AustLII Home | Databases | WorldLII | Search | Feedback

Melbourne University Law Review

Melbourne Law School
You are here:  AustLII >> Databases >> Melbourne University Law Review >> 2017 >> [2017] MelbULawRw 36

Database Search | Name Search | Recent Articles | Noteup | LawCite | Author Info | Download | Help

Hamer, David --- "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


THE UNSTABLE PROVINCE OF JURY
FACT-FINDING: EVIDENCE EXCLUSION,
PROBATIVE VALUE AND JUDICIAL RESTRAINT AFTER IMM V THE QUEEN

DAVID HAMER[*]

Intermediate appeal courts in Victoria and New South Wales recently diverged on the question of how a trial judge should approach the assessment of the probative value of evidence at the admissibility stage. NSW courts consider that trial judges should be wary of usurping the jury’s fact-finding role, while Victorian courts think that trial judge intervention is required to ensure a fair trial. Unfortunately, the High Court in IMM provided little resolution, splitting three ways, with a self-contradictory majority judgment. In an effort to make sense of IMM, this article examines other areas of criminal procedure — directed acquittals and appeals — that also demarcate fact-finding responsibilities between the judiciary and the jury. This broader jurisprudence reveals a range of underlying policies and interests, including efficiency, factual accuracy, and respecting the jury as the constitutional tribunal of fact. The diversity in these policies and their potential for conflict helps explain the unsettled nature of the law. However, appeal jurisprudence on the primary fact-finder’s epistemic advantage points to a reconciliation in the IMM majority’s self-contradiction, based on the distinction between evidence delivery and evidentiary context. Notwithstanding this resolution of the majority judgment, the minority judgments in IMM provide a clearer framework within which the sometimes competing forces and considerations can be balanced.

CONTENTS

I INTRODUCTION

While fact-finding in serious criminal matters is the traditional and constitutional province of the jury, it is subject to trial judge and appeal court regulation. One key aspect of the regulation is the trial judge’s responsibility to exclude certain evidence from the jury’s consideration.[1] This responsibility exists at common law and under the Uniform Evidence Law (‘UEL’).[2] Exclusion on the basis that evidence falls within a technical category, such as hearsay, may appear appropriately ‘legal’, causing little disturbance to the balance of power between judge and jury. However, in instances where evidence is excluded simply because it does not appear to the trial judge to be sufficiently strong or probative, the division of responsibilities between judge and jury are brought into sharp relief.[3]

The New South Wales Court of Criminal Appeal (‘NSWCCA’) has been wary of trial judges being too interventionist in assessing probative value at the admissibility stage, and has laid down tight principles of restraint. In R v Shamouil, Spigelman CJ indicated that ‘[t]o adopt any other approach would be to usurp for a trial judge critical aspects of the traditional role of a jury’.[4] Five judges of the Victorian Court of Appeal (‘VCA’) in Dupas v The Queen, however, described the NSWCCA approach as ‘manifestly wrong’,[5] observing that ‘trial judges should continue to have the power to exclude admissible evidence in order to minimise the risk of wrongful conviction. The effect of the decision in Shamouil is to undermine an important safeguard ... against an unfair trial.’[6] More recently, five judges of the NSWCCA provided an unsettled response in R v XY;[7] however, the Court appeared to maintain its restrictive, non-interventionist approach.[8]

Trial judge versus jury, fair trial versus institutional integrity, VCA versus NSWCCA; in IMM, the High Court of Australia (‘HCA’) had the opportunity to settle a multidimensional dispute.[9] The HCA favoured the NSWCCA’s more restrained approach by a narrow 4:3 majority, while the minority judges, in one respect, favoured even greater trial judge intervention than the VCA

in Dupas.[10] The majority judgment in IMM is elliptical and obscure, appearing to impose tight restrictions on the trial judge while at the same time implying that these do little to impede intervention. All the judgments question whether the difference between the two approaches is as great as may appear.[11]

In this article I attempt to make sense of the HCA’s reasoning in IMM. First, the majority and minority approaches are distinguished, along with other variations, and the seeming contradictions in the majority judgment are highlighted. Next, I place the admissibility issue in a broader context by examining other areas of procedure in which the fact-finding responsibilities of judiciary and jury are demarcated: directed acquittals and appeals on the facts. An understanding of policies and principles of judicial restraint invoked at these other stages provides insights into the conflicting positions regarding admissibility in IMM. The discussion traces the conflicts in IMM to deeper ideological tensions; for example, between protecting the province of the constitutional fact-finder on the one hand, and protecting the accused’s interest in avoiding wrongful conviction on the other. However, the appeal jurisprudence also offers a resolution to the seeming contradiction in the IMM majority judgment. Having regard to the jury’s epistemic advantage, the trial judge exercises considerable restraint in assessing the evidential source while being more willing to intervene with regard to evidential context. While helpful in resolving the contradictions in the majority judgment, this analysis nevertheless indicates that the majority in IMM has left the law in a needlessly complex state. The minority approach provides a cleaner and simpler framework for achieving the law’s policy goals.[12]

II RESTRAINT IN ASSESSING PROBATIVE VALUE AND ADMISSIBILITY ACCORDING TO IMM

The appellant in IMM had been found guilty in the Northern Territory Supreme Court on two counts of indecent dealing with a child and one count of sexual intercourse with a child under 16 years, all perpetrated against his step-granddaughter. According to the prosecution, these incidents occurred in the course of several years of recurrent sexual abuse.

The appellant appealed on the basis that two pieces of evidence had been wrongly admitted. The first challenge concerned tendency evidence from the complainant that the appellant had had sexual contact with her on an occasion not giving rise to charges. The appellant argued this was wrongly admitted as tendency evidence because it did not have ‘significant probative value’ as required by UEL s 97. The second challenge concerned evidence that the complainant had told relatives and a friend of the appellant’s sexual abuse some time prior to the matters being reported to police. This was admitted as hearsay evidence under an exception to the hearsay rule.[13] The appellant, however, argued that it should have been excluded under UEL s 137 on the basis that its ‘probative value is outweighed by the danger of unfair prejudice to the appellant’.[14]

The admissibility of both items of evidence turned upon trial judge assessments of probative value. On such questions, the Northern Territory courts adopted the trial judge restraints of the NSWCCA.[15] The trial judge assessed probative value on the basis that the evidence would be accepted by the jury. In the HCA, the appellant argued that this was incorrect. The trial judge should have taken a more interventionist approach, recognised the low probative value of the tendency and complaint evidence, and excluded both. In a joint judgment, a majority of four upheld the NSWCCA approach and held that the tendency evidence should have been excluded, while confirming the admissibility of the complaint evidence.[16] The minority, in two separate judgments, supported what appears to be a more interventionist approach. However, as discussed further below, the majority is far less restrained than first appears. Gageler J agreed with the majority orders,[17] and Nettle and Gordon JJ upheld the admissibility of the tendency evidence while questioning the admissibility of the complaint evidence.[18]

A Probative Value, Range and Acceptance in the Uniform Evidence Law

Despite the apparent divergence between the restrained majority and interventionist minority, there is an important element of common ground. Interventionists would agree that the UEL requires trial judges to exercise some restraint in assessing probative value. This is made clear in the definition of probative value in the UEL, which is expressed in terms of the capacity of evidence to prove the facts in issue. Probative value is defined as ‘the extent to which the evidence could rationally affect the assessment of the probability of the existence of a fact in issue’.[19] A leading advocate of the interventionist approach and defence counsel in IMM, Stephen Odgers, explains that ‘[t]he trial judge is required to accept the existence of a range of probative value assessments, none of which are unreasonable ... and assess probative value as at the highest point of that range’.[20] The majority and minority judges in IMM appear to agree with this approach.[21]

The point on which the restrained majority and interventionist minority differ is whether, in assessing probative value, the trial judge should assume that evidence will be ‘accepted’ by the jury. The majority held that this assumption should be made, inflating the probative value assessment and limiting the scope of the trial judge to intervene.[22] The minority held that the assumption should not be made, and that an important part of the trial judge’s role is to determine whether the evidence could reasonably be accepted, and to what degree.[23]

Both the majority and the minority supported their conclusions regarding the meaning of ‘probative value’ in the UEL by drawing on the UEL’s closely related concept of relevance.[24] Section 55 defines ‘relevant evidence’ as ‘evidence that, if it were accepted, could rationally affect ... the assessment of the probability of the existence of a fact in issue in the proceeding’. This definition is almost identical to the definition of probative value: ‘the extent to which the evidence could rationally affect the assessment of the probability of the existence of a fact in issue’. Whereas relevance is an all-or-nothing concept, probative value is measured by degrees.[25]

The minority judgments placed emphasis on another difference between the two definitions: ‘The statutory description of relevance requires making an assumption that evidence is reliable; the statutory definition of probative value does not provide for making that assumption.’[26] Their reasoning, in effect, is expressio unius est exclusio alterius: ‘An express mention to one matter indicates that other matters are excluded.’[27] The assumption that evidence is accepted is excluded from the trial judge’s assessment of

probative value.

The majority invoked a different maxim of statutory interpretation. They focused on the fact that the definitions of both relevance and probative value ask whether evidence ‘could rationally affect ... the assessment of ... the existence of a fact in issue’,[28] and found that ‘[t]he same construction must be given to the words ... where they appear in the definition of “probative value” as is given to those words in [the definition of “relevance”]’.[29] The determination of probative value, like the determination of relevance, should be made on the basis that the evidence is accepted. Like the minority, the majority approach is supported by established authority: ‘it is a fundamental rule of construction that ... as far as possible ... the same meaning [should be given] to the same words wherever those words occur in ... an Act of Parliament’.[30]

The minority interpretation is sounder as it takes greater account of the immediate context of the definitions of probative value and relevance.[31] While the same expression is used in the two definitions, it is accompanied by the difference in expression highlighted by the minority. Moreover, as Tim Smith and Stephen Odgers point out, having regard to the different roles played by relevance and the exclusionary principles, ‘there is every reason to view the absence of the words in the definition [of probative value] as intended and significant’.[32] As an ‘opening gate’ it makes sense that the relevance requirement imposes a less demanding test than ‘probative value’ at later admissibility stages, particularly ‘s 137 [which] provides the final critical safety net ... to minimise the risk of wrongful conviction’.[33]

B Four Attitudes to the Assumption that Evidence Is Accepted

The key distinction between the majority and minority in IMM is on the question whether, in assessing probative value, the trial judge should assume that evidence is accepted. Actually, there are at least four different approaches to this assumption. In order of increasing power: (1) no assumption; (2) accept that the evidence is truthful; (3) accept that the evidence is truthful and reliable; (4) accept that the evidence is truthful and reliable, and accept the inference invited from the evidence. In IMM the majority and minority adopted approaches (3) and (1) respectively. In this section I distinguish the four different approaches and explore their apparent points of divergence. However, the differences may not be as great as first appears. As explored in the next section, the majority principles of restraint are heavily, albeit unclearly, qualified.

The IMM minority takes approach (1) and makes no assumption that the evidence will be accepted. In assessing probative value, part of the trial judge’s task is to determine whether the evidence is capable of acceptance.[34] However, this does not mean that the trial judge’s assessment of probative value is totally unrestrained. As discussed above, the trial judge should recognise that there may be a range of reasonable interpretations of the acceptability of the evidence and take the evidence at its highest.

It appears the IMM minority judgments would give the trial judge greater scope for intervention than the VCA in Dupas, which takes approach (2). The Victorian court held that probative value should be assessed on the assumption that the witness has been truthful in providing the evidence.[35] This still allows the trial judge to consider whether the witness was mistaken. The challenged witnesses in Dupas were eyewitnesses who identified the accused as someone they had seen near the time and place of the murder.[36] Their honesty was not challenged, but the defence argued that, due to the lateness of the identifications and the witnesses’ exposure to media images of the accused linking him to the crime in the interim, there was the risk of a ‘displacement effect’ or an otherwise corrupted memory.[37] The VCA held that, while assuming witness truthfulness, these reliability considerations should be considered by the trial judge in determining the probative value of the evidence and its possible exclusion under UEL s 137.[38]

The majority in IMM held that the assessment of probative value should be based upon a stronger assumption. Taking approach (3), the majority held that the trial judge should assume that the evidence is both credible and reliable.[39] The trial judge should proceed on the basis that the jury ‘will ... accept it completely in proof of the facts stated’.[40] This complete-proof principle appears to impose a tight constraint on trial judge intervention. It seems to require the trial judge to give direct evidence, such as an eyewitness identification of the accused as the perpetrator, maximal value. Unlike the IMM minority, the majority would apparently not allow the trial judge to entertain the possibility that the eyewitness was dishonest or mistaken.[41]

The majority and minority diverge sharply over the trial judge’s assumption that evidence is accepted, but both reject the VCA’s distinction between truthfulness or honesty on the one hand, and reliability on the other.[42] The common law drew this distinction, reserving the term ‘credibility’ for witness honesty,[43] but the UEL defines ‘credibility’ to include the reliability of witnesses’ observations and memory:[44] ‘The Evidence Act itself creates a difficulty in separating reliability from credibility.’[45]

While the majority’s approach to evidence acceptance is far stronger than that of the minority in IMM, there is another version of acceptance that operates still more strongly. Approach (4) accepts not only the honesty and reliability of the witness, but also the inference that the prosecution invites from the evidence. As Heydon recently asked: ‘If in all other respects the evidence tendered by the prosecution is to be taken at its highest from the prosecution’s point of view, why should not [the] available inferences from it be taken at their highest as well?’[46] While some judgments of the NSWCCA lend support to this version of the assumption,[47] others reject it. For example, in XY, Hoeben CJ at CL suggested that ‘[w]hen assessing the probative

value of the prosecution evidence sought to be excluded ... a court can take into account the fact of competing inferences which might be available on

the evidence’.[48]

On its face, this strongest version of acceptance would give all evidence maximal force. According to the majority approach, direct evidence appears to acquire maximal force, but not circumstantial evidence.[49] Circumstantial facts are taken to be as narrated by the witness, but it is the role of the trial judge to then assess the extent to which the material facts may be inferred from those facts. A witness’s identification of the accused as the victim’s killer appears immune from admissibility challenge, but not so a witness’s identification of the accused as the person who threatened to kill the victim. A trial judge should weigh up the extent to which the accused’s making of the threat supports the prosecution case that the accused is the killer.

The majority’s treatment of the complaint hearsay evidence in IMM provides a more subtle illustration of their treatment of circumstantial evidence. Accepting the honesty and reliability of each complaint witness, the trial judge should proceed on the basis that the complaints were made at the time and in the circumstances related by the witnesses. However, the trial judge should then consider what inferences can be drawn from these complaints and whether they lead to the conclusion that the conduct complained of occurred.[50] In other words, the complaint witnesses’ evidence is accepted, but the trial judge should still assess the credibility and reliability of the complainant’s out-of-court complaints. In the majority’s view, ‘[g]iven the content of the evidence, the evident distress of the complainant in making the complaint and the timing of the earlier complaint, it cannot be said that its probative value was low’.[51] The majority agreed with the lower courts that the risk of prejudice was minimal and that the evidence need not have been excluded under s 137.[52]

C Applications, Contradictions and Explanations

The extreme force of the strongest approach to the assumption of acceptance seems inconsistent with the terms of the UEL. If all evidence is given maximal force, there would be no point in s 97 requiring the trial judge to determine whether tendency evidence has ‘significant probative value’.[53] Furthermore, there would be little room, under s 137, for probative value to be outweighed by the risk of unfair prejudice.[54] To a degree, it appears that this criticism extends to the majority’s complete-proof principle. The seeming breadth and strength of this principle threatens to enfeeble the trial judge’s powers of exclusion. Appreciating this, the majority does not present the principle as wholly unyielding and inflexible, and introduces certain qualifications.

The first qualification is relatively well defined. The majority refer to ‘some limited circumstances in which credibility will be taken into account’.[55] ‘[E]vidence which is inherently incredible or fanciful or preposterous’ would be subject to exclusion as irrelevant or under the general discretion.[56] While only a ‘limiting case’,[57] this qualification is difficult to square with the complete-proof principle. It should be noted that the majority would not reject this evidence on the basis that, under the definitions of relevance and probative value, it ‘could [not] rationally affect’ the probability of a fact in issue.[58] According to the majority, the trial judge’s acceptance of evidence is logically prior to, and not subject to, the consideration of its ‘rational’ effect.[59] But despite its difficulties, given that this first qualification is relatively well defined and of limited scope, it will not be considered further here.

The majority introduce a further qualification to their strong version of acceptance that is poorly defined, broader in scope, and altogether more problematic and puzzling.[60] Two paragraphs before stating their complete-proof principle, the majority briefly discuss an example, borrowed from Heydon: ‘an identification made very briefly in foggy conditions and in bad light by a witness who did not know the person identified’.[61] At the HCA appeal hearing, the parties thought that the majority would assess this evidence as having high probative value.[62] However, the majority described it as ‘an identification, but a weak one because it is simply unconvincing’.[63] In this situation, at least, the majority takes the view that a trial judge should give direct evidence far less than maximal probative value. The majority’s reasoning is unclear, but this qualification to the complete-proof principle appears to bring the majority’s approach much closer to that of the minority.[64]

How to reconcile the majority’s response to Heydon’s example with its complete-proof principle is a puzzle. The only explanation provided by the majority for its reasoning is that

the basis upon which a trial judge proceeds, that the jury will accept the evidence taken at its highest, does not distort a finding as to the real probative value of the evidence. The circumstances surrounding the evidence may indicate that its highest level is not very high at all.[65]

Odgers offers a reconciliation: on the majority approach, the identification should not be taken at face value as an unqualified positive identification. Instead,

the evidence may be seen as evidence of an opinion (‘in my opinion, the accused person is the offender’). Accordingly, it is to be assumed that the witness is being truthful when he or she testifies that this opinion is held and is reliably recounting the content of the opinion (thus, probative value may not be assessed on the basis that the witness actually holds a different opinion). This does not mean that the opinion itself must be assumed to be reliable. Other evidence, including ‘the circumstances surrounding the evidence’ of the witness, may indicate that it has low probative value.[66]

This rationalisation, while clever, is difficult to square with the majority judgment.[67] Odgers distinguishes two applications of the term ‘reliability’: first, reliability in recounting an opinion, and second, the reliability of the opinion. Odgers’ suggestion that the majority would have the trial judge accept the reliability of the evidence in the first sense, but not the second sense, is inconsistent with the tenor of the decision. Reliability in the first sense seems essentially the same as credibility or truthfulness. In IMM the term ‘reliability’ is used in the second sense.[68] The distinction Odgers draws is contradicted by the majority’s declaration that ‘[t]here can be no disaggregation of the two — reliability and credibility ... They are both subsumed in the jury’s acceptance of the evidence.’[69] Odgers’ rationalisation fails to dissolve the tension between the majority’s treatment of Heydon’s example and their complete-proof principle.

There is a further situation in which the majority appear to breach their own complete-proof principle — the tendency evidence provided by the complainant in IMM. The majority indicated that ‘it is difficult to see how a complainant’s evidence of conduct of a sexual kind from an occasion other than the charged acts can be regarded as having the requisite degree of probative value’.[70] Tendency evidence is circumstantial evidence rather than direct evidence and accordingly the majority’s complete-proof principle would not necessarily give the tendency evidence maximal value. However, it is clear that the majority’s doubts concerned the complainant’s truthfulness rather than any later step in the tendency inference:

Evidence from a complainant adduced to show an accused’s sexual interest can generally have limited, if any, capacity to rationally affect the probability that the complainant’s account of the charged offences is true. It is difficult to see that one might reason rationally to conclude that X’s account of charged acts of sexual misconduct is truthful because X gives an account that on another occasion the accused exhibited sexual interest in him or her.[71]

This reasoning of the majority, like the majority’s treatment of Heydon’s identification example, is puzzling. The majority’s analysis ‘replicates the very thing the majority’s statements of principle disavows, namely taking into account the reliability or credibility of the complainant’s evidence for the purposes of admissibility’.[72] If the complete-proof principle were applied, the trial judge would accept that the other sexual misconduct had occurred just as the complainant testified — as though the court were shown a video recording — and the concern regarding non-independence would not arise.

The majority’s treatment of the tendency evidence is not susceptible to the explanation Odgers provided for Heydon’s identification example. That explanation has the effect of assuming truthfulness and leaving reliability open to challenge, whereas here it is the witness’s truthfulness that is

being challenged. Odgers provides a different explication of the majority’s reasoning:

[I]t is important to focus carefully on the nature of the fact(s) in issue to which the evidence is relevant and whether the evidence may have significance or importance in establishing that fact or those facts ... [In IMM] the applicable ‘fact in issue’ was whether the complainant’s account of the commission of the offences was both truthful and reliable. When assessing the capacity of the tendency evidence to increase the probability that this account was credible, the fact that it came from the complainant was of critical importance in determining whether the evidence had significant probative value.[73]

However, this analysis makes no real effort to reconcile the majority’s discounting of the complainant’s truthfulness with the majority’s complete-proof principle. Instead, it carves out an exception to the complete-proof principle by reference to the nature of the facts in issue. On this view, the

trial judge may bypass the complete-proof principle where the witness’s credibility is of ‘critical importance’. The majority’s tight restraint on trial judge intervention does not apply to the most decisive interventions. This appears nonsensical.

Odgers’ rationalisations of these two problem cases — Heydon’s identification example and the tendency evidence in IMM — fail to cohere with each other and with other aspects of the majority judgment. This highlights the incoherency and opacity of the majority judgment. These problems undermine the authority of the majority judgment and reinforce the impression that

the HCA in IMM rehearses, rather than resolves, the conflicts between the

state courts.

III RESTRAINT IN DIRECTING ACQUITTALS

The NSWCCA and the VCA have taken different views regarding the level of restraint trial judges should exercise in excluding evidence on the basis of low probative value. Unfortunately, the HCA in IMM, rather than bringing much-needed clarity, has brought greater confusion. At this point it is worth drawing on other areas of adjectival law where the courts have considered the division of fact-finding responsibility between the primary fact-finder and the court. This part considers directed acquittals, and the next part examines criminal and civil appeals on the ground of factual error. In these two areas, as with admissibility, the (trial or appeal) court is required to judge the strength of incriminating evidence, potentially invading the province of the primary fact-finder. Extending the survey into these adjoining areas provides landmarks and guide-posts that clarify the contours of the admissibility principles.

On first impression, directed acquittals may appear to be a particularly useful source for understanding evidence exclusion since, in both areas, the trial judge pre-empts the jury’s determination of the strength of prosecution evidence (as opposed to an appeal court reviewing the primary fact-finder’s assessment). Directed acquittals, like admissibility determinations, present ‘the problem ... of striking a balance between, on the one hand, usurpation by the judge of the jury’s function, and on the other the danger of an unjust conviction’,[74] and authorities on directed acquittals prescribe trial judge restraint in very similar language to that used by admissibility authorities. It is not a matter of the trial judge ‘form[ing] the view [whether] ... he [or she] would entertain a reasonable doubt as to the guilt of the accused’.[75] Rather, ‘[i]t is the capacity of the evidence to lead to a conclusion of guilt that is to be the focus of attention’:[76] ‘the Crown case must be taken at its highest’.[77]

Despite these close similarities in expression, there are significant differences in the judicial tasks being performed. A directed acquittal is generally a far greater intervention than excluding a piece of evidence. In effect, a directed acquittal removes all evidence from the jury’s consideration; the jury is entirely displaced by the trial judge. By contrast, to exclude a piece of evidence from the jury’s consideration does not dictate the verdict; the jury remains the primary fact-finder. As the discussion below shows, recent Australian authorities suggest that the greater intervention — directing an acquittal — is more tightly restrained.

A Witness Truthfulness, Circumstantial Evidence and Conflicting Evidence

Interestingly, two leading state authorities on admissibility, the NSWCCA decision in Shamouil and the VCA decision in Dupas, although conflicting, both claim support from the leading Australian directed acquittal authority, Doney v The Queen.[78]

In Shamouil, Spigelman CJ recognised that Doney concerned a ‘different, but not irrelevant, context’.[79] However, Spigelman CJ suggested the connection in some cases could be strong. To exclude ‘evidence of critical significance ... would, in substance, be equivalent to directing a verdict of acquittal’.[80] And for the trial judge to take this step too readily ‘would be to usurp for a trial judge critical aspects of the traditional role of a jury’.[81] Trial judges should be tightly constrained in excluding evidence due to low probative value, as they are in directing acquittals.

In Dupas, the VCA suggested that Shamouil could not be ‘reconcile[d] ... with what was said in Doney about the traditional functions of the trial judge’.[82] Instead, the VCA claimed that Doney supported its approach, preserving only witness truthfulness as an exclusive jury issue and otherwise leaving probative value open to trial judge assessment.[83] But, in respect of directed acquittals at least, Doney does not support this level of trial judge intervention. The HCA in Doney expressed respect, not only for the jury’s ability to determine ‘whether, and in the case of conflict, what evidence is truthful’;[84] but also ‘the determination of factual matters’[85] more broadly, rejecting a more interventionist line of English authority.[86]

According to the current Australian approach, trial judges should be very restrained in directing acquittals. The trial judge should proceed on the basis of evidence and inferences favouring the prosecution,[87] ‘without regard to evidence which favours the accused as, for example, by contradicting, qualifying, or explaining the first-mentioned evidence in support of a conviction’.[88] The trial judge should not only assume that the prosecution evidence is accurate, but should make ‘the further assumption that all inferences most favourable to the prosecution which are reasonably open, are drawn’:[89] ‘the test of sufficiency is the same whether the prosecution evidence be direct or circumstantial’.[90] In terms of the four attitudes to acceptance of the prosecution’s evidence outlined above,[91] this aligns with the strongest approach, not the second weakest as the VCA suggested in Dupas.

However, contrary to Spigelman CJ’s suggestion in Shamouil, directed acquittal authorities do not necessarily support a similarly restrained approach at the admissibility stage. For example, Asche suggests that a trial judge should not direct an acquittal where a prosecution case rests upon ‘identification by a fleeting glance ... sworn to by a witness’,[92] but that it would be ‘open to the trial judge to have rejected the identification evidence’.[93] Strangely, perhaps, if exclusion of the identification evidence ‘left no evidence to go to the jury ... the trial judge could then have properly directed an acquittal on that basis’.[94] This appears paradoxical but the exclusion of evidence will not generally provide a way of bypassing the tighter restraints on directed acquittals. In most cases the exclusion of a single item of evidence will be a far less dramatic and decisive intervention than directing an acquittal. For this reason, looser restraints may be appropriate at admissibility.

B The Scintilla Principle

One extreme expression of the strength of judicial restraint operating at the directed acquittal stage is the scintilla principle.[95] According to the pure version of the principle, ‘if there were but a scintilla of evidence’ of guilt there is no scope for a directed acquittal.[96] On this view, the trial judge only considers ‘the existence of evidence’[97] and ‘whether there is evidence of each of the elements necessary to prove a conviction’.[98] The trial judge has no role in assessing its weight or sufficiency.[99]

It seems that the pure scintilla principle has few supporters. Courts have recognised that they do have some role in weighing evidence. An acquittal may be directed on the basis that ‘even if all the evidence for the prosecution were accepted and all inferences most favourable to the prosecution which are reasonably open were drawn, a reasonable mind could not reach a conclusion of guilt beyond reasonable doubt’.[100] The prosecution must present more than ‘such a faint scintilla that reasonable men could not act upon it’.[101]

The unworkability of the pure scintilla principle may provide a lesson for the complete-proof principle of the majority in IMM. The IMM majority hold that a witness must be assumed to be credible and reliable, and the witness’s evidence accepted as complete proof of the facts stated. This approach, in its absoluteness, resembles the pure scintilla principle, although within a narrower range. The pure scintilla principle would give all evidence maximal force, while the IMM majority’s complete-proof principle would only give direct evidence maximal force (the power to direct acquittals, as a stronger intervention than the exclusion of evidence, is subject to greater restraint). The difficulty in maintaining such an absolute approach to trial judge restraint is evident in both areas. Trial judges, in determining whether to direct an acquittal, should ask not only whether a scintilla of evidence exists but also whether it could satisfy a reasonable jury. And, in IMM, the majority qualified their own complete-proof principle. They recognised limiting cases in which trial judges may hold far-fetched or preposterous evidence to be irrelevant. Further, there are problem cases where the majority appear prepared to allow exclusion on the basis of credibility or reliability, seemingly in breach of their own complete-proof principle. Arguably the complete-proof principle should be recognised as unsustainable. As a matter of principle, trial judges, in determining probative value at the admissibility stage, should be permitted

to assess the extent to which the jury could reasonably accept the truth

and reliability of witness testimony. This, in effect, is the minority approach

in IMM.

IV PRINCIPLES OF APPELLATE RESTRAINT

Appeals on the facts,[102] like directed acquittals and the exclusion of evidence due to low probative value, raise questions regarding the division of responsibility between the primary fact-finder and a judicial overseer. Civil and criminal appeal courts have considered the appropriate level of restraint that they should apply. In broad outline, the approach on appeal resembles that taken in the directed acquittal and admissibility areas discussed above.

The common form criminal appeal legislation provides that a conviction should be overturned if it is ‘unreasonable, or cannot be supported, having regard to the evidence’.[103] According to the ‘authoritative guidance’ provided by the majority in M v The Queen,[104] the question this raises for the criminal appeal court is whether it was ‘open to the jury’ to be satisfied of the accused’s guilt beyond reasonable doubt.[105] While the court should ‘mak[e] its own independent assessment of the evidence’,[106] the court should be restrained: ‘the appellate court does not substitute its assessment of the significance and weight of the evidence for the assessment which the jury, properly appreciating its function, was entitled to make’.[107] The court is concerned with how a reasonable jury might view the evidence.[108] Further, ‘the boundaries of reasonableness within which the jury’s function is to be performed should not be narrowed in a hard and fast way’.[109]

These broad expressions of appellate restraint resemble statements of trial judge restraint in the context of directed acquittals and admissibility determinations. However, as was apparent in the comparison between directed acquittals and admissibility determinations, different levels of restraint may lie behind these surface similarities. Regard must be had to the nature and context of the different tasks.[110] Admissibility determinations and directed acquittals are similar in that, in both situations, trial judges pre-empt the jury’s consideration of evidence. However, as discussed above, stronger principles of restraint operate on the more drastic intervention — the

directed acquittal.

Overturning a conviction on appeal seems a comparable intervention to directing an acquittal. The successful appeal (on the facts) rejects the jury’s view of the evidence, while the directed acquittal deprives the jury the opportunity to hear it in the first place. There is little to choose between

them — and yet there is clear authority that appeal courts need not exercise the level of restraint exercised by a trial judge in directing an appeal. An appeal court should overturn a conviction where it appears to the court unsafe and unsatisfactory. However, a trial judge may not direct an acquittal on the basis that the trial judge considers a conviction would be unsafe and unsatisfactory.[111] Despite strong authority for this distinction, support is not universal,[112] and its supporters have not always provided persuasive justifications for it.

In Antoun, Gleeson CJ merely asserted that there are differences between directing an acquittal and finding a conviction to be unreasonable, and that these should be kept in mind to avoid confusion.[113] Gleeson CJ failed to say what the differences actually are. In Doney, the HCA suggested that, unlike the trial judge’s power to direct an acquittal, the appeal court’s power to overturn a conviction ‘is supervisory in nature ... [and] does not involve an interference with the traditional division of functions between judge and jury in a criminal trial’.[114] However, as Asche observes, ‘[i]f an appellate court’s supervisory powers extend to setting aside the verdict of a jury based on

the facts presented to it, it is difficult to see how this does not “involve an interference ...”’[115]

Glass recognises that while ‘the usurpation of the jury’s function of weighing evidence is contrary to accepted principle governing jury trials’, that proposition has been ‘eroded’ by the appeal court’s power to declare a conviction unsafe.[116] Glass distinguishes directed acquittals on the basis that they are a matter of law, whereas overturning unsafe or dangerous convictions is a matter of fact.[117] However, the application of these labels is conclusory. Glass suggests that ‘[t]he mark of a pure question of law is that it is decided by reference to legal materials, not to the evidence’.[118] However, directed acquittals do not satisfy this substantive criterion; acquittals are directed on the basis of evidence, not legal materials.[119] Directed acquittals are labelled questions of law for the tautological reason that they are determined by ‘that part of the court which alone has power to decide questions of law properly so called’ — the trial judge.[120]

These explanations for weaker judicial restraint on appeal than in directing acquittals are unpersuasive. Another fairly obvious argument for the distinction is that the appeal court, positioned higher in the judicial hierarchy and usually composed of three or more judges rather than one, is better equipped for intervention than a trial judge.[121] Against this, however, the trial judge will generally have a more complete appreciation of the evidence than the appeal court.[122] Perhaps the determining factor is simply that it is appropriate that an appeal court, operating as a final safety net, should have greater discretion and flexibility. It may appear premature for an acquittal to be directed without even hearing the defence evidence.[123]

The jurisprudence regarding civil and criminal appellate restraint is extensive. It reveals a range of epistemic and non-epistemic considerations operating on appeal courts as they determine whether to overturn a primary fact-finder’s verdict. While an appeal decision, in nature and context, differs substantially from an admissibility decision, the rich appellate jurisprudence offers valuable insights into the array of sometimes conflicting, policies

and interests underlying the High Court’s difficult decision on admissibility in IMM.

A The Primary Fact-Finder’s Epistemic Advantages

Primary fact-finders are recognised as having various epistemic advantages over appeal courts. Criminal appeal courts traditionally acknowledge ‘the jury’s advantage in seeing and hearing the evidence’.[124] This has led to a general reluctance to overturn convictions, particularly ones that may flow from demeanour-based credibility assessments.[125] Appeal courts have been more prepared to intervene ‘where the evidence lacks credibility for reasons which are not explained by the manner in which it was given’.[126] Examples include where ‘the evidence, upon the record itself, contains discrepancies, displays inadequacies, is tainted or otherwise lacks probative force’.[127] There are also cases which do not turn on witness credibility at all: ‘circumstantial case[s] resting on inferences equally able to be assessed by [the appeal court] as by the jury’.[128]

Unlike criminal appeals, civil appeals are rehearings.[129] However, proceeding on the basis of the record of trial evidence, civil appeal courts recognise their ‘natural limitations’[130] and, like criminal courts, have traditionally been reluctant to overturn trial court findings,[131] particularly those based on witnesses’ delivery of evidence.[132] Civil appeal courts, like criminal appeal courts, have been more prepared to intervene ‘where the conclusion of

the primary judge depends on inferences drawn from undisputed facts

or facts that have been found but can equally be redetermined by the appellate court, without relevant disadvantage’.[133] In determining whether to make

this intervention, civil appeal courts are better placed than criminal appeal courts in that trial judges explain their reasoning; jury findings, by contrast,

are ‘inscrutable’.[134]

This traditional deference to the primary fact-finders is not without its detractors. One recent line of civil appeal authority notes ‘a shift to some degree from the more extreme judicial statements commanding deference to the findings of primary judges said to be based on credibility assessments’.[135] This ‘important change’[136] is attributed, in large part, to ‘a growing understanding, both by trial judges and appellate courts, of the fallibility of judicial evaluation of credibility from the appearance and demeanour of witnesses in the somewhat artificial and sometimes stressful circumstances of the courtroom’.[137] On this view, trial and appeal judges increasingly seek to base their conclusions on ‘contemporary materials, objectively established facts and the apparent logic of events’ rather than ‘the appearances of witnesses’.[138] Appeal courts have been prepared to overturn credibility-based verdicts by reference to ‘incontrovertible facts or uncontested testimony’,[139] or simply on the basis that the verdict is ‘“glaringly improbable” or “contrary to compelling inferences” in the case’.[140] Clearly, on this approach, appeal courts have far greater liberty in weighing competing evidence and inferences against each other than trial judges at the directed acquittal stage who should limit their attention to evidence and inferences favouring conviction.[141]

Recent authorities have ‘not eliminated the necessity for appellate courts to give weight to the primary decision-maker’s advantages’,[142] and the value of demeanour-based findings of credibility can still be decisive.[143] In a 2016 decision, the HCA overturned a decision of the Queensland Court of Appeal (‘QCA’), reinstating the trial judge’s findings.[144] The HCA suggested that the meaning of an admission ‘depends as much on the way it is stated as on its content; and, in this case, the judge had the significant advantage of seeing and hearing [the witness] make the admission. The [QCA] majority appear to have overlooked that advantage’.[145] In another recent HCA decision, Callinan and Heydon JJ suggested that even where a trial judge has ‘made no express reference to ... demeanour’ of witnesses in resolving a conflict of evidence, the ‘subtle influence of demeanour’ is such that it may not be appropriate to overturn the trial findings.[146]

The civil appeal jurisprudence recognises a further epistemic reason for respecting the primary fact-finder’s determinations. As well as having viewed witness demeanour, ‘the trial judge ... has advantages that derive from the obligation at trial to receive and consider the entirety of the evidence and the opportunity, normally over a longer interval, to reflect upon that evidence and to draw conclusions from it, viewed as a whole’.[147] The appeal court is generally exposed only to a subset of the trial evidence. The extent of the primary fact-finder’s advantage in this respect will depend upon the quantity and complexity of the trial evidence, and the extent to which it has been covered in the appeal documentation and hearing. Appeal courts’ doubts about evidential completeness may make them reluctant to question the primary fact-finder’s inferences in general, not only those relating to demeanour and credibility.

To the extent that the jury is recognised as possessing genuine epistemic advantages, appeal court deference to the jury will be viewed as furthering factual accuracy. However, there will always be tension where the appeal court disagrees with the jury verdict. Notwithstanding the jury’s advantages, and that the appeal court should only intervene where the jury steps outside the bounds of reasonableness, an appeal court may hesitate in upholding a conviction that appears to be unproven. The appeal court’s obligation to avoid the ‘searing injustice and consequential social injury’[148] of a wrongful conviction may lead it to declare that ‘[i]t is the reasonable doubt in the mind of the court which is the operative factor ... If the court has a doubt, a reasonable jury should be of a like mind.’[149]

B Constitutional and Institutional Considerations

Beyond the supposed epistemic advantages of primary fact-finders having been exposed to the evidence in its entirety, including witness demeanour, appeal courts also defer to the primary fact-finder for broader constitutional and institutional reasons.

In criminal proceedings, the jury is the ‘constitutional tribunal for deciding issues of fact’.[150] It is a ‘little parliament’[151] which serves to ‘ensure a measure of democratic participation, and therefore democratic legitimacy’[152] in the administration of the criminal law. The democratic nature of the jury carries its own epistemic advantage: ‘[t]he strength of 12 jurors as a tribunal of fact derives also from their diversity and their opportunity to deliberate as a group in private throughout the trial’.[153] This is particularly important for factual issues requiring ‘the application of objective community standards, including ... an issue of reasonableness, negligence, indecency, obscenity or dangerousness’.[154]

However, quite apart from its epistemic functionality, the jury’s democratic composition contributes political and institutional functionality. In recognising ‘the political right’[155] of the community to play a role in criminal justice, it helps ‘ensure[] the application of the law consonant with the community conscience’.[156] Moreover, it ‘necessitates in practice that the trial be “comprehensible by both the accused and the general public and have the appearance, as well as the substance, of being impartial and just”’.[157]

Democratic and constitutional arguments for appellate restraint are obviously limited to jury trials which are predominantly criminal. The ‘legal system ... accords special respect and legitimacy to jury verdicts deciding contested factual questions concerning the guilt of the accused in serious criminal trials’.[158] Jury findings may receive ‘some slightly greater deference’ than those of trial judges.[159] However, there are further institutional reasons to ‘value the autonomy of the trial process’ and recognise the finality of trial judge findings of fact.[160] As a matter of effective resource allocation, appeals should be exceptional. The trial court is the primary decision-maker. It would strain resources for trial decisions to be routinely subject to review.[161] In addition, appeals are disruptive; in both criminal and civil spheres, they deny parties and victims closure and the ability to move on with their emotional, social and commercial lives.[162] ‘[I]n the interest of peace, certainty and security’ the law may set strict limits upon the challenges that may be made to the trial verdict.[163] After all, scepticism about the original decision may infect the appeal decision, spread to any further review, and pose the threat of an ‘infinite regress’ of second-guessing.[164] ‘[T]here are cases where the certainty of justice prevails over the possibility of truth ... and these are cases where the law insists on finality’:[165] ‘[t]he line must be drawn somewhere, and the least arbitrary place to draw it is under the original trial verdict’.[166]

These non-epistemic considerations — democratic involvement, efficiency, and finality — must be kept in proportion. While their influence cannot be ignored, they will often give way to the ‘overriding’[167] goal of factual accuracy. Avoiding wrongful convictions is a paramount goal of the criminal law.[168] From this perspective it has been argued that the courts should be concerned less about loving truth too much, and more with ‘loving finality too much’.[169] The accused enjoys a widely recognised right to have a conviction reviewed by a higher court.[170] Like many features of criminal procedure, this right is asymmetric. Traditionally, the accused has also enjoyed protection against the

double jeopardy of a prosecution appeal against an acquittal.[171] Such asymmetries recognise that the accused has more at stake and fewer resources than

the prosecution.[172]

V FROM APPEALS TO ADMISSIBILITY DECISIONS

The jurisprudence regarding judicial restraint at the appeal stage sheds some light on the principles applying at the admissibility stage, confusingly handled in IMM. This is not to promise a light-bulb moment. The shift between the different stages of criminal procedure is not straightforward, and the principles as they operate on appeals, while clearer than those expounded in IMM, are not perfectly structured and settled.

A Finality, Efficiency, Democracy and Wrongful Conviction

The desire for finality and its associated benefits is key to appeal jurisprudence. The bearing of this goal on evidence exclusion is less clear. In adding an extra stage to the trial, the challenge and exclusion of evidence may be seen as deferring finality and closure (while directed acquittals may be viewed as bringing a premature and unsatisfying finality and closure). Against this, evidence exclusion does sometimes promise efficiency. Section 135 of the UEL, a generalised relation of s 137 under consideration in IMM, indicates that evidence may be excluded where its probative value is outweighed by the risk that admission will be ‘misleading or confusing’, or ‘result in undue waste of time’. These same concerns attend the collateral issues raised by tendency evidence,[173] such as that under consideration in IMM. And yet, as endlessly proliferating appeals demonstrate, exclusionary rules and discretions themselves entail complexity and the consumption of considerable resources.[174] Sometimes it may appear preferable for the trial judge to stand back, let the evidence in, and let the jury to get on with it.[175]

Recognition of the jury’s important political and constitutional role

appears to carry a less ambiguous message: this role should not be diminished through the trial judge’s exclusion of evidence. Yet the jury’s fact-finding role is not exclusive — appeal courts can (and do) overturn the jury’s factual determinations.[176] The democratic function of the jury and the desire for finality can be outweighed by the accused’s right to review and the concern to avoid wrongful conviction. Arguably, the desire to ensure a fair and accurate trial may more readily lead to the exclusion of evidence, which is a less drastic intervention than overturning a conviction and raises less concern regarding intrusion into the jury’s domain. Against this, however, is the fact that evidence exclusion (like a directed acquittal) is an earlier intervention and may appear premature. Better to just allow the jury to consider the evidence. If the trial ends in conviction and this appears problematic, the safety net of the appeal can be relied upon.

Consideration of these competing policies, interests and rights is instructive but inconclusive. There is an irremovable ideological element in the search for a balance between the competing societal and individual interests. Inevitably, some judges and commentators will value the jury’s democratic value more highly, supporting greater trial judge restraint. Others will prioritise the accused’s right to avoid or correct wrongful conviction, advocating greater trial judge intervention. Spigelman CJ in Shamouil and the IMM majority may fall within the former camp, while the VCA in Dupas and the IMM minority may fall into the latter camp. The competing values may also explain the internal tensions within the majority judgment. The operation of these broad forces explains the persistence of these conflicting views without offering a resolution.

B The Jury’s Epistemic Advantages

Examination of the epistemic advantages of the primary fact-finder may provide clearer demarcation of the trial judge’s and jury’s respective fact-finding responsibilities. The existence and implications of such advantages raise questions that appear less ideological and more immediately connected with legal doctrine. Notwithstanding the sometimes conflicting views that have been expressed about the primary fact-finder’s epistemic advantages, this perspective may still shed light on the confusing majority judgment

in IMM.

The primary fact-finder is traditionally considered to have an advantage over the appeal court as a result of having seen and heard witnesses deliver evidence. To the extent that this advantage is enjoyed, the appeal court

should respect the primary fact-finder’s credibility assessments based on demeanour. However, an appeal court will intervene more readily based on its assessment of the objective features of the evidential record, unchallenged evidence, and inferences from circumstantial evidence: in some cases, it may overturn a credibility-based finding which appears outweighed by these other evidential features.[177]

The question arises whether the jury enjoys a similar epistemic advantage over the trial judge.[178] If so, this may serve to delineate the contours of trial judge restraint in excluding evidence. But in making the shift from appeal to admissibility stage, certain differences between the two stages need to be examined. First, it is easier for a trial judge to see and hear witnesses deliver testimony at the admissibility stage than it is for an appeal court. The challenged witness could be called during the voir dire, enabling demeanour to enter the admissibility determination. The parallel drawn between the trial judge and the appeal court would then be lost.

There are, however, practical and principled objections to conducting such elaborate voir dires. Heydon warns of an ‘increase in the length of trials and the delays of litigation’,[179] adding that it is ‘irritating to jurors whom it renders idle, destructive of trial rhythm’ and could even have an impact ‘on the survival of trial by jury itself’.[180] In IMM, Nettle and Gordon JJ, taking a more interventionist line, downplay such concerns, suggesting ‘such procedures were commonplace under the common law [and] ... were not productive of insurmountable or ordinarily undue difficulties’.[181] However, Heydon’s views resonate with contemporary concerns over efficiency and jury engagement.[182] To the extent that voir dires involving witness examination-in-chief and cross-examination are limited or undesirable, the jury will retain an epistemic advantage, lending support to the parallel between the trial judge and the appeal court.

A second key difference between an appeal on the facts and an admissibility determination needs to be taken into account in determining the limits of the jury’s epistemic advantage: an appeal court is concerned with the proof provided by an entire body of evidence, whereas the admissibility determination focuses on the probative value of a challenged item of evidence. In determining whether to overturn a trial verdict, an appeal court may be required to weigh demeanour-based credibility assessments against a wide body of unchallenged and circumstantial evidence. A trial judge, in weighing the probative value of a challenged item of evidence, has a narrower focus. The trial judge need only consider other evidence that sheds light on the connection between the challenged evidence and the fact in issue.

In having to deal with a narrower evidential context, the trial judge’s task may be easier than an appeal court’s. As discussed above, appeal courts recognise the primary fact-finder’s advantage in having been exposed to the evidence in its entirety and may be reluctant to second-guess primary factual findings by reference to the broader evidential record. Nevertheless, despite the narrower evidential context, trial judges face a similar difficulty. As the majority noted in IMM, ‘the evidence will usually be tendered before the full picture can be seen’.[183] It may be difficult, at this point, to assess ‘its credibility or reliability [which] will depend not only on its place in the evidence as a whole, but on an assessment of witnesses after examination and cross-examination and after weighing the account of each witness against each other’.[184] There are clear limits to the extent to which this context could be presented during the voir dire.[185] However, in some cases the trial judge will have enough evidential context to discount witness credibility or reliability.

C The Distinction between Evidential Source and Objective
Evidential Context in IMM

The epistemic advantages of primary fact-finders lead appeal courts to exercise restraint in questioning the primary factual findings. The same considerations may lead the trial judge to exercise restraint at the admissibility stage. At both points, the court may lack the primary fact-finder’s opportunity to view the source of the evidence presented. However, the appeal court and trial judge may still feel able to intervene having regard to the features of the broader evidential context.

The distinction between an assessment of evidence based on its source and an assessment based upon its place in the evidential record offers a resolution to the puzzle at the heart of the IMM majority judgment — the apparent contradiction between the majority’s complete-proof principle and its suggestion that Heydon’s identification evidence and the challenged tendency evidence have low probative value. The trial judge should avoid any assessment of the evidence based on the source of the evidence and its delivery. As Richard Lancaster suggests, the trial judge assessment is ‘undertaken with the [witness] out of view’.[186] Initially, the source of the challenged evidence should be ‘accepted’ as completely credible and reliable. However, that initial assessment may be discounted by the trial judge by reference to the broader evidential context.

Consider the majority’s treatment of direct evidence. In Heydon’s example, the eyewitness’s identification is initially accepted at face value as complete proof that the accused is the perpetrator. However, this assessment is then reconsidered having regard to the ‘[t]he circumstances surrounding the evidence’: it was ‘made very briefly in foggy conditions and in bad light by a witness who did not know the person identified’.[187] The objective features of the evidential context call the identification into question. As Lancaster observes, ‘quite apart from the truthfulness, eyesight, attention span, memory or ability to report of the particular witness making the identification ... the identification has a lower probative value than an identification made in

good conditions’.[188]

Something similar seems to be occurring with the majority’s assessment of the tendency evidence. The potential weakness with the evidence is not apparent on the face of the evidence; it emerges once it is placed in the context of the prosecution case. The probative value assessment is discounted because the tendency evidence carries the same potential weakness as the direct evidence at the heart of the prosecution case. Both consist of the complainant’s allegations of sexual contact by the accused. To attribute significant probative value to the evidence, enabling it to bolster the direct evidence, would be akin to self-corroboration or ‘bootstrap reasoning’.[189] ‘[T]he requisite degree of probative value is more likely to be met’, the majority explain, ‘[i]n cases where there is evidence from a source independent of the complainant’.[190]

This contextual approach may also lead to discounting probative value on the ground of redundancy — the challenged evidence replicating other evidence already admitted. As with non-independent evidence, redundant evidence ‘may have considerable probative value in respect of some fact when considered in isolation, [but it] adds little to what has already been admitted and ... its incremental relevance is minimal’.[191]

As such, the distinction between demeanour and the evidential record, or between the evidential source and the evidential context, may explain the IMM majority’s puzzling proposition that a trial judge should accept the witness’s honesty and reliability while ultimately doubting the accuracy of the witness’s evidence. On this view, the majority’s complete-proof principle has only provisional operation.

While providing a rationalisation for the seeming contradictions in the majority judgment, this account does not provide them with a justification or excuse. The majority could have explained itself far better. Moreover, the distinction between source and context can be accommodated within the IMM minority’s simpler framework. The IMM minority makes no initial assumption that evidence is credible and reliable. Instead, they simply accord the evidence the highest value that could reasonably be attributed to it. The jury’s evidentiary advantage in making source-based credibility assessments would tend to inflate the trial judge’s probative value assessment, while any apparent weaknesses in the evidential context would discount such an assessment. On the minority view, these are simply factors to be weighed in the balance (amongst others). The minority provide a more straightforward and realistic representation of the trial judge’s task than the majority’s provisional and qualified complete-proof principle.

VI CONCLUSION

IMM provided the HCA the opportunity to settle the question of how a trial judge should assess probative value at the admissibility stage. The NSWCCA had taken a pro-admissibility approach to this question, while the VCA favoured stronger trial judge regulation. Unfortunately, the HCA provided little resolution. The HCA split 4:3 with the majority supporting the NSWCCA while the minority apparently favour greater trial judge intervention than the VCA. More problematically, the majority judgment is confusing and seemingly self-contradictory. The trial judge should assume the challenged evidence is truthful and reliable and take it as complete proof of

the facts stated, but then, in certain situations, for unstated reasons, the challenged evidence may be taken to provide only weak support for the

facts stated.

In seeking clarification of this unsettled area, this article draws on directed acquittals and appeals jurisprudence regarding the division of responsibility between primary fact-finder and court. This broader jurisprudence suggests that the persistent difficulties with evidence exclusion are the product of complex, sometimes conflicting, policy goals and interests. It may be more efficient for the trial judge to take a relatively hands-off approach to admissibility, particularly if the trial judge is no better placed to assess probative value than the jury. However, this would be an abdication of the trial judge’s responsibilities to ensure a fair and accurate trial. For the trial judge to exclude evidence too readily may infringe the right of the jury, as community representatives, to participate in criminal justice. But for the trial judge to allow evidence of slight probative value to unduly sway the jury would fail to adequately respect the accused’s interest in avoiding wrongful conviction.

While abstract concepts like efficiency, accuracy, fair trial, community and individual rights may shed light on why the trial judge’s approach to probative value is such a fraught issue, they provide little precise guidance on matters like the self-contradictory majority judgment in IMM. At this point it is helpful to draw on appellate jurisprudence regarding the primary fact-finder’s epistemic advantage, and the distinction between demeanour-based assessments and assessments based on the objective features of the evidential record. The trial judge should leave assessment of the quality of the evidential source to the jury. However, the trial judge may be better able to assess weaknesses in the evidence, with reference to its context and the role it plays in the prosecution case.

In the case of circumstantial evidence, even if a witness’s account is accepted as complete proof of the facts stated, the strength of the inference to the ultimate facts in issue remains open. In the instance of direct eyewitness evidence, while the trial judge may raise no initial questions regarding the witness’s powers of observation, memory or honesty, the trial judge may still discount the identification because of poor visibility conditions, or the length of time between the witness’s observations and the identification of the accused. Certain items of prosecution evidence may, in themselves, carry considerable probative potential, but, viewed in context, add little to the prosecution case. The evidence may lack independence or be redundant.

This resolves the majority’s contradictory statements in IMM. The majority’s complete-proof principle only has provisional operation to source-based assessments. These assessments are then subject to contextual discounts. But it is unfortunate that the majority failed to make their reasoning clearer. In several respects the minority approach is preferable. The trial judge should consider the probative value that could reasonably be attributed to the evidence and take the evidence at its highest. This is a simpler structure within which relevant considerations can be brought to account, including any epistemic advantages of the jury. This approach avoids the unworkable absoluteness of the majority’s complete-proof principle, and is more consistent with the language of the UEL.


[*] BSc, LLB (Hons) (ANU), PhD (Melb); Professor, The University of Sydney Law School. The author is grateful for feedback from Geraint Howells and other colleagues at the Law School of City University, Hong Kong, Gary Edmond, Mehera San Roque, Stephen Odgers and Jeremy Gans.

[1] Other aspects include regulating the questions asked of witnesses (see, eg, Uniform Evidence Law s 46 (‘UEL’)), and guidance on how the jury should use the evidence (see, eg, at s 165) and determine whether it provides sufficient proof (see, eg, at ss 140–1).

[2] See, eg, Evidence Act 1995 (NSW) ch 3; Evidence Act 2008 (Vic) ch 3. On the jurisdictions with very similar legislation, see Stephen Odgers, Uniform Evidence Law (Lawbook, 12th ed, 2016) 1–4 (‘Uniform Evidence Law (12th ed)’); Jeremy Gans, ‘The Uniform Evidence Law in the Islands’ in Andrew Roberts and Jeremy Gans (eds), Critical Perspectives on the Uniform Evidence Law (Federation Press, 2017) 13.

[3] In reality, the exclusion of evidence on any ground can call this division of responsibility into question. Consider the broader opposition in evidence scholarship between free-proof Benthamites (see, eg, Larry Laudan, Truth, Error, and Criminal Law: An Essay in Legal Epistemology (Cambridge University Press, 2008) 18–19) and those favouring greater, more protective, regulation (see, eg, HL Ho, A Philosophy of Evidence Law: Justice in the Search for Truth (Oxford University Press, 2008) 49).

[4] (2006) 66 NSWLR 228, 238 [64].

[5] [2012] VSCA 328; (2012) 40 VR 182, 196 [63].

[6] Ibid 242 [226].

[7] (2013) 84 NSWLR 363.

[8] Extracting a ratio is not straightforward. A majority supported Shamouil: ibid 381 [66]–[67] (Basten JA), 385 [86]–[87] (Hoeben CJ at CL), 398 [162] (Simpson J); JD Heydon, ‘Is the Weight of Evidence Material to Its Admissibility?’ (2014) 26 Current Issues in Criminal Justice 219, 233; Stephen Odgers, Uniform Evidence Law (Lawbook, 11th ed, 2014) 856 (‘Uniform Evidence Law (11th ed)’). See also IMM v The Queen [2016] HCA 14; (2016) 257 CLR 300, 343 [152]–[153] (Nettle and Gordon JJ). However, Hoeben CJ at CL adopted a weaker version of Shamouil, allowing the trial judge to dispute the prosecution’s interpretation of the evidence: XY (n 7) 385–6 [88]–[91]; BC v R [2015] NSWCCA 327; (2015) 257 A Crim R 340, 367–8 [52], 369 [55] (Adams J).

Hoeben CJ at CL joined Blanch J and Price J in the majority, holding the evidence inadmissible: XY (n 7) 386 [92], 406 [208], 409 [225]–[226]. Basten JA and Simpson J dissented: XY

(n 7) 384 [82], 401 [178]. Price J, in particular, expressed support for Dupas: XY (n 7)

408 [224].

[9] IMM (n 8).

[10] See Parts II(A), (B).

[11] IMM (n 8) 323–4 [88]–[93] (Gageler J), 343–4 [154] (Nettle and Gordon JJ); see also at

314–15 [50] (French CJ, Kiefel, Bell and Keane JJ).

[12] The two minority judgments took the same broad approach, while differing on points of detail and in the ultimate conclusions: see nn 16–17 and accompanying text.

[13] See UEL (n 1) s 66.

[14] See ibid s 137. The foregoing summary of fact is adapted from IMM (n 8) 330–4 [115]–[125].

[15] See IMM v The Queen [2014] NTCCA 20 [46]–[48]; IMM (n 8) 309 [26].

[16] IMM (n 8) 315 [52], 320 [75].

[17] Ibid 330 [112].

[18] Ibid 353 [181], 354 [186].

[19] UEL (n 1) Dictionary pt 1 (definition of ‘probative value’) (emphasis added).

[20] Odgers, Uniform Evidence Law (11th ed) (n 8) 856 n 115.

[21] IMM (n 8) 313 [44] (French CJ, Kiefel, Bell and Keane JJ), 323 [90] (Gageler J). Unlike the others, Nettle and Gordon JJ did not use the term ‘highest’ but referred to ‘an assessment of the probative value which it would be open to a jury rationally to attribute to the evidence’: at 346–7 [162]. See also Dupas (n 5) 196–7 [63]; XY (n 7) 376 [46] (Basten JA). Tim Smith and Stephen Odgers, however, suggest the use of the term ‘at its highest’ in this context is ‘perhaps unfortunate’, because it is drawn from a different context — no-case submissions: Tim Smith and Stephen Odgers, ‘Determining “Probative Value” for the Purposes of Section 137 in the Uniform Evidence Law’ (2010) 34 Criminal Law Journal 292, 300. On the shift between these procedural stages, see Part III.

[22] IMM (n 8) 314 [49].

[23] Ibid 325–6 [96] (Gageler J), 337 [140] (Nettle and Gordon JJ).

[24] Ibid 312–15 [37]–[50] (French CJ, Kiefel, Bell and Keane JJ), 321 [79] (Gageler J), 336–7

[137]–[140] (Nettle and Gordon JJ).

[25] IMM (n 8) 313 [43] (French CJ, Kiefel, Bell and Keane JJ). French CJ, Kiefel, Bell and

Keane JJ considered this did not affect the interpretation of the words common to both definitions: at 314 [48].

[26] Ibid 325–6 [96] (Gageler J); see also at 337 [140] (Nettle and Gordon JJ).

[27] Dennis C Pearce and Robert S Geddes, Statutory Interpretation in Australia (LexisNexis

Butterworths, 8th ed, 2014) 178. See, eg, Salemi v MacKellar [No 2] [1977] HCA 26; (1977) 137 CLR 396, 402 (Barwick CJ).

[28] In the definition of relevance, ‘(directly or indirectly)’ has been omitted from the quote: see UEL (n 1) s 55; IMM (n 8) 314 [49]. This does not appear in the definition of probative value.

[29] IMM (n 8) 314 [49].

[30] Craig, Williamson Pty Ltd v Barrowcliff [1915] VicLawRp 66; [1915] VLR 450, 452, quoted in Pearce and Geddes

(n 27) 151.

[31] On the importance of context in statutory interpretation, see, eg: K & S Lake City Freighters Pty Ltd v Gordon & Gotch Ltd [1985] HCA 48; (1985) 157 CLR 309, 315, quoted in RS Geddes, ‘Purpose and Context in Statutory Interpretation’ (2005) 2(1) University of New England Law Journal

5, 18–19.

[32] Smith and Odgers (n 21) 296.

[33] Ibid 296, 304. There is a counterargument to the effect that the later admissibility provisions should not be allowed to undo the intended effect of the prior admissibility provisions: Papakosmas v The Queen (1999) 196 CLR 297, 325–6 [93] (McHugh J); Odgers, Uniform Evidence Law (12th ed) (n 2) 1157. This perspective may narrow the operation of the ‘safety net’: Em v The Queen [2007] HCA 46; (2007) 232 CLR 67, 104 [109] (Gummow and Hayne JJ). This counter-argument qualifies rather than negates Smith and Odgers’s point.

[34] IMM (n 8) 337 [140] (Nettle and Gordon JJ).

[35] Dupas (n 5) 196–7 [63].

[36] Ibid 187 [5].

[37] Ibid 250 [245]–[247]; see also at 220 [144].

[38] Ibid 249 [240]–[241]. The Court of Appeal ultimately agreed with the trial judge that the evidence was admissible.

[39] IMM (n 8) 314 [48], 315 [52].

[40] Ibid 315 [52].

[41] As discussed below, however, the majority qualify the complete principle so that direct evidence may be viewed by a trial judge as lacking probative value: see nn 63–4 and accopanying text.

[42] IMM (n 8) 315–16 [52]–[53] (French CJ, Kiefel, Bell and Keane JJ), 347–8 [163]–[165] (Nettle and Gordon JJ).

[43] Ibid 330 [114] (Nettle and Gordon JJ).

[44] UEL (n 1) Dictionary pt 1 (definition of ‘credibility’).

[45] IMM (n 8) 315 [53] (French CJ, Kiefel, Bell and Keane JJ); see also at 347–8 [164] (Nettle

and Gordon JJ).

[46] Heydon (n 8) 234.

[47] R v Sood [2007] NSWCCA 214, [38] (Latham J); see also at [40]. See R v SJRC [2007] NSWCCA 142, [38]–[39] (James J); XY (n 7) 381–2 [66]–[68] (Basten JA), 391 [122] (Simpson J); Odgers, Uniform Evidence Law (12th ed) (n 2) 1186–7.

[48] XY (n 7) 385 [88]. See also DSJ v The Queen [2012] NSWCCA 9; (2012) 84 NSWLR 758, 761 [10] (Bathurst CJ), 775 [78] (Whealy JA); Odgers, Uniform Evidence Law (12th ed) (n 2) 1187.

[49] This is an interpretation of the majority’s approach. The majority’s description of the difference between direct and circumstantial evidence is puzzling: IMM (n 8) 313–14 [45]. The High Court makes more sense of circumstantial evidence in Shepherd v The Queen [1990] HCA 56; (1990) 170 CLR 573, 579. See also Deane J’s dissent in Chamberlain v The Queen [1984] HCA 7; (1984) 153 CLR 521, 627.

[50] Nettle and Gordon JJ failed to appreciate this distinction between hearsay testimony and the out-of-court statement, and that acceptance of the hearsay evidence is not inconsistent with interrogation of the reliability of the out-of-court statement: see IMM (n 8) 337–8 [141]. See also The Law Reform Commission, Evidence (Report No 38, 1987) 80–1 [146]; Smith and Odgers (n 21) 298; Heydon (n 8) 235.

[51] IMM (n 8) 320 [73].

[52] Ibid 320 [74]; see also at 329–30 [110]–[111] (Gageler J). Nettle and Gordon JJ, dissenting, considered that had the trial judge not assumed that the evidence would be accepted, the evidence may have been excluded: at 353–4 [183]–[186].

[53] UEL (n 1) s 97(1)(b).

[54] One of the dominant forms of unfair prejudice is the jury giving evidence more weight than it deserves: Festa v The Queen (2001) 208 CLR 593, 609–10 [51] (McHugh J). This would not be possible if the evidence had maximal probative value: R v Handy [2002] 2 SCR 908, 945 [96] (Binnie J); David Hamer, ‘The Legal Structure of Propensity Evidence’ (2016) 20 International Journal of Evidence and Proof 136, 156–8.

[55] IMM (n 8) 316–17 [57], citing: R v Cook [2004] NSWCCA 52, [43] (Simpson J); Shamouil

(n 4) 236–7 [56]. See also IMM (n 8) 312 [39] (French CJ, Kiefel, Bell and Keane JJ).

[56] IMM (n 8) 317 [58]. See also UEL (n 1) ss 56, 135.

[57] IMM (n 8) 312 [39].

[58] Ibid 312 [38]–[39]; see also at 317 [58].

[59] ‘The reference to its “rational” effect does not invite consideration of its veracity or the weight which might be accorded to it’: ibid 312 [38]. For a strong critique of the HCA’s treatment of ‘rationality’ in IMM, see Andrew Roberts, ‘Probative Value, Reliability and Rationality’ in Andrew Roberts and Jeremy Gans (eds), Critical Perspectives on Uniform Evidence Law (Federation Press, 2017) 63.

[60] It seems clear that there are two distinct qualifications. The majority view the first as

primarily raising an issue of relevance rather than probative value: IMM (n 8) 312 [39],

317 [58]. However, the second goes to ‘a finding as to the real probative value of the evidence’: at 314–15 [50].

[61] Ibid 315 [50]; Heydon (n 8) 234.

[62] IMM (n 8) 324 [92] (Gageler J).

[63] Ibid 315 [50]; see also at 324 [92] (Gageler J).

[64] See n 11 and accompanying text.

[65] IMM (n 8) 314–15 [50].

[66] Odgers, Uniform Evidence Law (12th ed) (n 2) 1184–5.

[67] I put to one side any issues regarding the application of the opinion exclusionary rule: UEL (n 1) s 76. See also Richard Lancaster, ‘IMM v The Queen: A Response from Richard Lancaster SC’ [2016] (Winter) Bar News 40, 42.

[68] See, eg, IMM (n 8) 330 [114] (Nettle and Gordon JJ).

[69] Ibid 315 [52].

[70] Ibid 318 [62].

[71] Ibid 318 [63]. Despite the apparent strength and breadth of these observations, it is not clear that this would stand as a general principle. In HML v The Queen various judgments suggested that that the complainant’s other misconduct evidence may be distinguished sufficiently from the complainant’s direct evidence for it to provide some independent support: [2008] HCA 16; (2008) 235 CLR 334, 360–1 [32] (Gleeson CJ), 402 [182]–[183] (Hayne J), 427 [280] (Heydon J). See also IMM (n 8) 351 [176] (Nettle and Gordon JJ); David Hamer, ‘Admissibility and Use of Relationship Evidence in HML v The Queen: One Step Forward, Two Steps Back’ (2008) 32 Criminal Law Journal 351, 366; Lancaster (n 67) 43.

[72] Lancaster (n 67) 43.

[73] Odgers, Uniform Evidence Law (12th ed) (n 2) 671, citing IMM (n 8) 318 [62]–[63]

(French CJ, Kiefel, Bell and Keane JJ).

[74] R v R (1989) 18 NSWLR 74, 77 (Gleeson CJ), quoted in R v JMR (1991) 57 A Crim R 39,

42–3 (Lee CJ at CL). For an example of the two considerations on either side of Gleeson CJ’s balance, compare nn 4, 6.

[75] A-G’s Reference [No 1 of 1983] [1983] VicRp 101; [1983] 2 VR 410, 415. See also May v O’Sullivan [1955] HCA 38; (1955) 92 CLR 654, 658–9; R v PL [2012] NSWCCA 31, [33], [36] (Bathurst CJ).

[76] DPP (Cth) v Bradley [2009] ACTCA 5; (2009) 3 ACTLR 159, 167 [33] (emphasis added).

[77] PL (n 75) [27], [32] (Bathurst CJ), citing JMR (n 74) 44 (Lee CJ at CL).

[78] [1990] HCA 51; (1990) 171 CLR 207, cited in: Shamouil (n 4) 238 [64] (Spigelman CJ); Dupas (n 5) 233

[190]–[191].

[79] Shamouil (n 4) 238 [64].

[80] Ibid.

[81] Ibid.

[82] Dupas (n 5) 231 [189].

[83] Ibid 233 [190]–[191].

[84] Doney (n 78) 214.

[85] Ibid.

[86] Ibid 213–15, referring, for example, to R v Mansfield [1977] 1 WLR 1102, 1106–7. On Mansfield, see also Justice HH Glass, ‘The Insufficiency of Evidence to Raise a Case to Answer’ (1981) 55 Australian Law Journal 842, 844; Justice Austin Asche, ‘The Trial Judge, the Appeal Court and the Unsafe Verdict’ (1991) 15 Criminal Law Journal 416, 417–18.

[87] R v R (n 74) 81 (Gleeson CJ), citing Haw Tua Tau v Public Prosecutor [1982] AC 136, 151, both quoted in JMR (n 74) 42–3 (Lee CJ at CL).

[88] PL (n 75) [32] (Bathurst CJ), citing R v R (n 74) 81 (Gleeson CJ). In Doney, the HCA rejected the suggestion of the English Court of Appeal in R v Galbraith [1981] 1 WLR 1039 that an acquittal may be directed on the basis of a trial judge’s view of ‘inconsistent evidence’: Doney (n 78) 213–14. Earlier Australian authorities, like their English counterparts, were more interventionist: Glass (n 86) 845.

[89] Bradley (n 76) 166–7 [30], quoting R v Bilick and Starke (1984) 36 SASR 321, 337 (King CJ). See also Glass (n 86) 845, 851.

[90] Glass (n 86) 852. Glass’s views were accepted in R v R (n 74) 81 and confirmed in Doney

(n 78) 213.

[91] See Part II(B).

[92] Asche (n 86) 419. But see R v Turnbull [1977] QB 224, 229–30 (Lord Widgery CJ); Mezzo v The Queen [1986] 1 SCR 802, 818–21 (Wilson J), both discussed in R v R (n 74) 76–81

(Gleeson CJ).

[93] Asche (n 86) 419. Cf the IMM majority discussion of Heydon’s example: IMM (n 8)

314–15 [50]. See also Dupas (n 5) 202–12 [82]–[115].

[94] Asche (n 86) 419.

[95] ERS, ‘The Scintilla Rule of Evidence’ (1919) 18 Michigan Law Review 46.

[96] Doney (n 78) 213. On this point the Doney judgment appears to contain a typographical error — the negative is missing, and the opposite is stated. But from the context, it appears clear that the intended meaning is as presented in the text above. The HCA in Doney offered some support for the scintilla principle, at least to the extent of rejecting the ‘more robust’ alternative, which would permit a directed acquittal on the basis of the trial judge’s view

that a conviction would be unsafe and unsatisfactory: Doney (n 78) 214–15. See also Asche (n 86) 419.

[97] R v R (n 74) 84.

[98] Antoun v The Queen [2006] HCA 2; (2006) 224 ALR 51, 77 [86] (Callinan J). Note that Callinan J also expresses the view that the trial judge should be permitted to take a more robust approach:

at 77 [86] n 74.

[99] It follows that on the pure scintilla principle, the applicable standard of proof is irrelevant. Those who attribute significance to the standard of proof, must be taken as rejecting the pure principle: see, eg, Glass (n 86) 846, 852. See generally Richard Glover, ‘Codifying the Law on Evidential Burdens’ (2008) 72 Journal of Criminal Law 305, 310–11.

[100] Questions of Law Reserved on Acquittal [No 2 of 1993] [1993] SASC 4152; (1993) 61 SASR 1, 5 (King CJ), quoted in Bradley (n 76) 167 [32].

[101] R v Crooks [1944] NSWStRp 24; (1944) 44 SR (NSW) 390, 393 (Jordan CJ). See also R v R (n 74) 84–5

(Gleeson CJ); Ronald J Allen, ‘Structuring Jury Decisionmaking in Criminal Cases: A Unified Constitutional Approach to Evidentiary Devices’ (1980) 94 Harvard Law Review

321, 328.

[102] Appeals on matters of law are a different proposition. Where there has been an error of law, such as admitting inadmissible evidence, the appeal court is not assessing the strength of the same body of evidence as that assessed by the jury. See David Hamer, ‘Wrongful Convictions, Appeals, and the Finality Principle: The Need for a Criminal Cases Review Commission’ [2014] UNSWLawJl 12; (2014) 37 University of New South Wales Law Journal 270, 282–3.

[103] See, eg, Criminal Appeal Act 1912 (NSW) s 6(1).

[104] R v Baden-Clay (2016) 258 CLR 308, 330 [66], quoting M v The Queen [1994] HCA 63; (1994) 181 CLR 487, 494–5, where Mason CJ, Deane, Dawson and Toohey JJ noted that previously the test had been ‘variously expressed’.

[105] M (n 104) 493. Gaudron J also endorses this formulation: at 508; see also Brennan J, though with a little qualification: at 501–2. An alternative approach is to ask whether the jury

was ‘bound to have a reasonable doubt’: Chidiac v The Queen [1991] HCA 4; (1991) 171 CLR 432, 451 (Dawson J). While the latter is sometimes said to be more stringent, it is difficult to see much difference: R v Vjestica [2008] VSCA 47; (2008) 182 A Crim R 350, 369 [62] (Maxwell P).

[106] SKA v The Queen [2011] HCA 13; (2011) 243 CLR 400, 408 [20] (French CJ, Gummow and Kiefel JJ). See also Carr v The Queen [1988] HCA 47; (1988) 165 CLR 314, 331 (Brennan J), quoting Chamberlain v The Queen [No 2] [1984] HCA 7; (1984) 153 CLR 521, 534 (Gibbs CJ and Mason J).

[107] Carr (n 106) 331.

[108] See, eg, Knight v The Queen [1992] HCA 56; (1992) 175 CLR 495, 511 (Brennan and Gaudron JJ), quoting Chidiac (n 105) 451 (Dawson J).

[109] Baden-Clay (n 104) 329 [65].

[110] Including the language of the applicable appeal legislation: see, eg, Fleming v The Queen (1998) 197 CLR 250, 256 [12]; CSR Ltd v Della Maddalena [2006] HCA 1; (2006) 224 ALR 1, 6–7 [13]–[15], 8 [19] (Kirby J); Baini v The Queen [2012] HCA 59; (2012) 246 CLR 469, 476 [13].

[111] R v R (n 74) 76, 84; Doney (n 78) 213, 215; JMR (n 74) 43–4, quoting A-G’s Reference [No 1 of 1983] (n 75) 415–16; Questions of Law Reserved on Acquittal [No 2 of 1993] (n 100) 5

(King CJ); Antoun (n 98) 55 [16] (Gleeson CJ), 77 [86] (Callinan J).

[112] See n 88. See also Callinan J’s suggestion in Antoun that judges should be able to be more robust in directing acquittals: n 98.

[113] Antoun (n 98) 55 [16].

[114] Doney (n 78) 215.

[115] Asche (n 86) 421.

[116] Glass (n 86) 845.

[117] Ibid 843–4.

[118] Ibid 852.

[119] Doney (n 78) 214–15.

[120] Glass (n 86) 852. On the terminological distinction between fact and law, see Ronald J Allen and Michael S Pardo, ‘The Myth of the Law–Fact Distinction’ (2003) 97 Northwestern University Law Review 1769.

[121] Cf n 142.

[122] See n 147.

[123] The idea of a ‘safety net’ to protect against miscarriages of justice has also been raised in statutory construction of the UEL to justify greater judicial intervention at later stages of proceedings: see n 33.

[124] M (n 104) 494.

[125] See, eg, Singh v The Queen [2011] NSWCCA 100, [131]–[139].

[126] M (n 104) 494.

[127] Ibid.

[128] Hawi v The Queen [2014] NSWCCA 83, [483] (McCallum J).

[129] See, eg, Supreme Court Act 1970 (NSW) s 75A, discussed in Fox v Percy (2003) 214 CLR 118, 124–5 [20]–[22].

[130] Dearman v Dearman [1908] HCA 84; (1908) 7 CLR 549, 561 (Isaacs J).

[131] Coghlan v Cumberland [1898] UKLawRpCh 74; [1898] 1 Ch 704, 705; Justice Virginia Bell, ‘Appellate Review of the Facts’ (2014) 39 Australian Bar Review 132, 133.

[132] CSR (n 110) 7 [17] (Kirby J). See generally Justice Peter McClellan, ‘Who Is Telling the Truth? Psychology, Common Sense and the Law’ (2006) 80 Australian Law Journal 655, 657.

[133] CSR (n 110) 8–9 [22] (Kirby J). See also Warren v Coombes [1979] HCA 9; (1979) 142 CLR 531, 551

(Gibbs ACJ, Jacobs and Murphy JJ).

[134] Warren (n 133) 552, quoting Ward v James [1966] 1 QB 273, 301 (Lord Denning MR).

[135] CSR (n 110) 8 [19] (Kirby J), citing Fox (n 129).

[136] CSR (n 110) 8 [19] (Kirby J).

[137] State Rail Authority of NSW v Earthline Constructions Pty Ltd (in liq) [1999] HCA 3; (1999) 160 ALR 588, 617 [88] (Kirby J) (‘Earthline’). There is insufficient space here to examine the relevant literature in detail. Briefly, it appears that ‘there is evidence in the face and voice that someone is lying, particularly in high-stakes lies in which the liar faces benefits for successful lying and punishments for unsuccessful lying’: Mark G Frank and Paul Ekman, ‘Nonverbal Detection of Deception in Forensic Contexts’ in William T O’Donohue and Eric R Levensky (eds), Handbook of Forensic Psychology: Resource for Mental Health and Legal Professionals (Elsevier Academic Press, 2004) 635, 643. The problem is, however, that while signs of deception may be there, ‘laypeople are pretty poor at detecting lies’ and ‘most people involved in a trial, such as most expert clinicians, trial attorneys, and judges, are ... no better than laypeople’: at 644, 645. Added to this, there are various ‘ways in which the legal system works against a judge

or jury’s abilities to infer deception from a witness’, such as delay before the witness’s account is heard, the witness’s opportunity for rehearsal, and the fact that, in the courtroom situation, truthful witnesses may be just as anxious about being believed as lying witnesses:

at 646–7. Further research suggests ‘demeanor does not play a significant role in real-world deception detection’: Max Minzner, ‘Detecting Lies Using Demeanor, Bias and Context’ (2008) 29 Cardozo Law Review 2557, 2567. Instead, ‘the context surrounding the speaker’s statement does appear to matter. When observers have background information about the witness’s statement, they use it and lie detection improves considerably’. See also Frank and Ekman (n 137) 648.

[138] Fox (n 129) 129 [31] (Gleeson CJ, Gummow and Kirby JJ). See also Justice David Ipp, ‘Problems with Fact-Finding’ (2006) 80 Australian Law Journal 667, 672.

[139] Fox (n 129) 128 [28] (Gleeson CJ, Gummow and Kirby JJ).

[140] Ibid 128 [29], quoting: Brunskill v Sovereign Marine & General Insurance Co Ltd [1985] HCA 61; (1985) 62 ALR 53, 57; Chambers v Jobling (1986) 7 NSWLR 1, 10. Extrajudicially, David Ipp suggests that ‘the probabilities, together with external and internal consistency, should always be the touchstone of factual findings’ and that he ‘would consign demeanour to the bottom of the list’: Ipp (n 138) 672. However, a little inconsistently, Ipp has also expressed faith in judges’ ability to discern honesty from demeanour: ‘the time spent over many years observing evidence being led, and witnesses being questioned, is of great assistance to a trier of fact. One cannot help but develop antennae sensitive to deliberate untruths’: at 669 (emphasis in original). This is not entirely consistent with empirical evidence: Frank and Ekman (n 137) 644–5. Elsewhere Ipp cautions: ‘[o]verconfidence on the part of judges leads to illusions about the value and accuracy of their own judgment’: Ipp (n 138) 670.

[141] See n 88.

[142] Suvaal v Cessnock City Council [2003] HCA 41; (2003) 200 ALR 1, 19 [71] (McHugh and Kirby JJ).

[143] In CSR, Kirby J said ‘[i]t would be a misfortune for legal doctrine if, so soon after Fox v Percy corrected the non-statutory excesses of earlier appellate deference to erroneous fact-finding by primary judges, the old approach was restored, as, for example, by reversion to the previous formulae about the “subtle influence of demeanour”’: CSR (n 110) 9 [23]. He referred in this regard to the judgment of Callinan and Heydon JJ: at 44–5 [180].

[144] Robinson Helicopter Co Inc v McDermott [2016] HCA 22; (2016) 331 ALR 550.

[145] Ibid 561 [54] (citations omitted).

[146] CSR (n 110) 44–5 [180], quoting Abalos v Australian Postal Commission [1990] HCA 47; (1990) 171 CLR 167, 179 (McHugh J).

[147] Fox (n 129) 126 [23] (Gleeson CJ, Gummow and Kirby JJ). See also Housen v Rural Municipality of Shellbrook [2002] 2 SCR 235, 250 [14] (Iacobucci and Major JJ); CSR (n 110) 7 [17] (Kirby J); Bell (n 131) 140. Kirby J made similar comments in Earthline (n 137)

619 [90], although, somewhat inconsistently in the preceding paragraph he suggested the appeal court would have ‘more opportunity to evaluate particular facts than is possible in the midst of a trial and with the appellate advantage of viewing such facts in the context of the record of the complete trial hearing’: at 619 [89].

[148] Van der Meer v The Queen [1988] HCA 56; (1988) 82 ALR 10, 31 (Deane J).

[149] Ratten v The Queen [1974] HCA 35; (1974) 131 CLR 510, 516 (Barwick CJ).

[150] Hocking v Bell [1945] HCA 16; (1945) 71 CLR 430, 440 (Latham CJ), quoted in Baden-Clay (n 104) 329 [65]. Of course, strictly speaking, the Commonwealth Constitution only provides a ‘weak conditional guarantee’ of a jury trial: Alqudsi v The Queen [2016] HCA 24; (2016) 258 CLR 203, 216 [25]

(French CJ). Section 80 requires that trials on indictment of federal offences be heard by jury. Juries are not required for any summary trials, and are not required for non-federal trials on indictment. However, despite the limited reach of s 80 the strong constitutional tradition of trial by jury for serious criminal offences has been followed in large part across all Australian jurisdictions. There is provision for judge-alone trials in most Australian jurisdictions: see, eg, Supreme Court Act 1933 (ACT) s 68B; Criminal Procedure Act 1986 (NSW) s 132; Criminal Code Act 1899 (Qld) ss 614–15E; Juries Act 1927 (SA) s 7; Criminal Procedure Act 2004 (WA) s 118. However, these are subject to significant restrictions: Jodie O’Leary, ‘Twelve Angry Peers or One Angry Judge: An Analysis of Judge Alone Trials in Australia’ (2011) 35 Criminal Law Journal 154, 156–7. The vast majority of indictable trials are heard by jury: see, eg, Mark Ierace, ‘Trials in NSW by Judge Alone: Recent Legislative Changes’ (Conference Paper, AIJA Conference, 9 September 2011) 8–9.

[151] Alqudsi (n 150) 255 [131] (Gageler J), quoting Sir Patrick Devlin, Trial by Jury (Stevens, 3rd ed, 1966) 164.

[152] Alqudsi (n 150) 255 [131] (Gageler J).

[153] Hawi (n 128) [480].

[154] Criminal Procedure Act 1986 (NSW) s 132(5); Alqudsi (n 150) 240 [86]. Actually, these issues, with their moral content, have more of a legal than a factual nature. They are classified as factual because the tribunal of fact, the jury, is viewed as more representative of the community than the tribunal of law, the trial judge, and so better placed to make such judgments: see n 120.

[155] Alqudsi (n 150) 254 [129] (Gageler J).

[156] Ibid 255 [132] (Gageler J), quoting Mark Findlay, Jury Management in New South Wales (Australian Institute of Judicial Administration, 1994) 1.

[157] Alqudsi (n 150) 257 [135] (Gageler J), quoting Kingswell v The Queen [1985] HCA 72; (1985) 159 CLR 264,

301 (Deane J).

[158] MFA v The Queen (2002) 213 CLR 606, 624 [59] (McHugh, Gummow and Kirby JJ). However, Australian appeal courts appear less deferential than British courts. Recent

research indicates that in NSW over the period 2001 to 2007, 65 out of 315, or 20.6%, of successful conviction appeals were on the basis that the verdict was ‘unreasonable, or cannot be supported, having regard to the evidence’: Hugh Donnelly, Rowena Johns and Patrizia Poletti, Conviction Appeals in New South Wales (Monograph No 35, Judicial Commission

of New South Wales, June 2011) 51, referring to the first limb of the common appeal provision in the Criminal Appeal Act 1912 (NSW) s 6(1). In England and Scotland it can be difficult finding any successful appeals on the facts: Stephanie Roberts, ‘The Royal Commission on Criminal Justice and Factual Innocence: Remedying Wrongful Convictions in the Court of Appeal’ (2004) 1(2) JUSTICE Journal 86, 91; Peter Duff, ‘Straddling Two Worlds: Reflections of a Retired Criminal Cases Review Commissioner’ (2009) 72 Modern Law Review 693, 716–18; Andrew Ashworth and Mike Redmayne, The Criminal Process (Oxford University Press, 4th ed, 2010) 380; Hamer, ‘Wrongful Convictions, Appeals, and the Finality Principle’ (n 102) 303.

[159] Singh (n 125) [127].

[160] Bell (n 131) 140.

[161] Johnson v Gore Wood & Co [2002] 2 AC 1, 31 (Lord Bingham); Richard Nobles and David Schiff, ‘The Right to Appeal and Workable Systems of Justice’ (2002) 65 Modern Law Review 676, 685–7.

[162] See, eg, Legislative Review Committee, Parliament of South Australia, Report of the Legislative Review Committee on Its Inquiry into the Criminal Cases Review Commission Bill 2010 (Parliamentary Paper No 211, 18 July 2012) 82.

[163] The Ampthill Peerage [1977] AC 547, 569 (Lord Wilberforce), quoted in R v Carroll (2002) 213 CLR 635, 643 [22] (Gleeson CJ and Hayne J).

[164] Paul Roberts, ‘Double Jeopardy Law Reform: A Criminal Justice Commentary’ (2002) 65 Modern Law Review 393, 400, 421.

[165] The Ampthill Peerage (n 163) 569 (Lord Wilberforce), quoted in Carroll (n 163) 643 [22] (Gleeson and Hayne J).

[166] Hamer, ‘Wrongful Convictions, Appeals, and the Finality Principle’ (n 102) 281.

[167] William Twining, Theories of Evidence: Bentham and Wigmore (Stanford University Press, 1985) 117. See David Hamer, ‘Probabilistic Standards of Proof, Their Complements and the Errors that are Expected to Flow from Them’ [2004] UNELawJl 3; (2004) 1 University of New England Law Journal 71, 71 n 1.

[168] Van der Meer (n 148) 31 (Deane J).

[169] Burrell v The Queen [2008] HCA 34; (2008) 238 CLR 218, 236 [72] (Kirby J). See also Bell (n 131) 133, quoting Ras Behari Lal v The King-Emperor (1933) 50 TLR 1, 4 (Lord Atkins). Fact scepticism extends to the criminal trial. The accused’s special treatment may reflect a view of the trial as ‘a forensic game in which every accused is entitled to some kind of sporting chance’: R v TA [2003] NSWCCA 191; (2003) 57 NSWLR 444, 446 (Spigelman CJ). This view is a response to the perception that the prosecution’s greater resources mean that ‘the adversaries wage their contest upon a tilted playing field’: Daniel Givelber, ‘Meaningless Acquittals, Meaningful Convictions: Do We Reliably Acquit the Innocent?’ (1997) 49 Rutgers Law Review 1317, 1360. However, such fact-sceptical perspectives are generally descriptive and critical, presented in support of reforms furthering factual accuracy: see, eg, Jerome Frank, Courts on Trial: Myth and Reality in American Justice (Atheneum, 1963) ch 6; David Hamer, ‘Delayed Complaint, Lost Evidence and Fair Trial: Epistemic and Non-Epistemic Concerns’ in Paul Roberts and Jill Hunter (eds), Criminal Evidence and Human Rights: Reimagining Common Law Procedural Traditions (Hart Publishing, 2012) 215, 231–5.

[170] See, eg, International Covenant on Civil and Political Rights, opened for signature 16 December 1966, 999 UNTS 171 (entered into force 23 March 1976) art 14.5 (‘ICCPR’). In Australia, leave is required to appeal on matters of fact: see, eg, Criminal Appeal Act 1912 (NSW) s 5(1)(b). However, in practice, leave is not difficult to obtain: Krishna v DPP (NSW) [2007] NSWCCA 318; (2007) 178 A Crim R 220, 228 [43], 231 [53] (Rothman J); Donnelly, Johns and Poletti

(n 158) 23. By contrast the English Court of Appeal (Criminal Division) grants leave in fewer than half the cases: Ashworth and Redmayne (n 158) 375.

[171] See, eg, ICCPR (n 170) art 14.7; Carroll (n 163). This is now subject to limited exceptions in England and many Australian jurisdictions: see generally Paul Roberts, ‘Double Jeopardy Law Reform’ (n 164); David Hamer, ‘The Expectation of Incorrect Acquittals and the “New and Compelling Evidence” Exception to Double Jeopardy’ (2009) 2 Criminal Law Review

63, 63.

[172] Carroll (n 163) 643 [21] (Gleeson CJ and Hayne J).

[173] See, eg, Perry v The Queen (1982) 150 CLR 580, 586 (Gibbs CJ); Sutton v The Queen [1984] HCA 5; (1984) 152 CLR 528, 547 (Brennan J).

[174] For expressions of frustration with the complexity of the exclusion of tendency and propensity evidence, see Hamer, ‘The Legal Structure of Propensity Evidence’ (n 54) 137–8.

[175] Clearly there will be cases where trial judges need to control the quantity of evidence: see, eg, R v O’Dowd [2009] 2 Cr App R 16, a case described by the English Court of Appeal as ‘involving just one defendant and ... relatively simple issues’ which ended up taking six and a half months, almost half of it on tendency evidence: at 283 [1] (Beatson J).

[176] Interestingly, Australian appeal courts do this far more frequently than British appeal courts: see n 158.

[177] See, eg, Fox (n 129) 128 [28]–[29].

[178] Special considerations apply to the admissibility of evidence based upon specialised knowledge. Generally, the jury will have no advantage in judging the credibility or reliability of such evidence. This is particularly so where the tendering party provides no evidence of the formal evaluation of the witness’s purported knowledge, leaving probative value a matter of speculation: Gary Edmond, ‘Icarus and the Evidence Act: Section 137, Probative Value and Taking Forensic Science Evidence “at its Highest”’ [2017] MelbULawRw 22; (2017) 41 Melbourne University Law Review 106, 122.

[179] Heydon (n 8) 236.

[180] Ibid 237.

[181] IMM (n 8) 344 [156].

[182] Consider perennial but growing concerns with the intelligibility of jury directions and

early guilty pleas: see, eg, Victorian Law Reform Commission, Jury Directions: Final Report (Report No 17, 1 May 2009); New South Wales Law Reform Commission, Encouraging Appropriate Early Guilty Pleas (Report No 141, December 2014).

[183] IMM (n 8) 315 [51], citing XY (n 7) 400 [167], [170] (Simpson J).

[184] IMM (n 8) 315 [51].

[185] The earlier the admissibility voir dire the more acute the problem, and admissibility may be decided at a pre-trial hearing under, eg, Criminal Procedure Act 1986 (NSW) s 130(2).

[186] Lancaster (n 67) 41.

[187] IMM (n 8) 315 [50].

[188] Lancaster (n 67) 41.

[189] IMM (n 8) 351 [175]–[176] (Nettle and Gordon JJ), rejecting this argument by reference to HML (n 71) 427 [280] (Heydon J). See also n 71.

[190] IMM (n 8) 318 [62] (French CJ, Kiefel, Bell and Keane JJ). This resembles the rationale for the common law exclusion of coincidence evidence where there is a reasonable possibility it is the product of joint concoction: Hoch v The Queen [1988] HCA 50; (1988) 165 CLR 292, 296–7 (Mason CJ, Wilson and Gaudron JJ), 300–1 (Brennan and Dawson JJ); Hamer, ‘Admissibility and Use of Relationship Evidence in HML v The Queen’ (n 71) 359–60. However, without noting the connection, the majority queried whether Hoch has application under the UEL: IMM (n 8) 317 [59].

[191] Odgers, Uniform Evidence Law (12th ed) (n 2) 1182. Odgers notes that in Aytugrul v The Queen [2012] HCA 15; (2012) 247 CLR 170 the High Court dealt with contextual assessment of probative value inconclusively with regard to s 137: at 1182–3. See also David Hamer, ‘Expected Frequencies, Exclusion Percentages and “Mathematical Equivalence”: The Probative Value of DNA Evidence in Aytugrul v The Queen(2013) 45 Australian Journal of Forensic Sciences 271, 280–2. However, the probative value assessment in s 97 is expressly contextual, concerning the tendency evidence ‘either by itself or having regard to other evidence adduced or to be adduced by the party seeking to adduce the evidence’: UEL (n 1) s 97(1)(b). See also McIntosh v The Queen [2015] NSWCCA 184, [45] (Basten JA).


AustLII: Copyright Policy | Disclaimers | Privacy Policy | Feedback
URL: http://www.austlii.edu.au/au/journals/MelbULawRw/2017/36.html