Dica, R. v

This is an appeal by Mohammed Dica, with leave of the trial judge, against his conviction at Inner London Crown Court before HHJ Philpot and a jury on 14th October 2003 of two offences of causing grievous bodily harm, contrary to s.20 of the Offences Against the Person Act 1861. He was sentenced to consecutive sentences of 3 ½ years’ and 4 ½ years’ imprisonment, a total sentence of 8 years’ imprisonment. His appeal against sentence was referred to the full court by the Registrar.

The appeal raises issues of considerable legal and general public interest about the circumstances in which a defendant may be found guilty of a criminal offence as a result of infecting another person with a sexually transmitted disease. In the present case we are directly concerned with HIV. However we understand that there have been significant recent increases in the recorded rates of syphilis and gonorrhoea, and that a significant proportion of sexually active young women, and many young men, are infected with chlamydia. Accordingly, although we agreed to accept submissions from the Terence Higgins Trust, the George House Trust and the National AIDS Trust in relation to HIV, and some of the problems faced by those with this condition, for which we are grateful, the issues which arise in this appeal are not confined to that devastating disease.

The facts relevant to this appeal can be summarised briefly.

The appellant was told in December 1995 that he was HIV positive. Appropriate medication was then started.

The first complainant, L, was born on 12th December 1966. She was a refugee from Somalia who arrived in the United Kingdom in November 1994. She said that she was first introduced to the appellant in 1997, and they subsequently met on a number of occasions. She explained that she was having matrimonial difficulties, and he told her that he had left his wife. The relationship between them developed from there.

According to L, when they had sexual intercourse, the appellant would say “Forgive me in the name of God”. He however insisted that they should not use protection, telling her that she could not become pregnant because he had undergone a vasectomy. After a time she experienced thrush and swollen glands. She eventually went to hospital where she was diagnosed HIV positive.

She was cross-examined at trial, when it was suggested that she might have contracted HIV from sources other than the appellant.

The second complainant was D. She met the appellant in December 2000. In February 2001 they had protected sexual intercourse, but on subsequent occasions during their relationship, sexual intercourse was unprotected. When she found that she was developing similar symptoms to those suffered by L, she sought medical advice. She was diagnosed as being HIV positive. Apart from the appellant, her only other sexual partner during the previous 18 years had been her husband.

The appellant was arrested on 11th July 2002. When cautioned, he replied, “I am terminally ill, and need to go to hospital today for an operation, I will tell you everything, I did it.” A few days later he was interviewed in the presence of his solicitor. He said that he had first met L in Kenya in 1988 and had a casual relationship with her. He had met her again in the United Kingdom. He had told her he was HIV positive when their relationship restarted, and she responded by saying that she thought that she was also infected. He said that she had been involved with between six and ten different men. In relation to D, he asserted that he had met her in 1994, when they had had a “one night stand”. The relationship resumed in 2001, when she knew that he was HIV positive. Thereafter he was charged, and after caution he replied, “I’ve understood.”

It is perhaps important to emphasise at the outset that the prosecution did not allege that the appellant had either raped or deliberately set out to infect the complainants with disease. Rather, it was alleged that when he had consensual sexual intercourse with them, knowing that he himself was suffering from HIV, he was reckless whether they might become infected. Thus, in the language of the counts in the indictment, he “inflicted grievous bodily harm” on them both.

It was not in dispute that at least on the majority of occasions, and with both complainants, sexual intercourse was unprotected. Recklessness, as such, was not in issue. If protective measures had been taken by the appellant that would have provided material relevant to the jury’s decision whether, in all the circumstances, recklessness was proved.

Although both women were willing to have sexual intercourse with the appellant, the prosecution’s case was that their agreement would never have been given if they had known of the appellant’s condition. The appellant would have contended that he told both women of his condition, and that they were nonetheless willing to have sexual intercourse with him, a case which in the light of the judge’s ruling, he did not support in evidence. The suggestion would have been strongly disputed by them both.

At the end of the Prosecution case, Judge Philpot made two critical but distinct rulings. First, he concluded that notwithstanding the well-known decision by the Crown Cases Reserved in R v Clarence (1889) 22 QB 23, it was open to the jury to convict the appellant of the offences alleged in the indictment, on the basis that its standing as “an important precedent has been thoroughly undermined, and ... provides no guidance to a (first) instance judge”. His second conclusion, which in a sense was more far-reaching, was that whether or not the complainants knew of the appellant’s condition, their consent, if any, was irrelevant and provided no defence. Accepting the Crown’s argument as advanced to him, the judge believed that the decision in the House of Lords in R v Brown & ors [1994] 1 AC 212 deprived the complainants “of the legal capacity to consent to such serious harm”.

Following that ruling the appellant elected not to give evidence, and the issue whether the complainants consented to have sexual intercourse with him knowing of his condition was not left to the jury.

Mr Carter-Manning QC, arguing the case on behalf of the appellant before this Court, contends that both these rulings were wrong in law. We must therefore examine them both. We have been fortunate that Professor John Spencer QC of Selwyn College, Cambridge, had, as a result of these convictions, published two articles in the New Law Journal of 12th and 26th March 2004, entitled “Liability for reckless infection”, which were of considerable assistance to us.

Clarence had sexual intercourse with his wife when he knew, but she did not, that he was suffering from gonorrhoea. It was not suggested that he intended to cause her to become infected, and it was assumed that if she had known of the risk, she would not have had consensual sexual intercourse with him. In the result, she became infected with gonorrhoea, and accordingly suffered grievous bodily harm.

The indictment included two counts, the first alleging the infliction of grievous bodily harm, contrary to s.20 of the 1861 Act, and the second, assault occasioning actual bodily harm, contrary to s 47. The Recorder of London directed the jury that if the facts were proved the defendant could be convicted on either count, notwithstanding that the complainant was his wife. Clarence was convicted on both counts. By a majority of 9 to 4, his appeal was allowed. He had not committed an offence against either s.20 or s 47 of the 1861 Act. If Clarence remains authoritative, this case is indistinguishable and therefore this appellant should not have been convicted. His convictions, like Clarence’s, would have to be quashed.

In Clarence the main majority judgments were given by Wills and Stephen JJs. It is reasonable to infer that Manisty J agreed with them both, and Lord Coleridge CJ and Pollock B certainly agreed with both judgments, adding brief judgments of their own. The remainder of the majority, that is Matthew, AL Smith, Grantham JJs and Huddleston B expressly agreed with Stephen J.

Clarence has achieved notoriety as support for the proposition that a married woman is deemed to consent to sexual intercourse with her husband. A husband could not be indicted for rape of his wife. This “irrevocable privilege”, as Hawkins J described it, was finally identified as a fiction in R v R [1992] 1 AC 599. However the artificial notion that sexual intercourse forced on an unwilling wife by her husband was nevertheless bound in law to be treated as if it were consensual sexual intercourse permeated much of the reasoning of the majority, and was fundamental to the outcome in relation to both counts. For present purposes, it is sufficient to illustrate the impact of this artificial notion in relation to s 47 by considering Pollock B’s observations at p. 62:

“The second count charges an assault ... I should be inclined to hold that ... an assault must in all cases be an act which in itself is illegal and ... I cannot assent to the proposition that there is any true analogy between the case of a man who does an act which in the absence of consent amounts to an indecent assault upon his niece, or any woman other than his wife, and the case of a man having connection with his wife. In the one case the act is, taken by itself, in its inception an unlawful act, and it would continue to be unlawful but for the consent. The husband’s connection with his wife is not only lawful, but it is in accordance with the ordinary condition of married life. ... The wife as to the connection itself is in a different position from any other woman, for she has no right or power to refuse her consent.”

Many of the same considerations were thought to extend to the s.20 offence. Thus, for example, AL Smith J, having dealt with the assault issue on the basis of deemed matrimonial consent, turned to the offence under s.20, and went on:

“It appears to me that this offence cannot be committed unless an assault has in fact been committed, and indeed this has been so held ...”

Both Wills and Stephen JJs made the same point, Stephen J noting that although the word “assault” did not appear in s.20,

“I think the words imply an assault and battery of which a wound or grievous bodily harm is the manifest immediate and obvious result.”

Both believed that this conclusion was supported by the decision in R v Taylor (1869) Law Rep. 1 CCR 194. Manisty J, in his very short judgment considered it “contrary to common sense” to describe what Clarence did as an assault, and from his judgment, it looks as though this robust assertion was meant to apply to both convictions.

S.20 of the Offences Against the Person Act 1861 provides:

“Whosoever shall unlawfully and maliciously wound or inflict any grievous bodily harm upon any other person either with or without any weapon or instrument, shall be guilty of a misdemeanour and being convicted thereof shall be liable ... to imprisonment ... for not more than five years.”

Wills J suggested that s.20,

“... clearly points to the infliction of direct and intentional violence, with a weapon, or the fist, or the foot, or any other part of the person, or in any other way not involving the use of a weapon as, for instance, by creating a panic at a theatre ...”.

Without direct personal action of some kind, a conviction under s.20 would be wrong.

Stephen J thought that the section was dealing with,

“The direct causing of some grievous injury to the body itself with a weapon, as by a cut with a knife, or without a weapon, as by a blow with the fist, or by pushing a person down.”

He identified what seems to have been regarded as a crucial difference between the

“immediate and necessary connection between a cut or a blow and a wound or harm inflicted, and the uncertain and delayed operation of the act by which infection is communicated.”

It is perhaps significant that neither Wills nor Stephen JJs would have been prepared to accept that the administration of poison fell within the ambit of s.20 notwithstanding that grievous bodily harm was sustained.

Pollock B, consistently with both judgments, suggested that for the purposes of s.20, grievous bodily harm must represent:

“The natural consequence of some act in the nature of the blow, wound, or other violence which is in itself illegal, and not merely the result of conduct which is immoral and injurious by reason only of a fraud or breach of good faith; or to put the proposition in another form, ‘grievous bodily harm’ which is the ultimate effect of treachery in the doing of that which is not a ‘wounding or inflicting, etc, with or without any weapon or instrument,’ but is in the doing of an act of an entirely different character, is not within the terms of the statute.”

The requirement for an assault and an immediate connection between the violent action of the defendant and the onset of its consequences were plainly central to the decision that the conviction under s.20 should be quashed.

We have, so far, made no reference to any of the minority judgments. However we must now note the way in which Hawkins J approached the construction of s.20. He rejected the suggestion that bodily harm could not be “inflicted” unless it were brought about by an assault. He said:

“... the first count may be supported even assuming no assault to have been proved”.

He referred to the precise language of s 47 itself, commenting,

“Here it will be observed that where the legislature intends that an assault shall be the foundation of the offence, it says so in express terms. If in using the word ‘inflict’ in s.20 it had intended that it should be interpreted as ‘caused by means of an assault’, s 47 would have been superfluous; for by merely substituting the word ‘actual’ for ‘grievous’ in s.20, the whole object of both sections would have been attained; for the punishment awarded in each is the same, and the ‘actual’ harm of necessity includes ‘grievous’ harm.”

After a lengthy analysis, he concluded

“These considerations lead me to the conclusion that the word ‘inflicted’ when used in the statute was not intended to be construed as involving an assault.”

Hawkins J’s minority view has now been vindicated. In R v Wilson (Clarence) [1984] AC 242, the House of Lords was considering the problem of convictions on alternative counts under s 6(3) of the Criminal Law Act 1967. It was necessary for the decision that the true ambit of s.20 of the 1861 Act should be considered. In the only detailed speech, with which each member of the House of Lords agreed, Lord Roskill made plain that notwithstanding the absence of an assault, a conviction under s.20 could nevertheless be sustained. He said in terms that “there can be an infliction of grievous bodily harm contrary to s.20 without an assault being committed”. This decision undermined, indeed destroyed, one of the foundations of the reasoning of the majority in Clarence, based on the view that an offence under s.20, like that under s 47, required an assault resulting in a wound or grievous bodily harm. This represented a major erosion of the authority of Clarence in relation to the ambit of s.20 in the context of sexually transmitted disease.

This process has continued. Since R v Chan-Fook [1994] 99 CAR 147, as approved in the House of Lords in R v Ireland: R v Burstow [1998] 1 CAR 177, it has been recognised that for the purposes of both s.20 and s 47 “bodily harm” includes psychiatric injury, and its effects. Although the impact of Chan-Fook is reflected in that now well-established principle, it is perhaps worth noticing that

“... an injury can be caused to someone by injuring their health; an assault may have the consequence of infecting the victim with a disease or causing the victim to become ill. The injury may be internal and may not be accompanied by any external injury ...” (per Hobhouse LJ at p. 151)

This language, reflecting contemporary ideas, is entirely contrary to the reasoning adopted by the majority in Clarence. In argument in the House of Lords in Ireland and Burstow, Chan-Fook was strongly criticised. The challenge was robustly rejected. The ruling was said by Lord Steyn to mark “a sound and essential clarification of the law”. As he explained, the statute of 1861 was “always speaking”, and the ambit of the offences in ss 18, 20 and 47 had to be considered in circumstances which were never envisaged by the majority in R v Clarence.

In R v Ireland: R v Burstow, much argument also centred around the difference between the concept of inflicting grievous bodily harm in s.20 and causing it in s 18. Lord Steyn recognised that the two words, “inflict” and “cause”, are not synonymous. In relation to Clarence, he acknowledged that the possibility of inflicting or causing psychiatric injury would not then have been in contemplation, whereas nowadays it is. In his view the infliction of psychiatric injury without violence could fall within the ambit of s.20. Lord Steyn described Clarence as a “troublesome authority”, and in the specific context of the meaning of “inflict” in s.20 said expressly that Clarence “no longer assists”. Lord Hope similarly examined the consequences of the use of the word “inflict” in s.20 and “cause” in s 18. He concluded that for practical purposes, and in the context of a criminal act, the words might be regarded as interchangeable, provided it was understood that “inflict” implies that the consequence to the victim involved something detrimental or adverse.

Such differences as may be discerned in the language used by Lord Steyn and Lord Hope respectively do not obscure the fact that this decision confirmed that even when no physical violence has been applied, directly or indirectly to the victim’s body, an offence under s.20 may be committed. Putting it another way, if the remaining ingredients of s.20 are established, the charge is not answered simply because the grievous bodily harm suffered by the victim did not result from direct or indirect physical violence. Whether the consequences suffered by the victim are physical injuries or psychiatric injuries, or a combination of the two, the ingredients of the offence prescribed by s.20 are identical. If psychiatric injury can be inflicted without direct or indirect violence, or an assault, for the purposes of s.20 physical injury may be similarly inflicted. It is no longer possible to discern the critical difference identified by the majority in Clarence, and encapsulated by Stephen J in his judgment, between an “immediate and necessary connection” between the relevant blow and the consequent injury, and the “uncertain and delayed” effect of the act which led to the eventual development of infection. The erosion process is now complete.

In our judgment, the reasoning which led the majority in Clarence to decide that the conviction under s.20 should be quashed has no continuing application. If that case were decided today, the conviction under s.20 would be upheld. Clarence knew, but his wife did not know, and he knew that she did not know that he was suffering from gonorrhoea. Nevertheless he had sexual intercourse with her, not intending deliberately to infect her, but reckless whether she might become infected, and thus suffer grievous bodily harm. Accordingly we agree with Judge Philpot’s first ruling, that notwithstanding the decision in Clarence, it was open to the jury to convict the appellant of the offences alleged in the indictment.

We express no opinion, either way, whether the complainants did or did not have the requisite knowledge. That will be decided hereafter. For present purposes we have to address both possibilities, assuming for the purposes of the argument only that either may be correct, and bearing in mind that in this context the crucial question is whether the complainants were consenting to the risk of infection with HIV.

The Crown’s case

The judgments of the majority in Clarence included considerable discussion about the issue of fraud (in the sense of concealment), and the consequences if consent were vitiated. Again, however, the observations have to be put into the context of the perceived requirement that in the absence of an assault Clarence could not be guilty of the s.20 offence, and the deemed consent of the wife to have sexual intercourse with her husband. To illustrate the reasoning, two lengthy passages in the judgments must be cited.

Wills J suggested, at p. 27:

“That consent obtained by fraud is no consent at all is not true as a general proposition either in fact or in law. If a man meets a woman in the street and knowingly gives her bad money in order to procure her consent to intercourse with him, he obtains her consent by fraud, but it would be childish to say that she did not consent.”

Later, at p. 33 he added:

“If intercourse under the circumstances now in question constitute an assault on the part of the man, it must constitute rape ... it seems a strange misapplication of language to call such a deed as that under consideration either a rape or an assault. The essence of a rape is, to my mind, the penetration of the woman’s person without her consent ... if coition, under the circumstances in question, be an assault, and if the reason why it is an assault depends on any degree upon the fact that consent would have been withheld if the truth had been known, it cannot the less be an assault because no mischief then ensues to the woman, nor indeed where it is merely uncertain whether the man be infected or not ....”.

Stephen J addressed the issue of the defendant’s failure to tell his wife about his condition, and at p. 42 stated:

“The question here is whether there is an assault. It is said there is none, because the woman consented, and to this it is replied that fraud vitiates consent and that the prisoner’s silence was a fraud. ...”

He continued at p. 43,

“Is the man’s concealment of the fact that he was infected such a fraud as vitiated the wife’s consent to his exercise of marital rights, and converted the act of connection into an assault? It seems to me that the proposition that fraud vitiates consent in criminal matters is not true if taken to apply in the fullest sense of the word, and without qualification ...”

At p. 44 he went on:

“…. The only sorts of fraud which so far destroy the effect of a woman’s consent as to convert a connection consented to in fact into a rape are frauds as to the nature of the act itself, or as to the identity of the person who does the act. There is abundant authority to show that such frauds as these vitiate consent both in the case of rape and in the case of indecent assault. I should myself prefer to say that consent in such cases does not exist at all, because the act consented is not the act done.”

“… the woman’s consent here was as full and conscious as consent could be. It was not obtained by any fraud either as to the nature of the act or as to the identity of the agent. The injury done was done by a suppression of the truth. It appears to me to be an abuse of language to describe such an act as an assault.”

Clarence did not face a charge of rape or indecent assault, yet the concept of his wife’s notional consent to the act of sexual intercourse was inextricably linked with the quashing of his convictions for offences of violence. He was not charged with an offence under s 3(2) of the Criminal Law Amendment Act 1885, until recently, s 3 of the Sexual Offences Act 1956, and now in slightly different terms, s 4 of the Sexual Offences Act 2003. S 3(2) of the 1885 Act, enacted shortly before the decision in Clarence, provided that:

“Any person who … by false pretences or false representations procures any woman … to have unlawful carnal connexion … shall be guilty of a misdemeanour.”

In short, by 1885, quite separately from rape, it was already unlawful to procure sexual intercourse by deception. This provision was not considered in Clarence, no doubt because he was not charged with the offence, and presumably because on the then understanding of the principle of matrimonial privilege, sexual intercourse by a husband with his wife could never be unlawful.

The present case is concerned with and confined to s.20 offences alone, without the burdensome fiction of deemed consent to sexual intercourse. The question for decision is whether the victims’ consent to sexual intercourse, which as a result of his alleged concealment was given in ignorance of the facts of the appellant’s condition, necessarily amounted to consent to the risk of being infected by him. If that question must be answered “Yes”, the concept of consent in relation to s.20 is devoid of real meaning.

The position here is analogous to that considered in R v Tabassum [2000] 2 CAR 328. The appellant was convicted of indecently assaulting women who allowed him to examine their breasts in the mistaken belief that he was medically qualified. Rose LJ considered Clarence, and pointed out that in relation to the infection suffered by the wife, this was an additional, unexpected, consequence of sexual intercourse, which was irrelevant to her consent to sexual intercourse with her husband. Rejecting the argument that an “undoubted consent” could only be negatived if the victim had been deceived or mistaken about the nature and quality of the act, and that consent was not negatived “merely because the victim would not have agreed to the act if he or she had known all the facts”, Rose LJ observed, in forthright terms, “there was no true consent”. Again, in R v Cort [2003] 3 WLR 1300, a case of kidnapping, the complainants had consented to taking a ride in a motor car, but not to being kidnapped. They wanted transport, not kidnapping. Kidnapping may be established by carrying away by fraud.

“It is difficult to see how one could ever consent to that once fraud was indeed established. The ‘nature’ of the act here is therefore taking the complainant away by fraud. The complainant did not consent to that event. All that she consented to was a ride in the car, which in itself is irrelevant to the offence and a different thing from that with which Mr Cort is charged.”

In our view, on the assumed fact now being considered, the answer is entirely straightforward. These victims consented to sexual intercourse. Accordingly, the appellant was not guilty of rape. Given the long-term nature of the relationships, if the appellant concealed the truth about his condition from them, and therefore kept them in ignorance of it, there was no reason for them to think that they were running any risk of infection, and they were not consenting to it. On this basis, there would be no consent sufficient in law to provide the appellant with a defence to the charge under s.20.

The Defence Case

We must now address the consequences if, contrary to their own assertions, the complainants knew of the state of the appellant’s health, and notwithstanding the risks to their own, consented to sexual intercourse. Following Judge Philpot’s second ruling, this issue was not considered by the jury. In effect the judge ruled that in law such consent (if any) was irrelevant. Having listened to the exchanges on this topic between Mr Carter-Manning QC for the appellant, and the court, and on further reflection, Mr Gadsden for the Crown accepted that this issue should not have been withdrawn from the jury. Although we can take the issue relatively briefly, we must explain why this concession was right.

As a general rule, unless the activity is lawful, the consent of the victim to the deliberate infliction of serious bodily injury on him or her does not provide the perpetrator with any defence. Different categories of activity are regarded as lawful. Thus no-one doubts that necessary major surgery with the patient’s consent, even if likely to result in severe disability (e.g. an amputation) would be lawful. However the categories of activity regarded as lawful are not closed, and equally, they are not immutable. Thus, prize fighting and street fighting by consenting participants are unlawful: although some would have it banned, boxing for sport is not. Coming closer to this case, in Bravery v Bravery [1954] 3 All ER 59, Denning LJ condemned in the strongest terms, and as criminal, the conduct of a young husband who, with the consent of his wife, underwent a sterilisation operation, not so as to avoid the risk of transmitting a hereditary disease, or something similar, but to enable him to “have the pleasure of sexual intercourse without shouldering the responsibilities attaching to it”. He thought that such an operation, for that reason, was plainly “injurious to the public interest”. This approach sounds dated, as indeed it is. Denning LJ’s colleagues expressly and unequivocally dissociated themselves from it. However, judges from earlier generations, reflecting their own contemporary society, might have agreed with him. We have sufficiently illustrated the impermanence of public policy in the context of establishing which activities involving violence may or may not be lawful.

The present policy of the law is that, whether or not the violent activity takes place in private, and even if the victim agrees to it, serious violence is not lawful merely because it enables the perpetrator (or the victim) to achieve sexual gratification. Judge Philpot was impressed with the conclusions to be drawn from the well-known decision in R v Brown [1994] 1 AC 212. Sado-masochistic activity of an extreme, indeed horrific kind, which caused grievous bodily harm, was held to be unlawful, notwithstanding that those who suffered the cruelty positively welcomed it. This decision of the House of Lords was supported in the ECtHR on the basis that although the prosecution may have constituted an interference with the private lives of those involved, it was justified for the protection of public health (Laskey v United Kingdom [1997] 24 EHRR 34).

The same policy can be seen in operation in R v Donovan [1934] 2 KB 498, where the violence was less extreme and the consent of the victim, although real, was far removed from the enthusiastic co-operation of the victims in Brown.

R v Boyea [1992] 156 JPR 505 represents another example of the application of the principle in Donovan. If she consented to injury by allowing the defendant to put his hand into her vagina and twist it, causing, among other injuries, internal and external injuries to her vagina and bruising on her pubis, the woman’s consent (if any) would have been irrelevant. Recognising that social attitudes to sexual matters had changed over the years, a contemporaneous approach to these matters was appropriate. However, “the extent of the violence inflicted … went far beyond the risk of minor injury to which, if she did consent, her consent would have been a defence”. On close analysis, however, this case was decided on the basis that the victim did not in fact consent.

In R v Emmett (unreported, 18th June 1999), as part of their consensual sexual activity, the woman agreed to allow her partner to cover her head with a plastic bag, tying it tightly at the neck. On a different occasion, she agreed that he could pour fuel from a lighter onto her breasts and set fire to the fuel. On the first occasion, she was at risk of death, and lost consciousness. On the second, she suffered burns, which became infected. This Court did not directly answer the question posed by the trial judge in his certificate, but concluded that Brown demonstrated that the woman’s consent to these events did not provide a defence for her partner.

These authorities demonstrate that violent conduct involving the deliberate and intentional infliction of bodily harm is and remains unlawful notwithstanding that its purpose is the sexual gratification of one or both participants. Notwithstanding their sexual overtones, these cases were concerned with violent crime, and the sexual overtones did not alter the fact that both parties were consenting to the deliberate infliction of serious harm or bodily injury on one participant by the other. To date, as a matter of public policy, it has not been thought appropriate for such violent conduct to be excused merely because there is a private consensual sexual element to it. The same public policy reason would prohibit the deliberate spreading of disease, including sexual disease.

In our judgement the impact of the authorities dealing with sexual gratification can too readily be misunderstood. It does not follow from them, and they do not suggest, that consensual acts of sexual intercourse are unlawful merely because there may be a known risk to the health of one or other participant. These participants are not intent on spreading or becoming infected with disease through sexual intercourse. They are not indulging in serious violence for the purposes of sexual gratification. They are simply prepared, knowingly, to run the risk – not the certainty – of infection, as well as all the other risks inherent in and possible consequences of sexual intercourse, such as, and despite the most careful precautions, an unintended pregnancy. At one extreme there is casual sex between complete strangers, sometimes protected, sometimes not, when the attendant risks are known to be higher, and at the other, there is sexual intercourse between couples in a long-term and loving, and trusting relationship, which may from time to time also carry risks.

The first of these categories is self-explanatory and needs no amplification. By way of illustration we shall provide two examples of cases which would fall within the second.

In the first, one of a couple suffers from HIV. It may be the man: it may be the woman. The circumstances in which HIV was contracted are irrelevant. They could result from a contaminated blood transfusion, or an earlier relationship with a previous sexual partner, who unknown to the sufferer with whom we are concerned, was himself or herself infected with HIV. The parties are Roman Catholics. They are conscientiously unable to use artificial contraception. They both know of the risk that the healthy partner may become infected with HIV. Our second example is that of a young couple, desperate for a family, who are advised that if the wife were to become pregnant and give birth, her long-term health, indeed her life itself, would be at risk. Together the couple decide to run that risk, and she becomes pregnant. She may be advised that the foetus should be aborted, on the grounds of her health, yet, nevertheless, decide to bring her baby to term. If she does, and suffers ill health, is the male partner to be criminally liable for having sexual intercourse with her, notwithstanding that he knew of the risk to her health? If he is liable to be prosecuted, was she not a party to whatever crime was committed? And should the law interfere with the Roman Catholic couple, and require them, at the peril of criminal sanctions, to choose between bringing their sexual relationship to an end or violating their consciences by using contraception?

These, and similar risks, have always been taken by adults consenting to sexual intercourse. Different situations, no less potentially fraught, have to be addressed by them. Modern society has not thought to criminalise those who have willingly accepted the risks, and we know of no cases where one or other of the consenting adults has been prosecuted, let alone convicted, for the consequences of doing so.

The problems of criminalising the consensual taking of risks like these include the sheer impracticability of enforcement and the haphazard nature of its impact. The process would undermine the general understanding of the community that sexual relationships are pre-eminently private and essentially personal to the individuals involved in them. And if adults were to be liable to prosecution for the consequences of taking known risks with their health, it would seem odd that this should be confined to risks taken in the context of sexual intercourse, while they are nevertheless permitted to take the risks inherent in so many other aspects of everyday life, including, again for example, the mother or father of a child suffering a serious contagious illness, who holds the child’s hand, and comforts or kisses him or her goodnight.

In our judgement, interference of this kind with personal autonomy, and its level and extent, may only be made by Parliament.

This, and similar questions, have already been canvassed in a number of different papers. These include the efforts made by the Law Commission to modernise the 1861 Act altogether, and replace it with up to date legislation. In relation to sexually transmitted disease, much of the discussion initially focussed on the decision in Clarence, and its perceived consequences, which as we have now concluded is entirely bereft of any authority in relation to s.20 of the 1861 Act. In its report Non-Fatal Offences Against the Person No. 218 (1993), the Law Commission expressed the view that intentional or reckless transmission of disease should be capable of constituting an offence against the person (para 15.15-15.17). A second publication, Law Commission Consultation Paper No. 139 (1995) made a provisional proposal that precluded a defence of consent for the proposed offence of recklessly causing seriously disabling injury (para. 4.46-4.51). In 1998, in response to the activities of the Law Commission, the Home Office issued a consultation paper entitled Reforming the Offences Against the Person Act 1861. In this paper, the Home Office indicated that the Government had not accepted the recommendation that there should be offences to enable the intentional or reckless transmission of disease to be prosecuted. It pointed out that the issue had ramifications going beyond the criminal law into wider considerations of social and public health policy. It stated that the Government “is particularly concerned that the law should not seem to discriminate against those who are HIV positive, have AIDS or viral Hepatitis or who carry any kind of disease”. It then went on to say that there is a strong case for arguing that society should have criminal sanctions available for use to deal with evil acts, and that it was hard to argue that the law should not be able to deal with the person who gives the disease causing serious illness to others with intent to do them such harm. It then proposed that the criminal law should apply only to those whom it can be proved beyond reasonable doubt had deliberately transmitted a disease, intending to cause serious injury. It added “this aims to strike a sensible balance between allowing very serious intentional acts to be punished while not rendering individuals liable for prosecution of unintentional or reckless acts or for the transmission of minor disease” (see paras 3.13-318). On this approach it would seem that the policy at that stage would have been to criminalise conduct of the nature we are considering when it fell within s 18 of the 1861 Act, but not when it falls within s.20. In the Law Commission’s report in 2000, Consent in Sex Offences, no view was expressed on this topic, but it was assumed that any forthcoming legislation would not impose criminal liability for recklessly communicating HIV or other disease.

We have taken note of the various points made by the interested organisations. These include the complexity of bedroom and sex negotiations, and the lack of realism if the law were to expect people to be paragons of sexual behaviour at such a time, or to set about informing each other in advance of the risks or to counsel the use of condoms. It is also suggested that there are significant negative consequences of disclosure of HIV, and that the imposition of criminal liability could have an adverse impact on public health because those who ought to take advice, might be discouraged from doing so. If the criminal law was to become involved at all, this should be confined to cases where the offender deliberately inflicted others with a serious disease.

In addition to this material our attention has been drawn to the decisions in R v Mwai [1995] 3 NZLR 149, a decision of the Court of Appeal in New Zealand, and R v Cuerrier [1998] 27 CCC (3d) 1, in the Supreme Court of Canada. Both cases arose out of legislative provisions different to our own. Nevertheless, if we may say so, the judgments were illuminating, not least in the context of the views expressed in Cuerrier, which were inconsistent with some of the arguments put to us by the interested organisations. We also notice Professor Spencer’s illuminating conclusion on the question of recklessness. “To infect an unsuspecting person with a grave disease you know you have, or may have, by behaviour that you know involves a risk of transmission, and that you know you could easily modify to reduce or eliminate the risk, is to harm another in a way that is both needless and callous. For that reason, criminal liability is justified unless there are strong countervailing reasons. In my view there are not.”

Although we have considered these judgments, and the remaining material to which our attention was drawn, in this Court we are concerned only to decide what the law is now, and in this jurisdiction. Having done so, it is for Parliament if it sees fit, to amend the law as we find it to be.

In Judge Philpot’s second ruling, he accepted the Crown’s argument that the possible consent of the victims was irrelevant. That position, as we have already explained, was not maintained by the Crown before us. For the reasons we have now given, the ruling was wrong in law.

We repeat that the Crown did not allege, and we therefore are not considering the deliberate infection, or spreading of HIV with intent to cause grievous bodily harm. In such circumstances, the application of what we may describe as the principle in Brown means that the agreement of the participants would provide no defence to a charge under s 18 of the 1861 Act.

The effect of this judgment in relation to s.20 is to remove some of the outdated restrictions against the successful prosecution of those who, knowing that they are suffering HIV or some other serious sexual disease, recklessly transmit it through consensual sexual intercourse, and inflict grievous bodily harm on a person from whom the risk is concealed and who is not consenting to it. In this context, Clarence has no continuing relevance. Moreover, to the extent that Clarence suggested that consensual sexual intercourse of itself was to be regarded as consent to the risk of consequent disease, again, it is no longer authoritative. If however, the victim consents to the risk, this continues to provide a defence under s.20. Although the two are inevitably linked, the ultimate question is not knowledge, but consent. We shall confine ourselves to reflecting that unless you are prepared to take whatever risk of sexually transmitted infection there may be, it is unlikely that you would consent to a risk of major consequent illness if you were ignorant of it. That said, in every case where these issues arise, the question whether the defendant was or was not reckless, and whether the victim did or did not consent to the risk of a sexually transmitted disease is one of fact, and case specific.

In view of our conclusion that the trial judge should not have withdrawn the issue of consent from the jury, the appeal is allowed. Notwithstanding the arguments to the contrary, we unhesitatingly order a retrial, which should take place at the earliest possible date. Subject to witness convenience and availability, appropriate arrangements are in hand for a trial in early June before a High Court Judge at Inner London Crown Court. In these circumstances we shall not address the issue of sentence.