IpsofactoJ.com: International Cases  Part 7 Case 4 [SCC]
SUPREME COURT OF CANADA
- vs -
Attorney General of Quebec
18 OCTOBER 2001
Iacobucci J & Binnie J
(delivered the majority judgment of the court)
In 1986, the Prosecutor determined that there were insufficient grounds to charge the appellant with the murder of his former girlfriend, France Alain. There was no doubt that a murder had occurred. It was the Prosecutor's opinion, however, that there was no reliable identification evidence against the appellant or anyone else. The prosecution file was closed.
Some five years later, Radio Station CHRC broadcast sensational allegations linking the appellant to the murder, and the appellant retaliated in January 1991 with an action for defamation claiming damages of a million dollars against André Arthur, a journalist of the radio station, and John Tardif, the then retired police investigator who had worked on the closed file. The appellant had previously worked as a news reader at CHRC.
At this point, more than eight years after the murder, a journalist at CHRC (the appellant claims it was Arthur himself) contacted Tardif about a new witness who allegedly could supply the necessary eyewitness identification. The witness's name was Paul-Henri Paquet. According to Paquet's sworn declaration, [translation] "all the commotion" around the case, including the news reports involving Tardif and Arthur, is what convinced him to come forward. Subsequently, declared Paquet, he saw the appellant's picture in a newspaper article about the appellant's defamation suit against Tardif and Arthur and was suddenly struck by the eyes of the appellant whom he allegedly recognized as the man he encountered on the night of the murder. Paquet declared [translation]:
So at 8 a.m. I decided to go immediately to radio station CHRC to leave a message for Mr. Arthur and/or Mr. Tardif that I would probably have something to add to the France Alain case. (I told an announcer named Mr. Thibeault that the night of the murder someone had said to me "Don't go see, it's not a nice thing to see.")
Paquet also swore that a few days later (evidently through the intervention of someone at CHRC), he met with Tardif [translation] "who had taken charge of the case".
Under our criminal justice system, prosecutors are vested with extensive discretion and decision-making authority to carry out their functions. Given the importance of this role to the administration of justice, courts should be very slow indeed to second-guess a Prosecutor's judgment calls when assessing Crown liability for prosecutorial misconduct. Nelles v Ontario,  2 S.C.R. 170, affirmed unequivocally the public interest in setting the threshold for such liability very high, so as to deter all but the most serious claims against the prosecuting authorities, and to ensure that Crown liability is engaged in only the most exceptional circumstances. Against these vital considerations is the principle that the Ministry of the Attorney General and its Prosecutors are not above the law and must be held accountable. Individuals caught up in the justice system must be protected from abuses of power. In part, this accountability is achieved through the availability of a civil action for malicious prosecution. As stated by Lamer J. (as he then was) in Nelles, at p. 195:
.... public confidence in the office of a public prosecutor suffers greatly when the person who is in a position of knowledge in respect of the constitutional and legal impact of his conduct is shielded from civil liability when he abuses the process through a malicious prosecution.
The allegations in this case address one of those exceptional circumstances where it has been established on a balance of probabilities that the prosecutorial office has been found to have been used deliberately for purposes that we believe were improper and inconsistent with the traditional prosecutorial function (Nelles, supra, at pp. 196-97). Civil liability for malicious prosecution is therefore an appropriate remedy.
With all due respect to L'Heureux-Dubé J.'s statement of the application of "public" common law principles to actions brought in Quebec against the Crown for malicious prosecution, we prefer the analysis of LeBel J.A. (as he then was) in the Quebec Court of Appeal ( R.J.Q. 398), which delineated the respective roles of public law and private law. We think, as well, that his approach is more consonant with the approach to this issue taken in common law jurisdictions. However, nothing turns on this doctrinal difference in the present appeal. Our disagreement with our colleague turns on the facts of this particular case.
It is important to note that when the Quebec Court of Appeal overturned the appellant's conviction on the basis that no properly instructed jury could reasonably have found him guilty beyond a reasonable doubt, it not only criticized the trial judge for palpable errors but also pointed out the speculative and hypothetical nature of the theory which constituted the basis for initiating proceedings against the appellant:  R.J.Q. 2047, 76 C.C.C. (3d) 316. In a concluding statement, the Court of Appeal observed that (at p. 383 C.C.C.) [translation]:
.... the jury's verdict rests in part on evidence that the judge should have excluded. We are also of the view that the jury ended up at its verdict on the basis of inadequate instructions, in particular on the visual identification evidence and the question of motive. Finally, we consider that the identification evidence is so lacking in probative value that it would be unreasonable, even taking into account the other evidence called by the Crown, to find beyond a reasonable doubt that Mr. Paquet saw the appellant near the scene of the crime on the evening of October 25, 1982.
Like the trial judge and LeBel J.A., dissenting in the result in the Quebec Court of Appeal, we find that the record in this appeal suffices to satisfy the Nelles test because it reveals that:
the Crown Prosecutor did not have reasonable and probable cause upon which to found the charges brought against the appellant, and
the prosecution was motivated by an improper purpose which in law constitutes malicious conduct and intentional fault.
Thus, for the reasons that follow, we are of the view that this is one of the exceptional cases in which Crown immunity for prosecutorial misconduct should be lifted. We would therefore allow this appeal.
The starting point in any criminal prosecution is the presumption of innocence. The prosecutor must assess, in good faith and without any motive but the furtherance of the administration of justice, whether the presumption of innocence can be rebutted in a court of law. This is a practical decision based on the prosecutor's experience and knowledge, and on his or her assessment of all the potentially relevant evidence. A failed prosecution does not without more - much more - give rise to a viable claim for prosecutorial wrongdoing.
As we have noted above, Nelles, set out four requirements that must be established on a balance of probabilities by the claimant in an action in damages based on prosecutorial misconduct in order to avoid the Crown's relative immunity against such suits. In this appeal, the first two requirements, namely that the respondent initiated the prosecution, and that the prosecution resulted in the appellant's acquittal, are clearly satisfied. Our disagreement with our colleague relates to the additional two requirements, namely that the respondent lacked reasonable and probable grounds to prosecute the appellant, and that the prosecution was tainted with an improper motive.
A. ABSENCE OF REASONABLE AND PROBABLE CAUSE
The existence of reasonable and probable grounds was defined for present purposes in Nelles at p. 193 as follows:
This 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.
The best evidence that the Prosecutor lacked reasonable and probable grounds prior to 1991 is the Prosecutor's own admission in the stay application that there was no case to justify the laying of charges at that time. As our colleague states at para. 163: "He was aware of his role, his responsibilities and the law, and he closed the file". The analysis, then, turns to the events of 1991, and to a consideration of what changed the Prosecutor's decision and resulted in the charge of first degree murder being laid against the appellant on March 20, 1991. Two pieces of evidence appear to have been critical to that reversal. They constitute the bulk of the case against the appellant.
Firstly, the eye-witness "identification" by Paquet.
Secondly, the Prosecutor's re-evaluation, in light of Paquet's evidence, of the conversation between the appellant and the victim's father which took place on May 30, 1983 and which was surreptitiously recorded by the police.
We will deal with these items separately.
(1) The Evidence of Identification
The critical evidence, from the Crown's perspective, was the alleged identification of the appellant by the witness, Paquet. However, the process through which this identification was carried out was extremely flawed and unusual. As mentioned, Paquet came forward through the intervention of the defendants in the defamation action more than eight years after the murder. He claimed to have seen the probable perpetrator of France Alain's murder just minutes after she was shot. Although Paquet allegedly reported the encounter to three other people within hours of the murder, it is not explained why he did not go to the police eight years earlier. Indeed, even in 1991, rather than going straight to the police, Paquet went to the CHRC radio station, having been made aware of the well-publicized defamation action brought by the appellant. Paquet made this visit after seeing a photograph of the appellant featured in a newspaper article on February 9, 1991, about the appellant's initial allegation of defamation. (The appellant began a second action in defamation after his conviction was reversed by the Quebec Court of Appeal on August 20, 1992, but this second action came too late to have any relevance to the present proceedings.) The appellant says it was Arthur himself who initially sent Paquet not to the police but to speak to Tardif.
We wish to make it clear that we impugn neither the motives nor the trustworthiness of Paquet. The issue, simply, is the very limited nature of what he could honestly say about his encounter with a "bearded man" on the night in question. Such an encounter may have happened, but what if anything did it have to do with the appellant?
Paquet told Tardif that he had identified the appellant's eyes from the recently published newspaper photo, eight years after he supposedly saw him during a momentary encounter at night. Tardif then showed Paquet a photo of the appellant, but covered all but the appellant's eyes. Paquet affirmed that these were the eyes of his "bearded man", whom he had seen on the night of the murder. However, when the full face of the appellant was revealed, Paquet candidly stated that this was not the man he had seen. This was Paquet's evidence on this point [translation]:
Yes, at some point we were at the scene of the incident that I described this morning, earlier, and then at some point, he [Tardif] had me get into his automobile. Then he said to me: "I'll show you a photograph of Benoît Proulx." Then he said: "You tell me whether or not that was who you saw."
Then I immediately told him, I said: "Do you want to help me out, we can do it this way." I then said: "Begin by showing me the eyes, cover everything, and then begin by showing me the eyes and then afterwards you'll pull out your photograph."
He had a white envelope. It was quite a large photograph. Then he put -- he arranged things so as to show me just the eyes, and I then said: "It's my guy's eyes, again, there, it's my bearded guy's eyes."
So, when he showed me the photograph, he had a new permanent, I can say and it was black, a big beard and it's completely different; they weren't the same glasses. I said: "That's fine but it's not my guy."
He then had a look like he wanted to say: "Well, listen, it's him."
Well, I said, "No, it's not the guy I saw", after I had recognized only the eyes from the photograph and I didn't want to agree, accept it ....
Tardif did not disclose the existence of Paquet's evidence to the police for several weeks thereafter. The Prosecutor was then made aware of Paquet's existence and of Tardif's attempt to implicate the appellant.
Once Paquet's existence had been disclosed to the authorities, the Crown Prosecutor met with Paquet in the presence of Tardif, and Paquet was shown a series of eight more photographs of the appellant taken during a labour dispute, in which the appellant was protesting in front of his radio station. There was no proper reason for Tardif, with his own separate agenda, to be allowed to be present at this meeting. There was even less reason for the City of Ste-Foy then to hire Tardif as part of the criminal investigation team despite his obvious conflict of interest.
No photographs of similar-looking people were included in the eight photographs. If Paquet were to make an identification at all, it had to be of the appellant. On this point we agree with the observation made in R v Dwyer (1924), 18 Cr. App. R. 145, at pp. 147-48, referred to by our colleague at para. 191, but with respect, it seems to refute rather than support her conclusion. In that case, Lord Hewart C.J. wrote, at p. 148:
One distinction, however, is quite clear. It is one thing for a police officer, who is in doubt upon the question who shall be arrested, to show a photograph to another person in order to obtain information or a clue upon that matter; it is another thing for a police officer dealing with witnesses who are afterwards to be called as identifying witnesses to show to those persons photographs of those whom they are about to be asked to identify beforehand.
Our colleague suggests that Dwyer does not impugn the Crown Prosecutor's conduct in this instance, but rather supports her position because "[t]he identification of the appellant was merely one block in the structure comprised of the circumstantial evidence on which the charge was based" (para. 192). However, with respect, this understates its importance. The Prosecutor needed to know whether Paquet could provide credible evidence of identity to overcome the presumption of the appellant's innocence. It was the Prosecutor's own view that without Paquet he had no case. It was therefore critical for him to determine whether or not Paquet could supply the missing evidentiary link in court. The Prosecutor may have been persuaded of the appellant's guilt. The question for him in March 1991 was whether he could prove it.
Furthermore, in light of the publicity surrounding the defamation action, Paquet knew perfectly well the importance attached by Tardif to identifying the appellant as the "bearded man". In this case, unlike the situation contemplated by Lord Hewart C.J., the Prosecutor himself was present at this improper identification procedure.
Paquet then identified the appellant in one of the eight photos, stating that this was his "bearded man", whom he had seen on the night of the murder. However, as pointed out by the judges of the Court of Appeal in the criminal proceedings, the appellant's eyes, which Paquet claimed were critical to his identification, were not visible in this particular photograph.
The Prosecutor not only admitted at trial that a positive identification of the appellant was never made but also that this was not possible because the eyes and beard were the only traits of the appellant that Paquet was ever able to identify. The Prosecutor was familiar with the fundamental rules of eye witness identification. He also knew he had to prove identification beyond a reasonable doubt in order to obtain a proper conviction. His willingness to pursue the appellant with a charge of first degree murder in these circumstances is central to the determination of whether this is a case of mere recklessness or malice.
The deficiencies in Paquet's evidence must have been obvious to the Prosecutor from the outset, and indeed his view of the matter is shown by his treatment of Paquet at trial, where he deliberately chose not to ask Paquet to identify the appellant in the courtroom. He admitted that this was because Paquet could not do so, seeing as he recognized only the eyes and possibly only the shape of the eyes of a "bearded man". An obvious question arises in these circumstances: if a legitimate, unequivocal or even adequate identification of the appellant had already been made by Paquet, is it conceivable that he would be entirely incapable of identifying the appellant in person? The appellant may have discarded his beard and put on some weight in the intervening years but Paquet's identification allegedly centred on "the eyes", which were the same. In our view, such a flagrantly inadequate eyewitness identification could not have served as a legitimate basis for proceeding to trial.
Nor, in our view, do the cases cited by our colleague lend support to what was done here. In R v Swanston (1982), 65 C.C.C. (2d) 453 (B.C.C.A.), the witness had positively identified the accused as his assailant on previous occasions (at a line-up and at a preliminary hearing). In the case at bar, there was no such prior identification. Paquet said only that he encountered a man with a beard and distinctive eyes at night around the time of the murder and near the scene of the crime, and came forward more than eight years later.
In Mezzo v The Queen,  1 S.C.R. 802, as well, there was strong identification evidence before the irregularities took place during subsequent identification sessions. It is important to recall that Wilson J. said in Mezzo that the improprieties in the line-up do not necessarily destroy otherwise good evidence. The operative words here are "otherwise good evidence". Similarly, in Marcoux v The Queen,  1 S.C.R. 763, which is discussed by Wilson J. in Mezzo, this Court found that the trial judge properly put the identification evidence to the jury in spite of the problems with its quality because the complainant actually identified the accused on two occasions at the police station. In the case at bar, not only were there irregularities that were so "improper as to be beyond the power of the jury to handle given a proper direction by the trial judge" (to borrow the words of Wilson J. at p. 833) but no identification of the appellant by the only identification witness, Paquet, had ever taken place.
Even the majority judges in the Quebec Court of Appeal (Beauregard and Brossard JJ.A.) strongly disapproved of the manner in which the investigation was conducted. In fact, in his judgment, Brossard J.A. condemned the Prosecutor for:
lending any credibility at all to the evidence of identification provided by Paquet, and
failing to take into account the probable inadmissibility of the recorded conversation in light of existing case law.
Brossard J.A. added that on both of these points, the Prosecutor was wrong and negligent in carrying out his professional mandate.
(2) The Recorded Conversation
The Prosecutor says that the emergence of Paquet's evidence caused him to re-examine the file, and in particular the surreptitiously recorded conversation between the appellant and the victim's father. There were two problems with reliance on the recorded conversation.
Firstly, the Prosecutor must have known it was not properly admissible.
Secondly, even if admissible, it proved nothing.
In her analysis, L'Heureux-Dubé J. takes issue with the Court of Appeal's ruling in the criminal trial that the recordings should have been declared inadmissible under s. 24(2) of the Canadian Charter of Rights and Freedoms. She emphasizes that, at the time this conversation was recorded in 1983, the law had yet to develop to the point that the Prosecutor would have known that the tapes would be inadmissible. Further, by the time of the criminal proceedings in 1991, a trial judge would still have the discretion, pursuant to s. 24(2) of the Charter, to determine whether the admission of this evidence would bring the administration of justice into disrepute and thus, to find it admissible proof.
We respectfully disagree with this reasoning. In R v Duarte,  1 S.C.R. 30, this Court clearly held that the surreptitious electronic surveillance of an individual by an agency of the state constitutes an unreasonable search or seizure under s. 8 of the Charter. On this authority, it is clear that, by the time criminal charges were brought in 1991, the Crown Prosecutor would have known, or at least should be taken to have known, that unauthorized police recordings of the appellant's communications made without his knowledge or consent were not proper evidence.
Further, even if the tapes were admissible, the Crown would have known that they had no probative value. At best, they contained the appellant's speculations in regard to the crime, which were made in response to continued prompting by the victim's father. In addition, within this conversation the appellant denounced the killing and said he himself was incapable of such an act. He thus explicitly denied any complicity in the crime. Given these considerations, it is clear that the tapes, in and of themselves, could not serve to establish reasonable and probable cause for the prosecution initiated against the appellant. The way in which this recording was manipulated by the Prosecutor before the jury will be discussed below in connection with malice or improper purpose.
(3) Other Factors
Several other factors further reveal the absence of reasonable and probable cause for the appellant's prosecution. To begin, the prosecution did not advance any proof of motive. As LeBel J.A. noted in his reasons at the Court of Appeal, the Crown had no evidence to support its theory that the appellant was enraged after his relationship with the victim ended (p. 428). Further, there was no evidence that the appellant left work at any time on the night of the murder, or proof of the allegation that he knew facts and details about the murder before other media did. Finally, the police investigation never produced any weapon or evidence to suggest that the appellant owned a firearm.
To say that a Prosecutor must be convinced beyond a reasonable doubt of an accused person's guilt before bringing charges is obviously incorrect. That is the ultimate question for the trier of fact, and not the Prosecutor, to decide. However, in our opinion, the Crown must have sufficient evidence to believe that guilt could properly be proved beyond a reasonable doubt before reasonable and probable cause exists, and criminal proceedings can be initiated. A lower threshold for initiating prosecutions would be incompatible with the Prosecutor's role as a public officer charged with ensuring justice is respected and pursued. See Boucher v The Queen,  S.C.R. 16; R v Bain,  1 S.C.R. 91; Nelles, supra; R v Charest (1990), 28 Q.A.C. 258; and R v S. (F.) (2000), 47 O.R. (3d) 349 (C.A.). We think it must have been clear to the Prosecutor on March 20, 1991, when he authorized the charge of first degree murder against the appellant, that this evidence could not properly have resulted in a conviction.
Our colleague relies greatly on the facts that the preliminary inquiry judge concluded there was sufficient evidence to send the appellant to trial; that the trial judge did not direct a verdict of acquittal; and that the jury convicted. However, these events post-dated the Prosecutor's decision, and were in each instance decisions governed by different considerations. More importantly, the trial was, as found by the unanimous Court of Appeal in the criminal case, deeply flawed. In our opinion, the Prosecutor cannot bootstrap his own position on the basis of flawed court decisions that were swept away by the acquittal directed by the Court of Appeal.
Nor can the Prosecutor rely on consultations that he had with colleagues and superiors. He knew more about the case than they did and, as the holder of an important office under the Criminal Code, R.S.C. 1985, c. C-46, the decision to lay the charge was his and his alone: R v Campbell,  1 S.C.R. 565, at para. 33.
In our view, the charges brought against the appellant were based on fragments of tenuous, unreliable and likely inadmissible evidence. They were accurately characterized by Rioux J. at trial and LeBel J.A., dissenting, in the Court of Appeal, as being grounded in mere suspicion and hypotheses. As such, it could not serve to prove the appellant's guilt beyond a reasonable doubt. This being the case, we are of the view that the proceedings at issue were not based on reasonable and probable cause.
B. MALICE OR IMPROPER PURPOSE
In addition to an absence of reasonable and probable cause, a suit brought pursuant to an allegedly abusive prosecution may succeed only where malice or an improper purpose is shown. This criterion was discussed by Lamer J. in Nelles, supra. Writing for a majority of this Court, Lamer J. noted that cases of malicious prosecution involve serious allegations, which relate to the misuse and abuse of the criminal process and the office of the Crown Attorney. He stated (at pp. 193-94 and 196-97):
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".
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.
As such, a suit for malicious prosecution must be based on more than recklessness or gross negligence. Rather, it requires evidence that reveals a willful and intentional effort on the Crown's part to abuse or distort its proper role within the criminal justice system. In the civil law of Quebec, this is captured by the notion of "intentional fault". The key to a malicious prosecution is malice, but the concept of malice in this context includes prosecutorial conduct that is fueled by an "improper purpose" or, in the words of Lamer J. in Nelles, supra, a purpose "inconsistent with the status of 'minister of justice'.
The trial judge in this case held that the Prosecutor acted for an improper motive. It was the task of the trial judge, Rioux J., to find the facts, and in our view, after looking at all the circumstances, his findings are entitled to deference that the majority of the Court of Appeal was not prepared to give. In short, in our opinion, the Court of Appeal wrongly interfered with the findings of the trial judge.
In the case at bar, various significant factors stand out as indicators of an improper purpose underlying the Crown's decision to initiate proceedings against the appellant. In discussing these factors, we do not wish to emphasize the importance of one over another. In the final analysis, it is the totality of all the circumstances that are to be considered in cases of this kind.
At the outset, the absence of reasonable and probable cause, which has been discussed above, is particularly noteworthy. The Crown Prosecutor ignored the probable inadmissibility and lack of probative value of the recorded communications between the appellant and the victim's father. He also relied on the extremely tenuous identification evidence offered by Paquet as the primary basis for reopening the investigation and prosecuting the appellant. All of this occurred in the context of the well-publicized million dollar defamation action brought by the appellant. In these exceptional circumstances, in our opinion, no Prosecutor acting in good faith would have proceeded to trial on a first degree murder charge with such substandard and incomplete proof. The decision to recruit the retired policeman, Tardif, to assist in the resurrected prosecution file, notwithstanding Tardif's status as a defendant in the appellant's defamation suit, is further evidence of malice in the sense of the Prosecutor's apparent indifference to the improper mixing of public and private business.
The manner in which the Prosecutor dealt with the surreptitiously recorded conversation between the appellant and the victim's father further points to an improper purpose. We must assume that the Prosecutor knew the relevant law, and in particular that the surreptitiously recorded conversation was not properly admissible (Duarte, supra). His decision to proceed regardless, is evidence of malice, and his success in persuading the trial judge to erroneously admit the improper evidence is, if anything, an aggravating circumstance.
During the recorded conversation, in response to prodding by the victim's father, the appellant speculated on the type of person who may have committed the murder and, as mentioned earlier, expressly denounced the act in no uncertain terms. The Prosecutor, however, invited the jury to replace the word "he" in the appellant's conversation with "I", thereby creating the illusion that the appellant was speaking about the murder in the first person rather than in the third and thus confessing to be the killer, as follows (AR 955, p. 2) [translation]:
If we don't put his words into the third person, as he did so well, we can say the following, keeping in mind that these words came out of his own mouth during his conversation with Mr. Alain:
France made a mistake at one point that I understood to be out of spite ....
Now I'll replace "he" with "I".
Then something mean was done to me, that was premeditated, then it was something really insulting.
Because I did that, I mean, I'm a guy who, I don't know, I've always had that idea in my mind, and I don't know why I can't get rid of it.
As found by the Court of Appeal in the criminal case, this manipulative approach effectively distorted the appellant's words, and improperly transformed them into a full confession of guilt. The Crown's actions thus were more than careless. Rather, they represented an active effort to obtain a conviction at any price. This ran counter to the nature and spirit of the Crown Attorney's role, which was aptly described by Rand J. in Boucher, supra, at pp. 23-24, in a well-known passage:
It cannot be over-emphasized that the purpose of a criminal prosecution is not to obtain a conviction, it is to lay before a jury what the Crown considers to be credible evidence relevant to what is alleged to be a crime. Counsel have a duty to see that all available legal proof of the facts is presented: it should be done firmly and pressed to its legitimate strength but it must also be done fairly. The role of prosecutor excludes any notion of winning or losing; his function is a matter of public duty than which in civil life there can be none charged with greater personal responsibility. It is to be efficiently performed with an ingrained sense of the dignity, the seriousness and the justness of judicial proceedings.
The tangled relationship between the criminal proceedings initiated against the appellant, and the appellant's defamation suits against Tardif and André Arthur, also suggests that the prosecution was motivated by an improper purpose. The Prosecutor knew about the defamation suits, and that Tardif was retired by the time Paquet came forward. Nevertheless, he allowed Tardif to resume work on the case, even though he was in a conflict of interest and had no authority to conduct an investigation or to gather evidence. The Prosecutor also knew that Paquet had been in contact with Tardif for several weeks before police authorities were contacted. He was further aware that Paquet had "often" seen newspaper photographs of the appellant in the past, but chose to come forward only in 1991. The Prosecutor also knew about the first identification session with Tardif, in which Paquet was shown a photo of the appellant with all but his eyes covered, and Tardif was allowed to be present during the second identification session held in the Prosecutor's office.
One might ask why the Prosecutor did not question Tardif's involvement in the case or scrutinize the credibility of a witness who was brought forward after having first been in contact with the CHRC radio station, and then with Tardif. In our opinion, this juxtaposition of events shows the importance of the Prosecutor's duty not to allow the criminal process to be used as a vehicle to serve other ends, in this case the ends of Arthur and Tardif in attempting to defend against the appellant's defamation action. The Crown made the decision to prosecute with the full knowledge that prosecuting the appellant would potentially assist the defendants in the defamation actions. This was thus more than a simple abdication of prosecutorial responsibilities to the police or, in the case of Tardif, to a former police officer. Rather, the Prosecutor lent his office to a defence strategy in the defamation suits and, in so doing, was compromised by Tardif's manipulation of the evidence and the irregularities that took place during the 1991 investigation process.
To recapitulate on the question of motive, it was the Prosecutor's opinion in 1986 that there were no reasonable and probable grounds to lay the murder charge. The problematic evidence of Paquet and the "reconsidered" evidence of the surreptitious police recording did not, on any reasonable view, remedy the obvious deficiencies. What then motivated the Prosecutor? If it was a simple lapse of judgment, the appellant has no cause of action. But there is more. Despite knowledge of the private interest of CHRC and Tardif in defending the defamation action, the Prosecutor did not keep himself at arm's length but invited Tardif, the civil defendant, back on to the criminal prosecution team. The key witness, Paquet, was interviewed with Tardif's participation. This was despite the Prosecutor's knowledge of Tardif's previous efforts to obtain from Paquet an identification that Paquet could not in truth make. The Prosecutor might have been persuaded of the appellant's guilt, but he must have known that he lacked the credible evidence to prove it. This leads, on a balance of probabilities, to one of two conclusions. Either the Prosecutor allowed his office to be used in aid of the defence of a civil defamation action, which is a perversion of powers (détournement de pouvoirs) and an improper abuse of prosecutorial power, and thus malice in law; or, the Prosecutor decided in 1991 to go after the appellant to secure a conviction at all costs, despite his earlier decision that there was no case for the appellant to answer, and was quite willing to harness the tainted assistance of Tardif to this end. This was not only based on tunnel vision, but also tainted tunnel vision. In either case, there was a flagrant disregard for the rights of the appellant, fuelled by motives that were entirely improper. The facts of this case are thus highly exceptional. Nelles, supra, established a generous boundary within which prosecutors acting in good faith have immunity despite bad decisions. The mixed motives of the Prosecutor in this case carried him across that boundary. Unless Nelles is to be read as staking out a remedy that is available only in theory and not in practice, the appellant was entitled to hold the Prosecutor accountable in the civil action brought following the abusive prosecution.
Considering the Prosecutor's actions in their entirety, we would therefore conclude that they amount to "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'". His conduct constituted an abuse of prosecutorial power, and an attempt to mislead the court for the purposes of securing a conviction. In our view, this amounts to malice or, as Lamer J. stated in Nelles, a perpetuation of "a fraud on the process of criminal justice" (p. 194). As such, we agree with the following passage articulated by LeBel J.A. in his dissenting opinion in this case (at p. 431) [translation]:
The Prosecutor committed one illegal act after another, contrary to the principles of criminal law and the rules of procedure of the judicial system. Although he should have known the probable consequences of his acts, he abused his powers as an officer of the court and pursued an unlawful goal. That unlawful goal and thus bad faith may be inferred from the record as a whole. It would appear from both the circumstances prior to the laying of the complaint and from the conduct of the case that the objective was to obtain a conviction despite the rules of law, based on a deep-seated belief as to the accused's guilt that was not justified by an objective review of the case.
For the foregoing reasons, we would allow the appeal, set aside the judgment of the majority of the Quebec Court of Appeal, and restore the order of Rioux J. on the responsibility of the Crown, and the damages award ordered by Letarte J. The appellant is entitled to his costs in this Court and in the courts below.
The appellant brought an action for damages against the respondent, Attorney General of Quebec, following his acquittal by the Quebec Court of Appeal on a charge of first degree murder. This appeal invites us to determine what is the regime of immunity and of extra-contractual civil liability applicable in Quebec law to the Attorney General of Quebec and Crown attorneys in respect of malicious prosecutions. More specifically, we must decide whether Nelles v Ontario,  2 S.C.R. 170, applies in the Quebec civil law context, and if so, to what extent.
As regards the duality of the legal sources of Quebec law, Professor L.-P. Pigeon (later judge of the Supreme Court of Canada) observed that "[n]ot enough is made of the fact that Québec is not purely and simply a civil law province; it is a civil law province in private law but not in public law" (L.-P. PIGEON, Drafting and Interpreting Legislation (1988) at p. 65). The issue in this case concerns the public law /private law debate.
In order to determine the extent of the respondent's liability, should the respondent be found liable, it will be necessary to examine in detail the facts out of which the criminal charge against the appellant arose, as well as the evidence introduced by the prosecution at trial. To avoid repetition, I will set out here only the facts essential for the proper understanding of what follows.
A. THE CRIMINAL CHARGE
On October 25, 1982, France Alain, a mechanical engineering student at Laval University in Québec, was seriously wounded in the right hip by gunshot. Taken to hospital, she died a few minutes later.
On March 20, 1991, after a long and painstaking investigation, the Quebec Crown attorney authorized the laying of a charge of first degree murder against the appellant. On November 10, 1991, after a trial by judge and jury, the appellant was convicted as charged, and sentenced to life in prison without eligibility of parole for 25 years.
The appellant appealed his conviction, and on August 20, 1992, the Quebec Court of Appeal, in a unanimous decision rendered per curiam, set aside the guilty verdict and entered a verdict of acquittal: R v Proulx,  R.J.Q. 2047, 76 C.C.C. (3d) 316. In its decision, the Court of Appeal pointed out errors committed by the trial judge and concluded that [translation] "in the present case, a properly instructed jury `could not reasonably have decided that the accused was guilty beyond a reasonable doubt'" (p. 383 C.C.C.).
B. THE CIVIL ACTION
Following his acquittal, the appellant sued the Attorney General of Quebec, essentially claiming serious errors which had vitiated the trial as pointed out by the Court of Appeal. Essentially, the appellant claimed that such serious charge was laid without reasonable and probable grounds, and with malice.
By a first judgment dated August 7, 1997, the Quebec Superior Court found the respondent liable. In a subsequent judgment, the quantum of damages was established at $1,154,747.86, a judgment was entered against the Attorney General for that amount.
On appeal, on February 11, 1999, that judgment was set aside by a majority of the Court of Appeal and it is that decision that is now being appealed before us.
III. JUDICIAL HISTORY
To avoid repetition, I will refer here only to the judgments in the civil action. The verdicts in the criminal case are closely related to the facts that will have to be examined in detail when dealing with the extra-contractual civil liability of the Attorney General.
A. SUPERIOR COURT,  R.J.Q. 2509 (Extra-contractual Civil Liability)
With only passing reference to the Quebec regime governing the extra-contractual civil liability of the Attorney General for malicious prosecutions, Rioux J. relied almost entirely on the reproaches directed by the Court of Appeal to the trial judge when quashing the verdict of first degree murder. He concluded that (p. 2515) [translation]
the Crown charged the plaintiff based on suspicion, supposition and conjecture, and did so after relying on unlawful interrogations in which laid real traps. The Attorney General must, therefore, be held liable for the acts of his agents who approved and agreed to the police investigation conducted in this case, who went so far as to take part in it, and who decided to go ahead despite the flimsy nature of the evidence they had before them.
B. SUPERIOR COURT,  R.J.Q. 2516 (Quantum)
Since the parties had agreed to sever the process on account of the illness of the trial judge, Letarte J. took over the case and assessed the quantum of damages at $1,154,747.86, as follows (at p. 2525):
Loss of earnings capacity
Due to an error in addition, the total should read $1,179,747.86.
Letarte J., therefore, ordered the Attorney General of Quebec to pay that amount, plus interest and the additional indemnity provided by the Quebec Civil Code.
C. COURT OF APPEAL,  R.J.Q. 398
Each of the three judges of the Court of Appeal wrote separate reasons. While LeBel J.A. dissented as to the extra-contractual liability of the Attorney General of Quebec for malicious prosecution, he wrote the main opinion concerning the regime applicable in this area in Quebec law. Although with significant differences, his two colleagues agreed generally with him on this point, their disagreement related primarily to the application of the Nelles decision, supra, to the facts of the case.
(1) The Quebec Law Regime Governing the Extra-contractual Civil Liability of the Attorney General of Quebec for Malicious Prosecution
(a) LeBel J.A.
LeBel J.A. conducted an exhaustive analysis of the applicable law. He first examined the role of the Attorney General of Quebec in the Canadian criminal justice system and, more specifically, under the provisions of the Act respecting Attorney General's prosecutors, R.S.Q., c. S-35. He concluded that, despite the diversity and complexity of the duties assigned to them, prosecutors perform judicial or quasi-judicial functions (at p. 412) [translation]:
The multiple and complex functions of prosecutors, who examine, authorize, conduct or, where appropriate, withdraw charges, is not easily characterized in a comprehensive and definitive manner. The fact that those duties are closely connected and are of a discretionary nature has generally led the courts to characterize the function of prosecutors as judicial or quasi-judicial.
He went on to do an equally detailed analysis of the duties of prosecutors in relation to the laying of a criminal charge, and concluded that only [translation] "where there are sufficient grounds, which are characterized as reasonable and probable both by judicial practice and by the directives governing the conduct of Crown attorneys", must the decision to authorize prosecutions be made, based on "factors that are objectively verifiable and the probable state of the law" (p. 415).
LeBel J.A., as he then was, then turned to the rules of extra-contractual civil liability in Quebec law, and based his analysis on Laurentide Motels Ltd. v Beauport (City),  1 S.C.R. 705, concluding that the function of the prosecutor was determined primarily by public law and the legal rules that apply to [translation] "political authority" (p. 410). He made a distinction between purely administrative decisions and policy or "political" decisions; the latter being protected by relative immunity from civil liability actions, while the former are subject to the ordinary rules of civil liability in Quebec. LeBel J.A. observed that public law has moved away from the rule of absolute immunity, recognizing that suits may be brought against the state, its immunity being only relative.
At this point, LeBel J.A. considered Nelles, supra, a common law decision which, in his opinion, applied in Quebec public law, namely the relative immunity it confers on the state for the actions of its subordinates. On the other hand, because that common law decision, in his view, was based on the tort of malicious prosecution, a common law tort which he characterized as private law foreign to the Quebec law of extra-contractual civil liability, he applied the Quebec civil law as it relates to the concept of fault.
After comparing the concepts of gross negligence or deliberate fault to the types of faults for which, according to Nelles, the state may be found liable for, he concluded that (at p. 425) [translation]:
While the definitions generally assigned to the expressions "malice", "bad faith" and "gross negligence" are used in different contexts, they appear to be substantially the same as the expressions used in civil law. One common point appears to emerge. In Quebec, as in the common law provinces, it has been held that there is a distinction between the reckless and careless conduct associated with gross fault (gross negligence) and malicious or malevolent conduct or conduct in which a blameworthy state of mind was apparent.
Accordingly, in his view, the absence of reasonable and probable cause when laying a charge amounts to gross negligence in civil law, and, when combined with malice, it entails the extra-contractual civil liability of Crown attorneys. In his opinion, the Quebec civil law then corresponds to the common law test of Nelles (at p. 425) [translation]:
.... this solution recognizes a sphere of civil liability, by denying immunity in situations where malice in the exercise of prosecutorial functions or deliberate perversion of the objectives of those functions can be established.
(b) Beauregard J.A.
Beauregard J.A. although stating that he generally agreed with LeBel J.A. with respect to the regime of extra-contractual civil liability applicable in Quebec civil law in relation to malicious prosecution, clearly dissociated himself from the conclusion reached by LeBel J.A. on the question of the extent to which Nelles applies in Quebec. He adopted the integral test laid down in Nelles on the question of immunity, and thus on the liability of the Attorney General in the case. Unlike LeBel J.A., he did not resort to the concepts of fault in Quebec civil law. According to Beauregard J.A. (at p. 434) [translation]:
.... in order for an action in damages upon an acquittal to succeed, the plaintiff must prove not only that he was prosecuted when there were no reasonable and probable grounds for doing so, but also that in authorizing the charge the prosecutor exhibited malice, a concept that Lamer J. defined as follows:
.... a fraud on the process of criminal justice and in doing so has perverted or abused his office and the process of criminal justice.
He criticized the trial judge for [translation] "failing to properly apply Nelles and to ask whether the prosecutor exhibited malice in the sense of Nelles" (p. 434).
(c) Brossard J.A.
Brossard J.A. endorsed LeBel J.A.'s analysis and, like him, adopted the concept of fault as understood in civil law, while agreeing with his two colleagues that the Nelles test applies in Quebec. He held that the prosecutor was faulty and negligent in the performance of his professional duties. However, he concluded that the degree of fault alleged against the Crown attorney did not reach the threshold needed in order for the Attorney General to be liable in extra-contractual civil liability: [translation] "I do not agree with him [LeBel J.A.] regarding the application of the legal principles he stated to the facts of this case. On this conclusively important aspect of the case, I share the opinion of my colleague Beauregard J.A." (p. 435). Later in his reasons, he added (at p. 436) [translation]:
First, although the Crown has an obligation in criminal cases to prove the accused's guilt beyond a reasonable doubt, the burden in the civil trial is on the respondent to establish, by a preponderance of evidence, that the prosecutor committed a fault or faults that may be characterized as gross and malicious, or as being the result of reckless conduct. In other words, the respondent had to establish either malice on the prosecutor's part or conduct that deliberately disregarded the ultimate consequences of a criminal charge that he knew or ought to have known could not stand up to an objective analysis of the evidence.
(2) Application of the Law to the Facts of the Case
After a detailed analysis of the evidence and of the Court of Appeal's decision acquitting the appellant, LeBel J.A. wrote that [translation] "faults were committed by the prosecutor. The nature of those faults was such that he could not invoke the relative immunity recognized at common law" (p. 433). He concluded that the Attorney General's appeal had to be dismissed.
His two colleagues, Beauregard and Brossard J.A., allowed the appeal essentially on the ground that the facts, which were uncontested, as Beauregard J.A. mentioned, [translation] "cannot provide sufficient basis for Proulx's action in damages" (p. 435).
IV. THE ISSUE
The central issue in this case, as I noted at the outset, concerns the determination of which legal regime is applicable in Quebec as regards the extra-contractual civil liability of the Attorney General of Quebec and Crown attorneys for malicious prosecutions.
In order to dispose of the aforementioned issue, it is necessary to briefly review the sources of public law in Quebec and the role of the Attorney General and Crown attorneys in Canada and in Quebec in relation to criminal prosecutions, and a more thorough examination of the immunity they enjoy under Quebec law in relation to malicious prosecutions. The application of the Nelles decision, supra, to Quebec is central to this exercise.
A. LEGAL SOURCES OF PULIC LAW IN QUEBEC
No one disputes in this case that public law in Quebec derives from English law, as it has been amended over the years by statutes and case law. Both doctrine and jurisprudence have dealt with the subject at length. For a non-exhaustive study of the question, see, as to the doctrine: F. P. Walton, "The Legal System of Quebec" (1913), 33 Can. L. Times 280, at p. 281 ("Where a question is one which belongs to the public law it is the law of England which must supply the governing principle; and upon such matters the law of Quebec does not differ from that .... of any other province of Canada"); P. Garant, Droit administratif (4th ed. 1996), vol. 1, p. 10 ([translation] "the basis of our public law, in both Canada and Quebec, is the common law, which was imported from England at the time of the Conquest, as it has been amended by local statute and case law"). As for jurisprudence, see: Canadian Broadcasting Corp. v Quebec Police Commission,  2 S.C.R. 618, at p. 644 ("The source of this law is the common law, the principles of which are not set aside by statutes which do not mention it"); Senez v Montreal Real Estate Board,  2 S.C.R. 555, at p. 562 ("It need hardly be mentioned that in Quebec administrative law is of English origin"); Attorney General of Quebec v Labrecque,  2 S.C.R. 1057, at p. 1081; 2747-3174 Québec Inc. v Québec (Régie des permis d'alcool),  3 S.C.R. 919, at para. 81, per L'Heureux-Dubé J. ("the common law generally applies in Quebec public law, subject to specific legislative amendments").
The following passage from Pigeon, supra, at p. 65, is particularly relevant to this case:
[In Langelier v Giroux (1932), 52 B.R. 113,] libel proceedings had been instituted against a witness at a preliminary hearing .... The question at bar was whether the witness's immunity was absolute or relative .... The Court of Appeal decided that the matter was one not of civil law but of public law. It is true that libel is governed by civil law, under which it is an offence. It was decided, however, that the privileges arising from public law are public law, and the witness's privilege is one of these. It was consequently decided that the witness's privilege was governed by common law: common law, it was specified, as it existed in 1774.
At p. 66, the author discusses Alliance des professeurs catholiques de Montréal v Labour Relations Board of Quebec,  2 S.C.R. 140, as follows:
.... it was decided that the rule of audi alteram partem .... is a principle which, although not written down anywhere, must be applied. It is obviously a common-law rule, and it was considered a matter of public law, since it affected the Court's jurisdiction. In other words, while labour legislation is in principle considered civil law, the constitution of a court with jurisdiction in labour matters is considered public law because - and here is where the court of Appeal decision in Langelier v Giroux comes into play - the constitution of the courts, and the rules of immunity, are considered public law, not private law.
Accordingly, a brief review of sources of law in Quebec is in order. As this Court did in a different context, although that context is not applicable here: Laurentide Motels, supra, at pp. 737, 739-40 and 741:
The Quebec Act of 1774 sealed the fate of the two major legal systems that would govern the law applicable in Quebec. French civil law as it stood before 1760 with its subsequent amendments in Quebec for everything relating to property and civil rights, and the common law as it stood in England at that time, and as subsequently amended, for what related to public law.
[T]he common law which applies in Canada in the area of public law, in criminal as in administrative law, in the absence of legislation excluding it, is the common law as subsequently amended by statute and case-law ....
In everything not related to property and civil rights, then, common law is the fundamental law in the Province of Québec. [Pigeon, supra, at p. 66.]
Public law has its origin in the common law, and common-law decisions must thus be examined to determine the state of public law in the area applicable in Canada.
The nature and scope of the rules of public law and case law governing the liability of municipalities are particularly important here, where the dividing line between public and private law is crucial. This task is made all the more difficult as, at least initially, the common law courts made no clear distinction between public and private law, which are both derived from the same source, the common law.
Further, the following comments of Beetz J. in that case, are particularly relevant to this case (at p. 721):
The public law of Quebec is acknowledged to be composed of two elements: statute and the common law ....
The second component of the public law is the common law. Two clarifications must be made at this point. First, only that part of the common law which is of public character is applicable. Because the common law makes, in principle, no distinction between public and private law, the identification of the "public" common law can be a difficult task. Nonetheless, because Quebec is a jurisdiction of two juridical regimes, the civil law and the common law, the identification must be made. Second, it is the common law as it exists at present that is applicable in Quebec under art. 356 C.C.L.C.
(See also 2747-3174 Québec Inc., supra, at paras. 81-84, per L'Heureux-Dubé J.)
Professor J.-D. Archambault ("Les sources juridiques de la responsabilité extra-contractuelle de la Couronne du Québec: variations de droit public" (1992), 52 R. du B. 515), commenting on Laurentide Motels, supra, stated (at p. 537) [translation]:
The public law-private law dichotomy, which at first glance appears elementary, creates serious problems in terms of definition and application. The Quebec legal profession, however, like the profession elsewhere, does not have the option of ignoring or avoiding those problems. First, our history and our political institutions have bequeathed them both to us: the public common law, which comes from English roots, and the private civil law, which is continental and French by derivation. And neither of those root sources should infringe on its neighbour's domain, if it is not to risk jeopardizing the other's integrity.
I am thus in complete agreement with LeBel J.A. on this point when he states (at pp. 416-17) [translation]:
.... Quebec public law still seems to recognize the existence of a number of immunities that derive from the common law. They are precluded only where the legislature has demonstrated a clear intention to do so.
.... Apart from the legislation, we must then look to the principles of public law, which include the rules of the common law relating to immunity.
Given this brief overview of sources of public law applicable in Quebec, which are not disputed in this case, I now turn to the role of the Attorney General in Canadian and Quebec criminal law, similarly not in debate in this case.
B. THE ROLE OF THE ATTORNEY GENERAL IN CRIMINAL LAW
It is not disputed, either in Quebec or in this appeal, that the role of the Attorney General of Quebec and of the Crown attorneys is a matter of public law.
LeBel J.A. did a remarkable and very thorough study on this point, and I adopt it without hesitation. I would like to quote some extracts, which seem, in my view, to put that role in its proper context for the purposes of this appeal (at pp. 411-12) [translation]:
The Attorney General is traditionally responsible for administering justice and for initiating or terminating criminal and penal prosecutions. He also acts as the representative of the Crown in the courts and in various criminal proceedings. Those functions give him the status of constitutional guardian of the social peace, who has a duty to ensure that crimes and violations of the law are punished(Canada. Law Reform Commission of Canada, Criminal Procedure: Control of the Process, Ottawa: The Commission, 1975. Pp. 12 to 14).
The function of screening or authorizing prosecutions
arises at a particularly sensitive point in criminal law enforcement.
That is the point at which it is decided whether a criminal proceeding
will be initiated against an individual. It is the job of the prosecutor
to examine charges and to authorize them where there are reasonable and
probable grounds that an offence has been committed.
The Attorney General and the Attorney General's prosecutors are the guardians of the public interest, and assume a general responsibility for the efficient and proper functioning of the criminal justice system. Their role is not limited to that of private counsel who is responsible for an individual case (Canada. Law Reform Commission. Controlling Criminal Prosecutions: The Attorney General and the Crown Prosecutor, Ottawa: The Commission, 1990. P. 20).
(See: J. Sopinka, "Malicious Prosecution: Invasion of Charter Interests: Remedies: Nelles v Ontario; R v Jedynack; R v Simpson" (1995), 74 Can. Bar Rev. 366; D. Butt, "Malicious Prosecution: Nelles v Ontario: REJOINDER - John Sopinka  74 Can. Bar Rev. 366" (1996), 75 Can. Bar Rev. 335.)
I shall return later to this aspect of the analysis, which deals with the duties and obligations of the Attorney General and Crown attorneys in criminal prosecutions, more appropriately examined when applying the law to the facts of the case.
This brings us to the crux of this case: the analysis of the extra-contractual civil liability of the Attorney General for malicious prosecutions in Quebec law.
C. THE EXTRA-CONTRACTUAL CIVIL LIABILITY OF THE ATTORNEY GENERAL AND CROWN ATTORNEYS IN RELATION TO MALICIOUS PROSECTION.
The regime of liability which applies to the Crown is closely related to the rules governing relative or absolute immunity enjoyed by the Crown since times immemorial. It is clear that if the Crown enjoys absolute immunity from extra-contractual civil liability, both in private and public law, it cannot be successfully sued even for malicious prosecutions. On the other hand, if the Crown enjoys no immunity and is, in that respect, equal to any other respondent, as has been argued in the past, in both private and public law, there is nothing to prevent it from being sued for any type of extra-contractual civil liability. Is there a distinction to be made in that regard between civil liability as a matter of private law and as a matter of public law? If so, what of immunity? If there is immunity, does that immunity apply to both the Crown and the Attorney General and Crown attorneys? In the affirmative, is that immunity absolute or relative, and is there a distinction in that regard between the Crown and the Attorney General in the case of malicious prosecutions?
It is from this perspective that I propose, first, to identify the sources of Crown immunity, and to briefly describe how that immunity has evolved over the centuries, in order to determine the extent to which Nelles, supra, modified the rules of immunity. I shall then consider the extent to which that decision applies in Quebec public law, and consequently its impact on the liability of the Attorney General of Quebec and Crown attorneys in relation to malicious prosecutions.
(1) Crown Immunity
I propose to only briefly review the major phases of this almost millennium-long history since, here again, the historical aspect is not in dispute and given that there is a wealth of doctrine and jurisprudence on the subject. (See on this point P. W. Hogg and P. J. Monahan, Liability of the Crown (3rd ed. 2000); P. Lordon, Crown Law (1991); The King v Cliche,  S.C.R. 561; McArthur v The King,  Ex. C.R. 77; The King v Anthony,  S.C.R. 569; R v Canadian Broadcasting Corp.,  O.R. 55 (C.A.); R v Eldorado Nuclear Ltd.,  2 S.C.R. 551; Sparling v Québec (Caisse de dépôt et placement du Québec),  2 S.C.R. 1015; Alberta Government Telephones v Canada (Canadian Radio-television and Telecommunications Commission),  2 S.C.R. 225.)
In "Les sources juridiques de la responsabilité extra-contractuelle de la Couronne du Québec: variations de droit public", supra, at p. 517, Professor Archambault traces [translation] "the English genesis of the prerogative" which, in relation to tort liability, was expressed in the maxim that at one time became virtually axiomatic, "The King can do no wrong", to the point that [translation] "as political and constitutional mores changed, the concept of prerogative grew to encompass a range of situations".
According to H. Immarigeon, La responsabilité extra-contractuelle de la Couronne au Canada (l965), at p. 51, quoted by Professor Archambault in his article, supra, [translation] "whenever a case that involves prerogative must be decided, we must look to the principles of English law, having regard, however, to the Canadian statutory provisions that may have affected them, whether by amending them or by eliminating them".
Professor Archambault goes on to say (at pp. 519-20) [translation]:
The English law of prerogative was naturally incorporated in Canadian law at the time of the Conquest in 1760. Her Majesty's rights and privileges could then be exercised in the Colony in the same way as in the mother country, unless they were eliminated or modified by local legislation. A century later, the preamble to the Constitution Act, 1867 expressed the desire of the participating provinces to be federally united "with a Constitution similar in Principle to that of the United Kingdom". The House of Lords later held that the prerogative applied for the benefit of the provinces as well as the central government. Quebec had expressly recognized this legal heritage in art. 9 C.C.L.C.: "No act of the legislature affects the rights or prerogatives of the Crown, unless they are included therein by special enactment." Accordingly, absent legislation to the contrary, Quebec civil law as codified in 1866 did not run counter to the prerogative in public law. Thus in 1857 the Crown in Quebec enjoyed the prerogatives found in English public law, and more specifically immunity from prosecution and from proceedings in tort. The two prerogatives, which are separate but related, have been eroded, little by little, by the legislature and courts of Quebec together, and have finally become completely blunted. This metamorphosis of the law of the Crown in Quebec was accomplished on the foundation of British law.
Professor Archambault, supra, at p. 530, recalls the rocky road travelled by the prerogatives to our time, which includes the decision Cliche, supra, which, in his view, [translation] "approved the legislative repeal of the extra-contractual immunity of the Crown in Quebec" and ultimately led to art. 94 of the Code of Civil Procedure, R.S.Q., c. C-25.
Lamer J. referred specifically to art. 94 C.C.P. in Nelles, supra, while comparing the various schemes governing Crown immunity for malicious prosecutions, to the situation in Ontario, where the Crown had, until then, enjoyed absolute or quasi-absolute immunity. He wrote, in obiter (at p. 181):
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:
Lamer J. expressed no opinion as to the scope of that article in respect of public law, adding the following important caveat: "However, the substantive issue of immunity of Crown prosecutors has not been finally determined" (p. 181). He returned briefly to the subject to support his position that there are policy considerations which favour relative immunity for Crown attorneys in such cases, stating (at pp. 197-98): "In addition, since 1966, the province of Quebec permits suits against the Attorney General and Crown prosecutors without any evidence of a flood of claims."
Clearly, this is an obiter dictum. As well, those comments, made in passing, do not provide any substantive analysis of the scope of art. 94 C.C.P. in relation to actions against the Attorney General and Crown attorneys for malicious prosecutions. Lamer J. himself indicated that the question had not been finally decided and, in any event, was not necessary for the solution of Nelles, supra. In the circumstances, this obiter dictum cannot stand as authority on the issue before us. LeBel J.A. himself addressed it as follows (at p. 420) [translation]:
The incidental comments made by Lamer J. regarding the specific nature of the rules of Crown liability in Quebec law cannot alter those conclusions (see reasons of Lamer J., pp. 181, 197-198). The central issue is not Ontario private law, but the question of the immunity of the Attorney General and of prosecutors. As Lamer J. pointed out, the legal situation in Quebec differs from the situation in Ontario, since absolute immunity was abolished in Quebec long ago, given that the Code of Civil Procedure expressly provides for a right of action against the Crown, and no law has been enacted to clearly define the legal rules that apply to the Crown, the Attorney General and prosecutors. That does not mean, however, that the Supreme Court decided in that case that in Quebec the Attorney General and prosecutors enjoy no immunity, even relative. At most, we can say that in Quebec, the question has not been finally determined (see reasons of Lamer J., p. 181). In doing a comparative analysis of a number of legal systems, Lamer J. did leave civil law systems out of his analysis, because of the wide differences between those systems and the common law tradition (p. 191). It does not appear that Lamer J. dealt in that instance with the specific case of the legal system of Quebec; rather, his reasons appear to be concerned with the private law systems of major civil law nations such as France.
I would note in passing that this obiter dictum by Lamer J. does not meet the test proposed by Chouinard J. in Sellars v The Queen,  1 S.C.R. 527, at p. 530:
In Ottawa v Nepean Township et al.,  3 D.L.R. 802, Robertson, C.J.O. wrote for the Court of Ontario, at p. 804:
The opinion here is not a "considered opinion" on the subject; it is merely a comment that raises the question but is not intended to dispose of it, as Lamer J. himself states and as LeBel J.A. correctly points out
While, as I believe, art. 94 C.C.P. is intended to deal with actions in extra-contractual civil liability in private law against the Crown in Quebec, it does not, however, in my opinion, dispose of the question of actions in extra-contractual civil liability against public authorities in public law, and in particular, for the purposes of this appeal, of actions for malicious prosecution against the Attorney General of Quebec and Crown attorneys. Since Quebec has not legislated to abolish the immunities in public law that govern extra-contractual civil liability for malicious prosecutions by the Attorney General, we must look to the English public law on this point.
It is against this backdrop that we must now consider the Nelles decision, supra, and decide whether it applies in Quebec in whole or in part.
The Nelles decision dealt with a civil action for damages against the Ontario Crown, Attorney General and Crown attorneys and others, and arose out of a murder charge laid against Susan Nelles, who was acquitted after a preliminary inquiry. Ms. Nelles alleged that the Crown and its agents had committed the tort of malicious prosecution. The Ontario Crown responded to the action by making a preliminary motion to have the action dismissed on the ground that under Rule 126 of the Ontario Rules of Practice the plaintiff had no cause of action, because the Crown enjoyed absolute immunity. The motion was allowed both at first instance and on appeal, but that decision was set aside by the Supreme Court of Canada as against the Attorney General.
The question before this Court was whether the Crown, the Attorney General and Crown attorneys enjoy absolute immunity from an action for malicious prosecution. A majority of the Court concluded that while the Crown enjoys absolute immunity in this respect under the Ontario Rules of Procedure, the Attorney General and Crown attorneys do not enjoy such absolute immunity in respect of malicious prosecution. McIntyre and Lamer JJ., in turn, listed the various approaches that have been adopted around the world and in Canada and concluded that ultimately, there are policy considerations that support relative immunity in respect of the tort of malicious prosecution as it applies to the Attorney General and Crown attorneys.
LeBel J.A., with whom I agree completely on this point, adopted the principle of relative immunity, which, he wrote, [translation] "must be applied in Quebec" (p. 420). He also observed, on this point (at p. 420) [translation]:
As well, the administrative and judicial policy
considerations examined in Nelles v Ontario are matters of
public law. Because they relate to the application of a national law
that applies everywhere in Canada, they .... must be taken into account
in determining the nature of the public law rules governing an action in
liability for malicious prosecution by the Attorney General and the
Attorney General's prosecutors in Quebec.
The principles laid down by the Supreme Court of Canada in Nelles v Ontario are part of the public law that applies in Quebec. They mean that both the Attorney General and prosecutors enjoy relative immunity from civil suits, in respect of acts they perform in the course of their duties.
Because it is agreed that Nelles applies in Quebec, I shall move on immediately to the question of whether it applies integrally in Quebec, which is the question on which both the Quebec Court of Appeal and the parties to this case do not agree.
(3) Does Nelles apply integrally in Quebec?
While the appellant admits that Nelles applies in Quebec, he submits that it applies only in part; in taking this position, he relies both on the opinion of LeBel J.A. and on the following arguments, which he made in support of his contentions: (1) The present case is primarily an action in civil liability for malicious prosecution in which the plaintiff is seeking compensation for the infringement of his individual rights; (2) private law, and thus civil law, must be applied predominantly. Public law here only defines the context and the protection granted by the regime governing the civil liability of the Attorney General and of Crown attorneys. Accordingly, the principles which in Nelles derive from the private common law must give way to the application of the principles of Quebec civil law. The four criteria laid down in Nelles, therefore, cannot be applied mutatis mutandis in Quebec law. Proof of gross negligence is sufficient to ground liability against the Attorney General and Crown attorneys.
The Attorney General submits that the only onus compatible in Quebec law with the requirements of the burden of proof described in Nelles is intentional fault. Consequently, absence of reasonable and probable grounds and intent to harm must be demonstrated.
It is necessary here to refer to Nelles, and, in particular, to the statement by Lamer J. regarding the tort of malicious prosecution, which LeBel J.A. quotes (at p. 420): "Nelles v Ontario was indeed decided on the basis of the tort of malicious prosecution, a private common law action which is foreign to the Quebec law of civil liability." On the question of the tort of malicious prosecution, Lamer J. observed (at pp. 176-77):
I would like to point out that what is at issue here is not whether malicious prosecution is a reasonable cause of action. A suit for malicious prosecution has been recognized at common law for centuries dating back to the reign of Edward I. What is at issue is whether the Crown, Attorney General and Crown Attorneys are absolutely immune from suit for the well-established tort of malicious prosecution.
Here, Lamer J. drew a distinction that is crucial in the instant case between the tort of malicious prosecution in private common law, "which has existed for centuries", and the same tort of malicious prosecution in public common law, which, until then, had run up against the absolute immunity of the Crown, and the apparently absolute immunity of the Attorney General. In private common law, individuals against whom an action is brought may no longer rely on immunity. In public common law, on the contrary, until Nelles Crown attorneys seemed to enjoy absolute immunity.
While LeBel J.A. adopted the analysis of Lamer J. in Nelles and concluded that [translation] "[t]his principle of relative immunity must be applied in Quebec" (p. 420), he does not seem to have made that distinction when he described the tort of malicious prosecution as a private common law action and found it to be foreign to the Quebec law of civil liability. He did not pursue his analysis further, to determine whether there is a tort of malicious prosecution in public law, and if so, whether it applies in public law in Quebec. LeBel J.A. accordingly looked to Quebec's private civil law to determine the elements of the tort of malicious prosecution in public common law.
In my opinion, that is an incorrect interpretation of what Lamer J. said in Nelles when he referred to the common law tort of malicious prosecution "which has existed for centuries". By this, he could only have meant private prosecutions, and not public prosecutions, because at that time prosecutions were private. In fact, he stated in this respect, in Nelles (at pp. 189-90):
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 to a Canadian provincial Attorney General, was not created until 1979. In Riches v Director of Public Prosecutions,  2 All E.R. 935 (C.A.), the Court said the following in respect of suits against the D.P.P. (at p. 941):
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. As LeBel J.A. himself acknowledged in his reasons (at p. 420), and no one here is disputing, what we have in this case is a matter of public law and not of private law. In these circumstances, I believe that there is no need to bring in the notion of fault found in Quebec's private civil law and try to characterize that fault in such a way as to correspond to the Ontario common law tort of malicious prosecution, regardless of what equivalency there may be, as discussed by Professor Garant, cited by LeBel J.A. (See Garant, supra, vol. 2, at pp. 565 and 600-619)
LeBel J.A. wrote: [translation] "Apart from the statutory provisions, we must then look to the principles of public law, which include the rules of the common law governing immunities. When the delictual liability of the state or of a public body is in issue, the public law must be examined in order to determine the sphere in which the rules of liability that come from private law are to apply, and subject to what limits and conditions they will apply" (p. 417). He relied on Laurentide Motels, supra, in concluding that that case "laid down principles that are still relevant in defining the rules for stating the principles of public law and of the system of liability that comes from private law. An examination of the `public' common law will then be necessary, in order to determine whether it is legitimate to refer to the private law rules, and the limits on doing so" (p. 417). He discussed the reasons of Beetz J. in Laurentide Motels, supra, as follows (at p. 417) [translation]:
When the delictual liability of the state or of a
public body is in issue, the public law must be examined in order to
determine the sphere in which the rules of liability that come from
private law are to apply, and subject to what limits and conditions they
will apply. Although Beetz J., in his reasons in Laurentide Motels v Ville de Beauport,  1 S.C.R. 705, [pp. 724-725], was dealing
with the liability of a municipality, he laid down principles that are
still relevant in defining the rules for stating principles of public
law and of the system of liability that comes from private law. An
examination of the `public' common law will then be necessary, in order
to determine whether it is proper to refer to the private law rules, and
the limits on doing so:
Public law in Quebec has two sources, the statutory law and the common law. The meaning of the phrase "in certain respects" in art. 356 C.C.L.C. is therefore a problem to be resolved either by statutory provision or by the common law. As this Court observed in Adricon Ltée v Town of East Angus,  1 S.C.R. 1107, at p. 1120, in the matter of a municipality's contractual relations with individuals, the question is resolved by statute, since s. 28(1)(3) of the Cities and Towns Act confers upon a municipality the power to "contract transact, bind and oblige itself and others to itself, within its powers." In the matter of a municipality's delictual and quasi-delictual relations with individuals, no express statutory provision is made. Here the second source of public law, the "public" common law as it exists at present, must determine the respects in which a municipal corporation falls within the control of the civil law in its relations with individuals. Under this common law, a municipal corporation which acts within the operational sphere of its discretionary powers is subject to private law standards of conduct. In Quebec, the private law standards of conduct are those enunciated in the Civil Code, and particularly in arts. 1053 et seq. C.C.L.C.
The symmetry between the policy/operational rule drawn from the "public" common law and the collective relations/individual relations opposition established by art. 356 C.C.L.C. is no accident. When a municipal corporation exercises its discretionary powers to address political exigencies, it must make a judgment as to the interest of the community as a whole and is responsible to the community as a whole through the ballot box. However, once a municipality moves into the operational sphere of its powers, its negligent acts are susceptible of causing a distinct harm to an individual member of society, and it is responsible to that individual before the courts.
The application of the principles I have outlined above to the facts of the case at bar reveals that the responsibility of the city of Beauport for the damage caused to the appellants falls to be determined by arts. 1053 et seq. C.C.L.C.
In my opinion, as relevant as the passage from the opinion of Beetz J. cited by LeBel J.A. may be when it comes to the liability of municipal corporations in Quebec, which is governed by art. 356 of the Civil Code of Lower Canada (as amended, as to form but not as to substance, by the Civil Code of Québec, S.Q. 1991, c. 64, see arts. 300 and 1376 C.C.Q.), and given that this was an extra-contractual civil action in private law, that opinion cannot determine extra-contractual civil liability in public law for malicious prosecution, something that the judgment never considered and did not have to consider. The passages quoted supra may have misdirected the issue that is before us here, and particularly the position of the two parties on this point which LeBel J.A. himself adopted.
Unlike LeBel J.A., I do not consider it to be necessary for the purposes of this case to determine whether the decision to lay charges against an individual, which is one of the functions of the Attorney General and Crown attorneys, must be characterized as an act of administrative authority or an act of political authority (p. 418). The distinction between policy and operational spheres was borrowed by Quebec law from the public common law in Laurentide Motels for specific purposes and in a specific context: the tort of negligence, not the tort of malicious prosecution. (See J. D. Archambault, "Les sources juridiques des immunités civiles et de la responsabilité extra-contractuelle du Procureur général à raison d'accusations pénales erronées: le mixte et le mêlé (Québec c. Proulx)" (1999), 59 R. du B. 59, at pp. 82-83; P.-A. Côté, "La détermination du domaine du droit civil en matière de responsabilité civile de l'Administration québécoise -- Commentaire de l'arrêt Laurentide Motels" (1994), 28 R.J.T. 411, at p. 425; L. N. Klar, "Recent Developments in Canadian Law: Tort Law" (1991), 23 Ottawa L. Rev. 177, at p. 193.) On this point, it should be noted that this Court has applied that distinction only in situations relating to the tort of negligence (see Martel Building Ltd. v Canada,  2 S.C.R. 860, 2000 SCC 60, at para. 53; Ingles v Tutkaluk Construction Ltd.,  1 S.C.R. 298, 2000 SCC 12; Lewis (Guardian ad litem of) v British Columbia,  3 S.C.R. 1145; Brown v British Columbia (Minister of Transportation and Highways),  1 S.C.R. 420; Swinamer v Nova Scotia (Attorney General),  1 S.C.R. 445; Just v British Columbia,  2 S.C.R. 1228; Rothfield v Manolakos,  2 S.C.R. 1259).
It is instructive that while this common law rule had to be applied to the actions of the administration in the broad sense, the Supreme Court of Canada did not see fit to reiterate that distinction in relation to the tort of malicious prosecution, which was central to Nelles, a decision rendered after Laurentide Motels. (See also Ryan v Victoria (City),  1 S.C.R. 201, and Tock v St. John's Metropolitan Area Board,  2 S.C.R. 1181, in which the policy/operational dichotomy was ignored in cases relating to the tort of nuisance.) I believe that to refer here to the policy/operational dichotomy was misleading, because, as Professor Archambault has shown, [translation] "it suggests that it comprises a mandatory step in any extra-contractual liability action brought against any public authority" ("Les sources juridiques des immunités civiles et de la responsabilité extra-contractuelle du Procureur général à raison d'accusations pénales erronées: le mixte et le mêlé (Québec c. Proulx)", supra, at p. 84)
This leads me to conclude that the jurisprudence and doctrine relating to the civil liability of the administration in public law, in cases where the administration makes a policy decision, is of no assistance in this instance. (See on this point Québec (Procureur général) v Deniso LeBel Inc.,  R.J.Q. 1821 (C.A.), at p. 1838; P. Garant, "La responsabilité publique de la puissance publique: du clair obscur au nébuleux" (1991), 32 C. de D. 745; P. Giroux and S. Rochette, "La mauvaise foi et la responsabilité de l'État", in Service de la formation permanente du Barreau du Québec, Vol. 119, Développements récents en droit administratif et constitutionnel (1999), 117, at pp. 122-123.)
As I noted earlier, LeBel J.A. then reverted to Quebec private civil law to determine the elements of the public law tort of malicious prosecution in Quebec. He applied those elements, inter alia, to decide whether the Attorney General of Quebec and Crown attorneys enjoy the relative immunity in public law recognized in Nelles, supra. I cannot follow him down that path.
Professor Archambault, in "Les sources juridiques des immunités civiles et de la responsabilité extra-contractuelle du procureur général à raison d'accusations pénales erronées: le mixte et le mêlé (Québec c. Proulx)", supra, wrote on this point, at p. 97 [translation]:
By thus confusing private common law torts and public common law immunities, and simultaneously and erroneously rejecting them both, our Quebec private law has been invested with powers that it does not have, for which it was never designed. That leads to uncertainty and incoherence, in which, without actually admitting it, what is being done is to try to combine, to mix, common law considerations with principles that allegedly come from private law.
The role of the Attorney General and of Crown attorneys when they lay criminal charges, in Quebec as elsewhere, is a public law role. They represent the Crown and the public interest, and not their private interest or the interest of a private citizen, as LeBel J.A. himself stated (at p. 412) [translation]:
The Attorney General and the Attorney General's prosecutors are the guardians of the public interest, and assume a general responsibility for the efficient and proper functioning of the criminal justice system. Their role is not limited to that of private counsel who is responsible for an individual case (Canada. Law Reform Commission. Controlling Criminal Prosecutions: The Attorney General and the Crown Prosecutor. Ottawa: The Commission, 1990. P. 20).
Accordingly, it is public law that governs the actions of the Attorney General and of Crown attorneys in the performance of their duties.
Nor must it be forgotten that what was decided in Nelles was not so much the nature of the tort of malicious prosecution, but rather the scope of the immunity of the Crown and its agents as recognized in public law; this is a purely public law concept. No such concept exists in private law, except, perhaps, as Professor Archambault pointed out in "Les sources juridiques des immunités civiles et de la responsabilité extra-contractuelle du Procureur général à raison d'accusations pénales erronées: le mixte et le mêlé (Québec c. Proulx)", supra, at p. 74, in the statute law, for specific purposes: immunity is granted [translation] "to judges, commissioners, prosecutors, witnesses, parties and counsel". LeBel J.A. (at pp. 420-21) referred to this himself, as follows [translation]:
In fact, there are already other immunities for the benefit of participants in the judicial process, such as parties, witnesses or counsel (Langelier v Giroux (1932), 52 B.R. 113; Langlois v Drapeau,  B.R. 277), or justices (see: Doyon v Roussel,  R.R.A. 528 (C.A.)), and judges (see: Morier v Rivard,  2 S.C.R. 705).
Absolute immunity had been a prerogative of the Crown since time immemorial, as Professor Archambault stated in his article, portions of which I quoted, supra, in this respect ("Les sources juridiques de la responsabilité extra-contractuelle de la Couronne du Québec: variations de droit public", at pp. 519-520).
Each legal system, public and private, common law and civil law, has its unique rules, and I see no reason to intermingle them unless the legislature has itself so provided, as is the case for municipal corporations in Quebec law. How then can it be argued that private law concepts could be relevant in determining the scope of that public law immunity, however relative it may be? I believe that referring to private law concepts would take us down the wrong track.
It is important not to confuse torts, in private common law, and immunities, in public common law. While the tort of malicious prosecution is foreign to the private law of the Quebec Civil Code, the immunity defined by the public common law is not. Consequently, there is no need to translate the test that, according to Nelles, governs the relative immunity conferred on the Attorney General and Crown attorneys in public law into Quebec law, by trying to find some equivalent among the concepts of fault that are an inherent part of civil liability in Quebec. Application of the four criteria laid down in Nelles does not call for any reference to Quebec private civil law; the relative immunity of Crown counsel and of prosecutors originates in the public common law, which applies in its entirety in Quebec.
I therefore conclude from the foregoing that the extra-contractual civil liability of the Attorney General of Quebec and of the Attorney General's prosecutors for malicious prosecution is part of Quebec public law and is entirely governed in Quebec by the decision in Nelles, supra. The relative immunity enjoyed by the Attorney General and by Crown attorneys pursuant to Nelles applies to them. Consequently, the concepts of fault in the private civil law of Quebec are not applicable in deciding that liability; rather, the public law test laid down in Nelles applies.
That being said, I turn to the application of Nelles to the facts of this case, in order to determine whether the Attorney General of Quebec and the Crown attorneys are entitled to relative immunity in respect of the extra-contractual civil action brought against them. Prior to that, however, it is important to review the role of the Attorney General and Crown attorneys as regard criminal prosecutions, and the standards applicable to the performance of their duties, in accordance with Nelles.
D. APPLICATION OF NELES TO THE FACTS OF THE CASE
(1) Role and Duties of the Attorney General in Relation to the Laying of Criminal Charges
The leading decision in this respect is Boucher v The Queen,  S.C.R. 16, to which both Lamer and McIntyre JJ. referred in Nelles. Lamer J. dealt with it as follows (at pp. 19l-92):
The office of the Crown Attorney has as its main function the prosecution of and supervision over indictable and summary conviction offences. The Crown Attorney is to administer justice at a local level and in so doing acts as agent for the Attorney General. Traditionally the Crown Attorney has been described as a "minister of justice" and "ought to regard himself as part of the Court rather than as an advocate". (Morris Manning, "Abuse of Power by Crown Attorneys",  L.S.U.C. Lectures 571, at p. 580, quoting Henry Bull, Q.C.) As regards the proper role of the Crown Attorney, perhaps no more often quoted statement is that of Rand J. In Boucher v The Queen,  S.C.R. 16, at p. 23 and 24:
LeBel J.A. did a very thorough review of the question and there is no disagreement regarding the norms he identified. I believe that it is worth reproducing a passage from his opinion on this point here, with which I am in complete agreement (at p. 411) [translation]:
The critical nature of the role of the Attorney General and prosecutors in a province in the functioning of the criminal justice system in Canada is apparent even from a quick examination of their functions. The Attorney General is traditionally responsible for administering justice, and for initiating or terminating criminal and penal prosecutions. He or she also acts as representative of the sovereign, before the courts and in the various criminal proceedings. Those functions give the Attorney General the status of constitutional guardian of the social peace, and the duty to ensure that crimes and violations of statutes are punished (Canada. Law Reform Commission, Criminal Procedure: Control of the Process, Ottawa: The Commission, Pp. 12 to 14).
(2) The Nelles Test
In applying the Nelles decision, the respondent's conduct must be evaluated having regard to the role of the Attorney General. Lamer J. described the necessary cumulative elements to succeed in an action for malicious prosecution (at p. 193):
the proceedings must have been initiated by the defendant;
the proceedings must have terminated in favour of the plaintiff;
the absence of reasonable and probable cause;
malice, or a primary purpose other than that of carrying the law into effect.
(See J. G. Fleming, The Law of Torts (5th ed. 1977), at p. 598.)
Lamer J. pursued (at pp. 193-94):
The first two elements are straightforward and largely speak for themselves. The latter two elements require explicit discussion. Reasonable and probable cause has been defined 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" (Hicks v Faulkner (1878), 8 Q.B.D. 167, at p. 171, Hawkins J.)
This 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.
The required element of malice is for all intents, the equivalent of "improper purpose". It has according to Fleming, a "wider meaning than spite, ill-will or a spirit of vengeance, and includes any other improper purpose, such as to gain a private collateral advantage" (Fleming, op. cit., at p. 609). 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. (See for example breach of trust, s. 122, conspiracy re: false prosecution s 465(1)(b), obstructing justice, s. 139(2) and (3) of the Criminal Code, R.S.C. 1985, c. C-46.) [emphasis deleted.]
Lamer J. summarizes the plaintiff's burden in such actions (at p. l94):
By way of summary then, a plaintiff bringing a claim for malicious prosecution has no easy task. Not only does the plaintiff have the notoriously difficult task of establishing a negative, that is the absence of reasonable and probable cause, but he is held to a very high standard of proof to avoid a non-suit or directed verdict (see Fleming, op. cit., at p. 606, and Mitchell v John Heine and Son Ltd. (1938), 38 S.R. (N.S.W.) 466, at pp. 469-71). Professor Fleming has gone so far as to conclude that there are built-in devices particular to the tort of malicious prosecution to dissuade civil suits (at p. 606):
The test laid down in Nelles for lifting the immunity of Crown attorneys is very strict, and for a good reason. It must be recalled that before Nelles, immunity was absolute, as it has in fact always been the case in the United States. In Imbler v Pachtman, 424 U.S. 409 (1976), the United States Supreme Court held (at pp. 422-24):
The common-law immunity of a prosecutor is based upon the same considerations that underlie the common-law immunities of judges and grand jurors acting within the scope of their duties. These include concern that harassment by unfounded litigation would cause a deflection of the prosecutor's energies from his public duties, and the possibility that he would shade his decisions instead of exercising the independence of judgment required by his public trust. One court expressed both considerations as follows:
In Yaselli v Goff, 12 F.2d. 396 (2d Cir. 1926), Rogers J. wrote, at p. 406:
The public interest requires that the persons occupying such important positions and so closely identified with the judicial departments of the government should speak and act freely and fearlessly in the discharge of their important official functions. They should be no more liable to private suits for what they say and do in the discharge of their duties than are the judges and jurors, to say nothing of the witnesses who testify in a case.
Lamer J. recognized in Nelles that "[t]here is no doubt that the policy considerations in favour of absolute immunity have some merit" (p. 199). Those considerations include public confidence, diversion from duties, the choice between two evils and other available recourses. He concludes as follows (at p. 199):
Further, it is important to note that what we are dealing with here is an immunity from suit for malicious prosecution; we are not dealing with errors in judgment or discretion or even professional negligence. By contrast the tort of malicious prosecution requires proof of an improper purpose or motive, a motive that involves an abuse or perversion of the system of criminal justice for ends it was not designed to serve and as such incorporates an abuse of the office of the Attorney General and his agents the Crown Attorneys.
In that decision, I myself pointed out (at p. 223):
Attorneys General and Crown Attorneys are often faced with difficult decisions as to whether to proceed in matters which come before them. It is unfortunate that, like all human beings, they cannot be immune from error. However, the holders of such offices can and should be immune from prosecution for any such errors which occur in the course of the exercise of their functions. The freedom of action of Attorneys General and Crown Attorneys is vital to the effective functioning of our criminal justice system. In my view, the greater public interest is best served by giving absolute immunity to those agents.
This is the background against which the facts of this case must be examined having regard to the norms and test set out supra.
(3) The Judicial Framework
The facts that led to the prosecution of the appellant and the difficulties that arose at trial are not in dispute; the only issue concerns the assessment of those facts having regard to the applicable norms. It is useful, however, to recount at the outset the judicial history of this case, starting with the murder of France Alain.
Following the murder, officers John Tardif and Marcel Gauvin were put in charge of the investigation. The coroner summoned a number of witnesses, including the appellant. Early on, suspicion focused on the appellant on account of the facts established at the coroner's inquest.
In March 1986, after several weeks of inquiry, the coroner conceded that he could not conclude that the appellant had met France Alain on the evening of October 25, 1982 and no charges were laid. At that time, the Crown attorney was of the view that there was insufficient evidence.
Inactive since the coroner's inquest, the case was reactivated in early 1991, due to a new media coverage of an action in civil liability for defamation brought by the appellant against a journalist at the radio station CHRC and John Tardif, who had investigated the case. The publication of a photograph of the appellant in a local newspaper, alongside a news report on the civil action, revived anew Paul-Henri Paquet's memory. The Crown attorney at the coroner's inquest was informed of this new development. The police and the Crown then reopened the case of France Alain. The Crown attorney immediately requested the police officers to conduct a serious investigation of this new witness. They looked for additional verification concerning, among other things, written statements from the witness's relatives and on the information he would have provided in 1982. The account given by Mr. Paquet was corroborated by the testimony of a number of other people, three of whom affirming that, in the hours following the murder, Mr. Paquet had recounted the same incident he would recall in 1991.
In light of the facts recounted by this witness, the Crown attorney re-examined the whole of the evidence collected during the police investigation, including the transcripts of the police interrogations and of the coroner's inquest. Only after discussing the case with his colleagues and superiors, did the Crown attorney authorize the laying of a murder charge against the appellant, on March 20, 1991.
The appellant's preliminary inquiry was held in April 1991, at which time he was committed for trial. He was tried before judge and jury, and at the end of the trial, on November 10, 1991, he was found guilty of the first degree murder of France Alain. He was sentenced to life in prison, with no eligibility of parole for 25 years.
The Court of Appeal set aside the guilty verdict and substituted a verdict of acquittal, having concluded that the verdict was based on evidence that should have been excluded (at p. 383 C.C.C.) [translation]:
With respect, we consider that the jury's verdict rests in part on evidence that the judge should have excluded. We are also of the view that the jury ended up at its verdict as a result of inadequate instructions, in particular on the visual identification evidence and the question of motive. Finally, we consider that the identification evidence is so lacking in probative value that it would be unreasonable, even taking into account the other evidence called by the Crown, to find beyond a reasonable doubt that Mr. Paquet saw the appellant near the scene of the crime on the evening of October 25, 1982.
While we must take account of the advantages of the jury's position in regard to questions of credibility, we are on the other hand required, pursuant to the principles set out above, to consider the reasonableness of the verdict on the basis of the legally admissible evidence and what would be the appropriate instructions in law. In short, our conclusion must deal with a verdict that the jury could have reasonably rendered, not on the evidence admitted but on that which should have been admitted, after having been properly instructed by the trial judge.
Applying the principles set out in Yebes, Howard, S.(P.L.) and W.(R) to all of the evidence, we consider that in the present case, a properly instructed jury "could not reasonably have decided that the accused was guilty beyond a reasonable doubt".
(a) Preliminary Remarks
The Court of Appeal was particularly divided on the meaning and characterization of the test set out in Nelles. Beauregard J.A. strictly applied Nelles to the facts of the case, while LeBel J.A., dissenting on this point, redefined the test on the basis of the civil law definition of fault.
Applying the extremely rigorous test laid down in Nelles, which is, according to Lamer J., in the nature of a fraud on the law, I am in complete agreement with the conclusion reached by Beauregard J.A. based on the test.
As a preliminary point, it must be noted, as Brossard J.A. pointed out, that the burden of proof in civil matters is not subject to the same rules as in criminal matters. Proof beyond a reasonable doubt in criminal law is not the proof required in civil law, which is proof on a balance of probabilities. On this point, I would borrow the expression used by Professor J.-C. Royer: [translation] "Criminal law does not apply to civil law" (La preuve civile (2nd ed. 1995), at p. 473) or, as Brossard J.A. observed (at p. 436) [translation]:
In my opinion, and with respect for the contrary view, it seems to me that the trial judge's error gave the judgment of this Court acquitting the respondent, for good reason, too much weight in the civil trial. Not only is a civil trial not subject to the same rules of evidence, but also, and most importantly, the burden of proof is essentially reversed in a civil trial
.... while in the criminal case the Crown had the burden of proving the accused's guilt beyond a reasonable doubt, the burden is on the respondent, in the civil trial, [a burden of] proof by a preponderance of evidence....
On the other hand, the Crown attorney does not assume the role of the judge and does not have to anticipate all of the rulings that a judge may make in a criminal trial, those decisions often being determined by the arguments made by the defence, the gravity of the case, the gravity of the offence charged, and so on. The Crown attorney must know the law and apply it to the best of his or her knowledge.
Nor can the Crown attorney be held responsible for the errors which the judge at the criminal trial is alleged to have made as found by the Court of Appeal and on which the trial judge in the civil case relied entirely to uphold the appellant's action.
In addition, it is essential to point out that this was a prosecution based solely on circumstantial evidence, no one having witnessed the murder. Contrary to what my colleagues seem to be saying, the evidence and pleadings clearly show that neither the Crown nor the defence saw fit to compel the witness, Mr. Paquet, to identify the accused before or during the trial. In The Law of Evidence in Canada (2nd ed. 1999), J. Sopinka, S.N. Lederman and A.W. Bryant describe the requirements relating to circumstantial evidence (at pp. 39-40):
Circumstantial evidence in the criminal context is any circumstance which may or may not tend to implicate the accused in the commission of the offence for which the accused is charged .... Each piece of evidence need not alone lead to the conclusion sought to be proved. Pieces of evidence, each by itself insufficient, may however when combined, justify the inference that the facts exist.
John Robinette, "Circumstantial Evidence",  L.S.U.C. Spec. Lect. 307, said the following (at p. 307):
All that circumstantial evidence is, is that you are seeking to prove circumstances, subordinate circumstances, subordinate facts, from which a trial tribunal may draw the inference that a principal issue of fact vital to your case has been established. Therefore, as a matter of logic, if an inference may be drawn from a subordinate fact that the principal fact occurred then evidence is admissible to prove the subordinate fact and that is what is loosely called circumstantial evidence.
See also R. J. Delisle, Evidence: Principles and Problems (4th ed. 1996), at p. 23: "In cases of circumstantial evidence certain facts connected with the material fact are proved and the trier is asked to infer from those facts that the material fact exists"; A. F. Sheppard, Evidence (1988), at p. 158: "Circumstantial evidence is any item of evidence other than the testimony of an eyewitness to the material fact, which is called direct evidence."
141 As Hall J. (dissenting in part but not on this issue) wrote in Reference Re Truscott,  S.C.R. 309, at pp. 383-84:
I recognize fully that guilt can be brought home to an accused by circumstantial evidence; that there are cases where the circumstances can be said to point inexorably to guilt more reliably than direct evidence; that direct evidence is subject to the everyday hazards of imperfect recognition or of imperfect memory or both. The circumstantial evidence case is built piece by piece until the final evidentiary structure completely entraps the prisoner in a situation from which he cannot escape. There may be missing from that structure a piece here and there and certain imperfections may be discernible, but the entrapping mesh taken as a whole must be continuous and consistent.
See also R v Kaysaywaysemat (1992), 10 C.R. (4th) 317 (C.A.). Circumstantial evidence may therefore establish that the accused committed the crime: R v Bowles (1986), 21 C.C.C. (3d) 540 (C.A. Alta.); Genest v La Reine,  R.J.Q. 2387 (C.A.). The accused's conduct subsequent to the crime may also constitute circumstantial evidence of guilt: R v White,  2 S.C.R. 72; R v Ménard,  2 S.C.R. 109.
At the heart of the application of these rules of liability lies the decision by the Crown attorney to prosecute, which should be assessed having regard to the factual situation and all of the information available to him at the time he authorized the murder charge: March 20, 1991.
I will examine the four criteria set out in Nelles in turn having regard to the evidence on which the Crown attorney based his opinion that laying the information was justified.
(b) The Criminal Prosecution and the Acquittal
The first two criteria are easily met: the criminal proceedings in Proulx were initiated by the Crown attorney, who was the defendant in the civil case, Proulx v Québec (Procureur général). As well, "the [criminal] proceedings .... terminated in favour of the plaintiff": in R v Proulx, the Quebec Court of Appeal acquitted the appellant, Benoît Proulx. On the other hand, the last two criteria call for a more in-depth and detailed examination of the facts.
(c) Reasonable and Probable Grounds
LeBel J.A. stated the principles that must serve as guidelines in analysing this component of the extra-contractual civil liability for malicious prosecution in public law, which are not in dispute here (at p. 415) [translation]:
At the outset, at this critical stage of the criminal process, authorization for laying the information, objectivity and a cold and measured assessment of whether there is reasonable and probable cause are essential to the proper performance of the prosecutor's functions. The prosecutor's decision may not be based on any personal conviction of guilt. The prosecutor must retain a degree of detachment from the case, in order to evaluate all of the evidence available, as well as the impact of the rules found in the case law and the legislation regarding the admissibility of that evidence at trial, so that it can be determined whether it is objectively reasonable and in conformity with the law to initiate a prosecution. The prosecutor must not assume the role of the judge and conduct the trial in his or her head. However, the decision to authorize the charge must be based on objectively verifiable factors, and on the probable state of the law. The ultimate objective must not be to obtain a conviction at any cost. When it seems risky, at best, that a conviction could be obtained, based on an analysis of the case done with the essential professional detachment, setting a criminal proceeding in motion would violate some of the fundamental objectives and principles of the criminal justice system. That system does not allow the legal security and reputation of an individual, who is fundamentally protected by the presumption of innocence, to be jeopardized, unless there are sufficient grounds, which have been characterized as reasonable and probable both by judicial practice and by the directives governing the conduct of Crown attorneys.
Were there in fact no reasonable and probable grounds to charge the appellant when the charge of first degree murder was laid?
There are two important elements that would suggest, prima facie, that there were sufficient grounds.
First, the judge who presided at the preliminary inquiry concluded that there were reasonable and probable grounds for committing the appellant for trial.
Second, the trial judge found that there was sufficient evidence, since he did not direct a verdict of acquittal and the jury returned a guilty verdict.
A more in-depth analysis is however necessary.
On March 20, 1991, when he authorized the murder charge, did the Crown attorney have "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 [person], placed in the position of the accuser, to the conclusion that the person charged was probably guilty of the crime imputed"? (Hicks v Faulkner (1878), 8 Q.B.D. 167, at p. 171, Hawkins J., quoted by Lamer J. in Nelles, at p. 193).
(i) Appellant's Argument
The appellant submitted, first, that it was impossible to establish beyond a reasonable doubt that he had left his work station, had met the victim on the evening of the murder, owned a firearm, had had a quarrel with the victim a few days before the murder or knew important details about the victim before the police revealed them.
The appellant also contended that in its judgment of acquittal in 1992, the Court of Appeal was of the opinion that the Crown attorney's evidence was based solely on suspicion and supposition, fed by the prior relationship between the appellant and the victim, and the discovery of a polythene bag at the crime scene that was similar to those used by the appellant's employer.
(ii) Respondent's Argument
The respondent contends that the circumstantial evidence available to the Crown attorney at the time he laid the charges attests to the existence of reasonable and probable grounds for initiating a prosecution. That evidence related to: a motive, as per the conversation between the victim's father and the appellant; the knowledge of the circumstances of the crime evidenced by the content of the news item written and read on air by the appellant on the evening of the murder, which contained information not available at the time it was written; the guilty conscience on the appellant's part on the evening of the murder described by the witnesses Tanguay and Laberge; the real opportunity to commit the crime, on account of the appellant's schedule and the proximity to his place of work which gave him the necessary latitude; and the presence of evidence relating to the means used to commit the crime, such as the polythene bag found at the scene of the crime and some knowledge of how to handle a 12 calibre rifle.
The role of the Attorney General is not that of the judge, nor to be objectively satisfied, beyond a reasonable doubt, of the guilt of an accused, or even to ensure, in that respect, that the evidence he or she has will necessarily be sufficient to guarantee a guilty verdict. His or her role is limited to determining whether it is objectively reasonable, and in conformity with the law, to initiate a prosecution. In subjective terms, when the prosecutor exercises his or her discretion and lays a charge against an individual, he or she must believe in good faith in the guilt of the accused, and that certainty must be based on reasonable and probable grounds. While it is up to the prosecutor, where the evidence is circumstantial, to establish beyond a reasonable doubt that all of the facts, when considered in their entirety, lead logically to a conclusion of guilt or of exclusive opportunity, to the exclusion of any other logical solution (Truscott, supra, at pp. 383-84, per Hall J., and R v Yebes,  2 S.C.R. 168, at pp. 187 to 189), that burden of proof is applicable only at the criminal trial.
On the question of the objective analysis of the reasonableness of that belief, it must be determined whether a prudent and cautious person would have believed that the appellant was probably guilty of the crime. In applying that test, however, one must not forget the circumstances of the case, i.e. that the charges were laid on the basis of entirely circumstantial evidence to establish the identity of the author of the crime, given that the police investigation into the murder of France Alain had not led to the discovery of any eye witness. In that perspective, we need to examine both the content of the information available to the Crown attorney in 1986 and the events that occurred in 1991.
In 1986, the Crown attorney reviewed the coroner's report of the inquiry into the circumstances of the death of France Alain. In the course of that inquest, the coroner had summoned a number of witnesses, including the appellant. Although it could not be established with certainty that the appellant had met France Alain on the evening of October 25, 1982, the inquest nonetheless brought to light evidence incriminating the appellant, in particular the following:
Shortly before the murder, the appellant had been going out with the victim, France Alain.
They broke up three weeks before the evening of the murder.
Between the date when they broke up and the date of the murder, the appellant went to a party where France Alain was present. One person who was at the party said that he had witnessed friction between the appellant and France Alain.
The appellant admitted that between the time when they broke up and the time of the murder, he hung around in the immediate vicinity of the victim's home to see whether lights were on in her apartment.
The day before the murder, the victim tried to call the appellant. However, the appellant denied that he had returned France Alain's call.
The evening of the murder, the appellant was working between 4:00 p.m. and midnight at the radio station CHRC, which was located in premises near the place where the crime was committed.
France Alain was killed between 7:30 and 7:46 p.m. The investigating police officers determined that it took barely 13 minutes for a person to leave station CHRC, go by foot to the scene of the murder and come back, taking the longest route. The time needed to travel the same route by car is obviously shorter. The appellant had to be at the station for the 8:00 p.m. news broadcast; he therefore had sufficient time to make that return trip.
The appellant admitted that he had left work in the past to run errands.
When interviewed by the investigating police officers the day after the murder, the appellant was unable to confirm whether he had been away from his place of work the day before.
A green polythene bag, part of which had been torn away, was discovered near the scene of the crime, the same industrial type of bag in regular use on the CHRC premises. A ballistic examination subsequently showed that the tear was consistent with the tear caused on a specimen bag by firing a 12 calibre rifle.
The appellant had always maintained that he knew nothing about firearms. Nonetheless, he told the investigating police officers that it would have been easy for the murderer to dismantle his rifle in order to conceal it better. The appellant said that he had seen his brother dismantle a 12 calibre rifle in the past. The brother himself denied that he had ever owned a 12 calibre rifle, until he was confronted with the appellant's statement.
At about 9:00 p.m. on the evening of the murder, the appellant contacted the journalist Gilles Laberge to assign him to gather additional information relating to the murder that had just been committed near CHRC. He told Mr. Laberge that he was afraid that the victim was a [translation] "girlfriend" who lived in the neighbourhood. Gilles Laberge nonetheless assured the appellant that it was impossible to obtain any information from the Sainte-Foy municipal police from which the identity of the victim could be determined nor to learn anything more about the circumstances of her death.
The appellant stated that he had known about certain details relating to the murder as a result of an anonymous call received at about 9:30 on the evening of the murder. He wrote a news bulletin that he read on air at 10:00 p.m., and again at 11:00 p.m., which caught the attention of the investigating police officers. What was noteworthy about the bulletin was that it gave the exact age of the victim, 21, and the precise location where the victim's body was found, an [translation] "embankment", when the only one in a position to know these details were the police and the witnesses who had made the grim discovery. The victim had in fact not remained for long at the scene of the crime: the number of people who had seen the victim in the position in which she was found was therefore small. According to the Coroner's inquest, the bulletin exhibited a degree of prior knowledge of the circumstances of the crime on the part of the appellant, at a time when that kind of information was not available.
Another of the appellant's colleagues, Jean-Pierre Tanguay, was at CHRC on the evening of the murder. His work consisted of operating the console for broadcasting radio programs. He testified that the appellant told him that he had had [translation] "some very bad luck", that "his girl friend" .... "had got or had just got shot". The witness asked the appellant whether he had details and whether he knew what had happened. The appellant replied, [translation] "it happened in the street and there are no other details". At this precise time, this was the very first time that Tanguay had heard anything about the murder of France Alain. The appellant read a news bulletin at 10:00 p.m. concerning the murder of a girl whose body had been found in Sainte-Foy. The next day, the murder of France Alain made the headlines of most of the media in Quebec.
At about 11:15 on the evening of the murder, the appellant got a call from André Hébert, to whom he offered a lift back to Montmagny. The appellant took the usual route, Chemin Ste-Foy, although his attention was drawn by the light from the revolving flashers indicating the police presence at the scene of the crime. The appellant told his passenger that he was afraid it was [translation] "one of his friends who lives in the neighbourhood".
In the days following the murder, the appellant tried to obtain details concerning the circumstances of France Alain's death. He again approached the witness Mr. Laberge in an effort to obtain information about France Alain's last moments. He specifically wanted to know whether she had suffered, whether she had said something before dying. He even requested a copy of the autopsy report.
On May 30, 1983, seven months after the tragedy, the appellant visited the victim's father, Fernand Alain, and took part in a conversation that was recorded with the written consent of Mr. Alain, but without the appellant's knowledge. The appellant had a strange conversation concerning the profile of France Alain's murderer. Using the third person singular to refer to the murderer, the appellant said on that occasion that [translation] "he" had big problems, that "he" wanted to force France Alain to do certain things, that "he" had decided that she would not belong to anyone else, that "he" had acted impulsively and that "he" believed it was an accident so that he could avoid feeling guilty. The appellant added that the murderer was travelling on foot, that "he" had stayed near the scene of the crime and that "he" had had to conceal the firearm in a garbage bag in order to avoid being seen. In the course of that meeting, Fernand Alain gave the appellant a photo of the woman who the appellant had once said could have become his woman for life. The appellant could not recall that meeting until confronted with the tape recording at the coroner's inquest.
After the preliminary inquiry, other evidence surfaced which confirmed earlier testimony as to the appellant's behaviour on the evening of the crime, among them the following:
The witness Christian Thibault, a freelance journalist who was working for CHRC at the time of the murder, said he was at the radio station a few minutes after the murder took place, and placed the time at about 8:15 p.m. He recounted finding the appellant there in a state of extreme tension bordering on panic. When Thibeault questioned him about the source of this anxiety, the appellant answered that a girl had just been [translation] "shot" a few blocks from the station and he was trying to get details. Mr. Thibault was so surprised at the appellant's attitude and behaviour that when he got home, he told his companion who confirmed the conversation in 1982. He also discussed this event with a friend who also confirmed that fact.
Another witness, Suzanne Montminy, stated that she had known the appellant since 1979 and added that during 1983 she had travelled from Québec to Montmagny several times in his company. During one of those trips, the appellant told Ms. Montminy that a sixtyish man had been seen near the scene of the crime, shortly after the murder. However, the police investigators were not informed of the existence of such a sixtyish witness before 1991. This information could not have come from police sources.
At this stage, even though the investigation was not complete, as is often the case and as the evidence I have described indicates, there was a set of facts from which it was possible, if they were admissible and believed, to establish the appellant's behaviour toward the victim before the murder. This supplied a motive, and also the opportunity for the appellant to commit the crime, given his proximity to the scene of the crime and the possibility that he could have been away from his place of work at the time of the crime, as was the presence of an object on the scene of the crime that could be connected to the appellant's place of work.
The tape recording of the statement that the appellant gave to the police, which was ruled inadmissible at trial, and the tape recording of the appellant's conversation with the father of the victim, which was ruled admissible at the trial, were admissible at the civil trial, as was acknowledged by LeBel J.A. (p. 428), with whom his two colleagues agreed on this point, and, as Beauregard J.A. pointed out (at p. 435), [translation] "while this evidence was inadmissible at the criminal trial, it was entirely relevant in examining the conduct of the prosecutor".
LeBel J.A. placed great weight on these tape recordings in his analysis of the conduct of the Crown attorney. However, the conversation between the victim's father and the appellant, which he described as [translation] "veiled admissions", was, at the time it was recorded in 1983, admissible in evidence, since Hunter v Southam Inc.,  2 S.C.R. 145, and R v Duarte,  1 S.C.R. 30, were not decided until later.
In Duarte, which is of particular interest here, the Court held that evidence obtained by the police both without the knowledge and against the will of an accused to be inadmissible under s. 24(2) of the Canadian Charter of Rights and Freedoms. At the time when the conversation with France Alain's father was recorded, in 1983, however, there was no reason to believe that recording it was illegal and that evidence of the content of the recording was inadmissible.
As well, at the time the prosecution was initiated, the tape recording of that conversation would likely have been ruled admissible at trial, as it in fact was, having regard to the case law at that time, as we shall see later.
None of the facts disclosed by the investigation at that time, taken in isolation, was sufficient to establish the guilt of the accused. The question, however, is whether that evidence, when added up and taken in its entirety, could reasonably justify a finding of guilt under the test that applies to circumstantial evidence, which I set out earlier.
Clearly there was one essential element missing from this set of facts: the presence of the appellant at or near the scene at the time of the crime. With that evidence, however, the other elements would be viewed in a completely different light. Aware of that missing element, the Crown attorney was of the opinion that he did not have sufficient evidence to justify laying the charge, although in his view, the evidence against the appellant was overwhelming, as he testified at the civil trial: [translation] "[T]he evidence at that time .... was overwhelming against, of course, one witness - we often used the expression `the principal witness', and that is your client, Mr. Proulx - it was overwhelming, but not sufficient, of course, to authorize laying the charge."
After the coroner's inquest in 1986, and until the witness Paquet came forward, in 1991, no charge was laid. Mr. Paquet triggered the reopening of France Alain's case. In his testimony, the Crown attorney indicated that the insufficiency of the evidence available in 1986 was attributable to the difficulty in establishing that the appellant had been away from his place of work on the evening of the murder and that he was present at the scene of the crime at the time it was committed. The Crown attorney testified that, far from laying a charge solely on the basis of that new witness, in 1991 he reconsidered all of the already existing evidence [translation]:
Listen, Mr. Corriveau, when the witness Paquet came forward, obviously that shed a whole new light on the France Alain case.
And I won't conceal the fact that this meant that we opened up the whole case again and so, in a way, we did our homework over again.
We re-assessed all of the testimony that had been heard, all of the documents that had been seized, and also all of Mr. Paquet's testimony, to determine whether we should consider, in a way, whether a charge could, in a way, be authorized.
So, if I am to understand that it was the witness Paquet who .... who, if we could say, influenced you to lay murder charges against my client?
Listen, to answer your question, Mr. Paquet was, I'm telling you, the trigger that led to reconsideration of the whole case.
To answer your question, there were no witnesses other than Mr. Paquet who prompted us to do our homework over, that's true.
Up to that point, the Crown attorney's conduct is not open to criticism. He was aware of his role, his responsibilities and the law, and he closed the file. In 1991, Mr. P.H. Paquet came forward spontaneously neither solicited nor earlier interviewed by the investigators or the Crown attorney. He would say that when the photograph of the appellant was published in a local newspaper, reporting on the action in damages brought by the appellant after he was acquitted by the Court of Appeal, he was struck by the eyes of an individual whom he described as a [translation] "bearded man" whom he said he had met on the evening of the murder. He claimed to have spoken briefly with that person after his attention was drawn by a detonation while walking along chemin Ste-Foy near the scene of the crime and at the time of day when it was committed, facts that he had earlier reported to police officers but on which there had been no follow-up.
As noted earlier, Mr. Paquet contacted the radio station CHRC in 1991 and asked that his report be followed up, which Tardif, the investigator, did. After interviewing Paquet and recording his statement, the investigator took him to the scene of the crime and organized an identification session, at which he showed him a photograph and told him that it was that of the appellant. Looking solely at the eyes, Paquet said that these were the eyes of his "bearded man". When he examined the entire photograph, he corrected himself and said that it was not him. Tardif then informed the Ste-Foy police about his contact with Mr. Paquet.
The Crown attorney was informed by his superiors that he would have to interview Paquet. He said that he took [translation] "that event and that witness with a twenty-five-foot pole because, obviously, I had a lot of reservations".
On March 11, 1991, Paquet, Tardif, the police investigator and the Crown attorney met at the prosecutor's office, where another informal identification session was held. At that meeting, Paquet was shown eight photographs of the appellant taken during a union demonstration on May 18, 1983. Mr. Paquet spontaneously identified one of the photographs and said that he recognized his "bearded man" on it.
The Crown attorney did not readily accept Mr. Paquet's version. To satisfy himself as to this witness's credibility and good faith, he ordered an investigation of Mr. Paquet himself, his family and friends, and so on. Three of the people questioned said that Mr. Paquet had indeed informed them of his encounter with a "bearded man" during the hours following the murder, and this was the basis of the belief that Paquet would be able to identify the appellant as the person whom he had met on the evening of the murder in the circumstances already described.
Armed with this missing piece of evidence from the 1986 investigation, the Crown attorney reevaluated the whole of the evidence and particularly the transcripts of the police interviews, the Coroner's inquest report and Mr. Paquet's statement that he could identify the appellant. He was of the opinion that this evidence provided mutual corroboration and pointed inexorably to the appellant's guilt, and was likely to result in the conviction of the appellant. On March 20, 1991, after discussing it with his colleagues and superiors, he authorized the laying of a first degree murder charge against the appellant, relying on the circumstantial evidence revealed during the investigation and his conclusion that it constituted reasonable and probable grounds for laying the charge.
Having regard to the objective and subjective criteria for establishing reasonable and probable grounds which I described earlier, can it be said that, in the circumstances of the case, the evidence then in the Crown attorney's possession constituted reasonable and probable grounds for laying the charge in question?
With respect to the subjective element, from the record, it is apparent, in my opinion, that the Crown attorney had a sincere belief in the appellant's guilt. His decision was not taken lightly; the investigation was reopened, not by the Crown attorney of his own initiative, but only after Mr. Paquet spontaneously came forward. He appeared somewhat reluctant to reopen the case even after Mr. Paquet's discovery. There is nothing in the evidence that would lead me to believe that the Crown attorney had reason to believe that the appellant was innocent; quite the contrary: in his view, the evidence was overwhelming.
The objective element must be assessed from the time of the investigation and the state of the law at that time. At the outset, I do not share the opinion of the trial judge nor of LeBel J.A., that the case was based solely on suppositions and suspicion. First, where there is solely circumstantial evidence, the suppositions that are constructed are based on facts that are believed to be verifiable. Thus circumstantial evidence does not require an unassailable identification. As well, the facts that were uncovered by the investigation shed light on motive; knowledge of the circumstances of the crime; the accused's use of his time at the time of day when the crime was committed, that is, real opportunity to commit the crime; his knowledge of the use of a firearm; the fact that the weapon was carried in a polythene bag, i.e. means to commit the crime; his behaviour before and after the murder, that is, presence of a guilty conscience; the contradictions when he was questioned; the conversation with the victim's father; and so on - all factors which, taken together, could, if believed, constitute the basis of a circumstantial evidence case. (See on this point R v Ferianz (1962), 37 C.R. 37 (C.A. Ont.); R v Ruddick (1981), 57 C.C.C. (2d) 421 (C.A. Ont.).) I, therefore, reject this aspect of the analysis of LeBel J.A., who, in my view, have declined to give full weight at the outset to this set of facts.
On this point, the two pieces of evidence that raised questions dealt with the tape recording of the appellant's conversation with the victim's father and the identification of the appellant by Mr. Paquet.
1. The Tape Recording
With respect to the tape recording of the conversation in question, I discussed earlier the state of the law at the time when that conversation was recorded; the tape recording was admissible at that time. According to the judgment of the Court of Appeal in the criminal case, that recording should have been ruled inadmissible at the trial under s. 24(2) of the Charter, pursuant to Duarte, supra.
Evidence obtained in violation of a provision of the Charter is not automatically excluded in our law. Exclusion is an exceptional measure, as Zuber J.A. of the Ontario Court of Appeal wrote in R v Duguay (1985), 18 C.C.C. (3d) 289 (at p. 306):
Granted that the Charter has changed the law but it has not, overnight, transformed the healthy repute of the administration of justice into a fragile flower ready to wilt because of the admission of evidence obtained as a result of a violation of the Charter rights of an accused. The regard of the Canadian public for the administration of justice prior to the Charter, despite the fact that evidence illegally obtained was admitted as a matter of course, was, in my view, very high. The repute of the administration of justice has not now suddenly become highly vulnerable.
On the question of exclusion of evidence, R v Collins,  1 S.C.R. 265, and R v Duguay,  1 S.C.R. 93, were the leading decisions at that time in relation to the application of s. 24(2) of the Charter and its review by an appellate court. According to Lamer J., for the majority, at p. 283 of Collins: "the judge is directed by s. 24(2) to consider `all the circumstances'. The factors which are to be considered and balanced have been listed by many courts in the country". Lamer J. cited a number of those factors, which he then divided into three main categories, including the nature of the evidence and the nature of the right violated, and added: "Real evidence that was obtained in a manner that violated the Charter will rarely operate unfairly for that reason alone" (p. 284). The good faith of the police and the seriousness of the violation were important considerations, at the time, in assessing "all the circumstances", as Lamer J. in fact pointed out (at p. 285) in that decision, when he referred to R v Therens,  1 S.C.R. 613, in which Le Dain J. wrote:
The relative seriousness of the constitutional violation has been assessed in the light of whether it was committed in good faith, or was inadvertent or of a merely technical nature, or whether it was deliberate, wilful or flagrant. Another relevant consideration is whether the action which constituted the constitutional violation was motivated by urgency or necessity to prevent the loss or destruction of the evidence.
See also R v Thompson,  2 S.C.R. 1111; R v Wong,  3 S.C.R 36; R v Sobotiak (1994), 155 A.R. 16 (C.A.); and particularly, with respect to the Quebec case law at the time, Comeau v La Reine,  R.J.Q. 339 (C.A.); Shamirzadi v Polak,  R.J.Q. 1839 (C.A.)
The exclusion of evidence that is otherwise admissible, under s. 24(2) of the Charter, is a matter within the discretion of the trial judge. According to the case law at the time (Duguay (S.C.C.), supra), and since reiterated by this Court (R v Stillman,  1 S.C.R. 607; R v Belnavis,  3 S.C.R. 341), an appellate court should only intervene if the judge has made "some apparent error as to the applicable principles or rules of law" "or has made an unreasonable finding".
Stating the unanimous opinion of the Court on this point in Belnavis, supra, Iacobucci J., dissenting in part, wrote (at paras. 74-75):
This Court has emphasized on numerous occasions the importance of deferring to the s. 24(2) Charter findings of lower court judges, who hear evidence directly and are thus better placed to weigh the credibility of witnesses and gauge the effect of their testimony: see e.g., R v Duguay,  1 S.C.R. 93, at p. 98; R v Greffe,  1 S.C.R. 755, at p. 783; R v Mellenthin,  3 S.C.R. 615, at p. 625; R v Wise,  1 S.C.R. 527, at p. 539; R v Goncalves,  2 S.C.R. 3, at p. 3; R v Stillman,  1 S.C.R. 607. In Greffe, Lamer J. (as he then was) stated as follows at p. 783:
I agree with this statement, and I note that while it speaks of deference to the findings of Courts of Appeal, the same principles apply, a fortiori, to the findings of trial judges: Goncalves, supra; Stillman, supra, at para. 68. As Cory J. states in his reasons, appellate courts in general should not intervene with respect to a lower court's s. 24(2) analysis absent an error of law or unreasonable finding.
Cory J., writing for the majority in Belnavis, supra, stated (at para. 35):
The majority in Stillman, at para. 68, also reiterated the traditional position regarding appellate review of a trial judge's findings regarding s. 24(2):
It must be noted that R v Burlingham,  2 S.C.R. 206, and Stillman, supra, were decided after the appellant's criminal trial was held and are not proper guidelines for determining whether, at the time the criminal charge was laid, the Crown attorney had reason to believe that the conversation between the victim's father and the appellant was admissible.
In the appellant's criminal trial, however, the Court of Appeal reversed the trial judge's decision based solely on the test laid down in Duarte, supra, without having regard to the judicial deference in this matter affirmed by this Court in Duguay in 1989:  R.J.Q. 2047, 76 C.C.C. (3d) 316. At the same time, however, the Court of Appeal acknowledged that when the trial judge made that decision, there were particular circumstances that justified the conclusion he had reached (at p. 327 C.C.C.) [translation]:
It should be recognized that at the stage of the trial at which the voir dire was held, the judge did not have before him all the pieces of the "puzzle", to use the expression employed by counsel in the present case. These conversations between the appellant and Alain did not appear, at first glance, to be as incriminating as the prosecution subsequently showed them to be during the cross-examination of the appellant as well as in its address to the jury.
In this particular context, we cannot require more of the Crown attorney, at the time he laid the charge, than was required of the trial judge at the time his decision was made. It is one thing for the Court of Appeal to intervene in the context of an appeal from a murder conviction, it is quite another to judge the conduct of the Crown attorney at the time the charge was laid, in the context of an action for damages for malicious prosecution.
Moreover, there is nothing in the evidence from which it can be said that the police acted in bad faith in this case by suggesting that the conversation be recorded, any more than recording a similar conversation with a police informer would have been at the time (R v Hebert,  2 S.C.R. 151). Here, the additional fact that the tape recording was legal at the time it was made was an important consideration in evaluating its admissibility under s. 24(2). As the trial judge concluded, [translation] "this course of action in , given that the police were acting in good faith, in other words that it was permitted by the law .... does not bring the administration of justice into disrepute".
The Crown attorney should certainly have known what the law was when he authorized the laying of the charge. Even relying on Duarte, however, it was far from clear at the time that the evidence was inadmissible. The Crown attorney was legitimately entitled to believe that, having regard to all the circumstances of that recorded conversation, it could have been admitted in evidence, as it in fact was at the criminal trial. As the Court of Appeal pointed out with respect to the criminal trial, the trial judge was aware of the decision in Duarte. Exercising his discretion, he weighed the relevant factors in order to decide whether the admission of the evidence obtained in violation of a Charter right was likely to bring the administration of justice into disrepute, including the fact that: the recording had been made legally in 1983, with the written consent of one of the parties to the conversation; the conversation was entirely free and voluntary on the part of the appellant and there was no compulsion; it was not a statement made to a person in authority; the appellant had not been arrested or charged; and the police were acting in good faith. In those circumstances, even though the evidence was ultimately ruled inadmissible by the Court of Appeal in August 1992, could not the Crown attorney have had reason to believe, at that time, that the evidence would be ruled admissible, as it in fact was, by the trial judge? As the Court of Appeal pointed out in its judgment in the criminal case (at pp. 323-24 C.C.C.) [translation]:
The trial judge's understanding of the decision in R v Duarte ...., led him to conclude that even if these recorded conversations were obtained in violation of s. 8 of the Charter (which is admitted by the respondent), this evidence must be admitted because the police officers "acted in good faith". The good faith of the police officers, according to the judge, can be inferred from the fact that at the time when they intercepted the conversations, that is in 1983, "it was allowed by the law".
Having regard to the context in which the trial judge admitted this evidence, and the state of the law at that time, I conclude that the Crown attorney was entitled to believe that the tape recording of the conversation between Mr. Alain and the appellant would be ruled admissible at trial.
I note in passing that in the appeal relating to the criminal trial, the Court of Appeal acknowledged that the conversation between the appellant and Mr. Alain had an [translation] "incriminating" character (p. 327 C.C.C.).
2. The Identification
The second piece of evidence that the trial judge and LeBel J.A. found could not support a belief in reasonable and probable grounds to charge the appellant concerns the identification of the appellant by Mr. Paquet. On this point, the Crown attorney is criticized for improper identification, insufficient identification [translation] "by the eyes" and failure to produce identification evidence at trial. It is important to note, however, that the indictment was filed on the basis of circumstantial evidence and was not based on the identification of the appellant by Mr. Paquet, since the identification was only one of the elements of that circumstantial evidence.
To begin with, the identification must be placed in its proper context at the time when Mr. Paquet came forward. Mr. Paquet is a witness unrelated to anyone, a total stranger to the criminal trial and in fact unknown at that time, both to the coroner and to the Crown attorney. He spontaneously recognized the appellant from a photograph published in a newspaper, and it was his eyes that struck him. He testified to that effect at the trial. In the photograph published in the newspaper, the appellant had a beard; he had a beard in 1982, at the time of the murder, as the evidence at the criminal trial established. The individual whom Mr. Paquet said he had encountered and with whom he said he conversed briefly near the scene at the time of day when the crime was committed had a beard. The account of that encounter was corroborated by three witnesses to whom Mr. Paquet had spoken about it in the hours following it. Neither the credibility nor the good faith of this witness was impugned at any time or in any way, despite his having been rigorously cross-examined. A little over nine years after that encounter, when the criminal trial was underway, the appellant no longer had a beard and his physical appearance was no longer the same, as may be seen from photographs taken at that time.
The description that Mr. Paquet gave to the Ste-Foy municipal police investigator, Matte, also provided a number of additional facts regarding the height of the person he had seen (similar to the appellant's), his age (35-40), his beard, his somewhat long, curly hair, his round glasses, his checked sports clothes and his backpack. That description was on all fours with the description of the appellant in 1982, as confirmed by family and friends of the appellant.
Mr. Paquet added that in his brief conversation with the individual he encountered on the evening of the murder, that person told him [translation] "not to go and see it" .... "that it isn't nice to look at". In addition, he pointed out that when asked to identify him, he clearly recognized his bearded man, without hesitation, and that when encountered in the evening of the murder in 1982, the bearded man suddenly disappeared.
On the one hand, should the Crown attorney ignore this evidence? A prosecutor who would ignore it would have been remiss in the duties of his office, and would not have been acting in the public interest, which requires that persons who are guilty of a crime be brought before the justice system to answer for their crimes. The Crown attorney rightly took it into consideration. However, he did not content himself with the witness's account. He first wanted to ensure that the witness could actually identify the appellant. It was solely in that context that he showed him photographs of the appellant at around the time of the murder, one of which was selected by Mr. Paquet, and not for formal identification purposes at trial, as suggested by my colleagues in their reasons. There is nothing particularly surprising about the fact that the Crown attorney insisted on ascertaining that the witness could identify the appellant, especially given that the identification centered on the individual's eyes (although it also related to the other indicia described earlier).
At that time there was no reason to hold an identification parade: no one had been arrested or charged. This is an important detail. In R v Dwyer (1924), 18 Cr. App. R. 145, Heward L.C.J. made a distinction between the identification process before arrest and the formal identification process once the suspect has been arrested. On that point, he said (at pp. 147-48):
It is quite true, as counsel has urged on behalf of the Crown, that the police in showing those photographs to the persons who afterwards became witnesses, were not at all intending to influence them in the task of identification, or to equip them for it. On the contrary, they were seeking only to ascertain who were the persons proper to be arrested.
One distinction, however, is quite clear. It is one thing for a police officer, who is in doubt upon the question who shall be arrested, to show a photograph to another person in order to obtain information or a clue upon that matter; it is another thing for a police officer dealing with witnesses, who are afterwards to be called as identifying witnesses to show to those persons photographs of those whom they are about to be asked to identify beforehand.
This passage from Dwyer does not address formal identification, contrary to the manner in which my colleagues Iacobucci and Binnie JJ. apply it. The identification of the appellant was merely one element in the body of circumstantial evidence on which the charge was based. The Crown attorney's presence at the identification does nothing to change that fact. Mr. Paquet was not an eye-witness to the murder, and his testimony was not intended to identify a murderer, but rather simply to add an additional element to the body of evidence that was already available to the Crown attorney -- an important one it is true, but not the kind of identification evidence purporting to identify the author of a crime. The purpose of the identification resorted to by inspector Tardif and the Crown attorney was not to identify the suspect at trial or even to identify an accused: no one was arrested. In fact, at trial, the Crown attorney did not even proceed to identify the accused. That leads to the conclusion that the purpose of such identification was to ensure that the witness's identification was reliable.
Similarly, as indicated, and only for the purpose of the liberal interpretation courts here adopted on the question of identification, I reproduce a comment by Sopinka, Lederman and Bryant, supra, at p. 320, on R v Swanston (1982), 65 C.C.C. (2d) 453, where the British Columbia Court of Appeal admitted out-of-court identification evidence.
In R v Swanston, due to the accused's change in appearance, the victim of a robbery was able to testify at trial only that the accused resembled his attacker, although he had positively identified the accused as his assailant on previous occasions (at a line-up and at a preliminary hearing). The British Columbia Court of Appeal allowed the victim to give evidence of the earlier identifications and permitted police officers to testify that the person identified by the victim on those occasions was the accused. Accordingly, the law appears to be that evidence of extra-judicial identification is admissible not only to corroborate an identification made at trial, but as independent evidence going to identity, at least in situations where the witness who made the earlier identification is on the stand and can be subjected to cross-examination.
See also: R v Langille (1990), 59 C.C.C. (3d) 544 (Ont. C.A.).
The appellant submits that such identification was illegal and, therefore, not in conformity with the law, and consequently, that the prosecution was vitiated. An analysis of the case law on this point leads me to conclude that even an "irregular" identification can be legally admissible in evidence, as it was in this case. While its probative value may be weaker, it is then up to the judge to caution the jury as to the circumstances in which the identification was obtained and its probative value.
In Mezzo v The Queen,  1 S.C.R. 802, the headnote reads (at pp. 803-804):
In determining whether or not to direct a verdict in cases which turn on eyewitness testimony, the trial judge should address his mind to the factors going to the quality of the identification evidence and, where the frailties in the evidence can be remedied by a caution, he should leave the matter to the jury.
In the present case, the trial judge erred in law in directing a verdict of acquittal on the basis solely of the quality of the initial identification evidence. The substantial consistency in the complainant's description of the accused given on three separate occasions prior to any improper police procedures required him to put the evidence to the jury with a caution as to the inherent frailty of the identification evidence, coupled with an instruction to consider carefully the conditions under which the identification is made. A jury is in just as good a position as the trial judge to assess the witness' opportunity for observation and the strength of his evidence based on that opportunity.
In that same decision, Wilson J. did an exhaustive analysis of the impact of irregular police procedures in relation to identification, and came to the conclusion that the criteria elaborated in Canadian jurisprudence are less rigid than those adopted in the United States and England (at p. 831):
I think that several conclusions can be drawn from these authorities. First, the improprieties in the line-up do not necessarily destroy otherwise good evidence. Second, consistent with Shephard, damage can often be remedied by a proper caution. Third, at this stage of the analysis the assessment of the evidential basis for conviction depends not only on separate assessments of the weaknesses in the initial observation and the weaknesses in the conduct of the line-up but also on the impact of one or the other. As I stated at the outset, the stronger the initial identification evidence the higher the degree of subsequent impropriety it might take to undermine it.
In reference to Marcoux v The Queen,  1 S.C.R. 763, Wilson J. went on to say (at p. 832):
The discussion in Marcoux concerns the fairness or justice aspect of police procedures and of their impact on the accused's rights. There is no suggestion that apart from due process considerations the potential for inaccuracy generated by the station-house show-up rendered the evidence against Mr. Marcoux unsafe as a basis for conviction. The implication is that because Mr. Marcoux's refusal necessitated the show-up the fairness issue does not arise and that the trial judge properly put the evidence to the jury in spite of the problems in the quality of the initial observation and the subsequent identification.
In Pretrial Eyewitness Identification (1983) (a study paper prepared for the Law Reform Commission of Canada), Neil Brooks confirms that the rules set out in the case law regarding identification are very flexible. On that point, he said (at pp. 176-77):
When the police were justified in showing photographs, the courts have never suggested that this procedure ought to be followed by a corporeal identification test. In an Australian case, where defence counsel argued that the witness who had first identified the accused's photograph should later have been shown a lineup containing the accused, the Supreme Court of Australia stated: "If she had identified him in a line-up it would have been impossible to say how far she was relying on the photograph and how far on her recollection of her assailant on 17th June, and a line-up might have been harmful to his case. It was certainly not necessary.
In R v Gagnon (2000), 136 O.A.C. 116, Weiler J.A. of the Ontario Court of Appeal summarized state of the law regarding identification (at para. 91):
The trial judge was correct that the generally accepted state of the law is that, where evidence is tainted, either because identification was suggested by the accused's presence in the prisoner's box or as a result of inappropriate police procedures, the evidence is not thereby rendered inadmissible. Rather, the evidence of tainting is a factor going to the weight of the evidence which is exclusively the province of the jury. See Mezzo v The Queen (1986), 27 C.C.C. (3d) 97 (S.C.C.); R v Miaponoose (1996), 110 C.C.C. (3d) 445 at 458 (Ont. C.A.); R v Buric (1996), 106 C.C.C. (3d) 97 at 112 (Ont. C.A.) per Labrosse J.A., whose reasons were affirmed at (1997), 114 C.C.C. (3d) 95 (S.C.C.).
With respect to identification at the preliminary stage, she wrote (at para. 94):
An out-of-court description of the suspect given by a witness to the police, particularly a description that contains a distinctive personal characteristic, is admissible in court as a means of assisting the jury to assess what weight should be given to the identification that has been made in court: Tat, supra. Similarly, a videotape or photograph of a suspect in the vicinity at the time the crime was committed is also admissible provided that the accuracy and fairness of what the videotape or photograph purports to depict is established: R v Maloney (No. 2) (1976), 29 C.C.C. (2d) 431 (Ont. Co. Ct.) LeSage, Co. Ct. J. as cited in R v Brown,  O.J. No. 4864 (Gen. Div.) on the voir dire relating to the admissibility of the tape from the surveillance camera showing the homicide and robberies at the Just Desserts Café in Toronto. In these cases, there is a means for the jury to measure the accuracy or reliability of the identification made in court. As stated by Trafford J. in Brown, supra, there is an articulable basis for the jury to discount the frailties of the identification evidence. Because the jury will be able to weigh the evidence and to engage in a reasoning process respecting its admission or rejection, the admission of the evidence will not be unfair to the accused.
In R v Tat (1997), 35 O.R. (3d) 641, Doherty J.A. of the Ontario Court of Appeal also did a thorough analysis of the question, and wrote (at p. 657):
The second situation in which out-of-court statements of identification have been admitted arises where the identifying witness is unable to identify the accused at trial, but can testify that he or she previously gave an accurate description or made an accurate identification. In these circumstances, the identifying witness may testify to what he or she said or did on those earlier occasions and those who heard the description given by the witness or witnessed the identification made by the witness may give evidence of what the witness said or did.
I decipher from these decisions that the Crown attorney acted in conformity with the state of the law regarding identification. Regardless of its probative value, the identification in this instance was not illegal.
It was ultimately up to the jury to assess that identification evidence in light of Mr. Paquet's testimony at trial, and it was up to the judge to caution the jury regarding its probative value in his charge to the jury.
As regard the conduct of the Crown attorney when he laid the charge, however, having regard to the foregoing, it is reasonable to conclude that on the question of the identification by Mr. Paquet, the Crown attorney could hold a reasonable belief that this evidence was sufficient to meet the third criteria of Nelles.
At this point, and as an aside, I note that in cases where an identification is founded solely on a particular feature, such as a tattoo, a scar, a blemish, the shape of a nose -- identifications that are abundant in the case law, it is often impossible to proceed in the ordinary manner and hold an identification parade. Where would one find people who have the same characteristics? In the absence of those characteristics, the presumed perpetrator of the crime would immediately be recognized and the identification would be no more "proper" than what the witness did here. Had Mr. Paquet been shown a number of photographs of other "bearded men", and had he identified the appellant definitively in one of those photographs, the identification would certainly have carried more weight. But the fact that this recognized method was not used is not fatal, particularly in a context where Mr. Paquet had already spontaneously identified the appellant from a newspaper photograph at a time when he was unaware that he would be a witness at the trial and had not been involved in any earlier procedure in connection with the charge. Even in the absence of any other identification session organized by the investigators and the Crown attorney, the Crown attorney could have called Mr. Paquet to testify at the trial based on the original identification. I believe that the identification of the appellant using the photographs was intended only to reassure the Crown attorney before pursuing the investigation any further. In fact, the Crown attorney then pursued his investigation, so he could be satisfied, first, of the credibility and good faith of the witness, and second, of the convergence of the other evidence then in the file. He certainly cannot be criticized for doing this.
Did the fact that the identification was ruled to be insufficient because it related solely to the eyes and beard require the Crown attorney to close the case? Not in my opinion. This is true even if we believe that the method of identification chosen was inappropriate or negligent. The Crown attorney is neither judge nor jury, and he had the right, and I would go so far as to say the duty, to take that evidence to the justice system, provided that he himself believed that it was valid. There is nothing in the record to suggest that this was not the case and that he acted in bad faith. This was a very important piece of evidence in the structure of this case, and of itself, it put the appellant at the scene of the crime at the time of day when it was committed, and corroborated other evidence such as, in particular, the use of the polythene bag that was later found near the scene of the crime, the appellant's opportunity to leave work at that precise time of day, and so on.
Mr. Paquet testified at the trial about his encounter and all the facts that he had recounted to the investigators and the Crown attorney. There is no doubt that the jurors believed him. The Crown attorney was criticized for not asking Mr. Paquet to identify the appellant at trial. In his argument at trial, the Crown attorney did say that he had deliberately refrained from asking Mr. Paquet to identify the appellant for fear that he would be unable to recognize him, and he added: [translation] "If I had asked the question, I would have been making the witness contradict himself. That would have been brilliant: `Do you see your bearded man, sir? I don't see him.'" On the one hand, the explanation seems logical: in fact, at the date of the trial, the appellant no longer looked like the "bearded man" whom the witness had encountered more than six years earlier and identified from photographs that were more or less contemporaneous with the murder. On the other hand, this could have been a tactic on the Crown attorney's part, that we may or may not agree with. However, the opportunity opened to the defence to have the witness contradict himself was obvious, a fact the Crown attorney may well have counted on, all the while perhaps thinking that the witness would reaffirm his identification from the photograph published by the newspaper. It is also important to recall that the defence in cross-examination refrained from attempting to have the witness contradict himself, a matter for which the Crown attorney can surely not be criticized.
When the evidence in the Crown attorney's possession at the time he authorized the laying of the charge against the appellant for the first degree murder of France Alain is examined in its context and in the context of that time, in light of the nature of the evidence and the state of the jurisprudence and of the law at the time, as well as the duty of a Crown attorney who sincerely and in good faith believes that the perpetrator of a murder is guilty and that there are reasonable and probable grounds to justify laying a charge, I am satisfied that the Crown attorney could plausibly have believed, at that time, that he had sufficient reasonable and probable grounds to charge the appellant. It must not be forgotten, in this analysis, that at the preliminary inquiry the judge committed the appellant for trial. It is difficult to argue, in these circumstances, that the evidence in the Crown's possession was so non-probative, or so tenuous, as one would have us believe. It seems to me that if that was the case, there would have been no committal for trial. Moreover, the appellant was found guilty by the jury at the end of his criminal trial. The standard of proof beyond a reasonable doubt does not apply to the Crown attorney at the time the charge is laid, it applies only at trial; otherwise, more would be asked of the Crown attorney than of the judge at the preliminary inquiry and at trial (United States of America v Shephard,  2 S.C.R. 1067; R v Charemski,  1 S.C.R. 679).
As Ritchie J. said in Shephard, supra, at p. 1080: "a trial judge sitting with a jury [must decide] whether the evidence is "sufficient" to justify him in withdrawing the case from the jury and this is to be determined according to whether or not there is any evidence upon which a reasonable jury properly instructed could return a verdict of guilty."
In Charemski, McLachlin J., dissenting but not on this point, accurately summarized the state of the law concerning the role of the judge at the preliminary inquiry stage. On that question, she reiterated at para. 26 the principles laid down in Shephard, supra:
Until recently, no one questioned the rule that on a motion for a directed verdict the trial judge must determine whether there is sufficient evidence to permit a properly instructed jury, acting reasonably, to convict, with the implied correlative that the trial judge must weigh the evidence in the limited sense of determining whether it is capable of supporting essential inferences the Crown seeks to have the jury draw. However, in this case the Crown argues that the test has been altered in cases of circumstantial evidence by two decisions of this Court: Mezzo, supra, and Monteleone, supra. I do not agree. While some of the language of these cases is confusing, a closer reading suggests that the justices had no intention of discarding the time-hallowed and universally accepted test for directed acquittals.
Finally, recently, in R v Arcuri, 2001 SCC 54, at paras. 30 and 32, McLachlin C.J. discussed the task of the judge at a preliminary inquiry:
In performing the task of limited weighing, the preliminary inquiry judge does not draw inferences from facts. Nor does she assess credibility. Rather, the judge's task is to determine whether, if the Crown's evidence is believed, it would be reasonable for a properly instructed jury to find guilt.
This result would obviously be inconsistent with the mandate of the preliminary inquiry justice as is expressed in s. 548(1), which requires the preliminary justice inquiry to consider "the whole of the evidence". Further, it would undermine one of the central purposes of the preliminary inquiry, which is to ensure that the accused is not committed to trial unnecessarily: see R v Russell, 2001 SCC 53, at para. 20. Thus the traditional formulation of the common law rule should not be understood to foreclose consideration of defence evidence .... Whatever the evidence of the Crown and defence, the judge must consider "the whole of the evidence", in the sense that she must consider whether the evidence, if believed, could reasonably support a finding of guilt. The question is the same whether the evidence is direct or circumstantial.
One must assume that the judge at the preliminary inquiry was cognizant of the state of the law when he committed the appellant for trial. If he considered the evidence offered by the Crown attorney to be sufficient in that respect, the Crown attorney cannot be criticized for reaching the same conclusion.
The fact that the Court of Appeal concluded on appeal of the guilty verdict that the evidence presented at trial did not prove the appellant's guilt beyond a reasonable doubt, as it said, and that the trial judge committed errors, is one thing. Assessing the conduct of the Crown attorney when the criminal charge was laid is quite another.
In short, like Beauregard J.A., [translation] "I cannot say that when the prosecutor authorized the information, he deviated tangibly from the conduct of a reasonably competent and responsible prosecutor" (p. 435). I do not share the severe approach to the Crown attorney's conduct taken by LeBel J.A. In my opinion, he applied to the Crown attorney's conduct not only the concept of fault in Quebec civil law, but also the standards of the criminal law in 1999, i.e. at the time when the Court of Appeal rendered its judgment in the civil action. The time to which we should refer in evaluating that conduct is the time when the indictment was filed.
Strictly speaking, in view of my conclusion that the Crown attorney had reasonable and probable grounds for laying a charge, it is not necessary to address the question of malice. I consider it solely in case there is a difference of opinion with respect to my analysis concerning the existence of reasonable and probable grounds.
The definition of malice in Nelles, the essential element of which is the pursuit of an improper purpose, has been consistently applied by the courts in Canada. (See on this point: Boudreault v Barrett (1998), 219 A.R. 67 (C.A.); Thompson v Ontario (1998), 113 O.A.C. 82; Reynen v Canada (1995), 184 N.R. 350 (F.C.A.); Milgaard v Kujawa (1994), 118 D.L.R. (4th) 653 (Sask. C.A.); Prete v Ontario (1993), 16 O.R. (3d) 161 (C.A.); Deline v Kidd,  B.C.J. No. 645 (QL), 2000 BCSC 491; Monette v Owens (2000), 144 Man. R. (2d) 55 (Q.B.); Charemski v Ontario,  O.J. No. 5231 (QL) (S.C.J.); Fiset v Toronto (City) Police Services Board,  O.J. No.3731 (QL) (S.C.J.).) In Boudreault v Barrett, the Alberta Court of Appeal in fact concluded, at para. 14:
There is no indication of malice. The appellant has not produced any evidence suggesting that any of the respondents acted out of spite, ill-will or vengeance toward him. Nor is there evidence that any of the respondents were inspired to initiate and prosecute the charges against the appellant by any motive other than the fulfillment of their duties to enforce the criminal law.
The burden of proof to be met here is very high, as Lamer J. pointed out in Nelles, supra: "the burden of proof on the plaintiff is onerous and strict" (p. 197). As I noted in that decision: "Attorneys General and Crown Attorneys are often faced with difficult decisions as to whether to proceed in matters which come before them. It is unfortunate that, like all human beings, they cannot be immune from error .... [but the] freedom of action of Attorneys General and Crown Attorneys is vital to the effective functioning of our criminal justice system" (p. 223).
In any event, the appellant must show on a preponderance of evidence that there was an improper purpose and that the powers of the prosecutor were perverted to that end. The malicious use of the office may not have been accidental: it must be deliberate. As Lamer J. wrote in Nelles, at pp. 196-97: "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" (emphasis added).
This standard, the deliberate and malicious use of the office, is a high and clear one, in that it calls for proof of the subjective intent of the prosecutor to act out of malice or with an improper purpose by perverting the power of the office. It must be applied strictly, and we must avoid any interpretation that leaves any room for uncertainty in its application.
Thus a Crown attorney will not be personally uncertain as to whether a violation has occurred and his independence to decide and act according to his good judgment and the means available will be protected. Moreover, individuals will be spared involvement in pointless legal proceedings that could interfere with the proper administration of justice.
(i) Appellant's Argument
The appellant contends that the Crown attorney was looking for a conviction at any cost, and, therefore, that he acted with an improper purpose. He recites a number of facts in support of his argument.
The appellant asserted that the prosecution had a personal character, since the appellant had brought action in 1991 against Tardif, the retired investigator, and a journalist at CHRC for damages caused to the appellant by the defamatory statements they had allegedly made. The appellant alleged in that case that they had spread rumours publicly that he was guilty of the murder of France Alain.
The appellant relied first on the initial identification session in which Mr. Paquet participated, at which the police officer suggested the appellant's name, showed the witness only one photograph and argued with him when he denied recognizing the appellant. The appellant also impugned the second identification session at which Mr. Paquet was shown eight photographs, all of the appellant. In the appellant's submission, that identification session was utterly malicious, subjective and biased; he said that the Crown attorney knew the circumstances in which the first session had been held and he approved it nonetheless.
The appellant also impugned the conduct of the Crown attorney during argument, including his use of the tape recorded conversation between the victim's father and the appellant, to show that the Crown attorney was acting in pursuit of an improper purpose.
(ii) Respondent's Argument
The respondent argued that there was no evidence from which it could be concluded that the Crown attorney had adopted a conduct that denoted a desire to cause harm and an intent to cause prejudice to the appellant or to divert the process of the criminal justice system from its proper purposes. The respondent is of the view that, in this case, the Crown attorney had a sincere conviction of the appellant's guilt in his mind. In his submission, the evidence offers no support for the conclusion that the Crown attorney acted with the sinister purpose of obtaining a conviction with disregard for the law.
If suspicion and supposition there be, this is where they are found: there is simply no evidence of malice. On this point, I would refer to the appellant's assertions concerning the Crown attorney's malice.
With respect to the first allegation against the Crown attorney, the mere fact that he authorized the laying of a criminal charge which resulted in a verdict of acquittal cannot support the conclusion that there was malice on the part of the Crown attorney. For one thing, this would be completely contrary to the standard of conduct set out in Nelles; for another, it would place an obligation on the Crown attorney in respect of the result, and there is no justification for this here. In addition, a standard of that nature would amount to denying the Crown attorney any immunity.
The complaint on which the appellant mainly focused relates to the Crown attorney's lack of reasonable and probable grounds for laying the criminal charge, which he characterized as "improper purpose". I shall not revisit the question of whether there were reasonable and probable grounds here, as I have addressed it fully, other than to say that my conclusion on that point rules out the possibility that the Crown attorney was pursuing any "improper purpose" on this ground.
The appellant referred to the publicity surrounding the defamation action that he brought against various people, including Tardif, the investigator, and André Arthur, the radio host, who, the appellant alleged, had spread the rumour that he was guilty of the murder of France Alain. He suggested that the Crown attorney acted only to calm the public clamouring, with disregard for the rights of the appellant.
The Crown attorney had nothing to do with the defamation action, and there is nothing in the evidence to connect him to that action, directly or indirectly. The evidence in the record establishes that it was Mr. Paquet's coming forward that influenced the conduct of the Crown attorney, and not the publicity surrounding the civil defamation action.
While it is true that Mr. Paquet first went, on his own initiative, to CHRC, where André Arthur was working at the time, the Crown attorney cannot be criticized for this, as Mr. Paquet was completely unknown to him at the time.
In addition, contrary to the hypotheses stated by my colleagues Iacobucci and Binnie JJ. concerning the alleged collusion between Mr. Paquet and Messrs. Arthur and Tardif, there is not an iota of evidence in the record that could connect Mr. Paquet to Mr. Arthur. When he was questioned on that point, Mr. Paquet testified: [translation] "No, I have never spoken to him, I have never seen him in a restaurant, I have never, on any occasion, I have never spoken to him directly". I would point out that neither Mr. Paquet's credibility nor his good faith was ever doubted throughout this entire saga.
Relying on a brief three-page summary, my colleagues reach the conclusion that it was the media uproar surrounding the defamation action against Mr. Arthur and others that prompted Mr. Paquet to approach the CHRC radio station rather than the police. However, that statement needs to be considered in the context of Mr. Paquet's entire testimony, both at the criminal trial (102 pages) and at the civil trial (91 pages). When he was questioned by counsel for the appellant at the civil trial, Mr. Paquet testified that he had gone to CHRC in 1991 because he had in the past recounted to the Ste-Foy police his encounter on the evening of the murder, and told them: [translation] "I saw a bearded man at such a time and in such a place", and that on that occasion he had left his contact information with the constables, who never called him back.
When he testified at the civil trial on March 7, 1997, and was examined by counsel for the appellant concerning the reason why he did not go to the police on the morning that he visited CHRC, Mr. Paquet replied: [translation] "I had already been there". This is the context in which he came forward a second time, in February 1991, when he saw a photograph in the newspaper, and said that he had recognized the bearded man he had encountered on the evening of the murder. He then decided to go to the radio station on his own initiative. There, he met with a young trainee who introduced him to the journalist, Christian Thibault. After telling him about the encounter in question, he asked that they call him back. It was after that visit that Tardif, the investigator, contacted him and went to interview him at home, and it was Mr. Tardif to whom Mr. Paquet gave a complete tape recorded statement, which Mr. Tardif gave to the Ste-Foy police.
There is no evidence that the Crown attorney was aware of these facts before he was informed of them, later, by Mr. Tardif. The evidence is actually to the contrary. How then, in the total absence of any evidence, can the appellant connect the Crown attorney to the attention given to his defamation action in the media?
As I noted earlier, Mr. Paquet was unknown to all of the players in this case at the time of that appeal. He was a stranger to the process: he was not sought out by any party to the process. He was a party neither to the coroner's inquest nor to the police investigation; the Crown attorney was unaware of his existence. His good faith and his credibility were subsequently tested and were never doubted.
With respect to the Crown attorney, it was only at the request of his superiors that he reopened the case, and here again, with some reluctance and, given the evidence, with no great enthusiasm. In those circumstances, and given that the appellant has failed to provide any other evidence, it cannot be concluded that the publicity in question had any connection with the conduct of the Crown attorney, let alone with any malice on his part.
The appellant also attacked the identification session that Mr. Tardif had Mr. Paquet attend. At that time, Mr. Tardif was no longer in charge of the investigation for the Ste-Foy police; he was never employed by the Attorney General He was quite simply retired. Here again, the conduct of Mr. Tardif, the investigator, whatever it may have been, cannot be connected to the Attorney General, from whom he had received no mandate, nor to the Crown attorney, from whom he had received no instructions and who was even unaware of the existence of the new witness. Certainly the identification could have been conducted differently, but the charge was laid based on the circumstantial evidence, one element of which was the identification made by Mr. Paquet. In addition, the case law relating to the admissibility of improper identification evidence is very liberal.
The Crown attorney cannot be criticized for the acts of an officer who was at that time retired from the Ste-Foy police (and certainly no malice can be found on the Crown attorney's part based on those acts), when the Crown attorney was unaware of those acts, for which Mr. Tardif had received no mandate from the Crown attorney. There is nothing particularly surprising about the fact that the Attorney General, through the Ste-Foy police, subsequently obtained Mr. Tardif's services to pursue the investigation: he was the one who had conducted the entire investigation in this case from the beginning. It was logical for him to pursue the investigation rather than assigning it to another investigator who knew nothing at all about the case.
With respect to the identification of the appellant by Mr. Paquet, I share the opinion of Beauregard J.A. (at p. 434) [translation]:
.... even if the value of the identification made by Paquet was doubtful in itself, the prosecutor, rightly or wrongly but in good faith, was of the opinion that Paquet's story and the other evidence then in hand provided mutual corroboration. With respect for the contrary opinion, I am of the view that the prosecutor did not exhibit malice within the meaning of Nelles. The prosecutor acted in good faith, and given Paquet's story and the other evidence, I cannot say that there was a patent absence of the reasonable and probable grounds needed for authorizing the laying of the information.
The appellant contends that the Crown attorney's use of the conversation between the victim's father and the appellant in his argument at trial was malicious. I have discussed the admissibility of that conversation at the time it was recorded, and I will not revisit that question. If, as the Crown attorney believed at that time, the tape recording of that conversation was admissible in evidence, and since it had then been ruled admissible by the trial judge, the Crown attorney was entitled to invite the jury to replace the pronoun "he" with the pronoun "I" and cannot be criticized for doing so, since in so doing he was inviting the jury to [translation] "conclude that only the perpetrator of the crime could have talked like that", as Brossard J.A. pointed out (p. 440). In any event, there is no indication of malice on the part of the Crown attorney in making that argument.
There is one other fact that seems to me to be of considerable importance in examining malice, which does not appear to have attracted the attention of the Court of Appeal. The Crown attorney was not handling this case alone; rather, he was handling it with the full knowledge of his colleagues and his superiors, whom he consulted. The Crown attorney, who became counsel in this case only in 1986, was informed by his superiors that he had to meet with Mr. Paquet. He took [translation] "that event and that witness with a twenty-five-foot pole because, obviously, [he] had a lot of reservations". After the March 11 session with Mr. Paquet, the Crown attorney informed his superiors about the events in question and asked to be relieved of his day-to-day responsibilities [translation] "so that [he] could sift through the case from top to bottom". Before authorizing the laying of a murder charge against the appellant, the Crown attorney consulted his colleagues for about two hours, and followed the manual of guidelines relating to authorizing charges.
When examined objectively, the conduct of the Crown attorney in this case comes nowhere near the threshold required in order for relative immunity to be lifted, under the tests in Nelles.
In short, there is not an iota of evidence that the Crown attorney acted for personal purposes, out of vengeance or ill-will toward the appellant, in bad faith or beyond his mandate for improper purposes, or that he committed a fraud on the law. The fact that he may have lacked perspicacity in failing to anticipate all of the difficulties that arose at the trial does not get us to the rigorous threshold laid down in Nelles for lifting the relative immunity enjoyed by the Attorney General and by Crown attorneys in such matters. A careful examination of the facts in evidence in the civil action leads me to conclude that the Crown attorney acted within the bounds of his functions as a public officer, by prosecuting an individual whom he believed, in good faith, to be guilty of a crime. He was entitled to believe, at that time, that he had sufficient reasonable and probable grounds to charge him. The burden of showing that the Crown attorney acted with malice in laying the criminal charge rested on the appellant, and in my opinion he has failed to meet that burden. (See on this point a similar conclusion in Perron v Québec (Procureur général),  Q.J. No. 4700 (QL), at para. 950)
The appellant has failed to prove that he was prosecuted maliciously for the murder of France Alain, and so the Attorney General and the Crown attorney, who enjoy the immunity in this respect conferred on them by Nelles, cannot be found to have any extra-contractual civil liability.
Considering that the extra-contractual civil liability of the Attorney General of Quebec and of the Crown attorney for malicious prosecution is part of the public law of Quebec;
Considering that the public law of Quebec in this matter is governed by the public common law;
Considering that the decision in Nelles applies integrally in Quebec to malicious prosecution by the Attorney General and Crown attorneys;
Considering that as a consequence, the Attorney General and Crown attorneys enjoy relative immunity in respect of such actions;
Considering the protection that this immunity confers against extra-contractual civil liability for malicious prosecution;
Considering that the facts alleged against the Attorney General and the Crown attorney in this case do not meet the test set out in Nelles, having regard to the role and duties of the Attorney General and of Crown attorneys in respect of prosecutions, as well as the evidence in the record that establishes that, at the time the prosecution was initiated, the Crown attorney could reasonably have believed that he had reasonable and probable grounds to charge the appellant, and that he did not act with malice;
I would, therefore, affirm the majority decision of the Court of Appeal and I would dismiss the appeal. Like the Court of Appeal, I would do so without costs.
R v Bain,  1 S.C.R. 91; R v Charest (1990), 28 Q.A.C. 258; R v S. (F.) (2000), 47 O.R. (3d) 349; R v Campbell,  1 S.C.R. 565.; Nelles v Ontario,  2 S.C.R. 170; Laurentide Motels Ltd. v Beauport (City),  1 S.C.R. 705; Canadian Broadcasting Corp. v Quebec Police Commission,  2 S.C.R. 618; Senez v Montreal Real Estate Board,  2 S.C.R. 555; Attorney General of Quebec v Labrecque,  2 S.C.R. 1057; 2747-3174 Québec Inc. v Quebec (Régie des permis d'alcool),  3 S.C.R. 919; Alliance des professeurs catholiques de Montréal v Labour Relations Board of Quebec,  2 S.C.R. 140; The King v Cliche,  S.C.R. 561; McArthur v The King,  Ex. C.R. 77; The King v Anthony,  S.C.R. 569; R v Canadian Broadcasting Corp.,  O.R. 55; R v Eldorado Nuclear Ltd.,  2 S.C.R. 551; Sparling v Quebec (Caisse de dépôt et placement du Québec),  2 S.C.R. 1015; Alberta Government Telephones v Canada (Canadian Radio-television and Telecommunications Commission),  2 S.C.R. 225; Sellars v The Queen,  1 S.C.R. 527; Martel Building Ltd. v Canada,  2 S.C.R. 860, 2000 SCC 60; Ingles v Tutkaluk Construction Ltd.,  1 S.C.R. 298, 2000 SCC 12; Lewis (Guardian ad litem of) v British Columbia,  3 S.C.R. 1145; Brown v British Columbia (Minister of Transportation and Highways),  1 S.C.R. 420; Swinamer v Nova Scotia (Attorney General),  1 S.C.R. 445; Just v British Columbia,  2 S.C.R. 1228; Rothfield v Manolakos,  2 S.C.R. 1259; Ryan v Victoria (City),  1 S.C.R. 201; Tock v St. John's Metropolitan Area Board,  2 S.C.R. 1181; Québec (Procureur général) v Deniso LeBel Inc.,  R.J.Q. 1821; Boucher v The Queen,  S.C.R. 16; Imbler v Patchtman, 424 U.S. 409 (1976); Yaselli v Goff, 12 F.2d 396 (1926); Reference re Truscott,  S.C.R. 309; R v Kaysaywaysemat (1992), 10 C.R. (4th) 317; R v Bowles (1986), 21 C.C.C. (3d) 540; Genest v La Reine,  R.J.Q. 2387; R v White,  2 S.C.R. 72; R v Ménard,  2 S.C.R. 109; Hicks v Faulkner (1878), 8 Q.B.D. 167; R v Yebes,  2 S.C.R. 168; Hunter v Southam Inc.,  2 S.C.R. 145; R v Duarte,  1 S.C.R. 30; R v Ferianz (1962), 37 C.R. 37; R v Ruddick (1981), 57 C.C.C. (2d) 421; R v Duguay,  1 S.C.R. 93, aff'g (1985), 18 C.C.C. (3d) 289; R v Collins,  1 S.C.R. 265; R v Therens,  1 S.C.R. 613; R v Thompson,  2 S.C.R. 1111; R v Wong,  3 S.C.R. 36; R v Sobotiak (1994), 155 A.R. 16; Comeau v La Reine,  R.J.Q. 339; Amadzadegan-Shamirzadi v Polak,  R.J.Q. 1839; R v Stillman,  1 R.S.C. 607; R v Belnavis,  3 S.C.R. 341; R v Burlingham,  2 S.C.R. 206; R v Hebert,  2 S.C.R. 151; R v Dwyer (1924), 18 Cr. App. R. 145; R v Swanston (1982), 65 C.C.C. (2d) 453; R v Langille (1990), 59 C.C.C. (3d) 544; Mezzo v The Queen,  1 S.C.R. 802; Marcoux v The Queen,  1 S.C.R. 763; R v Gagnon (2000), 136 O.A.C. 116; R v Tat (1997), 35 O.R. (3d) 641; United States of America v Shephard,  2 S.C.R. 1067; R v Charemski,  1 S.C.R. 679; R v Arcuri, 2001 SCC 54; Boudreault v Barrett (1998), 219 A.R. 67; Thompson v Ontario (1998), 113 O.A.C. 82; Reynen v Canada (1995), 184 N.R. 350; Milgaard v Kujawa (1994), 118 D.L.R. (4th) 653; Prete v Ontario (1993), 16 O.R. (3d) 161; Deline v Kidd,  B.C.J. No. 645 (QL), 2001 BCSC 491; Monette v Owens (2000), 144 Man. R. (2d) 55; Charemski v Ontario,  O.J. No. 5231 (QL); Fiset v Toronto (City) Police Services Board,  O.J. No. 3731 (QL); Perron v Québec (Procureur général),  Q.J. No. 4700 (QL).
Act respecting Attorney General's prosecutors, R.S.Q., c. S-35.
Canadian Charter of Rights and Freedoms, ss. 8, 24(2).
Civil Code of Lower Canada, art. 356.
Civil Code of Québec, S.Q. 1991, c. 64, arts. 300, 1376.
Code of Civil Procedure, R.S.Q., c. C-25, s. 94.
Authors and other references
Archambault, Jean-Denis. "Les sources juridiques de la responsabilité extra-contractuelle de la Couronne du Québec: variations de droit public" (1992), 52 R. du B. 515.
Archambault, Jean-Denis. "Les sources juridiques des immunités civiles et de la responsabilité extracontractuelle du procureur général à raison d'accusations pénales erronées : le mixte et le mêlé (Québec c. Proulx)" (1999), 59 R. du B. 59.
Butt, David. "Malicious Prosecution: Nelles v Ontario: REJOINDER -- John Sopinka  74 Can. Bar Rev. 366" (1996), 75 Can. Bar Rev. 335.
Canada. Law Reform Commission. Pretrial Eyewitness Identification. By Neil Brooks. Ottawa: The Commission, 1983.
Côté, Pierre-André. "La détermination du domaine du droit civil en matière de responsabilité civile de l'Administration québécoise -- Commentaire de l'arrêt Laurentide Motels" (1994), 28 R.J.T. 411.
Delisle, Ronald Joseph. Evidence: Principles and Problems, 4th ed. Scarborough: Carswell, 1996.
Fleming, John G. The Law of Torts, 5th ed. Sydney, Aust.: Law Book Co., 1977.
Garant, Patrice. Droit Administratif, 4e éd. Cowansville: Yvon Blais, 1996.
Garant, Patrice. "La responsabilité civile de la puissance publique: du clair obscur au nébuleux" (1991), 32 C. de D. 745.
Giroux, Pierre, et Stéphane Rochette. "La mauvaise foi et la responsabilité de l'État". Dans Service de la formation permanente du Barreau du Québec, Développements récents en droit administratif et constitutionnel, vol. 119, Cowansville, Qué.: Yvon Blais, 1999, 117.
Hogg, Peter W., and Patrick J. Monahan. Liability of the Crown, 3rd ed. Scarborough: Carswell, 2000.
Immarigeon, Henriette. La responsabilité extra-contractuelle de la Couronne au Canada. Montréal: Wilson et Lafleur, l965.
Klar, Lewis N. "Recent Developments in Canadian Law: Tort Law" (1991), 23 Ottawa L. Rev. 177.
Lordon, Paul. Crown Law. Toronto: Butterworths, 1991.
Pigeon, Louis-Phillippe. Drafting and Interpreting Legislation. Toronto: Carswell, 1988.
Robinette, John. "Circumstantial Evidence",  Spec. Lect. L.S.U.C. 307.
Sopinka, John. "Malicious Prosecution: Invasion of Charter Interests: Remedies: Nelles v Ontario: R v Jednynack: R v Simpson" (1995), 74 Can. Bar Rev. 366.
Sopinka, John, Sidney N. Lederman and Alan W. Bryant. The Law of Evidence in Canada, 2nd ed. Toronto: Butterworths, 1999.
Walton, F. P. "The Legal System of Quebec" (1913), 33 Can. Law Times 280.
Christian Trépanier and Lawrence Corriveau, Q.C., for the appellant (instructed by Corriveau, Corriveau, Québec).
Claude Gagnon, Alain Loubier and Carole Soucy, for the respondent (instructed by St-Laurent, Gagnon, Québec).
all rights reserved