The admission of tendency evidence is governed by Pt 3.6 Evidence Act and in particular ss 97, 97A, 100 and 101.

Section 97(1) relevantly provides:

Taylor v R [2020] NSWCCA 355 at [85]–[122] contains a useful summary of the legislative history of, and caselaw on, tendency evidence. The following discussion of the caselaw must be read with the terms of s 94(4) in mind (which commenced after Taylor v R). This section puts “beyond doubt that any principle or rule of common law or equity preventing or restricting the admissibility of this kind of evidence is not relevant when applying Pt 3.6 of the Evidence Act”: see Second Reading Speech, Evidence Amendment (Tendency and Coincidence) Bill 2020, NSW, Legislative Assembly, Debates, 25 February 2020, p 1916.

Tendency evidence is a form and special class of circumstantial evidence: DPP v Roder (a pseudonym) [2024] HCA 15 at [23]. The trier of fact reasons from satisfaction that a person has a tendency to have a particular state of mind, or to act in a particular way, to the likelihood that the person had a particular state of mind, or acted in a particular way, on the occasion in issue. The starting point requires identifying the tendency and the fact(s) in issue which it is adduced to prove: Hughes v The Queen (2017) 263 CLR 338 at [16].

The tendency sought to be proved must be clearly articulated before a ruling on admissibility of the evidence for a tendency purpose can be made: New v R [2025] NSWCCA 32 at [180]. Otherwise, there will usually be three issues in determining the admissibility of tendency evidence:

Subject to s 97(2), a party seeking to adduce tendency evidence must give reasonable notice in writing in accordance with cl 5 of the Evidence Regulation 2020: ss 97(1)(a), 99. As to the date by which notice should be given, see s 99; District Court Rules 1973, Pt 53 r 10C; Supreme Court Rules 1970, Pt 75 r 3(6); and Uniform Civil Procedure Rules (UCPR), r 31.5.

The purpose and significance of the notice requirement (including giving the other party a reasonable opportunity to address and respond to it) were discussed in Bryant v R [2011] NSWCCA 26 at [50]–[51] and R v AC [2018] NSWCCA 130 at [23]–[25]. The question of whether or not notice is reasonable is to be evaluated by the court, having regard to the purpose of s 97 in the context of the Act as a whole: Wright v R [2025] NSWCCA 108 at [47]. Matters which may be relevant include:

In determining whether notice is reasonable it is not mandatory for the court to take into account any non-compliance with rules or regulations of court made under s 99 (including r 31.5 of the UCPR), although in an appropriate case such non-compliance might be relied upon: [47]–[54].

As for the contents of a properly drafted tendency notice, see R v Gardiner [2006] NSWCCA 190 at [128]. Section 100 sets out the circumstances in which notice may be dispensed with. The power conferred by s 100(1) is to “direct” that the tendency rule does not apply despite a party’s failure to give notice, and accordingly is subject to s 192 of the Act: R v AC at [22]; R v Harker [2004] NSWCCA 427 at [34]. When the power to dispense is engaged, the task of the court is to weigh the interests of justice, in the circumstances of the particular case, taking into account the purposes for the notice requirement, other ways in which those purposes might be achieved, and any reason for dispensing with the requirement notwithstanding that those purposes might not be achieved: R v AC at [26]ff.

Care must be taken by the court not to import elements into the asserted tendency in addition to those stated in a tendency notice or amended tendency notice. Reformulation of a tendency without providing notice under s 97(1)(a) may render evidence inadmissible: see TL v The King [2022] HCA 35 at [33]; see also New v R [2025] NSWCCA 32 at [180]. Usually, it is desirable that any amendment to an asserted tendency should be by way of an amended notice by the relevant party seeking to rely upon the tendency evidence. Otherwise, any disparity between the tendency asserted or evidence identified in the tendency notice and that asserted or relied on in the trial should be the subject of a direction under s 100(1): cf DP v R [2025] NSWCCA 45 at [63]–[67].

In determining the probative value of evidence for the purposes of s 97(1)(b), a trial judge should assume the jury will accept the evidence and, thus, should not have regard to the credibility or reliability of the evidence: IMM v The Queen (2016) 257 CLR 300 at [51]–[52], [54], [58]; The Queen v Bauer (a pseudonym) (2018) 266 CLR 56 at [69]; TL v The King [2022] HCA 35 at [28].

The test posed by s 97(1)(b) is whether the disputed evidence, together with other evidence, makes significantly more likely any facts making up the elements of the offence charged: Hughes v The Queen (2017) 263 CLR 338 at [40]. In the case of multiple counts on an indictment, it is necessary to consider each count separately to assess whether the tendency evidence which is sought to be adduced in relation to that count is admissible: Hughes v The Queen at [40].

The assessment of whether evidence has significant probative value involves consideration of two matters: first, the extent to which the evidence supports the tendency; and second, the extent to which the tendency makes more likely the facts making up the charged offence: Hughes v The Queen at [41]; see also DPP v Roder (a pseudonym) [2024] HCA 15 at [24]. There is likely to be a high degree of probative value where: (i) the evidence, alone or together with other evidence, strongly supports proof of a tendency, and (ii) the tendency strongly supports the proof of a fact that makes up the offence charged, in other words, the evidence must be “important” or “of consequence” to the assessment of the probability of the existence of a fact in issue: [41]; see also TL v The King at [28].

Evidence of a more generally expressed tendency is less likely to satisfy the threshold of “significant probative value”. While generalised tendency notices may be supported by a broader array of evidence, that evidence will often not be significantly probative of the fact(s) in issue and may not establish more than relevance: TL v The King at [29] (and see the cases cited there); Hughes v The Queen at [64].

There is no general rule requiring close similarity between the tendency evidence and the offence: TL v The King at [29]. For evidence to be admissible as tendency it is not necessary that it exhibit an “underlying unity”, “a modus operandi” or a “pattern of conduct”: Hughes v The Queen at [34] approving the approach in R v Ford [2009] NSWCCA 306; R v PWD [2010] NSWCCA 209; Saoud v R (2014) 87 NSWLR 481 and disapproving Velkoski v R (2014) 45 VR 680 at 682. It is not necessary that the common features be “striking”. What is needed is a sufficient link between the distinct events as to mean that one piece of conduct has significant probative value as regards another. That link need not be peculiar: Bektasovski v R [2022] NSWCCA 246 at [93]; The Queen v Bauer (a pseudonym) at [57]. Depending upon the issues in the trial, a tendency to act in a particular way may be identified with sufficient particularity to have significant probative value notwithstanding the absence of similarity in the acts which evidence it: Hughes v The Queen at [37]. Section 97(1) does not condition the admissibility of tendency evidence on the court’s assessment of operative features of similarity with the conduct in issue; however, commonly there may be a similarity between the tendency asserted and the offences charged: Hughes v The Queen at [39].

A “close similarity” between the tendency evidence and the charged offence will almost certainly be required where the evidence is adduced to prove the identity of an offender: Hughes v The Queen at [39]. However, this should be understood as referring to situations where there is little or no other evidence of identity apart from the tendency evidence and the identity of the perpetrator is “at large”: TL v The King at [30], [38].

See also The Queen v Bauer (a pseudonym), at [58], [60], where the Court considered the probative value of tendency evidence in sexual offence proceedings involving a single complainant, compared to multiple complainants. However, that discussion should now be read in light of s 97A — concerning the presumption of significant probative value for child sexual assault proceedings — see further below.

Section 94(4) and (5) were inserted into the Act in 2020 and affects hearings which commenced from 1 July 2020.

Section 94(4) states that any principle or rule of the common law or equity preventing or restricting the admissibility of tendency or coincidence evidence is not relevant when applying Pt 3.6.

Section 94(5) provides that in determining the probative value of tendency or coincidence evidence the court must not have regard to the possibility the evidence may be the result of collusion, concoction or contamination. Previously, in The Queen v Bauer (a pseudonym) (2018) 266 CLR 56 at [69]–[70], the Court left open the possibility that, where the risk of contamination, concoction or collusion is so great it would not be open to the jury rationally to accept the evidence, it could be taken into account in assessing probative value. In the Second Reading Speech, the Attorney General stated: “Proposed section 94(5) … closes that small gap left open by the courts …” (Evidence Amendment (Tendency and Coincidence) Bill 2020, NSW, Legislative Assembly, Debates, 25 February 2020, p 1917).

Section 97A applies to proceedings in which the commission by the defendant of an act that constitutes, or may constitute, a child sexual offence (as defined in s 97A(6)) is a fact in issue (s 97A(1)).

Under s 97A(2) there is a presumption that tendency evidence about the following will have significant probative value for the purposes of ss 97(1)(b) and 101(2):

Section 97A took effect on 1 July 2020. The Attorney General described the provision as altering the operation of s 97(1)(b) for child sexual abuse prosecutions in order to facilitate greater admissibility of tendency evidence (see Second Reading Speech, Evidence Amendment (Tendency and Coincidence) Bill 2020, NSW, Legislative Assembly, Debates, 25 February 2020, p 1914).

A court retains a discretion to determine such evidence does not have significant probative value if satisfied there are sufficient grounds to do so: s 97A(4). However, s 97A(5) lists the following matters (whether considered individually or collectively) the court is not to take into account in determining whether there are sufficient grounds, unless there are exceptional circumstances in relation to those matters:

The terms “sufficient grounds” (in s 97A(4)) and “exceptional circumstances” (in s 97A(5)) are not defined. As to the former, in the Second Reading Speech (NSW, Legislative Assembly, Debates, 25 February 2020, p 1915) the Attorney General said:

And of the latter:

For the purposes of determining whether “significant probative value” is satisfied in a child sexual assault case, s 97A creates a rebuttable presumption that that threshold is met for any tendency of the kinds described in subs (2) of s 97A: R v Clarke [2023] NSWCCA 123 at [16]. Section 97A evinces a clear legislative intention to make it easier for the prosecution to adduce tendency evidence: Davidson (a pseudonym) v R [2024] NSWCCA 60 at [46]. It represents a very significant departure from the previous state of the law: R v Clarke at [37]–[38].

In Stenner-Wall v R [2023] NSWCCA 163 at [14], Beech-Jones CJ at CL (agreeing with Button J; Hamill J also agreeing) noted that exceptional circumstances may be established where the particular circumstances of one or more of the matters identified in s 97A(5) are such as to demonstrate the absence of significant probative value on the part of the subject evidence: [14].

The transitional provisions for the amendment state s 97A does not apply where the hearing of proceedings began before the amendment commenced: Sch 2, cl 28. Where the application of the transitional provisions is in issue, it will be necessary to identify the relevant “hearing” of the proceedings and to determine when it began: JW v R [2022] NSWCCA 206 at [54].

Subject to s 101(3), tendency evidence about a defendant adduced by the prosecution cannot be used against the defendant unless the probative value of the evidence outweighs the danger of unfair prejudice to the defendant: s 101(2).

Since 1 July 2020, the test in s 101(2) is whether the probative value of the evidence outweighs the danger of unfair prejudice — the word “substantially” was removed.

In The Queen v Bauer (a pseudonym) (2018) 266 CLR 56 at [73], the High Court described prejudice as conveying the idea of harm to an accused’s interests by reason of a risk the jury would use the evidence improperly in some unfair way. See also Hughes v R (2015) 93 NSWLR 474 at [189]–[193]; Taylor v R [2020] NSWCCA 355 at [122 (xvii)–(xxviii)].

In Hughes v The Queen (2017) 263 CLR 338 at [17], the High Court articulated how tendency evidence may occasion prejudice to an accused:

In determining whether the probative value of tendency evidence outweighs the danger of unfair prejudice to an accused, the trial judge must have regard to the directions which can be given to remedy or ameliorate any unfair prejudice which might be occasioned in the absence of directions and whether those directions would be sufficient to remedy or ameliorate that prejudice: Davidson (a pseudonym) v R [2024] NSWCCA 60 at [51]. It is important that the unfair prejudice to a defendant be specifically identified for the purposes of the weighing exercise required by s 101 and in considering appropriate directions: BC v R [2015] NSWCCA 327 at [107]–[110]; Mol v R [2017] NSWCCA 76 at [36] (decided prior to amendment of s 101, however the principle still applies).

In circumstances where tendency evidence has been adduced by the accused, the jury should not be directed to exercise caution before drawing inferences based on that tendency evidence. Moreover, the tendency directions should not be tied to circumstantial evidence directions requiring that all other rational inferences need be excluded because it suggests that the accused bears an onus of proof: Waldron v R [2023] NSWCCA 128 at [39]–[48].

It is appreciated that the structure of a summing-up is a matter for the personal preference of judges. However, consideration should be given as to when a tendency direction might best be given to minimise the risk of confusion on the jury’s part as to any standard of proof to be applied. For example, it may be given before the directions about the onus and standard of proof and the essential elements of the offence(s) and shortly after directions concerning the drawing of inferences. The timing may vary depending on the issues in the particular trial.

The following suggested direction complies with s 161A(1) Criminal Procedure Act 1986 in not directing that tendency evidence needs to be proved beyond reasonable doubt. It will require modification as to the standard of proof where the exception in s 161A(3) applies. The observations in DPP v Roder (a pseudonym) [2024] HCA 15 (referred to above) and JS v R [2022] NSWCCA 145 at [40] should also be taken into account, including that the directions need to be crafted carefully to avoid undermining general directions concerning proof beyond reasonable doubt for each charge.

A tendency may be proved by evidence of “the character, reputation or conduct of a person, or a tendency that a person has or had”: s 97(1) Evidence Act. The suggested direction refers only to “conduct” and will require modification in a case in which it is sought to be proved in an alternative way.

Inferential reasoning is usually involved in deciding whether a tendency has been established so it will be helpful if the jury has already been directed as to the care required in the drawing of inferences generally.

Trial judges should be alive to any possible prejudicial misuse of tendency evidence that might arise in a particular case and add any further warning that may be required.

IMPORTANT NOTE TO TRIAL JUDGES: The suggested directions set out below are given by way of example. Where the prosecution relies on evidence for a tendency purpose, trial judges must formulate careful directions which must be fashioned to address the particular circumstances of the case and must guard against the improper use of tendency evidence and the potential for prejudice which is inherent in such evidence: Rassi v R [2023] NSWCCA 119 at [73]–[74].

The admission of coincidence evidence is governed by Pt 3.6 Evidence Act and in particular ss 98, 100 and 101.

Section 98(1) provides:

Coincidence evidence is a species of circumstantial evidence which invites the tribunal of fact to engage in probabilistic reasoning that two or more events could not have occurred coincidentally. It gains its probative value when it suggests the objective improbability of an event occurring accidentally: Sampson v R [2023] NSWCCA 239 at [55] and [56] citing Page v R [2015] VSCA 357 at [45]. Coincidence reasoning is not limited to the issue of identification: see the examples given in Addo v R [2022] NSWCCA 141 at [56]. Section 98 can be satisfied by a form of reasoning that relies on the improbability of two separate victims describing separate but similar crimes committed by an accused unless they are telling the truth. Reasoning in that form is predicated on the relevant “event” being the giving of the account of the crime by the victim rather than the crime itself: [57].

There is an “overlap” or “awkwardness” between tendency and coincidence reasoning that has been recognised, particularly where there is no dispute as to the identity of the alleged offender: Sampson v R at [56]; El-Haddad v R [2015] NSWCCA 10 at [46]; Saoud v R [2014] NSWCCA 136 at [43]; see also Addo v R at [66]–[68] and the discussion about s 98(1A) below at [4-234].

To apply s 98 it is necessary for the tendering party to identify the “basis” upon which the evidence is sought to be adduced. It can then be determined whether the mode of reasoning under s 98 is engaged: Addo v R at [52]. There will usually be three issues in determining the admissibility of coincidence evidence:

The series of steps involved in determining the admission of coincidence evidence were set out in R v Gale [2012] NSWCCA 174 at [31]; R v Duckworth [2012] NSWCCA 174 at [31]; see also Xie v R [2021] NSWCCA 1 at [780]; and Shah v R [2024] NSWCCA 113 at [92]–[98] where they were applied.

Subject to s 98(2), a party seeking to adduce coincidence evidence must give reasonable notice in writing in accordance with cl 5 of the Evidence Regulation 2020: ss 98(1)(a), 99.

The contents of a properly drafted coincidence notice were set out in R v Zhang [2005] NSWCCA 437 at [131]. See also the discussion in El-Haddad v R at [56].

For further commentary on the statutory provisions and rules governing notice, including when notice may be dispensed with, see [4-204] in relation to tendency evidence above.

For coincidence evidence to be admissible the court must be satisfied that the evidence will, either by itself or having regard to other evidence adduced or to be adduced by the party seeking to adduce the evidence, have significant probative value: s 98(1)(b). Section 94(4) provides that any principle or rule of the common law or equity that prevents or restricts the admissibility of evidence about propensity or similar fact evidence in a proceeding is not relevant when applying Pt 3.6 to tendency or coincidence evidence about a defendant.

In Xie v R [2021] NSWCCA 1 at [776]–[781], the Court reviewed multiple authorities on assessing probative value under s 98.

The observations in IMM v The Queen (2016) 257 CLR 300 as to the assessment of significant probative value for the purposes of s 97(1)(b) also apply to s 98(1)(b): A2 v R [2018] NSWCCA 174 at [1032]–[1033]; R v Matonwal & Amood [2016] NSWCCA 174 at [77]–[78]; Xie v R at [778]–[779]; Sampson v R [2023] NSWCCA 239 at [59]–[62]. See the discussion above at [4-206]. In R v Gale, the Court identified a “serious logical fallacy” in so far as the similarities relied upon as coincidence evidence assumed the very same facts the Crown sought to prove: [37]; distinguished in Sampson v R at [46].

It is necessary to give consideration to the evidence sought to be tendered as a whole, rather than separate consideration of each particular circumstance relied upon: R v Matonwal at [73], [75]; R v MR [2013] NSWCCA 236 at [9]–[10]; [78]–[79]. The degree of similarity between the events will be relevant to the assessment of the probability that the events did not occur coincidentally: Sampson v R at [59]. However, the improbability that something was a coincidence is not displaced by the fact that the two (or more) events bear some dissimilarities. The question is whether the dissimilarities are relevant in that they undercut the improbability of something being a coincidence and whether they detract from the strength of the inferential mode of reasoning permitted by s 98: Selby v R [2017] NSWCCA 40 at [24]–[26]; Sampson v R at [59].

Section 94(5) provides in determining the probative value of coincidence evidence, it is not open to the court to have regard to the possibility that the evidence may be the result of collusion, concoction or contamination. For further discussion of this provision and s 94(4), see above at [4-208].

Coincidence evidence about a defendant, that is adduced by the prosecution cannot be used against the defendant unless the probative value of the evidence outweighs the danger of unfair prejudice to the defendant: s 101(2). For a general discussion see [4-212] Unfairly prejudicial effect of tendency evidence.

See [4-214] Standard of proof for commentary on s 161A of the Criminal Procedure Act 1986.

The following suggested direction concerns where the coincidence reasoning is relied upon between counts on the indictment; identity is the issue; the coincidence evidence is not the only evidence relied upon in respect of each count to establish identity; and it is also not an essential intermediate fact. Note that in Davis v R [2018] NSWCCA 277 (at [77], [105]) the Court accepted it was an intermediate fact whether the same person killed all three victims.

This form of coincidence evidence is reflected in s 98(1A), which provides coincidence evidence includes the use of evidence from two or more witnesses claiming they are victims of offences committed by the defendant to prove, on the basis of similarities in the claimed acts or the circumstances in which they occurred, that the defendant did an act in issue in the proceedings: see also Addo v R [2022] NSWCCA 141 at [70]–[71].

The only available path of reasoning authorised by s 98 is (Addo v R at [87]):

If the evidence of two or more witnesses shows sufficient similarity to be admissible as coincidence evidence, it can be used to prove that the witnesses would not make up those versions independently and by chance. In such a case the issue of concoction may arise and require a direction to the jury that they should reject the possibility of concoction before using the evidence for coincidence reasoning.

The below suggested direction assumes there are two complainants.

Where evidence is not admitted as tendency or coincidence evidence, as is the case with context evidence (sometimes referred to as “relationship evidence”), then the issues around admissibility will likely be whether the evidence is relevant and whether it should be excluded under ss 137 or 135 of the Act.

Although not confined to particular offences, context evidence is most often admitted in child sexual assault cases. An example of its operation could be where the complainant in a sexual assault case is permitted to give evidence of other (uncharged) acts, often of a sexual nature, allegedly committed against them by the accused. In this way, context evidence is adduced to put into a meaningful context a complainant’s account about conduct which is the subject of a specific charge, without which the jury might have difficulty believing the complainant’s evidence: DP v R [2025] NSWCCA 45 at [50]; HML v The Queen (2008) 235 CLR 334 at [2]–[11].

It is essential to identify the purpose of the evidence tendered by the Crown and in this respect the Court is not bound by what the Crown asserts it to be: RWC v R [2010] NSWCCA 332 at [130]; RG v R [2010] NSWCCA 173 at [34]. Evidence is not admissible simply because it proves the relationship between the complainant and the accused: R v ATM [2000] NSWCCA 475 at [73]. There must be an issue which the evidence may explain or resolve by placing the alleged events in their true context: DJV v R [2008] NSWCCA 272 at [29]; Norman v R [2012] NSWCCA 230 at [33]; CA v R [2017] NSWCCA 324 at [65]. Otherwise, the evidence is irrelevant or may be relevant only as tendency: see, for example, DJV v R at [36]–[39]; RWC v R at [130].

In BRC v R [2020] NSWCCA 176 at [52]–[64], Simpson AJA reviewed a number of authorities to draw out the distinction between context and tendency evidence. See also Qualtieri v R [2006] NSWCCA 95 at [119]ff in relation to how context and tendency evidence bolster the credibility of the complainant in different ways.

Context evidence does not go directly to the guilt of the accused: RWC v R at [122]; KTR v R [2010] NSWCCA 271 at [92]. Rather, context evidence is relevant to:

For a discussion of the purpose and requirements of an anti-tendency direction see DP v R at [2025] NSWCCA 45 at [52]–[54]. See also the commentary as to when such a direction may be appropriate at [4-200] Introduction above. The timing at which a direction is given depends on the case. It may be given at the time the context evidence is led: Greenaway v R [2021] NSWCCA 253 at [56]. An anti-tendency direction should be given if there is a real risk of tendency reasoning being adopted: Medich v R [2021] NSWCCA 36 at [133].

Where evidence is not admitted as tendency or coincidence evidence, as is the case with background evidence, then the issues around admissibility will likely be whether the evidence is relevant and whether it should be excluded under ss 137 or 135 of the Act.

The term “background evidence” is adopted to specifically distinguish types of evidence from “tendency”, “coincidence” and “context evidence” described above.

Background evidence places the accused’s alleged conduct and/or state of mind within the surrounding events including the relationship between the accused and the complainant, or a series of other incidents which form part of chain of events. Background evidence tends to have a close temporal connection with the incident giving rise to the charge. Background evidence is admissible as circumstantial evidence to prove that the accused committed the offence charged.

Background evidence, however, is not tendency evidence. It does not require tendency reasoning to make it relevant although as circumstantial evidence it relies upon available inferences or conclusions arising from the background evidence to prove the charge.

Recognising there may be overlap, and although the language around these categories is not fixed, for the purposes of this chapter background evidence might be conveniently thought of as either:

The function of a direction in the case of background evidence is to inform the jury of the limited purpose for which the evidence is admitted and to direct them against using the evidence for tendency reasoning. The content of the direction will depend substantially upon the nature of evidence and the purpose it is being admitted. The direction should contain the following components: