Sexual Behaviour Evidence and Evidence of Bad Character in Sexual Offence Proceedings: Proposing a Combined Admissibility Framework

Introduction

The decision of the Court of Appeal in R v Evans (Chedwyn) ¹ (Evans), and the subsequent acquittal of the appellant following retrial in October 2016, received extensive press coverage ² and caused a great deal of public concern. The judgment provoked significant and high-profile criticism and led to unsuccessful attempts to introduce legislative measures, comprising, respectively, a Private Member’s Bill ³ introduced on 8 February 2017 and an amendment to the Government’s Prisons and Courts Bill introduced on 23 March 2017. ⁴

The proposed legislation would have introduced significant amendments to s 41 Youth Justice and Criminal Evidence Act (YJ&CEA) 1999. The Private Member’s Bill would have further restricted the admission of evidence of the complainant’s sexual behaviour with third parties. The proposed amendment to the Prisons and Courts Bill would have made sexual behaviour evidence inadmissible under any circumstances. ⁵ The Attorney-General’s Office and Ministry of Justice subsequently conducted a review into of the use of sexual behaviour evidence in over 300 cases. ⁶ Ultimately, the review concluded that a change in the law was unnecessary and that ‘section 41 is working as intended’. ⁷ This finding was subsequently supported by empirical research undertaken by Laura Hoyano on behalf of the Criminal Bar Association. The research revealed that of 179 responses from CBA members ‘[n]ot a single respondent considered that section 41 should be reformed to make it more restrictive’ (original emphasis). ⁸

The decision of the Court of Appeal in the Evans case also prompted significant academic debate, including exchanges between McGlynn and Dent and Paul in the Criminal Law Review ⁹ and other key contributions made by Stark ¹⁰ and Thomason. ¹¹ Dent and Paul call for recognition of the ‘progressive and pragmatic nature of s.41’ ¹² and for proposed amendments to ‘be based on evidence, rather than assumptions…’. ¹³ By contrast, McGlynn opines that reform of this area of law is ‘urgently required’ and that, ideally, a ‘comprehensive revision of the current law…from first principles’ ¹⁴ is necessary. Despite this, the consensus appears to be that the law on sexual behaviour evidence is both a ‘highly controversial’ and ‘extremely (and unnecessarily) complicated area of law’. ¹⁵

This article will provide critical insight into the admissibility of evidence of a complainant’s previous sexual behaviour under s 41 YJ&CEA 1999. Specific consideration will be given to issues raised in Evans to determine whether the case provides an appropriate basis on which to propose reforms to the law in this area. The article will also address the broader question of whether s 41 strikes an appropriate balance between the protection of complainants from unfair and unnecessary cross examination regarding their previous sexual behaviour, and the art 6 European Convention on Human Rights (ECHR) of the defendant to a fair trial. The Evans case has been selected for particular consideration, in part, because of the disproportionate attention that this case received in the context of proposed reforms of s 41 but also because, factually, it has exposed some of the deficiencies within the existing s 41 framework.

Finally, a case will be made for the adoption of an alternative statutory framework which departs from the ‘strict categories’ approach to admissibility currently adopted by s 41 YJ&CEA 1999. The new statutory regime proposes the creation of a combined admissibility framework encompassing both sexual history evidence and evidence of the complainant’s bad character in sexual offence proceedings. The framework draws not only on the lessons learned following the introduction of s 41 but also comparatively on the experiences of other jurisdictions in which ‘rape shield’ legislation has been enacted. Further, it seeks to add to the contribution made by commentators such as Stark ¹⁶ and Thomason ¹⁷ in developing a balanced legislative approach to the admissibility and restriction of sexual behaviour evidence.

The Evans Case

In April 2012, Chedwyn ‘Ched’ Evans (Evans) was convicted of rape and sentenced to five years’ imprisonment. His co-accused Clayton McDonald (McDonald) was acquitted. The allegation arose following sexual activity between the complainant, McDonald and Evans which took place at a hotel on 29 May 2011 while the complainant was heavily intoxicated. The following day the complainant awoke ‘naked, alone and confused’, ¹⁸ with no recollection of how she arrived at the hotel. After attempting ‘to piece together with friends what had happened’, ¹⁹ she eventually contacted the police. The complainant did not allege that she had been raped, ‘Her evidence was simply that she did not remember what happened in room 14…’. ²⁰ All ‘evidence as to sexual activity in room 14 came solely from [Evans] and McDonald themselves’. ²¹ The prosecution case at trial was that, by virtue of her level of intoxication, the complainant was not able to give a valid consent to sexual activity with Evans, nor could he have formed a reasonable belief in her consent.

After an unsuccessful appeal in 2012, ²² in 2016, ²³ the Criminal Cases Review Commission (CCRC) referred the case back to the Court of Appeal following ‘an in-depth, ten-month long investigation’, ²⁴ on the basis that fresh evidence had been obtained, ‘which was not raised at trial, and which in the view of the Commission, could have added support to Mr Evan’s defence at trial…’. ²⁵ The evidence in question related to statements from two witnesses, a Mr Owens (O) and a Mr Hughes (H), both of whom described engaging in sexual activity with the complainant in May and June 2011, which counsel for the appellant described as having ‘striking detail’ ²⁶ in common with the account given by the appellant. The CCRC considered that the evidence of O and H might have been admissible under s 41(3)(c)(i) YJ&CE Act 1999, because it was relevant to the issue of consent and was so similar to the sexual behaviour of the complainant which took place as part of the subject matter of the charge that the similarity could not reasonably be explained as coincidence. The defence sought to adduce the evidence on the basis that it supported the veracity of Evans’ account that the complainant, while heavily intoxicated, retained the capacity to consent to the sexual activity, despite the fact that she retained no memory of it the next day.

The CCRC further opined that the evidence of the two witnesses might also have been admissible under s 41(3)(a) (that is, that the evidence was relevant to an issue other than consent) on the basis that it was relevant to establishing whether the defendant had a reasonable belief in consent at the material time. Although this avenue was not seriously pursued on appeal, consideration will be given to the relevance of this gateway below.

It is noted at this early stage that significant criticism may be levelled against the way in which the sexual behaviour evidence was obtained, namely following the instruction of ‘professional investigators’, ²⁷ and through the offer of a financial reward for anyone who could provide evidence relevant to the defence. ²⁸ Aside from potential questions about the veracity of such evidence, there are three further issues which arise in the context of evidence obtained in this manner, in particular in sexual offence cases. First, it is inexplicable how such a general, public request for information of this nature is compatible with the complainant’s right to anonymity. ²⁹ Secondly, a question arises regarding the weight of such evidence where, as in the Evans case, the evidence was obtained by the promise of a reward. In a different context, Parliament has already recognised that hearsay evidence of a complainant is not admissible, ‘if the complaint…was made as a result of a threat or a promise’. ³⁰ Thirdly, it is unlikely that most defendants would possess the financial resources to take such action which may lead to forms of what Baird refers to as ‘do-it-yourself sleuthing’ ³¹ and which could have a seriously deleterious effect on complainants. This concern is supported by the fact that the complainant in Evans was subject to disgraceful treatment including receiving significant online threats and abuse, with nine people convicted in November 2012 of offences under s 5 of the Sexual Offences (Amendment) Act 1992 in respect of publishing the complainant’s identity on social media. ³²

Notwithstanding these issues, the Court of Appeal was faced with the existence of potentially relevant evidence. Absent an exclusionary discretion, ³³ the court was therefore required to determine the admissibility of such evidence against the existing statutory framework, s 41 YJ&CE Act 1999.

The Admissibility of Sexual Behaviour Evidence Under s 41 YJ&CEA 1999

The admissibility of sexual behaviour evidence gives rise to competing interests in sexual offence proceedings. Sexual offence complainants should not be subject to unnecessary questioning or cross-examination about their previous sexual conduct. Equally, the defence must be able to adduce cogent evidence to assist its case particularly where such evidence is central to the jury’s understanding of a key element of the case. The enactment of s 41 of the YJ&CEA 1999 followed extensive criticism of its predecessor, s2 of the Sexual Offences (Amendment) Act 1976, ³⁴ itself introduced following sustained criticism of the common law position, near universally condemned for its perceived failure to regulate effectively the introduction of sexual behaviour evidence. Often, questions or evidence concerning previous sexual conduct served little purpose other than to besmirch the moral character of the complainant in the eyes of the jury. ³⁵

The 1976 Act’s replacement marked a radical departure from its predecessor in two respects. First, s 41 of the 1999 Act makes no distinction between evidence of previous sexual behaviour between the complainant and the accused and that occurring between the complainant and a third party. ³⁶ Secondly, the 1999 Act, unlike the 1976 Act, all but eradicated judicial discretion to admit sexual behaviour evidence other than that permitted through four narrow exceptions. Admissibility is determined through strict gateways that have the effect of excluding evidence that does not reach the threshold of admissibility under one or more of the exceptions in s 41. This is so even if the evidence is otherwise relevant and potentially cogent. ³⁷ Like its predecessor, s 41 provoked considerable controversy. Initially, however, the criticisms levelled at s 41 were that it swung the balance too far in the opposite direction. Where s 2 attracted heavy criticism for being too generous to defendants, s 41 was initially described as having surpassed its legislative aim of protecting complainants from harassment in the courtroom by excessively curtailing the defendant’s right to adduce potentially vital cogent evidence. Indeed, the severity of the restrictions imposed by s 41 became the focus of an almost immediate challenge in the House of Lords in 2001 ³⁸ (discussed further below). The result was a requirement that s 41 be interpreted in a manner compliant with art 6 ECHR. Writing at the time, McEwan observed that, ‘it was a fair bet that the amendments to the rape shield comprised in section 41 would be an early target for a challenge under Article 6(1)’, ³⁹ but for some reformers, the interpretation of s 41 by their Lordships has rendered the provision difficult to apply in practice. ⁴⁰

Key Gateways to Admissibility

Section 41(1) sets out the basic prohibition on introducing sexual behaviour evidence:

If at a trial a person is charged with a sexual offence, then, except with the leave of the court—

(a) no evidence may be adduced, and

(b) no question may be asked in cross-examination:

by or on behalf of any accused at the trial, about any sexual behaviour of the complainant.

Only (iii)—and to a lesser extent (i)—were relevant in the context of the Evans decision. The relevance of both (i) and (iii) will, therefore, be considered below. It is important to stress that having satisfied one (or more) of the s 41 exceptions, this does not, on its own, guarantee that the evidence will be admitted. Additional restrictions or ‘hurdles’ must be surmounted. The evidence or questioning must constitute a specific instance of the complainant’s previous sexual behaviour ⁴⁶ and must not be adduced where the purpose (or main purpose) is to ‘impugn the complainant’s credibility as a witness’, ⁴⁷ and finally, the judge must not allow the evidence to be heard unless a refusal to do so ‘might have the result of rendering unsafe a conclusion of the jury or (as the case may be) the court on any relevant issue in the case’. ⁴⁸ Collectively and on a literal interpretation, the s 41 exceptions and restrictions constitute very difficult hurdles to negotiate before evidence of the complainant’s sexual behaviour can be heard.

McGlynn acknowledges that, while it might occasionally be necessary to adduce sexual history evidence, ‘it has been demonstrated that many unnecessarily intrusive questions have been asked of complainants, often as a matter of routine; potentially therefore, in breach of the complainant’s Article 8 rights’. ⁴⁹ This echoes the comments of Lord Hutton in R v A (No. 2) itself, which, while not specifically framed in terms of art 8 rights, highlights additional considerations in rape trials which go beyond ensuring that the guilty are convicted and the innocent are not. He makes reference to a ‘third objective’ ⁵⁰ to ensure that complainants are treated with dignity in court and that irrelevant evidence or question is prohibited. It is clear that complainants in sexual offence cases are not only asked to recount, but are also scrutinised about, traumatic experiences that they otherwise might not choose to share, particularly in the public environment of a courtroom. As Baird observes, ‘complainants in sexual violence cases give evidence as a public duty in the interests of the community [but] sexual violence cases are often treated as if they are trials between the complainant and defendant personally’. ⁵¹

While it is not disputed that complainants have endured appalling treatment in the form of irrelevant and excessive cross-examination about their sexual histories, some of the humiliation and distress suffered in court may be unavoidable if the questioning can be attributed to relevant evidence which, if excluded, would affect the defendant’s rights under Article 6 ECHR. ⁵² The need for a careful balance to be struck between the legitimate protection of complainants and the right of the accused to a fair trial is therefore paramount. After all, Temkin suggested that,

the primary purpose of rape shield legislation is not to save the blushes of the complainant. It is to exclude evidence that is of little or no relevance and which serves only to distract the jury from the issues in the trial. ⁵³

The Application of the s 41 Gateways in R v Evans

Two s 41 exception(s) aroused particular controversy in Evans—the ‘similarity’ exception contained in s 41(3)(c) of the 1999 Act and the non-consent gateway in s 41(3)(a) which, despite not being given detailed consideration in Evans, was recognised as a potential route to admissibility. Section 41(3)(c) of the 1999 Act applies where the issue is consent and the sexual behaviour of the complainant to which the evidence or question relates is alleged to have been, in any respect, so similar—

(i) to any sexual behaviour of the complainant which (according to evidence adduced or to be adduced by or on behalf of the accused) took place as part of the event which is the subject matter of the charge against the accused, or

(ii) to any other sexual behaviour of the complainant which (according to such evidence) took place at or about the same time as that event:

that the similarity cannot reasonably be explained as a coincidence.

A key objection raised by McGlynn about the Evans decision concerns the perceived liberal interpretation of the similarity exception where, arguably, the complainant’s behaviour did not amount to conduct that was deemed to be ‘so’ similar as to be not coincidental. ⁶³ The evidence was that C had consumed alcohol, instigated sexual activity, directed her partner into certain position and used specific words of encouragement and, according to Evans, at the original trial was ‘enthusiastic, awake and gave no indication that she was not capable of consenting’. ⁶⁴ When considered in isolation, it is difficult to disagree with McGlynn who suggests that ‘the everyday can constitute sufficiently similar behaviour’. ⁶⁵ A more forceful argument could be advanced for admissibility when a holistic approach to the evidence is taken. With this in mind, it was, perhaps, a culmination of similarities, each in themselves unlikely to satisfy the threshold for admissibility but together able to surpass it. ⁶⁶ Moreover, the Court of Appeal acknowledged that, while ‘it may be a rare case, as Lord Steyn envisaged in A (No.2) in which it will be appropriate to indulge in…forensic examination of sexual behaviour with others’, ⁶⁷ this was ‘potentially such a rare case’. ⁶⁸

Sexual Behaviour Evidence and Reasonable Belief in Consent

In Evans, it was suggested by the CCRC that the evidence could potentially have been admitted under s 41(3)(a) on the basis that it was relevant to E’s reasonable belief that X consented. ⁶⁹ Section 41(3)(a) of the 1999 Act isolates the relevant issue as one that does not involve ‘consent’. Section 42(1)(b) defines ‘issue of consent’ as:

any issue whether the complainant in fact consented to the conduct constituting the offence with which the accused is charged (and accordingly does not include any issue as to the belief of the accused that the complainant so consented).

McGlynn and Thomason both argue that sexual behaviour evidence could never be relevant to the reasonableness of a defendant’s belief in consent if the defendant had no prior knowledge of the evidence. ⁷⁷ Thomason further suggests that lack of knowledge means that the evidence ‘logically could have had no bearing whatsoever on their state of mind at the time of the sexual contact in question’. ⁷⁸ While the evidence provided by O and H could not have affected Evans’s state of mind, because it was fresh evidence and not available at the original trial, its significance derives from the fact that two independent witnesses came forward to provide similar accounts of sexual activity with C that mirrored Evans’s account, an account which, as a result of her impaired memory, was not subject to a contrary account given by C. As Stark suggests, while the evidence of O and H does not confirm whether Evans’s belief in consent was reasonable—this relied on his appreciation of C’s capacity to consent—O’s evidence, at the very least, demonstrates that C could give an impression to sexual partners that she was capable of consenting to sexual activity which she later had no recollection of. ⁷⁹ The question posed is thus: in the absence of a contrary account offered by C, does C’s previous sexual behaviour with O, which to O’s surprise C had no recollection of, support Evans’s account that he reasonably believed that C was consenting? This is of particular relevance given that the reasonableness or otherwise of Evans’s belief appears to depend on the jury’s acceptance of the veracity of his account.

In the context of s 41(3)(c), Thomason makes a strong case for suggesting that ‘Evans should be considered as an ECHR gloss case, rather than a similarity one’. ⁸⁰ This approach should allay fears that the Evans decision opens the ‘floodgates’ ⁸¹ for the admissibility of sexual behaviour evidence through the similarity gateway. As Mirfield noted, ‘what can be done to section 41(3)(c) can also be done to any other element of section 41 deemed capable of denying the accused his fair trial right’. ⁸² Thus, given the absence of a contrary account to the version of events presented by Evans, it is not difficult to see how exclusion of O’s evidence, as it relates to E’s reasonable belief in consent, ‘might have the result of rendering unsafe a conclusion of the jury’. ⁸³

Admissibility of Third-Party Sexual Behaviour Evidence

Section 41 does not discriminate between third-party sexual behaviour evidence and that occurring between the complainant and defendant. ⁸⁴ While Thomason suggests that this represents an ‘unintended oversight’ ⁸⁵ by the legislators, the specific wording, ‘whether or not involving any accused or other person’, clearly implies a deliberate intention by Parliament that both categories should be subject to the admissibility framework created by s 41. McGlynn cites Mukadi ⁸⁶ as an example of the ‘entirely disconnected sexual activity with third parties [that] can be adduced in evidence’. ⁸⁷ The evidence in Mukadi, which had been refused by the trial judge, consisted of an earlier incident which involved the complainant standing on Oxford Street before getting into a car and driving off with an unknown man who was much older than her and with whom she had exchanged phone numbers. It is not disputed that this decision demonstrated a liberal interpretation of s 41 and is ‘hard to justify’. ⁸⁸ As Page and Birch suggest, even if the complainant was open to the idea of engaging in sexual activity with a third party in the earlier incident, ‘it by no means follows that she found M an equally pleasing prospect’. ⁸⁹ There are suggestions that the complainant’s earlier conduct did not amount to ‘sexual behaviour’ ⁹⁰ and therefore should not have engaged s 41. The Court of Appeal merely commented that that the evidence should have been admitted ‘whatever label one attaches to the conduct’, ⁹¹ because its exclusion may have led the jury to take a different view of the complainant’s evidence on the issue of whether she had consented to full intercourse with the defendant. This leaves open the question of what types of behaviour can or cannot be labelled as ‘sexual’ so as to engage s 41 and serves as another example of s 41’s vagueness and complexity. However, as discussed above, to suggest that Mukadi sets a benchmark for s 41’s conventional application is misleading. In his comprehensive review of the law and procedure which governs the prosecution of sexual offences in Northern Ireland, Sir John Gillen disagreed that third-party sexual behaviour evidence should never be admitted, ⁹² suggesting instead that, ‘stricter scrutiny should be applied for applications for such third party evidence and it should be considered admissible only in relatively few cases’. ⁹³

Section 41 departs significantly from its predecessor by subjecting sexual behaviour evidence to strict categories of admissibility. This approach stands in stark contrast to the approach of the 1976 Act, which afforded wide discretion to trial judges to admit evidence, not merely based on similarity. It is perhaps for this reason that Lord Steyn posited:

Evidence or questions about sexual behaviour with third parties is likely to be much harder to justify on grounds of relevancy than evidence about sexual behaviour with the defendant. Nevertheless I think that the draftsman was right to avoid laying down an absolute rule on this point. ⁹⁴

Birch had previously warned against the dangers of excluding potentially probative material and remarked that ‘we need to think of ways of taking the jury into the light, rather than deliberately keeping them in the dark’. ⁹⁶ Sexual behaviour evidence might, therefore, have considerable value as explanatory evidence and should therefore, be made available to assist juries so that the potential for wrongful convictions can be avoided. This, of course, assumes that sexual behaviour evidence is adduced for ‘legitimate purposes’ ⁹⁷ and not merely deployed as a mechanism through which to tarnish the complainant’s moral character in the eyes of the jury. ⁹⁸ In the context of Evans, as Dent and Paul make clear:

…it was not suggested that the evidence of the complainant’s sexual behaviour meant that she could not be believed. Instead, it was simply suggested that the evidence in question was relevant as to whether she had consented to the intercourse—and the reasonableness of the defendant’s belief that she had consented to the intercourse. The complainant’s evidence was that she could not remember what had happened and the prosecution’s case was that she did not consent. The complainant’s credibility was not in question. ⁹⁹

Does the Decision in Evans Set a Dangerous Precedent?

The Evans decision has been described as a ‘throwback to the last century when women who reported rape were assumed to be lying and their sex life was on trial’. ¹⁰¹ In the aftermath of the decision, particular concerns were voiced that the decision sets a precedent, paving the way for future courts to adopt a more defendant-generous interpretation of s 41. Dent and Paul have attempted to allay fears, arguing that ‘the floodgates have not—and will not be opened’, ¹⁰² however, McGlynn suggests that the decision ‘does not simply open the “floodgates” but risks a tsumani (sic)’. ¹⁰³ As discussed above, Mukadi ¹⁰⁴ is often cited as evidence to support claims that sexual behaviour evidence ‘continues to be admitted in a considerable number of trials’. ¹⁰⁵ Several recent empirical studies undertaken in this area appear to have reached contrary conclusions. ¹⁰⁶

These findings, including that that the introduction of sexual behaviour evidence by the defence is exceptional, ¹⁰⁷ seem to provide cogent evidence that s 41 sets a high bar for defendants to surmount. As the Court of Appeal in Evans observed:

It was an unusual case. The only witnesses as to sexual activity and the only evidence as to sexual activity came from the accused. Unlike McDonald, the defendant [Evans] could rely on little more than his own account of events in the bedroom to advance his defence of capacity to consent and actual consent. X could not assist; she said she remembered nothing. ¹⁰⁸

Does the Decision in Evans Provide a Case for Legislative Reform?

Despite that fact that the decision in Evans appears to have limited precedent value, the case clearly highlights concerns that the current approach to the restriction (and admission) of sexual behaviour evidence is not fit for purpose. McGlynn’s criticism that this is ‘an extremely (and unnecessarily) complicated area of law’ ¹⁰⁹ appears to be a valid one. The rigidity of the ‘strict categories’ approach to admissibility, the uncertainty of how to read the categories in a convention compliant manner (as required by the decision in R v A (No. 2)), combined with the application of the unsafe conclusion test make the provisions of s 41 difficult to apply with any degree of certainty, thus leaving the provisions vulnerable to criticism from both sides of the argument.

The challenge for any new legislative approach (for example based on a departure from the strict categories approach to sexual behaviour evidence) is to successfully balance the interests of sexual offence complainants and defendants. This is demonstrated by the criticism of the current s 41 regime, simultaneously lambasted for being both too restrictive ¹¹⁰ and not restrictive enough. ¹¹¹ The difficulty encountered by various jurisdictions has been achieving the correct balance. Given the polarising views in this area of the law, it is clear that ‘what one sees as common sense, another sees as nonsense’. ¹¹² It is surely axiomatic that complainants must not be unduly harassed in the courtroom and should be able to give their best evidence; however, a strict categories approach for assessing relevance is flawed if, as is the case with s 41, such an approach has the potential to violate the defendant’s right to a fair trial. Similarly, it is flawed in equipoise if it allows the admission of the very type of evidence it was enacted to restrict.

Against this backdrop, Canada experienced similar problems after amendments were made to its first ‘rape shield’ statute. ¹¹³ The changes incorporated a pigeonholed approach to admitting sexual behaviour evidence. ¹¹⁴ The Canadian rape shield legislation was redrafted in the light of the decision in Seaboyer; ¹¹⁵ however, the UK Parliament decided to persevere with s 41 in England and Wales despite it being subject to a similar challenge, shortly after its enactment, in the form of R v A (No. 2). ¹¹⁶

The Scottish Government, when undertaking a review of its rape shield laws, published a pre-legislative consultation document in 2000, ¹¹⁷ which put forward a number of recommendations in which the rape shield might be strengthened. This exercise resulted in the replacement of ss 274 and 275 of the Criminal Procedure (Scotland) Act 1995 by the Sexual Offences (Procedure and Evidence) (Scotland) Act 2002. This was an example of ‘second wave of rape shield legislation sweeping the world in response to the perceived failure of most initial attempts to operate as intended’. ¹¹⁸ It is interesting, however, that Scotland decided not to follow the s 41 model, considering it to be ‘too inflexible’. ¹¹⁹

Options for Reform

Stark has proposed a replacement for s 41 that would provide for the admissibility of sexual behaviour evidence only where

if the evidence were not admitted or the question not asked, the jury or (as the case may be) the court would find it impossible or difficult properly to understand other evidence in the case, and the evidence or question’s value for understanding the case as a whole is substantial. ¹²⁰

While sexual behaviour evidence can be indirectly relevant to an issue in the proceedings—which, in the context of evidence concerning a non-defendant’s character, is the approach taken by s 100(1)(a) of the CJA 2003 in terms of its value as important explanatory evidence—it can also be directly relevant. It is on this basis that Thomason identifies an apparent gap in Stark’s proposed provision. Section 41(3) YJCEA 1999 stipulates that the 1999 Act applies ‘if the evidence or question relates to a relevant issue in the case’, which s 42(1)(a) defines as ‘any issue falling to be proved by the prosecution or defence in the trial of the accused’. Thus, the relevant issue can be consent under ss 41(3)(b) and (c) or, as provided for by s 41(3)(a), an issue other than consent (which includes evidence of the defendant’s reasonable belief in consent). So, while Stark’s provision would remove many of s 41’s complexities, it appears that only evidence which is capable of providing context to the jury would be admissible. As Thomason suggests, perhaps the best answer is a wholesale rewrite of s 41; ¹²⁴ however, no substantive alternative is suggested to address the omissions in Stark’s proposal. What is needed is a new provision which not only remedies that gap but also departs from the model upon which s 41 is based which has rendered it an inherently complex piece of legislation, with ambiguous terminology. ¹²⁵ The new statutory framework, proposed herein, draws upon the experiences of a number of common law jurisdictions including Scotland which decided not to adopt the s 41 model during subsequent reforms of its rape shield laws.

Thomason draws parallels between the admissibility of sexual behaviour evidence under s 41 and bad character evidence under s 100 of the CJA 2003 on the basis that both involve the admissibility of evidence concerning past conduct to infer some aspect of present behaviour. Although Thomason acknowledges that the complex machinery of the s 41 gateways and restrictions do not compare favourably with the more flexible ‘substantial probative value’ exception in s 100(1)(b) of the 2003 Act, further difficulties would potentially arise if the test for admitting sexual behaviour evidence mirrored the admissibility threshold provided by the ‘character model’ in s 100 CJA 2003. Section 112(3)(b) of the CJA 2003 expressly provides that nothing in the Act ‘affects the exclusion of evidence’ under s 41 of the YJ&CEA 1999. Similarly, s 41(8) of the 1999 Act provides that ‘[n]othing in this section authorises any evidence to be adduced or any question to be asked which cannot be adduced or asked apart from this section’. Using the ‘enhanced relevance test’ contained in s 100(1)(b) as an example, this test amounts to a less onerous threshold for admissibility than the ‘unsafe conclusion test’ in s 41(2)(b) of the 1999 Act. ¹²⁶ Indeed, where both s 41 and s 100 apply, for example, where there is evidence that the complainant has falsely claimed previous sexual behaviour was non-consensual, and thereby potentially amounting to evidence of the complainant’s character, the more burdensome obstacle confronting any application to adduce the evidence will be s 41. ¹²⁷ Indeed, the Law Commission has said, if s 41 is satisfied then it is inconceivable that the evidence will not satisfy the requirements of s 100. ¹²⁸

Thomason suggests that it would be simpler if each admissibility regime at least used the same operative concepts; ¹²⁹ however, there would be serious concerns about applying the same test for the admissibility of sexual behaviour evidence as that applied for character evidence under s 100 of the 2003 Act. By moving to the s 100 model, admissibility of sexual behaviour evidence would be assessed by way of its ‘substantial probative value’ rather than the higher hurdle faced by defendants under s 41. Given that the current law is castigated for allowing too much sexual behaviour evidence into the courtroom under the higher test in s 41—despite recent empirical evidence suggesting the contrary ¹³⁰ —there is clearly little appetite for moving to the s 100 admissibility threshold such as that advanced by Stark. Thomason has explored the possibility of adding a new gateway to s 41, akin to s 100(1)(a), to accommodate relationship evidence, but recognises that additional gateways would merely compound s 41’s complexity. ¹³¹ Kibble noted the possibility of adding further exceptions to s 41, such as providing for motive to fabricate; prior false allegations; and evidence of prior sexual history with the defendant. ¹³² However, he also observed that ‘while open-ended rules are not the answer, neither are rules which permit only mechanical application and do not permit trial judges to exercise judgment’. ¹³³ Rather than move the admissibility threshold of sexual behaviour evidence to an s 100 type model, any new legislative framework should retain the same higher test of admissibility that currently exists under s 41, but which operates in a way that is easier to apply in practice.

The Overlap Between s 41 of the 1999 Act and s 100 of the 2003 Act

The spectre of false allegations of sexual assault has given rise to a number of rape myths that are based on a general distrust of sexual offence complainants. Section 41(4) of the 1999 Act is designed to protect complainants from irrelevant and humiliating questioning and to safeguard against the twin myths which might suggest that previous sexual behaviour was indicative both of consent and lack of credibility. ¹³⁴ But while s 41 places heavy restrictions on the defendant’s freedom to ask questions about the complainant’s previous sexual behaviour, a distinction is currently drawn between questions about previous sexual behaviour, which fall under the Act, and questions relating to the complainant’s lies about sexual matters, which do not. Unlike rape shield legislation enacted in some other jurisdictions, ¹³⁵ s 41 does not cater for the admissibility of evidence concerning the complainant’s character. ¹³⁶ Rather, where the defence wishes to adduce evidence of or question the complainant about previous false allegations, leave will be required under s 100 of the Criminal Justice Act 2003 if this amounts to evidence of the complainant’s bad character. ¹³⁷ However, there is potential for s 41 and s 100 to overlap where the previous complaint, which is allegedly false, involves an allegation concerning the complainant’s sexual behaviour.

It was decided in R v T;MH ¹³⁸ that as far as s 41 is concerned, questions or evidence regarding a sexual offence complainant’s previous false statements are not questions about the ‘sexual behaviour’ of the complainant, because they relate not to the complainant’s previous sexual behaviour but to past statements regarding such behaviour. ¹³⁹ Moreover, the courts have concluded that where the defendant does not have such an evidential basis for the suggestion that a previous complaint was false, the proposed evidence or questioning will be regarded as a cynical attempt to elicit sexual behaviour evidence. Essentially, this means that the evidence or questioning will fall foul of s 41(4) of the 1999 Act. It is for this reason that defence counsel must obtain a ruling from the judge that s 41 does not prohibit questions put to a complainant regarding alleged previous false complaints. As Keene LJ observed in R v T;MH, ‘the judge is entitled to seek assurances from the defence that it has a proper basis for asserting that the statement was made and was untrue’. ¹⁴⁰

Of course, such evidence is now subject to an enhanced relevance test in s 100 of the 2003 Act which would not have applied at the time of the R v T;MH decision. Thus, if the defence does not have sufficiently probative material to satisfy an evidential basis for suggesting that the previous complaint is false, the material will not have ‘substantial probative value in relation to a matter which […] is a matter in issue in the proceedings and […] is of substantial importance in the context of the case as a whole’. ¹⁴¹

It is conceivable that in a sexual offence trial where the complainant has made false allegations in the past, either s 41 (if the evidential basis test is not satisfied) or s 100 (if the evidence amounts to evidence of the complainant’s bad character) might be brought into play, or the circumstances of the case might require the deployment of both provisions, or, indeed, as will be demonstrated below, of neither. In R v V, Crane J suggested that ‘cross-examination about an allegedly false sexual offence allegation may require a ruling in relation to section 41 as well as leave under s.100. In many cases section 41 will be the more formidable obstacle to overcome’. ¹⁴² Further support for the proposition that s 41 constitutes a more onerous provision is provided by the Court of Appeal’s interpretation of s 100(1)(b) in R v Brewster. ¹⁴³ This has been described as a ‘significant but unsatisfactory judgment’, ¹⁴⁴ because the court gave a liberal interpretation concerning the types of previous convictions having ‘substantial probative value’ for the purpose of s 100(1)(b). Pitchford LJ remarked that:

[t]he question is whether the evidence of previous convictions, or bad behaviour, is sufficiently persuasive to be worthy of consideration by a fair-minded tribunal upon the issue of the witness’s creditworthiness. ¹⁴⁵

Where issues of credibility arise in circumstances in which s 41 applies, it must be remembered that, in line with Parliament’s intended aim of not perpetuating the myth that unchaste women are less likely to constitute credible witnesses, s 41(4) provides for a ban on attacking the complainant’s credibility where this is ‘the purpose (or main purpose)’ of the cross-examination. Thus, unless the accused can show that the purpose or main purpose of the cross-examination was not to discredit the witness, which might be difficult for the court to determine in a sexual assault trial where the distinction between issue and credibility can become blurred, the evidence will be deemed inadmissible under s 41(4). This assumes that the questioning relates to the complainant’s previous sexual behaviour and not merely that the complainant’s lies, which would engage only s 100.

The following scenario illustrates one possible approach that may be taken by the courts. The defendant (D) wishes to cross-examine the complainant (C) about a previous allegation of rape made by C, which D has some evidential foundation for suggesting was false in circumstances in which the 2003 Act’s definition of bad character is engaged. Keene LJ laid down a requirement in R v T;MH that, ‘[t]he defence, wishing to put questions about alleged previous false complaints, will need to seek a ruling from the judge that section 41 does not exclude them’. ¹⁴⁶ Given Crane J’s interpretation of R v T;MH in R v V that s 41 does not apply if the questioning ‘goes to the lies rather than to the sexual behaviour itself’, ¹⁴⁷ it is conceivable that the court could grant leave under s 100 on the credibility issue on the basis that C has made previous false allegations but refuse D’s application under s 41 to cross-examine C on the sexual behaviour aspects surrounding the lie. This seems a logical approach given the effect of s 41(4) in restricting D’s ability to discredit the complainant and of s 100 in allowing D to discredit C on the alleged lie. Aside from the obvious dissonance, the way in which the two sections can potentially overlap is unsatisfactory and arises because of s 41’s inability to cater for the various contexts in which false allegations might be made. Thus, any statutory replacement for s 41 should cater for the admissibility of evidence concerning the complainant’s character as well as evidence concerning the complainant’s sexual behaviour.

Since the enactment of the Criminal Justice Act 2003, the extent to which complainants can be questioned about false allegations is determined by an awkward overlap of provisions in the 2003 and 1999 Acts (s 100 of the 2003 Act also, effectively, incorporating an evidential basis test). As has been identified the s 100 admissibility test will, potentially, be the governing provision in false allegation scenarios in which the evidential basis test is satisfied, whereas in circumstances where cross-examination is concerned with the complainant’s sexual behaviour, s 41 is likely to determine whether the accused is granted leave, whether or not the more generous s 100 test is applicable. What is clear is that the overlap between these two provisions continues to cause confusion. The unsatisfactory state of affairs surrounding the overlap between the two provisions is illustrated by the following commentary:

Although bad character evidence strictly falls outside the ambit of the YJCEA, in the context of trials of sexual offences it will normally constitute material of the kind with which section 41 is primarily concerned: sexual behaviour having an adverse bearing upon the complainant’s credibility. This is, of course an ambiguous area since the circumstances will naturally vary from one case to the next. However, statutory reform to remedy this issue could fall foul of the same criticism as met section 41: that trying to fit particular cases into predetermined, exhaustive categories of relevancies is not feasible. The problems cast up by the provision are capable of being remedied, therefore, not by statutory revision but rather by additional guidelines. ¹⁴⁸

Second, the authors refer to a fundamental criticism of s 41 that ‘trying to fit particular cases into predetermined, exhaustive categories of relevancies is not feasible’. On this point, however, a strict categories approach to admitting evidence is just one of a number of statutory models. This article advocates it is possible to cater for the admissibility of both sexual history and character evidence through more logical means than those which are provided for at present in the cumbersome overlap between the 1999 and 2003 Acts, namely in the form of a combined admissibility framework.

Finally, it is submitted that the implementation of a new rape shield statute, which departs from a strict categories approach to admitting evidence, is not necessarily bound to suffer the same criticism as s 41. Thus, the proposition that the present law can only be remedied ‘not by statutory revision but rather by additional guidelines’ is not accepted.

An Alternative Legislative Framework?

The proposed legislative framework, below, incorporates good practice from the Scottish, New Zealand and Canadian rape shield statutes. ¹⁵⁰ The new framework would depart from the strict categories approach to sexual behaviour evidence adopted by s 41 but would retain some of the important restrictions contained therein. The new provision aims to address the primary flaw with s 41, which is not the threshold at which sexual behaviour evidence is admitted, but rather s 41’s complexity of application in practice. The new provision would thus read as follows:

1. Restrictions on evidence about the sexual behaviour and character of the complainant in sexual offence proceedings.

(1) If at a trial a person is charged with a sexual offence, ¹⁵¹ [no evidence can be given and no question can be put to a witness relating directly or indirectly to] ¹⁵²

 (a) [any sexual behaviour of the complainant]; ¹⁵³ or

 (b) the character of the complainant (whether in relation to sexual matters or otherwise), ¹⁵⁴

except with leave of the court.

(2) The court may, on application made to it, admit such evidence or allow such questioning as is referred to in subsection (1) if:

 (a) the evidence or question relates to a specific instance (or instances) of the complainant’s sexual behaviour ¹⁵⁵ [or to specific facts ¹⁵⁶ demonstrating the complainant’s character] ¹⁵⁷ and

 (b) exclusion of the evidence might have the effect of rendering unsafe a conclusion of the jury (or as the case may be) the court. ¹⁵⁸

(3) In determining whether leave should be given under subsection (1), above, the court must have regard to—

 (a) the right of the accused to make a full answer and defence ¹⁵⁹ and

 (b) society’s interest in encouraging the reporting of sexual […] offences. ¹⁶⁰

(4) Subject to the requirements of s.1(2)(a), the court may admit evidence to rebut or explain evidence under subsection (1) above [if the evidence or question, in the opinion of the court, would go no further than is necessary to enable the evidence adduced […] to be rebutted or explained]. ¹⁶¹

(5) [Where cross examination of a sexual offence complainant is allowed under this section, the court] ¹⁶² must, [notwithstanding the terms of its decision under subsection (1) above limit the extent of questioning to be allowed] ¹⁶³ to that which is necessary to comply with the requirements imposed by s.1(2)(b) above.

(6) No agreement between parties to criminal proceedings may render admissible [any evidence to be adduced or any question to be asked which cannot be adduced or asked apart from this section]. ¹⁶⁴

(7) [Nothing in this section authorises any evidence to be adduced or any question to be asked which cannot be adduced or asked apart from this section]. ¹⁶⁵

A ‘Combined Admissibility Framework’ for Sexual Behaviour and Character Evidence

The legislative provisions proposed above would act as a combined admissibility framework, dealing both with the admissibility of sexual behaviour evidence and evidence of character in sexual offence proceedings. Under the new approach, s 100 of the 2003 Act would not apply to evidence or questioning concerning the bad character of sexual offence complainants. Section 100 would have no application where the complainant’s character is in issue and, as currently provided for by s 41, ‘the person is charged with a sexual offence’. This would require an amendment to the bad character provisions of the 2003 Act. Section 112(3)(b) of the 2003 Act, ¹⁶⁶ which applies to s 41, would be repealed and a new subsection to s 100 would be inserted, which would provide that, ‘nothing in this section authorises any evidence to be adduced or any question to be asked about the bad character of a witness where the witness is a sexual offence complainant in proceedings relating to that offence’.

Section 100(3) of the 2003 Act, which identifies a number of factors that must be considered to assess the probative value of the evidence, such as the ‘nature and number of the events, or other things, to which the evidence relates’, and ‘when those events or things are alleged to have happened or existed’ would have no application. This is because the ‘unsafe conclusion test’ adopted from s 41 would apply to both sexual behaviour evidence and character evidence, thereby providing an adequate safeguard by imposing a higher admissibility threshold than that imposed by s 100. The judge must not admit either type of evidence (sexual behaviour or character) unless refusing to do so might have the effect of rendering unsafe a conclusion of the jury. Restrictions on evidence relating to specific instances of sexual behaviour or character evidence, ¹⁶⁷ and the purpose for adducing that evidence, ¹⁶⁸ could be retained, and thereby reinforcing the position that the new approach would impose a higher test of admissibility. Along with strong adherence to the Criminal Procedure Rules ¹⁶⁹ and the Criminal Practice Direction, ¹⁷⁰ which provides more detailed guidance to judges and advocates for adducing sexual behaviour evidence, this should assist in the prevention of direct and covert attacks on the complainant’s character that had long been an unfortunate feature in sexual offence proceedings. This approach to admitting evidence of the complainant’s sexual behaviour and character provides a more logical basis for admitting both types of evidence than that provided for at present which may require the interplay of s 41 and s 100. ¹⁷¹ It would also remedy a potential problem which the court implicitly identified in R v V ¹⁷² when it suggested that cross-examination about an allegedly false sexual allegation may require a ruling in relation to s 41 as well as leave under s 100. ¹⁷³ Under a revised version of s 41, the judge could give leave to admit some evidence of the complainant’s character but refuse admission of false complaint or sexual behaviour evidence if it fails to satisfy other restrictions in s 41. Likewise, in the context of the complainant’s bad character, the exclusion of evidence concerning a previous false allegation or a previous conviction might not ‘render unsafe a conclusion of the jury’ if it amounted to an isolated episode which had occurred some years previously.

Explanation of Key Provisions

Conclusion

The Evans case is both unusual and challenging in equal measure. The prosecution of a high-profile defendant, the manner in which the evidence of the complainant’s sexual behaviour came to light and the deplorable treatment of the complainant (in particular via social media) was compounded by the media storm following the defendant’s successful appeal and subsequent acquittal. This complex factual background is precisely the reason that a dispassionate assessment of the effect of the Court of Appeal’s decision in Evans is imperative.

In respect of the evidence of the complainant’s previous sexual behaviour, acknowledging the potential difficulties created by the way in which the evidence was gathered, the court, in possession of that evidence, was nevertheless faced with a series of clear questions. Was the evidence of the complainant’s previous sexual behaviour relevant to the issues of consent; was it sufficiently ‘similar’ so as to pass through the gateway in s 41(3)(c)(i) YJ&CEA 1999 (as read in the context of the decision of the House of Lords in R v A (No. 2)), and, if so, would exclusion of the evidence have had the potential to render unsafe any conclusion of the jury? Alternatively, the court may have considered admitting the evidence on the basis that it was relevant to an issue other than consent, namely the reasonableness of the defendant’s belief, under s 41(3)(a) of the 1999 Act.

To answer these questions, consideration must be given to the purpose of the defence in seeking to adduce the evidence of the complainant’s previous sexual behaviour. No attempt was made by the defence to undermine the credibility of the complainant. In fact, this is a rare case in which the complainant offered no opposing account of what occurred; she simply had no memory of the incident in question. Absent a contrary narrative, the jury were faced with a stark question; was the defendant telling the truth when he asserted that the complainant, ‘was perfectly capable of exercising her choice as to whether to engage in sexual activity and took the lead in most of what took place’, ¹⁸⁶ or did they accept the prosecution version of events that the complainant’s level of intoxication meant that she was ‘incapable of behaving in that way’? ¹⁸⁷ Acknowledging the case of Dougal ¹⁸⁸ in which evidence of the complainant’s lack of memory due to intoxication appears to have proved fatal to the prosecution, the complainant’s lack of memory could be seen as a somewhat damning piece of evidence against the defendant’s account that the complainant ‘was enthusiastic, awake and gave no indication that she was not capable of consenting’. ¹⁸⁹ That is, unless the jury were able to hear that, while in a similarly intoxicated state the complainant engaged in consensual sexual activity of which she retained no memory the next day. Criticisms have been levelled against the determination that the complainant’s behaviour (for example, words of encouragement used and/or sexual positions adopted) were sufficiently similar so as to meet the requirements of s 41(3)(c)(i). However, as identified above, the level of similarity must be read in the light of the decision of the House of Lords in R v A (No. 2), and therefore, in a manner consistent with the defendant’s fair trial rights under Article 6 ECHR. ¹⁹⁰

An unusual aspect of the Evans case was that the evidence of previous sexual behaviour related to sexual activity with third parties rather than with the complainant. It is arguable that the veracity of Evans’ account was strengthened by corroborative third-party evidence, of which he had no knowledge, when that account was offered. Given the specific peculiarities of the facts of the Evans case, including its position as one of Lord Steyn’s ‘rarest cases’ ¹⁹¹ in which evidence of sexual behaviour with a third party is relevant, it is hard to avoid the conclusion that Evans was a case decided on its own facts and which, therefore, sets no precedent, other than to provide a rare example of admissible third-party sexual behaviour evidence.

The review of cases involving applications to admit evidence of a complainant’s sexual behaviour conducted by the Attorney-General’s Office and the Ministry of Justice and the acknowledgement that cases such as Mukadi, ¹⁹² and latterly Evans, are ‘outliers’ tend to suggest that s 41 YJ&CEA 1999 is operating as intended. On that basis, the Evans case does not appear to provide a sound basis on which to pursue legislative reform in this area. In addition, the newly implemented Practice Direction, ¹⁹³ which provides more detailed guidance to judges concerning decisions to admit sexual behaviour evidence, dispels the notion that such evidence will be easily admitted. This view is affirmed by research undertaken on behalf of the Criminal Bar Association; ¹⁹⁴ the findings of which were highly critical of previous empirical studies on s 41’s application. ¹⁹⁵ The Report, which considered the responses of 144 barristers, indicated that there was no support for amendments which would render s 41 even more restrictive than its current position. ¹⁹⁶

Much of the damage to the public perception of how the criminal justice system deals with allegations of serious sexual offences in general, and with complainants specifically, arose not from the decision of the Court of Appeal in Evans, but from the reporting of it. The way in which the judgment was presented by some elements of the news media (including social media) lacked clarity, nuance and understanding, and likely achieved the opposite of what s 41 YJ&CEA 1999 was designed for, to encourage victims of sexual offences to come forward and give evidence. Ensuring the accuracy of media narratives appears to be a consistent challenge in respect of sexual offence proceedings. This has again been highlighted by CPS, referring to the ‘serious inaccuracies’ ¹⁹⁷ in media reports about the use of complainant’s mobile phone data in sexual offence investigations.

As established, the Evans decision does not appear to be representative of the normative operation of s 41. On the other hand, although the decision does not provide a rationale for wholesale reform of s 41, it serves a useful purpose by highlighting the practical difficulties of applying the legislation. It is the complexity of s 41 that is problematic rather than the standard at which sexual behaviour evidence is considered sufficiently relevant to warrant its admission, which, according to recent research, ¹⁹⁸ is correct. The logical approach is to maintain the higher hurdle already provided by s 41, rather than to impose a lower threshold of admissibility as would result from the introduction of a model based on s 100 CJA 2003 as proposed by Stark, but to simplify the operation of the legislative provisions in practice. This could be achieved by the enactment of a combined admissibility framework for both sexual behaviour and character evidence based on the proposals advanced herein.