Article Text


Dangerousness, mental disorder, and responsibility
  1. J R McMillan
  1. Department of History and Philosophy of Science, University of Cambridge, Free School Lane, Cambridge CB2 3RH, UK; jm439{at}

    Statistics from

    While the UK Home Office’s proposals to preventively detain people with what it has called dangerous severe personality disorder (hereafter DSPD) have been subjected to debate and criticism the deeply troubling jurisprudential issues in these proposals have not yet entered into public debate in a way that their seriousness deserves.1 It is good that a commentator as well known as Professor Szasz is speaking out on this issue.

    Professor Szasz focuses upon a crucial question by calling into question the medicalisation of terms like dangerousness and mental illness.2 There is a great temptation for legislators and the public to treat these terms as if they are purely scientific terms and to think of risk assessment as a precise science. I don’t share Professor’s Szasz’s worry about how psychiatry is using these terms but I think there are some important questions about how these concepts function in the public sphere.

    This issue is important enough to justify the use of some rhetorical claims because such claims can serve to bring out what are neglected and important issues. In raising the following objections I am not attempting to refute the message that is in Szasz’s article or to say that he is not making an important point but rather to attempt to sharpen the issues and to suggest ways that we can respond to Professor Szasz’s challenge.

    Psychiatry has progressed in a number of ways since Professor Szasz wrote The Myth of Mental Illness.3 The science is much better, we have much better medications, and psychiatrists (at least all the psychiatrists that I know) are acutely aware of the tension between treating their patient and their obligations to the public interest.

    So when Professor Szasz says: “Psychiatrists offer to relieve the disturbed person of the burden of coping with his disturbed relative by incarcerating the latter and calling it ‘care’ and ‘treatment’”, this is a fairly prejudiced and outdated view of what psychiatrists do when placing a person under a compulsory treatment order. It would be more accurate to say that when care and treatment are appropriate families may be relieved of the burden of coping with a mentally unwell relative.

    The major thesis of The Myth of Mental Illness and Professor Szasz’s article is that mental illness is essentially linked with dangerousness. This link creates a huge tension within psychiatry between its obligation to heal patients and to protect the public.

    While I think there are some good reasons for taking this thesis seriously when considering the history of psychiatry it is not an accurate way to describe contemporary psychiatry. I have a number of major worries about the idea that there is some intrinsic link between mental disorder and dangerousness.

    While dangerousness, or perhaps more accurately social or economic disutility, was an important rationale for the incarceration of the insane4 it is no longer the telos of psychiatry. Dangerousness is no longer sufficient for incarcerating the mentally ill, except in the most extreme circumstances, under British and Commonwealth mental health legislation. The wording and actual effect of these acts means that people who are a danger to other people or unable to care for themselves but who do not fall into the legally defined categories of mental disorder are routinely not placed under compulsory treatment orders. This in itself is sufficient to refute the claim that there is an essential connection between dangerousness and mental disorder in the wording and effect of the law.

    Second, it’s not accurate to say that dangerousness is the overriding concept when thinking about compulsory treatment. The key moral concept (in fact you could almost call it a suppressed premise) in compulsory treatment is responsibility or the lack thereof. It is because of the effect that mental illness can have upon an individual’s ability to autonomously choose and act on a treatment decision when this treatment decision is related to a significant risk of harm to others or self that we might be justified in compelling people to be treated.

    Not all mental health acts make this point explicit. The 1983 UK Mental Health Act requires that a person meet one of the defined categories of mental disorder, be a danger to others or not be able to care for themselves and that their condition is treatable. While the majority of people sectioned under the UK act will lack responsibility or competence there is a theoretical possibility that a person may be competent to make treatment decisions, have a treatable condition, not be able to look after themselves, and have a disorder that meets the relevant definition of mental disorder.

    One interesting example of legislation that does make the link between responsibility and civil commitment explicit is the 1992 New Zealand Mental Health Act (for further discussion of this issue see Dawson).5 The New Zealand act requires that in order for a person to placed under a compulsory treatment order it must be the case that their inability to care for themselves or their dangerousness to others results from a disorder of mood, cognition, or volition. When a disorder of mind is causally related to harm to self or others then this is a strong moral reason for not treating this behaviour as an expression of autonomy. Furthermore this impairment of autonomy is tied to a serious risk of harm to self or others and this creates an important threshold for when people can be treated against their will. So in effect both dangerousness that results from diminished autonomy and diminished autonomy are necessary for justifiable compulsory treatment in New Zealand.


    The two classic explanations for the importance of autonomy and liberty come from Immanuel Kant and John Stuart Mill.6,7 While Mill and Kant had different explanations for why autonomy is important, both of them thought rationality was important for the autonomy of others being binding upon us. Mill thought that a full set of liberties should only be fully extended to

    . . . human beings in the maturity of their faculties. We are not speaking of children, or of young persons below the age which the law may fix as that of manhood or womanhood. Those who are still in a state to require being taken care of by others, must be protected against their own actions as well as against external injury (Mill, p 14).7

    Mill doesn’t tell us much about what constitutes a state in which a person must be protected from their own actions but it is fairly clear that he has in mind states in which the exercise of freedom would be at odds with what is important about having liberty. For Mill this would be when the exercising of a freedom does not further an individual’s interests, or capacity for interests and when this results in harm to self or others. So if we turn back to one of the classic justifications for the importance of liberty, impaired autonomy can reasonably be thought of as undermining the right to a full set of liberties. While mental disorders often do not give us reason to restrict liberty they can, and in such circumstances we may be justified in treating against an individual’s stated wishes.


    Just as considerations about diminished autonomy can underpin the morality of compulsory treatment they also furnish us with a response to Professor Szasz’s position on suicide.

    Rejecting life and wanting to kill oneself is defined as a severe mental illness characterised by “dangerousness to self”, and is treated as a quasicrime with coercions called “treatments” (especially involuntary “hospitalisation” and forced drugging) (p 228).

    While there is something to the idea that suicide becomes medicalised in psychiatry it seems very harsh to say that psychiatry views this as a quasicrime; it is more often a tragedy that we had sound moral reasons for wishing we had prevented.

    Although it is foolish to ignore the history of our views about the morality of suicide this claim doesn’t ring true for contemporary discussion about taking our own lives. Public opinion (at least in the UK) is, in general, sympathetic towards euthanasia and assisted suicide. Requests by an individual for help in bringing to an end an existence that has become intolerable to them (such as the request of Diane Pretty) are not treated as symptomatic of a disease nor a “quasicrime” but as reasonable requests given intolerable personal circumstances. When a person’s situation is such that nothing can be done to make their life better this strengthens our conviction that this is a reasonable request.

    There are, however, some crucial differences when the person who wants to commit suicide is mentally unwell. Successful suicide is irreversible so while there are instances of suicide where we should not interfere or criticise morally those who offer aid, there is an onus on being sure that an individual is not making a big mistake. Second, disturbances of mood, cognition or volition can make people want death when they would not if these disturbances were alleviated. In such cases it reasonable for us to take their preference when not influenced by a disturbance of mind as a better indication of what they want. A third disanalogy that will hold even when a person really does want to take their own life is whether anything can be done to make their existence more tolerable. While the predicament faced by Diane Pretty was one that offered her no hope that her condition might improve it is much more difficult to be so sure when the predicament is related to factors such as feeling that one’s life is of no value.

    Professor Szasz thinks that suicide can be reasonable and that it is the ultimate expression of personal autonomy. There are some famous dissenters to this view. Kant thought that suicide, far from being an expression of self rule, was the antithesis of self rule in that it involved undertaking an action that aims at the obliteration of the self. For what it’s worth I’m with Professor Szasz on this point and think that suicide is an act that exhibits the most significant amount of control over a life that an individual can exercise. But this shouldn’t distract us from being certain that a person knows what they are doing, is doing what they really want, and does not want this simply because of material conditions that are in the power of another person to alter.


    Irrespective of his views about suicide Professor Szasz is correct to question how dangerousness is used in the Home Office proposals for the preventive detention of people with DSPD.

    It is a fundamental principle of English and American law that only persons charged with and convicted of certain crimes are subject to imprisonment. Persons who respect other peoples’ rights to life, liberty, and property have an inalienable right to their own life, liberty, and property. Having a disposition or propensity to break the law is not a crime (p 228).

    It is important to clarify that nobody is saying that dangerousness or a disposition or propensity to break the law is sufficient for a person to be found guilty of a crime. The Home Office proposals do not recommend that preventive detention be like prison. One of the functions of imprisonment (which ought only to occur when a person has in fact committed a crime) is punitive. For this reason conditions in prisons are designed to deprive prisoners of some of the comforts that the rest of us enjoy. If, as we have reason to believe on the basis of the Home Office report, institutions for the preventive detention of the very dangerous are to be well funded and well staffed care institutions then they will not be prisons in many of the important senses. The fact that the detained person will not, however, be guilty of any crime, means they will not be sentenced and therefore will not have a release date. So those who are deemed to continue to present risk could in effect be preventively detained indefinitely.

    I’ve suggested that the answer to the challenge of Professor Szasz is to justify compulsory treatment on the grounds of diminished autonomy and risk to self and others. This suggestion seems to damn the Home Office proposals for managing those with DSPD. Making a case for those with DSPD having diminished autonomy is a much more difficult matter than making such a case for somebody who is in the grip of an acute schizophrenic episode. While there are often causes for the behaviour of those with the most severe personality disorders, using this as grounds for diminished autonomy places us in difficult territory where it may no longer be clear how any of us can be considered to be autonomous (for more on this idea see Elliott).8

    The difficulty of predicting risk with the degree of accuracy that is required for preventive detention is one of the major difficulties with the Home Office proposal. If we could know with a very high probability that we were preventing a rape or murder by preventively detaining an individual then this is a compelling reason in its favour. This degree of certainty might, however, be closer to science fiction than to the reality of risk assessment. Nigel Walker has suggested that, “the harm a person has done is more than a mere indicator of what he is capable of doing. It is our only sound justification for infringing his right to be treated as harmless”.9 The thought here is that all of us should be presumed to be harmless unless we have proven otherwise. It doesn’t follow from the fact that a person has performed terrible acts that the will do so again but it is the best indicator that we have.

    These two major challenges must be overcome by the wording of new legislation, the way in which proposed new institutions are organised, and, crucially, how dangerousness is assessed and threshold of risk required for preventive detention.


    I think that we need to tread very carefully on the issue of preventive detention. There is a risk that dangerousness becomes an all too easy justification for us to use. The deep anger and outrage at the recent abduction and murder of two young Cambridgeshire schoolgirls could easily move us towards profoundly illiberal approaches to those who may present a risk to others. If we were to ask the public about the appropriateness of preventively detaining individuals with a predisposition towards acts of that kind the reaction is likely to be quite predictable.

    The decision to preventively detain a person with DSPD involves a delicate balancing of their interests, the extent to which their ability to act autonomously is impaired, and the seriousness of the danger they present to other persons. A bare minimum for whatever policy is adopted is that it is carefully underpinned by awareness of the importance of respecting and enhancing autonomy and balancing this against the probability and severity of risk.


    View Abstract

    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.