R. v. Oickle,  2 S.C.R. 3
Her Majesty The Queen Appellant
Richard Floyd Oickle Respondent
The Attorney General for Ontario and
the Criminal Lawyers’ Association (Ontario) Interveners
Indexed as: R. v. Oickle
Neutral citation: 2000 SCC 38.
File No.: 26535.
1999: November 2; 2000 September 29.
Present: L’Heureux‑Dubé, McLachlin, Iacobucci, Major, Bastarache, Binnie and Arbour JJ.
on appeal from the court of appeal for nova scotia
Criminal law — Evidence — Admissibility — Confessions — Voluntariness — Common law limits on police interrogation — Proper scope of confessions rule — Police informing arson suspect that he had “failed” polygraph test — Suspect subsequently confessing to arson during skillful and lengthy interrogation — Whether police improperly induced suspect’s confession.
Criminal law — Evidence — Confessions — Voluntariness — Polygraph — Police informing arson suspect that he had “failed” polygraph test — Suspect subsequently confessing to arson during skillful and lengthy interrogation — Whether police’s failure to tell suspect that polygraph results inadmissible producing involuntary confession — Whether misleading of suspect concerning accuracy of polygraph rendered confession involuntary — Whether polygraph creating oppressive atmosphere — Whether tactical disadvantage to defence when suspect confessed following polygraph test relevant to voluntariness of confession.
Criminal law — Appeal — Court of Appeal — Standard of appellate review — Voluntariness of confessions — Court of Appeal disagreeing with trial judge as to weight to be given various pieces of evidence — Whether disagreement grounds to reverse trial judge’s finding on voluntariness of confession.
During the police investigation of a series of eight fires, the accused agreed to submit to a polygraph. The test took place in a motel and the police audiotaped the events. The accused was informed of his rights to silence, to counsel, and to leave at any time. He was also informed that while the interpretation of the polygraph results was not admissible, anything he said was admissible. At the conclusion of the test, around 5 p.m., the officer conducting the test informed the accused that he had failed it. The accused was reminded of his rights and questioned for one hour. At 6:30 p.m., a second officer questioned the accused and, after 30 to 40 minutes, the accused confessed to setting the fire to his fiancée’s car and provided the police with a statement. He appeared emotionally distraught at this time. The accused was arrested and warned of his rights. At the police station, he was placed in an interview room equipped with videotaping facilities where he was questioned about the other fires. Around 8:30 p.m. and 9:15 p.m., the accused indicated that he was tired and wanted to go home. He was informed that he was under arrest and he could call a lawyer but that he could not go home. A third officer took over the interrogation at 9:52 p.m. He questioned the accused until about 11:00 p.m., at which time the accused confessed to setting seven of the eight fires. The accused was then seen crying with his head in his hands. The police then took a written statement from the accused. He was placed in a cell to sleep at 2:45 a.m. At 6:00 a.m., a police officer noticed that the accused was awake and asked whether he would agree to a re‑enactment. On the tape of the re‑enactment, the accused was informed of his rights and was advised that he could stop the re‑enactment at any time. The police drove the accused to the various fire scenes, where he described how he had set each fire. The accused was charged with seven counts of arson. The trial judge ruled on a voir dire that the accused’s statements, including the video re‑enactment, were voluntary and admissible, and subsequently convicted him on all counts. The Court of Appeal excluded the confessions and entered an acquittal.
Held (Arbour J. dissenting): The appeal should be allowed and the conviction restored.
Per L’Heureux‑Dubé, McLachlin, Iacobucci, Major, Bastarache and Binnie JJ.: The confessions rule focuses not just on reliability but also on voluntariness broadly defined. The common law confessions rule can offer protections beyond those guaranteed by the Canadian Charter of Rights and Freedoms. While it may be appropriate in certain cases to interpret one in light of the other, it would be a mistake to assume one subsumes the other entirely. In defining the confessions rule it is important to keep in mind its twin goals of protecting the rights of the accused without unduly limiting society’s need to investigate and solve crimes.
The application of the confessions rule is of necessity contextual. Hard and fast rules simply cannot account for the variety of circumstances that vitiate the voluntariness of a confession. When reviewing a confession, a trial judge should therefore consider all the relevant factors. The judge should strive to understand the circumstances surrounding the confession and ask if it gives rise to a reasonable doubt as to the confession’s voluntariness, taking into account all the aspects of the rule. The relevant factors include threats or promises, oppression, the operating mind requirement and police trickery. While obviously imminent threats of torture will render a confession inadmissible, most cases will not be so clear. The use of veiled threats, for instance, requires close examination. The police may often offer some kind of inducement to the suspect to obtain a confession. This becomes improper only when the inducements, whether standing alone or in combination with other factors, are strong enough to raise a reasonable doubt about the voluntariness of the confession. An important consideration in all cases is to look for a quid pro quo offer by interrogators, regardless of whether it comes in the form of a threat or a promise. Oppressive conditions and circumstances clearly also have the potential to produce an involuntary confession. In assessing oppression, courts should consider whether a suspect was deprived of food, clothing, water, sleep, or medical attention; was denied access to counsel; was confronted with fabricated evidence; or was questioned aggressively for a prolonged period of time. The operating mind doctrine only requires that the accused knows what he is saying and that it may be used to his detriment. Like oppression, the operating mind doctrine should not be understood as a discrete inquiry completely divorced from the rest of the confessions rule. The operating mind doctrine is just one application of the general rule that involuntary confessions are inadmissible. Lastly, the police use of trickery to obtain a confession must also be considered in determining whether a confession is voluntary or not. This doctrine is a distinct inquiry. While it is still related to voluntariness, its more specific objective is maintaining the integrity of the criminal justice system. There may be situations in which police trickery, though neither violating the right to silence nor undermining voluntariness per se, is so appalling as to shock the community. In such cases, the confessions should be excluded.
In sum, because of the criminal justice system’s overriding concern not to convict the innocent, a confession will not be admissible if it is made under circumstances that raise a reasonable doubt as to voluntariness. Voluntariness is the touchstone of the confessions rule and a useful term to describe the various rationales underlying the rule. If the police interrogators subject the suspect to utterly intolerable conditions, or if they offer inducements strong enough to produce an unreliable confession, the trial judge should exclude it. Between these two extremes, oppressive conditions and inducements can operate together to exclude confessions. If the trial judge properly considers all the relevant circumstances, then a finding regarding voluntariness is essentially a factual one, and should only be overturned for some palpable and overriding error which affected the trial judge’s assessment of the facts.
In this case, the Court of Appeal applied the wrong standard of appellate review. Determining whether or not a confession is voluntary is a question of fact, or of mixed law and fact. Therefore, a disagreement with the trial judge regarding the weight to be given various pieces of evidence is not grounds to reverse a finding on voluntariness. The Court of Appeal also reached the wrong conclusion with regard to voluntariness. The police conducted a proper interrogation. The accused was fully apprised of his rights at all times. The police questioning, while persistent and often accusatorial, was never hostile, aggressive, or intimidating. In this context, the alleged inducements offered by the police do not raise a reasonable doubt as to the confession’s voluntariness.
In particular, the police did not improperly offer leniency to the accused by minimizing the seriousness of his offences. While the police did minimize the moral significance of the crimes, they never suggested that a confession would minimize the legal consequences of the accused’s crimes. As for the offers of psychiatric help, at no point did the police ever suggest that the accused could only get help if he confessed. There was never any insinuation of a quid pro quo. The police did suggest that confession would make the accused feel better, that his fiancée and members of the community would respect him for admitting his problem, and that he could better address his apparent pyromania if he confessed. However, read in context, none of these statements contained an implied threat or promise. As to the alleged threats against the accused’s fiancée, there were moments when the police intimated that it might be necessary to question her to make sure she was not involved in the fires. The relationship the accused had with his fiancée was strong enough potentially to induce a false confession were she threatened with harm. However, no such threat ever occurred. The most they did was promise not to polygraph her if the accused confessed. Given the entire context, the most likely reason to polygraph her was not as a suspect, but as an alibi witness. This is not a strong enough inducement to raise a reasonable doubt as to the voluntariness of the accused’s confession. The timing of the comments regarding his fiancée suggests that there was no causal connection between the police inducements and the subsequent confession. The police did not improperly abuse the accused’s trust to obtain a confession. The Court of Appeal criticized the police for questioning the accused in such a gentle, reassuring manner that they gained his trust. This does not render a confession inadmissible. Lastly, to hold that the police conduct in this interrogation was oppressive would leave little scope for police interrogation. They were always courteous; they did not deprive the accused of food, sleep, or water; they never denied him access to the bathroom; they fully apprised him of his rights at all times, and they did not fabricate evidence. While the re‑enactment was admittedly done at a time when the accused had had little sleep, he was already awake when they approached him, and was told that he could stop at any time.
The mere failure to tell a suspect that the polygraph is inadmissible will not automatically produce an involuntary confession. Courts should engage in a two‑step process. First, the confession should be excluded if the police deception shocks the community. Second, even if not rising to that level, the use of deception is a relevant factor in the overall voluntariness analysis. Here, the police made it abundantly clear to the accused just what was admissible and what was not. The accused was not confused on this point. Further, although the police exaggerated the accuracy of the polygraph, merely confronting a suspect with adverse evidence ‑‑ even exaggerating its accuracy and reliability ‑‑ will not, standing alone, render a confession involuntary. Finally, tactical disadvantage to the defence when a suspect confesses following a polygraph is not relevant to the voluntariness of the confession; instead, if anything, it simply suggests prejudicial effect. However, given the immense probative value of a voluntary confession, exclusion is inappropriate.
Per Arbour J. (dissenting): There were improper inducements held out by police officers who interrogated the accused and these inducements, considered cumulatively and contextually in light of the “failed” polygraph test, require the exclusion of the accused’s statements. Moreover, the proximity and the causal connection between the “failed” polygraph test and the confession also compels this result. While some deference is always required because of the privileged position from which the trial judge assesses credibility, including on a voir dire, appeal courts must ensure that the question of the voluntariness of a confession was correctly decided by the trial judge, in accordance with the applicable law, and on a reasonable view of the facts.
The statements were obtained as a result of fear of prejudice or hope of advantage held out by persons in authority. Repeated threats and promises were made. They were often subtle, but against the backdrop of the polygraph procedure, they overwhelmed the free will of the accused. The overall interrogation strategy was sound, and although it relied on considerable deception on the part of the police, that in itself is neither illegal nor sufficient to vitiate the voluntary nature of a confession. The line is crossed when, as here, improper inducements are put forward by persons in authority in an oppressive atmosphere, undermining the interrogated person’s control over his mind and will. First, the promise of psychiatric help was improper. While it is true that the police did not explicitly tell the accused that the only way he could get psychiatric help was if he confessed, this was the clear implication of what was said. Second, after the accused’s initial confession, the police minimized the seriousness of the additional legal consequences that would flow from a confession to all the fires, suggesting to the accused that it made no difference whether he had set one or ten fires and that if he confessed, the other fires could be bundled with the car fire. This was clearly improper. Third, the threat to interrogate the accused’s fiancée clearly suggested that the accused could spare her his predicament by admitting his sole involvement in all the fires. The relationship between her and the accused was such that the threats to implicate her placed impermissible pressure on the accused to confess. The foregoing representations constituted threats, promises and inducements within the meaning of the confessions rule and, when combined with the prevalent ambiguity concerning just what was and was not admissible in court against the accused, as well as the oppressive atmosphere created by the “infallible” polygraph test, they are sufficient to raise a reasonable doubt as to voluntariness of the accused’s confessions. The combination of the lies and misrepresentations, which are not impermissible, with the inducements, which are, caused the accused to make involuntary admissions. At the very least, the trial judge had to have a reasonable doubt about the application of the classic voluntariness rule.
In addition, the statements are inadmissible on the basis that the manner in which they were obtained by the police places the accused in the unfair position of having to lead prejudicial, unreliable and inadmissible evidence against himself in order to impeach the veracity of the statements obtained. The admission into evidence of a confession, given in circumstances where it is intimately linked to a “failed” polygraph test, as it was in this case, is wholly inconsistent with the holding of this Court in Béland and severely, and unjustifiably, prejudices an accused’s fair trial interests. Since the confession and the polygraph are so intimately linked, the accused will unavoidably have to reveal having failed the polygraph if he wishes to cast doubt on the truthfulness of his confession. This comes at too high a cost to the accused’s fair trial interests. He is forced to incriminate himself by introducing an otherwise inadmissible piece of evidence which cannot help but strengthen what is often the sole evidence against him. Given the unparalleled weight attributed to confessions, the prejudicial effect that flows from an accused’s reference to his “failed” polygraph test is overwhelming. Therefore, confessions should be excluded where, as here, the accused, because of the intimate causal and temporal connection between a “failed” polygraph and a subsequent confession, is unable to demonstrate fully the impact of the circumstances surrounding its making without inevitably introducing the polygraph evidence.
By Iacobucci J.
Applied: R. v. Ewert,  3 S.C.R. 161; Ward v. The Queen,  2 S.C.R. 30; R. v. Fitton,  S.C.R. 958; R. v. Murakami,  S.C.R. 801; referred to: R. v. Nugent (1988), 84 N.S.R. (2d) 191; R. v. Hebert,  2 S.C.R. 151; Ibrahim v. The King,  A.C. 599; Prosko v. The King (1922), 63 S.C.R. 226; Boudreau v. The King,  S.C.R. 262; R. v. Wray,  S.C.R. 272; Rothman v. The Queen,  1 S.C.R. 640; Horvath v. The Queen,  2 S.C.R. 376; R. v. Whittle,  2 S.C.R. 914; Hobbins v. The Queen,  1 S.C.R. 553; R. v. Liew,  3 S.C.R. 227; R. v. Broyles,  3 S.C.R. 595; R. v. Stillman,  1 S.C.R. 607; R. v. Collins,  1 S.C.R. 265; R. v. Precourt (1976), 18 O.R. (2d) 714; R. v. Mills,  3 S.C.R. 668; R. v. Leipert,  1 S.C.R. 281; Reilly v. State, 355 A.2d 324 (1976); R. v. Kalashnikoff (1981), 57 C.C.C. (2d) 481; R. v. Lazure (1959), 126 C.C.C. 331; R. v. Ewert (1991), 68 C.C.C. (3d) 207; R. v. Jackson (1977), 34 C.C.C. (2d) 35; Commissioners of Customs and Excise v. Harz,  1 A.C. 760; R. v. Smith,  2 Q.B. 35; R. v. Desmeules,  R.L. 505; Comeau v. The Queen (1961), 131 C.C.C. 139; R. v. Hanlon (1958), 28 C.R. 398; R. v. Puffer (1976), 31 C.C.C. (2d) 81, aff’d  1 S.C.R. 321 (sub nom. McFall v. The Queen); R. v. Hayes (1982), 65 C.C.C. (2d) 294; R. v. Rennie (1981), 74 Cr. App. R. 207; R. v. Hoilett (1999), 136 C.C.C. (3d) 449; R. v. Owen (1983), 4 C.C.C. (3d) 538; R. v. Serack,  2 W.W.R. 377; R. v. Clot (1982), 69 C.C.C. (2d) 349; Blackburn v. Alabama, 361 U.S. 199 (1960); Schwartz v. Canada,  1 S.C.R. 254; Stein v. The Ship “Kathy K”,  2 S.C.R. 802; R. v. James, Ont. Ct. (Gen. Div.), January 25, 1991; R. v. Ollerhead (1990), 86 Nfld. & P.E.I.R. 38; R. v. Fowler (1979), 23 Nfld. & P.E.I.R. 255; R. v. Alexis (1994), 35 C.R. (4th) 117; R. v. Béland,  2 S.C.R. 398; R. v. Amyot (1990), 58 C.C.C. (3d) 312; R. v. Romansky (1981), 6 Man. R. (2d) 408; R. v. Barton (1993), 81 C.C.C. (3d) 574.
By Arbour J. (dissenting)
R. v. Fitton,  S.C.R. 958; R. v. Middleton (1974), 59 Cr. App. R. 18; R. v. Béland,  2 S.C.R. 398; Phillion v. The Queen,  1 S.C.R. 18; R. v. Marquard,  4 S.C.R. 223; R. v. Mohan,  2 S.C.R. 9; R. v. Amyot (1990), 58 C.C.C. (3d) 312; R. v. L.E.W.,  O.J. No. 753 (QL); DeClercq v. The Queen,  S.C.R. 902; R. v. Murray,  1 K.B. 391; R. v. Charrette,  O.J. No. 2509 (QL); R. v. Whalen,  O.J. No. 3488 (QL); Bigaouette v. The King (1926), 46 C.C.C. 311; R. v. Hodgson,  2 S.C.R. 449; R. v. Warickshall (1783), 1 Leach 263, 168 E.R. 234; R. v. Hardy (1794), 24 St. Tr. 199; R. v. Baldry (1852), 2 Den. 430, 169 E.R. 568; R. v. Guidice,  W.A.R. 128; R. v. Egger,  2 S.C.R. 451; R. v. Thorne (1988), 41 C.C.C. (3d) 344; R. v. McIntosh (1999), 141 C.C.C. (3d) 97; R. v. Terceira (1998), 123 C.C.C. (3d) 1, aff’d  3 S.C.R. 866; R. v. Nugent (1988), 84 N.S.R. (2d) 191.
Statutes and Regulations Cited
Police and Criminal Evidence Act 1984 (U.K.), 1984, c. 60, s. 76(8).
Bedau, Hugo Adam, and Michael L. Radelet. “Miscarriages of Justice in Potentially Capital Cases” (1987), 40 Stan. L. Rev. 21.
Furedy, John J. “The ‘control’ question ‘test’ (CQT) polygrapher’s dilemma: logico‑ethical considerations for psychophysiological practitioners and researchers” (1993), 15 Int. J. Psychophysiology 263.
Furedy, John J., and John Liss. “Countering Confessions Induced by the Polygraph: Of Confessionals and Psychological Rubber Hoses” (1986), 29 Crim. L.Q. 91.
Gudjonsson, Gisli H., and James A. C. MacKeith. “A Proven Case of False Confession: Psychological Aspects of the Coerced‑Compliant Type” (1990), 30 Med. Sci. & L. 329.
Gudjonsson, Gisli H., and James A. C. MacKeith. “Retracted Confessions: Legal, Psychological and Psychiatric Aspects” (1988), 28 Med. Sci. & L. 187.
Kassin, Saul M., and Lawrence S. Wrightsman. “Coerced Confessions, Judicial Instructions, and Mock Juror Verdicts” (1981), 11 J. Applied Soc. Psychol. 489.
Kaufman, Fred. The Admissibility of Confessions, 3rd ed. Toronto: Carswell, 1979.
Lykken, David T. A Tremor in the Blood: Uses and Abuses of the Lie Detector. New York: Plenum Trade, 1998.
Marin, René J. Admissibility of Statements, 9th ed. Aurora, Ont.: Canada Law Book, 1996 (loose‑leaf updated January 2000, release 4).
Ofshe, Richard J., and Richard A. Leo. “The Consequences of False Confessions: Deprivations of Liberty and Miscarriages of Justice in the Age of Psychological Interrogation” (1998), 88 J. Crim. L. & Criminology 429.
Ofshe, Richard J., and Richard A. Leo. “The Decision to Confess Falsely: Rational Choice and Irrational Action” (1997), 74 Denv. U. L. Rev. 979.
Ofshe, Richard J., and Richard A. Leo. “The Social Psychology of Police Interrogation: The Theory and Classification of True and False Confessions” (1997), 16 Stud. L. Pol. & Soc. 189.
Patrick, Christopher J., and William G. Iacono. “Validity of the Control Polygraph Test: The Problem of Sampling Bias” (1991), 76 J. App. Psych. 229.
Sopinka, John, Sidney N. Lederman and Alan W. Bryant. The Law of Evidence in Canada, 2nd ed. Toronto: Butterworths, 1999.
White,Welsh S. “False Confessions and the Constitution: Safeguards Against Untrustworthy Confessions” (1997), 32 Harv. C.R.–C.L. L. Rev. 105.
Wigmore, John Henry. Evidence in Trials at Common Law, vol. 3. Revised by James H. Chadbourn. Boston: Little, Brown, 1970.
APPEAL from a judgment of the Nova Scotia Court of Appeal (1998), 16 C.R. (5th) 29, 164 N.S.R. (2d) 342, 491 A.P.R. 342, 122 C.C.C. (3d) 506,  N.S.J. No. 19 (QL), allowing the accused’s appeal, setting aside the convictions and entering acquittals. Appeal allowed, Arbour J. dissenting.
William D. Delaney, for the appellant.
Arthur J. Mollon, Q.C., and Marian Mancini, for the respondent.
Gary T. Trotter, for the intervener the Attorney General for Ontario.
Michael Code and John Norris, for the intervener the Criminal Lawyers’ Association (Ontario).
The judgment of L’Heureux-Dubé, McLachlin, Iacobucci, Major, Bastarache and Binnie JJ. was delivered by
Iacobucci J. –
1 This appeal requires this Court to rule on the common law limits on police interrogation. Specifically, we are asked to decide whether the police improperly induced the respondent’s confessions through threats or promises, an atmosphere of oppression, or any other tactics that could raise a reasonable doubt as to the voluntariness of his confessions. I conclude that they did not. The trial judge’s determination that the confessions at stake in this appeal were voluntarily given should not have been disturbed on appeal, and accordingly the appeal should be allowed.
2 In this case, the police conducted a proper interrogation. Their questioning, while persistent and often accusatorial, was never hostile, aggressive, or intimidating. They repeatedly offered the accused food and drink. They allowed him to use the bathroom upon request. Before his first confession and subsequent arrest, they repeatedly told him that he could leave at any time. In this context, the alleged inducements offered by the police do not raise a reasonable doubt as to the confessions’ voluntariness. Nor do I find any fault with the role played by the polygraph test in this case. While the police admittedly exaggerated the reliability of such devices, the tactic of inflating the reliability of incriminating evidence is a common, and generally unobjectionable one. Whether standing alone, or in combination with the other mild inducements used in this appeal, it does not render the confessions involuntary.
3 The facts surrounding the respondent’s interrogation are obviously central to the resolution of this appeal, and I will refer to them throughout my legal analysis. At this point, I will simply give an overview.
4 Between February 5, 1994 and April 4, 1995, a series of eight fires involving four buildings and two motor vehicles occurred in and around the community of Waterville, Nova Scotia. Most of the incidents occurred between 1:00 a.m. and 4:00 a.m. The vehicle fires involved a van belonging to the respondent’s father, and a car belonging to the respondent’s fiancée, Tanya Kilcup. The building fires occurred relatively close to where the respondent had lived when the various fires occurred. The fires appeared to have been deliberately set, with the possible exception of Ms. Kilcup’s vehicle. The respondent was a member of the Waterville Volunteer Fire Brigade, and had responded to each of the fires in that capacity.
5 The last fire involved Ms. Kilcup’s vehicle. The car was parked in the driveway of the apartment building where the respondent and Ms. Kilcup lived. The fire was discovered by a passerby who extinguished it. The Fire Marshall investigated the fire and concluded that since the car was subject to a prior recall for a possible faulty ignition switch, the fire may have been accidental owing to an electrical fault.
6 The police also conducted an extensive investigation of the fires. To help narrow the list of possible suspects, they asked a total of seven or eight individuals to submit to polygraph tests. Five or six individuals did so, passed the test, and were effectively removed from the list of suspects. Another person had agreed to take a polygraph, but was not examined after the respondent confessed to the crimes. The respondent, after initial doubts, agreed to submit to a test. Around 3:00 p.m. on April 26, 1995, the respondent went to the Wandlyn Motel for the test, according to a prior arrangement. The police audiotaped the events at the motel. Sergeant Taker administered the polygraph test. The respondent was fully advised of his rights to silence, to a lawyer (including the availability of Legal Aid), and to leave at any time. Sergeant Taker also advised him that while Sergeant Taker’s interpretation of the polygraph results was not admissible, anything said by the respondent was admissible. The respondent was given a pamphlet to review, which discussed the polygraph procedures, and he signed a consent form.
7 Before conducting the test itself, Sergeant Taker conducted a lengthy “pre-test” interview, which involved a variety of questions, many of them personal in nature. This interview was designed to provide a basis for the polygraph test itself, to help Sergeant Taker compose “control questions” for the polygraph exam, and to foster a sense of intimacy between examiner and subject. An exculpatory statement, which formed the basis for the polygraph test itself, was taken at the conclusion of the pre-test. Sergeant Taker then conducted the polygraph exam, which lasted only a matter of minutes. During the test Sergeant Taker did not ask about any specific fire, but instead asked if the respondent’s earlier statement had been truthful. At the conclusion of the test, around 5:00 p.m., Taker checked the charts and informed the respondent that he had failed the test. He reminded the respondent that his rights were still in effect, and proceeded to question him for approximately one hour. At one point the respondent asked “What if I admit to the car? ... Then I can walk out of here and it’s over.” Though Sergeant Taker replied “You can walk out at any time”, the respondent did not leave.
8 At 6:30 p.m. Sergeant Taker was relieved by Corporal Deveau, who reminded the respondent of his right to counsel. After 30 to 40 minutes, the respondent confessed to setting fire to his fiancée’s car. He appeared emotionally distraught at this time. After a recitation of his rights, and an acknowledgement that he understood them, the police took a written statement, in which he continued to deny any involvement in the other fires. The respondent was arrested, warned of his right to counsel, given the secondary police warning, and driven to the police station at 8:15 p.m. En route he was very upset and was crying. He was placed in an interview room equipped with videotaping facilities, which recorded the subsequent interrogation where Corporal Deveau questioned him about the other fires. Around 8:30 p.m. and 9:15 p.m. the respondent indicated that he was tired, and wanted to go home to bed. He was informed that he was under arrest, and he could call a lawyer if he wanted, but that he could not go home. Questioning did not cease.
9 Constable Bogle took over the interrogation at 9:52 p.m., after giving the respondent the secondary police warning. Constable Bogle questioned the respondent until about 11:00 p.m., at which time the respondent confessed to setting seven of the eight fires. He denied any involvement in the fire in his father’s van. At this time, Constable Bogle left the room, and the respondent was seen crying with his head in his hands. Constable Bogle returned with Corporal Deveau, and took a written statement. The respondent’s Charter rights and the police warning were on the statement, and were acknowledged by the respondent. The police warning stated that “[y]ou need not say anything. You have nothing to hope from any promise or favour and nothing to fear from any threat whether or not you do say anything. Anything you do say may be used as evidence.” The statement concluded at 1:10 a.m. on April 27. After the police attended to various administrative tasks, the respondent was placed in a cell to sleep at 2:45 a.m. At 6:00 a.m., Corporal Deveau noticed that the respondent was awake and asked whether he would agree to a re-enactment. On the tape of the re-enactment, the respondent was given a Charter warning, the secondary warning, and was advised that he could stop the re-enactment at any time. The police drove the respondent around Waterville to the various fire scenes, where he described how he had set each fire. The respondent was charged with seven counts of arson.
10 At trial, the trial judge held a voir dire to determine the admissibility of the respondent’s statements, including the video re-enactment. The trial judge ruled that the statements were voluntary and admissible, and subsequently convicted him on all counts. However, the Nova Scotia Court of Appeal found that the statements were involuntary and thus inadmissible, and allowed the respondent’s appeal. The Court of Appeal excluded the confessions, overturned the convictions, and entered acquittals.
III. Judicial Decisions
A. Nova Scotia Provincial Court
11 MacDonald Prov. Ct. J. concluded that the respondent’s confessions were voluntary. First of all, there was no evidence that the respondent did not understand the police warnings. After reviewing the law, the trial judge concluded that the respondent had an operating mind, and was aware of the consequences of his actions. The respondent “appeared to be coherent, capable of understanding the questions that were put to him and to be in complete and total control of his faculties”. Nor did the police unfairly deprive the respondent of his right to choose whether or not he wished to speak to the officers.
12 The trial judge also addressed the relevance of the polygraph test under the Nova Scotia Supreme Court – Appeal Division’s decision in R. v. Nugent (1988), 84 N.S.R. (2d) 191. MacDonald Prov. Ct. J. found that the respondent clearly understood that while the results were not admissible, his statements were. There being no confusion over the role of the polygraph, MacDonald Prov. Ct. J. concluded that the use of a polygraph did not render the ensuing confessions involuntary or otherwise inadmissible.
13 Finally, the trial judge addressed the respondent’s argument that the re-enactment was not voluntary because he was sleep-deprived at the time. The trial judge noted that the respondent appeared in full control of his faculties on the tape of the re-enactment. He was coherent. He understood what was being asked of him. There was no evidence to suggest that the re-enactments were anything but voluntary. MacDonald Prov. Ct. J. therefore admitted the statements and re-enactment into evidence.
B. Nova Scotia Court of Appeal (1998), 164 N.S.R. (2d) 342 (Pugsley and Cromwell
JJ.A., Flinn J.A. concurring)
14 Pugsley and Cromwell JJ.A. recognized that their role as an appellate court was not to retry the case or overturn the trial judge’s findings of fact. The decision of the trial judge should not be set aside unless she made a wrong assessment of the evidence, failed to consider the relevant circumstances, or failed to apply the correct legal principle.
15 The court first addressed the admissibility of the respondent’s initial confession to lighting the fire in Ms. Kilcup’s car. Confessions will only be admissible if the Crown proves, beyond a reasonable doubt, that they were made voluntarily. A statement will be involuntary if it is the result of either “fear of prejudice” or “hope of advantage” held out by persons in authority. Vigorous and skillful questioning, misstatements of fact by the police, and appeals to the conscience of the accused do not necessarily make a resulting statement inadmissible. The statement will not be excluded simply because the accused believes it will be to his or her advantage to confess. It is only when this belief is induced or confirmed by persons in authority that the statements should be excluded.
16 Turning to the facts of this case, the court found that the police did not advise the respondent that the results of the polygraph were not admissible as evidence. Sergeant Taker should also have advised the respondent that the polygraph was not infallible. The accused’s ensuing confusion about the significance of his failure of the test was an important aspect of the overall context that had to be considered in deciding whether the statements were voluntary.
17 Against the backdrop of the polygraph examination, the police questioning developed other themes that, in light of all the other circumstances, constituted improper inducements to confess. While any one of these might not be sufficient to produce doubt about the voluntariness of the statements, their cumulative effect was overwhelming. Shortly after the polygraph results were given to the respondent, Sergeant Taker suggested that confessing would prevent him from getting into any more trouble than he was already in, and that confessing would allow him to get the help he needed if he had done these things. The police also minimized the seriousness of the offence, and warned that they might have to interrogate Ms. Kilcup. These actions, where they result in a confession, are improper inducements.
18 Another factor was the police’s abuse of their relationship with the respondent. After gaining the respondent’s trust, Sergeant Taker and Corporal Deveau breached that trust by unfairly and aggressively exploiting the results of the test as a confession-inducing instrument. Immediately before the first confession, Deveau said to the respondent that he was probably his “best friend right now”. This statement was an abuse of confidence.
19 The court also disagreed with the trial judge’s conclusion that the respondent appeared coherent and in control of his faculties on the videotape. The court felt that the transcript of the audiotape revealed that the respondent did not appreciate the role the equipment played in the process, and the separate role played by Sergeant Taker in his role as interpreter of the result. The conclusions of the trial judge failed to take into account several comments of Sergeant Taker wherein he assured the respondent that the polygraph machine was reliable.
20 Although none of the inducements on their own would require exclusion, taken cumulatively and in light of the use of the polygraph test, the statements were clearly involuntary and should have been excluded.
21 The court allowed the appeal. Since the statements were the only evidence that directly implicated the respondent, and without them, no reasonable trier of fact could convict, the convictions were set aside and acquittals entered.
A. Standard of Review for Voluntariness
22 While determining the appropriate legal test is of course a question of law, applying this test to determine whether or not a confession is voluntary is a question of fact, or of mixed law and fact. See R. v. Ewert,  3 S.C.R. 161, at p. 161; Ward v. The Queen,  2 S.C.R. 30, at p. 42 (per Spence J.); R. v. Fitton,  S.C.R. 958, at pp. 983-84 (per Fauteux J.); R. v. Murakami,  S.C.R. 801, at p. 803 (per Rand J., Locke J. concurring). Therefore, as this Court held in Ewert, a disagreement with the trial judge regarding the weight to be given various pieces of evidence is not grounds to reverse a finding on voluntariness. Respectfully, I believe that the Court of Appeal did just that. Therefore, following Ewert, the appeal must be allowed.
23 While the foregoing might suffice technically to dispose of this appeal, I believe it is important to take this opportunity to set out the proper scope of the confessions rule. There was much argument among the parties and interveners in this appeal on this point, and this Court has not directly addressed the issue since the introduction of the Canadian Charter of Rights and Freedoms. Because of this lack of clarity, it has been often difficult to discern in various cases what standards have been applied. In addition, several arguments not addressed by the trial judge were raised before our Court. It is therefore necessary to broaden the discussion to deal with these issues.
B. The Development of the Confessions Rule
1. Two Elements of the Rule
24 As indicated by McLachlin J. (as she then was), in R. v. Hebert,  2 S.C.R. 151, there are two main strands to this Court’s jurisprudence under the confessions rule. One approach is narrow, excluding statements only where the police held out explicit threats or promises to the accused. The definitive statement of this approach came in Ibrahim v. The King,  A.C. 599 (P.C.), at p. 609:
It has long been established as a positive rule of English criminal law, that no statement by an accused is admissible in evidence against him unless it is shewn by the prosecution to have been a voluntary statement, in the sense that it has not been obtained from him either by fear of prejudice or hope of advantage exercised or held out by a person in authority.
This Court adopted the “Ibrahim rule” in Prosko v. The King (1922), 63 S.C.R. 226, and subsequently applied it in cases like Boudreau v. The King,  S.C.R. 262, Fitton, supra, R. v. Wray,  S.C.R. 272, and Rothman v. The Queen,  1 S.C.R. 640.
25 The Ibrahim rule gives the accused only “a negative right — the right not to be tortured or coerced into making a statement by threats or promises held out by a person who is and whom he subjectively believes to be a person in authority”: Hebert, supra, at p. 165. However, Hebert also recognized a second, “much broader” approach, according to which “[t]he absence of violence, threats and promises by the authorities does not necessarily mean that the resulting statement is voluntary, if the necessary mental element of deciding between alternatives is absent” (p. 166).
26 While not always followed, McLachlin J. noted at p. 166 that this aspect of the confessions rule “persists as part of our fundamental notion of procedural fairness”. This approach is most evident in the so-called “operating mind” doctrine, developed by this Court in Ward, supra, Horvath v. The Queen,  2 S.C.R. 376, and R. v. Whittle,  2 S.C.R. 914. In those cases the Court made “a further investigation of whether the statements were freely and voluntarily made even if no hope of advantage or fear of prejudice could be found”: Ward, supra, at p. 40. The “operating mind” doctrine dispelled once and for all the notion that the confessions rule is concerned solely with whether or not the confession was induced by any threats or promises.
27 These cases focused not just on reliability, but on voluntariness conceived more broadly. None of the reasons in Ward or Horvath ever expressed any doubts about the reliability of the confessions in issue. Instead, they focused on the lack of voluntariness, whether the cause was shock (Ward), hypnosis (Horvath, per Beetz J.), or “complete emotional disintegration” (Horvath, supra, at p. 400, per Spence J.). Similarly, in Hobbins v. The Queen,  1 S.C.R. 553, at pp. 556-57, Laskin C.J. noted that in determining the voluntariness of a confession, courts should be alert to the coercive effect of an “atmosphere of oppression”, even though there was “no inducement held out of hope of advantage or fear of prejudice, and absent any threats of violence or actual violence”; see also R. v. Liew,  3 S.C.R. 227, at para. 37. Clearly, the confessions rule embraces more than the narrow Ibrahim formulation; instead, it is concerned with voluntariness, broadly understood.
2. The Charter Era
28 The Charter constitutionalized a new set of protections for accused persons, contained principally in ss. 7 to 14 thereof. The entrenchment of these rights answered certain questions that had once been asked under the aegis of the confessions rule. For example, while the confessions rule did not exclude statements elicited by undercover officers in jail cells (Rothman, supra), such confessions can violate the Charter: see Hebert, supra, and R. v. Broyles,  3 S.C.R. 595.
29 In Hebert, supra, McLachlin J. interpreted the right to silence in light of existing common law protections, such as the confessions rule. However, given the focus of that decision on defining constitutional rights, it did not decide the inverse question: namely, the scope of the common law rules in light of the Charter. One possible view is that the Charter subsumes the common law rules.
30 But I do not believe that this view is correct, for several reasons. First, the confessions rule has a broader scope than the Charter. For example, the protections of s. 10 only apply “on arrest or detention”. By contrast, the confessions rule applies whenever a person in authority questions a suspect. Second, the Charter applies a different burden and standard of proof from that under the confessions rule. Under the former, the burden is on the accused to show, on a balance of probabilities, a violation of constitutional rights. Under the latter, the burden is on the prosecution to show beyond a reasonable doubt that the confession was voluntary. Finally, the remedies are different. The Charter excludes evidence obtained in violation of its provisions under s. 24(2) only if admitting the evidence would bring the administration of justice into disrepute: see R. v. Stillman,  1 S.C.R. 607, R. v. Collins,  1 S.C.R. 265, and the related jurisprudence. By contrast, a violation of the confessions rule always warrants exclusion.
31 These various differences illustrate that the Charter is not an exhaustive catalogue of rights. Instead, it represents a bare minimum below which the law must not fall. A necessary corollary of this statement is that the law, whether by statute or common law, can offer protections beyond those guaranteed by the Charter. The common law confessions rule is one such doctrine, and it would be a mistake to confuse it with the protections given by the Charter. While obviously it may be appropriate, as in Hebert, supra, to interpret one in light of the other, it would be a mistake to assume one subsumes the other entirely.
C. The Confessions Rule Today
32 As previously mentioned, this Court has not recently addressed the precise scope of the confessions rule. Instead, we have refined several elements of the rule, without ever integrating them into a coherent whole. I believe it is important to restate the rule for two reasons. First is the continuing diversity of approaches as evidenced by the courts below in this appeal. Second, and perhaps more important, is our growing understanding of the problem of false confessions. As I will discuss below, the confessions rule is concerned with voluntariness, broadly defined. One of the predominant reasons for this concern is that involuntary confessions are more likely to be unreliable. The confessions rule should recognize which interrogation techniques commonly produce false confessions so as to avoid miscarriages of justice.
33 In defining the confessions rule, it is important to keep in mind its twin goals of protecting the rights of the accused without unduly limiting society’s need to investigate and solve crimes. Martin J.A. accurately delineated this tension in R. v. Precourt (1976), 18 O.R. (2d) 714 (C.A.), at p. 721:
Although improper police questioning may in some circumstances infringe the governing [confessions] rule it is essential to bear in mind that the police are unable to investigate crime without putting questions to persons, whether or not such persons are suspected of having committed the crime being investigated. Properly conducted police questioning is a legitimate and effective aid to criminal investigation. . . . On the other hand, statements made as the result of intimidating questions, or questioning which is oppressive and calculated to overcome the freedom of will of the suspect for the purpose of extracting a confession are inadmissible. . . .
All who are involved in the administration of justice, but particularly courts applying the confessions rule, must never lose sight of either of these objectives.
1. The Problem of False Confessions
34 The history of police interrogations is not without its unsavoury chapters. Physical abuse, if not routine, was certainly not unknown. Today such practices are much less common. In this context, it may seem counterintuitive that people would confess to a crime that they did not commit. And indeed, research with mock juries indicates that people find it difficult to believe that someone would confess falsely. See S. M. Kassin and L. S. Wrightsman, “Coerced Confessions, Judicial Instructions, and Mock Juror Verdicts” (1981), 11 J. Applied Soc. Psychol. 489.
35 However, this intuition is not always correct. A large body of literature has developed documenting hundreds of cases where confessions have been proven false by DNA evidence, subsequent confessions by the true perpetrator, and other such independent sources of evidence. See, e.g., R. A. Leo and R. J. Ofshe, “The Consequences of False Confessions: Deprivations of Liberty and Miscarriages of Justice in the Age of Psychological Interrogation” (1998), 88 J. Crim. L. & Criminology 429 (hereinafter Leo & Ofshe (1998)); R. J. Ofshe and R. A. Leo, “The Social Psychology of Police Interrogation: The Theory and Classification of True and False Confessions” (1997), 16 Stud. L. Pol. & Soc. 189 (hereinafter Ofshe & Leo (1997)); R. J. Ofshe and R. A. Leo, “The Decision to Confess Falsely: Rational Choice and Irrational Action” (1997), 74 Denv. U. L. Rev. 979 (hereinafter Ofshe & Leo (1997a)); W. S. White, “False Confessions and the Constitution: Safeguards Against Untrustworthy Confessions” (1997), 32 Harv. C.R.-C.L. L. Rev. 105; G. H. Gudjonsson and J. A. C. MacKeith, “A Proven Case of False Confession: Psychological Aspects of the Coerced-Compliant Type” (1990), 30 Med. Sci. & L. 329 (hereinafter Gudjonsson & MacKeith (1990)); G. H. Gudjonsson and J. A. C. MacKeith, “Retracted Confessions: Legal, Psychological and Psychiatric Aspects” (1988), 28 Med. Sci. & L. 187 (hereinafter Gudjonsson & MacKeith (1988)); H. A. Bedau and M. L. Radelet, “Miscarriages of Justice in Potentially Capital Cases” (1987), 40 Stan. L. Rev. 21.
36 One of the overriding concerns of the criminal justice system is that the innocent must not be convicted: see, e.g., R. v. Mills,  3 S.C.R. 668, at para. 71; R. v. Leipert,  1 S.C.R. 281, at para. 4. Given the important role of false confessions in convicting the innocent, the confessions rule must understand why false confessions occur. Without suggesting that any confession involving elements discussed below should automatically be excluded, I hope to provide a background for my synthesis of the confessions rule in the next section.
37 Ofshe & Leo (1997), supra, at p. 210, provide a useful taxonomy of false confessions. They suggest that there are five basic kinds: voluntary, stress-compliant, coerced-compliant, non-coerced-persuaded, and coerced-persuaded. Voluntary confessions ex hypothesi are not the product of police interrogation. It is therefore the other four types of false confessions that are of interest.
38 According to Ofshe & Leo (1997), supra, at p. 211, stress-compliant confessions occur “when the aversive interpersonal pressures of interrogation become so intolerable that [suspects] comply in order to terminate questioning”. They are elicited by “exceptionally strong use of the aversive stressors typically present in interrogations”, and are “given knowingly in order to escape the punishing experience of interrogation” (emphasis in original). See also Gudjonsson & MacKeith (1990), supra. Another important factor is confronting the suspect with fabricated evidence in order to convince him that protestations of innocence are futile: see ibid.; Ofshe & Leo (1997a), supra, at p. 1040.
39 Somewhat different are coerced-compliant confessions. These confessions are the product of “the classically coercive influence techniques (e.g., threats and promises)”, with which the Ibrahim rule is concerned: Ofshe & Leo (1997), supra, at p. 214. As Gudjonsson & MacKeith (1988), supra, suggest at p. 191, “most cases of false confession that come before the courts are of the compliant-coerced type”. See also White, supra, at p. 131.
40 A third kind of false confession is the non-coerced-persuaded confession. In this scenario, police tactics cause the innocent person to “become confused, doubt his memory, be temporarily persuaded of his guilt and confess to a crime he did not commit”: Ofshe & Leo (1997), supra, at p. 215. For an example, see Reilly v. State, 355 A.2d 324 (Conn. Super. Ct. 1976); Ofshe & Leo (1997), supra, at pp. 231-34. The use of fabricated evidence can also help convince an innocent suspect of his or her own guilt.
41 A final type of false confession is the coerced-persuaded confession. This is like the non-coerced-persuaded, except that the interrogation also involves the classically coercive aspects of the coerced-compliant confession: see Ofshe & Leo (1997), supra, at p. 219.
42 From this discussion, several themes emerge. One is the need to be sensitive to the particularities of the individual suspect. For example, White, supra, at p. 120, notes the following:
False confessions are particularly likely when the police interrogate particular types of suspects, including suspects who are especially vulnerable as a result of their background, special characteristics, or situation, suspects who have compliant personalities, and, in rare instances, suspects whose personalities make them prone to accept and believe police suggestions made during the course of the interrogation.
And indeed, this is consistent with the reasons of Rand J. in Fitton, supra, at p. 962:
The strength of mind and will of the accused, the influence of custody or its surroundings, the effect of questions or of conversation, all call for delicacy in appreciation of the part they have played behind the admission, and to enable a Court to decide whether what was said was freely and voluntarily said, that is, was free from the influence of hope or fear aroused by them.
Ward, supra, and Horvath, supra, similarly recognized the particular circumstances of the suspects that rendered them unable to confess voluntarily: in Ward, the accused’s state of shock, and in Horvath, the psychological fragility that precipitated his hypnosis and “complete emotional disintegration” (p. 400).
43 Another theme is the danger of using non-existent evidence. Presenting a suspect with entirely fabricated evidence has the potential either to persuade the susceptible suspect that he did indeed commit the crime, or at least to convince the suspect that any protestations of innocence are futile.
44 Finally, the literature bears out the common law confessions rule’s emphasis on threats and promises. Coerced-compliant confessions are the most common type of false confessions. These are classically the product of threats or promises that convince a suspect that in spite of the long-term ramifications, it is in his or her best interest in the short- and intermediate-term to confess.
45 Fortunately, false confessions are rarely the product of proper police techniques. As Leo & Ofshe (1998), supra, point out at p. 492, false confession cases almost always involve “shoddy police practice and/or police criminality”. Similarly, in Ofshe & Leo (1997), supra, at pp. 193-96, they argue that in most cases, “eliciting a false confession takes strong incentives, intense pressure and prolonged questioning. . . . Only under the rarest of circumstances do an interrogator’s ploys persuade an innocent suspect that he is in fact guilty and has been caught.”
46 Before turning to how the confessions rule responds to these dangers, I would like to comment briefly on the growing practice of recording police interrogations, preferably by videotape. As pointed out by J. J. Furedy and J. Liss in “Countering Confessions Induced by the Polygraph: Of Confessionals and Psychological Rubber Hoses” (1986), 29 Crim. L.Q. 91, at p. 104, even if “notes were accurate concerning the content of what was said . . ., the notes cannot reflect the tone of what was said and any body language that may have been employed” (emphasis in original). White, supra, at pp. 153-54, similarly offers four reasons why videotaping is important:
First, it provides a means by which courts can monitor interrogation practices and thereby enforce the other safeguards. Second, it deters the police from employing interrogation methods likely to lead to untrustworthy confessions. Third, it enables courts to make more informed judgments about whether interrogation practices were likely to lead to an untrustworthy confession. Finally, mandating this safeguard accords with sound public policy because the safeguard will have additional salutary effects besides reducing untrustworthy confessions, including more net benefits for law enforcement.
This is not to suggest that non-recorded interrogations are inherently suspect; it is simply to make the obvious point that when a recording is made, it can greatly assist the trier of fact in assessing the confession.
2. The Contemporary Confessions Rule
47 The common law confessions rule is well-suited to protect against false confessions. While its overriding concern is with voluntariness, this concept overlaps with reliability. A confession that is not voluntary will often (though not always) be unreliable. The application of the rule will by necessity be contextual. Hard and fast rules simply cannot account for the variety of circumstances that vitiate the voluntariness of a confession, and would inevitably result in a rule that would be both over- and under-inclusive. A trial judge should therefore consider all the relevant factors when reviewing a confession.
(a) Threats or Promises
48 This is of course the core of the confessions rule from Ibrahim, supra. It is therefore important to define precisely what types of threats or promises will raise a reasonable doubt as to the voluntariness of a confession. While obviously imminent threats of torture will render a confession inadmissible, most cases will not be so clear.
49 As noted above, in Ibrahim the Privy Council ruled that statements would be inadmissible if they were the result of “fear of prejudice or hope of advantage”. The classic “hope of advantage” is the prospect of leniency from the courts. It is improper for a person in authority to suggest to a suspect that he or she will take steps to procure a reduced charge or sentence if the suspect confesses. Therefore in Nugent, supra, the court excluded the statement of a suspect who was told that if he confessed, the charge could be reduced from murder to manslaughter. See also R. v. Kalashnikoff (1981), 57 C.C.C. (2d) 481 (B.C.C.A.); R. v. Lazure (1959), 126 C.C.C. 331 (Ont. C.A.); R. J. Marin, Admissibility of Statements (9th ed. (loose-leaf)), at p. 1–15. Intuitively implausible as it may seem, both judicial precedent and academic authority confirm that the pressure of intense and prolonged questioning may convince a suspect that no one will believe his or her protestations of innocence, and that a conviction is inevitable. In these circumstances, holding out the possibility of a reduced charge or sentence in exchange for a confession would raise a reasonable doubt as to the voluntariness of any ensuing confession. An explicit offer by the police to procure lenient treatment in return for a confession is clearly a very strong inducement, and will warrant exclusion in all but exceptional circumstances.
50 Another type of inducement relevant to this appeal is an offer of psychiatric assistance or other counselling for the suspect in exchange for a confession. While this is clearly an inducement, it is not as strong as an offer of leniency and regard must be had to the entirety of the circumstances. A good example of this comes from R. v. Ewert (1991), 68 C.C.C. (3d) 207 (B.C.C.A.). In that case, the police made what Hinkson J.A. at the Court of Appeal described as a “bold offer to the accused to help him, in the sense of providing psychiatric help, if he told them what had happened” (p. 216). Reversing the Court of Appeal, this Court upheld the trial judge’s conclusion that, while the police conduct was an inducement, it was not a factor in the suspect’s decision to confess. Ewert thus recognizes the importance of a contextual approach.
51 Threats or promises need not be aimed directly at the suspect for them to have a coercive effect. For example, in R. v. Jackson (1977), 34 C.C.C. (2d) 35 (B.C.C.A.), McIntyre J.A. (as he then was) addressed a confession obtained in a case where the accused and his friend Winn had robbed and murdered a hitchhiker. The police suspected the murder was Jackson’s doing, and urged him to confess, lest his friend Winn be unjustly convicted of murder. The trial judge had concluded that:
[The police] were exerting a subtle form of pressure on Jackson, they were appealing to his concept of right and wrong. . . . They indicated that unless they got to the truth of the matter, it might be necessary to charge both, and this too was a very likely possibility. The officers were completely frank with him. The officers hoped that when Jackson was faced with what they had, and what might transpire if he didn’t speak up, that he would take Winn off the hook and confess. That is exactly what he did. I can see nothing in what they said or in what they did that can be construed by Jackson as holding out the possibility of any benefit to him should he confess.
McIntyre J.A. agreed that no hope of advantage that would render a confession inadmissible had been held out to the accused. He then presented the following very
helpful analysis of the law (at p. 38):
[Cases] must be considered in relation to their own facts. It is my opinion that for a promised benefit to a person other than the accused to vitiate a confession, the benefit must be of such a nature that when considered in the light of the relationship between the person and the accused, and all the surrounding circumstances of the confession, it would tend to induce the accused to make an untrue statement, for it is the danger that a person may be induced by promises to make such a statement which lies at the root of this exclusionary rule.
52 McIntyre J.A. offered, as examples of improper inducements, telling a mother that her daughter would not be charged with shoplifting if the mother confessed to a similar offence (see Commissioners of Customs and Excise v. Harz,  1 A.C. 760 (H.L.), at p. 821), or a sergeant-major keeping a company on parade until he learned who was responsible for a stabbing (see R. v. Smith,  2 Q.B. 35). In Jackson, by contrast, the accused had merely known Winn for a year in prison. The offence occurred a few days after their release. Neither testified to a relationship such that “the immunity of one was of such vital concern to the other that he would untruthfully confess to preserve it” (p. 39). The confession was therefore admissible.
53 The Ibrahim rule speaks not only of “hope of advantage”, but also of “fear of prejudice”. Obviously, any confession that is the product of outright violence is involuntary and unreliable, and therefore inadmissible. More common, and more challenging judicially, are the more subtle, veiled threats that can be used against suspects. The Honourable Fred Kaufman, in the third edition of The Admissibility of Confessions (1979), at p. 230, provides a useful starting point:
Threats come in all shapes and sizes. Among the most common are words to the effect that “it would be better” to tell, implying thereby that dire consequences might flow from a refusal to talk. Maule J. recognized this fact, and said that “there can be no doubt that such words, if spoken by a competent person, have been held to exclude a confession at least 500 times” (R. v. Garner (1848), 3 Cox C.C. 175, at p. 177).
Courts have accordingly excluded confessions made in response to police suggestions that it would be better if they confessed. See R. v. Desmeules,  R.L. 505 (Que. Ct. Sess. P.); Comeau v. The Queen (1961), 131 C.C.C. 139 (N.S.S.C.); Lazure, supra; R. v. Hanlon (1958), 28 C.R. 398 (Nfld. C.A.), at p. 401; White, supra, at p. 129.
54 However, phrases like “it would be better if you told the truth” should not automatically require exclusion. Instead, as in all cases, the trial judge must examine the entire context of the confession, and ask whether there is a reasonable doubt that the resulting confession was involuntary. Freedman C.J.M. applied this approach correctly in R. v. Puffer (1976), 31 C.C.C. (2d) 81 (Man. C.A.). In that case a suspect in a robbery and murder asked to meet with two police officers of his acquaintance. At this meeting, one officer said: “The best thing you can do is come in with us and tell the truth” (p. 95). Freedman C.J.M. held that while the officer’s language was “unfortunate”, it did not require exclusion (at p. 95): “McFall wanted to talk, he wanted to give the police his version of what had occurred, and above all he did not want Puffer and Kizyma to get away, leaving him to face the music alone” (emphasis in original).
55 In his reasons, Freedman C.J.M. referred to a passage from an article he had written earlier, “Admissions and Confessions”, published in Salhany and Carter, eds., Studies in Canadian Criminal Evidence (1972), at pp. 110-11, where he stated the following:
Risky though it be for a policeman to use words like “better tell us everything”— and an experienced and conscientious officer will shun them like the plague — their consequences will not always be fatal. There have been some instances where words of that type have been employed, and yet a confession following thereon has been admitted. That may occur when the court is satisfied that the offending words, potentially perilous though they be, did not in fact induce the accused to speak. In other words, he would have confessed in any event, the court’s enquiry on the point establishing that his statement was indeed voluntarily made. It is scarcely necessary to emphasize, however, that cases of the kind just mentioned will confront a prosecuting counsel with special difficulty. For words like “better tell the truth” carry the mark of an inducement on their very face, and a resultant confession may well find itself battling against the stream.
This Court upheld the Court of Appeal’s ruling. See McFall v. The Queen,  1 S.C.R. 321; see also R. v. Hayes (1982), 65 C.C.C. (2d) 294 (Alta. C.A.), at pp. 296-97. I agree that “it would be better” comments require exclusion only where the circumstances reveal an implicit threat or promise.
56 A final threat or promise relevant to this appeal is the use of moral or spiritual inducements. These inducements will generally not produce an involuntary confession, for the very simple reason that the inducement offered is not in the control of the police officers. If a police officer says “If you don’t confess, you’ll spend the rest of your life in jail. Tell me what happened and I can get you a lighter sentence”, then clearly there is a strong, and improper, inducement for the suspect to confess. The officer is offering a quid pro quo, and it raises the possibility that the suspect is confessing not because of any internal desire to confess, but merely in order to gain the benefit offered by the interrogator. By contrast, with most spiritual inducements the interrogator has no control over the suggested benefit. If a police officer convinces a suspect that he will feel better if he confesses, the officer has not offered anything. I therefore agree with Kaufman, supra, who summarized the jurisprudence as follows at p. 186:
We may therefore conclude that, as a general rule, confessions which result from spiritual exhortations or appeals to conscience and morality, are admissible in evidence, whether urged by a person in authority or by someone else. [Emphasis in original.]
57 In summary, courts must remember that the police may often offer some kind of inducement to the suspect to obtain a confession. Few suspects will spontaneously confess to a crime. In the vast majority of cases, the police will have to somehow convince the suspect that it is in his or her best interests to confess. This becomes improper only when the inducements, whether standing alone or in combination with other factors, are strong enough to raise a reasonable doubt about whether the will of the subject has been overborne. On this point I found the following passage from R. v. Rennie (1981), 74 Cr. App. R. 207 (C.A.), at p. 212, particularly apt:
Very few confessions are inspired solely by remorse. Often the motives of an accused are mixed and include a hope that an early admission may lead to an earlier release or a lighter sentence. If it were the law that the mere presence of such a motive, even if promoted by something said or done by a person in authority, led inexorably to the exclusion of a confession, nearly every confession would be rendered inadmissible. This is not the law. In some cases the hope may be self-generated. If so, it is irrelevant, even if it provides the dominant motive for making the confession. In such a case the confession will not have been obtained by anything said or done by a person in authority. More commonly the presence of such a hope will, in part at least, owe its origin to something said or done by such a person. There can be few prisoners who are being firmly but fairly questioned in a police station to whom it does not occur that they might be able to bring both their interrogation and their detention to an earlier end by confession.
The most important consideration in all cases is to look for a quid pro quo offer by interrogators, regardless of whether it comes in the form of a threat or a promise.
58 There was much debate among the parties, interveners, and courts below over the relevance of “oppression” to the confessions rule. Oppression clearly has the potential to produce false confessions. If the police create conditions distasteful enough, it should be no surprise that the suspect would make a stress-compliant confession to escape those conditions. Alternately, oppressive circumstances could overbear the suspect’s will to the point that he or she comes to doubt his or her own memory, believes the relentless accusations made by the police, and gives an induced confession.
59 A compelling example of oppression comes from the Ontario Court of Appeal’s recent decision in R. v. Hoilett (1999), 136 C.C.C. (3d) 449. The accused, charged with sexual assault, was arrested at 11:25 p.m. while under the influence of crack cocaine and alcohol. After two hours in a cell, two officers removed his clothes for forensic testing. He was left naked in a cold cell containing only a metal bunk to sit on. The bunk was so cold he had to stand up. One and one-half hours later, he was provided with some light clothes, but no underwear and ill-fitting shoes. Shortly thereafter, at about 3:00 a.m., he was awakened for the purpose of interviewing. In the course of the interrogation, the accused nodded off to sleep at least five times. He requested warmer clothes and a tissue to wipe his nose, both of which were refused. While he admitted knowing that he did not have to talk, and that the officers had made no explicit threats or promises, he hoped that if he talked to the police they would give him some warm clothes and cease the interrogation.
60 Under these circumstances, it is no surprise that the Court of Appeal concluded the statement was involuntary. Under inhumane conditions, one can hardly be surprised if a suspect confesses purely out of a desire to escape those conditions. Such a confession is not voluntary. For similar examples of oppressive circumstances, see R. v. Owen (1983), 4 C.C.C. (3d) 538 (N.S.S.C., App. Div.); R. v. Serack,  2 W.W.R. 377 (B.C.S.C.). Without trying to indicate all the factors that can create an atmosphere of oppression, such factors include depriving the suspect of food, clothing, water, sleep, or medical attention; denying access to counsel; and excessively aggressive, intimidating questioning for a prolonged period of time.
61 A final possible source of oppressive conditions is the police use of non-existent evidence. As the discussion of false confessions, supra, revealed, this ploy is very dangerous: see Ofshe & Leo (1997a), supra, at pp. 1040-41; Ofshe & Leo (1997), supra, at p. 202. The use of false evidence is often crucial in convincing the suspect that protestations of innocence, even if true, are futile. I do not mean to suggest in any way that, standing alone, confronting the suspect with inadmissible or even fabricated evidence is necessarily grounds for excluding a statement. However, when combined with other factors, it is certainly a relevant consideration in determining on a voir dire whether a confession was voluntary.
62 England has also recognized the role of oppression. Section 76(8) of the Police and Criminal Evidence Act 1984 (U.K.), 1984, c. 60, states that a confession must not be the product of “oppression”, which is defined to include “torture, inhuman or degrading treatment, and the use or threat of violence (whether or not amounting to torture)”. The Code of Practice for the Detention, Treatment and Questioning of Persons by Police Officers goes on to offer examples of what may amount to oppression, which are similar to what I described above.
(c) Operating Mind
63 This Court recently addressed this aspect of the confessions rule in Whittle, supra, and I need not repeat that exercise here. Briefly stated, Sopinka J. explained that the operating mind requirement “does not imply a higher degree of awareness than knowledge of what the accused is saying and that he is saying it to police officers who can use it to his detriment” (p. 936). I agree, and would simply add that, like oppression, the operating mind doctrine should not be understood as a discrete inquiry completely divorced from the rest of the confessions rule. Indeed, in his reasons in Horvath, supra, at p. 408, Spence J. perceived the operating mind doctrine as but one application of the broader principle of voluntariness: statements are inadmissible if they are “not voluntary in the ordinary English sense of the word because they were induced by other circumstances such as existed in the present case”.
64 Similarly, in concluding that the confessions rule cannot be limited to the negative inquiry of whether there were any explicit threats or promises, Beetz J. offered the following explanation of the rule, at pp. 424-25:
Furthermore, the principle which inspires the rule remains a positive one; it is the principle of voluntariness. The principle always governs and may justify an extension of the rule to situations where involuntariness has been caused otherwise than by promises, threats, hope or fear, if it is felt that other causes are as coercive as promises or threats, hope or fear and serious enough to bring the principle into play.
As these passages make clear, the operating mind doctrine is just one application of the general rule that involuntary confessions are inadmissible.
(d) Other Police Trickery
65 A final consideration in determining whether a confession is voluntary or not is the police use of trickery to obtain a confession. Unlike the previous three headings, this doctrine is a distinct inquiry. While it is still related to voluntariness, its more specific objective is maintaining the integrity of the criminal justice system. Lamer J.’s concurrence in Rothman, supra, introduced this inquiry. In that case, the Court admitted a suspect’s statement to an undercover police officer who had been placed in a cell with the accused. In concurring reasons, Lamer J. emphasized that reliability was not the only concern of the confessions rule; otherwise the rule would not be concerned with whether the inducement was given by a person in authority. He summarized the correct approach at p. 691:
[A] statement before being left to the trier of fact for consideration of its probative value should be the object of a voir dire in order to determine, not whether the statement is or is not reliable, but whether the authorities have done or said anything that could have induced the accused to make a statement which was or might be untrue. It is of the utmost importance to keep in mind that the inquiry is not concerned with reliability but with the authorities’ conduct as regards reliability.
66 Lamer J. was also quick to point out that courts should be wary not to unduly limit police discretion (at p. 697):
[T]he investigation of crime and the detection of criminals is not a game to be governed by the Marquess of Queensbury rules. The authorities, in dealing with shrewd and often sophisticated criminals, must sometimes of necessity resort to tricks or other forms of deceit and should not through the rule be hampered in their work. What should be repressed vigorously is conduct on their part that shocks the community. [Emphasis added.]
As examples of what might “shock the community”, Lamer J. suggested a police officer pretending to be a chaplain or a legal aid lawyer, or injecting truth serum into a diabetic under the pretense that it was insulin. Lamer J.’s discussion on this point was adopted by the Court in Collins, supra, at pp. 286-87; see also R. v. Clot (1982), 69 C.C.C. (2d) 349 (Que. Sup. Ct.).
67 In Hebert, supra, this Court overruled the result in Rothman based on the Charter’s right to silence. However, I do not believe that this renders the “shocks the community” rule redundant. There may be situations in which police trickery, though neither violating the right to silence nor undermining voluntariness per se, is so appalling as to shock the community. I therefore believe that the test enunciated by Lamer J. in Rothman, and adopted by the Court in Collins, is still an important part of the confessions rule.
68 While the foregoing might suggest that the confessions rule involves a panoply of different considerations and tests, in reality the basic idea is quite simple. First of all, because of the criminal justice system’s overriding concern not to convict the innocent, a confession will not be admissible if it is made under circumstances that raise a reasonable doubt as to voluntariness. Both the traditional, narrow Ibrahim rule and the oppression doctrine recognize this danger. If the police interrogators subject the suspect to utterly intolerable conditions, or if they offer inducements strong enough to produce an unreliable confession, the trial judge should exclude it. Between these two extremes, oppressive conditions and inducements can operate together to exclude confessions. Trial judges must be alert to the entire circumstances surrounding a confession in making this decision.
69 The doctrines of oppression and inducements are primarily concerned with reliability. However, as the operating mind doctrine and Lamer J.’s concurrence in Rothman, supra, both demonstrate, the confessions rule also extends to protect a broader conception of voluntariness “that focuses on the protection of the accused’s rights and fairness in the criminal process”: J. Sopinka, S. N. Lederman and A. W. Bryant, The Law of Evidence in Canada (2nd ed. 1999), at p. 339. Voluntariness is the touchstone of the confessions rule. Whether the concern is threats or promises, the lack of an operating mind, or police trickery that unfairly denies the accused’s right to silence, this Court’s jurisprudence has consistently protected the accused from having involuntary confessions introduced into evidence. If a confession is involuntary for any of these reasons, it is inadmissible.
70 Wigmore perhaps summed up the point best when he said that voluntariness is “shorthand for a complex of values”: Wigmore on Evidence (Chadbourn rev. 1970), vol. 3, § 826, at p. 351. I also agree with Warren C.J. of the United States Supreme Court, who made a similar point in Blackburn v. Alabama, 361 U.S. 199 (1960), at p. 207:
[N]either the likelihood that the confession is untrue nor the preservation of the individual’s freedom of will is the sole interest at stake. As we said just last Term, “The abhorrence of society to the use of involuntary confessions . . . also turns on the deep-rooted feeling that the police must obey the law while enforcing the law; that in the end life and liberty can be as much endangered from illegal methods used to convict those thought to be criminals as from the actual criminals themselves.” . . . Thus a complex of values underlies the stricture against use by the state of confessions which, by way of convenient shorthand, this Court terms involuntary, and the role played by each in any situation varies according to the particular circumstances of the case.
See Hebert, supra. While the “complex of values” relevant to voluntariness in Canada is obviously not identical to that in the United States, I agree with Warren C.J. that “voluntariness” is a useful term to describe the various rationales underlying the confessions rule that I have addressed above.
71 Again, I would also like to emphasize that the analysis under the confessions rule must be a contextual one. In the past, courts have excluded confessions made as a result of relatively minor inducements. At the same time, the law ignored intolerable police conduct if it did not give rise to an “inducement” as it was understood by the narrow Ibrahim formulation. Both results are incorrect. Instead, a court should strive to understand the circumstances surrounding the confession and ask if it gives rise to a reasonable doubt as to the confession’s voluntariness, taking into account all the aspects of the rule discussed above. Therefore a relatively minor inducement, such as a tissue to wipe one’s nose and warmer clothes, may amount to an impermissible inducement if the suspect is deprived of sleep, heat, and clothes for several hours in the middle of the night during an interrogation: see Hoilett, supra. On the other hand, where the suspect is treated properly, it will take a stronger inducement to render the confession involuntary. If a trial court properly considers all the relevant circumstances, then a finding regarding voluntariness is essentially a factual one, and should only be overturned for “some palpable and overriding error which affected [the trial judge’s] assessment of the facts”: Schwartz v. Canada,  1 S.C.R. 254, at p. 279 (quoting Stein v. The Ship “Kathy K”,  2 S.C.R. 802, at p. 808) (emphasis in Schwartz).
D. Application to the Present Appeal
72 Applying the foregoing law to the facts of this appeal, and having viewed the relevant video- and audiotapes, I find no fault with the trial judge’s conclusion that the respondent’s confession was voluntary and reliable. The respondent was fully apprised of his rights at all times; he was never subjected to harsh, aggressive, or overbearing interrogation; he was not deprived of sleep, food, or drink; and he was never offered any improper inducements that undermined the reliability of the confessions. As the Court of Appeal reached a contrary conclusion with respect to a number of these issues, I will address them in turn.
1. Minimizing the Seriousness of the Crimes
73 The Court of Appeal concluded that the police improperly offered leniency to the respondent by minimizing the seriousness of his offences and suggesting “that the same punishment would likely be given whether he confessed to one or a number of fires” (para. 156). This, in their opinion, was an improper inducement (at para. 126):
In the beginning, it was suggested that “there isn’t much in a car fire”. Once the admission relating to the car was obtained, then the suggestion was made — and on several occasions — that the accused was not really a criminal and that the police did not want to treat him as a criminal. In addition, it was stated to the accused — again more than once — that there was little difference between being found guilty of one fire as compared to 10.
74 Insofar as the police simply downplayed the moral culpability of the offence, their actions were not problematic. As even the Court of Appeal recognized (at para. 126), “minimizing the moral significance of the offence is a common and usually unobjectionable feature of police interrogation”. Instead, the real concern is whether the police suggested that “confession will result in the legal consequences being minimal” (para. 126). As discussed above, this is inappropriate.
75 However, and with the greatest respect to the Court of Appeal, I believe they have mischaracterized the police interrogators’ words. The offending passages are well represented by the following excerpt (A.R. at p. 552), made shortly after the respondent arrived at the police station subsequent to his initial confession:
If you done the other ones this -- or some of the other ones this is the time -- this is the time to just get them off your chest. This is the perfect opportunity because of what you’ve already told us, okay. And everybody can see this, that it’s --You didn’t do one fire and then years down the road you did -- this is a series of fires we’ve been having in Waterville. So we can look at it -- we look at it as a one-package type of thing. Okay. And it’s -- if you had a problem, I don’t know what it is yet. Maybe we’ll find out what it is, maybe you can help us on this. It’s not unrealistic that you would set some more things on fire especially when you would do your girlfriend’s vehicle, your fiancee’s vehicle but you don’t know why. So there’s something that -- there’s something that triggers you into setting that fire. [Emphasis added.]
76 The Court of Appeal focused on the underlined passage to suggest that the police were offering a “package deal”, whereby the respondent would not be charged with multiple crimes if he confessed to them all. However, as the rest of the passage makes clear, the police were doing nothing of the sort. Instead, they were simply pointing out their reasons for believing that he was responsible for all the fires, not just one: namely, that it was a series of fires in issue, not isolated incidents. The police therefore treated the fires as a “package”, all of which were likely set by the same person.
77 This interpretation is confirmed by the police’s consistent refusal to accept Oickle’s own suggestions of a “package deal”. Shortly before confessing to the vehicle fire, the following exchange took place between the respondent and Corporal Deveau (A.R. at pp. 519-20):
A: No, hang on, hang on. If [I] admit to her car and are the other ones looked at too?
Q: Richard, all I can tell you now is I want the truth out. I don’t want — you said, “If I admit to her car,” which leads me to believe that maybe you’re involved in that.
Q: Well, if you’re involved in it, tell me the truth and then we’ll look — you know, if — I don’t think for one minute that you’re involved in everything. Okay? But if you did the car, Richard, tell me you did the car. And if I believe that’s it, if we believe that you did not do the other one, I mean, we’re — remember I said, we’re not here to trick you into anything?
A: I trust you.
Q: I’m not here to bring everything down on you. The last thing I want to do, Richard. You’ve been good to me and I’m trying to be good to you.
Q: And I want you to tell me the truth. So if you did the car, tell me you did the car. But I want the truth. I just don’t want you to say, “I did the car, so I’m free from all the others.” Okay? That’s why it’s important here that —
Q: It’s the truth that we want. [Emphasis added.]
As this passage reveals, it was the respondent, not the police, who was seeking a “package deal”— a deal Corporal Deveau squarely rejected. While the police did minimize the moral significance of the crimes, there was never any suggestion by the police that a confession would minimize the legal consequences of the respondent’s crimes.
2. Offers of Psychiatric Help
78 The Court of Appeal also found that the police improperly offered psychiatric help in return for a confession. For example, at para. 121, the Court of Appeal noted passages wherein the police told the respondent “I think you need help”, and “[m]aybe you need professional help”. See also paras. 108 (pp. 363-64) and 122 (pp. 371-72). However, at no point did the police ever suggest that the respondent could only get help if he confessed. The distinction here is between the police suggesting the potential benefits of confession, and making offers that are conditional upon receiving a confession. The former is entirely appropriate — it is not an inducement because there is no quid pro quo. The latter is improper. However, the police made no such offer in the course of their interrogation of the respondent.
3. “It Would Be Better”
79 The transcripts are indeed rife with these sorts of comments. The police suggested that a confession would make the respondent feel better, that his fiancée and members of the community would respect him for admitting his problem (para. 120) and that he could better address his apparent pyromania if he confessed (para. 122). However, read in context, none of these statements contained an implied threat or promise. Instead, they were merely moral inducements suggesting to the respondent that he would feel better if he confessed and began addressing his problems. And indeed, after his confession, Corporal Deveau asked him “[s]o how do you feel now, Richard?” His answer was “[b]etter”.
80 To hold that the police officers’ frequent suggestions that things would be better if the respondent confessed amounted to an improper threat or inducement would be to engage in empty formalism. The tapes of the transcript clearly reveal that there could be no implied threat in these words. The respondent was never mistreated. Nor was there any implied promise. The police may have suggested possible benefits of confession, but there was never any insinuation of a quid pro quo. I therefore respectfully disagree with the Court of Appeal that these comments undermined the confessions’ voluntariness.
4. Alleged Threats Against the Respondent’s Fiancée
81 As discussed in connection with Jackson, supra, a threat or promise with respect to a third person could be an improper inducement. The Court of Appeal stated, at para. 128, that the police effectively told the respondent that “If he confessed, it would not be necessary to continue the investigation or put his fiancé [sic] through extensive interrogation.”
82 The majority of references during the interrogation to the respondent’s fiancée, Tanya Kilcup, centered on the respondent’s reliance on her as an alibi witness: see, e.g., A.R. at p. 570. However, the Court of Appeal is correct that there were moments when the police intimated that it might be necessary to question Ms. Kilcup to make sure she was not involved in the fires at all, either alone or in collaboration with the respondent:
Q. You know, this whole thing is — we might even ask Tanya if she would take a polygraph on this because we don’t know where she stands, okay.
A. Do I have to sit here for that?
Q. Oh, no, no, not until she takes the polygraph. She’s not going to take the polygraph tonight. But if you can tell us anything — [A.R. at p. 574]
. . .
Q. Do you realize the other reason is that we — that you’ve got to come clean with everything with us is for Tanya.
Q. We don’t want to put Tanya through any — I mean she’s going to be going through enough trying to — we don’t want to — and I’m sure you don’t want her to get — to go through half or what you went through today. It’s no fun.
A. No, no.
Q. It won’t be any fun for her. But in order for her to — in order for us to be one hundred percent we have to do it. So if there’s anything that you can tell us that can put her — that we say, okay, we don’t need you, Tanya, we have it here, you know, and we have some stuff. But we’re not convinced on everything else. So don’t put Tanya through that if there’s something you can tell me, okay. [A.R. at pp. 603-4]
83 The relationship the respondent had with Ms. Kilcup was, in my opinion, strong enough potentially to induce a false confession were she threatened with harm. However, I do not believe any such threat ever occurred. There were no pending charges against Ms. Kilcup that the police were offering to drop; they never threatened to bring charges against her; indeed, the police never seriously suggested her as a suspect. The most they did was promise not to polygraph her if the respondent confessed. Given the entire context, the most likely reason to polygraph her was not as a suspect, but as an alibi witness. In my opinion, this is not a strong enough inducement to raise a reasonable doubt as to the voluntariness of the respondent’s confessions.
84 Moreover, the timing of the comments regarding Ms. Kilcup suggests that there was no causal connection between the police inducements and the subsequent confession. After the statements quoted above, Corporal Deveau left the room, and told the respondent that he intended to speak to Tanya. Therefore the respondent’s actual confession was approximately two hours after he thought the police were already speaking to Tanya. Moreover, soon after Constable Bogle took over the interrogation, the respondent himself made it clear that he thought the police were only talking to Ms. Kilcup in order to verify his alibi (A.R. at p. 611):
Q. Okay. I mean we have to go and — we asked Cst. Taker to talk to Tanya, okay. (Inaudible).
A. But I didn’t tell her.
A. I didn’t tell her.
A. Totally by myself.
The “inducements” regarding the respondent’s fiancée lacked both the strength and causal connection necessary to warrant exclusion.
5. Abuse of Trust
85 The Court of Appeal suggests at para. 129 that the police in general, and Corporal Deveau in particular, improperly abused the respondent’s trust to obtain a confession. With respect, I cannot agree. In essence, the court criticizes the police for questioning the respondent in such a gentle, reassuring manner that they gained his trust. This does not render a confession inadmissible. To hold otherwise would send the perverse message to police that they should engage in adversarial, aggressive questioning to ensure they never gain the suspect’s trust, lest an ensuing confession be excluded.
6. Atmosphere of Oppression
86 To hold that the police conduct in this interrogation was oppressive would leave little scope for police interrogation, and ignore Lamer J.’s reminder in Rothman, supra, at p. 697, that “the investigation of crime and the detection of criminals is not a game to be governed by the Marquess of Queensbury rules”. Quite simply, the police acted in a proper manner. Viewing the videotapes and listening to the audiotapes reveal that at all times the police were courteous; they did not deprive the respondent of food, sleep, or water (at para. 119); they never denied him access to the bathroom; and they fully apprised him of his rights at all times (see, e.g., A.R. at pp. 370, 497 and 650). They did not fabricate evidence in an attempt to convince him denials were futile. They comforted him, with apparent sincerity, when he broke down in tears upon confessing. While the re-enactment was admittedly done at a time when the respondent had had little sleep, he was already awake when they approached him, and was told that he could stop at any time. And indeed, the Court of Appeal did not directly claim that the police created an atmosphere of oppression sufficient to exclude the statements.
87 The absence of oppression is important not only in its own right, but also because it affects the overall voluntariness analysis. In the preceding sections, I have concluded that the police offered the respondent, at best, extremely mild inducements. In particular, they suggested that “it would be better” if he confessed, and suggested that his girlfriend could be spared questioning if he confessed. However, given the entirely non-oppressive atmosphere maintained by the police, I do not believe that any of the alleged inducements are sufficient to render the confessions involuntary.
E. The Relevance of the Polygraph Test
88 In addition to the issues addressed above, the Court of Appeal found the police use of a polygraph particularly problematic. Because of the growing frequency with which police are using the polygraph as an investigative tool, and the absence of any direction thus far from this Court regarding the proper use of polygraphs in interrogations, I will now briefly discuss how polygraphs fit into the analytical framework set out above. The Court of Appeal identified several problems with the police’s use of a polygraph in this appeal. I will address each in turn.
1. Informing the Suspect of the Uses to Which the Polygraph Test Can Be Put
89 The Court of Appeal first stated that the police failed “to inform the accused clearly that the polygraph test was not admissible in court to show whether the accused was lying or telling the truth” (para. 156); see also R. v. James, Ont. Ct. (Gen. Div.), January 25, 1991; R. v. Ollerhead (1990), 86 Nfld. & P.E.I.R. 38 (Nfld. S.C.T.D.); R. v. Fowler (1979), 23 Nfld. & P.E.I.R. 255 (Nfld. C.A.).
90 To the contrary is R. v. Alexis (1994), 35 C.R. (4th) 117 (Ont. Ct. (Gen. Div.)). As noted at para. 159 of Hill J.’s lucid reasons in that case,
confrontation of a suspect with polygraph test results, in such circumstances, is not qualitatively dissimilar from such permissible techniques of persuasion as the police showing a detained suspect a co-accused’s confession inadmissible in evidence against the suspect, or police trickery, for example, the ruse of relating to the suspect that his or her fingerprint has been discovered at the scene of the crime.
On this view, police trickery or use of inadmissible evidence is not necessarily grounds for exclusion.
91 I agree that merely failing to tell a suspect that the polygraph is inadmissible will not automatically produce an involuntary confession. Courts should engage in a two-step process. First, following Rothman, supra, and Collins, supra, the confession should be excluded if the police deception shocks the community. Second, even if not rising to that level, the use of deception is a relevant factor in the overall voluntariness analysis. At this stage, the approach is similar to the one used with fabricated evidence, supra — though of course the use of inadmissible evidence is inherently less problematic than fabricated evidence. Standing alone, simply failing to tell the suspect that the polygraph results are inadmissible will not require exclusion. The most it can do is be a factor in the overall voluntariness analysis.
92 Moreover, in this particular appeal, the police made it abundantly clear to the respondent just what was admissible and what was not. For example, as recognized by the Court of Appeal at para. 81, Sergeant Taker told the respondent at the outset of the polygraph that his “opinion based on the results of your polygraph test is not admissible in court. However, anything said between you and I may be admissible.” Moreover, the respondent demonstrated during the interrogation that he understood this (A.R. at p. 464):
Q. . . . Because your heart has told me that you haven’t been truthful.
A. I don’t care what that thing says.
Q. That thing cannot say anything. (Inaudible)
A. I don’t care what you interpret from that thing.
Q. Just a minute now, Richard. Hear me out. Hear me out. That does not say anything, okay. Your body is what says it. That only records things, like I told you earlier –
A. I know that. [Emphasis added.]
93 This passage clearly demonstrates the respondent’s understanding that the bare polygraph readouts are irrelevant; what matters is the polygrapher’s opinion of these readings. Since Sergeant Taker clearly told the respondent that his interpretation of the readings was not admissible, I agree with MacDonald Prov. Ct. J. that “[t]here is no evidence here whatsoever that Mr. Oickle was confused on this point.”
2. Exaggerating the Polygraph’s Validity
94 The Court of Appeal also noted, correctly in my opinion, that the police made “repeated assertions to the accused that the polygraph was an infallible determiner of truth” (para. 156). Throughout the interrogation that produced the respondent’s initial admission that he set Ms. Kilcup’s vehicle on fire, both Sergeant Taker and Constable Deveau emphasized that the polygraph did not make mistakes, and that if Sergeant Taker interpreted it to indicate deception, then the respondent must have lied. For example, the Court of Appeal cited the following passage (at paras. 141-42):
[Oickle:] But if you read the chart and it says they are lying, then they are.
[Taker:] That’s right. That’s right.
Deveau: There’s no doubt in anybody’s mind now that you are involved in some of these fires.
Oickle: Because I failed that...
Deveau: Yes, very simple Richard... and when asked the question about these eight fires, the polygraph says that you are not truthful... the machine does not lie. You found that out today. [Emphasis added.]
95 I agree that the police exaggerated the accuracy of the polygraph. As many sources have demonstrated, polygraphs are far from infallible: see, e.g., D. T. Lykken, A Tremor in the Blood: Uses and Abuses of the Lie Detector (1998); J. J. Furedy, “The ‘control’ question ‘test’ (CQT) polygrapher’s dilemma: logico-ethical considerations for psychophysiological practitioners and researchers” (1993), 15 Int. J. Psychophysiology 263; C. J. Patrick and W. G. Iacono, “Validity of the Control Question Polygraph Test: The Problem of Sampling Bias” (1991), 76 J. App. Psych. 229. Similarly, this Court recognized in R. v. Béland,  2 S.C.R. 398, that the results of polygraph examinations are sufficiently unreliable that they cannot be admitted in court.
96 The Quebec Court of Appeal concluded in R. v. Amyot (1990), 58 C.C.C. (3d) 312, at p. 324, that representing the polygraph as infallible rendered a confession involuntary. In that case the polygrapher told the accused that [translation] “the test showed him that he is not telling the truth”. This, the court found, was inappropriate in that it
[translation] pushed what the examination consisted of much too far, into the absolute. The result was presented to the appellant as a certitude which obviously was going to shake him up and it made him say “but what is going to happen now?”. It seems to me that, as a result, the appellant was led into error on the infallibility of the test and this manner of proceeding could naturally induce a person to “confess”.
See also Fowler, supra. The Court of Appeal in Amyot put particular emphasis on the fact that the suspect confessed almost immediately after hearing the polygraph results, suggesting that his will was overwhelmed upon being confronted with the damning, supposedly incontrovertible evidence.
97 Without expressing an opinion as to whether Amyot was correctly decided, I note that the facts of the present appeal are very different. As the following passages demonstrate, the respondent repeatedly rejected the accuracy of the polygraph results:
A. I think you could bring a completely innocent person in here and with a bunch of nerves could do the same thing I just did. [A.R. at p. 495]
Q. So you’re telling me that this test today is a bunch of shit. Is that what you’re trying to tell me?
A. In my opinion, yeah. [A.R. at p. 505]
The respondent was not overwhelmed by the polygraph results. While the police clearly relied heavily on them to elicit a confession, this was not a situation like Amyot where the confession followed almost immediately after the announcement of the results.
98 Other courts have excluded confessions obtained through use of a polygraph only where the suspect took some time before eventually confessing. For example, in Ollerhead, supra, the court cited the following passage from R. v. Romansky (1981), 6 Man. R. (2d) 408 (Co. Ct.), at p. 421:
[T]he psychological tactics employed by him created an aura of oppression. The will of the accused quickly crumbled with his emotional disintegration. As evidenced by the concomitant amenability and/or responsiveness to suggestions, his will was overcome and overborne by the will of the person in authority.
Various lower courts have thus taken very different approaches to determining whether polygraphs create an oppressive atmosphere. The contrasting approaches in cases like Amyot and Ollerhead demonstrate that the timing of the confession vis-à-vis the polygraph cannot be determinative. Instead, it is but a piece of evidence for the trial judge to consider in determining whether the confession was voluntary.
99 Granted that the police misled the respondent with regards to the accuracy of the polygraph, the question remains whether, in light of the entire circumstances of the interrogation, this rendered the confessions inadmissible. In my opinion it did not. As discussed above, there was no emotional disintegration in this case. The mere fact that a suspect begins to cry when he or she finally confesses, as the respondent did, is not evidence of “complete emotional disintegration”; tears are to be expected when someone finally divulges that they committed a crime — particularly when the suspect is a generally law-abiding and upstanding citizen like the respondent.
100 Nor, as discussed above, do I believe that the police created an oppressive atmosphere. Simply confronting the suspect with adverse evidence, like a polygraph test, is not grounds for exclusion: see Fitton, supra. This holds true even for inadmissible evidence: see Alexis, supra. Nor does the fact that the police exaggerate the evidence’s reliability or importance necessarily render a confession inadmissible. Eyewitness accounts are by no means infallible; yet in Fitton, this Court ruled admissible a statement taken after the police told a suspect they did not believe his denials because several eyewitnesses had come forward against him. In short, merely confronting a suspect with adverse evidence — even exaggerating its accuracy and reliability — will not, standing alone, render a confession involuntary.
3. Misleading the Accused Regarding the Duration of the Interview
101 The final ground on which the Court of Appeal challenged the use of the polygraph, at para. 156, was the police’s
misleading the accused about the expected duration of the test procedure, particularly concerning the interrogation to follow and immediately commencing intense questioning upon informing the accused that he had “failed” the test....
A similar argument was made in Nugent, supra. Since this Court has ruled that polygraph results are not admissible in evidence, Béland, supra, “then the administering of a test must be clearly separated from questioning for the purpose of obtaining statements” (Nugent, supra, at p. 212). According to the Court of Appeal, a statement directly following a polygraph should not be admissible because the defence cannot adequately explain the context of the statement — which it might wish to do in order to attack the weight of the statement before the jury — without notifying the jury that the accused failed a polygraph test.
102 Drawing on these arguments, the intervener, the Criminal Lawyers’ Association, argued that the police have only two options when using polygraphs. One is to ensure that the suspect has consulted with counsel before consenting to the test. The other is to “clearly separate any post-test interrogation from the test itself”. I do not believe that it is necessary to limit the police’s discretion in this manner. It is true that the police procedures present the defence with the unpalatable choice of either trying to explain away the confession without using the polygraph, or admitting that the accused failed the test. However, this is true any time a suspect confesses after being confronted with inadmissible evidence, and it does not necessarily render the confession involuntary. Tactical disadvantage to the defence is not relevant to the voluntariness of the defendant’s confession; instead, if anything, it simply suggests prejudicial effect. However, given the immense probative value of a voluntary confession, I cannot agree that exclusion is appropriate.
103 The final argument in favour of separating the interrogation from the polygraph test is related to the alleged “abuse of trust” addressed above. It is submitted that the intimacy fostered during the pre-test interview improperly carries over to the post-test interrogation. Whether this is true or not, I do not believe it would be grounds to exclude the confession. On this point, I agree with the Ontario Court of Appeal in R. v. Barton (1993), 81 C.C.C. (3d) 574, at p. 575:
There is no question that the procedure is intrusive and purports to use expertise in psychology to create a relationship between the interviewer and the candidate which is conducive to making the technical analysis more accurate. It is also true that the appearance of intimacy carries over into the third stage when, in this case, the inculpatory statement was made. Yet, all police interrogations may include these features in one form or another. The “good cop, bad cop” routine is the best known.
Moreover, in this appeal the respondent did not confess until Corporal Deveau took over the questioning from Sergeant Taker. Therefore, any intimacy created by the pre-test interview could not have precipitated the respondent’s confessions.
F. Summary on Voluntariness
104 In summary, there were several aspects of the police’s interrogation of the respondent that could potentially be relevant to the voluntariness of his confessions. These include the comments regarding Ms. Kilcup; the suggestions that “it would be better” for the respondent to confess; and the exaggeration of the polygraph’s accuracy. These are certainly relevant considerations when determining voluntariness. However, I agree with the trial judge that neither standing alone, nor in combination with each other and the rest of the circumstances surrounding the respondent’s confessions, do these factors raise a reasonable doubt about the voluntariness of the respondent’s confessions. The respondent was never mistreated, he was questioned in an extremely friendly, benign tone, and he was not offered any inducements strong enough to raise a reasonable doubt as to voluntariness in the absence of any mistreatment or oppression. As I find no error in the trial judge’s reasons, the Court of Appeal should not have disturbed her findings.
105 In conclusion, in my view the Nova Scotia Court of Appeal applied both the wrong standard of appellate review, and reached the wrong conclusion with regard to voluntariness. I would therefore allow the appeal, set aside the judgment of the Court of Appeal, and restore the trial judge’s conviction of the respondent.
The following are the reasons delivered by
Arbour J.(dissenting) –
106 I have had the benefit of the reasons of my colleague, Justice Iacobucci, on this appeal. With respect, I believe that there were improper inducements held out by the police officers who interrogated the respondent and that these inducements, considered cumulatively and contextually in light of the “failed” polygraph test, require the exclusion of the respondent’s statements. Moreover, in my view the proximity and the causal connection between the “failed” polygraph test and the confession also compels this result. Accordingly, I would dismiss the appeal, set aside the convictions and enter acquittals on all counts.
107 This case involves exclusively the admissibility of a confession. We were greatly assisted, as was the Court of Appeal, by the existence of a rich record from which the factual issues can be reviewed. As I hope will increasingly be the case, the entire interrogation of the respondent, including the exchanges that preceded the administration of the polygraph test, was recorded. The first part, which lasted until the first incriminating statement, was audio recorded, and the several hours that followed were also video recorded. While some deference is always required because of the privileged position from which the trial judge assesses credibility, including on a voir dire, appeal courts must ensure that the question of the voluntariness of a confession was correctly decided by the trial judge, in accordance with the applicable law, and on a reasonable view of the facts.
108 The determination of the voluntariness of a statement obtained after a long
and protracted interrogation requires careful scrutiny of the record. This is particularly so where, as here, the improper inducements are interspaced throughout a web of skilful, effective and, I may add, permissible misrepresentations made to the respondent by the police. The words of Rand J., in R. v. Fitton,  S.C.R. 958, at p. 962, are instructive in this regard:
The cases of torture, actual or threatened, or of unabashed promises are clear; perplexity arises when much more subtle elements must be evaluated. The strength of mind and will of the accused, the influence of custody or its surroundings, the effect of questions or of conversation, all call for delicacy in appreciation of the part they have played behind the admission, and to enable a Court to decide whether what was said was freely and voluntarily said, that is, was free from the influence of hope or fear aroused by them.
109 I will not repeat the facts which are canvassed by my colleague. I simply wish to highlight the salient parts as they relate to the issues that I have identified.
II. The Administration of the Polygraph Test
110 Properly understood, this case involves two confessions obtained by the police following the “failure” of a polygraph test and a skillful interrogation which lasted nearly six hours. Repeated threats and promises were made. They were often subtle but in my view, against the backdrop of the polygraph procedure, they overwhelmed the free will of the respondent. These seemingly mild pressures make this case a difficult one in which to apply the confessions rule and demand an attentive appreciation of the full context in which the alleged voluntary, incriminating statements were made.
111 I fully agree with the summary of the applicable law provided by Justice Iacobucci at paras. 68-70. However, I take a different view of the proper legal characterization of what happened in the course of the many hours during which the respondent was interrogated and of the voluntary quality of his incriminating statements.
112 At the request of the police, on April 26, 1995, the respondent met Corporal Bruno Deveau and Sergeant Gregory Taker, both of the RCMP, at a motel room where the police were set up to administer a polygraph test to him. By then, the police had already administered several such tests in relation to their investigation of a series of fires in the Waterville area in the period from February 1994 to April 4, 1995. They administered two such polygraph tests in June of 1994, to two different subjects. In the opinion of the polygraph examiner (also called a polygraphist), Sergeant Taker, the results of these two tests were inconclusive. Other tests were administered in July and September 1994, and in January 1995. In all three cases, Sergeant Taker formed the opinion that the subject was telling the truth. See A.R. at pp. 735-54.
113 The last fire was set some two weeks before the respondent was asked to attend for his test; it involved a vehicle owned by his girlfriend, Tanya Kilcup. She was living with him at the time, and he said that they were planning to get married. On the form filled out by Sergeant Taker as an introduction to the polygraph procedure, under the heading “demeanor”, the police officer indicated the respondent appeared “nervous”. The form also indicates that the respondent understood his Charter rights, understood the warning, did not want a lawyer and understood that the purpose of the polygraph was “to see if I had anything to do with my girlfriend’s car fire”. He said he felt “[g]ood” about the test and thought it “should be ok”. See A.R. at pp. 746-48.
114 The respondent signed a consent form and was given a booklet which explained the testing procedure. The three-page booklet states that the polygraph, also known as the “Lie Detector” test, should be called a “Truth Verifier” because statistics show that in the majority of cases, the opinion of the examiner is that the subject was truthful. The booklet uses a question and answer format. It describes a two-step process: the first is the “Pre-Test Interview”, in which the subject’s legal rights are explained, his or her medical and psychological background is explored, and the facts under investigation and the workings of the instrument are reviewed; the second step is the “In-Test Phase”. It is described in the booklet as follows:
In-Test Phase: At this point, the polygraphist will review all pertinent test questions with you and then conduct a “Demonstration Test” to help he or she assess your suitability for additional tests. Several tests will then be conducted, which will include the reviewed questions. There will be no surprise questions on any of the tests. The polygraphist will then analyse the results of each test to arrive at an opinion.
Two other entries in the booklet are, in my view, of interest:
5. SUPPOSE THE INSTRUMENT SAYS I’M LYING WHEN I’M NOT?
The polygraph cannot say anything. It merely records on a chart certain bodily reactions to the questions you are being asked. It is up to the polygraphist to “Read” the information your body provides.
9. DOES THIS EXAMINATION MEAN THAT THE POLYGRAPH IS INFALLIBLE?
No. Most examinations readily reveal to the polygraphist whether or not the person is truthful. Possible outcome of any polygraph examination is Truthful, Deceptive or Inconclusive. “Inconclusive” means the examiner is unable to render an opinion of either Truthful or Deceptive in that particular case. The polygraphist may request a re-examination if the results are inconclusive. A re-examination is also voluntary.
It is interesting to me that this last answer is not very responsive to the real question raised. It suggests that the polygraph is not infallible because it sometimes does not indicate whether the subject is truthful or not, in which case the result may be inconclusive. The answer does not say whether mistakes can be made in identifying liars. In fact, it suggests that no such errors are likely since “[m]ost examinations readily reveal to the polygraphist whether or not the person is truthful”.
115 Although this was not described in the booklet, where, after the in-test phase, the subject is deemed to have been “truthful” the polygraph test comes to an end. However, if he is deemed to have been “untruthful”, it is standard procedure to move to a post-test “interview” and to interrogate the subject with a view to obtaining a confession. This was clearly the plan here and the post-test “interview” was strategically planned, and successfully executed.
116 The pre-test interview conducted here by the polygraph examiner, Sergeant Taker, laid critical groundwork for the post-test interrogation. It did so in two respects. First, in my opinion, it created ambiguity about the precise admissibility, or lack thereof, of polygraph-related evidence. Secondly, in conjunction with the wording of question 9, it conveyed the overall impression that the polygraph machine is a scientific and infallible lie-detector, akin to trustworthy medical procedures and tests, and, relatedly, that the polygraph machine yields objectively accurate information since it measures involuntary body reactions, independent of the controlled responses of the subject.
117 In the pre-test phase, Sergeant Taker informed the respondent of his rights. In the course of so doing, he provided the respondent with the following information concerning the admissibility of polygraph-related evidence. He stated that “[m]y opinion based on the results of your polygraph test is not admissible in court. However, anything said between you and I may be admissible” (A.R. at p. 370). Even when combined with the information provided in the booklet, to the effect that the polygraph cannot say anything, but that it is up to the polygraphist to “‘Read’ the information your body provides”, the explanation provided by Sergeant Taker, in my view, is hardly informative about the admissibility, or non-admissibility of polygraph-related evidence. In fact, it is incomprehensible. It gives no explanation of the distinction between an “opinion based on the results of your polygraph test”, the “results of your polygraph test” and the “reading” by the polygraphist of the recordings on the polygraph chart. It certainly does not convey accurately that the interrogators’ categorical assertions that the respondent had been untruthful, repeated to him at every opportunity during the post-test interview, reflected little more than the interrogators’ inadmissible opinions based on the bare graphs produced during the respondent’s test.
118 In fact, the record contains numerous statements by Sergeant Taker that may have been taken by the respondent as suggesting that the “results” of a polygraph test encompass more than the bare graphs produced by the machine’s pens and that these have meaning independent of the polygraphist’s reading or interpretation. The following assertions by Sergeant Taker are illustrative:
- It will only record that as a lie because to pass the polygraph you must be
100 percent truthful. [A.R. at p. 410]
- [I]t will show that you’re lying. [A.R. at p. 415]
- ...I would rather you explain that to me now than have the polygraph later
on tell me you’ve not been truthful. [A.R. at p. 416]
- They’re also questions that will show me when you’re telling the truth when I start off the polygraph okay. [A.R. at p. 442]
- [I]f you don’t answer 100 percent truthfully then you can’t pass the test. [A.R. at p. 445] [Emphasis added.]
119 Having explained to the respondent his rights, Sergeant Taker sought to impress upon him, both subtly and forcefully, the scientific and reliable nature of the polygraph instrument. For example, the police officer asked the respondent a series of medical questions (e.g., whether he had ever been treated by a psychiatrist, had rheumatic fever, heart problems, epilepsy, tuberculosis, etc.), ostensibly for the purpose of determining the respondent’s “suitability” to take the test, that could not help but serve to bolster the scientific aura of the polygraph in the respondent’s mind. See A.R. at pp. 381-89. Further, and more importantly, he repeatedly, though often implicitly, emphasized the polygraph’s infallibility as a lie-detecting instrument:
Q. Because right now you’re the only person in this room that knows if you have lit any of these fires that I referred to earlier, right? But before the day is over there’s going to be two of us that knows if you’ve lit any of these
fires, right? [A.R. at p. 376]
. . .
Q. . . . to pass the polygraph you must be 100 percent truthful. It’s like a girl who comes home and says to her mother, she says, “Mum, I’m a little bit pregnant.” Well, you and I both know that’s — I mean that doesn’t work,
Q. You either are pregnant or you’re not pregnant. That’s exactly how the polygraph works. You either are telling the truth or you’re not telling the
. . .
Q. . . . As good as a polygraph is it doesn’t have any human capabilities. It doesn’t have any ability to understand or doesn’t have the ability to reason.
And basically it’s a cold scientific instrument. . . . [A.R. at pp. 410-11] [Emphasis added.]
120 Sergeant Taker also impressed upon the respondent that the polygraph records the subject’s involuntary bodily responses, as opposed to his or her conscious, willed and therefore controlled verbal responses, to the questions posed. For example,
he stated (A.R. at p. 373):
I’m going to be using the polygraph to monitor your physiological responses is the — is your — any change in your movement in your chest cavity, your heart rate, and any changes in physiological activity in the things that I monitor, okay.
Later, he said (A.R. at pp. 404-5):
Q. The first set of nerves I’m going to talk about are the voluntary nerves.
And these are the nerves we can control at will.
. . .
Q. But as a polygraph examiner I’m not interested in that set of nerves. I’m interested in the... involuntary nerves or the autonomic nerves. Now the involuntary nerves control most of the muscles and the organs in your body. And the main muscle that they control is the heart muscle. And as polygraph examiner it’s the heart muscle that I’m mainly interested in, Richard.
. . .
Q. ... Now when I said to you, Richard, I’m mainly interested in the heart muscle it’s because I know when a person knowingly tells a lie and I must
stress that word “knowingly”. Your heart will rebel against that last. . . . [Emphasis added.]
III. The Post-Polygraph Interrogation
121 In my view, the information provided to the respondent during the pre-test interview served to convince him that the polygraph results would demonstrate to the police whether he had set fire to his girlfriend’s car. The interrogation that followed immediately on the heels of the announcement that the respondent had “failed” the polygraph test built very effectively on the groundwork laid during the pre-test interview. Sergeant Taker conducted approximately the first hour of the post-test interview. In that time, he further developed the “infallible polygraph” theme, introduced the possibility of psychiatric help and further obscured the distinction between an opinion based on the results of the test and the results of the test.
122 The respondent said very little during the nearly six hours that he spent being questioned. The three police officers who worked as a relay team to conduct the interview did most of the talking. Their overall approach and strategy were clear and proved effective. It consisted of persuading the respondent that during the polygraph test, his body had already betrayed him and that they now knew that he had some involvement in some of the fires. I believe that they were persuaded from the outset that all the fires had likely been caused by the same arsonist. However, they repeatedly told the respondent otherwise, focusing, at first, on the very recent burning of his girlfriend’s car. The polygraph test was set up in such a way that all eight fires were linked together and that he was asked to admit or deny his involvement in all of them collectively. This provided the police with the opportunity to persuade the respondent that he could have failed the test because of his involvement in only one fire, possibly the most trivial one — the girlfriend’s car — and that admitting to that one would dispel their concern that he might have done them all. It is clear, in my view, that the respondent came to be convinced that having failed the polygraph, he would not be left alone by the police, who were convinced of his guilt, until he conceded what his body had already revealed. On two occasions prior to the first confession, he asked what would happen if he admitted to setting fire to his fiancée’s car. This was clearly becoming the most attractive proposition to him.
123 In turn, that admission was to become the trigger for the next stage of the interrogation. At that point, the tables had turned substantially. To his surprise, it seems, the respondent was then placed under arrest, no longer free to leave, and was told that since he had no explanation for having set his girlfriend’s car on fire, it was likely that he had a problem with fires, and that he was involved in all the others.
124 I wish to stress that the overall interrogation strategy was sound, and that although it relied on considerable deception on the part of the police, that in itself is neither illegal nor sufficient to vitiate the voluntary nature of a confession. The line is crossed, and was crossed here in my view, when improper inducements are put forward by persons in authority in an oppressive atmosphere, undermining the interrogated
person’s control over his mind and will.
125 Here, the police created an atmosphere of trust, and persuaded the respondent, at the outset, that they thought he was a “good guy”, not a criminal, and that it was very unlikely that he was responsible for all the fires. This was not true, of course, but these representations did not violate the confessions rule. However, having left the respondent with no clear idea of the true nature — either scientific or legal — of the polygraph test, the police worked to persuade him that his continued denials were futile. With that in the background, they resorted to improper inducements which tipped the balance and served to convince the respondent to admit, first, that he set fire to his girlfriend’s car and, later, that he was also responsible for the building fires.
IV. The Voluntariness of the Statements: Threats, Promises and Inducements
A. Promise of Psychiatric Help
126 The first impermissible inducement was introduced by Sergeant Taker who suggested that the respondent needed psychiatric help and that he and/or the police could help the respondent get it (A.R. at pp. 506-8):
Q. . . . And I’m thinking and hoping that it’s only the car or the van or whatever the case may be, and it’s not those buildings. If it is, what I’m saying to you is that I’d rather see you be truthful with me now about that than lie to me and have me think that you burned those frigging buildings. Because if you burned the buildings, we need to get you some help, Richard. If you set those buildings on fire, we need to get you some help. And I’m talking maybe seeing a psychiatrist, Richard.
. . .
Q. . . . [W]hat I’m saying to you, if that’s the problem, let’s get it taken care of before someone gets hurt. Because it’s something that you can’t control if that’s what the problem is. You can’t control that.
A. I can’t control my own life?
Q. No, if you’re lighting fires, I’m saying, in those buildings... [you’re] the person that’s lighting those fires on the buildings maybe you better get some help, okay?
Now, if it’s you that’s lighting those buildings on fire, there’s people to see that can help you out.... And if that’s the case, Richard, let’s deal with that because I’ve got — I’d be really concerned with that.
In the final hour of the interrogation, then being conducted by Constable Dale Bogle, this help theme assumed a dominant role, particularly in the seconds just prior to the second confession:
Q. This fire here, you can just take this car one, for instance. I mean there’s
areas we have to place things. Is it a sickness? Would it happen again? Okay. How many times before? What can we do to help you so it can be
stopped? Okay. Do you understand?
. . .
Q. We have to talk about what do we have to — Is it a sickness?
Q. Do we have to have you sent to a hospital for assessment sort of thing?
. . .
Q. . . . Let’s treat it as a problem. Okay. I can say, Look, you know, look guys, I need some help. I don’t want to hurt anybody, property, you know,
(inaudible) been a car....
But if it’s pyromania and you think you’re suffering from it, for God’s sakes, come on guys give me a hand here. Okay. [A.R. at pp. 608-16]
. . .
Q. Okay. Let’s deal with the problem, okay. That problem you’ve got to come out and say, Look, boys, you know, give me help, I don’t know why I done it. Okay. [A.R. at p. 630]
. . .
Q. All right. So with that don’t you think you need some sort of help in
dealing with this or —
A. Someone to talk to, yeah, or maybe they can help me find out why. [A.R.
at p. 638]
. . .
Q. You’re a good guy. I know you’re a good guy.
A. I think I am. Yeah.
Q. Let us help you kind of deal with some of these things, okay?
. . .
Q. All right. Let’s not fabricate it anymore, okay. (Inaudible.) Let’s do it. Okay. [I’m really happy?]. Take that load off your back, okay. Okay, Richard. Okay. Just — let’s start a new chapter, okay. All right. Let us
help you. [A.R. at pp. 642-44]
In the midst of this last statement by Constable Bogle, the respondent collapsed emotionally. Leaning forward, with his head in his hands, he began to sob, while Constable Bogle comforted him, patting and rubbing his shoulder. With some additional prompting and encouragement from Constable Bogle, the respondent confessed to the building fires and provided police with a written statement to this effect.
127 In my opinion, the promise of psychiatric help was improper. While it is true that the police did not explicitly tell the respondent that the only way he could get psychiatric help was if he confessed, this was the clear implication of what was said. In the circumstances here, I believe that the offer by the police to provide the respondent with psychiatric assistance if he confessed his involvement in the fires constituted an improper inducement which affected the voluntariness of his confession.
128 The second improper inducement was introduced by Corporal Deveau, who replaced Sergeant Taker after the latter had interrogated the respondent for about an hour. Corporal Deveau had been sitting in an adjacent motel room listening to and recording the interrogation conducted by Sergeant Taker. After again emphasizing the certainty guaranteed by the results of the polygraph, the police officer downplayed the seriousness of the offences that were the subject of the interrogation, and in particular the seriousness of the car fire (A.R. at pp. 513-18):
Q. . . . You know why I feel for you? Because you’re not a criminal. You’re a good guy, Richard. You’re a good guy, and these fires — they’re not all that bad fires. They’re old empty buildings in Waterville. They look like the shits. [A.R. at p. 513]
. . .
But you know, if it’s Claude Cook’s store you’re involved with — it’s a piece of junk on the corner. It’s an eyesore. That burned down the first time.... But let’s assume you burned it down the third time. What is really wrong with that? Probably not a whole lot. That store was a piece of junk. It was already burned.
. . .
So somebody burned it the third time. What have you hurt? Basically nothing, except that you set fire, that it went down. But if that’s the case, you tell me. And the — was it burned down because it was an eyesore? [A.R. at p. 515]
. . .
If it’s just the motor vehicle — there’s not, you know, there’s not a whole lot in a motor vehicle fire. There’s not a whole lot to hold into that but you got to tell me about that. If it’s the old building, tell me about the old building. [A.R. at p. 518] [Emphasis added.]
129 After being interrogated by Corporal Deveau for 30 to 40 minutes, the respondent confessed to setting the fire in his girlfriend’s car and provided the police with a statement. He was distraught at this time. He continued, however, to maintain his innocence in relation to the other fires. In the face of these denials, Corporal Deveau returned briefly to the “infallible polygraph” theme, this time specifically with reference to the building fires. He then placed the respondent under arrest and transported him to the New Minas Detachment. En route, the respondent was visibly upset and was observed by Corporal Deveau to be crying. In the three-hour interrogation that followed the respondent’s initial confession to having set fire to his girlfriend’s car, the minimization theme crystallized. Corporal Deveau minimized the seriousness of the additional legal consequences that would flow from a confession to the building fires. He explicitly suggested that the fires could be “packaged” or “bundled”, the clear implication being that the respondent would not be that
much worse off by admitting to all the fires now that he had admitted to one:
Q. If you done the other ones this — or some of the other ones this is the time — this is the time to just get them off your chest. This is the perfect opportunity because of what you’ve already told us, okay. And everybody can see this, that it’s — You didn’t do one fire and then ten years down the road you did — this is a series of fires we’ve been having in Waterville. So we can look at it — we can look at it as a one-package type of thing. [A.R. at p. 552]
. . .
Q. [T]his is one in the bundle, we can all have – rope it together and it’s not an isolated situation here... Richard. This is a big bundle. This is a big cancer here and you’ve chopped a bit of it off. But if there’s more here this is the time to get rid of it because it’s just — it can be dealt almost as one
incident and people can understand that. [A.R. at p. 575]
. . .
Q. . . . And if you’ve got anything to do with it this is the time for you to tell us because like I said it’s all — it’s a big bundle there. It’s altogether and we can look at it as almost as one incident. There’s a number of incidents but it all happened in general — it’s all happened in the same area.... [A.R. at p. 577] [Emphasis added.]
130 Corporal Deveau persisted in minimizing the seriousness of the combined
offences (A.R. at pp. 630 and 643-44):
Q. . . . And you can help us in a big way (inaudible) ten, does not make a lot of difference.
. . .
Q. ... If we’re talking one or if we’re talking ten. You’ve climbed the first hurdle. You’ve done one good step. Okay. But you’re dealing with it like you done all the way if it’s one or if it’s ten, it doesn’t matter, okay.... You’re a good man. Let’s just deal with that. I know what you want, don’t you?
Q. You want to — don’t turn back now. We’re right here. Okay. We’re there. Let’s do it. Okay. I know all that. I’ve been at this for 17 years. Okay. (Inaudible) one or if it’s 10 or it’s 20. Okay.
A. Um. [Emphasis added.]
131 In short, it was suggested to the accused that it made no difference whether he had set one or ten fires and that if he confessed, the building fires could be bundled with the car fire. This was clearly improper.
C. Threat to Interrogate the Respondent’s Girlfriend
132 Shortly after the respondent’s initial confession, Corporal Deveau suggested to the respondent, for the first time, that confessing would be beneficial to his girlfriend
(A.R. at p. 574):
Q. You know, this whole thing is — we might even ask Tanya if she would take a polygraph on this because we don’t know where she stands, okay. [Emphasis added.]
This theme was pressed repeatedly. Specifically, the police officer was explicit that the respondent could spare his girlfriend the unpleasant experience of being polygraphed and interrogated if he would simply “come clean” (A.R. at pp. 603-4):
Q. Do you realize the other reason is that we — that you’ve got to come clean with everything with us is for Tanya.
Q. We don’t want to put Tanya through any — I mean she’s going to be going through enough trying to — we don’t want to — and I’m sure you don’t want her to get — to go through half of what you went through today.
It’s no fun.
A. No, no.
Q. It won’t be any fun for her. But in order for her to — in order for us to be one hundred percent we have to do it. So if there’s anything that you can tell us that can put her — that we say, okay, we don’t need you, Tanya, we have it here, you know, and we have some stuff. But we’re not convinced on everything else. So don’t put Tanya through that if there’s something you can tell me, okay.
A. About what?
Q. About anything — about any of the other fires or anything else at all. About your vehicle fire. There is something that you — [Emphasis added.]
133 It is apparent that it was known to the police that Tanya Kilcup was within the respondent’s “close circle”: see R. v. Middleton (1974), 59 Cr. App. R. 18 (C.A.), at p. 21. Earlier, referring to the respondent’s confession to setting fire to her car, Corporal Deveau had said (A.R. at p. 548):
Q. You’ve got a good girlfriend who you probably love very much.
A. (Nods “yes”.)
Q. Going to get married soon and — but this is not the end of — this is not the end. This is going to probably put you back a few pegs but it’s not
134 The police were also clearly aware that the respondent was already concerned about the implications of his earlier confession for his relationship with Ms. Kilcup. Consider the following exchange (A.R. at p. 555):
Q. . . . You’re going to have to live with your girlfriend. And I think that —
A. Maybe. She might leave me.
A. I can’t say for sure she’s not going to leave.
Q. No, but if there’s love there things are going to work out. . . .
135 Still more important, in providing a plain indication that the respondent wanted to avoid entangling his girlfriend with the police, are the respondent’s concerned reaction to the suggestion that she might be considered a suspect, despite his earlier, firm denial that Ms. Kilcup had any involvement in or knowledge of the car fire, see A.R. at p. 533, and his insistence that if she had any part to play it was limited to one of alibi:
Q. Okay. We’re going to have — on these things that you’ve told me, okay, so we’re going to have to check everything out, okay. So somebody is on the way up to talk with Tanya. Tanya has got to be spoken to, okay.... Looking at the situation we, you know, we don’t know what Tanya — if she knows anything or if she doesn’t. You have said she doesn’t know anything. We’re going to have to confirm that, okay.... What is Tanya going to say?
A. Probably it’s over.
Q. Well, just —
A. I don’t know. She’s not going to be impressed.
Q. Yeah. I’m sure she can — you two can probably work things out. Huh?
A. If she talks to me.
Q. I’m sure she’ll talk to you. . . .
A. More mad than anything probably. [A.R. at pp. 570-71]
. . .
A. She’s not involved in her car, and neither one of us are involved in the buildings. Well, if she does (inaudible), I can’t tell you. I highly doubt it.
Q. Yeah, I would doubt it too. And her car, she has no knowledge at all? You never talked to her about that or anything?
A. All she knew is verify where I was.
Q. Uh-hum. Well, we’ll check that out. [A.R. at p. 604]
. . .
Q. . . . I mean that’s got — from a law enforcement perspective, you know, was it you and her? Just you? Just whoever? Okay?
A. Just me.
Q. Okay. I mean we have to go and — we asked Cst. Taker to talk to Tanya, okay. (Inaudible).
A. But I didn’t tell her.
A. I didn’t tell her.
A. Totally by myself. [A.R. at p. 611]
136 The respondent’s concern to protect his girlfriend was known to the police
and they used it to induce him to confess. Corporal Deveau indicated that the police would have to consider her a suspect and that they would have to polygraph and interrogate her if the respondent’s confession was not forthcoming. Following the first confession, Corporal Deveau explored with the respondent the possibility that the car was set on fire in order to collect the insurance. It was her car. The threat to interrogate Ms. Kilcup remained effective throughout the post-test interview because, as Corporal Deveau had made clear to the respondent, the police did not intend to polygraph and interrogate Ms. Kilcup that very evening. The clear suggestion was that the respondent could spare her his predicament by admitting his sole involvement in all the fires. In my view, the relationship between the respondent and Ms. Kilcup was such that the threats to implicate her placed impermissible pressure on the respondent to confess.
137 In sum, it is my opinion that the foregoing representations constituted threats, promises and inducements, within the meaning of the confessions rule and, when combined with the prevalent ambiguity concerning just what was and was not admissible in court against the respondent, as well as the oppressive atmosphere created by the “infallible” polygraph test, they are sufficient to raise a reasonable doubt as to voluntariness of the respondent’s confessions, first, to the car fire, and, later, to the rest. The combination of the lies and misrepresentations, which are not impermissible, with the inducements, which are, in my view caused the respondent to make involuntary admissions. The few instances in which he appeared to reject the representations made to him by the police officers were little more than desperate bravado and vain attempts to delay what he seemed to view as the inevitable fact that he would have to confess. The following excerpts are indicative of the respondent’s state of mind:
Q. . . .You’re sitting there, Richard, and denying is not going to help this problem. I can tell you right now. It’s only going —
A. All I’m supposed to say, yeah, I did this, this, and this, and —
A. — whether I did it or not and —
Q. No, no, no.
A. — it doesn’t matter.
Q. No, that’s not right. That’s not right.... What I’m saying to you, something here that you’ve told me today is not truthful. I’m not saying all this is not truthful. I’m saying something on this piece of paper that you’ve told me here today isn’t — you haven’t been truthful with me. [A.R. at p. 483]
. . .
A. Uh-huh. You’ll tell (inaudible) that I failed it and —
Q. And these are the questions I asked.
A. And they’ll go off to wherever they want to go and away they go.
Q. Basically — right now that’s all I know.
A. So in other words, I’m screwed.
Q. No. What I’m saying to you — I think sometimes you’re missing what I’m saying, Richard. Just because you lit a car fire, doesn’t mean that you’re going to go to jail for the rest of your life. It definitely does not mean that. Right?
A. I didn’t think it would.
Q. Well, you’re thinking some awful strong things here. You know —
A. No, I’m thinking that unless I say okay, I did the car or whatever —
A. Bruno [Deveau] will be at my door until they can prove it. [A.R. at p. 502] [Emphasis added.]
V. Other Basis for Exclusion: Fair Trial Considerations
138 Polygraph evidence is inadmissible in a criminal trial. See R. v. Béland,  2 S.C.R. 398, at p. 416; see also Phillion v. The Queen,  1 S.C.R. 18. As the polygraph evidence was tendered by the defence in Béland, the majority of the Court excluded it on the basis of several well-established rules of evidence (i.e., the rule against oath-helping, the rule against past consistent statements and the rule relating to character evidence, at p. 417). More importantly, however, for the purpose of this appeal, the majority also rested its decision to exclude the polygraph evidence on the policy of law which dictates that, as a general rule, credibility is a matter solely for the trier of fact (see R. v. Marquard,  4 S.C.R. 223, at p. 248; R. v. Mohan,  2 S.C.R. 9, at p. 23); and on an assessment of the unacceptable cost, in relation to the fairness and efficiency of the trial process, of admitting polygraph evidence (see Mohan, supra, at p. 21). In this regard, McIntyre J. for the majority wrote (at pp. 415-18):
Here, the sole issue upon which the polygraph evidence is adduced is the credibility of the accused, an issue well within the experience of judges and juries and one in which no expert evidence is required. It is a basic tenet of our legal system that judges and juries are capable of assessing credibility and reliability of evidence....
[I]t is my view that the admission of polygraph evidence will serve no purpose which is not already served. It will disrupt proceedings, cause delays, and lead to numerous complications which will result in no greater degree of certainty in the process than that which already exists.
What would be served by the introduction of evidence of polygraph readings into the judicial process? To begin with, it must be remembered that however scientific it may be, its use in court depends on the human intervention of the operator. Whatever results are recorded by the polygraph instrument, their nature and significance reach the trier of fact through the mouth of the operator. Human fallibility is therefore present as before, but now it may be said to be fortified with the mystique of science. Then, it may be asked, what does it do? It provides evidence on the issue of credibility of a witness. This has always been a collateral issue and one to be decided by the trier of fact.
139 The admission into evidence of a confession, given in circumstances where it is intimately linked to a “failed” polygraph test, as it was in this case, is wholly inconsistent with the holding of this Court in Béland and the principles and policy that inform it. Further, it is my opinion that it severely, and unjustifiably, prejudices an accused’s fair trial interests.
140 Where, as here, the accused’s confession is intertwined with a “failed” polygraph test, he is presented with an impossible dilemma at trial. See R. v. Amyot (1990), 58 C.C.C. (3d) 312 (Que. C.A.), at p. 325; see also R. v. L.E.W.,  O.J. No. 753 (QL) (Gen. Div.), at para. 26. On the one hand, if the accused wishes to repudiate his confession, as he is entitled to do, he must be able to inform the trier of fact of all of the circumstances in which the confession was made. See DeClercq v. The Queen,  S.C.R. 902, at p. 921; L.E.W., supra, at para. 21; J. Sopinka, S. N. Lederman and A. W. Bryant, The Law of Evidence in Canada (2nd ed. 1999), at §§ 8.83-8.84. As indicated by the Court of Appeal, in the case at bar this would include: “all of the circumstances leading up to the taking of the test, the examination itself, the manner in which the failure of the test was conveyed, as well as the subsequent interrogation”: R. v. Oickle (1998), 164 N.S.R. (2d) 342 (C.A.), at para. 50. The trier of fact cannot otherwise properly appreciate whether the confession may be untrue and accept the accused’s repudiation of his out-of-court statement. See L.E.W., supra, at para. 26; see also Sopinka, Lederman and Bryant, supra, at § 8.84 (quoting R. v. Murray,  1 K.B. 391, at p. 393); R. v. Charrette,  O.J. No. 2509 (QL) (Gen.
Div.), at para. 28; R. v. Whalen,  O.J. No. 3488 (QL) (C.J.), at para. 26.
141 A confession has often been referred to as “the highest and most satisfactory proof of guilt”: Bigaouette v. The King (1926), 46 C.C.C. 311 (Que. K.B.), at p. 320 (quoting R. v. Lambe (1791), 2 Leach 552, 168 E.R. 379); see also R. v. Hodgson,  2 S.C.R. 449, at paras. 14 and 17; DeClercq, supra, at p. 922; R. v. Warickshall (1783), 1 Leach 263, 168 E.R. 234, at p. 235; R. v. Hardy (1794), 24 St. Tr. 199, at pp. 1093-94; R. v. Baldry (1852), 2 Den. 430, 169 E.R. 568, at p. 574; R. v. Guidice,  W.A.R. 128 (Ct. Crim. App.), at p. 130; Wigmore on Evidence (Chadbourn rev. 1970), vol. 3, §§ 820b-820c, at pp. 301-6. As Wigmore explains, § 820b, at p. 303, this is because
[t]he confession of a crime is usually as much against a man’s permanent interests as anything well can be; ... no innocent man can be supposed ordinarily to be willing to risk life, liberty, or property by a false confession. Assuming the confession as an undoubted fact, it carries a persuasion which nothing else does, because a fundamental instinct of human nature teaches each one of us its significance.
Where, as here, the confession has been recorded, there can be little doubt that it was made. The only remaining issue for the jury is whether it is reliable as proof of guilt. It is because of its “conclusive effect with respect to guilt” that the criminal standard of proof beyond a reasonable doubt is applied to the question of voluntariness, whereas proof on a balance of probabilities is the relevant standard with respect to the admissibility of evidence generally. See R. v. Egger,  2 S.C.R. 451, at p. 474; Sopinka, Lederman and Bryant, supra, at § 8.87. Given the overwhelming weight that jurors are likely to attribute to confessions, as a simple matter of human intuition, and, relatedly, given the skepticism and suspicion with which they will normally approach the proposition that a person may, absent torture, falsely confess to a serious crime, it must be open to the accused to exhaustively explain any inducements or “discrediting circumstances”, Wigmore, supra, at § 820c, at p. 306, which may cast doubt on its truthfulness.
142 Where, as here, the confession and the polygraph are so intimately linked, the accused will unavoidably have to reveal having failed the polygraph if he wishes to cast doubt on the truthfulness of his confession. This, in my opinion, comes at too high a cost to the accused’s fair trial interests. The accused is placed in a position where he cannot, as is his right, seek to repudiate his out-of-court confession without effectively being forced to adduce highly prejudicial evidence that the Crown could not tender, and that will appear to bolster, rather than impeach, the reliability of his confession. See R. v. Thorne (1988), 41 C.C.C. (3d) 344 (N.S S.C., App. Div.), at p. 353; cf. R. v. McIntosh (1999), 141 C.C.C. (3d) 97 (Ont. C.A.), at para. 70. It is difficult to imagine a more devastating, self-inflicted blow to the accused’s credibility, led through useless, unreliable and therefore prejudicial evidence. The accused is, in effect, forced to incriminate himself by introducing an otherwise inadmissible piece of evidence that cannot help but strengthen what is often, as here, the sole evidence against him.
143 In Béland, supra, the majority relied, in part, on the danger that polygraph evidence will be misused and will distort the fact-finding process as a ground for barring it from the courtroom. See Béland, supra, at pp. 417-18; see also Mohan, supra, at p. 21. La Forest J. (concurring in the result) expressed this concern in terms of “human fallibility in assessing the proper weight to be given to evidence cloaked under the mystique of science”: Béland, supra, at p. 434. Like some similar fact evidence, polygraph evidence is likely to generate a perception of reality which wholly over-reaches the mark.
144 While “[t]he possibility that evidence will overwhelm the jury and distract them from their task can often be offset by proper instructions” as to the limited use to which the evidence can be put, Mohan, supra, at p. 24, I cannot accept that this is so where, as here, the “expert” evidence, though not admitted for that purpose, goes directly to the credibility of the accused and, purportedly as an admission of guilt, to the ultimate issue for the jury to decide. See Thorne, supra; cf. McIntosh, supra, at paras. 68-70. To paraphrase the majority in Marquard, supra, at p. 248, the questions of credibility and guilt or innocence are difficult ones and polygraph evidence, cloaked as it is in the mystique of science, may be all too readily accepted by a frustrated or uncertain jury, in abdication of its duty to determine credibility and render a true verdict, as a convenient basis upon which to resolve its difficulties.
145 On the other hand, the alternative open to the accused, namely to deny the veracity of his confession without the benefit of informing the trier of fact of all the circumstances in which it was made, is equally problematic. In this case, for instance, no amount of editing could realistically remove the constant references, during the six-hour interrogation, to the respondent’s failed polygraph test. Moreover, there would be no basis upon which to understand why he would have confessed falsely without the
polygraph test as the trigger.
146 I see the “failed” polygraph test as distinct from other inadmissible evidence and from other highly incriminating evidence, such as DNA. If an accused confesses after having been confronted with incriminating DNA evidence, there is no unfairness in the accused having to relate the circumstances under which the confession was made, including having to refer to the DNA evidence itself. Unlike evidence of a “failed” polygraph test, DNA evidence is not as such inadmissible in a criminal trial. See R. v. Terceira (1998), 123 C.C.C. (3d) 1 (Ont. C.A.), aff’d  3 S.C.R. 866. In the DNA scenario the accused will merely be referring to evidence that may well be fully before the trier of fact in any event. In such a case, he lends little, if any, additional credibility or weight to his confession. In contrast, where an accused is forced to adduce “failed” polygraph evidence, he is effectively introducing a piece of evidence which will serve to bolster and confirm the probative value and credibility of his confession. I do not have to decide on the legality of deliberately using inadmissible evidence to secure a confession in all cases. Suffice it to say, for present purposes, that a “failed”
polygraph test is likely to be perceived as simply a confession by another name. Given the unparalleled weight attributed to confessions, I believe that the prejudicial effect that flows from an accused’s reference to his “failed” polygraph test is overwhelming.
147 In my view, therefore, confessions should be excluded where the accused, because of the intimate causal and temporal connection between a “failed polygraph” and a subsequent confession, is unable to demonstrate fully the impact of the circumstances surrounding its making without inevitably introducing the polygraph evidence (e.g., the fact of the test, the fact that the accused “failed” the test, the representations made to him, etc.). In this light, I agree that if the police [translation] “are to persist in having recourse to this investigation technique,... the administration of the [polygraph] test [must] be kept completely separate from the questioning process that may result in the obtaining of a confession”: Amyot, supra, at p. 325, per Proulx J.A.; see also R. v. Nugent (1988), 84 N.S.R. (2d) 191 (S.C., App. Div.), at pp. 212-13.
148 In the case at bar, the respondent’s statements were inextricably linked to his “failure” of the polygraph test. Crown counsel made this clear on the voir dire. He stated
(A.R. at p. 60):
The, the result is obviously no evidence of anything. It is no evidence of the truthfulness or lack of truthfulness of anything Mr. Oickle said clearly but it is, it is part of the scenario. It was because of the perceived result of the polygraph test that other things happened. It will be absolutely impossible for Your Honour to, to examine all of the communication between Mr. Oickle and the police without hearing, practically all of this is audio taped and video taped. It would be impossible for Your Honour to hear it without hearing over and over and over again references to the polygraph results. [Bold and underlining added.]
149 The record clearly supports Crown counsel’s position. The audio-tape recording of the interrogation leading up to the first confession is replete with references to the fact that respondent had not been truthful, that his heart had confessed his involvement and that his denials were futile. The “failed” test is therefore absolutely integral to a proper appreciation of the voluntariness, for voir dire purposes, or the veracity, for trial purposes, of the first confession. The same is true with respect to the second confession. Although there are only a limited number of references to the “failed” polygraph test after the first confession, it remained a critical element of the context and circumstances of the second confession. It powerfully enhanced the force and effect of the conventional inducements held out by the police by undermining the respondent’s will.
VI. Conclusion and Disposition
150 In my opinion, the respondent’s statements should be excluded on a dual basis: first, the statements were obtained as a result of fear of prejudice or hope of advantage held out by persons in authority. At the very least, the trial judge had to have a reasonable doubt about the application of the classic voluntariness rule. Second, I would also find the statements inadmissible on the basis that the manner in which they were obtained by the police places the respondent in the unfair position of having to lead prejudicial, unreliable and inadmissible evidence against himself in order to impeach the veracity of the statements obtained.
151 Having found that the respondent’s statements were involuntary and must be excluded, I agree with the Court of Appeal that “[t]he re-enactment was a continuation of the earlier statements” (para. 154) and was, therefore, also improperly admitted by
MacDonald Prov. Ct. J.
152 For these reasons I would dismiss the appeal, set aside the convictions and enter acquittals on all counts.
Appeal allowed, Arbour J. dissenting.
Solicitor for the appellant: The Attorney General of Nova Scotia, Halifax.
Solicitor for the respondent: The Nova Scotia Legal Aid, Sydney.
Solicitor for the intervener the Attorney General for Ontario: The Ministry of the Attorney General, Toronto.
Solicitors for the intervener the Criminal Lawyers’ Association (Ontario): Sack Goldblatt Mitchell, Toronto.