[1]               Over 40 years ago, on August 9, 1967, Leopold Roy, a 20-year veteran of the Ottawa Fire Department, was stabbed to death in an Ottawa apartment building. His wife, Mildred Roy, saw the assailant and provided the police with a description.  On August 10, Detective John McCombie of the Ottawa Police Department showed Mrs. Roy photographs of ten different men, including pictures of the appellant, Romeo Phillion, and his identical twin brother, Donald. Mrs. Roy picked Donald as the person she thought most resembled the attacker.

[2]               The police learned that the Phillion brothers were in New Liskeard, Ontario. On the evening of August 12, 1967, Detective McCombie travelled there with his partner, Detective Edward Coburn. They eliminated Donald as a suspect after confirming that he had been at work in a New Liskeard auto garage on August 9, 1967.  

[3]               On August 13, the appellant was arrested on an unrelated charge and Detectives McCombie and Coburn brought him to Ottawa to participate in a line-up before Mrs. Roy, which took place on August 15. While Mrs. Roy was in the viewing room, she identified the appellant as the perpetrator. However, when she was taken into the line-up room, she stated that the appellant looked like the man who killed her husband, but she was not sure.  In the absence of a positive identification, the appellant was released.

[4]               Over the next four-and-a-half years, the police made no meaningful progress in the investigation of Mr. Roy’s murder. The crime scene yielded no evidence and the murder weapon was never found.  Mrs. Roy identified three other men as suspects. All three were arrested and later released.

[5]               On the morning of January 11, 1972, the appellant was arrested for the armed robbery of a taxi cab driver in Ottawa.  Later that day while in police custody, he confessed to the murder of Mr. Roy and was charged with non-capital murder. 

[6]               The appellant retained Mr. J. Arthur Cogan as his defence counsel. Mr. Malcolm (Mac) Lindsay was the Assistant Crown Attorney assigned to prosecute the case.

[7]               A seven-day preliminary inquiry was held before Justice J.P. Beaulne on various days between March 2 and June 28, 1972, following which the appellant was committed to stand trial for non-capital murder.

[8]               The appellant’s three-week trial before Justice Mabel Van Camp and a jury began on October 16, 1972 and ended on November 7, 1972 with a guilty verdict. The trial judge sentenced him to life imprisonment with no chance of parole for ten years.  His appeal to the Court of Appeal was dismissed on September 3, 1974 and his further appeal to the Supreme Court of Canada was dismissed on March 22, 1977.   

[9]               On May 15, 2003, the appellant filed an application with the Minister of Justice under s. 696.1 of the Criminal Code requesting a review of his conviction based on an alleged miscarriage of justice.  On August 2, 2006, following an investigation and review into his conviction by John Briggs on behalf of the Minister, the Minister of Justice, the Honourable Vic Toews, ordered a reference to this court. 

[10]         The terms of reference require this court to provide its opinion on two questions.  The precise terms of reference will be set out below.  In general terms, the first question raises the issue of the admissibility on appeal of new information concerning the non-disclosure of an alibi for the appellant, as described in a police report of the lead investigating officer, Detective McCombie, as well as the non-disclosure of certain witness statements.  The second question relates to the admissibility on appeal of recent expert reports concerning the reliability of the appellant’s confession.

[11]         The terms of reference go on to provide that if this court determines that the information referred to in the first question and/or the recent expert reports referred to in the second question are admissible on appeal, then the court is to decide the case as if it were an appeal on the basis of fresh evidence.

[12]         For reasons that follow, I would admit the new information concerning the non-disclosure of an alibi for the appellant and the non-disclosure of certain witness statements. I would not admit the recent expert reports concerning the reliability of the appellant’s confession.  On the issue of remedy, I would allow the appeal, quash the conviction and order a new trial.

[13]          Before setting out my reasons for these conclusions, it is necessary to provide the factual background for this Reference, including the evidence that was heard by the jury at trial regarding Mr. Roy’s murder and the appellant’s alleged involvement in it, the theory of the Crown and the defence at trial, the nature of the appeal proceedings to this court and the Supreme Court of Canada, the events leading up to the ordering of the Reference and the terms of this Reference as specified by the Minister.

[14]         The evidence heard by the jury at the appellant’s trial regarding the details of Mr. Roy’s murder is succinctly stated by Ritchie J. in the majority judgment of the Supreme Court of Canada in R. v. Phillion, 1977 CanLII 23 (SCC), [1978] 1 S.C.R. 18, at pp. 21-22:

[15]         The jury also heard that some four years after the appellant’s initial arrest and subsequent release in August 1967, the appellant was brought into the Ottawa police station for questioning on January 11, 1972 on an unrelated matter (the jury did not hear that the appellant was brought into the station because he had been arrested on a charge of armed robbery). He was interviewed by Detectives Stephen Nadori and Roland Huneault. 

[16]         In his evidence in chief, Detective Huneault described an exchange with the appellant on the way to the cell area, in which the detective asked the appellant if he had anything else to tell them. In response, the appellant asked the detective what Neil Miller (the appellant’s lover) had told them. Detective Huneault responded that Miller had told the police “a lot about the things you bragged about to him”, following which the appellant asked if Miller had told them anything big.  When Detective Huneault indicated that he had, the appellant said, “I’ll confess”.  Detective Huneault asked him what he meant and he said, “Something big, like murder; the fireman, I did it.  Get me a coffee and we’ll talk about it.”  Detective Huneault testified that he had not taken any part in the investigation of Mr. Roy’s murder up to that time and that he was not thinking of the murder when he posed these questions to the appellant.

[17]         After Detective Huneault read the standard police caution and returned with the requested coffee, the appellant told the detective that he “did the murder of the fireman a long time ago on Friel Street.”  The appellant said he would make a full confession after he had spoken to Miller. 

[18]         The police then went to Neil Miller’s home and asked him to come to the police station. He agreed and the police obtained a signed statement from him, which stated that on January 7, 1972, the appellant had told him that:

[19]         After being permitted by police to meet with Miller, the appellant provided a signed confession to the murder.  Before providing the statement he asked, “Was that guy a fireman or a policeman?  I can’t hang for that, can I?”

[20]         The appellant’s typewritten, signed statement reads in part as follows:[1]

[21]         The jury also heard that, later on the evening of January 11, the appellant told another officer, P.C. Couture, that he had nothing to do with the murder.  He told the officer that he “had told Neil that if things came to worse, if I was caught in an armed robbery or if I would kill anyone in an attempt, that I was to confess to the murder so that Neil could report it and collect the reward[2] and that I would get half.  In other words, I wanted to get even and send them on a wild goose chase.” 

[22]         The sole issue before the jury was whether the appellant was responsible for Mr. Roy’s murder.  The Crown relied primarily on three pieces of evidence in support of a finding that the appellant was the perpetrator.

[23]         First, the Crown pointed to the identification of Mrs. Roy.  In closing argument, Crown counsel accepted that her identification was not without its problems.  However, he asked the jury “not to discard it altogether, but to keep it in your minds and give it what weight you feel is proper.” 

[24]         Second, and of the greatest significance to the Crown’s case, was the appellant’s confession to police on January 11, 1972.  Crown counsel reviewed the circumstances of the confession, including the police testimony that no force or pressure or inducement was used to extract it from the appellant. Crown counsel urged the jury to reject the defence suggestion that the appellant learned of the details of the murder by reading the newspapers at the time of the murder, arguing: 

[25]         As evidence in support of the veracity of the confession, the Crown pointed to the confirmatory evidence of Constable Bolger, a police officer with the New Liskeard Police Department, who found the appellant in New Liskeard, which was where the appellant said he went at around 11:30 p.m. on the night of the murder,[3] and Mrs. Barbe,[4] “who mentioned that she saw him around 11:30 and he left.”

[26]         Third, the Crown relied on the appellant’s confession to Neil Miller and the similarity of that confession to the one he gave to the police.  In this regard, Crown counsel pointed out that there was no evidence of any pressure on the appellant to lie to his friend.

[27]         The defence’s theory of the case was that the appellant’s personality, which was affected by an anti-social personality disorder, prompted him to falsely confess to the crime.  Defence counsel at the preliminary inquiry referred to the possibility of raising an alibi as a defence.  However, partway through the trial, during submissions to the court in the jury’s absence, defence counsel stated that he had previously advised the Crown that there was “no issue” that the accused was in Ottawa on the day in question.

[28]         The appellant did not testify on his own behalf. He called a psychiatrist, Dr. Arboleda, and a psychologist, Dr. Girodo, to testify that his confession was inherently unreliable. The details of the evidence of these experts will be provided in responding to the second question raised by the Minister. Suffice it to say for present purposes that both experts were of the view that the appellant’s psychological profile was such that the appellant had a propensity to lie and to invent stories to make him feel important.  The doctors were of the view that no reliability should be placed on any of his statements in the absence of independent supporting evidence. 

[29]         Dr. Arboleda further testified that he questioned the appellant about the truth of his confession after giving him sodium amytol, a so-called “truth serum”.  He also considered the results of a polygraph test during which the appellant said that he had lied to the police when he confessed.  Dr. Arboleda testified regarding the results of these tests:

[30]         In addition to the expert evidence tendered on behalf of the defence, defence counsel in closing pointed to the frailties of Mrs. Roy’s identification evidence, noting that she did not positively identify the appellant in the line-up and that on three different occasions, other people were arrested because of identifications she made. 

[31]         A further theme stressed by defence counsel in closing argument was the “shabby” police investigation of this crime.  For example, counsel pointed to the evidence that the police had lost the photographs that were shown to Mrs. Roy before the line-up, which included pictures of the appellant and his twin brother.  Defence counsel pointed out that the jury was denied the opportunity to see these photographs so that they might assess what impact they may have had on Mrs. Roy’s initial identification of the appellant in the line-up.

[32]         Defence counsel also asked the jury to resist a finding that the evidence supported the truth of the confession in a case where the police had lost fingernail scrapings, blood and hair samples removed from the deceased, as well as a hair sample voluntarily provided by the appellant.  He argued that these samples could have assisted the defence and asked the jury to look very carefully at the lack of a proper police investigation in arriving at their verdict. 

[33]         In his Notice of Appeal to the Ontario Court of Appeal, the appellant advanced numerous grounds of appeal, including several objections to the trial judge’s decision to admit both his confession and the identification evidence of Mrs. Roy.  He also alleged that he was deprived of the presumption of innocence because of the shortcomings of the police investigation.  In addition, he submitted that the trial judge erred in refusing to permit the polygraph examiner, Mr. John Edward Reid, to testify regarding his polygraph examination of the appellant and the results obtained.

[34]         In a brief decision by Jessup J.A. on behalf of the court, the appeal was dismissed.  The only ground of appeal discussed by the court was whether the trial judge erred in refusing to admit the evidence of Mr. Reid. The court concluded that the evidence was properly ruled inadmissible because the witness was “being asked to express his opinion directly, that the accused had not committed the act constituting the offence charged.”

[35]         The Supreme Court of Canada granted leave to appeal on the following question of law:

[36]         The court unanimously dismissed the appeal.  Ritchie J., writing on behalf of seven members of the court, stated at p. 25: 

[37]         Spence J. wrote concurring reasons on behalf of himself and Laskin C.J.  He agreed with the inadmissibility of Mr. Reid’s opinion that the accused was telling the truth during the polygraph examination.  At p. 20, Spence J. noted that the results of the polygraph test were before the jury through the evidence of Dr. Arboleda, yet the jury refused to accept the expert opinion.  He concluded that one could not have expected the jury to have had “any regard for the polygraph results which were only one of the materials upon which the expert witness Dr. Arboleda based his opinion.”

[38]         Following his failed appeals, the appellant wrote a series of letters to the Minister of Justice asking to have his case reopened.  In many of these letters, he insisted that he was in New Liskeard at the time of the murder. 

[39]         In 1998, the appellant’s parole office gave him a brown manila envelope containing Ottawa police reports from the appellant’s Canada Corrections security file.  The most significant document in the envelope was a police investigation report prepared by the investigating officer, Detective McCombie, on April 12, 1968 at 8:00 p.m. (“the April 12, 1968 report”).  The first question on the Reference to this court deals primarily with issues connected to this report. 

[40]         There are several important passages in the April 12, 1968 report.  The most salient is the passage containing Detective McCombie’s observation that a service station operator in Trenton verified that the appellant was at the service station between 12:00-1:00 p.m. on August 9, 1967, “therefore making it impossible for him to return to Ottawa by 2:45 p.m. at the time the murder was committed.” 

[41]         The appellant forwarded the envelope and its contents to the Innocence Project at Osgoode Hall Law School.  On May 15, 2003, counsel for the Innocence Project and Mr. James Lockyer, one of the appellant’s current counsel, submitted an application for ministerial review on the appellant’s behalf to the Minister of Justice.  The application was based on information obtained from the following sources: the Ottawa Police Department; the New Liskeard Police Department; the Ministry of the Attorney General; the Archives of Ontario; the Correctional Service of Canada; the Chief Coroner’s Office for Ontario; the Department of Justice, Convictions Review Group; the appellant’s trial counsel, Mr. Cogan; one of the appellant’s appellate counsel, J. Douglas Crane; the Penetanguishene Mental Health Centre, the Supreme Court of Canada archives; the archives of the Centre of Forensic Sciences; and printed media reports from 1967 and 1972.  Attempts to locate Detective McCombie’s police notes of his investigation and any physical evidence related to this case were unsuccessful. 

[42]         Also included in the materials presented to the Minister were reports prepared by Dr. Gisli Gudjonsson, an expert on false confessions, and Dr. Graham Turrall, a psychologist, who were retained by the appellant in 2002 to review his case.

[43]         On June 26, 2003, the appellant received bail pending a decision from the Minister on his application. By this time, he had served 31 years in prison, in part because he refused to apply for parole in connection with the Roy murder, for which he continued to maintain his innocence.

[44]         The Minister appointed Mr. John Briggs, Q.C., as his representative to review the application.  Mr. Briggs conducted sworn interviews of Mac Lindsay, Arthur Cogan, John McCombie, Edward Coburn, and Stephen Nadori.  In May 2005, Mr. Briggs submitted his investigation brief to the Minister.

[45]         Following his review of Mr. Briggs’ report, on August 2, 2006, the Minister of Justice ordered a reference to this court pursuant to s. 696.3(2) of the Criminal Code.[5]  He asked this court to first give its opinion on two questions.  The terms of reference state:

[46]         The terms of reference thus require this court to first provide an opinion on the two questions posed by the Minister.  If one or both of these questions are answered in the affirmative, then the court is to go on to determine the case as if it were an appeal. 

[47]         Before analyzing the first of the two questions, it is necessary to describe the new information referred to by the Minister in question one.

[48]         The first question posed by the Minister refers to “new information concerning the non-disclosure of Mr. Phillion’s alibi which is described in police reports authored by Detective McCombie”. The critical report referred to by the Minister is the two-page April 12, 1968 report of Detective McCombie.[6]  Indeed, the seeds of this reference are directly traceable to this report.

[49]         The opening words of the April 12, 1968 report reveal its origins: 

[50]         The April 12, 1968 report also records that the New Liskeard police provided Detective McCombie with a statutory declaration sworn by the appellant’s girlfriend, Gail Brazeau.  In this statement, dated April 4, 1968, Ms. Brazeau told the New Liskeard police that she had seen the appellant on Wednesday August 9, 1967, the day of the murder.  She also said that the appellant used to carry a knife with an eight inch blade, which he kept in the glove compartment of his car, and that she had not seen this knife since the day of the murder.  In addition, she recalled that on the day of the murder, the appellant was dressed in a black sweater and black pants and that she had not seen him with that clothing since. 

[51]         The statutory declaration records the following questions posed by police to Ms. Brazeau and her answers:

[52]         In addition to sending the statutory declaration, the New Liskeard police also sent Detective McCombie a photograph of a knife that Ms. Brazeau had picked out from a local store as being similar to the one that the appellant used to carry.  In the letter to Detective McCombie enclosing the statutory declaration and the photograph, P.C. Brown stated: “We will interview more persons here in New Liskeard who were with Phillion when he came to this area last August and who might be able to shed some light on the disappearance of the black clothing and the missing knife.  Because if my memory serves me right Phillion was wearing clothing that belonged to his brother at the time of his arrest on August 12, 1967 here in New Liskeard.” 

[53]         The April 12, 1968 report describes the investigative steps taken by Detective McCombie after receiving this information from the New Liskeard police. The report also describes certain statements that Detective McCombie said he obtained from various witnesses soon after the murder in August 1967.  The report concludes: “[I]t is felt by the writer and Lieut. Welsh that Phillion is not the man we are looking for and the New Liskeard Police Department will be so advised by letter.”

[54]          No such letter was found in the well-maintained files of the New Liskeard Police Department and thus there is no basis for concluding that the New Liskeard police received either the April 12, 1968 report or its contents.  However, there is also no indication that the New Liskeard police played any further role in the Roy murder investigation after April 12, 1968 – that is, until after the appellant confessed to the murder in January 1972.

[55]         The passage in Detective McCombie’s April 12, 1968 report relating to the appellant’s alibi as referred to in the first question posed by the Minister states as follows:

[56]         The report does not provide any further details surrounding the verification of the appellant’s alibi, such as the name of the service station operator or the service station in question, whether Detective McCombie went to Trenton to verify the alibi or if he telephoned the service station operator, or if another source passed on the verification information of the service station operator to Detective McCombie.

[57]         The first question posed by the Minister also refers to the non-disclosure of the statements of Mr. and Mrs. Barbe. The Barbes were neighbours of Gail Brazeau’s mother, with whom the appellant was staying in Ottawa around the time of the murder. Both Mr. and Mrs. Barbe testified at the preliminary inquiry and Mrs. Barbe testified at trial. Their evidence was that they last saw the appellant on Wednesday August 9, the night of the murder.  They both testified that they could recall it was a Wednesday because Mrs. Barbe had attended bingo that evening and she used to go to bingo every Wednesday.

[58]         In contrast to Mr. Barbe’s evidence at the preliminary inquiry, Detective McCombie’s April 12, 1968 report indicates that, after he received the information from the New Liskeard police in April 1968, Detective McCombie re-interviewed Mr. Barbe and learned the following:

[59]         The report also states that in 1967, shortly after the murder, Detective McCombie learned from Mrs. Brazeau’s neighbour (who presumably was Mr. Barbe): 

[60]         There is no reference in the April 12, 1968 report to any statement having been provided to Detective McCombie by Mrs. Barbe either in 1967 or 1968.  Nor is there any existing documentation of what Mrs. Barbe might have told Detective McCombie in August 1967. However, both Detective McCombie and Mrs. Barbe testified at the preliminary inquiry that Detective McCombie spoke to Mrs. Barbe in August 1967. There is no suggestion in the April 12, 1968 report that Mrs. Barbe’s initial recollection of the last night on which she and her husband saw the appellant differed from her husband’s recollection.

[61]         The first recorded statement of Mrs. Barbe is in Detective Huneault’s investigation report dated January 13, 1972, following the appellant’s confession, which reads:

[62]         The final piece of information referred to in the first question posed by the Minister relates to the non-disclosure of the statements of Mr. Loyer.  The only statement of Mr. Loyer is a willsay statement which is found in the Crown’s Brief.  The statement reads in its entirety as follows:

[63]         Mr. Loyer did not testify at the preliminary inquiry or the trial.  His name appeared on the back of the indictment and on the typed list of Crown witnesses.  Mr. Cogan saw the indictment on which his name appears.  What is left of Mr. Cogan’s file in this matter includes the transcript of the preliminary hearing.  Stapled to the transcript was a handwritten witness list, which included the name Mr. Loyer, and beside his name was written “bingo”.

[64]         Having described the new information referred to by the Minister in the first question, I now turn to assess whether that information is admissible on appeal.

[65]         Informing this assessment is the voluminous written record filed by the parties, as well as the written submissions of the appellant and the Crown totalling over 1000 pages, and oral argument that extended for two weeks. In addition, between January and March 2008, this court heard the testimony of six individuals, including Mr. Lindsay, Mr. Cogan and Detective McCombie. The ensuing analysis will refer to the pertinent aspects of this record.  

[66]         I start with the observation that the statutory declaration of Ms. Brazeau forwarded to Detective McCombie by the New Liskeard police, which prompted the authorship of the April 12, 1968 report, contained information that was potentially significant to the ongoing investigation of the Roy murder. If credible, it provided strong circumstantial evidence implicating the appellant as Mr. Roy’s killer and arguably served to bolster Mrs. Roy’s tentative but inconclusive identification of the appellant as her husband’s killer. Her inconclusive identification was the reason the police had freed the appellant following the line-up that took place on August 15, 1967. 

[67]         Instead of providing Detective McCombie with an important lead into a notorious and as yet unsolved murder case, the New Liskeard police’s efforts were met with Detective McCombie’s statements in the April 12, 1968 report that a service station operator in Trenton had verified that on August 9, 1967, the appellant was at the station between 12 and 1:00 p.m., “therefore making it impossible for him to return to Ottawa by 2:45 p.m. at the time the murder was committed” and that he and Lieutenant Welsh of the Ottawa police did “not believe” the appellant was “responsible for [the Roy] murder.” Detective McCombie also noted that Mr. Barbe confirmed having last seen the appellant on the night before the murder – a statement that was consistent with the information from the service station operator that the appellant had a substantiated alibi.

[68]         On this Reference, the appellant maintains that he had no knowledge of the April 12, 1968 report or its contents, either at his trial or on his appeals.  He claims that he first became aware of the substance of the report in 1998 when his parole officer handed him the envelope containing various Ottawa police reports, including the April 12, 1968 report.[8]

[69]         The appellant relies on the April 12, 1968 report to show that his conviction constitutes a miscarriage of justice. At the risk of oversimplification, his submissions are twofold.

[70]         First, he submits that the contents of the report and the lack of any convincing rebuttal evidence demonstrate that the police and the Crown proceeded against him in the face of a verified alibi, presumably under the mistaken belief that the appellant’s signed confession trumped a verified alibi. If he is correct that the Crown prosecuted him in the face of a verified alibi, the appellant maintains and the Crown agrees that his conviction constitutes a miscarriage of justice. The appropriate remedy would be for this court to set aside the conviction and enter a verdict of acquittal.

[71]         In advancing the argument that he was the victim of a miscarriage of justice because the Crown prosecuted him in the face of a verified alibi, the onus is on the appellant to show on a balance of probabilities that he had such an alibi. As this court explained in the Reference, R. v. Truscott (2007), 2007 ONCA 575 (CanLII), 225 C.C.C. (3d) 321, at para. 74: “The appellate nature of these proceedings dictates that the appellant carry the burden of demonstrating based on evidence admitted on this Reference that there has been a miscarriage of justice.” The court in Truscott further explained at para. 75 that, in addition to carrying the ultimate burden of persuasion of demonstrating a miscarriage of justice, “an appellant also bears the onus of establishing any factual assertions that are material to arguments advanced in support of a motion to adduce fresh evidence.”

[72]         The appellant’s second submission is that if, contrary to the evidence marshalled by him, he has failed to show, on balance, that the police did not discredit his verified alibi at some point after April 12, 1968, his conviction nonetheless constitutes a miscarriage of justice if we are satisfied, on balance, that the contents of the April 12, 1968 report were not disclosed to him either before or during his trial.  In support of that position, the appellant submits that the Crown was obliged to disclose the contents of the April 12, 1968 report and that the Crown’s failure to do so deprived him of his right to a fair trial. 

[73]         The Crown counters the appellant’s submissions with the following series of submissions. In regard to the first submission, the Crown submits that the appellant has failed to show, on balance, that Detective McCombie did not discredit the alibi referred to in the April 12, 1968 report at some point after writing this report.

[74]         As for the appellant’s second submission, the Crown submits that he has failed to show, on balance, that the contents of the April 12, 1968 report were not disclosed to him. Alternatively, the Crown maintains that in 1972, there was no obligation on the Crown to disclose the contents of the April 12, 1968 report. Hence, trial fairness concerns do not arise.

[75]         Finally, in respect of trial fairness, the Crown maintains that if this court should find that the contents of the April 12, 1968 report were not disclosed to the appellant and that the Crown had an obligation to do so, the undisclosed evidence does not have sufficient probity to warrant its reception on appeal. It is fanciful, the Crown submits, to think that there is a reasonable possibility that the result at the appellant’s trial could have been different had the jury been apprised of a once verified, but ultimately discounted alibi.

[76]         In assessing the relative merits of the parties’ submissions, this court must first assess whether the appellant has met the onus of establishing on a balance of probabilities that the verified alibi described in the April 12, 1968 report was not discredited by Detective McCombie sometime after he wrote that report.

[77]         In assessing this issue, I have considered the whole of the record. Regrettably, after doing so, I am unable to determine one way or the other whether Detective McCombie did or did not discredit the verified alibi as described in his April 12, 1968 report. Credible evidence pointing in one direction is counterbalanced by equally credible evidence pointing in the opposite direction.

[78]         From 1998, when the April 12, 1968 report surfaced, Detective McCombie has consistently maintained to those who have questioned him about the contents of his report (including members of the Innocence Project, Mr. Briggs on behalf of the Minister, and in his testimony before this court) that at some point, he travelled to Trenton and spoke to the owner of the service station where the appellant’s car was repaired in exchange for the car radio. According to Detective McCombie, he retrieved both the radio and the tow records relating to the appellant’s car from the service station operator. Those records, he claimed, discredited the information from the service station operator upon which he had based his assertion in the April 12, 1968 report that the appellant was in Trenton at a time that made it impossible for him to return to Ottawa by the time of the murder. The tow records showed that it was physically possible for the appellant to have been in Ottawa at the time of the murder. Hence, the so-called Trenton alibi was no longer sustainable. Detective McCombie also claimed in his testimony in 2008 before this court that he wrote a supplementary report following his trip to Trenton, which he attached to the bag containing the radio and tow records and placed in the property room of the Ottawa Police Department.  The radio, tow records and report have not been found.

[79]         Unfortunately, Detective McCombie’s post-1998 recollections give rise to as many questions as they do answers. Part of the difficulty lies in the fact that potentially confirmatory evidence that may once have existed is missing. My best assessment of the matter leads me to conclude that the radio, the tow records, and the occurrence report that Detective McCombie claims to have prepared upon his return from Trenton – if these items in fact existed and were deposited there – had gone missing from the property room by January 1972 when the appellant was charged with Mr. Roy’s murder. Otherwise, it would have been a simple matter to overcome the apparent conundrum created by the April 12, 1968 report and particularly its reference to the appellant’s having a verified alibi. This conundrum obviously drew the attention and concern of Crown counsel, Mr. Lindsay, and several Ottawa police officers including Detectives Huneault, Nadori and Lieutenant Lowry.

[80]         By way of illustration, an investigation report of Detectives Nadori and Huneault dated January 23, 1972, indicates that on January 21 and 22, 1972, these officers attempted to locate the service station near Trenton where the appellant claimed to have exchanged his car radio for gas and services.  They conducted this investigation while on their way to and from Richmond Hill, where they went to interview the former girlfriend of Donald Phillion, Gail Chartrand, in relation to the murder investigation.  The pertinent portions of their report state:

[81]         Had the occurrence report and tow records been available in January 1972, there would have been no mystery about the identity of the service station owner or the location of the service station.  As well, the police officers would have been able to track down the towing company that brought the appellant’s car to the service station. To the extent that Detectives Huneault and Nadori felt the need to do a follow-up investigation after the appellant’s arrest in January 1972, they could simply have gone to these sources instead of engaging in the wild and unproductive goose chase described in their investigation report.

[82]         Absent confirmatory evidence, we are essentially left to rely on Detective McCombie’s memory – a memory that has understandably faded with the passage of time and that accounts, as best as I can tell, for the confusing evidence he has given about the circumstances surrounding the initial verification of the Trenton alibi and its later discounting.  At least I am not prepared to find that his confusing evidence is the product of anything other than a faded memory, given the impressive body of evidence attesting to Detective McCombie’s distinguished career as a police officer and his unblemished reputation for professionalism and integrity.

[83]         Just as I find Detective McCombie’s memory of distant events to be largely unhelpful, so too is the evidence of other witnesses who testified before us and who purported to recall events dating back to 1968 and 1972. Without in any way doubting the personal or professional integrity of these witnesses or the sincerity of their testimony, the weight to be given to their evidence is greatly attenuated by the lengthy passage of time. In Mr. Lindsay’s case, he testified that he has virtually no memory of the pertinent events or documents surrounding the prosecution of the appellant in 1972. The great bulk of his evidence is therefore couched in terms of what he would have done or believes he would have done, as opposed to what he actually remembers doing. This is not meant as a criticism; it merely reflects the frailties of human memory.

[84]         Given my view that present-day memories are largely unhelpful and do not provide a fertile source for determining whether Detective McCombie did or did not discredit the Trenton alibi, I am left to decide the matter on the basis of the documentary record and legitimate inferences that can be drawn from it, as well as the personal and professional reputations of Mr. Lindsay and the several Ottawa police officers involved in the investigation.

[85]         The reputations of Mr. Lindsay and the Ottawa police officers are relevant despite the appellant’s insistence that he is not challenging their personal or professional integrity.  According to the appellant, his position is merely that the Crown and the police presumably thought that the appellant’s confession trumped his verified alibi.  Yet clearly, in my view, a finding by this court that the Crown and the investigating officers proceeded against the appellant on the belief that a signed confession trumped an alibi which had been verified by one of those same officers carries with it a connotation of unethical, if not unlawful, behaviour on their part. One need only state the proposition – the appellant could not have committed the murder because he was several hundred kilometres away from Ottawa when it occurred, but we will prosecute him anyway because we have his signed confession – to realize its highly unethical implications.

[86]          Leaving aside Detective McCombie for the moment, I find it virtually inconceivable that Mr. Lindsay and the Ottawa police officers involved in the investigation would have participated in prosecuting the appellant for the Roy murder knowing that he was innocent of the crime. I acknowledge that people of integrity have acted improperly in individual cases: see The Inquiry Regarding Thomas Sophonow:  The Investigation, Prosecution and Consideration of Entitlement to Compensation (Winnipeg: Manitoba Justice, 2001) (Commissioner: The Honourable Peter deCarteret Cory). However, in this case, I reject the appellant’s submission that one or more of these individuals may have been caught up in the notoriety of the case and viewed it as an opportunity to advance their careers. As the Crown points out, far from advancing their careers, detection would have spelled professional ruination. And detection was a real possibility considering that references to the appellant’s trip to the Trenton service station were all over the file – a file in which multiple copies of the April 12, 1968 report were located, including a copy that had Mr. Lindsay’s notations all over it. Mr. Lindsay and the police knew this and they knew as well that a judge could order production of any of the contents of the police’s investigation file at any time. 

[87]         As for Detective McCombie in particular, I cannot gainsay the possibility that he may, on his own, have misled his fellow police officers and Mr. Lindsay into believing that he had discredited the Trenton alibi when in fact he had not. The lack of any documentation confirming his trip to Trenton after April 12, 1968 is disconcerting; so too is the loss of the radio and tow records he claims to have retrieved. That said, it must be remembered that by 1972, when the appellant was charged, many exhibits in this case could not be located. Some had gone missing; others were apparently destroyed. As well, Detective McCombie’s notes could not be found.

[88]         Moreover, I find it hard to believe that Detective McCombie would have provided the Crown and his fellow officers with false information, knowing that in doing so he would be promoting the prosecution of a man whom he knew was innocent. In this regard, I note that by January 11, 1972 when the appellant was charged with the murder, Detective McCombie already had been promoted to the rank of Lieutenant and he was no longer in charge of the Roy murder investigation. He stood to gain little, if anything, from its successful resolution. I note as well that in the period from August 9, 1967 to January 1, 1972, when Detective McCombie had carriage of the file, the record reveals that he treated the appellant fairly and professionally. The record also demonstrates his continued willingness to follow-up on various leads relating to other possible suspects, thus foreclosing any accusation that he rushed to conclude that the appellant was the perpetrator of the crime. Detective McCombie’s even-handed investigation undercuts the notion that he would have misled Mr. Lindsay and his fellow officers in order to bring about the prosecution of a man he knew to be innocent of the crime.

[89]         For all of that, I acknowledge that there are disturbing features in the documentary record that give me cause for concern – not the least of which are a host of unanswered and now unanswerable questions, including:

[90]         These are but a few of the perplexing questions that remain unanswered. Both the appellant and the Crown have put forward hypotheses that legitimate the positions they seek to advance.

[91]         In a nutshell, it is the appellant’s contention that, shortly after he was arrested in New Liskeard on August 13, 1967, he advised Detective McCombie of the Trenton alibi and Detective McCombie drove to Trenton and verified it. According to the appellant, this sequence of events explains Detective McCombie’s emphatic language in the April 12, 1968 report. It also explains comments found in two earlier occurrence reports, one in September 1967 in which Detective McCombie reported that the Ottawa Police Department no longer “had any interest” in the appellant and the other in March 1968 in which a fellow Ottawa police detective reported that after an extensive investigation shortly after the murder, Detective McCombie “finally did eliminate [Romeo and Donald] as suspects.”

[92]         According to the appellant, those reports, along with the implausibility of a scenario in which Detective McCombie went to the service station after writing the April 12, 1968 report, the absence of any evidence confirming such a trip, and the absence of any reference to it in the January 1972 report prepared by Detectives Huneault and Nadori following their unsuccessful efforts to locate the relevant Trenton service station, provide strong proof that Detective McCombie did not discredit the Trenton alibi.

[93]         Equally strong proof, the appellant submits, is found in Detective McCombie’s less than forthcoming responses at the preliminary hearing to questions posed by Mr. Cogan that should, if Detective McCombie were being candid, have revealed the existence and origin of the April 12, 1968 report and any follow-up investigation he may have conducted. According to the appellant, Detective McCombie’s less than forthcoming responses reveal a concerted effort on his part to keep Mr. Cogan in the dark – something that would have been unnecessary had he in fact discredited the Trenton alibi.

[94]         The Crown, on the other hand, maintains that when Detective McCombie wrote the April 12, 1968 report, he had not been to the Trenton service station. It was only after April 12, 1968 that he made the trip, retrieved the radio and tow records, and realized that the appellant’s Trenton alibi was no longer substantiated.

[95]         In support of its position, the Crown points out that if, as the appellant contends, Detective McCombie personally attended the service station and verified the alibi prior to April 12, 1968, he would have told this to the New Liskeard police either before they had interviewed Gail Brazeau and the appellant or immediately after, when the New Liskeard police informed him of Gail Brazeau’s statutory declaration. A simple response to New Liskeard along the lines of “you are barking up the wrong tree” would have saved these officers a good deal of time and trouble.

[96]         Moreover, if Detective McCombie had personally gone to Trenton and verified the alibi before April 12, 1968, he would not have felt the need to re-interview Mr. Barbe and Gail Brazeau’s mother regarding the day on which they had last seen the appellant in Ottawa prior to the murder. That too would have been a wasteful exercise on his part.

[97]         In sum, according to the Crown, Detective McCombie’s reactions and responses lend credence to his position that while he may have received verification of the Trenton alibi by phone, or received information about the alibi and believed it to be true when he wrote the report, it was only after April 12, 1968 that he actually journeyed to Trenton and received the information that served to discredit the alibi.

[98]         As for Detective McCombie’s failure to resume the investigation against the appellant after the alibi was discredited, the Crown points out that the discounting of the Trenton alibi did not provide the police with any new evidence that could be used to prove the appellant’s guilt, except possibly as after-the-fact-conduct evidencing guilt if the appellant should raise the Trenton alibi as a defence. And regarding Detective McCombie’s failure to act on the information provided by Gail Brazeau, as Detective McCombie noted in the April 12, 1968 report, she had shown herself to be highly unreliable in the past and she now had an apparent motive for implicating the appellant in a crime he did not commit (she was alleging that he had forced her into prostitution). It followed that her statutory declaration would not have advanced the case against the appellant in any significant way.

[99]         Regarding the significance of Detectives Huneault and Nadori’s fruitless efforts to locate the Trenton service station, the Crown replies that Detective McCombie could not have known that they would come up empty-handed. Moreover, had he not discredited the Trenton alibi, it is unlikely he would have provided them with information about the whereabouts of the service station in an effort to assist them in locating it.

[100]      In addition to these factors that support Detective McCombie’s recent testimony and that rebut the appellant’s contention that he did not discredit the Trenton alibi, the Crown relies most heavily on an investigation report written by Detective McCombie on April 2, 1971. In that report, Detective McCombie recounted how on March 25, 1971, the appellant appeared at his office and informed him that his twin brother, Donald, who was serving a life sentence for murder in the Archambault Institute in Montreal, wished to confess to Mr. Roy’s murder. The appellant also told him that Donald would tell Detective McCombie where the knife was. 

[101]      Despite having cleared Donald of the murder as early as August 13, 1967 on the basis of his employment records, which showed that he was in New Liskeard at the time of the murder, Detective McCombie travelled to Archambault and met with Donald. Donald denied having made these statements to his brother and Detective McCombie returned to Ottawa empty-handed. In the concluding paragraph of his investigation report detailing those events, Detective McCombie wrote as follows:

[102]      According to the Crown, that statement – coming as it did well after the April 12, 1968 report and eight-and-a-half months before the appellant confessed – provides strong proof that Detective McCombie had indeed discredited the Trenton alibi sometime between April 12, 1968 and April 2, 1971.

[103]      The factors that I have identified are but a selected sample of the factors that the parties have referred to in support of their respective positions for and against a finding that Detective McCombie discredited the Trenton alibi. They are not meant to be exhaustive. Rather, they illustrate the conundrum that a court is faced with in viewing the matter some 40 years later.

[104]      In the final analysis, after considering all of the pertinent evidence and the arguments put forward by the parties, I am regrettably at a stalemate and find myself unable to determine one way or the other whether Detective McCombie did or did not discredit the Trenton alibi. The evidentiary scales are equally balanced. Considering that the appellant bears the onus of showing on a balance of probabilities that the alibi was not discredited, the appellant has failed to meet his burden.

[105]      That, however, does not end the matter. In my view, the issue whether the Trenton alibi was ever discredited is one of vital importance. It should have been resolved by the jury at the appellant’s trial in 1972. That did not occur because, as I shall endeavour to explain, Mr. Cogan had no knowledge of the April 12, 1968 report or its contents or any follow-up investigation that Detective McCombie may have conducted after writing the report. In my view, the evidence in that regard is overwhelming and given its importance, both in my response to question one and the ultimate disposition of this Reference, I propose to review it in some detail.

[106]      Arthur Cogan, the appellant’s lawyer at trial, testified before us in January 2008 to the effect that he had no knowledge of the April 12, 1968 report or its contents during his representation of the appellant; nor was he ever made aware of a follow-up investigation conducted by Detective McCombie in relation to the Trenton alibi.  The trial prosecutor, Malcolm Lindsay, also testified before us in January 2008 and swore that he could not remember one way or the other whether he disclosed the contents of the April 12, 1968 report to Mr. Cogan, but he believed that he would have done so based on his practice.

[107]      In line with what I have already said about the frailties of people’s memories dating back to 1972, I do not place a great deal of weight on their present-day recollections. What is known with certainty is that Mr. Lindsay was fully aware of the April 12, 1968 report and its contents. As he acknowledged before us, his markings appear on one of the copies of the report that was found in the archives. Before us, Mr. Lindsay confirmed that he had underlined the paragraph in the report relating to the Trenton alibi, as well as portions of the paragraph relating to Detective McCombie’s re-interview of Mr. Barbe, and those portions of the statement in which Detective McCombie asserted on behalf of himself and Lieutenant Welsh the opinion that the appellant was not the person responsible for Mr. Roy’s murder.[9]

[108]      What is also known with certainty is that neither the April 12, 1968 report nor the contents of that report were found in the Crown Brief.  I am satisfied that the Crown Brief, containing a synopsis of the case and the witness willsays would have been, and was, provided to Mr. Cogan under the disclosure regime that existed in 1972. In addition, it is known with certainty that the documents found in the Crown Brief did not refer to any follow-up investigation by Detective McCombie leading to the discounting of the Trenton alibi.

[109]      The willsay statements in the Crown Brief of Detectives McCombie, Huneault and Nadori are significant in this regard, not so much for what they say about the “once verified/later discredited” Trenton alibi but for what they do not say. Of all of the Crown witnesses, these three officers were most likely to have had the greatest familiarity with the contents of the April 12, 1968 report: Detective McCombie because he wrote it and because he would have been questioned about it in January 1972 (an inference which no one contests);  Detectives Huneault and Nadori because they referred to the April 12, 1968 report in their January 13, 1972 investigation report, which included a ten-item “to do list”, following which appeared the passage - “Attempt to clarify the contents of the investigation report submitted by Lt. McCombie dated April 12, 1968” - and because they attempted to locate the service station referred to in the April 12, 1968 report on January 21 and 22, 1972.

[110]      With that in mind, I reproduce here the passages from these officers’ willsay statements that bear on the April 12, 1968 report and its contents.

[111]      Detective McCombie’s willsay statement includes the following passage:

[112]      Turning next to Detective Huneault’s willsay statement, the following information appears:

[113]      As for Detective Nadori’s willsay statement, he merely recites that he will corroborate the evidence of Detective Huneault in relation to “all other investigation[s] conducted after the arrest of [the appellant]”.

[114]      Thus, there is nothing in the willsay statements of these three officers that would have alerted Mr. Cogan to the “once verified/later discredited” Trenton alibi. Indeed, Detective Huneault’s willsay statement refers to the radio being traded “on the evening or early morning of August 9th or 10th, 1967”, which would not have provided an air-tight alibi, whereas Detective McCombie’s April 12, 1968 report places the radio exchange at between 12 noon and 1:00 p.m. on August 9, which would have made it impossible for the appellant to have committed the murder given that Trenton is some 290 kilometres from Ottawa and the murder occurred at approximately 2:45 p.m.

[115]      Moving on to the evidence at the preliminary hearing, it is apparent from the record that Mr. Cogan was interested in exploring alibi, particularly in relation to Trenton. In his cross-examination of Detective McCombie, he asked the following question:

[116]      Before Detective McCombie could answer, the following exchange between Mr. Cogan and Mr. Lindsay occurred:

[117]      A similar foray into the realm of alibi occurred during Mr. Cogan’s cross-examination of Detective Norton. By the time of the preliminary inquiry, Mr. Cogan had been provided with the appellant’s statements to the police. One such statement was taken by Detective Norton at 7:30 a.m. on the morning of January 12, 1972 (the day after the appellant had provided a signed confession to Detectives Huneault and Nadori).  The January 12, 1972 statement concluded with the following question and answer:

[118]      Armed with that statement, Mr. Cogan asked Detective Norton in cross-examination:

[119]      At that point, Mr. Lindsay objected as follows:

[120]      Further discussions between the presiding judge and Messrs. Lindsay and Cogan ensued, in which Mr. Cogan stated, among other things, that:

[121]      In response, Mr. Lindsay repeated his position that it was for the Crown to determine whether statements of the accused would be led and that it did not lie with the defence to adduce such evidence under the pretext of putting the Crown on notice of an alibi. In the course of these submissions, Mr. Lindsay made the following comment:

[122]      In the end, the presiding judge ruled in favour of the Crown, holding: “I must agree with Mr. Lindsay that it would not be within the purview of [Mr. Cogan’s] cross-examination to obtain [a] conversation from Detective Sergeant Norton on the basis of an alibi as an exception to the general principle” that the Crown had the discretion to decide which statements of the accused it would tender in evidence.

[123]      These passages demonstrate that alibi was very much on Mr. Cogan’s mind at the preliminary hearing. It is useful, then, to consider both the manner of questioning and the content of the questions that Mr. Cogan posed to Detectives McCombie and Huneault for purposes of assessing whether he knew about the contents of the April 12, 1968 report and in particular, the facts and circumstances surrounding the “once verified/later discredited” Trenton alibi.  It is also useful to assess whether the answers he received put him any further ahead by the conclusion of the preliminary hearing.

[124]      Commencing with Detective McCombie, I propose to identify particular questions and answers that in my view shed light on the state of Mr. Cogan’s knowledge about the Trenton alibi, or his lack thereof, at the preliminary hearing. Before doing so, however, on a more general level, I find illuminating the open-ended nature of Mr. Cogan’s questions, certainly insofar as they relate to any investigative activities that Detective McCombie may have carried out in 1968 relating to the Trenton alibi.

[125]      I also note that Detective McCombie was giving evidence without the benefit of his notebook. His notes had gone missing and he could not recall when he had last seen them prior to January 1972, indicating that: “[i]t may have been two or three years ago.”  In response to a question from the preliminary hearing judge, he claimed that, apart from refreshing his memory as to the date of Mr. Roy’s murder, he was giving evidence from memory.  Be that as it may, there can be no doubt that sometime between the date of the appellant’s confession on January 11, 1972 and Detective McCombie’s testimony at the preliminary inquiry on April 6, 1972, he would, at a minimum, have reviewed the contents of the April 12, 1968 report. 

[126]      In cross-examination, Mr. Cogan established that Detective McCombie was the chief investigating officer in charge of the Roy murder investigation from August 9, 1967, the date of the murder, until January 1, 1972, when he was promoted to the rank of Lieutenant. Apart from his dealings with the appellant from August 13 to August 15, 1967, Detective McCombie acknowledged that he had no further dealings with the appellant other than in April 1971, when the appellant came to his office and provided him with the information about his twin brother, which resulted in the unproductive trip to see Donald Phillion at Archambault Institute.

[127]      Detective McCombie also testified about two further occasions when he was involved with the file between August 1967 and January 1972: one was during the Coroner’s inquest into the death of Mr. Roy held in November 1967; the other instance related to the investigation of a third party suspect in 1969 or 1970. The following exchange between Mr. Cogan and Detective McCombie captures the content of this evidence:

[128]      Mr. Cogan also asked Detective McCombie a number of questions about his dealings with the New Liskeard police. In this regard, it will be recalled that in his willsay statement, Detective McCombie mentioned some dealings in April 1968 with the New Liskeard police related to information New Liskeard had obtained from Gail Brazeau. According to his willsay, while that information led to further investigation, “no new evidence was discovered.” Presumably, with this passage from the willsay in mind, Mr. Cogan questioned Detective McCombie about the New Liskeard police’s involvement as follows:

[129]      Detective McCombie was later asked about the nature of the New Liskeard police’s involvement in the murder investigation, both in August 1967 and at any time after that:

[130]      Finally, towards the end of his cross-examination, Mr. Cogan put the following questions to Detective McCombie:

[131]      Three observations arise from those exchanges:

[132]      Turning next to Mr. Cogan’s cross-examination of Detective Huneault at the preliminary inquiry, it will be recalled that in his willsay statement, Detective Huneault mentioned a trip to Trenton on January 2l and 22, 1972, in which he and Detective Nadori had searched unsuccessfully for a Supertest service station at which the appellant alleged he traded a radio for gas and service. With that in mind, Mr. Cogan cross-examined Detective Huneault as follows:

[133]      Once again, three observations arise from that exchange:

[134]      In sum, as with Detective McCombie, the answers given by Detective Huneault could hardly be described as enlightening. They left Mr. Cogan no further ahead as to the contents of the April 12, 1968 report, the circumstances giving rise to it, or any subsequent steps Detective McCombie may have taken in relation to the information contained in it.

[135]      One final point from the preliminary inquiry illustrates Mr. Cogan’s lack of awareness of the contents of the April 12, 1968 report, certainly at that juncture.

[136]      Mr. and Mrs. Barbe gave evidence at the preliminary inquiry. Both testified that the appellant was in their company at about 10:00 p.m. on Wednesday August 9, the night of the murder. Mrs. Barbe testified that she did not see the appellant in Ottawa again for some days after that, while Mr. Barbe testified that he did not see him again after that night.

[137]      In cross-examination, Mr. Cogan tried unsuccessfully to have the Barbes admit that perhaps they had seen the appellant on a different Wednesday, not Wednesday the 9th. Significantly, at no time did Mr. Cogan confront Mr. Barbe with the suggestion that on two prior occasions, much closer to the time of the murder, he had told Detective McCombie that he had last seen the appellant on Tuesday August 8, the night before the murder, not Wednesday August 9, the night of the murder. Nor were similar questions put to Mrs. Barbe.

[138]      Surely, if Mr. Cogan had been made aware of the April 12, 1968 report or its contents, he would have questioned the Barbes and Detective McCombie about this important inconsistency. Once again, the silence of the transcript bears witness to his lack of awareness.

[139]      Whereas the preliminary inquiry transcript speaks eloquently through its silence to Mr. Cogan’s lack of awareness of the contents of the April 12, 1968 report, the trial record does so even more persuasively. Astoundingly, the jury sat through the entire trial hearing the word “Trenton” only once, when Detective Huneault in his examination in chief repeated the appellant’s confession and recounted him stating that he drove to Trenton at around 11:30 on the night of the murder. This is all the more astonishing when one considers that after the preliminary inquiry and before the trial, Mr. Lindsay believed that the appellant might be advancing alibi (as confirmed by his pre-trial notes found in the archives) and Trenton stood out among the various alibi locations identified by the appellant as the one most likely to be pursued. 

[140]      Nor, as the Crown acknowledges, had Mr. Cogan given up on raising alibi going into the trial. Indeed, as the trial record shows, he spent a good deal of effort trying to persuade the trial judge to admit the appellant’s statement to Detective Norton on the morning of January 12, 1972, in which the final question and answer between Detective Norton and the appellant read as follows:

[141]      In acknowledging that alibi was still in play at the trial, the Crown in its factum contends that Mr. Cogan’s purpose for seeking to admit the so-called Norton statement was as follows:

[142]      For reasons that need not be discussed, the trial judge refused to admit the Norton statement and the jury heard nothing of it. It was only then, after suffering that setback that Mr. Cogan apparently gave up on alibi. Again, as the Crown put it in its factum: “Once the Norton statement was excluded, then there was no way to raise the Trenton alibi or the lost Trenton tow records without the appellant testifying, which was not a viable option.” In out-of-court discussions with Mr. Lindsay sometime after the trial judge’s ruling, Mr. Cogan agreed not to contest the appellant’s presence in Ottawa at the time of the murder. I hasten to point out that this was not a formal admission of fact by Mr. Cogan.  As is apparent from the record, in return for that concession, Mr. Cogan understood that he was obtaining the Crown’s agreement not to lead Mrs. Barbe’s evidence concerning the appellant’s statement to her husband that he wanted to sell him his car and leave town.

[143]      Once it is accepted that Mr. Cogan commenced the trial with alibi still in play, several questions come to mind:

[144]      These questions remain equally perplexing if we are to assume that Mr. Lindsay and/or Detective McCombie disclosed the substance of the April 12, 1968 report and the details of any follow-up investigation at some point during the trial, before Mr. Cogan agreed that he would not contest the appellant’s presence in Ottawa at the time of the murder. If that disclosure had occurred, surely it would be reflected on the record. Mr. Cogan would have brought it to the attention of the trial judge and at the very least, sought an opportunity to test the newly-disclosed evidence on a voir dire.

[145]      But the record is silent. And it is silent because in my view, no such disclosure occurred.

[146]      In sum, the April 12, 1968 report would have been gold in Mr. Cogan’s hands. Had he known of its contents and the fact that not one shred of documentary or physical evidence existed to confirm Detective McCombie’s alleged discounting of the once verified alibi, he would have tracked it down to the last detail. But he did not – not at the preliminary hearing and not at the trial. And he did not because, as I have endeavoured to show, the evidence demonstrates overwhelmingly that he did not know about it.

[147]      With that, I turn to consider what consequences, if any, flow from the fact that Mr. Cogan had no knowledge of the contents of the April 12, 1968 report or any follow-up investigation that Detective McCombie may have conducted in relation to it.

[148]       In R. v. Dixon, 1998 CanLII 805 (SCC), [1998] 1 S.C.R. 244 and R. v. Taillefer, 2003 SCC 70 (CanLII), [2003] 3 S.C.R. 307, the Supreme Court of Canada established that a breach of the Crown’s duty to disclose relevant evidence to the defence at trial is one of two possible bases for admitting fresh evidence on appeal. The appellant submits that Mr. Lindsay had a duty to disclose the contents of the April 12, 1968 report to his counsel and that the failure to do so rendered the trial unfair because it hampered the appellant’s ability to make full answer and defence.  

[149]      Appellants who seek to tender this kind of fresh evidence on appeal must show an entitlement to the undisclosed evidence. If the Crown was not obliged to disclose it, unfairness will not have been made out and the application must fail. Thus, in this case, the appellant bears the burden of establishing that Mr. Lindsay was under a duty to disclose the April 12, 1968 report or its contents. If he fails to meet that burden, his application to admit the proposed fresh evidence on the basis of trial unfairness will not succeed.

[150]      The answer to the question whether Mr. Lindsay had a duty to disclose the report or its contents lies not in the dictates of our present-day disclosure regime but in the dictates of the disclosure regime that existed in 1972. In this regard, it would hardly be fair to criticize as “unfair” Crown disclosure practices that would not pass muster under today’s standards, but which were acceptable under the standards that prevailed in 1972.  As this court stated in Truscott, at para. 122:

[151]      With that in mind, I begin my analysis with the undisputed observation that even in 1972, it would have been incumbent on the Crown to disclose to the defence an alibi that was verified by the police. But that is not this case, necessarily. As explained earlier, whether Detective McCombie did or did not discredit the Trenton alibi is something that remains a mystery.  Regrettably, this question was not resolved when it should have been – by the jury at the appellant’s trial. Be that as it may, in my view, Mr. Lindsay was entitled to rely on the representations of the lead investigating officer that he had discredited the once verified Trenton alibi. Such reliance is particularly unobjectionable when the information that Mr. Lindsay received from Detectives Huneault and Nadori following their trips to Trenton in January 1972 would not have given him any cause to doubt the discounting of the alibi. In addition, the information conveyed to him by Detectives Huneault and Nadori regarding the Barbes’ statements to these officers in January 1972, indicating that the Barbes last saw the appellant on the night of the murder, was similarly consistent with the discounting of the Trenton alibi.

[152]      In these circumstances, the question to be decided is not whether Mr. Lindsay was obliged to disclose a verified alibi, but whether he was obliged to disclose an alibi that he believed had been discredited. In response, I know of no authority and none was presented to us, that would cause me to conclude that such an obligation existed in 1972 – and given the wide discretion that Crown counsel enjoyed in that era, I doubt that it did.

[153]      Manifestly, even in those times, Crown counsel could not withhold information that supported an accused’s innocence. But an apparently discredited alibi hardly fits that description. And, in my view, if an obligation could be said to exist it would only arise, if at all, once the accused had notified the authorities of the alibi upon which he or she proposed to rely.

[154]      That of course did not occur here. At no time did Mr. Cogan notify the police or Mr. Lindsay of a particular alibi upon which the appellant intended to rely. Absent such notice, Mr. Lindsay was under no obligation to disclose a piece of evidence that would, in the circumstances, have been irrelevant to a fact in issue.

[155]      And that, in my view, explains why Mr. Lindsay was so insistent at the preliminary inquiry that the appellant should not be permitted to give notice of an alibi through the back door, as it were, by means of out-of-court statements he had made to the police.  As Mr. Lindsay maintained in his submissions at the preliminary, if the appellant was going to rely on alibi, it was incumbent on him to raise it through admissible evidence or through proper notification procedures.

[156]      In so concluding, I have not ignored Mr. Lindsay’s testimony before us to the effect that he personally (and perhaps all Crown attorneys) would have felt obliged to disclose a discredited alibi, just as he would a verified alibi. While I do not doubt the sincerity of his testimony, I think, with respect, that present-day practices have crept into his thinking and have coloured his perception of the practices that he considered were appropriate in 1972.

[157]      My thinking in this regard is confirmed by Mr. Lindsay’s similar belief that he was also duty-bound to disclose prior inconsistent statements of witnesses and specifically, the alleged inconsistent statements made by Mr. Barbe concerning his last sighting of the appellant in Ottawa in relation to the murder.  Even the appellant does not contend that the Crown had any duty to disclose prior witness statements in 1972. Indeed, in his factum, he concedes, correctly in my view, that the decision to disclose prior inconsistent statements was a matter within the Crown’s discretion.

[158]      Regardless, it is not Mr. Lindsay’s personal beliefs about the dictates of the 1972 disclosure regime that count. Given my conclusion that he did not disclose the April 12, 1968 report or its contents to Mr. Cogan, what matters in my view, is the propriety of his conduct, having regard to the disclosure obligations that were binding on him in 1972.

[159]      In that regard, for reasons already stated, the appellant has failed to persuade me that Mr. Lindsay was obliged in the circumstances to disclose the April 12, 1968 report or its contents to Mr. Cogan. It follows that the proposed fresh evidence is not admissible on the basis that the trial process was unfair.

[160]      Even though the Crown was not obliged to disclose the contents of the April 12, 1968 report and thus the fresh evidence is not admissible on the basis of trial fairness, that conclusion does not end the fresh evidence inquiry.  The second basis for admitting fresh evidence on appeal is established in R. v. Palmer and Palmer, 1979 CanLII 8 (SCC), [1980] 1 S.C.R. 759, which requires this court to determine whether the proposed evidence is of sufficient cogency to warrant its admission on appeal, and if so, whether the interests of justice require that it be received. 

[161]      At its core, the Palmer test seeks to promote the interests of justice and to protect against miscarriages of justice. It acts as a safety valve in cases such as this where, through no fault of the Crown, cogent evidence comes to light that could reasonably be expected to have affected the result at trial.

[162]      This court in Truscott, at para. 92, described the Palmer test for the admission of fresh evidence on appeal as follows:

[163]       The first criterion – is the proposed evidence admissible under the usual rules of evidence governing criminal proceedings – is not contested. Detective McCombie could have been cross-examined on the contents of his April 12, 1968 report, the circumstances giving rise to it, and any investigative steps he may have taken thereafter. He and the Barbes could also have been questioned about the Barbes’ last sighting of the appellant in Ottawa in relation to the time of the murder.  Mr. Barbe could have been asked whether the information he gave to Detective McCombie in 1967 and 1968 differed from his post-charge 1972 recollections, and if so, why.

[164]      Turning then to the cogency criterion, this court in Truscott, at paras. 99 and 100, explained the nature and purpose of the cogency inquiry and the three questions that must be answered in relation to it:

[165]      Pausing here, questions one and two of the cogency inquiry pose no impediment to the appellant. The Crown does not suggest otherwise. What is at issue is the third question in this inquiry, namely whether the proposed evidence is sufficiently cogent that it could reasonably be expected to have changed the result at trial when considered in the context of the entirety of the evidence admitted on appeal and heard at trial. The Crown contends that it does not meet the required level of cogency, namely, that it could reasonably be expected to have changed the result at trial.

[166]      The Crown’s submissions hinge in large measure on the assumption that the jury would accept that Detective McCombie conducted a further investigation and that he discredited the Trenton alibi after writing the April 12, 1968 report. Assuming the jury were to make that finding, the Crown asks rhetorically: how could a discredited alibi reasonably be expected to have changed the result at trial?

[167]      As for the cogency of the prior inconsistent statement of Mr. Barbe, the Crown points out that Mr. Barbe did not testify at trial and the jury did not hear or rely on his evidence to convict the appellant. The Crown further submits that while Mr. Barbe’s alleged prior inconsistent statement could possibly have been used to cross-examine Mrs. Barbe at trial, it is questionable that the trial judge would have permitted such a cross-examination on an unsigned, prior notation that conflicted with Mr. Barbe’s sworn evidence at the preliminary inquiry. According to the Crown, the effectiveness of any such cross-examination is also dubious given that Mr. and Mrs. Barbe’s testimony regarding the night they last saw the appellant was solidly anchored in the Wednesday night bingo game that Mrs. Barbe always attended.  In any event, with the Trenton alibi discredited, some slight confusion on Mr. Barbe’s part as to the time he last saw the appellant in relation to the murder would not realistically have made any difference to the outcome of the trial.

[168]      I find the Crown’s submissions unpersuasive. They are inextricably linked to a finding that Detective McCombie travelled to Trenton after writing the April 12, 1968 report and obtained information that enabled him to discount the Trenton alibi. But the record is not at all clear on this. Indeed, as explained, my assessment of it leaves me unable to say one way or the other whether he did or did not discredit the alibi. Had he been cross-examined about this at trial, the matter would have been out in the open and the jury would have been in a good position to assess it. Unfortunately, that did not occur.

[169]      One thing, however, is certain. Had the jury been given the opportunity to assess the alleged discounting of the Trenton alibi and either disbelieved Detective McCombie or been unable to decide whether he had in fact travelled to Trenton and obtained information that enabled him to truly discredit the alibi, that circumstance alone could have had a dramatic impact on the outcome of the trial. Put simply, if the lead investigating officer was satisfied that the appellant was in Trenton near the time of the killing and that he and a superior officer were so certain of the appellant’s lack of involvement that they were prepared to forgo an important lead in the case presented by the New Liskeard Police Department, that evidence, on its own, could have left the jury in a state of reasonable doubt about the appellant’s guilt.

[170]      But, of course, that evidence would not have stood alone. It would have been considered with the once-clear recollection of Mr. Barbe, and potentially that of Mrs. Barbe as well, that the appellant was last seen by the Barbes at 10:00 p.m. on August 8, the night before the murder, not 10:00 p.m. on August 9, the night of the murder.

[171]      The possible evidence from the Barbes that may have thrown into doubt whether the appellant was in Ottawa after Tuesday August 8, 1967 could have been supported by the evidence of Mr. Loyer, as described in his willsay, that in August 1967, bingo games were held by a different organization on Tuesdays. In light of this body of evidence, the jury may well have been left in a state of reasonable doubt as to whether the appellant was in Ottawa on the day of the murder. 

[172]      Once again, however, the jury was not given the opportunity to decide this issue because part-way through the trial, Mr. Cogan agreed not to contest the appellant’s presence in Ottawa at the time of the murder. In my view, he would not have made that concession had he been aware of the substance of the April 12, 1968 report. But make it he did, thereby removing the critical issue of whether the appellant had the opportunity to commit the murder from the jury’s consideration.

[173]      In sum, there is a reasonable basis for concluding that the appellant’s trial would have proceeded very differently had Mr. Cogan been aware of the contents of the April 12, 1968 report. Had the contents been brought to the jury’s attention, there is reason to believe that the jury could well have entertained a reasonable doubt on the issue of opportunity. That being so, the proposed fresh evidence passes the cogency test in that it could reasonably be expected to have changed the result at trial.

[174]      In so concluding, I have not ignored the compelling nature of the appellant’s confessions. The level of detail and accuracy found in them is considerable. Indeed, it is tempting to conclude, as my colleague MacPherson J.A. has, that only the real killer would have been privy to that level of detail.

[175]      I acknowledge the possibility that a jury could view the appellant’s confessions in that light. But not necessarily, especially if Detective McCombie were to fare badly in cross-examination and the jury were left wondering whether he in fact ever did truly discredit the Trenton alibi.

[176]      Put succinctly, I do not consider the appellant’s confessions to be so compelling that the jury would inevitably act on them regardless of how Detective McCombie were to fare in cross-examination on the Trenton alibi and the Barbes on the apparent inconsistency in Mr. Barbe’s statements.

[177]      In that regard, while I acknowledge that the appellant’s statements contain a number of accurate details, many of them were reported by the press and thus they were in the public domain. Indeed, at trial, Detective Huneault agreed with Mr. Cogan that “the facts of this particular crime are really no mystery, if one read the newspaper clippings in 1967”. Of the twelve newspaper articles published in 1967, I have reproduced one from the Ottawa Journal, dated August 10, 1967, at Appendix D. Without reviewing it line by line, suffice it to say that it incorporates the vast majority of the twenty items my colleague has identified as “completely true or probably true”. Beyond that, there is no way of knowing what the appellant may have learned about the murder, initially from Detective McCombie and later from the New Liskeard police or any other persons with whom the police may have shared information, including for example, Gail Brazeau, the appellant’s former girlfriend.

[178]      But that is not all. As is the case with eye-witness identification, one wrong feature in a confession may overshadow or at least neutralize a host of other features that are accurate or near accurate. In this case, there is one telling feature of the appellant’s confession to the police that everyone, including the Crown, concedes is clearly wrong.

[179]      In his statement to the police, the appellant admitted to breaking into an apartment in the Friel Street building where Mr. Roy was murdered – an event which everyone agrees did not happen – and stealing a knife – another event that everyone agrees did not happen.

[180]      In my view, standing alone, that feature of the confession could leave a jury wary about the veracity and reliability of the confession as a whole.

[181]      And there is more. Two expert witnesses, Drs. Girodo and Arboleda about whom more will be said in my response to the second question posed by the Minister, testified at the appellant’s trial that he was a person who was prone to make things up and he did so to gain attention and make himself feel important. Both witnesses considered the appellant to be unreliable and they agreed that his personality was such that they would seek supporting evidence before believing anything he said.

[182]      Finally, it is noteworthy that within hours of confessing to the police, the appellant recanted his confession and denied any involvement in Mr. Roy’s murder – a position he has steadfastly maintained, without exception, for the past 37 years.

[183]      The body of evidence that I have summarized provided Mr. Cogan with the ammunition he needed to make a serious argument that the appellant’s confessions to Mr. Miller and the police were unreliable and that the jury should not act upon them, without more, to convict him. How much more persuasive Mr. Cogan could have been had he been able to combine that argument with the further argument that the appellant had, or may have had, an alibi that prevented him from being in Ottawa at the time of Mr. Roy’s murder.

[184]      Putting the matter as succinctly as I can, in my experience, that is the “stuff” that reasonable doubt is made of. Unfortunately, the jury was deprived of it in this case.

[185]      For those reasons, with full knowledge of the strength of the appellant’s confessions, I reiterate my conclusion that the proposed fresh evidence passes the cogency test in that it could reasonably be expected to have changed the result at trial.

[186]      That then leaves the third component of the Palmer test, sometimes referred to as the “due diligence” test. In its factum, the Crown does not suggest that Mr. Cogan, through diligence, could have learned about the contents of the April 12, 1968 report as it relates to the Trenton alibi. While there was some movement away from that concession in oral argument, I think the Crown’s original position is the correct one. Due diligence should not be equated with a lucky guess. The fact that Mr. Cogan might have happened upon the information related to the Trenton alibi had he asked the right questions of Detective McCombie or Detective Huneault begs the question. It smacks of Monday-morning quarterbacking and ignores that the answers he received at the preliminary inquiry, particularly from Detective McCombie, were less than forthcoming and hardly enlightening.

[187]      As for Mr. Barbe’s previous statements referred to in the April 12, 1968 report, the Crown submits that Mr. Cogan could have learned of these through due diligence. I question the validity of that submission. There is no indication in the record that Mr. or Mrs. Barbe provided a written statement to Detective McCombie at any time.  The only mention of Mr. Barbe’s original statements is found in the April 12, 1968 report. The record reveals that both the Barbes and Detective McCombie were vague in their recollection of their initial conversations and had little memory of specific details.

[188]      It follows, in my view, that the April 12, 1968 report was the only source from which Mr. Cogan could realistically have obtained the alleged prior inconsistent statements of Mr. Barbe and possibly Mrs. Barbe, if she indeed gave a prior inconsistent statement to Detective McCombie. 

[189]      Even if Mr. Cogan could somehow have ferreted out the prior inconsistent statements of Mr. Barbe and possibly Mrs. Barbe without the benefit of the April 12, 1968 report, and thus the due diligence criterion is not met, I am satisfied that the interests of justice would warrant the reception of the entirety of the April 12, 1968 report.  This is primarily because the significance of any inconsistency in the Barbes’ evidence is integrally linked to the information in the April 12, 1968 report relating to the verified alibi.  

[190]      Finally, in assessing the due diligence requirement, I feel obliged to address why Mr. Cogan needed the police to tell him about the Trenton alibi rather than being able to learn this information from the appellant himself. In many, if not most instances of alibi, it is for the client to inform counsel of his or her whereabouts at the time of the crime for purposes of leading an alibi defence. The client’s failure to do so would normally provide a full answer to the due diligence criterion, thereby rendering the proposed fresh evidence inadmissible.

[191]      But this case is anything but normal. By all accounts, the appellant at the best of times is an unreliable historian and is prone to making up tales. Indeed, at his trial, two doctors testified that absent independent confirmation, they would not place any reliance on what he had to say. That evidence was led in the context of the appellant’s confession, with a view to impressing the jury of the need to proceed carefully before relying on the appellant’s confession alone to convict him.

[192]      Without going into detail about the appellant’s psychiatric make-up or the bizarre behaviour he has engaged in, I accept Mr. Cogan’s evidence and am satisfied that the appellant was a difficult client (to say the least) and that Mr. Cogan would have had a hard time discerning when his client was telling him the truth and when he was lying to him. In that regard, I note, as mentioned earlier, that Trenton was but one of a handful of alibi locations that the appellant had mentioned in his dealings with the police.

[193]      Moreover, Mr. Cogan was hampered by the fact that the charge against the appellant came more than four years after the event. By then, as attested to by the unsuccessful efforts of Detectives Nadori and Huneault to track down the Trenton service station, Mr. Cogan would have had great difficulty confirming any information the appellant might have provided to him about Trenton. In short, it seems to me that anything that was going to be learned about the Trenton alibi after the appellant was charged in 1972 was locked away in Detective McCombie’s memory, apart of course from what appears in the April 12, 1968 report.

[194]      Thus, I would not cast aside the proposed fresh evidence on the basis of a lack of due diligence on Mr. Cogan’s part. To complete the circle, I can think of no tactical reason why Mr. Cogan would not have been interested in the contents of the April 12, 1968 report, and no reason why he would not have used its contents to challenge Detective McCombie and the Barbes had the report or its substance been made available to him.

[195]      For these reasons, I would answer question one in the affirmative.

[196]      Before leaving this matter, I wish to clarify that, although I have found that the information contained in the April 12, 1968 report meets the cogency test in Palmer, I do not want to leave the impression that the cogency test is an easy one to meet in historical cases – it is not. Here, the April 12, 1968 report derives its cogency from a constellation of factors that are unlikely to be repeated but which, in my view, coalesced in a way that make this case exceptional, if not unique.

[197]      The charge against the appellant was laid not at the time of the event, but more than four years later. By that time, important physical evidence, including a sample of hair likely belonging to the assailant, had gone missing. So too had Detective McCombie’s notes, as well as any documentation that could have confirmed the steps he allegedly took to discredit the Trenton alibi he once considered to have been verified. That the lead investigating officer was at one point convinced that the man ultimately charged with the offence had a verified alibi is itself a rare, if not unprecedented feature of this case. Added to those unusual features, the case for the Crown rested almost entirely on the appellant’s confessions – confessions that were challenged at trial as unreliable by two reputable experts retained by the appellant.

[198]      In addition to those factors, by the time the appellant was charged, his counsel would have had no effective means of exploring the Trenton alibi, even if he had wanted to do so. As I have attempted to show, by January 1972, whatever there was to know about the Trenton alibi was locked away in Detective McCombie’s mind and he did not readily give up the key to that information in the answers he provided to Mr. Cogan at the appellant’s preliminary inquiry.

[199]      These factors in combination coalesced to create the perfect storm as it were. They provide the backdrop against which I have measured the cogency of the April 12, 1968 report and the impetus for my conclusion that the cogency test has, in the exceptional circumstances of this case, been met.

[200]      The second question posed by the Minister relates to two expert reports – one prepared by Dr. Graham Turrall, a Canadian forensic psychologist, dated September 11, 2002, and the other authored by Dr. Gisli Gudjonsson, a British forensic psychologist, dated January 7, 2003. 

[201]      In 2002, Dr. Turrall completed a neuropsychological assessment of the appellant related to his conviction for Mr. Roy’s murder.  Dr. Turrall also reviewed a number of past psychiatric and psychological assessments that had been performed on the appellant while he was in custody. In his report, Dr. Turrall concluded that the appellant “continues to demonstrate some antisocial and borderline personality features.” He described the key features of the appellant’s personality traits as follows:

[202]      Dr. Gudjonsson, who examined the appellant in 2002, is put forward by the appellant as the world’s leading expert on the subject of the psychology of false confessions. His resume attests to the breadth and depth of his work in the area, which includes numerous publications and clinical studies, as well as consultations involving the reliability of confessions in over 700 cases worldwide. In addition, Dr. Gudjonsson has testified in more than 140 criminal proceedings, including prominent cases in the United Kingdom and elsewhere in which miscarriages of justice have been linked to false confessions.

[203]      Although not apposite to this case, Dr. Gudjonsson has developed two widely-used instruments for testing personality features relevant to the reliability of confessions: the Gudjonsson Suggestibility Scale and the Gudjonsson Compliance Scale. Pertinent to this case is Dr. Gudjonsson’s participation in the development of a three-part classification scheme or “typology” for describing the nature and cause of false confessions. The category relevant to the appellant is labelled, “voluntary false confessions”, the characteristics of which are outlined in the appellant’s factum as follows:

[204]       Dr. Gudjonsson’s testing of the appellant largely confirmed Dr. Turrall’s findings, as demonstrated in the following passage from his report:

[205]      Dr. Gudjonsson commented in his report on several motivating factors that may have led the appellant to falsely confess, including his relationship with Neil Miller, the “emotional build up” from his arrest for the armed robbery of a taxi driver, his “severe emotional and self-esteem problems”, as well as his “antisocial orientation, impaired rational judgment, dislike of the police, and disregard for the consequences of his actions.” Of these factors, Dr. Gudjonsson identified the appellant’s low self-esteem and need for notoriety to enhance his self-esteem as the “single best explanation for the confession, if he truly made a false confession.”

[206]      In both his report and testimony, Dr. Gudjonsson focused on the accuracy of the appellant’s confessions when tested against the known facts surrounding the murder and the possibility that the appellant’s knowledge about the crime was the result of contamination from outside sources, such as the media, the police or his visits to the crime scene with the police. He also compared and contrasted features of the appellant’s case that in his view were analogous to cases in which false confessions have been given.

[207]      Dr. Gudjonsson concluded that the appellant’s confessions to Neil Miller and the Ottawa police are “inherently unreliable” and “unsafe to rely on” in the absence of “good independent corroboration”. In reaching this conclusion, Dr. Gudjonsson wrote as follows:

[208]      In July 2008, the Crown cross-examined Dr. Gudjonsson about the contents of his report before Sharpe J.A. of this court.[12] When counsel asked him about the belief expressed in his report that the appellant’s confession was probably false, Dr. Gudjonsson conceded that, in expressing his personal view about the truth or falsity of the confession, he went beyond the limits of scientific evaluation because there is no psychological test

[209]      In further cross-examination, Dr. Gudjonsson expressed the view that the appellant’s antisocial and borderline personality disorders were interrelated causal factors that might have led to a voluntary false confession:

[210]      In response to the report tendered by Dr. Gudjonsson, the Crown retained Dr. Michael Welner, a forensic psychiatrist practising in the United States. Dr. Welner was retained for the sole purpose of offering a methodological critique of Dr. Gudjonsson’s opinion.

[211]      In his report, Dr. Welner commented disparagingly on some of the methods used by Dr. Gudjonsson to assess false confessions. In particular, he challenged the assertion that, through the work of Dr. Gudjonsson and others, the study of false confessions has advanced to the point that it provides a scientific methodology for determining the reliability of a confession. Dr. Welner was cross-examined at length on his views and the appellant is highly critical of his evidence, alleging that Dr. Welner lacks “expertise in both psychology and scientific methodology”.

[212]      A great deal of time was spent in oral argument debating, in particular, the admissibility of Dr. Gudjonsson’s evidence as expert evidence and whether his evidence meets the criteria of relevance and necessity identified in R. v. Mohan, 1994 CanLII 80 (SCC), [1994] 2 S.C.R. 9. On the issue of relevance, the Crown maintains that much of Dr. Gudjonsson’s proposed evidence lacks scientific reliability. The Crown submits:

[213]      In support of the claim that Dr. Gudjonsson’s evidence lacks scientific reliability, the Crown relies heavily on his acknowledgement that his research merely provides a “conceptual framework for understanding” why people falsely confess; it does not enable anyone to state scientifically that a particular confession is reliable or unreliable.  The Crown also relies on Dr. Gudjonsson’s further acknowledgment that the field of giving opinions on the reliability or unreliability of confessions is not one “where you can get error rates.”

[214]      The Crown points out that in seeking to exclude Dr. Gudjonsson’s evidence, it is not, as the appellant claims in his factum, seeking to “exclude the insights of psychology into the disordered personality of those making false confessions”. Rather, the Crown contends that:

[215]      On the issue of necessity, the Crown maintains that much of Dr. Gudjonsson’s proposed evidence relates to matters that ordinary people can understand and form a correct judgment about without the assistance of an expert. Moreover, there is the ever-present danger, especially with someone like Dr. Gudjonsson, that members of the jury could be overwhelmed by his credentials and would not be able to objectively assess his evidence and thus the proposed expert evidence might usurp the role of the jury.

[216]      As for the appellant’s contention that Dr. Gudjonsson’s evidence is necessary to shed light on the fact that false confessions do occur and to rebut the commonly-held view that people would not confess to a serious crime they have not committed, the Crown maintains that the trial judge could alert the jury to this possibility in the instructions to the jury, thereby removing the possible need for an expert.

[217]      I have identified the Crown’s objections to the admission of Dr. Gudjonsson’s evidence in considerable detail because, at the very least, they show that the admissibility of expert evidence on false confessions is anything but obvious and should be approached with considerable caution. Of particular concern is whether the proposed evidence reaches the level of scientific reliability required by Mohan to warrant its reception.

[218]      That said, I want to be clear that, in cases such as this where the reliability of a confession is in issue, expert evidence regarding an accused’s personality traits that is relevant to and probative of the issue will be admissible. As the Crown points out, that type of evidence was properly led at the appellant’s trial. I turn to that evidence now, since in my view, the expert evidence called at the trial is dispositive of the question posed to this court by the Minister of Justice.  This is because the expert evidence called at trial shows that the proposed new evidence is not “fresh” evidence and therefore not admissible under Palmer.

[219]      At the appellant’s trial, Mr. Cogan adduced a good deal of evidence about the appellant’s personality. The evidence came from both the appellant’s lover and from two experts who conducted examinations of him.

[220]      Neil Miller testified that the appellant was someone “who lived in … a world of fantasy”.  According to Mr. Miller, he was someone who “made up a lot of things” and who said “he did a lot of things that he didn’t do”. Mr. Miller agreed that “there was very little that you could believe him about.”

[221]      Dr. Michael Girodo, a clinical psychologist retained by the appellant, testified that the appellant suffered from an antisocial personality disorder. According to Dr. Girodo, the appellant was emotionally and cognitively immature. He had very little capacity to form attachments, he engaged in relationships based on self-gratification, he manipulated others for his own gain and he did not appreciate concepts like morality or legality.

[222]      Dr. Girodo told the court that during his psychological testing of the appellant, the appellant’s scores on his MMPI test [the Minnesota Multiphasic Personality Inventory] indicated that he was trying to present himself in the worst possible light. Dr. Girodo concluded that the appellant lacked any sense of identity or self-image and he saw himself as an individual at the extreme edge of society. For the appellant, having a negative identity was better than having no identity.

[223]      The following extract from Dr. Girodo’s trial testimony bears repetition. It tracks key elements of the so-called “fresh evidence” that the appellant seeks to introduce through Dr. Gudjonsson and to a lesser extent Dr. Turrall:

[224]      While Dr. Girodo expressed the opinion in cross-examination that the appellant was capable of telling the truth, he cautioned in re-examination that one would have to look for supporting evidence before believing him.

[225]        Dr. Julio Arboleda, a forensic psychiatrist, also testified for the appellant at trial. Like Dr. Girodo, he expressed the opinion that the appellant suffered from an antisocial personality disorder. He explained that persons with this type of disorder lack a sense of responsibility, are impulsive, have difficulty forming personal relationships, exploit others for their own gain, and are unable to foresee the consequences of their acts.

[226]      According to Dr. Arboleda, the appellant confided that his reason for falsely confessing was that “he just wanted to be on the radio, make the news”. In Dr. Arboleda’s opinion, the appellant engaged in boastful behaviour to enhance his self-image.  He further testified that the appellant was theatrical, he liked to exaggerate, and he was prone to using attention-seeking devices, including dramatic suicide attempts. Dr. Arboleda dismissed the possibility that the appellant may have confessed to the murder out of feelings of remorse, stating:  “He does not seem to be the type of person who can experience this type of thing.”

[227]      As the following extracts from his testimony reveal, Dr. Arboleda considered the confessions made by the appellant to be quite unreliable and testified that he would not believe anything the appellant said, absent supporting facts:

[228]      Mr. Cogan asked Dr. Arboleda if he had performed any tests on the appellant to assist him in assessing the reliability of his confessions. He indicated that he had personally performed a sodium amytol or “truth serum” test on the appellant and that he had also asked for a polygraph test to be performed on him.  When asked for the results of the sodium amytol test, Dr. Arboleda testified:

[229]      Dr. Arboleda was then asked for the results of the polygraph test and responded as follows:

[230]      Much like the evidence provided by Dr. Girodo, these extracts from Dr. Arboleda's evidence mirror critical elements of the so-called “fresh evidence” that the appellant seeks to introduce through Dr. Gudjonsson and to a lesser extent, Dr. Turrall. Indeed, the emphasized portion of Dr. Arboleda’s final response goes further than the proposed testimony of Dr. Gudjonsson, who candidly acknowledged that he had no scientific basis for expressing an opinion on whether the appellant’s confession was false.  Moreover, through Dr. Arboleda, the jury came to learn that the appellant had voluntarily undergone a sodium amytol or “truth serum” test and a polygraph test, and that the results were consistent with him having lied to the police when he confessed.

[231]      I have engaged in this exercise to demonstrate that the jury who tried the appellant had before it the substance of Drs. Turrall’s and Gudjonsson’s evidence. Drs. Girodo and Arboleda identified the same key personality traits that the new experts identified: the appellant’s antisocial personality; his poor self-esteem; his inclination to seek a negative identity; his difficulty in forming personal relationships; his impulsiveness; and his tendency to exaggerate or lie in order to gain attention and make himself feel important.

[232]      To the extent that Drs. Gudjonsson and Turrall added the formal diagnosis of borderline personality to the appellant’s personality profile, that label adds nothing to what the jury already knew about him, namely, that he was a person with low self-esteem who needed to make himself feel important. Although Drs. Girodo and Arboleda did not use the “borderline” label, they took this aspect of the appellant’s personality into account in their assessment of him. In the circumstances, I view the addition of this label as inconsequential and reject Dr. Gudjonsson’s position that the borderline label captures a feature of the appellant’s profile that was not explained to the jury at his trial.

[233]      As for the appellant’s contention that what the jury lacked was not the substance of Dr. Gudjonsson’s evidence but the fact that such evidence now has “the force of science” behind it, I have already expressed my view that this assertion is open to debate. But even if Dr. Gudjonsson’s proposed testimony does have the force of science behind it, I am not at all persuaded that his evidence if believed could reasonably, when taken with the other evidence adduced at trial, be expected to have affected the result.  In other words, Dr. Gudjonsson’s evidence, as well as that of Dr. Turrall, does not satisfy the third prong of the cogency test in Palmer and is therefore not admissible as fresh evidence. In the words of Doherty J.A. in R. v. M. (P.S.) (1992), 1992 CanLII 2785 (ON CA), 77 C.C.C. (3d) 402, at pp. 411-12, the proposed evidence is not “‘fresh’ evidence in any relevant sense” but only “a third voice which speaks to the exact issues addressed” at trial.

[234]      Those issues were aptly summarized by Ritchie J. for the majority of the Supreme Court of Canada in the appellant’s appeal in Phillion, at pp. 22-23:

[235]      In the same decision, Spence J., speaking for himself and Laskin C.J., referred at p. 20 to the nature and effect of the psychiatric and psychological evidence adduced at trial:

[236]      In many ways, it could be said that Mr. Cogan was ahead of his time in the defence he presented on behalf of the appellant. He provided the jury with psychological evidence designed to cast doubt on the reliability of the appellant’s confession. In the end, the jury did not accept that evidence. But the evidence heard by the jury at trial captured the essence of the testimony that the appellant now seeks to introduce through Dr. Gudjonsson and Dr. Turrall as fresh evidence. It therefore fails the cogency test in Palmer and must be rejected.

[237]      For these reasons, I would answer the second question posed by the Minister in the negative.

[238]      My response to the first question posed by the Minister requires that I go on “to determine the case as if it were an appeal by Romeo Phillion.”  In the words of the terms of reference:

[239]      A determination that fresh evidence should be received on appeal because it could reasonably be expected to have affected the verdict means that the verdict can no longer stand. Accordingly, the appellant’s conviction must be quashed. What remains to be decided is the appropriate remedy.

[240]      Although the court in these circumstances has several options, only two of them need be considered in the present case. The appellant seeks to have this court enter a verdict of acquittal, failing which he requests us to order a new trial.

[241]      The Crown maintains that this is not a case in which an acquittal can or should be entered and submits that a new trial should be ordered. In the event that a new trial is ordered, the Crown acknowledges that there will not be a retrial on the merits. Some of the key witnesses in this case are now deceased and, for those who remain, their memories of the relevant events have understandably faded with the passage of time.

[242]      In these circumstances, if a new trial is ordered, the Attorney General will have the following three options: stay the charge, withdraw the charge, or have the appellant arraigned before the trial court and offer no evidence against him. If the Attorney General were to choose the last option, the appellant would have his acquittal. On either of the other two options, there would be no final verdict in this case. In closing submissions, the Crown advised that it cannot commit to any of these possible options until it has had the opportunity to consider the court’s reasons for judgment.

[243]      Returning to the two options available to this court – an acquittal or a new trial – I have no hesitation in concluding that this is not a case in which a verdict of acquittal should be substituted; rather it is one that calls for an order for a new trial.

[244]      In so concluding, I have considered the fresh evidence that has been admitted against the backdrop of the evidence that was adduced at the appellant’s trial and I am satisfied, on balance, that the evidence in its entirety could reasonably support a conviction.  Depending on how the fresh evidence were to unfold at a new trial, it would be open to a jury to reject the defence of alibi and conclude, essentially on the basis of the appellant’s confessions to Mr. Miller and the police, that he was the person who murdered Mr. Roy. Hence, substituting a verdict of acquittal on the basis that the fresh evidence is “clearly decisive” of innocence is not a tenable position:  see R. v. Stolar, 1988 CanLII 65 (SCC), [1988] 1 S.C.R. 480 and R. v. Baltovich (2004), 2004 CanLII 45031 (ON CA), 73 O.R. (3d) 481 (C.A.), at para. 162.

[245]      Nor, in my view, is this a case in which it can be said that, on the basis of the information now available, it is “clearly more probable than not” that the appellant would be acquitted at a hypothetical new trial:  see Truscott, para. 268.  Much would depend on how the fresh evidence were to unfold at a new trial and how the jury were to assess it. One possibility is that the jury could find the appellant to be factually innocent.  Alternatively, the jury could entertain a reasonable doubt about his guilt. Another tenable possibility is that the jury could reject the defence of alibi and find the appellant guilty, essentially on the basis of his confessions.

[246]      To be clear, just because the fresh evidence referred to in the first question of the terms of reference passes the cogency test in Palmer, this does not mean that it is “clearly more probable than not” that the appellant would be acquitted at a new trial. All that can be said is that, in the context of the entirety of the evidence admitted on appeal and heard at trial, the fresh evidence could reasonably be expected to have changed the result at trial. In other words, if the jury had had the benefit of the information contained in the April 12, 1968 report, that information, considered with the entirety of the evidence heard at trial, may well have left the jury in a state of reasonable doubt as to whether the appellant was the person who killed Mr. Roy. However, the compelling nature of his confessions – particularly the level of detail and accuracy found in them – prevents me from concluding that the admission of the fresh evidence would make it “clearly more probable than not” that the appellant would be acquitted at a new trial.

[247]      In the result, I would allow the appeal, quash the conviction and order a new trial.

[248]      As a final note, I wish to thank all counsel for the assistance they have provided to the court and to commend them for their diligence and professionalism throughout.

[249]      I have had the opportunity to review the reasons prepared by my colleague, Moldaver J.A. (Laskin J.A. concurring).  I agree entirely with his analysis of and answer to the second question posed by the Minister of Justice.  Regrettably, I cannot agree with his analysis of and answer to the first question.

[250]      My colleague concludes that the undisclosed evidence relating to Detective McCombie’s April 12, 1968 report is admissible, not on the basis that it rendered the trial process unfair, but rather on the basis of the test for admission of fresh evidence set out in R. v. Palmer, 1979 CanLII 8 (SCC), [1980] 1 S.C.R. 759.

[251]      While I agree with my colleague’s legal conclusion on the trial fairness issue, I do not regard the evidence on the issue whether Detective McCombie discredited the alibi as being so evenly balanced that it gives rise to “a stalemate”.  In my view, the evidence supports a finding that Detective McCombie initially believed that the appellant had a verified alibi, but that he later discredited it. 

[252]      I also respectfully disagree with my colleague’s conclusion that the evidence relating to Detective McCombie’s April 12, 1968 report meets the Palmer test for admission of fresh evidence on appeal. In my view, this evidence is not sufficiently probative that it could reasonably, when taken with the other evidence at trial, be expected to have affected the jury’s verdict.  Hence the fresh evidence is not admissible on appeal. It is this latter point of disagreement that leads me to answer the first reference question in the negative. 

[253]      Against the backdrop of my colleague’s comprehensive factual review of the proceedings, I can set out my reasons for these conclusions relatively briefly.

[254]      I begin my discussion of the trial fairness issue with two introductory observations, one relating to the prosecutor and the defence counsel involved in the 1972 trial and the other relating to a jurisprudential factor which, in my view, must anchor this court’s assessment of the conduct of these men in 1972.

[255]      First, in 1972 Mac Lindsay and Arthur Cogan were young lawyers with promising careers and good reputations.  For both of them, the Phillion trial was the first murder trial they conducted on their own.

[256]      Fortunately, they were both able to testify before this court in January 2008.  Both were impressive witnesses. They struck me as honest, fair and deeply committed to assisting in the resolution of a reference relating to a trial that unfolded 37 years ago.

[257]      My impression of both men is confirmed by what they said about each other. In his testimony, Mr. Lindsay described Mr. Cogan in this fashion:

[258]      Mr. Cogan described Lindsay in a similar vein. He agreed that Mr. Lindsay was a Crown counsel with “excellent credibility”.  According to Mr. Cogan, “I have known Mr. Lindsay for a long time and he’s had a great career and he’s a nice man and he was a fair Crown” and “he had a fine reputation and he’ll end his career with a fine reputation.”

[259]      These descriptions of both counsel lead me directly to the jurisprudential point I want to make.  In R. v. Truscott (2007), 2007 ONCA 575 (CanLII), 225 C.C.C. (3d) 321, this court declined to consider the submission that Truscott’s trial almost 50 years ago was unfair because the Crown did not disclose crucial information to the defence.  This court offered five reasons for declining to consider the issue.  Two of those reasons, at paras. 116 and 123-24, are as follows:

[260]      This court cannot decline to consider the disclosure/trial fairness issue in this case; it is at the heart of the first reference question.  However, in considering this issue, it is crucial that this court take heed of the warning in Truscott as it reviews a trial, and the conduct of two young counsel with good reputations, that occurred 37 years ago.

[261]      It is true that Mr. Lindsay and Mr. Cogan were both able to testify before us.  However, as pointed out by my colleague, their testimony involved very little real memory of the trial proceedings.  As in Truscott, “[t]hose who were available spoke, quite understandably, in generalities and surmises rather than based on actual recollections of relevant events.”  Thus, on the question of trial fairness, as in Truscott, “[t]he court is left to speculate about what people knew or did not know and what they did or did not do almost fifty years ago.”

[262]      In assessing whether the appellant has established that his trial was rendered unfair by the Crown’s failure to meet its disclosure obligation in relation to either a verified alibi or a previously verified but subsequently discredited alibi, my colleague reaches four conclusions:

[263]      I agree with conclusions (2), (3) and (4).  I agree with my colleague’s legal conclusion in (1), that the appellant has failed to meet his burden of establishing that Detective McCombie did not discredit the alibi. However, unlike my colleague, I do not regard the evidence on this issue as being so evenly balanced that it gives rise to a stalemate, thus requiring resort to the default position that the onus of proof rests with the appellant.  Rather, although I acknowledge that there is strong evidence both ways, three components of the evidence lead me to conclude that Detective McCombie did in fact discredit the alibi that he referred to in the April 12, 1968 report.

[264]      First, based on the entire record, including his testimony in this court in January 2008, I believe Detective McCombie’s claim that he discredited the alibi, but that some 40 years later, the supplementary investigation report he wrote describing his trip to Trenton and the confirmatory physical evidence he obtained at the service station are no longer available. 

[265]      There is nothing in the record relating to Detective McCombie’s manner of investigating the Roy murder to suggest that he is lying about this. It should be recalled that on August 10, 1967, the day after the murder, Detective McCombie showed Mrs. Roy ten photographs and Mrs. Roy selected the photograph of Donald Phillion, the appellant’s identical twin brother, as the one most resembling the assailant.  Five days later, on August 15, 1967, Mrs. Roy identified the appellant when she viewed a line-up at the police station, but later expressed some doubt about her identification. Detective McCombie did not charge the appellant, which might be regarded as a cautious decision.  Moreover, for many months, Detective McCombie diligently pursued the investigation wherever it led, including towards other suspects. In addition, in the April 12, 1968 report, Detective McCombie recorded a verified alibi for the appellant in clear and strong language.

[266]      Based on these and many other examples, it cannot be said that Detective McCombie showed any signs at all of being out to get Romeo Phillion for the murder of Leopold Roy. As expressed by my colleague in his reasons:  “Detective McCombie’s even-handed investigation undercuts the notion that he would have misled Mr. Lindsay and his fellow officers in order to bring about the prosecution of a man he knew to be innocent of the crime.”  My colleague also refers to an “impressive body of evidence attesting to Detective McCombie’s distinguished career as a police officer and his unblemished reputation for professionalism and integrity.”  I would attach more weight to this evidence than does my colleague.

[267]      Second, based on my assessment of the recent testimony of Mac Lindsay, who impressed me with his candour and sincerity, and Arthur Cogan, who testified about Mr. Lindsay’s professionalism and fine reputation, it defies credulity to contemplate that Mr. Lindsay would have allowed a murder prosecution to proceed knowing that the lead investigating officer had previously believed that the accused had an alibi and that this alibi had not been properly discredited.  This conduct would have been a terrible abuse of process and contrary to the professionalism and integrity that animated Mr. Lindsay’s entire career as a Crown prosecutor.  As expressed by my colleague: “I find it virtually inconceivable that Mr. Lindsay…would have participated in prosecuting the appellant for the Roy murder knowing that he was innocent of the crime.” Again, I would attach more weight to this evidence than does my colleague.

[268]      Third, there is critical documentary evidence supporting a finding that Detective McCombie was satisfied that the Trenton alibi had been fully discredited prior to the appellant’s 1972 confession.  In Detective McCombie’s April 2, 1971 investigation report describing his trip to Archambault Institute to question Donald Phillion about the murder, he wrote: 

[269]      My conclusion on this issue is thus that Detective McCombie did discredit the alibi that he had referred to in his April 12, 1968 report.  Since, as my colleague explains, the Crown was not obliged in 1972 to disclose evidence of a discredited alibi to the accused, it follows that the appellant received a fair trial.

[270]      This court in Truscott, at para. 92, described the Palmer test for the admission of fresh evidence as follows:

[271]      I agree with my colleague that the first and third of these criteria are met in this case.  The Crown concedes that the proposed fresh evidence is admissible under the operative rules of evidence and, as expressed by my colleague, it would be quite wrong to “cast aside the proposed fresh evidence on the basis of a lack of due diligence on Mr. Cogan’s part” in 1972.

[272]      That leaves the second criterion, the cogency of the fresh evidence.  In Truscott, at paras. 99-100, this court explained the nature and purpose of the cogency inquiry and the three questions that must be answered in relation to it:

[273]      As noted by my colleague, only the third question is in issue.  He concludes that the information contained in the April 12, 1968 report could reasonably be expected to have affected the verdict, and thus gives an affirmative answer to the cogency question.  With respect, I disagree.

[274]      As the emphasized passages from the above quotation from Truscott indicate, in order to answer the cogency question, the court must engage in a qualitative assessment which measures the probative potential of the fresh evidence considered together with the evidence heard at the trial.  I am prepared to assume that the radio and tow receipt that Detective McCombie claims to have retrieved from the Trenton service station and his supplementary investigation report describing his trip to Trenton were missing by 1972.  I am also prepared to assume that Mr. Cogan would have vigorously cross-examined Detective McCombie about his alleged discrediting of the Trenton alibi in an effort to raise a doubt in the minds of the jury as to whether the appellant in fact had the opportunity to commit the murder.  In response to this questioning, Detective McCombie’s testimony presumably would have been that, more than three years prior, he believed that the appellant had an alibi placing him in Trenton around the time of the murder.  He then would have gone on to explain that he subsequently travelled to the Trenton service station and discredited the alibi and that the documentary evidence capable of supporting this discrediting was now missing.

[275]      While the jury may well have been troubled by the missing documentation, it must be pointed out that missing evidence would not have been a new factor for the jury to consider in this case.  By the time Detective McCombie testified in 1972, his own notes as well as several items of physical evidence that police had gathered were also lost.  Mr. Cogan cross-examined Detective McCombie at trial about his missing notes and the missing evidence. In Mr. Cogan’s closing submissions, he asked the jury to resist a finding that the evidence supported the truth of the confession in a case where selected items of evidence – the photographic line-up shown to Mrs. Roy, fingernail scrapings, blood and hair samples removed from the deceased, as well as a hair sample voluntarily provided by the appellant – had been lost.  He also asked the jury to look very carefully at the lack of a proper police investigation in this case.  Presumably, the jury considered this aspect of the case along with all the other evidence.

[276]      In short, in 1972 the jury had to evaluate important aspects of Detective McCombie’s testimony without some of the confirmatory evidence he said had existed but could no longer produce. The documentary evidence capable of confirming the discrediting of the Trenton alibi would simply have been another item that Mr. Cogan could point to in support of his position advanced in closing argument that the jury should look critically at why so many items of evidence were lost in this case. I am not satisfied that it could reasonably be expected that the jury would have reached a different verdict because of the inclusion of this further piece of missing evidence.

[277]      In addition to concluding that Mr. Cogan could have raised a reasonable doubt in the minds of the jury through his cross-examination of Detective McCombie on the alleged discrediting of the Trenton alibi, my colleague is of the view that the new information in the April 12, 1968 report concerning the inconsistent statements of Mr. Barbe, and potentially Mrs. Barbe, when combined with Mr. Loyer’s willsay, could also have assisted in raising a reasonable doubt.  Again, I am unable to share this view. 

[278]      Mr. Barbe did not testify at trial and the jury did not rely on his evidence to convict. If Mr. Barbe had testified and been cross-examined on his alleged prior statements to Detective McCombie to the effect that he last saw the appellant on Tuesday, August 8, 1967 (the night before the murder), his sworn evidence at the preliminary inquiry indicates that he would have responded that it could not have been Tuesday, August 8 that he last saw the appellant.  He would further have explained that his wife always went to bingo on Wednesday nights and the night that the appellant spoke to him about selling his car was his wife’s bingo night.

[279]      As for Mrs. Barbe, the record does not support my colleague’s assertion that if the defence knew the contents of the April 12, 1968 report, Mrs. Barbe potentially may have accepted that her last sighting of the appellant was on Tuesday, August 8, 1967 and not the night of the murder. While it is true that Detective McCombie spoke to her and her husband in August 1967, there is no record of what he asked her or how she responded. She may or may not have been asked when she last saw the appellant and she may or may not have had a different recollection than her husband. 

[280]      What is clear from the record is that Mrs. Barbe always anchored her last sighting of the appellant by reference to her bingo night. In her recorded statements to the police in January 1972 and in her evidence at the preliminary inquiry and at the trial, she was unwavering in her assertion that she attended bingo on Wednesday nights.  The suggestion that Mrs. Barbe told Detective McCombie in August 1967 that she last saw the appellant on Tuesday, August 8, 1967 is pure speculation and in direct contrast with her unsworn statements and her sworn testimony. 

[281]      I also see no reason to think that Mr. Loyer’s willsay could have assisted the defence in prompting Mrs. Barbe to accept that she may have attended bingo on a Tuesday rather than a Wednesday night. If anything, his willsay provides considerable support for Mrs. Barbe’s recollection that she last saw the appellant on August 9, 1967.  Mr. Loyer’s willsay – which was disclosed to Mr. Cogan in the Crown Brief – confirmed that his parish held Wednesday night bingos, including one on Wednesday, August 9, 1967.  The indication in his willsay that another organization usually held bingos on Friday nights, but moved them to Tuesday nights in July and August, could not reasonably be expected to have changed the Barbes’ testimony that Mrs. Barbe played bingo on Wednesday nights.    

[282]      Another factor to consider in assessing the cogency of the April 12, 1968 report is whether Mr. Cogan would have conceded his client’s presence in Ottawa on the day of the murder if he had known about the once verified and later discredited alibi.  My colleague is unequivocal on this point: “he would not have made that concession had he been aware of the substance of the April 12, 1968 report.”

[283]      I make two observations about this conclusion.  First, if Mr. Cogan had been made aware of the substance of the April 12, 1968 report, he also would have been informed about Detective McCombie’s claim that he discredited the alibi recorded in this report. 

[284]      Second, my review of the trial transcript suggests that the reason Mr. Cogan conceded his client’s presence in Ottawa on the day of the murder was not because he did not know of the once confirmed and later discredited Trenton alibi.  Rather, it was because he wanted to ensure that the jury would not hear Mrs. Barbe’s testimony regarding a conversation the appellant had had with her husband on the night of the murder in which he said he wanted to sell his car and leave town. As Mr. Cogan put it in his submissions to the trial judge, he was concerned that if the jury heard this “very prejudicial” evidence, they might conclude that his client had a motive to break into an apartment building, or infer that he was the kind of man who would commit murder. Mr. Cogan’s knowledge of a once confirmed and later discredited alibi would not, in my view, have changed the deal he made with Mr. Lindsay in 1972 to avoid this aspect of Mrs. Barbe’s testimony.

[285]      So far, I have attempted to suggest that the cogency of the information relating to the April 12, 1968 report is less compelling than my colleague asserts based on the factors he considers.

[286]      However, in my opinion, the significance of these factors pales in comparison to one factor that overwhelms the others in the assessment of the cogency criterion.  It is the tidal wave on the horizon.  It is a factor that sweeps all the others aside, yet is mentioned only briefly in my colleague’s cogency analysis.  This factor is the appellant’s confession.

[287]      Palmer and Truscott instruct that this court must consider the cogency of any fresh evidence in the context of the entirety of the evidence heard at the trial.  A full review of the evidence in this case must include the singular fact that the appellant confessed to the murder of Leopold Roy. 

[288]      What can be said about his confession to the police? 

[289]      The confession was spontaneous, voluntary, and took place 4 ½ years after the murder.

[290]      In terms of content, the confession was detailed and highly accurate. The appellant’s oral and signed confession to the police contains 24 factual assertions regarding the murder and his conduct before and after.

[291]      One assertion cannot be verified – the appellant said that after the murder he went to an Ottawa bar and had a few beers.

[292]      Three assertions in his confession appear to be false. There was no evidence that the perpetrator had broken into an apartment at 275 Friel Street, or that he had stolen a knife from that apartment.  (However, Mrs. Roy testified at trial that when she first saw the intruder, he was standing with his back to her, outside the door to Apartment 14 and, while keeping his back to her, he was moving his arms and hands “[a]s if he was putting something there.” She also testified that the inhabitant of Apartment 14 was an elderly lady who was in the hospital at the time of the murder.)  Also false, or at least inaccurate, is the appellant’s assertion that the man he stabbed “started coming down at me”, when in fact Mr. Roy was coming up the stairs when he encountered the intruder.

[293]      Every other assertion made by the appellant – all 20 of them – is either completely true or probably true given supporting information available from the record.  I set out these assertions in the order the appellant made them in his oral confession to Detective Huneault (the first three bullet points) and then in his signed confession (the remaining bullet points):

[294]      The appellant’s response to this remarkably long, specific and accurate list of assertions is two-fold: first, that he was always obsessed with the Roy murder, having been a suspect early on in the investigation; and second, that all of his accurate assertions in the confession had been reported in the local press and thus were known to him through outside sources.  Hence his confession was false.

[295]      The only evidence of the appellant’s obsession with the case is that he came to the police station almost four years after the murder and told Detective McCombie that his twin brother, Donald, who was already serving a life sentence for a different murder, wanted to confess to the Roy murder.  Detective McCombie went to Archambault Institute in Montreal and Donald denied his brother’s claim. This exceptional event strikes me as evidence of lying and disloyalty, but not of obsession. 

[296]      As for the appellant’s submission that he learned the details of the murder through a newspaper article, which appeared in the Ottawa Journal on the day after the murder (the appellant had left Ottawa by then) and which included certain facts about the murder, I might be impressed by the appellant’s submission if the confession had been made 4 ½ hours or 4 ½ days or even 4 ½ weeks after the murder. However, the appellant confessed some 4 ½ years later. In this context, it strains credulity to think that the number of accurate assertions (20), the level of detail provided in them (for example, that Mr. Roy had grabbed him), and even the mundaneness of the assertions in the confession (for example, he had seen the victim cutting grass) flowed from him having read, and remembered for 4 ½ years, details found in a newspaper article. Moreover, there is simply no evidence in the record to suggest that the police or anyone else told the appellant details about the murder at any time.

[297]      As noted above, my colleague briefly considers the appellant’s confession.

[298]      My colleague acknowledges that the confession contains “a number of accurate details”.  In fact, of the 24 factual assertions in the confession, 20 are accurate.

[299]      Regarding the accurate details in the confession, my colleague goes on to state that “many of them were reported by the press and thus they were in the public domain.”  However, the newspaper account of the murder that the appellant relied on in argument was an article in the Ottawa Journal that appeared on the day after the murder.  The appellant was not in Ottawa that day.  Moreover, his confession occurred 4 ½ years after the murder, which is a long period of time for a person to remember numerous and highly specific details about a murder he since claims he did not commit.

[300]      My colleague points to the inaccurate facts in the confession about a break-in and theft of a knife from an apartment at 275 Friel Street and states that “standing alone, that feature of the confession could leave a jury wary about the veracity and reliability of the confession as a whole.”  With respect, I disagree.  At the trial in 1972, the jury had before it all of the evidence about the confession – the circumstances leading up to it, its contents, Neil Miller’s confirmation that the appellant had also confessed to him, and the expert evidence about the falsity of his confession.  The jury obviously found that the confession was true.  In doing so, they must have considered and attached little weight to the assertions relating to the break-in and theft, which they knew were inaccurate.  Standing alone, that feature of the confession did not lead the jury to reject the veracity and reliability of the confession as a whole.  Nothing in the record has changed on this issue.

[301]      In conclusion, I recognize that some people confess to crimes they did not commit.  However, it is also true that many people, for moral and/or legal reasons, confess to crimes they did commit.

[302]      In a voluntary confession to the police in 1972, Romeo Phillion admitted to murdering Leopold Roy in August 1967.  The jury had before it his confession and the circumstances giving rise to it.  The jury heard Mr. Miller’s evidence confirming the confession.  Finally, the jury heard the expert evidence called by the defence to the effect that the confession was false and should not be believed.  The jury considered all this evidence and found the appellant guilty of murder. 

[303]      Thirty-seven years later, the appellant’s confession remains, in my view, as compelling today as it was to the jury in 1972.  The proposed fresh evidence relating to a once verified and later discredited alibi, and relating to the statements of the Barbes and Mr. Loyer, does not change this picture.  This evidence could not reasonably be expected to have affected the result.  Accordingly, I would not admit the fresh evidence relating to it.

[304]      I would answer Question 1 – No.  For the reasons of my colleague, I would answer Question 2 – No.