I have been asked to speak on the topic of “Evidence in prescribed sexual offence cases.” The first thing to be said about cases of this type is that they feature prominently in the appellate jurisdiction. The second thing to be said is that they also feature prominently in legislative reform. As a consequence of these two factors and the perception that the justice system has failed to provide procedural fairness for complainants, there has been a significant push in NSW for the creation of a specialist sexual assault court.
Leaving that issue for another day, what can be said is that “prescribed sexual offence cases” are a minefield through which even the most experienced judges cautiously and carefully plot their course. Even very experienced and capable criminal law judges suffer the embarrassment of having verdicts overturned from time to time, on the basis of so-called erroneous directions. Judges need to make a concerted effort to keep up to date with the latest legislative amendments and appellate decrees.
Dissatisfaction with appellate courts has been very evident in recent email correspondence between judges of the District Court and was raised during the first session of this conference. Unfortunately, problems with appellate courts seem to be the same around the world. For instance, a US trial judge is reported to have formulated the following jury direction in a criminal trial:
I must advise you as to the law in the case. These instructions are not to help you resolve the case. They are for the Court of Criminal Appeal so that they will know I have read their latest opinions. Therefore, the instructions will be convoluted and obtuse. You will not understand them.
General — admissibility and terminology
This paper focuses on the admissibility of contextual, tendency and coincidence evidence. The question of the admissibility of this type of evidence in prescribed sexual offence cases will usually arise in cases involving multiple counts on an indictment or cases involving evidence of multiple uncharged allegations.
The question of the admissibility of this type of evidence is very important because of the common law principle that counts should be severed where the evidence admissible to prove one is not admissible to prove the other, and the risk of prejudice is high: see De Jesus v The Queen (1986) 68 ALR 1; and Hoch v The Queen (1988) 165 CLR 292. This concern regarding the risk of prejudice to the accused where evidence on one count is not admissible on another, is also reflected in ss 21 and 29 of the Criminal Procedure Act 1986. If evidence is admissible on a contextual, tendency and/or coincidence basis, then the De Jesus concerns are not relevant as such evidence is likely to be admissible on all counts.
The threshold question of admissibility for any proposed evidence is that it must pass the relevancy test pursuant to s 55 of the Evidence Act 1995, that is “… could it rationally affect (directly or indirectly) the assessment of the probability of the existence of a fact in issue”: see R v Frawley (1993) 69 A Crim R 208 at 218; Papakosmas v The Queen (1999) 196 CLR 297; and Smith v The Queen (2001) 206 CLR 650. If the evidence is assessed as relevant, s 56 provides that it is admissible except as otherwise provided. For instance, tendency evidence that passes the relevancy test is not admissible unless it meets the criteria set out in ss 97 and 101.
“Relationship” is a nebulous term that has been associated with both sexual interest (previously guilty passion)/tendency evidence and contextual evidence. It is also used in s 293 of the Criminal Procedure Act 1986 to refer to a relationship between the complainant and the accused that is “existing or recent” at the time of the alleged prescribed sexual offence and is part of a connected set of circumstances in which the prescribed sexual offence was alleged to have occurred. The use of “relationship” in such a general and non-specific manner has led to the blurring of these vastly different legal heads of admissibility. This has generated misunderstanding and may well have resulted in a number of the problems discussed by the High Court in R v Tully (2006) 81 ALJR 39 (in particular, see the judgment of Callinan J at ). In Qualtieri v R (2006) 171 A Crim R 463, Howie J observed that it is preferable to refer to “context evidence” during the summing up, in accordance with the Criminal Trial Courts Bench Book, rather than “relationship evidence”: at  (Latham J agreeing at ).
In HML v The Queen (2008) 82 ALJR 723, the court clearly held that it no longer appropriate for a judge to describe other sexual incidents to the jury as “uncharged acts” as this may invite speculation about why no charges were laid (per Gleeson CJ at ; Hayne at ; Gummow J at  and Kirby J at  agreeing; Heydon J at ; Crennan J at  and Kiefel J at ). Further, it is not helpful to use the term “guilty passion” as this term obscures more than it illuminates (per Hayne J at ; Gummow J at  and Kirby J at  agreeing). Crennan J referred to the difficulty of identifying whether “guilty passion” referred to a pattern of conduct or indicated intention, motive or propensity: at .
If the Crown proposes to lead evidence of this type, which is potentially admissible on a number of different bases, it must clearly and precisely identify what evidence will be called and upon what basis or bases it is tendered. Anecdotally it seems to me that in many incidents of identified appellable error in this area, it has often been the case that insufficient attention was given to confining and defining the evidence and its purpose. Whatever the basis of the admission of such evidence, the jury should be clearly instructed on the use that can and can’t be made of the evidence. In order to do this effectively, the trial judge must be clear in his or her own mind as to the detail of the evidence, the purpose/s for which it can be used, and the limits to be placed on its use. If the trial judge is not clear, then there is very little prospect that it will ever be made clear to the jury, and if it is not, then the verdict is unlikely to stand.
In prescribed sexual offence cases evidence of this type is potentially admissible on a number of bases. It is fundamental to identify the basis on which evidence of other acts is admitted: DJV v R  NSWCCA 272 at . For example, and without attempting to cover the field:
As contextual evidence to establish the true nature and extent of the relationship between the complainant and the accused: see R v Wickham (unrep, 17/12/91, NSWCCA); B v The Queen (1992) 175 CLR 599; R v Tully (2006) 81 ALJR 391; HML v The Queen (2008) 82 ALJR 723; DJV v R  NSWCCA 272, or
as tendency (propensity or sexual interest) evidence: see ss 97, 101, R v AH (unrep, 27/11/97, NSWCCA); and R v Fraser (unrep,10/8/98, NSWCCA), or
as coincidence (similar fact) evidence: see ss 98, 101; R v Merritt  NSWCCA 29; and R v Bell  NSWCCA 2, or
as part of a connected set of circumstances (but not necessarily contemporaneous circumstances) in which the alleged prescribed sexual offence was committed. For instance, in the context of an existing or recent relationship between an adult complainant and the accused: see s 293 of the Criminal Procedure Act 1986, or
as part of the circumstances immediately before, or immediately after, the alleged prescribed sexual offence. For instance, in cases involving adult complainants where consent is a live issue, this evidence may be relevant to the question of whether the complainant had, or had not, consented and whether the accused knew the complainant had not been consenting. In some cases it might even be relevant to the issue of whether the charged sexual act actually took place at all, or
as relevant to subsidiary matters, such as rebuttal of good character, rebuttal of innocent explanation, rebuttal of other possible forensic issues raised by defence and to explain delay in, or lack of, complaint: see R v Fraser, above.
In this paper I will deal with the first three of these bases of admissibility.
One way that child sexual offence cases differ from other criminal cases is that, almost invariably, the allegation involves a course of conduct over a period of time. Complaints often lack specificity, as the frequency of offending is such that a young complainant may be unable to retain a specific recollection of each and every incident. This is compounded by the fact that developmental concepts such as time and sequence are lacking or still in a stage of immature development.
While many of the individual incidents tend to blur in the complainant’s mind, some can be recalled with enough specificity to be included as a count on an indictment. Examples of these may be the first or last incident of sexual abuse, incidents that occurred on particular or special days such as birthdays and incidents that may have occurred at different locations or in different circumstances than normal. In these cases, if a jury is to gain appropriate insight into the real context of the allegations it may be necessary for events not the subject of a charge to be referred to. The real context in this sense means a number of things, including but not confined to the history between the complainant and the accused and the circumstances in which the complainant was living at the time of the allegations.
The Court of Criminal Appeal and the High Court have considered the concept of contextual evidence on numerous occasions. In Plomp v The Queen (1963) 110 CLR 234 at 242, the High Court in dealing with a murder case where the question of “motive” was of critical significance, made the following comment regarding the admissibility of evidence going to the relationship or history between the accused and the deceased:
All the circumstances of the case must be weighed in judging whether there is evidence upon which a jury may reasonably be satisfied beyond reasonable doubt of the commission of the crime charged.
In the pre-Evidence Act 1995 case of B v The Queen (1992) 175 CLR 599 at 610, the High Court, this time when dealing with a sexual assault case where evidence of the accused’s earlier conviction for sexual assault against the same trial complainant was admitted, concluded that such evidence:
… constitutes part of the essential background against which both the daughter’s and the applicant’s evidence of the alleged offences necessarily fell to be evaluated.
This approach by the High Court was consistent with the approach espoused by McHugh J in the earlier case of Harriman v The Queen (1989) 167 CLR 590 at 631 where his Honour concluded:
… likewise in sexual cases, evidence of previous acts of misconduct by the accused in relation to the complainant will usually be admissible because it tends to prove why or how on the occasion in question the offence occurred in the circumstances alleged … (It may be) … circumstantial evidence tending to prove a fact in issue.
When sitting on the Court of Criminal Appeal Gleeson CJ, when dealing with this issue in R v Wickham (1991) (unrep, 17/12/91, NSWCCA), at p 8, explained the principle as follows:
… the evidence may constitute an essential aid to the understanding of an account or description of words or conduct. It will often be the case that it would be difficult, or impossible, to understand, or appreciate the full significance of, a complainant’s account of the events relating to a particular incident without knowing what had gone before. A bare description of the conduct the subject of a particular charge may be difficult to evaluate, and perhaps misleading, if it stood alone as an account of an apparently isolated incident.
This approach was confirmed in R v Beserick (1993) 66 A Crim R 419 at 422 when the court held:
… the evidence is admissible in order to place the evidence of the offence charged into a true and realistic context, in order to assist the jury to appreciate the full significance of what would otherwise appear to be an isolated act occurring without any apparent reason.
In the post-Evidence Act 1995 decision of R v AH (unrep, 27/11/97, NSWCCA) the court followed the pre-Evidence Act 1995 line of authority and concluded:
The relationship revealed may place the evidence of the events which give rise to a particular charge into their true context as part of the essential background against which the evidence of the complainant and of the accused necessarily fall to be evaluated.
In R v Fraser (unrep, 10/8/98, NSWCCA), at p 20 the court made specific reference to the Evidence Act 1995, and distinguished between the concept of relationship between the accused and the complainant, and the concept of contextual evidence, and explained the basis of the admissibility of contextual evidence as follows:
It may also be capable of explaining the nature of the relationship between an accused and his victim, and of placing any matter alleged against him in a true and realistic context … its admissibility depends upon its relevance, which in turn depends upon it being shown to be capable of rationally affecting (directly or indirectly) the assessment of the probability of the existence of a fact in issue: Evidence Act ss 55 and 56.
The conduct relied upon, proffered either as evidence supporting another specific indictment count or as an uncharged allegation, can be before or after the conduct alleged in the count being considered by the jury: see R v Fraser (unrep, 10/8/98, NSWCCA); R v Beserick (1993) 66 A Crim R 419 at 510; and R v Fordham (unrep, 2/12/97, NSWCCA).
The High Court’s decision in R v Tully (2006) 81 ALJR 391 deals with a Queensland case where the indictment contained eight counts of indecent dealing and two counts of rape. Evidence was admitted of numerous (probably in excess of 20) uncharged acts involving indecent acts and more serious acts involving penetration. The accused was convicted on seven counts of indecent dealing. The issues raised in the High Court were the need for a Longman warning, the need for a Murray warning and generally the situation regarding the evidence of, and directions given, in relation to uncharged acts. The case was a little unusual in that the defence adopted the forensic tactic of eliciting from the complainant during cross-examination as many incidents of uncharged acts as possible for the purpose of establishing gross exaggeration and therefore unreliability on her part.
The minority (Kirby and Hayne JJ) would have allowed the appeal on the basis that a Murray direction should have been given: see also Robinson v The Queen (1999) 197 CLR 162. Each concluded that the large issues regarding the admissibility of uncharged acts and the directions that should be given as to the use of such evidence should be reserved for a more appropriate case.
The majority (Callinan, Heydon and Crennan JJ) dismissed the appeal on the basis that the directions given were adequate. On the issue of uncharged acts Heydon and Crennan JJ agreed with Callinan J’s reasons and conclusion that there was no error that could have led to a miscarriage of justice. Callinan J acknowledged that the authorities of the High Court are not as clear as they might be in relation to uncharged acts. His Honour considered the decision of Gipp v The Queen (1998) 194 CLR 106 and acknowledged that “… the reasoning of the Justices in the majority (Gaudron, Kirby and Callinan JJ) was not all to exactly the same effect.” His Honour expressed a number of reservations regarding the reception of evidence of this type. However, his concerns seem to relate to the question of relevance and the weighing of the balance between probative value and prejudicial effect in any given case rather than to doubting the admissibility of relationship and/or context evidence that successfully negotiates the hurdles of ss 55, 135 and 137 of the Evidence Act 1995. His Honour ultimately concluded that the case was not one for the final resolution of the issues.
Callinan J pointed out at  of the judgment the difficulty of defining “relationship” and “contextual” and of the problems associated with imprecise use of these terms. For instance as he noted, “relationship” in the sense of a history between the complainant and the accused may also form part of the contextual material, while “relationship” in terms of marriage, boyfriend/girlfriend and friendship (either platonic or involving casual sexual activity) is often relevant in cases involving consent and knowledge. This is an entirely different meaning of “relationship” and it is relevant and therefore admissible on an entirely different basis than “relationship” in a contextual sense. A further complication is that both potentially raise tendency (propensity) or coincidence (similar fact) issues, as both are capable in any given case of being admitted on a tendency or coincidence basis.
Similarly, in Qualtieri v R (2006) 171 A Crim R 463, Howie J, following his earlier decision in R v ATM  NSWCCA 475, cautioned against the use of the term “relationship evidence”, particularly in child sexual assault cases, due to a lack of clear understanding of what the term means in any given case. Context evidence in child sexual assault cases is adduced to give context to the particular allegations made by the complainant that form the basis of the charges in the indictment and not to prove that the accused had any particular propensity or disposition towards the complainant (at ). Context evidence is normally adduced so that the complainant can give an intelligible narrative or history of events surrounding the particular allegations. Its relevance is found only in the extent to which it provides an understanding of the particular allegations before the jury: at .
The Court of Criminal Appeal in Qualtieri, above, held (at ) that where the Crown seeks to adduce evidence of other sexual conduct, the following steps should be taken:
identification of the evidence and the purpose of its tender should be clearly identified.
If the evidence is adduced to show a tendency on the part of the accused, it must be assessed in accordance with ss 97 and 101 of the Evidence Act 1995.
If evidence has been tendered merely to provide context to the charges laid, it will be admissible only if it is relevant and not unfairly prejudicial: R v ATM  NSWCCA 475 followed. Such evidence should be admitted in accordance with ss 135 and 137 of the Evidence Act. Because the evidence is prejudicial, great care must be exercised.
If the evidence is admitted, the judge should direct the jury as to its use at the time it is admitted and again in the summing up. The trial judge should direct the jury that context evidence has a confined use (outlined at , –) and cannot be used as tendency evidence.
The Court of Criminal Appeal referred to the model direction in the Criminal Trial Courts Bench Book (at [4-215]) as being in “the appropriate manner in which to instruct the jury” as to the confined use to be made of context evidence (at ).
The High Court gave extensive, but not always authoritative consideration, to the admissibility of evidence of other sexual conduct of the accused in the recent decision of HML, SB and OAE v The Queen (2008) 82 ALJR 723;  HCA 16. The court also considered the use to which such evidence might be put and the burden of proof in relation to that evidence. HML concerned three appeals from the South Australian Supreme Court where the common law applies and not the Evidence Act.
All of the Justices agreed that evidence of other sexual conduct is potentially admissible, but differed as to the basis for admission and the standard of proof that had to be established.
Gleeson CJ (at , ) and Crennan J (at ) held that evidence of a contextual nature is admissible if it is relevant, that is, if it bears upon “the assessment of the probability of the existence of a fact in issue” by assisting in the evaluation of other evidence. Contextual evidence may be admitted to explain a statement or an event that would otherwise appear curious or unlikely; to reinforce the plausibility of other evidence; to provide for an understanding of the complainant’s narrative; or to overcome a false impression that the event was an isolated one. Such evidence is subject to the general principle concerning probative value and prejudicial effect: per Gleeson CJ at . Gleeson CJ held (at –) that no separate question of the standard of proof arises where evidence is adduced to explain the context or assist in the evaluation of other evidence.
Crennan J (at –) held that evidence of other sexual conduct of the accused may be admissible as essential background against which evidence of the complainant and the accused falls to be evaluated, to show the continuing nature of the conduct, and to explain the offences charged so that the charges will not be considered in a vacuum. Crennan J held that the Pfennig test of probative value (see below) did not apply where the evidence was led for purposes other than to establish propensity as proving an element of, or fact essential to, the offences charged: at .
Keifel J held (at –) that “relationship” evidence is relevant, not generally, or by way of background or context, but for the limited purpose to answer questions that may arise in the minds of the jury were the jury limited to a consideration of evidence of the offences charged and thereby to rebut an inference which might otherwise be drawn by the jury. Hence, relationship evidence might explain why the complainant did not rebuff the accused or show distress or resentment, or why the complainant did not complain to another person.
Hayne J (at –, ; Gummow J at  and Kirby J at  agreeing) held that evidence of other sexual acts directed at the complainant by the accused, which are not acts the subject of the charges being tried, is not admissible in evidence unless the trial judge is satisfied that, in the context of the prosecution case, there is no reasonable view of the evidence which would be consistent with innocence, in accordance with Pfennig v The Queen (1995) 182 CLR 461 at 483–484. These Justices did not support a wider basis for the admission of such evidence such as to bolster the credibility of the complainant or to provide evidence of the general context. Hayne J said that admissibility is “… not to be determined by asking whether the evidence in question will put evidence about the charges being tried ‘in context’, or by asking whether it describes or proves the ‘relationship’ between the complainant and accused”. Hayne J cautioned against attempts to divide the uses to which evidence of other discreditable acts of an accused may be put: at –. Hayne J held (at ) that if evidence of other discreditable acts of the accused does not meet the Pfennig test, it is inadmissible. Hence, it is unnecessary to consider any division of uses.
After HML, the Court of Criminal Appeal in DJV v R  NSWCCA 272 at , confirmed that “context” evidence remains admissible in New South Wales. The court said that HML is “… a decision notable for its lack of a common approach to the problems”: DJV v R at .
A summary of the key legal principles in the decision was published in (2008) 20(11) Judicial Officers’ Bulletin 97.
The NSWCCA held in DJV v R that for context evidence to be admissible, it must be relevant. There must be an issue or issues relating to the charged acts that justifies the reception of the evidence: at , . McClellan CJ at CL said in DJV v R at : “… [the evidence] may provide the ‘context’ in which to understand a narrative in the sense suggested [in HML] by Gleeson CJ [at ] and Heydon J [at ]” and further: “In most cases relevance will be occasioned by an apparent lack of complaint by a complainant whose will has been overborne from a young age or who has feared the consequences of making a complaint about a family member”: at .
The trial judge in DJV v R and in JDK v R  NSWCCA 76 erroneously accepted that the evidence should be admitted merely because it described events in the relationship — even if unrelated to the particular charges — rather than asking “to what issue in the trial does the evidence go” and, in light of the answer to that question, determining its admissibility after also considering the probative value and any potential unfair prejudice (ss 135 and 137, Evidence Act): DJV v R at .
Where the evidence of other acts is remote in time and is adduced only for the purpose of showing the context in which the offence occurred, “the almost inevitable result” will be that it is rejected under s 137: at . When determining whether the evidence should be admitted, the judge must consider whether the directions can or cannot avert the risk of an impermissible line of reasoning by the jury that because the accused committed the other acts he/she committed the offences charged: DJV v R at ; JDK v R at .
If the evidence is admitted as context evidence, the dangers of its misuse are significant. The judge must clearly direct the jury as to the limited purpose for which they may use the evidence: DJV v R at . Where context evidence has been admitted, it may be desirable to give a tendency warning (see Criminal Trial Courts Bench Book at [5-1620]) but it is not mandatory as each case should be considered on its own facts. A tendency warning will be required if there is a significant risk that the jury might embark on impermissible tendency resoning having regard to the context evidence and arguments of counsel: Toalepai v R  NSWCCA 270 at .
In the rare case when context evidence is admitted which is remote in time from the charges, a direction to the jury in the form of “a caution as to the reliability of evidence remote to the charge is necessary”: DJV v R at .
Context evidence is not tendency evidence and need not comply with ss 97 and 101: R v AH (1997) 42 NSWLR 702 at 708. It does require close consideration of ss 55, 135 and 137: DJV v R at .
The court held that context evidence does not require a direction that it be proved beyond reasonable doubt: at .
While we must wait for the High Court’s final resolution of the law regarding evidence of other sexual conduct, it is advisable that the observations of Callinan J in Tully, above, at  be carefully considered:
… as a minimum that evidence of uncharged acts have some actual direct probative value relevant to the issues, that it be carefully scrutinized before it is admitted, that it may need to be characterized as propensity evidence, and that it almost always will require, if admitted over objection, directions appropriate for evidence of that kind.
Tendency evidence (propensity and sexual interest)
Having regard to s 97 and the Dictionary to the Evidence Act 1995, tendency can be defined as being evidence of the character, reputation or conduct of a person or a tendency that a person has or had which proves, or is capable of proving, that a person has or had a tendency, whether because of the person’s character or otherwise, to act in a particular way, or to have a particular state of mind.
A tendency to act in a particular way, or to have a particular state of mind at a certain time or times, leads to the inference that such a person has a propensity to act in a particular way, or have a particular state of mind thereafter, or at other times if the same circumstances exist. Therefore, evidence of an abnormal sexual interest in a young person for a period of time suggests the existence of sexual interest, being an abnormal or wrong passion, which is capable of amounting to a tendency as defined by the Evidence Act 1995.
Tendency evidence will include other acts alleged by the complainant and may also be established by other acts alleged against other complainants of a similar nature and involving an identical or similar “modus operandi”.
In a sense the tendency principle is a deceptively simple concept. However, the practical application of the principle is seemingly more problematic. Hence, clear and concise directions are required. For example, I simply direct the jury that if they are satisfied beyond reasonable doubt as to an accused’s conduct on other occasions and that that conduct establishes that the accused demonstrated an abnormal or unusual sexual interest in a complainant, then that evidence is available to prove that the accused had a tendency to have a particular state of mind, namely, an abnormal or unusual sexual interest in the complainant and that state of mind can be taken into account when assessing the credibility and reliability of the particular complainant.
As explained by Gleeson CJ in R v Wickham (unrep, 17/12/91, NSWCCA) at pp 7–8, evidence of sexual interest can provide direct proof of the commission of the actual offence charged:
… the evidence may establish a sexual relationship which makes the complainant’s allegations more likely to be true … the relationship in question need not be completely mutual, and the emotions or tendencies involved need not be charged for there to be a relevant relationship. The expression “guilty passion” is sometimes used in relation to an accused. It conveys an idea that most people would recognize as being of significance in determining whether a particular incident of a sexual nature has occurred. If past conduct shows that a child is an object of a parent’s sexual desire then that may well make more credible an allegation that a particular alleged sexual incident occurred between them.
In B v The Queen (1992) 175 CLR 599 at 610, Deane J referred to the use of evidence of established sexual interest and observed that such a state of mind was:
… capable of being seen by the jury as important evidence of motive, namely, sexual gratification. The jury was also entitled to accept it as providing corroboration of the daughter’s evidence; it supported the truth of the allegation which was implicit in the whole of the daughter’s evidence, namely, that the applicant entertained an unnatural passion for her.
The Court of Criminal Appeal in R v Beserick (1993) 66 A Crim R 419 at 422 confirmed the approach adopted in Wickham noting “… it makes it more likely that the offence charged was in fact committed.”
In Pfennig v The Queen (1995) 182 CLR 461 at 526, McHugh J noted that:
Since Ball [ AC 47], it has never been doubted that in a sexual offence case, evidence of the conduct of the parties is relevant to prove that the offence charged occurred, notwithstanding that the only reasoning process is propensity.
In HML v The Queen, above, (a common law case but of limited assistance in relation to the Evidence Act) the court held that evidence of other acts is admissible to show that the accused had a sexual interest in the complainant; that the accused had given expression to it by committing the uncharged offences and further, that this made it more probable that he had committed the charged offences: Gleeson CJ at ; Hayne J (with whom Gummow and Kirby JJ agreed) at , , –, , ; Heydon J at ; Crennan J at ; Kiefel J at  and . A majority held that at common law the Pfennig test applies to this evidence, requiring a consideration of whether there is a rational view of the evidence consistent with the accused’s innocence: Kirby J at ; Hayne J at , ,  (Gummow J agreeing at ); and Heydon J at . The terms “sexual interest” and “sexual attraction” were preferred to “guilty passion” in HML. Hayne J (Gummow J at  and Kirby J at  agreeing) said that the term “guilty passion” obscured more than than it illuminated: at .
If the Crown elects to use this type of evidence as “tendency,” as well as establishing “relevance” and “relationship”, the Crown must give reasonable notice of its intention (s 97(1)(a)), establish that the evidence has “significant probative value” (s 97(1)(b)) and that its “probative value substantially outweighs” any prejudicial effect: s 101(2)). Notice can be dispensed with pursuant to s 100 and is not required where the tendency evidence is led to “… explain or contradict tendency evidence adduced by another party”: s 97(2).
In R v Fraser (unrep, 10/8/98, NSWCCA) it was said that “… the expression ‘significant probative value’ has been interpreted to mean that its degree of relevance to the events giving rise to the offences charged is important or of consequence”: see also R v Lockyer (1996) 89 A Crim R 457 at 459; R v Lock (1997) 91 A Crim R 356; and R v AH (above).
Even where prosecution evidence has significant probative value, it is only admissible as tendency evidence if “… the probative value of the evidence substantially outweighs any prejudicial effect it may have on the defendant” (s 101(2)), unless it is led by the prosecution in reply to tendency evidence led by the defendant: s 101(3).
Evidence that is admitted for a non-tendency purpose cannot be left to the jury as tendency evidence unless it meets the requirements of ss 97 and 101: see s 95.
On the other hand, it should be noted that ss 97 and 101 do not apply to evidence that relates only to the credibility of a witness or to character when character has been raised as a fact in issue: see s 94.
While sexual interest can qualify as “important or of consequence,” clearly each factual situation must be considered separately and carefully. The High Court has held that providing the conduct is not too remote in time, evidence of sexual interest is of significant probative value: see Harriman v The Queen (1989) 167 CLR 590 at 597–599; S v The Queen (1989) 168 CLR 266 at 275; B v The Queen (1992) 175 CLR 599 at 618; and Gipp v The Queen (1998) 194 CLR 106.
The Court of Criminal Appeal in R v Beserick (1993) 66 A Crim R 419 held that evidence of sexual interest had to be proportional to the charges alleged. In that case the one count on the indictment related to an allegation right at the beginning of the “relationship” and the prosecution led evidence of sexual activity for some years subsequent to this charge. The court held (per Hunt CJ at CL at 428–429) that:
… remoteness of the other sexual activity from the time of the offence charged goes to the weight of that evidence. The more remote the other sexual activity is, the less will be its weight; and in general (as a matter of commonsense), the weight to be afforded to subsequent sexual activity will be less than that to be afforded to previous sexual activity.
In R v Fraser, above, the court followed Beserick and confirmed that the conduct could be before or after that which is charged. However, the circumstances and timing of the conduct is clearly very relevant to the question of the weight to be given: see Beserick, above, at 510; and R v Fordham (unrep, 2/12/97, NSWCCA).
Tendency evidence can also be used to establish that an accused has “groomed” a complainant or fostered a relationship by providing alcohol or drugs, employment, and recreational opportunities that might not normally be available to them. Such evidence of “grooming” is relevant generally as providing evidence of an accused’s “modus operandi”: see R v Fletcher (2005) NSWCCA 338; and R v Milton (2004) NSWCCA 195.
The decision of the specially convened five judge Bench of the Court of Criminal Appeal in R v Ellis (2003) NSWCCA 319 establishes that the admissibility of tendency and coincidence evidence is governed by the statutory regime found in the Evidence Act 1995. In summary this statutory regime, as set out in s 97 through to and including s 101, requires a court to conclude that the evidence either by itself or having regard to other evidence to be adduced has significant probative value. Further, s 101(2) requires a balancing process between probative force and prejudicial effect, during which the court must make a judgment that probative value substantially outweighs prejudicial effect. There is little guidance as to the meaning of substantial but the Macquarie Dictionary definition includes “real or actual”, “of ample or considerable amount,” and “of real worth or value.”
In the event that the court is satisfied that the proposed evidence has significant probative force and that that probative force substantially outweighs any prejudicial effect it may have on the accused, the court must still consider whether it is appropriate to limit the use made of such evidence pursuant to s 136 or whether the probative value of the evidence is outweighed by the danger of unfair prejudice to the accused pursuant to s 137 of the Evidence Act 1995. The tests in ss 101(2) and 137 are obviously different in that the former speaks of actual prejudicial effect while the later deals with the danger of unfair prejudice and s 101(2) raises the bar from outweighs to substantially outweighs. Accordingly, the s 101(2) requirement that the probative value must substantially outweigh any prejudicial effect is clearly a more difficult hurdle for the prosecution than that created by s 137 and it is difficult to imagine s 137 ever operating to reject tendency evidence, the probative value of which has been assessed as substantially outweighing any prejudicial effect.
On the other hand, s 136 does have a clear and important role to play. The use of s 136 should be carefully monitored as there may well be cases where limiting the use of tendency evidence might be required to reduce the danger of unfair prejudice. The section is used in such a way in cases where hearsay evidence (admitted through s 60) is limited to a non-hearsay purpose (see also situations where evidence is limited in use to avoid the disadvantage of a lost opportunity to cross examine: see Larrakia People v Northern Territory  FCA 1175 and generally s 65 regarding the use of statements when the maker is not available).
If tendency evidence is admitted, the jury must be warned that before they can use such evidence they must be satisfied beyond reasonable doubt that the allegations occurred as alleged. In HML v The Queen, above, a majority held that if evidence of other conduct of the accused is admitted to show the accused had a sexual interest in the complainant, the jury must be directed that they must only find that the accused has a sexual interest in the complainant if it is proved beyond reasonable doubt: Hayne J at , ,  (Gummow J at  and Kirby J at , ,  agreeing); Kiefel J at . However, Gleeson CJ at  and Crennan J at  held that the evidence of other uncharged offences, adduced to show the accused’s sexual interest in the complainant, need not be established beyond reasonable doubt unless it is an indispensable fact in the process of reasoning towards the guilt of the accused.
The application of HML to New South Wales on the question of whether the jury needs to accept evidence of other acts, adduced to show the accused’s sexual interest in the complainant, beyond reasonable doubt, was determined in DJV v R  NSWCCA 272 at . The court held, following the majority decision in HML, that “… unless after full argument of the issue this court or the High Court says otherwise, this court should accept that in sexual assault cases the appropriate standard of proof of tendency evidence is beyond reasonable doubt”.
The jury must also be warned that even if they accept the tendency evidence beyond reasonable doubt, nothing done on any other occasion can of itself establish that the accused is guilty of the charged offence.
It is also necessary to give clear and appropriate directions to the jury to ensure that the jury deals with tendency evidence correctly thus minimising the risk of unfair prejudice brought about by unjustified uses of this evidence or inappropriate emotional responses to the accused’s conduct on other occasions. (For unfair prejudice see Australian Law Reform Commission Report No 26, Evidence, Vol 1 at ; R v Yates  NSWCCA 520 at ; R v BD (1997) 94 A Crim R 131 at 139 and 151; and R v GK (2001) 53 NSWLR 317.) The danger of unfair prejudice is lowered if clear and strong directions are given to the jury identifying the dangers and how the evidence can and cannot be used both generally and in the context of the particular case.
Coincidence evidence (similar fact)
Coincidence evidence (known more commonly as similar fact evidence pre-Evidence Act 1995) relies on the improbability of two or more events occurring coincidentally. Evidence of related events that are substantially and relevantly similar and which occurred in substantially similar circumstances can be led to establish that a person did a particular act or had a particular state of mind, because of the improbability of the events occurring coincidentally. Note that the Evidence Amendment Act 2007 (which applies to proceedings commencing on or after 1 January 2009) has abolished the need for events to be “related” under s 98(1) and “substantially and relevantly similar” under s 98(2). The test now is expressed more simply as “having regard to any similarities in the events or the circumstances in which they occurred”.
In relation to the pre-Amendment Act test, where a jury is satisfied beyond reasonable doubt of evidence of conduct on some other occasion, and satisfied beyond reasonable doubt that such other conduct, when compared to the conduct alleged in the indictment, could not be coincidental, the jury may conclude that such improbability of coincidence establishes that the accused has demonstrated by his or her other conduct a particular state of mind, purpose or motive, and that that is capable of establishing that the accused had the same state of mind at the time of the incident alleged in the indictment.
Further, where a jury is satisfied beyond reasonable doubt of evidence of conduct on some other occasion and satisfied beyond reasonable doubt that such other conduct, when compared to the conduct alleged in the indictment, could not be coincidental, the jury may conclude that such improbability of coincidence establishes that the accused, having committed the other act, must have committed the act alleged in the indictment.
Therefore, the improbability of coincidence is also relevant to identification in that it may establish that two separate acts could only have been committed by the same person. Therefore proof that the accused committed the other separate act is strong evidence that he or she must have committed both crimes. For instance, where there is “… such a similarity between two different sets of acts and the circumstances in which they occurred”, it may be that the only acceptable explanation or conclusion available to the tribunal of fact is that the accused committed the offences against both complainants: Criminal Trial Courts Bench Book at [5-1640] (see now [4-237]).
Another example of the use of coincidence evidence (although not involving a prescribed sexual offence case) can be found in R v Anna Zhang (2005) 158 A Crim R 504  where Simpson J (Buddin J agreeing) concluded:
… that the evidence supporting each count was admissible in respect of the other count in order to prove that, because of the improbability of the events (that is, the presence of the drugs in two premises controlled by the appellant without her knowledge) occurring coincidentally, the appellant attempt to, in the one case, and did in fact in the other, possess the substance in question.
This case provides a good factual basis for examining the principles underlying the coincidence rule. For instance, to take a more extreme factual situation, it is hard to imagine that there could be any doubt that the owner of, and regular visitor to, four separate properties could be anything other than criminally involved in the cultivation of cannabis if sixty or so cannabis plants were located growing at each of the four properties. If the four plantations were grown sequentially, the first three in time would be available as tendency evidence in support of a prosecution for the cultivation of the fourth plantation. When the plantations are grown simultaneously at the four different locations that fact is admissible on a coincidence basis because of the improbability of four plantations occurring coincidently.
In R v Merritt (1999) NSWCCA 29 at 39, the Court of Criminal Appeal noted that:
The essence of coincidence evidence is that there is established to have been two events in circumstances where the existence of one event establishes the likelihood that the accused was responsible for, or involved in, the other separate event …
Coincidence evidence may establish a pattern of conduct: see R v Bell  NSWCCA 2. It may establish the credibility of witnesses who, completely independent of one another provide strikingly similar accounts. It may eliminate accident as an innocent explanation and it may establish a particular state of mind: see also R v Folbigg (2005) 152 A Crim R 35; R v WRC (2002) 130 A Crim R 139; and R v MM  NSWCCA 364.
An example of the improper use of coincidence reasoning can be found in Phillips v The Queen (2006) 158 A Crim R 431 where the High Court concluded that on the question of whether one complainant did not consent to sexual intercourse, evidence from other complainants (there were five complainants) that they did not consent, did not have sufficient probative value for admission, as it did not of itself prove any disposition on the part of the accused. Evidence that other complainants did not consent was said to say nothing about whether the first complainant had consented on the relevant occasion.
Apart from dealing with coincidence evidence rather than tendency evidence, s 98 is in precisely the same terms as s 97, and so precisely the same restrictions apply to the admissibility of coincidence evidence as apply to the admissibility of tendency evidence. Similarly, the restrictions on admissibility imposed by s 101 and as addressed in the tendency section of this paper apply with equal force to coincidence evidence.
Evidence that is admitted for a non-coincidence purpose such as motive, system or context cannot be left to the jury as coincidence evidence unless it meets the requirements of ss 98 and 101: see s 95. However, ss 98 and 101 do not apply to evidence that relates only to the credibility of a witness or to character when character has been raised as a fact in issue: see s 94.
The issues considered and noted in this paper highlight the complexity of the concepts raised by evidence of this type in prescribed sexual offence cases. A trial judge is required to work carefully through all the issues and it is not appropriate to justify rejection of such evidence because of what might be said to be the enormity of the judicial task, as justice will often require its admissibility, albeit in the context of all necessary jury directions and warnings.
*Paper delivered to the District Court of New South Wales Annual Conference, 10–12 April 2007, Leura; updated March 2010. Acknowledgment to Hugh Donnelly, Director, Research and Sentencing for preparing commentary on DJV v R  NSWCCA 272.