COURT OF APPEAL FOR ONTARIO
Rosenberg, Blair and Juriansz JJ.A.
BETWEEN:
Her Majesty the Queen
Respondent
and
Andre Jeanvenne
Appellant
Ian R. Smith, for the appellant
Alexander Alvaro, for the respondent
Heard: June 30, 2010
On appeal from convictions entered on two counts of first degree murder by Justice R. Kealey of the Superior Court of Justice, sitting with a jury, on October 24, 2005.
OVERVIEW
1Mr. Jeanvenne appeals from his convictions for first degree murder in relation to two killings of Donald Poulin and Michel Richard. The murder occurred more than 17 years apart; and the murders were strikingly dissimilar in nature. The motives were different. The method of killing and the weapons were different. The killings took place in different locations and settings; nothing about the killings qualified as similar fact evidence. No direct evidence connected the appellant to either crime. The only common connection was investigative; the police obtained a number of allegedly inculpatory statements from the appellant respecting each crime during the course of the same “Mr. Big” operation that was launched for that express purpose.
2The appellant sought a severance prior to trial. Justice Kealey refused the request. Respectfully, he erred in that regard, in my opinion, and I would allow the appeal on that ground.
3The appellant also argues that the trial judge erred in refusing to declare a mistrial following two prejudicial outbursts by an unsavoury, but charismatic, Crown witness to the effect that the appellant had murdered a third individual and attempted to murder a fourth (the witness). Were it necessary to do so, I would also allow the appeal on this ground with respect to the Poulin count.
4My reasons for arriving at these conclusions follow.
FACTS
5Donald Poulin was murdered on January 18, 1983, Michel Richard on May 17, 2000. Both deaths occurred in Ottawa. Mr. Poulin’s is described to be an assisted suicide or “mercy killing”, and Mr. Richard’s was an execution-style killing.
The Poulin Murder
6Mr. Poulin’s body was found late at night lying by the side of the Rideau Road in Ottawa. He had been shot twice with a shotgun, found lying some 8-10 feet from the body. Initially, the death was attributed to suicide, since Mr. Poulin had a history of depression and had previously attempted to shoot himself in the Spring of 1981. After the autopsy, however, the police formed the opinion that Mr. Poulin had been murdered in an assisted suicide or “mercy killing” context. The appellant was not at first a suspect. No forensic evidence associated him with the scene.
7As a result of certain statements made by the appellant during the Mr. Big operation in 2002-2003 – to which I shall return shortly – the police believed that the appellant was the killer, and he was charged with Mr. Poulin’s murder. At trial, the Crown called other evidence as well. The appellant’s ex-wife, Lisa Paquette, testified that the appellant admitted his involvement in the murder and told her that Mr. Poulin had asked him and another man, Jean Viau, to do a “mercy killing”, saying that he “didn’t want to live anymore.” One of Mr. Poulin’s brothers, Gilbert Poulin, also testified that the deceased had told him he had found a close friend of the family “to finish what he did to himself, to take his life.” The appellant was a close friend of the family, Gilbert Poulin said, although he would not have imagined Mr. Jeanvenne assisting in his brother’s death.
8There were some problems with the evidence relating to the initial police investigation of the Poulin murder. The police officers involved had lost their notes. The pathologist who had conducted the autopsy had died. Some of the photographs taken at the scene of the crime were missing. The shotgun found at the scene had been lost.
The Richard Murder
9Michel Richard’s murder was quite different. His body was found in his apartment amidst the blood-spattered signs of a significant struggle. The autopsy revealed 18 sharp force wounds to his head, neck, chest, arms and back, and a .32 calibre bullet through his left arm and chest which was said to have caused the death. Investigators found a blood-stained knife, but it bore the DNA of the deceased only. There was no DNA or other forensic link between the appellant and the scene.
10Again, the Mr. Big Operation yielded potentially damaging statements from the appellant regarding the Richard murder. In addition, the Crown called a number of arguably unsavoury witnesses to support its theory that the appellant had murdered Mr. Richard because he believed that Richard had “ratted” on him to the police in relation to a break-in and theft in which they had been involved several years before the murder. One witness, Jean-Claude Massie – whose outburst formed the basis for the mistrial application referred to above, and to which I will return – testified at some length as to the appellant’s involvement in that break-in and theft and other illegal activities; he also said the appellant had been going around five or six months before the murder saying that Richard was an informer for the police. Another witness, Charles Felteau, testified that the appellant had told him that Mr. Richard “got what he deserved and he deserved to be dumped (i.e., killed) ... for ratting him out and he (i.e., the appellant) dumped him.” A third witness, Denis Bruyère, gave evidence that he had visited the appellant to get some drugs shortly before the Richard murder and had found the appellant trying to take the cylinder out of .32 or .38 calibre firearm; the appellant told him he “had a score to settle.”
The Mr. Big Operation and Statements
11Between July 2002 and early 2003, the police conducted a “Mr. Big” operation designed to elicit incriminating statements from the appellant. Such operations have been held to be legitimate police investigative tools when conducted properly: R. v. Szanyi (2010), 2010 ONCA 316, 254 C.C.C. (3d) 528 (Ont. C.A.). In this case, the appellant was introduced to an undercover officer, Constable Lazenby, who played the role of a high-ranking member of a crime family and who – if the appellant were to become a member of the organization – could offer work in the criminal milieu. The appellant was also introduced to another undercover officer, Constable Stephane Grondin, who pretended to be a “soldier” in the organization. The police hoped that Grondin’s background as a native French Canadian would build a trusting relation with the appellant by offering him the inducement of membership in the criminal organization and, in the process, extract inculpatory confessions from him about both the Poulin and the Richard murders. For example, as part of the ruse, Constable Lazenby pretended he wanted the appellant to kill someone for him, and repeatedly used this subterfuge to discuss the appellant’s experience in killing people.
12The ploy worked, to some extent. While he denied, for the most part, any involvement in the two murders in question, the appellant eventually stated that he had “whacked” several people who were close to him and made certain other statements that connected him with the Poulin and Richard killings.
August 14, 2002
13On August 14, 2002, the appellant told Constable Lazenby that every time someone dies in Ottawa the cops look at him. In response to a question about how many homicides he was looking at, the appellant said “five or maybe eight ... two for sure.” He acknowledged that one of the murders the police were looking at him for was the Richard killing.
September 16, 2002
14On September 16, the appellant told Constable Lazenby about a robbery he had committed with a partner, “Michael.” Michael had been caught with hash sometime later, the appellant said, and had ratted the appellant out about the robbery, and shortly after that had died. The appellant said “that’s the way things go,” but, Constable Lazenby acknowledged, did not accept responsibility for the Richard murder at that time.
October 14, 2002
15On October 14, the police taped a conversation between Constable Lazenby and the appellant, in which the appellant provided details of a “mercy killing” he had committed “15, 20 years ago.” The appellant said that he had killed the guy – who was a close friend – as a “favour” because the guy “wanted to die” and had previously tried to shoot himself in the head with a .22-gauge rifle; he “popped him like he wanted me to pop him,” with a .10 gauge shotgun.
November 22, 2002
16On November 22, the appellant spoke with Constable Grondin about both the Poulin and Richard murders. The conversation took place while the two were driving from Sudbury to Montreal and after the police had staged a police stop of the car during which the appellant was made to hear over the police radio that he was suspected of the two killings. Constable Grondin later asked the appellant who Richard was. The appellant responded that he and Richard had robbed a safe together and Richard had ratted him out. He then said to Constable Grondin: “j’y ai faite la passe,” meaning – according to Constable Grondin – “I killed him.” The appellant also said that when the police found the body it was “toute déchrissé”, or “all messed up.” That homicide was about 2 ½ years before.
17In the same conversation, Constable Grondin asked about Poulin. The appellant replied that Poulin was a friend who had ratted him out and that he had to deal with three friends that ratted him out.
18The November 22 conversation was not recorded.
November 29, 2002
19In a subsequent conversation with Constable Lazenby on November 29, however – when asked to provide details of his previous experience committing murders – the appellant refused to provide further information, and denied that he had spoken about his involvement in the murders with Constable Grondin.
20On other occasions as well, during the M. Big conversations, the appellant denied any involvement in the Poulin and Richard murders.
The Massie Statements
21Jean-Claude Massie was, by everyone’s account, an unsavoury but somewhat charismatic witness. A former criminal associate of the appellant, he was called by the Crown – as noted above – to buttress its theory that the appellant had killed Mr. Richard in retaliation for Mr. Richard having ratted on him regarding the earlier break-in and theft in 1992. During the course of his testimony, however, Mr. Massie blurted out two unsolicited and highly prejudicial remarks. On the first occasion, he stated that the appellant had killed Jean Viau – the man Lisa Pasquette said the appellant had told her was with him at the time of the Poulin murder. On the second occasion, he suggested that the appellant had tried to kill him (Massie).
22The first outburst occurred during Mr. Massie’s examination in chief. The following exchange took place:
Q. So Michel [Richard] plead guilty and the charges were eventually withdrawn?
A. They were all dropped against us.
Q. All right. Did you know a man named Jean Viau?
A. Yeah, Jean Viau, I know him. He killed him.
Q. How long did you know Mr. Viau?
A. Mr. Viau, I knew him a long time. He’s from Lowertown.
They did a bank them, him and Viau.
Q. You can’t talk about anything about that.
A. Oh, I didn’t know, as soon as the Viau name comes out, I
was told I could – okay.
[Emphasis added.]
23There was some debate about whether, when Mr. Massie said “he killed him”, he was referring to the appellant or to Mr. Richard. It was unclear on the record whether the witness had pointed or motioned to the appellant in the prisoner’s dock when he made the statement.
24The second outburst occurred during cross-examination. Defence counsel suggested that the police had agreed not to charge Mr. Massie in relation to the break-in and theft incident in exchange for his testimony against the appellant. The witness denied this, but not without elaborating:
Q. Okay. And did they tell you that you can’t be charged with it because you’re going to give evidence against Mr. Jeanvenne?
A. No. Not for that.
Q. No, why did they tell you you can’t be charged for it?
A. Because Mr. Jeanvenne tried to have me killed here, I’m not gonna stand there like a nut and wait for him to come and kill me. So I told my son – I got five sons, and they ain’t small.
Q. But why did the detectives tell you you can’t be charged with this ...
A. He said everybody cannot be charged with this. It’s with everybody, not only me.
[Emphasis added.]
25How these outbursts evolved into the mistrial ground of appeal will be explored later in these reasons.
LAW AND ANALYSIS
The Severance Issue
General Principles
26The Supreme Court of Canada’s most recent clarification of the principles relating to the severance of counts in an indictment is found in R. v. Last, 2009 SCC 45, [2009] 3 S.C.R. 146. A trial judge has a broad discretion to sever under s. 591(3) of the Criminal Code where he or she is satisfied ‘‘the interests of justice so require.” Once the discretion is exercised, the judge’s decision must be afforded considerable deference; an appellate court may only interfere where the judge has “acted unjudicially” or “the ruling resulted in an injustice.” The onus lies on the accused who seeks the severance to persuade the court on the balance of probabilities that the interests of justice so require. See also, R. v. Arp, 1998 CanLII 769 (SCC), [1998] 3 S.C.R. 339, at para. 52; and R. v. Litchfield, 1993 CanLII 44 (SCC), [1993] 4 S.C.R. 333, at p. 352.
27As the Court pointed out in Last, at paras. 15 and 21, the two grounds for intervention involve different inquiries. In determining whether the judge acted unjudicially, the court inquires into the circumstances prevailing at the time the ruling was made, and should only intervene if the judge erred on a question of law or principle, or made an unreasonable decision. In determining whether the ruling resulted in an injustice, the court looks at the entirety of how the trial and the verdicts unfolded. In this regard, as Charron I.A. said in R. v. Rose (1997), 1997 CanLII 2231 (ON CA), 100 O.A.C. 67, at para. 17:
It is necessary to consider the entire trial, including the potential prejudicial effect of the evidence, the closing addresses of counsel, the judge’s instructions to the jury and any inference that may be drawn from the ultimate verdicts returned by the jury.
28At paras. 16-17 of Last, the Court canvassed the overall considerations bearing upon the judge’s exercise of discretion, as well as the factors to be considered:
The ultimate question faced by a trial judge in deciding whether to grant a severance application is whether severance is required in the interests of justice, as per s. 591(3) of the Code. The interests of justice encompass the accused’s right to be tried on the evidence admissible against him, as well as society’s interest in seeing that justice is done in a reasonably efficient and cost-effective manner. The obvious risk when counts are tried together is that the evidence admissible on one count will influence the verdict on an unrelated count. [Emphasis added.]
Courts have given shape to the broad criteria established in s. 591(3) and have identified factors that can be weighed when deciding whether to sever or not. The weighing exercise ensures that a reasonable balance is struck between the risk of prejudice to the accused and the public interest in a single trial. It is important to recall that the interests of justice often call for a joint trial. ...Severance can impair not only efficiency but the truth-seeking function of the trial.
29Finally, the Court identified a non-exhaustive list of particular factors to be considered. These factors are not in dispute, and include:
The general prejudice to the accused
The legal and factual nexus between the counts
The complexity of the evidence
Whether the accused intends to testify on one count but not another
The possibility of inconsistent verdicts
The desire to avoid a multiplicity of proceedings
The use of similar fact evidence at trial
The length of the trial having regard to the evidence to be called
The potential prejudice to the accused with respect to the right to be tried within a reasonable time, and
The existence of antagonistic defences as between co-accused persons.
Last, at para. 18; see also R. v. E. (L.) (1994), 1994 CanLII 1785 (ON CA), 94 C.C.C. (3d) 228 (Ont. C.A.), at p. 238; R. v. Cross (1996), 1996 CanLII 5992 (QC CA), 112 C.C.C. (3d) 410 (Que. C.A.), at p. 419; R. v. Cuthbert (1996), 1996 CanLII 8341 (BC CA), 106 C.C.C. (3d) 28 (B.C.C.A.), at para. 9, aff'd 1997 CanLII 397 (SCC), [1997] 1 S.C.R. 8 (sub nom. R. v. C. (D.A.)).
The Trial Judge’s Decision
30The trial judge rejected the appellant’s motion to sever. In brief reasons, he said:
There is no similar fact linkage between the two events except of course that they have – it is suggested a common perpetrator. Defence must satisfy this Court on a balance of probabilities that severance is required in the interest of justice, which test is considered under the following points: the general prejudice, factual and legal nexus, undue complexity of the evidence, the possibility of inconsistent verdicts, the desire to avoid a multiplicity of proceedings.
As I indicated during the submissions, it seems to me that it’s less compelling that severance be granted where the facts are not similar within the legal notion of similar fact evidence than when the converse exists, and this is the case before me. There is no element of inconsistent verdicts present, nor does it seem to me that there will be any unduly complicated evidence. The defence position seems to be that facing two charges of murder in itself creates an overwhelming prejudice to an accused person, which cannot be offset by jury instructions. I don’t agree with that proposition. In addition to some other individual testimony, the thrust of the Crown’s case emerges from the admissions allegedly made by Mr. Jeanvenne to two police undercover agents in the course of his involvement with the […] purported illegal activities. Whether Mr. Jeanvenne’s statements will be believed by the jury requires a consideration of the whole context within which they were made, together with of course all of the other evidence.
I’m satisfied that proper instructions can alleviate any concern for improper or impermissible reasoning on the part of the jury, or a misuse of the evidence on the two counts. And as such, at this juncture, I find that the defence has not met the onus. The severance application is dismissed. If and when it is known that Mr. Jeanvenne may wish to testify on one or the other of the counts that he faces in this indictment, then the question of severance may be revisited.
R. v. JeanVenne, (6 September 2005), Ottawa 03-G9319 (Ont. S.C.)
Analysis
31In my view the trial judge erred on both grounds in failing to grant a severance in the circumstances of this case, but in particular on the ground that he acted unjudicially. His decision was “unreasonable” on a number of grounds. It also resulted in an injustice, in the circumstances.
32First, his decision was based in a significant way on his mistaken view that the absence of similar fact evidence between the two murders militated against rather than in favour of the case for a severance. The trial judge made this observation both in his reasons and during the course of counsel’s arguments respecting severance. In fact, however, the contrary is generally the case. Mr. Alvaro acknowledges on behalf of the Crown that the trial judge erred in his assessment of how the absence of similar fact evidence impacts upon an application for severance, but argues that the error did not taint the entirety of the judge’s reasoning and was cured by the admittedly strong mid-trial and final instructions to the jury about the impermissibility of using bad character evidence for improper propensity reasoning.
33I do not agree.
34The appellant was facing two counts of murder – the most serious crime known to the Criminal Code and one evoking the greatest possible jeopardy in the justice system. While there is nothing in the Code that prohibits the joinder of more than one murder count in one indictment, there is very good reason, in such circumstances, to apply a more stringent application of the severance test in favour of the accused in such cases – particularly where, as here, the murders are completely unrelated: see R. v. Khan (1996), 1996 CanLII 7296 (MB CA), 108 C.C.C. (3d) 108 (Man. C.A.) (aff’d on other grounds, 2001 SCC 86, [2001] 3 S.C.R. 823. This flows from the common sense proposition that a jury may have great difficulty dissociating the evidence of one homicide from the other and in refraining from drawing the impermissible inference that because the accused may have committed one murder – bad character evidence of the highest degree – he or she is likely to have committed the other. The potential for serious prejudice to the accused is at its zenith in such circumstances. When there is no similar fact evidence connecting the two crimes, there is even less justification for refusing to sever.
35While the trial judge did consider the defence argument that facing two charges of murder created an overwhelming prejudice to the accused that could not be offset by directions to the jury, he dismissed it on the simple basis that he did not agree. He failed to give effect to the need to apply a more stringent application of the test in double-murder count circumstances. The error was compounded by his reverse application of the considerations surrounding the absence or presence of similar fact evidence.
36These errors in principle make the trial judge’s decision unreasonable in the sense called for by the severance jurisprudence, and are in themselves sufficient to justify appellate intervention. It therefore falls to this Court to decide, according to proper principles, whether severance should have been granted: see R v. Savoury (2005), 2005 CanLII 25884 (ON CA), 200 C.C.C. (3d) 94, at para. 26 (Ont. C.A.). In my view, a severance should have been granted. In addition to the foregoing considerations, there are other factors that militate in favour of granting severance in this case.
37First, there is no factual or temporal nexus whatsoever between the Poulin and Richard murders, except – as the trial judge acknowledged – a suspected common perpetrator. As I have noted earlier, the murders were committed 17 years apart, with no apparent connection between the two victims; the motives behind the murders differed significantly (a “mercy killing”/assisted suicide on the one hand, and an execution-style killing on the other hand); the weapons used and the method of killing were different; the locations of the killing were as well; and there was nothing about the two killings that could qualify as similar fact evidence. The trial judge gave little weight to these considerations, except to the extent that he viewed them in the context of his mistaken similar fact analysis referred to above.
38Nor is there much, if any, legal nexus between the murders. The two counts do not constitute overlapping transactions and, while the essential elements of the crime of murder remain the same, the application of the facts to those elements is entirely dissimilar in each case.
39The trial judge placed considerable emphasis on the fact that the Crown’s case depended significantly on the admissions allegedly made by the appellant during the course of the Mr. Big operation. He concluded that the jury required the whole context of all of those statements in order to be able to assess the credibility of the evidence and the reliability of the statements. Before us, Mr. Alvaro submitted that the evidence emanating from the undercover sting could not be separated neatly into little packets without removing from the jury consideration of the context in which the statements were made; reliability and context, he argued, were strongly interwoven. I am not persuaded by this argument, however.
40The trial judge failed to consider whether the various statements could be edited in the context of separate trials without potentially harming the probative value of the evidence as a whole and with a view to minimizing the prejudice potentially jeopardizing the appellant’s fair trial. Why was it irretrievably necessary that the jury considering the Poulin murder hear the appellant’s alleged confessions respecting the Richard murder, for example, and vice versa? With one exception – the November 22 conversation between the appellant and Constable Grondin en route from Sudbury to Montreal – the statements allegedly made respecting one murder were made during different conversations with Constables Lazenby and Grondin than the statements respecting the other, and at different times. The statements on August 14 and September 16, 2002, referred only to Mr. Richard, that of October 14 to Mr. Poulin. While the November 22 conversation covered both, I am not convinced that the evidence could not have been given effectively in separate trials with appropriate redactions of references to the other murder as may have been required. Again, the failure of the trial judge to consider these options constituted an error, in my view.
41Moreover, considerations relating to economy of trial time and administrative resources arising from the Mr. Big undercover officers having to testify more than once if a severance were granted – the multiplicity of proceedings issue – bear less weight here, given the serious nature of the charges. Nor is the possibility of inconsistent verdicts a concern.
42The appellant indicated to the trial judge that he had not yet decided whether he would testify at all, or with respect to one count rather than the other. This can be an important consideration, but the trial judge took this into account when he stated that the appellant could renew his mistrial application after that decision had been made, if so advised. The appellant never did so.
43Finally – and more importantly – however, the case as put forward by the Crown with respect to both murders was replete with very bad character evidence. But much of the bad character evidence pertaining to one murder would not have been pertinent to the other. Yet the jury was exposed to it all. The risk of impermissible propensity reasoning was overwhelming, notwithstanding the strongest jury instruction possible.
44Mr. Massie’s gratuitous remarks attributing two other murderous acts to the appellant are one example of this type of evidence. There are others.
45For instance, the evidence of Lisa Paquette related only to the Poulin murder. She testified as to the appellant’s admission to her that he had been involved in that murder, and that he had done so in the company of Jean Viau. None of this pertained to Mr. Richard’s death. Nonetheless, it was the witness Massie’s outburst – in the context of the Richard murder – about the appellant having killed Viau that gave rise to the mistrial application. A piece of evidence relating to one unrelated killing potentially poisoned the evidence in the other. In addition, Ms. Paquette gave extensive evidence about the appellant’s drug use and involvement in the drug trade, his volatile nature and the fact that she was afraid of him – none of which would have been heard in the context of the Richard charges had a severance been granted.
46Similarly, considerable evidence was called by the Crown in relation to the Richard case regarding the appellant’s involvement in a life of crime and, in particular, about his membership in the break and enter gang that had committed the crime giving rise to the theory that Mr. Richard had ratted the appellant out. This evidence was led not only through the unsavoury Mr. Massie, but also through a number of other arguably unreliable witnesses, including Charlies Felteau, Denis Bruyère and Andre Quesnel. Again none of that evidence would have been heard in the Poulin proceeding, had it been tried separately.
47Needless to say, the witch’s brew of all of this evidence in the cauldron of a single trial was potentially prejudicial to the appellant in a devastating manner, for all of the reasons outlined above. It is for this reason that I conclude the decision not to sever resulted in an overall injustice in this case. The review of the evidence at the outset of these reasons indicates that there was, indeed, evidence upon which the jury could have convicted the appellant on both counts of murder. However, the Crown’s case was not unassailable in either case. Much of the original investigatory evidence relating to the older Poulin murder was missing or difficult to establish. The balance of the case depended largely on the jury’s acceptance of Ms. Paquette’s evidence and that of the undercover officers. So, too, did the case against the appellant in relation to the Richard murder depend upon the acceptance of the Mr. Big testimony, buttressed by that of a series of admittedly – or at least arguably – unsavoury witnesses. Credibility was central to the resolution of all of these issues. Whether the appellant chose to testify or not – or on one count rather than another – the risk of cross-pollination by the jurors of the evidence from one count to the other and of the resort to prejudicial propensity reasoning cannot be understated.
48Balancing all of the relevant factors, and weighing the interests of justice as a whole, I am satisfied that the trial judge erred in refusing to sever the two unrelated murder counts, and that such a severance should have been granted. It was required in the interests of justice, in the circumstances.
The Mistrial Application
49In view of the foregoing, it is strictly speaking unnecessary to deal with the mistrial issue. For the reasons that follow, however, I am satisfied that the trial judge erred in failing to grant a mistrial, at least with respect to the Poulin count.
50The evidence of Mr. Massie giving rise to the mistrial issue is outlined earlier in these reasons (at paras. 21-25). His rash remarks were problematic, to be sure. Essentially, Mr. Massie portrayed the appellant as a murderer or attempt murderer with respect to two other alleged incidents. In the first instance, responding to a question whether he knew a man named Jean Viau, Mr. Massie said: “Yes, he killed him.” Although there was some ambiguity in the context whether “he” was a reference to the appellant or to the deceased, Michel Richard, there was an open inference that “he” meant the appellant. In the second incident, Mr. Massie said that the appellant had tried to kill him (i.e., Mr. Massie).
51At the break following the first statement, trial counsel for the appellant expressed concern about the remark and indicated he would be bringing a motion for a mistrial. The trial judge decided to hear the balance of Mr. Massie’s evidence and to deal with the mistrial motion at the end of the day or on the following day. The testimony of Mr. Massie continued, with no caution of any sort being given to the jury at that time.
52The second incident occurred during Mr. Massie’s cross-examination. Again, no caution was provided to the jury. Four more short witnesses gave evidence to complete the day.
53On the following day, October 4, 2005, the trial judge heard the mistrial submissions. At the end of argument he declined to rule on the mistrial motion, taking the position that he wanted to hear the whole of the evidence before doing so. He felt, in particular, that if the wiretap evidence the Crown was proposing to tender in the context of the Mr. Big operation were ruled admissible –it ultimately was – that evidence would take the edge off the significance of the Massie utterances: “Mr. Massie’s offhand remark[s]”, he said, “will pale.”
54Court did not sit on October 5. On October 6, the trial judge provided a strong mid-trial caution to the jury, saying:
On Monday, Mr. Massie, in his evidence, you may have noted, said that Mr. Jeanvenne killed a Mr. Viau and in cross-examination he may have suggested to you that Mr. Jeanvenne was also going to kill Mr. Massie, himself. In such evidence, which was blurted out by Mr. Massie, is not admissible and is not to be considered by you for any purpose. In addition to its – to what I’ve said about the impermissible reasoning process, this evidence has another layer, it is rank and gross hearsay evidence and is not to be used by you for any purpose in your deliberations at the end of the trial. I ask that you just put those remarks totally out of your minds. They should be completely ignored by you.
55At the same time, as requested by defence counsel, the trial judge gave a strong mid-trial Vetrovec warning to the jurors about the need to be cautious in considering the testimony of Mr. Massie, given his lengthy criminal record and lifestyle. This included a direction that “[they] must closely scrutinize his testimony and be careful, because it is dangerous for [them] to act on that evidence without confirmation, unless [they were] satisfied completely that it [was] trustworthy.”
56The trial judge did not revisit the mistrial issue until after his charge to the jury. At that point he declined deal with it again, taking the position that he had already ruled on it.
57Both of these unsolicited statements by Mr. Massie were inadmissible. No one contests this. The statements were unresponsive to the questions Mr. Massie was being asked and unrelated to the charges the appellant was facing. They constituted bad character evidence of the worst sort.
58I accept that trial judges are particularly well-placed to assess the impact of inadmissible and potentially prejudicial evidence in the context of the trial dynamic, including the effectiveness of any warning that may be issued. Their discretionary decision whether or not to grant a mistrial is entitled to great deference. Indeed, trial judges themselves are only to order a mistrial “as a last resort, in the clearest of cases and where no remedy short of that relief will adequately redress the actual harm occasioned”: R. v. Toutissani, 2007 ONCA 773, at para. 9. See also R. v. Khan, 2001 SCC 86, [2001] 3 S.C.R. 823, at para. 36, and R. v. Chiasson (2009), 2009 ONCA 789, 258 O.A.C. 50 (C.A.) at para 14. The court will only interfere where, in the exercise of his or her discretion, the trial judge is clearly wrong.
59Were it necessary to do so, I would hold that this is one of those rare instances where the court would be justified in interfering with the trial judge’s refusal to grant a mistrial. Respectfully, the trial judge erred in misapprehending, or failing to consider, the particularly damaging nature of the “Jean Viau” statement in relation to the Poulin murder.
60While it may be that there was some uncertainty on the record about whether Mr. Massie had pointed to the appellant when he said “he killed him”, it was a strong inference open to the jury that “he” meant the appellant. Jean Viau was the person Ms. Paquette said the appellant told her had been involved with him in the Poulin killing. Thus, there was an equally strong inference open to the jury that the appellant had killed Mr. Viau in order to silence him and cover up his own participation in that murder.
61Had the severance been granted, the Poulin jury would not have heard the evidence of Mr. Massie.
62The trial judge appears not to have considered this particularly dangerous chain of reasoning, and the complicating layer of his earlier refusal to grant the severance, in reaching his conclusion. I agree that the reference of the unsavoury witness, Massie, to the appellant trying to have him killed, might pale in comparison to the statements to the undercover officers that he had “whacked” or killed a number of people – or be subsumed in the generic nature of that reference. However, the same is not the case with the Jean Viau outburst. It pointed to the appellant in a specific and prejudicial way in relation to the Poulin murder and I am not satisfied that even the strong mid-trial and final instructions that were given would be adequate to enable the jury to disabuse themselves of that information.
63In addition, Mr. Smith argues that, even if no mistrial were declared, the jury had to be warned about the dangers of impermissible propensity reasoning and told to ignore such evidence “immediately” (see R v. D. (L.E.), 1989 CanLII 74 (SCC), [1989] 2 S.C.R. 111, at p. 127) and “in the strongest terms” (see R v. Woods (1989), 1989 CanLII 7192 (ON CA), 49 C.C.C. (3d) 20 (C.A.), at p. 32-33. Here, he submits, that did not happen. The effect of the sequence of events outlined above was that the jury was left for two days to absorb and think about the prejudicial remarks before any correcting instruction was provided to them. The evidence must have affected the jurors’ deliberations and thinking at least during the period of time before they were told to ignore the evidence, if not thereafter. It cannot be said with any assurance that the jurors would be able to disabuse their minds of the comments even with the strong mid-trial instruction that was given, he concludes.
64I would not give effect to this argument in the circumstances here. The evidence of the four witnesses who were called after Mr. Massie and before the Court adjourned on October 3 was short and uncomplicated. The mid-trial instruction was given at the first opportunity thereafter when court resumed with the jury. There is no magic in the word “immediately,” in my view. The mid-trial instructions – indisputably strong –were given clearly and sufficiently close in time to the impugned evidence, without waiting until the close of trial, to counter whatever limited prejudicial impact the statements may have had in the overall context of the trial, in my view: see R. v. Colas (2001), 2001 CanLII 3538 (ON CA), 161 C.C.C. (3d) 335 (Ont. C.A.), at paras 47-48.
65To the extent it means anything, then – and in the context of the refusal to grant a severance – I conclude that the trial judge erred, in the circumstances here, by failing to order a mistrial in relation to the Poulin murder charge. I do not think he erred in failing to do so with respect to the Richard count.
DISPOSITION
66For the foregoing reasons, then, I would allow the appeal and order a new trial with respect to both counts of murder, to be tried separately.
“R.A. Blair J.A.”
“I agree M. Rosenberg J.A.”
“I agree R.G. Juriansz J.A.”
RELEASED: October 26, 2010
