Relationship evidence

27.263 If there is only one complainant, the prosecution may want to lead evidence from other witnesses about the defendant’s criminal sexual behaviour with them, or it may wish to adduce relationship evidence to explain the nature of the relationship between the complainant and the defendant, as well as the context in which the sexual assault occurred.

27.264 Evidence of uncharged acts of sexual misconduct is commonly referred to as ‘relationship’, ‘context’ or ‘background’ evidence and is a type of circumstantial evidence. While relationship evidence describes all conduct ‘of a sexual kind’ which is often referred to as ‘uncharged acts’, it also includes grooming behaviours that do not amount to an offence, such as the purchase of gifts and non-sexual touching.[326]

27.265 Relationship evidence forms part of the background against which the complainant’s and the defendant’s evidence are assessed.[327] In a sexual assault trial, relationship evidence may be relevant for a number of different reasons including to: provide a context in, or background against, which to understand the charges laid against the defendant; explain why the complainant complied with the sexual demands of the defendant without surprise; explain why the complainant failed promptly to complain; and explain the defendant’s confidence in committing the sexual acts or his control over the complainant.[328]

27.266 Where such evidence is admissible, it ‘cannot be used by the jury to reason that, if the accused committed the uncharged acts, he or she is more likely to have committed the charged acts’.[329] However, the distinction between relationship evidence and tendency evidence has been described as ‘somewhat artificial’[330] since evidence which shows the ‘existence of a sexual relationship must surely tend to show that the accused [has a tendency] to do the sort of things the subject of the charge’.[331]

27.267 Nonetheless, many cases have held that evidence of uncharged sexual behaviour between a complainant and an accused is admissible.[332] Relationship evidence has a long history of being admitted in all types of criminal trials.[333]

27.268 In the last decade, however, the admissibility of relationship evidence has had a conflicted history with ‘sharp divisions’ in the High Court about when and for what purposes such evidence should be admissible.[334] This division saw a majority attempt to restrict the admission of such evidence at common law in Gipp v The Queen.[335] This case resulted in considerable uncertainty about the test that should be applied to admit relationship evidence.[336]

Case law

27.269 It is possible that this uncertainty has been resolved by the High Court decision in HML v The Queen (HML).[337] HML only applies to cases of child sexual assault, where lack of consent is not one of the issues to be decided[338] and in jurisdictions that have not overturned the ‘no rational view of the evidence’ (Pfennig) test.[339]

27.270 The issues in HML, a case involving sexual abuse by a father of his daughter, included when and in what circumstances evidence of uncharged acts of sexual misconduct is admissible in a child sexual assault trial and what test applies to its admission.[340] All seven judges accepted that there are important reasons why evidence of uncharged acts of sexual misconduct by the defendant ought to be admissible in child sexual assault trials. The public policy reasons recognised for permitting the child to give evidence of uncharged acts included:[341]

  • ‘multiple and repeated incidents over a period of time’ are typical of the sexual abuse of children (and this is particularly so in the context of sexual assault perpetrated by family members);

  • the impracticality or impossibility of being able to charge every multiple incident of sexual abuse;

  • the importance of legal procedure to facilitate the giving of a ‘fair and coherent account’ by the complainant of what has occurred;

  • the exclusion of evidence of uncharged acts would mean that the complainant’s evidence of the charged acts of sexual misconduct would be viewed in isolation and sometimes as if the complainant was sexually abused ‘out of the blue’;

  • the artificiality of attempting to ‘quarantine the charged acts’ from other incidents;

  • the seriousness of child sexual assault as a crime and its historical under-enforcement which ought not to be frustrated by ‘unjustifiably restrictive court procedures’;[342] and

  • because today’s juries are ‘better educated and more literate’, there is less need for restrictive rules for excluding relevant but prejudicial evidence.

27.271 In HML, all seven judges agreed that uncharged acts of sexual misconduct are likely to be relevant to the facts in issue in a child sexual assault trial, that is, whether the defendant committed the sexual acts that constitute the charges.[343] However, the relevance test was not considered to be a sufficient control on the admissibility of relationship evidence by Gummow, Kirby and Hayne JJ, all of whom agreed that, in addition to relevance, evidence of uncharged acts should not be admissible unless the Pfennig test is satisfied.[344]

27.272 The National Child Sexual Assault Reform Committee has noted that ‘lower courts may not find it all that easy to decide whether to apply the approach of Gummow, Kirby and Hayne JJ, or the approach of the other four judges’,[345] who either did not agree that the Pfennig test applied to relationship evidence at all,[346] or did not think the Pfennig test applied to the relationship evidence in the cases at hand, even though there may be other situations in which it would apply.[347]

27.273 In HML, Kiefel J considered that only where relationship evidence is being tendered for its tendency purpose should the Pfennig test be applied. This is the same approach that is taken in uniform Evidence Acts jurisdictions where the test under s 101(2) does not apply to relationship evidence which is tendered for a non-tendency purpose.[348]

Reform options

27.274 The National Child Sexual Assault Reform Committee proposed that, because of the different and inconsistent approaches taken by the judges in HML to the admissibility of relationship evidence, the Pfennig test at common law should be abolished in relation to the admissibility of relationship evidence.[349] This would bring common law jurisdictions in line with those that do not use the test (or use the balancing test under s 101 of the uniform Evidence Acts ) to admit relationship evidence—that is, Western Australia and the uniform Evidence Acts jurisdictions.

Consultation Paper

27.275 In the Consultation Paper, the Commissions asked whether the Pfennig test should be applied to determine the admissibility of relationship evidence at common law.[350] Few stakeholders responded to this question.[351] The Public Defenders Office NSW confirmed that the Pfennig test is not applied to relationship evidence.[352]

27.276 Cossins provided a detailed analysis of HML and associated law and submitted that the application of the Pfennig test to relationship evidence ‘is to interpose an extra test of relevance which is both illogical and unnecessary’. Cossins submitted that a specific provision that abrogates the Pfennig test should be enacted ‘to ensure that relationship evidence is admissible if it passes the relevance test, with appropriate warnings to be given to the jury about how they may use the evidence’. In her view, such a provision would ensure that relationship evidence at common law ‘is not subject to the rule in Pfennig where the evidence is tendered for a non-tendency purpose in a child sexual assault trial’.[353]

Commissions’ views

27.277 In the Commissions’ view, the ‘no rational view of the evidence’ (Pfennig) test should not apply to relationship evidence. As discussed above, in uniform Evidence Acts jurisdictions, the Pfennig test does not apply, even to tendency and coincidence evidence. There is no need for further provisions to deal specifically with relationship evidence, as defined in this discussion, beyond the general relevance test and other rules set out in the uniform Evidence Acts.

27.278 Reform directed to the admissibility of relationship evidence in common law evidence jurisdictions would be problematic given the uncertainty over the meaning and practical effect of the decision in HML. Complex legislative drafting would be required to remove doubt about the possible application of the test to relationship evidence in Queensland, South Australia and the Northern Territory. The reform would require the development of a statutory definition of relationship evidence and deal with the purposes for which the evidence may be used, directions to the jury and so on.[354]

27.279 The Commissions do not consider it would be appropriate, in the context of this Inquiry, to recommend reforms to evidence law in common law evidence jurisdictions directed to relationship evidence. Rather, the Commissions consider it is preferable that all Australian jurisdictions join the uniform Evidence Acts scheme.

[326]HML v The Queen (2008) 235 CLR 334, [492] (sexual offences against defendant’s daughter).

[327]B v The Queen (1992) 175 CLR 599, 610 (sexual offences against defendant’s daughter).

[328] A Cossins, Alternative Models for Prosecuting Child Sex Offences in Australia (2010), prepared for the National Child Sexual Assault Reform Committee, 235. See, eg, R v Vonarx [1999] 3 VR.

[329] Queensland Law Reform Commission, The Receipt of Evidence by Queensland Courts: The Evidence of Children, Report 55 (Part 2) (2000), 371.

[330] A Cossins, Alternative Models for Prosecuting Child Sex Offences in Australia (2010), prepared for the National Child Sexual Assault Reform Committee, 236.

[331]R v Knuth [1998] QCA 161, [22] (sexual offences against the defendant’s daughters).

[332] See Cook v The Queen (2000) 22 WAR 67; R v Nieterink (1999) 76 SASR 56 (sexual offences against a friend of the defendant’s step-daughter);R v Vonarx [1999] 3 VR (sexual offences against defendant’s son); R v Alexander [1996] SASC 5557; R v Beserick (1993) 30 NSWLR 510; B v The Queen (1992) 175 CLR 599.

[333] The reception of ‘this type of evidence has come to be routine and unsurprising’: HML v The Queen (2008) 235 CLR 334, [271]–[272].

[334] Ibid, [328].

[335] Gaudron J considered that the admissibility of propensity evidence in the form of past criminal conduct (including relationship evidence) was only warranted in a few specified situations, eg, where the defence raised specific issues such as evidence of good character or lack of surprise or failure to complain on the part of the complainant: Gipp v The Queen (1998) 194 CLR 106, 112. Callinan J stated that if such evidence is to be admitted ‘it must owe its admissibility to some, quite specific, other purpose, including eg, in an appropriate case, proof of a guilty passion, intention, or propensity, or opportunity, or motive’: Gipp v The Queen (1998) 194 CLR 106, [182]. See also Tully v The Queen (2006) 230 CLR 234, [144]–[145]. Kirby J stated that relationship evidence should only be ‘admissible if its probative value outweighs its prejudicial effect’: Gipp v The Queen (1998) 194 CLR 106, [141].

[336]HML v The Queen (2008) 235 CLR 334, [163]. This uncertainty has been evident, eg, in the different approaches taken by the Full Court of the Supreme Court of South Australia in R v Nieterink (1999) 76 SASR 56 and the Court of Appeal of Victoria decision in R v Vonarx [1999] 3 VR.

[337] Although ‘the six different judgments in the case may continue to confound rather than enlighten the law on relationship evidence’: A Cossins, Alternative Models for Prosecuting Child Sex Offences in Australia (2010), prepared for the National Child Sexual Assault Reform Committee, 237.

[338]HML v The Queen (2008) 235 CLR 334, [102].

[339] Ibid, [54]. The case does not apply to the admission of relationship evidence under the uniform Evidence Acts, nor under Evidence Act 1906 (WA) s 31A. The Pfennig test still applies in Queensland, SA and the NT.

[340] A Cossins, Alternative Models for Prosecuting Child Sex Offences in Australia (2010), prepared for the National Child Sexual Assault Reform Committee, 237–238.

[341]HML v The Queen (2008) 235 CLR 334, [56].

[342] This includes ‘the social interest in convicting those guilty of crimes against small children which are both grave and difficult to prove’: Ibid, [331].

[343] For example, Hayne J stated that ‘[p]roving that the accused not only had that sexual interest but had given expression to that interest by those acts, made it more probable that he had committed the charged acts’: Ibid, [103]. As discussed above, the evidence may also be relevant on other grounds, such as to explain the complainant’s delay in complaint, although some of those reasons may only be directly relevant to the credibility of the complainant: HML v The Queen (2008) 235 CLR 334, [426].

[344]HML v The Queen (2008) 235 CLR 334, [59] (Kirby J); [106] (Hayne J with whom Gummow J agreed). The purpose for adopting the Pfennig test is to place an extra control over ‘the discreditable facts’ that are admitted against an accused. Without such a control any relevant discreditable facts would be able to be tendered against an accused simply because they throw some light on the ‘context’ of the offences: HML v The Queen (2008) 235 CLR 334, [80]. Nonetheless, Hayne J considered that evidence of uncharged acts would usually satisfy the Pfennig test because ‘there will usually be no reasonable view of the evidence … other than as supporting an inference that the accused is guilty of the offence charged’: HML v The Queen (2008) 235 CLR 334, [107]. Although Hayne J recognised the disadvantage to the defendant by admitting such evidence, His Honour considered that ‘its admission would work no prejudice to the accused over and above what the evidence establishes’: HML v The Queen (2008) 235 CLR 334, [110].

[345] A Cossins, Alternative Models for Prosecuting Child Sex Offences in Australia (2010), prepared for the National Child Sexual Assault Reform Committee, 240.

[346]HML v The Queen (2008) 235 CLR 334, [455] (Crennan J); [511]–[512] (Kiefel J).

[347] See Ibid, [27] (Gleeson CJ); [264] (Heydon J).

[348] Ibid, [503], [505]. In ALRC Report 102 the ALRC, NSWLRC and VLRC considered whether the balancing test under s 101(2) of the uniform Evidence Acts should be extended to apply to any evidence tendered against a defendant which discloses disreputable conduct although tendered for a non-tendency or non-coincidence purpose—such as relationship evidence: see Australian Law Reform Commission, New South Wales Law Reform Commission and Victorian Law Reform Commission, Uniform Evidence Law, Report 102, NSWLRC Report 112, VLRC FR (2005), [11.76]–[11.93]. The Commissions concluded that the case had not been made out for change, which would be unlikely to result in different outcomes where questions arise as to the admissibility of evidence relevant for tendency or coincidence purposes and for other purposes.

[349] A Cossins, Alternative Models for Prosecuting Child Sex Offences in Australia (2010), prepared for the National Child Sexual Assault Reform Committee, Rec 3.14.

[350] Consultation Paper, Question 18–9.

[351] The Northern Territory Legal Aid Commission responded in the affirmative: Northern Territory Legal Aid Commission, Submission FV 122, 16 June 2010.

[352] Public Defenders Office NSW, Submission FV 221, 2 July 2010.

[353] A Cossins, Submission FV 112, 9 June 2010.

[354] See, eg, A Cossins, Alternative Models for Prosecuting Child Sex Offences in Australia (2010), prepared for the National Child Sexual Assault Reform Committee.