Bastarache J. took no part in the judgment.
This appeal concerns an accused’s motion for production of police disciplinary records and criminal investigation files relating to the Crown’s main police witness in the case against him. The respondent Lawrence McNeil brought his motion in accordance with the procedure set out in R v O’Connor,  4 S.C.R. 411, for the production of third party records. The proceedings against McNeil have since been discontinued, McNeil has withdrawn his participation in this appeal, and the production issue is now moot.
Issues concerning the production of police disciplinary records and criminal investigation files relating to third party accused occur frequently and, because the proceedings in which these questions generally arise are interlocutory in nature, production orders are often evasive of appellate review. In addition, the third parties in the present case have a subsisting interest in clarifying some of the uncertainties arising from the decision below. For these reasons, the Court appointed an amicus curiae and heard this appeal despite its mootness.
The question of whether production of the particular documents at issue in the court below should or should not have been ordered is no longer before us. The evidence and prior proceedings in the current case will therefore be reviewed simply to provide the necessary background for discussion of the issues canvassed by the parties on this appeal.
McNeil was arrested by Constable Rodney Hackett and other members of the Barrie Police Service in respect of an alleged drug transaction. He was subsequently prosecuted by the federal Crown and convicted on multiple drug charges, including possession of marijuana and cocaine for the purpose of trafficking. Hackett was the Crown’s main witness in the case against McNeil. He was the only witness who testified to the reasonable grounds supporting McNeil’s arrest. In addition, the trial judge’s ultimate finding that McNeil’s admitted possession of marijuana and cocaine was for the purpose of trafficking turned on Hackett’s credibility.
After his conviction but before sentencing, McNeil learned that Hackett was engaged in drug-related misconduct that had led to both internal disciplinary proceedings under the Ontario Police Services Act, R.S.O. 1990, c. P.15 and to criminal charges. Both proceedings against Hackett were ongoing at the time. Following an aborted application to reopen his trial to introduce evidence about Hackett’s misconduct, McNeil chose to proceed to sentence and appeal his conviction instead.
In a preliminary motion before the Court of Appeal for Ontario, McNeil sought production of all documents related to Hackett’s misconduct, claiming that he required this material to assist him in preparing an application to introduce fresh evidence on his appeal from conviction. The police disciplinary and criminal investigation documents at issue were intermingled, and were in the possession of both the Barrie Police Service and the provincial Crown prosecuting the criminal charges against Hackett. Both entities resisted production, and the federal Crown supported their opposition to the motion.
In O’Connor, this Court set out a two-part test for production of third party records. First, the applicant must demonstrate that the information contained in the records is likely relevant. In the appellate context, it was therefore incumbent on McNeil to show that the targeted documents were likely relevant to his proposed application to introduce fresh evidence on his appeal from conviction. Second, if the threshold test of likely relevance is met, the court may order production of the records for its inspection. With the targeted documents before it, the court weighs “the positive and negative consequences of production, with a view to determining whether, and to what extent, production should be ordered” (O’Connor, at para. 137). The second part of the O’Connor test essentially requires a court to conduct a balancing of the third party’s privacy interest in the targeted documents, if any, and the accused’s interest in making full answer and defence.
In the present case, the Court of Appeal found that some of the targeted records, including both disciplinary police records and criminal investigation files, met the requisite relevancy threshold, satisfying the first part of the O’Connor test. Rather than proceeding to the second part of the test, however, the court held that an “O’Connor-type procedure” is only required in cases where third party records attract a reasonable expectation of privacy ((2006), 218 O.A.C. 1). The court drew a distinction between criminal investigation files and police disciplinary records, and concluded that no expectation of privacy existed in respect of the former. Accordingly, subject to appropriate redactions and the resolution of any privilege claims, the court ordered the third parties to produce the criminal investigation files in their possession related to the charges against Hackett to the federal Crown prosecuting McNeil’s case. The court directed the prosecuting Crown in McNeil’s case to determine whether the documents were in fact relevant to McNeil’s appeal and, if so, to make disclosure to the accused. As for the police disciplinary records, the court noted the conflicting jurisprudence on whether such records are subject to a reasonable expectation of privacy and invited counsel to make further submissions on this issue. The balance of the motion was adjourned accordingly.
The Attorney General of Ontario, as third party record holder, was granted leave to appeal to this Court and the production order was stayed pending disposition of the appeal. Subsequently, Hackett pleaded guilty to one of the criminal charges brought against him. Evidence of Hackett’s conviction was admitted on McNeil’s appeal before the Court of Appeal for Ontario, and his convictions were set aside, following which the Crown undertook not to re-prosecute him. McNeil then withdrew his participation in the appeal. This Court acceded to the remaining parties’ request to proceed with the appeal despite its mootness and appointed an amicus curiae to maintain a proper adversarial context.
3. OVERVIEW OF THE ISSUES ON APPEAL
As stated earlier, the third party record holders in this case seek clarification in respect of some of the uncertainties arising from the decision below. The issues canvassed before this Court are the following.
First, by limiting the applicability of the O’Connor production regime to those cases where a third party has an expectation of privacy in the targeted documents, the decision under appeal raises some uncertainty concerning the appropriate mechanism for accessing third party records when it is unknown whether a reasonable expectation of privacy attaches. As I will explain, the procedure set out in O’Connor provides a general mechanism at common law for ordering production of any record beyond the possession or control of the prosecuting Crown. Whether or not the targeted record is subject to a reasonable expectation of privacy is one of the questions that must be determined at the hearing of an O’Connor application. For that pragmatic reason alone, the operation of the common law production regime cannot be premised on the existence of a reasonable expectation of privacy.
Second, to the extent that the decision in the court below suggests that there can be no expectation of privacy in the contents of a criminal investigation file, it is in error. As this Court stated in R v Mills,  3 S.C.R. 668: “Privacy is not an all or nothing right. It does not follow from the fact that the Crown has possession of the records that any reasonable expectation of privacy disappears” (para. 108). This principle holds equally in respect of criminal investigation files relating to third party accused that are not in the possession or control of the prosecuting Crown. There can be no assumption that criminal investigation files relating to third party accused persons do not attract an expectation of privacy absent consideration of their particular contents and other relevant factors. The existence of a reasonable expectation of privacy and its impact, if any, on a third party’s obligation to produce is always a contextual, fact-based inquiry. Likewise, no blanket ruling can be made in respect of privacy interests in police disciplinary records without regard to their contents.
Third, to the extent that the operative terms of the production order below may suggest that records in possession of one Crown entity are deemed to be in the possession of another, this interpretation should be discarded. The notion that all state authorities constitute a single indivisible Crown entity for the purposes of disclosure finds no support in law and, moreover, is unworkable in practice. Accordingly, Crown entities other than the prosecuting Crown are third parties under the O’Connor production regime. As I will explain, however, this does not relieve the prosecuting Crown from its obligation to make reasonable inquiries of other Crown entities and other third parties, in appropriate cases, with respect to records and information in their possession that may be relevant to the case being prosecuted. The Crown and the defence in a criminal proceeding are not adverse in interest for the purpose of discovering relevant information that may be of benefit to an accused.
In addition to clarifying these three uncertainties, this case provides an appropriate context within which to reiterate the respective obligations of the police and the Crown to disclose the fruits of the investigation under R. v Stinchcombe,  3 S.C.R. 326, and to consider the extent to which relevant police disciplinary records and third party criminal investigation files should form part of this “first party” disclosure package. The Crown’s obligation to disclose all relevant information in its possession to an accused is well established at common law and is now constitutionally entrenched in the right to full answer and defence under s. 7 of the Canadian Charter of Rights and Freedoms. The necessary corollary to the Crown’s disclosure duty under Stinchcombe is the obligation of police (or other investigating state authority) to disclose to the Crown all material pertaining to its investigation of the accused. For the purposes of fulfilling this corollary obligation, the investigating police force, although distinct and independent from the Crown at law, is not a third party. Rather, it acts on the same first party footing as the Crown.
As I will explain, records relating to findings of serious misconduct by police officers involved in the investigation against the accused properly fall within the scope of the “first party” disclosure package due to the Crown, where the police misconduct is either related to the investigation, or the finding of misconduct could reasonably impact on the case against the accused. The Crown, in turn, must provide disclosure to the accused in accordance with its obligations under Stinchcombe. Production of disciplinary records and criminal investigation files in the possession of the police that do not fall within the scope of this first party disclosure package is governed by the O’Connor regime for third party production.
I will first review the respective obligations of the Crown and the police to disclose the fruits of the investigation against the accused.
4. THE STINCHCOMBE DUTY TO DISCLOSE THE FRUITS OF THE INVESTIGATION
The Crown’s obligation to disclose all relevant information in its possession relating to the investigation against an accused is well established. The duty is triggered upon request and does not require an application to the court. Stinchcombe made clear that relevant information in the first party production context includes not only information related to those matters the Crown intends to adduce in evidence against the accused, but also any information in respect of which there is a reasonable possibility that it may assist the accused in the exercise of the right to make full answer and defence (pp. 343-44). The Crown’s obligation survives the trial and, in the appellate context, the scope of relevant information therefore includes any information in respect of which there is a reasonable possibility that it may assist the appellant in prosecuting an appeal.
While the Stinchcombe automatic disclosure obligation is not absolute, it admits of few exceptions. Unless the information is clearly irrelevant, privileged, or its disclosure is otherwise governed by law, the Crown must disclose to the accused all material in its possession. The Crown retains discretion as to the manner and timing of disclosure where the circumstances are such that disclosure in the usual course may result in harm to anyone or prejudice to the public interest. The Crown’s exercise of discretion in fulfilling its obligation to disclose is reviewable by a court.
As this Court confirmed in Mills, the Crown’s obligation under Stinchcombe to disclose the fruits of the investigation does not signify that no residual privacy interest can exist in the contents of the Crown’s file. It should come as no surprise that any number of persons and entities may have a residual privacy interest in material gathered in the course of a criminal investigation. Criminal investigative files may contain highly sensitive material including: outlines of unproven allegations; statements of complainants or witnesses – at times concerning very personal matters; personal addresses and phone numbers; photographs; medical reports; bank statements; search warrant information; surveillance reports; communications intercepted by wiretap; scientific evidence including DNA information; criminal records, etc. The privacy legislation of all 10 provinces addresses the disclosure of information contained in law enforcement files. See
Freedom of Information and Protection of Privacy Act, R.S.O. 1990, c. F.31, s. 14;
Access to Information and Protection of Privacy Act, S.N.L. 2002, c. A-1.1, s. 22;
Freedom of Information and Protection of Privacy Act, S.N.S. 1993, c. 5, s. 15;
Freedom of Information and Protection of Privacy Act, R.S.P.E.I. 1988, c. F-15.01, s. 18;
Right to Information Act, S.N.B. 1978, c. R-10.3, ss. 6(a) and 6(f);
An Act respecting Access to Documents Held by Public Bodies and the Protection of Personal Information, R.S.Q., c. A-2.1, s. 28;
The Freedom of Information and Protection of Privacy Act, S.M. 1997, c. 50, s. 25;
The Freedom of Information and Protection of Privacy Act, S.S. 1990-91, c. F-22.01, s. 15;
Freedom of Information and Protection of Privacy Act, R.S.A. 2000, c. F-25, s. 20;
Freedom of Information and Protection of Privacy Act, R.S.B.C. 1996, c. 165, s. 15.
See also the federal Privacy Act, R.S.C. 1985, c. P-21, s. 22.
Implicit in the Crown’s broad duty to disclose the contents of its file under Stinchcombe are not the absence of any residual expectation of privacy, but rather the following two assumptions. The first is that the material in possession of the prosecuting Crown is relevant to the accused’s case. Otherwise, the Crown would not have obtained possession of it (O’Connor, at para. 12). The second assumption is that this material will likely comprise the case against the accused. As a result, the accused’s interest in obtaining disclosure of all relevant material in the Crown’s possession for the purpose of making full answer and defence will, as a general rule, outweigh any residual privacy interest held by third parties in the material. These two assumptions explain why the onus is on the Crown to justify the non-disclosure of any material in its possession.
Although the common law regime of disclosure under Stinchcombe generally strikes the appropriate balance between the accused’s right to make full answer and defence and the residual privacy interests of other persons in the fruits of the investigation, it is not the only regime that meets constitutional standards. As this Court concluded in Mills, it was open to Parliament to enact, as it did, a statutory regime for the disclosure of records containing personal information of complainants and witnesses in proceedings for sexual offences under ss. 278.1 to 278.91 of the Criminal Code, R.S.C. 1985, c. C-46 (commonly referred to as the “Mills regime”). Absent an express waiver from the complainant or witness to whom the record relates, production of all records falling within the Mills regime, whether in the possession or control of a third person or of the prosecutor in the proceedings, can only be made on application to the court and in accordance with the balancing test set out in the Code provisions. This statutory regime therefore constitutes an exception to the common law regime of Crown disclosure under Stinchcombe. As we shall see, the Mills regime is also different from the common law regime for production of third party records under O’Connor. It is nonetheless constitutional (Mills, at para. 59).
The Stinchcombe regime of disclosure extends only to material in the possession or control of the Crown. The law cannot impose an obligation on the Crown to disclose material which it does not have or cannot obtain: R v Stinchcombe,  1 S.C.R. 754. A question then arises as to whether “the Crown”, for disclosure purposes, encompasses other state authorities. The notion that all state authorities amount to a single “Crown” entity for the purposes of disclosure and production must be quickly rejected. It finds no support in law and, given our multi-tiered system of governance and the realities of Canada’s geography, is unworkable in practice. As aptly explained in R v Gingras (1992), 120 A.R. 300 (C.A.), at para. 14:
If that line of reasoning were correct, then in order to meet the tests in Stinchcombe, some months before trial every Crown prosecutor would have to inquire of every department of the Provincial Government and every department of the Federal Government. He would have to ask each whether they had in their possession any records touching each prosecution upcoming. It would be impossible to carry out 1% of that task. It would take many years to bring every case to trial if that were required.
Accordingly, the Stinchcombe disclosure regime only extends to material relating to the accused’s case in the possession or control of the prosecuting Crown entity. This material is commonly referred to as the “fruits of the investigation”.
Under our Canadian system of law enforcement, the general duty to investigate crime falls on the police, not the Crown. The fruits of the investigation against an accused person, therefore, will generally have been gathered, and any resulting criminal charge laid, by the police. While the roles of the Crown and the police are separate and distinct, the police have a duty to participate in prosecutions: see, for example, s. 42(1)(e) of the Ontario Police Services Act. Of particular relevance here is the police’s duty to participate in the disclosure process. The means by which the Crown comes to be in possession of the fruits of the investigation lies in the corollary duty of police investigators to disclose to the Crown all relevant material in their possession. The police’s obligation to disclose all material pertaining to the investigation of an accused to the prosecuting Crown was recognized long before Stinchcombe. The state of the law was well summed up by the Honourable G. Arthur Martin, Q.C., in his Report of the Attorney General’s Advisory Committee on Charge Screening, Disclosure, and Resolution Discussions (1993), (“Martin Report”), at pp. 167-68:
It is well settled and accepted by all, including the police, that the police, although operating independently of Crown counsel, have a duty to disclose to Crown counsel all relevant information uncovered during the investigation of a crime, including information which assists the accused .... As one commentator has observed, “the duty of the police to disclose relevant information about a case, to the Crown, is a duty that existed before [Stinchcombe, supra]”.
The corollary duty of the police to disclose to the Crown the fruits of the investigation is now well recognized in the appellate jurisprudence. See R v Jack (1992), 70 C.C.C. (3d) 67 (Man. C.A.), at p. 94; R v T. (L.A.) (1993), 14 O.R. (3d) 378 (C.A.), at p. 382; R v Gagné (1998), 131 C.C.C. (3d) 444 (Que. C.A.), at p. 455; and Driskell v Dangerfield, 2008 MBCA 60,  6 W.W.R. 615, at para. 17. It is also widely acknowledged that the Crown cannot explain a failure to disclose relevant material on the basis that the investigating police force failed to disclose it to the Crown. See R. v MacPherson (1991), 105 N.S.R. (2d) 123 (S.C.), at paras. 37-38; R. v Oliver (1995), 143 N.S.R. (2d) 134 (S.C.), at para. 36; R v Campbell,  N.S.J. No. 702 (QL) (Prov. Ct.), at paras. 16-17.
Even though, in this narrow sense, the police and the Crown may be viewed as one entity for disclosure purposes, the two are unquestionably separate and independent entities, both in fact and in law. Hence, production of criminal investigation files involving third parties, and that of police disciplinary records, usually falls to be determined in the context of an O’Connor application. This is unsurprising because information about third party accused or police misconduct is not likely to make its way into the Crown’s Stinchcombe disclosure package unless such information is in some way related to the accused’s case. I will come back to this point later in discussing the circumstances in which information about third party misconduct should properly form part of the police disclosure package to the Crown. First, I will review the principles governing the production of third party records under the O’Connor regime.
5. THE O’CONNOR REGIME FOR PRODUCTION OF THIRD PARTY RECORDS
In O’Connor, this Court was concerned with the manner in which the accused, who was charged with multiple sexual offences, could obtain production of the therapeutic records of the complainants from third party custodians. O’Connor has been overtaken by Parliament’s subsequent enactment of the Mills regime contained in ss. 278.1 to 278.91 of the Criminal Code for the disclosure of records containing personal information of complainants and witnesses in sexual assault proceedings. In respect of any other criminal proceeding, however, the O’Connor application provides the accused with a mechanism for accessing third party records that fall beyond the reach of the Stinchcombe first party disclosure regime.
Stated briefly, the procedure to be followed on an O’Connor application is the following.
The accused first obtains a subpoena duces tecum under ss. 698(1) and 700(1) of the Criminal Code and serves it on the third party record holder. The subpoena compels the person to whom it is directed to attend court with the targeted records or materials.
The accused also brings an application, supported by appropriate affidavit evidence, showing that the records sought are likely to be relevant in his or her trial. Notice of the application is given to the prosecuting Crown, the person who is the subject of the records and any other person who may have a privacy interest in the records targeted for production.
The O’Connor application is brought before the judge seized with the trial, although it may be heard before the trial commences. If production is unopposed, of course, the application for production becomes moot and there is no need for a hearing.
If the record holder or some other interested person advances a well-founded claim that the targeted documents are privileged, in all but the rarest cases where the accused’s innocence is at stake, the existence of privilege will effectively bar the accused’s application for production of the targeted documents, regardless of their relevance. Issues of privilege are therefore best resolved at the outset of the O’Connor process.
Where privilege is not in question, the judge determines whether production should be compelled in accordance with the two-stage test established in O’Connor. At the first stage, if satisfied that the record is likely relevant to the proceeding against the accused, the judge may order production of the record for the court’s inspection. At the next stage, with the records in hand, the judge determines whether, and to what extent, production should be ordered to the accused.
The question of privilege is beyond the scope of this appeal. However, I will elaborate on each stage of the O’Connor test for production of third party records in turn.
5.1 First Stage: Screening for Likely Relevance
5.1.1 Burden is on the Applicant
The first step in any contested application for production of non-privileged documents in the possession of a third party is for the person seeking production – in this case the accused – to satisfy the court that the documents are likely relevant to the proceedings. This threshold burden simply reflects the fact that the context in which third party records are sought is different from the context of first party disclosure. We have already seen that the presumptive duty on Crown counsel to disclose the fruits of the investigation in their possession under Stinchcombe is premised on the assumptions that the information is relevant and that it will likely comprise the case against the accused. No such assumptions can be made in respect of documents in the hands of a third party who is a stranger to the litigation. The applicant must therefore justify to the court the use of state power to compel their production – hence the initial onus on the person seeking production to show “likely relevance”. In addition, it is important for the effective administration of justice that criminal trials remain focussed on the issues to be tried and that scarce judicial resources not be squandered in “fishing expeditions” for irrelevant evidence. The likely relevance threshold reflects this gate-keeper function.
5.1.2 Burden on Applicant is Significant but not Onerous
It is important to repeat here, as this Court emphasized in O’Connor, that while the likely relevance threshold is “a significant burden, it should not be interpreted as an onerous burden upon the accused” (para. 24). On the one hand, the likely relevance threshold is “significant” because the court must play a meaningful role in screening applications “to prevent the defence from engaging in ‘speculative, fanciful, disruptive, unmeritorious, obstructive and time-consuming’ requests for production” (O’Connor, at para. 24; quoting from R v Chaplin,  1 S.C.R. 727, at para. 32). The importance of preventing unnecessary applications for production from consuming scarce judicial resources cannot be overstated; however, the undue protraction of criminal proceedings remains a pressing concern, more than a decade after O’Connor. On the other hand, the relevance threshold should not, and indeed cannot, be an onerous test to meet because accused persons cannot be required, as a condition to accessing information that may assist in making full answer and defence, “to demonstrate the specific use to which they might put information which they have not even seen” (O’Connor, at para. 25, quoting from R v Durette,  1 S.C.R. 469, at p. 499).
5.1.3 O’Connor Common Law Threshold Significantly Different from Mills Statutory Regime
It is important to note that the common law likely relevance threshold in O’Connor differs significantly from the statutory likely relevance threshold set by Parliament for the production of records containing personal information in sexual assault proceedings under the Mills regime (see s. 278.3(4) of the Criminal Code). As this Court explained at some length in Mills, a range of permissible regimes can meet constitutional standards. It was therefore open to Parliament to craft its own solution to address the particular concerns arising from disclosure of third party records in sexual proceedings. In doing so, Parliament “sought to recognize the prevalence of sexual violence against women and children and its disadvantageous impact on their rights, .... and to reconcile fairness to complainants with the rights of the accused” (Mills, at para. 59). The following differences between the two regimes are particularly noteworthy.
First, the likely relevance standard adopted by Parliament under the Mills regime is tailored to counter speculative myths, stereotypes and generalized assumptions about sexual assault victims and about the usefulness of private records in sexual assault proceedings. Such generalized views need not be countered at large in respect of all third party records that fall outside the Mills regime. The general common law threshold of likely relevance under O’Connor is intended rather to screen applications to ensure the proper use of state authority in compelling production of third party records and to establish the appropriateness of the application so as to avoid squandering scarce judicial resources.
Second, while the Mills regime retains the two-stage framework set out in O’Connor, it differs significantly in that much of the balancing of the competing interests is effected at the first stage in determining whether production should be made to the court for inspection. This reflects Parliament’s assumption that a reasonable expectation of privacy exists in the types of records targeted by the statutory regime: see R v Clifford (2002), 163 C.C.C. (3d) 3 (Ont. C.A.), at paras. 48-49. An equivalent presumption of privacy does not attach in respect of all third party records that fall outside the Mills regime. Hence, any balancing of competing interests is reserved for the second stage of the O’Connor regime, when the documents can be inspected by the court to better ascertain the nature of the privacy interest, if any. Because of these significant differences, it is important not to transpose the Mills regime into the O’Connor production hearing in respect of documents to which the statutory dispositions do not apply.
5.1.4 Likely Relevance Under the Common Law Regime
“Likely relevant” under the common law O’Connor regime means that there is “a reasonable possibility that the information is logically probative to an issue at trial or the competence of a witness to testify” (O’Connor, at para. 22 (emphasis deleted)). An “issue at trial” here includes not only material issues concerning the unfolding of the events which form the subject matter of the proceedings, but also “evidence relating to the credibility of witnesses and to the reliability of other evidence in the case” (O’Connor, at para. 22). At this stage of the proceedings, the court cannot insist on a demonstration of the precise manner in which the targeted documents could be used at trial. The imposition of such a stringent threshold burden would put the accused, who has not seen the documents, in an impossible Catch-22 position.
5.2 Second Stage: Balancing the Interests at Play
If likely relevance is demonstrated by the applicant, the third party record holder may be ordered to produce the documents for inspection by the court in order to determine whether production should be ordered to the accused.
In O’Connor, this Court provided the following list of factors for consideration in determining whether or not to order production to the accused (at para. 156):
the extent to which the record is necessary for the accused to make full answer and defence;
the probative value of the record in question;
the nature and extent of the reasonable expectation of privacy vested in that record;
whether production of the record would be premised upon any discriminatory belief or bias; [and]
the potential prejudice to the complainant’s dignity, privacy or security of the person that would be occasioned by production of the record in question.
The factors set out in O’Connor should not be applied mechanically. It should be kept in mind that O’Connor involved the production of the complainant’s private records in proceedings for a sexual offence, an area of law subsequently overtaken by Parliament’s enactment of the Mills regime. Some of the factors listed in O’Connor, in particular items 4 and 5 above, were obviously tailored to meet the exigencies in sexual assault proceedings and, consequently, are unlikely to be of assistance in other contexts. Ultimately, what is required at this second stage of the common law regime is a balancing of the competing interests at stake in the particular circumstances of the case. No exhaustive list can be crafted to suit every situation; however, I will elaborate somewhat on the balancing process.
5.2.1 A Useful Starting Point: Assessing True Relevance
As we have seen, by the time the court embarks on the second stage of an O’Connor hearing, the application has been screened for likely relevance and the judge is satisfied that compelling the production of the documents for inspection by the court is warranted in the circumstances. Once the documents are before the court, the final question to be determined is whether production should be made to the accused.
The amicus curiae submits that the starting point for determining whether to order production to the accused of documents in the hands of third parties, including, as in this case, Crown briefs, police briefs and police disciplinary files, requires a determination of the nature of the records to define the existence or absence of a reasonable expectation of privacy (factum, at para. 19). There is some merit to the proposal of looking first at whether there is any expectation of privacy in the targeted records. Indeed, if it is clear upon inspection of the documents and their contents that there is no basis upon which the third party record holder or any other interested person could claim a reasonable privacy interest, there is no balancing of interests left to perform. Given that likely relevance has by this point been established, the sole remaining interest at stake is the accused’s right to make full answer and defence. A production order should therefore issue. It is in this sense that the Court of Appeal’s finding in the present case, that an “O’Connor-type procedure” is only required in cases where the third party records attract a reasonable expectation of privacy, must be understood.
The difficulty with this approach, however, is that it is largely premised on the conclusion that the existence of a reasonable expectation of privacy may be determined solely by characterizing the type of record at issue. While this may be possible in respect of some records, determining whether an expectation of privacy attaches to a particular record usually requires a more contextual approach. As we have seen, even in the context of the Crown’s Stinchcombe obligation of disclosure, there exists no presumption that all expectations of privacy are lost just because a record has found its way into the prosecuting Crown’s file.
The amicus curiae submits, and I agree, that determining the existence of a reasonable expectation of privacy requires a contextual assessment having regard to numerous factors, including but not limited to: how the record was created; who created the record; the purpose of the record; the context of the case in which the record would be used; who holds the privacy interest; how the record was obtained by the Crown or police; the presence or absence of waiver; any applicable legislation; and whether the privacy interest extends to all or part of the record. As one can readily appreciate from this non-exhaustive list, determining whether there is any residual expectation of privacy in a document may be a complex and time-consuming exercise that has the potential to significantly delay and detract from the actual proceeding before the court – the trial of the accused. For that reason, it is my view that in most cases, a more useful starting point for courts in balancing competing interests at the second stage of an O’Connor application will be to assess the true relevancy of the targeted record in the case against the accused. This approach allows the court to remain focussed on the trial of the accused and, given the competing interests at stake, the relevancy assessment will usually be largely determinative of the production issue. I will explain.
On the one hand, because the accused will not have seen the documents that he or she seeks to have produced, it may be readily apparent upon inspection by the court that the claim of likely relevance established at the first stage of the O’Connor application is simply not borne out. If the court is satisfied that the documents are clearly irrelevant, there is no basis for compelling production to the accused, and the application can be summarily dismissed.
On the other hand, if the claim of likely relevance is borne out upon inspection, the accused’s right to make full answer and defence will, with few exceptions, tip the balance in favour of allowing the application for production. Recall that at this stage of the proceedings, the court has confirmed that the production application concerns non-privileged documents. The existence of any privacy interest in third party records relevant to an accused’s defence of a criminal charge may well warrant, in appropriate circumstances, some redactions or the imposition of conditions to ensure that no unnecessary invasion of privacy follows from production to the accused. However, absent an overriding statutory regime governing the production of the record in question, a third party privacy interest is unlikely to defeat an application for production.
Once a court has ascertained upon inspection that third party records are indeed relevant to the accused’s case, in the sense that they pertain to an issue in the trial as described above, the second stage balancing exercise is easily performed. In effect, a finding of true relevance puts the third party records in the same category for disclosure purposes as the fruits of the investigation against the accused in the hands of the prosecuting Crown under Stinchcombe. It may be useful to pose the question in this way: if the third party record in question had found its way into the Crown prosecutor’s file, would there be any basis under the first party Stinchcombe disclosure regime for not disclosing it to the accused? If the answer to that question is no, there can be no principled reason to arrive at a different outcome on the third party production application. As we have seen, the Crown’s obligation under Stinchcombe to disclose to the accused the fruits of the investigation in its possession does not signify that no residual privacy interest exists in the contents of the file. However, it does mean that, with few exceptions (including the Mills statutory scheme), the accused’s right to access information necessary to make full answer and defence will outweigh any competing privacy interest. The same applies with respect to relevant material in the hands of third parties. This is particularly so in respect of criminal investigation files concerning third party accused. As Professor Paciocco aptly puts it:
.... it would be perverse to hold that investigative information is not private enough to impose any constraints on the ability of the police to gather it and the Crown to use it to seek the conviction of the target of that investigation, yet that same information is protected by privacy interests when it contains relevant information that would help someone defend himself.
(David M. Paciocco, “Filling the Seam between Stinchcombe and O’Connor: The “McNeil” Disclosure Application” (2007), 53 Crim. L.Q. 161, at pp. 199-200)
5.2.2 Guarding Against Unnecessary Intrusions Into Privacy Interests
That is not to say, however, that residual privacy interests in the contents of criminal investigation files should be disregarded. As concluded in the Martin Report, at p. 181: (reference at AG factum, at para. 24):
The privacy of the victim and any other witnesses must yield to preparing a full answer and defence. But it need not yield any further. The Committee considers that provided the making of full answer and defence is not impaired, it is desirable to permit limitations on the use of disclosure materials that recognize the privacy interests of victims and witnesses.
The same applies in respect of police disciplinary records, or any other third party records. The court should ensure that a production order is properly tailored to meet the exigencies of the case but do no more. As we have seen, likely relevance for disclosure purposes has a wide and generous connotation and includes information in respect of which there is a reasonable possibility that it may assist the accused in the exercise of the right to make full answer and defence. In considering the ambit of the information that can assist in the trial, regard must be given to the particular issues in the case and to the governing rules of evidence and procedure. This does not mean that only material that would be admissible at trial should be produced. Material that would not, on its own, be admissible may nonetheless be of use to the defence, for example, in cross-examining a witness on matters of credibility or in pursuing other avenues of investigation. An O’Connor application must, however, lay a foundation for the relevance of the material sought, having regard to the issues in the case. In large part, this is the reason why such applications should be brought before the judge seized with the trial. Doing so allows production issues to be effectively disposed of by the judge and counsel who are sufficiently appraised of the issues in the case. Of course, production applications can be heard prior to the commencement of the trial, and the judge, when appropriate to do so, may enlist the assistance of counsel in determining the relevance of a particular record for the purposes of production.
Ascertaining the true relevancy of records targeted for production may become particularly important when the information on the production application concerns police disciplinary records. The contentious nature of police work often leads to public complaints, some legitimate and others spurious. Police disciplinary proceedings may also relate to employment issues or other matters that have no bearing on the case against the accused. The risk in this context is that disclosure, and by extension trial proceedings, may be sidetracked by irrelevant allegations or findings of police misconduct. Disclosure is intended to assist an accused in making full answer and defence or in prosecuting an appeal, not turn criminal trials into a conglomeration of satellite hearings on collateral matters.
Hence, to ensure that only relevant material is produced and that no unwarranted invasion of privacy interests occurs, the court may find it necessary to make a production order subject to redactions or other conditions. In addition, when just and appropriate to do so, the court may well impose restrictions on the dissemination of the information produced for purposes unrelated to the accused’s full answer and defence or prosecution of an appeal: see, for example, recommendations 34 and 35 in the Martin Report (at p. 179) and P. (D.) v Wagg (2004), 239 D.L.R. (4th) 501 (Ont. C.A.), at para. 46. Absent unusual circumstances, however, the crafting of a production order should not necessitate any detailed inquiry into the precise nature or extent of the privacy interest in question. The O’Connor hearing should remain focussed on the criminal proceeding at hand.
6. BRIDGING THE GAP BETWEEN FIRST PARTY DISCLOSURE AND THIRD PARTY PRODUCTION
As the preceding discussion makes clear, once the true relevance of the targeted records has been ascertained on an O’Connor application for production of third party records, the ultimate question of production is essentially governed by the same principles that apply to the disclosure of material in the possession of the Crown under Stinchcombe. To the accused, however, the distinction between these two regimes is significant. While the accused will receive automatic disclosure of relevant material that finds its way into the hands of the prosecuting Crown, accessing relevant material in the hands of third parties will often be more happenstance. To a certain extent, that is inevitable. Third parties are under no obligation to come forth with relevant information to assist the accused in his defence. However, the prosecuting Crown and the investigating police force are in a different position and can assist in bridging the gap between first party disclosure and third party production. I will deal firstly with the Crown.
6.1 Crown Counsel’s Duty to Inquire
As stated earlier, the suggestion that all state authorities constitute a single entity is untenable and unworkable. In order to fulfill its Stinchcombe disclosure obligation, the prosecuting Crown does not have to inquire of every department of the provincial government, every department of the federal government and every police force whether they are in possession of material relevant to the accused’s case. However, this does not mean that, regardless of the circumstances, the Crown is simply a passive recipient of relevant information with no obligation of its own to seek out and obtain relevant material.
The Crown is not an ordinary litigant. As a minister of justice, the Crown’s undivided loyalty is to the proper administration of justice. As such, Crown counsel who is put on notice of the existence of relevant information cannot simply disregard the matter. Unless the notice appears unfounded, Crown counsel will not be able to fully assess the merits of the case and fulfill its duty as an officer of the court without inquiring further and obtaining the information if it is reasonably feasible to do so. Ryan J.A. in R v Arsenault (1994), 153 N.B.R. (2d) 81 (C.A.), aptly described the Crown’s obligation to make reasonable inquiries of other Crown agencies or departments. He stated as follows [para. 15]:
When disclosure is demanded or requested, Crown counsel have a duty to make reasonable inquiries of other Crown agencies or departments that could reasonably be considered to be in possession of evidence. Counsel cannot be excused for any failure to make reasonable inquiries when to the knowledge of the prosecutor or the police there has been another Crown agency involved in the investigation. Relevancy cannot be left to be determined by the uninitiated. If Crown counsel is denied access to another agency’s file, then this should be disclosed to the defence so that the defence may pursue whatever course is deemed to be in the best interests of the accused. This also applies to cases where the accused or defendant, as the case may be, is unrepresented ....
The same duty to inquire applies when the Crown is informed of potentially relevant evidence pertaining to the credibility or reliability of the witnesses in a case. As the amicus curiae rightly states, “[t]he Crown and the defence are not adverse in interest in discovering the existence of an unreliable or unethical police officer” (factum, at para. 62). Doherty J.A. made the point forcefully in R. v Ahluwalia (2000), 138 O.A.C. 154, commenting on the Crown’s failure to inquire further when confronted with the perjury of its own witness as follows (at paras. 71-72):
For reasons not shared with this court, the Crown does not appear to have regarded itself as under any obligation to get to the bottom of this matter ....
The Crown has obligations to the administration of justice that do not burden other litigants. Faced with its own witness’s perjury and the fact that the perjured evidence coincided with the incomplete disclosure that the Crown says it innocently passed to the defence, the Crown was obliged to take all reasonable steps to find out what had happened and to share the results of those inquiries with the defence. In my view, the Crown did not fulfill its obligations to the administration of justice by acknowledging the incomplete disclosure discovered by the defence, and after making limited inquiries, professing neither a responsibility for the incomplete disclosure nor an ability to provide any explanation for it. The Crown owed both the appellant and the court a fuller explanation than it chose to provide.
Hence, by properly fulfilling its dual role as an advocate and officer of the court, Crown counsel can effectively bridge much of the gap between first party disclosure and third party production. I now turn to the police corollary duty to participate in the disclosure process.
6.2 The Police’s Corollary Duty to Disclose Relevant Information
We have already seen that the police have a corollary duty to disclose to the prosecuting Crown all material pertaining to the investigation of an accused. This disclosure obligation accords with police codes of conduct enacted in many jurisdictions across the country. For example, under s. 2(1)(c)(vi) of the Code of Conduct found in O. Reg. 123/98, Sch., enacted pursuant to the Ontario Police Services Act, a police chief or other police officer commits neglect of duty if he or she “fails to report anything that he or she knows concerning a criminal or other charge, or fails to disclose any evidence that he or she, or any person within his or her knowledge, can give for or against any prisoner or defendant”. See also Code of Professional Conduct Regulation, B.C. Reg. 205/98, s. 5(e); Police Service Regulation, Alta. Reg. 356/1990, s. 5(2)(h)(vii); Code of ethics of Québec police officers, (1990) 122 G.O. II, 1760, s. 7(2); and Code of Professional Conduct Regulation – Police Act, N.B. Reg. 2007-81, s. 36(1)(d)(iii).
While the obligation itself is firmly established, the difficulty lies in identifying the contours of relevance for the purposes of the police’s first party disclosure obligation. The particular question that this case exemplifies is whether information of misconduct by a police officer involved in the case against the accused should form part of the first party disclosure package provided to the Crown for its assessment of relevance according to the edicts of Stinchcombe. Obviously, the accused has no right to automatic disclosure of every aspect of a police officer’s employment history, or to police disciplinary matters with no realistic bearing on the case against him or her. However, where the disciplinary information is relevant, it should form part of the first party disclosure package, and its discovery should not be left to happenstance.
When the police misconduct in question concerns the same incident that forms the subject-matter of the charge against the accused, the police duty to disclose information concerning police disciplinary action taken in respect of that misconduct is rather self-evident. To state an obvious example, if a police officer is charged under the applicable provincial legislation for excessive use of force in relation to the accused’s arrest, this information must be disclosed to the Crown. Where the misconduct of a police witness is not directly related to the investigation against the accused, it may nonetheless be relevant to the accused’s case, in which case it should also be disclosed. For example, no one would question that the criminal record for perjury of a civilian material witness would be of relevance to the accused and should form part of the first party disclosure package. In the same way, findings of police misconduct by a police officer involved in the case against the accused that may have a bearing on the case against an accused should be disclosed.
The need for guidance in this area prompted a review commissioned by the Chief of the Toronto Police Service. The review was conducted by the Honourable George Ferguson, Q.C., and culminated in his January 2003 report entitled Review and Recommendations Concerning Various Aspects of Police Misconduct, vol. I (the “Ferguson Report”). The terms of reference included the mandate to review “when, in what manner and under what circumstances does the Police Service have an obligation to bring to the attention of the Crown, alleged or proven acts of misconduct of a police officer who will be a witness or was otherwise involved in an investigation that has led to a criminal proceeding” (p. 1).
The Ferguson Report noted that the question of disclosure of such information was usually left to be determined in the context of an O’Connor application for third party production and that this often gave rise to a “Catch-22” problem, stating as follows (at p. 12):
Once into O’Connor, it is difficult, if not impossible, for the defence to meet the threshold required to access the records. Unless the officer is notorious or the defence personally knows details of the officer’s files, the defence will be found to be on nothing more than a fishing expedition and access will be denied. A diligent officer-in-charge will not likely know the full employment history of each police witness, and, currently, is not required to make any inquiries. Short of a criminal conviction for dishonesty, nothing is likely to be provided to the Crown for disclosure analysis. Even convictions for dishonesty may not always be revealed. This is clearly unsatisfactory.
The Ferguson Report concluded that leaving the entire question of access to police disciplinary records to be determined under the O’Connor regime for third party production “is neither efficient nor justified” (p. 15). In order to assist in bridging the gap between first party disclosure and third party production, the Ferguson Report made a number of recommendations, including the automatic disclosure by the police upon request by the Crown of the following information regarding acts of misconduct by a member of the Toronto Police Service who may be a witness or who was otherwise involved in a case before the court (at p. 17):
Any conviction or finding of guilt under the Canadian Criminal Code or the Controlled Drugs and Substances Act [for which a pardon has not been granted].
Any outstanding charges under the Canadian Criminal Code or the Controlled Drugs and Substances Act.
Any conviction or finding of guilt under any other federal or provincial statute.
Any finding of guilt for misconduct after a hearing under the Police Services Act or its predecessor Act.
Any current charge of misconduct under the Police Services Act for which a Notice of Hearing has been issued.
The Ferguson Report recommended that upon receiving this information from police, the Crown act as “gate-keeper”, sorting out what parts of this material, if any, should be turned over to the defence in compliance with the Crown’s Stinchcombe obligation of disclosure. The Ferguson Report made the further recommendation that any concerned officer who was the subject of disciplinary records produced to the Crown be notified in writing and be given the opportunity to make submissions to the Crown.
I agree that it is “neither efficient nor justified” to leave the entire question of access to police misconduct records to be determined in the context of the O’Connor regime for third party production. Indeed, as discussed earlier, the disclosure of relevant material, whether it be for or against an accused, is part of the police corollary duty to participate in the disclosure process. Where the information is obviously relevant to the accused’s case, it should form part of the first party disclosure package to the Crown without prompting. For example, as was the case here, if an officer comes under investigation for serious drug-related misconduct, it becomes incumbent upon the police force, in fulfilment of its corollary duty of disclosure to the Crown, to look into those criminal cases in which the officer is involved and to take appropriate action. Of course, not every finding of police misconduct by an officer involved in the investigation will be of relevance to an accused’s case. The officer may have played a peripheral role in the investigation, or the misconduct in question may have no realistic bearing on the credibility or reliability of the officer’s evidence. The kinds of information listed in the Ferguson Report can provide useful guidance on those types of matters in respect of which a police force may well be advised to seek the advice of Crown counsel.
With respect to records concerning police disciplinary matters that do not fall within the scope of first party disclosure obligations, procedures such as those recommended in the Ferguson Report, tailored to suit the particular needs of the community in which they are implemented, can go a long way towards ensuring a more efficient streamlining of O’Connor applications for third party production. Trial courts seized with motions for disclosure under Stinchcombe or applications for third party production are well placed to make appropriate orders to foster the necessary cooperation between police, the Crown and defence counsel.
The appeal is allowed and the order in the court below is set aside. The application having become moot, the Court makes no further order.
R v O’Connor,  4 S.C.R. 411; R v Mills,  3 S.C.R. 668; R v Stinchcombe,  3 S.C.R. 326; R v Stinchcombe,  1 S.C.R. 754; R v Gingras (1992), 120 A.R. 300; R v Jack (1992), 70 C.C.C. (3d) 67; R v T. (L.A.) (1993), 14 O.R. (3d) 378; R v Gagné (1998), 131 C.C.C. (3d) 444; Driskell v Dangerfield, 2008 MBCA 60,  6 W.W.R. 615; R v MacPherson (1991), 105 N.S.R. (2d) 123; R v Oliver (1995), 143 N.S.R. (2d) 134; R. v Campbell,  N.S.J. No. 702 (QL); R v Chaplin,  1 S.C.R. 727; R v Durette,  1 S.C.R. 469; R v Clifford (2002), 163 C.C.C. (3d) 3; P. (D.) v Wagg (2004), 239 D.L.R. (4th) 501; R v Arsenault (1994), 153 N.B.R. (2d) 81; R v Ahluwalia (2000), 138 O.A.C. 154.
Access to Information and Protection of Privacy Act, S.N.L. 2002, c. A-1.1, s. 22.
Act respecting Access to Documents Held by Public Bodies and the Protection of Personal Information, R.S.Q., c. A-2.1, s. 28.
Canadian Charter of Rights and Freedoms, s. 7.
Code of Conduct, O. Reg. 123/98, Sch., s. 2(1)(c)(vi).
Code of ethics of Québec police officers, (1990) 122 G.O. II, 1760, s. 7(2).
Code of Professional Conduct Regulation, B.C. Reg. 205/98, s. 5(e).
Code of Professional Conduct Regulation – Police Act, N.B. Reg. 2007-81, s. 36(1)(d)(iii).
Criminal Code, R.S.C. 1985, c. C-46, ss. 278.1 to 278.91, 278.3(4), 698(1), 700(1).
Freedom of Information and Protection of Privacy Act, R.S.A. 2000, c. F-25, s. 20.
Freedom of Information and Protection of Privacy Act, R.S.B.C. 1996, c. 165, s. 15.
Freedom of Information and Protection of Privacy Act, R.S.O. 1990, c. F.31, s. 14.
Freedom of Information and Protection of Privacy Act, R.S.P.E.I. 1988, c. F-15.01, s. 18.
Freedom of Information and Protection of Privacy Act, S.M. 1997, c. 50, s. 25.
Freedom of Information and Protection of Privacy Act, S.N.S. 1993, c. 5, s. 15.
Freedom of Information and Protection of Privacy Act, S.S. 1990-91, c. F-22.01, s. 15.
Police Service Regulation, Alta. Reg. 356/90, s. 5(2)(h)(vii).
Police Services Act, R.S.O. 1990, c. P.15, s. 42(1)(e).
Privacy Act, R.S.C. 1985, c. P-21, s. 22.
Right to Information Act, S.N.B. 1978, c. R-10.3, s. 6
Authors and other references
Ontario. Report of the Attorney General’s Advisory Committee on Charge Screening, Disclosure, and Resolution Discussions. Toronto: The Committee, 1993.
Ontario. Review and Recommendations Concerning Various Aspects of Police Misconduct, vol. I. Toronto: Toronto Police Service, 2003 (online: http://www.torontopolice.on.ca/publications/files/reports/ferguson1.pdf).
Paciocco, David M. “Filling the Seam between Stinchcombe and O’Connor: The ‘McNeil’ Disclosure Application” (2007), 53 Crim. L.Q. 161.
Christine Bartlett-Hughes and Elise Nakelsky (instructed by M/s Attorney General of Ontario, Toronto), for the appellant.
James C. Martin and Rick Visca (instructed by M/s Public Prosecution Service of Canada, Halifax), for the respondent Her Majesty the Queen.
H. Reginald Watson and Jill Sexsmith (instructed by M/s Caswell & Watson, Toronto), for the respondent the Chief of Barrie Police Service, 3rd Party Record Holder.
Gary R. Clewley and Henry S. Brown, Q.C. (instructed by M/s Gary R. Clewley, Toronto), for the intervener Matthew Marshall.
James A. Bowron (instructed by M/s Attorney General of Alberta, Edmonton), for the intervener the Attorney General of Alberta.
Ian J. Roland and Danny Kastner (instructed by M/s Paliare Roland Rosenberg Rothstein, Toronto), for the intervener the Police Association of Ontario.
David M. Porter and Christopher A. Wayland (instructed by M/s McCarthy Tétrault, Toronto), for the intervener the Criminal Lawyers’ Association (Ontario).
Marie Henein and Jordan Glick (instructed by M/s Henein & Associates, Toronto), for the amicus curiae.
all rights reserved