A right to sex? (Part II)
A Controversial Decision
Under section 39 of the Sexual Offences Act 2003 (‘the Act’), it is an offence for a care worker to “intentionally cause or incite” a service user with a mental disorder to engage in sexual activity. The court was asked to decide whether, if a service user wants to visit a sex worker* and has the capacity to make that decision, a care worker could be guilty of the offence by facilitating the visit. Hayden J ruled that they could not and we described his reasoning on that central point as “unassailable”. It was perhaps inevitable, therefore, that a strong Court of Appeal unanimously came to the opposite conclusion!
We remain of the view that Hayden J got it right and the Court of Appeal got it wrong, and in this blog we will endeavour to explain why.
But first, a disclaimer: the Court of Appeal’s decision has been widely celebrated by people who believe that sex work is something the courts should do everything in their power to discourage, people who believe that the state should not be indirectly funding prostitution, and people who are concerned about the rights of care workers. Those are legitimate, well-founded arguments that deserve serious consideration and discussion. In this blog, however, we are concerned only with the proper interpretation of section 39. If the correct interpretation leads to a socially undesirable outcome, then the appropriate remedy is for Parliament to amend the Act.
It is first necessary to distinguish between factual and legal causation. A defendant’s act cannot be the cause of an event if the event would have occurred in precisely the same way had the act not occurred. This is often referred to as the ‘but for’ test: the prosecution must prove that but for the defendant’s act or omission, the event would not have occurred. This is a necessary but not sufficient condition for the defendant’s act to be a legal cause of the event: but for the births of murderers, no murder would ever occur, but we do not say that murderers’ mothers caused the deaths of their sons’ victims.
Acres of academic texts have been written about the extra ingredients that distinguish a legal cause from a factual cause. We cannot possibly do justice to that scholarship in this blog, but it is helpful to start with this from Glanville Williams:
What a person does (if he has reached adult years, is of sound mind and is not acting under mistake intimidation or similar pressure) is his own responsibility and is not regarded as having been caused by other people.
(Textbook on Criminal Law, 2nd edition)
That quote comes from a discussion of supervening causes, but it is not inapt in the circumstances of this case: if we are serious about respecting human autonomy, we should hesitate before attributing the cause of C’s (hypothetical) sexual activity to anyone other than C.
True it is that there are circumstances in which we hold others liable for the principal defendant’s intended and freely-chosen act, typically alongside the principal defendant. If another participates in the crime to the same extent as the principal defendant, he may be liable as a joint principal; if he assists the principal defendant in the commission of the offence, he may be liable as an accessory. But we do not usually say that such a person has “caused” the principal defendant’s act.
The Court of Appeal briefly touched on the way in which causation has been analysed in driving cases. Those cases are of limited assistance because deaths caused by dangerous or careless driving are never consequences intended or desired by the defendant or anyone else (if they were, the appropriate charge would be murder). It is unfortunate that the court did not consider a driving offence rather closer to the offence in this case, namely that of causing a person to use a motor vehicle on a road in a dangerous condition. That offence is now contained in s. 40A of the Road Traffic Act 1988, but a similarly worded provision existed in the Motor Vehicles (Construction and Use) Regulations 1951. In Shave v Rosner  2 QB 113, it was held that a mechanic who had failed to tighten the wheel nuts on a van before returning it to its unwitting owner had not caused the vehicle to be used on the road in an unsafe state. Lord Goddard CJ said that “’causes’ involves a person who has authority to do so ordering or directing someone.”
This case features in a discussion by Hart and Honoré in ‘Causation in the Law’ of the circumstances in which a person can be held criminally liable for causing the act or omissions of another. The principle which the authors draw from their study of many similar offences and cases is that:
The accused is only held to have caused another to act if he makes use of threats, lies or the exercise of authority and thereby induces the second to act … it is indeed this which distinguishes ‘causing’ from other cases of inducing such as advising, persuading etc., though the decision in R v Wilson is to the contrary”.
That principle, it is submitted, remains sound.
The strongest argument in favour of the Court of Appeal’s expansive interpretation of “cause” in section 39 is the analogy drawn by Burnett LCJ with offences concerning abuse of positions of trust. It is worth quoting the relevant passage in full:
47. I have referred to the offences created by sections 16 to 19 of the 2003 Act relating to abuse of positions of trust: see para  above. It is difficult to conceive that it would be a defence to a charge under section 16 (sexual activity with a child) for the adult in a relevant position of trust to say that a 16 or 17 year old child was a consenting participant in the sexual activity who actively desired the sexual contact. Once more, Parliament has chosen to protect a group of people who are vulnerable by drawing bright lines which reduce risk to the cohort generally. There have been many instances of teachers and pupils falling in love, running away together and the like. It is immaterial that if the adult were not in a position of trust no offence would be committed.
48. Section 17 is the precise analogue of section 39. A person over 18 in a position of trust commits an offence if he intentionally causes or incites a child to engage in sexual activity. That is so even where the child is over the age of consent. Would a teacher who engaged a sex worker at the request of one of his 17 year old students, made all the arrangements including for payment and transport be able to avoid a conviction on the basis that there was no breach of trust and that he was not repressing the student’s autonomous sexual expression but advancing it? Unless the word “causes” means different things in sections 17 and 39 that line of defence would be open to him on the judge’s interpretation. No argument was advanced to the effect that the words “causes or incites” should mean different things in different places in Part 1 of the 2003 Act.
Criminalising Care Workers?
It is of course right to say that very few people would mourn the conviction of a teacher for facilitating his 17 year old student’s visit to a sex worker, but we should be cautious about working backwards from the desired result. Furthermore, there is a good reason that Burnett LCJ did not use a child of say, 15, in his example: the obvious charge for a teacher who facilitated sex for a student under the age of consent would be the offence of arranging or facilitating the commission of a child sex offence, contrary to section 14 of the Act. We might well ask why there is no similarly-worded provision for those suffering from mental disorders if Parliament intended to criminalise care workers for arranging or facilitating their sexual activity.
Hayden J’s analysis of the mischief at which section 39 is directed is hard to fault:
The mischief of Section 39 SOA 2003, as elsewhere in the legislation, is exploitation of the vulnerable. The provision is perhaps not drafted with pellucid clarity, but its objectives are identifiable. It is intended to signal unambiguous disapprobation of people employed in caring roles (i.e. care workers) who cause or incite sexual activity by a person for whom they are professionally responsible. The legislative objective is to criminalise a serious breach of trust and, as I have commented, attracts a significant custodial sentence. The words of the statute need to be given their natural and obvious meaning. They are intending to criminalise those in a position of authority and trust whose actions are calculated to repress the autonomy of those with a mental disorder, in the sphere of sexual relations. Section 39 is structured to protect vulnerable adults from others, not from themselves. It is concerned to reduce the risk of sexual exploitation, not to repress autonomous sexual expression. The language of the section is not apt to criminalise carers motivated to facilitate such expression. In my judgement, the expanded interpretation of this provision, contended for on behalf of the Secretary of State, requires the language of the section to be distorted and the philosophy of the Act to be disregarded.
There is much to be said for bright lines when it comes to child protection, but it not self-evident that the same approach can be transferred unproblematically to vulnerable adults. At the very least there is a real tension between such an approach and the individualised, granular focus on autonomy that lies at the heart of the Mental Capacity Act 2005.
Whatever one may think of the present state of the law, it is not an offence for A to arrange or facilitate B’s visit to a sex worker if he is not disabled, nor is it an offence for A to arrange or facilitate B’s visit to a sex worker if he is B’s carer and B’s only disability is physical. It is difficult to see why criminal liability should turn on the fact that B’s disorder is mental unless his disorder deprives him of the capacity to make the relevant decisions.
*We understand that the term “sex worker” is, in itself, controversial. We adopt it here because it feels more respectful, but in doing so we do not intend to prejudge the question of whether “sex work” is truly akin to work.