Court of Appeal for Ontario

Date: June 7, 2018

Docket: C63537

Judges: Laskin, Sharpe and Fairburn JJ.A.

In the Matter of a Reference

Pursuant to sections 696.3(2) and 696.3(3)(a)(ii) of the Criminal Code, R.S.C. 1985, c. C-46 by the Minister of Justice to the Court of Appeal for Ontario concerning the criminal convictions of Mr. Eric Biddle rendered on December 21, 1987, in Toronto, Ontario.

Between

Her Majesty the Queen Respondent

and

Eric Biddle Applicant/Appellant

Counsel:

• Leo Adler, for the applicant/appellant
• Jamie Klukach, for the respondent

Heard: January 26, 2018

On a reference from convictions entered on December 21, 1987 by Justice D.C.J. Sheard of the District Court of Ontario, judicial district of York, in the Municipality of Metropolitan Toronto, sitting with a jury.

Fairburn J.A.:

A. OVERVIEW

1On July 11, 1986, P.B. was in her apartment bathroom, splashing water on her face, when she looked up from the sink and saw a man standing in the doorway. She attempted to fight him off, but he beat and strangled her into unconsciousness. P.B. later awoke in a pool of blood on her bedroom floor. Although her attacker was still in the apartment, she managed to escape.

2In December 1986, Eric Biddle was charged in P.B.'s assault. The backbone of the prosecution's case was P.B.'s identification of Biddle as her attacker. Although the identification evidence suffered from indisputable frailties, including a procedurally flawed courtroom identification procedure, Biddle was convicted of: break and enter a dwelling and cause bodily harm; assault cause bodily harm; and overcoming resistance by strangulation. In dismissing his appeal, this court concluded that the evidence was not "so faulty" as to render it "valueless": R. v. Biddle (1993), 65 O.A.C. 263, at para. 44, leave to appeal dismissed (1994), 69 O.A.C. 47 (S.C.C.). The court placed particular emphasis on the fact that Biddle had a full opportunity to cross-examine P.B. on the problems with her identification evidence and to comment on its frailty during closing submissions.

3What the trial Crown, defence and jury did not know at the time of trial, nor this court on appeal, was that P.B. and G.M., a police officer who was involved in the flawed courtroom identification procedure, were in a personal relationship. The existence of the relationship by the time of trial is not in dispute on this reference, although when it started is.

4On February 22, 2017, the Minister of Justice and Attorney General of Canada, The Honourable Jody Wilson-Raybould, referred the matter to this court under s. 696.3(2) of the Criminal Code, requesting that the court consider whether the new information "concerning the non-disclosure of the relationship between … [P.B.] and … [G.M.]" should be admissible as fresh evidence. If the court concludes that the information is admissible, then Minister Wilson-Raybould asks the court, pursuant to s. 696.3(3)(a)(ii), to determine the case as if it were an appeal brought by Biddle.

5Using the test set out in R. v. Dixon, [1998] 1 S.C.R. 244, I conclude that fairness demanded that the fact of the relationship be disclosed and that, had it been disclosed, there is a reasonable possibility that it would have tipped the balance in this extraordinarily weak Crown case. The information regarding the relationship should be admitted as fresh evidence, the convictions set aside and a new trial ordered.

B. BACKGROUND

(1) Attack on P.B.

6P.B. was attacked in her bathroom. Her attacker had women's underwear on his head. He stood there for about four to five seconds before placing a bra around her neck. She struggled with him and managed to strike him in the face two to three times before losing consciousness.

7When she awoke on her bedroom floor, she got up and proceeded to the kitchen area, where she saw her assailant again. He was blocking her way to the outside door. P.B. "charged" him, pushed him out of the way and ran into the parking lot. P.B. acknowledged that the injuries to her face, suffered as a result of the beating, interfered with her ability to see her assailant very well as she escaped from the apartment. From the beginning to end, she agreed that she may have seen her attacker for no more than a total of eight to ten seconds.

(2) Other Attacks and Arrests

8Two other women were attacked in the months following the assault on P.B., one on September 25 and the other on October 28, 1986. Although these later attacks were said to be similar to the attack on P.B., neither involved a home entry.

9Biddle was arrested in the vicinity of the October attack. He was charged and detained in custody on that matter. Over a month later, he was also charged in relation to the attack on P.B. and the September attacks.

10The grounds for arrest on the attack on P.B. only developed after the police took P.B. to a courtroom where Biddle was making an in-custody appearance on the October attack. She knew that she was being taken there to see if she could identify her attacker. G.M. sat with P.B. in the courtroom. A number of prisoners, including Biddle, were then brought into the courtroom. She identified Biddle as her attacker.

(3) Legal Proceedings

11The Crown attempted to prosecute all of the offences together. On the eve of trial, the home invasion attack on P.B. was severed. The September and October 1986 offences were tried in November 1987 and convictions were entered. The trial involving P.B. proceeded in December 1987 and Biddle was convicted.

12Although Biddle testified at the first trial, denying having committed the offences, he did not testify at the second. There is evidence before this court suggesting that his decision to remain silent was rooted in counsel's advice that he not testify given that he would likely be confronted with the convictions that had been entered a month previously. Since R. v. Corbett, [1988] 1 S.C.R. 670, which modified the law on cross-examining accused on prior convictions, had not yet been decided, it is plausible that Biddle may have been cross-examined on those convictions. Relying on a letter Biddle wrote to the Minister, Mr. Adler maintains that had Biddle testified, he would have denied committing the offence. However, Biddle has filed no affidavit to this effect.

13Biddle was sentenced for all three attacks at the same time, receiving a total sentence of 12 years: seven years for the attack on P.B., seven years concurrent on the September attack and five years consecutive on the October attack.

14Biddle then started the appeal process. As for the September and October attacks, Biddle was ultimately successful and a new trial was ordered: R. v. Biddle (1993), 84 C.C.C. (3d) 430 (Ont. C.A.) ("Biddle 1"), rev'd , [1995] 1 S.C.R. 761 ("Biddle SCC"). As noted before, his appeal in relation to the attack on P.B. was dismissed. I will refer to that judgment as "Biddle 2".

15Biddle was granted bail pending his new trial on the September and October attacks: R. v. Biddle (1995), 82 O.A.C. 107. By the time he received bail, however, he had already been in custody for over eight and a half years and had completed his sentence in relation to the attack on P.B. Ultimately, the trial Crown withdrew the charges relating to the September and October attacks. There is some suggestion that the Crown did so on the basis that the complainant in the October attack did not wish to testify again and that without the similar fact evidence of the October attack the Crown was left without a reasonable prospect of conviction on the September attack.

16Biddle continued to maintain his innocence in relation to the July attack on P.B. Having exhausted his appeal routes, he brought an application for ministerial review in late 1995. The evidence gathering process commenced with interviews in 1999. It is in the context of those interviews that the information forming the subject of the fresh evidence in this reference – the information about P.B. and Officer G.M.'s relationship – was discovered.

C. ANALYSIS

(1) The Test for the Admissibility of the Fresh Evidence

17There are two ways in which the proposed fresh evidence may become admissible on this reference: (1) on the basis of non-disclosure giving rise to a breach of the right to make full answer and defence, governed by the test first set out in Dixon (the "Dixon test"); or (2) on the basis that the cogency of the evidence is such that it warrants admission and the interests of justice require that it be received, governed by the test first set out in R. v. Palmer, [1980] 1 S.C.R. 759 (the "Palmer test"). Biddle maintains that the fresh evidence is admissible under both the Dixon and the Palmer tests. The respondent maintains that neither test affords admission.

18There are two components to the Dixon test. First, the court asks whether there has been a breach of the Crown's duty to disclose. A time-sensitive approach governs this inquiry as it would be unfair to consider allegations of failed disclosure through the lens of current day rules and practices. As this court cautioned in R. v. Truscott, 2007 ONCA 575, 225 C.C.C. (3d) 321, at para. 122, applying present-day disclosure standards to past actions"is akin to using present-day medical standards to decide whether a diagnostic protocol followed in 1959 amounted to malpractice". See also: R. v. Phillion, 2009 ONCA 202, 241 C.C.C. (3d) 193, at para. 150.

19If the court concludes that, at the relevant time, the Crown failed in its disclosure obligations, then Dixon requires that the court go on to consider whether there exists a "reasonable possibility" that the non-disclosure: (a) impacted the outcome of the trial; or (b) impacted the overall fairness of the trial process: R. v. Taillefer, 2003 SCC 70, [2003] 3 S.C.R. 307, at paras. 71, 78; see also Dixon, at paras. 34-35. Although a reasonable possibility must be more than "entirely speculative" in nature (Dixon, at para. 34), the "mere existence of such a possibility constitutes an infringement of the right to make full answer and defence": Taillefer, at para. 78.

20For the reasons that follow, I find that the Dixon test governs and that a new trial is required. This means that there is no need to address the Palmer argument, nor Biddle's additional contention that non-disclosure amounted to an abuse of process.

21This was a classic eyewitness identification case that hung by a thread. It is against the backdrop of this exceptionally weak case that I will go on to discuss the nature of the fresh evidence and how it intersects with the case. I will then consider the fresh evidence against the Dixon test, ultimately arriving at my conclusions that: (a) there was failed disclosure; and (b) there is a reasonable possibility that had the disclosure been made, the verdicts would have been different. Accordingly, there is no need to go on and consider the second prong of the Dixon test involving the overall fairness of the trial process.

(2) The Fragile Nature of the Case at Trial

22There is no doubt that P.B. was brutally assaulted in her apartment, in what must have been a terrifying and life-altering experience.

23The prosecution case turned on P.B.'s identification of Biddle as her attacker. Without that identification, there was no case. As fairly acknowledged by the respondent"[P.B.'s] identification of the appellant was the centrepiece of the Crown's case against [Biddle]".

i. The non-identification evidence was weak

24There were only four pieces of evidence that existed beyond the realm of eyewitness identification evidence at trial. All combined, they amounted to little.

25First, the jury heard that Biddle lived in an apartment building close to P.B.'s building in the City of Toronto. P.B.'s neighbour saw a man of "average height", who was "stocky" and "thick" running away from P.B.'s apartment, in the same direction as Biddle's apartment building. The defence maintained that the path of travel was also consistent with running towards the parking lot for P.B.'s building.

26Second, blood was found on the underwear that the assailant placed on his head. If the blood was from a single source, it was neither Biddle's nor P.B.'s blood. If it was a mixture of blood, it could have come from P.B. and another person sharing the same blood type as Biddle. The appellant's blood type is shared by 40 percent of the population. In a rather understated way, the trial judge pointed out in his charge that the blood evidence was "not in any sense conclusive evidence".

27Third, one of Biddle's co-workers saw Biddle wearing "dark glasses" three days after the attack, which occurred in July. This evidence was led to show that Biddle might have been hiding a facial injury that he incurred during the assault. The witness was asked in cross-examination how long he was with Biddle that day and he answered: "Just a pass … He was coming in and I was heading in the other direction."

28Finally, P.B. described her assailant as being clean-shaven. By the end of October, Biddle had a beard. The trial Crown maintained that this was evidence of Biddle's attempt to disguise his appearance. Biddle's neighbour gave evidence that Biddle was clean-shaven until at least early October. Again, in a rather understated way, the trial judge cautioned the jury that "if one were considering disguising one's appearance after having committed a criminal act in July … October would be rather a late date to begin the process of a disguise". In Biddle 2, this court commented, at para. 22, that the beard "evidence was very weak and one would expect that a jury would place very little weight on this submission".

29In summary, the non-identification evidence included the facts that Biddle: (i) lived in the same area of Toronto as P.B.; (ii) was part of 40 percent of the total population that might have contributed to a blood sample, if it was a mixed blood sample; (iii) was wearing dark glasses in mid-July; and (iv) grew a beard a few months after the assault. This is not the stuff that proof beyond a reasonable doubt is made of. Indeed, it offered little to the trial.

30Undoubtedly, this case turned on eyewitness identification evidence, inherently dangerous evidence.

ii. Eyewitness identification cases contain inherent dangers

31It is widely acknowledged that there are intrinsic frailties within eyewitness identification evidence. As Doherty J.A. recognized a decade after the convictions in this case, there is a "well recognized potential for injustice" in eyewitness identification cases: R. v. Tat (1997), 117 C.C.C. (3d) 481 (Ont. C.A.), at paras. 99-100. The difficulty is that the trier of fact is susceptible to conflating the concepts of credibility and reliability. Perfectly honest and well-meaning identification witnesses can provide convincing evidence that is unreliable and often impermeable to attack on cross-examination: R. v. Goran, 2008 ONCA 195, 234 O.A.C. 283, at para. 19; R. v. Hibbert, 2002 SCC 39, [2002] 2 S.C.R. 445, at para. 52.

32The dangers linked to eyewitness identification evidence increase exponentially when it comes to in-dock identifications. In these circumstances, the identification witness has a person presented to them, a person who the police obviously believe committed the offence. Given the suggestive nature of in-dock identifications, they are renowned for being of little to no value as reliable identification evidence: R. v. Bailey, 2016 ONCA 516, 339 C.C.C. (3d) 463, at para. 48; Hibbert, at paras. 47-53; R. v. Jack, 2013 ONCA 80, 294 C.C.C. (3d) 163, at paras. 17, 30-31.

33P.B. was a confident witness. I do not doubt her genuine belief that Biddle was the man who so viciously attacked her. The central question for the trier of fact, though, was not the sincerity of her evidence. It was the reliability of her identification of Biddle as her attacker.

34I will now explain why the reliability of that evidence was so troubling. In doing so, I should not be taken as questioning this court's judgment in Biddle 2. Indeed, this court recognized in 1993 that the identification evidence was "faulty". Read in context, though, the judgment conveys that the convictions were upheld because the court was satisfied, despite the flawed evidence, that the jury had the tools necessary to decide the case fairly. What was unknown at that time was that the defence was missing an important tool that could have been used to test the already fragile eyewitness identification evidence.

iii. The eyewitness identification evidence was riddled with problems

The hair evidence

35P.B. told the first officer on scene that her attacker had "shoulder" length hair, but the jury heard evidence that Biddle did not have shoulder-length hair in July 1986. Nor did he have shoulder-length hair at trial. He was also bald on top.

36Although P.B. could not remember telling the officer that her attacker had shoulder-length hair, she did not disavow the comment at trial. Instead, relying on her father's "military" approach, she explained that, to her, anything more than a "brush cut" was shoulder-length hair.

37She also explained that her attacker's hair may have looked shoulder-length because the underwear he had placed on his head pressed his hair down. In Biddle 2, this court referred to P.B.'s reference to her attacker's shoulder-length hair as "the most important inconsistency" between what Biddle actually looked like and P.B.'s description of her attacker: at para. 32.

Exclusion from photo line-up

38On November 5, 1986, about four months after the offence, P.B. was shown a photo line-up containing Biddle's picture. The officer who conducted the line-up testified at trial that P.B. identified four photos as containing images of men who looked similar to her attacker. Biddle's photo was in that group. Ultimately, though, she excluded Biddle's picture on the basis that her attacker had "more hair" than the man in the photo.

The courtroom identification procedure: "not the fairest line-up"

39On December 3, 1986, about a month after excluding Biddle from the photo line-up, P.B. was taken to a courthouse where Biddle was scheduled to make an in-custody appearance on the charges arising from the October attack. P.B. sat with G.M. in the courtroom. The complainant from the September attack was taken to the courthouse at the same time. Although P.B. was told that she did not have to identify anyone, she acknowledged at trial that she knew that the police were taking her there to see if she could identify her attacker.

40At the preliminary inquiry she said that she was told to look in the prisoner's dock. At trial she said that she was simply asked to look around the courtroom. Once P.B. and G.M. were seated, a number of prisoners, including Biddle, were brought into the courtroom. Six of the prisoners were male. Although G.M. and P.B. testified that they could not recall what any of the other prisoners looked like, another police officer, sitting in the courtroom with the complainant from the September attack, acknowledged that none of the other prisoners looked even vaguely like Biddle.

41A copy of the pictures of the other prisoners who were in the courtroom at the same time as Biddle demonstrate that they differed in age, ethnicity, size, hair length and colour. None of the men look anything like Biddle. It appears that the jury did not see these photos.

42Importantly, one of the very first things that P.B. said after she was attacked was that her assailant wore glasses. Biddle was wearing glasses in the courtroom that day, yet the pictures of the other inmates with him suggest that none of them wore glasses. Accordingly, it appears that he was the only inmate present in the courtroom who presented with this critical identifying feature.

43To call this a flawed identification procedure would seriously undershoot the mark in terms of the powerful dangers it created for misidentification. As another police officer present during the courtroom identification fairly acknowledged at trial, it was "not the fairest line-up" and he would not do it again.

44In fact, this was not a line-up at all, as there was no attempt to ensure that the people presented looked somewhat alike. P.B. knew that the police believed that her attacker would be in the courtroom. Biddle was the only one wearing glasses, the first identifying feature that P.B. gave of her attacker. He was in custody relating to an attack on another woman. The danger of a mistaken identification was obvious.

45P.B. gave convincing identification evidence at trial. She described how she identified Biddle in court on December 3, 1986 and the strong emotional impact it had on her. When the trial Crown was interviewed as part of the ministerial review, over ten years after trial, he still spoke expressively of P.B. being the "strongest of the witnesses" and his view of her identification of Biddle as being "visceral". G.M. also testified before the jury, attesting to P.B.'s visceral reaction to seeing Biddle.

In-dock Identification at trial

46P.B. also gave convincing and unwavering in-dock identification evidence at trial, saying that Biddle's face is the face that she saw in her "nightmares, time and time again".

The evolving description of her attacker

47P.B. gave descriptive details about her attacker. Those details morphed between the time that she originally spoke to the police and her trial testimony. For instance, prior to trial, she said her attacker was between 25 to 30 years of age and later said he was 27. At trial, she said that he was between 25 to 40 years of age (consistent with Biddle who was 37 years old at the time of trial). Prior to trial, she also said that her attacker had a medium to heavy build. At trial, she said he was slender and athletic (consistent with Biddle).

48P.B. was also able to provide details about her attacker at trial that she had never mentioned before. For instance, although she had not mentioned it to the police before, by the time of trial, P.B. stated that her attacker wore wide-rimmed glasses with clear lenses"not shaded" (consistent with Biddle's glasses); and his hair was sandy blonde (consistent with Biddle's hair).

iv. Conclusion

49I agree with this court's assessment in Biddle 2 that the eyewitness identification evidence was faulty. Indeed, having regard to the intervening 25 years between Biddle 2 and today, and the experience with eyewitness identification cases having gone wrong, one might reasonably choose an even stronger word to characterize that evidence today. Despite its frailties, though, this court concluded that the verdicts were safe. Read in context, this conclusion was informed by: (a) the full cross-examination that had been conducted; and (b) the fact that the trial judge had strongly cautioned the jury about the dangers of the eyewitness identification evidence.

50Unfortunately, at the time that this conclusion was reached, no one but the police and P.B. knew about the relationship between P.B. and G.M. I will now discuss that fresh evidence, followed by an application of that evidence to the Dixon test. I find that whether the relationship started before or after the first in-dock courtroom identification, in accordance with the Dixon test: (a) the relationship should have been disclosed; and (b) there is a reasonable possibility that, had it been disclosed, the verdicts may well have been different.

(3) The Nature of the Fresh Evidence

51The Criminal Conviction Review Group ("CCRG") commenced an investigation after Biddle applied for ministerial review under then s. 690 of the Criminal Code. A lawyer at the CCRG first interviewed G.M. on April 29, 1999. The lawyer made handwritten notes about what G.M. told him at that meeting, including that G.M. and P.B. had been in a relationship.

52The handwritten notes from the April 29 meeting contain the following: "Married [P.B.] but was at arm's length during (now divorced) 3 months after the assault but there was no info exchanged." If the relationship started "3 months after the assault", that would place the start of the relationship around mid-October, 1986, before the centrally important December 3, 1986 courtroom identification involving P.B. and G.M.

53In 2017, the CCRG lawyer who originally interviewed G.M. was asked to provide a typed version of his handwritten notes for purposes of this reference. In doing so, he commented that the notes were not created simultaneously with the interview. Between the interview and writing his notes, he recalls attending the Art Gallery of Ontario and having drinks with a friend. He was also recovering from surgery at the time and may have been taking medication. He concludes: "[a]ll of which is to say that my notes of the timing of the relationship could have been faulty."

54The lawyer's current uncertainty as to the veracity of his handwritten notes from 1999 should be considered against the fact that he twice committed his then understanding of what G.M. had told him to two typed documents. First, in a typed memo to file, dated April 29/30, 1999, the lawyer wrote:

At the beginning of the interview, [G.M.] disclosed that he started to date [P.B.] about three monthes [sic] after the assault. He married her and subsequently they were divorced. He told me that he was well aware how their relationship might jeopardise [sic] the investigation and was very careful not impart [sic] any information about the case to her. When they first went out she suspected him of trying to dig out a motive for the crime; eg. that the attacker had been an old boyfriend.

55Then, in a draft "Investigation Brief", prepared in 1999, the lawyer commented:

At the beginning of the interview, Officer [G.M.] candidly disclosed to Departmental counsel that he started to date [P.B.] about three months after the assault. He married her and subsequently they were divorced. He told me that he was well aware how their relationship might jeopardize the investigation and was very careful not to impart any information about the case to her.

56While being examined by another CCRG counsel in 2004, G.M. denied that he told the original CCRG lawyer that the relationship started three months after the offence. Alternatively, G.M. maintained that, if he told the first lawyer that, he must have been mistaken at the time. G.M. was adamant that the relationship started in the late spring or into the summer of 1987, after the committal to trial and after he says he was no longer involved in the case.

57P.B. was also interviewed by CCRG counsel in 2004. She too was adamant that the relationship did not commence until the spring or summer of 1987. If this timing is correct, it would place the start of the relationship about six months after the flawed courthouse identification procedure.

58I would make the following two observations about the timing issue. First, it is one thing for someone to make a mistake in a scratchy, handwritten note after an interview. It is quite another to commit one's understanding of what one has been told to more formal, typed documents, prepared at a later date. Although the original lawyer to interview G.M. has some present concern about whether his handwritten notes are accurate, this does not answer why, at a time when his memory would have been much fresher, he would have typed in a memo to file and a draft investigative report what he believed he had been told by G.M.

59Second, in the memo to file, it seems clear that G.M. told the interviewer that when he and P.B. "first went out [P.B.] suspected him of trying to dig out a motive for the crime; eg. that the attacker had been an old boyfriend". Of course, on P.B. and G.M.'s version, their relationship started long after P.B. had identified Biddle as her attacker – indeed, on their version, Biddle had already been committed to stand trial by the time that P.B. and G.M. started to date. There would quite simply be no reason for G.M. to be trying to figure out a "motive" of an "old boyfriend" at that point in time. Accordingly, one is left wondering why G.M. would say that P.B. "suspected him of trying to dig out a motive for the crime" when they "first went out".

60Although the police were concerned early on in the investigation about whether P.B. was holding back information about her attacker, that concern was being actively addressed in the late summer and into September of 1986. At that time, the officer-in-charge of the investigation asked G.M. to meet with P.B. to try and figure out if there was anything to that theory. This was before Biddle had even been identified as a suspect in the attack. Once Biddle was identified, the police were no longer concerned with whether P.B. was holding back information that could assist in the investigation.

61Although there are questions about when the relationship between P.B. and G.M. started, the timing issue is not one that needs to be resolved for purposes of this reference. While the relevance of the fresh evidence increases if G.M. was with P.B. as both a police officer and her personal partner when she identified Biddle for the first time in court in December 1986, the relevance of the fresh evidence is not strictly limited to that identification procedure. There is no dispute that, by the time of trial, they had been involved in a serious relationship, at least for six months. As discussed previously, the full slate of P.B.'s identification evidence – including the evolution of the details about her attacker – could have been tested against her personal relationship with a police officer who had been an investigator on the case and who had access to the full investigative file.

62Accordingly, the question for this court is not when the relationship started, but whether the existence of the relationship was such that disclosure should have been made so that the defence could explore whether that relationship had any impact on an already fragile identification case.

63As I will now discuss, simple principles of fairness required disclosure.

(4) Applying Dixon

i. Disclosure obligations in 1986 and 1987

64The respondent urges this court to consider the Crown's disclosure obligations as they existed in 1986 and 1987, about five years before R. v. Stinchcombe, [1991] 3 S.C.R. 326 was released. Having regard to the disclosure rules, customs and practices at the time, the respondent maintains that, even if the trial Crown had known the relationship information, he would not have been required to disclose it. That submission leans heavily on the following factual assertions:

1. the relationship information did not arise from the police investigation;
2. the relationship commenced only after the courthouse identification; and
3. P.B. and G.M. were adamant that they did not discuss the case when their personal relationship began and there is no evidence to contradict that assertion.

65I disagree. At a minimum, the undisclosed information could have been used to challenge the reliability of the eyewitness identification evidence. Even prior to Stinchcombe, basic principles of fairness required that this information be disclosed.

66I agree with the respondent that Stinchcombe was an important development in the law and that, since Stinchcombe, we have developed a firmer idea about what must be disclosed: subject to some limited exceptions, anything not clearly irrelevant.

67As the respondent fairly recognizes, although Stinchcombe elevated the Crown's disclosure obligations to a constitutional right, it did not create those obligations. Even before Stinchcombe, the Crown had a duty to disclose "all relevant evidence to the defence", regardless of whether that evidence would assist the accused or not: Taillefer, at para. 64.

68The pre-Stinchcombe common law duty to disclose "all material evidence whether favourable to the accused or not" was recognized in R. v. C.(M.H.), [1991] 1 S.C.R. 763, at p. 337, a case released before Stinchcombe and one that reviewed the pre-existing law of disclosure dating back many decades. Although McLachlin J. (as she then was) found it unnecessary to "establish the exact ambit of the Crown's duty of disclosure", she emphasized that the Crown's duty to disclose was rooted in principles of fairness.

69In McTaggart v. Ontario, 2000 CarswellOnt 4808 (S.C.), the court heard evidence, both from the Crown and defence, that the prevailing disclosure obligations in 1986 were tied to the Crown's obligation to be fair. The classic quote from R. v. Boucher, [1955] S.C.R. 16, at p. 24, was relied upon:

The role of prosecutor excludes any notion of winning or losing; his function is a matter of public duty than which in civil life there can be none charged with greater personal responsibility. It is to be efficiently performed with an ingrained sense of the dignity, the seriousness and the justness of judicial proceedings.

70Although decided some years after this case, the words of L'Heureux-Dubé J. in R. v. O'Connor, [1995] 4 S.C.R. 411, at para. 101, are applicable: "full and fair disclosure is a fundamental aspect of the Crown's duty to serve the Court as a faithful public agent, entrusted not with winning or losing trials but rather with seeing that justice is served." Although what constitutes "full" disclosure has grown over time, the Crown's commitment to fairness has always been a guiding principle.

71The trial Crown acknowledged as much in his interview with the CCRG 17 years after the verdict in this case. He recognized that while he was uncertain whether the police would have told him about the relationship in 1986 and 1987, had he known about it, he would have disclosed it. He candidly expressed that he would have been"concerned about … whether there might be some sort of suspicion that the witness was being tampered with because of the close relationship with the officer".

72Of course, if the Crown had a duty to disclose, the police had a corollary obligation to provide the discloseable information to the Crown. As noted by the Honourable G. Arthur Martin, Q.C., in his Report of the Attorney General's Advisory Committee on Charge Screening, Disclosure, and Resolution Discussions (Toronto: Ministry of the Attorney General, Queen's Printer for Ontario, 1993), at p. 167-68:

It is well settled and accepted by all, including the police, that the police although operating independently of Crown counsel, have a duty to disclose to Crown counsel all relevant information uncovered during the investigation of a crime, including information which assists the accused. … As one commentator has observed"the duty of the police to disclose relevant information about a case, to the Crown, is a duty that existed before [Stinchcombe, supra]. (See also: R. v. McNeil, 2009 SCC 3, [2009] 1 S.C.R. 66, at para. 24).

73I agree with the trial Crown's observation that, had he known about the relationship information, he would have disclosed it. Fairness required that approach. Disclosure was not required because P.B.'s evidence was, in fact, tampered with, that she and G.M., in fact, spoke about the case, or that they were, in fact, involved in a relationship at the time of the December 3 courtroom identification. It is not this court's function to decide these factual issues on the reference. Rather, disclosure was required so that the defence could have used the information to test P.B. and G.M.'s evidence in this already exceedingly frail eyewitness identification evidence case.

74The respondent relies upon the fact that P.B. and G.M. have been unequivocal that they "scrupulously avoided discussing the case prior to the trial when their personal involvement began" and that their evidence on this point stands uncontradicted. Whether their evidence stands uncontradicted, though, does not resolve the issue. The question is whether the defence should have been granted the opportunity to explore the potential impact of that relationship on P.B.'s evidence. The answer to this question must be yes.

75This conclusion is supported by G.M.'s own evidence from 2004, where he acknowledged that, in "hindsight""perhaps" he regretted not telling the trial Crown. He placed emphasis, though, on having told his superior officer. As well, G.M. and P.B. both emphasized that their relationship was not a secret within the police community. Indeed, P.B. told the CCRG in 2004 that she and G.M. attended the police "Christmas" party together in 1987. (This would have been around the time of Biddle's trials.) G.M. told the CCRG that, although he may regret not telling the Crown about the relationship, he felt that by telling his superior officer, he had done all that was required of him.

76Further support for the conclusion that the information had to be disclosed lies in P.B.'s and G.M.'s acknowledgement of their discussions surrounding whether the relationship was a problem for the prosecution's case. These discussions demonstrate that they knew that there were trial fairness concerns, ones that they turned their minds to and then dismissed. As P.B. noted:

We talked about it, like: my conscience is clear, your conscience is clear, fine. This is no different than dating a lawyer that might be, you know, with a law firm, the law firm that's representing Biddle. So we looked at our conscience, and no, no, no, we're clean, we've done nothing wrong and certainly nothing to compromise the integrity of the investigation. We were conscious of that. I was painfully conscious of that. I wanted him behind bars for my own safety.

77What they took upon themselves to decide was the wrong question. This was never about whether they were doing something personally wrong. It is possible that they may not have done anything to compromise the case, but that misses the point. The question is what the defence was entitled to know about their relationship so that the already frail eyewitness identification evidence could be put to the test. I conclude that as a matter of fairness, the information ought to have been disclosed.

78Having determined the need for disclosure, I will now turn my attention to what difference it would have made.

ii. Reasonable Possibility that the Undisclosed Information Impacted the Outcome of the Trial

79As I earlier explained, finding a failure to disclose does not result in the breach of full answer and defence on its own. The burden is on the appellant to show that had the information been disclosed, there is a "reasonable possibility" that the jury would have been left with a reasonable doubt about his guilt: Taillefer, at para. 81. It is not the appellate court's function to assess the probative value of the evidence"item by item". Rather, it is this court's function to reconstruct the overall picture of the evidence that would have been presented to the jury, had the undisclosed information been given: Taillefer, at para. 82.

80It is the respondent's position that there is no reasonable possibility that the verdict would have been different even if the fresh evidence had been disclosed having regard to the fact that: (1) P.B.'s evidence was strenuously challenged at trial; and (2) the jury charge on the dangers of identification evidence was strong. The respondent maintains that to reach the opposite conclusion requires speculation that is unsubstantiated on the record.

81The respondent's position aligns with this court's view in Biddle 2 that the identification procedure was faulty, but not so faulty as to render the evidence of identification valueless. P.B. was fully cross-examined, she was confronted with her inconsistencies and she gave explanations. This was all available for the jury's consideration in the context of a strong jury charge on eyewitness identification evidence, at least strong in 1987 terms.

82In my view there exists a reasonable possibility that the jury would have acquitted Biddle had they known then what we know today.

83The fact is that the jury could not convict unless they accepted P.B.'s identification evidence as both credible and reliable. The problem is that her evidence was never tested against a critical fact: she was personally involved in a relationship with one of the investigators on the case, an investigator who would have known everything about the case, and the one who sat with her in the courtroom when the first critical and inherently suspect identification was made.

84I do not agree that the potential impact of the relationship on her evidence is entirely speculative. I say this for several reasons.

85First, although P.B. and G.M. are adamant that they never discussed the case, P.B. clearly had information that was not derived from her personal knowledge. For instance, at one point during cross-examination at trial, she acknowledged that she knew that Biddle lived in her area and that this information came from someone else:

Q. You know that Mr. Biddle lives in your area, don't you?

A. I know that now.

Q. How do you know that?

A. You just told me.

Q. Well, have you known that beforehand?

A. I believe I was told that he lived in the area, yes.

Q. … By whom?

A. I'm sorry, I have no idea.

Q. One of the policemen?

A. I don't know.

Q. Well, I'm quite interested in what you know about Mr. Biddle's background, ma'am. Would you think a little harder about that? You seem to have been told something about Mr. Biddle apart from your own personal knowledge. Could you think about who could have told you this?

A. I'm sorry, but I don't know. I'm the last person anyone tells me. Nobody tells me anything. I want to know.

Q. You were told by someone that he lives in the area, right?

A. I remember being told there's a possibility he lives in this area. There had been a similar attack afterwards.

Q. By a policeman, right?

A. I'm not sure. I don't know.

86It is worth noting that the first question G.M. was asked in cross-examination was whether he had told P.B. that Biddle lived in her area. He denied having done so. Although I accept that P.B. may have learned this information from someone other than G.M., it takes little imagination to think about how this cross-examination would have gone had defence counsel known about the personal relationship. The defence was entitled to that cross-examination.

87Second, although P.B. and G.M. deny having discussed the case, they admit that G.M. prepared P.B. for her testimony at trial, coaching her on matters of courtroom procedure. For instance, he testified that he told P.B. to think about questions before answering, take her time with exhibits and collect her thoughts. The defence was entitled to explore the exact nature of those discussions, particularly in light of P.B.'s strength and conviction as a witness.

88Third, although P.B. and G.M. are clear in their own minds on the timing of the start of their relationship, the defence was entitled to explore that issue. In a rather prescient way, defence counsel suggested to P.B. that Biddle had been pointed out to her as her attacker. She remained firm in her evidence and denied that occurred:

Q: And you were taken to a courtroom and shown Mr. Biddle in custody and had him pointed out to you?

A: You are wrong. It was the face that I saw in that room was the same face I saw in my nightmares prior to the photograph and after I saw him in person. You are wrong.

89Again, one can readily imagine how that cross-examination might have gone had the defence known about the relationship. The reliability of the in-dock identification on December 3, the day before Biddle was charged with the July attack, may well have been badly shaken. This would be particularly true if the defence had been able to establish, or even raise a doubt, about whether G.M. and P.B. were involved at the time of the courtroom identification.

90In short, the evidence of P.B. and G.M.'s relationship may well have tipped the balance of an already extraordinarily thin prosecution case. I find that there exists a reasonable possibility that had the information been disclosed, the trial may have ended differently.

D. CONCLUSION

91P.B. was brutally attacked and undoubtedly believed that the appellant was her attacker. I accept the honesty of that belief. The jury also accepted the honesty of that belief. But the jury had to do more. The jury had to accept the reliability of her identification evidence.

92We cannot know with any certainty what the result of this case would have been had the disclosure been made and had the Crown's case been considered against the fresh evidence. What we do know, though, is that in a case already hanging so finely by a thread, information potentially relevant to the strength of the eyewitness identification evidence was withheld from the trial Crown and, therefore, not disclosed to the defence. Accordingly, the matter was decided without reference to this important information.

93I would answer the questions posed by the Minister of Justice as follows. First, for the reasons given, I would admit the fresh evidence. Having regard to my conclusion on the fresh evidence, the convictions must be quashed, leaving the question as to what remedy should be imposed.

94The appellant leaves the issue of remedy in this court's hands. This is not a case in which a verdict of acquittal should be entered. The appeal was dismissed in 1993. Although the fresh evidence could shake a trier of fact's confidence in the identification evidence, this is not a determination to be made here. All that can be said at this stage, is that there is a reasonable possibility that the fresh evidence could have changed the result at trial.

95I would therefore allow the appeal, quash the convictions and order a new trial. While this may be a difficult, if not impossible prosecution to remount, fairness requires this result.