to a grouping of numerous similar hair fragments.

[68] On the other hand, the potential for prejudice was significant given that the appellant is a black man and the evidence was that several type C hairs, including the one found on the victim's perineum, exhibited Negroid characteristics. Even though Mr. Crocker testified that most of the six C hairs exhibited dissimilarities to the appellant's known samples, the jury could well have inferred that as the only known black suspect in the case, these hairs must have been his. For these reasons, the trial judge erred in not excluding the evidence regarding the racial characteristics of the C hair from the jury's consideration.

Issue 2: Did the Trial Judge err in his Charge by Failing to Instruct the Jury that they must not Draw the Speculative and Prejudicial Inference that the Crown Invited them to Draw with Respect to the Trace Evidence?

[69] There is a real possibility that the jury misused the hair comparison evidence in this case.

[70] In his ruling on the voir dire, Glithero J. expressed confidence that any prejudice from the type 2 hair evidence could be reduced or eliminated at trial:

Such potential prejudice [of the microscopic hair comparison evidence] can in my opinion be very much reduced, if not eradicated, by the proper examination and cross-examination of Mr. Crocker, by the introduction of such other expert evidence as the defence may see fit to elicit, and by my best efforts to carefully instruct the jury as to the limitations on the probative value of this evidence.

[71] The potential for prejudice from the type 2 hair comparison evidence was dramatically increased by the way that Crown counsel in his closing address asked the jury to use this evidence. The Crown urged the jury to draw inferences unsupported by Mr. Crocker's testimony based on a "common sense approach" to the hair evidence:

But as a matter of common sense, if you find DNA on one of those B-2 hairs, as was done here, that would seem to make it pretty compelling that it was from Mr. Bennett. Were you to find two hairs together, one being a B-1 [page280] and one being a B-2, for instance, it would seem highly likely that the same person deposited those there.

And as Mr. Crocker said in his evidence that while you may have this . . . this indeterminate class is the phraseology that's used, if you find a number of hairs that vary between each other, but are all very similar to a known sample, it's the fact we have different hairs all narrowing toward a single sample that tends to narrow the range of this indeterminate group.

And he also went on to say that, in response to a suggestion by my friend, if you had six hairs in exactly the same place, what that would suggest to you, using this logic, is that all six individual hairs probably came from a single individual. And, of course, he's being very careful not to say they would come from Mr. Bennett, or anybody else, but a single person. And that's a pretty logical, common sense approach.

And in this particular case, we not only have that type of thing, but we also have DNA on some of those hairs, and the fact they're Negroid.

(Emphasis added)

[72] The Kaufman Report noted at p. 315 that the Crown used the type of argument based on common sense in the Morin trial:

Any trier of fact, hearing an exhaustive detailing of the minutia of hair similarities found, could easily (and understandably) conclude that only some legal or professional restraint prevents the experts from saying that the compared hairs come from a common source. Indeed, Mr. McGuigan very persuasively suggested in his jury address that, apart from the experts, a "common sense" approach to the hair and fibre evidence led inexorably to the conclusion that Christine Jessop had been in the Morin Honda.

(Emphasis added)

[73] Even more problematic in this case is that Crown counsel was misleading and imprecise in the way that he dealt with the hair evidence in his closing argument. For example, he told the jury: "And the B-2's were, I believe, perhaps one exception, all DNA'ed to Mr. Bennett". Having said that, he went on to tell the jury that they could double check for themselves the numbers and locations of the hairs, as the jurors evidently had with them a chart of the evidence. However, at another point in his closing, the Crown stated in quite an inflammatory manner that all of the B-2 hairs were attributed by DNA to the appellant:

And so let's then use common sense. And where these items are found when you look at the B2 hairs, etcetera, and in conjunction with B1 hairs, or DNA'ed hairs, and the lack of other Negroids in Jen's life. And then look as well, when you look at the DNA, at the other commonsensical type approach that these hairs, the body hairs, the pubic hairs left behind with Mr. Bennett's DNA profile on them and bearing some, if not all microscopic similarities to his hair, are attributed, at least by DNA, to a person whose dick ached every time he thought about Jennifer, and who was in fact over there looking for her. [page281]

[74] And again near the end of his closing, the Crown told the jury:

And then you look at all . . . the fact that of all the hairs you find in the apartment, you find two A-2 hairs, which some would suggest we don't even look at A-2 hairs, scalp hairs. I think it's obvious you should look at them.

But then you find all of these body hairs. And you find them only on the body, in the bedclothes, or in the body bag, or under, with the exception of one in the washroom, with DNA that's linked to him. And they are pubic hairs.

(Emphasis added)

[75] In his charge, the trial judge told the jury that in Mr. Crocker's opinion, type 2 levels of comparison ought not be included in reports that go to the jury and that they are not probative. He said the following about the Crown's address to the jury on using this evidence:

Mr. Russell made submissions to you as to the significance of the type 2 comparison hairs and the C hairs. He urged you to draw certain inferences and use the evidence you have in a way that may go beyond Mr. Crocker's opinion as to the use that should be made of it.

I give you this caution, acknowledging that you are the finders of facts, not me.

[76] The trial judge went on to tell the jury in regard to Mr. Crocker's evidence:

I would suggest to you that his opinions you can accept, you can reject, but I do not know that you can go adding much to them because you were not given the tools which he used in order to formulate his opinions so as to be able to build your own blocks, if I can put it that way, and go beyond what he gave you in terms of opinion.

[77] He asked the jury to look at the charts before them on the hair evidence and proceeded to summarize the location and type of hairs found in the apartment as indicated in Mr. Crocker's report.

[78] In fairness to the trial judge, crafting an appropriate instruction on the use of the hair evidence in this case was not a simple task and the trial judge went to great lengths to be fair in his charge. However, in my view, the trial judge failed to provide the careful instruction regarding the hair evidence, which he acknowledged on the voir dire was required.

[79] The hair evidence in this case was a morass. It is not safe to assume that the jury was not misled on its significance. The trial judge instructed the jury that some of the Crown's suggested inferences "may have gone beyond Mr. Crocker's opinion". Yet this instruction significantly understated the unfairness of Crown counsel's treatment of the hair evidence. Crown counsel misstated the expert evidence by suggesting to the jury that they [page282] could infer that finding one B-1 hair together with a B-2 hair makes it "highly likely that the same person deposited those there". Mr. Crocker's evidence did not support such an inference. In addition, Crown counsel's suggestion that all of the type 2 hairs had been "attributed, at least by DNA, to a person whose dick ached every time he thought about Jennifer" misstated and oversimplified the evidence in a way that was capable of seriously misleading the jury to draw unwarranted inferences adverse to the appellant.

[80] The case against the appellant, which was based entirely on circumstantial evidence, cannot be described as an overwhelming one. In this context, the hair comparison evidence was a very important element of the Crown's case. In view of my ultimate disposition directing a new trial, I think it might be of value to give some direction on how the hair evidence ought to be dealt with in this case.

[81] First of all, as explained above, the evidence that certain C hairs revealed Negroid characteristics should not be admitted as evidence for the jury to consider. All that Mr. Crocker should be permitted to say is that various C hairs were found, but that they are not useful for comparison purposes.

[82] Second, the type 2 hair comparison evidence related to the three body hairs and three scalp hairs that were not linked by DNA to the appellant should not be admitted.

[83] Third, I question the wisdom of a concession by defence counsel that the type 1 hairs without a DNA match to the appellant are his hairs. One of these hairs was on the washroom floor and one was found on the body bag. This concession appears to be unwarranted in light of the expert's evidence that a type 1 comparison is not evidence that a hair is definitely that of the known donor, only that it may or may not have come from him. Mr. Crocker acknowledged that he used language that was frowned on by the Kaufman Report when he defined the type 1 hairs in his report as hairs that "could have originated from the same source as the known sample" (emphasis added). As noted, the Kaufman Report recommends that the phrase "may or may not" have originated from a particular person is preferable to "could have" originated from that person. The Report also recommends that:

Evidence that shows only that an accused cannot be excluded as the donor of an unknown hair (or only that an accused may or may not have been the donor) is unlikely to have sufficient probative value to justify its reception at a criminal trial as circumstantial evidence of guilt.

[84] Kaufman concluded at p. 323 that it was not appropriate for him "to articulate any hard and fast rules" as to when hair [page283] comparison evidence should be admitted in a criminal trial. I am not prepared to say that the type 1 comparison evidence should not have been admitted in the circumstances of the other forensic evidence in this case.3 However, in my view, the jury should be instructed that Mr. Crocker's evidence that certain hairs found at the crime scene are "microscopically similar" to hairs donated by the accused is not to be taken as expert testimony that the found hairs definitely came from the accused. The jury should further be instructed that the inconclusiveness of this evidence is a matter of weight for them to assess in the context of the other pieces of circumstantial evidence relied on by the Crown.

[85] Fourth, as the trial judge properly did, the jury should be reminded of the four hairs that yielded a DNA profile matching that of the accused as well as the fifth hair that had a DNA profile matching his at three of four loci. They should also be reminded of Ms. Johnston's testimony of the population frequency statistics regarding DNA. And further, as the trial judge did in this case, the members of the jury should be instructed to apply their own common sense and good judgment in assessing this evidence.

[86] Fifth, the jury should be instructed that the presence on the body and bed sheets of hair linked by DNA to the appellant had to be considered along with the evidence that a number of head and body hairs that did not come from either the victim or the appellant were also found on the victim's body and her bedding. The trial judge should remind the jury of the number and location of these hairs, as well as the expert's opinion on the type of such hairs, i.e., whether they are body or scalp hairs. [page284]

[87] Sixth, with respect to the numerous hairs found in the apartment that did not exhibit any similarities to the appellant's known hair samples, Mr. Crocker could be permitted to give evidence regarding which of these hairs showed a type 1 relationship to the other known donors of the hair samples, including the deceased. For those hairs that showed a type 2 relationship to the other known donors of the hair samples, Mr. Crocker should only be permitted to testify that he cannot be certain who these hairs came from. For the remaining hairs, he should simply be permitted to testify that these hairs were dissimilar to all of the known hair samples.

[88] Finally, the record indicates that the jury had a chart that set out the hair evidence in this case. The chart will of course need to be modified in light of my decision. It is worth emphasizing that some sort of written aid, preferably one that is less confusing than the one found on this record, is required to help the jury in sorting out the admissible hair evidence in this case.

Issue 3: Did the Trial Judge Err in his Charge in Regard to Identification Evidence and did he Erroneously Allow the Jury to Consider as Evidence Against the Appellant an Identification which Excluded the Appellant from Being the Person who was Following the Victim on the day Before she was Killed?

[89] This ground of appeal relates to the evidence of Dennis Morningstar, a Crown witness with an extensive criminal record and a history of drug abuse. Morningstar testified that on May 6, 1992, he saw the appellant holding Ms Ueberschlag's hand at a youth drop-in centre called Oasis. He also testified that in the late afternoon or evening of May 7, 1992, Ms Ueberschlag visited him at his house on Waterloo St. and appeared to be scared and nervous. She called a friend to pick her up. He testified that he saw the appellant standing across the street, leaning against a post and wearing a Walkman. Morningstar agreed with the proposition put to him in cross-examination that the man across the street was "clean shaven, no beard". However, the evidence of the other trial witnesses was consistent that at the time in question, the appellant had a beard.

[90] In his statement to police in July 1992, Morningstar mentioned having seen the appellant at Oasis on May 6, but he did not mention having seen the appellant outside his house on the following day. He first told police about the May 7th sighting over four years later, in December 1996. At the preliminary inquiry, he was not able to identify the appellant as being the man he saw [page285] standing across the street on May 7, 1992. At trial, he identified the appellant as that man.

[91] The Crown, in his closing address, conceded with respect to Morningstar that "there were obviously a number of huge gaps or errors in his evidence" and acknowledged that the jury may find his "evidence to be totally worthless in that sense, in terms of his identification". Crown counsel went on to give several reasons for not rejecting his evidence out of hand, including that Morningstar did not hear a lot of defence counsel's questions, that Morningstar had mentioned seeing the appellant at Oasis in his July 1992 statement to police and that he testified "in this court that he was certain that that was the man and said that he hadn't seen the beard because the man had his head down across the way. But that that was the guy, and that as he walked away, he turned his face back, and he recognized him."

[92] In the trial judge's charge to the jury, after bringing up the topic of Morningstar's evidence, the trial judge warned the jury in general terms of the dangers of identification evidence. He summarized Morningstar's testimony and commented as follows:

I am instructing you that you ought to be exceedingly careful with respect to this evidence because I suggest it has a number of real concerns to it. It is very easy to point to someone who is sitting in an enclosure in the courtroom and say that is the man because you are picking out of a line up of one.

On a previous occasion, this witness was unable to pick the same man out in the same circumstances at a time that was closer to the event. . . .

I would suggest that those are matters of real concern and, in addition, he has sworn before you that the man across the street did not have a beard. All of the evidence you had before you, as I appreciate it, is that Mr. Bennett had a beard back in May of 1992. Where a witness describes a person in a way inconsistent with a prominent feature of their appearance, I would suggest to you that it is identification which is worth very, very little and that you ought to be very cautious in accepting the rather bare assertion of that is the man when he points to a line up of one here in the courtroom.

(Emphasis added)

[93] Counsel for the appellant contends that the trial judge's charge was inadequate because he did not instruct the jury that Morningstar's evidence exonerated the appellant in the limited sense that his evidence indicated that Ms Ueberschlag appeared to be afraid of a black man other than the appellant on the day before her death. Trial counsel had asked for such an instruction during the pre-charge conference, relying on Chartier v Quebec (Attorney General), 1979 17 (SCC), [1979] 2 S.C.R. 474, 9 C.R. (3d) 97. In Chartier, at p. 494 S.C.R., p. 138 C.R., Pigeon J. for the majority stated: [page286]

The appellant was not "positively identified" by the witness Holland . . . , since the latter said: "I recognize him by his posture, build, facial features, stomach, etc., but the only thing that is different is that his hair was grey at the time of the incident."

In order for this statement to be an identification, it would have been necessary to establish that the Appellant had had grey hair at the time of the incident, otherwise the witness Holland was not identifying the Appellant, but rather exonerating him. Regardless of the number of similar characteristics, if there is one dissimilar feature there is no identification.

(Emphasis added)

[94] This court in R. v. Boucher (2000), 2000 3270 (ON CA), 146 C.C.C. (3d) 52, [2000] O.J. No. 2373 (QL) (C.A.), at pp. 57-58 C.C.C. applied Chartier to conclude that a notable dissimilarity in identification evidence, absent some other inculpatory evidence, renders the identification evidence of a resemblance of no probative value. In Boucher, this court quashed a committal against three accused on charges of robbery and related offences. The Crown's case depended on a nexus being shown between certain pants worn by one of the robbers and the clothing worn by one of the accused on the day of the robbery. If the description of the pants matched, the pants could provide the requisite link to the robbery. However, the pants worn by the robber were described by a bank customer as being all black, while the pants worn by the accused were described by another witness as having a white stripe running the length of the leg. At p. 58 C.C.C., Rosenberg J.A. stated:

In view of the dissimilar feature of the pants, there was no identification, merely a resemblance. In the absence of some other inculpatory evidence, a resemblance is no evidence. If there were other inculpatory evidence it may be that a trier of fact would have good reason for finding that the customer's testimony was unreliable. Since there was no other evidence, the dissimilarity at worst renders the resemblance of no probative value and possibly stands as an exculpatory feature.

[95] In my view, the trial judge erred in leaving it open for the jury to conclude that the appellant was the person Morningstar saw across the street the day before Jennifer was killed. Morningstar's evidence that the man across the street was "clean shaven, no beard" constitutes a notable dissimilarity in the identification evidence. Again, quoting Boucher, at p. 58 C.C.C., "In the absence of some other inculpatory evidence, a resemblance is no evidence . . . Since there was no other evidence, the dissimilarity at worst renders the resemblance of no probative value and possibly stands as an exculpatory feature."

[96] There was no evidence other than Morningstar's testimony that the man across the street was the appellant. Accordingly, the [page287] trial judge ought to have instructed the jury that Morningstar's identification of the appellant was worthless, not that it was "worth very, very little". Further, the trial judge ought to have instructed the jury that it was open to them to infer from Morningstar's evidence that he had seen a clean-shaven black man, and not the appellant, standing across the street on the day before the killing and that Jennifer appeared to be afraid of that man. In fairness to the trial judge, the Boucher decision was released after the trial took place.

Issue 4: Did the Trial Judge Err in his Charge in Regard to After-the-Fact Conduct by Erroneously Allowing the Jury to use After-the-Fact Conduct Evidence Against the Appellant that was Prejudicial and had Little or no Probative Value?

[97] The appellant contends that the trial judge erred in leaving with the jury various items of after-the-fact conduct evidence from which they could infer a consciousness of guilt on the part of the appellant. The first area of evidence relates to the appellant's two statements to police, while the other areas relate to the appellant's behaviour and reaction to various events after the killing.

I. The alibi evidence

[98] Crown counsel led the appellant's two statements to police through his examination-in-chief of Detective Close. The statements were introduced as exhibits over defence counsel's objection.

[99] In the first statement, which was given in response to the police's expressed intention of trying to rule out the appellant as a suspect, the appellant said that the last time he talked to Jennifer was on Thursday (May 7). He said that he went to her house on Friday (May 8) with Dan (Stewart) and Justine (Meekison) with a bottle of tequila and no one was there. They waited for half an hour and no one showed up so they left. He tried to call her Friday night, a few times on Saturday and once on Sunday afternoon but there was no answer. He went to Elora Gorge with the kids (whom he lived with) on Sunday. On the afternoon of May 10, Nicole Rowe had called him and asked if he had seen Jennifer. He told her that he last talked to Jennifer on Thursday and that he had not seen her since. He said he then got kind of worried about her and so he phoned a mutual friend, Michelle, to ask if she had seen her. She said that she thought she saw Jennifer on the bus headed for Fairview (a shopping mall) on Friday but she wasn't sure. He called Rowe and told her this and that was the last he heard until he heard her address on the radio. [page288]

[100] In response to police questioning about when he was at Jennifer's apartment on Friday, he indicated that "it was still light -- 7 or 7:30, I'm not sure". He was asked what he did after leaving the apartment and replied, "We came here [the home on Mill Street] and we drank for a bit. We went and dropped Dan and Justine off at Wizards and I came back here." He said that this was "at 9:30-10".

[101] After responding negatively to the question if he ever had sexual relations with Jennifer, the appellant said "An hour or so later I went back down to Wizards to find Chris Corbett. She lives at the Y too. I didn't find Chris, so I talked to Justine for a while and then I came back here about 11:30-12." He was asked if he had ever been in Jennifer's apartment and he replied "Ya, a couple of times." Finally, he was asked if he knew who would want to kill Jennifer and he replied that he had "no idea . . . she seemed alright to me."

[102] The appellant's second statement was given at police headquarters after the police told him that he was still a suspect and in response to their request that he alibi himself from Friday morning until Sunday evening. The appellant wrote that he got up Friday at about 8:30 to get the kids he had been watching off to school and then went back to sleep until about 1 p.m. He watched TV most of the afternoon, got supper ready, cleaned up around the house until about 7 and went to Fairview Mall to see Michelle. They went to a friend of her's house until about 8 or 8:30. He left and went downtown to the pool hall (Wizards), where he saw Justine and Dan and asked if they wanted to get drunk because he had tequila. They went to the park where he and Dan drank, but they left because there were too many cops. They went to Jen's house, he knocked, but there was no answer so they sat in the car for awhile. A man he had seen before (this was Jennifer's landlord) offered them pizza and asked a lot of questions. So they left and went to the house where he was staying, drank some more, and then he dropped them off at Wizards. He then went to his friend Bozer's house on Waterloo St., but no one was home. He returned to the pool hall and stayed for about 15 minutes and then went home and watched T.V. the rest of the night. He stayed around the house all day Saturday and Saturday night and went to bed at about 10 p.m. because he, the kids and Miz (Mitsy Roy) were to go to Elora on Sunday. They got up at 4 a.m. and left the house at 7 a.m. and got back at 1:30 p.m. He went to sleep until supper at 8 p.m.

[103] A comparison of the two statements reveals some minor differences in the appellant's version of his whereabouts on Friday May 8, 1992. He indicated in the first statement that [page289] he went to Jennifer's apartment when it was "still light", around 7 or 7:30. In contrast, in the second statement he did not give a time when he arrived at her apartment, but his chronology indicates that it was some time after 8 or 8:30. In addition, in the first statement the appellant said that he went to the pool hall to find Chris Corbett, but he did not refer to this in his second statement. In the second statement, the appellant said that he went to his friend Bozer's house, which he had not mentioned in the first statement. Finally, in the first statement he told police that he returned home at about 11:30-12, whereas in the second statement, he did not indicate what time he went home.

[104] The trial judge in his charge to the jury stated there were various items of evidence of things done by the appellant from which the Crown was urging them to draw an inference of guilt, one of which involved the discrepancies in the appellant's two statements regarding his whereabouts on the evening of May 8th. The trial judge described the Crown's position on the significance of the statements as follows:

The Crown says that the explanations by Mr. Bennett for his whereabouts on the evening of May 8 are different in those two statements. The Crown argues that the inconsistency in the explanations is such that at least one of them cannot be true. The Crown argues that at least one of them is therefore a deliberate falsehood as to his whereabouts during the evening of the killing and the Crown argues that is something from which you can infer guilt because the Crown says it is more likely that a guilty person would lie about his whereabouts than would an innocent person.

[105] After going over the gist of the statements, the trial judge gave the following instruction:

If, having regard to all of these circumstances, you are satisfied that Mr. Bennett was deliberately fabricating or concocting a false alibi in making these statements to the police, and if you are also satisfied that this fabrication or concoction emanated from a sense of guilt for the crime on his part, then you would be entitled to consider that as circumstantial evidence of involvement.

[106] The trial judge stressed that disbelief of the alibi evidence is not proof of concoction. He told the jury that the Crown must prove deliberate concoction. He also told them to consider all of the circumstances surrounding the creation of the statements, whether there were material inconsistencies between them and if so, whether the inconsistencies were deliberate or innocent. He finished his instruction on this point as follows:

It is only if you are satisfied that the evidence is actually concocted, that is deliberately fabricated, that you would be entitled to draw an inference of guilt from that evidence. [page290]

[107] Appellant's counsel argued that the appellant did not give himself an alibi in the sense of being with someone else at the time of the murder. Counsel further argued that, considering the Crown's theory was that the killing took place at approximately 11:55 p.m. on the Friday night, the appellant did not give himself a clear alibi in either statement because he left open the possibility that he was out on the town at the time of the murder. In the first statement, the appellant placed himself at home at "around 11:30-12", while in the second statement, he did not give a time when he got home.

[108] While I accept that the appellant did not give himself a strong alibi in the sense of being with someone else at the time of the killing, or being in a location far removed from the scene of the crime at the critical time, I would agree with the respondent that the appellant did give an alibi in the sense that his statements indicate that he was not at the crime scene at the time of the murder.

[109] Counsel for the appellant went on to argue that even if the statements could be taken as providing an alibi, the jury ought to have been told that they could not infer guilt from the statements because there was no extrinsic evidence of concoction in this case. Counsel pointed to a series of cases from this court, which he says establish that the issue of concoction in connection with an alibi should only be raised when there is extrinsic evidence that the alibi has been concocted, citing R. v. Coutts (1998), 1998 4212 (ON CA), 40 O.R. (3d) 198, 126 C.C.C. (3d) 545 (C.A.), leave to appeal refused [1998] S.C.C.A. No. 450; R. v. Krishantharajah (1999), 1999 3701 (ON CA), 43 O.R. (3d) 663, 133 C.C.C. (3d) 157 (C.A.); R. v. Campbell (1999), 1999 2688 (ON CA), 139 C.C.C. (3d) 258 (Ont. C.A.); R. v. Wristen (1999), 1999 3824 (ON CA), 47 O.R. (3d) 95, 141 C.C.C. (3d) 1 (C.A.), leave to appeal refused [2000] S.C.C.A. No. 419; R. v. Price (2000), 2000 5679 (ON CA), 144 C.C.C. (3d) 343, 33 C.R. (5th) 278 (Ont. C.A.); R. v. Diu (2000), 2000 4535 (ON CA), 49 O.R. (3d) 40, 144 C.C.C. (3d) 481 (C.A.); R. v. Blazeiko (2000), 2000 14726 (ON CA), 48 O.R. (3d) 652, 145 C.C.C. (3d) 557 (C.A.); and R. v. O'Connor (2002), 2002 3540 (ON CA), 62 O.R. (3d) 263, 170 C.C.C. (3d) 365 (C.A.).

[110] The respondent submits that there was no need to adduce extrinsic evidence of concoction in this case because Crown counsel was relying on a material inconsistency between the two statements, citing the decisions of this court in R. v. Ruddick (1980), 1980 2941 (ON CA), 57 C.C.C. (2d) 421 (Ont. C.A.), at p. 440, leave to appeal to S.C.C. refused (1981), 57 C.C.C. (2d) 421n; R. v. Andrade (1985), 1985 3502 (ON CA), 18 C.C.C. (3d) 41 (C.A.) at pp. 66-69 C.C.C.; and R. v. McInnis (1999), 1999 2671 (ON CA), 44 O.R. (3d) 772, 134 C.C.C. (3d) 515 (C.A.), at pp. 786-87 O.R., pp. 533-34 C.C.C.

[111] The cases cited by the appellant draw a distinction between statements made by an accused which are disbelieved [page291] and therefore rejected and statements that can be found to have been concocted in an effort to avoid conviction. The former have no evidentiary value, while the latter can constitute circumstantial evidence of guilt. In Coutts, supra, Doherty J.A. observed, at p. 203 O.R., p. 551 C.C.C. that this distinction is:

. . . essential to ensure that the trier of fact properly applies the burden of proof in cases where statements of an accused are tendered or an accused testifies. If triers of fact were routinely told that they could infer concoction from disbelief and use that finding of concoction as evidence of guilt, it would be far too easy to equate disbelief of an accused's version of events with guilt and to proceed automatically from disbelief of an accused to a guilty verdict. That line of reasoning ignores the Crown's obligation to prove an accused's guilt beyond reasonable doubt. By limiting resort to concoction as a separate piece of circumstantial evidence to situations where there is evidence of concoction apart from evidence which contradicts or discredits the version of events advanced by the accused, the law seeks to avoid convictions founded ultimately on the disbelief of the accused's version of events.

(Emphasis added; citations omitted)

Doherty J.A. went on to hold that a jury should only be instructed that they may find that an accused's statement is concocted and, therefore, capable of constituting circumstantial evidence of guilt, where there is some evidence of actual concoction.

[112] In the case at bar, the differences between the two statements in this case, as even Crown counsel acknowledged in his closing address to the jury, were slight. The Crown in his closing treated the inconsistency as being that the appellant mentioned going to Bozer's place only in his second statement:

And his alibi, if you will, is slightly different. Not a huge difference, but his whereabouts are clearly differentiate . . . different to a certain extent from the one statement to the other in that he has himself going to Waterloo Street to look for his friend, Bozer, in the second statement, which hadn't been in the first statement.

[113] On appeal, the Crown now contends that the inconsistency lies in the fact that in the first statement, the appellant potentially had himself out on the town at 11:30-12, and thus he would have had the opportunity to kill the victim, whereas in the second statement, he tightened up the timing to have himself home earlier. I cannot agree with this interpretation. In his second statement, the appellant gave no indication of the timing of his whereabouts any time after 8:30 p.m. on the night of the murder. He said that "he went home and watched TV for the rest of the night", but did not try to pin down when he returned home on the night of the killing.

[114] In his charge, the trial judge said on several occasions that the Crown's position was that the two accounts were different. [page292] However, the trial judge did not attempt to indicate to the jury the nature of the inconsistency between the two statements. Instead, he left it for the jury to determine whether there was any inconsistency, and if so, whether it was material.

[115] In my view, if an instruction permitting a jury to find concoction and to infer guilt based on that finding is to be given, the trial judge is required to identify for the jury what inconsistency in the accused's statements could warrant a finding of concoction. The inconsistency must be compelling in the sense, for example, that there is an indication that the accused was attempting to mislead investigators by fabricating an alibi. As Martin J.A. stated in R. v. Andrade, supra, at p. 67 C.C.C.: "the giving of contradictory statements by an accused with respect to his whereabouts at the critical time may in some circumstances constitute evidence upon which the jury is entitled to find that one or both statements are fabricated".

[116] The trial judge, in my view, erred in instructing the jury that they were entitled to find that the appellant had fabricated an alibi, and that it was open to them to infer guilt from his two statements. The slight differences between his statements were not capable of warranting a finding of concoction. While the trial judge could have told the jury that any differences in the two statements was a factor they could take into account in deciding whether to believe either of them, there was no basis for instructing them that they could go further and use the statements as a separate piece of circumstantial evidence from which guilt could be inferred. Rather, the jury ought to have been told that if they disbelieved the alibi, they should simply discard that evidence, without more: see R. v. Hibbert, 2002 SCC 39, [2002] 2 S.C.R. 445, 163 C.C.C. (3d) 129, at paras. 61 and 67.

II. Other after-the-fact conduct evidence

[117] The appellant identified six other items of after-the-fact conduct that the trial judge left with the jury as being evidence from which the Crown asked them to infer guilt. The items of evidence were: the appellant's anger at a neighbour for telling police that he thought the appellant was capable of killing Jennifer; the appellant telling Nicole Rowe that Jennifer was seen going to a mall on the afternoon of her death; his failure to make certain phone calls upon learning of Jennifer's death; his failure to ask police if Jennifer was the victim; his desire to have copies of statements made to police by his acquaintances, Justina Meekison and Dan Stewart; and his reaction to the report that police had found a silver letter opener that was missing from Jennifer's apartment. [page293]

[118] In my view, the trial judge should not have left three of these items of evidence with the jury as evidence from which they could infer consciousness of guilt: the appellant's anger at being described as someone capable of killing the victim, his failure to make certain phone calls and his failure to ask police if Jennifer was the victim. These forms of conduct are examples of demeanour evidence that is highly suspect and easily misinterpreted. Regarding the remaining items, the trial judge erred in not instructing the jury more fully on the limitations of this evidence and in what circumstances it is permissible to draw an inference of consciousness of guilt from it.

[119] Given the importance of this evidence in the context of the Crown's circumstantial case against the appellant, I will describe in some detail the impugned areas of evidence. I then set out the trial Crown's closing arguments to the jury in regard to each, as well as the trial judge's instructions on the use that the jury could make of this evidence.

(i) Anger at being described by Doug Hiltz as someone who could kill

[120] The Crown witness, Doug Hiltz, lived across the street from the home where the appellant was staying and he also knew Ms Ueberschlag. Hiltz testified to having two encounters with the appellant after Jennifer's death. The first was on May 12, 1992. That day, the police arrived at Hiltz's house and questioned him. Hiltz told the police that he thought the appellant was capable of killing Jennifer. The appellant, who had seen the police across the street, came over to ask Hiltz what he told them. Hiltz testified that the appellant got upset when Hiltz told him what he said, and then left.

[121] The second encounter occurred a week or two later. Hiltz testified that there had been rumours circulating downtown that he was telling people that the appellant had killed Jennifer. The appellant confronted Hiltz about the rumours and told him that once it was all done and over with, he and Hiltz were going to have a talk and it was not going to be a friendly talk.

[122] The Crown in closing said the following about the evidence of the appellant's conduct towards Doug Hiltz:

And you look at how he approached Doug Hiltz, a fellow he'd drunk tequila with the day before, when he knows that the police were over there and . . . and Hiltz candidly said well sure he'd be mad at me. I said anybody's capable of a homicide. But then later, when he hears Hiltz may have been starting rumours, he says well we're going to . . . we're going to talk about this later, and it's not going to be nice. [page294]

[123] In his charge to the jury, the trial judge repeated Hiltz's evidence regarding his two encounters with the appellant and instructed them as follows:

It is for you to say whether these things were said and, if so, what importance, if any, attaches to them. Take all the circumstances into account. Question the reasonableness of the reactions attributed to Mr. Bennett.

All of this evidence is for you to consider. I must say I would suggest, in my submission to you, it is not very hard to understand why someone would be upset if the police had been told by a neighbour of theirs that well, he might have done it because anybody can do it if they have their buttons pushed, and anyone might be upset if they find out somebody else has been spreading rumours about them being involved in a killing.

It may well be that not only anybody might be upset by these things. It may well be that an innocent person would be more upset than a guilty one. These are all matters for you to consider and upon which to apply your good sense and judgment.

(ii) Conversation about Ms Ueberschlag going to a mall

[124] Nicole Rowe testified that on the Sunday after Jennifer's death and before her body was found, she paged the appellant to ask if he had heard from Jennifer. The appellant called her about 15 minutes later and said that he went by to drop off a television for Jennifer on Friday at around 7 p.m. and that no one answered. In response to questions put to her in cross-examination, Rowe recalled the appellant telling her that he had heard from Michelle that she thought she had seen Jennifer on a bus going to the mall. Rowe admitted that the conversation about Michelle could have occurred in a second phone call, but she remembered it being in the same call. She testified that the appellant did not tell her anything more about Michelle, or what her last name was, or who she was.

[125] Michelle Klobucar, another Crown witness, was a friend of the appellant's who testified that she had never known of Jennifer and that the appellant had not called her to ask if she had seen her.

[126] In his closing address, Crown counsel said:

Was that just an instance of Bennett feigning, faking, an interest in Jennifer to let it appear that he was concerned for her well being and making the inquiries that Nicole Rowe would so obviously want him to do, and that this call back in regard to talking to a mutual friend was merely to convince Nicole of that fact?

Because if Bennett was the killer, of course, he already knew that Jennifer was dead. And so this would be pure pretext.

[127] The trial judge discussed the Crown's position on this evidence and instructed the jury as follows on how to use it: [page295]

The Crown asks you to infer that Mr. Bennett was lying when he allegedly told Nicole Rowe when she called him on the Sunday afternoon that he had talked to some girl named Michelle who said she had seen Jen going towards Fairview mall on a bus on the Friday. The Crown's position is that this Michelle must be Michelle Klobucar and that Michelle Klobucar did not testify to any such conversation with Mr. Bennett but rather said that when she spoke to him on the Saturday, he did not have much time to speak to her and sort of brushed her off.

It is for you to say whether or not these remarks were made. It is for you to say what weight, if any, to attach to them. The Crown's requested inference is that Mr. Bennett said this just to sort of throw everybody off the track and make him look as though he was concerned when in fact he had been the killer.

I do not know what benefit there would be to telling somebody that you had been told that Jen was seen during the day Friday when that is before the time of the murder. It does not seem relevant in terms of trying to suggest to anyone that that means the murder did not occur or anything like that.

These are matters that you are asked to consider and that you are asked to use as a basis upon which to draw an inference. As I have said now several times, consider all of the circumstances and my instructions to you on how to handle circumstantial evidence.

(iii) Failure to make certain phone calls

[128] In his closing, Crown counsel said the following about the appellant's failure to telephone certain people upon learning of Jennifer's death:

And in terms of his reaction, does it make a great deal of sense when Nicole Rowe had phoned him up asking if he has some . . . seen Jennifer around, when Bulldog had asked him to look after Jen? And my friend brought out in cross- examination that he could have found out Bulldog's phone number and that type of thing, that he doesn't talk to Nicole Rowe again after that, doesn't say what happened Nicole? Or I just found out, this is terrible. Or that he doesn't call Bulldog and let him know what's happened, the fellow that introduced them, and the fellow who asked him to look after Jennifer, and he makes no reference to that? Does that seem at all logical?

[129] The trial judge dealt with this evidence in his charge as follows:

Lastly, the Crown argues that the accused man did not act appropriately after the death by not phoning certain people, for instance, Bulldog, who was as you will know Jen's street dad before he moved to Chatsworth, and the inference you are asked to make is that it would be reasonable that Mr. Bennett would have called Bulldog, that Mr. Bennett would have called Nicole Rowe, that Mr. Bennett would have asked people about what had happened.

Those are all matters for you to consider. Consider, on the basis of the evidence you have, what reasonable inferences arise, if any, from such evidence. Consider whether or not any explanation suggests itself to you from [page296] all of the circumstances. Again, it is for you to say what, if any, inferences you are prepared to draw from such evidence, but you should only do so if in your view the requested inference is a reasonable one arising from all of the evidence.

You will bear in mind that a person might, when faced with circumstances such as you may feel were facing Mr. Bennett at the time, namely that it is clear that in respect to some of these matters he by that time had been questioned by police, indicated that he was a suspect, told that he was not believed. It may well be that there are some elements of panic or embarrassment or fear of false accusation or some other such considerations that should be taken into account.

(iv) Utterances to the police on May 11, 1992

[130] Crown counsel in his closing attached much significance to the appellant's reaction to a call from Detective Close on Monday, May 11, 1992 advising that he would like to talk to the appellant about the death of a friend of his, the identity of whom he testified he did not disclose over the phone. Crown counsel asked the jury to infer guilt because, although the appellant asked Det. Close who the deceased was, he did not ask specifically whether it was Jennifer. Counsel's argument on the point is quite lengthy, but particularly significant are the following passages:

When a police officer calls you up and says I want to talk to you about the death of a friend of yours, well that narrows the indeterminate class presumably to friends. And if Bennett was concerned about Jen, as he indicated in his statement to the police, if he called her Saturday, had called her Friday, and now you're getting a call from Sunday night from somebody that obviously wouldn't be particularly fond of you, Nicole Rowe, who's asking if he's seen her, and obviously worried about her, and he says he's worried, what would the first reaction be when a police officer calls? It's not Jennifer. . . .

But wouldn't you at least twig to that and say something, please don't tell me it's Jennifer? I've been worried about her. Her friend, Nicole's worried about her. Unless you already knew exactly who it was . . .

But the lack of curiosity on his part as to what happened to her seems inconceivable. . . . Wouldn't you want to know something about what happened? The police are calling you about the death of a friend of yours. What would be the first and most logical thing you would want to ask? What happened?

[131] The trial judge instructed the jury on the use of this evidence as follows:

As with the written statements, it is for you to determine whether these things were said. It is for you to determine, if said, whether there is anything about them that warrants the drawing of any inference as requested by the Crown. In so considering, take into account any explanation which in your view reasonably arises from the evidence, having regard to all of the circumstances. It is for you to say what weight, if any, is to be attached to this evidence. [page297]

(v) Appellant's desire to have copies of statements made to police by Meekison and Stewart

[132] Justina Meekison testified that the appellant approached her and said he wanted to get a copy of the statements that she and Dan Stewart had given to police.4 She recalled telling the appellant that he was able to see the statements from his lawyer. She further recalled the appellant saying that his lawyer needed the statements from her because the police could change what she had said. She testified that the appellant told her that he didn't do it and the only reason he was a suspect was because he had charges against him.

[133] Crown counsel said the following about this encounter to the jury:

And Bennett's been told that Meekison supports his story, and he's been asked by the police officers to flesh out the rest of the weekend. We need to alibi you for the rest of the weekend. So why does he go back to Meekison on May the 20th in order to get her statement to the police, and that of her boyfriend, Dan Stewart? Why focus on her? Because the killer knows that the important time is Friday night, Saturday morning. Not Saturday. Not Sunday.

But when Bennett's been told that you're okay, Ms. Meekison supports your story. She thinks you're a good guy on the . . . when they come to talk to him on May 14th, why go back and talk to her? Why worry about the police changing his statement? She says to him, well can't your lawyer get this? He says, well no, I have to, the police could change it around, presumably to suit their purposes.

Why go back to her unless he's really worried about the people he's with Friday night, and not to Russell, whom he . . . could alibi him for Saturday and Sunday as well, unless he knew when Jennifer was killed?

[134] In his charge, the trial judge told the jury that the Crown asked them to draw an inference adverse to the accused based on the evidence that he asked Meekison for copies of her statements to the police and those of her boyfriend Stewart. He repeated Meekison's evidence and then instructed the jury as follows:

It is for you to say whether these things were said. It is for you to consider all of the circumstances, consider any explanations that arise as to why things might be said. It is for you to determine whether these comments mean anything or merit the drawing of any inference. [page298]

. . . You may well be of the view that it is not surprising that Mr. Bennett would want to see what a witness or witnesses has told the police about an event in which he is alleged to have been involved, given that the police had already told him that they did not believe his denial.

These are matters for you to consider and upon which to apply your good judgment and common sense. You are entitled to draw the requested inference if you think it to be a reasonable one, having considered all of the circumstances.

(vi) Letter opener

[135] Following the murder, rumours were circulating in the community that the victim's silver letter opener with a grim reaper on the handle, which was missing from her apartment, was the murder weapon. A photo of a similar letter opener was in the local newspaper on June 23, 1992. The missing letter opener was never recovered.

[136] In August 1992, police purchased a similar letter opener. They showed a photograph of it to Renee Biddiscombe, the appellant's girlfriend at the time of the murder, who was 15 years old and heavily addicted to crack cocaine and heroin. Police informed Ms Biddiscombe that they were investigating the murder of Jennifer Ueberschlag and told her that the appellant was the prime suspect in the killing. Police asked if she had seen the letter opener in the appellant's possession. They also told her that the appellant was married and showed her a copy of his marriage licence.

[137] Ms Biddiscombe testified that she was upset to learn that the appellant was married. She also testified that she assumed the letter opener was the murder weapon and confronted the appellant. She told him that the police thought he was the prime suspect in the killing and that they showed her the photo of a letter opener with a grim reaper handle. His response was that the police were trying to frame him and break them up.

[138] Ms Biddiscombe further testified that one morning in November 1992, when "she was very strung out", police stopped her and showed her a grim reaper letter opener in what looked like an evidence bag. They asked if she had seen it before, and she said only in the photo that they had shown her before. Police told her that the victim "didn't die pretty".

[139] Ms Biddiscombe said she was upset and went to see the appellant, who was in jail on unrelated charges. She told him that the police had shown her the letter opener. Ms Biddiscombe testified that after telling the appellant this, he replied "how the fuck did they find that, how the fuck did they find that?", looked very shocked and became agitated. She asked him if he didn't do [page299] it, why was he so upset and he remarked that the police "are trying to frame me and, now they've found that, now they can".

[140] The Crown in his closing referred to Renee Biddiscombe's evidence as "perhaps the most important evidence of all" and, later, as "absolutely a critical piece of evidence". He went into considerable detail in his closing about this evidence, but the most pertinent passages are the following:

But it's Mr. Bennett's reaction that's critical to this. What he does is he goes how the fuck did they find that? How the fuck did they find that? Not, where did they find that. For somebody who might have been worried about being framed for something he didn't do, it's a pretty significant distinction. . . .

He says how the fuck did they find that? In other words, I thought I'd hidden it so well, it was gone forever. How the fuck did they find that?

Now, why would anybody make the link between the finding of the letter opener, and him being in trouble, and him being framed, unless they knew what the link was? Why would he automatically assume that the finding of the letter opener would in any way shape or form, connect him to the crime, unless he had taken it and he knew its significance . . .

[141] The trial judge put to the jury the Crown's position that the appellant's reaction supported the inference that he had taken the letter opener and hidden it. He also repeated the defence position that there was nothing to this evidence given the rumours on the street that the letter opener was supposedly tied in with the killing and that the appellant had said to people on many occasions that he was worried about being framed by the police. The trial judge then said:

These are all matters for you to consider. Consider all of the circumstances. Consider what inferences reasonably arise from those circumstances and be careful in drawing inferences in the sense that you are taking that extra step of making sure that the inference you are requested to draw is indeed a reasonable one arising from the circumstances.

III. Analysis of the charge on the after-the-fact conduct evidence

[142] Counsel for the appellant submits that if the trial judge had left only some of this after-the-fact conduct evidence to the jury, his error might have been harmless in isolation. However, in leaving all these items to the jury as evidence from which they could infer consciousness of guilt, the cumulative effect was highly prejudicial, particularly in a case where the Crown's case was entirely circumstantial.

[143] The respondent acknowledges that the first and third items of evidence relied on by Crown counsel as circumstantial evidence of guilt were weak arguments (i.e., the appellant's anger at being described by Doug Hiltz as someone who could kill and [page300] his failure to make certain telephone calls) and that the reliance on the fourth item was a doubtful argument (i.e., the appellant's failure to ask police if Jennifer was the victim in the May 11th phone conversation). According to the respondent, it was open for the Crown to argue and for the jury to conclude that the remaining items of evidence constituted evidence of consciousness of guilt.

[144] The trial judge did not have the benefit of this court's recent decisions on consciousness of guilt evidence in R. v. Diu (2000), 2000 4535 (ON CA), 49 O.R. (3d) 40, 144 C.C.C. (3d) 481 (C.A.), R. v. Levert (2001), 2001 8606 (ON CA), 159 C.C.C. (3d) 71 (Ont. C.A.) and R. v. Baltrusaitis (2002), 2002 36440 (ON CA), 58 O.R. (3d) 161, 162 C.C.C. (3d) 539 (C.A.). In Diu, Sharpe J.A. observed, at p. 74 O.R., p. 519 C.C.C. that, "[i]t has long been recognized that in certain circumstances, the conduct of an accused after a crime has been committed may provide circumstantial evidence of the accused's culpability with respect to that crime". He went on to state, at p. 74 O.R., p. 520 C.C.C. that "it has also been repeatedly recognized that evidence of after-the-fact conduct is often highly ambiguous . . . There is often a serious risk that the jury may fail to consider alternative explanations for the after-the-fact conduct and erroneously infer guilt." Sharpe J.A. concluded:

It is, therefore, important for the trial judge to ensure, by careful jury instructions, that the jury does not misuse the evidence. Accordingly, there is a well-developed body of jurisprudence to the effect "that juries be carefully instructed that there may be alternative explanations for the accused's conduct and that, in such cases, the accused's conduct is not capable of supporting an inference of consciousness of guilt." See Jenkins, supra, at p. 471 [(1996), 107 C.C.C. (3d) 440 (Ont. C.A.)].

In general, the trial judge should instruct the jury that the evidence of the accused's after-the-fact conduct has only an indirect bearing upon the issue of guilt, and that the jury should exercise caution in inferring guilt because the conduct might be explained in an alternative manner: Arcangioli, supra, at pp. 299-300 [(1994), 87 C.C.C. (3d) 289], citing Gudmondson v. The King (1933), 60 C.C.C. 332 (S.C.C.). The trial judge should also instruct the jury that the evidence of the accused's after-the-fact conduct can only be used to support an inference of guilt where they have rejected any innocent explanation for the conduct: Peavoy, supra, at p. 238 [(1997), 117 C.C.C. (3d) 226 (Ont. C.A.)].

[145] In Baltrusaitis, supra, the court considered whether the trial judge erred in leaving with the jury evidence of the accused's demeanour on learning of his brother's death, his failure to ask questions about the circumstances of the death and his failure to inform police of a recent meeting with his brother, as after-the-fact conduct capable of supporting an inference of guilt. Moldaver J.A. wrote, at p. 182 O.R., pp. 561-62 C.C.C.: [page301]

I agree with the appellant that the three impugned items of evidence should not have been left to the jury as after-the- fact evidence capable of supporting an inference of guilt because the probative value of this type of evidence is highly suspect and easily misinterpreted. The point was recently addressed by this court in R. v. Levert (2001), 2001 8606 (ON CA), 159 C.C.C. (3d) 71 (Ont. C.A.), at p. 81 where Rosenberg J.A. stated as follows:

The probative value of this type of evidence [unusually calm reaction by the accused upon being confronted with an allegation of sexual abuse] is highly suspect. In the two recent cases of Susan Nelles and Guy Paul Morin use of the accused's demeanour was found to have played a part in the wrongful prosecution. The Report of the Commission on Proceedings Involving Guy Paul Morin, 1998, vol. 2, pp. 1142 to 1150, contains an extensive discussion of the dangers of admitting such demeanour evidence. The expert and other evidence introduced at the Commission strongly suggests that this evidence can be highly suspect and should be admitted at a criminal trial with caution. Perceptions of guilt based on demeanour are likely to depend upon highly subjective impressions that may be difficult to convey to the jury and in any event the significance of the reaction will often be equivocal.

(footnotes omitted)

The concerns expressed by Rosenberg J.A. apply with equal force to this case. In my view, rather than leaving the impugned items of evidence to the jury as evidence capable of supporting an inference of guilt, the trial judge should have told the jury to ignore them. With respect, his failure to so instruct the jury constituted error.

(Emphasis added)

[146] As explained in Diu, it has long been recognized that the accused's after-the-fact conduct can give rise to a circumstantial inference of guilt; however, it is incumbent on the trial judge to ensure by careful instruction that the jury does not misuse such evidence. Moreover, as explained in Levert and Baltrusaitis, there are some types of post-offence conduct evidence that the trial judge is required to remove from the jury's consideration altogether.

[147] In my view, as I stated earlier, the trial judge should not have left the following items of demeanour evidence with the jury as being capable of supporting an inference of guilt: the appellant's anger at being described as someone who could kill, the appellant's failure to call certain people after Jennifer's killing and the appellant's failure to ask police if the victim was Jennifer. The trial judge ought to have told the jury that these items of evidence were not capable of supporting the inference of guilt urged by the Crown and that they should ignore them. These items of demeanour evidence are of the type that this court in Levert and Baltrusaitis described as having highly suspect probative value and are easily misinterpreted.

[148] With respect to the remaining items of after-the-fact conduct evidence, in my view, it was open to the trial judge to leave [page302] these items with the jury. However, the trial judge ought to have instructed the jury that the after-the-fact conduct evidence relied on by the Crown has only an indirect bearing upon the issue of guilt, and that the jury should exercise caution in inferring guilt because the conduct might be explained in an alternative manner. In addition, he ought to have instructed the jury that they must not use this conduct to support an inference of guilt unless they rejected any innocent explanation for the conduct: Diu, supra.

[149] Such a carefully worded instruction was required because two of the three remaining items of evidence had limited probative value on the issue of the identity of the killer. Regarding the evidence that the appellant asked for copies of Meekison and Stewart's statements to police, Crown counsel's argument was that the appellant asked Meekison for her statement because he knew that the important time was Friday night. The Crown's argument ignores that Meekison and Stewart were not providing an alibi for the appellant at the time of the killing, but only for earlier that evening. Moreover, there was no evidence to the effect that the appellant was asking Meekison to change her story or to assist him in providing an alibi for the time of the killing. While there may be some probative value in the appellant's act of asking a witness for statements made to police, in my view, the trial judge needed to more carefully instruct the jury on the limitations of this evidence as a basis for drawing the inference requested by the Crown. It was not enough to simply instruct the jury that they were entitled to draw the requested inference if they thought it was a reasonable one.

[150] The trial judge indicated in his charge that the appellant's statement about someone named Michelle having possibly seen Jennifer on the bus going to Fairview Mall on Friday afternoon was of questionable relevance to the identity of the killer. The trial judge nonetheless left it as evidence for the jury to consider and to decide whether to draw the inference that the appellant was trying to look concerned when in fact he was the killer. In my view, the jury should again have been told to be very cautious in drawing this inference, particularly considering that it had not been established which "Michelle" the appellant was referring to in his statement to police and considering that it was clearly established that Jennifer was alive at the time of the reported trip to the mall.

[151] The evidence of the appellant's reaction to the finding of the letter opener was characterized by Crown counsel as being perhaps the most important piece of evidence and absolutely critical to its case. In my view, it was incumbent on the trial judge to [page303] carefully instruct the jury of the need for caution in using this evidence to draw an inference of guilt. The trial judge repeated to the jury the defence position that an innocent explanation of the evidence of the appellant's reaction was available given the widespread (albeit erroneous) rumours that the letter opener was the murder weapon and in light of his belief that the police were trying to frame him for the killing. The trial judge ought to have gone further and instructed the jury that they must not use this conduct to support an inference of guilt unless they rejected any innocent explanation for the conduct.

[152] The respondent submits that if it is found that the trial judge ought not to have left any of these items with the jury as potential inculpatory after-the-fact conduct, this court should apply the curative proviso in s. 686(1)(b)(iii) of the Criminal Code, R.S.C. 1985, c. C-46. I am not prepared to do so. The Crown's case was entirely circumstantial and the after-the-fact conduct evidence was a critical aspect of its case against the appellant. The jury was improperly permitted to make a finding of a concocted alibi and to draw an inference of guilt therefrom. As well, three items of after-the-fact conduct evidence were improperly allowed to go to the jury as inculpatory evidence and three other items were put to the jury without an adequate cautionary instruction. I am unable to say in light of these errors, and also in light of the errors that I previously identified in connection with the hair comparison evidence and the identification evidence, that the result of the trial would necessarily have been the same.

Issue 5: Any Verdict of Culpable Homicide is Unreasonable

[153] Counsel for the appellant submits that any finding that the appellant killed Jennifer Ueberschlag is unreasonable. At the close of the Crown's case, defence counsel brought a motion for a directed verdict of acquittal on the basis that there was no evidence upon which a reasonable jury properly instructed could conclude that the appellant was the person who caused the death. In the alternative, defence counsel moved for an order directing that the appellant be acquitted of first degree murder and that the case proceed as one of second degree murder.

[154] On the motion for a directed verdict, the trial judge observed that in a case involving circumstantial evidence, the assessment whether or not there is a rational explanation for the circumstantial evidence other than the guilt of the accused is a question for the jury, citing the majority's opinion in R. v. Charemski, 1998 819 (SCC), [1998] 1 S.C.R. 679, 123 C.C.C. (3d) 225, at pp. 683-84 S.C.R., pp. 229-30 C.C.C. After reviewing the evidence at some [page304] length, the trial judge concluded that given the possible inferences that could be drawn from the evidence, it was possible that the jury could find that the appellant was the attacker.

[155] I am not able to say that the trial judge erred in this assessment. Although this was not a strong circumstantial case, there was evidence capable of supporting the inference that the appellant killed Jennifer. I would not give effect to this ground of appeal.

Issue 6: The Verdict of First Degree Murder is Unreasonable and Unsafe

[156] Counsel for the appellant submits in the alternative that the verdict of first degree murder is unreasonable. In dismissing the defence motion for an order directing that the case proceed as one of second degree murder only, the trial judge concluded that there was some evidence upon which a properly instructed jury acting reasonably could find that the killing occurred while the killer was committing or attempting to commit one of the forms of sexual assault or forcible confinement delineated in s. 231(5) of the Criminal Code.

[157] Again, I am not prepared to interfere with this finding. Although the evidence that a sexual assault had occurred was not overwhelming, there was some evidence upon which a properly instructed jury acting reasonably could conclude that there was a sexual assault during which the victim was killed. The trial judge identified this evidence as the accused's interest in the victim, the finding of the trace of semen, the body was nude, the state of the bedroom and the finding of hairs consistent with pubic hairs of the accused on the nude body.

[158] I reach the same conclusion with respect to the issue of unlawful confinement. The trial judge was mindful that evidence of assaultive behaviour does not automatically equate to unlawful confinement. He concluded that it was open to the jury to conclude that the killer had assumed control over the liberty of the deceased in a way over and above that which would flow as purely part of a straight assault. I am not prepared to interfere with that finding, given the nature of the injuries to the deceased, and particularly the multiple superficial cuts to the neck, the significant bruising to the face and upper body and the evidence of asphyxiation.

CONCLUSION

[159] For the reasons given, I am satisfied that the appellant is entitled to a new trial. Accordingly, I would allow the appeal, [page305] quash the conviction and order a new trial on the charge of first degree murder.

Appeal allowed.

Notes

[^1]: Trial judge's Pre-trial Ruling on the Hair Comparison/ DNA Evidence at p. 44.

[^2]: The hairs sent for testing had apparent tissue associated with the root and were thus more likely to reveal a DNA profile.