NOTE: This article has been divided into two parts due to its length. If you wish, you can read the FIRST PART.

(1996) 1 Canadian Criminal Law Review 31-56
Trial by Ordeal
Marilyn MacCrimmon*

The Threshold Test

Both the majority and the dissent adopt the standard of "likely relevance", but their views as to its meaning appear to differ. The dissent identifies the standard as a "significant" one.¹ The majority state, "While this is a significant burden, it should not be interpreted as an onerous burden upon the accused."² They differ as to the factors to be considered in assessing "likely relevance". The majority equate it with "logically probative".³ The dissent incorporate the concepts of necessity⁴ and collateral issues⁵ and would scrutinize the assessment for the operation of discriminatory beliefs.⁶ The majority apparently do not foresee the operation of discriminatory beliefs by trial judges at the first stage since they only list this as a possibility in the decision to disclose the records to the defence. Such confusion requires legislative clarification.

Fair trials require that courts and legislatures scrutinize the commonsense generalizations underlying relevance assessments. This requires a recognition of the cognitive limitations of the decision maker. What is the underlying reasoning process? What irrationalities might be operating? Two irrationalities of concern with regard to confidential records of sexual assault complainants are errors in estimating the probabilities of the underlying generalizations and reliance on discriminatory generalizations.⁷ The first raises the issue of the probative force of the inferential link between the information and the fact in issue.⁸ The second raises the difficult question: how do we identify discriminatory generalizations which should be eliminated from the judicial fact finding process?

First, a threshold test sets out the necessary probative value or in other words, the strength of the inferential link. Requiring a higher probative value than "likely relevance" may be justified in cases in which disclosure will adversely affect the individual witness and the fact determination process. Underlying generalizations which are improbable, and the likelihood of which are likely to be overestimated would be eliminated from the reasoning process, One step towards this goal is the identification by Parliament of its empirical assumptions such as disclosure to trial judges may deter reporting of sexual assaults and cause psychological harm, many victims of sexual assault seek counselling, evidence of sexual history is prejudicial, fear of sexual assault affects the lives of people, and evidence of therapeutic records is rarely relevant.

Second, another aspect of a threshold test is the reliability of the inference. For instance, assessments of the relevance of expert opinion includes an assessment of the reliability of the inferential link.⁹ Rejecting information because it is unreliable is analogous to concluding, as does David Schum, that some information is not "evidence" for the purposes of making decisions with significant consequences such as by hospitals, courts, laboratories or boardrooms.¹⁰ Thus he would exclude from his definition of evidence testimony of a person who claims to be clairvoyant and claims to be able to tell whether some event will happen in the future. There is some information which is too unreliable to be classified as "evidence" for the purposes of certain types of decisions. L'Heureux-Dubé J. argues that therapeutic records are in most cases unreliable sources of information:

(M)edical records concerning statements made in the course of therapy are both hearsay and inherently problematic as regards reliability. A witness's concerns expressed in the course of therapy after the fact, even assuming they are correctly understood and reliably noted, cannot be equated with evidence given in the course of a trial. Both the context in which the statements are made and the expectations of the parties are entirely different. In a trial, a witness is sworn to testify as to the particular events in issue. By contrast, in therapy an entire spectrum of factors such as personal history, thought, emotions as well as particular acts may inform the dialogue between therapist and patient. Thus, there is serious risk that such statements could be taken piecemeal out of the context in which they were made to provide a foundation for entirely unwarranted inferences by the trier of fact.¹¹

Third, another goal of a threshold test should be to eliminate discriminatory beliefs from disclosure decisions. Assessment of relevance and probative value of therapeutic records is particularly prone to the operation of discriminatory beliefs. A principal difficulty is that much of the harm to equality and privacy rights occurs at the time of disclosure of the record even if this disclosure is to a judicial decision maker. Therefore protection of the equality and privacy rights of complainants requires, at least at the first level of decision, that relevance be assessed without examining the records. This approach results in reliance on hypothetical reasoning. Because judges do not know what is in the record, they develop hypotheses about what might be there. Judges will conclude therapeutic records may contain relevant information if they can think of hypothetical situations in which the information would be relevant. For instance, the judge may conclude there may be an inconsistent statement in the record or that there is support for false memory syndrome. Thus the rule on disclosure has depended on judges' creativity in creating hypotheticals about what information may be in the record. And such hypotheticals have come easily to mind in sexual assault cases. The image of the lying women or the emotionally disturbed woman underlie doubts about the credibility of complainants. There is a real danger of biases and stereotypes distorting fact finding process. Reliance on hypotheticals leads to a general rule of disclosure since they are not case specific but apply to all cases. What looks like a case by case analysis is really adoption of a general rule of disclosure.

Eliminating unreliable and improbable inferences as a ground for findings of "relevance" may not filter out discriminatory stereotypical reasoning because stereotypes are "not readily falsifiable either in theory or in practice"¹³--they may in their general form be seen to be true.¹⁴ Reliance on stereotypes is objectionable, not because they are never true, but because using heuristics based on gender or race even if generally plausible tends to perpetuate a system in which legal and social arrangements place "one social group defined in terms of a morally irrelevant characteristic, systematically below another."¹⁵ Stereotypes determine the range of interpretations thereby limiting and closing off alternative interpretations.¹⁶ We reject some factual claims based on biases and heuristics because the costs of error is too high--even if using race or gender as a proxy is rational in that it is statistically valid, the costs of permitting that heuristic to operate are too high both in the case in which the proxy is inaccurate and in terms of social costs. Cass Sunstein states: A "prohibition on discrimination is designed, not to root out irrationality, but to eliminate a system with castelike features."¹⁷

Which stereotypes should be eliminated from judicial fact determination? The task is to identify commonsense generalizations which if generally applied are likely to have a discriminatory effect. The beginning point is to ask why the private, personal background of complainants of sexual assault are assumed to be relevant when other witnesses are typically not asked to reveal their confidential, private records. The foundational discriminatory myths in sexual assault cases are that complainants frequently fabricate allegations of sexual assault and such allegations are easily brought. These myths underlie arguments that therapeutic records contain relevant information. The belief that there may be an inconsistent statement in the records draws on the traditional generalization that women frequently make false accusations of sexual assault. Drawing on this myth, the scenario that the complainant has told her therapist she lied although unsupported by any other information seems plausible. There are also conflicting assumptions about therapists which lead to disclosure. On the one hand, there may be an assumption that therapeutic records might assist rather than distort the search for truth. On the other hand, there may be a suspicion about therapists and their methods. Have the allegations arisen because of the therapeutic process?

Disclosure of unreliable and discriminatory information adversely affects the administration of justice. An additional concern present when private records are disclosed pre-trial is the role of the information in the pre-trial fact determination process. The Supreme Court of Canada in R. v. Stinchcome stated that there was an expectation disclosure would result in more pre-trial settlements. The impact of disclosure of therapeutic records on this process should be a factor in designing disclosure rules. In my view, in the light of the above analysis, there is a real possibility that therapeutic records will have a prejudicial effect on the pre-trial fact determination process. The safeguards of the trial process do not operate. The impact of discriminatory beliefs on the plea bargaining process and in police decisions to continue a case has been documented.¹⁸

How can Parliament regulate the inferential process? Legislative identification of impermissible grounds of inference will help to eliminate the operation of discriminatory beliefs. As pointed out above, the British Columbia Court of Appeal and the dissenting Supreme Court justices in O'Connor included such a list in their disclosure procedures.¹⁹ The effect is to require that the defence show the probative value of the record in the specific case as stated by the dissent in O'Connor.

Whether setting out impermissible grounds of inference conflicts with the right to a fair trial is debatable. The Supreme Court of Canada has held that assessments of relevance, at least for the purposes of admission at trial, must be examined for discriminatory biases and stereotypes.²⁰ Presumably, although not expressly mentioned by the majority, assessments of "likely relevance" must not be based on such assumptions.²¹ The objection by the defence will be that these generalizations are sometimes true.²² But as pointed out above, the fact a stereotype may sometimes be true does not reduce its discriminatory effects. Requiring the defence to show the generalization is true in the specific case may lessen the harm to equality rights and reduce the operation of prejudicial generalizations in the fact determination process. In the view of the dissent in O'Connor:

The accused's demonstration that information is likely to be relevant must be based on evidence, not on speculative assertions or on discriminatory or stereotypical reasoning.²³

A witness is presumed competent to testify until otherwise shown. Incompetence to testify can be shown in many ways, such as calling a doctor who has treated the witness, which do not require disclosure of private medical records. If competence is the basis for defence counsel's application for production of private medical records, then the court should first consider if there are any other reasonable alternatives of testing the witness's competence which would constitute a lesser invasion into the witness's privacy.²⁴

I would add that where the defence seeks to justify disclosure on the basis of anticipated relevance to particular issues, some inquiry is warranted into whether or not these issues are collateral to the real issues at trial. Since the defence cannot pursue inconsistencies on collateral issues, the defence is really no better off having production on that issue. It follows that failure to produce information relating only to collateral issues will not impair the accused's right to full answer and defence.²⁵

Nor does the accused have any right to search the records for potential leads. The sole ground on which third parties may be compelled to produce the records to the defence, is if they have probative value in respect of the issues in the trial, or the competence to testify of the subject of the records, that is not significantly outweighed by prejudice to the administration of justice or to the subject's privacy and equality rights.²⁶

Legislative Balancing of Rights

However, a higher threshold will still infringe the privacy and equality rights of complainants of sexual assault. In my view, disclosure of any therapeutic records created for "the purpose of assisting the survivor of sexual assault in dealing with the trauma caused by sexual assault"³⁰ infringes the equality and privacy rights of complainants. When one or more Charter rights are infringed by a legal rule, the conflicting rights must be balanced applying the principles in Dagenais v. Canadian Broadcasting Corp.³¹ The Court in considering a conflict between the accused's right to a fair trial and the freedom of expression of the media held there was no hierarchy of Charter rights.³² Once an infringement of more than one right is shown or rights conflict, the analysis moves to section 1 where the benefits of the disclosure will be weighed against the deleterious effects. Assessing the benefits of disclosure requires among other things an assessment of probative value, reliability and admissibility all of which in my view are likely to be low. On the other hand, as described above, the deleterious effects are great. .

Constitutionality turns on assessments of the degree of harm to the various rights affected. With regard to the right to full answer and defence, it can be argued that disclosure is not necessary because the complainant is subject not only to full cross-examination at trial, but also to scrutinizing questions by police and Crown counsel prior to the trial. Defence counsel will be able to use these pre-trial statements to attack credibility. On the other hand disclosure imposes severe costs on complainants. An essential aspect of fair procedures is they promote equal access to justice. Evidence rules which give preference to information more or less likely to be produced by a specific class of persons is contrary to the concept of a fair trial when the information operates to the prejudice of that group by, for instance, imposing extraordinary barriers to justice or distorting the fact determination process. In the case of children's evidence, the Supreme Court of Canada has recognized that a preference for in-court testimony and concomitant exclusion of reliable hearsay statements may deny child victims of sexual abuse access to the justice system.³³

A rule resulting in the general disclosure of therapeutic records results in the disclosure of information that is most likely to be found in sexual assault cases. Canadian experience has shown that most of the defence applications for confidential, private information of third parties has been in sexual violence cases.³⁴ An examination of the reasoning about the relevance of information in sexual violence cases indicates that myths and stereotypes about sexual assault and complainants of sexual assault is pervasive. In particular there is a foundational presumption that complainants fabricate allegations or are mistaken. Women who are Aboriginal, disabled or Black are more vulnerable to sexual exploitation. Children are especially vulnerable. The more socially powerless the victims, the more likely their lives will be documented and the more easily they will be discredited.

The majority in O'Connor have not yet engaged in this balancing. The majority do not mention equality rights although the Attorneys General of Ontario and Canada and the Coalition argued that disclosure infringes equality rights. In fact, the majority do not appear to engage in any balancing of rights in order to arrive at the threshold test of "likely relevance" for disclosure to the trial judge. Privacy rights play little role in the adoption of the threshold test. While the majority acknowledge that "a constitutional right to privacy extends to information contained in many forms of third party records", this right is not further considered in setting the threshold.³⁵ Rather the higher threshold for records in the hands of third parties than for information in the possession of the Crown is based on 1) the information is not part of the state's case to meet and 2) "third parties have no obligation to assist the defence."³⁶ If privacy rights are infringed by disclosure to the trial judge, there should be a balancing of constitutional rights as required by Dagenais before documents can be disclosed to the trial judge. The majority, however, reject the dissenters' view that records which are "likely to be relevant", should not be produced to the court until the accused has shown "that the salutary effects of ordering the documents produced to the court for inspection outweigh the deleterious effects of such production.³⁷

Balancing: The Threshold

Placing an evidentiary burden on accused to establish the relevance of the contents of records they have not seen may seen to be unfair if viewed in isolation. In a perfect world of fully rational decision makers and adequate investigatory resources for both the prosecution and the defence, it may well be that all information should be disclosed.³⁸ But in the current system where the accused does not have the right to all information, any proposal to disclose or not to disclose must be placed in the context of other limits on disclosure of information and investigatory resources. There are existing limits on the accused's right to information and "the traditional understanding of the role of the police...has not required the police to become defence investigators."³⁹ The information must be relevant. There is no right to compel witnesses to talk to the defence before the trial. There is no right to a preliminary inquiry.⁴⁰ There are several instances in which a higher threshold than a mere assertion of relevance by the defence has been required. The level of the threshold varies depending on the importance attached to the interests adversely affected by automatic disclosure. The accused must establish that a publication ban is necessary for a fair trial.⁴¹ The defence may not call Crown counsel as a witness unless they can show "there is a real basis for believing that it is likely the witness can give material evidence."⁴² There is a higher standard for intrusion in places where there is a high expectation of privacy such as law or media offices which requires, for instance, an exploration whether alternative sources have been exhausted.⁴³ When the existence of materials alleged to be in the possession of the Crown is disputed by the Crown, the defence must lay a foundation showing their existence.⁴⁴ A high threshold of need must be satisfied for the disclosure of solicitor-client communications or the identify of informers.⁴⁵ A notable instance in which no threshold is required is access to information underlying an authorization for a wiretap.⁴⁶ However, access to private information may be distinguished from information created as part of the criminal investigation by state authorities.⁴⁷ Access to the latter information is clearly necessary for accused to present their case adequately.

McLachlin J., dissenting in O'Connor on the issue of disclosure procedures, concludes:

(T)he Canadian Charter of Rights and Freedoms guarantees not the fairest of all possible trials, but rather a trial which is fundamentally fair...What constitutes a fair trial takes into account not only the perspective of the accused, but the practical limits of the system of justice and the lawful interests of others involved in the process, like complainants and the agencies which assist them in dealing with the trauma they may have suffered. Perfection in justice is as chimeric as perfection in any other social agency. What the law demands is not perfect justice, but fundamentally fair justice.⁴⁸

Balancing: Timing

(A) judge is not in a position, before the beginning of the trial, to determine whether the records in question are relevant, much less whether they are admissible, and will be unable to balance effectively the constitutional rights affected by a production order.⁵¹

Privilege

How would a qualified class privilege differ from legislation adopting the procedure set out by the dissent in O'Connor, but incorporating a higher threshold test than "likely relevance"? Under a statutory privilege, the constitutional balancing of rights would be the same for all cases since the balancing underlies the general rule. A class privilege rests on the assumption that the harm from disclosure outweighs the benefits in all cases except those where disclosure is necessary, for instance, to show the innocence of the accused. Disclosure in the individual case would depend on probative value. Under the modified O'Connor procedure, the constitutional rights would be balanced in each case by the trial judge once the information is found to be highly relevant.⁵⁷ Disclosure would depend on probative value and a balancing of rights in each case.

Is either procedure more likely to limit disclosure? Theoretically, they appear to have the same result--only highly probative and reliable records necessary to show the accused's innocence would be disclosed. But the outcome in practice will depend, first, on whether the balancing of rights under the O'Connor procedure is a meaningful one which gives weight to all Charter rights including equality rights. Second, the effectiveness of a statutory privilege will depend on judicial attitudes towards privilege. The rules of evidence construct privilege as excluding relevant evidence and thus necessarily interfering with the search for truth. As L'Heureux-Dubé J. states in L.L.A. v. A.B., "(a)ll claims of privilege hinder the search for the truth since they result in the concealment of what may be helpful and, indeed, crucial information."⁵⁸

But if judges begin with the presumption that these records are rarely relevant or necessary, a statutory privilege for records of sexual assault counsellors has advantages over the O'Connor procedure. The importance of the therapeutic relationship to the administration of justice will not be questioned just as the solicitor-client relationship is not questioned. There will not be an examination whether the relationship should be protected in the individual case. There is also no need to investigate the degree of infringement of privacy rights in the individual case. An inquiry into the witness's expectation of privacy based on, for instance, whether the communications were understood to be confidential would be avoided.

The concept of logical relevance underlying the O'Connor procedure provides little protection against disclosure. A generalization linking information to an issue can usually be constructed to make even the remotest information appear relevant. It is virtually impossible to say in advance that information will never become logically relevant in this minimal sense in the course of an inq uiry. The threshold of "likely relevance" if interpreted as logical relevance will in effect prevent disclosure to the trial judge of only the most tangential information. A class privilege would focus attention at the other end of the probative value spectrum--the issue is not whether the evidence is irrelevant, but whether it is significantly probative. An O'Connor procedure which incorporated the factors suggested above such as reliability, necessity and elimination of discriminatory beliefs would counter some of the prejudicial tendencies of the "logically probative" test and would be more likely to balance the competing interests fairly than the majority's procedures, but it is subject to the uncertainties inherent in the standard of relevance.

An examination of the constitutionality of a class privilege is beyond the scope of this paper, but I will mention a decision which will be central to the debate. In L.L.A. v. A.B. three members of a seven person hold there is no common law class privilege for counselling records of sexual assault complainants.⁵⁹ The other four justices do not address the issue of privilege, but hold that the proper procedure for production of therapeutic records in the hands of third parties is set out in R. v. O'Connor. The implications of L.L.A. v. A.B. for the constitutionality of, for instance, a statutory privilege for therapeutic records of a sexual assault crisis centre, in some respects, is ambiguous. First, L.L.A. v. A.B. considers a common law rule, not a statute, and although the common law rule is interpreted in accordance with Charter values, the opinion does not engage in a balancing of rights under section 1. Second, the class privilege considered is a very broad one which appears to include all private records of sexual assault complainants. A more narrowly defined class will raise fewer concerns about unforeseen consequences.⁶⁰ Third, the opinion does not clearly make the distinction between a qualified class privilege and an absolute class privilege. At some points the opinion considers an absolute class privilege under which no records would be disclosed, and at other times, a qualified class privilege.⁶¹ The justices do not discuss why common law qualified class privileges for solicitor-client communications and informer's identity which require disclosure only if it is necessary to show the innocence of the accused and therefore result in non-disclosure of "likely relevant" information are consistent with the right to full answer and defence. They also do not consider why counselling communications to sexual assault crisis centre therapists are not analogous to solicitor client communications as was argued by the Attorney General of Canada. The Supreme Court has held that "the relationship and the communications between solicitor and client are essential to the effective operation of the legal system....Such communications are inextricably linked with the very system which desires the disclosure of the communication."⁶² Similarly, the communications between a sexual assault survivor and her counsellor may be seen to be necessary to a properly functioning judicial system because, among other reasons, the reporting of the crime and the participation by the complainant are essential parts of the criminal process.⁶³

In my view, a qualified class privilege for records created for "the purpose of assisting the survivor of sexual assault in dealing with the trauma caused by sexual assault"⁶⁴ is well worth considering if a constitutional underpinning can be found for it. It is a well understood doctrine which avoids many of the conceptual and practical uncertainties of relevancy analysis.

Conclusion: Trial by Ordeal

An aspiration of the justice system as expressed by Wigmore in the quotation at the beginning of this comment is that we should progress from "inferior" modes of proof to those we have "come to know as better". Rational procedures have evolved even in sexual assault trials with, for instance, the abolition of corroboration requirements and the regulation of sexual history evidence. But, the O'Connor procedures are a step backwards. It is time to recognize that the modes of proof which require disclosure of private, confidential records are "not the best",⁶⁹ but a regression. Wigmore's recommendation that no sex offence should go to the jury unless the "female complainant's social history and mental makeup" have been examined by a qualified physician has emerged in a new form:--a fair trial cannot be held unless the complainant's therapeutic records have been examined by the court.⁷⁰ The next step may be that a fair trial is not possible if therapeutic records do not or no longer exist.⁷¹

Constitutional limitations on Parliament's ability to respond to community concerns should not be based on faith. Decisions which severely limit Parliament's powers must be based on full reasons. The majority fail to fulfill this duty. In my view, Parliament should design its own resolution and perform the balancing process of all Charter rights--a process which still remains to be done by the Supreme Court. The procedures created in O'Connor arise in a parliamentary vacuum. At the time the decision was released a review of disclosure of private records was underway by the Department of Justice. Continuation of this process and adoption of legislation will enable the Court to examine Parliament's balancing of interests and its measure of the broad societal view of what is fair and just in these circumstances.

*

Professor, Faculty of Law, University of British Columbia.

¹

Para. 142.

²

Para. 24.

³

Para. 22.

⁴

Para. 149.

⁵

Para. 161.

⁶

See e.g., Para. 148.

⁷

Irrationalities can be distinguished from the elimination of reasoning which results in a system with castelike features. See infra, text at note 69 and Bartlett, Katherine, "Feminist Legal Methods" (1990), 103 Harv. L. Rev. 829.

⁸

The process is that described by among others, Stephen Toulmin, The Uses of Argument (1958). The generalizations underlying the inferential link must be supported by backing. The theory must be reliable and valid. See also Schum, David, Evidence and Inference for the Intelligence Analyst (1987).

⁹

In R. v. Mohan [1994] S.C.J. No. 36, the Supreme Court of Canada held that the relevance of novel expert opinion evidence depends not only on its logical relevance but also its reliability. In summary, expert evidence which advances a novel scientific theory or technique is subjected to special scrutiny to determine whether it meets a basic threshold of reliability and whether it is necessary in the sense that the trier of fact will be unable to come to a satisfactory conclusion without the assistance of the expert.

¹⁰

Schum, David A., The Evidential Foundations of Probabilistic Reasoning (1994) at 12.

¹¹

Para 109 quoting L'Heureux-Dubé J. dissenting in R. v. Osolin [1993] 4 S.C.R. 595, at pp. 622-2.

¹²

Smith, Dorothy, Conceptual Practices of Power: A Feminist Sociology of Knowledge (1990).

¹³

Code, Lorraine, What Can She Know (1991) at 203.

¹⁴

See discussion of David Paciocco infra, note 74 who argues that we should be wary of eliminating generalizations alleged to be discriminatory from judicial decision making if they are sometimes true.

¹⁵

Sunstein, Cass, "On Finding Facts" in James Chandler, Arnold I Davidson, And Harry Harootunian, Questions Of Evidence: Proof, Practice, And Persuasion Across The Disciplines (1994) at 190.

¹⁶

Code, Lorraine, supra, note 65 at 200-201.

¹⁷

Sunstein, supra, note 67 at 195: "Frequently disagreements about 'the fact' are really disagreements about the social world that would be created if we treated the facts one way rather than another." at 194.

¹⁸

Ekos Research Associates Inc., Report on the treatment if Sexual Assault Cases in Vancouver, British Columbia (September 1988) (Study funded by the Research Section, Department of Justice, Canada) (Found that the "perceived moral character and conduct of the complainant appear to influence the founding decision" by the police at 50 and 174. Where the police report "indicated that the complainant was vulnerable due to physical or mental disabilities, the report almost three times more likely to be considered unfounded" at 53.)

¹⁹

Supra, notes 24 and 52.

²⁰

R. v. Seaboyer supra, note 24.

²¹

The majority, however, can be interpreted as deferring any consideration of the operation of discriminatory beliefs to the second stage of the decision. See supra, notes 87 to 89.

²²

Paciocco, David, "Techniques For Eviscerating The Concept Of Relevance: Reply And Rejoinder To 'Sex With The Accused On Other Occasions: The Evisceration Of Rape Shield Protection'" (1994), 33 C.R. (4th) 365: "..inferences that constitute overgeneralizations, ...may, in some circumstances, contain valuable information; we must not throw the baby out with the bath water".

²³

Para. 140.

²⁴

Para. 149.

²⁵

Para. 161

²⁶

Para. 183.

²⁷

Para 183: "Nor does the accused have any right to search the records for potential leads."

²⁸

R. v. O'Connor supra, note 3 at 148-9.

²⁹

See e.g., R. v. O'Connor (No.1) (1994), 89 C.C.C.(3d) 109 (B.C.C.A.).

³⁰

Joo, Anita, "Broadening the Scope of Counselor-Patient Privilege to Protect the Privacy of the Sexual Assault Survivor" (1995), 32 Harv. J. on Legislation 255 at 298.

³¹

[1994] 3 S.C.R. 835.

³²

Id. Freedom of expression must not always give way to fair trial rights of the accused. The court adopted the following test: "A publication ban should only be ordered when: (a) Such a ban is necessary in order to prevent a real and substantial risk to the fairness of the trial, because reasonably available alternative measures will not prevent the risk; and (b) The salutary effects of the publication ban outweigh the deleterious effects to the free expression of those affected by the ban" at Para. 73.

³³

See e.g., R. v. L.(D.O.) [1993] 4 S.C.R. 419.

³⁴

See the conflicting inferences drawn from this fact by the majority and dissent, supra, notes 46 and 48.

³⁵

Para. 17.

³⁶

Para. 19.

³⁷

LLA v. AB supra, note 28 at Para. 90

³⁸

I am indebted to Christine Boyle for this argument.

³⁹

Report of the Attorney General's Advisory Committee on Charge Screening, Disclosure and Resolution Discussions Ontario, 1993 (The Martin Report) at 150.

⁴⁰

Atrens, Jerome, The Charter and Criminal Procedure: The Application of Sections 7 and 11 (1989)at 11-23 citing Re Arviv (1985), 19 C.C.C.(3d) 395 (Ont.C.A.).

⁴¹

C.B.C. v. Dagenais supra, note 83.

⁴²

R. v. Harris (1994) (Ont. C.A.).

⁴³

Descoteaux v. Mierzwinski [1992] 1 S.C.R. 860. C.B.C. v. New Brunswick [1991] 3 S.C.R. 459; C.B.C. v. Lessard [1991] 3 S.C.R. 421.

⁴⁴

R. v. Chaplin (1995) 36 C.R.(4th) 201 (S.C.C.) (The Crown refused to say whether wiretaps relating to other investigations existed. The court did not require the Crown to conduct a search to determine if the records existed. Sopinka J. characterized this a case in which "the existence of material which is alleged to be relevant is disputed by the Crown.")

⁴⁵

See R. v. Garofoli [1990] 2 S.C.R. 1421 (S.C.C.) (The relevance of the identity of the informer to the accused's case is balanced against prejudice to the informer and to the public interest in law enforcement.) R. v. Scot [1990] 3 S.C.R. 979.

⁴⁶

R. v. Durette [1994] 1 S.C.R. 469.

⁴⁷

The dissent at Para. 141 addresses this issue: "However, in those cases, the accused sought access to by the State as part of its investigation; that situation can hardly be compared to the situation of an accused who demands access to therapeutic or other private records created and held by a third party. The records here in question are not within the possession or control of the Crown, do not form part of the Crown's "case to meet", and were created by a third party for a purpose unrelated to the investigation or prosecution of the offence. In my opinion, it cannot be assumed that such records are likely to be relevant, and if the accused is unable to show that they are, then the application for production must be rejected as it amounts to nothing more than a fishing expedition."

⁴⁸

Para. 193 referring to R. v. Harrer [1995] S.C.J. No. 81 ( File No. 24141).

⁴⁹

R. v. Gingras (1992), 71 C.C.C.(3d) 53 (Alta.C.A.).

⁵⁰

Para. 181

⁵¹

Para. 812.

⁵²

The Ontario Court of Appeal in Allen v. Morissette (1993), 69 O.A.C. 76, (sub nom. R. v. Morissette) 79 C.C.C. (3d) 444, at p. 445: "The admissibility of the documents for which privilege is claimed is a matter for the trial judge to decide. The trial judge is in the best position to assess the impact of the statutory privilege, the relevance of the documents and, if necessary, to conduct a voir dire in order to determine whether the information for which privilege is claimed should be disclosed."

⁵³

Other rules which exclude information can play a role such as the regulation of admissibility of sexual conduct of the complainant: Criminal Code s. 276, 276.1-276.5. See the quotation supra, note 24.

⁵⁴

Criminal Code s. 700.

⁵⁵

See L'Heureux-Dubé J. in R. v. Osolin: quoted supra at note 63.

⁵⁶

Supra, note 82.

⁵⁷

There is still a role for second stage of balancing even if the evidence is highly probative because judicial examination of the records will enable the trial judge to re-assess probative value and to accurately measure the prejudicial effects to Charter rights.

⁵⁸

Para. 66.

⁵⁹

LLA v. AB supra, note 28.

⁶⁰

Supra, note 28 at para 71: "To extend a class privilege to (all actors involved in sexual assault counselling relationships) might mean that information gathered in group therapy or communications relating to mental health exchanged on pay-per-call telephone service lines could attract such privilege.".

⁶¹

The discussion of United States state laws appears to confound absolute with class privilege. While some absolute privileges have been held to be unconstitutional, class privileges with defined exceptions have been approved. See Joo, supra, note 108

⁶²

R. v. Gruenke[1991] 3 S.C.R. 326 (S.C.C.).

⁶³

See the factum of the intervener, the Attorney General of Canada in LLA v. AB, supra, note 28.

⁶⁴

Supra, note 82

⁶⁵

R. v. Plant, [1993] 3 S.C.R. 281 at 293 quoted by the dissent in O'Connor at para. 118.

⁶⁶

Bartlett, supra, note 1 at 19.

⁶⁷

See for instance the dissent's discussion of three cases at paras. 125-127 involving applications for private records in a murder case and two cases of assault on a police officer in which disclosure was denied. The reasons for denying the applications included: the application was no more than a fishing expedition, it was not based on evidence and the information was unreliable.

⁶⁸

"The medieval ordeal is a subject of great intrinsic interest and fascination. It is one of the more dramatically alien practices of medieval society and, as such, it demands and yet resists explanation. For those concerned to make the imaginative leap into a past society, the order is a hurdle and a challenge. Its 'otherness' represents an explanatory problem." Bartlett, supra, note 43 at 1.

⁶⁹

"The contrast, it may be noted, between employing rational and non-rational modes of proof is after all not between the use of scientific reasoning and the employment of superstitious ordeals; it is rather between employing the best standards we know and those which we realize are not the best." Wigmore On Evidence, Vol. I, para. 9, at 34, (1905, Canadian Edition).

⁷⁰

Wigmore, J.J., Evidence in Trials at Common Law, vol. 3A, s. 924, Chadbourne Rev. Ed. 1970 at 736-7.

⁷¹

But see R. v. Carosella [1995] O.J. No. 2927. (The Ontario Court of Appeal held that the accused was not denied the right to full answer and defence by the destruction of records by a sexual assault crisis centre. The Court concluded that a realistic appraisal of the probable effect of the lost notes could not support the conclusion the accused's rights were compromised. The court concluded that "the report is not a verified account of what the complainant said; it is not a written statement of the complainant and other material was available to the defence such as the complainant's statements to the police, statements of other Crown witnesses and the preliminary inquiry evidence.")

NOTE: This article has been divided into two parts due to its length. If you wish, you can read the FIRST PART.

---

Top

This page was created 13 February 1997 and updated 4 June 1997.