University of New South Wales Law Journal
|
|
Home
| Databases
| WorldLII
| Search
| Feedback
University of New South Wales Law Journal |
|
DAVID HAMER[1]*
The admissibility of evidence often turns on its probative value. Probative value measures the strength of connection between the challenged evidence and the fact in issue. It may be assessed with differing degrees of atomism/holism. The High Court, at common law and under the Uniform Evidence Law, from Pfennig v The Queen to Phillips v The Queen to TL v The King, has adopted an extremely holistic approach – the trial judge should assess the challenged evidence together with other evidence. This introduces incoherence and uncertainty into the law. It contradicts the ‘importance’ admissibility criterion, under which the presence of other prosecution evidence may work against admissibility. The High Court’s holism appears to be the product of a fallacious conflation of proof and probative value. It risks the trial judge, at admissibility, trespassing on the jury’s fact-finding province.
Proof is the ultimate object of the criminal trial. The jury, as tribunal of fact, should weigh up all the admitted evidence and, taking account of applicable burdens and standards of proof, and the trial judge’s comments and directions, determine the defendant’s guilt. The trial judge’s function is to ensure the trial proceeds according to law. In testing the admissibility of challenged evidence, the trial judge is often required to assess its relevance and probative value.
In this article, I consider how Australian trial judges should approach relevance and probative value assessments at the admissibility stage. More specifically, to what extent are these assessments atomistic, focusing on the challenged evidence in isolation, as opposed to holistic, viewing the challenged evidence in the broader evidential context? A decade ago, Jennifer Mnookin, commenting on United States (‘US’) law, said ‘the distinction between atomism and holism is a central – perhaps even the central – tension pervading judicial determinations of evidence’.[1] Despite differences between the law of the US and Australia, Mnookin’s observations have resonance here. ‘Much of the time, there is neither explicit language within the text of the Rule nor definitive caselaw with which to provide guidance to courts’,[2] and ‘[s]urprisingly little scholarship has looked at issues of atomism and holism with respect to judicial determinations’.[3]
Broadly speaking, in Australian law, relevance and probative value assessments are atomistic, focusing on the connection between the challenged evidence and the fact in issue. Of course, prior to assessing the relevance or probative value of the challenged evidence, the trial judge must identify the fact(s) in issue, an exercise that draws in the broader case context. However, it then appears appropriate for the trial judge to focus narrowly on the challenged evidence, to avoid straying into the jury’s more holistic proof territory. As discussed in Part II, this orthodox approach to relevance was followed by the High Court at common law in the sexual assault cases Phillips v The Queen (‘Phillips’)[4] and Stubley v Western Australia (‘Stubley’),[5] although in these cases the Court failed to recognise a clear connection between the evidence of other alleged victims and the complainants’ claimed absence of consent. The High Court has also taken this focused approach to probative value in Uniform Evidence Law (‘UEL’)[6] cases such as Kadir v The Queen (‘Kadir’),[7] which concerned improperly obtained surveillance evidence. The High Court also applied a sharp focus to the related admissibility element of reliability in Sio v The Queen (‘Sio’),[8] which concerned first-hand hearsay evidence. In relation to certain types of evidence, such as similar fact evidence of further poisonings linked to the defendant in Perry v The Queen (‘Perry’),[9] a degree of holism may enter. Probative value was assessed by reference to the improbability of the various poisonings occurring coincidentally. But the assessments remained atomistic, with evidence of some poisonings admissible and evidence of other poisonings excluded.
Problematically, however, the High Court has not consistently followed this more or less atomistic approach to probative value. Part III examines a series of cases concerning challenges to the admissibility of propensity and tendency evidence. In assessing probative value, the challenged evidence is weighed together with the other evidence. The High Court developed this holistic approach at common law in cases like Pfennig v The Queen (‘Pfennig’)[10] and Phillips. It influenced the 2020 amendments to the UEL following the Royal Commission into Child Sexual Assault,[11] and has continued with the recent UEL case, TL v The King (‘TL’).[12] This holistic conception of probative value is incoherent. It is at odds with the atomistic approach taken by the High Court in other cases and contradicts another admissibility element – ‘importance’ – a further factor in Kadir. As examined in Part IV, the High Court’s occasional unorthodox holism also confounds its other efforts to constrain the trial judge’s admissibility determinations and maintain a clear demarcation with the jury’s proof function. The High Court’s probative value holism may stem from a conflation of proof and probative value, a fallacy that manifests in other areas of the High Court’s proof jurisprudence.
Part V concludes. The issue addressed in this article – where admissibility determinations of probative value lie on the atomism/holism spectrum – is narrow, technical, and formal. However, it is crucial to the design and operation of the criminal trial, the role of the jury, the pursuit of factual accuracy, and the enforcement of the criminal law. The persistent incoherence in the High Court’s admissibility jurisprudence is deeply troubling.
In this Part, I examine the concepts of relevance and probative value together with related concepts: the fact in issue, admissibility, and proof. Generally, in definition and application, relevance and probative value are atomistic concepts. The next Part considers a troubled line of High Court authority which construes probative value holistically.
In the criminal trial, the jury is responsible for determining whether the defendant’s guilt has been proven ‘beyond reasonable doubt’.[13] According to the High Court, ‘the purpose and the genius of the jury system is that it allows for the ordinary experiences of ordinary people to be brought to bear in the determination of factual matters’.[14] The jury should not engage in ‘objective analysis’:[15]
[T]he jury should not be told to look at the evidence of each witness ‘separately in, so to speak, a hermetically sealed compartment’; they should consider the accumulation of the evidence ... the jury must consider ‘the weight which is to be given to the united force of all the circumstances put together’.[16]
The High Court’s view of jury reasoning as an intuitive holistic process is consistent with the dominant descriptive model in psychology: the story model. Jurors approach their proof task by ‘generating a plausible, coherent story out of the evidence ... According to this model, the jurors do not simply weigh the items of evidence individually and piecemeal; rather, they endeavor to make sense of the evidence as a whole’.[17]
The trial judge’s job is to ensure that the trial proceeds in accordance with the law. A key function is to ensure that jurors are not exposed to inadmissible evidence. As discussed in Part II(C), under many exclusionary principles, the admissibility of challenged evidence largely turns on its probative value. Probative value, like proof, is concerned with the strength of evidence, but the two are clearly distinguishable. The proof issue for the jury is whether all of the evidence, together, proves guilt beyond reasonable doubt. The probative value of challenged evidence measures the capacity of ‘the evidence [to] rationally affect the assessment of the probability ... of a fact in issue’.[18] The trial judge should focus on the challenged evidence and determine its contribution. This focus has a practical benefit. As a majority of the High Court appreciated in IMM v The Queen (‘IMM’),[19] the trial judge must determine admissibility before the ‘the trial is complete’[20] and the ‘full picture has emerged’;[21] admissibility tests must ‘acknowledge these limitations’.[22] As discussed in Part IV(B), at the admissibility stage the challenged evidence is assumed credible and reliable, limiting the trial judge’s focus considerably. Whereas the proof concept is holistic and intuitive, probative value for the trial judge should be approached with particularity and precision.
Mnookin observes that ‘[a]bsolute atomism, therefore, cannot exist ... no item of evidence can be wholly judged as “an island, Entire of itself”’.[23] While probative value, as a matter of principle, focuses on the contribution of the challenged evidence, the assessment is not wholly acontextual. Probative value is a relational concept, like the closely related concept of relevance.[24] ‘Relevance does not inhere to a piece of evidence but rather exists as a relation between the item and something of consequence in the case.’[25] Evidence is relevant if it ‘could rationally affect ... the assessment of the probability of the existence of a fact in issue in the proceeding’.[26] Any degree of impact suffices for relevance.[27] Probative value measures the ‘extent’ to which evidence affects the probability of a fact in issue.[28]
A prior step in determining relevance and probative value is identifying the live issues. This is highly contextual, depending ‘on a consideration of what happened in the trial as a whole’.[29] However, tracing a connection from the challenged evidence to the fact in issue is then a focused exercise, requiring close attention to the content of the challenged evidence and the nature of the fact in issue.
Sexual assault cases provide a useful illustration of how the issues in a trial can narrow. In ‘acquaintance rape’ cases, where the complainant testifies that the defendant committed the offence, identity will generally not be in issue.[30] Instead, the defendant will usually deny having committed the act, or claim consent or belief in consent.[31] In acquaintance rape cases with a child complainant, consent is immaterial and commission is often the only fact in issue.[32] Delay in reporting is not uncommon in acquaintance rape cases, leading to a loss of forensic and other evidence, and complainant credibility can be key.[33] In ‘stranger rape’ cases, there is often less delay in reporting, commission may be clear cut, and identity is typically the main issue.[34]
The nature of the fact in issue can determine the relevance and admissibility of challenged evidence.[35] Traditionally, evidence of a defendant’s other misconduct is viewed as carrying the risk of ‘strong prejudice [and should be excluded] unless it is plainly necessary to prove something which is really in issue’.[36] In a child sexual offence case with commission in issue, and where the defence challenges the complainant’s credibility on the basis of delayed report, evidence of the defendant’s other misconduct with the complainant ‘may provide a context helpful, or even necessary, for an understanding of a narrative’.[37] The ‘series of incidents [may] make it believable or understandable that the victim might not have complained about the incidents charged until much later in the piece’.[38]
Depending upon what is in issue, evidence of a defendant’s other misconduct may have a crucial role to play in sexual offence cases via propensity reasoning. Consider a sexual assault case where the complainant testifies that the perpetrator had forced her to engage in unusual sexual acts, and the prosecution adduces evidence that other women had voluntarily engaged in these unusual sexual acts with the defendant. This would be relevant (and may well be admissible) if identity were in issue.[39] The defendant has engaged in these unusual sexual acts on other occasions and has a tendency to engage in these unusual sexual acts, which increases the probability that it was the defendant that engaged in these unusual sexual acts on the charged occasion. But if the defence concedes the commission of the charged unusual sexual acts,[40] leaving consent the only live issue, evidence of the defendant engaging in these unusual sexual acts with consenting partners on other occasions would no longer assist the prosecution. If anything, the evidence could assist the defence.[41]
In the 2006 acquaintance rape case Phillips, the High Court held that, in relation to the issue of a complainant’s consent, evidence that the defendant’s other alleged victims did not consent is irrelevant.[42] According to the High Court, the non-consent of other alleged victims ‘does not itself prove any disposition on the part of the accused: it proves only what mental state each of the other [alleged victims] had on a particular occasion affecting them, and that can say nothing about the mental state of the ... complainant on a particular occasion affecting her’.[43] This decision received swift and just criticism.[44] Evidence of other alleged victims’ non-consent has a clear connection with the complainant’s non-consent. The connecting element is ‘an inclination in the accused to have intercourse with girls despite their protests and, if necessary, by threats of violence’.[45] It was artificial for the High Court to view each victims’ non-consent in a vacuum.
Courts subsequently struggled with the artificiality of the High Court’s reasoning. Occasionally, courts sought to avoid the Phillips principle[46] by recognising that the defendant’s conduct is in issue as well as the complainant’s consent.[47] In 2007 in Western Australia v Osborne (‘Osborne’),[48] the Western Australian Court of Appeal noted that ‘evidence which ... concerns whether an accused person behaved in a particular way may also bear upon the question of consent’.[49] Phillips may have precluded the other victims’ evidence being admitted on consent, but the evidence was still admissible to ‘show a tendency on the part of the accused to take sexual advantage of drunken [persons] while asleep’.[50] In 2008 in R v Wallace,[51] the South Australian Court of Criminal Appeal held that ‘the relevance of the evidence went beyond the issue of consent’.[52] It went to prove that the defendant ‘behaved in a threatening and aggressive manner towards the women’;[53] ‘what could be viewed as a formatted pattern of sexual behaviour on the part of the appellant, indifferent to the wishes or consent of the women in question’.[54]
The High Court had the opportunity to consider the Osborne strategy in 2011 in Stubley.[55] The defendant, a consulting psychiatrist, was charged with a series of sexual offences against two of his female patients. The defence conceded that sexual contact had occurred but claimed that it was consensual.[56] The prosecution sought to adduce evidence of three other female patients who alleged the defendant had had sexual contact with them without their consent. The trial judge, following Osborne, admitted the evidence to show how the defendant exploited his relationship with the women to have non-consensual sexual contact.[57] On appeal to the High Court, a majority held that the propensity evidence was irrelevant.[58] However, the majority did not endorse the broad Phillips principle and did not outrightly reject the Osborne strategy. Instead, the majority pointed out that the defendant’s alleged conduct with the other victims was different from the defendant’s alleged conduct with the complainants. According to the majority, evidence of the defendant’s manipulation of the other alleged victims ‘could not rationally affect the likelihood’ that the complainants’ consent was vitiated by force, threats, or intimidation.[59]
In Stubley, the High Court did not rely on the bald Phillips principle that no connection can exist between the non-consent of different victims. However, the majority’s reasoning is still problematic. Modern evidence law adopts the notion of ‘logical relevance’, which ‘requires a minimal logical connection between the evidence and the “fact in issue”’.[60] Differences in the defendant’s methods may have weakened the connection between the non-consent of the other victims and the non-consent of the complainants, but a connection still remained.[61] Both the complainants and the other victims portrayed the defendant as a sexual predator prepared to exploit (whether by manipulation or threats) the vulnerability of his female patients. It may be that, rather than ‘logical relevance’, the majority was applying a notion of ‘legal relevance’, requiring, for unstated policy reasons, ‘something more than a minimum’.[62]
Subsequent cases appear to follow Stubley in this respect, demanding more than a minimal connection. In Jacobs v The Queen (‘Jacobs’)[63] there was evidence that the defendant met the alleged victims through an online dating site and used manipulation and alcohol to overcome his sexual prey.[64] The Victorian Court of Appeal held these commonplace circumstances were ‘unremarkable’,[65] ‘the asserted similarities were either irrelevant or only marginally relevant’, and evidence of other women’s non-consent was inadmissible.[66] By contrast, in R v Little (‘Little’),[67] the Queensland Court of Appeal held evidence of the other sexual assaults admissible even though consent was ‘the only real issue’.[68] The defendant’s methods involved breaking into the victims’ homes wearing a balaclava and threatening to kill them with a knife.[69] The more extreme methods employed in Little may have greater probative value than those in Jacobs, both because they are more unusual and distinctive,[70] and because they provide a stronger explanation for non-consensual sexual contact. But the non-consent of the defendant’s other victims in Jacobs, as in Phillips and Stubley, still has at least some bearing on the complainant’s non-consent. Properly understood, logical relevance merely requires ‘some connection, even if tenuous, between the item and something legitimately at issue in the case’.[71]
Relevance is the threshold admissibility requirement. Relevant evidence may still fall foul of some other exclusionary principle. Various types of evidence – such as hearsay evidence,[72] credibility evidence,[73] opinion evidence,[74] and tendency evidence[75] – should be excluded unless they satisfy certain admissibility requirements. The trial judge may also exclude evidence not caught by these exclusionary rules. The trial judge is given both general exclusionary powers[76] and more specific ones relating to admissions[77] and unlawfully obtained evidence.[78]
These exclusionary principles serve a range of overlapping goals. A key consideration is efficiency. ‘[L]itigation is a practical enterprise that must seek finality within reasonable time, money, and other resource constraints.’[79] Section 135 of the UEL allows the trial judge to exclude evidence on the basis that admitting it would be an ‘undue waste of time’.[80] Efficiency is also one of the concerns underlying the exclusion of credibility evidence and propensity evidence of the other activities or characteristics of a witness or a defendant. Both types of evidence may generate collateral issues beyond the central facts in issue, heightening ‘the need to confine the trial process’.[81]
A variety of other goals also motivate exclusionary principles. Relevant evidence may be excluded out of concern over how it was obtained – for example, section 84 of the UEL excludes ‘admissions influenced by violence and certain other conduct’, while ‘improperly or illegally obtained evidence’ more broadly may be excluded under section 138. The prohibition on propensity reasoning from a defendant’s other misconduct may, in part at least, reflect the ‘moral imperative of respect for personal autonomy’,[82] or the ‘ancient and deeply embedded moral proscriptions against the act of speaking ill of others’.[83]
Among the various goals served by the exclusionary principles, factual accuracy is ‘paramount’.[84] Hearsay evidence is subject to exclusion because of reliability concerns; the previous representations are generally not given under oath and cannot be tested by cross-examination.[85] A major consideration in the exclusion of evidence is ‘unfair prejudice’: the risk that ‘the jury are likely to give the evidence more weight than it deserves or ... the nature or content of the evidence may inflame the jury or divert the jurors from their task’.[86] Concern about unfair prejudice motivates the exclusion of propensity and tendency evidence,[87] and the general exclusionary powers at common law[88] and sections 135 and 137 of the UEL.
Of course, the exclusion of relevant evidence may also mislead the jury and threaten factual accuracy. The greater the probative value, the more critical it is that evidence be admitted in ‘the public interest in bringing to conviction those who commit criminal offences’.[89] The propensity exclusionary rule at common law was qualified by the recognition that the evidence can be ‘so very relevant that to exclude it would be an affront to common sense’.[90] Evidence excluded regardless of its probative value is rare, and invites the question of whether system goals can be pursued ‘by means less corrosive of the judicial system’s ability to ascertain the truth’.[91] Many exclusionary principles admit evidence if it has a required level of probative value,[92] or where probative value outweighs competing considerations in a kind of cost-benefit analysis.[93] A number of other exclusionary principles draw in other measures related to factual accuracy, such as importance[94] and reliability.[95]
Where admissibility turns upon the outcome of a balancing test, concerns may be raised about the commensurability of the various considerations.[96] Probative value and the risk of unfair prejudice arguably possess a unitary metric;[97] the ‘obvious common value ... is the accuracy of fact-finding’.[98] But many admissibility balancing tests involve incommensurables.[99] For example, improperly or illegally obtained evidence will be excluded under section 138 of the UEL unless ‘the desirability of admitting the evidence outweighs the undesirability of admitting evidence’.[100] Probative value is placed in the balance among considerations, such as the gravity of the impropriety or illegality.[101] The desire for factual accuracy[102] is weighed against concerns about ‘judicial legitimacy, protection of individual rights or the deterrence of police misconduct’.[103] There are clear challenges balancing these diverse social goods against each other.[104] However, as Cass R Sunstein observes, ‘[t]his problem does not entail paralysis, indeterminacy, or arbitrariness. Decisions are made all the time among incommensurable goods, at the personal, social, and legal levels.’[105]
In determining the admissibility of challenged evidence, the trial judge often considers the capacity of the evidence to contribute to an accurate verdict. In answering this question, it appears appropriate to focus closely upon the challenged evidence.
The High Court decision on the admissibility of hearsay evidence in Sio illustrates the focused approach. The defendant was charged with armed robbery and constructive murder under the doctrine of joint criminal enterprise. Mr Filihia admitted to the police that he had committed the robbery and stabbing, but implicated the defendant as the driver who had put him up to the robbery and gave him the knife. Admissibility, under the relevant hearsay exception,[106] required that the representation was ‘against the interests of [Mr Filihia] at the time it was made’, and ‘made in circumstances that make it likely that the representation is reliable’.[107] These requirements provide assurance that ‘the dangers which the [hearsay rule] seeks to prevent are not present or are negligible’.[108]
Upholding the defence appeal, the High Court held that Mr Filihia’s hearsay representations were inadmissible.[109] The Court of Criminal Appeal had given insufficient weight ‘to the circumstance that Mr Filihia’s representations were those of an accomplice in the commission of the crimes in question’.[110] And the High Court criticised the Court of Criminal Appeal’s ‘compendious approach’[111] to admissibility ‘whereby an overall impression was formed of the general reliability of the statements made by Mr Filihia and then all his statements were held to be admissible against Mr Sio’. [112] Reference may be had to ‘other representations which form part of the context’, such as ‘a specific retraction of the assertion of the relevant fact’.[113] However, section 65 ‘direct[s] attention to the particular representation which asserts the relevant fact’.[114]
While ... generally speaking, the totality of Mr Filihia’s statements were against his own interest, his statement that [the defendant] gave him the knife and put him up to the robbery was, given the circumstances in which that statement was made, plainly apt to minimise his culpability by maximising that of [the defendant].[115]
Admissibility tests more commonly turn on ‘probative value’[116] than on ‘reliability’.[117] Often, probative value is placed in the balance with a range of other considerations. These considerations generally focus on the challenged evidence, but occasionally extend more broadly. For example, in relation to improperly and illegally obtained evidence under section 138(3) of the UEL, two considerations extend beyond the challenged evidence: sub-section (c) ‘the nature of the relevant offence’ – the more serious the offence, the more desirable that the jury have access to the evidence;[118] and sub-section (g) ‘whether any other proceeding ... has been or is likely to be taken in relation to the impropriety or contravention’ – in which case the ‘sanction’ of exclusion may not be required.[119] The other factors focus closely on the evidence, including sub-section (a) ‘the probative value of the evidence’, and sub-section (d) ‘the gravity of the impropriety or contravention’ by which the evidence was obtained.[120] In weighing these factors in Kadir, the High Court focused closely on the challenged evidence.
In Kadir, the defence challenged the admissibility of several sets of evidence – illegally obtained surveillance footage, and also search warrant evidence and subsequently obtained admissions. The latter two sets of evidence were not themselves illegally obtained but were still covered by section 138 since they had been obtained ‘in consequence of’ the earlier illegality.[121] The High Court rejected the view that the three sets of evidence should be subject to a single compendious admissibility ruling:
Self-evidently, ... the probative value of the evidence ... cannot be picked up from findings made with respect to the surveillance evidence and applied to the search warrant evidence or the admissions. ... Moreover, the weighting of the factors that are concerned with the impropriety or illegality to the balancing of the public interests may differ as between the surveillance evidence, the search warrant evidence and the admissions.[122]
Ultimately the illegally obtained surveillance evidence was excluded,[123] but the search warrant evidence, ‘obtained by a regulator acting lawfully and without prior knowledge of the contravention’,[124] and the admissions, which had only a ‘bare connection [with] the contravention’,[125] were held to be admissible.
In Mnookin’s terms, the High Court’s reasoning in Kadir was highly atomistic; ‘breaking [the challenged section 138 evidence] into smaller units ... rather than evaluating it as a single whole’.[126] In some situations, a less atomistic approach may be more appropriate. It depends upon the nature of the evidence and how it acquires probative value. Mnookin criticises the US Supreme Court majority’s ‘strikingly atomized’[127] approach to the assessment of the reliability and admissibility of expert evidence in General Electric Co v Joiner (‘General Electric’).[128] At issue was whether chemicals manufactured by the defendants caused the plaintiff’s cancer. The majority judgment ‘briefly [considered] each of a number of studies that the plaintiffs’ experts relied on ... picking them off one by one, individually, and leaving [Chief Justice Rehnquist] by the end of his analysis with no admissible studies that could warrant the experts’ claims’.[129] Mnookin argues that ‘holism is the more intellectually legitimate perspective for the assessment of expert evidence, both as a matter of evidence law and as a matter of scientific epistemology’.[130] As the dissenting judgment recognised, the plaintiff’s experts, themselves, ‘did not suggest that any one study provided adequate support for their conclusions, but instead relied on all the studies taken together’.[131]
Gary Edmond cautions against undue holism where the validity of expert evidence is challenged: ‘[T]he consideration of other evidence ... is likely to distract or mislead, especially where a procedure or ability has not passed formal evaluation. Other evidence ... reveal[s] nothing conclusive about whether or how well a procedure works.’[132] However, these comments are directed to other evidence beyond the body of expert evidence.[133] The larger point here is that challenged evidence may be subject to differing degrees of granularity in admissibility determinations. Importantly, the degree of atomism/holism should itself be determined with care and precision.
The nuance involved in determining an appropriate degree of atomism/holism is well illustrated by the 1982 High Court decision in Perry. The defendant was charged with the attempted murder in 1978 of her third husband by arsenic poisoning. The prosecution adduced evidence that her second husband, brother, and a de facto husband had died of poisoning in 1961, 1962, and 1970 respectively. The prosecution wished to rely upon similar fact or coincidence reasoning – ‘the improbability that a number of deaths would occur in similar circumstances merely by coincidence’.[134] According to the dominant view, the similar fact evidence could gain admission if it possessed ‘strong probative force’.[135] The probative value assessments of evidence of the three other poisonings combined elements of atomism and holism.
To a degree, the prosecution’s coincidence reasoning is inherently holistic in that, as recognised in the High Court’s judgments, it relies on the ‘cumulative effect of the evidence relating to the four’.[136] ‘[I]t would be a mistake to consider the evidence relating to [the various alleged poisonings] as separate pieces of evidence which [must] be admitted or rejected in isolation from each other.’[137] ‘[T]he frequency of the occurrence of the similar facts enhances the probative force of the evidence, though the necessary probative force would be lacking if the similar fact had occurred but once or on a few occasions only.’[138] However, a degree of atomism was retained. Only Murphy J held that all of the similar fact evidence was inadmissible.[139] Gibbs CJ would have excluded the evidence relating to the de facto husband and the brother while admitting the evidence relating to the second husband,[140] and Wilson J and Brennan J would only have excluded the evidence relating to the de facto husband.[141] While viewing the evidence together, the strength of the evidence relating to each death was considered separately according to the type of poison involved, the defendant’s opportunity to administer the poison, and the strength of her motive.
The High Court’s close focus on the reliability and probative value of the challenged evidence in Sio, Perry and Kadir appears principled and orthodox. However, a line of High Court authority regarding propensity evidence favours an unfocused holistic assessment, conflating probative value and proof. I consider the common law origins of this disordered jurisprudence – the Pfennig line of authority – in section A. Section B examines how the UEL, rather than resolving the difficulties, retained and amplified them. Section C observes that the High Court’s holism from Pfennig contradicts the contextual notion of importance, which has been expressly adopted as an admissibility factor in the UEL. Section D discusses the latest manifestations of holistic probative value in the Royal Commission reforms to the UEL tendency and coincidence rules, and the High Court decision in TL.
In many criminal cases, the prosecution tenders evidence of a defendant’s other misconduct and argues that the defendant has a propensity to commit that kind of misconduct, increasing the probability that the defendant committed the charged offence. Propensity reasoning is related to the similar fact reasoning relied upon in Perry.[144]
Propensity and similar fact evidence have been traditionally viewed as unfairly prejudicial and subject to exclusion.[145] Lord Hailsham described this area of law as a ‘pitted battlefield’ half a century ago,[146] however, it became widely accepted that, despite the exclusionary rule, sufficiently probative propensity evidence could gain admission.[147] In the latter part of the 20th century, two different admissibility tests emerged. One line of authority supported a simple balancing test. In 1989 in Harriman v The Queen,[148] Dawson J held that, to be admissible, propensity evidence must be ‘of sufficient strength to outweigh the prejudice which it inevitably carries with it’.[149] This balancing test gained dominance in some jurisdictions,[150] but Australian law grew more complex. In Sutton v The Queen,[151] Gibbs CJ indicated that the discretionary balancing test was one element of a ‘double safeguard against the injustice that may be caused by evidence of this kind’.[152] As well as satisfying the discretionary balancing test, a ‘rule of law’ required that the evidence be ‘strongly probative’ to gain admission.[153]
In Pfennig, the majority, objecting to the discretionary quality of the balancing test,[154] endorsed a particular version of the ‘strongly probative’ requirement. The evidence will only be admissible if ‘there is no reasonable view of the evidence consistent with the innocence of the accused’.[155] This admissibility test was derived from a jury direction, sometimes given in circumstantial cases, that the jury should acquit ‘if there is any reasonable hypothesis consistent with ... innocence’.[156] This direction, in turn, is an ‘amplification’[157] of the criminal standard of proof requiring proof of guilt beyond reasonable doubt. The Pfennig majority, with heightened concern about the risk of prejudice, employed the criminal standard of proof as an admissibility test: ‘Only if there is no such view can one safely conclude that the probative force of the evidence outweighs its prejudicial effect.’[158] With the ‘no rational view’ admissibility test, the risk of prejudice leading to a wrongful conviction appears to have been all but eliminated.[159]
The Pfennig majority’s ‘no reasonable view’ test is problematic. First, it seems to assume that ‘unfair prejudice’ is at an extremely high level.[160] This assessment may be plausible in a case like Pfennig, where the prosecution sought to show the defendant’s propensity to abduct, sexually abuse, and murder young boys, disposing of the bodies so that they are never found.[161] Jurors may respond to this evidence with strong revulsion. But Pfennig is an extreme case. At the other end of the spectrum is Martin v Osborne,[162] where evidence showed the defendant’s history of carrying passengers for reward without a licence. Jurors may respond to this with a shrug of the shoulders.
A further concern is that the ‘no reasonable view’ test is overly stringent. Applied atomistically, this test would demand that the propensity evidence by itself prove guilt beyond reasonable doubt, an ‘extraordinarily high threshold for admissibility’.[163] In Phillips in 2006, the High Court responded to this concern by indicating ‘Pfennig ... does not require the judge to conclude that the similar fact evidence, standing alone, would demonstrate the guilt of the accused’.[164] The Court emphasised ‘the necessity to view the similar fact evidence in the context of the prosecution case’.[165] It makes sense to construe the test holistically given that it is derived from the criminal standard of proof which the jury should apply to the whole of the evidence.[166] But to adopt such holism for an admissibility test raises further questions. As explored in Part IV below, this blurs the line between the trial judge’s admissibility assessment and the jury’s proof determination. More immediately, the holism fails to address the prejudicial risk of a specific item of challenged evidence. Operating in the context of an otherwise strong prosecution case, the test may admit highly prejudicial evidence that adds little to the prosecution case.[167] The holistic version of the test will often lack bite.
The probative value holism of Pfennig and Phillips is quite extreme.[168] It extends beyond the limited holism discussed in Part II(C) above, which may allow a body of challenged evidence – the similar fact evidence in Perry, or the expert evidence in General Electric – to be considered together. The ‘no reasonable view’ test draws in other available prosecution evidence.[169] These two holisms may operate independently.[170] Challenged evidence atomism could combine with other evidence holism. Where several items of propensity are challenged, the trial judge could pose the admissibility question whether the other incidents, taken individually, but each in the context of the prosecution case, exclude the defendant’s innocence as a reasonable possibility.[171] However, this combination would be incongruous.
For the ‘no reasonable view’ test to avoid the charge of excessive holism it would somehow need to retain focus on the challenged evidence. The High Court jurisprudence is unclear, but two focused interpretations may be open. Neither is entirely plausible, and they are inconsistent with each other. The first possibility is that Gibbs CJ’s ‘double safeguard’ has continuing operation. Pfennig replaced the balancing test element with the holistic requirement that there is ‘no reasonable view’ of the evidence consistent with innocence. The other element of the safeguard focuses more closely on the challenged evidence itself. While defending a highly contextualised Pfennig test in Phillips, the Court also said ‘[i]t is necessary to find “a sufficient nexus” between the ... particular charge and the similar fact evidence ... Admissible similar fact evidence must have “some specific connection with or relation to the issues for decision in the subject case”’.[172] ‘[S]triking similarity, underlying unity and other like descriptions of similar facts are not essential ... though usually the evidence will lack the requisite probative force if the evidence does not possess such characteristics.’[173] The difficulty with this interpretation is that the High Court in Pfennig and Phillips did not clearly distinguish between the two elements of the double safeguard. Instead, the Court suggested that the ‘no reasonable view’ test subsumes the ‘strongly probative’ requirement – the challenged evidence must ‘[possess] a particular probative value or cogency such that, if accepted, it bears no reasonable explanation other than the inculpation of the accused in the offence charged’.[174]
A second elaboration of the ‘no reasonable view’ test does a better job in maintaining a discrete role for the ‘strongly probative’ requirement. In Pfennig, the majority indicated that, in applying the test, ‘the trial judge ... must regard the evidence as a step in the proof of [the prosecution] case’.[175] Hodgson JA provided a two-step account of this approach[176] that has received High Court endorsement.[177] First, it is ‘assumed that all the other evidence in the case left the jury with a reasonable doubt about the guilt of the accused’.[178] Second, it is asked whether ‘the propensity evidence ... when it is added to the other evidence, ... would eliminate any reasonable doubt which might be left by the other evidence’.[179] The problem with this approach is its lack of precision; it fails to make any definite demand of the challenged evidence in isolation. In R v Joiner, Hodgson JA considered that the other evidence only left innocence as a ‘very remote [possibility]’.[180] Very little was then demanded of the propensity evidence ‘to regard such a highly improbable scenario as ... not a reasonable possibility’.[181]
These two versions of the ‘no reasonable view’ test are inconsistent with each other. If the other evidence is considered sufficiently strong, propensity evidence that is merely relevant would satisfy Hodgson JA’s test. But the High Court in Phillips, in emphasising the ‘sufficient nexus’ requirement, expressly stated that ‘it is not enough that the evidence merely has some probative value of the requisite kind’.[182] Indeed, in Phillips, a serial sexual assault case, the context for the other allegation evidence was strong – the direct evidence of the complainant. And the High Court still held that the other allegations lacked sufficient nexus – the ‘similarities relied on were not merely not “striking”, they were entirely unremarkable’.[183] The common law exclusionary rule in Australia is even more of a ‘pitted battlefield’ than it was half a century ago. As discussed in the next section, the UEL has done nothing to resolve these difficulties.
In 1979, the Commonwealth Attorney-General identified ‘the need for modernization of the law of evidence’.[184] The Law Reform Commission was instructed to conduct a ‘comprehensive review’,[185] which produced valuable research[186] and recommendations which received the broad approval of the New South Wales Law Reform Commission.[187] A series of Evidence Bills took form over the early 1990s, culminating in the passage of almost identical Evidence Acts by the Commonwealth and New South Wales (‘NSW’) Parliaments in 1995. The UEL now covers all Australian jurisdictions other than South Australia (‘SA’), Queensland, and Western Australia (‘WA’).[188]
In some areas, such as hearsay, the UEL implemented ambitious structural reform.[189] However, in relation to propensity and similar fact evidence – ‘tendency’ and ‘coincidence’ evidence in the UEL – the reforms were tentative. While the Law Reform Commission and the Parliaments were working towards the UEL, the common law rule was undergoing the messy evolution discussed in the previous section.[190] The UEL adopted a variation of the unsettled and problematic common law and it remained, in the words of the Victorian Court of Appeal, ‘exceedingly complex and extraordinarily difficult to apply’.[191] A review of the UEL by the Australian, NSW and Victorian Law Reform Commissions in 2005 offered only tweaks around the edges.[192] The tendency and coincidence provisions were subject to major reforms following the Royal Commission into Child Sexual Abuse, however, as discussed in the next section, this has only added to the underlying problems.
The UEL adopted a version of Gibbs CJ’s ‘double safeguard’. For tendency evidence or coincidence evidence to gain admission – under sections 97 and 98 respectively[193] – it must possess ‘significant probative value’ (UEL sections 97(1)(b) and 98(1)(b)), and the probative value of prosecution evidence against the defendant must also ‘substantially outweigh’ the danger of unfair prejudice (UEL section 101(2)). The Royal Commission reforms eliminate the asymmetry in the balancing test by removing the word ‘substantially’.[194]
The drafting of the UEL expresses the uncertainty then prevailing as to whether the assessment of probative value should be focused or contextual. In determining whether the evidence has ‘significant probative value’ in sections 97(1)(b) and 98(1)(b), the trial judge is to consider the evidence ‘either by itself or having regard to other evidence’.[195] On a literal interpretation this drafting may suggest that either requirement will suffice for admissibility. The problem with this interpretation is that the contextual requirement is weaker and easier to satisfy, leaving the focused requirement no role to play. If the challenged evidence has significant probative value when considered by itself, clearly it will have significant probative value together with the other evidence.[196] The focused requirement is redundant. In 2017 in Hughes v The Queen (‘Hughes’), a majority of the High Court emphasised the contextual aspect: ‘It is sufficient if the disputed evidence together with other evidence makes significantly more likely any facts making up the elements of the offence charged.’[197] A few years later, the High Court in TL interpreted the words to require ‘an assessment of the evidence both by itself and “having regard to other evidence ...”’.[198] But this appears contrary to the legislative language; the legislative ‘either/or’ structure implies a ‘choice between alternatives ... only one of two’ is required.[199] In the UEL, as at common law, if the probative value of challenged evidence is to be assessed with ‘other evidence’, it is uncertain what is demanded of the challenged evidence ‘by itself’.
Sections 97 and 98 are the only ones in the UEL to expressly refer to both focused and holistic conceptions of probative value; this choice does not even appear in the second probative value admissibility test in section 101.[200] Elsewhere in the UEL there is no mention of the challenged evidence being assessed ‘by itself’ or ‘having regard to other evidence’. However, in 2018 in R v Falzon (‘Falzon’),[201] the High Court casually extended its holism beyond tendency and coincidence evidence. The defendant was charged with cannabis trafficking. The defendant appealed to the High Court, arguing that the evidence of a large amount of cash that had been secreted in his home was inadmissible under section 137 of the Evidence Act 2008 (Vic) on the basis that ‘its probative value is outweighed by the danger of unfair prejudice to the [defendant]’. The cash did not come from the sale of the cannabis that gave rise to the charges; the implication was that it came from other drug sales.[202] The High Court assessed the evidence holistically and found it highly probative and admissible:
Combined with the other circumstantial evidence of the respondent’s carrying on of a business of drug trafficking, including ... the physical paraphernalia of drug trafficking and the large quantities of cannabis ... evidence of the cash ... constituted a powerful circumstantial case.[203]
The role of context in the assessment of ‘probative value’ and admissibility in the UEL is unclear, even in sections 97 and 98 which make express mention of ‘other evidence’. A number of other admissibility[204] provisions require consideration of the evidential context from a different perspective, directing the trial judge to consider the ‘importance’ of the challenged evidence.[205]
The High Court recently commented on the importance factor and its relation to probative value in Kadir. As discussed above, this case concerned the probative value and admissibility under section 138 of the UEL of illegally obtained surveillance footage, and also search warrant evidence and admissions which were obtained in consequence of the illegal surveillance. In relation to the importance factor, the Court said:
Evidence may possess high probative value but not be important in the proceeding in a case in which other equally probative evidence is available to the prosecution. In this case, the importance of the search warrant evidence, and ... the admissions, is greater by reason of the exclusion of the surveillance evidence.[206]
The UEL distinguishes between the ‘probative value’ and ‘importance’ of evidence as distinct admissibility factors. Where ‘probative value’ is listed as an admissibility consideration but not ‘importance’, this implies the latter contextual principle should not be brought to account.[207] In the 2012 High Court case of Aytugrul v The Queen (‘Aytugrul’),[208] the defence, in effect, argued to the contrary – that the contextual importance of evidence is a component of its probative value assessment under section 137 of the UEL. Aytugrul was a murder case with identity in issue. The appeal concerned two different expressions of the probative value of a DNA profile match: (1) the frequency ratio – only 1 in 1,600 people would be expected to have the profile; and (2) the exclusion percentage – 99.9% of people would not be expected to have the profile.[209] Section 137 provides for the exclusion of evidence when its probative value is outweighed by danger of unfair prejudice. The defence argued that the exclusion percentage was highly prejudicial and should be excluded. The jury might round up from 99.9% to 100% and conclude no-one else would have the profile.[210] Further, the defence argued, given evidence of the frequency ratio, the exclusion percentage evidence had limited, if any, incremental value.[211] In other words, the exclusion percentage lacked importance because other (less prejudicial) evidence – the frequency ratio – covered the same ground.
The Court rejected the appeal, on the basis that the exclusion percentage was not prejudicial, without considering the importance argument.[212] However, Heydon J noted that a version of the defence’s argument had received support from the US Supreme Court in Old Chief v US[213] in relation to the broadly similar Federal Rules of Evidence (1975) rule 403.[214] This case may have limited persuasive power in relation to section 137 of the UEL because rule 403, unlike section 137, expressly incorporates an importance factor. Probative value is weighed, not only against ‘the danger of unfair prejudice’, but also the ‘needless presentation of cumulative evidence’ – a concept closely related to that of importance. In this respect, rule 403 more closely resembles UEL section 135, which weighs the probative value of the evidence against the admission of the evidence being an ‘undue waste of time’, among other things.[215]
As a matter of statutory interpretation, it may be hard to make a case for ‘incremental probative value’. Nevertheless, Stephen Odgers suggests section 137 may be employed where ‘the prosecution tenders gruesome photographs of the deceased in a murder trial, where a pathologist has already described the injuries and there is little forensic assistance to be derived from the photographs’.[216] And the policy argument is persuasive. As Mnookin observes, ‘[i]f, given the availability of alternative evidence, the marginal or incremental probative value of the new item of evidence approaches zero, why should the jury be allowed to hear it, if it risks unfairly prejudicing them?’[217] Clearly, a trial judge will need to exercise care in making this determination, as Dyson Heydon notes extrajudicially, ‘the issue is still open, and there is no predicting what view the jury might take of any particular piece or pieces of evidence’.[218]
While incremental probative value – or importance – requires a consideration of context, it provides no support for the High Court’s holism discussed in the preceding sections. Indeed, importance pulls in the opposite direction.[219] A holistic probative value assessment of challenged evidence, and the likelihood of its admissibility, will increase with support from other evidence.[220] However, the existence of such other evidence diminishes the importance of the evidence and the likelihood of admissibility.
As discussed in the preceding sections, there is incoherence and uncertainty in both the common law and the UEL relating to admissibility probative value assessments and the extent to which the challenged evidence is assessed by itself or together with other evidence. This section considers the latest developments regarding the admissibility and probative value assessment of tendency and coincidence evidence under the UEL, namely, the reforms following the Child Sexual Abuse Royal Commission, and the recent High Court decision in TL. Unfortunately, these developments do little to clarify the position.
The Royal Commission considered that the criminal justice system was not responding adequately to high rates of child sexual assault.[221] In part this was due to the exclusionary rule operating too stringently in relation to evidence of the defendant’s other sexual misconduct.[222] The Royal Commission downplayed traditional concerns about such evidence: ‘[T]he probative value of tendency and coincidence evidence generally has been understated [and] the risk of unfair prejudice to the accused arising from tendency and coincidence evidence has been overstated.’[223] The Royal Commission recommended the exclusion be relaxed. The Council of Attorneys-General endorsed draft UEL reforms in 2019, and they were adopted by NSW in 2020 with other jurisdictions following.[224] In child sexual offence proceedings, tendency evidence may now gain the benefit of section 97A, a strong (but rebuttable) presumption of ‘significant probative value’.
So far the reforms appear to be effective in facilitating the admission of prosecution tendency evidence in child sexual offence prosecutions.[225] However, rather than clarifying uncertainties and incoherencies, the reforms add further complexity.[226] Sections 97(1)(b) and 98(1)(b) continue to apply the requirement for ‘significant probative value’ to the challenged evidence ‘either by itself or having regard to other evidence’ (emphasis added).[227] The section 97A presumption of significant probative value operates ‘for the purposes of [the tests in] sections 97(1)(b) and 101(2)’,[228] even though there is no reference to ‘other evidence’ in the latter section.[229] However, the presumption applies only to tendency evidence, and not coincidence evidence.[230]
Interestingly, the Royal Commission drew upon the supposedly holistic nature of the probative value assessment in criticising courts for being too stringent in their application of the admissibility tests.[231] In PNJ v Director of Public Prosecutions (Vic), for example, the Victorian Court of Appeal excluded other allegations since the shared features were ‘unremarkable’, ‘commonplace’, and lacked any ‘distinctive feature’.[232] The Royal Commission indicated that this ‘overlook[s] the fact that the probative value of the tendency or coincidence evidence should be assessed in the context of the issues and the other evidence in the trial ... How much work it has to do will depend on the strength of the other evidence.’[233] Where the other evidence makes a significant contribution ‘it is not clear why any particular level of similarity ... [or] any distinctiveness in the offending would be required’.[234] As discussed in Part III(A) above, this tension between holistic and focused approaches to probative value is present in the High Court’s judgment in Phillips.
The Royal Commission is correct to highlight this tension. However, its discussion is open to two criticisms. First, the Royal Commission failed to grapple with the underlying concern considered above – that the holistic approach is unprincipled and conflates probative value and proof. There is a strong case for tendency and propensity evidence being admitted more readily,[235] but this does not justify the holistic approach to probative value. Second, the Royal Commission misstated the implications of the holistic approach, suggesting that other evidence will provide significant support only where commission rather than identity is in issue, as in many child sexual offence cases where the complainant knows the defendant.[236] ‘[A] search for additional similarities or distinctiveness in circumstances where the tendency or coincidence evidence is not being relied on to prove the identity of the accused is unwarranted.’[237] As the Royal Commission’s final report was being drafted, a majority of the High Court in Hughes made the same error suggesting that ‘close similarity’ may be required ‘to prove the identity of the offender for a known offence’, but not ‘where the fact in issue is the occurrence of the offence’.[238] This commission/identity distinction appears to have made its way into the Royal Commission tendency evidence reforms. The strong probative value presumption is restricted to cases ‘in which the commission by the defendant of an act ... is a fact in issue’.[239] But the holistic approach to probative value does not sustain this broad distinction between identity and commission cases. In most sexual offence cases with commission in issue, propensity evidence operates in conjunction with the complainant’s direct testimony.[240] But it is not necessarily the case that, where identity is in issue, there is a dearth of other evidence leaving more work for the propensity evidence.[241]
In the recent case of TL, the High Court corrected the error it had made in Hughes.[242] The defendant was charged with the murder of his two-and-a-half-year-old stepdaughter. Identity was in issue and the defence relied upon Hughes to argue that tendency evidence lacked close enough similarity with the charged offence. The tendency evidence concerned the defendant’s alleged infliction of injuries on the stepdaughter of a different kind and lesser degree than the fatal injury. The High Court rejected this argument.[243] It reiterated its strong support for the holistic assessment of probative value: ‘It is sufficient if the disputed evidence together with other evidence makes significantly more likely any facts making up the elements of the offence charged.’[244] And this contextual support may be available in identity cases as well as commission cases. In Hughes, in suggesting that ‘close similarity’ would be required in an identity case, the majority had in mind ‘a situation in which there is little or no other evidence of identity apart from the tendency evidence, and the identity of the perpetrator is “at large”’.[245] In TL, however,
there was [other] important evidence of identity, including the evidence that the appellant was one of only three persons who had the opportunity to inflict the fatal injuries and the evidence pointing against the likelihood that either the mother or the nephew was the perpetrator.[246]
Close similarity was not required. Admissibility was upheld.
Assuming that probative value is assessed holistically, the High Court’s reasoning in TL is unobjectionable. However, the holistic approach appears unprincipled. As explored further in the next Part, it conflates probative value and proof,[247] and appears to trespass on the province of the jury.
As discussed in Part IV(B) below, in related areas of evidence law courts seek to maintain a clear demarcation between the judicial and jury functions. And yet the holistic approach to probative value requires the trial judge to weigh the challenged evidence together with the rest of the prosecution case. How, then, did the notion of holistic probative value assessment arise? As explored in the next section, it may be the result of a persistent fallacy in judicial reasoning.
Proof is concerned with the weight of the evidence in its entirety, while probative value is the contribution of a challenged piece of evidence. The highly contextualised approach to probative value assessment advocated in cases like Pfennig, Phillips and TL may be an instance of the ‘division fallacy’ – a failure to properly distinguish between the parts and the whole.[248] A majority of the High Court committed this fallacy in Hoch v The Queen (‘Hoch’),[249] a precursor of Pfennig, in suggesting that ‘the [propensity] evidence, being circumstantial evidence, has probative value only if it bears no reasonable explanation other than the happening of the events in issue’.[250] This is clearly wrong. Evidence can be probative notwithstanding it leaves innocence as a reasonable possibility.[251] Evidence of a defendant’s motive in a murder case, for example, is probative even though it may be entirely consistent with the defendant’s innocence: ‘[T]he vast majority of people with a motive to kill do not go on to commit murder.’[252] The majority in Hoch appears to have simply got confused between probative value and proof; between a brick and the wall.[253]
Another variant of this division fallacy has appeared on occasions over recent decades, in relation to the jury’s approach to proof. In 1984 in Chamberlain v The Queen [No 2] (‘Chamberlain’), Gibbs CJ and Mason J held that ‘the jury can draw inferences only from facts which are proved beyond reasonable doubt’.[254] As with the principles developed in Hoch and Pfennig, the Chamberlain principle inappropriately extends the criminal standard of proof from the larger question of guilt to a component in the prosecution case. However, whereas Hoch and Pfennig were concerned with the degree to which the challenged evidence must prove guilt, the Chamberlain principle is concerned with the degree to which underlying circumstantial facts must be proven.
In Chamberlain, Deane J disagreed with the majority,[255] arguing that the application of the criminal standard to underlying facts is inconsistent with the ‘cumulative’ nature of the prosecution case.[256] Deane J distinguished cumulative proof from the less common situation in which proof of guilt is ‘contingent’ upon certain matters; exceptionally, those matters would have to be proved beyond reasonable doubt. In 1990 in Shepherd v The Queen [No 5] (‘Shepherd’), the majority confined the Chamberlain principle to the latter situation in which ‘intermediate facts ... constitute indispensable links in a chain of reasoning towards an inference of guilt’.[257] The principle has no application where ‘the evidence consists of strands in a cable rather than links in a chain’.[258]
The notion of ‘indispensable’ inferential links may be open to different interpretations. The logical, tighter interpretation is that links are indispensable where they make up the only inferential chain leading to an ultimate fact in issue. In the absence of that inferential chain, there would be no case to be left to the jury.[259] On a looser interpretation, an inferential chain may be indispensable because, although not alone in proving an ultimate fact, without it the other inferential strands are insufficient.[260] This looser interpretation is illogical. Where an inferential chain is not alone it does not need to satisfy the criminal standard of proof by itself. It can share the work with other strands of the larger, stronger cable.
In Shepherd, McHugh J pointed out that the majority in Chamberlain applied the criminal standard to an underlying fact which was not indispensable in the narrower logical sense.[261] Lindy Chamberlain appealed against her conviction for the murder of her infant Azaria while on a family camping trip near Uluru.[262] Part of the prosecution’s case was that infant blood had been found in the Chamberlains’ car; according to the prosecution this is where Lindy had killed Azaria with a knife. The defence case was that Azaria had been taken by a dingo. Gibbs CJ and Mason J held that, in view of ‘conflicting evidence’ regarding the nature of the substance found in the car, the jury could not ‘safely accept as a primary fact’ that this was foetal blood.[263] However, Gibbs CJ and Mason J clearly did not view this fact as indispensable. On the contrary, they recognised that the prosecution’s circumstantial case operated cumulatively:
None of [the] facts, regarded in isolation, would have entitled the jury to infer that Azaria had been murdered or that Mrs Chamberlain was responsible for the murder. When the evidence of all these matters is considered together, however, its probative force is greatly increased.[264]
Ultimately, dismissing the appeals, Gibbs CJ and Mason J held the conviction safe even without the circumstantial fact of infant blood in the car.
The Chamberlain fallacy is a trap into which courts keep falling. Despite Shepherd, the Chamberlain principle re-emerged in child sexual offence cases. In Gipp v The Queen (‘Gipp’)[265] in 1998 and HML v The Queen (‘HML’)[266] in 2008, the Chamberlain principle was inappropriately applied by the High Court to propensity evidence of a defendant’s other sexual misconduct towards the complainant. In HML, Hayne J held that the jury should be told they can only use the evidence for this propensity purpose ‘if ... they are persuaded beyond reasonable doubt that some or all of the other acts did occur’.[267] He noted that a majority of the court in HML supported this position.[268] The majority appeared to consider the Chamberlain principle applicable because of the sequential nature of propensity reasoning.[269] Propensity reasoning entails a finding that the other sexual misconduct had occurred, and that this demonstrates the defendant’s tendency to commit sexual misconduct of this nature. These were viewed as ‘step[s]’[270] or ‘intermediate fact[s]’[271] in ‘that chain of reasoning’[272] which must then be proved beyond reasonable doubt.[273] However, the sequential nature of propensity reasoning is beside the point. Given the complainant’s direct evidence of the charged offence (and assuming this is independent from the other misconduct evidence),[274] the propensity inference is not ‘indispensable’, and under Shepherd, the Chamberlain principle is inapplicable.
The Chamberlain fallacy is remarkably persistent.[275] In Doyle v The Queen,[276] the NSW Court of Criminal Appeal inappropriately applied the principle to tendency evidence in a multiple-complainant child sexual assault case with commission in issue. In R v Matonwal[277] and Ilievski v The Queen[278] the Court inappropriately applied the principle to tendency evidence in relation to robbery charges with identity in issue. In each of these cases there was other independent evidence of the charged offence and so the propensity inference was not indispensable to the prosecution case.
An earlier case that did involve an indispensable inference is O’Keefe v The Queen.[279] The defendant was charged with a series of sexual assaults against four complainants. In this case, in respect of the charges relating to one complainant, JG, the tendency evidence was ‘the substantial, if not only, evidence’ implicating the defendant.[280] Accordingly, in order to convict the defendant on the JG offences, ‘the jury had to be satisfied beyond reasonable doubt that he had committed offences involving one or more of the other complainants and then reason that, because he committed those offences, he was the person who committed the JG offences’.[281] Such indispensable inferences are rare as the High Court appreciated in R v Bauer (a pseudonym) (‘Bauer’)[282] in 2018. Reversing the majority approach of Gipp and HML, the Court held that ‘trial judges ... should not ordinarily direct a jury that, before they may act on evidence of uncharged acts, they must be satisfied of the proof of the uncharged acts beyond reasonable doubt’.[283] Following Shepherd, such a direction would be appropriate only where ‘there is a significant possibility of the jury treating the uncharged acts as an indispensable link in their chain of reasoning to guilt’.[284]
The Chamberlain/Shepherd direction has been the subject of recent legislative intervention. Section 161A of the Criminal Procedure Act 1986 (NSW) has ‘similar effect’[285] to Bauer. In relation to tendency and coincidence evidence, only if matters are ‘essential’ may the jury be directed that they must be proven beyond reasonable doubt.[286] In Victoria, section 62 of the Jury Directions Act 2015 (Vic) goes much further. It is not limited to tendency and coincidence evidence, and it precludes a direction based on not only the broad Chamberlain fallacy, but also the narrower logical proposition from Shepherd.
The High Court’s holistic approach to probative value appears contrary to principle and the UEL definition. It also raises concerns about institutional harm. As Mnookin observes in her US study, ‘an aggressively holistic approach to evaluating evidence could well be seen as an invasion of the province of the jury’.[287] The High Court has recognised the jury as ‘the constitutional arbiter of guilt’,[288] the ‘little parliament’[289] with the ‘political right’[290] to issue verdicts which then carry ‘a special authority and legitimacy’.[291] In certain areas, care has been taken to keep the trial judge’s admissibility determinations within bounds and separate from the jury function. However, this care is absent from Pfennig, Phillips, Falzon and TL.
The risk of the trial judge ‘usurp[ing] ... critical aspects of the traditional role of a jury’[292] has been addressed in relation to another aspect of the assessment of probative value and admissibility. In 2016 a majority of the High Court in IMM held that, in assessing the admissibility of evidence under UEL sections 97 and 137, the trial judge should take the evidence ‘at its highest’.[293] ‘[Q]uestions as to the reliability or otherwise of evidence are matters for a jury.’[294] The trial judge would face practical challenges assessing the reliability of challenged evidence since this ‘will depend ... on its place in the evidence as a whole’.[295] The trial judge may only consider reliability, assigning ‘nil’ probative value, in ‘a limiting case in which the evidence is so inherently incredible, fanciful or preposterous that it could not be accepted by a rational jury’.[296] A couple of years later in Bauer, the High Court applied these principles to sexual assault cases. ‘[T]he risk of contamination, concoction or collusion goes only to the credibility and reliability of evidence and, therefore, is an assessment which must be left to the jury.’[297] Again, an exception was created for limiting cases,[298] however, a statutory version of the principle, section 94(5) – introduced as part of the Royal Commission reforms – appears to ‘[close] that small gap’, at least as far as tendency and coincidence evidence is concerned.[299]
A couple of admissibility provisions in the UEL, relating to hearsay[300] and admissions,[301] expressly refer to reliability. However, as highlighted in the 2005 UEL review, these do not require the trial judge to ‘form a view about the actual reliability of the representation ... [which would be] likely to require the trial judge to consider the whole of a prosecution case and determine guilt before admitting the representation as reliable’.[302] The trial judge should focus on the ‘circumstances in which a representation was made ... bearing on reliability’.[303] Actual reliability ‘will ultimately be a question for the tribunal of fact’.[304] In Sio, discussed above in Part II(D), the High Court highlighted the circumstances that the maker of the representation was an accomplice, and the representation tended to lessen his culpability.[305]
It should be acknowledged that the constraints on trial judge admissibility determinations are contested. Some argue that giving trial judges greater scope to consider the reliability of challenged evidence addresses ‘the risk of wrongful conviction’, and provides ‘an important safeguard ... against an unfair trial’.[306] In IMM, Nettle and Gordon JJ rejected practical concerns[307] and argued ‘it is a misconception of the traditional division between the functions of judge and jury to suppose that it denies the judge any role in the assessment of reliability’.[308] Gageler J said that ‘an assessment of probative value necessarily involves considerations of reliability’,[309] and the three minority judges denied that this involved ‘in any sense a usurpation of the jury’s function’.[310] On the minority’s more expansive approach, the trial judge would have the power to exclude a greater amount of evidence from the jury. The impact of the High Court’s holistic probative value assessment is less clear. The Pfennig ‘no reasonable view’ test was initially viewed as favouring exclusion but, as exemplified by TL, the test engendered a holistic approach favouring admission.
The holistic probative jurisprudence fails to display much awareness of a potential intrusion into the jury’s fact-finding province even though the risk appears particularly acute. McHugh J, who opposed the ‘no reasonable view’ test in Pfennig, observed in Melbourne v The Queen that in upholding admissibility ‘[t]he judge has then, in effect, determined that the accused is guilty of the charges’.[311] This is not to say that the trial judge exactly stands in the shoes of the jury. McHugh J recognised, ‘of course, it is for the jury to determine the ultimate question of the guilt or innocence of the accused on the whole of the evidence’.[312] In making a holistic probative value assessment, the trial judge is necessarily quite cursory. The High Court indicated in Phillips that the trial judge ‘assume[s] that the [challenged] evidence would be accepted as true and that the prosecution case (as revealed in evidence already given at trial or in the depositions of witnesses later to be called) may be accepted by the jury’.[313] Admissibility would, most likely, be determined in the absence of the jury,[314] and the jury would be instructed to the effect that ‘you are quite correctly called the judges of the facts. I have nothing to do with those facts or your decisions in relation to them.’[315]
Nevertheless, the holistic probative value assessment retains a close resemblance to a jury determination. In Falzon, in ruling on admissibility, the High Court held that the challenged evidence ‘[c]ombined with the other circumstantial evidence ... constituted a powerful circumstantial case’.[316] Such an observation by a trial judge could be construed as encouraging the jury to convict. The High Court’s probative value holism, as well as being conceptually incoherent, raises significant concerns about the trial judge/jury demarcation.
In resolving issues of guilt in the criminal trial, responsibilities are shared between the trial judge and jury. The jury ultimately determines whether guilt is factually proven, while the trial judge ensures that the trial operates according to law. A key function of the trial judge is to exclude inadmissible evidence. The exclusion of relevant evidence may hinder the pursuit of factual accuracy, and probative value is often a key admissibility factor. In many situations, the more probative the evidence, the more likely it will gain admission.
This article has explored the atomistic/holistic dimensions of probative value assessments. As a matter of principle and definition, the jury’s approach to proof is holistic, and the trial judge’s probative value assessments are more atomistic. The evidence is assumed reliable, and the trial judge focuses on its strength of connection with the fact in issue. Probative value assessments are not wholly divorced from context. Like the related notion of relevance, probative value concerns the relationship between evidence and the facts in issue. The trial judge takes account of the broader trial context in identifying the live issue(s), but in then determining the existence and strength of the relationship between the evidence and the issue, the trial judge should adopt a tighter focus. These propositions are relatively settled, although, as discussed in Part II(B), in relation to the consent issue in sexual assault cases, the High Court has done a poor job in recognising the relevance of other allegations.
Probative value is a measure of the strength of connection between challenged evidence and a fact in issue. One issue that can arise for the trial judge is the extent to which a body of challenged evidence should be viewed holistically as opposed to being broken down into its constituent parts. As discussed in Part II(D), the answer may depend upon the type of evidence under consideration. A more atomistic approach may suit improperly obtained evidence for example, while expert evidence may demand a more holistic approach. Coincidence evidence may require greater nuance. The improbability of the coincidence increases as the incidents accumulate, but the admissibility of the different incidents should be treated separately. Care is required in determining the degree of atomism or holism to be applied to the challenged evidence.
Parts III and IV of this article have focused on a more extreme and troubling holism in the High Court’s admissibility jurisprudence. According to a line of High Court authority, both at common law and under the UEL, running from Pfennig through Phillips to TL, the probative value of challenged evidence should be assessed, not ‘standing alone’, but ‘view[ed] in the context of the prosecution case’.[317] In this instance context is not used, as it is with the prior question of relevance, to identify the live factual issues. Nor is it a limited holism going to the granularity of the challenged evidence. The challenged evidence is ‘[c]ombined with the other ... evidence’.[318] Subsumed by the larger prosecution case, it is unclear whether and to what extent any focus on the challenged evidence remains.
The High Court’s holistic approach to probative value raises a number of problems. It contradicts another kind of probative value holism which has a more secure policy basis and is expressly noted in the UEL – assessing the ‘importance’ of evidence having regard to other prosecution evidence. If there is other evidence available on the same issue excluding challenged prejudicial evidence will be less of a concern. With ‘importance’, other prosecution evidence weakens the admissibility of challenged evidence. According to the High Court in Pfennig, Phillips and TL, other prosecution evidence enhances the probative value and admissibility of challenged evidence.
The High Court’s probative value holism is ill-considered. It may be the product of fallacious reasoning – a conflation of probative value with proof – which also crops up in the High Court’s jury direction jurisprudence. The High Court has developed this approach to admissibility with little regard for the risk of trespassing on the jury’s fact-finding province. Probative value holism has brought conceptual incoherence and institutional uncertainty. The High Court should reconsider its position, paying greater attention to the logic of proof, the practicalities of admissibility determinations, and the jury’s role as constitutional fact finder.
* Professor of Evidence Law, University of Sydney Law School. I am grateful for feedback obtained from colleagues at a presentation of an earlier version of this article at the Sydney Institute of Criminology in September 2023.
[1] Jennifer L Mnookin, ‘Atomism, Holism, and the Judicial Assessment of Evidence’ (2013) 60(6) University of California Los Angeles Law Review 1524, 1543 (emphasis in original).
[2] Ibid 1555.
[3] Ibid 1539.
[4] [2006] HCA 4; (2006) 225 CLR 303 (‘Phillips’).
[5] [2011] HCA 7; (2011) 242 CLR 374 (‘Stubley’).
[6] Uniform Evidence Law (‘UEL’): see Evidence Act 1995 (Cth) (‘Cth Evidence Act’); Evidence Act 2011 (ACT) (‘ACT Evidence Act’); Evidence Act 1995 (NSW) (‘NSW Evidence Act’); Evidence (National Uniform Legislation) Act 2011 (NT) (‘NT Evidence Act’); Evidence Act 2001 (Tas) (‘Tas Evidence Act’); Evidence Act 2008 (Vic) (‘Vic Evidence Act’). In this article, a reference to the UEL (n 6) is a reference to the corresponding section(s) in each Act unless otherwise stated.
[7] [2020] HCA 1; (2020) 267 CLR 109 (‘Kadir’).
[8] [2016] HCA 32; (2016) 259 CLR 47 (‘Sio’).
[9] (1982) 150 CLR 580 (‘Perry’).
[10] [1995] HCA 7; (1995) 182 CLR 461 (‘Pfennig’).
[11] Royal Commission into Institutional Responses to Child Sexual Abuse (Criminal Justice Report, August 2017) pt VI, 411 (‘Criminal Justice Report’). See below n 224 on the UEL (n 6) amendments.
[12] [2022] HCA 35; (2022) 275 CLR 83 (‘TL’).
[13] See, eg, Green v The Queen [1971] HCA 55; (1971) 126 CLR 28 (‘Green’); UEL (n 6) s 141(1).
[14] Doney v The Queen [1990] HCA 51; (1990) 171 CLR 207, 214 (Deane, Dawson, Toohey, Gaudron and McHugh JJ).
[15] Green (n 13) 33 (Barwick CJ, McTiernan and Owen JJ).
[16] Chamberlain v The Queen [No 2] [1984] HCA 7; (1984) 153 CLR 521, 535 (Gibbs CJ and Mason J) (‘Chamberlain’), quoting Weeder v The Queen (1980) 71 Cr App R 228, 231 (Lane CJ for the Court), Re Belhaven and Stenton Peerage (1875) 1 App Cas 278, 279 (Lord Cairns LC). See also R v Hillier [2007] HCA 13; (2007) 228 CLR 618, 638 [48] (Gummow, Hayne and Crennan JJ), quoting Chamberlain (n 16) 535 (Gibbs CJ and Mason J).
[17] Mnookin (n 1) 1540, citing Nancy Pennington and Reid Hastie, ‘A Cognitive Theory of Juror Decision Making: The Story Model’ (1991) 13(2–3) Cardozo Law Review 519.
[18] UEL (n 6) Dictionary (definition of ‘probative value’).
[19] [2016] HCA 14; (2016) 257 CLR 300 (‘IMM’).
[20] Ibid 310 [30] (French CJ, Kiefel, Bell and Keane JJ).
[21] Ibid 311 [36].
[22] Ibid 312 [36].
[23] Mnookin (n 1) 1536.
[24] Consider also the notion of provisional relevance. The relevance of one item of evidence may be contingent on the acceptance of a fact which can only be proven by another item of evidence: UEL (n 6) s 57; Law Reform Commission, Evidence (Report No 26, 1985) vol 1, 353–5 [646] <http://www.austlii.edu.au/au/other/lawreform/ALRC/1985/26.pdf?stem=0 & synonyms=0 & query=alrc%20evidence> (‘Evidence No 26’). As Mnookin notes, the notion is of considerable theoretical interest but ‘judges in practice do not seem to face significant difficulties in its application’: ibid 1553.
[25] Mnookin (n 1) 1543, citing Federal Rules of Evidence (1975) r 401 (‘Federal Rules of Evidence’).
[26] UEL (n 6) s 55(1).
[27] According to the modern notion of ‘logical relevance’: see below n 60.
[28] UEL (n 6) Dictionary (definition of ‘probative value’) (emphasis added).
[29] Stubley (n 5) 409 [117] (Heydon J).
[30] In R v Little [2018] QCA 113 (‘Little’), the perpetrator wore a balaclava and blindfolded the complainant. Even still, the complainant knew the defendant and ‘identity had ceased to be a real issue’; consent became the dominant issue: at [7], [24]–[26] (Fraser JA).
[31] See, eg, Phillips (n 4) 311–12 [27] (Gleeson CJ, Gummow, Kirby, Hayne and Heydon JJ).
[32] See, eg, Pell v The Queen [ 2019] VSCA 186, [15] (Ferguson CJ and Maxwell P).
[33] Royal Commission into Institutional Responses to Child Sexual Abuse (Final Report, December 2017) vol 4 (‘Final Report’); Vaughn I Rickert, Constance M Wiemann and Roger D Vaughan, ‘Disclosure of Date/Acquaintance Rape: Who Reports and When’ (2005) 18(1) Journal of Pediatric and Adolescent Gynecology 17 <https://doi.org/10.1016/j.jpag.2004.11.006>.
[34] See, eg, O’Keefe v The Queen [2009] NSWCCA 121 (‘O’Keefe’).
[35] Phillips (n 4) 311 [26] (Gleeson CJ, Gummow, Kirby, Hayne and Heydon JJ).
[36] R v Bond [1906] UKLawRpKQB 91; [1906] 2 KB 389, 417 (Bray J).
[37] HML v The Queen [2008] HCA 16; (2008) 235 CLR 334, 352 [6] (Gleeson CJ) (‘HML’).
[38] R v Nieterink (1999) 76 SASR 56, 65 [43] (Doyle CJ).
[39] See, eg, R v Butler (1987) 84 Cr App R 12.
[40] Whether or not something remains a live issue may not be straightforward. This is the point on which Heydon J dissented in Stubley (n 5).
[41] See, eg, R v Tweed [1992] NI 269.
[42] See Phillips (n 4).
[43] Ibid 318 [47] (Gleeson CJ, Gummow, Kirby, Hayne and Heydon JJ).
[44] Jeremy Gans, ‘Similar Facts after Phillips’ (2006) 30(4) Criminal Law Journal 224, 229–31; Bill Pincus, ‘Recent Cases: Phillips v The Queen’ (2006) 80(8) Australian Law Journal 504, 510; David Hamer, ‘Similar Fact Reasoning in Phillips: Artificial, Disjointed and Pernicious’ [2007] UNSWLawJl 36; (2007) 30(3) University of New South Wales Law Journal 609, 616 (‘Similar Fact Reasoning’); R v Healy [2007] NZCA 451 [69] (France J for the Court). ‘[T]he question whether it is proved that one of them did not consent may in part be answered by proving that another of the women did not consent if the circumstances bear a striking resemblance’: R v Wilmot (1989) 89 Cr App R 341, 345 (Glidewell LJ for the Court). See also David Hamer, ‘The Structure and Strength of the Propensity Inference: Singularity, Linkage and the Other Evidence’ [2003] MonashULawRw 7; (2003) 29(1) Monash University Law Review 137, 191 (‘Structure and Strength’).
[45] Pincus (n 44) 510. See also R v PS [2004] QCA 347, [59], [74] (Williams JA).
[46] Strictly speaking, the Phillips (n 4) holding on consent and relevance should just be viewed, not as a legal principle, but as a relevance ruling on the facts of the case, and one that probably was better left to the trial judge: Smith v The Queen [2001] HCA 50; (2001) 206 CLR 650, 658–9 [23] (Kirby J) (‘Smith’); Andrew Ligertwood and Gary Edmond, Australian Evidence: A Principled Approach to the Common Law and Uniform Acts (LexisNexis Butterworths, 6th ed, 2017) 81; David Hamer, ‘The Significant Probative Value of Tendency Evidence’ [2019] MelbULawRw 5; (2019) 42(2) Melbourne University Law Review 506, 518–21 (‘Significant Probative Value’).
[47] This may not be a legitimate basis for distinguishing Phillips (n 4) since the defendants’ conduct on some counts was also in issue in that case: at 321 [55] (Gleeson CJ, Gummow, Kirby, Hayne and Heydon JJ). This did not prevent the Court considering the other allegations irrelevant to consent.
[48] [2007] WASCA 183 (‘Osborne’).
[49] Ibid [27] (Wheeler JA, Miller JA agreeing at [51]).
[50] Ibid [21].
[51] [2008] SASC 47; (2008) 100 SASR 119.
[52] Ibid 137 [98] (Vanstone J).
[53] Ibid 136 [95].
[54] Ibid 128 [41] (Bleby J).
[55] See Stubley (n 5).
[56] Ibid 392 [64]–[65] (Gummow, Crennan, Kiefel and Bell JJ), 402 [96], 415 [138] (Heydon J dissenting).
[57] Ibid 381 [15] (Gummow, Crennan, Kiefel and Bell JJ). See also Stubley v Western Australia [2009] WASC 57, [39] (Johnson J), citing Osborne (n 48) [27] (Wheeler JA, Miller JA agreeing at [51]). A majority of the Western Australian Court of Appeal agreed ‘that Phillips did not preclude the admission of the evidence of the propensity witnesses for purposes unrelated to consent, even though the evidence may have borne indirectly on this issue’: Stubley (n 5) 390 [60] (Gummow, Crennan, Kiefel and Bell JJ), citing Stubley v Western Australia [2010] WASCA 36, [375] (Buss JA).
[58] Heydon J dissented on the basis that consent was not the only live issue: Stubley (n 5) 402 [96]. He did not discuss Phillips (n 4) in his judgment, but at the appeal hearing noted it was ‘one of the most criticised decisions of the High Court of all time’: Transcript of Proceedings, Stubley v Western Australia [2010] HCATrans 269, 22.
[59] Stubley (n 5) 395 [74] (Gummow, Crennan, Kiefel and Bell JJ). Force, threats, or intimidation against the complainants needed to be proven since ‘by reference to the [substantive] law [of sexual assault] as it stood at the material time [ie, the 1970s] ... manipulating a person into sexual intercourse by exploiting that person’s known psychological vulnerability would not, without more, vitiate their consent’: at 393 [69] (Gummow, Crennan, Kiefel and Bell JJ), 397 [82] (Heydon J). See also at 394 [70] (Gummow, Crennan, Kiefel and Bell JJ). It does not follow, as a matter of proof or evidence law, that evidence of manipulation is irrelevant to consent.
[60] Evidence No 26 (n 24) vol 1, 350 [641]. See also Smith (n 46) 663–4 [41] (Kirby J).
[61] The prosecution failed to press this point: Stubley (n 5) 394 [72] (Gummow, Crennan, Kiefel and Bell JJ). Empirical evidence suggests that the connection between offending behaviours may be quite strong despite variation in the defendant’s modus operandi. Many sex offenders, like other criminals, are generalists rather than narrow specialists: Criminal Justice Report (n 11) pt VI, 595, 603.
[62] Mnookin (n 1) 1544.
[63] [2017] VSCA 309 (‘Jacobs’).
[64] Ibid [16]–[17] (Maxwell P, Ashley JA and Forrest AJA). Though there were also claims of unlawful imprisonment: at [13] (Maxwell P, Ashley JA and Forrest AJA). In R v Collins [2013] QCA 389, adopting similar reasoning, the Queensland Court of Appeal held the other alleged victims’ evidence admissible in respect of the occasions where the young women claimed the defendant had given them a stupefying drink, but inadmissible where the defendant merely isolated them and plied them with (unspiked) alcohol: at [37], [49]–[50] (McMurdo P).
[65] Jacobs (n 63) [4] (Maxwell P, Ashley JA and Forrest AJA).
[66] Ibid [48].
[67] Little (n 30).
[68] Ibid [24] (Fraser JA).
[69] Ibid [5], [21].
[70] See, eg, Hughes v The Queen [2017] HCA 20; (2017) 263 CLR 338 (‘Hughes’). ‘[A] tendency expressed at a level of particularity will be more likely to be significant’: at 363 [64] (Kiefel CJ, Bell, Keane and Edelman JJ).
[71] Mnookin (n 1) 1544. See also John Henry Wigmore, A Treatise on the Anglo-American System of Evidence in Trials at Common Law (Little, Brown and Company, 2nd ed, 1923) 233–6 with reference to Federal Rules of Evidence (n 25) r 401, which also adopts the ‘logical relevance’ notion: Ligertwood and Edmond (n 46) 83–4.
[72] See, eg, UEL (n 6) pt 3.2; Walton v The Queen [1989] HCA 9; (1989) 166 CLR 283 (‘Walton’).
[73] See, eg, UEL (n 6) pt 3.7; Palmer v The Queen [1998] HCA 2; (1998) 193 CLR 1 (‘Palmer’).
[74] See, eg, UEL (n 6) pt 3.3; Ramsay v Watson [1961] HCA 65; (1961) 108 CLR 642.
[75] See, eg, UEL (n 6) pt 3.6; Pfennig (n 10).
[76] See, eg, UEL (n 6) ss 135, 137; R v Christie [1914] UKLawRpAC 20; [1914] AC 545 (‘Christie’).
[77] See, eg, UEL (n 6) ss 84–5, 90; R v Swaffield (1998) 192 CLR 159.
[78] See, eg, UEL (n 6) ss 138–9; Bunning v Cross [1978] HCA 22; (1978) 141 CLR 54 (‘Bunning’).
[79] Dale A Nance, ‘The Best Evidence Principle’ (1988) 73(2) Iowa Law Review 227, 233.
[80] UEL (n 6) s 135(c).
[81] Palmer (n 73) 22 [52] (McHugh J), quoting Natta v Canham [1991] FCA 470; (1991) 32 FCR 282, 298 (French, O’Loughlin and Higgins JJ). See also Pfennig (n 10) 513 (McHugh J); UEL (n 6) s 192(2)(a).
[82] Ho Hock Lai, A Philosophy of Evidence Law: Justice in the Search for Truth (Oxford University Press, 2008) 337.
[83] David P Leonard, ‘In Defense of the Character Evidence Prohibition: Foundations of the Rule against Trial by Character’ (1998) 73(4) Indiana Law Journal 1161, 1188. Cf Vic Evidence Act (n 6) s 135 (as amended in 2014) allowing the exclusion of evidence ‘if its probative value is substantially outweighed by the danger that the evidence might ... unnecessarily demean the deceased in a ... homicide [trial]’.
[84] Marvin E Frankel, ‘The Search for Truth: An Umpireal View’ (1975) 123(5) University of Pennsylvania Law Review 1031, 1033, 1055 <https://doi.org/10.2307/3311524>. See also Jack B Weinstein, ‘Some Difficulties in Devising Rules for Determining Truth in Judicial Trials’ (1966) 66(2) Columbia Law Review 223, 243 <https://doi.org/10.2307/1120774>.
[85] Nance (n 79) 281–3.
[86] Festa v The Queen (2001) 208 CLR 593, 609–10 [51] (McHugh J) (‘Festa’).
[87] Pfennig (n 10) 512 (McHugh J); Hughes (n 70) 365–8 [72]–[78] (Gageler J). Cf Criminal Justice Report (n 11) pt VI, 633–4. See also Part III(C) below.
[88] Christie (n 76); Ligertwood and Edmond (n 46) 94–9.
[89] Bunning (n 78) 80 (Stephen and Aickin JJ). See also at 64 (Barwick CJ).
[90] Boardman v DPP (UK) [1975] AC 421, 456 (Lord Cross) (‘Boardman’), quoted in HML (n 37) 482 [443] (Crennan J).
[91] Weinstein (n 84) 228 n 23. In the UEL (n 6), an admission is absolutely excluded if influenced by ‘violent, oppressive, inhuman or degrading conduct’: at s 84. But improperly or illegally obtained evidence more generally may be admissible if sufficiently probative and important: at s 138.
[92] UEL (n 6) ss 97–8. See also at ss 103, 108C(1)(b)(ii).
[93] Ibid ss 101, 126B, 135, 137, 138, 190 but note that section 126B only appears in the ACT Evidence Act (n 6), NSW Evidence Act (n 6) and Tas Evidence Act (n 6). See also UEL (n 6) s 18(7)(b), which requires the trial judge to consider ‘the weight that is likely to be attached’ to the challenged evidence. Other provisions refer to the related notions of reliability (at sections 65, 85), and importance (at sections 18, 114, 115, 126B, 130, 138, 169, 190, 192 (but note that sections 114 and 115 do not appear in the Tas Evidence Act (n 6)); see below Part III(B)), or incorporate pre-determined indicia of reliability (see, eg, hearsay exceptions in UEL (n 6) ss 62, 65(2)–(3), 66(2), 69(1)–(2), 81–2).
[94] See Part III(C) below.
[95] See below nn 106–17, 300–5 and accompanying text.
[96] Cass R Sunstein, ‘Incommensurability and Valuation in Law’ (1994) 92(4) Michigan Law Review 779 <https://doi.org/10.2307/1289693>.
[97] Notwithstanding McHugh J’s view to the contrary in Pfennig (n 10) 528. See David Hamer, ‘The Legal Structure of Propensity Evidence’ (2016) 20(2) International Journal of Evidence and Proof 136, 154–5 <https://doi.org/10.1177/1365712716628540> (‘Legal Structure’).
[98] Mike Redmayne, Character in the Criminal Trial (Oxford University Press, 2015) 134 <https://doi.org/10.1093/acprof:oso/9780199228898.001.0001>.
[99] Hamer, ‘Legal Structure’ (n 97) 154–5. Similar balancing tests appear in sections 101 and 135 of the UEL (n 6).
[100] UEL (n 6) s 138(1).
[101] Ibid ss 138(3)(a), (d).
[102] Evidence No 26 (n 24) vol 1, 534–7 [964].
[103] Ibid 260 [468]. See also Kadir (n 7) 125 [13] (Kiefel CJ, Bell, Keane, Nettle and Edelman JJ).
[104] Sunstein (n 96) 796.
[105] Ibid 860.
[106] First-hand hearsay, criminal proceedings, maker not available: UEL (n 6) s 65. Mr Filihia, tried separately, was called by the prosecution but efforts to compel him to testify failed.
[107] Ibid s 65(2)(d).
[108] Sio (n 8) 66 [63] (French CJ, Bell, Gageler, Keane and Gordon JJ), quoting Walton (n 72) 293 (Mason CJ).
[109] Sio (n 8) 69 [73] (French CJ, Bell, Gageler, Keane and Gordon JJ).
[110] Ibid 66 [62].
[111] Ibid 63 [51], 64–5 [58]–[59], [62].
[112] Ibid 64 [58].
[113] Ibid 68 [71].
[114] Ibid 64 [55].
[115] Ibid 67 [68].
[116] See above nn 92–3 and accompanying text.
[117] ‘[Sections] 65(2)(c) and (d) and 85 provide “[t]he only occasion for a trial judge to consider the reliability of evidence, in connection with the admissibility of evidence”’: Sio (n 8) 68 [72] (French CJ, Bell, Gageler, Keane and Gordon JJ), quoting IMM (n 19) 316 [54] (French CJ, Kiefel, Bell and Keane JJ).
[118] Stephen Odgers, Uniform Evidence Law (Lawbook, 18th ed, 2023) 1407 [EA 138.510], citing R v MM [2004] NSWCCA 364, [54] (James J).
[119] Odgers (n 118) 1416 [EA 138.630], quoting Parker v Comptroller-General of Customs [2007] NSWCA 348; (2007) 232 FLR 362, 379 [57], 381 [64] (Basten JA).
[120] UEL (n 6) ss 138(3)(a), (d).
[121] Ibid s 138(1)(b).
[122] Kadir (n 7) 135 [42] (Kiefel CJ, Bell, Keane, Nettle and Edelman JJ).
[123] Ibid 133 [37].
[124] Ibid 135 [41].
[125] Ibid 139 [51].
[126] Mnookin (n 1) 1534.
[127] Ibid 1572.
[128] [1997] USSC 88; 522 US 136 (1997) (‘General Electric’).
[129] Mnookin (n 1) 1572.
[130] Ibid 1576.
[131] Ibid 1573, quoting General Electric (n 128) 152–3 (Stevens J).
[132] Gary Edmond, ‘Icarus and the Evidence Act: Section 137, Probative Value and Taking Forensic Science Evidence “at Its Highest”’ [2017] MelbULawRw 22; (2017) 41(1) Melbourne University Law Review 106, 133.
[133] In any event, in Australia, unlike the US, reliability is not a requirement for the admissibility of expert evidence: see Gary Edmond, ‘Regulating Forensic Science and Medicine Evidence at Trial: It’s Time for a Wall, a Gate and Some Gatekeeping’ (2020) 94(6) Australian Law Journal 427, 436.
[134] Perry (n 9) 588 (Gibbs CJ).
[135] Ibid 586. As discussed in Part III, authorities regarding the admissibility test were and remain very unsettled.
[136] Ibid 591 (Murphy J). This includes the fourth alleged poisoning giving rise to the charges. To the extent that this involved an assumption regarding the defendant’s guilt, this raised concerns about circularity. ‘To seek to prove a fact in issue by a chain of reasoning which assumes the truth of that fact is, of course, a fallacy’: at 612 (Brennan J). See also at 589–90 (Gibbs CJ), 594–5 (Murphy J). However, circularity is avoidable if the reasoning merely notes that the defendant is implicated in the charged poisoning, leaving the defendant’s guilt as a possible conclusion: see Evidence No 26 (n 24) vol 1, 219–20 [400]; Pfennig (n 10) 530 (McHugh J).
[137] Perry (n 9) 612 (Brennan J) (‘must’ has been inserted in place of ‘much’, which appears to be a printing error).
[138] Ibid 610.
[139] Ibid 600 (Murphy J).
[140] Ibid 589–91 (Gibbs CJ).
[141] Ibid 605–8 (Wilson J), 611–13 (Brennan J). The fifth member of the Court, Aickin J, died before the decision was published: at 614.
[142] Sutton v The Queen [1984] HCA 5; (1984) 152 CLR 528, 534 (Gibbs CJ) (‘Sutton’).
[143] See, eg, Pfennig (n 10) 516 (McHugh J).
[144] See David Hamer, ‘“Tendency Evidence” and “Coincidence Evidence” in the Criminal Trial: What’s the Difference?’ in Andrew Roberts and Jeremy Gans (eds), Critical Perspectives on the Uniform Evidence Law (Federation Press, 2017) 158 (‘Tendency and Coincidence Chapter’). Whether the common law exclusion extends to other uses of evidence showing a defendant’s bad character is unclear: Ligertwood and Edmond (n 46) 248, discussing the division of opinion in HML (n 37). Under the UEL (n 6) the specific exclusionary rules are limited to evidence adduced for these purposes: at ss 97–8. If evidence is admissible for a non-propensity purpose it cannot be used for a propensity purpose without satisfying the part 3.6 admissibility tests (at s 95) and may be liable to exclusion because of the risk of propensity-based prejudice (at s 137).
[145] Pfennig (n 10) 482 (Mason CJ, Deane and Dawson JJ), 512–13 (McHugh J). See also Perry (n 9) 603–4 (Wilson J).
[146] Boardman (n 90) 445 (Lord Hailsham).
[147] See, eg, ibid 456–7 (Lord Cross); LH Hoffmann, ‘Similar Facts after Boardman’ (1975) 91 (April) Law Quarterly Review 193; Sweitzer v The Queen [1982] 1 SCR 949, 953.
[148] [1989] HCA 50; (1989) 167 CLR 590 (‘Harriman’).
[149] Ibid 598 (Dawson J). See also at 593–4 (Brennan J), 610 (Toohey J); B v The Queen [1992] HCA 68; (1992) 175 CLR 599, 618–19 (Dawson and Gaudron JJ); Pfennig (n 10) 478 (Mason CJ, Deane and Dawson JJ), 515, 528 (McHugh J); BRS v The Queen [1997] HCA 47; (1997) 191 CLR 275, 305 (McHugh J); Gipp v The Queen (1998) 194 CLR 106, 157 [142] (Kirby J) (‘Gipp’).
[150] In England, the balancing test was adopted at common law: see, eg, DPP (UK) v P [1991] 2 AC 447, 460–1 (Lord Mackay LC). It was subsequently replaced by a more complex and permissive statutory framework: Criminal Justice Act 2003 (UK) pt 11; JR Spencer, Evidence of Bad Character (Hart Publishing, 3rd ed, 2016). The balancing test operates at common law in Canada (R v Handy [2002] 2 SCR 908, 932 [55] (Binnie J)) and under legislation in New Zealand (Evidence Act 2006 (NZ) s 43(1)).
[151] Sutton (n 142).
[152] Ibid 534 (Gibbs CJ).
[153] Ibid.
[154] Pfennig (n 10) 483 (Mason CJ, Deane and Dawson JJ).
[155] Ibid 484. This alternative formulation received support in earlier decisions: see, eg, Sutton (n 142) 564 (Dawson J); Hoch v The Queen [1988] HCA 50; (1988) 165 CLR 292, 296 (Mason CJ, Wilson and Gaudron JJ) (‘Hoch’).
[156] Peacock v The King [1911] HCA 66; (1911) 13 CLR 619, 630 (Griffith CJ). See also Shepherd v The Queen [1990] HCA 56; (1990) 170 CLR 573, 578 (Dawson J) (‘Shepherd’); David Hamer, ‘Probabilistic Standards of Proof, Their Complements, and the Errors that Are Expected to Flow from Them’ [2004] UNELawJl 3; (2004) 1(1) University of New England Law Journal 71, 99–104.
[157] Grant v The Queen [1905] ArgusLawRp 161; (1975) 11 ALR 503, 505 (Barwick CJ) (‘Grant’).
[158] Pfennig (n 10) 483 (Mason CJ, Deane and Dawson JJ). See also at 515–16 (McHugh J).
[159] See Hoffmann (n 147) 194.
[160] Pfennig (n 10) 516 (McHugh J).
[161] See, eg, ibid 469–70, 489–90 (Mason CJ, Deane and Dawson JJ).
[162] [1936] HCA 23; (1936) 55 CLR 367.
[163] Velkoski v The Queen [2014] VSCA 121; (2014) 45 VR 680, 686 [29] (Redlich, Weinberg and Coghlan JJA) (‘Velkoski’). See also David Hamer, ‘Proof of Serial Child Sexual Abuse: Case-Law Developments and Recidivism Data’ in Thomas Crofts and Arlie Loughnan (eds), Criminalisation and Criminal Responsibility in Australia (Oxford University Press, 2015) 242, 245–6.
[164] Phillips (n 4) 323 [63] (Gleeson CJ, Gummow, Kirby, Hayne and Heydon JJ). See also Pfennig (n 10) 478, 485 (Mason CJ, Deane and Dawson JJ); Hamer, ‘Similar Fact Reasoning’ (n 44) 628. Cf Melbourne v The Queen (1999) 198 CLR 1, 17 (McHugh J) (‘Melbourne’); Festa (n 86) 623–4 [97] (McHugh J); Hamer, ‘Structure and Strength’ (n 44) 186.
[165] Phillips (n 4) 323 [63] (Gleeson CJ, Gummow, Kirby, Hayne and Heydon JJ).
[166] Grant (n 157) 505 (Barwick CJ); Knight v The Queen [1992] HCA 56; (1992) 175 CLR 495, 502 (Mason CJ, Dawson and Toohey JJ); Shepherd (n 156) 578 (Dawson J).
[167] Hamer, ‘Structure and Strength’ (n 44) 163–4, 183–4.
[168] In Mnookin’s diagram of the atomism/holism continuum, the only thing more extreme than considering the challenged evidence together with ‘other evidence in the case’ is to also include ‘other imaginable evidence’: Mnookin (n 1) 1539.
[169] See ibid 1534.
[170] Mnookin recognises the different dimensions but then puts them on the same continuum: ibid.
[171] With propensity and similar fact evidence, the degree of other evidence holism has varied between jurisdictions and over time. However, a degree of atomism has regularly been adopted with regard to the challenged evidence, with some parts being admitted while other parts are excluded: see, eg, Perry (n 9) 612–14 (Brennan J); Velkoski (n 163); R v Clarke [2023] NSWCCA 123; (2023) 111 NSWLR 501 (‘Clarke’).
[172] Phillips (n 4) 320–1 [54] (Gleeson CJ, Gummow, Kirby, Hayne and Heydon JJ), quoting Hoch (n 155) 301 (Brennan and Dawson JJ), Pfennig (n 10) 483 (Mason CJ, Deane and Dawson JJ).
[173] Pfennig (n 10) 484 (Mason CJ, Deane and Dawson JJ), quoted in Phillips (n 4) 320 [53] (Gleeson CJ, Gummow, Kirby, Hayne and Heydon JJ). See also at 322 [58].
[174] Pfennig (n 10) 481 (Mason CJ, Deane and Dawson JJ) (emphasis added). See also Phillips (n 4) 320–1 [54] (Gleeson CJ, Gummow, Kirby, Hayne and Heydon JJ).
[175] Pfennig (n 10) 483 (Mason CJ, Deane and Dawson JJ).
[176] R v WRC [2002] NSWCCA 210; (2002) 130 A Crim R 89 (‘WRC’); R v Joiner [2002] NSWCCA 354; (2002) 133 A Crim R 90 (‘Joiner’); R v Folbigg [2003] NSWCCA 17. Prior to R v Ellis [2003] NSWCCA 319; (2003) 58 NSWLR 700 (‘Ellis’), some courts continued to apply the Pfennig test under section 101 of the UEL (n 6).
[177] BBH v The Queen [2012] HCA 9; (2012) 245 CLR 499, 546–8 [155]–[159] (Crennan and Kiefel JJ); HML (n 37) 359 [27] (Gleeson CJ), 429–30 [285] (Heydon J). Heydon J has approved of this approach extrajudicially and claims that it was approved by the High Court in Phillips (n 4) at 323–4 [63] (Gleeson CJ, Gummow, Kirby, Hayne and Heydon JJ): see, eg, JD Heydon, Cross on Evidence (LexisNexis, 14th ed, 2024) 812 [21035] (‘Cross on Evidence’). However, any such approval has not been explicit. Hodgson JA’s approach continues to be utilised in Queensland, the sole remaining common law jurisdiction: see, eg, R v Thomson [2022] QCA 36; (2022) 296 A Crim R 510, 533 [132] (McMurdo JA).
[178] WRC (n 176) 102 [29] (Hodgson JA).
[179] Ibid.
[180] Joiner (n 176) 99 [39] (Hodgson JA).
[181] Ibid 99 [40].
[182] Sutton (n 142) 534 (Gibbs CJ), quoted in Phillips (n 4) 320–1 [54] (Gleeson CJ, Gummow, Kirby, Hayne and Heydon JJ). See also Pfennig (n 10) 528–9 (McHugh J).
[183] Phillips (n 4) 321 [56] (Gleeson CJ, Gummow, Kirby, Hayne and Heydon JJ); Hamer, ‘Similar Fact Reasoning’ (n 44) 624–7.
[184] Evidence No 26 (n 24) vol 1, xxviii.
[185] Ibid.
[186] Particularly ibid, which is the interim report, and Law Reform Commission, Evidence (Report No 38, 1987) (‘Evidence No 38’), which is the final report.
[187] New South Wales Law Reform Commission, Evidence (Report No 56, June 1988).
[188] Cth Evidence Act (n 6); ACT Evidence Act (n 6); NT Evidence Act (n 6); NSW Evidence Act (n 6); Tas Evidence Act (n 6); Vic Evidence Act (n 6).
[189] See, eg, Justice TH Smith and OP Holdenson, ‘Comparative Evidence: The Uniform Evidence Acts and the Common Law’ (1998) 72(5) Australian Law Journal 363.
[190] Evidence No 26 (n 24) vol 1, 81–6 [164]–[165], 219–20 [400]. In the period between Evidence No 26 (n 24) in 1985 and the passage of the Cth Evidence Act (n 6) on 7 February 1995, the High Court considered propensity evidence on a number of occasions, including the important decisions Hoch (n 155), Harriman (n 148) and Thompson v The Queen [1989] HCA 30; (1989) 169 CLR 1. Pfennig (n 10) was decided 10 days after the legislation passed. The propensity provisions finally adopted in 1995, while keeping the same structure, diverged in some respects from those proposed by the Law Reform Commission in the interim and final reports: Evidence No 26 (n 24) app A cls 91–3; Evidence No 38 (n 186) app A cls 87–9. The provisions in the interim and final reports were essentially the same. They distinguished tendency evidence from coincidence evidence, though applying the same double admissibility test to both. The other misconduct or events must be ‘substantially and relevantly similar’ (Evidence No 26 (n 24), app A cls 91(3)(b), 92(3)(b); Evidence No 38 (n 186) app A cls 87(b), 88(b)) and, if tendered by the prosecution against the accused, also have ‘substantive probative value’ (Evidence No 26 (n 24), app A cl 93(2)(b); Evidence No 38 (n 186) app A cl 89(2)(b)). The draft legislation contained guidance on the factors that should be considered in determining probative value: Evidence No 26 (n 24) app A cl 93(3); Evidence No 38 (n 186) app A cl 89(3). The reforms first entered Parliament as the Evidence Bill 1991 (Cth). The tendency rule in clause 103 of the 1991 Bill bore little resemblance to the Law Reform Commission’s draft legislation, containing a symmetrical balancing test, and a ‘no rational view’ test. The Bill made no reference to coincidence evidence. The Evidence Bill 1993 (Cth) and Evidence Bill 1994 (Cth) contained clauses 97, 98 and 101 that became sections 97, 98 and 101 in the UEL (n 6). See also Ellis (n 176) 714–17 [65]–[84] (Spigelman CJ); Hughes (n 70) 411–15 [185]–[192] (Nettle J). It is difficult at this point to find out the reasoning behind the various detailed changes that were made through the reform and legislative process: at 368 [80] (Gageler J). There is no explanation in the various explanatory memoranda, second reading speeches, parliamentary debates, nor in the Report of the Senate Standing Committee on Legal and Constitutional Affairs of June 1994.
[191] Velkoski (n 163) 687 [33] (Redlich, Weinberg and Coghlan JJA).
[192] Australian Law Reform Commission, NSW Law Reform Commission and Victorian Law Reform Commission, Uniform Evidence Law (ALRC Report No 102, NSWLRC Report No 112, VLRC Final Report, December 2005) ch 11 (‘Uniform Evidence Law Report’). Following this review, section 98 was amended by the Evidence Amendment Act 2007 (NSW), Evidence Amendment Act 2008 (Cth) (both commencing 1 January 2009) and Evidence Amendment Act 2010 (Tas) (commencing 1 January 2011) to broaden the definition of evidence subject to the coincidence exclusionary rule. It was originally defined very narrowly; evidence would only be caught by the exclusion if the evidence related to events which are ‘substantially and relevantly similar’ and occurred in ‘substantially similar’ circumstances. The amendment extended the scope of the exclusion to evidence of events where the adducing party relies upon ‘similarities in the events or the circumstances in which they occurred’ to establish ‘it is improbable that the events occurred coincidentally’. See also R v MR [2013] NSWCCA 236, [59]–[79] (Beech-Jones J).
[193] This distinction between tendency and coincidence evidence also adds to the unnecessary complexity of the UEL (n 6). Arguably, it serves no purpose, but is an accident of history. A major issue that had emerged at common law was the fallacious notion that propensity reasoning was absolutely forbidden, and evidence had to be adduced for non-propensity purposes in order to be admissible. On one view, adducing evidence for a similar fact purpose, relying on coincidence reasoning or the doctrine of chances, would avoid the prohibition: Evidence No 26 (n 24) 83 [165]; Edward J Imwinkelried, ‘An Evidentiary Paradox: Defending the Character Evidence Prohibition by Upholding a Non-character Theory of Logical Relevance, the Doctrine of Chances’ (2006) 40(2) University of Richmond Law Review 419 <https://doi.org/10.2139/ssrn.795725>. As discussed in the text, the common law developed so that evidence for a propensity purpose was admissible if it had sufficient probative value, and the UEL (n 6) followed the common law in this respect. However, a hangover from the forbidden reasoning notion became ossified in the UEL (n 6) – the unnecessary and confusing distinction between tendency and coincidence evidence: see Hamer, ‘Tendency and Coincidence Chapter’ (n 144); David Hamer, ‘Current Issues: The Case for Principled and Practical Propensity Evidence Reform’ (2020) 94(4) Australian Law Journal 239, 241 (‘Principled and Practical’); David Hamer, ‘Myths, Misconceptions and Mixed Messages: An Early Look at the New Tendency and Coincidence Evidence Provisions’ (2021) 45(4) Criminal Law Journal 232, 247 (‘Mixed Messages’). In Perry (n 9), Murphy J observed that the ‘supposed rigid division between prohibited use of previous criminality to show propensity and its use to establish a chain of coincidences so remarkable that it excludes the accused's innocence as “an affront to common sense”, is unsatisfactory’: at 592.
[194] Another change was made to section 101 of the UEL (n 6). Previously section 101 referred to ‘any prejudicial effect’ rather than ‘unfair prejudice’ as appears, for example, in section 137. In the latest reforms, section 101 is brought into line with section 137 in this respect. However, the difference in wording was without significance: R v Bauer (a pseudonym) [2018] HCA 40; (2018) 266 CLR 56, 93–4 [73] (Kiefel CJ, Bell, Gageler, Keane, Nettle, Gordon and Edelman JJ) (‘Bauer’).
[195] UEL (n 6) ss 97(1)(b), 98(1)(b) (emphasis added).
[196] The majority in IMM (n 19) appear to suggest that circumstantial evidence depends upon other evidence to gain probative value: ‘Taken by itself, the evidence may, if accepted, support an inference to a high degree of probability that the fact in issue exists. On the other hand, it may only, as in the case of circumstantial evidence, strengthen that inference, when considered in conjunction with other evidence’: at 313 [45] (French CJ, Kiefel, Bell and Keane JJ). The majority’s meaning is obscure and unexplained, however, this appears wrong. Classic types of circumstantial evidence – such as motive, means, and opportunity – may gain probative value independently. The majority adds: ‘The evidence, if accepted, may establish a sufficient condition for the existence of the fact in issue or only a necessary condition.’ This is similarly obscure and difficult to accept.
[197] Hughes (n 70) 356 [40] (Kiefel CJ, Bell, Keane and Edelman JJ).
[198] TL (n 12) 95 [28] (Kiefel CJ, Gageler, Gordon, Steward and Gleeson JJ) (emphasis added).
[199] Oxford English Dictionary (online at 10 May 2024) ‘either-or’ (defs 1 and 2). ‘Or’ may be read as ‘and’ in some situations, but this is problematic where the disjunction is of the ‘either-or’ variety: see D Pearce, Statutory Interpretation in Australia (LexisNexis Butterworths, 9th ed, 2019) 72–3 [2.50].
[200] The second safeguard – the balancing admissibility test for prosecution tendency and coincidence evidence in section 101(2) of the UEL (n 6) – simply directs the trial judge to weigh up the ‘probative value of the evidence’. This is particularly odd because, as discussed in the previous section of this article, at common law it seems to be the other way around. It was the balancing test, which in Pfennig (n 10) transmuted into the ‘no reasonable view’ test, that became highly contextualised.
[201] (2018) 264 CLR 361 (‘Falzon’).
[202] Ibid 377–8 [41]–[43] (Kiefel CJ, Bell, Keane, Nettle and Gordon JJ).
[203] Ibid 379 [45].
[204] Here using the term loosely to cover also privileges and other more procedural decisions affecting the admission of evidence.
[205] UEL (n 6) ss 18(7)(b), 114(3)(b), 126B(4)(b), 130(5)(a), 138(3)(b), 142(2)(a), 169(5)(a), 190(4)(a), 192(2)(c). To presumably similar effect is the consideration whether there is ‘other evidence concerning the matters’ to which the challenged evidence relates: at ss 18(7)(c), 126B(4)(d) (but note that section 126B does not appear in the Cth Evidence Act (n 6), NT Evidence Act (n 6) or Vic Evidence Act (n 6)). It is unclear why two forms of expression are used to cover essentially the same notion, particularly when they sit side-by-side in section 18. The importance of the evidence will also be a consideration in determining whether admitting the evidence will ‘cause or result in undue waste of time’: UEL (n 6) ss 53(3)(c), 135(c).
[206] Kadir (n 7) 135 [42] (Kiefel CJ, Bell, Keane, Nettle and Edelman JJ).
[207] The reasoning is captured by the Latin maxim expressio unius est exclusio alterius (an express reference to one matter indicates that other matters are excluded): Pearce (n 199) 174 [4.43]. Another situation in the criminal trial where courts may assess probative value without regard to importance is in relation to the credibility of the criminal defendant. The defendant’s baseline credibility is invariably low because of the strong interest the defendant has, whether guilty or innocent, in obtaining an acquittal: R v Campbell [2007] EWCA Crim 1472; [2007] 1 WLR 2798, 2808 [30] (Lord Phillips CJ for the Court). Arguably, any further evidence of a defendant’s credibility has slight incremental probative value. Statements that, for example, evidence of a defendant’s prior conviction for corruption is ‘highly probative’ or has ‘substantial probative value’ in relation to the defendant’s credibility (eg, R v El-Azzi [2004] NSWCCA 455, [12] (Santow JA), [189], [200] (Simpson J), [272], [273] (Sperling J)) appear to disregard the importance notion.
[208] [2012] HCA 15; (2012) 247 CLR 170 (‘Aytugrul’).
[209] The arithmetic adopted in the case is imprecise: David Hamer, ‘Expected Frequencies, Exclusion Percentages and “Mathematical Equivalence”: The Probative Value of DNA Evidence in Aytugrul v The Queen’ (2013) 45(3) Australian Journal of Forensic Sciences 271, 272–4 <https://doi.org/10.1080/00450618.2013.790478> (‘Expected Frequencies’).
[210] Aytugrul (n 208) 185–6 [30] (French CJ, Hayne, Crennan and Bell JJ).
[211] Ibid 185 [29] (French CJ, Hayne, Crennan and Bell JJ), 189 [41] (Heydon J).
[212] There was said to be no prejudice because the exclusion percentage was mathematically equivalent to the frequency ratio, and the derivation had been explained: ibid 186 [30], 187 [34] (French CJ, Hayne, Crennan and Bell JJ). Like the majority, Heydon J considered the exclusion percentage equivalent to the frequency ratio: at 198 [65], 203–4 [75]–[76]. Further, psychological literature regarding the dangers of juries misunderstanding statistical evidence had not been properly admitted: at 199 [66], 201–3 [71]–[74] (Heydon J).
[213] [1997] USSC 2; 519 US 172 (1997), discussed in Mnookin (n 1) 1557–63.
[214] Aytugrul (n 208) 190–1 [44] (Heydon J).
[215] Ibid. Is ‘waste’ not inherently ‘undue’? See also Odgers (n 118) 1325 [EA.135.210]. As well as putting more factors into the balance, rule 403 contains the same asymmetry as section 135 in requiring probative value to be ‘substantially outweighed’ for exclusion.
[216] Odgers (n 118) 1342 [EA.137.60]. Odgers expresses uncertainty about the implications of Aytugrul (n 208) for the notion of ‘incremental probative value’: at 1349 [EA.137.90].
[217] Mnookin (n 1) 1560. Here Mnookin considers other available prosecution evidence, not only other admitted prosecution evidence.
[218] Heydon, Cross on Evidence (n 177) 834 [21130]. See also Odgers (n 118) 1325–6 [EA.135.210]. ‘Importance’ may depend, in part, on the extent to which the other evidence leaves the ultimate fact a live issue.
[219] Hamer, ‘Structure and Strength’ (n 44) 187; Donald K Piragoff, Similar Fact Evidence: Probative Value and Prejudice (Carswell, 1981) 149.
[220] If the other evidence is so strong that the issue is no longer live, however, then the challenged evidence will become irrelevant. This aspect of relevance is captured by the ‘importance’ notion but appears inconsistent with holistic probative value: see Hamer, ‘Significant Probative Value’ (n 46) 527.
[221] Criminal Justice Report (n 11) 7, 9.
[222] Ibid pt VI, 639–42. See generally David Hamer, ‘Propensity Evidence Reform after the Royal Commission into Child Sexual Abuse’ (2018) 42(4) Criminal Law Journal 234; Hamer, ‘Mixed Messages’ (n 193).
[223] Criminal Justice Report (n 11) pt VI, 633–4. There is a formal connection between these two findings. Let X represent legitimate probative value and Y represent unfair prejudice or illegitimate probative value, as in Part II(A). If X and Y measure increases in the probability of guilt flowing from the evidence, then X + Y < 1, because the probability scale lies on the unit interval between 0 and 1. If propensity evidence is more probative than traditionally thought, ie, X is greater than previously thought, then there is less room on the unit interval for unfair prejudice to operate, ie, less room on the unit interval for Y: Hamer, ‘Significant Probative Value’ (n 46) 516–18. It should be noted, however, that this argument fails if the likelihood of guilt is measured on the odds scale (which runs from zero to infinity) or the log odds scale (which runs from negative infinity to positive infinity): Hamer, ‘Expected Frequencies’ (n 209) 276–9.
[224] So far, section 97A has been introduced into the ACT Evidence Act (n 6), NSW Evidence Act (n 6), NT Evidence Act (n 6), and Tas Evidence Act (n 6). Due to concerns raised about the limited operation of the reform and its failure to cover adult sexual assault cases which raise similar issues, the NSW legislation was subject to a review after two years: Evidence Amendment (Tendency and Coincidence) Act 2020 (NSW) s 30. The Victorian Government indicated it would wait for the outcome of the NSW Review before settling on the reforms: Victoria, Parliamentary Debates, Legislative Council, 8 March 2022, 585 (Jaclyn Symes, Attorney-General). The NSW Review concluded that it was too early to assess the operation of the provisions and recommended a further two-year review: New South Wales Department of Communities and Justice, Statutory Review Report: Evidence Amendment (Tendency and Coincidence) Act 2020 (Report, September 2022) [4.26]–[4.29].
[225] See, eg, R v Brookman [2021] NSWDC 110; R v Young (a pseudonym) [2021] NSWDC 622; R v IW (2021) 38 DCLR (NSW) 148; R v QX [No 5] [2021] ACTSC 247; (2021) 292 A Crim R 193. Cf Clarke (n 171), where the Court applied section 101 and held that one complainant’s allegations were not admissible in relation to charges relating to two other complainants because the first complainant’s allegations were far more serious than those of the other two complainants.
[226] Hamer, ‘Principled and Practical’ (n 193); Hamer, ‘Mixed Messages’ (n 193).
[227] The amendments adopted in the UEL (n 6) differ from those proposed by the Royal Commission (drafted by the New South Wales Parliamentary Counsel’s Office), but the Royal Commission’s draft legislation also failed to address disjunction issue: Criminal Justice Report (n 11) pt VI, 649, app N, 595 cls 97(1A)(a)–(b), 98(1A)(a)–(b).
[228] UEL (n 6) s 97A(2), but note that section 97A does not appear in the Vic Evidence Act (n 6) or Cth Evidence Act (n 6).
[229] See above nn 199–200 and accompanying text.
[230] Accentuating the distinction between tendency and coincidence evidence appears unjustified: Criminal Justice Report (n 11) pt VI, 594–6 cls 96A, 97(1A), 98(1A)(a), 100A. See also at 642–3; Hamer, ‘Mixed Messages’ (n 193) 245–50.
[231] Criminal Justice Report (n 11) pt VI, 594–5.
[232] [2010] VSCA 88; (2010) 27 VR 146, 151–2 [22] (Maxwell P, Buchanan and Bongiorno JJA).
[233] Criminal Justice Report (n 11) pt VI, 594; Hamer, ‘Structure and Strength’ (n 44) 181–5; Hamer, ‘Similar Fact Reasoning’ (n 44) 628; Hamer, ‘Significant Probative Value’ (n 46) 524–5.
[234] Criminal Justice Report (n 11) pt VI, 594.
[235] See above nn 221–3 and accompanying text; Hamer, ‘Similar Fact Reasoning’ (n 44); Hamer, ‘Significant Probative Value’ (n 46).
[236] Criminal Justice Report (n 11) pt VI, 594.
[237] Ibid 595.
[238] Hughes (n 70) 356 [39] (Kiefel CJ, Bell, Keane and Edelman JJ). See Hamer, ‘Significant Probative Value’ (n 46) 525. In Phillips (n 4), the High Court took a very stringent approach to admissibility in relation to commission and consent: see above nn 43, 183.
[239] UEL (n 6) s 97A(1), but note that section 97A does not appear in the Cth Evidence Act (n 6) or Vic Evidence Act (n 6); Hamer ‘Mixed Messages’ (n 193) 251. The drafting is ambiguous. Given the background of Hughes (n 70) and the Royal Commission report, the most plausible interpretation is that the presumption is restricted to commission cases – where the defendant’s ‘commission ... of an act’ is in issue. But it could be interpreted to extend to identity cases where it is in issue whether the commission was ‘by the defendant’. Perhaps, given the developments in TL (n 12) discussed in the text, section 97A would be interpreted to apply to identity cases. But this would still exclude, for no good reason, cases involving defence arguments of accident and medical necessity: Hamer ‘Mixed Messages’ (n 193) 251. Limiting words also appeared in the Royal Commission’s Draft Provisions: Criminal Justice Report (n 11) app N, 594 cl 96A(1)(a) (‘evidence that shows a propensity of the defendant to commit particular kinds of offences if the commission of an offence of the same or a similar kind is in issue in the proceeding’).
[240] See above nn 30–3 and accompanying text.
[241] Hamer, ‘Structure and Strength’ (n 44) 183–5; Hamer, ‘Significant Probative Value’ (n 46) 525; Hamer, ‘Mixed Messages’ (n 193) 251; Hamer, ‘Similar Fact Reasoning’ (n 44) 628.
[242] Though the High Court did not admit to the error: TL (n 12) 96 [30] (Kiefel CJ, Gageler, Gordon, Steward and Gleeson JJ).
[243] The High Court could have upheld admissibility on the alternative basis that the other alleged misconduct, like the charged offence, was directed by the defendant against the same victim: ibid 98–9 [37]. See also Bauer (n 194) 71–2 [20], 82–3 [48]–[49] (Kiefel CJ, Bell, Gageler, Keane, Nettle, Gordon and Edelman JJ).
[244] TL (n 12) 95 [28] (Kiefel CJ, Gageler, Gordon, Steward and Gleeson JJ).
[245] Ibid 96 [30].
[246] Ibid.
[247] Hamer, ‘Significant Probative Value’ (n 46) 523. In TL (n 12), the High Court directed this charge against the defence for making strong demands of the tendency evidence by itself: at 96 [31] (Kiefel CJ, Gageler, Gordon, Steward and Gleeson JJ). However, as is clear from the discussion in this article, support can be found for a stringent and focused admissibility test.
[248] See generally Jacob E Van Vleet, Informal Logical Fallacies: A Brief Guide (University Press of America, 2011) 1–2; Roy T Cook, A Dictionary of Philosophical Logic (Edinburgh University Press, 2009) ‘fallacy of division’ <https://doi.org/10.1515/9780748631971>.
[249] Hoch (n 155).
[250] Ibid 296 (Mason CJ, Wilson and Gaudron JJ).
[251] As the NSW Court of Criminal Appeal appreciates, ‘although [tendency] evidence ... would not suffice to make out the charge ... it increased the probability of the [charge] ... and thus possessed significant probative value’: Hughes v The Queen [2015] NSWCCA 330; (2015) 93 NSWLR 474, 513–14 [172] (Beazley P, Schmidt and Button JJ).
[252] Redmayne (n 98) 16–17, quoted in Criminal Justice Report (n 11) pt VI, 607. See also Hamer ‘Mixed Messages’ (n 193) 238.
[253] See Edward W Cleary (ed), McCormick’s Handbook of the Law of Evidence (West Publishing, 2nd ed, 1972) 436; Mnookin (n 1) 1543–4; Hamer, ‘Significant Probative Value’ (n 46) 522.
[254] Chamberlain (n 16) 538 (Gibbs CJ and Mason J).
[255] Deane J’s position on this point favoured the prosecution. However, unlike the majority, he would have upheld the defendants’ conviction appeal: ibid 630. See also Murphy J’s dissenting judgment: at 577.
[256] Ibid 627 (Deane J); Shepherd (n 156) 584–5 (Dawson J); David Hamer, ‘The Continuing Saga of the Chamberlain Direction: Untangling the Cables and Chains of Criminal Proof’ [1997] MonashULawRw 5; (1997) 23(1) Monash University Law Review 43, 44–5.
[257] Shepherd (n 156) 579 (Dawson J, Mason CJ agreeing at 575–6, Toohey and Gaudron JJ agreeing at 586).
[258] Ibid 579 (Dawson J); Edwards v The Queen [1993] HCA 63; (1993) 178 CLR 193, 204 (Brennan J), 210 (Deane, Dawson and Gaudron JJ).
[259] See Davidson v The Queen [2009] NSWCCA 150; (2009) 75 NSWLR 150, 165 [75] (Simpson J).
[260] Ligertwood and Edmond (n 46) note that ‘evidence [may be] “essential to the process of reasoning”, as a crucial circumstantial intermediate fact, even where it is not a logically indispensable link in a chain of proof’: at 160, discussing Evidence Act 1929 (SA) s 34R, which may imperfectly capture the Shepherd (n 156) restriction on the Chamberlain principle. See also Velevski v The Queen [2002] HCA 4; (2002) 187 ALR 233, 244–5 [43]–[44]; Stephen Odgers, ‘Editorial: The Burden and Standard of Proof’ (2021) 45(3) Criminal Law Journal 139, 140.
[261] Shepherd (n 157) 591 (McHugh J).
[262] Lindy’s husband Michael Chamberlain also appealed against his conviction as an accessory after the fact. Among other things, the case spawned a best-selling book (John Bryson, Evil Angels (Penguin Books, 1985)), and a Hollywood movie based on the book (Evil Angels (Warner Brothers Pictures, 1988)).
[263] Chamberlain (n 16) 559 (Gibbs CJ and Mason J). The jury could, however, proceed on the basis that the substance was blood. This more modest finding, along with much other expert evidence, was criticised in the Morling Royal Commission in 1987.
[264] Ibid 568 (Gibbs CJ and Mason J).
[265] Gipp (n 149) 115 [21] (Gaudron J), 155–7 [139]–[142] (Kirby J). McHugh and Hayne JJ would have applied the Chamberlain (n 16) principle had the relationship evidence been adduced for a propensity purpose, but considered it inapplicable here because the relationship evidence was only adduced as background: at 132–3 [76]–[79]. Callinan J did not address this issue. See also Penney v The Queen [1998] HCA 51; (1998) 72 ALJR 1316, 1321 [26] (Callinan J), which, without citing Shepherd (n 156), applied the Chamberlain (n 16) principle to proof of motive.
[266] HML (n 37).
[267] Ibid 390 [132]. See also at 406 [200], 416 [244].
[268] Ibid 416–17 [247] (Hayne J), citing at 362 [41] (Gummow J), 371 [63] (Kirby J), 500 [506] (Kiefel J). Gleeson CJ noted that the evidence was relied upon for context and did not involve indispensable links. While wary of applying the Chamberlain (n 16) principle too broadly, he seemed to support it where the evidence is relied upon for motive or propensity: at 360 [29]–[31], 361–2 [37]. Heydon J did not decide whether a direction in terms of the Chamberlain (n 16) principle was required since, if required, it had been provided: at 452–3 [339], 470–1 [395]–[396]. Crennan J agreed with Heydon J on this: at 490 [479]. See also at 491 [483] (Kiefel J). See generally David Hamer, ‘Admissibility and Use of Relationship Evidence in HML v The Queen: One Step Forward, Two Steps Back’ (2008) 32(6) Criminal Law Journal 351 <https://doi.org/10.2139/ssrn.1422206>.
[269] There may be a rationale for the Chamberlain (n 16) principle in some child sexual assault cases. Where there is very little evidence beyond that of the complainant, the complainant’s credibility could be viewed as indispensable. As the majority later observed in IMM (n 19), ‘[i]t is difficult to see that one might reason rationally to conclude that [the complainant’s] account of charged acts of sexual misconduct is truthful because [the complainant] gives an account that on another occasion the accused exhibited sexual interest in him or her’: at 318 [63] (French CJ, Kiefel, Bell and Keane JJ). But in HML (n 37), this rationale was rejected, on the basis that the complainant’s evidence of the other misconduct can be viewed as independent of the complainant’s direct evidence of the charged misconduct: at 402 [183] (Hayne J). See also at 361 [32] (Gleeson CJ), 427 [280] (Heydon J).
[270] HML (n 37) 416 [244] (Hayne J).
[271] Ibid 500 [506] (Kiefel J).
[272] Ibid 416 [244] (Hayne J).
[273] Ibid 405–6 [195] (Hayne J), 500 [506] (Kiefel J). See also at 360 [29]–[31], 362 [37] (Gleeson CJ).
[274] See above n 269.
[275] See also Hamer, ‘Tendency and Coincidence Chapter’ (n 144) 164–5.
[276] [2014] NSWCCA 4, [128]–[130] (Bathurst CJ).
[277] [2016] NSWCCA 174; (2016) 94 NSWLR 1, 17 [92] (Bathurst CJ).
[278] [2018] NSWCCA 164, [93] (Bathurst CJ).
[279] O’Keefe (n 34).
[280] Ibid [5] (Howie J).
[281] Ibid.
[282] Bauer (n 194).
[283] Ibid 98 [86] (Kiefel CJ, Bell, Gageler, Keane, Nettle, Gordon and Edelman JJ). Unusually, this statement was expressly directed to NSW trial judges even though Bauer (n 194) was a Victorian case where the issue had not arisen.
[284] Ibid.
[285] JS v The Queen [2022] NSWCCA 145, [46] (Basten AJA).
[286] The provision is poorly drafted. It may leave open the looser illogical interpretation of essential facts: Criminal Procedure Act 1986 (NSW) s 161A(3). And it suggests the direction may be given where tendency evidence is adduced ‘as proof of an element or essential fact’: at s 161A(2) (emphasis added). Evidence may be adduced as proof of an element without, in any sense, being essential. Courts continue to misapply the law. R v Khorami [2021] NSWDC 760 is a multiple complainant sex offence case with cross-admissibility between the various allegations. Despite the existence of the multiple strands of evidence, the trial judge wrongly viewed them as ‘essential’ and disregarded one as not sufficiently proven: at [68] (Weinstein DCJ).
[287] Mnookin (n 1) 1537. A similar demarcation issue arises for the trial judge on no case submissions, and for the appeal court on conviction appeals: see David Hamer, ‘The Unstable Province of Jury Fact-Finding: Evidence Exclusion, Probative Value and Judicial Restraint after IMM v The Queen’ [2017] MelbULawRw 36; (2017) 41(2) Melbourne University Law Review 689 (‘Unstable Province’).
[288] Jones v The Queen [1997] HCA 56; (1997) 191 CLR 439, 442 (Brennan CJ). See also M v The Queen [1994] HCA 63; (1994) 181 CLR 487, 502 (Brennan J); Pell v The Queen [2020] HCA 12; (2020) 268 CLR 123, 145 [38] (Kiefel CJ, Bell, Gageler, Keane, Nettle, Gordon and Edelman JJ); Hamer, ‘Unstable Province’ (n 287) 716–17.
[289] Alqudsi v The Queen [2016] HCA 24; (2016) 258 CLR 203, 255 [131] (Gageler J) (‘Alqudsi’); Cheng v The Queen [2000] HCA 53; (2000) 203 CLR 248, 290 [123] (McHugh J), citing Patrick Devlin, Trial by Jury (LawBook, 1966) 164.
[290] Alqudsi (n 289) 254 [129] (Gageler J).
[291] MFA v The Queen (2002) 213 CLR 606, 621 [48] (McHugh, Gummow and Kirby JJ).
[292] R v Shamouil (2006) 66 NSWLR 228, 238 [64] (Spigelman CJ) (‘Shamouil’). See also DSJ v The Queen [2012] NSWCCA 9; (2012) 84 NSWLR 758, 761 [10] (Bathurst CJ), 778 [93] (Whealy JA).
[293] IMM (n 19) 315 [50] (French CJ, Kiefel, Bell and Keane JJ).
[294] Ibid 316 [54].
[295] Ibid 315 [51], citing R v XY (2013) 84 NSWLR 363, 400 [167], [170] (Simpson J) (‘XY’). See also JD Heydon, ‘Is the Weight of Evidence Material to Its Admissibility?’ (2014) 26(2) Current Issues in Criminal Justice 219, 236–7 <https://doi.org/10.1080/10345329.2014.12036016> (‘Weight of Evidence’).
[296] IMM (n 19) 312 [39] (French CJ, Kiefel, Bell and Keane JJ). An example may be an eyewitness identification based upon a sighting in poor conditions which should be assessed by the trial judge as ‘weak [and] simply unconvincing’: at 315 [50] (French CJ, Kiefel, Bell and Keane JJ), discussing Heydon, ‘Weight of Evidence’ (n 295) 234.
[297] Bauer (n 194) 92 [69] (Kiefel CJ, Bell, Gageler, Keane, Nettle, Gordon and Edelman JJ). At common law, Hoch (n 155) is authority that the reasonable possibility of collusion is enough to preclude the admission of evidence of other alleged victims: at 297 (Mason CJ, Wilson and Gaudron JJ); Bauer (n 194) 92 [70] (Kiefel CJ, Bell, Gageler, Keane, Nettle, Gordon and Edelman JJ).
[298] Bauer (n 194) 91–2 [69] (Kiefel CJ, Bell, Gageler, Keane, Nettle, Gordon and Edelman JJ).
[299] New South Wales, Parliamentary Debates, Legislative Assembly, 25 February 2020, 1911–18 (Mark Speakman, Attorney-General); Odgers (n 118) 741 [EA.94.150]. For non-UEL (n 6) jurisdictions see: Evidence Act 1977 (Qld) s 132A; Evidence Act 1929 (SA) s 34S(b); Evidence Act 1906 (WA) s 31A(3).
[300] UEL (n 6) ss 65(2)(c)–(d).
[301] Ibid s 85.
[302] Uniform Evidence Law Report (n 192) 239 [8.58].
[303] Ibid.
[304] Ibid. See also at 341 [10.69], [10.71], 345 [10.87], [10.90]; UEL (n 6) s 189(3).
[305] See above nn 106–15 and accompanying text.
[306] Dupas v The Queen [2012] VSCA 328; (2012) 40 VR 182, 242 [226] (Warren CJ, Maxwell P, Nettle, Redlich and Bongiorno JJA), criticising the NSW Court of Criminal Appeal’s constrained approach in Shamouil (n 292). The NSW Court of Criminal Appeal subsequently revisited the issue in XY (n 295) but was divided on which approach was appropriate. See IMM (n 19) 342–3 [151]–[153] (Nettle and Gordon JJ). Cf at 310 [29] (French CJ, Kiefel, Bell and Keane JJ).
[307] IMM (n 19) 344 [156] (Nettle and Gordon JJ), citing XY (n 295) 408 [224] (Price J).
[308] IMM (n 19) 345 [157] (Nettle and Gordon JJ). See also at 323 [88], 325–6 [96] (Gageler J).
[309] Ibid 325 [96] (Gageler J).
[310] Ibid 346 [161] (Nettle and Gordon JJ). See also at 323 [88] (Gageler J).
[311] Melbourne (n 164) [40] (McHugh J). Note that McHugh J considered that the ‘no reasonable view’ test was focused rather than holistic, requiring that there be ‘no reasonable explanation for the disputed evidence other than the accused’s guilt’ (emphasis added).
[312] Ibid.
[313] Phillips (n 4) 323 [63] (Gleeson CJ, Gummow, Kirby, Hayne and Heydon JJ).
[314] In a pre-trial hearing or on a voir dire, though this is not mandatory for tendency and coincidence evidence: UEL (n 6) ss 189(2), (4).
[315] Judicial Commission of New South Wales, ‘Criminal Trial Courts Bench Book’ (Bench Book, October 2002) [7-020].
[316] Falzon (n 201) 379 [45] (Kiefel CJ, Bell, Keane, Nettle and Gordon JJ).
[317] Phillips (n 4) 323 [63] (Gleeson CJ, Gummow, Kirby, Hayne and Heydon JJ).
[318] Falzon (n 201) 379 [45] (Kiefel CJ, Bell, Keane, Nettle and Gordon JJ).
AustLII:
Copyright Policy
|
Disclaimers
|
Privacy Policy
|
Feedback
URL: http://www.austlii.edu.au/au/journals/UNSWLawJl/2024/29.html