Corps de l’article

Introduction

The common law confessions rule—the requirement that the prosecution prove the voluntariness of an accused’s statements to persons in authority—has a close relationship with the constitutional principles that govern criminal trials. Thus, the Supreme Court of Canada’s recent recognition of the confessions rule as a principle of fundamental justice[1] is unsurprising, particularly since doctrines that were less securely rooted in the common law—such as the right to silence—have already been recognized as principles of fundamental justice under section 7 of the Canadian Charter of Rights and Freedoms.[2] But the constitutionalization of the confessions rule raises questions of both structure and content. The structural question involves the relationship between the common law rules of evidence and the accused’s constitutional rights. The Supreme Court of Canada has adopted two different approaches to this structural question. Under what I will call the “rule of evidence” approach, the common law rules governing the admissibility of the accused’s statements are understood as an aspect of the accused’s procedural right to a fair trial under section 7. Under what I will call the “rights violation” approach, the conduct of the state in obtaining an involuntary statement is treated as a violation of the accused’s substantive constitutional rights. The question of content has to do with the strength of the protection offered by the constitutionalized confessions rule and with its relationship to other constitutional rights.

The question of structure and the question of content may seem separate in principle, but in this paper I argue that the Supreme Court of Canada may have weakened the content of the confessions rule in part because it has structured the constitutional version of the rule as a rights violation. Specifically, in Singh, the Court appears to have constitutionalized the confessions rule by linking it tightly to the pre-trial right to silence. This approach conflates the distinct protections offered by the right to silence on the one hand, and the confessions rule on the other. While both doctrines are branches of the overarching principle against self-incrimination, neither one is reducible to the other. The pre-trial right to silence protects a detainee’s right to choose whether to speak to the police at all; the confessions rule protects the accused from the prosecutor’s use of his or her involuntary statements used at trial. While both rules lead to the exclusion of statements at trial, they do so by very different routes: the exclusion of a statement obtained in violation of the right to silence is a constitutional remedy under section 24 of the Charter, while the exclusion of an involuntary statement is simply a rule of evidence intended to protect the trier of fact from hearing potentially unreliable information. Thus, if the confessions rule is to be constitutionalized as a principle of fundamental justice under section 7 of the Charter, it should be treated as an aspect not of the right to silence but of the right to a fair trial. Fortunately, the Court’s recent decisions concerning the confessions rule do not foreclose this method of incorporating the confessions rule into section 7 because these decisions can be read not as changing the content of the common law rule but as instances of appellate deference to fact-finding on voir dires.[3] As a constitutionalized rule of evidence, a robust common law confessions rule can still take its proper place—alongside the pre-trial right to silence under section 7—as an aspect of the constitutional principle against self-incrimination.

I. The Principle Against Self-Incrimination

It has often been said that respect for human dignity has been an important organizing principle of constitutional law since the Charter came into force.[4] Although courts have been unwilling to recognize respect for human dignity as a principle of fundamental justice in itself under section 7, the idea of human dignity provides a normative benchmark for specific Charter rights.[5] The precise meaning and scope of the requirement that the state respect human dignity are neither wholly clear nor uncontested. But, at a minimum, respecting human dignity must mean that the state has an obligation to treat each individual as an end and not as a means to, or a resource to be exploited for, achieving the state’s ends or the ends of other individuals.

The principle against self-incrimination is a very basic norm for a system of criminal justice in a constitutional order that is committed to human dignity. It is a well-recognized “principle of fundamental justice” within the meaning of section 7.[6] The core idea of the principle is that when the state uses its power to prosecute an individual for a criminal offence, the individual ought not to be required to assist the state in the investigation or trial of the offence. And it is striking that the Supreme Court of Canada has identified the principle against self-incrimination as “[p]erhaps the single most important organizing principle in criminal law,”[7] and has linked it with many rules of evidence and procedure. Instances of this principle include the express protection against self-incrimination in section 13 of the Charter,[8] the rule against the Crown’s splitting its case,[9] the common law confessions rule,[10] the Charter right to silence,[11] the doctrine of derivative-use immunity under section 7,[12] and the line between permissible and impermissible uses of the state’s power to compel the production of information.[13] This constellation of common law and constitutional rules provides powerful protection against the state’s use of the testimony of suspects and accused persons against their will in the investigation and prosecution of criminal offences.[14]

The limit on state power imposed by the principle against self-incrimination is supported by both a principled and a pragmatic consideration. The principled consideration is rooted in the requirement that the criminal process respect the basic human dignity of those who are subject to it. To allow the state to force a suspect or an accused person to testify in support of the state’s case against him or her would be to treat this person as a mere means to the state’s objectives (particularly the prevention and punishment of crime), rather than as an end. This consideration, though obviously applicable to the factually innocent, may not seem very compelling where the accused individual is guilty of a serious crime, particularly one that involved an attack on the dignity and worth of others. But the presumption of innocence means precisely that the system of criminal justice must assume that the accused individual is not guilty. It must therefore protect the dignity to which an innocent person is entitled until guilt is established by proper means. The fact that the person may, at the end of the process, be found guilty of a serious crime does not change the normative requirement that the state respect the principle against self-incrimination during the process itself.

The pragmatic consideration supporting the principle against self-incrimination is simply that forcing the accused to provide testimony is unlikely to produce a verdict that is more accurate or just than it would be without his or her forced testimony. The common law has long recognized that coercive police tactics are likely to produce statements from suspects that are designed to satisfy the police and alleviate the coercion rather than to be truthful and reliable.[15] More recently, social scientists and other observers of the justice system have learned that some modern techniques of interrogation, though not as obviously coercive as violence and threats, may also have the kind of coercive effect that can produce unreliable confessions.[16] A coerced confession admitted into evidence at trial has a substantial prejudicial effect on the trial process because jurors find it difficult to accept the possibility that a person would falsely admit to a crime. They therefore tend to overvalue a coerced confession.[17]

This normative understanding of the principle against self-incrimination should be kept in mind in any discussion of the content and the constitutional status of the confessions rule. The principle against self-incrimination does not mean that a suspect is prohibited from providing incriminating statements; it means that the state cannot require a suspect to incriminate himself or herself. Where a suspect freely chooses to co-operate with the state’s investigation, even to the point of admitting full responsibility for the crime, the principle against self-incrimination is not infringed. The difficulty, of course, is in deciding what exactly it means for the suspect to freely choose to co-operate. The confessions rule provides part of the answer. It is intended to ensure that the content of a suspect’s statement to a person in authority reflects his or her own reasons for speaking, such as a genuine desire to confess or at least to talk about the offence at issue, and not by extraneous reasons such as responding to improper conduct by state agents. The common law confessions rule therefore excludes statements made in response to threats and promises because of the likelihood that the suspect was motivated by the desire to avoid the threatened consequence or to obtain the promised benefit, rather than by the desire to speak for his or her own reasons. The constitutional right to silence provides another part of the answer. It is intended to protect the accused’s right to choose whether to speak to the authorities at all.

The overarching principle against self-incrimination cannot, by itself, determine the precise content of the confessions rule, or of any other rule of evidence or procedure. There may be institutional differences in the way that legal systems manifest this principle. There may also be reasonable disagreement about the extent to which any given rule of evidence or procedure is consistent with this principle. Consider, for example, the Canadian holding that neither the common law confessions rule nor the section 7 right to silence applies to a suspect’s statement where the suspect is not in custody and does not know that the listener is a person in authority.[18] On the one hand, it might be argued that this rule of evidence complies with the principle against self-incrimination in that the suspect who freely, albeit foolishly, confesses to an undercover state agent is not being used by the state; rather, the state is simply taking advantage of the suspect’s voluntary decision to furnish evidence against himself or herself, as if he or she had been speaking to another private individual. On the other hand, it might be argued that this rule of evidence violates the principle against self-incrimination in that the state uses deceit and trickery to deprive the suspect of the ability to make a properly informed decision about whether to speak.

The principle against self-incrimination does not fully determine this issue, so there may be room for pragmatic considerations to come into play. For example, if undercover police tactics are likely to produce unreliable statements, that fact would count against the Canadian rule. Furthermore, there is undoubtedly room for other rights and institutional factors to influence the content of the rule. The significant duties imposed on the police as a consequence of the subsection 10(b) right to counsel, for example, may affect the content of the section 7 right to silence. Once the police have complied with their subsection 10(b) duties and the detained suspect has spoken to counsel, it is assumed that he or she has been vigorously advised to remain silent. Consequently, the positive duties on the state to respect the right to silence are perhaps less demanding than they would be if the suspect did not have a strong right to confer with counsel on arrest.[19] Pragmatic factors may even influence the content of a right; if a particular way of specifying the right eases the task of law enforcement without encroaching on the core of the principle against self-incrimination, there would be a pragmatic argument for specifying the right that way.

But it would be a mistake to take this pragmatic point as a general warrant to articulate the confessions rule, or any other rule, in a way that makes it easier to get a statement from a suspect. The goal of law enforcement is not to get statements as such, but to get reliable information. There is reason to believe that the more pressure the police exert on suspects and the more deceit and trickery they use, the less reliable the resulting statements will be.[20] Moreover, the pragmatic point is secondary. Since the principle against self-incrimination is part of the structure of a rights-based system of criminal justice, whatever minimum demands it places on investigative techniques have to be respected.[21] Any system of justice that takes the dignity and worth of the individual seriously must uphold some version of the principle against self-incrimination.

Thus, while there may be different ways of specifying the confessions rule, either at common law or as a constitutional principle, the point is not to make it easier for the police to obtain statements, but to ensure that the suspect is able to make his or her own informed decision whether to speak to the police. In light of this normative goal, I consider the recent history of the confessions rule in Canada in Part II, two ways in which the confessions rule might be constitutionalized in Part III, and the current status of the confessions rule under the Charter in Part IV.

II. The Confessions Rule Restated

The common law confessions rule, as it stood in Canada before 2000, can be briefly stated as follows: if an accused person has given a statement to a person in authority, the Crown must prove beyond a reasonable doubt that the statement was made voluntarily before using the statement for any purpose in the accused’s trial.[22] Whether the person receiving the statement is a “person in authority” is assessed on the basis of the accused’s reasonable perception of that person’s role in the criminal process.[23] There are three, or possibly four, ways in which a statement can be involuntary: (1) it was obtained by means of an inducement (a threat or a promise);[24] (2) it was not the product of the accused’s operating mind;[25] (3) it was obtained by oppressive conduct by the police;[26] or possibly (4) it was obtained via a police trick that would shock the conscience of Canadians.[27] If involuntary, the statement is inadmissible for any purpose. However, physical evidence derived from an involuntary statement is admissible.[28] Moreover, even an involuntary statement is admissible to the extent that it is confirmed by that physical evidence.[29]

In R. v. Oickle, decided in 2000, the Supreme Court of Canada restated the confessions rule on a principled basis.[30] The accused was charged with several counts of arson. He made two detailed statements to the police describing his involvement in the fires in question. Without these statements, the Crown’s case would not have risen much above the level of suspicion. The methods the police used to obtain these statements are not central to the current discussion, though they are fascinating and illustrative of the controversial interrogation tactics now routinely employed to obtain confessions.[31] The trial judge found that the statements were voluntary, admitted them, and convicted the accused. On appeal, the Nova Scotia Court of Appeal ordered an acquittal. Justice Pugsley and Justice Cromwell, for the court, found that there were several factors that, taken singly or in combination, made the statement involuntary. These factors fell under the first (inducement) and third (oppression) branches of the common law rule.

The Supreme Court of Canada allowed the Crown’s appeal and restored the conviction. The Court was cautious in its characterization of the relationship between the common law rule and the Charter, but was quite bold in reformulating the confessions rule in accordance with the principled approach to the common law rules of evidence. Justice Iacobucci, writing for the majority, did not reject the four branches of the confessions rule as they have developed in Canadian law. He confirmed that a threat or promise, an atmosphere of oppression, a failure of the operating mind, or an unacceptable police trick could make a statement involuntary.[32] But he did state that the first three branches of the rule should not be considered in isolation from one another. In accordance with the principled approach, a trial judge’s analysis of voluntariness should be “contextual” and should consider all branches of the rule together:

[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 ... On the other hand, where the suspect is treated properly, it will take a stronger inducement to render the confession involuntary.[33]

Moreover, Justice Iacobucci emphasized the need for appellate deference to the trial judge’s ruling on voluntariness: “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.’”[34]

Oickle clearly requires trial judges to consider all of the circumstances in making a ruling on voluntariness. It also makes it easier for the Crown to establish voluntariness by requiring proof, not of the absence of a threat or promise, but of the absence of a quid pro quo for the accused’s statement.[35] But, apart from that, Oickle does not change the test for voluntariness. The questions that the judge must ask remain the same: Was the statement induced by a threat or promise, or by oppressive circumstances? Did the accused have an operating mind? Did the police use an unacceptable trick to obtain the statement? The adoption of a principled approach to voluntariness means only that these questions have to be addressed in light of all the circumstances. As Justice Iacobucci’s examples suggest, an inducement that appears insignificant on its own might, in the context of harsh treatment, render a statement involuntary; similarly, an inducement that seems important on its own might not make a statement involuntary if there is no suggestion of oppression.[36] Moreover, the two main purposes of excluding involuntary statements remain unchanged: to keep unreliable evidence away from the trier of fact and to protect suspects from improper investigative techniques.[37]Oickle does, however, emphasize the first purpose over the second.

In Spencer, a majority of the Supreme Court of Canada suggested a very different reading of Oickle. The accused, Spencer, was charged with offences arising out of eighteen robberies. While in custody, he admitted to a police officer his involvement in these robberies. At trial, he argued that the statement was involuntary because it was induced by a threat that the police would charge his girlfriend unless he made a statement, and a promise that he would be allowed to see her if he made a statement. After a lengthy voir dire, the trial judge found that the statement was voluntary. He stated the test for voluntariness as follows: “The question is whether the inducement, standing alone or in combination with other factors, is strong enough to raise a reasonable doubt about whether the free will of the accused was overborne.”[38] With respect to the alleged threat, the trial judge found that the constable had merely “appealed to Mr. Spencer’s common sense and knowledge of the justice system.”[39] With respect to the alleged promise, the trial judge found that, in the context, the inducement was not strong enough to make the statement involuntary. Accordingly, the statement was admitted and the accused was convicted as charged.

The British Columbia Court of Appeal allowed the accused’s appeal and ordered a new trial. Justice Donald, for the majority, held that the trial judge had applied the wrong test for voluntariness. He focused squarely on the trial judge’s invocation of the “overborne will” and held that this language was inappropriate where an inducement was in issue. The question was not whether the suspect’s will had broken down but whether there was an inducement: “Whether the suspect is strong or weak, robust or timid, a clear inducement of leniency gives rise to the same concern of a false confession.”[40]

The Supreme Court of Canada allowed the Crown’s appeal and restored the conviction. Both the majority and the minority framed the issue as whether the trial judge had correctly applied Oickle. Justice Deschamps, for the majority, deferred to the trial judge’s findings, commenting that “[g]iven the highly fact-sensitive nature of this area of the law and the duration of the interview, the precise content of the exchanges [between the accused and the constable] is important.”[41] She did not specifically restate the test for voluntariness; her reasons may be read as an endorsement of either the trial judge’s view that the issue is whether the will of the detainee has been overborne[42] or of the more traditional view that the issue is whether the statement was induced by a fear of prejudice or a hope of advantage.[43] Justice Deschamps concluded that “it is the strength of the inducement, having regard to the particular individual and his or her circumstances, that is to be considered in the overall contextual analysis into the voluntariness of the accused’s statement.”[44] Since the trial judge had not applied the wrong test, his factual determinations were entitled to deference on appeal.[45]

Justice Fish, for the dissent, held that the trial judge had indeed applied the wrong test; the question of whether the accused’s will was overborne related only to the operating mind branch of the test.[46] As this case involved an inducement, the question was whether it “would not have been made but for an improper inducement.”[47] Justice Fish then reviewed the record and found that the statement was involuntary because it was induced by the threat that the accused’s girlfriend would be charged unless he made a statement, and by the promise that he would be allowed to see her if he made a statement, which he characterized as “an implicit but unmistakable threat accompanied by an implicit but unmistakable promise.”[48]

Spencer is significant because the Court upheld the trial judge’s finding that the statement was voluntary, notwithstanding his arguably erroneous statement of the test for voluntariness. Since the majority’s reasons are not very clear about the proper test, Spencer can be read either as endorsing the trial judge’s statement of the test or as merely deferring to his factual findings. On the former reading, all four traditional branches of the confessions rule now reduce to one question: when the accused made the statement in issue, was his or her will overborne? On the latter reading, the question of whether the accused’s will was overborne may be part of the inquiry, particularly on the operating mind branch of the rule, but it is not determinative.[49]

Since the confessions rule is an aspect of the general principle against self-incrimination, it is clearly preferable to read Oickle and Spencer as emphasizing deference to the trial judge rather than as changing the test for voluntariness. Equating involuntariness with “overborne will” inadequately reflects the norm against self-incrimination because it permits a finding of voluntariness precisely when the accused’s decision to speak, or the content of his or her speech, is a considered response to a threat or promise or to coercive circumstances. In this situation, the accused’s will is not overborne, but his or her genuine reasons for speaking to a person in authority have been undermined by inducements or oppression. But before considering this point further, it is necessary to consider the constitutional status of the confessions rule in Canada.

III. Two Approaches to Constitutionalizing the Confessions Rule

There are several ways in which the common law confessions rule could be recognized as a constitutional principle, but the Supreme Court of Canada’s decisions present two main alternatives: the rule of evidence approach and the rights violation approach. The first alternative is to constitutionalize the confessions rule as a rule of evidence at trial: the accused’s statement to a person in authority would be inadmissible unless the Crown proved it voluntary beyond a reasonable doubt, as a principle of fundamental justice under section 7 of the Charter. Put another way, the common law exclusionary rule would be restated as an aspect of the accused’s constitutional right to a fair trial. The second alternative is to characterize the state’s conduct in obtaining an involuntary confession as a violation of the accused’s rights in the investigative process. The statement would then be evidence “obtained in a manner that infringed or denied”[50] the accused’s Charter rights, so its admissibility would be determined under subsection 24(2) of the Charter. These two methods of recognizing the common law confessions rule as a constitutional principle are concerned with the accused’s rights at different points in the process: under the rule of evidence approach, the constitutionalized confessions rule is linked to the accused’s rights at trial, while under the rights violation approach, the constitutionalized confessions rule is linked to the accused’s rights in the investigative process.

A. G.(B.): The Rule of Evidence Approach

The rule of evidence approach is central to the Supreme Court of Canada’s reasoning in R. v. G.(B.), the only case in which the Court has considered a conflict between a statute and the confessions rule.[51] The Court resolved the conflict by an exercise of statutory interpretation premised on the assumption that the confessions rule had been constitutionalized as a rule of admissibility under section 7, not on the assumption that obtaining an involuntary statement violated the accused’s Charter rights. In other words, the Court effectively adopted the rule of evidence approach.[52]

The accused was charged with sexual offences. On arrest, he made a statement to the police. The court referred the accused to psychiatric assessment pursuant to section 672.11 of the Criminal Code in order to determine his fitness to stand trial.[53] During the assessment, a psychiatrist asked him to explain his earlier statement to the police. In response, the accused made an incriminating statement. Two psychiatrists found the accused fit to stand trial. Following a voir dire, the trial judge found the accused’s statement to the police to be involuntary and inadmissible. His incriminating statement to the psychiatrist was a “protected statement” under subsection 672.21(1) of the Criminal Code and, pursuant to subsection 672.21(2), was inadmissible during the Crown’s case in chief. The accused testified in his own defence at trial and denied that the sexual activity in question had occurred. Paragraph 672.21(3)(f) provides that, notwithstanding subsection 672.21(2), a protected statement “is admissible for the purpose of ... challenging the credibility of an accused in any proceeding where the testimony of the accused is inconsistent in a material particular with a protected statement.” Since the accused’s testimony was inconsistent with the protected statement, the Crown cross-examined him on it pursuant to paragraph 672.21(3)(f). The trial judge found, among other things, that the accused’s testimony was not credible, and the accused was convicted.

At common law, the protected statement would have been inadmissible for any purpose because it was tainted by the earlier involuntary statement; however, according to paragraph 672.21(3)(f) of the Criminal Code, the protected statement was admissible for the purpose of impeaching his credibility. Thus, on the particular facts of G.(B.), there was a conflict between the common law rule and the statutory rule. Since statutes prevail over the common law, it would seem that this conflict should have been resolved in favour of the statute, and the trial judge’s ruling should have been upheld. But Justice Bastarache, for a majority of the Court, held that paragraph 672.21(3)(f) should be read as if Parliament intended it to operate consistently with the common law confessions rule.[54] While this conclusion might have been reached as a matter of statutory interpretation, Justice Bastarache took the stronger view that the admission at trial of an involuntary statement would violate section 7 of the Charter: “Since the protected statement in the instant case was inadmissible because of its degree of connection with the prior inadmissible confession, Parliament could not make it admissible for any purpose whatsoever without violating s. 7 of the Charter.”[55]

It was the admission of the protected statement at trial, not the conduct of the police in obtaining the involuntary statement (or of the psychiatrist in obtaining the protected statement), that offended section 7;[56] moreover, Justice Bastarache made no declaration of invalidity in respect of paragraph 672.21(3)(f) itself.[57] So G.(B.) supports the proposition that it is a principle of fundamental justice that a statement by an accused to a person in authority is inadmissible for any purpose at trial unless the Crown proves beyond a reasonable doubt that the statement was obtained voluntarily. According to this view, obtaining an involuntary confession is not, in itself, a violation of an accused’s Charter rights for which a remedy must be sought under section 24. Rather, the rule of evidence excluding involuntary confessions is a principle of fundamental justice, and it is the admission of the statement at the trial that creates a Charter violation—a violation that is avoided simply by applying the common law rule.

Chief Justice McLachlin, in dissent, held that the trial judge properly admitted the protected statement to challenge the accused’s credibility under paragraph 672.21(3)(f). She offered a number of reasons, each sufficient in itself, to support this conclusion.[58] For present purposes, the most significant holding was that section 7 of the Charter would not prevent the tainted protected statement from being admitted into trial. She saw the act of obtaining an involuntary statement as a violation of the section 7 right against self-incrimination.[59] The admissibility of the tainted statement would therefore be considered under subsection 24(2) of the Charter, and the protected statement would be admissible for the purpose of impeaching the accused’s credibility.[60] Her reasoning on the last point is in stark tension with the Court’s subsection 24(2) jurisprudence, but the point is that the majority of the Court rejected her adoption of the rights violation approach to the constitutionalization of the confessions rule.[61]

Thus, G.(B.) is a powerful authority in support of the rule of evidence approach. The majority not only relied on this approach, but also specifically rejected the rights violation approach adopted by the dissenting judges. Notwithstanding the adoption of the rule of evidence approach in G.(B.), more recent decisions on the confessions rule suggest that the Court may now favour the rights violation approach.

B. Singh: The Rights Violation Approach

If G.(B.) represents the rule of evidence approach to constitutionalizing the confessions rule, then Singh represents the rights violation approach. It does so despite (or perhaps because of) the fact that the argument in Singh was framed as a constitutional claim about the right to silence, not as a common law claim about voluntariness. The accused was charged with second degree murder and identity was in issue. The victim had been struck by a bullet fired during a dispute between three men and the employees of a pub. The police had a photograph and a videotape of a man whom eyewitnesses had identified as the shooter, and they believed that this man was the accused. The police arrested the accused and charged him with the murder. A police sergeant interviewed the accused twice, for a total of about 117 minutes. He directed his questioning at obtaining admissions that would establish the accused’s identity as the shooter. Justice Charron, writing for the majority, summarized the facts surrounding the accused’s statements as follows:

Before the interviews, Mr. Singh was given proper Charter and official police warnings and spoke to counsel by phone and in person ... [When] the discussion turned to the incident in question, ... he provided some information regarding his presence at the pub on the night of the shooting, [but] he repeatedly denied his involvement and asserted his right to silence. He indicated that he did not want to talk to the police, that he had nothing to say, that he knew nothing about the shooting, or that he wanted to return to his cell. Before Mr. Singh was shown the photographs in question and made the impugned admissions, he asserted his right to silence 18 times. Each time, the interviewing officer would either affirm that Mr. Singh did not have to say anything, or would explain to Mr. Singh that he had a duty or desire to place the evidence before him and he continued the interview.[62]

Ultimately, the accused made statements identifying himself in the photographs. The trial judge found that the accused’s statements were voluntary and that his section 7 right to silence had not been infringed. In particular, “Mr. Singh’s admission that he was in the still photograph came freely and did not result from the police systematically breaking down his operating mind or undermining his right to silence.”[63] The statements were therefore admissible (though only one of them was actually put before the jury), and the accused was convicted. On appeal, the accused conceded that the statements were voluntary, but continued to maintain that his right to silence had been violated. The British Columbia Court of Appeal rejected this submission and upheld the conviction. The Supreme Court of Canada dismissed the accused’s further appeal by a five-to-four majority, with Justice Charron writing for the majority and Justice Fish writing for the dissent. The majority found no legal error in any of the lower courts’ decisions and no reviewable error in the trial judge’s factual findings on the voir dire. The dissent found that the police conduct violated the accused’s section 7 right to silence. Although Justice Charron did not articulate a clear test to establish a breach of the right to silence, she did comment that

the law as it stands does not permit the police to ignore the detainee’s freedom to choose whether to speak or not ... Under both common law and Charter rules, police persistence in continuing the interview, despite repeated assertions by the detainee that he wishes to remain silent, may well raise a strong argument that any subsequently obtained statement was not the product of a free will to speak to the authorities.[64]

Since in this case the police repeatedly asked the accused to make a statement despite his eighteen assertions of his right to silence, we must infer that the test is not quantitative but qualitative: the accused must show that the police conduct somehow deprived him of the freedom to choose whether to speak.[65]

Since Singh did not contend on appeal that his statement was involuntary, it was, strictly speaking, unnecessary for the Court to state anything about the confessions rule. But Justice Charron used her reasons for judgment in Singh as an occasion to comment on the relationship between these two rules in cases where the accused person is detained and knows he or she is speaking to a person in authority:

[I]n the context of a police interrogation of a person in detention, where the detainee knows he or she is speaking to a person in authority, the two tests are functionally equivalent. It follows that, where a statement has survived a thorough inquiry into voluntariness, the accused’s Charter application alleging that the statement was obtained in violation of the pre-trial right to silence under s. 7 cannot succeed. Conversely, if circumstances are such that the accused can show on a balance of probabilities that the statement was obtained in violation of his or her constitutional right to remain silent, the Crown will be unable to prove voluntariness beyond a reasonable doubt.[66]

This holding moves away from the rule of evidence approach implicit in G.(B.) and towards a rights violation approach. It is, of course, plainly inconsistent with the admonition in Oickle that neither rule subsumes the other.[67] This inconsistency can perhaps be explained by Justice Charron’s resolution of the burden of proof problem mentioned in Oickle. If the Crown can prove voluntariness beyond a reasonable doubt, it is irrebuttably presumed that the accused would not be able to prove a breach of the right to silence on a balance of probabilities. If the accused proves a breach of the right to silence on a balance of probabilities, it is irrebuttably presumed that the Crown would not be able to establish voluntariness beyond a reasonable doubt. But note that Singh does not expressly hold that if the Crown cannot prove voluntariness, the right to silence is necessarily violated (or, equivalently, that if there is no violation of the right to silence, the statement must be voluntary). Rather, Justice Charron holds that if the Crown cannot establish voluntariness beyond a reasonable doubt, there is “no point” in inquiring into the right to silence because the statement will be excluded anyway.[68] If the accused is detained and knows he or she is speaking to a person in authority, the two rules are “functionally equivalent”.[69] The question of whether obtaining an involuntary statement necessarily breaches the accused’s Charter right to silence is left open in Singh.

The approach in Singh does have two apparent advantages. First, it simplifies the trial judge’s task in a complex voir dire where both voluntariness and the right to silence are at play. Second, it grants a kind of constitutional status to the common law confessions rule. But, as I demonstrate in the next part of this paper, these advantages may be illusory.

IV. Voluntariness and Silence: The Confessions Rule after Singh

Any beneficial simplification of a trial judge’s task after Singh depends on the proposition that the common law confessions rule and the Charter right to silence are “functionally equivalent”.[70] Yet this proposition is not true in an important class of cases, so it may be necessary for a trial judge to undertake the “double-barrelled” analysis that the majority disapproved of in Singh (discussed in Part IV.A).[71] Any benefit to the constitutional status of the confessions rule after Singh depends on the rule being sufficiently robust to adequately reflect the principle against self-incrimination. But depending on exactly how Singh and Spencer are read, there is a real danger that the common law confessions rule has been substantially weakened even as it has been constitutionalized (discussed in Part IV.B). This weakening of the rule is due in part to the Court’s decision to constitutionalize the rule in accordance with the rights violation approach rather than the rule of evidence approach. Happily, the cases are also open to a reading that would permit the Court to constitutionalize the confessions rule as a rule of evidence, enabling it to preserve the robust content of the common law rule (discussed in Part IV.C).

A. Functional Equivalence

Justice Charron’s holding in Singh that the confessions rule and the right to silence are “functionally equivalent” (at least in cases where the detained accused knows he or she is speaking to a person in authority) must mean that the two rules have the same implications for the admissibility of evidence. If the only evidence in issue is the accused’s statement, this holding is correct: under the confessions rule, an involuntary statement is inadmissible for any purpose, while a statement obtained in violation of the right to silence is, for all intents and purposes, automatically inadmissible under subsection 24(2) of the Charter. Although evidence is to be excluded under subsection 24(2) only if “the admission of it in the proceedings would bring the administration of justice into disrepute,”[72] statements obtained in violation of the right to silence will normally be excluded.[73] Indeed, the Supreme Court of Canada has never admitted a statement that was obtained in violation of the right to silence, and has rarely admitted a statement obtained after a violation of any other Charter right.

By contrast, if the accused’s statement leads the police to other evidence—in particular, to physical evidence that tends to confirm the statement—the common law confessions rule and the Charter right to silence do not lead to the same decision about admissibility. To better understand this point, consider the following scenario, modelled on the facts of Wray: The accused is charged with murder. The victim was shot with a rifle, but the police have not yet discovered the murder weapon. Under interrogation, the accused denies shooting the victim, but admits to disposing of a rifle. He tells the police, “My rifle is in the swamp. I threw it in after the shooting because I was afraid you’d think I did it.” The police search the swamp, discover the rifle, and link it to the shooting. Assume that the police would never have found the rifle without the accused’s statement. Assume that the trial judge is seriously troubled by the way the police obtained the statement from the accused; perhaps there was an inducement of some kind, combined with oppressive behaviour, such as keeping the accused awake into the early hours of the morning when he had clearly indicated that he wanted to sleep. In accordance with the approach in Oickle, the trial judge finds that the inducement, in context, made the statement involuntary.

In this scenario, the finding of involuntariness is not “functionally equivalent” to a finding that the accused’s right to silence was violated because the admissibility of both the rifle and the statement would be treated very differently at common law and under the Charter. At common law, real evidence derived from an involuntary statement would undoubtedly be admissible. Moreover, the involuntary statement itself would be admissible to the extent that it was confirmed by the real evidence. So at common law, the rifle would be admissible, and so would that portion of the accused’s involuntary statement that is confirmed by the finding of the rifle. Specifically, his statement “my rifle is in the swamp” would be admissible because it is confirmed by the finding of the rifle in the swamp, but his statement “I threw it in ... ” would remain inadmissible because the finding of the rifle does not show how or why it got there. But if the accused’s Charter rights were violated, both the rifle and the statement would have to be considered under subsection 24(2). Thus, given that there is something troubling about the way the statement was obtained, it would (curiously enough) be in the Crown’s interest to characterize the issue as involuntariness and in the accused’s interest to characterize it as a Charter breach.

For the common law confessions rule and the Charter right to silence to be truly “functionally equivalent”, the trial judge in this scenario would have to decide the question left unanswered in Singh: does a finding of involuntariness necessarily imply a breach of the right to silence? Although this proposition does not follow logically from the holdings in Singh,[74] it is tempting to say yes for at least two reasons.[75] First, there are hints in the judgment that this is the Court’s view. Second, accepting this proposition would finally resolve the question of the relationship between the common law confessions rule and section 7 of the Charter. If Singh implicitly holds that whenever the statement is involuntary, there must be a violation of the right to silence, then the obtaining of an involuntary statement would be construed as a Charter breach, and the rule in St. Lawrence would be abrogated in favour of an analysis of the admissibility of derivative evidence under subsection 24(2) of the Charter.[76]

Although Singh expressly leaves the question open, Justice Charron’s reasons, combined with Justice Deschamps’s reasons in Spencer, strongly suggest that only one fact can ultimately support a finding of involuntariness or a finding that the right to silence has been violated: the accused’s will was overborne. This suggestion is reinforced by the following passages from Justice Charron’s reasons:

[V]oluntariness, as it is understood today, requires that the court scrutinize whether the accused was denied his or her right to silence. The right to silence is defined in accordance with constitutional principles.[77]

[T]he confessions rule effectively subsumes the constitutional right to silence in circumstances where an obvious person in authority is interrogating a person who is in detention because, in such circumstances, the two tests are functionally equivalent.[78]

If a statement could be found to be involuntary without a breach of the right to silence, it would not be necessary to scrutinize the right to silence in every case. Similarly, if the confessions rule “subsumes” the right to silence, then a finding of involuntariness must amount to a violation of the right to silence. To “subsume” is to “include ... in a larger group, class, or category” or to “incorporate [or] absorb”.[79] If the right to silence is included in, incorporated in, or absorbed by the confessions rule, then it is hard to see how a finding of involuntariness could not lead to the conclusion that the right to silence was violated. So on a radical reading of Singh, if there is no breach of the right to silence, the statement must be voluntary, and if there is a breach of the right to silence, the statement must be involuntary. Each implies the other.

It is difficult to believe that the Court intended to make such a fundamental change to the confessions rule in Singh, a case in which the accused conceded that his statement was voluntary. Moreover, there is no indication that Justice Charron meant to dispense with the four branches of involuntariness that have been recognized in Canadian law—indeed, she explicitly refers to them.[80] And there are at least two additional reasons to resist this reading of Singh: First, the factual issues involved in determining voluntariness are different from those involved in determining a breach of the right to silence, and neither should be reduced to the question of whether the accused’s will is overborne. Second, it appears that all of the Court’s recent confessions cases are instances of appellate deference to factual findings by trial judges. Despite some of the alarming language in Oickle, Spencer, and Singh, it is still possible to save the confessions rule from being subsumed by the right to silence.[81]

B. The Overborne Will

There are many cases where a breach of the right to silence will cast doubt on the voluntariness of a statement (especially where the oppression branch of involuntariness is in issue), and many cases where scrupulous respect for the right to silence will support a finding of voluntariness. But these two issues should be kept separate, as they involve distinct factual findings and different burdens and quanta of proof. The right to silence is concerned with the conditions under which the accused decides to speak at all; the voluntariness rule is concerned with specific features of the interaction between the police and the accused that tend to make the content of the accused’s statements unreliable.

The right to silence can be violated when the police improperly persuade the accused to speak, but without any inducement or other factor that would make the ensuing statement involuntary. To better understand this possibility, consider the following example: The accused is arrested and charged with an offence. He is informed of, and exercises, his right to counsel. His counsel advises him to remain silent. He decides to follow this advice, and tells the police that he is not going to say anything. But he is young, vulnerable, and very upset, and the police refuse to accept this assertion of the right to silence; they repeatedly ask him to make a statement, suggesting rather vaguely that it would be to his moral advantage to do so. The accused comes to believe that the police officers are really on his side and that his counsel’s advice is unsound. He decides that there is no point in remaining silent, and he makes a statement admitting some involvement in the offence. There is no fear of prejudice or hope of advantage; there is no oppression, and there is no trickery that would shock the conscience of Canadians. Thus, the statement is untainted by any factor that would render it involuntary. Nevertheless, a trial judge might find that the accused’s right to silence, as explicated in Singh, has been violated.

Alternatively, the accused might choose to speak to the police quite deliberately, in which case there is no violation of the right to silence. Parts of the ensuing statement might nevertheless be involuntary because the police offer an improper inducement for a confession. Suppose the accused is arrested and charged with an offence. He is informed of, and exercises, his right to counsel. His counsel advises him to remain silent. After this consultation, the police remind the accused once that he can talk if he wants to, but they do not in any way denigrate counsel’s advice. The accused decides to talk to the arresting officers. He offers an alibi. The police refuse to accept the proffered alibi, and press the accused to say something about the crime. In the course of the discussion, the police tell the accused that unless he confesses, there is no possibility that he will get bail, but if he does confess, the police assure him, “We’ll make sure you get out.” The accused does not confess to the crime, but abandons his alibi and admits that he was at the scene of the crime. It is artificial to describe this as a situation in which the right to silence is respected until the moment the inducement is uttered and is then violated. It seems even more artificial to describe it as a situation in which the accused’s will is overborne. Rather, he consciously and deliberately chose to respond to the inducement by abandoning his alibi. The most plausible way to describe this situation is to acknowledge that the accused’s right to silence has been respected, but that the statement is involuntary and inadmissible at trial because it was induced by a threat and by a promise.

These two examples are meant to suggest that the facts relevant to voluntariness and those relevant to the right to silence are sufficiently distinct to preclude the holding that one subsumes the other. More fundamentally, collapsing both the Charter right to silence and the common law confessions rule into the question of whether the accused’s will is overborne would be a serious conceptual mistake.

The concept of the “overborne will” is no more coherent in the law of criminal procedure[82] than it is in the law of contract[83] or in the criminal law defence of duress.[84] The language of the “overborne will” suggests that the conduct in question resulted from the person’s loss of control over his or her decision-making power, or that the pressure exerted by another rendered the person akin to an automaton. But in the typical case of wrongful pressure, the conduct of the person whose will is said to have been overborne is fully voluntary and rational; it is a deliberate, though highly constrained, effort to avoid an unpleasant consequence. In the particular context of an improper police interrogation, the police conduct does not deprive the suspect of control over the power of speech; rather, the accused makes a conscious decision to speak, with the goal of avoiding the threat, taking advantage of the promise, or relieving the oppressive police conduct. This choice is comparable to a typical case of duress in the law of contract or in criminal law, in which the threatened person makes a conscious and, in the circumstances, rational decision to comply with the threat. It is not plausible to speak of the will being overborne in these situations. Though the accused has consciously exercised his or her will, that exercise is tainted by police behaviour that is liable to render the statement unreliable. It would be ironic indeed if the very effectiveness of the inducement (persuading the accused to choose to speak) were to establish the voluntariness of the statement, when the objective of the common law confessions rule is to render involuntary and inadmissible a statement that was induced by a threat or promise.[85] A finding that the police conduct induced the statement by one of the means contemplated by the traditional rule should make the statement involuntary, as this finding is more than sufficient to raise the twin spectres of improper police conduct and unreliability that underlie the confessions rule. It should not be necessary to go further and ask whether the police conduct not only induced the statement but also overwhelmed the accused’s will.

C. Deferring to Trial Judges

Fortunately, Singh and the cases that preceded it are amenable to a reading that preserves the common law confessions rule and prevents it from being completely assimilated into the Charter right to silence. On this reading, founded on the facts of the cases, the Supreme Court of Canada was merely deferring to the factual findings of trial judges who had applied the correct legal test for voluntariness.[86] I will consider the cases in reverse chronological order. In Singh, the accused argued at trial both that his statement was involuntary and that his right to silence had been violated. The voluntariness challenge was based on a claim of oppression, not a claim of inducement or a lack of an operating mind. The voluntariness challenge was abandoned on appeal. On the right to silence, the trial judge found as fact that the police officer’s repeated refusal to accept the accused’s assertions of his right to silence did not deprive the accused of the ability to choose whether to speak. In assessing the reasonableness of this finding, it is worth noting that the interrogation was relatively brief and that no oppression was found. A claim of oppression, though not the same as a claim that one has been deprived of the ability to choose whether to speak, is more plausibly connected to the idea of the overborne will than a claim of inducement is. This is so because oppressive physical conditions are more likely than verbal threats or promises to lead to a loss of voluntary control over one’s actions.[87] As such, the factual statement in Singh that the accused’s will was not overborne can be read as no more than an unfortunate shorthand for the finding that there was no oppression.

Spencer is admittedly less conducive to a reading of this kind in that the accused did argue that his statement was induced by a fear of prejudice and a hope of advantage (both relating to his girlfriend). As noted above, however, Justice Deschamps did not expressly hold that “overborne will” was the exclusive criterion for determining whether a statement is involuntary. Thus, it is possible that Justice Deschamps was using “overborne will” as a shorthand expression for an accused’s response to any police conduct that improperly induces a statement, whether that conduct is a threat or promise, oppression, trickery, or some combination thereof. On this reading, a finding that the accused’s will was overborne would not establish that he or she was utterly deprived of the power to choose to speak or act; rather, it would affirm that the police had improperly altered the accused’s reasons for speaking in the way he or she did,[88] or that the accused lacked an operating mind.

These fact-based readings of Spencer and Singh are reinforced by the Supreme Court of Canada’s brief decision in Tessier.[89] In that case, the accused was charged with second degree murder in the death of his girlfriend. She had disappeared in 1989 and the accused had been a suspect for some time. In 2000, he was arrested and he gave a lengthy statement that included an admission that would probably have established that he was guilty of manslaughter. The trial judge, without expressly stating that there was either an inducement or oppression, found that the statement was involuntary. This finding appears to have been based on two concerns: the police suggested that the charge would be reduced from murder to manslaughter if the accused would admit to manslaughter, and there was something oppressive about the circumstances in which the statement was made. The exclusion of the statement caused the Crown’s case to collapse, and the accused was acquitted. The Crown appealed, arguing on the basis of Oickle (decided after the trial judge’s ruling) that the trial judge had applied the wrong test. The majority of the New Brunswick Court of Appeal agreed, holding that the trial judge “did not direct his mind to the question that must be answered if tolerable persuasion is to be distinguished from vitiating inducement: what quid pro quo offer did the interrogating officers make to Mr. Tessier that might have caused his will to be overborne?”[90] In dissent, Justice Deschênes held that the trial judge had done precisely what Oickle demanded: although the trial judge had not expressly found an inducement or an atmosphere of oppression, he had considered the voluntariness of the statement “on the basis of a consideration of all the circumstances surrounding the taking of the statements.”[91] In particular:

[I]t was open to the trial judge in this case to look upon the comments of the interrogators with respect to what a confession to manslaughter might attract as a sentence as an improper inducement in the nature of an assurance that state authorities would treat him with leniency if he admitted some involvement in the death of [the victim]. In addition, it was open to the trial judge to conclude that the process of interrogation was oppressive ...

Finally ... the trial judge must have been influenced to some extent in his decision by what might be described, to put it lightly, as an improper inducement relating to the detainee’s right to silence when the interrogator told him that to continue his silence was a “dangerous” option to adopt, bearing in mind the unfavorable inference a jury would draw upon viewing the taped interview.[92]

The Supreme Court of Canada agreed with Justice Deschênes’s holding that the trial judge had applied the correct test; thus, the Crown’s appeal raised no question of law and the New Brunswick Court of Appeal should have dismissed it.

Tessier supports the fact-based reading of Spencer and Singh in at least two ways. First, in Tessier, the majority of the Court of Appeal explicitly stated that the test for voluntariness was whether the accused’s will was overborne; yet the Supreme Court of Canada endorsed Justice Deschênes’s view that the Oickle approach should be understood in terms of whether all the circumstances surrounding the statement rendered it involuntary. Thus, Tessier supports the assertion that the phrase “overborne will”, which figures so prominently in the reasoning in Spencer and Singh, can be understood as a shorthand (albeit a rather misleading shorthand) for the question of whether the accused was improperly induced to speak. Second, Tessier upholds a trial judge’s decision that might well have been considered vulnerable to reversal in light of the reasoning in Oickle and Spencer.

Finally, Oickle is recognized as the new leading case on the common law confessions rule. It basically preserves the traditional structure of the rule and holds that trial judges should consider all the circumstances in determining voluntariness. Justice Iacobucci’s references to the “overborne will” are fleeting and peripheral to his reasoning. Further, Justice Iacobucci was largely concerned with the standard of review, and he began his reasons by stating that the case could be resolved on that basis alone.[93] Indeed, in all four cases—Oickle, Tessier, Spencer, and Singh—the Supreme Court of Canada upheld the trial judge’s ruling. Despite the alarming implications of the language in some of these cases, they may be read as indicating, not a change in the test for voluntariness, but a high degree of deference to a trial judge’s consideration of all the factors relevant to voluntariness.

V. The Confessions Rule and the Principle Against Self-Incrimination

In Part IV of this article, I argued that Oickle, Tessier, Spencer, and Singh can be read in various ways. On the most radical reading, these cases combine the Charter right to silence and the common law confessions rule as follows: If the accused knows he or she is dealing with a person in authority, the common law confessions rule and the Charter right to silence collapse into the single question of whether the accused’s will was overborne. If so, the statement is inadmissible both because it is involuntary and because the accused’s right to silence is violated. If not, the statement is voluntary and there is no violation of the right to silence.

On the most conservative reading, these cases maintain the traditional structure of the confessions rule. There are still four ways in which a statement can be involuntary (though three of them need to be considered together in a principled and contextual way), and none of them necessarily implies a violation of a Charter right. If the statement is involuntary, it is excluded at common law; but if the statement is voluntary, there is also no violation of the right to silence and the statement is admissible.[94] The reference in these cases to “overborne will” is just an unfortunate shorthand used to articulate the question of whether the accused was prompted to speak by any of the traditional factors that make a statement involuntary.

The radical reading suggests that the confessions rule should be constitutionalized in accordance with the rights violation approach outlined in Part IV.B, above. The Charter rights of the accused are not violated until his or her will is overborne. Once that has occurred, the court deems the statement involuntary and excludes it at common law. While this exclusion may be correct, if the rights violation approach is adopted in the form presented in Singh, it will not be long before the Crown begins to argue that instead of being automatically excluded, the statement should be considered under subsection 24(2) of the Charter.

The conservative reading suggests that the confessions rule should be constitutionalized as a rule of evidence applicable at trial. This reading recognizes that the inquiry into voluntariness is distinct from the inquiry into recognized Charter rights such as the right to silence or the right to counsel. The question of voluntariness is whether the accused’s decision to say what he or she did was prompted by his or her own reasons or by outside factors such as an inducement, the absence of an operating mind, or oppression. But the conduct of the police in obtaining the statement need not violate the accused’s Charter rights, and in particular need not violate the right to silence. The accused’s operating mind can fail without any improper state conduct whatsoever. An offer of leniency in exchange for a statement does not, in itself, violate any recognized Charter right. Further, an inducement or an oppressive circumstance does not imply a violation of the right to silence (even in the weak form articulated in Singh). Thus, the best way to constitutionalize the confessions rule is to treat the exclusionary rule itself as a principle of fundamental justice under section 7 of the Charter, as in G.(B.).

The ultimate choice between these two readings of the Court’s recent decisions on the confessions rule, and between these two ways of constitutionalizing the confessions rule, depends on which better reflects the principle against self-incrimination. As discussed above, the principle against self-incrimination prevents the state from compelling a suspect to participate in an investigation, but permits the suspect to choose freely to assist the investigation.[95] On this understanding of the principle, the conservative reading is clearly preferable. The principle against self-incrimination is a very broad normative idea that has, as the Supreme Court of Canada has recognized, several different aspects.[96] Collapsing those aspects into one factual inquiry, as the radical reading does, threatens to discount some serious infringements of the principle against self-incrimination because they no longer fit into the narrowly defined category of police misconduct that violates Charter rights. By contrast, the conservative reading of the cases preserves distinct aspects of the principle against self-incrimination. If the conduct of the police is so egregious that the accused’s will is overborne, then the accused’s statement should certainly be excluded on constitutional or common law grounds. However, overbearing the accused’s will is not the only way that the principle against self-incrimination can be infringed. Consider the following examples.

The Crown’s use, at trial, of statements that were prompted by a loss of the accused’s operating mind need not involve any state conduct that can be characterized as overbearing the will or infringing the right to silence. Nevertheless, as the common law confessions rule recognizes, it does infringe the principle against self-incrimination by using the accused’s inability to control his or her incriminating utterances. The best way to protect the principle against self-incrimination in this context is simply to exclude the statement at trial.

Scrupulous respect for the section 7 right to silence—a factor which will certainly weigh in favour of a finding of voluntariness and will, on one reading of Singh, determine the issue of voluntariness in the Crown’s favour—can easily co-exist with a significant inducement that would make the statement inadmissible on common law grounds. Again, the best way to preserve these different manifestations of the principle against self-incrimination is to adopt the conservative reading of the cases and to treat the common law confessions rule as a constitutionalized rule of evidence.

It may be argued that the root of the problem in Singh is not the Court’s effort to constitutionalize the confessions rule, but the minimal content that the Court gave this rule in Oickle and Spencer. Put another way, perhaps we should worry less about the constitutional status of the confession rule and more about whether its content is sufficiently robust. There is certainly something to be said for this view: linking the confessions rule to the right to silence would be less troubling if both doctrines adequately protected the accused from self-incrimination. But the constitutional status of the rule affects its content. As I have tried to demonstrate, if obtaining an involuntary statement is treated as a violation of the accused’s rights, analogous or even equivalent to a breach of the right to silence, it is necessary to characterize police conduct as a rights violation before a confession can be found to be involuntary. But the common law confessions rule, both before and after Oickle, considers a range of factors as relevant to voluntariness, many of which have nothing to do with rights violations. Treating the confessions rule as an aspect of the constitutional right to a fair trial is much more compatible with the traditionally robust content of the rule; it asks not whether the police have violated the accused’s rights but whether anything that happens in the interaction between the police and the accused might cast doubt on the accused’s reasons for speaking as he or she did and therefore on the reliability of his or her statement. By avoiding the intractable concept of the “overborne will”, the conservative reading of the cases preserves this aspect of the common law rule and thus maintains the confessions rule and the section 7 right to silence as distinct aspects of the principle against self-incrimination. It is therefore less likely than the radical reading to lead to the admission of statements that are not the product of the suspect’s true desire to speak.

Conclusion

In this paper, I have offered two readings of the Supreme Court of Canada’s recent decisions concerning the common law confessions rule. These cases are plausibly read as linking the confessions rule with a rather weak version of the Charter right to silence, or more specifically as understanding both the right to silence and the confessions rule as a right against having one’s will overborne by state agents. This reading would significantly weaken the common law confessions rule and thereby weaken the principle against self-incrimination. Fortunately, the cases can equally be read as deferring to trial judges’ factual assessments of voluntariness while preserving the common law confessions rule and constitutionalizing it, if at all, as an exclusionary rule of evidence at trial. This second, more conservative reading is preferable because it would preserve the right to silence and the rule that a statement should be voluntary as distinct aspects of the principle against self-incrimination. The common law confessions rule is a manifestation of the principle that a person should not be required to incriminate himself or herself. It is related to several constitutional doctrines, including the section 7 right to silence. The best way to continue to develop the principle against self-incrimination, however, would be to preserve the traditional structure and content of the common law confessions rule, and to constitutionalize it as a rule of evidence rather than to reduce it to an aspect of the Charter right to silence. The common law confessions rule and the Charter right to silence have distinct roles to play in protecting an accused person’s right against self-incrimination, and it is a mistake to conflate them.