Malicious Prosecution Report 2007


by Judy Mungovan

Readers are cautioned that the ideas or conclusions set forth in this paper, including any proposed statutory language and any comments or recommendations, may not have not been adopted by the Uniform Law Conference of Canada. They may not necessarily reflect the views of the Conference and its Delegates. Please consult the Resolutions on this topic as adopted by the Conference at the Annual meeting.

Charlottetown, Prince Edward Island, September 2007


[1] Following the Annual Meeting of the Uniform Law Conference of Canada (ULCC) in Edmonton, in August 2007, the Executive Committee, on behalf of both the Civil Section and the Criminal Section, constituted a joint working group to give further consideration to the need for uniform legislation to respond to concerns being reported across Canada regarding common law developments in the intentional tort of Malicious Prosecution. The working group began its work in the winter of 2007 through telephone conferences culminating in one in-person meeting in Toronto on March 29, 2007. The membership of the working group included:

- Alberta: Kate Bridgett, Sarah Dafoe, Tim Hurlburt

- Canada: Kathryn Sabo, Nancy Irving, Robert Frater

- Manitoba: Lynn Romeo, Robin Finlayson

- Ontario: Bill Manuel, Judy Mungovan

- Quebec: Michel Breton

- Saskatchewan: Darcy McGovern, Dean Sinclair

- ULCC/CHLC: Clark Dalton

[2] Crowns across the country have been following the case law in the area of malicious prosecution actions very closely, as have their respective provincial associations. Mindful of Crowns’ concerns, the working group identified early on three dangers that stem from an apparent loosening of the criteria that must be met before a claim of malicious prosecution against a Crown attorney is allowed to proceed. First, there is an increased risk of frivolous prosecution claims that demoralize both the Crown attorney named and Crown attorneys in general. Second, there is an increased risk that the loosening of the criteria will lead not only to more malicious prosecution claims, but also to other actions in tort to which Crown attorneys have traditionally been immune. Third, the lack of clarity in recent jurisprudence has left Crowns unsure how to best fulfill their quasi-judicial role as “ministers of justice”, due to an apparent gap between the standard that compels a Crown to proceed with a prosecution that is in the public interest and the standard a court uses when subsequently reviewing that same decision to proceed.

[3] This paper reviews in detail the developments in the common law tort of malicious prosecution through the Nelles v. Ontario[1] decision and its subsequent interpretations by the judiciary up to the recent Saskatchewan Court of Appeal decision in Miazga v. Kvello Estate.[2] In response to these developments and to the concerns outlined above, the working group has considered in this paper the merits of making Attorneys General solely liable for malicious prosecution torts, the development of methods to more effectively weed out frivolous lawsuits, and the prospect of tightening (via uniform law) the criteria that are used to determine whether a malicious prosecution action can be commenced against a Crown attorney.

Malicious Prosecution: The Context

[4] If we look to Ontario as our working example, prior to 1989 by virtue of Section 5(6) of the Proceedings Against the Crown Act,[3] Crown attorneys in Ontario were believed to be immune from civil liability flowing from the discharge of their prosecutorial responsibilities and decisions. However, in Nelles the Supreme Court (largely on policy grounds) brought an end to the notion of complete immunity when it held that the tort of malicious prosecution could lie against a Crown Attorney upon proof of the following four necessary elements:

1. That the proceedings were initiated by the defendant;

2. That the prosecution was terminated in favour of the plaintiff;

3. That there was an absence of reasonable and probable cause; and

4. That there was malice or that the primary purpose of the prosecution was other than that of carrying the law into effect.

[5] The Court reasoned that these elements were sufficient to ensure accountability on the part of Crowns for bad faith or malicious acts while at the same time being sufficiently onerous to prevent frivolous claims by disgruntled plaintiffs. The Court predicted that Crowns could continue to exercise their discretion in good faith, unfettered by any chilling effect that might flow from a less balanced approach. Recent jurisprudence indicates that there is cause for concern for the continued proper administration of justice – Crowns may no longer be able to operate both as zealous prosecutors in the public interest and as quasi-judicial ministers of justice[4] without fear about how their decisions in the course of their work could impact upon their own and their family’s personal lives.

[6] In Ontario, one in eight Crowns are named during their careers in a suit of malicious prosecution; virtually every Crown knows a colleague who has been named in such a suit. This is despite ongoing training for Crowns in all jurisdictions across Canada, which features professional programs with core curricula, and despite the fact that Crowns are guided by standardized and lengthy policy manuals that guide the exercise and limits of Crown discretion. Furthermore, the courts are reviewing the general exercise of Crown discretion in new ways – in addition to allowing actions alleging malicious prosecution, courts have also reviewed the decisions of Crown attorneys to not prosecute, in one case allowing a third party to allege (though unsuccessfully) that the Crown was liable for the criminal activities of an individual who was not prosecuted.[5]

[7] This paper reviews the Nelles decision and its subsequent interpretations in detail. It concludes that the third and fourth elements have been interpreted in a manner that is contrary to the policy rationale stated by the Supreme Court in Nelles, and absent judicial reinterpretation or statutory reform, this is likely to pose an increasing risk of frivolous malicious prosecution actions against Crown attorneys, and lead to a corresponding drop in the effectiveness and morale of Crown attorneys as a group. Any dip in prosecutorial resolve should be a cause for concern to the public and would impact the administration of justice.

[8] Courts use the third element (absence of reasonable probable cause) to review whether or not a prosecution should have been initiated. This has created a gap through which malicious prosecution actions can slip due to its difference with the standard that Crown attorneys are required to follow when initiating or continuing a prosecution (whether there is a reasonable prospect of conviction). Of even more concern, where courts have determined that no reasonable and probable cause exists, some have used this to infer malice on the part of the Crown, conflating the fourth element (requiring malice or some improper purpose) with the third. The fourth element was intended by the Supreme Court to be a bulwark against frivolous actions and actions based solely on negligence, but the jurisprudence has not evolved in this manner.

Statistics – The National Picture

[9] Tracking rates of malicious prosecution claims across Canada is challenging: not all jurisdictions keep detailed statistics and those that do use a variety of metrics (for reference, see Appendix I). However, all three provinces that do keep annual records show an increasing rate of malicious prosecution civil suits. Between 1992 and 2002, Alberta Crown attorneys were the subjects of 16 new malicious prosecution actions, an average of 1.6 per year. Between 2002 and 2006, there were also 16 new actions, a doubling to 3.2 per year. In Quebec, between 1995 and 2002 there were approximately 40 new malicious prosecution actions commenced, a rate of 5.7 per year. From 2002 to 2006 (when formal statistics were tracked), the rate jumped to 18.2 per year, for a total of 91 new actions. In Ontario, 167 malicious prosecution actions were commenced between 1992 and 2002, equalling 16.7 per year. Between 2002 and 2006, the rate increased to 23.2 per year, for a total of 119 new actions. Of the other provinces, only Saskatchewan and Prince Edward Island could provide rough estimates of the rate of malicious prosecution actions; Saskatchewan estimates approximately 15 to 20 actions over the last 20 years, while Prince Edward Island estimates only 3 over the last 10 years. Of the territories, only the Northwest Territories has any record of a malicious prosecution action, and it is ongoing.

[10] In Alberta, there remain 15 open files, 7 of which are active. In Ontario, there are 83 open files, 72 of which are active, naming a total of 98 separate Crown attorneys. In Prince Edward Island, there are 3 open files, all of which are active. In the Northwest Territories, there is one open file, which is active.

[11] Regardless of the jurisdiction, malicious prosecution actions are rarely successful. Alberta and Saskatchewan each reported only a single successful action, while Ontario and Prince Edward Island reported none. One successful action was reported by Quebec, and it was acknowledged that five were terminated through settlement.

[12] Faced with a burgeoning number of malicious prosecution actions, Ontario and Alberta have begun to track them with greater detail. On November 1st, 2005, there were 107 identified open files in Ontario alone. Over the following year, 36 of those actions were dismissed against 46 separate Crown attorneys on consent, by motion, or at a status hearing. Eighteen files were closed against 14 Crown attorneys due to a lack of a Statement of Claim being issued within one year. Eleven Rule 21[6] motions were brought, of which one was adjourned sine die, two were successful but subsequently appealed, and two were unsuccessful. Six Rule 20[7] motions were brought (one was successful but subsequently appealed, one was unsuccessful). Four actions went to trial involving four Crown attorneys; all four were dismissed.

[13] In Alberta, four claims between 1992 and 2005 were closed as a result of a successful summary dismissal (Rule 159) application brought by the Crown, five were closed after applications for dismissal for want of prosecution (Rule 244.1(1)), five were closed when the plaintiffs filed discontinuances of the action (usually pending an application by the Crown), and two were closed because the statements of claim, while filed, were never served.

The Tort of Malicious Prosecution as it Relates to Crown Attorneys Policy Reasons Against Absolute Immunity in Favour of Limited Exposure

[14] It is clear from the Supreme Court’s reasons in Nelles that the exception it intended to carve out from the existing doctrine of absolute immunity for Crowns was to be sufficiently narrow and onerous so as to catch only Crown conduct that was truly maliciously motivated. Lamer J. (as he then was) writing for the majority, said:

“It is also said in favour of absolute immunity that anything less would act as a “chilling effect” on the Crown attorney's exercise of discretion. It should be noted that what is at issue here is not the exercise of a prosecutor's discretion within the proper sphere of prosecutorial activity as defined by his role as a “minister of justice”. Rather, in cases of malicious prosecution we are dealing with allegations of misuse and abuse of the criminal process and of the office of the Crown attorney. We are not dealing with merely second-guessing a Crown Attorney's judgment in the prosecution of a case but rather with the deliberate and malicious use of the office for ends that are improper and inconsistent with the traditional prosecutorial function.

Therefore it seems to me that the “chilling effect” argument is largely speculative and assumes that many suits for malicious prosecution will arise from disgruntled persons who have been prosecuted but not convicted of an offence. I am of the view that this "floodgates" argument ignores the fact that one element of the tort of malicious prosecution requires a demonstration of improper motive or purpose; errors in the exercise of discretion and judgment are not actionable.”[8]

[15] In addition, Lamer J. described the fundamental flaw in a policy of absolute immunity, as follows:

“The fundamental flaw with an absolute immunity for prosecutors is that the wrongdoer cannot be held accountable by the victim through the legal process. As I have stated earlier, the plaintiff in a malicious prosecution suit bears a formidable burden of proof and in those cases where a case can be made out, the plaintiff’s Charter rights may have been infringed as well. Granting an absolute immunity to prosecutors is akin to granting a license to subvert individual rights.”[9]

[16] Lamer J. concluded that a narrowly defined tort was needed in those cases where public confidence would be eroded by virtue of truly malicious conduct. For him, the type of malice or ‘improper purpose’ that must be proved by the plaintiff is akin to that which perpetrates a fraud on the criminal justice system:

“In my view this burden on the plaintiff amounts to a requirement that the Attorney General or Crown attorney perpetrated a fraud on the process of criminal justice and in doing so has perverted or abused his office and the process of criminal justice. In fact, in some cases this would seem to amount to criminal conduct.”[10]

[17] In order to ensure reasonable accountability, the Court preferred the use of effective striking out provisions (such as Rules 20 and 21 of the Ontario Rules of Civil Procedure) to absolute immunity. The efficacy of those rules in the present context will be discussed later.

The Tort of Malicious Prosecution: An Analysis of the Four Nelles Requirements

[18] The focus of the following analysis is upon the third and fourth elements of the test in Nelles, as their consideration and interpretation in the ensuing jurisprudence are the primary reasons for the expression of concern in this paper. However, the first and second elements have not gone unnoticed and deserve some attention as part of a comprehensive review.

That the proceedings were initiated by the defendant

[19] The test for “initiating” a prosecution goes back to the time in England when private individuals commenced all prosecutions. Under Ontario law, individuals may lay an information, but there are no longer private prosecutions of criminal charges, as the Crown takes over all prosecutions and exercises discretion in the usual course. In the modern context, the word “initiated” applies to the police or individual who laid the charges, and to each Crown attorney who continues the prosecution.

[20] In Nelles, Lamer J. said that the meaning was “straightforward” and spoke for itself.[11] Nelles and a variety of subsequent cases refer to Fleming’s Law of Tort for further explanation, which reads:

“To incur liability, the defendant must play an active role in the conduct of the proceedings, as by ‘instigating’ or setting them in motion... The defendant must have been ‘actively instrumental’ in setting the law in motion. Simply giving a candid account, however incriminating, to the police or magistrate is not the equivalent of launching a prosecution: the critical decision to prosecute not being his, ‘the stone set rolling [is] a stone of suspicion only’.”[12]

[21] Fleming notes that “candid” in this context implies if the account is intentionally false, then the accuser may be liable. Private parties can therefore “initiate” too – in Sabourin v. Ontario,[13] the judge concluded that (if proven true at trial) an individual who provided false information to the police resulting in a charge against another individual would have “initiated” the charges against that person.

[22] In Miazga, it was also held that the prosecution was “initiated” once the information was sworn and filed, and to be held as responsible for “initiating”, the test was whether or not the charges would have not been filed “but for” the actions of the initiator. In Walker v. Ontario,[14] the motions judge declined to decide if “initiating” the proceedings could only apply to the person who made the ultimate decision to lay charges. However, Fleming has updated his Torts text since it was referred to in Nelles – the Ninth Edition now notes that “a defendant may be liable not only for initiating, but also for adopting or continuing proceedings”;[15] this meaning was cited with approval in Bairnard v. Toronto Police Services Board.[16]

[23] In Wiche v. Ontario[17] the court cited with approval Bernan v. Jenson[18] as holding that a strict definition of “initiate” was unhelpful. The court in Bernan asked, “was the defendant actively instrumental in putting the law in force?... Mere passive knowledge or acquiescence or consent in the acts of another is not sufficient to make one liable; in order to impose liability there must be some affirmative action by way of advice, encouragement, pressure, etc., in the institution, or causing the institution, of the prosecution or in affirmatively encouraging its continuance after it has been instituted” (emphasis added).[19]

[24] Similarly, in Wilson v. Toronto (Metropolitan) Police Service,[20] Drambrot J. concluded that Lamer J.’s use of Fleming’s test in Nelles was not intended to restrict the liability of Crown attorneys to cases where that counsel was actually involved in the initiation of the prosecution, since typically in Ontario the prosecution is initiated when a police officer swears an information; only subsequently does the prosecutor review the case and determine whether or not to proceed.[21]

[25] The totality of the jurisprudence seems to suggest then that where a Crown attorney adopts or continues a case that was previously a file of a different counsel, they would still be liable for malicious prosecution even though they did not technically “initiate” the case; indeed, this was the result in Miazga.

That the prosecution was terminated in favour of the plaintiff

[26] Recently, the Court of Appeal for Ontario rendered its decision on an appeal from a decision by the Superior Court of Justice summarily dismissing, respectively, tort claims in malicious prosecution against certain Crown Attorneys and malicious prosecution and negligence claims against the Niagara Regional Police Department in Mammoliti v. Niagara Regional Police Service (“Ferri”).[22] Until that decision, it was fairly settled law that to meet the second Nelles criteria, the prosecution must have been terminated in one of three ways: the charges were withdrawn, stayed, or the accused was found not guilty after trial. Criminal matters dealt with by way of plea bargain or a restorative process (whereby charges are withdrawn in exchange for some form of restitution by the plaintiff in the criminal matter) were deemed to be settlements that were not considered to have terminated the prosecution in the plaintiff’s favour.[23]

[27] A “favourable termination” includes an acquittal (whether at a preliminary enquiry, trial, or appeal), unilateral withdrawal of charges or a stay of proceedings by the Crown even if the Crown retains the right to later proceed with the prosecution if it so desires.[24] However, the jurisprudence also indicates that where a compromise or settlement (such a peace bond, conditional sentence, or plea bargain) results in the conclusion of a prosecution, the prosecution is not “terminated in favour” within the requirements of a malicious prosecution action.[25] In this vein, in Hunt v. Ontario[26] it was found that prosecution was not terminated in favour of the accused when the charges were withdrawn on the condition they pay an amount to charity.

[28] This “arrangement” logic also applies to guilty pleas. In Ramsay v. Saskatchewan,[27] Zarcenzy J. found that the plaintiff could not succeed in his malicious prosecution claim since he had pled guilty to the original charges, and it could not therefore be said that the proceedings had terminated in his favour. In Hainsworth v. Ontario (Attorney General),[28] Sutherland J. found that a plea bargain that resulted in guilty pleas in two original charges in exchange for three further charges being dropped also did not mean that the proceedings had been terminated in the accused’s favour.

[29] The function of the second element of the Nelles test is to screen out cases where the accused has been found guilty beyond a reasonable doubt, and is therefore properly estopped from making an allegation of malicious prosecution. Where there is a finding of guilt, the proper recourse for the offender is the criminal appellate route, and not the civil courts.

[30] In Ferri, Mammoliti was charged with extortion and possession of property over $5,000. Ferri was initially charged with attempted extortion and theft over $5,000. The two matters were joined, and following a preliminary hearing, Ferri was committed on the charge of theft, while Mammoliti was committed on both the theft and attempted extortion charges. On review, the committal against Ferri was quashed, and the committal on both charges against Mammoliti was upheld. On the eve of trial, the charges against Mammoliti were withdrawn on certain condition that bank documents would be dealt with in a certain way, pending the outcome of an anticipated civil suit, and that Mammoliti would bring such a suit within 30 days. It appears clear from the conditions, all parties were aware that Mammoliti intended to pursue a malicious prosecution claim against TD Bank.

[31] Writing for the majority in Ferri, LaForme J.A held that before a settlement is deemed to be a bar to a claim of malicious prosecution, the Court must conduct “further analysis into the underlying reasons on the part of the Crown and police for entering into such an arrangement, agreement or compromise.”[29] Policy reasons dictate, argued that court, that a settlement in the form of a withdrawal on certain conditions should not automatically bar a malicious prosecution suit. Were this to happen,

“the Crown or police could avoid scrutiny by simply entering into an agreement, arrangement or compromise with an accused person, no matter how trivial, before withdrawing the charges against that person. This would have the effect of completely undermining the purpose of the tort.”[30]

[32] The Court found that the NRPS could not use the existence of a settlement to thwart a claim of malicious prosecution because it was consulted about the potential resolution, and therefore knew of Mamolliti’s intention to seek civil redress. LaForme J.A. noted in this regard that police and Crowns are always acting from a position of strength vis a vis the accused:

“The accused person, who believes he or she was the subject of a malicious prosecution, would effectively be giving up his or her right to sue even in the clearest of cases in exchange for his or her freedom, to avoid the stigma of a criminal conviction, or perhaps, to avoid the costs of a trial.”[31]

[33] This scenario overlooks the other safeguards that protect the accused and ensure the continued proper administration of justice, including the overarching and supervisory role of the Judge over criminal proceedings, and the existence of a high burden on the Crown to prove the accused’s guilt beyond a reasonable doubt. The Martin Report succinctly described the nature of the differing roles of Crown, defence and judge, as follows:

“It cannot be forgotten that Crown counsel is only one of the participants in an adversarial system of justice. Crown counsel performs his or her duties in the context of a system that ascribes fundamentally important, and counterbalancing responsibilities to defence counsel, and to the judge. It is the interaction of these three parties in the criminal process, and not the action of Crown counsel alone, that ensures just outcomes to criminal proceedings.” [32]

[34] LaForme J.A.’s reasoning is also unsettling given the distance that the court traveled beyond the facts of the case before it, to conjure up a scenario whereby it could be argued that a settlement should not act as an estoppel to a claim of malicious prosecution, because the settlement itself is furtherance of the improper conduct. The court appears concerned that Crown attorneys and police could conspire together to manufacture a trivial resolution to cover their tracks with respect to other abusive conduct in the course of the same prosecution, when the facts of the case before the court reflected a settlement which was entered into by Mamolitti with the assistance of counsel, and which produced a result which was responsive to the victim’s concerns, provided no unjust enrichment, and which guaranteed immediate protection of the Bank records. The TD Bank files were to be immediately removed from Mamolitti to a place of safekeeping pending the outcome of a civil process. There is nothing in the decision to suggest that the NRPS (or the Crown for that matter) even knew that a settlement would bar a claim of malicious prosecution.

[35] Indeed, before a prosecution can proceed, a Crown must determine that not only is there a reasonable prospect of conviction, but that the prosecution is also in the public interest. The consideration of the public interest is often case specific, and requires consideration of a broad number of diverse factors[33] such as the gravity of the incident, the circumstances and attitude of the victim, the availability of compensation, restitution, or reparation and the availability and efficacy of alternatives to prosecution. It is arguable on the facts of Ferri before the Court of Appeal, the settlement proposed by the Crown and approved and adopted by the Court in the criminal proceedings was entirely consistent with a proper application by the Crown of these kind of public interest factors and was therefore entirely consistent with a proper exercise of the Crown’s prosecutorial discretion. Yet, the court’s analysis of the policy ramifications indicates an undercurrent of suspicion that Crowns and police have sinister motives, which are real enough to require that even settlements entered into by Crowns that result in withdrawal of charges are not beyond scrutiny as a potential forum for misconduct. If a settlement automatically disqualifies an accused from a future claim of malicious prosecution without any analysis of the legitimacy of the settlement and the Crown’s conduct in relation to it, then, suggests the Court of Appeal, police and Crowns are at liberty to convert what could have been an outright withdrawal into one with conditions for the sole purpose of shielding them from a legitimate claim of malicious prosecution. There is no deference shown to the Crown in recognition of the many competing factors that they must consider in the determination of whether there is a reasonable prospect of conviction, and whether the prosecution is in the public interest.

[36] What does this mean for Crowns in the context of their other obligations as ministers of justice? The Martin Report described as a fundamental principle of justice the idea that there must be sufficient evidence of a crime for a prosecution to be initiated, and that the prosecution must also be in the public interest. Even if there remains a reasonable prospect of conviction, it may be in the public interest to opt for a pre-trial resolution that will satisfy the ends of justice having regard to all of the circumstances of the case, input from the investigating officer and the victim. Indeed, a pre-trial resolution which has the best likelihood of restoring the victim to his or her position before the alleged incident through some form of restitution (even when the quid pro quo for requiring the accused to provide restitution is to withdraw the charge against the accused) may be the result which is best characterized as in the public interest. In light of the Ferri decision, however, a Crown’s effort to seek a restorative pre-trial solution may now be complicated by an overriding concern that the settlement may be scrutinized by a reviewing court in a malicious prosecution claim against that Crown to assess whether the conditions upon which the withdrawal of charges is premised, are necessary in the circumstances.

[37] LaForme J.A.’s reasoning leaves open the risk that if conditions to be met by an accused before a Crown withdraws charges are not deemed to be necessary by a reviewing court, then that too may be evidence of improper motive, which could support a finding of malice. The Crown may not only be preoccupied by the risk of an allegation of malicious prosecution when proceeding with a criminal prosecution, but may now likewise be preoccupied by the same risk when considering whether the continued prosecution is in the public interest. This creates the real danger that a Crown will no longer consider whether a potential settlement of a criminal matter before trial is in the public interest, but rather be preoccupied as to whether the proposed settlement may be dissected as potential evidence of malice in a suit by the accused against the Crown personally in a future action. Moreover, the ease in which the court is willing to second guess the exercise of Crown discretion, belies an appreciation of the complexity of the application of the test of reasonable and prospect of conviction, and the public interest factor to any given case by a Crown in both its role as an advocate and as a Minister of Justice. Lamer J. in Nelles particularly cautioned against this second-guessing of the exercise of Crown discretion.

That there was an absence of reasonable and probable cause

[38] Reasonable and probable cause to prosecute was defined in Hicks v. Faulkner as

an honest belief in the guilt of the accused based upon a full conviction, founded on reasonable grounds, of the existence of a state of circumstances, which, assuming them to be true, would reasonably lead any ordinarily prudent and cautious man, placed in the position of the accuser, to the conclusion that the person charged was probably guilty of the crime imputed.[34]

[39] The test contains both a subjective and objective element. There must be both actual belief on the part of the prosecutor and that belief must be reasonable in the circumstances. The existence of reasonable and probable cause is a matter for the judge to decide as opposed to the jury.

[40] There are two areas of concern in the way in which the courts have applied this test. First, the cases indicate an increasing judicial willingness to second-guess the Crown’s reasoning in determining that the prosecution should go forward. The difficulty with this approach to the determination of whether there is reasonable and probable cause is that criminal proceedings often involve complex charges and principles, and reasonable people may reasonably disagree about the sufficiency or quality of evidence, based upon their respective biases, style and experience. Where reasonable people can fairly disagree, it seems overly harsh to allow a malicious prosecution claim to be founded on those disagreements.

[41] For example, an acquittal in a criminal proceeding does not imply (in the absence of specific evidence to the contrary) that the system failed somehow to properly administer justice. Without anything further it simply means that the Crown did not displace the accused’s right to be presumed innocent; that is, the Crown did not prove to the satisfaction of a judge or jury the accused’s guilt beyond a reasonable doubt. As Sopinka J. has noted,

“The rationale of a tort of action for damages for malicious prosecution is that the court’s process has been abused by wrongfully invoking the law on a criminal charge. The tort has been restricted, however, to ensure that criminals can be brought to justice without making prosecutors fear an action for damages if a prosecution fails.” [35]

[42] Nor should an acquittal imply that the Crown improperly approached its assessment of whether there was sufficient evidence to proceed, or whether the prosecution was in the public interest: witnesses may not have testified at trial in accordance with their earlier interviews, cross-examination may have been particularly effective in revealing inconsistencies, and scientific evidence may have been faulty. These are all weaknesses that may only be exposed in the course of the trial. Indeed, even when a verdict is set aside on appeal, there should not be an automatic assumption that the Crown therefore must not have had a reasonable prospect of conviction. As the Martin Report notes, “the Court of Appeal has, of course, the benefit denied to a prosecutor of reviewing all of the evidence as it unfolded at trial, including any evidence called by the defence.”[36]

[43] An example of the danger of substituting the Crown’s views with that of the reviewing judge can be found in the Ferri decision. LaForme J.A. for the majority held that on his review of the facts “the constituent elements of the offences contemplated, and with which Mammoliti was subsequently charged [with theft over $5,000 and attempted extortion], were almost entirely absent.”[37] This finding directly contradicted the view of the motions judge on the original application for summary judgement, who found that the third element – a lack of reasonable and probable clause – was not satisfied. The motions judge held (in the claim by Mammoliti against NRPS) that Mammoliti’s request for $500,000 for the return of the bank’s files was sufficient to establish reasonable and probable cause for the police to charge him with extortion and theft. Juriansz J.A’s dissent drew a different conclusion again. Not only did Juriansz J.A. find that there was evidence sufficient to support reasonable and probable cause to believe that Mammoliti had committed the offences of theft and attempted extortion, but additionally, on the totality of the information the respondents possessed during the course of the prosecution, they were justified in believing they possessed sufficient evidence to prove guilt beyond a reasonable doubt. In his dissent, Juriansz J.A. cautioned about the inherent danger of reviewing a constellation of past events retrospectively:

“The determination of whether the respondents had reasonable and probable grounds to prosecute the appellants must be made on the basis of the information the respondents had at the time or could have obtained by reasonable inquiry. New facts that the appellants may offer now are not relevant in determining if the respondents had reasonable and probable grounds at the time.”[38]

[44] This example of reasonable judges making opposite findings reflects the reality that the application of the test is more of an art than a science. Any difference of emphasis in the review of evidence, especially in a motion for summary judgement process, can render a different finding. As Juriansz J.A. noted:

“On the view of the facts I take, the actions of the appellants had a character not apparent in LaForme J.A.’s reasons and lead me to conclude the appeals should be dismissed entirely…With this in mind, I offer the following account of the essential facts, which differs from the account set out by LaForme J.A. only in tone. I add detail to certain events and highlight some of the evidence the respondents had when they initiated and continued the prosecution of the appellants.”[39]

[45] For these reasons, a judge on a summary judgement motion should approach this element of the test cautiously; the tort of malicious prosecution is a serious one, and some deference should be given to Crowns in their decisions to pursue or abandon prosecutions.

[46] The second (and more pressing) concern is that there is a disconnect between the standard a court uses to review the decision to prosecute (reasonable and probable cause), and the standard a Crown is instructed to follow when deciding to prosecute (reasonable prospect of conviction).[40] As noted, the test as set out in Nelles to show absence of reasonable and probable cause has both a subjective and objective component, in that the Crown must not only believe in the guilt of the accused, but that belief must be reasonable on the facts before him. However, a Crown is duty bound to proceed with a prosecution in Ontario when there is a reasonable prospect of conviction, and so long as that prosecution is in the public interest. This latter threshold was crafted and recommended by the Martin Report after a careful review of the role of the Crown in the proper administration of justice, as well as existing threshold tests in other jurisdictions. There is no subjective element to this screening requirement. Indeed, the Martin Report cautions prosecutors about the inherent danger of making a threshold decision about the future of a prosecution on the basis of one’s own personal opinion:

“If only those case were prosecuted in which Crown counsel firmly believed in the guilt of the accused, the settled notion that “the purpose of a criminal prosecution is not to obtain a conviction” may well be compromised in practice by prosecutors who, having formed the opinion that the accused is guilty, would therefore see it as their duty to obtain a conviction. …

Crown counsel need not and ought not to be substituting his or her own views for those of the trial judge or jury, who are the community’s decision makers. It cannot be forgotten that much of the public’s confidence in the administration of justice is attributable to the trial court process that ensures that justice is not only done, but is seen to be done… Granting Crown counsel the power to initiate or discontinue prosecutions based on a subjective assessment of whether or not the accused is guilty would, in some circumstances, be tantamount to replacing these open, impartial, and community-based processes with the unexplained, unreviewable decisions of prosecutorial officials, who have no direct accountability to the public (emphasis added).”[41]

[47] The Martin Report also addresses situations where the Crown may personally harbour some doubt about the guilt of the accused. In such cases, the Crown:

“ …is duty bound to carefully explore the reasons for that doubt as they might be revealed in the Crown Brief or investigative file, and to recommend any further investigations as appear necessary... If, however, following such a review and investigation, there remains a reasonable prospect of conviction, and if the prosecution is otherwise in the public interest, the prosecution should usually proceed. On the other hand, the prosecutor's belief in the guilt of the accused counts for nothing if, on the evidence, there is no reasonable prospect of conviction... “[42]

[48] The Martin Report also cautions that at different stages in the criminal process, there are different standards of proof. Police need only have reasonable and probable grounds to believe that an offence has been committed in order to lay a charge. Crowns must meet a higher standard of ensuring that there is a prima facie case and that the prosecution is otherwise in the public interest in order for the case to be prosecuted. But since an accused cannot be convicted unless there is a proof beyond a reasonable doubt, as determined by a Judge or jury of one’s peers, for a Crown to assess a prosecution and make decisions to prosecute based on his or her personal views as to the accused’s guilt is inappropriate:

“As proof beyond a reasonable doubt is a higher standard than whether there is a reasonable prospect of conviction, it may well be that a reasonable doubt about the guilt of the accused can exist notwithstanding that the latter standard is met. However, in the Committee’s view, the existence of such a doubt cannot, generally speaking, justify preventing the community from passing judgment as to the guilt or innocence of an accused where there is a reasonable prospect of conviction and the prosecution is otherwise in the public interest.”[43]

[49] Finally, the Martin Report speaks to the extent to which character assessment should factor in to a determination of whether there is a reasonable prospect of conviction:

“Accordingly, assessments of demeanour and other characteristics of prosecution witnesses are best left for the public forum of the trial court, where any errors in judgment can be best left for the public forum of the trial court, where any errors in judgment can best be prevented or corrected. Ultimately criminal justice should be administered even-handedly, which can best be ensured by having an objective test focused on the quality of the Crown’s case, rather than on the subjective assessments of particular Crown attorney. Thus, the review of credibility or other capacities of witnesses undertaken for the purpose of the threshold test should be founded on objective indicators, such as incontrovertible evidence from an independent source that a particular witness is mistaken or lying. A decision as important as who is or is not to be prosecuted should not depend on the happenstance of who is assigned to the case.”[44]

[50] Furthermore, in recent years, there has been a trend toward greater involvement by senior Crown attorneys in decisions to prosecute. This may happen in several ways. First, with increasingly complex cases, as well as concerns about mal pro actions, junior Crowns often consult with senior Crowns about whether to proceed. It is desirable that juniors are properly mentored, but the result is often that their decisions are not based on their own experience and assessment of a case, but on the opinion of someone more experienced. In this situation, it would be improper to rely upon the junior counsel’s “subjective” belief as to the guilt of the accused, as it played no role in whether or not the case was prosecuted. The same issue raised by Crown attorneys who are taking a more active role in managing assistant Crown attorneys, and may in some cases direct that a prosecution go forward, even though the assistant Crown attorney has a different view of the evidence or would prefer to withdraw the charges for various reasons, including fear of a civil action. This is a very real problem, especially in smaller cities and towns.

[51] Interestingly, as it stands now, the test to intervene to review a decision by a Crown not to proceed with a prosecution is higher than the test to establish that the Crown proceeded with a prosecution maliciously (see Respondent’s factum in Huixia chen et al.). An applicant must establish ‘flagrant impropriety’ on the part of the Crown in an application for judicial review to compel the Crown to lay charges. Flagrant impropriety as a threshold for intervening with Crown discretion not to proceed with a prosecution is the opposite side of the same coin that allows for court review after the fact, when there is an allegation of malicious prosecution. On that point, the Supreme Court noted in Krieger v. Law Society (Alberta) that “to this point, it appears that there has been no case in which any Court has actually set aside either a refusal to give consent to prosecute or a decision to enter a stay of proceedings where a prosecution had already been commenced.”[45]

[52] In R. v. Power[46] the Supreme Court said that a court must exercise extreme caution before embarking on a review of prosecutorial discretion, finding that it should only be reviewed in cases of flagrant impropriety or malicious prosecution. The Court noted:

“…that courts have been extremely reluctant to interfere with prosecutorial discretion is clear from the case law. They have been so as a matter of principle based on the doctrine of separation of powers as well as a matter of policy founded on the efficiency of the system of criminal justice and the fact that prosecutorial discretion is especially ill-suited to judicial review.”[47]

[53] The Court also explained what kind of evidence would be sufficient to constitute flagrant impropriety, holding that it would have to “shock the conscience of the community and [be] so detrimental to the proper administration of justice that it warrants judicial intervention” or that it showed “conspicuous evidence of improper motives or of bad faith.”[48] Such findings sufficient to warrant court intervention compelling a Crown to proceed with a prosecution will necessarily be extremely rare. It is inconsistent to require that a court have clear conspicuous evidence of impropriety when questioning the discretion of a Crown to not proceed with a prosecution, yet require something less to permit a court to second guess Crowns when exercising the same discretion to proceed.

[54] The danger now is that in order for Crowns to protect themselves from allegations of malicious prosecution solely on the basis of whether they believe in the guilt of the accused, they will approach cases not with a dispassionate eye, but will only prosecute those in which they feel a personal belief that the accused has in fact committed the crime in question. Such an approach is inherently in conflict with the Crown attorney’s recognized quasi-judicial role as a Minister of Justice.

[55] It is arguable that reasonable prospect of conviction standard is a more stringent than the Nelles criterion, which comprises the lower standard of “reasonable and probable grounds” (for the police) to lay a charge and (for Crown attorneys) to prosecute. While “reasonable prospect of conviction” is a higher threshold to cross with regard to the objective quality of the available evidence, the fact that it does not contain a subjective view of the accused’s guilt, unlike the Nelles standard, creates a problem. The differing standards between Nelles’ ‘reasonable and probable cause to prosecute’ (both a subjective and objective standard) and ‘reasonable prospect of conviction’ (an objective standard only) have created a dangerous gap which can allow a Crown who believed he or she was duty bound to continue a prosecution because the evidence showed a reasonable prospect of conviction to later be named in a malicious prosecution suit if they lacked subjective belief in the guilt of the accused. While the fourth Nelles factor (requiring malice) was intended to guard against this possibility, the post-Nelles cases have increasingly conflated the third and fourth criteria, removing this safeguard.

That there was malice or that the primary purpose of the prosecution was other than that of carrying the law into effect.

[56] In Nelles, Lamer J. was clear that malicious prosecution requires not only absence of reasonable and probable cause, but also specific evidence of malice:

“To succeed in an action for malicious prosecution against the Attorney General or Crown Attorney, the plaintiff would have to prove both the absence of reasonable and probable cause in commencing the prosecution, and malice in the form of a deliberate and improper use of the office of the Attorney General or Crown Attorney, a use inconsistent with the status of ‘minister of justice’. In my view this burden on the plaintiff amounts to a requirement that the Attorney General or Crown Attorney perpetrated a fraud on the process of criminal justice and in doing so has perverted or abused his office and the process of criminal justice. In fact, in some cases this would seem to amount to criminal conduct.”[49]

[57] It is the distinctiveness of this fourth element that serves to shield Crowns from liability in cases where decisions are made in good faith but based upon poor judgement or lack of experience; it should also shield them where they have followed their training to prosecute based on a “reasonable prospect of conviction”, supra.

[58] Unfortunately, post-Nelles, courts have increasingly conflated the third and fourth factors so as to remove this final safeguard that the Supreme Court viewed as so important, culminating in the Miazga decision (infra).

[59] How did we arrive at Miazga? The Supreme Court itself is not to blame – since Nelles, the Court has only revisited the tort of malicious prosecution once, in Proulx v. Quebec (Attorney General)[50] in which the majority of the Court held the Crown attorney liable for malicious prosecution. In Proulx, the defendant Crown attorney had originally concluded in 1985 that he did not have sufficient evidence to prosecute Proulx, but reopened the case in the early 1990s after a witness came forward with new identification evidence identifying Proulx as someone he saw near the scene of the crime. The evidence was that the Crown knew that there were frailties in this new identification evidence. This same Crown attorney employed a police detective (Tardif) to look into the evidence against Proulx. The Crown was aware that Tardif was entangled in a civil action for defamation launched against him by Proulx that could be perceived to cloud the officer’s judgment. Notwithstanding this knowledge, the Crown allowed Tardif to gather evidence against Proulx for use in criminal proceedings. The Court inferred malice from the fact that Crown attorney knew about the conflict which existed given Tardif’s prior relationship with Proulx and yet allowed his involvement in any event, thereby allowing the prosecutorial process to be used as a forum for Tardif to potentially collect evidence to be used as leverage in his civil case. This was in keeping with the Nelles decision, which Proulx specifically upheld, because the evidence of the relationship between the Crown, Proulx and Tardif, was, according to the Court, specific evidence of malice beyond mere or gross negligence. Iacobucci and Binnie JJ. wrote that a “suit for malicious prosecution must be based on more than recklessness or gross negligence.”[51] The Court also found that the Crown attorney, in his address to the jury, had deliberately manipulated a speculative comment made by the Plaintiff into a purported confession. This was deemed to be evidence of malice; importantly, there is nothing in Proulx which talks of inferring malice from a mere absence of reasonable and probable cause.

[60] The Court of Appeal for Ontario decision in Oniel v. Toronto[52] has arguably provided fodder for later findings in support of the notion that malice can be inferred from the absence of reasonable and probable grounds to prosecute. It was rendered prior to Proulx but was not referred to in the higher Court’s decision, perhaps because Oniel was an allegation of malicious prosecution brought against police officers as opposed to Crowns.

[61] In Oniel, the plaintiff had been investigated by several police officers after they received a complaint from an individual that the plaintiff had stolen his watch and assaulted him. The police investigated and charged Oniel. Oniel retained counsel who in turn advised the police that his client had been misidentified, asking them to contact two witnesses who could confirm the accused’s story. The police made no further calls and the accused was acquitted at trial. Oniel brought an action against the investigating officers. He did not sue Crown attorney. The Court of Appeal held that the police officers had prosecuted the plaintiff without cause, notably because they had not verified the plaintiff’s evidence corroborating the fact that he was not at the scene of the crime, despite the fact that they were explicitly asked to do so. In the court’s consideration of the malice component of the malicious prosecution test, it held that it was malicious for the officers to continue the prosecution after they deliberately or recklessly ignored the plaintiff’s evidence proving the unreliability of the complainant’s evidence. The court determined that malice could be proven by either pointing to the improper purpose motivating the officers, or by showing tha they were reckless in their investigation.[53] Borins J.A. stated that, in addition to the more traditional ways in which malice could be established, malice can be inferred from police conduct which is recklessly indifferent to the truth or to the guilt or innocence of the accused and further that it could be inferred from proceeding with a prosecution in the absence of reasonable and probable grounds to do so.[54]

[62] In holding that proof of recklessness in the face of an absence of reasonable and probable cause constitutes proof of malice, Oniel directly contradicted Nelles, in which the Court held that proof of recklessness is not sufficient on its own to constitute malice. It is arguable that the lower bar set by Oniel for proving malice is related to the fact that, unlike in Nelles (and later Proulx), the action was against police officers rather than Crowns. While Borins J.A. accepted the Nelles test, it was with the following caveat:

“The explanation of malice by Lamer J. in Nelles … is a useful starting point in considering what the authorities have accepted as constituting malice … it is helpful to reproduce what Lamer J. had to say about the tort at pp. 192-93, omitting, however, his view of what a plaintiff must establish respecting malice where the defendant is an Attorney General or a Crown Attorney …”[55]

[63] Oniel recognized that a distinct malice test is required because of the role police officers play in the investigation and prosecution of crime. Borins J.A. accepted that it is police officers who have the training and experience to investigate alleged crimes and the powers needed to conduct a proper investigation in order to critically weigh the evidence. He noted that although Crowns make the final decision whether to prosecute, they rely on the police to conduct a thorough investigation and to provide accurate and complete results of such investigation. This reflects the distinctive roles played by the police and Crown attorneys and supports the fact that such distinctive roles require different approaches to the consideration of when Crowns and police should be found liable for malicious prosecution.[56]

[64] One of the first cases to consider Oniel and Proulx together was Dix v. Canada (Attorney General),[57] in which Crown attorney was found liable for malicious prosecution. In considering the malice factor, the court said:

“(a) there are sufficient indicia of malice on the facts to prove that element of the test discretely, for example intentional Charter breaches and misleading the Court on a bail application, without requiring recourse to lack of reasonable and probable cause or any of the inferences of malice which may arise therefrom;

(b) even if there were not sufficient indicia of malice on the facts to prove the fourth element apart from lack of reasonable and probable cause, several of the factors contributing to lack of reasonable and probable cause, for example the pursuit of, and reliance on, clearly inadmissible evidence, can be considered as factors going to proof of malice and, once coupled with the other indicia of malice not already counted as being related to lack of reasonable and probable cause, are, all tolled, sufficient to prove malice under the fourth part of the test; and

(c) even if there were insufficient indicia of malice on the facts under "a)" above, or insufficient indicia of malice on the facts when coupled with inferences of malice based on certain factors relating to the absence of reasonable and probable cause under "b)" above, prosecution in the face of evidence that the Plaintiff was probably not guilty permits an inference of malice under Oniel.

Put shortly, the conduct of the police and prosecutors in this case not only establishes pursuit of an improper purpose apart from any lack of reasonable and probable cause; but also that, as there was in fact a lack of reasonable and probable cause, several of the factors causing that lack also contribute to proof of malice which, when coupled with other factors, is sufficient to prove malice; and apart from these first two proofs, prosecution of the Plaintiff in the face of evidence that he was probably not guilty itself permits an inference of malice. Finally, although each of these three alternative proofs overlap somewhat in their factual and inferential foundations relating to proof of malice, the point is that I find that the Plaintiff has, regardless, proved, discretely under each of these alternatives, that malice exists. The defendants are, quite simply, legally cloaked in malice.”[58]

[65] The court therefore found that the lack of reasonable and probable cause could indicate the existence of malice, though it seemed careful to say that it was the particular factors that meant there was no reasonable and probable cause (in this case, clear evidence of innocence) that meant malice could be inferred. This seems distinct from saying any lack of reasonable and probable cause can be inferred as evidence of malice.

[66] This can be contrasted with the decision of Mazumder v. Ontario,[59] in which the defendant Crown and four Crown attorneys moved for summary dismissal of the plaintiff Mazumder's claim on the grounds that there was no genuine issue for trial and the action was an abuse of process. Mazumder sued for malicious prosecution, claiming damages for his arrest and wrongful conviction on charges of harassment. The charges related to harassing calls to Mazumder's ex-wife's place of employment. Mazumder had been convicted at trial, but was acquitted on appeal on the basis of new evidence. The motion was dismissed. The actions as they related to two of the Crown attorneys were dismissed, as they had been involved in the very early stages and only peripherally. However, there were genuine issues to be tried in Mazumder's claim for malicious prosecution, which could only be decided by a trial. It would be open to a trial judge to find the action allowable, if the facts pleaded were proven. Therefore, the actions against the other two Crown attorneys were not dismissed. The court held that in Oniel, it was found that malice must be proved by the plaintiff, but that it can be done in one of two ways: either by identifying the specific motive, or by showing that the the prosecution can only be accounted for through a finding of malice, even if the specific improper purpose is unknown.[60]

[67] Despite the fact that the ratio of Oniel was in the context of malicious proscution actions against police officers, the Court in Mazunder stated that the approach taken by the Court in Oniel applied equally to the Crown as well. Unfortunately, there is nothing to indicate that the Court turned its mind to the inherent differences between the role of the Crown and the role of a police officer, or the different standards with which these different justice players must approach the evidence when deciding whether to move the matter forward.

[68] In Scott v. Ontario,[61] the court ruled that a committal to trial at a preliminary inquiry precludes an inference of malice based on the absence of reasonable grounds. While this would seem to be a positive development, Scott was followed by Coulter v. Toronto Police Services Board et al.,[62] in which the court ruled that reasonable and probable cause must exist at every stage of the prosecution, as the evidence and other circumstances may change. In essence then, a valid initial prosecution could later result a malicious prosecution charge.

[69] The problems with the continual devaluing of Nelles were most clearly revealed in the recent Saskatchewan Court of Appeal decision Miazga, which ultimately concluded that a Crown’s subjective views about the accused’s guilt or innocence spoke directly to the existence of reasonable and probable cause and may in turn be evidence of malice. The majority of the court dismissed the appeal of the Crown attorney, Miazga, who had contended that the trial judge had erred in finding that there was no reasonable and probable cause to charge the respondents, and that there was malice in the prosecution. The trial judge had found that Miazga had no subjective honest belief in the guilt of the accused and that objectively, reasonable and probable grounds to prosecute did not exist, and, as a corollary, this meant malice must have been the driving reason behind the prosecution. The majority of the Court of Appeal relied on Proulx for the proposition that in determining the issue of malice, the totality of all circumstance is to be looked at. The trial judge had attributed malice to Miazga for a variety of actions, including failing to investigate the matter properly. The Court of Appeal, however, found that this was an error in law – it was the responsibility of the police, not the Crown, to investigate the accused. With regard to other actions by Miazga (for example, failing to interview certain witnesses prior to charges being laid, or giving pre-charge advice to the police, failing to apologize to the accused), the Court of Appeal found that the trial judge had failed to understand the role of the Crown and had also failed to distinguish between malice and errors in judgment or negligence, which are not actionable.

[70] The majority judgment therefore focused on the lack of reasonable and probable cause as proof of malice. The trial judge had made a finding of fact that Miazga did not have an honest belief that the respondents were guilty of the offences charged. The majority found that this finding of fact could not be interfered with in the absence of palpable and overriding error, which they did not find. They argued that it was reasonable for the trial judge to have found statements that had led to the prosecution to be so unbelievable that Miazga could not have had an honest belief in the guilt of the respondents, and therefore no reasonable and probable cause to prosecute. Of note, children made the “statements” in question, and Miazga himself admitted that he was not sure about the veracity of all the statements. The Martin Report points out, however, that the nature of testimony of child witnesses is necessarily difficult to gauge, and in such a situation, “it may not be possible to say that a conviction is more likely than not.”[63] Nonetheless, in the majority’s view, not subjectively believing some of the statements to be true meant that there was no reasonable and probable cause, which was the same as proof of malice, and thus the Nelles elements of malicious prosecution were made out, and Miazga’s appeal was dismissed.

[71] By taking the absence of reasonable and probable cause as evidence of malice, the majority therefore collapsed the third and fourth factors of the Nelles test: the third element is an objective test, while the fourth is subjective. Both must be made out to constitute malicious prosecution. Here, there was no evidence to support the conclusion that Miazga prosecuted out of malice; that is, there was no evidence that he had a primary purpose other than that of carrying the law into effect, nor was there evidence he had used his office for an improper purpose.

[72] The decision in Miazga therefore removes the final safeguard that Lamer J. created in Nelles to avoid a floodgate of unmeritorious claims against Crown attorneys. If courts in other jurisdictions follow Miazga, Crown attorneys will be increasingly exposed to frivolous claims even when they prosecute individuals based upon their belief that, on the evidence, there is reasonable prospect of conviction. Further, Crown attorneys will be exposed to charges of malicious prosecution even when guilty only of professional negligence. While such negligence is indeed a serious matter, there are actions that can be taken against negligent counsel through the various law societies and professional bodies across Canada. The tort of malicious prosecution should be reserved for what its very name implies – a malicious intent on the part of a Crown to prosecute an innocent individual.

[73] In Coulter, the Crown bought a motion for summary judgment supported by an affidavit from the Crown attorney attesting that she had acted in good faith in proceeding with the prosecution and setting out the reasons why she proceeded. The plaintiff’s supporting affidavit simply stated that he believed the Crown attorney had acted with malice, but gave no particulars. This was enough for the motions judge to say that there was a genuine issue for trial on the question of malice. This is further evidence of an erosion of the Crown’s ability to get motions struck at an early stage.

[74] It is submitted then that the recent jurisprudence has tipped the delicate balance sought by Lamer J. in Nelles between the need to provide protection for the community from specific maliciousness on the part of the Crown in the exercise of his or her duties, and the need to ensure sufficient freedom for the Crown to focus on the difficult task of determining whether there is a reasonable prospect of conviction, and that the prosecution is in the public interest, at all points along the criminal justice spectrum, from point of charge through to resolution.

[75] Ferri indicates a willingness of the Court to second-guess the Crown’s exercise of discretion, not only to assess whether the Crown had reasonable and probable grounds to believe in the guilt of the accused, but as well in the exercise of Crown discretion in determining how the prosecution should be dealt with in the public interest. Miazga suggests that malicious conduct by a Crown attorney can be inferred from a reviewing judge’s finding of lack of reasonable and probable cause. This belies the fact that with no discrete evidence of malice while a finding of a lack of reasonable and probable cause may be consistent with malicious conduct, it may equally be consistent with inexperience or systemic issues such as a lack of proper resources that prevent a more fulsome Crown review of the matter. It may equally be consistent with negligence, a tort from which the Crown is supposedly immune. Finally, it may simply reflect a different interpretation of the facts vis a vis the prosecuting Crown and the reviewing judge. Indeed, it appears from Miazga that the Crown’s personal view of the guilt of the accused may weigh heavily in the assessment of whether reasonable and probable cause existed to prosecute. This is a particular risk if judges approach the analysis through their subjective view of the facts, rather than on an assessment of whether the exercise of the Crown’s discretion was unreasonable on the facts. If the tort of malicious prosecution against a Crown no longer requires discrete evidence of malice, then Crowns can have no comfort that so long as they exercise their discretion in good faith, they need not worry about allegations of malicious prosecution. This represents an enormous shift in the tort of malicious prosecution. It reduces evidence of malice from a necessary element to a sufficient element in the malicious prosecution equation and ignores the difference between unintentional negligence and deliberate malice. The policy balance in the Nelles test is now imbalanced, and Crowns have no clarity about the proper exercise of their discretion.

[76] The evolution of the jurisprudence post-Nelles threatens to undermine the important quasi-judicial role Crown attorneys play as “Ministers of Justice”. There is a very real risk that the “chilling effect” will become real, as Crown attorneys avoid difficult prosecutions in order to protect themselves from frivolous malicious prosecution actions. To prevent the administration of justice from declining in such a manner, and to protect Crown attorneys from the embarrassment and stress of fruitless malicious prosecution actions, it is proposed that serious consideration be given to implementing solutions that restore the policy rationale of Nelles. The solutions may be legislative or non-legislative, with corresponding benefits and drawbacks.

[77] (It should be noted that the Ontario Attorney General has sought leave to appeal[64] Ferri from the Supreme Court, and while leave has not yet been sought in the more recent case of Miazga, the time to do so has not yet expired (as of the writing of this paper).

The Availability of Other Torts Against Crown Attorneys

[78] In Driskell v. Dangerfield,[65] the court refused to strike out a claim in negligence against a Crown attorney who was sued for failure to disclose evidence in a case where the plaintiff had been convicted of murder. The Court held that disclosure was not within the scope of activities for which there was prosecutorial immunity, and the law as to whether an action in negligence lay against prosecutors was unsettled.

[79] The courts appear to be moving away from the Nelles criteria and allowing cases to proceed where malice, misfeasance in public office, abuse of office and similar claims are made against Crown attorneys. The most important development in this regard was Folland v. Ontario[66] in which the Court of Appeal held that the jurisprudence was not fully settled as to whether the four elements for the tort of malicious prosecution must always be proven in every civil action against a prosecutor. The court said it was also not clear that, if the four elements were established, Folland would be restricted to framing his action as one of malicious prosecution rather than as one of conspiracy, abuse of process or intentional infliction of harm. Worryingly, the court concluded that prosecutorial misconduct was a broad term that could encompass more than malicious prosecution. In its application for leave to appeal to the Supreme Court, Ontario provided affidavit evidence from the Deputy Attorney General about the consequences to Crown attorneys of this decision, but the Court nonetheless declined to grant leave.

[80] The general approach of Folland was used in McNutt v. Canada (Attorney General)[67], in which the court found that prosecutorial misconduct included abuse of process, conspiracy to injure and intentional infliction of harm. Wilson J. (who heard the initial application to have the plaintiff’s claims dismissed) noted that “abuse of process” is a generic term that encompasses the torts of conspiracy, intentional breach of duty to disclose information, malicious breach of duty and abuse of statutory power.

[81] In the last few years, both before and after Folland, claims have significantly expanded to include misfeasance in public office, breach of fiduciary duty, conspiracy and interference with economic relations. The growing list of torts can convey an impression that Crown jeopardy is growing rather than shrinking; this is likely to further increase the uncertainty Crowns may perceive about the parameters of claims that are brought against them now and in the future


Non-legislative response: “Case Management” as a tool to mitigate stress of Crowns named in malicious prosecution suit

[82] Ministries of the Attorney General across Canada might reduce the collective anxiety of Crowns by developing an intensive case management and client service protocol that would apply to all suits against Crowns. The protocol could be developed in consultation with Crowns so that it would be sufficiently responsive to their ongoing need for information and communication, and most importantly, for a timely resolution of the matter. In most cases, there is little strategic reason for cases to languish for any lengthy period of time. Therefore, a protocol might include aggressive movement of cases that include prompt motions for summary judgment or motions to strike as showing no reasonable cause of action, where appropriate.

[83] Further, this strategy might include ongoing outreach efforts to ensure that Crowns feel supported and informed by their Ministry. For instance, time should be set aside at annual Crown conferences for an update about the current lay of the land with respect to evolving legal issues and trends in this area.

[84] In addition, a link ought to be established on the Ministry intranets to provide Crowns with information relating to the protocol, legal issues and trends with respect to the tort of malicious prosecution. In this way, Crowns may feel better served. Also, this strategy may send a clear message to potential plaintiffs in frivolous claims about the consistently forceful way in which the Ministry intends to respond to such claims, and their heavy exposure to costs.

[85] However, better service cannot address the increased exposure Crowns have in the wake of Ferri, Miazga, Dix, Klassen, Coulter, and Folland.

Legislative response:

i. Require Claimants to Name the Attorney General as Defendant rather than the Individual Crown

[86] Making an Attorney General solely liable for the torts of Crown attorneys would provide protection to individual Crown attorneys from frivolous suits: since the Nelles decision, the number of malicious prosecution actions in Ontario has dramatically risen, without a single finding of liability. There is often tremendous pressure on those named to retain private legal representation because of the seriousness of the charge and concerns that at some point the interests of the individual Crown and the interests of the Ministry may diverge. Even in the absence of a finding of liability, the mere accusation of malicious prosecution can be demoralizing to both the individual named and to Crown attorneys in general, and be perceived as an impediment to a judicial appointment. Making the Attorney General solely liable would improve morale amongst Crown attorneys and reduce the exposure they feel when personally named as a defendant. In the hierarchy of the Crown attorney system, superiors may often authorize the prosecution of a particular case that will actually be prosecuted by a junior Crown, even if that junior is ambivalent. When the junior Crown is later named as a defendant in a malicious prosecution action, making the Attorney General solely liable would ease the tension this creates. Similarly, when multiple Crown attorneys have responsible for a file at different points, it is natural that differing viewpoints can arise. Sole liability to the Attorney General would eliminate any real or perceived conflict between the Crowns.

[87] There are, of course, associated drawbacks to a statutory reform making the Attorney General solely liable for the torts of individual Crown attorneys. By removing individual responsibility from Crown attorneys, it may be easier for a court to make a finding of liability against a ‘corporate’ defendant. It would also provide individual Crown attorneys with immunity from malicious prosecution suits, thereby implicating some of the concerns raised by the Court in Nelles, which concluded that such immunity would be contrary to public policy. However, it is arguable that the accountability mechanisms provided through Law Societies as regulators of proper professional conduct, and through available forms of discipline provided through the employer/employee relationship, are sufficient to regulate the individual Crown even if the Attorney General is the only named plaintiff.

[88] Despite removing a Crown attorney’s name from the style of cause, the workload created by frivolous malicious prosecution suits would not be reduced in any significant manner. In any given suit, the Crown attorney whose actions are in question will still have to participate in defence of these suits and any finding of liability will still turn on the conduct of the individual Crown attorney. Creating sole liability in the Attorney General would also have the effect of removing Crown immunity from these actions, and there would need to be consequential amendments to provincial/territorial legislation relating to proceedings against the Crown, in order to put the Attorney General in the same position as Her Majesty the Queen. Additionally, special rules would have to be put in place regarding who may be examined for discovery and how affidavits of documents should be provided if the Attorney General is the named defendant.

[89] Again, making the Attorney General solely liable for malicious prosecution suits against his agents does not address the increased exposure that the Attorney General would then assume in place of individual Crowns in the wake of the jurisprudence discussed in this paper.

ii. Statutorily Entrenching the Nelles Test

[90] The collapsing of the third and fourth Nelles elements goes against the very principle behind Nelles: that there should be a high threshold for any malicious prosecution action. Where there is simply an absence of reasonable and probable cause but no malice, a Crown attorney may indeed have been negligent. However, professional negligence is not an actionable tort against a Crown attorney. While evidence of the absence of reasonable and probable cause to prosecute can be consistent with an improper motive, it cannot alone also be the only indicator of malice, except possibly in extreme cases where there is simply no evidence to justify the prosecution. In general though, the collapse of the third and fourth factors essentially makes Crown attorney negligence an actionable tort. This lowering of the bar threatens to open the floodgates to frivolous litigation, and would implicate the public policy rationale stated in Nelles.

[91] Entrenching Nelles would give greater clarity to the requirements of each factor. For example, to give proper effect to the fourth Nelles factor (malice or improper purpose), a plaintiff should be required to first identify the improper purpose the Crown attorney is alleged to have had, and second to point to specific evidence beyond absence of cause that justifies drawing an inference of malice.

[92] Improper purpose within the meaning of the fourth factor should be understood as one that is personal to the Crown attorney accused of malicious prosecution. For example, a Crown attorney may have been improperly motivated to prosecute out of anger or vengeance, or by a desire for career advancement. If such a purpose is in fact motivating Crown attorney, then the tort of malicious prosecution is appropriate, and can restore public confidence in the administration of justice. At the same time, since such instances would be rare, the bar to an action would remain high, consistent with the public policy rationale laid out in Nelles.

[93] With regard to the third Nelles factor of “reasonable and probable cause” to prosecute, some thought should be given to eliminating the gap that currently exists between the standard a court uses to review whether there was cause to prosecute, and the standard a Crown uses when deciding to prosecute in the first place. The courts currently review decisions to prosecute on a mixed subjective/objective standard, while Crowns are instructed to apply strictly objective standards to their decisions to prosecute (generally in the form of whether there is a reasonable prospect of conviction, or words to that effect). Legislative clarification would most likely have to take the form of employing an objective standard only in reviewing decisions to prosecute, since pushing Crowns towards requiring a subjective belief in the guilt of the accused before deciding to prosecute would be in conflict with their quasi-judicial role as impartial “Ministers of Justice”.

[94] This legislative option would directly respond to the conflating of the third and fourth elements of the Nelles test. Once legislation is passed entrenching Nelles with clear guidance on the malice issue (that is, requiring specific evidence of malice), any motions to strike for lack of cause or for summary judgment would be addressed in that context.

[95] Consideration should also be given to modifying the Nelles criteria in light of recent jurisprudence, including Folland, and the discussion in this paper. In particular, it might be wise to clarify precisely which torts are available as against Crown attorneys (or to whomever liability is attached) for actions performed in the scope of their professional duties (and, what precisely that scope is). Naturally, the Working Group is in favour of limiting torts against Crown attorneys to malicious prosecution only, in the belief that it covers the improper use of a Crown attorney’s office.

iii. The “Drop-Dead” Option

[96] Under R.244.1(1) of the Alberta Rules of Court, A.Reg. 390/68, where five[68] or more years have expired from the time that the last thing was done in an action that materially advances the action, the Court shall, on the motion of a party to the action, dismiss that portion or part of the action that relates to the party bringing the motion. This rule prevents cases that have been constructively abandoned by the plaintiff from clogging the judicial system.

[97] In the context of a malicious prosecution action, the advantages of such a rule would be two-fold. First, it would be a powerful tool to assist in attempts to clear the backlog of frivolous malicious prosecution actions that are going nowhere. The implementation of such a rule in Ontario and other provinces would not impact any legitimate malicious prosecution actions in any other way than to encourage plaintiffs to make their cases. Only frivolous actions that plaintiffs declined to pursue would be cleared. Given that the vast majority of malicious prosecution actions in Ontario do not even make it to trial, it is likely that a substantial number of older cases could be quickly and efficiently cleared in this manner.

[98] Second, the speedy resolution of long-standing claims would remove the burden that Crown attorneys currently feel when a malicious prosecution action is “hanging over their head”. Even if the action is widely viewed as frivolous, simply being named in the action can create personal and professional stress, stress that is compounded the longer the action is remains unaddressed. This stress is not limited to the specific Crown involved, but can have wider negative impacts on the morale of Crown attorneys in general, all of whom aware that they too could be saddled with the burden of a frivolous action with little hope of immediate resolution.

[99] Ontario has, in some regions, a provision for a status notice, whereby the court sends a notice to plaintiffs when an action has not been set down for trial within two years. If the plaintiff fails to take steps to list the action for trial or to request a status hearing within 60 days of the date of the notice, the registrar may dismiss the claim. This is a better solution in our view because it does not put the onus on the Crown to bring a motion and to appear in court to have the action dismissed. Many actions used to be dismissed through this procedure that, for some reason, seems to have fallen out of use.

[100] Another option would be to require leave of the court before an action can be brought against a Crown attorney or the Attorney General, similar to the procedure in s. 140 of the Courts of Justice Act. This would place the onus on the plaintiff to satisfy the court that there is a basis for the action (that they can meet the Nelles requirements) It would be very helpful in screening out frivolous claims, especially those brought by unrepresented plaintiffs, and would save resources that would otherwise be devoted to preparing and conducting motions to strike and defending against frivolous actions.

[101] These rule changes alone, however, would not reduce the exposure created by jurisprudence as described. They are best thought of as options to use in combination with entrenching Nelles, as proposed above.


[102] Despite the policy rationale stated in Nelles regarding the balance between preventing absolute immunity for Crown attorneys in malicious prosecution actions while ensuring a healthy respect for Crown discretion in prosecution decisions, the subsequent jurisprudence has eaten away at the safeguards the Supreme Court created. Although the focus of this paper has been on recent interpretation of the Nelles test, the Working Group has identified other issues that warrant further consideration. For example, do public policy considerations support suggestions in jurisprudence that prosecutorial liability can or should be founded on torts other than malicious prosecution? Is there a need for uniform rules of court that effectively and fairly screen out frivolous lawsuits against prosecutors? Is there a need to develop uniform legislation restricting the ability of plaintiffs to sue prosecutors in their personal capacity for professional decisions made as agents of the Attorney General?

[103] Accordingly, the Working Group recommends that the Uniform Law Conference of Canada receive this paper and consider:

the preparation of a uniform law entrenching the Nelles criteria as the exclusive basis on which Crown prosecutors may be sued for malicious prosecutorial acts;

the preparation of a uniform law making Attorneys General solely liable for the torts committed by prosecutors as agents of the Attorneys General; and

the preparation of other uniform jurisdictional responses that would fairly and effectively limit the harm caused by frivolous malicious prosecution lawsuits.

Next Annual Meeting

2020 Annual Meeting

Place to be Announced

August 9 – 13, 2020