DATE: 20051130
DOCKET: C38022
COURT OF APPEAL FOR ONTARIO
DOHERTY, MOLDAVER and ARMSTRONG JJ.A.
B E T W E E N :
HER MAJESTY THE QUEEN
Alan D. Gold for the appellant
Respondent
- and -
IVAN KIRK KLYMCHUK
Renee Pomerance and Rick Schwarzl for the respondent
Appellant
Heard: September 22, 2005
On appeal from a conviction on a charge of first degree murder returned by a jury presided over by Justice B. Wein of the Superior Court of Justice on December 8, 2000.
DOHERTY J.A.:
I
Overview
[1] The appellant was convicted by a jury of the first degree murder of his wife. He received the mandatory sentence of life imprisonment without eligibility for parole for twenty-five years.
[2] The principal ground of appeal concerns the admissibility of expert evidence of “staging” offered by the Crown. I am satisfied that parts of the expert evidence should not have been admitted at trial. The Crown has not demonstrated that the verdict would necessarily have been the same had the expert evidence been properly limited. I would quash the conviction and direct a new trial.
II
Factual Background
[3] The deceased, Maria Klymchuk, had been married to the appellant for nine years. They had two young children. She was murdered some time between 10:00 p.m. and 11:00 p.m. on Easter Sunday 1998. The murder occurred in a drive shed about thirty feet northwest of the Klymchuk family home. Mrs. Klymchuk used the drive shed to train her dogs and had taken one of the dogs out to the drive shed for a training session at around 10:00 p.m. The home was situated on a one acre lot in a semi-rural area near Bolton, Ontario.
[4] The appellant telephoned 911 at about 11:00 p.m. and reported that he had found his wife unconscious on the drive shed floor with obvious head injuries. The appellant was instructed to perform CPR and emergency personnel were dispatched. When they arrived, it was obvious that Mrs. Klymchuk was dead.
[5] The post-mortem examination revealed that Mrs. Klymchuk had been struck on the neck and the back of the head several times. A bloodied axe found on the drive shed floor could have been used to inflict some of the head and neck injuries. There were also indications that pressure had been applied to Mrs. Klymchuk’s neck before she died. There were no signs of sexual assault and Mrs. Klymchuk’s jewellery, including her wedding ring and watch, was not taken.
[6] The police interviewed the appellant three times between Easter Sunday and June of 1998. The appellant gave exculpatory statements. He was arrested in December 1998.
[7] The Crown’s case rested primarily on evidence of opportunity and motive. The appellant was admittedly at home when his wife was attacked and killed in the drive shed. The Crown argued that the appellant’s extra-marital relationship with a woman named Robin Mays provided the motive. The affair began in late 1997. Ms. Mays moved to the Toronto area in January 1998 and she and the appellant saw each other regularly. The appellant told Ms. Mays that he would divorce his wife so they could be together by the spring. In April of 1998, Ms. Mays told the appellant that she was considering a reconciliation with her former common law husband. On April 7th, she told him that she wanted to end their relationship. The appellant seemed upset, although he appeared to accept the decision. Within a day or two, the appellant called Ms. Mays and assured her that his wife had agreed to a divorce and that she planned to tell her parents on Good Friday. The appellant called Ms. Mays on Good Friday and told her that the meeting with the parents had been postponed because his wife was ill. Ms. Mays did not hear from the appellant again until after she learned of Mrs. Klymchuk’s death.
[8] The relationship between the appellant and Ms. Mays continued after Mrs. Klymchuk’s death. The appellant told Ms. Mays that he wanted to marry her. Ms. Mays became concerned that she might be considered a suspect in Mrs. Klymchuk’s murder, so she went to the police and told them about her relationship with the appellant. She agreed to keep the police informed of anything the appellant said to her that may have relevance to the murder. The telephone conversations between the appellant and Ms. Mays were surreptitiously intercepted pursuant to judicial authorizations granted in June 1998.
[9] In August 1998, Ms. Mays ended her relationship with the appellant, having decided to reconcile with her former common law husband. Ms. Mays had a brief chance encounter with the appellant in September and spoke to him on the telephone in November 1998. He told her that he still hoped to marry her.
[10] In addition to the evidence of motive and opportunity, the Crown relied on inconsistencies between the appellant’s account of the events following his discovery of his wife’s body and the physical evidence. Specifically, the Crown argued that the appellant’s statement to the authorities that he performed CPR on his wife for some twenty minutes was false given the virtual absence of any blood on the appellant’s face, hands and clothing, or in the area of the kitchen around the telephone the appellant had used to place calls to 911 after he had performed CPR on his wife’s bloodied body.
[11] The Crown also relied on evidence that the crime scene had been staged to make it appear as though someone had broken into the drive shed through a window. The Crown argued that the apparent break-in had been staged by the appellant to mislead the police into thinking that his wife had been killed by a burglar who was caught in the act by Mrs. Klymchuk. Expert evidence from Special Agent Allan Brantley of the Federal Bureau of Investigation (“F.B.I.”) figured prominently in this part of the Crown’s case.
[12] The appellant testified and denied that he had anything to do with his wife’s death. His testimony was consistent with the statements he gave the police. The appellant acknowledged the affair with Robin Mays, but claimed that he never intended to leave his wife, but was only “stringing Robin along” because he wanted to continue the sexual relationship.
[13] According to the appellant, his wife went out to the drive shed to put one of their dogs through a training session at about 10:00 p.m. The appellant remained in the house. About an hour later when his wife had not returned to the home, he went to the shed looking for her. The door to the shed was locked. The appellant used his keys to gain entry and found his wife face down on the floor. He reached down and gave her a light shake but got no response. He returned to the home and called 911. He then returned to the drive shed with his cell phone and performed CPR for several minutes. He returned to his home at least twice to speak to the 911 operator.
[14] The defence relied on the absence of any forensic evidence implicating the appellant and the absence of any incriminating statements by the appellant despite months of audio surveillance. The defence contended that the crime scene was consistent with the conclusion that a burglar had broken into the shed, come upon Mrs. Klymchuk and killed her.
III
The admissibility of the expert evidence of “staging”
(a) The law
[15] This court has considered the admissibility of expert evidence that a crime scene was “staged” in R. v. Ranger (2003), 2003 32900 (ON CA), 178 C.C.C. (3d) 375 (Ont. C.A.) and R. v. Clark (2004), 2004 12038 (ON CA), 182 C.C.C. (3d) 1 (Ont. C.A.). In the present context, staging refers to a deliberate alteration of the crime scene by the perpetrator of the crime to mislead the police as to the identity of the perpetrator. The decisions in Ranger and Clark go a long way to resolving the admissibility of the expert evidence tendered in this case. The trial judge, whose reasons reflect a careful consideration of the issues before her, did not have the benefit of this court’s reasons in either Ranger or Clark.
[16] In Ranger, Charron J.A distinguished between expert evidence of crime scene analysis offered to assist the jury in deciding what had happened at the crime scene (the WHAT question) and expert evidence based on crime scene analysis offered to assist the jury in deciding why a crime was committed (the WHY question), or who committed the crime (the WHO question). She observed at paras. 71 and 72:
Crime scene analysis (which I find useful to label as the “WHAT” referred to earlier) results in many forms of expert opinion evidence that regularly meets the legal requirements for admissibility. A few examples readily come to mind: an expert’s opinion in an arson case that a fire was not accidental but, rather, deliberately set; opinion evidence explaining the significance of blood splatters; a pathologist’s opinion about the likely cause of death or of injuries observed on a deceased victim; an expert’s opinion on how a motor vehicle accident happened. There are many more examples. This kind of evidence assists the trier of fact in understanding WHAT the crime scene shows. The admissibility of that kind of evidence will usually turn on questions of relevance or the witness’s particular expertise. Of course, issues may also arise under any other aspect of the Mohan test. However, the scientific basis for this kind of evidence is usually not contentious.
By contrast, attempts to adduce expert opinion evidence about WHY an offence was committed in a particular manner and, more particularly, about WHO is more likely to have committed the offence, that is, the kinds of evidence that I have labelled more particularly as criminal profiling, have generally not met with success, either in this jurisdiction or elsewhere. I will refer to some American and English jurisprudence later in this judgment. However, in so far as Canadian jurisprudence is concerned, the best example is Mohan itself [emphasis added].
[17] In Clark at paras. 83 and 84, Moldaver J.A. applied the distinction drawn in Ranger. He characterized expert evidence addressed to the WHAT question as “crime scene reconstruction evidence”. That evidence was potentially admissible if relevant and offered through a properly qualified witness. He characterized expert evidence aimed at answering the WHO or the WHY questions as “impermissible criminal profiling evidence”.
[18] The distinction drawn in Ranger and Clark between expert evidence based on crime scene observations and reconstructions on the one hand, and expert evidence based on assessments of the type or category of person who committed a particular offence on the other, is rooted in the established reliability of the former and the unproven nature of the latter. Expert evidence describing the type of person who would or would not engage in a particular criminal activity is admitted only in that narrow class of cases where psychiatric knowledge has advanced sufficiently to allow a qualified expert to testify that the crime in question would in all likelihood have been committed by a person with a particular mental makeup, or alternatively that a person with the accused’s mental makeup would in all likelihood not have committed the offence. Psychiatric profiling evidence is admissible only if the profile relied on by the expert is sufficiently precise and detailed to provide a meaningful distinction between the group said to fit the profile and the rest of the population. The evidence must also demonstrate that the profile has been developed and confirmed through an application of proper scientific methodology: R. v. Mohan (1994), 1994 80 (SCC), 89 C.C.C. (3d) 402 at 423 (S.C.C.); R. v. J.(J.L.) (2000), 2000 SCC 51, 148 C.C.C. (3d) 487 at 500-507 (S.C.C.).
(b) The argument
[19] Mr. Gold in his able argument accepts, albeit with little enthusiasm, the law as developed in Ranger and Clark. He submits that to the extent that the expert evidence of staging offered in this case addressed the WHAT question, it should have been excluded because its manner of presentation produced prejudice far beyond any potential probative value. Mr. Gold argues that to the extent that the opinion had relevance to the jury, it amounted to little more than common sense inferences which could be drawn from the physical evidence found at the scene. He contends that by eliciting the evidence through an expert with long and impressive credentials, the Crown was attempting to elevate possible inferences to the standard of scientific truths.
[20] Alternatively, Mr. Gold argues that if the expert evidence going to the “WHAT” question was admissible, the expert’s testimony went beyond that question and included evidence of impermissible criminal profiling. Mr. Gold argues that the improper admission of the expert’s evidence going to the WHO and the WHY questions caused significant prejudice to the appellant requiring a new trial.
[21] I will address Mr. Gold’s second argument first. In my view, the expert’s evidence, while concerned primarily with the WHAT question, did seep into the prohibited area of criminal profiling identified in Ranger and Clark.
(c) The evidence
[22] Special Agent Brantley had worked for the F.B.I. for seventeen years. He had been trained in and become an expert in what he described as criminal investigation analysis. Agent Brantley testified that criminal investigation analysis is an umbrella designation referring to a number of investigative services offered to police agencies to assist them in their investigations and sometimes offered to the courts as expert evidence. These investigative services included profiling and crime scene analysis. Agent Brantley testified that when performing a crime scene analysis, he was always concerned with whether the scene had been staged or manipulated to create a misleading impression. He defined staging as:
The intentional alteration or manipulation of the crime scene by the offender to divert attention away from that individual as a logical suspect and/or to divert attention away from the most logical motive.
[23] In response to a long hypothetical, that was rooted in the evidence, Agent Brantley opined that the scene in the drive shed had been staged to make it appear as though there had been a break-in. He emphasized that his opinion was based on the combined consideration of many circumstances and not on any one factor. He testified that there were many “behavioural, forensic and investigative contradictions” that told him there had not been a real break-in, but rather an attempt to make it appear as though there had been a break-in.
[24] The factors referred to by Agent Brantley can be grouped into five categories. First, he opined that the drive shed was a “high risk” target for a break-in (in the sense that the risk to a burglar of being caught was substantial), and not one likely to be selected by a burglar. This statement of the risk posed to a burglar was based on many considerations, including the ample lighting around the drive shed, the locks on the doors, the alarm system, the presence of dogs in the home, the activity in the area of the home on that evening, the Klymchuks’ presence in the home that evening, the close proximity of neighbours, the relative unlikelihood that there would be valuable, easily portable property in the drive shed, and the availability of easier targets in the vicinity.
[25] Second, Agent Brantley described Mrs. Klymchuk as at a very low risk to be the victim of crime. She was security conscious, lived in a good neighbourhood and did not engage in any activities, such as drug dealing, that would make it more likely that she would become a victim of crime.
[26] Third, based on a statistical review of break and enters in the United States and in the five years between June 1995 and October 2000 in the area of the Klymchuks’ home, Agent Brantley concluded that confrontations between the burglar and the victim were rare, and that in those rare cases where a confrontation occurred, there was seldom any violence directed at the victim. The burglar preferred to flee the scene. Brantley said:
When you consider the incidents where contact is made between a burglar and the resident, when that happens for there to be a confrontation or the offender remains in the area and does not flee immediately, that is very rare. Even more rare is when a confrontation does occur and it turns violent. It is even more rare when that violent confrontation is also murder or homicide.
[27] Brantley testified that his review of the five years for which he had statistics of break and enters in the area of the Klymchuk home revealed no other case where a break and enter had resulted in a homicide.
[28] Fourth, Brantley considered the nature of the violence inflicted on Mrs. Klymchuk, the absence of any evidence of sexual assault, or theft from her person, and the indication that the perpetrator had quickly gained control over Mrs. Klymchuk in a confined area as contra-indicative of homicide by an unknown intruder.
[29] Fifth, Agent Brantley focused on the window which was the apparent point of entry by the burglar. In Brantley’s view, a burglar would not have entered the drive shed through that window. The window was in plain view and there were other, less exposed windows in the drive shed. The cutting of the screen on the window also, according to Agent Brantley, seemed unnecessary to gain entry through the window since the screen could be easily removed. The location of the cut on the screen and the manner in which the cut was made were viewed by Agent Brantley as inconsistent with the screen having been cut by someone who was trying to gain entry through the window. He also noted that the area around the window where the screen had been cut did not show any indications of entry through that window (i.e. fingerprints). Finally, Brantley found it significant that the window had been left open. In his opinion, a burglar would close the window after the burglar gained entry to avoid the risk that the open window could attract someone’s attention.
[30] While I will have more to say about parts of this evidence later in these reasons, for the moment I observe that this evidence was all directed at the WHAT question and Brantley’s opinion that this was not a break-in, but an attempt to make it appear as though there had been a break-in.
[31] At the end of his examination-in-chief, Brantley gave a series of answers which Mr. Gold contends went beyond the permissible limits expert evidence of staging. I will quote those questions and answers in full:
Q. Sir, you indicated in this case that one of the factors that you considered as one of the more highlighted ones, and you pointed it out initially, was the apparent maximum human injury or human loss to the minimal property loss, did I understand that as being your evidence?
A. That’s correct.
Q. What, if any significance, does that have to do from a crime scene analysis point of view?
A. It is important in terms of what was the focus of the offender. Was the focus of the offender assault and killing of the victim and was there more time spent accomplishing those acts than any other acts that we assessed at the scene.
Certainly we consider what was done to the victim. The numerous forms of trauma and the length of time that that process took. You compare that with the rather minimal movement of that snow blower. Clearly, that the focus in this particular situation was on her and not on that piece of equipment.
Q. Is there a name for that sort of focus?
A. Well, we would refer to this, because there was no indication of sexual activity or that the sexual parts of the body were traumatized or any semen or sperm or body fluids present and the fact that there was nothing taken from the scene we would describe this or classify this as a “personal cause homicide”.
Q. And what does a personal cause homicide mean?
A. Well, the victim is killed, because of who that victim is, he or she and not necessarily because of what that victim possesses. This generally includes reasons of revenge, anger, elimination of an obstacle to a goal. Those kinds of things are part and parcel to the personal cause homicide.
Q. Agent Brantley are you aware of any homicide case that you have been involved with in which staging was found to exist wherein the victim and the offender were strangers?
A. I’m aware of none [emphasis added].
[32] The answers quoted above provide Brantley’s opinion as to the possible motives for the murder and offer his opinion that there was a prior relationship between Mrs. Klymchuk and her killer. These answers are directed at the WHY and the WHO questions and not the WHAT question. They offer Agent Brantley’s opinion that the killer was someone who knew Mrs. Klymchuk and had a personal motive for killing her. That profile of the killer fit the appellant.
(d) The application of Ranger and Clark
[33] In Ranger, the expert was allowed to offer an opinion as to the killer’s motive, the existence of a prior relationship between the killer and the victims, and which of the victims was the true target of the killer. This court held that none of that evidence was properly admitted stating at para. 82:
Detective Inspector Lines’ [the expert] opinions about the perpetrator’s likely motivation for staging the crime scene and his characteristics as a person associated with the victims and having a particular interest in Marsha [one of the victims] constituted evidence of criminal profiling. Criminal profiling is a novel field of scientific evidence, the reliability of which was not demonstrated at trial. To the contrary, it would appear from her limited testimony about the available verification of opinions in her field of work that her opinions amounted to no more than educated guesses. As such, her criminal profiling evidence was inadmissible. The criminal profiling evidence also approached the ultimate issue in the case and, hence, was highly prejudicial [emphasis added].
[34] In Clark, the expert (the same expert who testified in Ranger) was allowed to advance the opinion that the killer knew the victims and was familiar with the residence in which the homicides occurred. In holding that this evidence went beyond the pale of permissible expert evidence of staging, this court said at para. 87:
To the extent that the Detective Inspector’s evidence about the phone and the lighting may have conveyed the impression that the offender was someone familiar with the layout of the Tweeds’ [the victims] apartment, it was offensive. She was not entitled to testify about the characteristics of the likely offender, characteristics which in this case comfortably fit with the appellant. That constituted criminal profiling evidence. As such, for reasons stated earlier, it was inadmissible and should not have been received [emphasis added].
[35] No meaningful distinction can be drawn between Agent Brantley’s answers quoted above and the evidence found to be inadmissible in Ranger and Clark. In all three cases, the evidence was offered to identify the killer by reference to the killer’s motive and his prior association with the victim(s). In each case, the accused fit the profile of the killer provided by the expert.
[36] There is nothing in the basis of Agent Brantley’s opinion that renders it inherently more reliable as expert evidence than the similar opinions rejected in Ranger and Clark. The Crown did not offer any evidence that Agent Brantley’s opinions as to the motives and prior connection between the killer and victim of those who stage a break-in in the course of committing a homicide, had been or could be tested according to the generally accepted scientific methodology identified in Daubert v. Merrell Dow Pharmaceuticals Inc. 509 U.S. 579 (1993) and quoted with approval in R. v. J.(J.L.), supra, at 501-502.
[37] Agent Brantley’s opinions as to the killer’s motive and prior relationship with the victim were not founded on any scientific process of inquiry, but on his own experience as augmented by his review of similar case files and interviews with incarcerated felons. Agent Brantley’s experience and review of the other sources led him to conclude that those who staged break-ins as part of a homicide probably had a personal motive for the homicide and probably had a prior association with the victim. Even if those opinions accurately reflect the statistical probabilities that a killer who stages a break-in as part of a homicide has a personal motive for the homicide and a prior relationship with the victim, conclusions based on statistical probabilities can offer no insight as to what happened in a specific case. For example, evidence from a homicide investigator that in his experience, his review of similar cases, and his interviews of killers, 85 percent of spousal homicides (a hypothetical figure) not involving a sexual assault or theft from the victim were committed by the surviving partner, could not be offered as evidence (expert or otherwise) that a specific spousal homicide was committed by the surviving partner. To borrow the words of Charron J.A. in Ranger, Agent Brantley’s opinion as to the killer’s motive and prior relationship with the deceased were “educated guesses” and not scientifically based opinions. As Charron J.A. indicated, those “educated guesses” can play a valuable role in the investigation of crime by directing the police to fruitful areas of investigation. They cannot, however, be admitted as evidence under the guise of expert opinion.
[38] The trial judge erred in law in allowing Agent Brantley to give opinion evidence as to the killer’s prior relationship with Mrs. Klymchuk and the possible motives for her killing. These errors were compounded by the failure to limit Agent Brantley’s opinions as to the WHAT question to evidence based on his examination and reconstruction of the crime scene.
(e) The proper scope of Agent Brantley’s evidence of “staging”
[39] Agent Brantley’s opinion was that the scene was staged to make it appear as though there had been a break-in when in fact there was no break-in. As indicated above, the factors relied on to support that opinion can be grouped under five headings:
• the drive shed was a “high risk” burglary target;
• Mrs. Klymchuk was at low risk to be a victim of crime;
• statistics relating to break and enters and the incidence of associated violence;
• observations of Mrs. Klymchuk’s body and the immediate vicinity in which it was found; and
• observations of the window with the cut screen.
[40] I think only the factors gathered under the fourth and fifth of these headings are properly viewed as crime scene reconstruction evidence upon which an expert could base the opinion that a crime scene was staged.
[41] The first two headings, the high risk nature of the drive shed as a target for burglary and Mrs. Klymchuk’s low risk as a target of crime, are in essence opinion evidence based on forms of profiling. Agent Brantley testified that the drive shed did not meet the profile of the kind of location likely to be the target of a break-in, and inferred from that assessment (in combination with other evidence) that the drive shed had indeed not been the target of a break-in. Similarly, Agent Brantley offered his opinion of the kind of person who was most often the victim of crime, noted that Mrs. Klymchuk did not fit that profile, and inferred, again in combination with other evidence, that she had not been the victim of a break-in.
[42] There is nothing in Agent Brantley’s evidence to suggest that his profiles of burglary locations or victims were any more scientifically based than his profile of persons who staged break-ins as part of a homicide. His evidence that places targeted by burglars fit a common profile that is sufficiently accurate and detailed to provide evidence that a burglary did or did not occur at a specified location cannot withstand even modest scrutiny. Burglars pick the places to be burgled. Burglars range from the drug addled teenager desperate for money to the sophisticated professional “second storey” man. It is difficult to imagine that the targets selected by this diverse population of burglars would have very much in common.
[43] Agent Brantley’s opinion that the drive shed was a “high risk” burglary target amounted to no more than his assessment of how an intelligent, careful burglar would assess the risks and benefits inherent in breaking into the drive shed. Even if this assessment had some validity if one assumed that the burglar was a careful, intelligent house burglar, it had no validity when that assumption was removed.
[44] Agent Brantley’s evidence that Mrs. Klymchuk did not fit the profile of a person likely to be the victim of a crime suffers from the same failings as his evidence concerning the likelihood of a burglary at the drive shed. The mere identification of factors that can make a person more susceptible to crime does not amount to an admissible profile of victims of crime according to the standards developed in R. v. J.(J.L.), supra. See also R. v. F.(D.S.) (1999), 1999 3704 (ON CA), 132 C.C.C. (3d) 97 at 111 (Ont. C.A.). Agent Brantley’s opinion as to the profile of victims of crime has not been subject to the kind of rigorous scientific scrutiny and review that is a prerequisite to the admissibility of this kind of opinion evidence.
[45] There is a further difficulty with permitting Agent Brantley to give expert evidence as to the profiles of burglary sites and victims of crime. Both profiles offered by Agent Brantley were based in large measure on inferences from facts that could be drawn by any reasonably intelligent person. For example, Agent Brantley’s opinion that the drive shed did not fit the profile of a good burglary target was based in part on the ample lighting around the drive shed, the dogs in the home, and the presence of an alarm system. It seems rather self-evident that these facts would make the drive shed a less attractive target to a burglar, or at least to a burglar who is aware of these facts.
[46] The third heading, the statistical evidence, also offered no basis for the opinion that this crime scene was staged. As I explained earlier, statistical evidence of probabilities based on prior similar events, while useful in many disciplines, offers no admissible evidence as to what happened on a specific occasion in a criminal trial.
(f) Can the curative proviso be applied?
[47] Having concluded that parts of Agent Brantley’s evidence were improperly admitted, the appeal must be allowed unless the Crown can demonstrate that the legal error did not occasion any substantial wrong or miscarriage of justice. In assessing the application of s. 686(1)(b)(iii), the curative proviso, I must consider the cumulative effect of the improper admission of opinion evidence as to the killer’s motive and his prior connection with the victim as well as the failure to limit the “staging” evidence to observations and inferences based on examination and reconstruction of the crime scene.
[48] Section 686(1)(b)(iii) eliminates the need for a new trial despite a legal error at the first trial where a new trial is not necessary to ensure the proper administration of criminal justice. If the legal error casts no doubt on the reliability of the verdict, or the essential fairness of the trial, justice is not served by a new trial. Where the essential fairness of the trial is not undermined by the legal error, s. 686(1)(b)(iii) can only be applied to dismiss an appeal where the appellate court is satisfied, based on a review of the entirety of the trial record, that there is no reasonable possibility that the verdict would have been different had the legal error not been made: R v. Bevan (1993), 1993 101 (SCC), 82 C.C.C. (3d) 310 at 328-329 (S.C.C.).
[49] Where the error lies in the improper admission of evidence, that error will be harmless within the meaning of s. 686(1)(b)(iii) if the improperly admitted evidence was so insignificant that it could not have affected the verdict, or if the Crown’s case apart from the improperly received evidence can be characterized as so overwhelming as to render a conviction almost inevitable: R. v. Khan (2001), 2001 SCC 86, 160 C.C.C. (3d) 1 at paras. 26, 31 (S.C.C.). If an appellate court finds legal error, it cannot pre-empt an accused’s right to a trial according to law by an appellate evaluation of the appropriate or even likely verdict: R. v. S.(P.L.) (1991), 1991 103 (SCC), 64 C.C.C. (3d) 193 at 199; R. v. C.(W.B.) (2000), 2000 5659 (ON CA), 142 C.C.C. (3d) 490 at 513 (Ont. C.A.), aff’d (2001), 2001 SCC 17, 153 C.C.C. (3d) 575 (S.C.C.).
[50] The Crown’s case, while formidable without the improper opinion evidence of Agent Brantley, cannot be described as so overwhelming as to render the improper admission of that evidence harmless. This is not a case like Clark, supra, at paras. 135, 138, where the defence evidence can be characterized as “preposterous” and the Crown’s case as “virtually insurmountable”.
[51] Nor can I describe the improperly admitted evidence as insignificant. It was central to the Crown’s case that the crime scene was staged to make it appear as though there had been a break-in when in fact there had been no break-in. A finding by the jury that the crime scene was staged would go a long way to the conviction of the appellant. On the evidence, he was the only person with a motive to kill Mrs. Klymchuk and a reason to stage the crime scene to make it appear as though there had been a break-in. Agent Brantley’s opinion evidence that the scene was staged was an integral part of the Crown’s theory. While I accept that he could give that evidence, he improperly bolstered his opinion with evidence that the drive shed was a “high risk” burglary target, evidence that Mrs. Klymchuk was at a low risk to be a victim of crime, and evidence based on statistics of prior break and enters and the incidence of violence associated with those break and enters. In my view, all of this evidence potentially gave his opinion that the crime scene was staged added weight. I cannot say that the jury would necessarily have found that the crime scene was staged had Agent Brantley’s evidence concerning the staging of the crime scene been properly limited.
[52] I am also satisfied that Agent Brantley’s evidence addressing the killer’s motive and the prior relationship between the victim and the killer could well have affected the jury’s verdict. Ms. Pomerance, for the Crown, in her excellent oral argument, contended that if the jury found that the crime scene was staged, Agent Brantley’s evidence as to the killer’s motive and the prior relationship between the victim and the killer added little to the Crown’s case. She submitted that as a matter of common sense, if the jury found that the crime scene was staged, it would infer that the killer was not a burglar and that the homicide was motivated by something other than the break-in. She further submitted, as a matter of common sense, that a stranger to the victim would have no reason to stage a break-in as part of a homicide. Ms. Pomerance closed the circle of her argument by contending that no one other than the appellant had a personal motive to kill Mrs. Klymchuk. It follows, according to this line of reasoning, that once the jury decided that the crime scene was staged, the evidentiary path led inevitably to the appellant.
[53] The logic of Ms. Pomerance’s submissions cannot be denied. Even accepting, however, that the inferences she describes were all available without reference to Agent Brantley’s evidence concerning motive and the relationship between the killer and the victim, that does not mean that the admission of those parts of his evidence constituted harmless error.
[54] Courts have repeatedly expressed the concern that juries will overestimate the value of expert opinion evidence, especially when given by an obviously well qualified expert. This concern animates the gatekeeper function assigned to judges when expert evidence is tendered. Recognition that jurors may well place undue weight on expert opinion evidence is not an attack on the intelligence or work ethic of jurors, but simply a recognition that jurors, like most intelligent people, are sufficiently aware of their own limitations to willingly yield to the opinions of those who have a specialized expertise in an area. I would think this willingness to accept an expert’s opinions is highest when those opinions appear consistent with a juror’s notions of the dictates of common sense.
[55] Agent Brantley was presented as a person with long and impressive credentials. He works for the F.B.I., one of the preeminent law enforcement agencies in the world and at a location, F.B.I. headquarters at Quantico, Virginia, that in the eyes of many is the forensic equivalent of the Mayo Clinic. Current popular culture ascribes to criminalists like Agent Brantley a level of knowledge and objectivity that all but demands acceptance of their opinions.[^1] Crown counsel presented Agent Brantley as an eminently qualified expert, called in from the outside to objectively examine the material produced by the local investigators. When Agent Brantley advanced his opinion as to the killer’s motive and the prior connection between the killer and Mrs. Klymchuk, it is entirely possible that the jury would have taken this evidence as confirmation by the expert from F.B.I. headquarters that the local authorities had arrested the right man.
[56] Crown counsel also relied on the trial judge’s jury instructions on Agent Brantley’s evidence to support her harmless error argument. While I do not agree with Mr. Gold’s contention that the instructions did not adequately caution the jury against an unquestioning reliance on those opinions, the instructions did nothing to devalue the potential significance of Agent Brantley’s evidence in the jury’s assessment.
[57] The Crown has not satisfied me that Agent Brantley’s improperly admitted evidence going to bolster his opinion that the crime scene was staged, and his improperly admitted evidence going to the motive of the killer and the relationship between the killer and the victim could not reasonably have affected the verdict. The curative proviso cannot be applied. There must be a new trial.
(g) Admissibility at the new trial
[58] If on the new trial Agent Brantley, or some other qualified witness, is offered to give evidence that the crime scene was staged to make it appear as though there had been a break-in, that opinion must be based on observations of the crime scene and information gathered at the crime scene. The trial judge will have to decide the admissibility of any opinion offered using the principles set out in R. v. Mohan, supra; R. v. D.(D.) (2000), 2000 SCC 43, 148 C.C.C. (3d) 41 (S.C.C.); and R. v. J.(J.L.), supra. If both the opinion and the basis for the opinion are properly limited to crime scene reconstruction evidence, I think the potential prejudicial effect of the opinion evidence is significantly decreased. Opinion evidence based on observations of the crime scene and inferences drawn from those observations by experienced crime scene investigators is commonly placed before juries. These opinions are not generally based on any novel science or theory, but on well accepted practices and the appreciation that experience and training can educate individuals to see things and draw inferences from those observations that the uneducated will neither see nor appreciate. Nor do I see any significant danger that an opinion properly limited to an examination of the crime scene will produce expert evidence that is so difficult to understand or explain as to defy effective cross-examination or invite acceptance of the opinion as a matter of faith.
[59] The admissibility of expert evidence of staging may well come down to the trial judge’s determination of whether that evidence is necessary: R. v. D.(D.), supra, at 63-64. It is not for this court to preempt that ruling, which must be made on the basis of the record put before the trial judge on the new trial. I will, however, address one argument made to this court, as in my view it is wrong in law and could lead to error if applied at the subsequent trial.
[60] Mr. Gold submitted that even if expert opinion evidence that the crime scene was staged is admissible, the Crown should be required to elicit that opinion through its own investigators who are qualified to give that evidence, rather than through an expert like Agent Brantley. He argued that it was much less likely that an opinion given by investigators involved in the actual homicide would overwhelm the jury and distort their fact finding function.
[61] The trial judge’s gatekeeper function does not extend to determining which of various qualified experts counsel should call. If the expert opinion is necessary in the sense that it will provide the jury with access to information that is important to the jury’s determination and would not otherwise be available to the jury, I see no good reason for requiring the Crown (or the defence) to elicit the expert opinion through a witness who, because of his or her credentials, or for any other reason, might be less impressive than another witness who is available to give that opinion evidence. This approach smacks of a compromise that could well deny the jury the most reliable evidence available to it.
[62] Mr. Gold’s concerns that the jury will be overwhelmed by an expert like Agent Brantley can be dealt with by a proper jury instruction and, where appropriate, an editing of the expert’s qualifications as elicited before the jury. For example, many of Agent Brantley’s qualifications had nothing to do with his expertise in crime scene reconstruction. The trial judge could well limit evidence of Agent Brantley’s qualifications to those which are relevant to his expertise in crime scene reconstruction. In addition to a proper editing of the expert’s credentials, cross-examination can effectively expose an attempt to give artificial weight to an opinion by introducing it through an overly qualified witness.
IV
(a) The other grounds of appeal
[63] There were three additional grounds of appeal. I need make only brief reference to each. The appellant made three statements to the police. The Crown led the first two in their entirety without objection. The Crown sought to tender only four brief references from the lengthy third statement that had been made during a polygraph interview and test. The Crown contended that two of the passages demonstrated inconsistencies that were relevant to the appellant’s credibility and that the other two constituted admissions made by the appellant in response to suggestions made to him by his police interrogators. The trial judge admitted the four extracts of the statement.
[64] The appellant’s objection comes down to a claim that the extracts, taken out of context, had no probative value. I would go so far as to describe the probative value as slight. I agree with the trial judge’s assessment, however, that there was virtually no prejudicial potential attached to the admission of this evidence. I see no error in allowing the Crown to adduce the four extracts from the third statement and leaving it to the jury to assess their probative value, if any.
[65] The Crown also adduced certain statements made by Mrs. Klymchuk prior to her death. These were said to demonstrate that she was very security conscious and also that her marriage was in trouble. On appeal, counsel argued that the statements were inadmissible hearsay. It was open to the trial judge to find that the circumstances in which the statements were made gave them sufficient reliability to justify their admission into evidence. The statements had some probative value. I see no potential prejudicial effect. It is hard to see how evidence of statements made by Mrs. Klymchuk to support the Crown’s contention that the Klymchuks’ marriage was in trouble could have prejudiced the appellant when part of his defence was an acknowledgement of an ongoing affair with Ms. Mays. The evidence of the statements made by Mrs. Klymchuk played a small part in this trial.
[66] The final ground of appeal arises out of the trial judge’s admission of evidence that a trained tracking dog had failed to pick up any scent in the area around the drive shed some two or three hours after the alleged homicide. The Crown relied on this evidence to support its theory that there was no burglar.
[67] Evidence that a properly trained tracking dog followed a scent may is admissible if relevant to a fact in issue: R. v. Holmes (2002), 2002 45114 (ON CA), 169 C.C.C. (3d) 344 at 357 (Ont. C.A.). Mr. Gold contends that the situation in which the trained tracking dog fails to pick up a scent is distinguishable from the situation in which the dog tracks a scent. He argues, as acknowledged by the Crown witnesses responsible for training the tracking dog, that there is no way to distinguish between the dog’s failure to pick up a track when a track was present (a false negative) and the dog’s failure to pick up a scent when there was none present (a true negative). On the evidence, a dog’s effectiveness as a tracking animal could be destroyed if it was asked to track a scent when there was in fact no scent. Consequently, trainers cannot test the accuracy of tracking dogs to determine the incidence of true negatives. Mr. Gold argued that in the absence of any method of distinguishing between the incidence of false and true negatives, evidence that a tracking dog did not pick up a scent had no probative value.
[68] Counsel also submitted that there were features specific to this case which further significantly undermined the probative value the dog’s failure to pick up a scent. For example, the dog could not search the area directly in front of the drive shed leading to the road as that area had been contaminated by emergency personnel who were on scene before the dog.
[69] It was part of the Crown’s case that the killer had made it appear as though the murder had occurred in the course of a break-in. It was part of the defence that Mrs. Klymchuk may have been killed by a burglar. Evidence that the police investigated the possibility of a break-in from the outset was relevant to both the Crown and the defence theory. Evidence that the police used a trained tracking dog in an effort to pick up the scent of the killer was admissible to show that the police considered and investigated the possibility of a break-in. The Crown was entitled to lead evidence of the steps it took to investigate the possibility of a break-in and the results of those efforts. In this sense, evidence that the trained tracking dog did not pick up a scent was no different than evidence that the officers examining the area around the window of the drive shed could find no indication that an intruder had entered through that window. Neither proved that there was no burglar, but both, placed in the context of the rest of the evidence, could help the jury decide whether there had in fact been a burglary.
[70] Evidence that the tracking dog did not pick up any scent offered some support for the Crown’s theory that there was no intruder. Evidence that the dog may have missed a scent that was there, or may not have detected a scent for a variety of other reasons, went to the weight to be given to the failure to pick up a scent. I see no reason to doubt the jury’s ability to accurately evaluate that evidence, especially in the light of the thorough jury charge, which included the following:
In the end you may well find that apart from confirming the thoroughness of this investigation, the evidence may be of limited weight in this case. That is for you to decide, but I must warn you of these dangers that are present with tracker dog evidence. At best, it is simply one factor for you to consider in assessing the overall circumstances.
V
Conclusion
[71] I would allow the appeal, quash the conviction and direct a new trial.
RELEASED: “DD” “NOV 30 2005”
“Doherty J.A.”
“I agree M.J. Moldaver J.A.”
“I agree Robert P. Armstrong J.A.”
[^1]: Current television programming is replete with fictional dramas in which all manners of crime are solved through the industry, resourcefulness and expertise of various criminal forensic experts. One of those dramas, entitled Criminal Minds, is a fictional account of crime solving by a group of expert criminal analysts who work out of the very section of the F.B.I. in which Agent Brantley works.