Jonathan Clough* PFENNIG v THE QUEEN: A RATIONAL VIEW OF PROPENSITY EVIDENCE? INTRODUCTION [Tlhe law is nearly always most obscure in those fields in which judges say the principle is plain, but the difficulty lies in its application to particular facts .I 0 area of the law lends more truth to this statement than that relating to the admissibility of what may generically be described as "propensity e ~ i d e n c e " . ~ For while it is true that the principles governing the admissibility of such evidence are clear, the search for a rule which will facilitate their consistent application has been the one constant feature of cases in this area. N * 1 The admissibility of propensity evidence reflects a conflict between two clear but competing principles. The first is that all evidence which is relevant is prima facie admissible. The second is that all evidence which is unduly prejudicial should be BSc, LLB (Hons) (Mon), LLM (Cantab); Lecturer in Law, Monash University. I am grateful for the helpful comments and suggestions of Professor CR Williams, Associate Professor SB McNicol, Mr Jonathon Moore and the anonymous referee for this journal. Ilkiw v Samuels [I9631 1 WLR 991 at 1004 per Diplock LJ. Lord Herschel1 prefaced his famous statement of the law relating to propensity evidence in almost precisely these terms: "the principles which must govern the decision of the case are clear, though the application of them is by no means free from difficulty": Makin v Attorney-General for New South Wales [I8941 AC 57 at 65. While recognising that it is not a definitive term, the phrase "propensity evidence" will be adopted in this article to describe "evidence which is received notwithstanding that it discloses the commission of offences other than those with which the accused is charged. [Such evidence] is always propensity evidence but it may be propensity evidence which falls within the category of similar fact evidence, relationship evidence or identity evidence. Those categories are not exhaustive and are not necessarily mutually exclusive": Pfennig v The Queen [1995] HCA 7; (1995) 182 CLR 461 at 464-465 per Mason CJ, Deane and Dawson JJ. It is beyond the scope of this article to consider the difficult and still unresolved issue of the exact scope of the rule which presumptively excludes such evidence. See generally Ligertwood, Australian Evidence (Butterworths, Sydney, 3rd ed 1998) pp109-114, Heydon, Cross on Evidence (Butterworths, Sydney, 5th ed 1996) [21001]-[21010] and Palmer, "The Scope of the Similar Fact Rule" [1994] AdelLawRw 4; (1994) 16 Adel LR 161. The use of "propensity evidence" as a generic term is reflected in the recent decision of the High Court in Gipp v The Queen (1998) 72 ALJR 1012 at 1015 per Gaudron J, at 1027 per McHugh and Hayne JJ, and at 1045 per Callinan J. 2 288 CLOUGH - A RATIONAL VIEW OF PROPENSITY EVIDENCE'? excluded. The rule against propensity evidence resolves this conflict in favour of the latter by rendering such evidence presumptively inadmissible. It is excluded not because it is irrelevant, but because its admissibility is considered to be unfairly prejudicial to an accused. This presumption may, however, be displaced by demonstrating that the evidence has sufficient probative force. This has traditionally involved the trial judge in an illdefined "balancing" of probative force and prejudicial effect in order to determine admissibility. A majority of the High Court has now adopted a test which reduces this balancing process to a single question: is there "a rational view of the evidence that is consistent with the If innocence of the a c ~ u s e d " . ~ there is another rational view that is consistent with innocence, the propensity evidence is inadmissible. If, however, there is no other rational view, then it will be admissible. Whether this test, which will be referred to as the "nc) rational view" test, should continue to be adopted is the subject of this article. Given the difficulties inherent in determining the admissibility of propensity evidence, it is easy to be seduced by the apparent simplicity of this approach. While it has the advantage of providing a trial judge with a precise question to determine the admissibility of propensity evidence, it is submitted that it should not be adopted on two grounds. First, the precise application of the test is unclear. Secondly, and more importantly, it fails to address adequately the competing principles which operate whenever the admissibility of such evidence is considered. The resolution of this conflict is a matter of judgment for the trial judge which, it is submitted, cannot and should not be resolved by such a formulaic approach. It will be argued that the preferable approach is for the High Court to reject the "no rational view" test and to return to its earlier formulation of the test of admissibility as one involving a comparison of the probative force and prejudicial effect of the evidence. However, the test should be phrased not in terms of a "balancing" process, but in terms of whether the evidence is sufficiently probative that it would not be unfair to admit it. "Unfair" in this sense refers to the possibility of an unfair trial. The trial judge is therefore required to consider the probative force of the evidence together with its prejudicial effect, and determine whether there is an unacceptable risk of prejudice to the accused such that his or her trial would be unfair if the evidence were to be admitted. Although propensity evidence will inevitably carry with it some degree of prejudice, the question is whether the degree of prejudice is such that it can only be dealt with by exclusion of the evidence. THE EVOLUTION OF THE "NO RATIONAL VIEW" TEST The "no rational view" test is the product of necessity. As with so many issues in this area of the law, that necessity arose out of the need to give effect to Lord Herschell's 3 Pfennig [1995] HCA 7; (1995) 182 CLR 461 at 483 per Mason CJ, Deane and Dawson JJ. formulation of the applicable principles in Makin v Attorney-General for New Soutlz Wales: It is undoubtedly not competent for the prosecution to adduce evidence tending to shew that the accused has been guilty of criminal acts other than those covered by the indictment, for the purpose of leading to the conclusion that the accused is a person likely from his criminal conduct or character to have committed the offence for which he is being tried. On the other hand, the mere fact that the evidence adduced tends to shew the commission of other crimes does not render it inadmissible if it be relevant to an issue before the jury, and it may be so relevant if it bears upon the question whether the acts alleged to constitute the crime charged in the indictment were designed or accidental, or to rebut a defence which would otherwise be open to the a c c u ~ e d . ~ Although this passage has been described as stating the principles with "crystal clarity"? the history of propensity evidence has been concerned largely with giving effect to these inherently contradictory sentences. While the first sentence sets out a rule of exclusion, the second sets out a rule of inclusion which is wider than the first. In order for any evidence to be admissible it must be relevant to an issue before the jury. If it is not relevant then it is inadmissible on that basis. Such an interpretation would render the special exclusionary rule superfluou~.~In general, the courts have reconciled this contradiction by adopting one of two approaches. The first is concerned with types of relevance, the second with degrees of relevance. Under the first approach, a distinction is drawn between evidence which is relevant merely because it demonstrates a criminal propensity, and evidence which is relevant other than by showing propensity. The evidence will be inadmissible "if it shows only that the accused had a propensity or disposition to commit crime, or crime of a particular kind, or that he was the sort of person likely to commit the crime charged? If, however, it can be shown that the evidence is otherwise relevant to a fact in issue, then it may be admissible to prove that fact notwithstanding that it also reveals the accused's criminal pr~pensity.~ Examples of such additional relevance include where the evidence shows the existence of a relevant relationship or where the unusual nature of the prior criminal conduct identifies the accused as the offender? 4 5 6 7 8 9 [I8941 AC 57 at 65. Director of Public Prosecutions v Boardman [I9751 AC 421 at 461 per Lord Salmon. Williams, "The Problem Of Similar Fact Evidence" (1979) 5 Dalhousie W 28 1 at 283-290. Also see Perry v The Queen [1982] HCA 75; (1982) 150 CLR 580 at 593 per Murphy J . Markby v The Queen [1978] HCA 29; (1978) 140 CLR 108 at 1 16 per Gibbs ACJ. As above. Perry [1982] HCA 75; (1982) 150 CLR 580 at 586-587 per Gibbs CJ. 290 CLOUGH - A RATIONAL VIEW OF PROPENSITY EVIDENCE? Such an approach is difficult to reconcile with the authorities, as in many cases the evidence clearly derives its probative force from the accused's criminal propensity.1° More importantly, this approach fails to take into account the inherent prejudicial effect of propensity evidence as a factor in deterinining admissibility. The balancing of probative force and prejudicial effect must take place, if it is to occur at all, in the exercise of the trial judge's general discretion to exclude otherwise legally admissible evidence.l This may be contrasted with the second approach in which the balancing of probative force and prejudicial effect is the focus of the test of admissibility. This approach reconciles the inherent contradiction in Lord Herschell's formulation by admitting propensity evidence only where its probative force exceeds its prejudicial effect.12 The conflict between these two approaches is evident in a number of decisions of the High Court. In Markby a majority of the Court clearly adopted the first approach and held that propensity evidence would only be admissible if it tended to show that the accused was guilty for some reason other than that he or she had committed crimes in the past or had a criminal disposition.13 However, the fact that the evidence was relevant in some other way would not of itself be sufficient to secure its admissibility. It must, in addition, possess a "strong degree of probative force."l4 The admissibility of the evidence is therefore determined in two stages. First, does the evidence possess relevance other than by showing the accused's criminal propensity? Secondly, even if it has such additional relevance, is it sufficiently probative? Although this view was affirmed by the Chief Justice in Perry,ls other judgments demonstrated an important shift towards the second approach. While Justices Wilson and Brennan (as he then was) both acknowledged that the applicable principles had been stated in Markby, the focus of their judgments moved away from relevance other than via propensity towards a balancing of probative force and prejudicial effect,I6 a trend which continued in Sutton v The Queen.17 For example, Justice Brennan, although citing Markby with approval, stated that "[ilt is the probative force (or cogency) of the evidence in 10 11 12 13 14 15 16 17 Harriman v The Queen [1989] HCA 50; (1989) 167 CLR 590 at 600-601 per Dawson J. For example, see R v Straffen [I9521 2 QB 91 1 and R v Ball [I91 11 AC 47. Markby [1978] HCA 29; (1978) 140 CLR 108 at 117 per Gibbs ACJ and Perry [1982] HCA 75; (1982) 150 CLR 580 at 585 per Gibbs CJ. Williams, "The Problem Of Similar Fact Evidence" (1979) 5 Dalhousie W 281 at 289. [1978] HCA 29; (1978) 140 CLR 108 at 116 per Gibbs ACJ, with whom Stephen, Jacobs and Aickin JJ agreed. Justice Murphy decided the application on other grounds. At 117 per Gibbs ACJ, quoting Lord Wilberforce in Boardrnan [I9751 AC 421 at 444. Perry [1982] HCA 75; (1982) 150 CLR 580 at 585 per Gibbs CJ. At 603-604 per Wilson J and at 608-609 per Brennan J. Pre-empting the "no rational view" test, Justice Murphy held at 596 that, "[tlhe evidence being circumstantial, it is admissible only if ... there is no rational explanation ... consistent with the accused's innocence". [1984] HCA 5; (1984) 152 CLR 528. comparison with the impermissible prejudice that it may produce which determines admissibilityw.'8 There were therefore two important issues to be resolved. The first was whether, to be admissible, the evidence had to possess relevance other than via propensity.19 The second issue, and that with which this article is concerned, relates to the degree of probative force which the evidence must possess in order to be admissible. Irrespective of which approach is adopted, it is clear that, assuming the propensity evidence to be relevant, "the criterion of its admissibility is the strength of its probative force".20 However, the phrases used .to describe the requisite degree of probative force have varied considerably. For example, it has been variously and often interchangeably described as "so very relevant that to exclude it would be an affront to common sense", "a really material bearing upon the issues to be decided", "strong probative force" and "probative force [which] clearly transcends its merely prejudicial effect".Z1 The way in which the test is phrased is important in two respects. First, it is relevant to the question of who bears the onus of proving this issue. Secondly, it affects the manner in which the issue may be dealt with by an appellate court.22 For example, in order to be admitted under the approach adopted in Markby the evidence must possess "strong probative force". Because the evidence must necessarily be highly probative, it may be implicit in such an approach that the general discretion to exclude will rarely be exercised.23 Nonetheless, the balancing of probative force and prejudicial effect technically plays no part in determining the admissibility of the evidence. Therefore the prejudicial effect of the evidence must be considered, if at all, in the exercise of the general discretion to exclude. This has two important consequences. First, the onus is on the accused to justify the rejection of the evidence in the exercise of the general discretion. 18 19 At 547-548. Also see Justice Deane at 559-560 who, while supporting the need to show 20 21 22 23 relevance other than via propensity, held that the prosecution must show that the probative force of the propensity evidence clearly transcends its prejudicial effect. It is beyond the scope of this article to consider whether there remains in Australia any vestige of this requirement. Although the High Court has not expressly overruled Markby on this point, there are numerous statements to the effect that evidence of propensity may in itself be sufficiently relevant to be admissible. For example, see Perry [1982] HCA 75; (1982) 150 CLR 580 at 592-593 per Murphy J; Harriman [1989] HCA 50; (1989) 167 CLR 590 at 597-601 per Dawson J and at 613 per Gaudron J; S v The Queen [1989] HCA 66; (1989) 168 CLR 266 at 275 per Dawson J; B v The Queen [1992] HCA 68; (1992) 175 CLR 599 at 618 per Dawson and Gaudron JJ; Pfennig [1995] HCA 7; (1995) 182 CLR 461 at 484 per Mason CJ, Deane and Dawson JJ, at 505 per Toohey J and at 517-518 and 527-528 per McHugh J; and BRS v The Queen [1997] HCA 47; (1997) 71 ALJR 1512 at 1527 per McHugh J. Hoch v The Queen [1988] HCA 50; (1988) 165 CLR 292 at 294 per Mason CJ, Wilson and Gaudron JJ. Sutton [1984] HCA 5; (1984) 152 CLR 528 at 563 per Dawson J. Pfennig [1995] HCA 7; (1995) 182 CLR 461 at 5 15 per McHugh J. Sutton [1984] HCA 5; (1984) 152 CLR 528 at 534 per Gibbs CJ. 292 CLOUGH - A RATIONAL VIEW OF PROPENSITY EVIDENCE? Secondly, an appellate court will be more reluctant to interfere with the exercise of a judicial discretion as opposed to the application of a rule of law ?4 It is for this reason that subsequent judgments of the High Court have tended to express the balancing of probative force and prejudicial effect as part of the test of admissibility itself, rather than as an aspect of the general discretion to exclude otherwise admissible evidence.25 This places the onus on the prosecution and makes the trial judge's decision more susceptible to appellate scrutiny. However a statement of competing principles does little to assist a trial judge in determining whether the probative force of the evidence can be said to outweigh its prejudicial effect. This issue was specifically addressed by Justice Dawson in Sutton and it was in this context that the "no rational view" test evolved. After referring to the various standards which had been suggested in the cases, his Honour concluded that a trial judge may find assistance by applying the same standard as the jury must ultimately apply in dealing with circumstantial evidence. 'This is to say no more than that to be admissible the similar fact evidence must give rise to a clear inference tending to establish an element of the offence charged and must not be reasonably explicable upon some other basis."26 Justice Dawson derived support for this approach from the earlier decision of the High Court in Martin v Osborne?' The accused in this case had been convicted before a magistrate of driving a commercial passenger vehicle without a licence. While there was ample evidence that the accused had driven the vehicle and carried passengers, it was necessary to prove that he had done so for reward. The prosecution therefore tendered evidence that on previous occasions the accused had been seen picking up passengers and their luggage. This was tendered to suggest that it was extremely improbable that the accused would have carried passengers other than for reward. The Magistrate convicted the accused but his decision was reversed by the Supreme Court of Victoria which held that the disputed evidence should not have been admitted. The High Court allowed the appeal and held that the evidence had been rightly admitted. There are a number of statements in the judgments which, if looked at in isolation, appear to support Justice Dawson's approach. For example, Justice Evatt concluded that the evidence was rightly admitted as the "only rational explanation" of the evidence was that the accused was carrying passengers for reward.28 Similarly, Justice Dixon, with whom 24 25 Pfennig [1995] HCA 7; (1995) 182 CLR 461 at 525 per McHugh J. For example, see Hoch [1988] HCA 50; (1988) 165 CLR 292 at 300 per Brennan and Dawson JJ; Harriman [1989] HCA 50; (1989) 167 CLR 590 at 593-594 per Brennan J and at 610 per Toohey J; Thompson v The Queen [1989] HCA 30; (1989) 169 CLR 1 at 16 per Mason CJ and Dawson J; and B v The Queen [1992] HCA 68; (1992) 175 CLR 599 at 618 per Dawson and Gaudron JJ. Sutton [1984] HCA 5; (1984) 152 CLR 528 at 564 per Dabvson J. [1936] HCA 23; (1936) 55 CLR 367. Also see Sutton [1984] HCA 5; (1984) 152 CLR 528 at 564 per Dawson J and Hoch [1988] HCA 50; (1988) 165 CLR 292 at 294 per Mason CJ, Wilson and Gaudron JJ. Martin v Osborne [1936] HCA 23; (1936) 55 CLR 367 at 382. 26 27 28 Chief Justice Latham agreed, held that, to be admissible, "the evidentiary circumstances must bear no other reasonable e~planation".2~ However, it is submitted that these statements must be seen in context. Although this was a case involving propensity evidence, as it was decided before Boardman there is no suggestion in any of the judgments that in order to be admissible the evidence must possess a certain degree of probative force. In particular, there is no suggestion that the probative force of the evidence must outweigh its prejudicial effect. The appellant argued, on the basis of Makin, that such evidence is inadmissible if it is relevant only to show that the accused was the kind of person likely to commit the offence charged. The admissibility of the evidence therefore depended upon its relevance?(' and it is in this context that the judgments must be understood. In considering the admissibility of the evidence their Honours had to determine whether the evidence possessed relevance beyond showing propensity, not whether it possessed a particular degree of probative force?' Secondly, as the trial was heard before a magistrate, the judges were concerned to ensure not only that the evidence was rightly admitted but that the magistrate's decision was ~ supportable on the e ~ i d e n c e ?Therefore statements to the effect that the probative force of the evidence was extremely high related to the correctness of the magistrate's decision, As rather than to admi~sibility.~~ this was a case involving circumstantial evidence, the magistrate had to apply the same standard which a jury would apply in such a case. That is, he had to be satisfied on the whole of the evidence that there was no rational explanation of the evidence consistent with the innocence of the accused. This test relates to the ultimate issue of guilt?4 and it is in this context that the phrase "no rational explanation" was used in this case. There is no suggestion in any of the judgments that the evidence must satisfy the same test in order to be admissible. Justice Dawson therefore took an extraordinary step by taking the statements in these judgments and applying them in the context of the admissibility of propensity evidence. Unlike the judges in Martin v Osborne, his Honour was not applying these statements to the question of relevance, nor to the determination of the ultimate issue of guilt. Rather, he was applying this test to determine whether the propensity evidence had sufficient probative force to outweigh its prejudicial effect. Such an approach was presumably based on the assumption that if the evidence satisfies this test then it must be so probative that any prejudicial effect must necessarily be outweighed. Such an approach is clearly not supportable on the basis of Martin v Osborne, where it was held that "the test of 29 30 31 32 33 34 At 375. At 369 per Fullagar KC. At 375 per Dixon J, at 384-385 and 392 per Evatt J and at 403 per McTiernan J. At 380 per Evatt J. At 383 per Evatt J and at 404 per McTiernan J . Knight v The Queen [1992] HCA 56; (1992) 175 CLR 495 at 502-503 per Mason CJ and Dawson and Toohey JJ. 294 CLOUGH - A RATIONAL VIEW OF PROPENSITY EVIDENCE" admissibility of evidence is satisfied if the evidentiary fact points to the required conclusion as either the more plausible explanation out of those which are conceivable or at any rate a plausible one among those conceivable"?5 This is in contrast to the ultimate question of guilt, where the jury must be satisfied that the evidence as a whole bears no rational explanation consistent with the innocence of the accused.36 Nor, as will be argued below, is such an approach supported as a matter of logic. It does not follow that, because propensity evidence is a species of circumstantial evidence, the test to be applied by the jury in determining guilt is somehow appropriate to determine the admissibility of that evidence.37 In particular, it is incorrect to assume that by applying that test the probative force of the evidence must necessarily outweigh its prejudicial effect. Nonetheless, the "no rational view" test was subsequently adopted by a majority of the High Court in H o ~ and~ ~ h Pfennig?9 In both decisions the same reasoning as that adopted by Justice Dawson in Sutton is evident. That is, in order to be admissible the propensity evidence must possess a "strong degree of probative force" or the probative force must "clearly transcend its prejudicial effect". Because propensity evidence is a special type of circumstantial evidence the trial judge must apply the same test as a jury must apply in dealing with circumstantial evidence and ask whether there is a rational view of the evidence that is consistent with the innocence of the accused. Here "rational" must be taken to mean "reasonable" and the trial judge must ask himself or herself the question in the context of the prosecution case. ... Only if there is no such view can one safely conclude that the probative force of the evidence outweighs its prejudicial e f f e ~ t . ~ It is important to note that, according to the majority, it is only where there is no other rational view of the evidence that the probative force of the propensity evidence can be said to outweigh its prejudicial effect. In this way the "no rational view" test becomes a test of admissibility as opposed to a factor to be taken into account in determining the 35 36 37 38 39 Martin v Osborne [1936] HCA 23; (1936) 55 CLR 367 at 391 per Evatt J. At 375 per Dixon J. Odgers, "Proof and Probability" (1989) 5 Aust Bar Rev 137 at 141. [1988] HCA 50; (1988) 165 CLR 292 at 296 per Mason CJ, Wilson and Gaudron JJ. [1995] HCA 7; (1995) 182 CLR 461 at 483 per Mason C J , Deane and Dawson JJ. Also see S v The Queen [1989] HCA 66; (1989) 168 CLR 266 at 287 per Gaudron and McHugh JJ; Thompson [1989] HCA 30; (1989) 169 CLR 1 at 18 per Mason CJ and Dawson J; and Harrirnan [1989] HCA 50; (1989) 167 CLR 590 at 602 per Dawson J, at 607 per Toohey J and at 614 per Gaudron J. Pfennig [1995] HCA 7; (1995) 182 CLR 461 at 483 per Mason CJ, Deane and Dawson JJ. Justice Toohey, although also dismissing the appeal, showed only qualified support for the "no rational view" test (at 506-507). Justice McHugh, while showing some support for such a test, did not consider that it was appropriate in all cases (at 530). 40 probative force of the evidence. The advantage of such an approach is twofold. First, it provides a question which, if answered in the affirmative, ensures that the propensity evidence is sufficiently probative. Secondly, it avoids the criticism that the balancing of probative force and prejudicial effect resembles "the exercise of a discretion rather than the application of a p r i n ~ i p l e " . ~ ~ the test does not involve a balancing of probative force As and prejudicial effect, it appears to be premised on the assumption that if the test is satisfied the propensity evidence must necessarily be so probative that any prejudicial effect is outweighed. To determine whether this premise is correct it is necessary to examine the application of the test in particular circumstances. THE APPLICATION OF THE "NO RATIONAL VIEW" TEST Rather than requiring the trial judge to engage in a balancing of probative force and prejudicial effect, the "no rational view" test substitutes a question: is there a rational view of the evidence that is consistent with the innocence of the accused? However, the judgments do not make clear precisely which evidence they are referring to. There would appear to be two possibilities: 1. the trial judge must look for another rational explanation of the impugned propensity evidence that is consistent with innocence. Only if there is no rational explanation for the propensity evidence itself will it be admissible; or 2. the trial judge must consider the propensity evidence together with the other prosecution evidence and determine whether there is any "rational explanation of the prosecution case that is consistent with the innocence of the ac~used".~2 In the first situation, the explanation that is consistent with innocence relates to the propensity evidence itself. That is, the accused disputes the reliability or the existence of that evidence, thereby attempting to destroy its probative force. The most common example of such a situation is where the accused alleges that the evidence is fabricated. In the second situation, the existence or reliability of the propensity evidence is not disputed and the other rational explanation is directed toward the prosecution case as a whole, including the propensity evidence. It is therefore submitted that the test is applied differently according to whether the propensity evidence is disputed by the accused or not. 41 42 At 483 per Mason CJ, Deane and Dawson JJ. At 530 per McHugh J (emphasis added). An alternative formulation was adopted by Pincus JA in Wackerow v R [I9981 1 Qd R 197 at 204 and R v Ingram (Unreported, Qld Court of Appeal, 27 August 1996) at 3. His Honour, quoting from Pfennig, held that there must be no other reasonable view of the evidence other than as supporting an inference that the accused is guilty of the offence charged. If such an approach were to be followed, it would potentially be far more favourable to the admissibility of propensity evidence. 296 CLOUGH - A RATIONAL VIEW OF PROPENSITY EVIDENCE? Where the Evidence is Not in Dispute The first interpretation is clearly inapplicable in those cases where the propensity evidence is undisputed. For example, where the propensity evidence consists of the accused's prior convictions and the accused does not challenge those convictions. In such cases the accused does not put forward another rational view of the undisputed evidence. For the accused to say that he or she has an alibi or was wrongly identified does not provide another rational view of the propensity evidence itself; rather it provides another view of the prosecution case. In this context the "no rational view" test requires the trial judge to consider the propensity evidence in the context of the prosecution case and determine whether there is a rational view of the prosecution case, including the propensity evidence, which is consistent with innocence. This approach is illustrated by the facts in Pfennig itself. This case involved the murder of a 10 year old boy, Michael Black, at Murray Bridge in South Australia. Michael was last seen at Sturt Reserve on the Murray River and, despite intensive searches, his body was never recovered. Michael's belongings were found neatly stacked further upstream at Thiele Reserve. There was evidence that the accused had been at Sturt Reserve and had been seen in conversation with Michael. The accused had been seen leaving Sturt Reserve at about the same time as Michael was last seen and there was evidence from which it could be inferred that his white Kombi van was at Thiele Reserve after Michael's belongings were placed there. The propensity evidence consisted of the appellant's conviction, following a plea of guilty, for the abduction and rape of a 13 year old boy some 12 months after Michael Black disappeared. The circumstances of that offence were that the complainant had been riding his bicycle past the appellant's van when the appellant lured him into the van. The appellant also took the bicycle into the van but later left it at the top of a The complainant was bound, gagged and blindfolded and the appellant committed various sexual offences until the following day when the complainant managed to escape. The trial judge held that this evidence was admissible. On all of the evidence it was possible to exclude drowning as an alternative explanation for Michael's disappearance, and it was conceded by the appellant that the only rational alternative to drowning was abduction. The trial judge held that abduction would most likely have been for a sexual purpose and therefore, given the unusual nature of the crime, it would be an affront to common sense to postulate that there could have been another person at the reserve at that time, with a propensity to abduct and sexually assault young boys, who also had the means and the opportunity to do so. Considered in the light of the prosecution case, the evidence showed the accused's criminal propensity, in particular his modus operandi in abducting 43 The Crown case was that this was an attempt by the accused to lay a false trail, and a connection was drawn between this and the fact that Michael's clothing was found piled neatly near the river: at 508 per Toohey J. and sexually assaulting young boys, and this was likely to have been similar to that adopted by the abductor in this case. In dismissing the appeal, the majority agreed that there was no reasonable view of the evidence which was consistent with the innocence of the accused. Therefore, in determining the admissibility of the propensity evidence, the trial judge must consider the entire prosecution case (including the propensity evidence) together with any other innocent explanation put forward by the accused and determine whether, in his or her view, it is reasonable to suppose that the accused could be innocent. In other words, in order for the evidence to be admissible the trial judge must be of the view that the accused is guilty. To require the trial judge to determine the admissibility of evidence on the basis of his or her assessment of the accused's guilt is unique in the law of evidence.44 "Admissibility depends on cogency to prove another fact, not upon the likelihood that the jury will convict the accused person if they accept the similar fact evidence."45 It has been suggested that such an approach is justifiable on the basis that, if there is no rational explanation for the evidence consistent with innocence, then "a reasonable jury would, if they accept the ~~ evidence, regard it as being inconsistent with i n n ~ c e n c e " .There is therefore no prejudice to the accused even though an irrational jury may also have convicted the accused, because "the irrational jury would only be doing, for the wrong reasons, what a rational jury would do for the right reasons".47 This rationale is objectionable on a number of grounds. First, it is inappropriate for admissibility to be dependent upon the trial judge's finding as to the guilt of the accused. This seems to suggest that we need not be concerned about the prejudicial effect of the evidence on the jury because the trial judge is satisfied that the accused is guilty. It also assumes that the trial judge will be immune to the influence of the prejudicial effect of the evidence. In addition, such an approach raises obvious concerns as to perceptions of bias in the trial judge's handling of the trial, given that he or she has already been required to form a preliminary view that the accused is guilty. It would be extremely prejudicial were jurors to become aware that because the evidence was admitted the trial judge had necessarily formed such a view. Secondly, it assumes, incorrectly, that the trial judge has made his or her decision based on all of the available evidence. The trial judge will almost invariably be deciding the admissibility of propensity evidence on a voir dire either before or during the prosecution case. The defence is therefore placed in an invidious position. To what extent must they disclose their case in order to challenge the admissibility of the evidence? The defence may not wish to disclose their case at such an early stage, as to do so may fortuitously 44 45 46 47 At 517 per McHugh J. Also see R v H [I9951 2 AC 596 at 61 1 per Lord Mackay LC. Sutton [1984] HCA 5; (1984) 152 CLR 528 at 543 per Brennan J. At 564 per Dawson J. Palmer, "The Scope of the Similar Fact Rule" [1994] AdelLawRw 4; (1994) 16 Adel LR 161 at 168. 298 CLOUGH - A RATIONAL VIEW OF PROPENSITY EVIDENCE? advantage the prosecution. Further, the strength of the defence case may only be apparent after the witnesses have been fully examined. However, not to disclose the substance of the defence is to run the risk that the trial judge will more easily be satisfied that there is no other rational view of a prosecution case that is not effectively contradicted. However, it is submitted that the more fundamental flaw in the "no rational view" test is that it does not achieve what it is intended to achieve. As outlined above, the adoption of the "no rational view" test appears to be premised on it being such a stringent test that, if satisfied, the evidence must be so probative that it necessarily outweighs any attendant prejudicial effect.48 Therefore it can only be suitable as a general test of admissibility if this premise is correct in all cases. It is submitted that this premise is false. While it is true that the "no rational view" test evolved out of the direction which the trial judge may give to the jury in cases involving circumstantial evidence, it is incorrect to state that the trial judge is applying the same test in determining the admissibility of propensity evidence. The two tests are fundamentally different. When a jury considers a case based on circumstantial evidence they must be satisfied that the evidence bears no reasonable explanation other than the guilt of the accused.49 They make this assessment based on all of the evidence. In contrast, the trial judge will almost invariably be deciding the admissibility of propensity evidence on a voir dire either before or during the prosecution case. Therefore, unlike the jury, he or she will not have the benefit of hearing all of the defence evidence. For this reason it is submitted that it is incorrect to assume that evidence which satisfies this test must necessarily be extremely probative. Because the trial judge hears the propensity evidence in the context of the prosecution evidence and with limited exposure to the defence case, it may be relatively easy to be satisfied that the only rational inference to be drawn from that evidence is guilt. The view that it is a difficult test to satisfy seems to be based on the incorrect assumption that to be admissible the propensity evidence must by itself prove the guilt of the accused beyond reasonable doubt.50 Of course the probative force of propensity evidence must be considered in the context of the prosecution case.5' However, there is a fundamental 48 A number of commentators and judges have criticised the "no rational view" test on the basis that it sets the standard of admissibility too high. See Odgers, "Proof and Probability" (1989) 5 Aust Bar Rev 137 at 144; Palmer, "Pfennig v R: Two Versions of the Similar Fact Rule" [1995] MelbULawRw 32; (1995) 20 MULR 600 at 604; Palmer, "The Scope of the Similar Fact Rule" [1994] AdelLawRw 4; (1994) 16 Adel LR 161 at 168; Harris, "Propensity Evidence, Similar Facts and the High Court" (1995) 11 QUT W 97 at 113; Nair, "Weighing Similar Facts and Avoiding Prejudice" (1996) 112 LQR 262 at 267; and Pfennig [1995] HCA 7; (1995) 182 CLR 461 at 5 16 per McHugh J. Also see R v Best (Unreported, Victorian Court of Appeal, 23 July 1998) at 7 per Callaway JA. Plomp v The Queen [1963] HCA 44; (1963) 110 CLR 234. Odgers, "Proof and Probability" (1989) 5 Aust Bar Rev 137 at 144. Sutton [1984] HCA 5; (1984) 152 CLR 528 at 532-533 per Gibbs CJ and at 549 per Brennan J. 49 50 51 difference between determining the probative force of the evidence in the context of the prosecution case, and determining whether there is another rational view of the prosecution case including the propensity evidence. The view that there is no rational view consistent with innocence becomes a reflection of the prosecution case as a whole. This does not necessarily warrant the conclusion that the propensity evidence must therefore be sufficiently probative so as to transcend its prejudicial effect. For example, for the propensity evidence in Pfennig even to be considered relevant, the following inferences had to be drawn by the trial judge. As Michael's body was never recovered he could be presumed to be dead. Drowning could be excluded as a reasonable possibility because searches of the river were unsuccessful, and there was evidence which suggested that it was highly unlikely he would have gone swimming alone at that spot. Therefore, it could be concluded that he was murdered. It was conceded by the defence that if drowning were excluded as the cause of death the most likely scenario was that he was abducted. The trial judge held that extortion could be ruled out as a possible motive as no ransom demand was received, as could a "mindless killing" which remained as a "mere theoretical possibility" .52 All of these inferences were clearly supportable on the facts. However, it was the next inference which was crucial, bearing in mind that whether there is another rational view is determined on the basis of the prosecution case including the propensity evidence. His Honour concluded that the most likely motive for the abduction was sexual, a conclusion Once which Justice McHugh described as never rising above "surmise or spe~ulation".~~ this inference is drawn the propensity evidence is seen in a different light and it appears that there is no reasonable explanation other than that the accused committed the offence. However, this conclusion stems from the probative force of the prosecution case as a whole, including the propensity evidence, rather than considering the probative force of the propensity evidence in the context of the prosecution case. The crucial inference that Michael Black was abducted for sexual purposes becomes inseparable from the propensity evidence and both become mutually supportive. It is hardly surprising that the trial judge, given his knowledge of the accused's conviction, considered abduction for sexual purposes to be the most likely explanation. The trial judge clearly directed the jury that the propensity evidence was only relevant to the identify of the person who committed the offence, and that they should not regard it as relevant unless they were satisfied beyond reasonable doubt that Michael Black had in fact been abducted and murdered. However, the potential prejudicial effect of the evidence in this context is apparent. It is difficult to believe that the jury would not be influenced by this evidence in determining whether Michael Black was abducted, particularly when they 52 53 Pfennig [1995] HCA 7; (1995) 182 CLR 461 at 486 per Mason CJ, Deane and Dawson JJ. At 535. 300 CLOUGH - A RATIONAL VIEW OF PROPENSITY EVIDENCE'? were told by the trial judge that if he was abducted, the most likely motive for the abduction was a sexual one.54 Therefore it is submitted that the "no rational view" test does not necessarily ensure that the prejudicial effect of the propensity evidence is outweighed by its probative force. However, it is equally possible that it may exclude propensity evidence which is in fact sufficiently probative. For example, the accused has an alibi and claims that the offence charged is a case of mistaken identity. The trial judge must consider whether this is a reasonable explanation given the prosecution case and the propensity evidence. While the prosecution case may be very strong, this does not mean that the alibi evidence is not a reasonable possible explanation. If it is a reasonable explanation consistent with innocence then the propensity evidence must be excluded. However, it does not follow that the possibility of another reasonable explanation necessarily destroys the probative force of the propensity evidence. The test avoids the fundamental consideration of the probative force of the propensity evidence itself, and becomes an evaluation of the strength of the prosecution case, including the propensity evidence, against which the reasonableness of the proposed innocent explanation is measured. Where the Evidence is Disputed The situation becomes more complex where the accused is able to dispute the propensity evidence itself. The most common situation in which this will arise is where an accused is charged with several offences which are tried together, and the prosecution seeks to admit the evidence of each complainant as evidence in respect of the other offences. Applying the "no rational view" test to such cases, the prosecution must satisfy the trial judge that there is no rational explanation of the evidence other than the guilt of the accused. As the probative force of such evidence "lies in the improbability of the witnesses giving accounts of happenings having the requisite degree of similarity unless the happenings o~curred"?~ the rational view must be directed at explaining those similarities. Broadly speaking, such evidence is said to be admissible on the basis that the statements are so similar that, when judged by common sense, they must "either all be true, or have arisen from a cause ~~ common to the witnesses or from pure c ~ i n c i d e n c e " .The prosecution must therefore satisfy the trial judge that there is no reasonable possibility of the similarities being coincidental or the product of collusion. This is illustrated by the decision of the High Court in Hoclz. The applicant, who had been a part-time recreation officer at a home for young boys in Brisbane, was convicted on three counts of unlawfully and indecently dealing with a boy under the age of 14 years. Two of 54 55 56 At 474 per Mason CJ, Deane and Dawson JJ. Note that in this case the trial judge considered the admissibility of the propensity evidence after hearing most of the prosecution case. Hoch [1988] HCA 50; (1988) 165 CLR 292 at 295 per Mason CJ, Wilson and Gaudron JJ. Boardman [I9751 AC 421 at 444 per Lord Wilberforce. the complainants were brothers and the third was a friend of one of the brothers. There was apparently some antipathy on the part of the boys towards the applicant, which may have been independent of the alleged offences. The trial judge rejected an application that each of the three counts be tried separately, and it was this decision which formed the basis of the appeal. In a joint judgment, Chief Justice Mason and Justices Wilson and Dawson held that the probative force of such evidence stems from the "improbability of similar lies".s7 Consequently, if there is a possibility of concoction, the evidence is inadmissible, as concoction provides a rational explanation for the similarities that is inconsistent with the guilt of the accused. "It thus destroys the probative value of the evidence which is a condition precedent to its admissibility."58 On the facts of this case concoction was clearly a "reasonable explanation" and therefore the evidence was inadmis~ible.~~ In applying the "no rational view" test in such cases the trial judge must, as a matter of logic, first consider the possibility of concoction and then consider the possibility of coincidence. This is because whether coincidence can be discounted as a reasonable possibility will depend to a large extent on the similarities between the allegations. This will in turn be effected by the possibility of concoction. If there is a chance that there was some collusion it places the apparent similarities in an entirely different light. Therefore the possibility of concoction becomes a threshold issue. If there is a reasonable possibility of concoction, the evidence is inadmissible. Although it was not necessary for the court to address the issue, it would seem that, where the possibility of concoction is discounted, the "no rational view" test must then be applied to the evidence in the same way as it would apply if the evidence were not disputed.60 The absence of concoction only addresses one possible innocent explanation. It remains for the trial judge to consider the possibility of coincidence. The test of admissibility is therefore in two stages. First, in assessing the possibility of concoction the trial judge is concerned with the propensity evidence itself and not the prosecution case as a whole. Although evidence of concoction helps explain the similarities in the evidence, other aspects of the defence case are irrelevant to this issue. For example, the fact that the accused has an alibi does not destroy the probative value of the evidence, based as it is on the improbability of similar lies. It merely provides an alternative view of the prosecution case. It would, however, be relevant to the "second stage" of the test in such cases. That is, the possibility of an alibi may provide another rational view of the prosecution case as a whole, including the propensity evidence, which is consistent with the innocence of the accused. 57 58 59 60 Hoch [1988] HCA 50; (1988) 165 CLR 292 at 295. At 296. At 297. R v H [I9951 2 AC 596 at 615 per Lord Mustill. 302 CLOUGH - A RATIONAL VIEW OF PROPENSITY EVIDENCE.? While it is accepted that such an approach is a logical application of the "no rational view" test, it is submitted that it gives rise to a number of significant practical difficulties which further undermine this test as a general test of admissibility. First, the test is extremely difficult to satisfy, as there need only be the possibility of concoction for the probative value of the evidence to be destroyed. Although the other rational view must be a "reasonable" explanation,61 their Honours clearly rejected any requirement that there must be a "probability" or "real chance"62 of concoction. Indeed, ~f the "no rational view" test is to be applied, this must be the case, as the test reflects the criminal standard of proof. So long as the doubt is reasonable, even if it is only a possibility, the evidence is inadmissible. It would be inconsistent with the nature of the test to introduce a lower standard.'j3 However, it does not follow that a reasonable possibility of concoction, no matter how slight, necessarily destroys the probative force of the evidence. This is particularly the case in circumstances not involving deliberate concoction or collusion, but "innocent i n f e ~ t i o n " .That is, in circumstances where details of the offences may have innocently ~~ been shared between complainants. In such cases the evidence must also be excluded, as the possibility of innocent infection also explains the similarities in a way that is consistent with innocence. However, the possibility of innocent infection may be present in a large number of cases, particularly those involving sexual offences within families. It may also arise as a result of the complainant becoming aware, for example through the media, of similar allegations having been made against the accused.65 To automatically exclude such evidence on the basis of a possibility of concoction is to place an unrealistic and unnecessary obstacle in the path of the prosecution, as it effectively requires the prosecution to prove beyond reasonable doubt that there was no innocent infection.66 This issue was specifically addressed by the House of Lords in R v H. The appellant had been convicted of various sexual offences against his adopted daughter and step-daughter and there was clearly the possibility of innocent infection between the two complainants. The House of Lords unanimously rejected the argument, based partly on Hoclz, that the evidence of one complainant would not be admissible in support of the other unless the trial judge was satisfied that there was no possibility of the evidence being the product of collusion, whether deliberate or innocent. It was held that credibility may be affected in a number of ways, collusion being one of them, and such an approach would require the trial 61 62 63 64 65 66 Hoch [1988] HCA 50; (1988) 165 CLR 292 at 297 per Mason CJ, Wilson and Gaudron JJ. At 296. R v Best (Unreported, Victorian Court of Appeal, 23 July 1998) at 9 per Callaway JA. R v H [I9951 2 AC 596 at 616 per Lord Mustill. BRS v The Queen [1997] HCA 47; (1997) 71 ALJR 1512 at 1525 per Gaudron J. At 1525. judge to determine a question of fact which should rightly be determined by the jury?' Of course there may be exceptional cases where the risk of concoction is so apparent that the evidence should not go to the jury. This would arise in those cases where evidence of collusion is so apparent that the trial judge is of the view that no reasonable jury could possibly accept the evidence as being free from c o l l ~ s i o n ?In such cases it is clear that ~ the probative force of such evidence is not sufficient to outweigh its prejudicial effect.69 In cases falling short of this extreme the trial judge should assume that the witness will come up to proof.70 The question of whether they are ultimately to be believed is a matter for the jury. Secondly, there is the practical issue of how the trial judge is to be satisfied that such a possibility does not exist. The majority in Hoch held that the determination of this issue would not necessarily involve an examination on a voir dire. The trial judge is not required to determine whether or not there was concoction, but rather whether concoction is a reasonable explanation. This will often be determined by the relationship between the witnesses, which in many cases will be apparent from the dep0sitions.7~ However, concoction, and particularly the possibility of innocent infection, will often not be so apparent. It would seem that in such cases it would be necessary for the witnesses to be examined on a voir dire as well as during the This of course adds to the length of the trial but more importantly increases the considerable difficulties already faced by complainants in sexual offence cases. A PREFERABLE APPROACH If, as is suggested, the "no rational view" test is rejected as the test of admissibility for propensity evidence, the underlying basis upon which such evidence is said to be admissible will not alter. The trial judge will still be required to determine the admissibility of the evidence based on a comparison of its probative force and prejudicial e f f e ~ t . 7To reject the "no rational view" test is not to reject this basic premise, but rather ~ to reject the proposition that this test gives effect to it in all cases. 67 68 69 R v H [I9951 2 AC 596 at 617 per Lord Mustill. See Nair, "Weighing Similar Facts and Avoiding Prejudice" (1996) 112 LQR 262 at 271. R v H [I9951 2 AC 596 at 61 1 per Lord Mackay LC. In Hoch [1988] HCA 50; (1988) 165 CLR 292 at 300-301, Justices Brennan and Dawson did not adopt the "no rational view" test, but rather held that where there is a "real chance" of conspiracy the evidence cannot be said to possess the requisite degree of probative force which would justify its admission. R v H [I9951 2 AC 596 at 604-605 per Lord Mackay LC and at 621-622 per Lord Mustill. Hoch [1988] HCA 50; (1988) 165 CLR 292 at 297 per Mason CJ, Wilson and Gaudron JJ and at 303-303 per Brennan and Dawson JJ. Also see R v McKellin (Unreported, Victorian Court of Appeal, 19 December 1997) at 10 per Vincent AJA. R v H [I9951 2 AC 596 at 615 per Lord Mustill. That this is the criterion of admissibility has been accepted in Australia (Hoch [1988] HCA 50; (1988) 165 CLR 292 at 294 per Mason CJ, Wilson and Gaudron JJ), England (Director of Public Prosecution v P [I9911 2 AC 447 at 460 per Lord Mackay LC), Canada (R v B(CR) [I9901 70 71 72 73 CLOUGH - A RATIONAL VIEW OF PROPENSITY EVIDENCE? In what form, then, should the test of admissibility be expressed? It is submitted that the courts should not revert to speaking of probative force "outweighing" or "transcending" prejudicial effect, as such a formulation does not accurately reflect the exercise in which the trial judge is engaged. It is misleading to speak of probative force "outweighing" the prejudicial effect of the evidence, as the two are "incommensurable". One goes to the issue of proof while the other is concerned with the fairness of the The prejudicial effect which the law seeks to guard against is the "undue impact, adverse to an accused, that the evidence may have on the mind of the jury over and above the impact that it might Prejudicial be expected to have if consideration were confined to its probative f o r ~ e " ? ~ effect in this sense does not "disappear" because that evidence has strong probative f0rce.7~As there is no direct correlation between the two factors, an increase in probative force does not necessarily result in a decrease in prejudicial effect. On the contrary, the prejudicial effect may well increase as the probative force of the evidence increases. What the test of admissibility requires is for the trial judge to assess the likely prejudice to the accused if the evidence is admitted. This is a matter of judgment rather than precise calculation. "If there is a real risk that the admission of such evidence may prejudice the fair trial of the criminal charge before the court, the interests of justice require the trial The judge to make a value judgment, not a mathematical cal~ulation."~~ trial judge must therefore determine to what extent the propensity evidence is likely to have an unacceptable prejudicial effect on the jury. By "unacceptable" is meant that the prejudicial effect is such that it cannot be addressed by other means such as an appropriate direction to the jury, but rather can only be dealt with by exclusion of the evidence. This is a question of judgment, not balancing, requiring the trial judge to determine whether it is "the peg of relevance or the dirty linen hung thereon, upon which the jury is going to ~oncentrate".~~ It is therefore submitted that the test of admissibility for propensity evidence should be formulated in terms of whether the evidence is sufficiently probative that it would not be unfair to admit it despite its inherent prejudicial effect. "Unfair" in this context refers to the fairness of the accused's trial. The evidence will only be admissible if the trial judge forms the view that there is no significant risk that the admissibility of the evidence will 1 SCR 717) and New Zealand ( R v Accused (1991) 7 CRNZ 604). The possibility of another reasonable explanation consistent with innocence would of course continue to be relevant to the probative force of the evidence. Pfennig [1995] HCA 7; (1995) 182 CLR 461 at 528 per McHugh J. Also see Tapper, "Trends and Techniques in the Law of Evidence" in Birks (ed), Pressing Problems in the Law Volurnr 1: Criminal Justice and Human Rights (Oxford University Press, Oxford 1995) pp28-29. Pfennig [1995] HCA 7; (1995) 182 CLR 461 at 487-488 per Mason CJ, Deane and Dawson JJ. At 5 12-5 13 per McHugh J. Cf Hoffman, "Similar Facts After Boardman" (1975) 9 1 LQR 193 at 194, quoted with approval in Sutton [1984] HCA 5; (1984) 152 CLR 528 at 534 per Gibbs CJ. Pfennig [1995] HCA 7; (1995) 182 CLR 461 at 528-529 per McHugh J. Also see R v H [I9951 2 AC 596 at 621 per Lord Mustill. Stone, "The Rule of Exclusion of Similar Fact Evidence: England" (1932) 46 Harv L Rev 954 at 984. 74 75 76 77 78 result in a miscamage of justice. Such an approach does not alter the basis on which the evidence is to be admitted, or the factors that are relevant to the question of admissibility. The change in terminology does no more than describe more accurately the essential process that lies belzlnd the admissibility of propensity evidence. An alternative formulation, and that which is adopted in England and Victoria, is that in order to be admissible the probative force of the propensity evidence must be "sufficiently great to make it just to admit the evidence, notwithstanding that it is prejudicial to the accused".79 While this test focuses correctly on the potential for injustice if the evidence is admitted, it is submitted that the use of the word "just" is ambiguous. On the one hand it may be interpreted as being concerned with fairness to the accused. That is, it will not be just to admit the evidence unless it would not be unfair to admit the evidence.80 However it may equally be argued that the word "just" should be interpreted more broadly. For example, Justice Toohey has held that it is not only the interests of the accused which are involved, "the legitimate interests of the Crown and of the community cannot be 0verIooked".8~ Similarly, Justice McHugh has held that the admissibility of propensity evidence can only be justified if the trial judge concludes that the probative force of the evidence when compared with its prejudicial effect is such that "fair minded people would think that the public interest in adducing all relevant evidence of guilt must have priority over the risk of an unfair With respect, it is submitted that this view is incorrect. The public interest in an offender being convicted only extends to that person being convicted after a fair trial. The admissibility of propensity evidence is concerned precisely with the issue of whether there If is a significant risk of an unfair trial in the circ~mstances.8~ there is no significant risk that the trial will be unfair then the evidence is rightly admitted and there is no need to consider any other public interest. If, however, there is a significant risk that the trial will be unfair then the interests of the community in the conviction of offenders is irrelevant. "[Tlhe public interest in holding a trial does not warrant the holding of an unfair 79 80 81 82 83 84 Director of Public Prosecutions v P 119911 2 AC 447 at 460 per Lord Mackay LC with whom the other Law Lords agreed and s398A Crimes Act 1958 (Vic). This approach was also approved of by Justice Toohey in Pfennig [1995] HCA 7; (1995) 182 CLR 461 at 507. This is to some extent supported by the interchangeable use of the words "just" and "fair" by members of the House of Lords. See Director of Public Prosecutions v P 119911 2 AC 447 at 462-463 per Lord Mackay LC and R v H [I9951 2 AC 5% at 603 per Lord Mackay LC and at 613 per Lord Griffiths. Pfennig [1995] HCA 7; (1995) 182 CLR 461 at 507. At 529. Sutton ( [1984] HCA 5; 1984) 152 CLR 528 at 547 per Brennan J. Jago v District Court of New South Wales [1989] HCA 46; (1989) 168 CLR 23 at 31 per Mason CJ. 306 CLOUGH - A RATIONAL VIEW OF PROPENSITY EVIDENCE'? IS THERE A ROLE FOR THE GENERAL DISCRETION TO EXCLUDE? In the past, the characterisation of the test of admissibility as one of balancing probative force and prejudicial effect has given rise to concern over the apparently discretionary nature of such a process. This was clearly illustrated by the majority's rationalisation of the "no rational view" test in Pfennig: unless the tension between probative force and prejudicial effect is governed by such a principle, striking the balance will continue to resemble the exercise of a discretion rather than the application of a principle.85 As outlined above, whether the test of admissibility is a question of law or the exercise of a discretion is important for two reasons. First, if it is a matter of discretion the onus of demonstrating that the prejudicial effect of the evidence outweighs its probative force rests on the accused. Secondly, appellate courts will be more reluctant to interfere with the exercise of a discretion as opposed to the application of a rule of law.86 Although the admissibility of propensity evidence is quite clearly a question of law and not discretion?' it is arguably an example of what may be described as a "concealed as ~~ opposed to an "overt" d i ~ c r e t i o n .An overt discretion is one in which the trial judge is truly able to decide between two or more legally permissible alternative courses of action. A concealed discretion is not truly a discretion at all. It is a term used to describe those situations where, in applying a rule of law, the trial judge has "considerable freedom of choice ... because [the rules] contain value-qualified precepts which require a personal assessment of the circumstances".89 In comparing the probative force and prejudicial effect of the evidence the trial judge in fact has no true discretion. If he or she concludes that the evidence is sufficiently probative the evidence must be admitted. If it is unduly prejudicial, it must be excluded?O This is not to ignore the reality that in "balancing" the two factors the trial judge is exercising a value judgment which allows him or her to determine the outcome as much as if the discretion were overt. To say this is merely to acknowledge the inherent flexibility of such a test; it is not to say that it is a true discretion. 85 86 87 88 89 90 Pfennig [1995] HCA 7; (1995) 182 CLR 461 at 483 per Mason CJ, Deane and Dawson JJ. At 515 per McHugh J. Markby [1978] HCA 29; (1978) 140 CLR 108 at 117 per Gibbs ACJ. Pattenden, Judicial Discretion and Criminal Litigation (Oxford University Press, Oxford, 2nd ed 1990) pp 1-2. At p2. Tapper, "Trends and Techniques in the Law of Evidence" in Birks (ed), Pressing Problems in the Law Volume 1: Criminal Justice and Human Rights p29. Because the test is a rule of law, both concerns relating to onus of proof and appellate review are addressed. In terms of onus of proof, this is quite clearly on the prosecution. As for appellate review, it appears that where the trial judge applies a rule of law, no matter how vague the criteria on which it is based, an appellate court is more likely to interfere and substitute its own finding. This may be contrasted with appellate review of a true discretion, in which case the court will not usually interfere unless the trial judge has taken into account irrelevant factors or has ignored relevant factors. This is illustrated by the decision of the English Court of Criminal Appeal in R v Viola.91 The issue arose in the context of s2 of the Sexual Offences (Amendment) Act 1976 ( U K ) which prohibited cross-examination as to the sexual history of a complainant in a rape case without leave of the trial judge. Such leave could only be given "if and only if [the trial judge] is satisfied that it would be unfair to that defendant to refuse to allow the evidence to be adduced or the question to be asked. According to the Court, it is: wrong to speak of a judge's "discretion" in this context. The judge has to make a judgment as to whether he is satisfied or not in the terms of section 2. But once having reached his judgment on the particular facts, he has no discretion. If he comes to the conclusion that he is satisfied it would be unfair to exclude the evidence, then the evidence has to be admitted and the questions have to be allowed. ... [I]s this Court entitled to differ from the conclusions of the learned judge? As already pointed out, this is the exercise of judgment by the judge, not an exercise of his discretion. ... So what we have to decided is whether the judge was right or wrong in the conclusion which he rea~hed.9~ It is submitted that this quotation is equally applicable to appellate review of decisions relating to the admissibility of propensity evidence. The trial judge has to exercise his or her judgment as to whether it would not be unfair to admit the propensity evidence notwithstanding its prejudicial effect. Once that judgment is reached, he or she has no discretion but to act in accordance with it. It is therefore open to an appellate court to substitute its own judgment on the facts before it. One suspects, however, that what concerns the courts is the "discretionary" nature of the test, not in the sense of it being a true discretion, but in the sense that it requires a judgment to be and that judgment may be made differently by different judges. While consistency in judicial decisions is of course desirable, the dilemma posed by the competing principles of probative force and prejudicial effect is "fundamentally insoluble 91 92 93 (1982) 75 Cr App R 125. At 130-131 per Lord Lane CJ. "It is true that in deciding whether the evidence is admissible, questions of degree arise and that the judgment to be made is to some extent discretionary": Perry [1982] HCA 75; (1982) 150 CLR 580 at 585-586 per Gibbs CJ. Also see Markby [1978] HCA 29; (1978) 140 CLR 108 at 117 per Gibbs ACJ. 308 CLOUGH - A RATIONAL VIEW OF PROPENSITY EVIDENCE:' By and ine~capable"?~ its very nature, the test of admissibility for propensity evidence requires the trial judge to exercise his or her judgment based on the circumstances of the particular case. It is submitted that this is unavoidably a matter of judgment and not capable of resolution by such a formulaic approach as the "no rational view" test. What then, if any, is the role to be ascribed in this context to the general discretion to exclude otherwise admissible evidence the prejudicial effect of which outweighs its probative force?95 It is submitted that, once it is accepted that the admissibility of propensity evidence is determined in this way, then there can be no room for the exercise of the general discretion to e x ~ l u d e ?Despite cautious statements by judges leaving open ~ it the p~ssibility?~ is difficult to see how the position can be otherwise when "the whole purpose of restricting the admissibility of similar fact evidence is to ensure that it cannot be used unless its probative force is sufficiently strong to outweigh or transcend its prejudicial effect".98 In practical terms, how could a trial judge, after considering the probative force and prejudicial effect of the evidence and concluding that it would not be unfair to admit the evidence, then exclude the same evidence on the basis that it is unduly prej~dicial.9~ This conclusion is neither surprising nor should it give rise to any concern. There is no need for the general discretion to continue to operate, given that the same concerns have been addressed as a matter of admissibility. If anything, the position is more favourable to the accused, the onus being on the prosecution. LEGISLATIVE REFORM It has been said that one of the underlying problems in this area of the law has been "the absence of statutory intervention throwing more weight upon the precise formulation of appellate opinions than they can reasonably bear".loO Yet despite recent legislative activity in this area, the problems outlined in this article are likely to remain, or to be replaced by new problems of interpretation. 94 95 % 97 98 99 100 Heydon, Cross on Evidence at [21005]. Driscoll v The Queen [1977] HCA 43; (1977) 137 CLR 517 at 541 per Gibbs J. Pfennig [1995] HCA 7; (1995) 182 CLR 461 at 5 15 per McHugh J. Markby [1978] HCA 29; (1978) 140 CLR 108 at 116-117 per Gibbs ACJ; Perry [1982] HCA 75; (1982) 150 CLR 580 at 585 per Gibbs CJ and at 605 per Wilson J; Sutton [1984] HCA 5; (1984) 152 CLR 528 at 534 per Gibbs CJ and at 565 per Dawson J; Harriman [1989] HCA 50; (1989) 167 CLR 590 at 594-595 per Brennan J; and R v Bullen (Unreported,Victorian Court of Appeal, 23 July 1998) at 17 per Callaway JA. Sutton [1984] HCA 5; (1984) 152 CLR 528 at 565 per Dawson J. Tapper, "Trends and Techniques in the Law of Evidence" in Birks (ed), Pressing Problems in the Law Volume 1: Criminal Justice and Human Rights p29. This view has also been expressed in relation to sslOl and 137 of the Evidence Act 1995 (NSW): R v Lock (1997) 91 A Crim R 356 at 363 per Hunt CJ and R v Fordham (Unreported, NSW Court of Appeal, Howie AJA, 2 December 1997) at 16. Heydon, Cross on Evidence at [21005]. Uniform Evidence Law At Commonwealth level, reform of the law relating to propensity evidence has occurred as part of the move toward the enactment of uniform evidence legislation in all Australian jurisdictions to be modelled on the Evidence Act 1995 (Cth).lol Part 3.6 of the Commonwealth Act deals with what is referred to as "tendency" and "coincidence" evidence. Although the legislation makes some significant changes to the admissibility of such evidence, the focus of this article is on slOl which provides that in criminal proceedings such evidence will not be admissible against a defendant unless "the probative value of the evidence substantially outweighs any prejudicial effect it may have on the defendant".1°2 Apart from the addition of the word " ~ u b s t a n t i a l l y " the ~legislation ~~ therefore adopts the terminology which prevailed prior to the decisions in Hoclz and Pfennig. This was understandable given that the evolution of the "no rational view" test occurred after the Australian Law Reform Commission's final report.lo4 However, rather than taking this opportunity to revert to a balancinglo5 of probative force and prejudicial effect, which is clearly what was envisaged by the drafters of the legislation, the courts have felt constrained to follow the common law and have applied the "no rational view" test to the admissibility of evidence under s101 .I06 The problems associated with such a test therefore remain. 101 102 103 104 105 106 The Evidence Act 1995 (NSW) has already been enacted in essentially identical terms to the Commonwealth Act. The adoption of similar legislation was also recommended in Victoria: see Victoria, Parl, Scrutiny of Acts and Regulations Committee, Review of the Evidence Act 1958 (Vic) (1996). The enactment of the legislation in Victoria may, however, be open to question given the reforms outlined below. This requirement is in addition to, and effectively subsumes, the general requirement that in order for such evidence to be admissible it must have "significant probative value" (ss97(l)(b) and 98(l)(b)). It is unclear to what extent the addition of the word "substantially" will alter the basic nature of the balancing test. It was obviously intended to ensure that the probative force did more than merely outweigh the prejudicial effect (R v Lock (1997) 91 A Crim R 356 at 363 per Hunt CJ), although given the imprecise nature of the balancing test it seems somewhat artificial to speak in terms of different degrees of "outweighing". The Commission referred to the "extreme" position of Justice Dawson in Sutton: Australian Law Reform Commission, Evidence (Report No 26, 1985) para 165. Although it has been argued that the use of the word "balancing" is inappropriate, it does not preclude the test from being applied in the manner suggested in this article. R v Lock (1997) 91 A Crim R 356 at 363 per Hunt CJ and R v AH (1997) 42 NSWLR 702 at 709 per Ireland J. Also see Harris, "Propensity and Similar Fact Evidence under the New Evidence Legislation" (1996) 16 Qld Lawyer 117 at 120. 3 10 CLOUGH - A RATIONAL VIEW OF PROPENSITY EVIDENCE? Victoria The law relating to propensity evidence in Victoria was recently amended by the Crimes (Amendment) Act 1997 (Vic), s14 of which introduced a new s398A into the Crimes Act 1958 (Vic) dealing with "propensity evidence".lo7 Section 398A(2) states: Propensity evidence relevant to facts in issue in a proceeding for an offence is admissible if the court considers that in all the circumstances it is just to admit it despite any prejudicial effect it may have on the person charged with the offence. The section therefore adopts the test propounded by the House of Lords in Director of Public Prosecutions v P.108 The motivation behind these amendments appears to have been the recommendation of the Rape Evaluation Advisory Committeelogthat the rules of evidence should be reviewed as a matter of urgency to take account of the difficulties associated with prosecuting offences involving multiple counts of sexual offences.Il0 As a result, the government expressed concern that the effect of the "no rational view" test in cases of two or more complainants in sexual offence cases was that the evidence of one complainant would be inadmissible in respect of the other where there was even a possibility of concoction or collusion. The Victorian Government therefore acted to abrogate the "no rational view" test."' The relevant provision is s398A(3) which states that: The possibility of a reasonable explanation consistent with the innocence of the person charged with an offence is not relevant to the admissibility of evidence referred to in sub-section (2). However, under s398A(4): 107 108 109 110 111 The term "propensity evidence" is not defined in the legislation. It has, however, been interpreted as "evidence which is received notwithstanding that it discloses the commission of offences other than those with which the accused is charged": R v Best at 5 per Callaway JA, with whom Phillips CJ and Buchanan J agreed. While this definition will no doubt be subject to refinement, it is clear that the section is given a very wide operation. Consequently the "no rational view" test is unlikely to have any continuing application in Victoria. Victoria, Parl, Debates (9 October 1997) Vol 436 at 431 per Jan Wade, Attorney-General and R v Best at 5 per Callaway JA. Heenan and McKelvie, The Crimes (Rape)Act 1991: An Evaluation Report (Department of Justice, Melbourne 1997). At p370. Victoria, Parl, Debates (9 October 1997) Vol436 at 429 per Jan Wade, AttorneyGeneral. Although the concerns expressed by the Rape Evaluation Advisory Committee related to trials of sexual offences the provisions apply to all trials involving propensity evidence. Nothing in this section prevents a court taking into account the possibility of a reasonable explanation consistent with the innocence of the person charged with an offence when considering the weight of the evidence or the credibility of a witness. The intention behind these provisions was to ensure that the mere possibility of concoction would not render propensity evidence inadmissible. However, "implicit in the provision is the notion that where the court is satisfied that there is a substantial risk of concoction having occurred it would not be just to admit the evidence in a single trial".l12 However, it is arguable that the sections do not achieve this compromise because they are mutually inconsistent. A crucial factor in determining whether it will be just to admit propensity evidence is the weight of that evidence. Relevant to that issue is, as s398A(4) acknowledges, the possibility of another reasonable explanation. However, s398A(3) specifically states that the possibility of another reasonable explanation consistent with innocence is irrelevant to that issue. No distinction is drawn between a "mere possibility" as opposed to a "substantial risk".l13 The issue is simply said to be irrelevant. It therefore seems to preclude the possibility of another reasonable explanation from being taken into account in determining whether it is just to admit the evidence, even if there is a "substantial risk" of collusion. An alternative interpretation is that the possibility of another reasonable explanation is relevant to weight and in that way is relevant to the primary question of admissibility. That is, the possibility of another reasonable explanation may so diminish the weight of the evidence that it would be unjust to admit it. However, it is arguable that this interpretation may also fail to give effect to the objective of the legislation, as it would be open to the trial judge to find that the mere possibility of another reasonable explanation destroys the probative force of the evidence and it is therefore unjust to admit it.l14 In this way the possibility of another reasonable view would preclude the admissibility of the evidence, which is precisely what the legislation was trying to avoid. The Victorian Court of Appeal has resolved these inconsistencies by interpreting the reference to "explanation" in ss398A(3) and (4) as referring only to "explanations, like collusion and unconscious influence, that affect the truth of the propensity evidence sought .~~ to be adduced and not to extend to explanations like c ~ i n c i d e n c e " That~is, a distinction is drawn between explanations which affect the reliability of the propensity evidence, and those which affect the probative force of the evidence in the context of the prosecution case. The former are irrelevant to the question of admissibility and may only be considered by the trier of fact in determining the weight of the evidence, while the latter are relevant to determining the admissibility of the evidence under s398A(2). 112 113 114 115 Victoria, Parl, Debates (9 October 1997) Vol436 at 43 1 per Jan Wade, Attorney-General R v Best at 10 per Callaway JA. Hoch [1988] HCA 50; (1988) 165 CLR 292 at 296 per Mason CJ, Deane and Dawson JJ. R v Best at 10 per Callaway JA. . 3 12 CLOUGH - A RATIONAL VIEW OF PROPENSITY EVIDENCE? Accordingly, the phrase "all the circumstances" in s398A(2) is interpreted as referring to all the circumstances "bearing on probative value and prejudicial effect ... but not factors impugning the reliability of the evidencew.1l6 The effect of this is that "the possibility. even a strong possibility of collusion or any other matter affecting the reliability of the evidence is a matter for the jury". 117 CONCLUSION For the reasons outlined, it is submitted that the High Court should reject the "no rational view" test as the test of admissibility for propensity evidence.ll* In cases where the propensity evidence is disputed, it sets the standard of admissibility too high, particularly in cases involving sexual offences against more than one complainant. Of even greater concern is its application in cases where the propensity evidence is not disputed. While superficially appearing to be a more stringent test, it avoids the fundamental dilemma posed by the conflicting principles of probative force and prejudicial effect. To require the trial judge to determine admissibility by evaluating the strength of the prosecution case as a whole, including the propensity evidence, merely ensures that the prosecution case is highly probative. It avoids the question of whether the propensity evidence is, of itself, sufficiently probative to be safely admitted as part of that prosecution case notwithstanding its inherent prejudicial effect. While in the past an accused could challenge the admissibility of the evidence by questioning its probative force and highlighting its potentially prejudicial effect, an accused is now effectively required to explain how he or she could possibly be innocent in light of the prosecution case including the propensity evidence. Such an approach is an abdication of the trial judge's responsibility for determining the fairness of admitting such evidence which is, the courts continue to assure us, highly prejudicial and therefore only exceptionally admissible. It now appears to be uniformly accepted that the admissibility of propensity evidence depends on the strength of its probative force in comparison with its prejudicial effect. The rejection of the "no rational view" test, far from altering this basic premise, places the focus clearly upon these key issues. While it has been suggested in this article that the test should be reformulated in terms of whether the evidence is sufficiently probative that it would not be unfair to admit it notwithstanding its inherent prejudicial effect,l19 the change 116 117 118 119 At 10. At 20. Of the majority judges who supported the "no rational view" test only Justice Gaudron remains. Her Honour expressed views consistent with her earlier judgment in BRS v The Queen [1997] HCA 47; (1997) 71 ALJR 1512 at 1524-1525. Both Justice Toohey and McHugh have shown limited support for the test (Pfennig [1995] HCA 7; (1995) 182 CLR 461 at 506-507 and at 530). Also see BRS v The Queen at 1520 per Toohey J. In BRS v The Queen at 1538 Justice Kirby did not seem averse to a rearguing of the decision in Pfennig in an appropriate case. As a result of legislative changes this formulation could not be adopted in the Federal, ACT, New South Wales or Victorian courts (s101 Evidence Act 1995 (Cth), applicable in in terminology is secondary to the rationale behind it. The potential for a miscarriage of justice as a result of the admissibility of propensity evidence is always present. It is for the trial judge to determine the degree of risk and the appropriate manner in which it should be addressed. This is, unavoidably, a question of judgment, a question which the "no rational view" test fails to answer. the ACT by virtue of s4 Evidence Act 1995 (Cth), slOl Evidence Act 1995 (NSW) and s398A Crimes Act 1958 (Vic)).