How should English law deal with deceptive sex?

The SOA

Before the coming into force of the SOA on 1 May 2004, it was clear that a man performed the actus reus of rape if he had ‘unlawful sexual intercourse’ with a woman who was, at the relevant time, not consenting. ¹⁴¹ It was also clear, following the EWCA's decision in R v Olugboja, ¹⁴² that non-violent coercion, as well as force or violent threats, could render a complainant's ‘consent’ unreal. Less clear, however, was when precisely consent would be absent. For, while the offence of rape was defined in statute, ¹⁴³ ‘consent’ was not; and, while Olugboja attempted to give some guidance about the meaning of this term, the Court ultimately conceded that – at least in cases ‘not involving violence or the fear of it’ – ‘the dividing line … between real consent on one hand and mere submission on the other may not be easy to draw’. ¹⁴⁴ ‘Where it is to be drawn in a given case’, the Court concluded, ‘is for the jury to decide’. ¹⁴⁵ Predictably enough, this led to ‘absence of clarity and consistency as to the meaning of consent’ ¹⁴⁶ – and, eventually, to a recommendation in a review carried out in 2000 for the Home Office that greater statutory detail be provided about this matter. ¹⁴⁷ According to the relevant report: ‘rape and sexual assault are primarily crimes against the sexual autonomy of others’; ¹⁴⁸ ‘consent’ should be defined in law as ‘free agreement’; ¹⁴⁹ and the relevant statute should also provide for a non-exhaustive list of circumstances in which consent is absent. ¹⁵⁰ Relevantly, two of those proposed circumstances would cover cases where, respectively, the complainant was ‘deceived as to the purpose of the act’ or ‘was mistaken or deceived as to the identity of the [accused] … or the nature of the act’. ¹⁵¹

In 2002, the government accepted that legal clarity about ‘what consent means’ is ‘vital’ ¹⁵² and stated its intention ‘to make statutory provision on this issue that is clear and unambiguous’. ¹⁵³ Likewise, it said that its new legislation would provide that, consistently with the then current legal position, ¹⁵⁴ a person would be taken not to have been consenting to sexual activity where she was deceived as to its ‘nature’, or where her ‘consent’ was ‘induced by impersonation of someone else’. ¹⁵⁵ Two years later, the SOA delivered on these promises – although, in certain respects, that Act deviates from the approach recommended by the Home Office review.

Between them, ss 1–4 of the SOA create four non-consensual sexual offences. According to s 1(1), a person commits rape if, without a reasonable belief that the complainant is consenting, he penetrates with his penis the vagina, anus or mouth of a non-consenting person; and s 2(1) likewise exposes a person to the possibility of life imprisonment if, without consent and a reasonable belief in it, she sexually penetrates the vagina or anus of another person with something other than a penis. ¹⁵⁶ Section 3(1) provides that a person is guilty of sexual assault if she intentionally sexually touches a non-consenting person and lacks a reasonable belief that he is consenting; and s 4(1) makes it criminal for a person intentionally to cause another, non-consenting, person to engage in sexual activity ¹⁵⁷ – unless, relevantly (and ignoring the onus of proof), he reasonably believes that that person is consenting. The maximum penalty for the s 4(1) offence depends on the intrusiveness of the non-consensual sexual conduct that the accused has caused. If the Crown can prove that he has caused conduct of the type caught by ss 1 and 2 offences, his crime will be punishable by life imprisonment. If, on the other hand, the accused has caused non-penetrative sexual touching, the maximum penalties are the same as those for the s 3(1) offence. ¹⁵⁸

‘Consent’ for the purposes of each of these four offences is defined in s 74, although in terms that differ from those recommended by the Home Office review. ¹⁵⁹ According to the section, a person consents if she ‘agrees by choice, and has the freedom and capacity to make that choice’. Moreover, as essentially recommended by the Home Office review, and consistently with the government's 2002 assurance, s 76 provides for two circumstances in which consent is conclusively presumed to be absent (although the section adds that the accused is also conclusively presumed to have the requisite mens rea). Section 76 broadly ¹⁶⁰ reflects the common law position as stated in cases such as Linekar and Papadimitropoulos. If the defendant either ‘intentionally deceived the complainant as to nature or purpose of the relevant act’ ¹⁶¹ or ‘intentionally induced the complainant to consent to the relevant act by impersonating a person known personally to the complainant’, ¹⁶² she will be guilty of the relevant non-consensual sexual offence.

Finally, it is necessary to note s 75. While the Home Office review had of course recommended the creation of a non-exhaustive list of circumstances in which consent was absent, s 75 adopts a different approach. Like s 76, the section creates certain presumptions – though, unlike the s 76 presumptions, these presumptions are evidential, not conclusive. If, for example, at the time of the relevant sexual activity or immediately before it, any person was using violence against the complainant, ¹⁶³ or if at the time of the act the complainant was asleep or unconscious ¹⁶⁴ or unlawfully detained, ¹⁶⁵ the accused will be presumed to be guilty of the relevant non-consensual sexual offence – unless ‘sufficient evidence is adduced’ to raise an issue about consent and/or the accused's reasonable belief in it. Once sufficient evidence has been adduced about both matters, however, the s 75 presumptions have no application – which has led one commentator to describe them as ‘about as forceful as a feather’. ¹⁶⁶ We shall return to this point in part 4, when discussing how the SOA might be reformed.

The deceptive sex case law under the SOA

What is the precise reach of s 76 of the SOA? Further, if a complainant participates in sexual activity because of a different deception from those to which s 76 refers, or a mistake or misapprehension that the accused has not intentionally caused her to make, in which circumstances, if any, will s 74 render such participation non-consensual? At the time of the SOA's enactment, there was much uncertainty about how these questions should be answered ¹⁶⁷ – but, in the intervening years, a greater level of clarity has been achieved. That said, one criticism of Lawrance is that its answer to the second of these questions leaves the law in an unpredictable state. ¹⁶⁸

If we deal first with the s 76 categories, s 76(2)(b) ‘extends the old law beyond cases of impersonating spouses or partners’, ¹⁶⁹ but in one way takes a narrower approach than does the common law. The provision applies where the complainant has participated in sexual activity because of his mistaken belief that the accused is his spouse or partner, or someone else whom he actually knows (although not where, e.g. the accused has personated ‘a film star or sports star whom the accused has not met’. ¹⁷⁰ ) But, in contrast to the position at common law, it requires the accused to have ‘intentionally induced’ the complainant to participate by personating someone known personally to him. In other words, where the complainant is mistaken as to the accused's identity, but this mistake has been caused by something other than the accused's intentional deception, s 76(2)(b) has nothing to say about whether she is consenting. ¹⁷¹

Section 76(2)(a) has been said to ‘re-state … and slightly extend the common law’ ¹⁷² – a reference, it seems, to its providing that consent is absent where the accused has ‘intentionally deceived’ the complainant as to the ‘nature or purpose’ of the act. But, as with s 76(2)(b), s 76(2)(a) applies only if there has been an intentional deception – and, more significantly, it is unclear when precisely the accused will have intentionally deceived the complainant as to the ‘purpose’ of the act. ¹⁷³ No doubt, there will have been such deception in a case such as Mobilio – or R v Green ¹⁷⁴ – where the accused deliberately causes the complainant mistakenly to believe that his purpose is medical, not carnal. Seemingly, too, s 76(2)(b) will apply where the accused intentionally leads the complainant to believe that his purpose is to assess ‘her modelling statistics’ ¹⁷⁵ (when it is really to gain sexual gratification) ¹⁷⁶ or to obtain sexual gratification (when it is really to humiliate the complainant). ¹⁷⁷ But the position is perhaps less clear where the accused has dual purposes. Consider a case where a private investigator engages in sexual activity with a sex worker because she has been instructed to determine whether the ‘massage business’ at which the sex worker is working is in fact a brothel. ¹⁷⁸ If the investigator has two purposes in so acting – to obtain sexual gratification and information about the complainant's employer – has she intentionally deceived her as to the ‘purpose’ of the act? The better view would seem to be that she has not – it would seem to be enough that the sex worker knows of one of the accused's purposes ¹⁷⁹ – but the matter is not free from doubt. That said, the Courts have made it clear that, because, once it operates, s 76 will normally mandate a finding of guilt, it ‘will be a rare case in which [it] … should be applied’. ¹⁸⁰

We can now move to the second question posed above. It is now clear that s 76 does not ‘cover the field’ insofar as deceptive sexual activity is concerned. In other words, if a complainant participates in sexual activity because of her mistake or misapprehension about some matter, but this mistake or misapprehension was not caused by any of the deceptions with which s 76(2) deals, the Crown might be able to use s 74 to prove that she was not consenting. This seemed ¹⁸¹ to be the effect of the EWCA's decision in R v Jheeta, ¹⁸² a case in which the appellant, concerned that the complainant (his girlfriend) would leave him and posing as various police officers, told her via text message that the appellant was suicidal and that ‘she should do her duty’ and have sexual intercourse with him (and that she would be liable to a fine if she did not). ¹⁸³ The sex that ensued was non-consensual, the Court held, because of s 74. ¹⁸⁴ While the accused had not deceived the complainant about the ‘nature or purpose’ of the acts ¹⁸⁵ (or his identity), the complainant had not made a ‘free choice’ ¹⁸⁶ to participate in them. Such an approach was confirmed in Assange v Swedish Prosecution Authority – and for much the same reasons. If, as the complainant had alleged, the appellant had surreptitiously removed the condom that she insisted he wear during sexual intercourse, it could be argued that he had deceived her as to the ‘nature’ of the act. ¹⁸⁷ But, the Court held, ‘s 76 should be given a stringent construction, because it provides for conclusive presumptions’. ¹⁸⁸ ‘The question of consent in the present case’, their Lordships concluded, ‘is to be determined by reference to s 74’. ¹⁸⁹ If the allegations could be proved, it would ‘plainly be open to a jury’ to convict the appellant of rape. ¹⁹⁰ Surely, the Court said, it was Parliament's intention that both non-deceptive and deceptive conduct could be taken into account when assessing whether a person had consented within the meaning of s 74. ¹⁹¹

Where the latter kind of conduct is at issue, however, exactly how far does s 74 reach? In R v B, ¹⁹² the EWCA had held that, where the accused has a sexually transmissible disease and fails to disclose this to a sexual partner, ‘any consent that may have been given … by the other party is not thereby vitiated’. But, in R(F) v Director of Public Prosecutions, the Divisional Court concluded without much hesitation that the claimant's consent was ‘negated’ ¹⁹³ if she had expressly conditioned her participation in sexual intercourse on her husband's not ejaculating inside her, and if – unbeknownst to her, and at least some time before he did ejaculate inside her – he formed an intention to breach this condition. ¹⁹⁴ The Court noted the terms of s 74. ¹⁹⁵ It thought that Assange ‘underlined’ that ‘“choice” is crucial to the issue of “consent”’. ¹⁹⁶ And it held that, in cases such as the one with which it was dealing, the ‘evidence relating to “choice” and the “freedom” to make any particular choice must be approached in a broad common sense way’. ¹⁹⁷ Two months later, in R v McNally, ¹⁹⁸ this approach reached its high-water mark.

In McNally, the defendant had been convicted of six counts of s 2 SOA offending, in circumstances where she had digitally penetrated a girl who said that she participated in such activity because she mistakenly believed that McNally was a cisgender boy. ¹⁹⁹ In upholding the convictions, the EWCA rejected a defence argument that seems very similar to the approach taken subsequently in Lawrance. According to that argument, a distinction had to be drawn between ‘deceptions as to qualities or attributes’ and deceptions ‘as to the features of the act itself’. ²⁰⁰ A deception as to age, marital status, wealth or HIV status, it was said, was a deception of the former kind. ²⁰¹ So was a ‘deception as to gender’ ²⁰² such as had occurred here. But the deceptions in Assange and R(F) were different. ²⁰³ Those deceptions were deceptions ‘as to the features of the act itself’. Only these latter deceptions, the argument concluded, were capable of negating consent. ²⁰⁴

In rejecting this argument, Leveson LJ made one noteworthy suggestion and one important statement of principle. The suggestion was that, when dealing with the question of consent under s 74, cases of ‘[a]ctive deception’ ²⁰⁵ might have to be treated differently from cases of non-disclosure. Whereas the appellant's failure to disclose his HIV infection in B did not vitiate the complainant's consent, his Lordship noted, the position might have been different if he had ‘positively assured’ her that he was not HIV+. ²⁰⁶ The statement of principle was that, while certain deceptions – ‘such as, for example in relation to wealth’ – will ‘obviously’ not vitiate consent, the remarks in R(F) about ‘choice’ and ‘freedom’ and ‘common sense’ provided the key to the present case. ²⁰⁷ The physical acts here, Leveson LJ conceded, were no different from those to which the complainant had agreed. ²⁰⁸ But ‘on any common sense view’, he said, ‘the sexual nature of the acts is … different where the complainant is deliberately deceived by a defendant into believing that the latter is a male’. ²⁰⁹ The complainant here had not freely chosen ‘to have a sexual encounter with a girl’. ²¹⁰

In Lawrance, however, the EWCA disapproved Leveson LJ's suggestion and distanced itself from his statement of principle. In that case, the Crown argued that Lawrance's positive deception concerning his fertility was materially indistinguishable from the deceptions in Assange and R(F). As in those cases, it was said, the complainant ‘sought to avoid the risks of pregnancy’. ²¹¹ And, as in those cases, ‘the deception deprived the complainant of having a free exercise of choice for the purposes of section 74’. ²¹² But the EWCA thought that Assange and R(F) could be distinguished. ‘Unlike the woman in Assange, or in F’, the Court ruled, ‘the complainant agreed to sexual intercourse with the defendant without imposing any physical restrictions. She agreed to penetration of her vagina and to ejaculation without the protection of a condom’. ²¹³ And that, said the Court, was enough. Their Lordships quoted ²¹⁴ a passage from Monica, ²¹⁵ in which Lord Burnett CJ – who also authored the Lawrance decision – had indicated that the R(F) and McNally ‘broad common sense’ approach was insufficiently rigorous and excessively vague. And they associated themselves ²¹⁶ with the Monica statement that, in the SOA deceptive sex case law, a principle could be discerned. That principle is that, in a deception case not covered by s 76, a person will be consenting within the meaning of s 74 unless the deception is ‘closely connected to the performance of the sexual act, or … [is] so fundamental, owing to that connection, that they can be treated as cases of impersonation’. ²¹⁷ Contrary to what Leveson LJ had suggested in McNally, their Lordships held that the deception need not be ‘express’. ²¹⁸ Passive deception resulting from non-disclosure is enough, so long as it is ‘sufficiently closely connected to the performance of the sexual act’. ²¹⁹ But the defendant's lie here, the Court concluded, had no such sufficient connection. ²²⁰ It related, not to the act, but to ‘to the broad circumstances surrounding it’. ²²¹ The defendant's rape convictions were set aside.

Criticisms of Lawrance

A statement made by the Court in Lawrance goes a long way towards demonstrating why that decision has been met with almost ²²² universal disfavour. ‘In terms of section 74’, their Lordships said, ‘the complainant was not deprived by the defendant's lie of the freedom to choose whether to have the sexual intercourse which occurred’. ²²³ Yet, of course, she was. The sexual intercourse that occurred in Lawrance was unprotected sexual intercourse with a fertile man. If the defendant had told her the truth about his fertility, the complainant would have chosen not to engage in such intercourse. Therefore, the defendant's lie about his fertility did deprive the complainant of her freedom to choose whether to participate. That said, not all of the criticisms of Lawrance have been justified. Contrary to what some have argued, as a matter of statutory interpretation the decision seems right. It is submitted that the real problem with Lawrance is normative. In other words, to accept that Parliament's intention was to criminalise roughly the same amount of deceptive sexual activity as Lawrance criminalises is certainly not to accept that it was wise so to provide. Put bluntly, the line that Lawrance holds Parliament to have drawn between criminal and non-criminal deceptive sex, is arbitrary – and it fails to maximise the protection of sexual autonomy.

It will have been observed that the Lawrance ‘closely connected’ principle ties the deceptions that will negate consent because of s 74 as closely as possible to the s 76 categories. According to it, deceptions about condom use, or an intention to ejaculate, are sufficiently similar to deceptions about the ‘nature … of the act’. It seemingly regards deceptions as to gender as closely analogous to cases of personation. ²²⁴ But why did the Court find that, on its proper construction, s 74 effects only a ‘modest’ ²²⁵ change to the common law position regarding deceptive sexual activity? What motivated this decision? The answer to this question is simple. Despite the broad terms of s 74, the SOA evinces no clear parliamentary intention to bring about a ‘profound’ ²²⁶ alteration to the law relating to sexual consent. Or, in the Court's words, ‘there is no sign that Parliament intended a sea change in the meaning of consent when it legislated in 2003’. ²²⁷ In the absence of such a clear intention, the Court found that no radical change was intended.

Murray and Beattie have criticised the EWCA's approach. ²²⁸ So too has Sir Richard Buxton, who has argued that it ‘impossible to agree’ ²²⁹ with the ‘sea change’ remarks just noted. But, with great respect to these commentators, the EWCA was right.

Murray and Beattie argue that, by tying the s 74 negating deceptions so closely to the s 76 categories, Lawrance has taken a wrong turn. Section 76, ‘not s 74’, they say, was intended to restate (at least broadly) the common law position regarding consent and deception. ²³⁰ Accordingly, ‘while the pre-2003 case law may be instructive for the purposes of interpreting s 76’, ²³¹ s 74 should be ‘taken on its own terms’. ²³² When regard is had to those terms, Murray and Beattie continue, it is hard to see why any restrictive approach should be taken to the question of which deceptions negate consent. ²³³ Murray and Beattie note that s 74's language places the focus ‘squarely on the choice of the complainant’, without qualifying this in any way. ²³⁴ Moreover, the authors conclude, the notion that the words of s 74 should be given their plain meaning is fortified by the ‘policy intention behind the 2003 reforms’. ²³⁵ It was noted above ²³⁶ that, in its 2000 report, the Home Office stated that ‘rape and sexual assault are primarily crimes against sexual autonomy of others’. ²³⁷ Murray and Beattie point to this statement, arguing that Parliament evidently intended that a person's choice not to engage in sexual activity be more fully protected than it is under Lawrance. ²³⁸

This argument is unpersuasive. It fails essentially because, as Munro has noted, terms such as ‘freedom’ and ‘choice’ ‘can be interpreted in radically divergent … ways’; ²³⁹ and there is barely anything in the extrinsic materials preceding the 2003 reforms to indicate that, when it used these terms in s 74 SOA, Parliament intended them to bear the meaning suggested by Murray and Beattie. Indeed, as I shall now argue, there is much in those materials that points in the opposite direction. In these circumstances, it was ‘sensible’ for the EWCA in Lawrance to find that s 74 was not intended to bring about ‘radical social reform’. ²⁴⁰

Sometimes, lawyers see statutory language – language such as ‘a person consents if he agrees by choice’ ²⁴¹ – and leap to the conclusion that it must be given its plain meaning. In fact, ‘even apparently plain words’ must be interpreted purposively. ²⁴² And it is far from clear that Parliament's purpose, when it enacted the SOA, was to deviate significantly from the common law's approach to deceptive sex. It is true that the Home Office made it clear in 2000 that, in its view, the law must protect sexual autonomy. But, when proposing that consent ‘should be defined in law as “free agreement”’, ²⁴³ it said absolutely nothing to suggest that it thought that such a definition would radically expand the circumstances in which a complainant's mistake or misapprehension negated her consent. Indeed, it almost immediately suggested that it held a very different view from this.

Once it had made the proposal just noted, the Home Office indicated that the law should provide some specific detail about what exactly ‘free agreement’ means. ²⁴⁴ ‘The law should set out’, it said, ‘a non-exhaustive list of circumstances where consent was not present’. ²⁴⁵ It was open to the Home Office to provide on its proposed list – as certain ‘Australian states’ ²⁴⁶ have provided on theirs – that consent is absent if the complainant participates in sexual activity due to ‘any fraudulent means’. ²⁴⁷ But it did not propose this. Rather, as noted above, ²⁴⁸ it merely proposed that the law should state that a person who is ‘deceived as to the purpose of the act’ or who ‘was mistaken or deceived as to the identity of the person or the nature of the act’ is not consenting. ²⁴⁹

When compiling its report, the Home Office was assisted by a Law Commission Report entitled Consent in Sex Offences. ²⁵⁰ In that Report, the Law Commission had recommended that ‘consent’ be defined in law as ‘subsisting, free and genuine agreement to the act in question’. ²⁵¹ But it clearly regarded much deceptive sex as being consistent with ‘free and genuine agreement’. The Commission thought that the law should provide that a person has not genuinely agreed to an act if his apparent agreement was ‘obtained by a deception as to the other's identity … or the nature of the act’. ²⁵² It also stated ²⁵³ that there should be a statutory reversal of Richardson. ²⁵⁴ But it rejected the idea that ‘gender deceptions’ should be capable of negating consent ²⁵⁵ – and it implied ²⁵⁶ that, in many cases where a person participated in sexual activity because of deception, the proper charge would be ‘procuring sexual intercourse by deception’. ²⁵⁷ The Home Office did not dissent from the latter view. Indeed, in proposing that the SOA should contain a procurement offence, ²⁵⁸ it clearly implied that it did not consider that its definition of consent would render very much deceptive sex non-consensual. The Home Office indicated that, where a complainant's participation in sexual intercourse was caused by ‘deception of a lower level’ than a deception as to the accused's ‘identity … or the nature or purpose of the act’ – as in Papadimitropoulos – the offence should be procurement, not rape. ²⁵⁹ And, crucially, in its White Paper, ‘which culminated in the SOA 2003’, ²⁶⁰ the government took a similar approach. While strangely silent about a procurement offence, ²⁶¹ the government stated that its intention was to enact a provision similar to the present s 76. ²⁶² No mention was made of any other deceptions. The relevant Minister did not mention deceptions at all in the Second Reading Speech for the Sexual Offences Bill. ²⁶³

In these circumstances, it seems clear that the ‘expansive’ interpretation of s 74 suggested by Murray and Beattie is not ‘the correct interpretation’. ²⁶⁴ Quite simply, Parliament evinced no clear intention to effect such a radical change to the existing law; ²⁶⁵ and, as Plaxton has explained, there are good reasons for the Courts to exercise restraint where this is so. Parliamentary sovereignty requires Parliament, not the courts, to decide what conduct is criminal. For, if the courts were too free to attribute to Parliament an intention to bring about ‘radical social change’, there would be a danger of their criminalising conduct that Parliament in fact had not set itself against. ²⁶⁶

This brings us to the normative objection to Lawrance. Again, even if it is accepted that the ‘closely connected’ principle draws the line between consensual and non-consensual deceptive sexual activity essentially where Parliament intended it to be drawn, that does not make such a solution right. Indeed, assuming that Lawrance's deception was not ‘sufficiently closely connected to the performance of the sexual acts’, but Assange's and F's alleged deceptions were, ²⁶⁷ why should that matter? More important, surely, is that in each of these cases the complainant's sexual autonomy was (allegedly) violated? In each, the complainant said that she explicitly conditioned her participation in sex on X. In each, the accused was alleged to have deceitfully created the impression that X was fulfilled. A principle that treats two of these allegations as allegations of rape, and the other as an allegation of non-criminal conduct, is not really a principle at all. It is an arbitrary rule that ignores the substance of what has happened. A truly principled approach – an approach that protects a person's exercise of her sexual autonomy – is required. That said, while it is widely acknowledged that the SOA must be reformed, ²⁶⁸ scholars fundamentally disagree about the direction that such reform should take.

A return to the pre-2003 approach

One prominent reform suggestion has been ‘to turn back the clock’ ²⁶⁹ – that is, ‘restrict the type of deceptions that can vitiate consent’ ²⁷⁰ and, possibly, reinstate the offence of procuring a person to engage in sexual activity by fraud. ²⁷¹ It is submitted, however, that this solution is undesirable. That is partly because, under it, much factually non-consensual sexual activity would be treated as consensual.

Rogers challenges this claim. According to him, a person generally is factually consenting when she participates in sexual activity because of her mistake or misapprehension about some matter. But he reaches this conclusion only after deploying a highly idiosyncratic account of what sexual autonomy involves. After conceding that ‘views about what “sexual autonomy” should mean may be disputed’, ²⁷² Rogers argues that a person's sexual autonomy is only violated when he is ‘used for the sexual gratification of another in a way’ that fails to show ‘regard to [his] … own sexual preferences’. ²⁷³ In other words, for Rogers, a complainant like the sex worker in Linekar is in fact consenting. She is consenting, he thinks, because she is willing to perform the precise act (penile-vaginal sex) that occurs. It is irrelevant that she is unwilling to perform that act with a person, like the accused, who lacks an intention to pay her.

This reasoning is misconceived. ²⁷⁴ It was noted above ²⁷⁵ that the common law's approach to ‘mistaken sex’ ²⁷⁶ was based on two outdated ideas – that the wrong in rape is defilement; and that rape involves force or the threat of it. Both ideas seem to influence Rogers’ account. According to Rogers, it was enough that the complainant in Assange was ‘prepared to offer … her partner … sexual stimulation from vaginal sex’. ²⁷⁷ In other words, by consenting to sex, she had – if we return to Rubenfeld for a moment – ‘done all of the consenting she needed to do’. ²⁷⁸ But, also, Rogers criticises the contention in McNally that the defendant's deception had removed the complainant's ‘freedom’. ²⁷⁹ This, he thinks, ‘makes a nonsense of the word “freedom”, which should be about the measure of resistance to perceived coercion (whether physical or otherwise)’. ²⁸⁰ But even if that were right, ²⁸¹ it would fail to establish what Rogers needs to establish if he is to demonstrate that sex in cases like McNally was consensual. It does not establish that a person who agrees to X but gets Y, has agreed ‘by choice’ ²⁸² (i.e., autonomously) to participate in Y.

There are other reasons why the clock should not be turned back. Let us focus for a moment on proposals essentially to reinstate the common law's approach to mistakes and misapprehensions that negate consent – and supplement this with a procurement offence. As Sjolin has argued, one problem with this solution is that, once the law says (however wrongly) that most deceptive or materially mistaken sex is consensual, the justification for the criminalisation of such conduct seems to disappear. ²⁸³ Take a case such as Lawrance. Under the approach that we are considering, the law would hold the complainant there to have been consenting. But if she was consenting, her sexual autonomy was not breached – and it seems that no other interest of hers was set back sufficiently to warrant the involvement of the criminal law. Of course, a person could object to what has just been argued by noting that, in fact, this complainant was not consenting. But that just returns us to the point suggested at the beginning of this section. If she was not consenting, surely the law should say so?

Another of Sjolin's arguments warrants attention. According to her, a ‘broad deception offence’ such as the procurement offence that we are considering ‘risks catching those who currently, rightly, are not criminalised’. ²⁸⁴ Sjolin seems to have in mind cases where, for example, the accused has procured the complainant's ‘consent’ by telling him that she is unmarried, is not having an affair, is wealthy, or is an ‘Arab … who was seeking no long-term relationship’. ²⁸⁵ If it were a crime to obtain sexual activity by ‘deception in any part of the world’, ²⁸⁶ these accused would be guilty of a serious offence. Sjolin evidently believes that this is not right. But this places her at odds with Herring (as she notes ²⁸⁷ ) – although, as we shall now see, Herring contends that such accused should be guilty of non-consensual sexual offending, not procurement.

Herring's proposal

In a well-known article, Herring noted that non-consensual sexual offences protect the right to sexual autonomy and that the person who has engaged in sexual activity due to a mistake about some matter has had her sexual autonomy violated. ²⁸⁸ Accordingly, he thought, a person should be guilty of a non-consensual sexual offence if the Crown could prove that he: (i) engaged in sexual activity with a materially mistaken complainant; and (ii) knew or ought to have known that the complainant was materially mistaken. ²⁸⁹ This proposal has been forcefully criticised. ²⁹⁰ Yet, as a matter of principle, I can see no flaw in it. At the risk of repetition, surely a person has made no genuine choice to engage in sexual activity if he would not have engaged in it had he known the truth?

That said, Herring's critics have rightly pointed out that most people would be unhappy with some of the consequences that his proposal would mandate. ²⁹¹ Some such cases have been noted in two paragraphs above. If a person induces another to participate in sexual activity by lying about her marital status, or about whether he is having an affair, or about her wealth or race – or, it can be added, about his biological sex at birth, ²⁹² or her feelings for the other person, or his sexual history – many would baulk at calling that person a sex offender. And this, it is submitted, is significant. It is of the utmost importance that the law is as principled as possible. Where possible, that is, it should treat a person who has had non-consensual sex with another person as being guilty of a non-consensual sexual offence. Nevertheless, the italicised words are important. Principle, and the truth, are very important; but they are not the only considerations that the law must take into account. To be more specific, it must never be forgotten that, to achieve its purposes, the law must retain its authority – and, to retain its authority, the law cannot stray too far beyond what the public will accept. This is by no means to dispute Williams’ contention that the ‘criminal law can lead public opinion as well as follow it’. ²⁹³ Again, to the extent possible, it should be principled; and it should do what is right and true, not what is popular. But sometimes people consider an idea so risible or unacceptable that, however morally correct it is, the law should not act in accordance with it.

To exemplify the point: in 1874, in New South Wales, a Member proposed an amendment to a Bill that had been introduced into Parliament. Under the clause he proposed, adultery would be made a crime in the colony. But the proposed amendment was soundly defeated – mainly because, as Woods has noted, the ‘legislators of late nineteenth century Sydney were not all practising monogamists’. ²⁹⁴ Indeed, Woods thinks that Sir Henry Parkes, for one, might have been incarcerated for lengthy periods had the amendment been carried. ²⁹⁵ This is directly relevant to the present discussion. For, if the law were to criminalise sexual activity engaged in by a married person who induced ‘consent’ by falsely saying that (a) he was not married, or (b) she was not having an affair, it would tend to achieve indirectly what a crime of adultery would achieve directly. As Pundik and his co-authors have put it, under such an approach, many individuals ‘who had experienced infidelity’ could ‘enlist society to impose severe penal sanctions on their unfaithful partners’. ²⁹⁶ To be clear, adultery arguably satisfies the conditions that, on a liberal approach, must be satisfied before conduct may be regarded as criminal. ²⁹⁷ So does non-consensual sex. ²⁹⁸ However, once enough socially powerful people do X, or at least regard X as non-criminal, the law must not be too quick to criminalise it.

A middle way

Most commentators agree that not all deceptive sex should be criminalised. But they cannot agree on what the law in this area should state. Kennedy thinks that, if a person expressly conditions her consent to sexual activity on the fulfilment of a condition, and the accused proceeds with the sexual activity despite knowing that that condition is not fulfilled, the accused should be liable for a sexual offence. ²⁹⁹ She further contends that even if there is no such ‘express conditional consent’, ³⁰⁰ the accused should generally be guilty of sexual offending if she has deceived the complainant in a manner that fails to recognise his identity – even if the deception did not ‘actually matter to the complainant’. ³⁰¹ But Kennedy's proposal is both too narrow and too broad. It is too broad because it would criminalise: (a) the adulterous husband who knew that his wife expressly conditioned her participation in sex on his assurance that he was not having an affair; and (b) the person who proceeded with sex after deceiving her partner about a matter crucial to that person's identity – even if the deceived person would have engaged in sexual activity with the person anyway. ³⁰² It is too narrow because it would seem not to criminalise the very serious deceptive conduct in Monica. ³⁰³ According to Kennedy, ‘deceptions about religious or political views should not count’ as instances of identity non-recognition. ³⁰⁴ Yet, in Monica, the claimant was clear. She would not have engaged in any sexual relations with the police officer had she known that, rather than being an activist like her, he was a police officer who was investigating her. ³⁰⁵ In short, if Kennedy's proposal were enacted, the law would not properly pursue the aim – the protection of sexual autonomy – that it should be pursuing.

Dsouza's proposal is equally questionable. According to him, sexual activity should be treated as non-consensual if, at the time of agreeing to it, a person: (a) conditions her consent on X; or (b) in a considered or unreflective manner, frames the object of her consent in a particular way; and (c) the sexual activity fails to respect her condition(s) or stay within her parameters. ³⁰⁶ But consider a case imagined by Tadros. ³⁰⁷ In it, X says to Y that he will have sexual intercourse with her only if she has never been raped. Y falsely states that she has never been raped, and sex ensues. Because X has clearly conditioned his consent on Y's not having been raped, Y would be a sex offender if Dsouza's proposal were enacted. Yet many would object to such a prosecution. To use Williams’ words, they would think that ‘where a condition placed on the act by the victim is … potentially socially undesirable’ ³⁰⁸ – that is, where the complainant's condition(s) reflect(s) an invidiously discriminatory outlook – the law should not be seen to connive at his prejudice.

Moreover, under Dsouza's proposal, the more extraordinary the accused's deception, the less likely it might be that he would be guilty of non-consensual sexual offending. Dsouza is explicit: because Monica did not condition her consent on the police officer's having a ‘genuine commitment to her own core beliefs’, ³⁰⁹ and because it ‘seems unlikely’ that she ‘either consideredly [sic] or unreflectively defined the object of her consent by reference to whether … [he] was truly an animal rights activist’, ³¹⁰ the law should treat her as consenting. But the only reason why she did not set such a condition, or define the object of her consent in such a way, is because it could not have occurred to her that she was being deceived so elaborately. As with Kennedy, Dsouza's proposal misses what is crucial about Monica: she did not make a genuine choice to engage in sexual relations with the police officer.

The way ahead is suggested, separately, by Tadros and Scheidegger. Tadros notes the ‘familiar intuition that trivial errors of consenters do not make the conduct of consentees wrong, even where the conduct falls outside the scope of consent’. ³¹¹ Likewise, Scheidegger indicates that ‘trivial deceptions’ should not give rise to sexual offence liability. ³¹² Furthermore, both commentators suggest that, where a person has only participated in sexual activity because of her ignorance as to the accused's membership of a group that she is irrationally prejudiced against, the law should refrain from regarding her as a sexual offence victim. ³¹³ These remarks point to the two categories of case in which the law should hold that no sexual offence liability arises despite the accused's material deception. That said, not all of Tadros's reasoning seems sound.

For Tadros, even if a person's ‘trivial lie’ has caused another person to participate in sex with him, he is permitted to engage in the sexual activity. ³¹⁴ Such lies ‘do not seem to matter much’, Tadros thinks; they ‘do not significantly undermine the sexual autonomy of consenters’. ³¹⁵ ‘Where X consents to sex with Y, and X is in error so that Y having sex with them does not fall within the scope of consent’, he reasons, ‘Y may nevertheless be released from their duty not to have sex with X because the sex that X gets is close to what they consent to’. ³¹⁶ But this does not seem right. One way in which Tadros exemplifies his point is by using the example of Y, who is Jewish but does not disclose this to X. In the example, Y cleans the pool of X, who is anti-Semitic and who would never have allowed this to happen if she had known of Y's cultural background. Because what X gets is ‘close to what [she] … consents to’, Tadros concludes, ‘Y is permitted to clean X's pool’. ³¹⁷ A problem, however, is that Tadros readily concedes that, if Y had disclosed that he was Jewish, X had objected to his cleaning the pool, and Y had still cleaned it, it would have been wrong for Y to do so. ³¹⁸ Another problem is that, in X's view, she has not got something that is close to what she consented to. ³¹⁹ Because of her strongly-held, pernicious views, she would presumably consider that there is a great distance between pool cleaning by a Jewish person and pool cleaning by someone else. In short, there are strong reasons to doubt Tadros's claim that Y is permitted to do what he did. And there are strong reasons to believe that, if the law is to refrain from imposing liability where there have been material deceptions of this kind, it is for the reasons that I have stated above. Reactions to the Kashur case in Israel, where the accused's rape conviction was partly based on a lie that he had told about his race, ³²⁰ seem to indicate that such prosecutions would bring the law into disrepute.

The CLRNN's proposal

Before setting out my reform proposal, it is necessary to do one more thing. As noted in the introduction to this article, in January 2023, the CLRNN Committee published its final reform proposal. As also noted in the introduction, while stating that no other changes should be made to the SOA, the Committee recommended the creation of a new, s 4A, offence of inducing a person to engage in sexual activity by deception. ³²¹ We must now consider this proposed offence in more detail.

A person would commit the new offence if, without any ‘reasonable excuse’ for deceiving another person, she were deceptively to induce him to engage in sexual activity, intending by this deception to induce such participation. ³²² A deception could amount either to a false representation about a matter that the accused ‘knows is important’ to the complainant's decision whether to engage in sexual activity or a failure to disclose a matter that the accused ‘knows is important, or that [she] … believes would be important’ to that decision. ³²³ In deciding whether the accused might have had a ‘reasonable excuse’, the jury might have regard to the following factors, where relevant: the ‘age and vulnerability’ of the complainant; the ‘age and immaturity’ of the accused; the ‘risks of serious consequences’ for the complainant if he were to engage in the sexual activity; any purpose of the accused in deceiving the complainant beyond the intention to induce him to engage in sexual activity; and ‘the personal or private nature of the matter’ about which the accused has deceived the complainant. ³²⁴ As with the s 4 SOA offence, the maximum penalty would depend on the nature of the sexual activity that the accused had induced the complainant to perform. ³²⁵ If the activity were penetrative, the penalty would be same as for the ss 1 and 2 SOA offences. If the activity amounted to touching, the penalties would correspond to those for the s 3 offence.

There are a number of problems with the CLRNN's proposal.

First, as with proposals to reinstate the procurement offence, so here. When an accused deceptively induces another person to engage in sexual activity, that sexual activity is non-consensual. Accordingly, the law should treat such an accused as being guilty of a non-consensual sexual offence. The CLRNN's reasons for taking a different approach are unpersuasive. According to it, while the new offence is no less serious than the corresponding offences in ss 1–4 – as reflected by the various maximum penalties for it – ‘we are convinced by consultation authors who present sex … by deception as a separate and distinct wrong from that of sex without consent: whereas both represent an attack on B's sexual autonomy, those attacks are of a different kind’. ³²⁶ This all sounds very much like an argument made by Gibson. ³²⁷

According to that author, while ‘deceptive sexual relations and the relations prohibited by the principal sexual offences (concerning incapacity and coercion) are equally harmful to V's right to sexual autonomy’, ³²⁸ such harm generally ³²⁹ arises differently, ‘such that deceptive sexual relations represent a different wrong to the relations proscribed by the principal sexual offences’. ³³⁰ In the case of sex that is coercive or involves a complainant who lacks capacity, Gibson explains, the complainant's negative sexual autonomy is set back – and the same is true of deceptive sexual relations. ³³¹ That is, in all three cases the complainant's ‘right not to engage’ ³³² has been breached. But, in the case of deceptive sexual relations, the complainant's positive sexual autonomy has also been impeded. She has a ‘vision of sexual liberty’, but the accused's deception has prevented her from achieving this vision. ³³³ For Gibson, the fact that, in the case of deceptive sex, there is a violation of negative and positive sexual autonomy does not make it more seriously wrongful than other non-consensual sex. ³³⁴ What it does mean is that a different wrong has been committed, and this should be reflected in the relevant offence label. ³³⁵ If it had not been for the accused's deception, Gibson says, ‘the relations would have proceeded along the lines that [the complainant] … sought]’. ³³⁶ This can be contrasted, he thinks, with other non-consensual sex. There, because of the complainant's unconditional unwillingness to engage in sexual activity, the relations could never have ‘come to any good’. ³³⁷

This reasoning is suspect. Indeed, to an extent, it is reminiscent of the ‘rape as defilement’ reasoning noted above. True, Gibson argues that there is no less a violation of sexual autonomy in deceptive sex scenarios than in other situations where sexual consent is absent. But he rather undermines this when he attaches the significance he does to the deceived person's willingness to engage in sexual activity. One is left to ask whether relations between her and the accused could in fact have ‘come to any good’. In a case like Lawrance, the answer might be yes. If the accused had told the truth and put on a condom, the complainant might have secured positive sexual autonomy. In the case of personation, the answer would seem to be no. Without a total change of identity on the accused's part, the complainant could not have realised her sexual vision. But why is this morally relevant anyway? More relevant, surely, is that, as in other cases of non-consensual sexual relations, the materially mistaken complainant has made no real choice to do what she has done.

Secondly, while, as just noted, the CLRNN has argued that deceptive sexual activity involves a distinct wrong, under its proposal certain such activity would still give rise to ss 1–4 SOA liability. The most obvious examples are cases of personation and those involving deceptions as to the ‘nature’ or ‘purpose’ of the accused's act. ³³⁸ But cases like Assange, R(F) and McNally ³³⁹ would also seemingly be capable of giving rise to a s 1 or 2 conviction, as the case may be. Indeed, given the ‘reasonable excuse’ ‘defence’, it would appear to be more prudent for the Crown to seek a s 2 conviction in a case like McNally than a s 4A one. ³⁴⁰ But, in any case, a question arises. If deceptive sexual activity really does almost ³⁴¹ always involve a distinct wrong, should not all such sexual activity be criminalised under the one offence label? Alternatively, if, as I think, all non-consensual sexual activity, without more, involves the commission of the same wrong, ³⁴² should not all that sexual activity be criminalised in the same way? In short, rather than rejecting the capricious line drawn by Lawrance, the CLRNN's proposal tends to perpetuate it. Lawrance would continue not to be guilty of rape. Defendants in cases like Assange and R(F) seemingly would be. ³⁴³ But, again, what is the difference between these cases?

Thirdly, and relatedly, the CLRNN's proposal does not deal with the SOA as a whole. As we have seen, under it, a new section would be added to that Act; no other changes would be made. Yet there are other provisions in the Act that are, and should be, directly relevant to deceptive sexual activity. As just suggested, ss 74 and 76 are two such provisions. So too are ss 1–4. Because certain of those sections – and s 75 ³⁴⁴ – are problematic in certain respects (and in ways directly touching deceptive sex regulation), they too must be reformed. I shall return to this matter below.

Finally, the ‘reasonable excuse’ ‘defence’ would create too vague a standard for juries to apply. The CLRNN is right to provide that not all deceptive sexual activity should be criminal. But, as it concedes, the ‘reasonable excuse exemption [would] … not provide a certain answer’ in particular cases. ³⁴⁵ Take the case where ‘person A is HIV positive, but has been assured that his viral load represents no danger to his sexual partner’. ³⁴⁶ Certainly, the CLRNN is right to say that a jury ‘could find that person A acted with a reasonable excuse’. ³⁴⁷ But it might not; and, even if it did, should this person be capable of being tried for serious offending at all? ³⁴⁸ And the same goes for many other cases, too. Perhaps a jury would find that the person who induced ‘consent’ by lying about his marital status, or about whether she was having an affair, had a ‘reasonable excuse’ for the deception. But even if a jury were to acquit in such a case, should such defendants be in court in the first place? Might not such prosecutions bring the law into disrepute? That said, while the law can – and should – be more prescriptive than proposed s 4A is, it cannot provide for every eventuality. The CLRNN's list of factors that a jury might consider when applying the ‘reasonable excuse’ test does direct attention to some of the pertinent matters where the law is by necessity silent about whether liability arises. This point is reflected in my own proposal, which I shall now set out.

An alternative proposal

Three values underlie and are reflected in, my solution to the ‘notoriously difficult and controversial’ ³⁴⁹ problem of how best to regulate deceptive sexual activity. The first is the desirability of the law's being as principled and honest as possible. ³⁵⁰ The second, referred to in the paragraph immediately above, is the desirability of certainty in the law – and, in particular, in the criminal law. ³⁵¹ A criminal statute should set out as precisely as possible the circumstances in which a person is, and is not, criminally liable. The judicial expansion of liability is neither democratic nor fair to the accused. Admittedly, it is more democratic for a jury to determine that an accused is guilty of serious offending in circumstances where a statute does not clearly make her so. Nevertheless, fairness does require that the criminal law be as ‘intelligible, clear and predictable’ as possible. ³⁵² The third value relates to fair labelling. In agreement with the CLRNN, ³⁵³ but dissenting from its characterisation of the wrong in deceptive sex, I believe that the criminal law should generally punish separate wrongs separately. This, too, essentially comes back to truth and fairness. An offence label should convey that a person has done what in fact he has done. Concomitantly, it should distinguish between greater and lesser wrongs. The reasons for this are simple. An offence label would stigmatise the offender excessively if it were to convey that she had committed some greater wrong. It would stigmatise her inadequately if it were to convey that she had committed some lesser wrong. ³⁵⁴

With this in mind, I make the following reform suggestions.

First, s 74 of the Act (defining consent) should be retained in its current form, but s 75 (dealing with evidential presumptions) and s 76 (dealing with conclusive presumptions) should be repealed. Those two sections should be replaced with a single section, new s 75, which would provide that: ‘For the purposes of the non-consensual sexual offences in the Act, and without limiting the grounds on which it may be established that a person does not consent to sexual activity, a person does not consent to sexual activity in the following circumstances’. ³⁵⁵ It is unnecessary for the purposes of this article to state precisely what those circumstances should be, though one matter that should obviously be on the list is (something like): ‘If the person participates in the sexual activity because of force or the threat or fear of force’. ³⁵⁶ And this points to why current s 75 should be repealed. The s 75 evidential presumption approach has been widely criticised. ³⁵⁷ One major problem with it is that it gives very little guidance about when a person is, and is not, consenting. ³⁵⁸ For example, it does not assist the jury very much in a particular case to know that there is an ‘evidential presumption’ that a person is not consenting, and the accused had the requisite mens rea, if (for example) violence was being used against the complainant at the time of, or immediately before, the act. ³⁵⁹ It would surely be better for the law to provide – as it does in many other common law jurisdictions ³⁶⁰ – that there is no consent where a complainant has participated because of certain factors.

Secondly, with the repeal of s 76, the question of mistakes and misapprehensions should be dealt with in new s 75. That section should provide, relevantly, that one of the circumstances in which a person is not consenting is: ‘If the person participates in sexual activity because of a mistake or misapprehension about one of the following matters’. There should follow a lengthy non-exhaustive list of such mistakes or misapprehensions. That list should be based on cases that have actually arisen. I shall not attempt to be exhaustive here, but it should include cases where the complainant was mistaken as to: the nature or purpose of the act; ³⁶¹ the identity of the accused (regardless of whether the person for whom the complainant mistook the accused was personally known to the complainant); ³⁶² whether the accused, who was purporting to provide medical or like treatment to the complainant, possessed a medical or like qualification; ³⁶³ whether the complainant was married to the accused; ³⁶⁴ whether the accused intended to provide a benefit to the complainant for engaging in the sexual activity; ³⁶⁵ whether the accused was fertile; ³⁶⁶ whether the accused had a grievous bodily disease that she posed a real risk of transmitting to the complainant; ³⁶⁷ whether the accused was wearing a (non-sabotaged) condom during the sexual activity; ³⁶⁸ whether the accused intended to ejaculate inside the complainant's body; ³⁶⁹ whether the sexual activity was part of a traditional cultural practice; ³⁷⁰ and whether the accused was investigating the complainant and/or an organisation or business to which she was connected. ³⁷¹ In each of these scenarios, the conduct is factually non-consensual; and there seems to be no reason why the law should refrain from allowing sexual offence liability to arise if the accused has the requisite mens rea. ³⁷²

That will not always be the case, however; and that is why there should also be a proviso included in the new section. That proviso should read:

But there is to be no conviction for an offence to which this section applies where the person participates in sexual activity because of a mistake or misapprehension about a matter, but: (a) the matter was insufficiently serious to give rise to liability for a sexual offence; and/or (b) the matter had a decisive influence on her decision to participate in sexual activity because of her irrational prejudice.

It is necessary to explain certain features of what I have just proposed.

First, the proposed law just outlined seeks to be as faithful as possible to the requirements of legal honesty and principle. It would acknowledge the truth that scholars such as Herring ³⁷⁹ – and Dougherty ³⁸⁰ and Madhloom ³⁸¹ – have revealed, namely, that a person is not in fact consenting if she engages in an activity due to a mistake or misapprehension of any kind. It would do this by stating that, in cases of ‘trivial’ mistakes or misapprehensions, or irrationally prejudiced complainants, ‘there is to be no conviction’. It would implicitly accept, in other words, that the sexual activity in such cases – and indeed in all other material mistake or misapprehension cases – was non-consensual. ³⁸² Moreover, the proposed law would give full effect to the principle, noted by the HCA in Papadimitropoulos, that where deceptive sex is concerned, it is a complainant's material mistake or misapprehension – and not the accused's fraud – that renders such sexual activity non-consensual. ³⁸³ Relatedly, it would acknowledge – as the CLRNN does, and as English ³⁸⁴ and Canadian ³⁸⁵ law do – that it is not just in cases of ‘active deception’ that a person's sexual autonomy can be violated. In Clarence, for example, the accused never lied to his wife about his infection. But surely his failure to disclose the fact rendered the resulting sexual activity factually non-consensual? Similarly in Monica. Did not the police officer's failure to disclose that he was a police officer – and was investigating the claimant and her friends – render her apparent consent unreal?

Admittedly, if the law were to focus on the complainant's mistake or misapprehension, rather than the accused's fraud, juries might be drawn into a rather difficult mens rea enquiry in some cases of alleged deceptive sex. ³⁸⁶ Take, for instance, a case where the accused failed to disclose to the (as it turned out) ‘pro-life’ complainant that she had had an abortion years before. If the jury were to find that the complainant's misapprehension vitiated his apparent consent, ³⁸⁷ it would then have to consider whether it might nevertheless have been reasonable for the accused to believe that her past reproductive choices were immaterial to the complainant's decision to engage in sexual activity with her. ³⁸⁸ There is perhaps no obvious answer to that question. ³⁸⁹ Then again, in most non-disclosure cases, the answer would be far clearer. Surely the accused in B ought to have known that his HIV+ status was material to the complainant's decision to have sexual intercourse with him? ³⁹⁰ And surely the police officer in Monica ought to have known that the claimant would have withheld her consent had she known his true identity? Accordingly, I doubt whether, when it comes to deceptive sex, there are strong enough reasons to have a ‘subjective’ mens rea requirement of the sort recommended by the CLRNN. ³⁹¹ Rather, I tend to think that, as in all other cases of non-consensual sex, the question should be an ‘objective’ one: might the accused reasonably have believed that the complainant was consenting?

Secondly, the proposed law would also be as faithful as possible to the requirements of legal certainty. Legislatures have traditionally shown some reluctance to deal very specifically with the circumstances in which deceptive sexual activity does, and does not, amount to non-consensual sexual offending. And, undoubtedly, such wariness is understandable. For one thing, when Parliament states that certain mistakes or misapprehensions are too ‘trivial’ to give rise to sexual offence liability, it does substitute its ‘appraisal’ of the gravity of such errors for that of the complainant. ³⁹² Moreover, as the Western Australian Law Reform Commission (‘WALRC’) has recently noted, ‘people have different views about whether a matter is trivial’. ³⁹³ But once it is accepted that there are certain cases of deceptive sex that should not give rise to criminal liability, it seems that there are two options. One is a ‘reasonable excuse’ provision of the kind favoured by the CLRNN. The other is for the law to list those mistakes and misapprehensions that can, and cannot, give rise to sexual offence liability – and to provide for how unforeseen mistake or misapprehension cases should be resolved. The advantage of the latter approach is that, provided that the lists are as exhaustive as possible – which they can only be if the legislature carefully examines all cases that have actually arisen – a large measure of guidance is provided about the circumstances in which deceptive sex is, and is not, criminal. The objection to which the WALRC refers is a serious one: there is disagreement about which mistakes and misapprehensions are serious and which (if any) are trivial. But provided that those who draft the lists keep firmly in mind why they are drafting them – to maintain the reputation of the criminal law and, relatedly, to prevent it from being seen to endorse invidiously discriminatory attitudes – this problem seems largely to disappear. That is because any resulting moral arbitrariness will enhance the law's authority. ³⁹⁴

Thirdly, something must be said about the method that the proposed law would set down for resolving unforeseen mistake cases. As reflected in that proposal, it should be a question for the trial judge whether a novel mistake or misapprehension can give rise to sexual offence liability. It is true that, because the jury represents the community, it might be thought more democratic to have it decide whether liability should be capable of arising in such cases. But, as discussed above, legal certainty is also an important value; and it would seem that it should take precedence here. Even with statutory guidance about how to exercise the relevant discretion, different juries would be liable to reach different conclusions in like cases. And it must also be recalled that it is a question of law whether the complainant's alleged mistake or misapprehension was insufficiently serious, or her conduct too prejudiced, for a sexual offence prosecution to succeed. It should accordingly be answered by a lawyer. ³⁹⁵ In answering it, the judge should have to compare the complainant's mistake or misapprehension with the statutory lists of mistakes and misapprehensions that cannot lead to liability. And, if the sexual activity that actually occurred between the complainant and the accused risked causing serious consequences for the complainant, the judge should be required to consider how high that risk was. Such guidance would help to focus the judge's mind on the relevant question. Once she resolved it, her decision would have precedential value. There would be a fair measure of legal consistency and predictability.

My final reform proposal is this. The ss 1–2 SOA offences (rape and assault by penetration) should be repealed and replaced by a new offence that would criminalise all of the penetrative acts currently caught by ss 1 and 2. ³⁹⁶ Essentially this is because there is no justification for the SOA's current approach, namely, criminalising penile ³⁹⁷ and non-penile ³⁹⁸ penetration separately. Or, to put the same point in different terms, the non-consensual acts covered by the s 2 offence are no less degrading or intrusive than those covered by s 1. ³⁹⁹ Moreover, and to be clear, the person who deceptively caused another to engage in one of those acts would usually be guilty of this offence. ⁴⁰⁰ As argued throughout this article, he would have committed the same wrong as the person who has used other means – for example, force or the threat of it, coercion or intimidation, or ‘stealth’ ⁴⁰¹ – to cause an unwilling actor to participate in such sexual activity. He would have had non-consensual sexual intercourse with the complainant ⁴⁰² – and the law should acknowledge this.

Pundik disputes the idea that ‘deceptive sexual relations’ should in any circumstances be ‘criminalized with the same offence that is used to criminalize coercive sexual relations’. ⁴⁰³ According to him, that is, all coercive sexual activity – including sexual activity that takes place due to non-violent threats or intimidation – is ‘fundamentally different’ ⁴⁰⁴ from deceptive sexual activity and should be criminalised separately to it. ‘The coercer's conduct’, Pundik says, is ‘especially wrongful because of its cruelty and dismissiveness’. ⁴⁰⁵ Such conduct is ‘cruel’, Pundik says, because the coercer is face-to-face with the victim and her suffering, yet remains unmoved. ⁴⁰⁶ The coercer is ‘dismissive’, he says, because – at least in many such cases – he witnesses the victim's ‘negative reactive attitudes’ ⁴⁰⁷ and fails to respond as he should. In other words, by not desisting immediately, the coercer shows indifference to the victim's interests; in important ways, he fails to treat her as a human being. ⁴⁰⁸ Pundik concludes that such cases can be contrasted with cases of deception. In those latter cases, because the victim is a willing participant – albeit only due to the accused's deception – the accused is never face-to-face with her suffering and is, therefore, never ‘cruel’ or ‘dismissive’ in a relevant way. A ‘hypothetical bystander who knows nothing of the coercer's character’, Pundik argues, ‘can still notice that the coercer is not moved in the way they ought to be moved by the victim's suffering and negative reactive attitudes’. ⁴⁰⁹ That is not so in the case of deceptive sex.

This argument is unpersuasive. Even if we accept that factors such as ‘cruelty’ and ‘dismissiveness’ are relevant not just to culpability but also to wrongdoing, is the deceiver's conduct any less cruel or dismissive than the coercer's? Consider, for example, the events at issue in Papadimitropoulos. In that case, the sentencing judge considered the prisoner's conduct to have been ‘particularly cruel’; ⁴¹⁰ and it is hard to see why he was wrong. An objective bystander viewing the sex that took place between the accused and ‘Mrs Papadmitropoulos [sic]’ ⁴¹¹ might not have seen the accused's contemptuous disregard for the complainant's interests. But why is that morally relevant? More important, surely, is that such a disregard was present. Like the coercer, Papadimitropoulos dehumanised his ‘wife’ by violating her sexual autonomy. ⁴¹² It is submitted that he thus committed the wrong targeted by non-consensual sexual offences – a wrong that seems to be supplemented by no other wrong that warrants criminalisation in the case of non-violent coercion. ⁴¹³