Nelles v. Ontario

The Ontario Court of Appeal in the case at bar found that an absolute immunity exists, and in reaching this conclusion relied extensively on the decision by the Supreme Court of the United States in Imbler v. Pachtman, 424 U.S. 409 (1976). The Court of Appeal found the idea of an absolute immunity "troubling" but determined that it was justified by the following policy concerns. First, the rule encourages public trust in the fairness and impartiality of those who act and exercise discretion in the bringing and conducting of criminal prosecution; the rule is designed for the benefit of the public not the benefit of the individual prosecutor. Second, the threat of personal liability for tortious conduct would have a chilling effect on the prosecutor's exercise of discretion and third, to permit civil suits against prosecutors would invite a flood of litigation which would deflect a prosecutor's energies from the discharge of his public duties. In short, the absence of an absolute immunity would open the door to unmeritorious claims and would be a threat to prosecutorial independence. The Court also relied on two decisions of the Ontario High Court, Owsley v. The Queen in right of Ontario (1983), 34 C.P.C. 96 and Richman v. McMurtry (1983), 41 O.R. (2d) 559. Both these decisions rely extensively on the American position as found in Imbler, supra. The case law in Ontario therefore, uniformly stands for the proposition that the Attorney General and Crown Attorneys enjoy absolute immunity from civil liability for malicious prosecution. Outside of Ontario, the issue is somewhat more ambiguous.

In Levesque v. Picard (1985), 66 N.B.R. (2d) 87, the New Brunswick Court of Appeal held on the authority of the Ontario cases, especially the case at bar, that an absolute immunity shielded a provincial Crown prosecutor from suit for malicious prosecution. By contrast the appellate courts of Nova Scotia and Alberta have cast some doubts on the existence of an absolute immunity. First, in Curry v. Dargie (1984), 28 C.C.L.T. 93 (N.S.C.A.), the Crown was sued as being vicariously liable for the action of a residential tenancy officer. Hart J.A. held that while the Proceedings Against the Crown Act, R.S.N.S. 1967, c. 239, might absolve the provincial Crown from civil liability, a Crown servant could still be personally liable for misconduct. In the course of his decision Hart J.A. considered the Ontario decisions especially that of Galligan J. in Richman, supra (at p. 110):

I am not prepared to go as far as Galligan J. in holding that an officer of the Crown cannot be liable for a proceeding commenced maliciously, but it is not necessary to consider that issue at the moment. I do not believe that in the case at bar it can be said that the respondent in laying the information against the appellant was in fact carrying out a judicial function similar to those carried out by Attorneys General and prosecutors.

In German v. Major (1985), 39 Alta. L.R. (2d) 270, a Crown prosecutor was sued for alleged misconduct in the preferment of a charge of tax evasion, a charge on which the accused was acquitted. Kerans J.A. speaking for the Alberta Court of Appeal assumes throughout that a suit for malicious prosecution is possible and disposes of the case on the ground that there had been "reasonable and probable cause" to initiate the prosecution. The case was dismissed pursuant to Rule 129 of the Alberta Rules of Civil Procedure, a rule similar to the old Ontario Rule 126. In this context Kerans J.A. said the following (at p. 276):

The rule upon which I rely has much to commend it. It falls short of the absolute immunity suggested by Major and accepted by the Supreme Court of the United States in Imbler v. Pachtman . . . but offers some protection from the harassment which he says would otherwise afflict prosecuting counsel because suit would not be permitted to proceed if utterly without merit. It would indeed be a curious thing if we chose a stern immunity rule in preference to an effective striking-out rule.

Further support for the view that Kerans J.A. is not inclined to accept the existence of an absolute immunity for prosecutors can be found in the following statements (at pp. 277 and 286):

I will assume, for the sake of argument, that, if counsel, with malice, continues a prosecution he once thought sound but now knows is unsound, he may be sued.

. . .

Counsel for the Attorney General who acts as his agent in the prosecution of a criminal case is not accountable in civil proceedings to the accused except possibly to the extent that it is alleged against him that he has not acted in good faith, and to that extent the allegation falls within the nominative tort of malicious prosecution . . . [Emphasis added.]

Therefore the Canadian position ranges from a strong assertion of absolute immunity in Ontario to an acceptance of the possibility of suing the Attorney General and Crown Attorneys if bad faith or malice can be proven as evidenced by the cases from Nova Scotia and Alberta. The situation in Quebec differs in that since 1966 the Code of Civil Procedure, R.S.Q., c. C-25, specifically provides for claims against the Crown in the following terms:

94. Any person having a claim to exercise against the Crown, whether it be a revendication of moveable or immoveable property, or a claim for the payment of moneys on an alleged contract, or for damages, or otherwise, may exercise it in the same manner as if it were a claim against a person of full age and capacity, subject only to the provisions of this chapter.

No provisions in this chapter prevent a suit for malicious prosecution against the Crown. However, the substantive issue of immunity of Crown prosecutors has not been finally determined.

A consideration of the position in respect of prosecutorial immunity in the United States is vital both because it is relied extensively upon by the Court of Appeal in the case at bar, and because it has been the source of a healthy debate in courts and among academics in that country. This position is furthermore interesting since a variety of approaches have been proposed and many critical comments have been made.

i)The Functional Approach -- Imbler v. Pachtman: "The Powell Judgment"

In 1972 Paul Imbler filed a claim under 42 U.S.C. {SS} 1983 alleging that the prosecutor and various members of the police force conspired to cause him loss of liberty by allowing a witness to give false testimony, suppressing evidence, prosecuting with knowledge of an exculpatory lie-detector test and introducing an altered police artist's sketch. Section 1983 of the Civil Rights Act creates a federal damage action against anyone who acts under colour of state law to deprive a person of his civil rights as protected by the U.S. Constitution. Powell J., speaking for five members of the Supreme Court, held that a prosecutor is absolutely immune from s. 1983 actions when the actions arise out of the prosecutor's initiation of prosecution and presentation of the State's case. In addition, the Court seemed to suggest that absolute immunity also attached to activities that "were intimately associated with the judicial phase of the criminal process" (p. 430). The Court then adopted what has become known as the "functional approach" of prosecutorial immunity.

The Imbler decision recognizes that prosecutors perform many functions in the course of fulfilling their duties, among them being the decision to initiate a prosecution, which witnesses to call, what other evidence to present, and obtaining, reviewing and evaluating evidence. The Court accepts that drawing a line between these functions is a difficult task but concludes that prosecutorial functions of a quasi-judicial or advocatory nature should be afforded absolute immunity. The Court refused to comment on whether a similar immunity attaches to what it called the "administrative" or "investigative" role of the prosecutor. In the course of justifying its position, the Court noted that the same policy considerations that afford absolute immunity to judges acting within the scope of their duties support a prosecutor's common law absolute immunity. The Court simply extended that line of reasoning to s. 1983 claims.

The policy considerations canvassed by the Court are familiar ones and can be summarized as follows:

1. Public Confidence

"The public trust of the prosecutor's office would suffer if he were constrained in making every decision by the consequences in terms of his own potential liability in a suit for damages."

2. Diversion from Duties

" . . . if the prosecutor could be made to answer in court each time such a person charged him with wrongdoing, his energy and attention would be diverted from the pressing duty of enforcing the criminal law."

3. Balancing of Evils

" . . . we find ourselves in agreement with Judge Learned Hand, who wrote of the prosecutor's immunity from actions for malicious prosecution:

" . . . it has been thought in the end better to leave unredressed the wrongs done by dishonest officers than to subject those who try to do their duty to the constant dread of retaliation." Gregoire v. Biddle, 177 F. (2d) 579, 581 (CA2 1949) cert. denied, 339 U.S. 949 (1950)."

"Even judges . . . could be punished criminally for willful deprivations of constitutional rights. . . The prosecutor would fare no better for his willful acts. . . Moreover, a prosecutor stands perhaps unique, among officials whose acts could deprive persons of constitutional rights, in his amenability to professional discipline by an association of his peers."

(Imbler, supra, at pp. 424-29)

Therefore, Powell J. affirmed the judgment of the Court of Appeal for the Ninth Circuit and held that a prosecutor is absolutely immune from suit in initiating a prosecution and in presenting the State's case.

ii)The Functional Approach -- Imbler v. Pachtman: "The White Judgment"

While concurring with the judgment of Powell J. and much of his reasoning, White J. (Brennan and Marshall JJ. joining) would carve out an exception to the rule of absolute immunity for the unconstitutional suppression of evidence. In doing so White J. examined the rationale for granting absolute immunity to prosecutors at common law (at p. 442):

The absolute immunity . . . is designed to encourage [the prosecutors] to bring information to the court which will resolve the criminal case. . . . Lest they withhold valuable but questionable evidence or refrain from making valuable but questionable arguments, prosecutors are protected from liability for submitting before the court information later determined to have been false to their knowledge.

According to White J. immunity from suit based on the unconstitutional suppression of evidence would "stand this immunity rule on its head" (p. 442) by discouraging precisely the disclosure of evidence sought to be encouraged by the rule (at p. 443):

A prosecutor seeking to protect himself from liability for failure to disclose evidence may be induced to disclose more than is required. But this will hardly injure the judicial process. Indeed, it will help it. Accordingly, lower courts have held that unconstitutional suppression of exculpatory evidence is beyond the scope of "duties constituting an integral part of the judicial process" and have refused to extend absolute immunity to suits based on such claims. Hilliard v. Williams, 465 F. 2d 1212, 1218 (CA6), cert. denied, 409 U.S. 1029 (1972) . . . .

White J.'s position then would limit the scope of absolute immunity but would not eliminate the theoretical underpinning of the Powell majority judgment, namely the functional approach to absolute immunity.

The functional approach has been criticized on a number of grounds. First, there is the ever present problem of line-drawing between functions that are quasi-judicial and those that are administrative or investigative. Drawing the line is made more difficult by multi-faceted functions, functions that simultaneously serve quasi-judicial, administrative and investigative functions. (See Anthony Luppino, "Supplementing the Functional Test of Prosecutorial Immunity" (1982), 34 Stan. L. Rev. 487, at pp. 493-94.) Aside from the problem of distinguishing between prosecutorial functions, there is the conceptual difficulty in justifying differential treatment of malicious acts based on the criterion of function. If a prosecutor acts maliciously in the course of the prosecution of an accused, does it really matter whether the function being carried out is characterized as "quasi-judicial" or "administrative"?

An example of the difficulty with the functional approach is the disagreement in the lower courts in the United States over whether quasi-judicial absolute immunity extends to investigative functions of a prosecutor. In addition, and in light of the White concurring judgment in Imbler, there is disagreement over whether leaks of information and destruction or alteration of evidence are acts that are protected by absolute immunity: see cases cited by J. C. Filosa, "Prosecutorial Immunity: No Place for Absolutes," [1983] U. Ill. L. Rev. 977, at pp. 985-86. In my view, these disagreements demonstrate the futility of attempting to differentiate between functions of a prosecutor in a principled way. The result is often arbitrary line-drawing which leads to seemingly unresolvable conflict and the diversion of attention from the central issue, namely whether or not a prosecutor has acted maliciously.

Second, it has been argued that the policy rationales supporting absolute immunity for prosecutors, derived as they are from judicial immunity, rely on an inaccurate reading of history. Filosa in his article challenges the derivation of the prosecutor's quasi-judicial immunity from s. 1983 claims from the absolute immunity of judges at common law (at pp. 980-81):

In the sixteenth century, English judges were typically liable for their torts. Throughout the nineteenth century, judges remained liable for malicious conduct done without reasonable and probable cause. In America before Bradley v. Fisher [80 U.S. (13 Wall.) 335 (1872)], courts held many judicial officers liable for their wrongful acts. . . . Of the thirty-seven states in existence in 1871, thirteen had judicial immunity, six states held judges liable for malicious actions, nine had not taken a clear position, and nine had not faced the question.

Filosa goes on to argue that Congress could not have meant to incorporate a doctrine of absolute immunity into s. 1983 because Bradley, which firmly entrenched judicial immunity in the common law, was not decided until 1872, one year after the Civil Rights Act of 1871 that contained s. 1983.

4. Alternatives to Imbler i) The Functional Approach Reapproached

The difficulties in applying the functional test have led American courts and academic commentators to suggest alternatives or reassessments of the test. One such attempt has been described by its proponent as the "functional approach reapproached". (See Note, "Delimiting the Scope of Prosecutorial Immunity from Section 1983 Damage Suits" (1977), 52 N.Y.U. L. Rev. 173, at pp. 190-91.) This approach seeks to avoid a judicial hearing to determine whether a prosecutor's action is quasi-judicial. As such the test states that "the only duties clearly not entitled to quasi-judicial immunity are those so divorced from the judicial process that they could readily be assigned to another official who could be completely independent of the prosecutor" (see Note, loc. cit., at p. 191). This approach seeks to grant to the prosecutor absolute immunity in a wider sphere of activities in the hopes of clarifying the distinction between quasi-judicial and investigative activities. In my view, this modification still has the drawback of requiring a line to be drawn between prosecutorial functions, a difficult task in itself. The modification, in seeking to make that task easier, errs on the side of including more activities within the realm of absolute prosecutorial immunity, a modification that, with respect, offers an immunity considerably wider than that given to judges from which prosecutorial immunity is allegedly derived.

ii) General Features Test: Wilkinson v. Ellis

In Wilkinson v. Ellis, 484 F. Supp. 1072 (E.D. Pa. 1980), the plaintiff alleged that a prosecutor destroyed a tape recorded interview with a man who admitted involvement in the alleged criminal activity, thereby exonerating the plaintiff. The prosecutor moved to dismiss the action, arguing that the destruction of evidence is a quasi-judicial act shielded by absolute immunity. The Wilkinson court refused to characterize the destruction as either investigative or quasi-judicial. Rather, it resolved the difficulty of classifying activities by asking whether the activity contained features "which generally characterize quasi-judicial activity" (p. 1083). In deciding that the destruction did not have the "general features" of quasi-judicial activity, the court identified three factors to be taken into account: (1) the activity's physical and temporal proximity to the judicial process; (2) the degree of dependence upon legal opinions and prosecutorial discretion involved in the conduct; and (3) whether the activity is primarily advocatory (p. 1080). This approach in my view, does little to get away from the inherent problems involved in categorizing prosecutorial actions.

iii) The Imbler "Umbrella"

This variation of the functional approach involves limiting the scope of the prosecutor's quasi-judicial function to conduct that falls within the narrowest confines of the Imbler test: in other words within the "umbrella" of coverage defined by the language of Imbler. Acts that are under the "umbrella" attract absolute immunity; all others receive at most qualified immunity. (See Marrero v. City of Hialeah, 625 F.2d 499 (5th Cir. 1980), cert. denied, 450 U.S. 913 (1981).) This approach merely re-states the categorization problem found in Imbler. The test requires a determination of what constitutes the coverage of the so-called "Imbler umbrella" and thereby takes us back to the original problem of line-drawing.

iv) The Harm Test

This variation of Imbler construes that decision broadly by granting absolute immunity to prosecutorial conduct that causes a defendant to "face prosecution, or to suffer imprisonment or pretrial detention". (See Taylor v. Kavanagh, 640 F.2d 450 (2d Cir. 1981), at p. 453.) The test denies absolute immunity to prosecutorial conduct that inflicts harm independent of the prosecution itself. This approach looks to the effects of prosecutorial conduct and as such purports to reduce the issue to a factual determination of harms. If the harm is unrelated to the judicial phase of the criminal justice process then the prosecutorial act causing the harm is not quasi-judicial.

v) The Supplemental Functional Approach

This approach involves a two-step process: first, determining what conduct normally merits absolute or qualified immunity and second, in the remaining cases, identifying the substantive values affected by conduct that is not susceptible to traditional categorization. (See Luppino, loc. cit., at p. 505.) This variation recognizes that there will be occasions when conduct does not clearly fall into one of the two traditional categories: quasi-judicial and non-quasi-judicial. When conduct does not fall into either category explicit balancing of competing interests becomes necessary. In this respect, courts should weigh the cost to the judicial system resulting from the unredressed civil wrong against the cost to the efficiency of the criminal justice system. This approach recognizes that the Imbler functional approach cannot account for all prosecutorial functions; there will be some conduct that is multi-faceted and uncategorizable. As a result the approach resorts to a consideration of first principles, namely a balancing of the policy considerations both in favour and opposed to prosecutorial immunity in the first place. In short, we have come full circle.

The American position, in any of its forms, demonstrates the impracticality of the functional approach to prosecutorial immunity. In my view, the functional approach leads to arbitrary line drawing between prosecutorial functions. This line drawing exercise is made nearly impossible by the reality that many prosecutorial functions are multi-faceted and cannot be neatly categorized. Further, it must be noted that however one categorizes a prosecutor's function it is still that of the prosecutor. If it can be demonstrated that a prosecutor has acted without reasonable cause and has acted with malice then does it really matter which functions he was carrying out? In my view to decide the scope of immunity on the basis of categorization of functions is an unprincipled approach that obscures the central issue, namely whether the prosecutor has acted maliciously. If immunity is to be qualified it should be done in a manner other than by the drawing of lines between quasi-judicial and other prosecutorial functions.

The position in respect of prosecutorial immunity in England is somewhat unique in that jurisdiction owing in part to the tradition of private prosecution. Private prosecutors have always been liable to suit for malicious prosecution though few, if any, reported cases exist. The Director of Public Prosecutions, who performs the same or similar function as a Canadian provincial Attorney General, was not created until 1879. In Riches v. Director of Public Prosecutions, [1973] 2 All E.R. 935 (C.A.), the Court said the following in respect of suits against the D.P.P. (at p. 941):

I do not wish to be taken as saying that there may never be a case where a prosecution has been initiated and pursued by the Director of Public Prosecutions in which it would be impossible for an acquitted defendant to succeed in an action for malicious prosecution, or as saying, that the existence of the Attorney General's fiat where required conclusively negates the existence of malice and conclusively proves that there was reasonable and probable cause for the prosecution. There may be cases where there has been, by even a responsible authority, the suppression of evidence which has led to a false view being taken by those who carried on a prosecution and by those who ultimately convicted.

The English position then, at the very least, leaves the door open for suits against the equivalent of our Attorneys General and Crown Attorneys when what is at issue is the suppression of evidence. It is apposite to note that this position is reflective of White J.'s concurring opinion in Imbler, supra, wherein he carved out an exception to the rule of absolute immunity for the unconstitutional suppression of evidence.

It would appear that in Scotland the equivalent of our Attorney General and Crown Attorneys are absolutely immune from civil liability. In Hester v. MacDonald, [1961] S.C. 370, the court said at p. 377:

It is, therefore, an essential element in the very structure of our criminal administration in Scotland that the Lord Advocate is protected by an absolute privilege in respect of matters in connexion with proceedings brought before a Scottish Criminal Court by way of indictment . . . . Never in our history has a Lord Advocate been sued for damages in connexion with such proceedings. On the contrary, our Courts have consistently affirmed the existence of such immunity on his part.

The rationale underlying this comment has been disputed by Professor Edwards in The Attorney General, Politics and the Public Interest (1984) in which he argues that the Scottish rationale is based upon the idea that the Lord Advocate and his agents enjoy a constitutional trust which assumes good faith in commencing a prosecution, a rationale far removed from that invoked by the Ontario courts.