A woman with dementia is living in care home. Her husband comes calling on Saturday afternoons. Upon arrival, he closes the door to her room, and the care home staff are reasonably sure that sexual congress is taking place. The partners have been in a long and happy marriage, and as far as anyone can tell, the wife enjoys the sexual activity. Because of her dementia, however, she has no real understanding of the sexual nature of the husband’s behaviour. What if anything should the care staff do?

As part of my job, I speak with a wide variety of care professionals, advocacy organisations and service users on issues about law relating to mental health and mental capacity. Beginning in mid-2007, the above scenario was put to me on a number of occasions in a period of a few months. This may be in part because an Oscar-nominated film starring Julie Christie, Away from Her, was in circulation at the time, although that film concerns sexual activity between residents with dementia, a situation slightly different from the scenario above. In any event it became clear that the above scenario, or situations very similar to it, was confronted by care home staff on a frequent basis.

The legal response to the situation is clear. The woman lacks the capacity to consent to the sexual activity. The husband’s actions are therefore criminal. Depending on the specific circumstances, this will be either or both under the general offences contained in the opening sections of the Sexual Offences Act 2003, and under the more specific offences relating to sex with people with mental disorders impeding choice in sections 30–37 of that statute. Assuming the care home staff were not actively promoting the activity, nothing in that statute would make them criminally liable; but tort law does impose a duty on carers to ensure that vulnerable people are safe in their care. There is therefore a significant possibility that the staff (and, by vicarious liability, the home) could be liable in damages for the activity, if they did indeed know that it was happening and failed to prevent it.

The overwhelming response when I have given this answer in these seminars and discussions (and, indeed, with friends and many law colleagues) is that this is the wrong result, both on ethical and practical terms. Responses have ranged from the charitable (‘it’s one of the few pleasures she has left’), to the pragmatic (‘he’s paying for the care home; what are we supposed to do?’). A few of my feminist friends have focussed on the violative nature of sexual relations (‘just because she’s old, it doesn’t mean it’s not rape’), but more frequently in seminars feminist arguments are used to support the sexual activity (as with women saying ‘what do you mean I can’t decide in advance that I will allow this? It’s my body, isn’t it?’). On a number of occasions, I have been told that the law is adopting an approach that is simply inhuman, or manifestly absurd.

I have some sympathy with this critical view, but I am unable to see the issue as simply as the seminar participants present it. This article will discuss some of the ambiguities and problems both in the legal approach, and in the permissive approach.

The Current Legal Approach

The criminal law in this area is covered by the Sexual Offences Act 2003. The activity at issue may be criminal under that statute in either of two ways.

First, the general criminal law, codified in the initial sections of that statute, defines a variety of offences based on a lack of consent. Most significant for the fact situation above are likely to be rape (s 1), assault by penetration (s 2), and sexual assault, an offence which in this statute includes intentional sexual touching (s 3). Perpetrators of these offences are guilty if the victim does not consent, and if the perpetrator does not reasonably believe that the victim consents. A person without capacity cannot consent, and if the perpetrator is aware of this, he or she will be guilty of one of these offences, depending on the sexual act in question.

Second, the 2003 statute includes a specific set of offences relating to people with mental disorders. Some of these offences relate to sex between carers and persons with mental disorders (s 38–42). An exception is made in these cases for spouses (including civil partners) of the person with a mental disorder (s 43), and for sexual relationships that pre-date the care relationship (s 44). These offences are presumably primarily directed at ensuring professionalism in care relationships. Other offences relate to the use of inducement, threats or deception to engage or involve a person with mental disorder in sexual activity (s 34–37). All these offences may be committed in the context of a care home for people with dementia, but they involve facts that are outside the scenario noted above and are thus of limited relevance here.

Much more significant for current purposes are the offences in the 2003 statute that concern a ‘person with a mental disorder impeding choice’. These offences occur when the victim has a mental disorder of a type that renders them ‘unable to refuse’ the activity in question, with inability to refuse further defined as ‘lack[ing] the capacity to choose whether to agree to [the activity in question] (whether because he lacks sufficient understanding of the nature or reasonably foreseeable consequences of the activity, or for any other reason), or (b) [being] unable to communicate such a choice to A.’ [s. 30(2), 31(2), 32(2), 33(2)] The offences extend to engaging in sexual activity with such a person (s 30), causing or inciting such a person to engage in sexual activity (s 31), engaging in sexual activity in the presence of such a person (s 32), and causing such a person to watch a sexual act (s 33). While there may be some room for discussion as to what constitutes ‘sexual’ activity,Footnote 1 it seems likely from the scenario that the husband’s activity is within the ambit of this term, and it seems equally likely that the wife lacks capacityFootnote 2 and therefore has a mental disorder impeding choice. Unlike the offences relating to sexual activity with care staff, no exception is made for spousal relationships or pre-existing sexual relationships. As such, it would seem that the husband’s actions are criminal.

The Mental Capacity Act 2005 offers no assistance in these cases. While that statute may provide some broad guidance as to how to approach questions of mental capacity in this area,Footnote 3 it specifically does not provide any mechanism to allow any form of advance decision-making or substitute decision-making in sexual matters when a person lacks capacity (s 27(1)(b)). If the person with dementia is unable to consent to sexual activity, no one can consent in his or her stead.

The sentencing guidelines impose fairly stiff terms of imprisonment for such crimes relating to persons with a mental disability impeding choice. They require the judge to consider a public protection order, and convictions under these sections are notifiable offences for purposes of the sex offenders register. Custodial sentences are to be the norm, with penetration with aggravating factors generally resulting in a sentence of eleven to seventeen years, a single offence of penetration with no aggravating/mitigating factors eight to thirteen years, mutual genital contact four to eight years, and contact that is non-genital from four weeks to eighteen months. Aggravating factors include the predictable collection of issues relating to compulsion or coercion and transmission of disease, but also ejaculation or causing the victim to ejaculate; mitigating factors include a relationship of genuine affection and the perpetrator also having a mental disorder that affected his or her culpability.Footnote 4 The inclusion of a relationship of genuine affection among the mitigating factors is notable here. While the husband in the scenario above apparently has such a relationship with the wife, and therefore could expect to receive some mitigation, its inclusion as such a factor makes it clear (if it were necessary to do so) that such a relationship does not constitute a defence: the husband is still guilty.

The early cases regarding these sections of the Sexual Offences Act 2003 concerned persons with learning difficulties, and not in the context of old age.Footnote 5 R v. Adcock,Footnote 6 from early 2010, appears to be the first reported case concerning dementia in a care home for senior citizens. The case involved a man whose wife was in the care home, but who was found engaging in sexual activity with another woman resident. The sexual activity in question involved the stroking of the woman’s vagina through her underwear and rubbing her right breast. Staff noted that both she and the accused were smiling, and indeed the woman had a history of sexual disinhibitedness in the home. The accused admitted that this was part of a series of sexual encounters going back a few weeks, which had included insertion of his finger in her vagina. There was no evidence of physical or emotional harm to the woman. The accused pleaded guilty, albeit not immediately, and was of previously unblemished character.

Adcock was sentenced to concurrent sentences of a maximum four years, reduced to three on appeal. This is an ambiguous sentence. On one hand, three years is a significant sentence. On the other, the sentence tariff for a single act of penetration should have been eight to thirteen years, suggesting a markedly light sentence. The good character of the accused, and the fact that his crime was one of acquiescence to the sexual request of the victim (albeit an incompetent request) appear to have been significant factors in the apparently low sentence. The overall question of the gravity of sexual activity with a person with dementia is not discussed. The significant but less than tariff sentence perhaps reflects some of the ambiguities of the area, identified in this article: the court takes this sort of thing very seriously, sort of.

Sex, Dementia and the Elderly

Until recently, there has been little explicit discussion of sex among the elderly. The legal literature related to mental disability and capacity to engage in sexual relationships has tended to focus on people with learning disabilities, rather than people with age-related incapacity.Footnote 7 While there are clearly some similarities between these groups, there are also differences. Discussions of sex and people with learning disabilities tends to focus on people who are socially expected to be sexually active, typically late adolescents and young adults. For this group, issues relating to the right to become parents are often relevant to the discussion. Often, these discussions also concern people who are not in institutional care. These issues do not transfer simply to the institutionalised elderly, for whom the right to become a parent is unlikely to be a concern.

Empirical literature has similarly been slow to develop. As Kessel points out, sex between elderly people has traditionally been perceived either not to exist, or to be a topic of humour, or to be morally disgusting or otherwise grotesque.Footnote 8 This silence has begun to be broken regarding the elderly population generally. While it is true that sexual activity tends to decrease in frequency as people age, it is still not uncommon. A study by DeLamater and Moorman found that more than a quarter of couples over the age of 75 had sexual relations at least monthly, and of those that did, the average was three times per month.Footnote 9 Their study suggests that medications and illnesses have a relatively small effect on sexual activity.Footnote 10 In a study by Gott et al, thirteen of the twenty-one interviewees over the age of 70 considered sexual activity was moderate to extremely important to them, and this included a majority of both men and women interviewed.Footnote 11 In Helgason’s study, almost half of Swedish men aged between 70 and 80 had an orgasm at least once a month.Footnote 12 Kesel notes in a study of California retirement homes, 62 per cent of men and 30 per cent of women continued to engage in sexual intercourse, and 87 of men and 68 per cent of women to engage in physical intimacy.Footnote 13

Certainly, therefore, it cannot be assumed that older adults are not sexually active. It further appears that sexual and other intimate contact is healthy among older people. Tenenbaum cites literature correlating sexual activity in this population with an enhanced feeling of self-worth, longer life expectancies, better cognitive functioning, and enhanced independence. The absence of such relationships, by comparison, she correlates to loneliness, depression, a higher usage of medical and psychological services, and greater difficulties in acclimatising to life in an institution, should such a move be necessary.Footnote 14 The human side of this is evident from an interviewee in Roach’s study of staff responses to sexual behaviour of people in nursing homes:

Everyone, everyone in nursing home, no matter how many visits from the family, they are lonely. They’re crying for love, that’s what they want.Footnote 15

These figures are helpful in dispelling the image of elderly adults as being asexual, but they are of limited assistance in considering sex among people with dementia, let alone who lack capacity to consent to sexual activity, as the studies do not generally focus on these categories. Most do not focus on people in nursing homes, let alone people with dementia. Sex among people with dementia is much less studied. Ehrenfeld et al used an observational study to evaluate the sorts of sexual activity in which people with dementia living in care homes engage. Seventy per cent of the activity was between men and women, and it ranged from loving and caring to eroticism.Footnote 16 The study did not quantify the frequency of sexual contact, however. Tabak states that with the onset of Alzheimer’s disease, desire for sexual contact ‘frequently increases or, alternatively, there is complete disinterest’Footnote 17 but offers no empirical support for this claim. A study by Alagiakrishnan et al shows that sexually inappropriate behaviour is rare among elderly people with dementia, occurring in less than two per cent of that population, with rates not differing between institutionalised people and those living in the community,Footnote 18 but once again no overall prevalence of sexual activity more generally is provided for these populations. Nonetheless, it would appear from the limited empirical literature that some sexual interests do remain among the elderly with dementia, and the quote from Roach’s study above seems intuitively to be all the more compelling.

Indeed, the problem has an additional facet when one (or, perhaps, both) of the partners lack capacity. With the progress of dementia, the nature of the relationship will change. Conversation, upon which so much of relationship and expressions of affection are based, will become less relevant. If a meaningful relationship between the husband and wife is to be maintained, it seems likely that some form of physical intimacy will be of enhanced importance. If this is correct, then too rigid a restriction of such intimacy may have an adverse effect on the relationship, to the detriment of both partners.

All this would support a view that sex and/or intimacy among the elderly is something that ought at least in general to be encouraged.

At the same time, there is a serious risk of abuse. The incidence of abuse of elderly people as a whole is well-known and, appropriately, a matter of public concern.Footnote 19 That said, it would seem that the incidence of sexual abuse among the elderly is relatively small. In a study by O’Keeffe et al of 1784 persons in the United Kingdom over the age of 66, only .2 per cent of those sampled had been subjected to sexual abuse in the previous year. This is notably lower than any of the other categories measured in the study—neglect, and financial, psychological and physical abuse.Footnote 20 Further, the abuse uncovered was relatively uninvasive—being talked to in a sexual way, and being touched in a sexual way, complaints the study authors viewed as classifiable as harassment.Footnote 21 While this study is useful in counteracting any image that sexual abuse of the elderly is epidemic in society, this survey is problematic for present purposes, in that it excluded both the institutionalised elderly and people with sufficiently severe dementia that they were unable to engage in the research. It would seem that the institutionalised and incapable elderly are more likely to be subject to abuse, suggesting that this is not a question that should be simply dismissed.

The question of abuse is arguably circular, in that any sexual congress with a person lacking capacity is arguably abusive. That would be the starting point in the general criminal law, and any movement from that standard requires us to ask a range of very difficult questions about sexuality, and why sexual behaviour with older adults lacking capacity would be approached differently from other persons lacking capacity. Our current law is clear: sexual activity with someone unable to consent is a sexual offence. Issues such as whether the person ‘enjoyed it at the time’ may be relevant to sentence, but they are not relevant to culpability. Why would we create an exception for older people? Further, overlaying such a structure onto existing law would be problematic. Tennenbaum and Ehrenfeld both note instances of an institutionalised woman with dementia in a sexual relationship with a man living in the care home, thinking incorrectly that the man in question was her husband.Footnote 22 Elsewhere in law, such a mistake of identity would vitiate consent; is there any reason that we should adopt a different standard because the woman has dementia? Does it matter if the person with dementia does not really understand who the other party to the encounter is?

The current law takes a purely prohibitive approach to sex with the incapable adult. If it is allowed that this is at least potentially counterintuitive, in that some sexual activity may be helpful and healthy for such persons, the question becomes how law could approach these difficulties.

Attempts to Escape the Dilemma

The 1957 Approach

The 2003 Act superseded the Sexual Offences Act 1957, which prohibited sexual activity with any woman with a mental impairment, whether that woman had capacity or not. Certainly, it would be theoretically possible to return adopt a status-based prohibitory approach, where any sexual activity with a person with a mental illness or learning disability would be criminal.

This direction is mentioned merely for completeness, and can be dismissed summarily. It is not consistent with the developments in the rights of persons with mental disabilities over the last half century. It would also not solve the difficulties identified above. The question is whether too many people are currently precluded from sexual activity, not too few, and a move to a status-based approach would exacerbate rather than alleviate the difficulties discussed above.

The road to the future is thus not a return to the past.

The Definition of CapacityFootnote 23

One way to minimise the effect of the tensions identified above would be to apply the definition of capacity to maximise the number of people who can consent to sexual activity. Such an approach is consistent with the approach of the Mental Capacity Act 2005, with its requirements that a person ‘is not to be treated as unable to make a decision unless all practicable steps to help him to do so have been taken without success’ (s 1(3)) and should not be held to lack capacity if that person ‘is able to understand an explanation of [the matter of the decision] given to him in a way that is appropriate to his circumstances (using simple language, visual aids or any other means)’ (s 3(2)). While this Act is not directly relevant to the interpretation of the Sexual Offences Act 2003, its approach is consistent with the direction of human rights law in the last decades.Footnote 24 If this is correct, a procedural standard is created: a person who can make a capable decision with appropriate support is capable of consenting to the sexual activity.

The jurisprudence suggests that the substantive knowledge required for capacity to consent to sexual activity is relatively low. Certainly, to have capacity the individual must understand that the act in question is sexual, but the court in Re MAB holds that this is not to be read in too restrictive a way:

Crucially, the question is whether she (or he) lacks the capacity to understand the sexual nature of the act. Her knowledge and understanding need not be complete or sophisticated. It is enough that she has sufficient rudimentary knowledge of what the act comprises and of its sexual character to enable her to decide whether to give or withhold consent.Footnote 25

This raises the question of how the ‘sexual character’ is to be understood. The court in MAB quotes R v Morgan [1970] VR 337 on this point:

The better view is that, in order to protect those with a mental disorder, a person is to be treated as understanding the “nature” of the touching if they understand its sexual nature, as opposed to the fact that the physical touching is taking place. This interpretation focuses on the complainant’s knowledge or understanding of the meaning or consequences of sexual relations, and so moves firmly away from the heavily criticised and old-fashioned view that a mentally disordered person can legally consent to sexual activity if their “animal instincts” take over.Footnote 26

The court in MAB expressly does not reach a decision as to whether the individual must understand the nature and character of the sexual act, or instead (or in addition?) understand the reasonably foreseeable consequences of that act. In particular, it does not decide whether a knowledge of possible pregnancy or sexually transmitted diseases is necessary to have capacity. The court’s concern here is in placing capacity determination in a formalistic straight-jacket.Footnote 27 The same judge, in Re MM, elaborates on this concern, that the traditional framework of capacity determination—understanding, retention, belief and appreciation of relevant information—was unduly complex for questions relating to capacity to marry, or to engage in sexual activity.Footnote 28 Whether this less formalistic approach by the court will stand over time remains to be seen. The fourfold structure of capacity determination was well-established in common law,Footnote 29 and (with the exception of the removal of belief and the addition of the ability to communicate the decision) is contained in the Mental Capacity Act 2005 (s 3(1)). Nonetheless, it would seem that a simplified framework has so far been adopted by the Court in the context of capacity to consent to sexual activity.

Capacity is not to be determined by whether it is a good idea to engage in the behaviour in question, but rather of having the requisite understanding of that behaviour. Nonetheless, an ability to understand specific issues that may be relevant to the desirability of the sexual activity proceeding may be relevant, but only, of course, if it relates to the specific relationship in question.Footnote 30 Thus it is usual that sexual congress between men and women can lead to pregnancy. Insofar as this is a factor that one would normally expect to be a prerequisite to capable decision-making in this area, an exception must surely be made if one or both of the parties are unable to have children by reason, for example, of being beyond child-bearing years. Insofar as the assessment is based on the ramifications of the activity, the assessment must surely be based on the actual ramifications that are possible from the liaison in question. In the case of elderly people beyond child-bearing years, this may well limit the understanding required for capacity, and correspondingly increase the number of capable people.

The courts remain coy however on what it means to understand that the activity is ‘sexual’. As will be noted in the next section, the line between sexual and other physically intimate behaviour can be illusive. It might mean that the behaviour does or is intended to provoke a reaction of sexual arousal. It will be argued below that this is a problematic definition in its own terms, but it is certainly problematic to be included in an assessment of capacity. Some people come from religious or other social or cultural traditions where this is certainly well-understood, but not the topic of polite conversation. It is difficult to see that people with dementia who come from such traditions, even if they have capacity, will be happy to discuss their understandings of these private matters with care home staff in an assessment of capacity. It cannot follow from this that these people are held to lack capacity. It may also be the case that care home staff are uncomfortable probing residents on these topics.

Alternatively or in addition, ‘sexual’ can be an abstract and complex concept. In some psychological and sociological theories, it is a matter of social conditioning or knowledge production. Consistent with this, Denno notes that seven American states require that an individual must understand the ‘moral dimensions’ of the decision to engage in sexual activity in order to have capacity to do so,Footnote 31 and Tenenbaum gives detailed consideration to religious factors in the assessment of whether sexual relationships involving a person with dementia are appropriate.Footnote 32 Such considerations suggest that the understanding of the social meaning of sexuality is complex, and their inclusion in a test of capacity might well result in a much-restricted set of people able to make sexual decisions. They also raise theoretical problems as to what religious belief, or what moral code. People with dementia will usually have had capacity at an earlier time in life, so it would be possible to use this as a guide to decide what the individual must now understand in order to have capacity, but this is counterintuitive. Generally, previous views are a guide to best interests,Footnote 33 not to a decision as to whether an individual now has capacity. Adoption of a more objective approach is similarly problematic, however, as it is difficult to see a consistent set of morals or social beliefs about sex shared across society. The social nature of definitions of sex may be inescapable; but they do not seem in any meaningful way shared, so they would be difficult to incorporate into a definition or assessment of capacity.

Indeed, even though they may be social, conceptions of the sexual are often unarticulated. Frequently, it is a case of we know it when we see it (or, alternatively, feel it), and our decisions about it are often instinctive rather than academic. If that is the case, it is once again difficult to see how that is to be tested, to determine whether someone is able to understand that behaviour in question is indeed sexual.

The Meaning of ‘Sex’

One way to escape part of the potential harshness of the law is to interrogate more closely what is meant by ‘sexual’ activity. It is clear that the offences are broader than penetrative sex and genital contact, as the sentencing guidelines provide for sentences of four weeks to eighteen months for non-genital touching. Particularly in the realm of this conduct, what is it that makes conduct sexual, rather than merely intimate?

Section 78 of the 2003 statute defines ‘sexual’ in the following terms:

For the purposes of this Part (except section 71), penetration, touching or any other activity is sexual if a reasonable person would consider that–

(a) whatever its circumstances or any person’s purpose in relation to it, it is because of its nature sexual, or

(b) because of its nature it may be sexual and because of its circumstances or the purpose of any person in relation to it (or both) it is sexual.

It would seem that this says that behaviour is sexual if it either (a) is sexual, or (b) might be sexual and the perpetrator intends it to be sexual. In the current context, it is difficult to see that this helps much.

Certainly, one might understand the sexual nature of an act in the context of the inducement of or the intent to induce sexual arousal in a genital, physical sense. Such a definition would be problematic in practice. Assume in the scenario above that the husband greets his wife with an affectionate cuddle. If he finds himself becoming aroused, he would on this test be required instantly to stop the cuddle, whatever the effect of that on his wife. This seems both unrealistic and unduly unkind both to him and to the wife.

Further, such an approach constitutes a remarkably impoverished notion of sexuality. As it is experienced, sexuality is not merely about physical arousal. It is about a particular form of intimacy. That is much more difficult to define, particularly when one presumably wants the intimate relationship between the husband and wife in the scenario above to continue. Defining when that is sexually intimate and when that is merely emotionally intimate is likely to be either absurd (since the distinction is nonsensical in the way the parties experience it) or impossible (as the two meld together).

The secondary literature shows some of the ambiguities and arbitrariness of the scope of the ‘sexual’ as opposed to the emotionally intimate. Tenenbaum, for example, states:

Although it may at first seem counterintuitive, sexual contact is especially important in nursing homes. Nursing homes are places of isolation and loss, especially for dementia patients. By the time Alzheimer’s patients are placed in nursing homes, they may be suffering from loss of memory, self-esteem, and some ability to express themselves. Physical contact is an important method of communicating with Alzheimer’s patients. It is a good way to calm and reassure them and to show love and caring.Footnote 34

This may well be true—but what about this contact makes it sexual, rather than merely intimate? The question may well be nonsensical, as the participants may not divide their feelings or their activity in this way. The uncertainty of the ‘sexual’ as a category is further evident in Ehrenfeld’s study:

Several cases could not be classified, such as eight of the 46 (17%) reports of residents found in their neighbours’ beds. This happened between women, between men, and between men and women. It was not possible to conclude that lying together was for sexual stimulation or an erotic interlude; perhaps the intent was closeness and intimacy, resulting from fear of loneliness.Footnote 35

Alternatively, the authors might presumably have added, it might be for all of the above.

Given the complexity of sexuality as a concept, one could imagine that the legislators might have been advised to take a different course, clarifying more specifically what activities are permitted and what not when a person lacks capacity in these circumstances. Thus one could imagine that acts of penetration might appropriately be prohibited, but non-genital touching permitted unless there were evidence of objection by the party lacking capacity, irrespective of whether or not that touching induced a physical response of sexual arousal.

For better or worse, this is not the approach that the legislator took. It may be that litigation may bring about some helpful clarification of section 78, but that may prove problematic, as this section applies not only to the current context, but to all sexual offences. If behaviour would be non-sexual for purposes of the current sections, it would also be non-sexual, for example, in the context of the sexual offences relating to children and minors, as section 78 applies to both. This is likely to militate against restricting the scope of ‘sexual’ acts.Footnote 36

Balancing of Interests, and Analysing Harm

Pressing the boundaries of capacity to consent and of the range of behaviours covered by the 2003 act in a sense constitutes tinkering at the edges of the problem. Certainly, effective jurisprudence may increase the range of people who may consent to sexual activity, and the range of quasi-sexual activity that does not offend the statute. The results are problematic however. First, the point of those who advocate a more permissive policy is that some people who lack capacity by any meaningful measure should be allowed to engage in activity which, on any sensible reading of the 2003 Act, would be illegal. Their argument would be that such activity is not merely not harmful, but is actually beneficial to the individual lacking capacity. That core problem is not addressed by frittering at the edges of the problem. Second, and equally problematic, a lowering of these thresholds may expose more vulnerable people of marginal capacity to relationships which are abusive. To protect against such abuse, it would be better to allow a more robust threshold of capacity, and to allow non-abusive relationships to continue or develop with people who lack capacity. A move to such a higher standard of capacity would be criticised, in that it suggests a return to the old paternalism that has been under attack in much of the disability rights movement for the last half century. While this would be true for intrusive sexual behaviour, the paternalist tendency would be offset by allowing less invasive and non-abusive activity to be engaged in by people who did not have capacity.

Taking account of these concerns would mean a move away from capacity as the sole determinant of the permissibility of sexual activity—a significant change from the status quo, but one proposed by a number of commentators in the field. Tenenbaum suggests a four-step approach:

  1. 1.

    Determination of whether the individual has the ability to express a choice, capable or otherwise;

  2. 2.

    Determine the interests or values might be affected by acting on these desires, including the effects on the feelings of others;

  3. 3.

    Determine whether the resident can consider these interests in making a decision. If so, the relationship should be allowed to continue.

  4. 4.

    If not, determine whether the values of the relationship outweigh the critical ssues.Footnote 37

Everett proposes an approach that would balance a variety of factors including risks and benefits to the individual, and offence to others.Footnote 38 Like Tenenbaum, Everett argues that interference is only acceptable if the individual lacks capacity, but this of itself is not enough. The activity must also pose an unreasonable risk, and any interference will be effective, not creating more harm than it prevents, as mild as possible, non-discriminatory, and if at all possible, considered justifiable by the person upon whom it is imposed.Footnote 39

These approaches have their appeal, but are potentially problematic. Both allow people who have capacity to make their own decisions as to whether or not to engage in sexual conduct: to that extent, both approaches must be right. A balancing approach when the individual lacks capacity is more problematic, in that the factors to be taken into account in the balancing process are not clearly defined. Certainly, relations that caused physical or emotional distress to the person lacking capacity should not be permitted; that too is surely uncontroversial, and constitutes a ‘trump card’ in the balancing calculation. Equally convincingly, any direct positive physical or emotional effects of the activity on the individual would weigh in favour of allowing continuation of the activity. Both Tenenbaum and Everett move beyond these relatively narrow parameters, however, considering for example the views and feelings of others in determining the balance. This is potentially of concern, in that the decision as the balancing act ceases to be focussed on the person lacking capacity alone. Is it obvious for example whether the continuation of sexual activity ought to be determined by whether the children of the person lacking capacity are uncomfortable with the behaviour? Certainly, the literature concerning sexual conduct between elderly care home residents contains accounts of cases where family members are supportive of the sexual behaviour, but this is by no means universal. Ehrenfeld’s study is interesting in this regard, in that it finds that ‘in almost all cases in which a woman was erotically active, her relatives reacted angrily and demanded that the staff should protect her.’Footnote 40 This suggests a gendered view of sexuality by the family, expressed in the notion of the sexual behaviour as uniformly abusive. Insofar as these findings are representative (and Ehrenfeld’s was a relatively small study), it is fair to ask how far such factors should be taken into account in the balancing process.

A similar set of questions arises regarding care home staff. The literature suggests that they may have firm views on the appropriateness of sexual behaviour in care homes. Among the staff in the nursing homes studied by Ehrenfeld, romantic behaviour apparently provoked mixed reactions, with some staff members seeing it as amusing: ‘[t]o a certain extent, the staff treated these elderly people like small children.’Footnote 41 Generally such romantic activity was allowed to continue, but eroticism aroused anger and objections. In Roach’s study, the responses of staff were found to be a function of the staff member’s own comfort with sexuality issues.Footnote 42 It is at best doubtful whether, in such a fundamental area of personal life as sexuality, such attitudes should be permitted to determine the permissibility of behaviour. At the same time, this may be a practical reality: the administration of the balancing process would presumably rest with the institutional staff of the care home.

An alternative approach would be to remove the restriction in the Mental Capacity Act 2005 that takes sexual decision-making out of its ambit. If this occurred, decisions regarding sexual behaviour could be made according to the best interest test of that statute, a test which focuses on the best interests specifically of the person lacking capacity rather than others, and requires consideration of-

  • The person’s past and present wishes and feelings, and in particular any written directions provided while he or she had capacity

  • The beliefs and values that would be likely to influence his or her decision if he or she had capacity;

  • The other factors he or she would be likely to consider if able to do so (s 4(6))

  • All other relevant circumstances (s 4(2))

The open-ended criteria permitted to be considered by s 4(2) would allow objective harm or benefit to the individual to be considered, and the remaining factors would require factors specific to the individual and his or her previous beliefs to be expressly taken into account. In evaluating these factors, the Mental Capacity Act 2005 requires the decision-maker to consult persons involved in the care of the person lacking capacity, holders of a lasting power of attorney for that person, and consultees identified by the person lacking capacity. The views of these individuals are intended to be sought merely as they relate to the best interests of the individual lacking capacity. The consultee’s own views of (in this case) the morality or propriety of the sexual activity would be irrelevant, unless those views were also those of the person lacking capacity.

These criteria would if properly applied move a considerable distance towards an approach centred on the person lacking capacity, rather than the professional and family group surrounding that person. If the approach of the Mental Capacity Act 2005 were carried to its full remit, the person while competent could include sexual decision-making in a lasting power of attorney, designating the individual named in that power to make decisions regarding sexual behaviour on his or her behalf, in the event of the donor’s incapacity. This would take the decision-making right out of the hands of the care home staff.

In some of the seminars I have given, this is greeted as the obvious and desirable way forward. I am more sceptical. My concern is whether the best interests test above can be successfully and accurately applied in the context of sexual decision-making. As Tenenbaum notes, it is not obvious even outside a sexual context that families are particularly adept at predicting how one of their members would decide in a specific situation, or what criteria that person would bring to the situation, and I am sympathetic to her suspicion that these inaccuracies will be greater in matters of sexual choice.Footnote 43 Sexuality is so often not discussed in families, let alone the question of consent to sexual activity in the event of subsequent incapacity. Can we know what values the individual would have brought to the decision, if that individual did not discuss those values with third parties prior to losing capacity? This creates particular problems regarding sexuality. If an individual does not discuss the values he or she would bring to consent to medical treatment with family members, there is the objective test to fall back on, and in the context of provision of medical treatment, that seems acceptable. Would we be comfortable relying on an objective test to decide whether sexual behaviour was in the best interests of an individual? And if so, what would that test look like? We are back to the Tenenbaum and Everett approaches, discussed above.

The unreliability of evidence for choice is a theme which pervades any movement away from the current law. Consider the facts in the scenario with which this paper began. The length of the couple’s marriage can presumably be verified, but how reliable will the information be that it was happy? The attraction of the scenario is its unspoken suggestion that the woman would, if competent, consent to the sexual activity, as it is implicit that she freely did throughout her cohabitation with her husband. Once again, it is difficult to see that this will be known by anyone other than the husband, if indeed by him; is his word to be taken uncritically on this matter?

A problem with the sort of test in the Mental Capacity Act is that it does not take account of the fact that people change as they age. What was once unthinkable becomes the norm. This is a problem with many decisions under the Act, but sexual behaviour is a clear example. Even if we had a good sense of the views of the individual at some point in the past, while of robust capacity, does it follow that the individual’s views would remain unchanged at this later time?


The current situation is certainly problematic. Indications are that care staff are uncertain how to deal with sexual behaviour involving people incapacitated by dementia, and deal with it in different ways. In the Adcock case, the care home staff alerted the care home manager, albeit it would seem after letting the sexual conduct continue for some time, and the police were called. By comparison, in its report, The Human Rights of Older People in Healthcare, the Joint Committee of the House of Commons and House of Lords on Human Rights cites the case of an eighty year old woman who was seriously sexually assaulted by another resident in 2004. While the assault was reported in the institutional log book, it was not reported to the victim’s daughter until the following year, and it was she, rather than the care home, that alerted the police.Footnote 44 My experience in speaking with practitioners suggests that much sexual activity, but no doubt not all, is not reported to the authorities; but equally it is clear from those discussions that this is an area where they are left in real doubt as to how to respond.

As noted above, some wiggle room may be possible around the edges of the problem, through juridical manipulation of the definitions of capacity and of sexual and non-sexual behaviour, but these are unlikely to be sufficient to provide clear and defensible indicators of what activity is and is not permitted. The possibilities discussed in the final section, on ‘balancing’, would require amendments to the law. It is equally difficult to see that the law can come up with a particularly convincing answer to the problem. The reality is that some of the sex that would occur in these settings would be healthy, happy and desirable; some would be abusive. It is difficult to see that the law can come up with a sufficiently clear set of criteria that could be enacted in legislation that would sensibly permit the good sex, and outlaw the bad. This would be the case even if there were the political will to enact such legislation, which there almost certainly is not: any move away from a capacity and consent threshold for sexual conduct might well provoke a political firestorm that would not necessarily encourage considered legislative change.

A possible way forward is perhaps provided by R (Purdy) v Director of Public Prosecutions.Footnote 45 That case concerned assisted suicide, a matter coming within article 8 (right to privacy and family life) of the ECHR. The House of Lords held in that case that the uncertainties relating to prosecution in that situation were such that the applicant, Ms Purdy, could not be sure that her ECHR rights would in the event be respected. The House of Lords therefore required the Director of Public Prosecutions to publish detailed guidelines on how prosecutorial discretion would be exercised.

A parallel case can be made out in the current context. Sexual activity comes within the purview of article 8.Footnote 46 The argument can therefore be made that those who are engaging in this illegal sexual activity—activity which may be desirable and healthy for the person the law is designed to protect—are likely to be prosecuted or not.

The current CPS advice regarding prosecution for offences relating to an individual with a mental disorder impeding choice reads as follows:

Code for Crown Prosecutors—considerations

A prosecution will usually take place unless there are public interest factors tending against prosecution which clearly outweigh those tending in favour. Given the seriousness of these offences a prosecution will almost certainly be required in the public interest.Footnote 47

Unless one takes the extreme hard line that all sex with a person lacking capacity is abusive and harmful, this guidance must be incorrect. It is difficult to see that prosecution of intimate cuddles between an elderly married couple will really ‘almost certainly be required in the public interest.’

If, as seems likely, statutory amendment is unlikely and would not improve matters in any event, perhaps an improved guidance on prosecutorial discretion is the best way forward.