Abstract
It is widely agreed and well understood that the uniqueness of the mental health care setting gives rise to its own distinctive versions of ethical and policy dilemmas over informed consent, treatment refusal, harm to self, involuntary treatment and competency. Less well understood is the way the uniqueness of the mental health care setting and of psychiatric practice calls for a distinctive professional ethics which goes above and beyond that provided by the principles of bioethics applicable to all fields of medicine [1, 2]. Psychiatric practice is and must also be governed by those principles. But psychiatric practice differs from other forms of medicine, and an additional set of ethical demands derives from its uniqueness as a social and medical practice.
Some of the elements distinctive to psychiatry have been noted in the literature, others have not. These elements can be sorted into three broad categories: those pertaining to (i) the relationship or therapeutic ‘alliance’ forged between treater and patient; (ii) certain characteristics of the psychiatric patient; and (iii) the goals and purposes of the therapeutic enterprise itself. Taken individually none of these elements is unique to psychiatric practice; each is present, to greater and lesser degrees, in other types of medicine. But the presence of all three elements together in psychiatry serves to make psychiatry very different from any other branch of medicine in terms of the ethical demands it places on practice.
Psychiatric ethics here concerns itself with the values, ideals and distinctive prescriptions instilled through professional education and other informal kinds of professional acculturation. An ethics for psychiatric refers to the set of those values, ideals and prescriptions. The term ‘ethics’ is also used of the quasi-legal and legal codes regulating professional conduct and governing matters of legal liability, but such codes will not be the primary focus of this discussion.
Only preliminary groundwork is undertaken here. By identifying features distinguishing psychiatry as medical and social practice it will be possible to determine what kind of ethical framework is required of an ethics for psychiatry and some of the emphasis and focus demanded of that ethics. This analysis aims to illustrate why the uniqueness of psychiatry calls for a unique ethics; the larger task of formulating the substance of that unique ethics is beyond the scope of this paper.
After introducing the relevant literature, this paper proceeds in two stages. First, the elements differentiating psychiatry (i–iii listed earlier) are examined. These elements are shown to impose requirements on both the structure and the content of an ethics for psychiatry. Three of these requirements are introduced and discussed: a focus on character, a rubric for acknowledging boundary violations, and an emphasis on gender.
Earlier discussions
The American Psychiatric Association's Principles of medical ethics: with annotations especially applicable to psychiatry [3] is an important starting point for these ideas. The presumption is that psychiatrists and other physicians share goals, this document notes, but special ethical problems arise in psychiatric practice which differ in ‘colouring and degree’ from the ethical problems shared by all branches of medicine (p.1). Some features of psychiatric practice which introduce distinctive and additional ethical constraints on its practitioners are alluded to in the APA document, although only briefly and casually. Four of these features are particularly important. First, the preservation of optimal conditions for development of sound doctor-patient relationships should take precedence over all considerations because the relationship is ‘such a vital factor in effective treatment’ in psychiatry (p.8). Second, extra ethical vigilance over the relationship by the psychiatrist is necessitated by ‘the essentially private, highly personal, and sometimes intensely emotional’ nature of the relationship established with the psychiatrist (p.3). Third, an especially high standard of confidentiality is required of the psychiatrist because of the ‘sensitive and private’ nature of the information with which the psychiatrist deals (p.6). Fourth, the intensity of the relationship may tend to activate sexual and other needs and fantasies on the part of both patient and psychiatrist ‘while weakening the objectivity necessary for control’ (p.4).
Although these four features partially explain why psychiatric ethics must go above and beyond the ethical obligations incumbent on all doctors, these brief remarks understate the distinctness of an ethics for psychiatry and the importance of that distinctness. Further explanation is provided by Fulford and Hope, noting psychiatry's status as a bioethical ‘ugly duckling’ [4]. Relative to values about physical illness, values about mental illness are unsettled and unresolved, these authors emphasize. There are deep disagreements over what sort of thing mental illness is and how to understand it. Because of this, psychiatric practice cannot be subsumed into the standard model of bioethics, a model which often makes the presumption (rightly or wrongly) that a value-free medicine is possible. More recently Dickenson and Fulford have developed related ideas by showing that compared with cases in the rest of medicine, cases from psychiatric practice routinely implicate deep philosophical questions of metaphysics, epistemology and the philosophy of science as well as ethics in ways that inextricably tie psychiatry to philosophy [5].
The following exploration into the way in which psychiatry differs from other medical practice develops on both the remarks of the APA and on Fulford's, Hope's and Dickenson's work.
Psychiatry's uniqueness as social and medical practice
The several features differentiating psychiatric practice can be usefully categorized into three kinds, those ascribable to the therapeutic relationship, those to the psychiatric patient and those to the psychiatric or psychotherapeutic project. Each category requires analysis.
Features of the therapeutic relationship
These are fairly thoroughly canvassed in the APA document [3]. The therapeutic relationship is private, personal and raises intense emotions. Perhaps most important, it is widely believed that the therapeutic relationship is a key ingredient in therapeutic effectiveness. Varying with the type of therapy undertaken, and perhaps most obviously true with psychodynamic psychotherapy, the relationship is a treatment tool, analogous to the surgeon's scalpel [6–8].
Features characterizing the psychiatric patient
Some vulnerability to exploitation, dependence, and inherent inequality characterize not just the psychiatric patient but any client in a professional relationship [9]. But vulnerability to exploitation, dependence and inherent inequality are markedly present in psychiatric patients. The special vulnerability of this patient group derives from several sources, including these patients' diminished judgement; the stigma and controversy associated with their condition, and the salient place of gender in psychiatry.
1. Diminished judgement: the model of the patient as an autonomous agent capable of giving or withholding informed consent is not, or at least not consistently, applicable with the psychiatric patient. If only temporarily and partially, psychiatric patients are deprived of their best defenses against exploitation – their reasoning ability, their judgement in matters concerning their longterm self interest, their self-control, and their capacity to make their needs and concerns known to others. (Nonpsychiatric conditions, including the organic dementias and injuries to the brain also deprive their sufferers of these capabilities, of course.)
2. Stigma and controversy: psychiatric patients are also particularly vulnerable to exploitation because they are the subjects of stigmatizing attitudes towards their condition, and because – this is the point made by Fulford and Hope – their condition itself is so conceptually controversial. There is fundamental disagreement over the nature and ontological status of mental disorder, and over moral and social attitudes towards it [10–15]. For example, it remains controversial whether and to what degree the lifting of responsibility associated with the sick role is applicable to those with mental disorder.
3. The link with gender: the patient's gender and gender identity are inescapable aspects of psychiatric practice. Historically mental disorder and psychiatric concepts have been associated with gender, and psychiatric categories, diagnosis, treatment and research have reflected these associations. In addition, many disorders remain strongly gender linked, and much therapy utilizes cultural categories such as gender (it attempts to effect a mentally healthy man or woman, for example, rather than a mentally healthy ‘person’).
Features of the therapeutic project
At its most modest, psychiatric treatment seeks to restore some earlier level of functioning and to relieve debilitating signs and symptoms. At its most ambitious, however, the therapeutic project is a deeply significant one – that of re-forming the patient's whole self or character, when these terms are understood in holistic terms as the set of a person's long-term dispositions, capabilities and social and relational attributes. In respect to moral seriousness, the therapeutic project is matched by few other projects undertaken in our culture except the raising of children, adding immeasurably to the patient's vulnerability and to the responsibility imposed on the treater [6–8].
The above arrangement into features pertaining to therapeutic relationship, psychiatric patient and therapeutic project must be recognized to be fairly artificial. The element of gender and ‘gendering’ affects and is affected by the therapeutic relationship, for example; and the significant and far-reaching nature of the therapeutic project affects the patient in therapy rendered vulnerable by undertaking that project.
These and other features distinguishing psychiatry suggest at least some of what an ethics for psychiatry might need to include. The discussion which follows is focused around three illustrations of the way in which the features which combine to account for the uniqueness of psychiatric practice dictate the type of ethics required for psychiatric practice.
Illustrations
Psychiatric ethics as a character-based ethics
Psychiatric practice approaches the patient as a character. (Character sometimes refers to attributes relevant to the moral life only, but ‘character’ here functions as an equivalent of ‘self’ in referring to a person's longterm dispositions, capabilities and social and relational attributes.) The notion of character presupposes a holistic conception of the person, in the sense that a person is understood as an embodied psychosocial whole embedded in a social and relational nexus. Moreover, the notion of character presupposes the person is an identity across stretches of time; it concerns a person's more enduring states and dispositions rather than particular actions.
The psychiatric patient typically seeks to change more permanent and pervasive states of being, traits, dispositions and habits of mind rather than seeking advice on particular courses of action. Thus it might be said that the project of psychiatry is a normative one which involves re-forming by improving the patient's character. Even when adherence to a strictly neurochemical model reduces psychiatry to little more than the administration of psychoactive drugs therapeutic success is usually portrayed and measured in holistic terms: the patient as a whole feels better or behaves in closer conformity to mental health norms [6, 16]. (Some cognitive behavioural therapy which identifies goals and outcomes in limited, behaviourally defined terms may be an exception to these claims about the emphasis on character [17]).
As well as focusing on the character of the client, psychotherapeutic practice involves the character of the treater [and cognitive behavioural therapy is no exception in this respect]. The treater's character enters into the psychotherapeutic relationship instrumentally, the APA annotations emphasize: the relationship between doctor and patient is so ‘vital’ a factor in effective treatment of the patient that preservation of optimal conditions for development of a sound working relationship between doctor and his/her patient ‘should take precedence over all other considerations.’ (3, p.8) Because therapeutic effectiveness is widely seen to depend on it the treater's having a self or character of a certain kind is usually understood, as in that passage, as a sine qua non of the therapeutic engagement.
That a virtuous character in the treater is requisite for psychotherapeutic practice has been implied in several recent discussions of ethics for psychiatry [2, 5, 18–20]. This suggestion seems to encompass both the tenet that more effective psychotherapeutic practice results from a virtuous practitioner and that more ethical practice requires a virtuous treater regardless of treatment outcome. Whether valued instrumentally for their alleged role in effective therapy or valued intrinsically because they are called for by the particular setting of mental health care, reference to certain virtues as ‘mental health virtues’ indicates that there is a special place for character in an ethics for psychiatry.
To the extent that we adopt a consumer or a negotiated contract model of the relationship, the force of this good character requirement may be expected to diminish [21, 22]. Within psychiatry consumer or negotiated contract models have been slow thus far to unseat more traditional models of the relationship in healing practice however, perhaps with good reason. and while it may strike the reader that the exigencies of managed psychiatric care will soon render the model of practice outlined here obsolete, it should be remembered that this work sketches guidelines for ethical practice. If the model of practice here presupposed does not prevail, perhaps managed care practice cannot be ethical practice. [In fact, features distinguishing the patient in mental health care such as those noted above call into doubt the application of consumer or negotiated contract models to psychiatric practice [23].
Summing up then, the fact that psychiatry is by its nature holistic in its approach, that the goals of psychotherapeutic practice frequently involve the character of the client, and, in particular, that the treater's character is widely believed to play this central part in psychotherapeutic relationships, all invite us to speculate that the appropriate ethical framework for psychotherapeutic practice is some version of a character ethics.
Different ethical frameworks distinguished by philosophers include ‘virtue ethics’, Utilitarianism and Kantianism. Because virtue ethics is portrayed as focused on character and more permanent states of being rather than action and particular doings [24], we might expect virtue ethics to be particularly fitting when we formulate an ethics for psychiatry. Recent versions of both Utilitarianism and Kantianism also attempt to accommodate emphasis on character however, arguing, respectively, that the nature and structure of a person's dispositions and character must be such as to maximize utility [25], or that something like Kant's categorical imperative is a normative disposition in the character of a good agent [26]. Some recent analyses have argued that a virtue ethics best captures other aspects of the particular role-generated responsibilities and sensitivities required of professionals such as doctors [27, 28]. But for our purposes here any ethical framework placing emphasis on character may be sufficient to accommodate the centrality of character to psychiatric practice.
There are a number of ways in which the patient's attribution of good character to the treater can be expected to affect treatment outcomes. One of these is suggested by the APA annotations [3], which note that the patient tends to ‘model’ his or her behaviour after the treater ‘by identification’ (p.2). But the nature and significance of this influence are each greater than the notions of modelling and identification convey. Again, depending on the techniques employed and other features of the context such as the constraints imposed by managed care, psychiatric practice involves a dialectical process, an exploration on the part of patient and treater into experience, values and existence. In this process, the treater's role is that of guide, teacher and mentor as much as model. The project undertaken in psychotherapy is also of more significance than the APA's discussion indicates. It is a project which while too often incompletely realized, particularly with the severely mentally ill, has as its end self or character change or re-formation. It is a normative project: the patient wishes and often strives very hard to become a new and better person, more stable, more happy, more capable, more self-determining, more subjectively unified, more insightful, or just more normal (and, often, more of all of the above). In this potentially momentous normative project undertaken on the patient's behalf the treater is model, teacher and guide.
Turning from the treater's character as it is used in the therapeutic relationship to the traits of character in the treater called for by other aspects of the mental health care setting, we find a long tradition emphasizing the virtuous character required of the doctor. The virtues called for by the practice of medicine include the socalled ‘health care virtues’: trustworthiness, honesty, kindness, humility and patience [29]; compassion, phronesis, justice, fortitude, temperance, integrity and self effacement [9]. Recent discussions have identified virtues of particular importance to the mental health care setting [30]. As well as cultivating the virtues required in any medical setting, it has been claimed, psychiatrists should cultivate extra virtues including compassion, humility and fidelity [19]; trustworthiness and respect for confidentiality [18]; fidelity, veracity, and prudence [31]; warmth and sensitivity [2]; humility and perseverance [5]. (These proposed candidates for mental health virtues await analysis and critical evaluation (a task not undertaken here). On closer scrutiny they may not all prove to be virtues (warmth, for instance, may be no more than a sign of a virtue, perhaps Aristotle's philia); or they may not be efficacious in, or called for by, the mental health care setting.)
Several features of the mental health care setting were noted earlier: the special vulnerabilities of mental patients, the central part attributed to the relationship in the healing process, and the potentially far-reaching normative project undertaken in therapy. Recognized not for their instrumental but for their intrinsic value, and because they are called for by such features of the mental health care setting, it is fairly apparent why the psychiatrist should cultivate some of the virtues named above. A patient group especially vulnerable to exploitation will require extra sensitivity, trustworthiness and fidelity on the part of the treater, for example; a healing practice in which such a central part is attributed to the relationship will likely call for extra sensitivity, humility and perseverence. (The psychiatrist is also a medical practitioner and, most generally construed, a member of the professions. As medical practitioners and as professionals, it will be incumbent on psychiatrists to cultivate the virtues called for by these broader professional roles as well, some named above, such as compassion and trustworthiness, others perhaps different.)
Certain moral imperatives derive from the particular goods of each of the professions. Medicine most generally understood demands the promotion of health, for example [9, 21, 27]. For the treater in psychiatry, this includes engaging in therapy which is effective in its promotion of mental health. Thus, assuming the importance of the treater's character in therapeutic effectiveness, the cultivation of these instrumental virtues will be no less morally incumbent on the treater than is the cultivation of these and other virtues for their intrinsic value. For the effective therapy it is the treater's professional obligation to foster, the treater must strive to acquire and cultivate virtues effective in therapeutic practice.
Psychiatric ethics as boundary violation discourse
The special function of the therapeutic relationship in psychiatric practice is the feature most emphasized by the APA document [3], which notes both aspects of the relationship itself (essentially private, sensitive, highly personal, intensely emotional), and the relationship's causal role in effective therapy. These features suggest psychiatry is a practice requiring a heightened ethical attention to the niceties of the interpersonal relationship, and such ethical attention can be identified in the conceptually confused, but potentially valuable, discourse which makes reference to boundary violations.
The therapeutic rationale for boundary restrictions has been explained: boundaries ‘allow the therapist to interact with warmth, empathy, and spontaneity within certain conditions that create a climate of safety… the external boundaries of the treatment are established so that the psychological boundaries between patient and therapist can be crossed through a number of means that are common to psychotherapeutic experience…’ including ‘identification, empathy, projection, introjection, and projective identification’[32], p.143].
An ethics for psychiatry can usefully employ boundary violation language; however, only when some of the fundamentals of that language are clarified. Several conceptual confusions attach to boundary violation discourse. First, there are ambiguities and unclarities in its fundamental metaphor. In some alleged boundary violations the boundaries appear to be literal bodily boundaries, in others that which is bounded is the psyche, person or self of the patient, or her secrets; in yet other behaviour described as boundary violating it seems clearest to see the dyadic relationship as bounded. Only with uniform interpretation of what is bounded will coherence be achieved [33]. Second, there are difficulties formulating or defining criteria for boundary violating behaviour, which (allegedly) includes everything from physical contact between treater and patient, to forms of self disclosure on the part of the treater over personal matters, to breaches of confidentiality by the treater, to conflicts of interest, and ‘role reversal’ allowing the patient to provide support or other gratification for the therapist, to improprieties or irregularities arising out of fee-setting, gift-giving, appointment times and places, and such things. and finally, there are concerns over the explanatory status of claims about boundary violations, which are frequently, and mistakenly, introduced by way of explanation, rather than evaluation. (In actuality, the force of the judgement ‘X is a boundary violation’ depends on implicit assumptions, observational, theoretical and ethical [23]. Rather than ‘Action X is wrong because it is a boundary violation’ we need to recognize that because action X is judged wrong it is judged to be a boundary violation.)
Some writing reserves ‘boundary crossing’ as a nonevaluative description to emphasize that not all boundary crossings are egregious boundary violations [34]. This seems to be a helpful convention which should be adopted more broadly. The difficulty of defining which boundary crossings are unacceptable boundary violations can be further explained, if not resolved. It rests in part on the vague and metaphorical nature of the language of boundaries and violations, but also on the critical importance of context in judgements about boundary violations [34]. The theoretical and perhaps moral orientation of the person making the judgement accounts for part of this context. But in addition, contextual features of the therapy itself determine the judgement: the nature, including the history, of the relationship between treater and patient, the particular boundary crossing's social and cultural meaning, and so on. While this recognition of the part played by context does not yield a satisfactory definition of which boundary crossings are boundary violations, it allows us to see why such a definition eludes us. Moreover, recognition of the part played by context also allows some progress in understanding, if not resolving, the frequent disputes which arise over alleged boundary violations. Identifying the underlying context and contextual assumptions which are the source of difference is the first step to reaching agreement over the ground rules for decision making in this difficult arena.
Despite the above clarification of the underlying metaphor, a definition of behaviour judged boundary violating, and a means of resolving disagreements over alleged violations are still required. Only when those are completed will the full usefulness of this discourse for psychiatric ethics be realizable.
Psychiatric ethics and gender
The APA annotations [3] make reference to sex in the psychiatrist-patient relationship but are surprisingly silent on the question of gender, and sexism. This is a notable omission, for gender is inescapably tied to psychiatric practice through epidemiology, associations, theories of psychosexual development, and as part of a systemic patriarchal culture.
1. Epidemiology: from their beginnings psychiatric diagnosis, practice and research, at least in the West, have been influenced by widespread associations and attitudes which saw women as particularly prone to mental disorder [35–41]. Even in our day women make up a great, and perhaps (the data is ambiguous) disproportionate number of those seeking help from psychiatry and subject to psychiatric diagnosis, and recent research reveals gender links – some in incidence, others in age of onset, course, and response to therapy, in several mental disorders [31, 42].
2. Associations: in our Western traditions, the central categories within which mental disorder was understood – rationality and the reasoning capabilities, the passions, the mind, cognition, the imagination, feelings, beliefs, moods, emotions, the will, the self and selfcontrol – were all for hundreds of years deeply ‘gendered,’ i.e. strongly associated with one sex or the other. The feminine, like madness itself, was associated with irrationality, lack of control, unbridled passion, immaturity, and so on, linking the feminine with madness by powerful strands of cultural influence [37], as the once common diagnosis of hysteria reminds us. Other categories as well, such as race, class and ethnicity, are also implicated in the normative project in psychotherapy and must be acknowledged by the therapist. But the legacy of this long and multi-stranded cultural association in which madness and its seat in the mental faculties were all gendered categories, leaves gender unavoidable for a discussion of mental disorder in the way that no other category can be.
3. Normative Theories of Health: psychological theorizing has typically endorsed stereotypical assumptions about sex roles and supported differently valued criteria of mental health for men and women. Traits such as dependency and passivity were regarded as normal qualities for women, while the more desirable traits of assertiveness and independence were judged normal for men [43]. Gender bias has been identified in categories such as Premenstrual Dysphoric Disorder, Dependent Personality Disorder, Histrionic Personality Disorder [44, 45]. Some theories also designated women's dissatisfaction with her traditional role as a form of psychopathology [46].
Progressive psychiatric thinking has sometimes attempted to replace gender-specific mental health norms with an androgenous ideal of the mentally healthy, mature (genderless) person [47]. When women's capabilities are appropriately acknowledged, women also can be seen to be strong, competent, logical and selfcontrolled, for example [48, 49]. However, recent research suggests that gender should be seen not as an attribute of one's individual personality and self identity, but as an emergent property of social groups, the product of biological predispositions which vary across the sexes (for instance, differences in maturation rates for different kinds of behaviour), which have in turn stimulated patterns of differentiated socialization [50]. If men and women are best seen to be members of distinct subcultures then the challenge may not be to forge androgenous ideals so much as to revise or reconstruct sex-specific mental health norms. Some norms can be cleansed of their older, misogynistic associations. (This particular reconstructive task has been undertaken by socalled cultural feminists who want to revive appreciation of long-denigrated feminine traits such as emotionality [51, 52].) In a society such as ours, where gender differentiation structures so many activities and ways of viewing the world these recent findings also suggest that getting help to feel, be, or behave better, must be getting help to feel, be, or behave a better woman or man.
4. Patriarchy: the therapeutic relationship in which male therapists help female patients, still common today despite the increased number of women training in psychiatry and the availability of non-medical feminist psychotherapies, replicates within the therapeutic relationship the power arrangement in which women usually find themselves in the patriarchal structure of the broader society [53–56]. Because of this broader context gender has a profound effect on how power is experienced (thus, receiving care from a male practitioner is for a woman a different experience from that experienced by a male patient with the same therapist, for example). Raised in the therapeutic context, therefore, gender will enter two ways: discussion of the patient's gender identity – how she(he) situates her(him)self within those of her(his) own traits that she(he) regards as feminine(masculine) – but also discussion of the gendered social structuring within which she(he) lives her(his) life.
The pervasive presence of gender within psychiatric theory and practice suggests a range of additional ethical strictures incumbent on the psychiatric professional, as this discussion indicates. An ethics of psychiatry may be required to address the several respects in which gender affects, and women are affected by, psychiatric diagnosis, for example. It may be required to adopt a theory of gender, and to identify and resolve the above noted tension between androgenous and gendered ideals. It may have to answer the challenge that psychiatry upholds patriarchal power arrangements, and supports an undesirable individualism particularly unfitting to women's socialization and gender identity. Because psychiatry works so directly with self identity, it may need to acknowledge certain value and metaphysical assumptions about the nature of the self, explicitly defending or rejecting the isolated, independent individual self of traditional psychiatric norms – a self which many today regard as not even realistic, let alone desirable [57].
Conclusion
While the ethical principles guiding other medical subspecialties may be necessary in the development of an ethics for psychiatry, they are not alone sufficient for such an ethics. Attributes of the psychiatric patient, the patient's condition, and the therapeutic project, each partly explain why psychiatry needs a distinctive professional ethic over and above the medical ethics incumbent on its practitioners. These attributes, together with some of the consequences for such an ethics which flow from recognition of those considerations, were outlined here.
This was not intended as an exhaustive explanation of the uniqueness of psychiatry; nor was it implied that any one of these attributes is unique to psychiatric practice. When taken together, nonetheless, these considerations intensify the distinctness of psychiatric practice in such a way as to call for distinctive ethics. At least three issues, the focus on character, the language of boundary violation, and the special prominence of gender, will be integral to such an ethics.
Footnotes
Acknowledgements
Versions of this paper were presented at Monash University's Center for Human Bioethics, and at the Department of Psychiatry, Massachusetts General Hospital. I am grateful to commentators on those occasions and to several people who have helped me sort out these ideas: Roger Crisp, John Sadler, Justin Oakley, David Brendel, Janet Farrell Smith, Jane Roland Martin, Beatrice Kipp Nelson, Barbara Houston, and Ann Diller. Finally, I wish to acknowledge help from an anonymous assessor for this Journal.
∗Versions of this paper were presented at Monash University's Center for Human Bioethics, and at the Department of Psychiatry, Massachusetts General Hospital.