Experts, Rights and Precaution

Phil Bielby, LLB, PCHE, PhD

Lecturer in Law,
Law School,
The University of Hull,
Hull, UK

p.bielby@hull.ac.uk

and

Tony Ward, LLB, LLM, PhD, Barrister-at-Law

Reader in Law,
Law School,
The University of Hull,
Hull, UK

a.ward@hull.ac.uk

An early version of this article was presented at the ‘Experts, Authority and Law’ conference held at The University of Hull in 2009 to inaugurate the Experts and Institutions Research Centre. A later version was presented at a Philosophy Research Seminar at The University of Hull in 2010. Our thanks go to those in attendance at both who offered helpful comments and insights.

Copyright © 2012 Phil Bielby and Tony Ward
First published in the Web Journal of Current Legal Issues.

---

Summary

In this article, we offer an account of the epistemological and moral principles that should govern decisions where judges and other official decision-makers are asked to authorize courses of action which would amount to a violation of someone’s rights in the absence of justifying circumstances. Our argument is in four sections. In Section I, we outline the normative framework for our analysis. We draw on the theory of rights expounded by Alan Gewirth and developed by Beyleveld, Brownsword and Pattinson, but rather than relying on the reader’s being fully convinced by Gewirth’s argument for the ‘dialectical necessity’ of such rights, we suggest an alternative, contractarian defence of Gewirth’s ‘Principle of Generic Consistency’ (PGC) as a basis for social co-operation that all reasonable citizens could accept. Section II explores the epistemological implications of the principle of public justification: specifically, how the knowledge-claims of experts are made cognitively accessible to, and open to evaluation by, ordinary citizens. Section III brings together the epistemological and moral arguments to set out a framework of ‘rights precautionism’ by which non-consensual expert intervention in the lives of citizens can be regulated. Section IV illustrates the application of the principles developed in Sections II and III to two fictional examples drawn from mental health and capacity law.

---

Contents

Introduction

I. Rational Acceptability: (i) Moral acceptability

II. Rational acceptability: (ii) Cognitive accessibility

III. Competing Rights and Rights Precautionism

IV. Applications in mental health and capacity law

Conclusion

Bibliography

---

Introduction

Judges and other official decision-makers are often asked to authorize courses of action which involve an infringement of someone’s rights, i.e. what would amount to a violation of those rights in the absence of justifying circumstances. Such interventions are often advocated by experts who claim either that no-one’s rights will in fact be violated - for example, because a patient lacks decisional competence to refuse medical treatment - or that infringing a person’s rights (a justifiable restriction of rights)(1) is warranted in order to protect more important rights of others or themselves - for example, the compulsory detention of a person in a psychiatric hospital on the grounds of presenting serious risk of harm to others or oneself. Such acts require justification; and if the state claims to be one that can be accepted as legitimate by all citizens, the justification must be one that all citizens can rationally accept. It is not enough, or so we shall argue, that the infringement of rights be justified from the perspective of some group that possesses, or claims to possess, specialized knowledge. Those knowledge-claims must be rationally acceptable to citizens in general.

Where the person whose rights are to be invaded (for example by surgery) is a decisionally competent adult, the question of justification will typically be reduced to whether informed consent has been provided. Just how ‘informed’ the consent should be is debatable (see, for example, Beyleveld and Brownsword 2007, pp. 170-183, Manson and O’Neill 2007, chapters 2 and 3 and McLean 2010, chapter 2, but where the patient has genuinely consented to the invasion of her bodily and/or psychological integrity, the public justification of the intervention is, generally speaking, a fairly straightforward matter. More difficult questions arise where the affected person is, or is alleged to be, incompetent to decide by reason of youth, severe mental disorder or intellectual disability, or where a decisionally competent person can be detained and treated against her will on the grounds of severe mental disorder. These are the cases on which we focus here.

What we offer in this article is an account of the epistemological and moral principles that should in our view govern such decisions, rather than any detailed blueprint for legal reform. Briefly, we argue for two main principles:

A principle of public justification, which we will call ‘rational acceptability’, which requires decisions that impinge upon citizens’ rights to be justified in terms that citizens in general and where possible, the particular citizen directly affected) can rationally accept; and

A moral heuristic which we call ‘rights-precautionism’, which requires that when an error in one direction on a question of fact (e.g. determining that a person is incompetent to make a particular decision when in fact she is competent) is more likely to lead to a violation of rights than an error in the other direction, there should be a strong presumption in favour of the view which, if erroneous, will lead to the lesser violation of rights. The level of evidence required for this will depend on the probability of serious of the respective violations.

Our argument in this article is in four sections. In Section I, we outline the normative framework for our analysis. We draw on the theory of rights expounded by Alan Gewirth and developed by Gewirthian legal philosophers Beyleveld, Brownsword and Pattinson, but rather than relying on the reader’s being fully convinced by Gewirth’s argument for the ‘dialectical necessity’ of such rights, we suggest an alternative, contractarian defence of Gewirth’s ‘Principle of Generic Consistency’ (PGC) as a basis for social co-operation that all reasonable citizens could accept. In our view, the PGC as developed by Beyleveld et al. provides the best analysis of the rights of severely mentally disordered and intellectually disabled adults (see, in particular, Beyleveld 2012, Beyleveld and Pattinson 2000 and 2010, and Beyleveld and Brownsword 2001, as well as Bielby 2008). Section II explores the epistemological implications of the principle of public justification: specifically, how the knowledge-claims of experts are made cognitively accessible to, and open to evaluation by, ordinary citizens. Section III brings together the epistemological and moral arguments to set out a framework of ‘rights precautionism’ by which non-consensual expert intervention in the lives of citizens can be regulated. Section IV illustrates the application of the principles developed in Sections II and III to two fictional examples drawn from mental health and capacity law.

With the exception of the final section, our discussion in this paper is pitched almost entirely at the level of general principles. Anyone familiar with the way decisions about competence and compulsory treatment are actually made will recognize that there is a wide gap between the position we defend and current medico-legal reality (for a good example of the current reality, see Peay 2003). We do not believe that the normative argument we advance in this paper, however, is so far-fetched that it could not provide a basis for practical prescriptions for reform. Nonetheless, we must leave detailed consideration of that gap and the institutional changes that would be required to close it for another day.

I. Rational Acceptability: (i) Moral acceptability

The two principles we propose draw from two strands within a broadly Kantian approach to ethics and political philosophy, those of Gewirthian ethics (see, seminally, Gewirth 1978) and contractualism (again, see seminally, Rawls 1993, Rawls 1999 and Scanlon 1998 as well as the overview in Darwall 2003). The combination of the two gives us a picture of deliberative bodies constrained by a strong framework of rights. Our starting point is the widely held view in democratic theory, which we endorse, that the state must justify its use of coercive power in terms that all reasonable citizens could rationally accept (epitomised by Rawls 1993, Habermas 1990 and Estlund 2008). Rational acceptability has two elements: the normative grounds for coercion must be morally acceptable, and the factual grounds on which a norm is held to apply to a particular situation must be cognitively accessible to citizens in general, so that citizens can know that the situation is one in which coercion is authorized in accordance with the norm. We shall discuss moral acceptability in this section and cognitive accessibility in the next.

Our central claim is that no rational citizen would consent to their human rights, properly understood as protecting extremely important interests, being invaded without very good reasons that she could understand.(2) Evidently, not all citizens are rational all the time, and when significant departures from this standard occur that risk harmful consequences for others and occasionally unintended harmful consequences for self, we sometimes must invade citizens’ rights without their consent, either because this is refused or because they are incapable of offering consent or refusal at the material time.

In addition, of course, rational citizens can reasonably disagree. Even if one accepts, as we do, that there are substantive moral principles that all rational citizens must endorse, any plausible theory will still leave scope for highly significant disagreements (see, for example, Beyleveld and Brownsword 2006). How can legitimate decisions be reached in the face of such disagreements? A standard liberal answer is that they are legitimate if they are made

“in accordance with a constitution the essentials of which all citizens as free and equal may reasonably be expected to endorse in the light of principles and ideals acceptable to their common human reason.” ('Rawls 1993, p 137).

These “constitutional essentials” (Rawls ibid., p 137) include not only procedural arrangements but also a framework of rights, since reasonable citizens cannot be expected to agree to a ‘tyranny of the majority’. And, what is particularly pertinent to our argument, an adequate framework of rights must include some principles to guide decision-making when rights conflict or their application is unclear. If all reasonable citizens accept a decision-making body as legitimate, they cannot reasonably deny that its decisions should be followed, even if they reasonably regard some of its decisions as wrong (Estlund 2008, p 49).

At this stage, it is important to clarify what we mean by ‘rational’ and ‘reasonable’. In recent years, liberal political theory has sought to distinguish the two, epitomised in Rawls’s Political Liberalism. Rawls claims that a reasonable person is “ready to propose principles and standards as fair terms of cooperation and to abide by them willingly, given the assurance that others do so” (Rawls 1993, p 49) 'whereas a rational person is “a single unified agent . . . with the powers of judgment and deliberation in seeking ends and interests peculiarly its own” (Rawls ibid., p 50). For Rawls, “there is no thought of deriving the one from the other” (Rawls ibid., p 51). Some other philosophers, notably Alan Gewirth, take the view that the reasonable can indeed be derived from the rational: that to recognise one’s own right to the freedom and well-being that make successful agency possible while denying that others have rights to those necessary goods as well is self-contradictory (Gewirth, 1983 p 244). We need not resolve the precise relationship between the reasonable and the rational here, though we may at least understand the latter as a function of the former. What we do claim is that, whether reasonableness is strictly entailed by rationality or adds something to it, rational citizens who deliberated about the moral and political framework of their society would necessarily endorse a supreme principle of moral and political rights on the lines of Alan Gewirth’s Principle of Generic Consistency: the principle of respecting the rights of others as well as of oneself to the ‘generic conditions’ or ‘generic features’ of agency (the seminal statement of this argument is in Gewirth 1978, with a detailed reconstruction and defence in Beyleveld 1991). That is to say, they would agree that each citizen had rights to the elements of freedom and well-being that constitute the generic conditions of agency, i.e. action in pursuit of freely chosen purposes (see Gewirth 1982, especially at pp. 46-47 and pp. 52-53.). These rights comprise not only negative rights, for example to freedom from interference with bodily integrity, but positive rights to receive assistance from one’s fellow-citizens (delivered in practice through a state), for example, to mental and physical care. It is beyond the scope of this paper to defend that principle in detail, though below we indicate two ways in which the justification of rights might proceed – one which is dialectical necessary and another which is related to the acceptance of rights on grounds more familiar with contractarianism.

Gewirth’s dialectically necessary argument for the PGC involves three main stages (Beyleveld 1991, pp. 13-14 and pp. 21-46), which he claims must be accepted by anyone who acts for voluntary purposes of any kind. This proceeds from the first person standpoint of the agent. In stage (1), the agent recognises that she must necessarily have the ‘generic conditions’ of agency - freedom and well-being - which are the core features of any purposive action whatsoever, in order to be able to fulfil her particular purposes (whatever these may be) with any chance of success. Freedom and well-being include both needs for absence of interference and assistance, such as freedom from external constraint and adequate levels of shelter, nutrition and mental stability to facilitate the pursuit of her chosen goals ( see Gewirth 1978, pp. 22-63). In stage (2), the agent claims that she has these rights to freedom and well-being simply by virtue of being an agent (known as “the Argument from the Sufficiency of Agency” (Gewirth ibid., p 110)) and recognises that she contradicts herself as an agent if she tries to justify her having the generic rights on any other criterion ( see Gewirth ibid., pp. 63-103 and also Beyleveld 1991, pp. 23-42).(3) In stage (3), the agent moves from prudential to moral reasoning through universalizing these rights-claims: she accepts that since she has rights to freedom and well-being simply by virtue of being a agent, the same is true of all other agents by applying the ‘”logical principle of universalizability” (see Beyleveld 1991 p 44 and Gewirth 1978, esp. pp. 104-106). This leads to a categorical prescription for moral action, which is the essence of the PGC “Act in accord with the generic rights of your recipients as well as of yourself [emphasis in original]” (Gewirth ibid., p 135).

The controversy about Gewirth’s theory mainly concerns whether he can demonstrate a strictly logical progression from stages (1) to (2) to (3). As Beyleveld (2012) notes, stage (1) is relatively uncontroversial (but for an important criticism targeting stage (1), see Regan 1999). Many commentators, however (including one of the present authors) remain unpersuaded by stage (2). For the purposes of this article, we take an agnostic stance as to whether Gewirth’s argument succeeds on its own dialectically necessary terms. (One of us (PB) believes it does, the other (TW) does not.)mso-spacerun:yes'>  Irrespective of whether or not an agent is compelled to make those moves as a matter of rational necessity, we submit that a reasonable agent – one who is motivated to seek and abide by fair terms of social co-operation – would do so. In other words, we claim that it follows from Stage (1) that she could not rationally agree to political arrangements that would deny her those rights (unless she were willing to compromise her capacity for agency to some degree - for the qualification, see Beyleveld 2012, though we need not consider here what arrangements might be accepted by people who do accept significant limitations on their own agency, e.g. for religious reasons). A fortiori she could reasonably reject such arrangements(4). This approach is influenced by the idea of “the reasonable self”, developed by Gewirth (1988 p 144), in which the rational reflection that being reasonable entails takes precedence “over egoistic or particularistic rationality” (Gewirth 1983 p 244). Recognising her own needs for freedom and well-being and those of others, she would accept the creation of a system of mutually recognised rights to freedom and well-being. Being reasonable (in the Rawlsian sense (1993 p 49)), she would also be disposed to support even those rights that were irrelevant to her particular position and chosen purposes. If one grants that legitimate political arrangements must be rationally acceptable to all citizens, it follows that legitimate political arrangements must enshrine the rights that flow from the PGC.(5) Even if one wishes to contend that the universalization of agency rights which occurs at stage (3) of the argument is a move the agent could rationally deny, it is a move which all moral theories that incorporate a notion of equal treatment based upon a principle of relevant similarity require (or assume) the moral actor to make (for a defence of this principle, see Levvis 1991).(6) In this respect, Gewirth’s argumentis not susceptible to any greater objection than any other position in moral philosophy (and if the dialectically necessary method is successful, it is superior to one that appeals to a standard of reasonableness independent of rationality(7)).

We have set out this argument as concisely as we can because it is not the purpose of this article to develop or defend the first principles of a substantive moral or political theory. Rather, we have sought to offer plausible grounds for the moral acceptability limb of the principle of rational acceptability that we argue should govern the use of expert evidence in decisions affecting rights (the second limb, that of ‘cognitively accessibility’ we will turn to next). In setting out our stall – which behoves any normative argument - we aim merely to demonstrate that our core premises are not tied to one particular theory, and are less controversial than might be thought had they have been grounded in Gewirth’s dialectically necessary method alone. Since we cannot pursue these meta-ethical questions any further in a short paper, we are willing to accept that readers who are unmoved by our conception of reasonableness (or are not among the admittedly small number of those convinced by Gewirth’s dialectically necessary argument), will probably dissent from at least some of what follows.(8)

II. Rational acceptability: (ii) Cognitive accessibility

It is a fundamental democratic principle that the exercise of state power must be justified in terms that citizens can understand. This does not mean that, where the purported justification for an exercise of power depends on an expert’s opinion, the citizen has to understand every detail of the expert’s reasoning. Rather, it means that the citizen must have good reason to believe the expert’s conclusions. As Gutmann and Thompson put it:

“Citizens are justified in relying on experts if they describe the basis for their conclusions in ways that citizens can understand; and if the citizens have some independent basis for believing the experts to be trustworthy (such as a past record of reliable judgments, or a decision-making structure that contains checks and balances by experts who have reason to exercise critical scrutiny over one another).” (Gutmann and Thompson 2004, p. 5.)

Although this is on the right lines, it raises the difficulty that a completely independent basis for believing the experts to be trustworthy may be difficult to come by. For example, if we look for evidence that the medical profession has a past record of reliable judgments, we are likely to find it in research conducted by members of the same profession. And few citizens outside any particular group of experts can have a detailed first-hand knowledge of that profession’s decision-making structure. So we are thrown back either on the testimony of the experts themselves or of some group of meta-experts such as sociologists of medicine - and why should we trust them? The difficulty of finding solid ground for our judgments of trustworthiness does not arise only in our dealings with experts: it is a fundamental issue in the epistemology of testimony (see, for example, Coady 1992, pp. 46-47, Lehrer 2006 and Foley, 2001). If we did not start with some predisposition to believe that people who tell us things are trustworthy more often than not, our epistemic predicament would be dire (the classic defence of this view is offered by Reid: see Coady 2004). To some extent, experts have to be trusted to tell the truth about the matters from which they invite us to infer the reliability of their judgments - trusted not qua experts, but simply as responsible human beings. In any case, these issues are of more theoretical than practical interest in the context of institutional structures that are built on the assumption that professions such as medicine, psychiatry and social work are at least somewhat trustworthy.

The really difficult question is not whether citizens are justified in having some degree of trust in experts, but rather what degree of trust is appropriate when the rights of fellow-citizens are at stake. On the approach we advocate here, professionals should be required to satisfy some independent tribunal that the publicly accessible reasons for the proposed course of action justify an infringement of the patient's rights. Almost invariably, the publicly accessible reasons will be something less than the full range of reasons considered by the professionals. Professionals will rely to some extent on ‘tacit knowledge’, that is, on judgments for which they cannot fully articulate the reasons (Polyani 1983). They will also rely on scientific knowledge for which they cannot be expected to articulate the complete body of justificatory evidence and reasoning; indeed they will often not know this themselves, having accepted the knowledge from other experts. Experts will necessarily be selective in what they tell decision makers: they cannot recount every moment and every nuance of a conversation with a patient, for example. The decision-maker has to decide whether these reasons, together with the general reasons to regard the expert as a reliable assessor of the matter in question, are sufficient to justify authorising the proposed rights infringement. In other words, do the publicly accessible reasons, including the publicly accessible reasons to believe that the expert has good reasons that are not publicly accessible, suffice to justify the infringement? In this situation the decision-maker will often be justified in according a degree of epistemic or theoretical authority to the expert. The decision-maker is not as well placed as the expert to weigh up all the factors considered by the expert. Thus, if the decision-maker has good reasons to trust the expert, she may well have good reasons to accept the expert’s assessment of the balance of reasons rather than attempting to weigh up the reasons independently. To this extent, judicial deference to expert opinion may appear inevitable.

What we are dealing with here, however, are situations in which the decision-maker is concerned not simply to maximise the chances of forming true beliefs, but with adopting a basis for decision that minimises the risk of violating another person’s rights. Such a situation sets up a complex interplay between epistemic and moral reasons. Even if a decision-maker is justified in believing that an expert is more likely than she is to arrive at a correct view of whether a patient is decisionally competent, poses a genuine risk of harm to others or has any conscious experience, it by no means follows that the decision-maker is justified in accepting that conclusion for practical purposes, given the momentous ethical and legal issues at stake. This point is a familiar one in the context of criminal justice. Perhaps it is the case that the police are better placed than jurors to arrive at sound beliefs about the identity of criminals. But even a jury that accepted that view would not be justified in convicting a defendant unless they could be sure on the basis of the evidence presented to them that the defendant was guilty.

In certain respects, the epistemic or theoretical authority parallels the practical authority of law-makers as analysed by Raz. Raz himself has explored this parallel in a number of writings (most importantly in Raz 1986 pp. 52-53 and Raz 2009 pp. 154-158). In both cases the authority should reach a conclusion based on the same reasons that would lead the deferring party to a sound belief or decision if she had access to those reasons and weighed them correctly. In both cases the authority’s better access to the reasons gives the other party a reason to defer to his judgment. The difference is that in the case of practical authority, one who defers to the authority’s judgment cannot consistently do anything except act as directed, unless there are reasons not considered by the authority that defeat the directive. By contrast, someone who accepts another party as an epistemic authority on some matter has a number of options open to her. She may fully believe the authority. She may believe the authority but with less than complete confidence, given that her reasons to believe the authority’s conclusions can only be as strong as her reasons to trust the authority. She may suspend judgment, since she may believe the authority is more likely to be right than she without forming a positive belief that the authority is right. She may accept what the authority says for practical purposes without believing it; or she may believe it (with less than full confidence) but nevertheless accept the contrary proposition for contrary purposes, as a jury may acquit a defendant they believe to be guilty but not to the extent of being sure. Reliance on expert authority, then, is not inconsistent with the independent exercise of reasoning by the citizen. On the contrary, deference to expert authority, when it is justified by critical examination of the grounds for that authority, greatly extends the range of matters on which the citizen can form rational beliefs.

Reliance on authority, however, is only one way of using expert opinion. In many cases the citizen can and should understand the expert’s reasoning rather than deferring to her conclusions. This particularly true in two types of case that are relevant to our present concerns. One is moral or ethical expertise. Perhaps the most obvious example of an ‘ethics expert’ is an individual conversant in medical ethics who is appointed to sit on a hospital ethics committee (e.g. a clinical bioethicist) or an academic philosopher undertaking research and teaching in a university. In such cases, we agree with Varelius that expertise in ethics can exist and that it is analogous to other roles and professions about which we can plausibly speak of the presence of expertise (Varelius 2008). However, this amounts to only one possible sense of expertise in ethics.

There are several different kinds of moral expertise (see Weinstein 1994 ), but the one that is relevant for our purposes is expertise in producing cogent moral arguments. If the reasoning is cogent, as is characteristic of a strongly justified argument, it should (in the absence of strong contrary arguments) be capable of persuading any reasonable citizen to act upon it, provided that she accepts the general principles on which it is based. But these are decisions that reasonable citizens must, generally speaking, take for themselves. This does not mean that the reasonable citizen is expected to have the philosophical training and acuity of the moral philosopher - expertise in normative ethics is, after all “a claim to command knowledge in respect of the making of normative judgments not commanded by others” (Archard 2011, p 121). (Archard refers to “moral expertise” in the context of this passage, though from the preceding discussion it is clear he is referring to it in a sense similar to Weinstein.) What it does mean is that such individuals have the capacity for critical reflection on the ethical precepts offered by others as reasons for action, accepting those which cohere with the sense of reasonableness we have outlined above and rejecting those which, following Ronald Dworkin, rely upon prejudice, overwhelming emotional force, factual inaccuracy or unreflective conformity to received moral opinion (“parroting”) for their justification (Dworkin 1978, pp. 248-253).(9) Using Weinstein’s scheme, we might understand the moral expertise of the reasonable citizen as “performative expertise in ethics” – i.e. “to live well” (1994 p 70) - such that she is responsive to the legitimate rights-claims of herself and others.(10)

In one’s private life, of course, there may be exceptions to this – a person might feel too exhausted, or too emotionally involved, to trust himself to resolve a particular moral dilemma, and might defer to the judgment of someone he has good reason to trust. But these exceptions do not apply to persons who have been specifically entrusted with the task of taking difficult moral decisions on behalf some community. Moreover, moral expertise is quite different from expertise on the factual questions which decision-makers have to take into account. Most moral philosophers are laypeople in relation to medicine, and vice versa. So unless we want to entrust vital decisions to an elite of scientist-philosopher or medico-philosopher-kings – which, for reasons already given, we do not – we have to entrust them to people who, however good or even expert they are at moral reasoning, are laypeople in respect of the technical information on which they have to act.

The second kind of expertise to which we should be wary of according authority is what we may call hermeneutic expertise; that is expertise in interpreting texts, works of art, or human behaviour. An expert may help the citizen to understand a difficult poem, or a seemingly irrational piece of human behaviour, in a new light. (The analogy between art criticism and psychotherapy is also used by Habermas (1986, pp.20-21).) But to achieve this understanding the citizen must understand why the expert’s interpretation makes sense. If she simply accepts that the poem or the behaviour must mean whatever the expert says it means, without knowing why, then she has understood nothing. Much psychiatric evidence is best considered as a form of hermeneutic expertise. (The locus classicus of the hermeneutic approach to psychiatry is Karl Jaspers, General Psychopathology (1997, originally published in 1913).)

One form of psychiatric – as well as philosophical - hermeneutic expertise is the attribution of mental states such as intention to individuals. To say that a person’s action was done, or words spoken, with a particular intention or motive is to interpret it in a certain way; to describe it in a way that ‘makes sense’ in the light of reasons for action that the agent is taken to have had (Anscombe 1963). Whether or not an expert’s account of a person’s action ‘makes sense’ of it seems to us to be always a question that a reasonable citizen is competent to judge. It may be that the expert makes no sense of the action qua reasoned action, and instead explains it in purely causal terms, but in this case, the reasonable citizen can at least assess the senselessness of the action for himself. Even supposing that there is ‘one right answer’ to the question of what a person’s intentions or motives were, which could be determined by a precise knowledge of the sequence of neural events that led up to the act (this is a controversial view, and not one we endorse: see Hornsby 1993), such knowledge is not accessible even to the most expert observers. Given the epistemic elusiveness of past mental states, there is no reason for the reasonable citizen to take experts’ pronouncements about them as authoritative (Slobogin, 2000).

The difficult question is where the line is to be drawn between ‘moral’ and ‘hermeneutic’ aspects of expert judgment on the one hand and ‘technical’ or pragmatic aspects on the other. This is particularly salient in the area of psychiatry, where psychiatric classification schemes (such as the DSM and ICD nosologies) are constituted by a problematic mix of scientific and value judgments (see, for a detailed discussion, Sadler 2005 and Cooper 2005). Whether or how a patient’s speech or behaviour ‘makes sense’, whether it is contrary to her interests and whether it manifests a condition that will respond to a particular kind of drug are respectively hermeneutic, moral and technical/pragmatic judgments but they may all be made together in a single diagnoses. One possible strategy is to adopt an integrated approach, which draws together the ‘technical domain’ and value judgments (deriving from all relevant perspectives on the decision) into the decision-making procedure. In the context of psychiatric ethics, for example, an integrated approach – values-based medicine - has been proposed by Bill Fulford (2004), who argues that decision-making in health care must “stand on two feet” – that of facts and that of values (Fulford, ibid., p 206). According to this approach, the “process” of reaching a decision (ibid., p 205) is crucial to investigate the “space of values” (ibid., pp. 221-222) that are manifest within “evaluatively complex” (ibid., p 212) decision-making scenarios. We agree with Fulford that such a departure from “the traditional fact centred medical model” (ibid. p. 207 ff) is welcome and indeed necessary for a comprehensive understanding of decisions steeped in an array of facts and values. However, we would question how consistently this could be achieved when the criteria against which achieving “a balance of legitimately different value perspectives [emphasis in original]” (Fulford ibid., p 216) is not entirely clear and a vivid separation of facts and values may at times be difficult and perhaps impossible in some circumstances, such as in the diagnosis and treatment of personality disorders (for a discussion of the ethical as opposed to medical nature of a range of personality disorders, see Charland 2004).

The capacity of non-experts to examine and critically evaluate the reasoning of experts is not confined to moral and hermeneutic expertise. As Deidre Dwyer argues, the basic norms of rational inference are common to different disciplines, so anyone who has grasped those principles has at least a “limited epistemic competence” to test the soundness of a chain of inferences drawn in any discipline (Dwyer, 2008). Dwyer also stresses the important point that expert evidence in litigation is usually only part of a larger evidential matrix, and the inferences drawn by an expert can be tested for their coherence with inferences drawn by other experts. Dwyer does not, however, seem to us to give sufficient weight to some of the severe limitations in the competence of a non-expert to assess many kinds of expert reasoning such as experts’ reliance on tacit knowledge and the accumulated learning of an expert community, and the inevitable selectivity of the data and background generalizations that form the basis of an expert’s judgment (although Dwyer does, however, discuss the last point: ibid., pp. 144-145). We are readier than Dwyer to accept that there are good epistemological (and not just ‘political’) (cf Dwyer ibid., p 312) reasons for decision-makers to defer to the theoretical authority of experts, but we stress that such deference leaves a great deal of cognitive and practical autonomy to the court or the citizen.

In any case, it is unclear to what extent the human sciences, particularly in their applied forms such as psychiatry and social work discourse, can claim epistemic authority. The suspicion is that these disciplines (the double entendre is intentional) do not exemplify what Longino calls ‘tempered equality’ of intellectual authority (Longino 2002, p 131). As Habermas insists, rational discourse is one in which all relevant points of view are taken into consideration (Habermas 1990, pp. 65-66), so that mental health services users, for example, though not considered medical experts, are recognised as having knowledge derived from their own experience which is given respectful consideration in the formation of psychiatric knowledge. To what extent this is true of psychiatry today is not easy for an outsider to judge. (The importance of including the voice of service users in the formation of psychiatric knowledge is addressed in Bracken and Thomas (2005) and its contribution is also brought out vividly in Cresswell (2005).) If tribunals and other decision-makers accord too much deference to psychiatrists rather than listening to what their patients have to say, the result may be “epistemic injustice” (Fricker 2007), a failure to give due weight to people’s testimony on account of their status or the difficulty of understanding them. This is a further reason for treating psychiatric testimony, so far as possible, not as authoritative knowledge but as a form of hermeneutic expertise which can help decision-makers to understand the behaviour and testimony of those affected by their decisions.

In the sort of situations that we are concerned with here, where any available course of action risks bringing about a prima facie rights violation and we are unavoidably making decisions under conditions of uncertainty, the use of a heuristic to evaluate the moral consequences of acting to protect one right where the other is invaded seems inevitable. In principle, this seems to lead to a straightforward distinction between the prediction of consequences and their moral evaluation. In practice, under conditions of unavoidable uncertainty and with competing rights-claims in play, the picture is more complex. This leads us to our favoured approach to resolving these issues, which we elaborate in the next section.

III. Competing Rights and Rights Precautionism

Whether decision-makers are entitled to accept the experts’ testimony or advice as a basis for their decisions depends not only on epistemic factors but also on the moral issues at stake. In situations where accepting the erroneous advice of an expert would lead to the violation of a person’s rights, the level of confidence required (the standard of proof, in legal terms) will be high, and the graver the proposed rights violation the higher the standard.

The choice a decision-maker faces may be one between a course of action which will violate a person’s right in the absence of some fact asserted by the expert and a course which, whether the expert is right or wrong, will violate no rights. Other cases, which we focus on in this article, involve a conflict of rights, either between two or more rights of the same person (which we will call an intra-agent conflict) or between the rights of two or more person (which we will call an inter-agent conflict), For example, where someone is said by an expert to be in a persistent vegetative state with no prospect of recovering consciousness, continuing to feed or care for that person will prima facie violate no-one’s rights, even if the expert is correct. On the other hand , withdrawing nutrition will, if the expert is wrong, constitute an extremely serious violation of the patient’s right to subsistence and ultimately to life. But if the patient has made an advance directive refusing feeding in the event of being in PVS, feeding him will, if the expert is correct, violate the patient’s right to self-determination, and we have an intra-agent (or, strictly speaking, an intra-possible-agent) conflict of rights. An example of an inter-agent conflict would be where some compulsory intervention is said to be necessary to protect others from serious harm by a mentally ill patient. We discuss similar examples to these in Section IV, but first we examine the general principle that we advocate in such cases, that of rights precautionism.

When rights conflict, we must have regard to how important these rights are within the hierarchy determined by a substantive moral theory. As we argued above, the rights that must be respected in any legal order that all citizens could rationally accept are those entailed by the PGC. These can be understood as rights to the goods which are needs of agency (described above), and include both positive and negative rights, such as, for instance, the right to bodily and psychological integrity, the right not to be tortured and the right to education and healthcare (Gewirth 1978, pp. 54-58). These positive and negative rights can be ordered in a way that reflects their needfulness for the very possibility of action (Gewirth 1982, pp. 55-56; Gewirth 1978, pp. 343-344).

However, the likelihood of an action actually encroaching on one or more rights-claims also needs to be accounted for in deciding whether a particular action is permissible. Whereas the ranking of rights can be determined a priori, the weighting of risks depends upon specific contextual factors relevant to a particular case and takes place under conditions of uncertainty. Decisions of this kind must involve some form of weighing of the consequences for rights of alternative courses of action, in terms of what human rights law appropriately terms ‘proportionality’. In the last analysis, this involves weighing incommensurables: as Beyleveld and Brownsword (2006 p. 148) and Beyleveld and Pattinson (2010 pp. 268-269) acknowledge, there is no formula that will tell us that an x per cent risk of violating a right to a basic good (i.e. one which is essential to the very possibility of successful agency (Gewirth, 1982, pp. 55-56 ) outweighs a y per cent risk of violating a right to a non-subtractive good (i.e. one which preserves an individual’s existing possibilities for agency, such as sufficient information with which to make a choice relevant to her action, Gewirth, ibid., p 56). Moreover, the balance of risks will almost inevitably look different from the perspective of the expert and from that of the decision-maker.

Despite the absence of uniquely ‘right answers’ to problems that involve prima facie rights violations no matter what course of action is taken, breaking the decision down into two elements – the relative importance of the rights at stake and the relative risks of violations of each right - is a potentially useful starting point for this heuristic. As outlined in Section I, we take the view that all prospective purposive agents, i.e. all those who are capable of prospectively valuing purposes and thus of self-consciously having reasons for action,(11) have rights to the generic conditions of agency. However, it is not always clear whether a being or entity is an agent or not. We accept the philosophical problem of other minds and the theoretical possibility that one or other of the present authors (or you, the reader) is the only genuine human agent in a world of fiendishly clever simulacra contrived by some Cartesian demons (see Beyleveld and Brownsword 2001, p 120, Beyleveld and Pattinson 2010, p 260; and Beyleveld 2012, pp. 9-10), but for practical purposes, the existence of other minds, and other agents, is a fairly straightforward instance of inference to the best explanation. Such an inference, of course, cannot afford an absolutely conclusive demonstration but it can provide a strong warrant for believing that another agent exists (and can warrant a rebuttable presumption of the non-existence of agency in machines, so long as any agent-like behaviour they display short of ‘apparent’ or ‘ostensible’ agency can be explained by their design or programming) (see Hyslop, 2010). In some cases the degree of probability that a being or entity is an ‘apparent’ or ‘ostensible’ agent truly is an agent may be less than typically warrants the inference. For practical purposes, then, ‘ostensible’ or ‘apparent’ agents must be treated as having rights, even if this rights attribution might be mistaken in some cases (see Beyleveld 2012, pp. 9-11).(12)

To be an agent with rights does not necessarily equip one with the capacities required for full citizenship. We can draw the distinction between agents who can be assumed to have the required capacities and those who cannot in the following terms. The vast majority of adults possess a set of reasoning, behavioural and communicative capacities (much in excess of the bare threshold required for agency) that allow them to participate meaningfully in the life of their society. Education, socialization and love, operating in conjunction with the usual stages of psychological development, are key to developing the capacities required fully to exercise one’s rights and to participate in civil and political life.(13) We can think of adults having undergone such processes as agents of full societal competence (for a more detailed discussion of this point, see Bielby 2008, chapters 1 and 5). By contrast, other human beings either will temporarily, permanently or intermittently lack this competence whilst still retaining the capacity for agency. They may be considered ‘vulnerable’ in the sense that their restricted sphere of decisional competences places them at risk either of direct physical harm of long-term detriment to their development as societally competent agents. For example, children of primary school age have a right to education but do not yet have sufficient knowledge or decision-making capacity to decide whether to avail themselves of that right. Human beings with heightened vulnerabilities of a primarily cognitive nature, for example, adults with severe mental disorder or intellectual disability as well as young children, must, in an appropriate way and to a proportionate extent, receive the protections entailed by their having rights to protect them against suffering and wherever possible to nurture their own autonomy (see for further discussion of this point, Bielby, ibid. chapter 5). It would therefore be permissible - indeed, in many cases, obligatory - to treat them as societally incompetent agents and to restrict their participation in fundamental deliberative processes, as well as acting paternalistically to further their interests in ways they do not choose. This illustrates how the minimal threshold of agency - proactive purpose valuation - is far lower than that required for citizenship in this sense. So, whilst all agents possess rights under the ‘will’ conception, not all agents possess the decisional competence to exercise or waive all of their rights. Those who lack these citizenship capacities of course remain ‘citizens’ in a broader sense (being entitled to social welfare provisions, passports, etc.). Indeed, those with intermittent or partial capability for societal competence may be competent to participate in specific fora, e.g. as representatives on patient and children’s panels in health and social care provision. But these differences are irrelevant to why we should treat both those who possess and those who lack these citizenship capacities as agents.

At the extreme, of course, agency itself may be questionable. Should medical experts convince us that the individual in question is brain-dead or irreversibly comatose, thereby lacking all capacity for agency, then in principle that individual, though still morally valuable, ceases to be a subject of rights - at least of the ‘will conception’ rights that can be enjoyed only by actual agents (Beyleveld 2012, p 11). As Beyleveld and Pattinson put it, “precaution requires agents to treat other beings as agents only if and to the extent that it is possible/meaningful to do so” (Beyleveld and Pattinson 2010, p. 262). The moral stakes involved in withholding ascriptions of agency, or failing to take seriously that its capacity may somehow be retained, are so high, however, that we may seriously question whether the requisite standard of proof that agency is lacking in such cases could be met at all.(14) This is because the consequence of error would be among the gravest moral wrongs: to treat an entity as a non-agent (and thus as a non-rights-holder) when the entity actually possesses the capacity for agency, however dormant - and where to treat that entity as an agent would not undermine the agency rights of others who, on the evidence, are more probably agents (Beyleveld and Pattinson 2000, p 42 ). It is on this precautionary ground of attaching moral weight to evidence of agency, rather than by virtue of mere species membership, that beings or entities merit being treated as subjects of rights.

To be clear, the rights precautionism approach adopted here is not the same as the currently dominant ‘precautionary principle’ as used in environmental and biotechnology regulation, but rather has a determinate moral content influenced by Gewirthian rights theory. (For a critical discussion of the precautionary principle in the former context, see Feintuck (2005) and in the latter, see Engelhardt and Jotterand (2004).) As already indicated, our thinking about precaution follows similar lines to Beyleveld, Brownsword and Pattinson, the first two of whom have pointed recently to the criminal standard of proof as an example of precautionary reasoning in law (Beyleveld and Brownsword 2009, esp. pp. 183-185). The criminal standard of proof rests on the right of every person not to be punished for acts of which they are in fact innocent (Dworkin 1985). Because punishing an innocent person would violate that right, whereas not punishing them would not directly violate anyone’s rights (though it might place others at risk of having their rights violated), the presumption of innocence requires that they be treated as belonging to the class who enjoy the right not to be punished unless the contrary is proved to a very high degree of probability – in an old phrase, to the level of ‘moral certainty’ (Shaprio 1986).

An important feature of standard of proof rules is that they do not allow a consequentialist balancing of risks on a case-by-case basis. For example, faced with a strong but not conclusive case against an alleged serial killer, it might be reasonable to think that the probability that if he is set free and is guilty, he will kill again, outweighs the risk of imprisoning an innocent man. To allow such discretion (or at least to allow it overtly – there is no guarantee that juries will not in fact reason in this way) would, given the inevitable pressures on decision-makers to err on the side of public protection, erode the presumption of innocence to the point of destruction. The benefits of a uniform rule in protecting the innocent outweigh the seriously suboptimal consequences of applying the rule in a small number of cases. Similarly in the family courts, the facts relied upon to establish that a child is likely to suffer significant harm must be proved on the balance of probabilities, without taking account of the degree or risk or the gravity of the harm to which the child will be exposed if the allegations are true (Re B (Children: Standard of Proof) [2008] UKHL 35, [2009] 1 AC 11). This leads to a suboptimal allocation of risk in some cases, but is deemed necessary “to protect both the children and their parents from unjustified intervention in their lives” (ibid., para. [54] (Baroness Hale)). Of course, since a substantive moral theory determines the importance that attaches to the competing rights, other controversial cases would invite a range of different responses. In the next section, we will explore our approach towards weighing rights in the context of expert judgments about compulsory psychiatric hospitalisation and refusal of psychiatric treatment.

Rights-precautionary reasoning in one sense involves a “rights-consequentialist” or “deontological consequentialist” approach, in which the process of deciding upon the right action in any context requires us to take account of the consequences of choices available to us in that context for respecting the rights of agents (see Beyleveld and Brownsword 2001, p 184, n 19 and Gewirth 1978, p 216). On one level, it is a form of rule-consequentialism, though not of rule-utilitarianism,(15) since the undesirable outcomes to be avoided are identified by a rights-based moral theory and act as ‘trumps’ over any merely utilitarian reason for decision (Dworkin 1981). (A more difficult question, which we need not pursue here, is whether rights always ‘trump’ non-rights-based moral reasons.) As such, our approach to rights precautionism would reject the optionality of rights which remains open under standard consequentialist accounts. The rules must, however, themselves conform to a principle of proportionality. The risk of morally undesirable decisions resulting from the application of the rule must be proportionate to the risk of undesirable decisions that would result from allowing decision-makers greater discretion.

On these grounds, it seems possible to justify the following general principles regarding standards of proof:

1. Where it is foreseeable that in the great majority of cases the undesirable consequences of wrongly finding a fact proved will greatly outweigh the consequences of not finding it proved when it is in fact true, and where allowing decision-makers to weigh the risk of error in particular cases carries a high risk that they will reach unjustified decisions, a ‘beyond reasonable doubt’ standard of proof should be required;
2. Where it is foreseeable that in the great majority of cases the undesirable consequences of an error in either direction are of approximately equal gravity, and where allowing decision-makers to weigh the risk of error in particular cases carries a high risk that they will reach unjustified decisions, proof on the balance of probabilities should be required;
3. Where the relatively gravity of the consequences of different errors varies greatly between different cases, decision-makers should be permitted to take account of the relative consequences of different errors in each case, provided that procedures can be designed that will result in fewer undesirable decisions, or less gravely undesirable decisions, than adherence to a uniform standard of proof.

The really difficult question, however, is what to do about types of case where the consequences of error are very variable and there is no procedure that will allow wide discretion in particular cases without producing a high risk of morally undesirable decisions. Unfortunately, this is precisely the situation that obtains in many mental health (and child care) decisions. On the one hand, it is impossible to generalize about whether the consequences of compulsorily detaining or treating somebody (or removing a child from its family) are more or less serious than the consequences of not doing so, because the consequences vary so greatly from one case to another. On the other hand, particularly in mental health cases,(16) the pressures on decision-makers to avoid decisions that result in a mentally ill person harming someone else are much greater than the pressure to avoid unnecessary detention or compulsion, so that unless decision-makers are remarkably virtuous, any system that allows them broad discretion will almost inevitably result in injustice. (Then again, given the balance of pressures on legislators, any rule-based system is very likely to be unjust as well.)

Supposing, however, that decision-makers are sufficiently virtuous to do justice in such situations, they could rationally accept the following Principle of Rights Precautionism, which develops Beyleveld, Brownsword and Pattinson’s formulation of precautionary reasoning, especially their “Criterion of Avoidance of More Probable Harm” (see Beyleveld and Pattinson 1998 (unpublished paper, on file with PB) and Beyleveld and Brownsword 2001, p 123 as well as the discussion in Bielby 2008, pp. 103-105.):

The justification of any course of action or inaction in a case that involves the unavoidable prima facie violation of rights hinges upon:

1. the relative importance of the rights at stake;
2. the strength of the evidence available to suggest that one or more of these rights would be violated were a certain course of action or inaction taken;
3. ensuring that the person(s) whose rights are affected has been given a fair opportunity to challenge that assessment, either on their own behalf if they have the mental capacity to do so or, if not, through an advocate or proxy;
4. the relative severity of the consequences false positive and false negative judgments as to the existence of sufficient grounds for infringing a right.

Thus, to satisfy the moral acceptability criterion of our principle of rational acceptability outlined in section I, and the cognitive accessibility criterion outlined in section II, one should respond in such situations in a way that minimizes the negative consequences of the possible rights violations, determined by the relative weight of the rights of the individual(s) so affected and the probability of violations actually occurring should one option be chosen over another (including the probability of the consequences of error), informed by the most comprehensive expert information available at the relevant time.

IV. Applications in mental health and capacity law

To illustrate the points made so far, let us consider two examples: the example of judging capacity and the example of involuntary detention and treatment for mental disorder.(17) We will approach these through the fictional cases of Emma and Paul.

Clearly, on our theory there would be no question of the mentally ill not having rights, and the particular ‘heightened vulnerabilities’ of this group would require us to reflect on how to respect their rights when their capacities for exercising these might be significantly impaired (for discussion, see Cavadino 1997, Matthews 2000 and Bielby, 2008). In cases like these, then, we have an ‘intra-agent’ conflict of rights (i.e. between the bundle of rights that agents have) that requires ethical resolution: the negative right not to be detained and medicated against one’s will and the positive right the person has to receive the support of others (that includes mental health professionals) to experience good mental health (and by this we mean rights which are also engaged outside of the psychiatric context). In certain circumstances, the above rights need to be further weighed against ‘inter-agent’ rights (i.e. the rights other agents have against the agent in question). Most obvious in this context is the right of other members of the public to be protected from unintended harm that may arise were the person with mental illness not to be so detained. However, this also may engage both an intra-agent as well as an inter-agent right justified by the mentally ill person’s own interest that she has in being restrained from engaging in such action when this is a product of her disorder which compromises her capacity for responsibility.

Though the ethics of compulsory psychiatric hospitalisation has been debated widely (e.g. Chodoff 1976, Culver and Gert 1981 and Szasz, 1997), the role of expert and lay influences on decision-making in such contexts has been rarely discussed. In one of the few papers that considers this question explicitly, Stephen Morse claims:

“The use of experts encourages courts, legislatures, and legal decisionmakers to avoid the hard social, moral, and legal questions posed by mental health laws by responding as if there were scientific answers to them . . . laypersons are perfectly competent both to provide most of the observational data necessary for mental health decisions and to make such decisions.” (Morse 1978, pp. 602-603)

We share Morse’s objections to unduly technocratic approaches to decision-making, which tend towards giving experts broad discretion, in making difficult choices about rights invasions in psychiatry. However, Morse’s failure to develop a perspective on which, if any, layperson’s perspective should be prioritized over another deprives his account of a deeper normative resonance. Indeed, his approach seems to proceed from little more than intuition:

“Using lay as well as expert testimony about the actor's behavior, the decisionmaker can then decide if the person is sufficiently crazy to be an appropriate candidate for the application of mental health laws. If the factfinder's response to the behavioral data it hears is "so what," then the actor probably does not meet the legal criterion of mental disorder; if the response is "that's crazy" or "he's crazy," then the criterion of mental abnormality may be met.” (Morse ibid., pp. 612-613)

We find this approach unsatisfactory for two reasons. First, the absence of an ethical framework to guide decision-making other than intuition is very likely to lead to inconsistencies and a lack of transparency in the decision-making process. It is also not clear why intuition should represent a more attractive approach than one grounded in a moral theory capable of rational acceptance, such as that we outlined in Section I. Second, despite Morse’s keenness to reduce the significance attached to expert testimony in making such decisions (with which we are, of course, sympathetic), his approach does not make clear whether lay interpretations of the mentally ill persons’ behaviour are always the standard by which the rightness of any subsequent action should be judged, including those which may infringe the mentally ill person’s rights. As stated, Morse’s account could be used to justify the primacy of the citizen’s perspective only where the fact-finder or decision-maker sees fit; in other cases, especially where the decision-maker is a specialist (e.g. a psychiatrist or a judge) or the body making legal decisions is comprised of specialists alone (e.g. a review tribunal), the citizen’s perspective could be taken into account only to be set aside when reaching a verdict. This possibility appears to be left open by Morse when he later goes on to claim “[w]hether a person cannot or will not think straight or control himself is a moral and commonsense judgment that should be made by the legal decisionmaker”. (Morse, ibid, p 618) We do not think that ‘commonsense judgments’ (if such a phenomenon exists) can be used as a synonym for the perspective of public justification that is rationally acceptable.

If previous approaches bear little resemblance to our own framework, when would psychiatric detention be a justifiable infringement rather than a violation of someone’s rights under rights precautionism? In our view, the following stages would need to be satisfied. First, the rights that would prima facie be violated as a result of the decision to detain or not detain would need to be identified and taken into account. Commonly, this would amount to one’s rights to freedom of movement and association, along with the right not to be subject to personality-affecting psychoactive medication against one’s will, the latter which is best understood as a species of the more widely known right against torture.(18) Second, there would need to be sufficient expert evidence to support the claim that the person in question is suffering from a mental illness of such a degree as to warrant the judgment that they were incapable of making competent decisions about their own treatment, and/or were posing a substantial risk of harm to themselves or to others.(19) It is in this respect that evidence is most likely to be presented in the specialized language of psychiatry: nosological classifications, case histories, prescribed medications and recommended treatments. As discussed in section II, some of the more ‘technical’ aspects of psychiatric knowledge, for example about appropriate medication and how it should be administered, may need to be treated as authoritative for practical purposes here, after both the scientific and ethical aspects of the knowledge have been tested by appropriate procedures that allow for dissenting voices to be heard. However, we would need a critical awareness of how descriptive explanations of syndromes in psychiatric judgments are intertwined with normative judgments about ‘normal’ and ‘deviant’ behaviour in order to avoid the ambiguity identified by Weisstub that “it is often unclear, when psychiatric evidence is used, whether the role of psychiatry is merely factual rather than constituting the substance of the standard” (Weisstub 1978, p 45).

Third - and this goes to the heart of rights precautionism - the moral significance of the rights identified at the first stage and the consequences of erring either way would need to be considered. To illustrate how this stage might be addressed, let us consider two hypothetical scenarios that involve some degree of risk for rights flowing from a false positive and false negative decision. The first concerns the potential of harm to self only, and the second potential of harm to others as well as oneself.

The case of Emma

Emma, a middle-aged woman, has suffered from severe depression for some time. Having been detained under mental health legislation, she is subject to a course of electro-convulsive therapy (ECT) for her mental disorder against her will in the belief that it is clinically indicated and that she lacks the competence to consent to or refuse treatment.(20) (Under current English mental health law, a patient who “is capable of understanding the nature, purpose and likely effects of the treatment” can normally consent to or refuse ECT, unless in an emergency (Mental Health Act 1983 s. 58A(3)(c) and s. 62.)).(21) On subsequent reassessment, Emma’s refusal is found to be the product of competent decision-making. The use of ECT is discontinued and she is returned to standard (less efficacious) antidepressant medication. A few months later, her healthcare treatment team proposed that a course of novel antidepressant medication is used instead, to which Emma agrees, and which proves beneficial. Over time, this makes Emma well enough to undergo psychotherapy, which eventually leads to her being discharged from the hospital. However, the trauma induced by the experience of coercive ECT causes Emma to become particularly distrustful of psychiatrists, and her recovery, though ultimately still achieved, is delayed.

The case of Paul

Paul, a quiet and gentle young man, has been diagnosed with paranoid schizophrenia. His delusions cause him to believe that those in control of the IT systems at his former workplace are sending emails encouraging his former colleagues to persecute and harm him. Since Paul retains some insight into his condition and is compliant, his consultant psychiatrist initially recommends anti-psychotic medication outside of hospital, rather than compulsory treatment. However, this generally suppresses rather than dispels the delusion. One day, Paul returns to his former workplace and confronts the Head of IT, who he accuses of masterminding a hate campaign against him. Paul becomes agitated though not physically aggressive and is led away from the premises by security staff. There are no witnesses to the altercation, but the employee reports that he felt physically threatened by Paul’s conduct. When being questioned by police, Paul denies any intention to harm anyone, states he does not wish to return and claims he just wants the persecution to stop. As it happens, Paul never had such an intention. However, the decision is taken to detain him for assessment under mental health legislation on grounds of potential harm to others.

From the perspective of rights precautionism, is the error of the initial competence assessment which led to the coercive ECT in Emma’s case more or less significant than the error of believing that Paul presents harm to others when he does not? In Emma’s case, the competing rights in play are her right to be free from unwilled interventions into her psychological and physical integrity and her right to be helped to experience good mental health. In Paul’s case, the competing rights are Paul’s to have his treatment managed in circumstances which present the fewest restrictions on his freedom and the right others have against Paul that he does not violate (or threaten to violate) their physical integrity. However, the judgment that Emma was incompetent to consent to a treatment she would otherwise have the right to refuse (in English mental health legislation, at least) and that Paul presented such a risk were in both cases erroneous. Let us suppose that had the error in each case not been made, Emma would have had her refusal of ECT respected and her subsequent trauma would very likely not have arisen. Paul would have continued with consensual medication outside of hospital, and honoured his pledge not to return to his former workplace. In order to make our conclusions less obvious, let us further suppose that despite the error and the consequent implications for the rights of both Emma and Paul respectively, the decision-makers in each case took steps to establish and evaluate relevant information before arriving at their decision, presenting it in a way that would be rationally acceptable. These steps would clearly diminish the procedural shortcomings of the decision in ethical terms.

Nevertheless, in each case it remains that a significant right held by Emma and Paul (on our rational acceptability argument) was overridden. Let us now look to the consequences of deciding the other way. Were it possible to be sure of Emma’s competence and that Paul would remain non-threatening and compliant with treatment, then there would be no question that the decisions in both cases were wrong. The problem arises because we cannot know this – it is possible that Emma could have indeed been incompetent all along, in which case allowing a putatively competent refusal could have been detrimental to her treatment (and thus to her right to be helped to achieve mental health), especially had the novel treatment not become available. Equally, Paul could have been prepared to lie about his true intentions and, were he not detained for assessment and possible treatment, return to his former workplace to commit a violent attack on the employee.

At the level of probability, on one level it seems there is little to distinguish Emma’s actually being incompetent from Paul’s actually posing a threat of harm to others. At most, we might only be able to rely upon the existence of reliable empirical data in both contexts to establish the likelihood of competence in severely depressed patients when incompetence has been found or the likelihood of ‘dangerousness’(22) in an apparently compliant patient with some insight into their illness. However, in English mental health law, at least, compulsory hospitalisation and treatment is not subject to a test of ‘impaired decision-making capacity’ or ‘considerable risk of serious harm to others or self’, along the lines of that which has been proposed by commentators arguing for reform (see, e.g. Doyal and Sheather 2005, Dawson and Szmukler 2006, and Richardson 2007). It is therefore possible to detain Paul, at least temporarily, on a weaker standard of suffering from a mental disorder “of a nature or degree which warrants . . . detention . . . in a hospital for assessment (or for assessment followed by medical treatment) . . . in the interests of his own health or safety or with a view to the protection of other persons” (Mental Health Act 1983 (as amended by the Mental Health Act 2007), s.2(2)a-b). We will return to this point shortly. It will usually remain the case, though, if the empirical evidence gleaned is itself inconclusive, or if no such studies exist, we then need to turn to the remaining variable: the significance of the possible harm.

In the first case, Emma’s ‘inter-agent’ rights to be free from unwanted psychological and physical interference might be thought to be quite finely balanced against her right to be helped to achieve mental well-being – both are, after all, a manifestation of the negative and positive fundamental rights to psychological well-being, which are a precondition of effective agency (see Gewirth 1982, pp. 55-56). Though the rights denied by the decision were significant and the consequences for her of being coerced into treatment distressing, we need to weigh these against the implications for Emma’s positive right in mental health which would have been violated had a ‘false positive’ decision of competence been reached. While Emma’s case involves a difficult balance of rights, it involves a relatively simple question of fact: does she have mental capacity to decide whether to consent to ECT, as defined by the Mental Capacity Act 2005 (hereafter MCA)? What is involved is essentially an interpretation of what Emma says about her refusal and the reasons (if any) for it. We suggested above that in such questions of interpretation, decision makers should not simply defer to expert opinion but should seek to understand the person’s words and actions for themselves, taking account of the expert’s views.(23)

Emma’s case appears to involve a scenario where an error either way will be of comparable seriousness: on the one hand, detrimental consequences would follow from the false negative denial of competence whereas wrongly deeming her to be competent to refuse ECT would involve continuing violations to her right to mental health. This suggests that a decision-maker ethically need only be satisfied on a balance of probabilities that Emma is incompetent – which reflects the current English legal position under s.5(1-2) of the MCA. If after making all reasonable enquiries the decision-maker is satisfied that Emma is incompetent, an infringement of her right to physical and psychological integrity might appear to be justified as a way of helping her to secure her equally important (and closely related) right to good mental health.

But matters are not quite as simple as that. Even if the balance of probabilities tips marginally towards incompetence, there remains a significant risk that Emma’s competent refusal is being overridden; and even if Emma is truly incompetent, in English law, her wishes remain a factor that can properly be taken into account in assessing whether ECT is an “appropriate” treatment.(24) If the benefits of ECT are uncertain in relation to the side-effects or only marginally superior to alternative treatments, the invasion of her bodily integrity by compulsory treatment might be unjustified even though she was probably incompetent to refuse. Conversely, if there is evidence which suggests that on balance Emma may be competent, but this is short of clear evidence to support such a judgment, and it is plain that the treatment would be enormously beneficial and her prognosis without it extremely bleak, it might be justifiable to reach a judgment of incompetence and override her refusal if the consequences of doing so would be less serious than reaching a ‘false positive’ finding of competence.

A relevant factor in resolving the dilemma is the consequence for the patient of the decision over time – here, but for the emergence of the novel treatment which could not have been foreseen, Emma could have remained for a much more extensive time in psychiatric care awaiting recovery to the point where she could meaningfully undergo psychotherapy – justified perhaps if we could be clear that she has the competence to refuse but not, if as was suspected, she had not. Since the investigation of Emma’s competence was not procedurally deficient – and we think a ‘reasonable belief’ of a lack of competence is a morally sound principle from the perspective of rational acceptability to absolve mistake - then we can say that Emma’s rights here were infringed but not violated.

In the second case, it seems that Paul’s ‘intra-agent’ right to be helped to full mental health (which he has shown to value through consenting to treatment outside of hospital) is being overshadowed by an undue emphasis placed upon ‘inter-agent’ rights, given the facts known. This is facilitated by the discretion afforded to mental health professions (in England and Wales as in other many other jurisdictions) to detain on the grounds of potential of causing harm to others. Where the risk of harm to others is real and the type of harm threatened is morally significant (e.g. serious physical harm rather than, say, harm to property), the rights precautionary ground for overriding liberty rights temporarily (albeit on a clearly articulated and reviewable basis) for the sake of upholding rights against physical attack – and sometimes the right to life – is decisive. The proviso is that the right infringed (e.g. liberty) should be justified by the risks presented to more important rights of other agents (e.g. to their bodily integrity and life) (see further on this point, Bielby 2008, p. 200). In addition, it is possible that psychiatric detention on the grounds of harm to others could be further justified with reference to the principle set out at the beginning of this section that one has a right to be prevented from causing harm to others for which one could not be held responsible. As Monahan and Wexler argue, provided we understand the prediction of dangerousness as a statement of probability, “rather than an absolute claim that violent behaviour will occur” (Monahan and Wexler 1978, p 38), then even the highest standard of proof available - beyond a reasonable doubt – may be satisfied on these terms (ibid.). This means that, notwithstanding the severity of the limitations placed on the rights of the patient detained, detention of a person with a mental disorder whose behaviour presents a demonstrable risk to the basic rights of other agents is justifiable on a balance of probabilities standard of proof.

However, in Paul’s case there are compelling grounds for doubt as to whether he poses a risk of harm to others on the balance of probabilities, despite his delusions and behaviour at his former workplace. Since Paul’s ‘intra-agent’ rights are unlikely to be advanced in the hospital setting, and he retains some insight into his condition, we can conclude that the decision to prima facie violate his liberty rights can only be shown to be a justifiable infringement if the threat that he poses to other agents is real. From the perspective of rational acceptability, we doubt that it could be shown. First, there is no compelling evidence that Paul is causing a risk of harm to others – despite the remark made by the employee that he found Paul’s agitated behaviour to be threatening. Second, it is not clear what assessment would achieve in terms of the rights at stake in the scenario – if the risk of harm to others Paul presents is low, and his treatment has been provided with reasonable success outside of hospital, then it is likely he would be discharged swiftly in any event. Third, there is a risk that the experience may cause Paul to feel stigmatized, and like Emma, to erode his trust in the psychiatric profession, thus hampering his recovery. Though rights precautionism could readily conceive of situations where a patient should be detained on grounds of the potential to cause serious harm to others – à la the Richardson recommendations - Paul’s actions or intentions would have had to pose a more palpable threat than they were here for that criterion to be met.

What both these cases tell us is that the particular vulnerabilities faced by people with mental health problems, both cognitive and circumstantial, should encourage us to reflect upon how acts that affect their rights can leave them especially open to further psychological harm. This is consistent with acknowledging that there needs to be a sensitive appreciation of the consequences for all rights holders concerned of deciding one way or the other. Such insights need not lead us to ignore the practical constraints on the decision makers but to ensure the rights at stake in the decision are interpreted with due regard to the plight of especially vulnerable citizens. In terms of decision-making about competence to refuse treatment and compulsory hospitalization , this may mean taking urgent decisions that are believed to accord with the perspective of the reasonable citizen and then having them retrospectively scrutinized, where this represents the earliest opportunity for the grounds of the decision to be assessed. Already, through the system of mental health review tribunals, mental health experts can offer persuasive testimony when a decision is challenged that might lead to the decision ceasing to have effect. However, our account would go further insofar as it would favour greater lay and service user participation in review tribunals and in ethics committees at psychiatric hospitals. The opportunities offered by such approaches are topics for future research but should be taken very seriously if we are not, as Bartlett and Sandland warn, “to treat mental health law as a set of academic constructs, and ignore the people contained within the system” (Bartlett and Sandland 2007 p 31).

Conclusion

In this article, we have argued that a normative principle of public justification with a moral acceptability and cognitive accessibility limb which we have called ‘rational acceptability’ would support a theoretical account of human rights and a heuristic for resolving conflicts of rights under conditions of uncertainty - ‘rights precautionism’. This builds upon the work of Alan Gewirth and Gewirthian legal philosophers Deryck Beyleveld, Roger Brownsword and Shaun Pattinson, although we have argued that Gewirthian rights theory also can be supported from a contractualist position. Following this approach, we have argued that it is from the perspective of rational acceptability that assessments of expert evidence should be made that inform decisions that involve prima facie rights violations. This approach divides responsibilities between experts and the representatives of the citizenry in a way that does not lead either into technocracy or to a ‘decisionist model’ which sees the non-expert as making essentially arbitrary or non-rational choices between values.(25) The official decision-maker will sometimes be dependent on experts for information bearing on the question of whose rights are affected by a decision (e.g. the psychiatric patient alone or members of the public as well) and, more often, for information bearing on the assessment of the risks of prima facie rights-violations flowing from various alternative courses of action (e.g. to what extent, if at all, does a person diagnosed with paranoid schizophrenia present a genuine risk of harm to others or herself and will this risk be minimized if she is deprived of her liberty and subjected to compulsory treatment?). However, the fundamental normative question of weighing rights is to be judged from the perspective of rational acceptability (e.g. whether the degree of risk posed is too low to justify compulsory detention and treatment).

On this account, on the one hand, the weighing of such alternative consequences is not arbitrary, since as we have seen, there are good reasons to regard some rights as more fundamental than others, and thus possible violations of these rights can be viewed in proportion to their importance and the likelihood of their occurrence (Beyleveld and Brownsword 2006, p. 148). Neither, on the other hand, are the weights in question susceptible to any precise quantification (Beyleveld and Brownsword, ibid and Beyleveld and Pattinson 2010, pp. 268-269): no amount of expert knowledge can tell us definitely, for example, whether an individual will go on to cause serious harm to others or herself if she is not compulsorily detained and treated. People with certain kinds of expertise, such as psychiatrists, medical ethicists and lawyers, may be better able than most citizens to formulate cogent arguments on these questions; but the acceptance or rejection of these arguments, unlike that of some kinds of scientific or medical judgment, must depend ultimately on their rational acceptability in accordance with the moral and epistemological criteria we have presented here. Through bringing these criteria together in a framework of rights precautionism, and by illustrating its application to two fictional mental health/capacity law cases, a potentially useful heuristic emerges which can justify an ethically sound resolution to such cases, given the inevitable conditions of uncertainty in which it is applied. Clearly, this points to further work to be done beyond the groundwork we have laid here, and so we must leave the more practical question of how such a model may be implemented, and the challenges it may face, for another day.

Bibliography

Anscombe G E M, Intention, (2nd ed Cambridge: Mass: Harvard University Press 1963)

Archard D, ‘Why Moral Philosophers Are Not and Should Not Be Moral Experts.’ (2011) 25 Bioethics 119

Bartlett P and Sandland R, Mental Health Law: Policy and Practice, (3rd ed Oxford: OUP 2007)

Bersoff D N, ‘Judicial Deference to Nonlegal Decisionmakers: Imposing Simplistic Solutions on Problems of Cognitive Complexity in Mental Disability Law.’ (1992) 46 S M U L Rev 329

Beyleveld D, The Dialectical Necessity of Morality: An Analysis and Defense of Alan Gewirth’s Argument to the Principle of Generic Consistency (Chicago: The University of Chicago Press 1991)

— — ‘Legal Theory and Dialectically Contingent Justifications for the Principle of Generic Consistency.’ (1996) 9 Ratio Juris 15

— — ‘The Principle of Generic Consistency as the Supreme Principle of Human Rights.’ (2012) 13 Human Rights Review 1

— — and Brownsword R, Human Dignity in Bioethics and Biolaw. (Oxford: OUP 2001)

— — and — —, ‘Principle, Proceduralism and Precaution in a Community of Rights’. (2006) 19 Ratio Juris 141

— — and — —, Consent In The Law. (Oxford: Hart Publishing 2007)

— — and — —, ‘Complex Technology, Complex Calculations: Uses and Abuses of Precautionary Reasoning in Law’ in: P. Sollie and M. Düwell (eds.) Evaluating New Technologies. (Dordrecht: Springer 2009) 175

— — and Pattinson S, ‘Proportionality, Precaution and Duties in Relation to “PPAs-in-Part”’ (unpublished paper on file with PB 1998)

— — and — —, ‘Precautionary Reasoning as a Link to Moral Action’, in: M. Boylan (ed.), Medical Ethics. (Upper Saddle River, New Jersey: Prentice-Hall 2000) 39-53

— — and — —, ‘Defending Moral Precaution as a Solution to the Problem of Other Minds: A Reply to Holm and Coggon.’ (2010) 23 Ratio Juris 258

Bracken P and Thomas P, Postpsychiatry: Mental health in a postmodern world. (Oxford: OUP 2005)

Cavadino M, ‘A Vindication of the Rights of Psychiatric Patients’. (1997) 24 Journal of Law and Society 235

Charland L C, ‘Character: Moral Treatment and the Personality Disorders’. in. J. Radden (ed.). The Philosophy of Psychiatry: A Companion. (New York: OUP 2004) 64-77.

Chodoff P, ‘The Case for Involuntary Hospitalization of the Mentally Ill’. (1976) 133 American Journal of Psychiatry 496

Coady C A J, Testimony: A Philosophical Study. (Oxford: Clarendon Press 1992)

— — ‘Reid and the Social Operations of Mind.’ in: T. Cuneo and R. van Woudenberg (eds.) The Cambridge Companion to Thomas Reid. (Cambridge: Cambridge University Press 2004) 180-203.

Cooper R, Classifying Madness: A Philosophical Examination of the Diagnostic and Statistical Manual of Mental Disorders. (Dordrecht: Springer 2005)

Cresswell M, ‘Psychiatric “survivors” and testimonies of self-harm.’ (2005) 61 Social Science and Medicine 1668

Culver C M and Gert B, ‘The Morality of Involuntary Hospitalization’, in: S. F. Spicker, J. M. Healey Jr and H. T. Engelhardt, Jr (eds.). The Law-Medicine Relation: A Philosophical Exploration. (Dordrecht: D. Reidel 1981) 159

Darwall S, ‘Introduction’ in S. Darwall (ed.) Contractarianism/Contractualism. (Oxford: Blackwell 2003) 1

Dawson J and Szmukler G, ‘Fusion of mental health and incapacity legislation’, (2006) 188 British Journal of Psychiatry 504

Donnelly M, ‘Capacity assessment under the Mental Capacity Act 2005: Delivering on the functional approach?’ (2009) 29 Legal Studies 464

Doyal L and Sheather J, mso-bidi-font-weight: bold'>‘Mental health legislation should respect decision making capacity’, (2005) 331 British Medical Journal 1467

Driver J, ‘Autonomy and the Asymmetry Problem for Moral Expertise’, (2006) 128 Philosophical Studies 619

Dworkin R, Taking Rights Seriously (Duckworth, London 1978)

— —, ‘Is there a Right to Pornography?’ (1981) 1 OJLS 177

— —, ‘Principle, Policy Procedure.’ in: R. Dworkin (ed.) A Matter of Principle. (Cambridge, Mass.: Harvard University Press 1985) 72

Dwyer D, mso-bidi-font-weight:bold'>Themso-bidi-font-weight:bold'> Judicial Assessment of Expert Evidencemso-bidi-font-weight:bold'>. (Cambridge: Cambridge University Press 2008)mso-bidi-font-weight:bold'>

Engelhardt Jr H T and Jotterand F, ‘The Precautionary Principle: A Dialectical Reconsideration’, (2004) 29 Journal of Medicine and Philosophy 301

Estlund D M, Democratic Authority: A Philosophical Framework. (Princeton, NJ: Princeton University Press 2008)

Feinberg J, ‘Voluntary Euthanasia and the Inalienable Right to Life’, (1978) 7 Philosophy and Public Affairs 93

— —, Harm to Self (The Moral Limits of the Criminal Law: Volume 3). (New York: OUP 1986)

Feintuck M, 'Precautionary Maybe, but What's the Principle? The Precautionary Principle, the Regulation of Risk, and the Public Domain', (2005) 32 Journal of Law and Society 371

Foley R, Intellectual Trust in Oneself and Others. (Cambridge: Cambridge University Press 2001)

Fricker M, Epistemic Injustice: Power and the Ethics of Knowing. (Oxford: OUP 2007)

Fulford K W M, ‘Facts/Values: Ten Principles of Values-Based Medicine’, in J. Radden (ed.) The Philosophy of Psychiatry: A Companion. (New York: OUP 2004) 205

Gerhardt S, Why Love Matters (Hove, East Sussex: Brunner-Routledge 2004)

Gewirth A, Reason and Morality (Chicago: The University of Chicago Press 1978)

— —, ‘The Basis and Content of Human Rights’. in: A. Gewirth (ed.) Human Rights: Essays on Justification and Applications (Chicago: The University of Chicago Press 1982) 41

— —, ‘The Rationality of Reasonableness’, (1983) 57 Synthese 225

— —, ‘Human Rights and Conceptions of the Self’. (1988) 18 Philosophia 129

Gostin L O, ‘Beyond Moral Claims: A Human Rights Approach in Mental Health’. (2001) 10 Cambridge Quarterly of Healthcare Ethics 264

Grisso T and Appelbaum P S ‘Is It Unethical to Offer Predictions of Future Violence?’ in: R. B. Edwards (ed.). Ethics of Psychiatry: Insanity, Rational Autonomy and Mental Health Care. (Prometheus Books, Amherst, NY 1997, orig. pub. 1992) 446

Gutmann A and Thompson D, Why Deliberative Democracy? (Princeton: Princeton University Press 2004)

Habermas J, Toward a Rational Society: student protest, science and politics (trans. J.J. Shapiro). (London: Heinemann 1971)

— —, The Theory of Communicative Action, Volume 1: Reason and the Rationalization of Society. (Cambridge: Polity Press 1986)

— —, Moral Consciousness and Communicative Action (trans. C. Lenhardt and S. Weber Nicholsen). (Cambridge, Mass.: MIT Press 1990)

Harris J and Holm S, ‘Extending human lifespan and the precautionary paradox’, (2002) 27 Journal of Medicine and Philosophy 355

Hornsby J, ‘Agency and Causal Explanation.’ in: J. Heil and A. Miele (eds.) Mental Causation. (Oxford: Clarendon Press 1993) 161

Hyslop A, ‘Other Minds’ in E. N. Zalta (ed.) The Stanford Encyclopaedia of Philosophy (Fall 2010 Edition) <http://plato.stanford.edu/archives/fall2010/entries/other-minds/> [Accessed 26 July 2012]

Jaspers K, General Psychopathology, Volume I (trans. J. Hoenig and M. W. Hamilton). (Baltimore, MD: The John Hopkins University Press 1997 [1913])

— —,General Psychopathology, Volume II (trans. J. Hoenig and M. W. Hamilton). (Baltimore, MD: The John Hopkins University Press 1997 [1913])

La Torre M, ‘On Two Distinct and Opposing Versions of Natural Law: “Exclusive” versus “Inclusive”’, (2006) 19 Ratio Juris 197

Lehrer K, ‘Testimony and Trustworthiness’ in J. Lackey and E. Sosa (eds.) The Epistemology of Testimony. (Oxford: OUP 2006) 145

Levvis G W, ‘The Principle of Relevant Similarity’, (1991) 25 The Journal of Value Inquiry 81

Liao S M, ‘The Right of Children to Be Loved’, (2006) 14 Journal of Political Philosophy 420

Longino H E, The Fate of Knowledge. (Princeton, NJ: Princeton University Press 2002)

Manson N C, and O’Neill O, Rethinking Informed Consent in Bioethics. (Cambridge: Cambridge University Press 2007)

Matthews E, ‘Autonomy and the Psychiatric Patient’, (2000) 17 Journal of Applied Philosophy 59

McLean S A M, Autonomy, Consent and the Law. (London: Routledge-Cavendish 2010)

Monahan J and Wexler D B, ‘A Definite Maybe: Proof and Probability in Civil Commitment’. (1978) 2 Law and Human Behavior 37

Morse S J, 'Crazy Behavior, Morals, and Science: An Analysis of Mental Health Law’. (1978) 51 Southern California Law Review 527

Oberdiek J, ‘Lost In Moral Space: On the Infringing/Violating Distinction and Its Place in the Theory of Rights’. (2004) 23 Law and Philosophy 325

Owen A M, and others, ‘Detecting Awareness in the Vegetative State’, (2006) 313 Science 1402

— —, and Coleman M R, ‘Functional neuroimaging of the vegetative state’, (2008) 9 Nature Reviews Neuroscience 235

Peay J, Decisions and Dilemmas: Working with Mental Health Law. (Oxford: Hart Publishing 2003)

Polyani M, Personal Knowledge: Towards a Post-Critical Philosophy. (London: Routledge and Kegan Paul 1983)

Rawls J, Political Liberalism. (New York: Columbia University Press 1993)

— —,A Theory of Justice, revised edition. (Oxford: OUP 1999)

Raz J, The Morality of Freedom. (Oxford: Clarendon Press 1986)

— —,Between Authority and Interpretation: On the Theory of Law and Practical Reason. (Oxford: OUP 2009)

Regan D, ‘Gewirth on Necessary Goods: What is the Agent Committed to Valuing?’ in M. Boylan (ed.) Gewirth: Critical Essays on Action, Rationality and Community. (Lanham, Maryland: Rowman and Littlefield Publishers 1999) 45

Richardson G, ‘Balancing Autonomy and Risk: A failure of nerve in England and Wales?’ (2007) 30 International Journal of Law and Psychiatry 71

Sadler J Zmso-spacerun:yes'>  Values and Psychiatric Diagnosis (Oxford: Oxford University Press).

Scanlon T M, ‘Contractualism and Utilitarianism’, in: A. Sen and B. Williams (eds.) Utilitarianism and Beyond (Cambridge: Cambridge University Press 2005) 103

— —, What We Owe to Each Other. (Cambridge, Mass: Harvard University Press 1998)

Shapiro B J, ‘“To a Moral Certainty”: Theories of Knowledge and Anglo-American Juries 1600-1850.’, (1986) 38 Hastings Law Journal 153

Slobogin C, ‘Doubts about Daubert: Psychiatric Anecdata as a Case Study’. (2000) 57 Washington and Lee Law Review 919

Szasz T S, ‘Involuntary Civil Commitment: A Crime Against Humanity’. in: R. B. Edwards (ed.). Ethics of Psychiatry: Insanity, Rational Autonomy and Mental Health Care (Amherst, NY: Prometheus Books 1997, orig. pub. 1970) 437

Talbott W J, Which Rights Should Be Universal? (New York: OUP 2005)

Thomson J J, ‘Some Ruminations on Rights’, in: W. Parent (ed.), Rights, Restitution, and Risk: Essays In Moral Theory. (Cambridge, Mass: Harvard University Press 1986) 49

UK Department of Health Report of the Expert Committee: Review of the Mental Health Act 1983. UK Department of Health 1999 Available at: <http://www.dh.gov.uk/en/Publicationsandstatistics/Publications/PublicationsPolicyAndGuidance/DH_4009576> [Accessed 26 July 2012]

UK Department of Health Code of Practice Mental Health Act 1983. (London: TSO 2008) Available at:

<http://www.dh.gov.uk/prod_consum_dh/groups/dh_digitalassets/@dh/@en/documents/digitalasset/dh_087073.pdf> [Accessed 26 July 2012]

Varelius J, ‘Is Ethical Expertise Possible?’ (2008) 11 Medicine, Health Care and Philosophy 127

Weinstein B, ‘The Possibility of Ethical Expertise’, (1994) 15 Theoretical Medicine 61

Weisstub D N, ‘The Theoretical Relationship Between Law and Psychiatry’, in: D N Weisstub (ed.) Law and Psychiatry: Proceedings of an International Symposium Held at the Clarke Institute of Psychiatry, Toronto, Canada, February 1977 (New York: Pergamon Press 1978) 30

---

(1)Our use of the distinction between infringing and violating a right is taken from Feinberg 1978 and Thomson 1986. For an interesting critical discussion of this distinction, see Oberdiek 2004.

(2) For reasons we will go on to discuss, we take those principles to include most of the rights on standard lists of human rights, including civil, economic and social rights as well as basic liberties. In terms of international law, would take us beyond the rights enumerated in the Universal Declaration of Human Rights (1948) to those given in the International Covenant on Civil and Political Rights (1966) and in the International Covenant on Economic, Social and Cultural Rights (1966). Of course, this only gives an idea of the scope of the rights we would endorse – the ways in which we think they ought to be implemented may not be clear on the face of these instruments, which, like the existence of the rights themselves, would require independent moral justification for their force.

(3)To couch the need for freedom and well-being in terms any less forceful than a right would be to diminish the necessity of claiming freedom and well-being to any purpose we may wish to pursue (Beyleveld 1991 pp. 97-98). Agents are “logically required to be opposed to such interference” with their rights (ibid., p 97), a claim which is supported by “the argument from attitudinal consistency” (ibid., pp. 95-101).

(4) For Scanlon’s view, see Scanlon 1982, p 112: “it is the reasonableness of rejecting a principle, rather than the reasonableness of accepting it, on which moral argument turns.”

(5) For instance, a reasonable person who has enough wealth to remain financially self-sufficient for the rest of her life and who does not wish to have children would recognise the moral force of contributing to a progressive system of wealth taxation (e.g. on her income or the value of her home) at a high level to fund expansive social welfare provision and would accept that a share of these funds should be channelled into activities such as family support schemes (e.g. ‘Sure Start’ programmes in the UK) and infertility treatments.

(6) Gewirth and his followers have been criticised for adopting an unduly ‘monological’ approach to moral and political reasoning (see, for example, La Torre 2006). In our view, adopting a more ‘dialogical’ approach, in which we consider not what an agent would accept in solitary reflection but what she could reasonably accept in seeking to reach agreement with others, does not undermine the PGC but rather provides a different route to the same conclusion. The view that citizens engaged in rational discussion must accept a framework of rights is one that we share with Habermas – indeed, Habermas hints at such an argument (Habermas 1990, p 101).

(7) We refer readers interested in an analysis of the dialectical necessity of the PGC to Beyleveld 1991 and for a discussion of the dialectally contingent approach to Beyleveld 1996.

(8) In any case, from the perspective of English law at least, the European Convention of Human Rights and the Human Rights Act 1998 commit the English courts to a set of rights not dissimilar to those derived from the PGC. What is important is that such rights are interpreted as moral rights derived from the needs of agency, and not as inconvenient legal barriers to politically-calculated decision-making.

(9)Founding the acceptability requirement along these lines also allays some of the concerns Driver raises about “borrowing moral knowledge” (2006 p 639) that could arise in relationships between two or more non-experts in ethics (e.g. a bigoted newspaper columnist and his uncritically devoted readership) as well as between an ethics expert and a non-expert.

(10)Where we might depart from Weinstein is in his relativism towards the epistemology of the ‘good life’ (Weinstein 1994, p 71) – whilst we too would advocate a pluralist approach at the level of non-moral and morally permissible choices, the extent of this pluralism would be determined fundamentally by an egalitarian, liberal rights-based moral epistemology like the PGC.

(11) In the words of Beyleveld and Pattinson (2010, p 270): “the list of generic rights holders includes every vulnerable being with the capacity to act (to do something voluntarily for a purpose that it treats as the reason for its behaviour)” but it is not the case that we thereby must “treat logically possible agents as agents” (ibid.), such as “rocks” (ibid., p. 263) – we can only treat as agents those beings and entitles who ostensibly behave as agents (ibid., p. 261).

(12) As Beyleveld argues (2012, pp. 9-11), only actual agents can have rights under the ‘will’ conception, i.e. rights that they can choose either to waive or to enforce; but beings that appear as if they may be agents (i.e. those whose evidence for agency is limited or has apparently extinguished) can have interests that others are under a duty to protect, and thus they have rights under the ‘interest’ conception (i.e. (“interest rights” as they uphold the important interests such beings or entities would have if they are or remain agents). (ibid.)

(13) There is increasing neuroscientific evidence to support the claim that love makes a significant difference to a child’s ability to engage in healthy social interaction in later life, which is clearly and powerfully presented in Gerhardt 2004. For the case as to why children have a human rights claim to be loved, see Liao 2006.

(14)Especially in light of recent neuroscientific evidence that challenges assumptions about lack of awareness and intention in pvs and the minimally conscious state, chiefly Owen et al. 2006 and Owen and Coleman 2008. In the former study, which sought to ascertain an unconscious woman’s response to being invited to undertake imagining activities such as hitting a tennis ball after at least five months of being in a vegetative state, it was concluded “her decision to cooperate with the authors by imagining particular tasks when asked to do so represents a clear act of intention, which confirmed beyond any doubt that she was consciously aware of herself and her surroundings” (Owen et al. 2006, p. 1402).

(15)It is important to stress that in endorsing this claim we are not espousing a position in consequentialist ethics, where the salience or even the mere existence of a right would depend upon its ability to secure a non-moral end (e.g. welfare, happiness, economic efficiency etc., see, e.g. Talbott, 2005). In our view, the moral worth of the rights to freedom and well-being does not depend on the perceived value of the goal to be pursued. Rather, where rights conflict, we agree with Beyleveld and Brownsword that these rights need to be weighed against one another in terms of their relative importance for the possibility of successful action qualified by the greater or lesser probability of their being wrongly denied in contexts where prima facie violation is inevitable. See Beyleveld and Brownsword (2006 p. 148).

(16)In family cases the pressures are more evenly balanced: hence the saying that social workers are ‘damned if they do and damned if they don’t’.

(17) There are other possible examples, different from these, in which an expert might be called upon to provide evidence to support the decision of a legal decision-maker, such as where a patient’s best interests need to be determined in end of life cases, e.g. whether life-sustaining treatment should be withdrawn and withheld for an adult in a minimally conscious state (W (by her litigation friend, B) v M (by her litigation friend, the Official Solicitor) and others [2011] EWHC 2443 (Fam)) and provision of coercive feeding for an adult with anorexia nervosa (A Local Authority v E (by her litigation friend, the Official Solicitor) and others [2012] EWHC 1639 (CoP)). For reasons of space, these cannot be addressed here, suffice to say that we do not envisage any difficulties in applying the principle of rights precautionism outlined to cases such as these.

(18) Which could engage Articles 5, 11 and 3 of the European Convention on Human Rights respectively, subject to the provision in Article 5(e), which allows “the lawful detention  . . . of persons of unsound mind”. The seminal case on the application of this provision is Winterwerp v The Netherlands (1979-80) 2 EHRR 387.

(19) This is broadly in line with the position contained in the recommendations proposed by the Expert Committee established to consider reform of the Mental Health Act 1983 in England and Wales (the ‘Richardson Committee’), which were not accepted by the British Government following publication of its report in 1999. See UK Department of Health (1999).

(20) Let us assume that the legislation allowing for the compulsory detention and treatment of persons with mental disorder was made in accordance with principles of human rights (and that in certain circumstances ECT may be clinically justified, Emma’s being one such case) which we think is a viable assumption provided the legislation is founded on patient-centred grounds and subject to rigorous procedural safeguards. For a discussion of how human rights principles can be used to shape mental health law, see Gostin 2001.

(21)The wording of the decisional competence requirement in relation to ECT under s.58A(3)(c) reflects the 'understanding limb' of the pre-MCA common law test for mental capacity set out in Re C (adult: refusal of medical; treatment) [1994] 1 All ER 819, although para. 23.28 of the 2008 revision to the Mental Health Act 1983 Code of Practice states, "[f]or people aged 16 or over, capacity to consent is defined by the MCA" (UK Department of Health, 2008, p. 186). Later, in a footnote to para. 24.7, it is stated explicitly "in sections 58 and 58A, the Act refers to the patient being "capable of understanding the nature, purpose and likely effects" of the treatment. However, for all practical purposes this can be understood to mean the same as the test of whether the patient has the capacity to consent (or, if under 16, the competence to do so)" (ibid. p 201).font-size:10.0pt;

(22) The assessment of ‘dangerousness’ as a suitable aim for psychiatric prediction is questioned by Grisso and Appelbaum (1997, p. 457, n. 3). This would appear to be connected with anxiety over psychiatry involving itself in making value judgments rather than confining itself to ‘scientific’ descriptions of probable behaviour. We expect that their anxieties would also extend to the often inter-changeable use of the term ‘harm’. We do not see a problem with the prediction of either ‘dangerousness’ or ‘harm’, provided it is judged in accordance with a theory of human rights capable of rational acceptance as set out in this article. This then makes a definition of ‘dangerousness’ or ‘harm’ less contentious: in our view, it can be understood as unjustifiable detrimental consequences for interests other agents have in freedom and well-being. In this sense, our understanding of harm is close to that of Feinberg (1986), who understands it as “wrongly to set back interests” [emphasis in original] (ibid., p. 11).

(23) Bersoff also shares this view, questioning the justifiability of deferring to psychiatric experts in legal decision-making: “The psychiatric and psychological components of decisions affecting the rights of mentally disabled persons are not so proportionately great nor is the ability of mental health professionals to arrive at accurate judgments so demonstrable, that their expertise should dominate all other considerations” (Bersoff 1992, p 371). Yet, as Donnelly highlights, under English law, there is evidence of deference by the courts towards professionals (such as psychiatrists and psychologists) who assess capacity, including giving only brief reasons for endorsing the medical professional’s judgment of incapacity (Donnelly 2009, pp. 469-470) and where competing evidence arises, often expressing a preference for that provided by “independent experts” above those familiar with the patient in question (ibid. p 470).

(24) See the 2008 revision to the Mental Health Act 1983 Code of Practice (UK Department of Health 2008, para. 24.58), which states that the second opinion appointed doctor (SOAD) should, inter alia, “give due weight to the patient’s views, including any objection to the proposed treatment and any preference for an alternative” (ibid., p 212).

(25) On technocratic and decisionist models of the relations between politics and scientific expertise, see Jürgen Habermas, Toward a Rational Society: student protest, science and politics (trans. J.J. Shapiro, London: Heinemann, 1971) pp. 61-66.