R. v. Youvarajah, 2013 SCC 41, [2013] 2 SCR 720

 

SUPREME COURT OF CANADA

 

Citation: R. v. Youvarajah, 2013 SCC 41, [2013] 2 S.C.R. 720

Date: 20130725 Docket: 34732

 

Between:

Yousanthan Youvarajah

Appellant

and

Her Majesty The Queen

Respondent

- and -

Criminal Lawyers’ Association

Intervener

 

Coram: McLachlin C.J. and Fish, Abella, Rothstein, Cromwell, Karakatsanis and Wagner JJ.

 

Reasons for Judgment: (paras. 1 to 72)   Dissenting Reasons: (paras. 73 to 153):

Karakatsanis J. (McLachlin C.J. and Fish, Abella and Cromwell JJ. concurring)   Wagner J. (Rothstein J. concurring)

 

 

 

 

R. v. Youvarajah, 2013 SCC 41, [2013] 2 S.C.R. 720

Yousanthan Youvarajah                                                                                Appellant

v.

Her Majesty The Queen                                                                              Respondent

and

Criminal Lawyers’ Association                                                                    Intervener

Indexed as:  R. v. Youvarajah

2013 SCC 41

File No.:  34732.

2013:  February 20; 2013: July 25.

Present:  McLachlin C.J. and Fish, Abella, Rothstein, Cromwell, Karakatsanis and Wagner JJ.

on appeal from the court of appeal for ontario

                    Criminal law — Evidence — Admissibility — Hearsay — Murder trial — Co-accused witness recanting previous statement implicating accused in murder — Trial judge finding prior inconsistent statement not meeting threshold reliability test — Whether prior inconsistent statement was sufficiently reliable to be considered by jury for truth of its contents. 

                    The accused, Y, and D.S., the co-accused shooter, were charged with first degree murder after a failed drug deal.  D.S. was tried separately as a young offender.  He pleaded guilty to second degree murder.  As part of his plea agreement, D.S. signed a written Agreed Statement of Facts (ASF) drafted by Crown counsel, with input from defence counsel.  The ASF directly implicated Y in the murder.  At the time of his plea, D.S. acknowledged the accuracy of the ASF.  The ASF was neither videotaped nor preceded by sworn oath or affirmation.  

                    During Y’s trial, the Crown asked D.S. to adopt the ASF.  D.S. testified that he could not remember signing the document, but acknowledged that it bore his signature.  D.S. further denied the facts in the ASF implicating Y.  In response to D.S.’s denials, the Crown sought to adduce the ASF for the truth of its contents.  Following a voir dire, the trial judge found insufficient means for the jury to assess the reliability of the ASF as a prior inconsistent statement and ruled the signed ASF did not meet the threshold reliability required for it to be admissible as evidence for the truth of its contents.  When D.S. asserted solicitor-client privilege, the ability to cross-examine was curtailed and the trial judge found the opportunity for an effective cross-examination at trial was to a large extent illusory.  The trial judge granted the defence’s application for a directed order acquitting Y.  The Court of Appeal allowed the appeal, set aside the acquittal and ordered a new trial.

                    Held (Rothstein and Wagner JJ. dissenting):  The appeal should be allowed and the acquittal restored.

                    Per McLachlin C.J. and Fish, Abella, Cromwell and Karakatsanis JJ.:  A prior inconsistent statement of a non-accused witness may be admitted for the truth of its contents if the following reliability indicia are met:  (1) the statement is made under oath or solemn affirmation after a warning as to possible sanctions if the person is untruthful; (2) the statement is videotaped or recorded in its entirety; and (3) the opposing party has a full opportunity to cross-examine the witness on the statement.  The prior inconsistent statement’s threshold reliability may also be established by:  (1) the presence of adequate substitutes for testing truth and accuracy (procedural reliability); and (2) sufficient circumstantial guarantees of reliability or an inherent trustworthiness (substantive reliability).  A trial judge is well-placed to assess the hearsay dangers in a particular case and the effectiveness of any safeguards to assist in overcoming them.  Thus, absent an error in principle, the trial judge’s determination of threshold reliability is entitled to deference.  

                    In this case, the trial judge did not err in finding that there were insufficient safeguards to establish threshold reliability to admit the ASF as evidence for the truth of its contents.  Only a full and complete opportunity to cross-examine would have provided a genuine basis on which to assess the reliability of D.S.’s statements.  D.S.’s invocation of solicitor-client privilege, however, curtailed significantly the cross-examination available to assess the threshold reliability of the prior inconsistent statement.  The Crown could not have probed the conversations between D.S. and his counsel about legal advice in connection with his decision to plead guilty or to accept the ASF.  The trial judge may have overstated the scope of solicitor-client privilege and its consequences for the cross-examination of D.S.; however, any overstatement would not have had a material impact on the conclusion that cross-examination was not, in this case, a sufficient means to satisfy threshold reliability.

                    The Crown cannot ask for a new trial on the basis that the prosecution should have been conducted differently.  The trial judge did not preclude the Crown from conducting a more probing cross-examination of the witness or from calling other witnesses about the creation of the ASF.  The trial judge made no comments suggesting that there should be no further cross-examination of the witness on the voir dire.  The judge was not asked for a ruling on this issue.  Had Crown counsel wished to call further evidence, he could have sought direction.  Since this was the same Crown counsel who had prosecuted D.S. at his separate youth trial, and who negotiated the plea bargain and drafted the ASF, he knew what D.S. or his lawyers could say about those matters, outside the scope of solicitor-client privilege.  

                    The trial judge did not err in finding that the circumstances surrounding the making of the prior inconsistent statement did not provide sufficient guarantees of substantive reliability.  The circumstantial guarantees of trustworthiness asserted by the Crown — the thorough process in creating the ASF, the involvement of counsel, and the solemnity of the guilty plea proceeding — do not establish threshold reliability for the statements from which D.S. recanted, which served to minimize his involvement in the murder and shift responsibility to Y.  In the circumstances of this case, the formality of the process and the involvement of counsel only provide comfort in respect of D.S.’s statements admitting his own culpability for the murder.  The administration of justice would not be enhanced in permitting admissions made by a co-accused in his own interest, as part of a plea bargain for a conviction of a lesser crime and favourable sentence, to be used against a co-accused, in circumstances where the reliability of the statements cannot be adequately tested.

                    Per Rothstein and Wagner JJ. (dissenting):  Threshold reliability should be generously interpreted in the admissibility inquiry.  As a general rule, if a statement is accepted for its truth by the courts, and used to balance the liberty interests of the accused with societal considerations such as deterrence and retribution, then that statement provides a level of implicit reliability that warrants consideration in the admissibility inquiry in a subsequent third-party trial.  That is not to say that a statement’s use and judicial acceptance at a prior proceeding is sufficient in and of itself to meet the reliability requirement of the principled approach.  Rather, it is merely a factor that must be considered in the admissibility inquiry, a factor that goes towards meeting the threshold reliability standard.  Each case will have to be examined on its particular circumstances when determining threshold reliability, but consideration should certainly be given to the general acceptance of and reliance on statements read in as part of a guilty plea.  Limits on cross-examination should not be applied rigidly and absolutely where solicitor-client privilege is asserted.  Rather, issues of privilege should be addressed as they arise on cross-examination.  This way, relevant information that may not be prohibited by the assertion of privilege may still be elicited, furthering the underlying truth-seeking function of the proceedings.

                    The question of admissibility in this case must be redetermined in a new trial where threshold reliability can be properly assessed.  The trial judge erred in his determination that the ASF did not meet the standard of threshold reliability.  He failed to adequately assess the available indicia of reliability and erroneously found the opportunity to cross-examine D.S. to be illusory.  That is not to say that the circumstances in this case provide sufficient evidence to conclusively determine that the ASF should be admitted.  Rather, the admissibility inquiry was incomplete and did not provide a proper basis for the trial judge to exclude the relevant evidence.  The opportunity to cross-examine D.S. was not entirely closed and the trial judge misapprehended the scope and impact of solicitor-client privilege on the opportunity for cross-examination.  Had the trial judge permitted the cross-examination to proceed and addressed issues involving solicitor-client privilege as they arose, information sufficient to meet the threshold reliability standard may have been brought forward.  D.S. was available for cross-examination, he recalled having made many of the statements within the ASF, he did not dispute understanding the key facts in the ASF which he subsequently denied at Y’s trial, and he recalled the ASF being read in at his guilty plea proceeding. 

                    Other factors were available in this case that could have satisfied the trial judge that the circumstances surrounding the drafting and reading in of the ASF provided assurances of reliability equivalent to an oath and presence.  If these indicia had been considered and cross-examination had been permitted, it is possible that threshold reliability would have been met.  Perhaps the most persuasive indicia of reliability are the nature of the statement and how it was constructed.  The ASF tendered as evidence in D.S.’s guilty plea proceeding provides just such a record.  Having been reduced to writing, signed not only by D.S. but by both Crown and defence counsel, and having been read in to the record in the context of D.S.’s guilty plea, the source of the document cannot be disputed.  This was not a statement that was unilaterally drafted and forced upon D.S. to adopt.  Rather, the drafting of the ASF was a collaborative effort where there is evidence that at least one crucial component to the Crown’s case against Y was supplied by defence counsel — namely the statement that Y supplied D.S. with the gun.  Moreover, D.S. had the opportunity to review the ASF with his counsel prior to signing it and adopting it.  This is an important aspect as it refutes any negative inferences that could be drawn against the ASF on the grounds that it was not captured in D.S.’s own words or that D.S. did not understand its contents.  Perhaps most importantly, D.S. testified that he understood the three crucial components of the ASF, namely that Y gave D.S. the gun, Y told D.S. to shoot the victim, and demanded that D.S. return the gun.  Without more persuasive evidence to the contrary, this essentially forecloses any argument that D.S. lacked understanding of the ASF’s contents.  Looking beyond the construction of the ASF, the context of the proceedings under which it was read in to court and the contents of the ASF itself provide additional indicia supporting threshold reliability. 

Cases Cited

By Karakatsanis J.

            Referred to:  R. v. B. (K.G.), [1993] 1 S.C.R. 740; R. v. Conway (1997), 36 O.R. (3d) 579; R. v. Khelawon, 2006 SCC 57, [2006] 2 S.C.R. 787; R. v. Khan, [1990] 2 S.C.R. 531; R. v. Smith, [1992] 2 S.C.R. 915; R. v. U. (F.J.), [1995] 3 S.C.R. 764; R. v. Blackman, 2008 SCC 37, [2008] 2 S.C.R. 298; R. v. Hawkins, [1996] 3 S.C.R. 1043; R. v. Devine, 2008 SCC 36, [2008] 2 S.C.R. 283; R. v. Couture, 2007 SCC 28, [2007] 2 S.C.R. 517; R. v. Graveline, 2006 SCC 16, [2006] 1 S.C.R. 609; R. v. Briscoe, 2010 SCC 13, [2010] 1 S.C.R. 411; R. v. Khela, 2009 SCC 4, [2009] 1 S.C.R. 104; R. v. Brooks, 2000 SCC 11, [2000] 1 S.C.R. 237.

By Wagner J. (dissenting)

                    R. v. B. (K.G.), [1993] 1 S.C.R. 740; R. v. Conway (1997), 36 O.R. (3d) 579; R. v. Khelawon, 2006 SCC 57, [2006] 2 S.C.R. 787; R. v. U. (F.J.), [1995] 3 S.C.R. 764; R. v. Couture, 2007 SCC 28, [2007] 2 S.C.R. 517; R. v. Trieu (2005), 195 C.C.C. (3d) 373; R. v. Blackman, 2008 SCC 37, [2008] 2 S.C.R. 298; R. v. Hawkins, [1996] 3 S.C.R. 1043; R. v. Tran, 2010 ONCA 471, 103 O.R. (3d) 131; R. v. D.P., 2010 ONCA 563, 268 O.A.C. 118; R. v. McGee, 2009 CanLII 60789; R. v. McClure, 2001 SCC 14, [2001] 1 S.C.R. 445; R. v. S.G.T., 2010 SCC 20, [2010] 1 S.C.R. 688.

Authors Cited

Paciocco, David M., and Lee Stuesser.  The Law of Evidence, 6th ed.  Toronto:  Irwin Law, 2011.

Proulx, Michel, and David Layton.  Ethics and Canadian Criminal Law.  Toronto:  Irwin Law, 2001.

Wigmore, John Henry.  Evidence in Trials at Common Law, vol. 8.  Revised by John T. McNaughton.  Boston, Mass.:  Little, Brown, 1961.

            APPEAL from a judgment of the Ontario Court of Appeal (Winkler C.J.O. and Moldaver and Simmons JJ.A.), 2011 ONCA 654, 107 O.R. (3d) 401, 284 O.A.C. 300, 278 C.C.C. (3d) 102, 90 C.R. (6th) 184, [2011] O.J. No. 4610 (QL), 2011 CarswellOnt 11167, setting aside the accused’s acquittal entered by Flynn J. on March 17, 2010, and ordering a new trial.  Appeal allowed and acquittal restored, Rothstein and Wagner JJ. dissenting.

                    Philip R. Campbell and Jonathan Dawe, for the appellant.

                    James K. Stewart and Nadia Thomas, for the respondent.

                    Marie Henein and Matthew Gourlay, for the intervener.

 

                    The judgment of McLachlin C.J. and Fish, Abella, Cromwell and Karakatsanis JJ. was delivered by

                    Karakatsanis J. —

I.    Introduction

[1]                              The issue in this appeal is whether a co-accused’s prior inconsistent statement, implicating the appellant in a murder, was sufficiently reliable to be considered by a jury for the truth of its contents.

[2]                              The Crown’s theory was that the appellant planned the murder and provided the murder weapon to the co-accused shooter.  The shooter, in a separate proceeding in youth court, pleaded guilty to second degree murder.  In doing so, he adopted a written Agreed Statement of Facts (ASF) in which he shifted responsibility for planning the murder and for obtaining the murder weapon to the appellant.

[3]                              As a witness in the appellant’s murder trial, the shooter resiled from those facts inculpating the appellant and refused to adopt the prior statement.  The Crown’s case against the appellant collapsed when the trial judge found that the ASF did not meet the threshold reliability required to be admitted for the truth of its contents.

[4]                              The question before us is whether the trial judge erred in finding that the ASF did not meet threshold reliability to be put before the jury.

[5]                              The Court of Appeal concluded that the trial judge erred.  However, for the reasons that follow, I would allow the appeal and reinstate the acquittal.

II.     Background

[6]                              Yousanthan Youvarajah, the appellant, and D.S., a minor, were charged with first degree murder after a failed drug deal.  On the Crown’s theory, Youvarajah planned the murder and provided the murder weapon for D.S. to shoot the victim.

[7]                              D.S. was tried separately as a young offender.  He pleaded guilty to second degree murder and was sentenced as a youth.  As part of his plea agreement, D.S. signed the ASF drafted by Crown counsel, with input from defence counsel.  In it, D.S. directly implicated the appellant in the murder:  the appellant gave D.S. the handgun that was used in the shooting; he ordered D.S. to shoot the victim; and he demanded the return of the handgun after the shooting.

[8]                              At the time of his plea, D.S. acknowledged the accuracy of the ASF.  At the appellant’s trial, he testified that he did not understand the words “acknowledged” or “accurate” as used by his counsel at the guilty plea proceedings.

[9]                              The ASF was neither videotaped nor preceded by sworn oath or affirmation.  D.S. was told that, for endorsing the agreed statement, no further statement regarding the murder would be required from him.  During his testimony at the appellant’s trial, he said that was one of the reasons that motivated him to plead guilty.

[10]                          During examination-in-chief, the Crown asked D.S. to adopt the ASF.  D.S. testified that he could not remember signing the document, but acknowledged that it bore his signature.  D.S. further denied the facts in the statement implicating the appellant.  Instead, D.S. testified that the gun was his own; he shot the victim because of the way he was talking; and he threw the gun into the river after the shooting.

[11]                          In response to D.S.’s denials, the Crown sought to adduce the ASF for the truth of its contents.  The trial judge rejected the Crown’s application.  At the close of the Crown’s case, the trial judge granted the defence’s application for a directed order acquitting the appellant.

III.    Decisions Below

[12]                          Following a voir dire, Flynn J. ruled that the signed ASF — hearsay evidence by virtue of being an out-of-court statement adduced for the truth of its contents — did not meet the threshold reliability required for it to be admissible as evidence.  None of the safeguards identified in this Court’s decision in R. v. B. (K.G.), [1993] 1 S.C.R. 740 (K.G.B.), for reducing the dangers associated with hearsay had been undertaken.  The statement was not videotaped; there had been no recital of an oath or affirmation; and the transcript of the guilty plea proceedings was not a suitable substitute to assess D.S.’s demeanour and credibility at the time the statement was made.  Furthermore, the statement had been drafted by counsel; it was not offered spontaneously and was not in D.S.’s own words.

[13]                          Referring to R. v. Conway (1997), 36 O.R. (3d) 579 (C.A.), Flynn J. noted that the opportunity for an effective cross-examination at trial would be “to a large extent illusory” where the declarant experiences significant memory lapses or where he recants (A.R., vol. I, at p. 61).  D.S.’s refusal to waive solicitor-client privilege precluded questions to D.S. or his counsel about how D.S. came to implicate the appellant (A.R., vol. I, at p. 59).

[14]                           Flynn J. found insufficient means for the jury to assess the reliability of the ASF as a prior inconsistent statement and held that it was not admissible as evidence for the truth of its contents.

[15]                          The Ontario Court of Appeal concluded that the trial judge erred in his understanding of the scope of solicitor-client privilege.  Solicitor-client privilege would not have precluded all questioning of D.S. or his counsel regarding the ASF or his choice to implicate the appellant.  The Court of Appeal also held that the trial judge further erred in strictly focussing on the K.G.B. requirements for establishing reliability and in failing to consider other factors, such as the solemnity of the occasion upon which D.S. adopted the ASF and the involvement of counsel in preparing it.

[16]                          Satisfied that without such errors the trial judge might have admitted the ASF into evidence, the Ontario Court of Appeal allowed the appeal, set aside the acquittal, and ordered a new trial.

IV.    Issues

[17]                          The question in this appeal is whether the trial judge erred in finding that there were insufficient safeguards to establish threshold reliability to admit the ASF as evidence and, if he did, whether the error had a material bearing on the result.  Answering this question requires the consideration of two issues:

(a)        Did the trial judge err in finding that cross-examination did not provide a sufficient basis for the jury to assess the truthfulness of the prior inconsistent statement?

(b)        Did the trial judge err in finding that the circumstances surrounding the making of the prior inconsistent statement did not provide sufficient guarantees of substantive reliability?

V.     Legal Principles

A.     Hearsay and Threshold Reliability

[18]                          Hearsay evidence — an out-of-court statement tendered for the truth of its contents — is presumptively inadmissible.  This is because the dangers associated with hearsay evidence may undermine the truth-seeking function of a trial or its fairness.  These dangers typically include an inability to test and assess a declarant’s perception, memory, narration, or sincerity:  R. v. Khelawon, 2006 SCC 57, [2006] 2 S.C.R. 787, at para. 2.

[19]                          The law has conventionally favoured the evidence of witnesses who give evidence in court because they can be observed, under oath or affirmation, and their credibility and reliability can be tested by cross-examination.  These elements help the trier of fact assess the credibility of the declarant or witness, the reliability of the evidence, and the degree of probative force it carries.  When these elements are absent, as is the case with a statement made outside of the court, it is more difficult for the trier of fact to make these assessments.

[20]                          Over time, however, the law has recognized that in certain circumstances, it may be safe to rely on out-of-court statements for the truth of their contents.  Exceptions to the hearsay rule developed for statements carrying certain guarantees of inherent trustworthiness, often because of the circumstances in which they were made (for example, dying declarations and declarations that are adverse in interest).

[21]                          In addition to the traditional exceptions, however, this Court developed a principled approach that permits trial judges to admit hearsay evidence if it meets the twin threshold requirements of necessity and reliability.  This is a flexible case-by-case examination.  See especially R. v. Khan, [1990] 2 S.C.R. 531; R. v. Smith, [1992] 2 S.C.R. 915; K.G.B.; R. v. U. (F.J.), [1995] 3 S.C.R. 764; and R. v. Blackman, 2008 SCC 37, [2008] 2 S.C.R. 298, at para. 38.  With the increased flexibility provided by the principled approach, however, the gatekeeper function of the trial judge becomes more complex and nuanced.

[22]                          Where a witness recants from a prior statement, necessity is established: Khelawon, at para. 78.  The focus in this case is on whether the prior inconsistent statement meets threshold reliability.

[23]                          The trial judge, as the evidentiary gatekeeper, assesses the threshold reliability of the hearsay statement.  The decision as to the ultimate reliability of the statement is left to the trier of fact:  Khelawon, at para. 2. Even if the necessity and reliability of the hearsay evidence are proven, the trial judge maintains discretion to exclude the evidence where the “prejudicial effect is out of proportion to its probative value”:  Khelawon, at para. 3.

[24]                          Why not simply let the trier of fact determine both threshold and ultimate reliability?  Professors D. M. Paciocco and L. Stuesser provide the following explanation, with which I agree:

In considering “reliability”, a distinction is made between “threshold” and “ultimate” reliability.  This distinction reflects the important difference between admission and reliance.  Threshold reliability is for the trial judge and concerns the admissibility of the statement.  The trial judge acts as a gatekeeper whose function “is limited to determining whether the particular hearsay statement exhibits sufficient indicia of reliability so as to afford the trier of fact a satisfactory basis for evaluating the truth of the statement.”  So long as it can be assessed and accepted by a reasonable trier of fact, then the evidence should be admitted.  Once admitted, the jury remains the ultimate arbiter of what to do with the evidence and deciding whether or not the statement is true.

 

(The Law of Evidence (6th ed. 2011), at pp. 122-23)

 

See R. v. Hawkins, [1996] 3 S.C.R. 1043, at para. 75; and Khelawon, at paras. 50-52.

[25]                          Threshold reliability serves an important function.  Rules of evidence and principles governing the admissibility of evidence exist in the first place because experience teaches that certain types of evidence can be presumptively unreliable (or prejudicial) and can undermine the truth-seeking function of a trial.  Rules of admissibility of evidence address trial fairness and provide predictability.  They also provide the means to maintain control over the scope of criminal trials to keep them manageable and focussed on probative and relevant evidence.

B.     Admissibility of Prior Inconsistent Statements

[26]                          Historically, an out-of-court prior inconsistent statement of a non-accused witness was admissible only to impeach the credibility of the witness.  A prior inconsistent statement — hearsay evidence — was not admissible for the truth of its contents unless the witness adopted the prior statement in court.  Otherwise, the jury was limited to rejecting the viva voce evidence of the recanting witness; the jury could not substitute the contents of the out-of-court statement.

[27]                          This traditional rule excluding prior inconsistent statements was altered in K.G.B. to conform with the evolving principled approach to hearsay.  On an exceptional basis, a prior inconsistent statement is admissible for the truth of its contents, provided the threshold criteria of necessity and reliability are established.

[28]                          In K.G.B., at p. 787, Lamer C.J. stated that the focus of the reliability inquiry, when dealing with prior inconsistent statements, “is on the comparative reliability of the prior statement and the testimony offered at trial, and so additional indicia and guarantees of reliability . . . must be secured in order to bring the prior statement to a comparable standard of reliability before such statements are admitted as substantive evidence”.

[29]                          Accordingly, Lamer C.J. held, at pp. 795-96, that a prior inconsistent statement of a non-accused witness may be admitted for the truth of its contents if the so-called K.G.B. reliability indicia are met:  (1) the statement is made under oath or solemn affirmation after a warning as to possible sanctions if the person is untruthful; (2) the statement is videotaped or recorded in its entirety; and (3) the opposing party has a full opportunity to cross-examine the witness on the statement.  Such K.G.B. statements have become prevalent, especially in murder investigations.

[30]                          However, the K.G.B. indicia are not the only means of establishing threshold reliability.  The prior inconsistent statement’s threshold reliability may be established by:  (1) the presence of adequate substitutes for testing truth and accuracy (procedural reliability); and (2) sufficient circumstantial guarantees of reliability or an inherent trustworthiness (substantive reliability):  Khelawon, at paras. 61-63. These two principal ways of showing threshold reliability are not mutually exclusive:  R. v. Devine, 2008 SCC 36, [2008] 2 S.C.R. 283, at para. 22.

[31]                          The admissibility of hearsay evidence, such as the prior inconsistent statement in this case, is a question of law.  Of course, the factual findings that go into that determination are entitled to deference and are not challenged in this case.  As well, a trial judge is well placed to assess the hearsay dangers in a particular case and the effectiveness of any safeguards to assist in overcoming them.  Thus, absent an error in principle, the trial judge’s determination of threshold reliability is entitled to deference:  R. v. Couture, 2007 SCC 28, [2007] 2 S.C.R. 517, at para. 81.

[32]                          To obtain a new trial following an acquittal, the Crown must show that the trial judge erred and that this error “might reasonably be thought . . . to have had a material bearing on the acquittal”:  R. v. Graveline, 2006 SCC 16, [2006] 1 S.C.R. 609, at para. 14.  The Crown is not required to establish “that the verdict would necessarily have been different”:  Graveline, at para. 14.  This is still, however, a “heavy onus” for the Crown:  R. v. Briscoe, 2010 SCC 13, [2010] 1 S.C.R. 411, at para. 26.

VI.     Application

[33]                           In light of the trial judge’s obligation to act as evidentiary gatekeeper, I turn first to the specific hearsay dangers posed by the prior inconsistent statement in this case as noted by the trial judge.  In the context of the plea bargain, D.S. had a strong incentive to minimize his role in the crime and to shift responsibility to the appellant, a co-accused, in order to obtain a favourable outcome.  There was no opportunity to observe the demeanour of D.S. or his own choice of words, as the ASF was drafted by counsel and was not spontaneous.  When D.S. asserted solicitor-client privilege, the ability to cross-examine was curtailed.

[34]                          The trial judge therefore considered whether cross-examination of D.S. at the appellant’s trial would provide an adequate procedural basis to permit the jury to assess the veracity of the prior statement and whether there were other circumstances that provided circumstantial assurances of trustworthiness.

A.     Procedural Substitutes:  The Opportunity to Cross-Examine the Recanting Witness

[35]                          The most important factor supporting the admissibility of a prior inconsistent statement of a non-accused witness for the truth of its contents is the availability of the non-accused witness for cross-examination.  In Couture, Charron J. held that “the availability of the declarant for cross-examination goes a long way to satisfying the requirement for adequate substitutes” for testing the evidence (at para. 92) and that “the opportunity to cross-examine is the most powerful factor favouring admissibility” (para. 95).

[36]                          In assessing the means by which a jury could rationally evaluate the truthfulness and accuracy of the statement, the trial judge correctly noted the importance of cross-examination.  He stated that it was an important means for the jury to determine whether a previous statement from a witness was ultimately reliable.  However, referring to Conway, the trial judge noted that the opportunity for an effective cross-examination of D.S. at the appellant’s trial would be “to a large extent illusory” due to D.S.’s memory lapses with respect to the ASF and his invocation of solicitor-client privilege.

[37]                          With respect to solicitor-client privilege, the trial judge concluded:

Solicitor/client privilege precluded questions to or from [D.S.] or indeed of his counsel about how it came to be that this confession from [D.S.] as the shooter and his implication of the accused as the person who provided him with the gun and directed him to do the shooting was made on the day of his guilty plea almost two years after his arrest when he had never confessed or implicated Yousanthan Youvarajah in all that time. [A.R., vol. I, at p. 59]

[38]                          The respondent agrees with the Ontario Court of Appeal that the trial judge erred in overstating the scope of solicitor-client privilege and failed to recognize that there would be significant areas of cross-examination open to counsel at trial.  The Court of Appeal found that the trial judge erred in relying on Conway to conclude that cross-examination of D.S. would be “to a large extent illusory”.  As a result, the court concluded that the trial judge may have admitted the ASF if he had appreciated the broader scope of cross-examination available to test its reliability.

[39]                          The Court of Appeal concluded that both counsel and the trial judge appeared “to have been of the mistaken view” that solicitor-client privilege would preclude certain questions being asked of D.S. with respect to the ASF (2011 ONCA 654, 107 O.R. (3d) 401, at para. 93).  Further, at para. 95, the Court of Appeal cited the trial judge’s statement that “no legal witnesses that have [ever] spoken to [D.S.]” should be called (A.R., vol. II, at p. 132).

B.  Solicitor-Client Privilege

[40]                          On a strict reading of the passage quoted above, I agree that the trial judge overstated the scope of solicitor-client privilege.  It would have been more accurate for the trial judge to say that it precluded many questions about his decision to accept the plea bargain and his reasons for implicating the appellant. Simmons J.A., for the unanimous panel of the Court of Appeal, outlined a number of areas of questioning that remained available to the Crown (see para. 84).

[41]                          However, for the reasons that follow, on my reading of the reasons and the voir dire record, I am not persuaded that the trial judge erred in concluding that cross-examination was “to a large extent illusory” and insufficient to overcome the hearsay dangers in this particular case.  Further, I do not agree that the trial judge precluded the Crown from conducting a more probing cross-examination of the witness or from calling other witnesses about the creation of the ASF.  Finally, I am not persuaded that any overstatement of the scope of solicitor-client privilege would have had a material impact on the conclusion that cross-examination was not, in this case, a sufficient means to satisfy threshold reliability.

[42]                          Once D.S. confirmed that he would not waive solicitor-client privilege, the Crown chose to circumscribe its cross-examination and had no further questions of the witness.  The trial judge made no comments suggesting that there should be no further cross-examination of the witness on the voir dire.

[43]                          The following exchange between Crown counsel and the trial judge occurred after independent counsel confirmed that D.S. would not waive solicitor-client privilege (A.R., vol. II, at p. 132):

[CROWN COUNSEL]:  And I don’t anticipate even now given the comments of [D.S.’s independent counsel] Mr. Marentette I don’t anticipate needing to call further witnesses.

THE COURT:  Well, no legal witnesses that have ever spoken to this man anyway.

[CROWN COUNSEL]:  Right.

THE COURT:  Right.  I mean we’re going to take Mr. Marentette’s word as the final word on that.

[CROWN COUNSEL]:  Right.

[44]                          This exchange between the trial judge and Crown counsel must be read in context.  It does not amount to a direction or a ruling not to call other witnesses.  The trial judge merely affirmed Crown counsel’s statement that he would not be calling any “legal witnesses that have ever spoken to” the witness (emphasis added).  The judge was not asked for a ruling.  Had Crown counsel wished to call further evidence, he could have sought direction.  The judge would no doubt have sought submissions and considered the extent of the restrictions imposed by solicitor-client privilege before making a ruling.

[45]                          Since this was the same Crown who had prosecuted D.S. at his separate youth trial, and who negotiated the plea bargain and drafted the ASF, he knew what D.S. or his lawyers could say about those matters, outside the scope of solicitor-client privilege.  Thus, I reject the submission that the trial judge curtailed the scope of evidence on the voir dire.

[46]                          I conclude that the voir dire unfolded as it did primarily due to the Crown’s prosecutorial decisions.  Crown counsel was not precluded by the trial judge from calling further witnesses or from posing further questions to D.S.  The Crown cannot ask for a new trial on the basis that the prosecution should have been conducted differently.

[47]                          Further, it should not be inferred from this exchange that the trial judge did not correctly understand the scope of solicitor-client privilege when he made his ruling on the voir dire.  Trial judges are presumed to know the law.  Further, given the context of this informal exchange, I am not persuaded that it provides a basis to conclude that the trial judge did not understand the scope of solicitor-client privilege.  He was neither stating a legal principle nor making a ruling.

[48]                          Finally, notwithstanding the aspects of cross-examination that would have remained available, the reality is that solicitor-client privilege would curtail significantly the cross-examination available to assess the threshold reliability of the prior inconsistent statement.  The Crown could not have probed the conversations between D.S. and his counsel about legal advice in connection with his decision to plead guilty or to accept the ASF.

[49]                          Simply put, D.S. recanted the portions of the ASF that implicated the appellant and replaced them with assertions that exonerated the appellant.  Cross-examination of D.S. did not elicit an explanation for his about-face from the assertions that tied the appellant to the murder.  This Court stated in U. (F.J.), at para. 46, that if the witness “provides an explanation for changing his or her story, the trier of fact will be able to assess both versions of the story, as well as the explanation”.  Paciocco and Stuesser similarly state that “[a] testing of the witness’s recantation is only possible when the witness admits making the earlier statement and provides a story for his or her recantation” (p. 131 (emphasis added)).  See also Khelawon, at para. 76.

[50]                          In this case, solicitor-client privilege would hinder the fact finder’s opportunity to fully explore any explanation offered.  The jurisprudence emphasizes the presence of a “full opportunity to cross-examine the witness [at trial] respecting the statement”:  K.G.B., at p. 796 (emphasis added).  In Devine, Charron J. held:

It is important to note that the availability of the declarant to be cross-examined will not necessarily tip the scales in favour of admissibility.  In order for this factor to weigh in favour of admission, there must be a “full opportunity to cross-examine the witness” at trial (K.G.B., at p. 796). [para. 26]

[51]                          The trial judge referred to Conway, a case in which the witness could not recall making the prior statement and in which cross-examination therefore would not assist in determining which version was true.  Unlike Conway, two versions of the events were before the trial judge in this case that could have been the subject of cross-examination (i.e., the ASF and D.S.’s testimony at the appellant’s trial).

[52]                          The trial judge noted in this case that there were significant memory lapses in addition to D.S.’s recantation and direct contradiction.  I am not satisfied that a fair reading of the trial judge’s reasons suggests that he misunderstood the law.  While he quoted from Conway, and the circumstances in that case were somewhat dissimilar, the words he quoted were apt.

[53]                          As noted above, solicitor-client privilege would have significantly limited the effectiveness of the cross-examination.  Thus, even if the trial judge had erred in overstating the scope of solicitor-client privilege or the analogy to Conway, I am not persuaded that it would have materially affected the outcome in these circumstances.

[54]                          Given the nature of the hearsay dangers in this particular case, the trial judge did not err in concluding that nothing short of full cross-examination could overcome them.  Here, the cross-examination at trial would be significantly limited by the claim of solicitor-client privilege.  The statement was not videotaped.  There had been no oath or affirmation when the statement was made.  The transcript of the guilty plea proceedings established the words of the prior statements but was not a suitable substitute to assess D.S.’s demeanour and credibility at the time the statement was made.  Lastly, the ASF was not spontaneous and was not in D.S.’s own words.  Under the circumstances, I cannot conclude that the trial judge erred in finding that there were inadequate substitutes to test and assess the truth of the evidence.

C.     Inherent Trustworthiness:  The Circumstances Surrounding the Preparation and Presentation of the ASF

[55]                          If the circumstances in which the out-of-court statement came about do not give rise to a concern over whether the statement is true or not, the statement possesses an inherent trustworthiness or substantive reliability.

[56]                          The Ontario Court of Appeal concluded that the trial judge erred in failing to consider whether the circumstances surrounding the preparation and presentation of the ASF provided an inherent trustworthiness to establish threshold reliability.

[57]                          In my view, the circumstantial guarantees of trustworthiness asserted by the respondent — the thorough process in creating the ASF, the involvement of counsel, and the solemnity of the guilty plea proceeding — do not establish threshold reliability for the statements from which D.S. recanted, which served to minimize his involvement in the murder and shift responsibility to the appellant.  In the circumstances of this case, the formality of the process and the involvement of counsel only provide comfort in respect of D.S.’s statements admitting his own culpability for the murder.

[58]                          D.S.’s guilty plea was, indeed, a solemn occasion.  The ASF was drafted by the Crown and defence counsel.  It was signed and acknowledged in open court by both D.S. and his defence counsel at the time D.S. pleaded guilty to second degree murder.  D.S. was represented by counsel when he endorsed the ASF and it formed the basis upon which the judge entered a conviction for second degree murder and fashioned his sentence.

[59]                          To the extent that the ASF incriminated D.S., was against his interests, and admitted his own culpability in court, these circumstances provide a compelling inference that those statements were in fact reliable in establishing D.S.’s criminal conduct.  However, the underlying rationale for the admissibility of admissions as against the party making them falls away when they are sought to be used against a third party.

[60]                          Here, the portions of the ASF that the Crown sought to rely upon at the appellant’s trial are statements that shifted responsibility for the murder from D.S., the shooter, to his co-accused, the appellant.  They were elements that would support a plea to the lesser offence of second degree murder as well as support a more advantageous sentence.

[61]                          Furthermore, the involvement of defence counsel provides no meaningful check on the danger of an accused acknowledging false allegations against a third party in order to obtain a favourable plea bargain.  Counsel have an ethical duty to not knowingly mislead the court.  However, it does not require them to verify or investigate the truth of information they present; and the duty is triggered only where counsel has information leading to the “irresistible conclusion” that something is false.  See M. Proulx and D. Layton, Ethics and Canadian Criminal Law (2001), at pp. 40-47 and 460.

[62]                          The suggestion that the solemnity of the occasion or the involvement of counsel increases the inherent trustworthiness of the statement does not resonate to the extent that it incriminates a third party — in this case a co-accused.  Criminal law is generally and rightfully suspicious of allegations made by a person against an accomplice.  It has long been recognized that evidence of one accomplice against another may be motivated by self-interest and that it is dangerous to rely on such evidence absent other evidence which tends to confirm it.  The fact that such statements are contained in an ASF does not provide any reassurance of reliability.  Indeed, statements by a co-accused or accomplice are recognized as inherently unreliable.

[63]                          Vetrovec warnings issued by trial judges, for example, provide jurors with direction to consider “all of the factors that might impair [the] credibility [of the testimony of co-accused or accomplices]”:  R. v. Khela, 2009 SCC 4, [2009] 1 S.C.R. 104, at para. 31.  Triers of fact, under such conditions, are able to determine whether the “evidence properly weighed [overcomes] its suspicious roots”: R. v. Brooks, 2000 SCC 11, [2000] 1 S.C.R. 237, at para. 69.

[64]                          The administration of justice would not be enhanced in permitting admissions made by a co-accused in his own interest, as part of a plea bargain for a conviction of a lesser crime and favourable sentence, to be used against a co-accused, in circumstances where the reliability of the statements cannot be adequately tested.

[65]                          In this case, the trial judge was alive to the circumstances surrounding the preparation and presentation of the ASF.  He described the circumstances in some detail, including those that weighed in favour of and against reliability.  He considered whether the solemnity of D.S.’s guilty plea proceeding demonstrated the trustworthiness of the ASF.  He appreciated that defence counsel assisted Crown counsel in the drafting of the statement and that, according to testimony during the voir dire, the information about the source of the gun came from the defence.  He also noted the evidence from D.S. that he did not understand all of the words in the ASF and that the assurance he would not have to give any further statement to the police was one of the reasons he agreed to the plea bargain.

[66]                          The trial judge identified D.S.’s incentive to minimize his own involvement in order to obtain a youth sentence for second degree murder.  Indeed, D.S. testified that he agreed to facts outside of his knowledge or which he knew to be wrong in order to secure his plea bargain. In the context of this plea bargain, D.S. was motivated to minimize his own conduct and maximize that of the co-accused.

[67]                          In the circumstances of this case, I cannot conclude that the trial judge erred by failing to consider whether the solemnity of the occasion or the involvement of counsel provided circumstantial guarantees of substantive reliability.

VII.  Summary and Conclusion

[68]                          The trial judge may have overstated the scope of solicitor-client privilege and its consequences for the cross-examination of D.S.  However, I am not persuaded that the Crown has satisfied its burden to show that any errors would have materially affected the conclusion reached by the trial judge.

[69]                          The circumstances identified by the trial judge raise significant concerns about the threshold reliability of the portions of the ASF upon which the Crown sought to rely at the appellant’s trial, all of which minimized D.S.’s involvement in the murder.  D.S. endorsed the ASF as part of a plea bargain for second degree murder and a sentence in youth court. In these circumstances, there was motivation to shift responsibility to his co-accused.  D.S. was also assured that he would not have to make any further statements to police and he testified at the appellant’s trial that this was one of the reasons that he had accepted the plea agreement.   D.S. further testified that he agreed to some facts in the ASF that he said he did not or could not know and that he did not understand everything that he read before agreeing to the statement’s contents.  Those portions of the ASF that shifted responsibility for the murder to the appellant are inherently unreliable.

[70]                          Further, the trial judge identified the following factors that made it difficult to assess the veracity of the prior inconsistent statement.  The statement was not videotaped or taken under oath.  There was no opportunity to assess D.S.’s demeanour or rely upon any spontaneous choice of words.  The police and Crown counsel chose not to videotape, under oath, as is routinely available, especially in homicide cases.

[71]                          In light of these difficulties, only a full and complete opportunity to cross-examine would have provided a genuine basis on which to assess the reliability of D.S.’s statements.  His invocation of solicitor-client privilege, however, would preclude a full opportunity to cross-examine him.

[72]                          The trial judge did not err in finding that the jury would not have the tools to assess or test the truthfulness of the exculpatory portions of the ASF.  I would allow the appeal and restore the acquittal.

                     The reasons of Rothstein and Wagner JJ. were delivered by

                    Wagner J. (dissenting) —

I.       Background

[73]                          At its core, this case deals with the admissibility of hearsay statements under the principled approach to hearsay rooted in necessity and reliability.  The hearsay evidence in question is a prior statement which was inconsistent with the testimony of a witness (also the co-accused) at the trial of the accused.  The prior inconsistent statement was an Agreed Statement of Facts (“ASF”) which was read in as part of the guilty plea of the co-accused who was being tried separately as a youth. 

[74]                          While there is a great deal of case law that has dealt with prior inconsistent statements under the principled approach, there are two wrinkles which make its application to this case unique.  First, unlike the police and out-of-court statements that are dealt with in much of the existing jurisprudence, the ASF at issue was accepted by the courts as part of a guilty plea and sentencing relating to the co-accused’s participation in the crime in question.  Second, the analysis under the principled approach was fraught with procedural errors stemming from the conduct of the trial judge as well as both Crown and defence counsel.  With these two differences in mind and for the reasons that follow, I believe that the question of admissibility must be redetermined in a new trial.  

II.     Background

[75]                          The appellant, Yousanthan Youvarajah, stands accused of the first degree murder of Andrew Freake who was shot and killed during a drug deal that went wrong on October 11, 2007.  While Youvarajah did not pull the trigger, the Crown contends that the appellant orchestrated the shooting of Freake as retribution for being short-changed during two earlier sales of cocaine.

[76]                          The events leading to the fatal shooting were initiated when Youvarajah called Freake to arrange for the purchase of one and a half pounds of marijuana.  After a series of delays, including an initial meeting in an apartment building and two changes of venue for the exchange due to excessive crowds, the parties met in Clyde Park to complete the transaction.  Freake, who was accompanied by two friends, followed the appellant into the park.

[77]                          Youvarajah was accompanied by a young person, D.S., and two other acquaintances, Abhishaik Shinde and Raibeen Mohammad.  They went to the meeting place in an SUV driven by Shinde.  D.S. sat in the front passenger seat with the appellant sitting directly behind him.  Mohammad sat in the back seat behind Shinde.

[78]                          Once inside the park, Shinde drove the SUV onto a grassy area and Freake stopped in a nearby parking lot.  Freake and one of his acquaintances approached the SUV by foot carrying three bags of marijuana.  Two bags of marijuana were passed into the car to be tested.  When Freake refused to pass the third bag, D.S. pulled out a handgun and shot Freake in the chest.

[79]                          Within days of the shooting, both D.S. and Youvarajah were arrested and charged with first degree murder.  The charges against D.S. related to his role as the shooter while charges against Youvarajah were for planning and orchestrating the shooting. Since D.S. was a youth, his trial proceeded separately.  He pleaded guilty to second degree murder with the consent of the Crown.  As part of his guilty plea proceeding, an ASF which was signed by D.S., his two trial counsel, and the Crown, was read in and filed. 

[80]                          In the ASF, D.S. claimed that the appellant (1) gave D.S. the handgun used in the shooting; (2) told D.S. to shoot Freake at some point during the drug deal; and (3) demanded that D.S. return the handgun to him after the shooting.  These three elements of the ASF were crucial components of the Crown’s case against Youvarajah.

[81]                          At the appellant’s trial, D.S. was called to testify about Youvarajah’s alleged involvement in planning the shooting.  While D.S. confirmed that he was the shooter, his testimony regarding the appellant’s involvement contradicted portions of the ASF, namely the three elements previously outlined.  Once on the stand, D.S. claimed that (1) he obtained the handgun from a prior break and enter; (2) he did not shoot Freake on Youvarajah’s instructions but rather because Freake was being disrespectful; and (3) he threw the murder weapon into a river after the shooting.

[82]                          In light of D.S.’s testimony, the Crown applied to have the ASF admitted for the truth of its contents.  The trial judge dismissed the Crown’s application, finding that the ASF did not meet the standard of threshold reliability that is necessary for evidence to be admitted under the principled approach to hearsay evidence.  Without the relevant evidence implicating the accused, the trial judge granted Youvarajah’s application for a directed verdict of acquittal. 

[83]                          In reaching his conclusions, the trial judge focused on the three indicia of reliability outlined in R. v. B. (K.G.), [1993] 1 S.C.R. 740 (“B. (K.G.)”).  After noting that D.S. did not take an oath, he found that the jury would not have the opportunity to assess the witness’ demeanour since the guilty plea was not videotaped.  Accordingly, the trial judge’s reliability analysis hinged on the opportunity of cross-examining D.S.

[84]                          During the trial proceedings, defence counsel suggested that the examination of D.S. regarding the ASF could be restricted by solicitor-client privilege.  D.S. was granted the opportunity to obtain independent legal advice.  After consulting with independent counsel, Mr. Marentette, D.S. informed the court that he would not waive solicitor-client privilege in any way.  In light of D.S.’s assertion of solicitor-client privilege, the trial judge found that the opportunity to cross-examine D.S. about the ASF was illusory. 

[85]                          In his reasons, the trial judge also noted that the ASF lacked additional indicia of reliability beyond the three markers laid out in B. (K.G.).  Namely, the ASF was not spontaneous, it was not crafted in D.S.’s own words, and the circumstances of the ASF allowed the court to infer a motive to lie in order to secure a lesser sentence as part of the guilty plea.  The trial judge further noted that D.S. was never warned of the potential penal consequences of later recanting. 

[86]                          The Crown appealed the decision to the Ontario Court of Appeal where the acquittal was set aside and a new trial ordered.  In her reasons, Simmons J.A. addressed three issues.  First, she held that the trial judge erred in finding that the opportunity to cross-examine D.S. was illusory.  Specifically, Simmons J.A. found that the trial judge “put his stamp of approval” on the position of both defence and Crown counsel that a cross-examination of D.S. or D.S.’s counsel about the circumstances giving rise to the ASF was precluded due to D.S.’s assertion of privilege (2011 ONCA 654, 107 O.R. (3d) 401, at para. 95).  Simmons J.A. also distinguished the instant case from R. v. Conway (1997), 36 O.R. (3d) 579, where the Ontario Court of Appeal found the possibility of cross-examining a recanting witness to be illusory since he could not recall the prior statement.  Simmons J.A. found that D.S. could recall portions of the ASF and, in fact, the three elements crucial to the Crown’s case against Youvarajah.

[87]                          Second, Simmons J.A. found that the trial judge erred by unduly focusing on the B. (K.G.) recommended safeguards while failing to consider other factors relevant to the reliability analysis.  She found that the lack of “presence” at the time of the prior statement, such as in the form of a video, was irrelevant due to D.S.’s minimal involvement in the proceeding and that the solemn circumstances of D.S.’s guilty plea added to the reliability of the statement.  Simmons J.A. also held that it could be inferred that D.S. was warned about the penal consequences of recanting since experienced defence counsel was involved in the preparation and presentation of the ASF. 

[88]                          Third, Simmons J.A. briefly addressed the issue of inherent motive for D.S. to lie in the ASF.  She found that this was an issue for the trial judge to address at the new trial if the other indicia considered were not sufficient to meet the standard of threshold reliability.

III.  Issues

[89]                          The issue before this Court is simply whether the ASF was sufficiently reliable to be admitted into evidence for the truth of its contents under the principled approach to hearsay based on necessity and reliability.  Since necessity is not at issue, the focus of the analysis that follows will be on the reliability of the ASF.  The reliability analysis raises a series of sub-issues on the facts before this Court:

1.         Did the trial judge err in holding that the opportunity to cross-examine D.S. was illusory as a result of his invoking solicitor-client privilege?

 

2.         Did the nature of the ASF and the circumstances under which it was prepared and presented provide sufficient indicia of reliability?

IV.    Analysis

A.     Overview of Admissibility of Prior Inconsistent Statements Under the Principled Approach

[90]                          The principled approach to the admissibility of hearsay evidence was adopted by this Court in B. (K.G.). Prior to B. (K.G.), the use of prior inconsistent statements, or any form of hearsay evidence, was restricted to impeaching witness credibility.  This restriction was to ensure that untested and unreliable evidence that could lead to an unfair verdict was excluded and that litigants were provided with the opportunity to confront witnesses who were adverse to their position. 

[91]                          In certain circumstances, hearsay evidence presents minimal dangers to trial fairness.  In those situations, its exclusion would be more detrimental to the fact-finding function of the court than its admission.  By endorsing the principled approach in B. (K.G.), this Court sought to strike a better balance between trial fairness and the truth-seeking function of the judicial process.  Under the principled approach, hearsay statements remain presumptively inadmissible, but may be admitted where necessity and reliability are sufficiently demonstrated.

[92]                          Necessity arises from the fact that important evidence leading to the truth would otherwise be lost when a witness recants.  This aspect of the principled approach requires no further analysis on these facts.

[93]                          The reliability requirement is aimed at identifying cases where the dangers of hearsay evidence may be overcome.  Indeed, in R. v. Khelawon, 2006 SCC 57, [2006] 2 S.C.R. 787, at paras. 62-63, Charron J. indicated that the reliability requirement will generally be met either (1) by showing that there is no real concern about a statement’s truth owing to the circumstances in which it arose; or (2) by showing that there is no real danger that the statement is made in hearsay form since, in the circumstances, its reliability can be sufficiently tested by means other than contemporaneous cross-examination and presence.  These categories are not mutually exclusive and merely provide guidance in identifying the factors to be considered in an admissibility inquiry.

[94]                          In B. (K.G.), this Court addressed the presence of adequate substitutes for traditional safeguards relied upon to test hearsay evidence. The admissibility question in B. (K.G.) dealt with the prior inconsistent statements made by the accused’s friends in which they told police that he was responsible for stabbing and killing the victim.  The three friends recanted at trial and the Crown sought to have their prior statements admitted for their truth.  While a new trial was ordered to determine the admissibility of the prior inconsistent statements, this Court outlined the principled approach to be used in the analysis.

[95]                          Lamer C.J. indicated that sufficient circumstantial guarantees of reliability could be found where three indicia were satisfied: if (1) the statement was made under oath, solemn affirmation or solemn declaration following an explicit warning as to the criminal consequences of making false statements; (2) the entire statement was videotaped; and (3) the opposing party has an opportunity to cross-examine the witness at trial.

[96]                          Importantly, Lamer C.J. left the door open for alternative indicia of reliability, stating, at p. 796:

Alternatively, other circumstantial guarantees of reliability may suffice to render such statements substantively admissible, provided that the judge is satisfied that the circumstances provide adequate assurances of reliability in place of those which the hearsay rule traditionally requires.

[97]                          Indeed, subsequent cases have found hearsay evidence to be admissible even when all three of the indicia from B. (K.G.) are not present.  R. v. U. (F.J.), [1995] 3 S.C.R. 764, was such a case.  In that case, the accused was arrested for reported sexual activity with his 13-year-old daughter.  During separate police interviews, both the accused and his daughter provided matching and specific details of the same sexual activities and occurrences, including having had intercourse the previous night.  The issue on appeal was whether the trial judge erred in allowing the jury to compare the accused’s unadopted statement with his daughter’s unadopted prior inconsistent statement.  Despite the absence of an oath and videotaped statement, this Court found that the reliability threshold was met.  In addition to the ability to cross-examine the accused, the Court found reliability in the “strikingly similar” statements provided by the accused and his daughter.

[98]                          In Khelawon, this Court further clarified that B. (K.G.) was never intended to create categorical exceptions.  Rather, admissibility was to be determined using the principled approach on a case-by-case basis, allowing the indicia of oath, presence (or video), and cross-examination to provide guidance in the reliability assessment as opposed to rigid requirements.

[99]                          Regardless of the circumstances of a particular case and the indicia that may be available for the reliability analysis, it is crucial not to lose sight of the truth-seeking function of an admissibility inquiry under the principled approach.  In the instance of a prior inconsistent statement, the purpose of the admissibility inquiry is not to make an absolute determination of the reliability of the hearsay statement as compared with the subsequent testimony.  Rather, the inquiry is intended to ensure that the statement is sufficiently reliable in order for the trier of fact to assign the appropriate weight to the evidence. As Charron J. states in Khelawon, the “general exclusionary rule is a recognition of the difficulty for a trier of fact to assess what weight, if any, is to be given to a statement made by a person who has not been seen or heard, and who has not been subject to the test of cross-examination” (para. 35).

[100]                      This principle is articulated in B. (K.G.) where this Court draws the distinction between threshold reliability and ultimate reliability, stating:

The ultimate reliability of the statement and the weight to be attached to it remain, as with all evidence, determinations for the trier of fact.  What the reliability component of the principled approach to hearsay exceptions addresses is a threshold of reliability, rather than ultimate or certain reliability. [p. 787]

[101]                      As Charron J. stresses in Khelawon, at paras. 92-93, this distinction does not mean that relevant factors should be categorized in terms of threshold and ultimate reliability.  Rather, a more functional approach should be adopted by which the context will determine whether certain factors go to threshold reliability.  The focus of the inquiry should be on the dangers of hearsay evidence and the circumstances relied upon to overcome those dangers.  However, “it is crucial to the integrity of the fact-finding process that the question of ultimate reliability not be pre-determined on the admissibility voir dire” (para. 93).

[102]                      Given the truth-seeking function of the judicial process, I am of the view that threshold reliability should be generously interpreted in the admissibility inquiry.  Trial fairness and protection against the dangers of hearsay are not sacrificed where the trier of fact is ultimately able to make reasonable determinations of the weight to be accorded to evidence before the court.  The reliability inquiry does not need to ensure absolute reliability.  Rather, the circumstances surrounding the hearsay evidence must provide a sufficient basis for the trier of fact to assess and assign the appropriate weight and eventually determine ultimate reliability.

B.     Application to the Case at Hand

[103]                      As the following analysis will disclose, the trial judge erred in his determination that the ASF did not meet the standard of threshold reliability.  That is not to say that the circumstances in this case provide sufficient evidence to conclusively determine that the ASF should be admitted. Rather, the admissibility inquiry was incomplete and did not provide a proper basis for the trial judge to exclude the relevant evidence.  As articulated by the Ontario Court of Appeal, the issue of admissibility should be addressed at a new trial where threshold reliability can be properly assessed.

                    (1)     The Nature of the Hearsay Evidence in Question

[104]                      Not all out-of-court statements are created equal.  Indeed, this idea is captured by the principle from Khelawon that hearsay evidence will generally be admissible under the principled approach where there is either inherent reliability or circumstances that allow reliability to be tested by the trier of fact.   While much of this analysis hinges on the factors that can reduce the dangers of hearsay, the nature of the hearsay statement before this Court cannot be ignored.

[105]                      Much of the jurisprudence pertaining to the admission of prior inconsistent statements deals with statements made to police (see Khelawon; B. (K.G.); U. (F.J.); R. v. Couture, 2007 SCC 28, [2007] 2 S.C.R. 517; R. v. Trieu (2005), 195 C.C.C. (3d) 373 (Ont. C.A.)) and to a lesser degree with out-of-court conversations (see R. v. Blackman, 2008 SCC 37, [2008] 2 S.C.R. 298) or statements made at preliminary hearings for the trial where admissibility is in question (see R. v. Hawkins, [1996] 3 S.C.R. 1043). 

[106]                      The hearsay statement at issue in this case is an ASF which was drafted with the contribution and approval of Crown counsel, defence counsel and the accused.  More importantly, it was accepted by a court in the sentencing of a youth for the very serious crime of second degree murder. While the ASF and guilty plea proceedings did not include an explicit indication by the Crown that it consented to a lesser charge against D.S. because of his anticipated subsequent cooperation in proceedings against Youvarajah, it is certainly implicit in the nature of the plea agreement and the facts therein.  Accordingly, the ASF was used in some capacity to reduce the sentence of D.S. for a very serious crime. 

[107]                      The characteristics of the ASF in this case are somewhat unique when compared to other statements in previous case law. Statements to police and out-of-court conversations are not generally admitted as evidence in proceedings that may limit another individual’s freedom through sentencing, or that impact on society’s interest in deterrence, rehabilitation and retribution, not to mention public protection from dangerous offenders.

[108]                      An ASF presented in court carries an inherent level of reliability that the judicial system accepts and, in fact, implicitly relies upon in the workings of the administration of justice.  It follows that not allowing an ASF to be admitted for its truth in a subsequent hearing, to be weighed and evaluated by the trier of fact, is problematic when viewed through the wider lens of the repute of the system of administration of justice.  What does it say about the administration of justice if the courts accept the reliability of a statement to convict an individual and to vary the length of his sentence for a crime as serious as murder, but are unwilling to place that same statement before the trier of fact in another proceeding to be weighed and considered against a recantation of that statement?

[109]                      As a general rule, if a statement is accepted for its truth by the courts, and used to balance the liberty interests of the accused with societal considerations such as deterrence and retribution, then that statement provides a level of implicit reliability that warrants consideration in the admissibility inquiry in a subsequent third-party trial.  That is not to say that a statement’s use and judicial acceptance at a prior proceeding is sufficient in and of itself to meet the reliability requirement of the principled approach.  Rather, it is merely a factor that must be considered in the admissibility inquiry, a factor that goes towards meeting the threshold reliability standard.

[110]                      Certainly, there are potential dangers that could arise with the use of such statements at subsequent trials.  Indeed, these concerns are captured in a line of Ontario cases that has suggested that care should be taken when considering the admissibility of a statement for its truth under the principled approach when it is read in at a witness’s earlier guilty plea (see R. v. Tran, 2010 ONCA 471, 103 O.R. (3d) 131; R. v. D.P., 2010 ONCA 563, 268 O.A.C. 118; and R. v. McGee, 2009 CanLII 60789 (Ont. S.C.J.)).  There is a whole host of reasons for which an accused could provide untruthful information in the context of a guilty plea, including a reduced sentence, expediency of multiple proceedings, retribution against the target of a subsequent trial through cooperative testimony, or even the lack of understanding of the meaning of the statement or of the charges laid against the individual. 

[111]                      However, these potential dangers do not apply in every case and the resulting blanket assertions of unreliability do not take account of the nuanced approach that should take stock of factors that may add to or detract from the reliability inquiry in the particular context of the statement in question.  While these cases correctly highlight some of the reliability concerns with this type of statement, they do not address some of the potential indicia of reliability that may be found in certain circumstances.  It requires little imagination to envision circumstances where making a statement against friends, family, or professional acquaintances could be socially or even physically detrimental to a recanting witness. These social or physical ramifications against the interests of the witness provide an indication of reliability that may be weighed against any motive to lie. But these are considerations that must be borne out by the evidence and by the unique circumstances of each case and may not even be necessary to explore in depth at this stage of the analysis.  Certainly, motive need not be considered if reliability can be established by other means.  Nevertheless, motive is something that can provide context to the threshold reliability analysis.  

[112]                       I also do not accept the argument that only the essential elements of the crime detailed in the ASF are accepted for their truth by the courts in a guilty plea.  The implication of this proposition is that the courts are willing to accept a statement that is only partially true so long as it provides the judiciary with the necessary elements to convict an individual. Beyond the obvious concerns of threatening individuals’ liberty or reducing sentencing based on documents that are deemed to be at least partially unreliable or untrue, this creates an arbitrary distinction about which aspects of a guilty plea are accepted for their truth.

[113]                      Certainly an argument could be made that the reliability of a guilty plea rests on the admission against the interests of the accused. However, as I have stated above, what constitutes a statement against the interests of the accused may vary from case to case depending on the circumstances.  Indeed, an individual may place his safety or even his life in jeopardy by implicating himself in a serious crime.  This factor should be considered on the circumstances of each case.  A rigid rule finding that all third-party guilty pleas are unreliable is potentially detrimental to the truth-finding function of the judicial process.

[114]                      Further, accepting that only the essential elements of a guilty plea need to be accepted for their truth ignores the interrelated nature of a guilty plea and sentencing.  Factors that do not form part of the essential elements of a crime may nonetheless prove to be pertinent when considering mitigating or aggravating factors in sentencing.  This is particularly true where the trial judge is presented with a joint submission by the Crown and the accused concerning sentencing and must determine whether the recommendation is appropriate within the entire context of the crime in question.

[115]                      Looking at the case before this Court, the three crucial facts from the ASF which the Crown relies on are certainly not essential elements of D.S.’s crime.  The source of the gun, D.S.’s reasons for shooting Freake, and the manner in which the gun was disposed of do not form part of the actus reus or mens rea relating to D.S.’s crime.  Nonetheless, these factors are relevant to motive and the degree of planning and deliberation, all of which are properly considered during sentencing. Allowing a judge to consider these factors in sentencing, while finding that they are not sufficiently reliable to even be weighed by the trier of fact in another case, allows for an inconsistent application of the reliability of evidence and brings the administration of justice into question.

[116]                      In summary, when dealing with threshold reliability, the fact that courts accept the statements read in on a guilty plea when considering convictions and sentencing is a factor that points to the reliability of the prior inconsistent statement, even when such statement implicates a third party.  It would not be prudent to make blanket conclusions about the dangers associated with this sort of statement, nor about its inherent reliability.  Each case will have to be examined on its particular circumstances when determining threshold reliability, but consideration should certainly be given to the general acceptance of and reliance on statements read in as part of a guilty plea.

[117]                      While the nature of the statement in question warrants consideration, it is still necessary in the instant case to determine whether the ASF meets the requirements of necessity and reliability under the principled approach.  Since necessity is not at issue, the balance of my analysis will explore whether the threshold of reliability is met.

                    (2)      Reliability

 

[118]                      The essential question before this Court is whether the trial judge erred in finding that the ASF was not sufficiently reliable to be admitted for its truth.  For the reasons that follow, I agree with the conclusion of the Ontario Court of Appeal that the trial judge failed to adequately assess the available indicia of reliability and erroneously found the opportunity to cross-examine D.S. to be illusory.  While the ASF may in fact prove to be too unreliable to be admitted, the trial judge’s analysis was insufficient to permit him to come to that conclusion.  As highlighted by Charron J. in Khelawon, the most important contextual factor in the reliability inquiry is the availability of the declarant for cross-examination.  Accordingly, my analysis will begin there.

                    (a)      Cross-Examination

 

[119]                      The opportunity to cross-examine the declarant in relation to his prior inconsistent statement warrants strong consideration in a reliability analysis.  While the concerns of hearsay evidence still exist even where the declarant is available at trial, cross-examination can go a long way towards alleviating those concerns.  As Lamer C.J. noted in B. (K.G.), “the focus of the inquiry in the case of prior inconsistent statements is on the comparative reliability of the prior statement and the testimony offered at trial” (p. 787).  The availability of cross-examination allows the opposing party to test the current version and simultaneously address the declarant’s initial statement.  As Charron J. states in Khelawon, at para. 41:      

         Hence, although the underlying rationale for the general exclusionary rule may not be as obvious when the declarant is available to testify, it is the same — the difficulty of testing the reliability of the out-of-court statement.  The difficulty of assessing W’s out-of-court statement is the reason why it falls within the definition of hearsay and is subject to the general exclusionary rule.  As one may readily appreciate, however, the degree of difficulty may be substantially alleviated in cases where the declarant is available for cross-examination on the earlier statement, particularly where an accurate record of the statement can be tendered in evidence.  I will come back to that point later.  My point here is simply to explain why, by definition, hearsay extends to out-of-court statements tendered for their truth even when the declarant is before the court. [Emphasis added.]

[120]                      The issue before this Court is whether D.S.’s assertion of solicitor-client privilege led to the opportunity for cross-examination becoming so illusory that it could not be used to establish reliability.  For the reasons that follow, I agree with the Ontario Court of Appeal’s finding that the trial judge erred in finding the opportunity for cross-examination to be illusory.

[121]                      The first error committed by the trial judge was to overextend the breadth of solicitor-client privilege and its impact on the ability to cross-examine D.S.  The trial judge found that solicitor-client privilege precluded questioning D.S. and his former counsel about the circumstances of the guilty plea, the drafting of the ASF, and D.S.’s implication of Youvarajah as the person who provided him the gun and directed him to do the shooting.  While I am loath to provide a road map for the Crown to follow at a new trial, the Ontario Court of Appeal was correct in finding that the opportunity to cross-examine D.S. was not entirely closed and that the trial judge misapprehended the scope and impact of solicitor-client privilege on the opportunity for cross-examination. 

[122]                      Solicitor-client privilege is intended to prohibit the disclosure of any form of communication made in confidence between a lawyer and his client for the legitimate purpose of obtaining lawful professional advice or assistance, unless the privilege is waived by the client or a recognized exception to the privilege applies.  This definition is captured succinctly by J. H. Wigmore in Evidence in Trials at Common Law (McNaughton rev. 1961), vol. 8, at p. 554, and adopted by this Court in R. v. McClure, 2001 SCC 14, [2001] 1 S.C.R. 445, at para. 36:

                    Where legal advice of any kind is sought from a professional legal adviser in his capacity as such, the communications relating to that purpose, made in confidence by the client, are at his instance permanently protected from disclosure by himself or by the legal adviser, except the protection be waived.

[123]                      This definition does not encompass all communications between a lawyer and his client.  More importantly to the admissibility inquiry in this case, solicitor-client privilege does not foreclose cross-examining D.S. in relation to a number of areas concerning the ASF.  Specifically, the Ontario Court of Appeal identified six areas about which D.S. could have been cross-examined, had the trial judge not misunderstood the scope of solicitor-client privilege:

(1)               D.S.’s understanding of the ASF;

 

(2)               D.S.’s understanding of any conditions that were attached to the plea deal;

 

(3)               D.S.’s understanding of the consequences of misleading the police, the court and the Crown;

 

(4)               D.S.’s motive for making what he now asserts were false statements;

 

(5)               D.S.’s instructions to counsel regarding communications to the court and the Crown; and

 

(6)               D.S.’s knowledge of the Crown’s theory at the time he provided counsel with instructions.

[124]                      Admittedly, these six areas of potentially permissible lines of questioning could eventually be curtailed by issues of solicitor-client privilege.  However, that is an issue for the trial judge to address as those possible issues arise.  In any event, there is certainly the possibility of cross-examination in relation to these six areas.  Indeed, both defence counsel and Crown counsel cross-examined D.S. on some of these very issues, eliciting information that could have been relevant in the reliability assessment.

[125]                      The Crown elicited evidence from D.S. regarding his understanding of the ASF.  Specifically, D.S. acknowledged that at his guilty plea he agreed that “Yousanthan Youvarajah gave [D.S.] a loaded small black handgun” and that he understood what that meant.  Further, the Crown read the following statement to D.S. from the ASF:  “When he was given the gun, [D.S.] was told by Yousanthan Youvarajah to shoot Mr. Freake sometime during the course of the drug rip off” (A.R., vol. V, at pp. 62-63). During the Crown’s cross-examination, D.S. admitted that he remembered the statement being read at his guilty plea and that he agreed to those facts through his counsel.

[126]                      Similarly, defence counsel was able to elicit substantial information regarding the motive behind D.S.’s guilty plea and the ASF.  Specifically, during defence counsel’s cross-examination, D.S. conceded that a factor in his decision to plead guilty was his impression that he would not have to provide further statements.  He also agreed to defence counsel’s assertion that D.S. signed an ASF that he knew contained false information to “get to the end game that day which was pleading to second degree murder” (A.R., vol. III, at p. 92).  Defence counsel was also able to elicit D.S.’s assertions that he had never implicated Youvarajah in the killing until his guilty plea on September 17, 2009.  

[127]                      These examples are merely illustrative of the information that was potentially available on cross-examination without breaching D.S.’s asserted solicitor-client privilege.  Had the trial judge permitted the cross-examination to proceed and addressed issues involving solicitor-client privilege as they arose, information sufficient to meet the threshold reliability standard may have been brought forward. 

[128]                      The appellant claims that any cross-examination on the six aforementioned areas would either skirt the critical issues or be entirely blocked by solicitor-client privilege.  I cannot agree with this position.  The appellant’s position presupposes that D.S.’s recantation arose from direct or indirect communications with counsel.  While this is certainly possible, and maybe even likely, an admissibility inquiry should not rest on assumptions.  Further, and as I have already alluded to, limits on cross-examination should not be applied rigidly and absolutely where solicitor-client privilege is asserted.  Rather, issues of privilege should be addressed as they arise on cross-examination.  This way, relevant information that may not be prohibited by the assertion of privilege may still be elicited, furthering the underlying truth-seeking function of the proceedings. 

[129]                      While the issue is not before this Court, it bears mentioning that this approach is not an invitation for counsel to attempt to skirt privilege by inducing the waiver of privilege by an unsuspecting witness who has clearly asserted solicitor-client privilege.  Solicitor-client privilege is a fundamental principle of the Canadian legal system.  Any conduct that unscrupulously attempts to erode solicitor-client privilege should not be encouraged.  While solicitor-client privilege is not absolute and circumstances may arise where it will be subordinated, duping a witness into waiving privilege is never an acceptable practice.

[130]                      Given that there were several subjects about which D.S. could have been cross-examined and that this could have potentially assisted in establishing reliability, the trial judge erred in determining that cross-examination was effectively foreclosed by the claim of solicitor-client privilege. 

[131]                      The appellant argues that the mistaken view of the scope of solicitor-client privilege rested solely on the shoulders of Crown counsel and that the trial judge cannot be held accountable for the tactical decisions made by counsel.  I cannot agree with this argument, given the facts before this Court.  Crown counsel certainly misunderstood the implications of D.S.’s asserted privilege, but the transcripts provide clear indication that the mistake was shared.  Indeed, defence counsel was the first to state that any cross-examination was meaningfully foreclosed once D.S. asserted privilege, a position that, coincidentally or not, was advantageous to the accused’s case.  Crown counsel seemingly concurred by indicating that he did not anticipate calling further witnesses after the privilege was asserted.  The trial judge did not correct these erroneous assumptions.  Rather, as Simmons J.A. correctly notes, “the trial judge put his stamp of approval on this approach — and effectively prevented the Crown from reconsidering calling D.S.’s counsel” (para. 95).  This is evident from the following exchange:

[CROWN COUNSEL]:  And I don’t anticipate even now given the comments of Mr. Marentette I don’t anticipate needing to call further witnesses.

 

THE COURT:  Well, no legal witnesses that have ever spoken to this man anyway.

 

[CROWN COUNSEL]:  Right.

 

THE COURT:  Right.  I mean, we’re going to take Mr. Marentette’s word as the final word on that.

 

[CROWN COUNSEL]:  Right. [A.R., vol. II, at p. 132]

[132]                      The trial judge’s misapprehensions were also evident when he stated: “Well, you bump into the solicitor/client . . . privilege every which way here” (A.R., vol. III, at p. 1).  Perhaps most telling was the following excerpt from the trial judge’s reasons where he summarized the impact of solicitor-client privilege without having heard any argument on the matter:

Solicitor/client privilege precluded questions to or from [D.S.] or indeed of his counsel about how it came to be that this confession from [D.S.] as the shooter and his implication of the accused as the person who provided him with the gun and directed him to do the shooting was made on the day of his guilty plea almost two years after his arrest when he had never confessed or implicated Yousanthan Youvarajah in all that time. [A.R., vol. I, at p. 59]

[133]                      The appellant is correct in asserting that a judge must remain an impartial arbiter and refrain from intervening where parties do not properly raise evidentiary issues (see R. v. S.G.T., 2010 SCC 20, [2010] 1 S.C.R. 688, at para. 36).  However, it cannot be said that the trial judge in the present case remained impartial by weighing in with a clear expression of prohibition.  The appellant’s argument would be more compelling had the trial judge remained silent on the matter and merely allowed Crown counsel to proceed, but this was clearly not the case.  At the very least, the trial judge’s remarks improperly influenced the conduct of counsel.

[134]                      While the appellant did not present the argument before this Court that the instant case was analogous to the situation in Conway, a brief discussion of that case is warranted to support the conclusion that the opportunity to cross-examine D.S. was not illusory. 

[135]                      In Conway, two men were charged with the second degree murder of an elderly disabled man while committing a robbery.  They pleaded not guilty to murder charges but guilty to the lesser charge of manslaughter on the grounds that they did not have the requisite mens rea, having ingested intoxicating substances prior to committing the offence.  One witness provided a written statement to the police denying knowledge of the killing and provided a subsequent written statement implicating the accused.  This statement had none of the indicia of reliability outlined in B. (K.G.).  At trial, the witness could not recall providing either written statement to police.  In finding the second written statement inadmissible for its truth, Labrosse J.A. found that the opportunity to cross-examine a witness who recanted and denied all recollection of the prior inconsistent statement was illusory. 

[136]                      The important distinction to be made between that matter and the present one is that two versions of the witness’s story could not be said to be before the trier of fact.  The evidence given by the witness amounted to a declaration of “I don’t remember” (p. 589).  As Labrosse J.A. stresses, the value of cross-examining a recanting witness lies in assessing the comparative reliability of the two versions.  Where both versions cannot be tested by the trier of fact, this cannot be accomplished.   

[137]                      In the instant case, the potential cross-examination of D.S. as to his statements does not suffer from the same deficiency as that encountered in Conway.  Simmons J.A. correctly states that while D.S. initially claimed not to recall signing the ASF, he later confirmed that he remembered a number of events set out within the document.  Indeed, Simmons J.A. astutely points out, at para. 74, that the trial judge acknowledged this very fact when he stated:  “When cross-examined by the Crown, [D.S.] admitted making many of the statements in the [ASF] and conceded that many of those statements were true” (A.R., vol. I, at pp. 49-50).

[138]                      So while it is possible for a recanting witness’s cross-examination to prove too illusory to ground a reliability analysis, this may not be the situation in the present case.  D.S. was available for cross-examination, he recalled having made many of the statements within the ASF, he did not dispute understanding the key facts in the ASF which he subsequently denied at Youvarajah’s trial, and he recalled the ASF being read in at his guilty plea proceeding.  As a result, Simmons J.A. correctly summarized that the trial judge erred by failing to assess “whether it could properly be said that two versions of the events were before the court; and whether the extent of D.S.’s recollection would afford sufficient scope to assist the jury in assessing which, if either, version was true” (para. 79).

[139]                      Accordingly, the trial judge erred in precluding a cross-examination which possibly could have gone towards establishing the threshold reliability of the ASF.  Given the importance of cross-examination in the admissibility inquiry, this error is sufficient in and of itself to order a new trial.  However, in the interest of completeness and of providing guidance for the new trial, I will briefly address the errors of the trial judge with regard to other indicia of reliability. 

                    (b)      Other Indicia of Reliability

[140]                      The two other indicia of reliability detailed in B. (K.G.), namely a statement made under oath and captured by video, were not present in the context of the ASF in this matter.  However, the inquiry does not end there.  A reliability analysis is not restricted to the three indicia detailed in B. (K.G.).  In fact, as stated by Lamer C.J. in B. (K.G.) itself, “other circumstantial guarantees of reliability may suffice to render such statements substantively admissible” (p. 796).  This was clearly understood by the Ontario Court of Appeal in this case where Simmons J.A. stated that “the focus of the threshold reliability inquiry can change depending on the circumstances of the case” (para. 123). I also agree with the Ontario Court of Appeal’s finding that other factors were available in this case that could have satisfied the trial judge that the circumstances surrounding the drafting and reading in of the ASF provided assurances of reliability equivalent to an oath and presence, in the form of video.  If these indicia had been considered and cross-examination had been permitted, it is possible that threshold reliability would have been met.   

[141]                      In the instant case, perhaps the most persuasive indicia of reliability are the nature of the statement and how it was constructed.  As quoted above, in Khelawon, at para. 41, Charron J. indicated that where the declarant is available for cross-examination, an accurate record of the statement can do much to alleviate the dangers of hearsay.

[142]                      The ASF tendered as evidence in D.S.’s guilty plea proceeding provides just such a record.  Having been reduced to writing, signed not only by D.S. but by both Crown and defence counsel, and having been read in to the record in the context of D.S.’s guilty plea, the source of the document cannot be disputed. 

[143]                      The nature of how the ASF was constructed further supports its reliability.  This was not a statement that was unilaterally drafted and forced upon D.S. to adopt.  Rather, the drafting of the ASF was a collaborative effort where there is evidence that at least one crucial component to the Crown’s case against Youvarajah was supplied by defence counsel — namely the statement that Youvarajah supplied D.S. with the gun. 

[144]                      Moreover, D.S. had the opportunity to review the ASF with his counsel prior to signing it and adopting it.  This is an important aspect as it refutes any negative inferences that could be drawn against the ASF on the grounds that it was not captured in D.S.’s own words or that D.S. did not understand its contents. 

[145]                      The appellant correctly argues that a lawyer’s ethical obligation not to knowingly present false statements to the court does not ensure the truth of the contents of the ASF.  However, it does provide some assurances that D.S. was advised of the consequences of providing a false statement.  Before entering a guilty plea, counsel has an ethical obligation under rules of professional conduct to ensure that the client (1) is voluntarily admitting the elements of the offence; and (2) understands the implications and possible consequences of a guilty plea, particularly with regards to the authority and discretion of the court.  In the absence of evidence to the contrary in the instant case, it is not unreasonable to assume that counsel would have advised D.S. about the contents of the ASF, the potential consequences of implicating the appellant, and the potential penal consequences of subsequently recanting.  This assumption is also supported on the record by D.S.’s admission that he understood that the basis of his guilty plea was the ASF.

[146]                      Perhaps most importantly, D.S. testified that he understood the three crucial components of the ASF, namely that the appellant gave D.S. the gun, told him to shoot Freake, and demanded that D.S. return the gun.  Without more persuasive evidence to the contrary, this essentially forecloses any argument that D.S. lacked understanding of the ASF’s contents.

[147]                      Looking beyond the construction of the ASF, the context of the proceedings under which it was read in to court and the contents of the ASF itself provide additional indicia supporting threshold reliability.  As the Ontario Court of Appeal stated, “the solemnity of the occasion on which the [ASF] was presented” was a feature that favoured threshold reliability (para. 137).  It certainly would have been ideal for the ASF to have been adopted under oath or solemn affirmation, but the nature of the guilty plea can help fill that gap when considered with the other indicia of reliability before the court.

[148]                      The Ontario Court of Appeal also correctly rejected the trial judge’s assertion that the lack of spontaneity of the statement and the fact that it was not in D.S.’s own words detracted from the reliability of the ASF.  While spontaneity can be an indication of reliability, its absence does not necessarily detract from it.  With regards to an ASF prepared in the context of a guilty plea, reliability is not found through spontaneity but through the carefully crafted language which is subsequently adopted by the declarant after careful consideration with his counsel.

[149]                      Finally, the Ontario Court of Appeal concluded that the lack of “presence” when the ASF was read in at the guilty plea, for example by way of video, was of no consequence since D.S.’s only contribution was to reply “yes” in response to two separate questions by counsel (para. 128).  I respectfully disagree with this assessment.  It is not appropriate to dismiss presence as indicia of reliability merely because of the declarant’s limited level of involvement or because of the brevity of his statement.  Had there been video presence, D.S.’s demeanour could have been assessed throughout the reading in of the ASF as well as in his answers to the two questions posed to him.  Even if one were to focus solely on D.S.’s two short responses, such presence could provide invaluable insight into D.S.’s demeanour and ultimately into the veracity of the statement.

[150]                      While I cannot agree with the Ontario Court of Appeal’s assessment on the issue of presence, I do not think that lack of presence is detrimental to potentially establishing reliability.  If the cross-examination is found sufficiently meaningful and other indicia support the reliability of the ASF, presence may be superfluous.  As I have detailed above, the reliability indicia from B. (K.G.) are not rigid and the absence of one factor can be overcome if other factors are sufficient in establishing threshold reliability.  It is not necessary for me to draw any firm conclusion in this regard in the instant matter, and this is a question best left for a new trial where all of the factors can be assessed in concert.

V.     Conclusion

[151]                      While there are certainly concerns with admitting evidence from a guilty plea proceeding at the trial of a third party, I find it unpalatable to apply a blanket prohibition at the admissibility inquiry stage.  A proper admissibility inquiry is only intended to determine whether the prior inconsistent statement has sufficient markers of reliability to allow the trier of fact to reasonably assess the appropriate weight it should be given.  This is a contextual question that will change with the particular circumstances of each case.

[152]                      Based on my analysis above, I do not think that a proper analysis was conducted in this case — particularly in relation to the opportunity for cross-examination and the consideration of other potential indicia of reliability.  Accordingly, the decision of the Ontario Court of Appeal should be upheld and a new trial ordered.  Reliability may be properly assessed at the new trial and the evidence may be appropriately admitted or excluded upon consideration of all the relevant factors.

[153]                      For these reasons, I would dismiss the appeal.

                    Appeal allowed, Rothstein and Wagner JJ. dissenting.

                    Solicitors for the appellant:  Lockyer Campbell Posner, Toronto; Dawe & Dineen, Toronto.

                    Solicitor for the respondent:  Attorney General of Ontario, Toronto.

                    Solicitors for the intervener:  Henein Hutchison, Toronto.