Three for two, and two for three

The common law may become overwrought by distinctions without a difference. On reading Tofilau v R [2007] HCA 39 (30 August 2007), one might wonder whether this has happened in respect of the discretionary exclusion of evidence in Australia.

In Australian common law there are three so-called “discretions” the exercise of which may lead to the exclusion of evidence that would otherwise be admissible. These are summarised at paras 245 to 248 of the joint judgment of Callinan, Heydon and Crennan JJ (I have put the footnotes in square brackets):

“245 In order to appreciate the significance of the appellants' arguments, it is desirable to place them in the context of the law relating to the admissibility of confessions as a whole. An admission by an accused person "must be voluntary in order to be admissible"[ R v Lee [1950] HCA 25; (1950) 82 CLR 133 at 144 per Latham CJ, McTiernan, Webb, Fullagar and Kitto JJ]. It is common to divide involuntary statements into two categories. One concerns the "inducement rule": an admission by an accused person "is not voluntary if it is preceded by an inducement, such as a threat or promise, held out by a person in authority, unless the inducement is shown to have been removed"[R v Lee [1950] HCA 25; (1950) 82 CLR 133 at 144 per Latham CJ, McTiernan, Webb, Fullagar and Kitto JJ]. The other category concerns those caught by the "basal involuntariness" rule …. But even if an admission is voluntary, it may be excluded on "discretionary" grounds. In R v Swaffield [[1998] HCA 1; (1998) 192 CLR 159 at 189 [52]] Toohey, Gaudron and Gummow JJ grouped these "discretionary" grounds under three heads.

“246 The first in time to emerge was that which was stated in the cases summarised by Lord Sumner delivering the advice of their Lordships in Ibrahim v The King [[1914] AC 599 at 611-614] about impropriety in police questioning. The correctness of excluding evidence on this ground in Victoria was left open in Cornelius v The King [[1936] HCA 25; (1936) 55 CLR 235 at 247-248 per Dixon, Evatt and McTiernan JJ], but was approved for New South Wales by Dixon J in McDermott v The King [[1948] HCA 23; (1948) 76 CLR 501 at 513]. Dixon J said of it:
"there has arisen almost in our own time a practice in England of excluding confessional statements made to officers of police if it is considered upon a review of all the circumstances that they have been obtained in an improper manner".

In R v Lee [[1950] HCA 25; (1950) 82 CLR 133 at 149-151] Latham CJ, McTiernan, Webb, Fullagar and Kitto JJ declined to interfere with the Victorian Full Court's recognition of the discretion in that State. The application of this head was given fresh life from 1982, for in Cleland v The Queen [[1982] HCA 67; (1982) 151 CLR 1] the discretion to exclude illegally or improperly obtained real evidence enunciated in Bunning v Cross [[1978] HCA 22; (1978) 141 CLR 54 at 75] was applied to confessions. It has since been common to refer to this as a "policy" discretion.

“247 The second "discretionary" head to emerge arose where it could be said of a voluntary confession that "in all the circumstances it would be unfair to use it in evidence against" the accused. The words are those of Latham CJ in McDermott v The King [[1948] HCA 23; (1948) 76 CLR 501 at 506-507] summarising R v Jeffries [(1946) 47 SR (NSW) 284], but they were approved by Latham CJ, McTiernan, Webb, Fullagar and Kitto JJ in R v Lee [[1950] HCA 25; (1950) 82 CLR 133 at 151]. This is commonly called a "fairness" discretion.

“248 The third "discretionary" head to emerge was the application to confessions of the "discretion" to exclude evidence the prejudicial impact of which is greater than its probative value, which had begun to be recognised in R v Christie [[1914] AC 545 at 560 per Lord Moulton, 564-565 per Lord Reading (Lord Dunedin concurring)] and continued to develop in various fields of the law of evidence throughout the 20th century. The application of that principle to confessions, but not its description as a "discretion", was accepted as legitimate by Toohey, Gaudron and Gummow JJ in R v Swaffield [[1998] HCA 1; (1998) 192 CLR 159 at 191-193 [61]- [65]…].”

Any difficulty will concern the distinction between the first two discretions. It should be acknowledged that a distinction is discernable, at least in respect of what gives rise to the exercise of each. The first, the “policy” discretion, applies where the public interest in enforcement of the law is outweighed by unfairness to the defendant in the manner in which the evidence was obtained (per Barwick CJ in Bunning v Cross). Here, fairness means society’s right to insist that those who enforce the law themselves respect it (ibid, per Stephen and Aickin JJ). The second, the “fairness” discretion, applies where improper methods in obtaining evidence mean that it would be unfair to admit it; examples are where the defendant was not mentally alert when questioned, or where he fails to understand and appreciate the effect of questions and answers (examples from R v Jeffries, referred to in R v Lee). Those examples of the second discretion might suggest that the underlying concern here is the reliability of the evidence, but that would be inaccurate. The common law resists allowing a rule of exclusion of evidence for unreliability, insisting that the reliability of the evidence is a matter for the fact-finder. This was emphasised in Tofilau in footnote 255, as follows:

“In [Swaffield] at 194 [69] it was further suggested that, subject to a qualification, admissibility at common law turns ‘first on the question of voluntariness, next on exclusion based on considerations of reliability and finally on an overall discretion which [takes] account of all the circumstances of the case to determine whether the admission of the evidence or the obtaining of a conviction on the basis of the evidence is bought at a price which is unacceptable, having regard to contemporary community standards’. Whether or not this is so was not crucial to the outcome in R v Swaffield, was not argued in the present appeals, and is not crucial to their outcome. The correctness of the suggestion thus need not be decided in these appeals (cf reasons of Kirby J at [127]). Subject to that, it is desirable to say the following about the fact that Tofilau appeared to submit to Osborn J that confessions can be excluded merely because they are unreliable, as did Marks and Clarke [other appellants in the present case]. The submissions were rejected on the facts (R v Tofilau [2003] VSC 188; (2003) 13 VR 1 at 17 [57]- [58]; R v Marks [2004] VSC 476 at [83]- [92]; R v Clarke [2004] VSC 11 at [54]- [65]), but in any event their correctness in law is highly questionable: a fear of unreliability may underpin the "inducement rule", and unreliability may be a factual circumstance relevant to basal involuntariness and to discretionary exclusion, but the fact that a judge thinks that a confession is unreliable is not in itself a ground of automatic exclusion.”

The real nature of the unfairness with which this second discretion is concerned arises from the inability of the defendant to decide whether to exercise his rights. It will not necessarily be the fault of the police that this occurs: they may, for example, be quite unaware that the suspect is unable to understand what he is told about his right to obtain legal advice and to refrain from answering questions. Where the police are at fault, however, misconduct will be in issue, and this discretion will tend to merge with the first.

Does the decision process differ as between these two discretions? If the grounds giving rise to their exercise merge and the decision process is the same for each, there is little point in distinguishing between them.

The decision process for the first discretion, the public policy discretion, is a weighing exercise: the public interest in enforcement of the law is weighed against the public interest in having officials obey the law, in the circumstances of the particular case; the question would come down to whether exclusion of evidence would be a proportionate response to the misconduct that had occurred. It would seem that the decision process for the second discretion, the fairness discretion, involves asking whether admission of evidence obtained in the face of a failure of rights would bring the administration of justice into disrepute, or would amount to an abuse of process. I have borrowed these phrases from the common law in Canada and New Zealand, but they seem apt for the Australian context. Essentially, then, these two discretions can be combined; the disrepute to the administration of justice criterion has its origins in the prevention of abuse of process. This permits the conclusion that revision of the Australian analysis of the discretions is justified.

There are some indications in Swaffield that a revision of the common law to this effect would be appropriate, but the opportunity to do this was not taken in Tofilau. This point was alluded to in Tofilau at para 399:

“Counsel's submissions in this Court. Counsel for Clarke in this Court submitted that while it was conventional to analyse discretionary exclusion of confessions as involving two "discretions" - to reject a confession the reception of which would be unfair [R v Lee [1950] HCA 25; (1950) 82 CLR 133], and to reject a confession that was illegally or improperly obtained on public policy grounds [Cleland v The Queen [1982] HCA 67; (1982) 151 CLR 1] - in truth there was but a single "discretion" [He cited R v Swaffield [1998] HCA 1; (1998) 192 CLR 159 at 202 [91] per Toohey, Gaudron and Gummow JJ]. It is not necessary to resolve this question, since the outcome of the appeal will be the same whatever the answer. He also submitted that in any event what was involved was not a "discretion" but a rule of law, and relied on an analogy with the rule excluding evidence the prejudicial effect of which exceeded its probative value. It is not necessary to resolve that question either, for the same reason, although the appellant's submission is open to question, and quite out of line with past linguistic usage in this field.”

The Swaffield analyses of the discretions are rather complicated, and the Tofilau summary avoids this complexity.

It is useful to compare the Australian common law with the position in New Zealand under the Evidence Act 2006. There are two discretions under the Act: the weighing exercise in relation to improperly obtained evidence, s 30, and the weighing exercise in relation to prejudice and probative value, s 8. The scope of the improperly obtained evidence discretion is such that it covers the same ground as the first two of the Australian discretions. The prejudice and probative value discretion is equivalent to the third Australian discretion.

Tofilau is also about the use of undercover police officers to encourage admissions of crimes, and it contains discussions of matters such as voluntariness, police use of deception, reliability, and the right to silence. Its result follows that of the Supreme Court of Canada case R v Grandinetti [2005] SCC 5 (27 January 2005) which held that where the accused did not think that the people he was talking to were police officers, they were not persons in authority, the state’s coercive power was not involved, and his admissions to them were voluntary; in the particular circumstances of the present appeals the evidence was held to be admissible.

Update: For doubts as to the existence of a residual common law discretion to exclude evidence for unfairness, see Police v Dunstall [2015] HCA 26 (5 August 2015) at [47]-[48]. For discussion of the use of undercover officers as cellmates to obtain admissions from the defendant, see R v Kumar [2015] NZSC 124 (6 August 2015). In New Zealand the "discretions" referred to above are no longer called "discretions", but are recognised to be the application of rules by the exercise of judgment, so an appeal is on a question of law, not against the exercise of a discretion.

“He jests at scars that never felt a wound.”

Romeo’s words, or similar, might have occurred to the prisoners whose claims for compensation for mistreatment in prison were under the scrutiny of the New Zealand Supreme Court in Taunoa v Attorney General [2007] NZSC 70 (31 August 2007). The Court has decided the appropriate amounts of money to be paid by the state to several prisoners for the circumstances in which they had been subjected to breaches of their right to be treated with humanity and respect for their inherent dignity: s 23(5) New Zealand Bill of Rights Act 1990. In relation to one prisoner, the question was how much should be awarded for breach of his right not to be subjected to disproportionately severe treatment: s 9. The central problem was how to assess public law damages. These are discretionary. Thus they are in contrast to tortious damages which are awarded as of right upon proof of the cause of action, and which are compensatory, can be aggravated, and may be exemplary. Public law damages have a symbolic importance that is seen as diminishing their need to be high in magnitude.

Consequently, the amounts awarded were small (“moderate” was the adjective used by Blanchard J at para 265). I have previously noted here cases in other jurisdictions concerning compensation for breach of another right, the right to a fair hearing, where amounts were also small: see entries for 21 February 2005, 6 and 30 March 2005, 13 April 2005. Here, the Supreme Court reduced those of the awards that the Attorney General appealed.

One would never expect much agreement on amounts that the state should pay to its serious criminals who as a result of their difficult conduct in prison are subjected to efforts to control their behaviour, where those efforts go too far and breach basic rights. Here, the Chief Justice would have been the most generous. Blanchard and McGrath JJ agreed on a middle figure, and Tipping and Henry JJ would have awarded lesser amounts. The result was that the majority of the Court agreed that the amounts decided by Blanchard and McGrath JJ were the minimum that should be ordered, and those amounts became the outcome of this aspect of the case.