Article Text

Download PDFPDF

Ignorance is bliss? HIV and moral duties and legal duties to forewarn
  1. Rebecca Bennett,
  2. Heather Draper,
  3. Lucy Frith
  1. University of Manchester, University of Birmingham and University of Liverpool respectively


    In 1997, a court in Cyprus jailed Pavlos Georgiou for fifteen months for knowingly infecting a British woman, Janet Pink, with HIV-1 through unprotected sexual intercourse. Pink met Georgiou in January 1994 whilst on holiday. She discovered that she had contracted the virus from him in October 1994 but continued the relationship until July 1996 when she developed AIDS. She returned to the UK for treatment and reported Georgiou to the Cypriot authorities.1

    There have been a number of legal cases involving deliberate transmission of HIV, but most have involved forced exposure to infected bodily fluids for example, rape or biting, and have been dealt with using the existing legislation for rape or assault. While it is often difficult to prove responsibility for transmission in cases of forced exposure to HIV, it is even more contentious in cases like those of Janet Pink where an individual has consented to sex but claims that he/she was not forewarned of his/herpartner's HIV-positive status. At present there is no specific criminal offence of having unprotected sexual intercourse without disclosing one's HIV-positive status but a prosecution could possibly be brought under any one of a number of existing offences.2 Perhaps a change of policy needs to be considered. The Home Office has issued a consultation document which outlines a proposal that will allow the criminalisation of intentional transmission of diseases, like HIV, that are likely to cause serious harm. This revised legislation would cover all other potentially fatal diseases (including salmonella and legionnaire's disease, for instance) but seems primarily to be targeted at HIV transmission. Should transmission of HIV through consensual sex, without the HIV-positive status of the individual being disclosed, be an offence? This question, and that of whether there is a moral obligation to disclose a positive HIV status prior to having a sexual relationship is the subject of this paper.

    • HIV
    • duty to warn
    • enforcing obligation
    • infectious disease
    • ethics

    Statistics from

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Criminalisation and moral obligations

    It has been argued that “the generally acknowledged threat of a steady and inexorable growth in the incidence of HIV infection represents a sufficient threat to require all reasonable methods of containment to be seriously examined including use of the criminal law”.2 In jurisdictions such as the USA the criminal law has been used to punish those who have deliberately or recklessly transmitted HIV. Smith suggests that the criminal law in the UK should be used to punish deliberate transmission of HIV.3 First, he argues that the harms caused through HIV transmission in the form of physical and emotional trauma are considerable. Second, although the detection/deterrence rate would be low, he believes that specific criminalisation would underscore the social rejection of what are highly dangerous practices.

    The criminalisation of such behaviour is fraught with practical problems, not least that of enforcement. It would be exceedingly difficult to establish in court that the offence had been committed because it would be difficult to prove whether a sexual partner had or had not been adequately forewarned of a partner's HIV positivity. However, that prosecution for intentional HIV transmission would probably be relatively rare and difficult to prove, does not provide an argument against prosecution itself. Prosecution for rape within marriage is also relatively rare and difficult to prove, however, by putting into place legislation criminalising such an act, not only can justice be seen to be done in at least a few cases but also an important message about the absolute wrongness of rape within marriage is established. Criminalising the non-consensual transmission of HIV could be justified along the same lines. If those who knowingly put sexual partners at risk of HIV infection without forewarning of this known risk are behaving in a morally abhorrent way, it may be that the status of this behaviour as unacceptable should be signalled in the law.

    Thus, when considering the possibility of a legal duty of HIV-positive individuals to forewarn sexual partners of their positive status, it is important to determine whether it is reasonable to suppose that individuals have a moral duty to forewarn sexual partners of their HIV positivity. A convincing argument establishing the likelihood of a moral duty to forewarn in these circumstances will drive any policy debate towards considering criminalisation of HIV transmission. Determining the extent of any moral duty to forewarn may also inform other areas of policy, for instance, contact tracing, or even forewarning by health care professionals when those with HIV refuse to forewarn. If it can be established that strong moral obligations can be attributed to an HIV-positive individual then it may be that these moral obligations should be transformed into legal obligations.

    HIV and moral obligations

    It seems uncontentious to assert that individuals have a general moral obligation to avoid harming or wronging others whenever possible. This is a general obligation and not one restricted to the transmission of HIV. What needs to be examined is the question of whether this general obligation not to harm or wrong others implies specific obligations with regard to possible HIV transmission. What is not clear in the case of HIV transmission is whether the obligation to prevent transmission falls solely (or even largely) on the shoulders of those who are already infected, or whether it is down to everyone to do their best to protect themselves against transmission.


    In a liberal society it is difficult to formulate an uncontentious argument in support of a moral obligation on the part of HIV-negative individuals to protect themselves from HIV infection. Granted the common liberal structure of societies with the emphasis on liberty and autonomy, then a “self-protection obligation” would seem irretrievably incompatible. Promoting individual autonomy and freedom seems necessarily to entail promoting the legitimacy of risky choices. The decision to run the risk of HIV infection seems on a par with engaging in dangerous sports or careers, or heavy drinking and smoking. Protecting oneself from HIV infection is clearly the prudent thing to do, but there may be no moral obligation to do so.

    The relationship between self-protection and the protection of others

    Alternatively, it may be that a duty of self-protection arises from a duty to protect others. Those on whom others rely for financial or other support may well have a duty to protect themselves from HIV in order to maintain this support. It may also be the case that individuals have a duty to protect themselves from HIV infection in order not to make avoidable calls on health care resources. But, an obligation to protect oneself in order to protect others from financial or other hardship or distress, if it exists, is not specific to HIV. If such a duty exists, it exists generally, and all individuals would be obliged to do everything possible to ensure their good health, not only by avoiding communicable diseases but also by modifying lifestyle choices from motor racing to eating junk food. Even granted that a general obligation of self-protection exists, it is something that should be heeded in wider policy considerations regarding individuals' responsibility for health, or should not be considered at all. To consider such an obligation only within the framework underlying policy on HIV is an act of discrimination against those with HIV.

    Self-inflicted harm?

    Evidence of a self-preservation duty relating specifically to HIV infection would weaken the case for specific obligations on the part of HIV-positive individuals to warn others and would provided a strong argument against policies which diminish the freedom of those infected with HIV. However, evidence for a specific self-preservation duty relating to HIV is not forthcoming. For a self-preservation obligation to seem plausible it would have to be more generally applicable and exist in a society which upheld vastly different values and expectations than a society based on a general commitment to liberalism.

    But even granted we accept this conclusion, if it can be shown that HIV infection via consensual sexual contact is an instance of self-inflicted harm this will impact on analysis of the responsibilities of HIV-positive individuals, and in turn on recommendations for policy. If the responsibility for infection lies with those who become infected then criminalisation of HIV again seems inappropriate.

    For HIV transmission via consensual sexual contact to be categorised as an instance of self-inflicted harm, it would have to be shown that an individual consented to the risk of infection. In cases where an individual was unaware of the specific risk involved, a categorisation of self-harm could not be confidently applied. Thus, the question that needs to be addressed is now one of how much information must be available for consent to risk of HIV infection to be valid, and whether valid consent would indeed render any resulting harm self-inflicted?

    Informed consent

    Charles Erin and John Harris claim that one has an absolute duty to forewarn sexual partners of one's HIV-positive status. Their claim is based on their high regard for autonomy and informed consent.4 They argue that forewarning allows those considering a sexual relationship to make an informed, autonomous choice about whether or not to run the risk of infection. Erin and Harris argue that, once forewarned of the risk of HIV infection, the individual is free to take an autonomous decision and is accordingly responsible for his or her actions. Individuals who become infected as a result of fully forewarned consensual sex are responsible for this harm, the harm is self-inflicted. Accordingly, HIV-infected individuals can be said to have fulfilled their moral obligations towards sexual partners, regarding HIV, as long as they fully forewarn of their infection.

    However, even if this is accepted, it is not clear that an individual is necessarily being deceived when a sexual partner does not disclose his or her HIV positivity because a failure to disclose need not preclude an autonomous choice. With sex education in schools and extensive educational campaigns and media coverage it is reasonable to suppose that most adults are aware of the existence of HIV and have some elementary knowledge of its transmission routes. It is arguable then, that it can be assumed that any consent given to, say, “high-risk” sexual activity includes consent to the background risk of HIV infection without specific information about the particular sexual partner in question.

    Erin and Harris reject this line of thought.5 They argue that consent to sexual contact is invalidated if any information which may cause a potential partner to refuse consent to sexual activity is withheld. But this view does not translate readily to other circumstances. For instance, it would be difficult for a woman who becomes pregnant to blame her partner for this on the grounds that he did not warn her of this possibility. It is not unreasonable for men to suppose that women who are competent to consent to sex are also aware of the risk of pregnancy and it is not, therefore, morally irresponsible of them to fail to make a specific warning about this risk.

    The basis of Erin and Harris's claim is that the risk assessment individuals may make when they believe that the risk of their sexual partner being HIV-positive is low, is very different from the risk assessment after their partner has told them that he or she is HIV-positive. For Erin and Harris, consent to sexual contact is invalidated if any information which may cause a sexual partner to refuse consent to sexual activity is withheld. It would, however, seem difficult to establish that an individual's failure to disclose his or her HIV-positive status would legally be considered to negate the validity of the consent to sexual contact. In the case of Hegarty v Shine, for instance, a claim in battery brought by a woman who contracted syphilis from her lover failed.6 The court held that she had consented to sexual intercourse and that the disease did not affect the validity of her consent. It is perhaps necessary to draw a distinction here between consent to sex and consent to the risk of infection. It is difficult to argue that failure to disclose infection with a potentially fatal sexually transmitted disease (STD) renders an otherwise consensual sexual act as rape or battery. However, it may be that while the consent to the sexual contact is valid in such cases, the consent to risk of infection is invalid. Consent to the possibility that a partner is infected may well be significantly different from consent where the partner is in fact infected, especially where the sexual partner is aware of his or her HIV positivity.


    So far, we have looked at a general obligation to forewarn and have rejected this. We will now turn to the question of whether there might be a duty to forewarn in specific circumstances.

    Levels of risk

    One specific circumstance which might have a bearing on the obligation to forewarn is the level of risk to which someone is being exposed. Erin and Harris tend to lump together all kinds of sexual relationships and then describe all as carrying an absolute risk because HIV disease will ultimately result in premature death. Equally, because not everyone who suspects that they may be HIV-positive takes an HIV test, there are good reasons to claim that those who have “reason to believe”8 that they may have contracted the virus be treated on a par with those who know that they have. Taken together, this means that the scope for a general obligation to forewarn is very wide indeed.

    As a recent, high profile trial in the USA has illustrated, the term “sexual relationship” can mean many different things. To restrict the term “sexual partner” to those who engage in coital7 relations is far too conservative. Those who engage in very intimate physical activity for pleasure which does not involve penetration are surely more than good friends! This said, there is a legal tradition of sexual partners in terms of coital relations. According to Black's Law Dictionary, oral sex, for instance, is not technically adultery.9 It was, of course, this issue of whether oral sex constituted “sexual relations” that formed the basis of President Clinton's legal defence in the Monica Lewinsky scandal. Yet any definition of a sexual partner based on coital relations, while being easy to apply, would seem woefully inadequate. On this definition, for instance, there could be no such thing as lesbian sexual partners.

    All sexual activity is not equally risky. Many sexual liaisons only involve a very low risk of infection, for example, kissing, mutual masturbation or oral sex using an appropriate barrier to infection ie condoms or dental dam. In sexual liaisons where risks are extremely low it is possible to act in a responsible and morally justifiable way without forewarning partners of one's HIV infection. The duty to forewarn, cannot, therefore apply equally to all sexual partners.

    To accept that those who have reason to believe that they are HIV-positive have a moral obligation to disclose this fact where there is any risk to others would entail widespread forewarning, even where the risks are low. An HIV-infected individual's duty to forewarn would cover not only both high and low risk sexually activity, but also any other kind of activity involving close contact no matter how low the actual risk of transmission actually was. For Erin and Harris it is not just potential sexual partners to whom we owe a duty of disclosure, but whomsoever we risk infecting, however slight this risk is - or indeed however unjust or beyond our control their own activities may be. For instance, one becomes obliged to forewarn one's flatmates of one's status just in case one of them, or one of their guests, decides to use one's toothbrush or razor without permission. Effectively, Erin and Harris seem committed to the view that if one suspects one is HIV-positive, this information must be made public.

    Such a conclusion is incompatible with a liberal society. The level of risk of transmission must be allowed to influence the strength of any moral obligation to forewarn others of HIV infection. Whilst one may be justified in withholding information about one's HIV status from one's flatmates as long as one minimises the risk of infection by disposing carefully of any spilled blood etc, it is much less acceptable that one withholds information about one's HIV-positive status from an individual with whom one intends to have unprotected penetrative sex. Perhaps there is a moral duty to forewarn where there is a significant risk of infection.

    Levels of trust

    There are other problems raised by the Erin and Harris's stance on absolute disclosure. The duty to forewarn, according to this view, in addition to not being influenced by the very different levels of risk involved, does not take into account the significance of different levels of trust and expectation that may exist in a sexual relationship. According to Erin and Harris, irrespective of how casual the partner, the kind of activity being undertaken and however small the risk, those who believe themselves to be HIV-positive must forewarn. Even in the context of other views Erin and Harris hold about social justice, this insistence on an absolute duty to disclose seems unnecessarily stringent, and, if the aim is to reduce the spread of infection, seems likely to fail because even if it is observed (which seems unlikely) it will result in some kind of bland, universal disclaimer prior to sexual activity which, very quickly, no one will take seriously.

    But the strength of their view lies in the intuition that there are at least some circumstances where disclosure is morally required, such as between long term partners or those contemplating unsafe sex - perhaps in order to have children. So, whilst there might be good grounds for dismissing an absolute duty to disclose, it is undesirable to move from this to a position to one where there is never an obligation to disclose HIV positivity.

    The key point is that our obligations to our sexual partners may change over time and depending upon the circumstances. This means that our moral obligations shift from one of self-protection - which in effect requires us to assume that everyone is HIV-positive until proven negative - to an obligation to disclose when, within a relationship, there is a false assumption of negativity. Once partners - for whatever reason - come to assume that each is HIV-negative, then there is an obligation on a partner who believes or knows herself to be HIV-positive to disclose this, particularly if some change in risk-taking is about to occur - such as, for instance, a change from the use of condoms to unprotected intercourse. This understanding of the moral obligation to disclose addresses several of the problems inherent in the Erin and Harris argument.

    First, it permits couples to arrive at an assumption of negativity by a variety of routes which reflect the different ways in which sexual relationships are formed in contemporary society, and the different ways in which couples develop longer term relationships. For instance, some individuals may take positive steps to ascertain their HIV status and insist that partners do the same. Others practise unsafe sex without apparent adverse effects and come to assume that each is negative. For still others, the decision to have unsafe sex is a statement about what they are prepared to risk to achieve closeness with the other, or is a statement of trust in the other, or of the emotional investment in the relationship. Attaching a moral value to the assumption of negativity provides a flexible point at which trust in relationships can be said to have developed. It is a point which is not based either on duration of the relationship, type of sexual activity or particular risks taken. But it is the point at which it might be argued that a relationship has moved from the casual to one of mutual moral obligation. False assumption of negativity also requires disclosure in established relationships, like those formed before HIV was considered to be a risk. Because the criteria of assumption of negativity is flexible and enables individuals to distinguish between different kinds of relationships, it does not place unrealistic demands on comparative strangers. There is a duty to disclose HIV positivity to sexual partners only when there has been an assumption of negativity, and one knows this assumption to be false. There remains, however, a general duty upon everyone not to make rash assumptions of negativity. To make a rash assumption would be to fail in one's obligation to protect one's self.

    Second, there are good moral reasons to distinguish between different kinds of relationships. In arguing that there is an absolute moral obligation to disclose even a suspicion of HIV positivity, in effect, one is arguing that most cases of HIV transmission are not just cases of harm to others but wrongs or serious injustices to others. Clearly, HIV infection is always a harm, but not so obviously a wrong, especially as there is such a range of risk, and risks (like that in the example of the toothbrush or razor) which might be taken without the consent of the infected person. Under this view of obligations, HIV transmission would only be a wrong as well as a harm if it occurred as a result of a failure to correct a false assumption of negativity. This correction of a false assumption need not necessarily take the form of a disclosure of HIV positivity. In the context of casual relationships, a more appropriate correction might be to remind one's partner that such assumptions should not be made about anyone. Likewise, modification of behaviour might also be a response to a false assumption.

    Is there an overriding moral obligation not to harm?

    Erin and Harris argue that: “If an HIV seropositive individual does inform any sexual partner of his HIV status, and the partner nevertheless indulges in sexual intercourse, whether protected or unprotected, he is a volunteer to the respective risks. This latter individual had made an informed consent and must bear responsibility for any consequences detrimental to himself. The HIV seropositive person who discloses his HIV status has discharged his responsibilities to his partner.”10

    The claim here is that while deliberate or reckless transmission of HIV or attempts at such transmission are “not only a serious moral wrong but also, in all probability, a crime”11 they cease to be morally wrong if consent to risk of infection is obtained. However, if knowing transmission of HIV is viewed as morally wrong then it would seem to follow that it is wrong whether the uninfected person is forewarned of the risk or not.

    Assuming an individual has a general obligation to do no harm where s/he has the option of doing no harm, can this obligation to do no harm be nullified by another's consent to be harmed? Can it ever be morally justifiable to harm someone?

    Clearly, there are cases where consent to harm makes the infliction of the harm morally justifiable. For instance, if a woman can save her sister by donating one of her kidneys, there would seem to be good reasons to suppose that while the removal of her kidney would be a harm, if she consents, the infliction of the harm would be morally justifiable as the consequences of not donating it may be a much greater harm. In cases where the possible harm involved is temporary and not debilitating then it may be that the benefits to the individuals involved are such that the risk of harm is worth taking.

    Consent to the risk of HIV infection may be given for many reasons, all of which may be very significant in, say, the context of a long term relationship. It may be that the uninfected partner wants to prove how much he/she cares about his/her partner, or to increase the quality of their sex lives or even to have a child together. However, HIV disease is still ultimately fatal. It is arguable that there are few benefits that could outweigh the harm of a fatal disease. One of the strongest moral obligations must be an obligation not to kill others. This obligation is clearly not absolute but where killing is deemed morally permissible it is usually as the lesser of two evils, for instance in war or instances of euthanasia. However, it is difficult to imagine a circumstance which absolves one of one's obligation not to kill other people by infecting them with a fatal condition like HIV, and it is not clear that gaining their consent to this harm makes a significant moral difference.

    Acknowledging that concern for consent seems to miss the point here suggests that if an HIV- infected individual consistently avoids “high risk” activities in an attempt to protect his sexual partners from infection, then he has discharged at least some of his moral responsibility towards his partners. So, could it be that instead of having a moral obligation to forewarn others who may be at risk of an HIV infection before they consent to any sexual contact, the infected individual actually has an obligation to protect others from infection. If an HIV-positive individual does have a moral obligation to refrain from activities that expose others to a high risk of infection, then the argument for forewarning appears redundant.

    On the wrongs of knowing transmission, we offer one further and final observation. At some point in the widespread debate about HIV transmission, we must explore the relative worth which is to be attached to sexual expression and sexual gratification. This relative worth needs to be assessed both in terms of different relationships, and in terms of the value to the individual versus his/her moral obligation to avoid avoidable harms. Without such an analysis, we are unable to pass judgment either on the behaviour of those who engage in risky activity following a failed attempt to rectify a false assumption of negativity, or on the behaviour of those who continue to have risky sex despite knowing the danger to themselves of HIV transmission. If sexual expression and gratification are extremely valuable, then some known or unknown risk-taking may be justified with reference to the good that such sex bestows on the individuals concerned. If sexual expression and gratification can be afforded only very conditional value, then it may indeed be wrong knowingly to take risks with HIV transmission to achieve even substantial sexual expression and gratification. Likewise, it may also be wrong to place one's concerns to protect one's highly valued sex life above one's concerns to protect one's partner against transmission.

    Between committed couples, the question about the value of sexual expression might be even more difficult to determine, but it is certainly something that couples themselves need to - and certainly do - address. But, until this question of value is resolved, it is difficult to argue that the decisive issue is that of disclosure.

    HIV, moral obligations and legal sanctions

    Although Erin and Harris suggest that HIV-positive individuals have a strong moral duty to disclose their status to their intended sexual partners they reject the use of the criminal law as a means by which such disclosure could be encouraged. Their concern is that such legislation would discourage those who believe they may be infected from coming forward to be tested. Since Erin and Harris assert a general obligation to disclose, is it reasonable for them to shy away from translating this obligation into a legal one?

    There are good reasons to suppose that if criminal sanctions were applicable to all cases of HIV transmission, whatever the circumstances, this would not only be morally unacceptable but also unjustifiable in terms of public health efficacy. As this paper has demonstrated the ascription of moral responsibility in many cases of HIV transmission are complex, especially those where forewarning has taken place, or “high risk” activities have been avoided, or the “at risk” individual has an obligation to attempt to protect him/herself. In such cases it will be very difficult, if not impossible to establish a compelling case that the partner who is aware of his/her HIV positivity is solely responsible for the HIV transmission. In many cases of HIV transmission, at least via consensual acts, it seems probable that the responsibility for any resulting harm is, in some sense, held jointly by both partners. In such cases it is unhelpful to prosecute one or both of the parties involved. We suggest that the criminalisation of HIV transmission is justifiable only where there is a good case for placing the responsibility for the resulting harm with one partner, that is, where the “plaintiff” is both harmed by HIV infection and wronged by not being forewarned of her partner's known HIV-positive status. If criminalisation of deliberate or reckless HIV transmission is restricted to those cases where there is clear and unequivocal evidence of both harming and wronging, it is hoped that such legislation would allow the punishment of a serious moral crime, while avoiding any unhelpful consequences in terms of public health.


    While it seems there may be a good case for criminalising some extreme cases of reckless or deliberate HIV transmission, the criminalisation of transmission of HIV alone is unjustifiable. If criminal sanctions are used then they should be available where there has been the deliberate or reckless transmission of any communicable disease which leads to death or serious injury. This is in line with the recommendations of the Law Commission which suggests that the deliberate or reckless transmission of disease should be potentially subject to criminal prosecution.11


    We have suggested that if criminalisation of HIV transmission is to be seriously considered, its applicability should be restricted to cases where reckless or deliberate transmission has occurred. Criminalising cases where forewarning occurs is not advocated. While we may disagree with Erin and Harris's assertion that by forewarning an HIV-positive individual has fulfilled his/her moral obligations regarding HIV to his/her partner, the complexity of the moral obligations involve mean that criminalising his/her behaviour will not always be appropriate and thus should be avoided. However, this needs to be clarified with regard to existing legislation. It is as important for unhelpful legislation to be repealed as it is for appropriate legislation to be introduced. At present even if disclosure takes place before the parties have penetrative sex, if the partner of an HIV-positive person becomes infected as a result, a criminal offence may have been committed. English law does not allow a person to consent to the infliction on him/herself of any harm, however grave.12 The Law Commission in recent recommendations supports this approach. It recommends that a person should not be able to consent to the infliction of a “seriously disabling injury” and that both perpetrator and victim of such an injury would be guilty of a criminal offence.13

    It can be argued that there is a public policy interest in avoiding such harm which may arise through deliberate consensual action which leads to the transmission of HIV. Indeed, ultimately society as a whole has to bear the financial burden of counselling and medical care required in the future. But on the other hand, to criminalise sexual conduct, even where the infliction of harm is involved, in a situation in which both parties are competent, consenting adults, is intrusion upon an individual's right of privacy, a right recognised in article 8 of the European Convention on Human Rights.


    There is a real danger that the criminalisation of HIV transmission may produce consequences that are not only morally unjustifiable but also unhelpful in terms of public health aims. To avoid these undesirable consequences, criminalisation of HIV should be restricted to the small minority of cases of HIV transmission that constitute serious moral wrongs. It is our assertion that criminal punishment of HIV transmission should only be applicable where there is clear evidence that an individual was not only harmed but also wronged when s/he became infected with HIV. That is, where an individual was infected with HIV by a person who was aware of his infection but failed to warn his partner of this fact and also failed to attempt to protect against transmission or where it was reasonable for the partner to assume negativity. Likewise, where transmission was a deliberate attempt to kill the subject. If and only if legislation could be developed that would restrict criminalisation to such cases should it be enacted. In addition, if criminal sanctions are to be available in these cases of HIV transmission, then they should also be available in comparable cases, that is, where there has been deliberate or reckless transmission of any communicable disease that leads to death or serious injury.


    The authors wish to thank Jean V McHale for helping with earlier drafts of this paper and for generously contributing her thoughts on the legal situation. Any errors remaining are entirely our own.



    • Rebecca Bennett is Lecturer in Bioethics, Centre for Social Ethics and Policy, University of Manchester. Heather Draper is Lecturer in Biomedical Ethics, Centre for Biomedical Ethics, The Medical School, University of Birmingham. Lucy Frith is Lecturer in Health Care Ethics, Department of Primary Care, University of Liverpool. Corresponding address: Dr Heather Draper, Lecturer in Biomedical Ethics, Centre for Biomedical Ethics, The Medical School, University of Birmingham, Edgbaston, Birmingham B15 2TT.