Toronto (City) v. Canadian Union of Public Employees, Local 79

Toronto (City) v. Canadian Union of Public Employees, Local 79 Collection Decisions of the Court of Appeal Date 2001-08-10 Docket numbers C35112 Judges Labrosse, Jean-Marc; Doherty, David H.; Feldman, Kathryn N. Subject Civil Decision Content DATE: 20010810 DOCKET: C35112 COURT OF APPEAL FOR ONTARIO LABROSSE, DOHERTY and FELDMAN JJ.A. BETWEEN: ) ) Douglas J. Wray and CITY OF TORONTO ) Craig Flood ) for the appellant Applicant/ ) Respondent in Appeal ) Kari M. Abrams and ) Jason Hanson - and - ) for the respondent ) CANADIAN UNION OF PUBLIC ) Leonard Marvy, EMPLOYEES, LOCAL 79 ) Lucy McSweeney and ) Karin Rasmussen Respondent/ ) for the intervener Appellant in Appeal ) ) ) - AND - ) ) DOUGLAS C. STANLEY ) Heard: April 24, 2001 ) Respondent ) ) - and - ) ) ONTARIO MINISTRY OF THE ) ATTORNEY GENERAL ) ) Intervener ) On appeal from the decision of the Superior Court of Justice (Div. Ct.) (O’Driscoll, MacFarland and Crane JJ.) released May 5, 2000. DOHERTY J.A.: [1] This is another case in which the Court must decide whether a litigant can relitigate an issue finally determined in a prior judicial proceeding. [1] [2] At a grievance arbitration, the employer relied on the employee’s criminal conviction in order to establish that the employee had engaged in conduct which justified his dismissal. The arbitrator ruled that while the employee’s criminal conviction was admissible as evidence that he did the acts alleged in the criminal charge, the Union, on the employee’s behalf, could relitigate that very issue in the grievance arbitration. After hearing evidence and reviewing the record of the criminal proceedings, the arbitrator decided that the employee had not committed the acts for which he had been previously convicted. He ordered the employee reinstated. [3] The Divisional Court, for reasons reported at (2000), 187 D.L.R. (4th) 323, allowed the employer’s application for judicial review and quashed the order reinstating the employee. This court granted leave to appeal from the order of the Divisional Court. I would dismiss the appeal. I [4] Glenn Oliver (Oliver) worked for the respondent, City of Toronto (Toronto), as a recreation instructor. He began his employment in 1970. In July 1994, Oliver was charged with sexually assaulting a young boy (the complainant) who had met Oliver while participating in recreational programs offered by Toronto and supervised by Oliver. The charges alleged that the offences had occurred between 1985 and 1987. [5] Mr. Oliver pleaded not guilty. His trial proceeded in January 1996 before Ferguson J. of the Ontario Court (Gen. Div.), sitting without a jury. The complainant testified and was cross-examined. Oliver testified in his own defence and while acknowledging a close relationship with the complainant, denied any sexual conduct. Oliver also called several defence witnesses, including witnesses who testified to his good character. [6] Ferguson J. observed that the outcome of the criminal proceedings turned very much on the credibility of the complainant and of Oliver. He reviewed the evidence and concluded that the complainant was credible, but that Oliver was not. Ferguson J. found that over a prolonged period, Oliver “slowly groomed” the complainant to participate in various sexual activities, including oral sex. He found that sexual activity occurred in the community centre where Oliver worked and in his apartment. Ferguson J. found Oliver guilty and imposed a sentence of 15 months. [7] Oliver’s appeal to this court from his conviction and sentence was dismissed. He did not seek leave to appeal to the Supreme Court of Canada. [8] Toronto fired Oliver on January 16, 1996, a few days after he was convicted. Mr. Zanetti, Oliver’s supervisor, wrote to Oliver and informed him that his employment could not continue in light of his conviction for sexual assault. In Mr. Zanetti’s view, that conviction was directly linked to Oliver’s employment with Toronto. [9] Oliver grieved his dismissal pursuant to the terms of the Collective Bargaining Agreement between the appellant, Canadian Union of Public Employees, Local 79 (CUPE), and Toronto. The grievance arbitration started in June 1998, some 2 1/2 years after the conviction. [10] In the grievance proceedings, Toronto had to establish on the balance of probabilities that it had reasonable cause to fire Oliver. It relied on Oliver’s conviction as conclusive proof that he had sexually assaulted a young child under his charge during the course of his employment. [2] [11] CUPE acknowledged that Toronto had just cause to fire Oliver if he had sexually assaulted the complainant as found in the criminal proceeding. It contended, however, that the grievance proceeding was entirely separate from the criminal proceeding, as neither CUPE nor Toronto had been a party to the criminal trial. CUPE argued that Toronto was obliged to prove on the balance of probabilities that Oliver had sexually assaulted the complainant, and that Toronto could not rely on the criminal conviction which counsel for CUPE described as no more than Ferguson J.’s “opinion” that Mr. Oliver had sexually assaulted the complainant. [12] The arbitrator was asked to decide the admissibility of Oliver’s prior conviction at the outset of the arbitration. He held that the conviction was admissible as prima facie, but not conclusive proof that Oliver had sexually assaulted the complainant during the course of Oliver’s employment. The arbitration proceeded following that ruling. Toronto called Mr. Zanetti, but did not call the complainant. Toronto relied on the conviction of Oliver to establish that he had in fact sexually assaulted the complainant during the course of his employment. Counsel for CUPE called Oliver and proposed to ask him whether he had sexually assaulted the complainant. Counsel for Toronto objected, contending that Oliver could not deny his guilt in the face of his criminal conviction. The arbitrator rejected this submission holding: … I have jurisdiction under the Labour Relations Act, I am obliged to exercise that jurisdiction, and I will confine myself to the exercise of that jurisdiction. It is not an abuse of process under the collective agreement and the Labour Relations Act for the grievor to maintain, in these proceedings, that he did not do things that he was found to have done in other proceedings. My reading of the Del Core case [Re Del Core and Ontario College of Pharmacists (1985), 51 O.R. (2d) 1 (Ont. C.A.), leave to appeal to S.C.C. refused, [1986] 1 S.C.R. viii] is that the Court accepts that it is open to any person in a situation like that of the grievor to do that. Further, the criminal trial transcript can be used to challenge his credibility in that testimony. My conclusion is that the law recognizes that it is open to me in these proceedings to weigh all the evidence and come to a different conclusion on the facts than was arrived at in some other proceedings. [Emphasis added.] [13] Oliver, as he had at the criminal trial, denied that he had sexually assaulted the complainant and attempted to explain his relationship with him. Although the complainant had not testified before the arbitrator, his testimony at the criminal trial, [3] the letters he had written to Mr. Zanetti [4] and Mr. Zanetti’s notes of a meeting with the complainant were filed on the arbitration. The arbitrator thoroughly reviewed Oliver’s testimony, the transcript of the criminal proceedings, including the complainant’s evidence, and the letters written by the complainant. He decided that the complainant, whom he had never seen, was a liar, and that Oliver was telling the truth. [14] The arbitrator gave detailed reasons for his credibility findings and for his rejection of the findings made in the criminal trial. He also accepted an in terrorem argument advanced on behalf of CUPE: I accept the argument made by Counsel for the Union to the effect that if we were to conclude on the evidence in this case that the grievor [Mr. Oliver] had committed the alleged assaults, then no man who, in the course of his employment working with children, develops a close but professional relationship with a child, and on occasion is alone with that child, will be safe from false accusations of impropriety. In my view, in the proper weighing of the evidence, and the application of the tests for credibility, there must be something more to support the accusation than there is in this case. He then said: On reviewing all of the evidence before me, I have to conclude that the grievor [Mr. Oliver] is being truthful in denying the allegation made by [the complainant]. I have to conclude that [the complainant’s] story is dramatic and contrived and is simply not consistent with the probabilities which a practical and informed person would readily recognize as reasonable in light of all the evidence. I conclude that the [complainant’s] story is lacking in harmony with the preponderance of probabilities disclosed by all the facts and circumstances in this case. Because I conclude that the grievor is telling the truth and [the complainant] is not, I must find that there are no grounds for the grievor’s termination. The grievance is allowed. I direct that the grievor be reinstated in employment with full compensation. … [Emphasis added.] [15] In allowing Toronto’s application for judicial review, the Divisional Court held that the arbitrator should not have gone behind the criminal conviction and made his own assessment of whether Oliver had sexually assaulted the complainant. Writing for the court, O’Driscoll J. relied on the rule against collateral attack on orders of a superior court, issue estoppel and abuse of process. II The Standard of Review [16] O’Driscoll J. found it unnecessary to decide the applicable standard of review, indicating at p. 342: Whether the benchmark is “correctness” or “patently unreasonable” … the Applicant [Toronto] has met the standard. [17] I think it is necessary to determine the appropriate standard of review. The arbitrator’s finding that Oliver’s conviction for sexual assault was admissible in the arbitration proceedings as evidence that Oliver sexually assaulted the complainant, but did not foreclose relitigation of that very issue, enjoys considerable support in the case law, [5] particularly in the reasons of Houlden J.A. in Re Del Core and Ontario College of Pharmacists, supra, dissenting on another ground, at pp. 14-15, 17-18, leave to appeal to S.C.C. refused, [1986] 1 S.C.R. viii. Although for reasons I will develop, I think the arbitrator was wrong in law, his view of the law was not patently unreasonable. [18] Counsel for CUPE submits that the arbitrator’s decision should be afforded maximum deference and must stand unless found to be patently unreasonable. Counsel submits that the grievance process is designed to provide for the expeditious and fair resolution of labour disputes arising out of collective bargaining agreements by those with expertise in that field, and that any type of judicial intervention other than the most limited form of judicial review unduly interferes with that process. Counsel also contends that for the same reason, the arbitrator is given a wide latitude on questions of the admissibility of evidence and the weight to be given to evidence. [19] In making his case for limited review, counsel characterizes the issue before the arbitrator as “whether there was just or reasonable cause for the discharge”. Counsel contends that as in any arbitration, this arbitrator had to resolve a variety of legal and factual issues in the course of deciding whether Toronto had shown reasonable cause for Oliver’s discharge. Those issues included the admissibility and effect of Oliver’s previous criminal conviction. Counsel submits that the patently unreasonable standard applies not only to the ultimate determination of whether reasonable cause was shown, but also to other decisions made by the arbitrator in the course of deciding whether Toronto had established reasonable cause. He relies on Toronto (City) Board of Education v. O.S.S.T.F., District 15, [1997] 1 S.C.R. 487, where Cory J., for the majority, said at p. 505: A decision as to whether there is “just cause” for discipline of an employee comes within the jurisdiction of an arbitration board, and therefore can only be set aside if it is patently unreasonable. … [20] Cory’s J.’s words must, however, be placed in their proper context. In that case, the employee had engaged in misconduct. The sole question for the Tribunal was whether his discharge was an appropriate penalty. As Cory J. observed, at p. 510: … The only question was whether the disciplinary measure of dismissal was too harsh. [21] A review of the appropriateness of disciplinary action taken by an employer is at the very core of the arbitrator’s function and fully engages the arbitrator’s expertise. Not every issue that arises in the course of a grievance arbitration can be so described. Cory J. said, at p. 506: … It has been held on several occasions that the expert skill and knowledge which an arbitration board exercises in interpreting a collective agreement does not usually extend to the interpretation of “outside” legislation. The findings of a board pertaining to the interpretation of a statute or the common law are generally reviewable on a correctness standard. [citations omitted] An exception to this rule may occur where the external statute is intimately connected with the mandate of the tribunal and is encountered frequently as a result. [Citation omitted.] [Emphasis added.] [22] In this case, the arbitrator’s decision that CUPE could relitigate the issue decided in the criminal trial turned on his understanding of the common law rules and principles governing the relitigation of issues finally decided in a previous judicial proceeding. Far from supporting the appellant’s submission that every decision made by an arbitrator in the course of a grievance arbitration is due maximum deference, the judgment of Cory J. suggests that a decision like the one made by this arbitrator is “generally reviewable” on a correctness standard. See also United Brotherhood of Carpenters and Joiners of America, Local 579 v. Bradco Construction Ltd., [1993] 2 S.C.R. 316 at 336. [23] The appellant’s submission erroneously assumes that the standard of review applicable to a tribunal’s ultimate decision must be applied to every decision made by the tribunal in the course of arriving at its ultimate conclusion. It is implicit in the decision of Cory J. in Toronto (City) Board of Education v. O.S.S.T.F., District 15, supra, that different standards of review may apply to different decisions made by a tribunal in the course of a proceeding. The same point is expressly made in Canadian Broadcasting Corporation v. Canada (Labour Relations Board), [1995] 1 S.C.R. 157 at 187-8, where Iacobucci J., for the majority, said: While the Board may have to be correct in an isolated interpretation of external legislation, the standard of review of the decision as a whole, if that decision is otherwise within its jurisdiction, will be one of patent unreasonableness. Of course, the correctness of the interpretation of the external statute may affect the overall reasonableness of the decision. Whether this is the case will depend on the impact of the statutory provision on the outcome of the decision as a whole. [24] While Iacobucci J. spoke of the “isolated interpretation of external legislation”, his words have equal application to a tribunal’s interpretation of common law principles that do not engage the expertise of the tribunal. Consequently, while it is correct to assert, as CUPE does, that the arbitrator’s decision that Toronto had not established reasonable cause to fire Oliver is owed maximum deference on judicial review, that does not mean that the same standard of review should be applied to the arbitrator’s decision that CUPE could relitigate issues decided in the criminal trial. If that decision is reviewable on a correctness standard, the court must determine whether the arbitrator was correct and if he was incorrect, whether his error made his determination that Toronto had not established just cause patently unreasonable: see also Ontario (Attorney General) v. Ontario Public Service Employees Union (2000), 52 O.R. (3d) 77 (C.A.), per Moldaver J.A. concurring at p. 88. [25] The standard of review cannot be fixed by reference to any single isolated factor, but requires a consideration of the interaction of a variety of factors. Apart from constitutional limits which need not be explored here, it is for the legislature to decide the basis upon which courts may review the decisions of tribunals. In deciding the appropriate standard of review, the court must determine the extent to which the legislature intended that a particular decision of a tribunal should be final and beyond judicial review. The search for the appropriate standard of review is a search for that sometimes elusive legislative intent: Pezim v. British Columbia (Superintendent of Brokers), [1994] 2 S.C.R. 557 at 589-90; Canadian Broadcasting Corporation. v. Canada (Labour Relations Board), supra, at 178-79; Pasiechnyk v. Saskatchewan (Workers’ Compensation Board), [1997] 2 S.C.R. 890 at 904-905; Committee for the Equal Treatment of Asbestos Minority Shareholders v. Ontario (Securities Commission), 2001 S.C.C. 37, at paras. 46-49; and D. Jones, A. DeVillars, Principles of Administrative Law, 3rd ed. (Carswell, 1999) at pp. 449-450. [26] The methodology to be used in determining the extent to which the legislature intended that courts should review tribunals’ decisions was set out in some detail in Canada (Director of Investigation and Research) v. Southam Inc., [1997] 1 S.C.R. 748 at 765-75. That case involved a statutory appeal from a tribunal’s decision. The same methodology has been applied to judicial review of tribunal decisions: Pushpanathan v. Canada (Minister of Citizenship and Immigration), [1998] 1 S.C.R. 982. In Pushpanathan, supra, at pp. 1003-1012, Bastarache J., for the majority, identified four categories of factors that should be considered in determining the appropriate standard of review: · the interpretation of any privative clause which may apply; · the extent to which the decision engaged the expertise of the Board; · the purpose of the Act as a whole and the purpose of any particular provisions implicated in the challenged decision; and · the nature of the problems giving rise to the decision under review. [27] The categories identified by Bastarache J. should not be treated as airtight compartments. Nor will the various factors necessarily send the same signal as to the appropriate level of judicial review. Bastarache J. observed, at p. 1004: … [T]his court has determined that the task of statutory interpretation requires a weighing of several different factors, none of which are alone dispositive, and each of which provides an indication falling on a spectrum of the proper level of deference to be shown to the decision in question. … [28] I begin my consideration of the factors identified by Bastarache J. with s. 48(1) of the Labour Relations Act 1995, S.O. 1995, c. 1, Schedule A. That section shelters the arbitrator’s decision under a finality clause which reads: Every collective agreement shall provide for the final and binding settlement by arbitration, without stoppage of work, of all differences between the parties arising from the interpretation, application, administration or alleged violation of the agreement, including any question as to whether a matter is arbitable. [Emphasis added.] [29] The language of s. 48(1) offers support for a deferential standard of review of decisions made by arbitrators. It is not, however, a full privative clause and may be usefully compared to s. 116 of the Labour Relations Act 1995. That section applies to decisions of the Ontario Labour Relations Board and specifically interdicts review of those decisions by the court. 116. No decision, order, direction, declaration or ruling of the Board shall be questioned or reviewed in any court, and no order shall be made or process entered, or proceedings taken in any court, whether by way of injunction, declaratory judgment, certiorari, mandamus, prohibition, quo warranto, or otherwise, to question, review, prohibit or restrain the Board or any of its proceedings. [Emphasis added.] [30] The predecessors to ss. 48(1) and 116 were considered in Dayco (Canada) Ltd. v. CAW-Canada, [1993] 2 S.C.R. 230 at 264, 268. LaForest J. saw both provisions as manifesting the legislature’s intent that judicial review of the decisions affected by those sections should be limited. He described the predecessor to s. 48(1), however, as providing a more “limited shield against judicial review for decisions of an arbitrator” than that provided to decisions of the Labour Relations Board by the predecessor to s. 116. In Bradco, supra, at pp. 331-35, Sopinka J., for the court, also drew a distinction between finality clauses and full privative clauses. He described a finality clause framed in terms very similar to s. 48(1) as contemplating a standard of appellate review falling “somewhere between a full privative clause and a clause providing for full review by way of appeal”. [31] Although s. 48(1) of the Labour Relations Act does not manifest an intention to exclude judicial review entirely, it does indicate an intention that courts should not become readily involved in the review of an arbitrator’s decisions. In so doing, it is consistent with the overall purpose of the grievance process. Mandatory arbitration of grievances aims at arriving at an efficient and cost-effective resolution of disputes arising out of a collective bargaining agreement, while enabling the parties to that agreement to maintain an ongoing working relationship. Unfettered judicial review of grievance decisions is inconsistent with those goals: Bradco, supra, at p. 334. The finality clause reflects the legislature’s determination that disputes arising out of collective bargaining agreements should be resolved by arbitrators and not by the courts. [32] Standing alone, s. 48(1) of the Labour Relations Act 1995 would suggest a deferential standard of review in the case at bar. The other factors outlined in Pushpanathan, supra, however, point towards a correctness standard of review. The most important of these factors is the relative expertise of the arbitrator and the courts: Pushpanathan, supra, at p. 1006. The arbitrator’s decision that the appellant could relitigate an issue finally decided in the previous criminal case had nothing to do with labour law principles, the interpretation of the collective bargaining agreement or the interpretation of any labour law statute. The arbitrator’s decision turned on his understanding of the common law principles derived from judge-made law having no special application in the labour law field. While arbitrators are no doubt competent to apply common law principles and must do so on a daily basis, they have no special expertise in that area. [33] Res judicata and related principles which operate to preclude relitigation of decided issues are judge-made doctrines. They reflect a balancing of fairness and finality concerns, both of which are fundamental to the maintenance of the rule of law and the effective administration of justice. The development and application of these principles are quintessentially judicial functions. I find nothing in the applicable legislation or in the terms of this collective bargaining agreement that would suggest that the legislature or the parties intended that the courts, as guardians of the common law, should defer to the arbitrator’s view of the common law. The court’s relative expertise in the determination and application of the principles in issue on this appeal speaks strongly in favour of a correctness standard of review. [34] The second factor favouring the correctness standard of review is the nature of the problem under review. As Bastarache J. observed in Pushpanathan, supra, at p. 1007, this factor is “closely interrelated” with expertise. Decisions that turn on questions of law alone, especially questions having broad application and, therefore, significant jurisprudential value are often subject to a correctness standard of review. In Pushpanathan, supra, at p. 1012, Bastarache J. put it this way: … In the usual case, however, the broader the propositions asserted, and the further the implications of such decisions stray from the core expertise of the tribunal, the less likelihood that deference will be shown. Without an implied or express legislative intent to the contrary as manifested in the criteria above, legislatures should be assumed to have left highly generalized propositions of law to courts. [35] In this case, the arbitrator had to decide whether CUPE could relitigate an issue decided in Oliver’s criminal trial. As a first step in making that decision, he had to settle on the legal principles to be applied. Once he identified those principles, he then had to apply them to the facts before him. The first step in this two-step process involves a question of law alone: Canada (Director of Investigation and Research) v. Southam Inc., supra, at p. 766-67. Not only is it a question of law alone, it is, as the recent decisions of this court demonstrate, [6] a question of general significance with implications that reach into many areas of the law. To the extent that the arbitrator was required to identify the appropriate legal principles to be applied to CUPE’s claim that it could relitigate the issue decided against Oliver in the criminal proceedings, the nature of the problem faced by the arbitrator strongly suggests a correctness standard of review. [7] [36] The final factor identified in Pushpanathan, supra, offers some support for a correctness standard of review on this application and is also closely connected to expertise. Courts are best suited to decide issues that settle specific disputes between clearly identifiable parties. Tribunals whose decisions are not the product of an adversarial process, reflect far ranging non-legal policy considerations and address the interests of the variety of constituencies, are much less amenable to judicial review. This is particularly true if the tribunal’s decision rests on an expertise possessed by it and not the court. In Pushpanathan, supra, at p. 1009, Bastarache J. said: …While judicial procedure is premised on a bipolar opposition of parties, interests, and factual discovery, some problems require the consideration of numerous interests simultaneously, and the promulgation of solutions which concurrently balance benefits and costs for many different parties. Where an administrative structure more closely resembles this model, courts will exercise restraint. The polycentricity principle is a helpful way of understanding the variety of criteria developed under the rubric of the “statutory purpose”. [37] The decision challenged here arises from a proceeding convened to decide whether Toronto had just cause for firing Oliver. The dispute was “bipolar” and turned on a very specific inquiry. There is nothing in the nature of the decision rendered by the arbitrator or the process involved in arriving at that decision that suggests a high level of judicial deference. [38] Having considered the factors analyzed in Pushpanathan, supra, I am satisfied that a correctness standard of review must be applied to the arbitrator’s decision in so far as it involved a determination of the legal principles to be applied in deciding whether Oliver could relitigate the issue decided against him in the criminal proceeding. III The Substantive Issue [39] There are two related but distinct questions that arise from Toronto’s reliance on Oliver’s prior conviction to establish just cause for dismissal: · Was the prior conviction admissible as evidence that Oliver sexually assaulted the complainant? · If the conviction was admissible for that purpose, could CUPE relitigate the finding that Oliver sexually assaulted the complainant? [40] The first question is a relatively narrow evidentiary one and is now governed by statute. The second question is a much broader question of public policy. The second question addresses the interaction of the various adjudicative processes which may be brought to bear on a specific fact situation. It asks how decisions made in one adjudicative context should be treated when the same issue arises for determination in a different adjudicative context. [41] Section 22.1 of Ontario’s Evidence Act, R.S.O. 1990, c. E.23, governs the admissibility of Oliver’s prior conviction as evidence that he had sexually assaulted the complainant: 22.1(1) Proof that a person has been convicted or discharged anywhere in Canada of a crime is proof, in the absence of evidence to the contrary, that the crime was committed by the person, if, (a) no appeal of the conviction or discharge was taken and the time for an appeal has expired; or (b) an appeal of the conviction or discharge was taken but was dismissed or abandoned and no further appeal is available. (2) Subsection (1) applies whether or not the convicted or discharged person is a party to the proceeding. [8] [Emphasis added.] [42] Section 22.1 provides that the prior conviction is “proof” that the convicted person committed the crime absent “evidence to the contrary”. [9] While s. 22.1 clearly contemplates that a convicted person will be able to challenge the validity of a conviction in a subsequent proceeding, the section does not speak to the circumstances in which that challenge will be permitted. The distinction between admissibility of evidence of a prior conviction and a convicted person’s entitlement to relitigate that conviction in a subsequent proceeding was clearly drawn by Blair J.A. in Re Del Core, supra, at p. 22: Some confusion appears to have arisen between the rule that evidence of prior convictions is admissible and the doctrine of abuse of process. … It was argued that such evidence could only be admitted where it would be an abuse of process to challenge the conviction. This is not so. The admissibility of such evidence is not dependent on a determination that it would be an abuse of process to attack the conviction. As I have explained above, evidence of prior convictions is admissible in all cases, where it is relevant. The abuse of process doctrine can only be evoked, in particular cases, to prohibit rebuttal of such evidence. [Emphasis added.] [43] In his reasons, O’Driscoll J. at pp. 335-36 correctly observed that s. 22.1 of the Evidence Act applied on the arbitration. It does not appear to me that his conclusion that Oliver could not relitigate the finding that he had sexually assaulted the complainant turned on his interpretation of s. 22.1. The arbitrator did not refer to s. 22.1, but relied on cases decided before the enactment of s. 22.1, including Re Del Core, supra, per Blair J.A. at 20-22, per Houlden J.A. at 13-17, to find that proof of a prior conviction was admissible as “strong prima facie” but not “conclusive” proof that Oliver had sexually assaulted the complainant. [44] The arbitrator’s ruling that the prior conviction was admissible as evidence that Oliver had sexually assaulted the complainant accords with s. 22.1(1). His further conclusion that the prior conviction is “strong prima facie evidence” is not quite in line with the section. It creates a rebuttable presumption of law. In the absence of evidence to the contrary, the prior conviction is “proof” that the person committed the crime. If there is no evidence to the contrary, the court or tribunal must conclude that the person committed the crime: R. v. Proudlock, [1979] 1 S.C.R. 525, per Pigeon J. at 547-49; and J. Sopinka, S. Lederman, A. Bryant, The Law of Evidence in Canada, 2nd ed. (Butterworths, 1999) at pp. 104-106. [45] The arbitrator’s slight understatement of the evidentiary effect of evidence of a prior conviction in the absence of any evidence to the contrary is of no consequence if the arbitrator was correct in holding that CUPE could challenge the conviction. If CUPE could relitigate Oliver’s culpability, there was “evidence to the contrary” before the arbitrator that was capable of rebutting the legal presumption created by s. 22.1. The correctness of the arbitrator’s decision to allow CUPE to relitigate the conviction turns on the answer to the public policy question described above. (ii) The public policy question (a) The Divisional Court’s Answer [46] The Divisional Court came down strongly in favour of a policy which would preclude relitigation in the arbitration context of issues decided in the criminal courts. O’Driscoll J. made it clear that in his view the arbitrator had no business reviewing the correctness of a conviction and substituting his own view for that of the criminal court. In reaching that result, O’Driscoll J. relied on issue estoppel, abuse of process and the doctrine prohibiting collateral attack on orders of superior courts. [47] In this jurisdiction, issue estoppel applies only where the parties to the second proceeding where issue estoppel is said to preclude relitigation, were parties to or privies of parties to the first proceeding. Even if it could be said that CUPE was Oliver’s privy, Toronto played no role in the criminal proceedings and had no relationship to the Crown: Angle v. Minister of National Revenue (1974), 47 D.L.R. (3d) 544 at 555 (S.C.C.); Minott, supra at 330, 336-40 (C.A.); Danyluk, supra, at 238 (C.A.). Issue estoppel did not foreclose CUPE’s relitigation of the finding made in the criminal proceeding. [48] I also do not agree that the rule against collateral attack on orders of superior courts provides a free-standing basis upon which to preclude relitigation. Not all collateral challenges are offensive: R. v. Consolidated Mayburn Mines Ltd. (1998), 123 C.C.C. (3d) 449 (S.C.C.); Braithwaite v. Nova Scotia Public Service Long Term Disability Plan Trust Fund (1999), 176 N.S.R. (2d) 173 (C.A.); and D. Lange, The Doctrine of Res Judicata in Canada, (Butterworths, 2000) at pp. 369-374. Describing CUPE’s attempt to relitigate Oliver’s culpability as a collateral attack on the order of the criminal court does not assist in determining whether that relitigation should be permitted. [49] I do agree with O’Driscoll J.’s ultimate conclusion that public policy should not permit CUPE to relitigate Oliver’s guilt in the grievance arbitration. I will develop my reasons for coming to that conclusion after I address the arbitrator’s reasons for holding that CUPE could relitigate. (b) The Arbitrator’s Answer [50] I have set out above at paragraph 12 the arbitrator’s reasons for concluding that Oliver could relitigate the issue decided in the criminal proceedings. For convenience, I will repeat them: … I have jurisdiction under the Labour Relations Act, I am obliged to exercise that jurisdiction, and I will confine myself to the exercise of that jurisdiction. It is not an abuse of process under the collective agreement and the Labour Relations Act for the grievor to maintain, in these proceedings, that he did not do things that he was found to have done in other proceedings. My reading of the Del Core case is that the Court accepts that it is open to any person in a situation like that of the grievor to do that. Further, the criminal trial transcript can be used to challenge his credibility in that testimony. My conclusion is that the law recognizes that it is open to me in these proceedings to weigh all the evidence and come to a different conclusion on the facts than was arrived at in some other proceedings. [Emphasis added.] [51] The arbitrator’s conclusion was driven by his interpretation of Re Del Core, supra, and Demeter v. British Pacific Life Insurance Co. (1983), 43 O.R. (2d) 33 (H.C.J.) aff’d. (1984), 48 O.R. (2d) 266 (C.A.). The arbitrator read Re Del Core, supra, as holding that it was open to “any person in the situation of the grievor” to relitigate an issue decided against that person in a criminal proceeding. He interpreted Demeter, supra, as limiting the abuse of process doctrine, as it applied to attempts to relitigate previously decided issues, to cases in which a convicted person instituted a subsequent proceeding for the purpose of challenging a finding made in the prior criminal proceedings. I think the arbitrator misread both cases. His misreading led him to an erroneously narrow view of the circumstances in which public policy will preclude relitigation of previously decided issues. [52] The result in Re Del Core, supra, turned on whether the Discipline Committee of the College of Pharmacists had made a finding against the pharmacist that was not supported by the record. That finding was germane to the seriousness of the pharmacist’s misconduct. The majority in this court held that the Committee had not fallen into that error (Finlayson J.A., Blair J.A.). Houlden J.A., in dissent, and agreeing with the Divisional Court, found that the Committee had made a finding that was not supported by the evidence. [53] The issue on which Re Del Core, supra, turned had no jurisprudential significance. However, in the course of the discipline hearing, the Committee had admitted into evidence a certificate of conviction indicating that the pharmacist had been convicted on three counts of fraud arising out of his work as a pharmacist. These frauds were the basis for the discipline proceedings against the pharmacist. The pharmacist led no evidence to challenge the convictions and the Committee found that he had committed the frauds. Section 22.1 of the Evidence Act had not been enacted and the admissibility and effect of the evidence of the prior convictions became the focal point of the appeal. [54] In this court, each member of the panel wrote reasons. All three agreed that the certificate of conviction was admissible in the discipline proceedings as evidence that the pharmacist had committed the frauds alleged in the criminal charges. There was no clear consensus beyond that finding. [55] Finlayson J.A. held that the pharmacist could not challenge the finding of the criminal court in the discipline proceedings. He said, at p. 9: … [T]o insist that the civil court or other tribunal is required to have proven to its satisfaction by independent evidence that the offence did in fact take place, is to relitigate the issue in a collateral proceeding. This is so even where, as here, the pharmacist is not initiating the proceeding. By insisting, in his defence, on having the substance of his criminal misconduct retried, he is initiating a collateral attack on a final decision of a criminal court of competent jurisdiction. … [56] As I understand the reasons of Finlayson J.A., he would hold that where the act underlying the criminal conviction is identical to the act underlying the allegation of misconduct in the discipline proceedings, the convicted person cannot relitigate in the discipline proceedings the issues determined against him in the criminal proceeding. O’Driscoll J., at pp. 341-42, referred with approval to the words of Finlayson J.A. [57] Houlden J.A. took a very different view of the law in Re Del Core, supra. He held that evidence of the convictions was prima facie evidence that the pharmacist had committed the frauds, but that the pharmacist was entitled in the discipline proceeding to relitigate his culpability. Houlden J.A. rejected the argument that such relitigation constituted a collateral attack on the criminal convictions or an abuse of process. He read the latter doctrine as limited to cases in which a plaintiff institutes a subsequent proceeding for an oblique motive. He concluded his reasons with these words, at p. 19: … I cannot agree with Finlayson J.A. that by insisting in his defence on having the substance of his misconduct retried, the respondent is initiating a collateral attack on a final decision of a criminal court of competent jurisdiction. If that proposition is correct, the Demeter case has brought about an astounding amalgamation of criminal and civil law which has heretofore been unknown in Canadian law. … [58] The reasons of Houlden J.A. offer strong support for this arbitrator’s conclusion. Houlden J.A. would apparently impose no public policy limitation on a defendant’s right to relitigate the merits of a criminal conviction in a subsequent proceeding in which a plaintiff relied on the conviction to establish a claim against the defendant. [59] Blair J.A., the third member of the panel, agreed with his colleagues that evidence of the conviction was admissible as evidence that the pharmacist had committed fraud. After observing that since there was no evidence to the contrary, it was unnecessary to decide whether the conviction was conclusive proof, Blair J.A. went on to say, at p. 21: I agree with my brother Houlden that such evidence constitutes prima facie and not conclusive proof of the fact of guilt in civil proceedings. The prior conviction must of course be relevant to the subsequent proceedings. Its weight and significance will depend on the circumstances of each case. … [60] If one stopped reading the reasons of Blair J.A. at this juncture, they would appear to support those of Houlden J.A. and the arbitrator in this case. Blair J.A. went on, however, to acknowledge that the abuse of process doctrine would prohibit attempts by defendants to relitigate facts underlying previous convictions in some circumstances. He was not prepared to limit the doctrine as Houlden J.A. had in his reasons. Blair J.A. said, at p. 22: The right to challenge a conviction is subject to an important qualification. A convicted person cannot attempt to prove that the conviction was wrong in circumstances where it would constitute an abuse of process to do so [citations omitted]. Courts have rejected attempts to relitigate the very issues dealt with at a criminal trial where the civil proceedings were perceived to be a collateral attack on the criminal conviction. The ambit of this qualification remains to be determined in future cases. [Emphasis added.] [61] The reasons of Blair J.A. stand between the absolute positions staked out by Finlayson J.A. and Houlden J.A. Blair J.A. recognized that while evidence of a prior conviction was not conclusive proof that the convicted person committed the crime, there would be circumstances where policy considerations would prohibit the convicted person from leading evidence to challenge that conviction. Where those considerations operated, the evidence of the conviction would stand unchallenged and thereby prove that the person had committed the crime. [62] Blair J.A. was not prepared to limit the circumstances in which relitigation would offend public policy to those cases in which a plaintiff sought to relitigate the validity of a conviction by instituting subsequent proceedings. He was also not prepared to attempt to provide a definitive description of the circumstances in which relitigation would offend public policy. He left that for future cases. The future has arrived and with it, a plethora of such cases. They fully vindicate Blair J.A.’s refusal to limit the court’s power to prevent relitigation of decided issues to the specific fact situation presented in cases like Demeter, supra. [63] A close reading of Re Del Core, supra, reveals that only the reasons of Houlden J.A. support the position taken by the arbitrator. Like Finlayson J.A., Houlden J.A. did not speak for a majority of the court on this issue. I believe that the arbitrator erred in law in concluding that Re Del Core, supra, decided the issue before him and opened the door to CUPE’s relitigation of the issue decided against Oliver in the criminal proceeding. Re Del Core, supra, determined the admissibility of the convictions as evidence in a subsequent proceeding, but provided no majority opinion as to when public policy would preclude relitigation of the issues determined in the criminal proceeding. Subsequent cases have addressed that issue. [64] The arbitrator’s second reason for concluding that Oliver could relitigate the issue decided in his criminal proceedings rested on his view of the abuse of process doctrine in so far as that doctrine operates to prevent relitigation of decided issues. As indicated above, the arbitrator, like Houlden J.A. in Re Del Core, supra, saw the doctrine as limited to those situations in which a previously convicted plaintiff initiates a proceeding for the purpose of challenging that conviction. [65] It is true that the two authorities most often cited to support the court’s power to prevent relitigation of decided issues in circumstances where res judicata does not apply, are cases where a convicted person commenced a civil proceeding for the purpose of attacking a finding made in a criminal proceeding against that person: Demeter, supra; and Hunter v. Chief Constable of West Midlands Police, [1982] A.C. 529 (H.L.). Both cases recognize that the court’s power to prevent relitigation reaches beyond the limits of the res judicata doctrine. Both also recognize that the power is founded on public policy and is an expression of the court’s broad power to act to protect the integrity of its own process. [66] I do not read either Demeter, supra, or Hunter v. Chief Constable of West Midlands Police, supra, as limiting the court’s power to prevent abuses brought about by relitigation to the specific circumstances revealed by the facts of those two cases. In Hunter v. Chief Constable of West Midlands Police, supra, at p. 536, Lord Diplock said this in the first paragraph of his reasons: … The circumstances in which abuse of process can arise are very varied; those which give rise to the instant appeal must surely be unique. It would, in my view, be most unwise if this House were to use this occasion to say anything that might be taken as limiting to fixed categories the kinds of circumstances in which the court has the duty (I disavow the word discretion) to exercise this salutary power. [Emphasis added.] [67] In the course of his reasons, Lord Diplock cited with approval the speech of Lord Halsbury L.C. in Reichel v. Magrath (1889), 14 App. Cas. 665. In that case the court struck a statement of defence as an abuse of process holding that it was an attempt to relitigate an issue finally decided in a previous action in which the defendant in the second action had been the unsuccessful plaintiff. The plaintiff in the second action was not a party to the first action. [68] The contention that the abuse of process doctrine operates to prohibit relitigation only in the fact situation presented in Hunter v. Chief Constable of the West Midlands Police, supra, is contrary to the caution sounded by Lord Diplock at the outset of his reasons, and to at least one of the authorities relied on by him. [69] In Demeter, supra, the plaintiff (Demeter) commenced an action against his insurers claiming the proceeds of his wife’s life insurance policies. Demeter had been convicted of murdering his wife, and made it clear that he intended to use his action against the insurers to demonstrate that he had been wrongly convicted. His motive for instituting the action played a significant role in Osler J.’s finding that the action constituted an abuse of process. It is not clear to me, however, that Osler J. regarded that motive as essential to his conclusion that Demeter’s attempt to relitigate his murder conviction constituted an abuse of process: see Demeter, supra, per Osler J. at p. 50. In this court, McKinnon A.C.J.O. emphasized Demeter’s motive. Nothing in his brief reasons suggests that, apart from the doctrine of res judicata, the court could not prevent relitigation of already decided issues except in the specific fact situation presented in Demeter, supra. To dispose of the appeal before him, the Associate Chief Justice did not have to describe the full reach of the court’s power to prevent relitigation of decided issues and he did not purport to do so. [70] A review of the Canadian case law reveals that the courts have prohibited relitigation of decided issues in cases that are not captured by the res judicata doctrine, and do not involve plaintiffs who have commenced subsequent proceedings for some improper motive. In these cases defendants were precluded from relitigating issues decided against them in a prior proceeding. Some of the cases involve prior criminal convictions: Nigro v. Agnew-Surpass Shoe Stores Ltd. (1977), 18 O.R. (2d) 215 at 218 (H.C.J.) aff’d. without reference to this point, (1978), 18 O.R. (2d) 714 (C.A.); Bjarnarson v. Manitoba (1987), 38 D.L.R. (4th) 32 (Man. Q.B.) aff’d. (1987), 21 C.P.C. (2d) 302 (Man. C.A.); Simpson v. Geswein (1995), 25 C.C.L.T. (2d) 49 at 61 (Man. Q.B.); Bomac Construction Ltd. v. Stevenson, [1986] 5 W.W.R. 21 at 26-27 (Sask. C.A.); Re Roenisch and Roenisch (1991), 85 D.L.R. (3d) 540 at 546 (Alta. Q.B.); Saskatoon Credit Union v. Central Park Enterprises Ltd. (1988), 47 D.L.R. (4th) 431 at 438 (B.C.S.C.); Canadian Tire Corporation Limited v. Summers (1995), 23 O.R. (3d) 106 at 115 (Gen. Div.); Franco v. White, supra; Germscheid v. Valois (1989), 34 C.P.C. (2d) 267 (Ont. H.C.J.); see also, P. Perell “Res Judicata and Abuse of Process” (2001), 24 Advocates Quarterly 189 at 196-97; and G. Watson “Duplicative Litigation: Issue Estoppel, Abuse of Process and the Death of Mutuality” (1990) 69 Can. B.R. 623 at 648-651. [71] The terminology used in these cases to describe the power to prevent relitigation has varied. Many cases have described the purported relitigation as an abuse of process. Others have expanded the conventional issue estoppel component of the res judicata doctrine to encompass situations which deviate to some extent from the accepted criteria for the application of issue estoppel. It may be that the phrase “abuse of process” is best limited to describe those cases where the plaintiff has instituted litigation for some improper purpose: Demeter, supra; Foy v. Foy (1978), 20 O.R. (2d) 747 (C.A.). It is also established in this jurisdiction that res judicata is limited to those cases which fall squarely within the established criteria: Minott, supra. Concerns with terminology cannot, however, obscure the recognition by the courts of a power to prevent the relitigation of decided issues in circumstances which extend beyond the res judicata doctrine and beyond cases where a plaintiff has commenced a proceeding for an improper purpose. These cases hold that where relitigation would undermine the integrity of the adjudicative process, it should not be permitted. Plaintiffs who commence a proceeding for the purpose of attacking the validity of an issue finally determined in a prior criminal proceeding, clearly undermine the integrity of the criminal justice process. It ignores reality, however, to pretend that only those plaintiffs commit that sin. [72] The arbitrator erred in law by limiting the scope of the power to prohibit relitigation of issues previously decided in criminal proceedings to circumstances in which the convicted person initiates the subsequent proceeding for the purpose of challenging a finding made in the criminal proceeding. The arbitrator took what is perhaps the clearest example of abusive relitigation and declared it to be the only situation in which relitigation would offend public policy. His erroneous and unduly restrictive reading of the authorities from this court led the arbitrator to fail to address the crucial policy considerations which had to be assessed and balanced in order to determine whether relitigation of Oliver’s conviction should be allowed in the grievance proceeding. (b) My Answer [73] When a controversy arises over a party’s right to relitigate a decided issue, important principles that are at the core of the due administration of justice clash. Parties who seek to relitigate stress their “right” to a decision on the merits made after they have had a full opportunity to present their case before the tribunal charged with the responsibility of rendering a decision. Those who resist relitigation assert their “right” to rely on the previous determination. All parties insist that “justice” is achieved by the position they assert. [74] The adjudicative process in its various manifestations strives to do justice. By the adjudicative process, I mean the various courts and tribunals to which individuals must resort to settle legal disputes. Where the same issues arise in various forums, the quality of justice delivered by the adjudicative process is measured not by reference to the isolated result in each forum, but by the end result produced by the various processes that address the issue. By justice, I refer to procedural fairness, the achieving of the correct result in individual cases and the broader perception that the process as a whole achieves results which are consistent, fair and accurate. [76] Justice, must of course, look through a much broader lens than that worn by the party who seeks to relitigate. Other interests, including those of the previously successful litigant, those collaterally affected by the proposed relitigation (eg. witnesses), the institutions responsible for controlling and administering the adjudicative process, and the community at large, are all affected when an issue that has been finally decided is relitigated. To those interests, finality is a central feature of justice and to the extent that relitigation uproots finality, it invites injustice. This is so whether the relitigation occurs in the same forum as the first adjudication, or in a different forum. [77] In deciding whether to permit relitigation, a court or tribunal must decide whether finality concerns should outweigh an individual litigant’s claim that the justice of the specific case warrants relitigation. That determination is fact-specific and requires that the court or tribunal weigh these competing considerations in the context of the facts of the particular case. Not all relitigation compromises finality concerns to the same extent and finality is not as important in some facets of the law as in others. Similarly, a claim that the justice of the individual case requires relitigation is much stronger in some situations than in others. [78] The finality principle and the values underlying that principle are well understood and have been stressed in various contexts by this court: Tsaoussis (Litigation Guardian of) v. Baetz (1998), 41 O.R. 257 at 264 (C.A.), leave to appeal to S.C.C. refused, [1998] S.C.C.A. No. 518; R. v. H.(E.F.) (1997), 115 C.C.C. (3d) 89 at 101 (Ont. C.A.), leave to appeal to S.C.C. refused, [1997] S.C.C.A. No. 256. Those values warrant reiteration. [79] Finality is essential to the maintenance of a fair and effective adjudicative process. The values it serves are fundamental to that process and to the community. Firstly, relitigation raises the spectre of inconsistent results. Such results create a myriad of problems for those involved in the proceedings, including, but not limited to those arising from attempts to enforce conflicting orders. Inconsistent results are also capable of bringing the administration of justice into disrepute in the eyes of reasonable and well informed members of the community. In this case, the conflicting results arrived at in the criminal proceedings and in the arbitration would, in my view, cause reasonable persons to wonder about the rationality of the adjudicative process. [80] Secondly, relitigation diminishes the overall authority of the adjudicative process. What value is a result if that result can be challenged and relitigated in another forum at any time? Thirdly, relitigation breeds uncertainty. How can those drawn into the adjudicative process determine when they have reached the end of that process, get on with the rest of their lives and arrange their affairs in reliance on the decisions reached in that process? Finally, relitigation drains individual and institutional resources. Neither individuals, nor the community as a whole, have the resources or the lifespan required to permit the continual relitigation of decided issues. [81] In emphasizing the importance of finality, I do not pretend that the decision made in the first proceeding is always correct. It must, of course, be observed that relitigation is not a guarantee of a more accurate result. The simple truth is that finality is so essential that it is routinely given priority over the possibility that relitigation would achieve a more accurate result. The importance of finality is best seen in those cases where finality has trumped individual constitutional rights: Reference re Manitoba Language Rights, [1985] 1 S.C.R. 721; R. v. Thomas, [1990] 1 S.C.R. 713; and R. v. Sarson, [1996] 2 S.C.R. 223. In those cases, even though the decision is wrong in the most important sense in that it denies someone a fundamental right, the courts have held that finality precludes relitigation. A desire to avoid the harm caused by permitting relitigation is given paramountcy over the rights of the individual and the accuracy of the result in the particular case. [82] As indicated above, finality concerns are given more prominence in some areas of the law than others. For example, finality carries less weight in cases involving child support and custody orders than in cases involving the ownership of real property. Finality is very important in criminal cases. A criminal verdict is not simply a determination of an issue as between two private litigants. The prosecution is taken on behalf of the community and the verdict is the verdict of the community. The community has a real stake in that verdict. To permit relitigation of issues decided in criminal cases in subsequent private litigation is to seriously diminish the force of the criminal law in the community. [83] The arbitrator made no reference to the finality principle or the values it serves. He dismissed any concerns about the effect of relitigation on the integrity of the adjudicative process by observing: … I cannot see that the grievor’s protestations of innocence before this arbitration tribunal is an affront to, or serves to under the criminal justice system. [84] This observation misses the point. The problem was not Oliver’s protestations, but the arbitrator’s willingness to relitigate a decided issue and reach an inconsistent result. Despite the arbitrator’s insistence that he was not passing on the correctness of the decision made by Ferguson J., that is exactly what he did. One cannot read the arbitrator’s reasons without coming to the conclusion that he was convinced that the criminal proceedings were badly flawed and that Oliver was wrongly convicted. This conclusion, reached in proceedings to which the prosecution was not even a party, could only undermine the integrity of the criminal justice system. The reasonable observer would wonder how Oliver could be found guilty beyond a reasonable doubt in one proceeding and after the Court of Appeal had affirmed that finding, be found in a separate proceeding not to have committed the very same assault. That reasonable observer would also not understand how Oliver could be found to be properly convicted of sexually assaulting the complainant and deserving of 15 months in jail and yet also be found in a separate proceeding not to have committed that sexual assault and to be deserving of reinstatement in a job which would place young persons like the complainant under his charge. [10] [85] The arbitrator took no account of other adverse effects occasioned by his decision to allow CUPE to relitigate Oliver’s conviction. He did not consider the drain on institutional resources occasioned by the relitigation. [11] He did not address the uncertainty created by the relitigation. If the arbitrator’s view that CUPE was entitled to relitigate Oliver's guilt was to prevail, employers like Toronto would be in a difficult position when an employee is convicted of a serious crime in the course of his employment. If Toronto had not discharged Oliver, it would have exposed itself to potential liability should Oliver re-offend, and most certainly would have exposed itself to legitimate criticism from the community. If, however, Toronto discharged Oliver, it would have exposed itself to a grievance arbitration at which it may or may not have been able to prove that Oliver committed the sexual assault for which he was convicted. It cannot be assumed that an employer is positioned to prove that an employee committed a crime in the same way as the prosecutorial authorities are able to prove that crime. [86] The arbitrator did not consider the adverse effects of relitigation on complainants. All witnesses are potentially inconvenienced by relitigation. For victims of sexual assaults, the harm can go well beyond mere inconvenience. For these victims, testifying about the abuse is a difficult and traumatic experience. For some, it is physically and emotionally harmful. No one is more sympathetic than I to arguments that concerns for the well-being of complainants cannot diminish the rights of persons against whom such serious allegations are made to fully defend themselves. Concern for the harm done to the complainant by exposure to the adjudicative process must, however, take on added weight when the issue is not whether the accused person should have a full opportunity to challenge that allegation, but whether that person should have more than one full opportunity to challenge that allegation. I think the potential harm done to complainants by the relitigation of allegations of sexual abuse where those allegations have been finally determined in the criminal courts is a significant consideration in determining whether relitigation of those allegations should be permitted. [87] Nor is it any answer to suggest that the complainant need not testify in the second proceeding. It seems to me that unless the complainant does testify, the contention that the second proceeding is more likely than the first to yield an accurate result loses much of its force. Absent the testimony of the complainant, the risk of an unjust and inaccurate result is very real. In my view, this case demonstrates that danger. The arbitrator found the complainant to be incredible without the benefit of seeing or hearing the complainant. The risk inherent in credibility assessments based only on a written record are well recognized. [88] The arbitrator also failed to recognize the dangers inherent in relitigation where that relitigation includes an assessment of prior proceedings which occur in a forum with which the second adjudicator is not familiar. The arbitrator’s ultimate conclusion that Oliver had not sexually assaulted the complainant was based in part on two misconceptions of the rules governing admissibility of evidence in criminal proceedings. [89] The arbitrator observed that there was no evidence that the police investigation uncovered any other complainant. Referring, with apparent approval to counsel for CUPE’s submission, the arbitrator observed that “multiple victims is a common feature of sex assault cases involving men and young boys”. He accepted that if there were other boys who had made allegations against Oliver, they would have testified at the criminal trial. He suggested that the absence of any such evidence compelled the conclusion that no one else had complained about Oliver’s conduct. This line of reasoning ignores the strict rules of evidence governing the admissibility of “similar fact” evidence in criminal proceedings. The arbitrator’s inference that because no other complainants testified at the criminal trial, there were no other complainants, is not one that anyone familiar with the rules of criminal evidence would readily draw. [90] The arbitrator’s second misapprehension of the criminal rules of evidence is revealed by this observation: …[T]here is no evidence that the grievor had any sexual interest in young boys. Indeed the evidence we have before us indicates that he does not. [91] Evidence that an accused person had “a sexual interest in young boys” is propensity evidence of the rankest sort and would be inadmissible as part of the Crown’s case in a criminal proceeding. The arbitrator’s inference that the absence of such evidence supported the conclusion that Oliver had no interest in young boys and hence did not commit the acts alleged by the complainant, reveals a misapprehension of the evidentiary rules governing criminal proceedings. [92] In referring to these two aspects of the arbitrator’s ruling, I do not suggest that he should have a perfect understanding of the rules of criminal evidence. I refer to them only to demonstrate that relitigation has its own pitfalls when it comes to getting at the truth in an individual case. Those pitfalls re-enforce the value of finality. The arbitrator did not appreciate these risks when he undertook a critical analysis of the criminal proceedings in the course of making his own determination of Oliver’s culpability. [93] The dangers associated with the relitigation of decided issues were fully engaged in the circumstances of this case. Finality concerns should have been in the forefront of the arbitrator’s analysis. Instead, he did not factor those concerns into his decision at all. [94] As important as finality is in the adjudicative process, it does not always block relitigation. Finality concerns must be tempered by a search for justice in each individual case. Sometimes, finality will yield completely to the interests of an individual litigant. For example, if subsequent to a final decision, new evidence is discovered which conclusively demonstrates that a finding of fact made in the first proceeding was incorrect, the justice of that case could demand an opportunity to relitigate. The integrity of the adjudicative process, the rationale underlying finality, may not be served by adhering to an undeniably erroneous decision. Similarly, if an initial decision is obtained by fraud or other dishonest practice, relitigation must be allowed. In those cases, relitigation restores the integrity of the adjudicative process. [95] The balancing of finality concerns with the individual litigant’s claim to access to justice is best seen in recent res judicata jurisprudence. This court has held that the traditional criteria set out in the res judicata doctrine provide the starting point for an analysis of any claim that a prior determination precludes relitigation. If those criteria are met, relitigation will be foreclosed in the vast majority of cases. However, even where the criteria are met, a court or tribunal may refuse in exceptional circumstances to apply the doctrine and may permit relitigation where the circumstances dictate that finality interests should yield to the justice of the individual case: Danyluk v. Ainsworth Technologies Inc., supra, at paras. 62-67 (S.C.C.); Minott, supra; Schweneke v. Ontario, supra; Ontario (Attorney General) v. Bear Island Foundation et al. (1999), 126 O.A.C. 385 at 392-93 (C.A.); D.J. Lange, The Doctrine of Res Judicata in Canada, (Butterworths, 2000) at pp. 31-34. [96] In Minott, supra, Laskin J.A., at p. 340, put it this way: Issue estoppel is a rule of public policy and, as a rule of public policy, it seeks to balance the public interest in the finality of litigation with the private interest in achieving justice between litigants. Sometimes these two interests will be in conflict, or at least there will be tension between them. Judicial discretion is required to achieve practical justice without undermining the principles on which issue estoppel is founded. Issue estoppel should be applied flexibly where an unyielding application of it would be unfair to a party who is precluded from relitigating an issue. [12] [97] A second feature of the res judicata jurisprudence is the court’s willingness to preclude relitigation in circumstances which do not quite come within the res judicata criteria, but in which the policy considerations barring relitigation operate as strongly as they do in cases captured by the res judicata doctrine: Wernikowski, supra; Franco v. White, supra. [98] After thorough and careful analysis of the case law, Sharpe J.A. said in Franco v. White, supra, at p. 413: Current Canadian law, through the doctrines of abuse of process and the application of the prima facie evidence standard, results in giving preclusive effect to prior determinations in many situations falling outside the rules of traditional issue estoppel. In this regard, Canadian law may be seen as paralleling the discretionary approach to offensive non-mutual issue estoppel in American law. The effect of a prior determination is not determined through an inflexible application of mechanical rules, but rather on the basis of the important underlying procedural values and an element of judicial discretion is applied to ensure that justice is done in each case. [Emphasis added.] [99] Those cases which have refused to apply res judicata to preclude relitigation even though the criteria for the application of the doctrine were met, and those cases which have refused to permit relitigation even though all of the res judicata doctrine were not met, reflect the same principled approach to the problem of relitigating decided issues. The ultimate question must be whether finality concerns preclude relitigation or whether the justice of the individual case demands relitigation. [100] The following approach should be taken when weighing finality claims against an individual litigant’s claim to access to justice: q Does the res judicata doctrine apply? q If the doctrine applies, can the party against whom it applies demonstrate that the justice of the individual case should trump finality concerns? q If the doctrine does not apply, can the party seeking to preclude relitigation demonstrate that finality concerns should be given paramountcy over the claim that justice requires relitigation? [101] For reasons set out above (supra para. 47), issue estoppel, the applicable component of the res judicata doctrine, does not foreclose relitigation by CUPE of Oliver’s culpability. While acknowledging that issue estoppel does not apply, it is important to recognize that save for the mutuality requirement of that doctrine, all of the other criteria for its application are met in this case. The decision in the criminal proceeding is a judicial decision and a final decision. Perhaps most importantly, the issue decided in the criminal case is exactly the issue that had to be decided in the arbitration. Thus, while the res judicata doctrine does not apply, the facts of this case put it at the very edge of that doctrine. [13] [102] Even though issue estoppel does not apply, I am convinced that in these circumstances, finality concerns must be given paramountcy over CUPE’s claim to an entitlement to relitigate Oliver’s culpability. Several factors lead me to that conclusion. [103] There is no suggestion that the criminal proceedings were tainted by fraud or any procedural failing that could cast any doubt on the fairness of the criminal trial or the accuracy of the result. Oliver had all of the procedural and constitutional rights which quite properly protect those charged with a crime. These included the right to full disclosure, the right to counsel, the right to silence, the presumption of innocence, and the right to make full answer and defence. Oliver also had the benefit of the heavy burden imposed on the prosecution in a criminal proceeding. Despite these protections and the onus imposed on the Crown, Oliver was found beyond a reasonable doubt to have sexually assaulted the complainant. That finding was upheld on appeal. In my view, the protections afforded an accused and the heavy burden of proof imposed on the Crown are relevant to both the fairness of the process and the accuracy of the result. They fully justify investing a criminal conviction with a very high degree of finality when the same issue arises in subsequent litigation. [104] The fact that CUPE is not a party to the criminal proceedings is of no consequence in assessing the competing values outlined above. CUPE presented the grievance on behalf of Oliver. Its interest was the same as Oliver’s. CUPE cannot argue that the outcome of the grievance arbitration had some significance to a broader constituency beyond Oliver’s personal interest. Nor can CUPE claim that its participation in the arbitration altered the essential nature of the inquiry. In both the criminal proceeding and the arbitration, the question was whether Oliver sexually assaulted the complainant. [105] This is also not a case in which the prior determination of guilt arose out of some relatively minor prosecution in which Oliver could argue that the stakes of that proceeding did not warrant his full attention. The charge against Oliver was serious and he obviously knew it was serious. The sentence imposed at trial and affirmed in this Court demonstrates that the consequences of Oliver’s conviction were very serious indeed. [106] Counsel for CUPE also argued that the arbitrator had evidence before him that was not adduced in the criminal proceeding. Counsel submitted that for this reason alone, the arbitrator was correct in allowing CUPE to relitigate Oliver’s culpability. I would reject this argument. To justify discounting finality, the evidence heard by the arbitrator that was not produced in the criminal court must be evidence that was not available at the criminal trial through the exercise of due diligence. It must also be evidence that is sufficiently cogent to significantly, if not conclusively, alter the evidentiary landscape from that which existed in the first proceeding. There is nothing in this record to suggest that anything that was put before the arbitrator (eg. the complainant’s letters to Mr. Zanetti) was not available to counsel conducting Oliver’s criminal defence. [14] [107] In any event, the arbitrator did not purport to justify his finding that Oliver had not sexually assaulted the complainant by reference to evidence that was available to him and not to the criminal court. To the contrary, his reasons consist of a detailed and microscopic critique of the evidence offered by the Crown in the criminal proceedings, the arguments advanced by the Crown and some of the conclusions drawn by Ferguson J. He said: When I balance the evidence in this case and subject it to the tests of credibility outlined in the case law above, I simply cannot believe [the complainant’s] story. I accept that this is a different finding on credibility than that of the Trial Judge and because of that I have taken pains to outline the evidence which I rely on. Counsel for the Employer submitted that Oliver appealed the conviction and that the Court of Appeal did not think that the grievor was wrongly convicted. Where a Trial Judge makes a finding of credibility, Appeal Courts are loathe to set aside such a finding. I am not in a position where I must pay the same deference to the Trial Judge’s finding of credibility. … [108] As I am satisfied that CUPE has not shown that any evidence placed before the arbitrator could qualify as “new” evidence, I need not come to any firm conclusion as to the degree of cogency required of any such “new” evidence before that evidence could justify relitigation of an otherwise final criminal conviction: Wernikowski, supra, at p. 636. [109] This is also not a case where even if Oliver’s conviction is given pre-emptive force, there would still be a need to hear much of the same evidence in the arbitration proceeding. In cases like that, finality arguments lose some of their force as preventing relitigation will do little to preserve personal and institutional resources. Here, the savings would be very real. Had Oliver not been allowed to relitigate his criminal conviction, that would have effectively ended the arbitration proceedings. [15] [110] Lastly, I can see nothing in the arbitration process that suggests that the result arrived at through that process is more likely to be accurate than the result arrived at in the criminal process. In this particular case, given that the complainant did not testify in the arbitration, there is good reason to doubt the accuracy of the results arrived at in the arbitration proceeding. [111] The fact-specific approach to determinations of whether previously decided issues may be relitigated creates some uncertainty. On that approach it cannot be said definitely that a prior conviction will always foreclose relitigation of culpability in a subsequent proceeding. It seems to me, however, that some uncertainty is the inevitable cost where flexibility is favoured over bright line rules. [112] Flexibility does not mean, however, that each case stands alone and provides no guidance for subsequent cases. In the context of a grievance arbitration, the principles establish that where a prior conviction is proved and the facts underlying the conviction are the same as the facts in issue on the arbitration, the party seeking to relitigate those facts must demonstrate compelling circumstances which warrant overriding the strong public interests served by treating a criminal conviction as a final determination of those facts. IV [113] For the reasons set out above, the arbitrator erred in law in permitting CUPE to relitigate Oliver’s culpability. The arbitrator should have held that Oliver’s conviction established that he had sexually assaulted the complainant for the purposes of the arbitration. Based on that finding, any conclusion other than that Toronto had established just cause for Oliver’s dismissal would be patently unreasonable. [114] I would dismiss the appeal with costs. “D. H. Doherty J.A.” “I agree J.M. Labrosse J.A.” “I agree K. Feldman J.A.” RELEASED: August 10, 2001 [1] The recent decisions of this Court include Danyluk v. Ainsworth Technologies Inc. (1998), 42 O.R. (3d) 235 (C.A.), rev’d. 2001 SCC 44; Wernikowski v. Kirkland, Murphy & Ain (2000), 181 D.L.R. (4th) 625 (Ont. C.A.), leave to appeal to S.C.C. refused, [2000] S.C.C.A. No. 98; Rasanen v. Rosemount Instruments Limited (1994), 17 O.R. (3d) 267 (C.A.), leave to appeal to S.C.C. refused, [1994] S.C.C.A. No. 152; Minott v. O’Shanter Development Co. (1999), 42 O.R. (3d) 321 (C.A.), leave to appeal to S.C.C. filed, discontinued, May 14, 1999, [1999] S.C.C.A. No. 128; Schweneke v. Ontario (2000), 47 O.R. (3d) 97 (C.A.), application for leave to appeal to S.C.C. filed October 2, 2000; Heynen v. Frito-Lay Canada Ltd. (1999), 179 D.L.R. (4th) 317 (Ont. C.A.), leave to appeal to S.C.C. refused, [1999] S.C.C.A. No. 569; Franco v. White (2001), 53 O.R. (3d) 391 (C.A.). [2] Toronto also argued that Oliver’s status as a person convicted of sexually assaulting a young child in the course of his employment, warranted his dismissal without any inquiry beyond the fact of the conviction. The arbitrator rejected this argument in part because the collective bargaining agreement did not specifically provide for the discharge of employees convicted of criminal offences. I need not address this argument. [3] A transcript of the entire criminal proceeding, including sentencing was filed in the arbitration. [4] One of the letters sent to Mr. Zanetti was a copy of a letter sent to Mr. Zanetti’s superior. [5] The following authorities support the arbitrator’s position: Taylor V. Baribeau (1985), 51 O.R. (2d) 541 (Div. Ct.); Roth v. Roth (1991), 4 O.R. (3d) 740 (Gen. Div.); Q. & Q. v. Minto Management Ltd. et al. (1984), 46 O.R. (2d) 756 (H.J.C.); Re Canada Post Corporation and C.U.P.W. (Cashen) (1993), 29 C.L.A.S. 516; Re Canada Post Corporation v. C.U.P.W. (Leavere) (1990), 73 L.A.C. (4th) 129. Counsel for the respondent also points out that the arbitrator’s decision is consistent with that reached by Stevenson J. sitting as an umpire under the Unemployment Insurance Act on an appeal brought by Mr. Oliver after he was denied unemployment insurance benefits (April 22, 1999). [6] See supra, n. 1. [7] The second part of the arbitrator’s task, the application of the correct principles to a given set of facts, may be subject to a more deferential standard of review. I need not address that issue, given my conclusion that the arbitrator did not correctly identify the relevant principles: see Canada (Director of Investigation and Research) v. Southam Inc., supra, at 767, 770-71. [8] Section 22.1(3) is not reproduced. It provides for proof of the prior conviction by way of a certificate and is not germane to this appeal. Section 2 of the Evidence Act makes the Act applicable to the arbitration proceedings. [9] The section speaks of proof of the conviction as being proof “that the crime was committed by the person”. Sometimes there may be a dispute as to exactly what findings of fact made in the criminal proceedings are included in the phrase “proof that the crime was committed”. That difficulty does not arise here. If the conviction was admissible, it was evidence that Oliver had, over a period of time, sexually assaulted a young boy. There was no dispute that if the assaults occurred, they were closely related to Oliver’s employment. Those facts provided ample cause for the discharge of Oliver. [10] In his reasons, the arbitrator suggested that it would be Kafkaesque if Oliver could not maintain his innocence in the arbitration proceedings. In my view, the more appropriate literary allusion is to Dickens’ Mr. Bumble. The inconsistent results achieved in the two proceedings could lead the reasonable person to conclude that Mr. Bumble was correct when he described the law as “an ass” and “an idiot”. [11] In this case, the allegation of sexual assault has spawned three proceedings: the criminal proceeding, the grievance arbitration, and the proceeding under the Unemployment Insurance Act. It takes little imagination to envision further proceedings, for example, if Oliver happened to be a member of a professional body, a discipline proceeding. [12] The same flexibility exists where cause of action estoppel would appear to preclude relitigation: Ontario (Attorney General) v. Bear Island Foundation et al., supra. [13] Professor Watson, among others, favours abandonment of the mutuality requirement for the application of issue estoppel. See G. Watson “Duplicative Litigation: Issue Estoppel, Abuse of Process and the Death of Mutuality”, supra. As Sharpe J.A. observed in Franco v. White, supra, at p. 407, this court has declined to follow that path. [14] Zanetti testified that the City provided everything to the police. There was no reason to assume that the prosecution did not make disclosure of this information to the defence. [15] The arbitrator appeared to accept CUPE’s argument that if the conviction could not be relitigated, there would be no point to the arbitration. He also seemed to accept that this was a strong argument in favour of permitting CUPE to relitigate. Avoiding subsequent proceedings where issues that have already been decided would be relitigated is of course one of the primary values underlying the finality principle.