In an important editorial in the British Journal of Psychiatry, Kendell suggested that the practice of dichotomising illness into mental and physical categories is ‘archaic and deeply misleading’ and ‘incompatible with contemporary understanding of disease’ (Reference KendellKendell, 2001: p. 492). Kendell argued that since the various taxonomies of mental disorder are essentially a list of conditions treated by psychiatrists, it is a matter of pragmatism whether (for example) Alzheimer's disease is classified as mental illness and treated by a psychiatrist, or as a physical illness and treated by a neurologist. In sanctioning this approach Kendell was reiterating the claims made by Carl Gustav Hempel, who after articulating his celebrated deductive–nomological model of scientific explanation (Reference Hempel and OppenheimHempel & Oppenheim, 1948) went on to crucially influence the development of the Diagnostic and Statistical Manual (DSM) classifications of mental illness in his 1959 landmark paper, ‘Fundamentals of taxonomy’ (see Reference Hempel and HempelHempel, 1965).

In a related and equally important article, also in the British Journal of Psychiatry, Cheng furthered the cause of scientific psychiatry by expressing the opinion that ‘cultural variation in mental health is mainly in the presenting features rather than in the nature and frequency of the underlying [my italics] neuropsychiatric impairments and disorders’ (Reference ChengCheng, 2001: p. 3). He suggested that an approach that recognised the pathoplastic effects of culture on illness behaviour and the existence of culture-general psychopathology, while at the same time respecting the biological basis of mental illness, was feasible providing that ‘standardised diagnostic interview[s]... incorporated psycholinguistic equivalents from different cultures’ (Reference ChengCheng, 2001: p. 3). In articulating these opinions Cheng was echoing the intention of the architects of the International Classification of Diseases (ICD) and DSM classifications to create a transcultural scientific psychiatry.

Kendell and Cheng are not saying exactly the same thing, but both are promoting a scientific psychiatry, the details of which go almost unquestioned within the profession. It is, however, possible to cast doubt on the validity of their views by considering an alternative approach to the mental which is much discussed in the philosophy of mind, language and anthropology, and with which most psychiatrists are unfamiliar. The approach in question can explain two things: first, why the distinction between mental and physical illness is a necessary and unavoidable one, and second, why the claim that mental illness is the same in all cultures is an implicit assumption, rather than an empirical finding, in Cheng's cross-cultural psychiatry.

MENTAL SCIENCE

The alternative to the views of Kendell and Cheng can be seen as a distillation of ideas that have their origins in predominantly German hermeneutical works such as Johann Gustav Droysen's Grundrisse der Historik (Reference Droysen and Andrews1858), Wilhelm Dilthey's Einleitung in die Geisteswissenschaften (Reference Dilthey, Makkreel and Frithjof1883) and, more recently, Hans-Georg Gadamer's Wahrheit und Methode (Reference Gadamer, Weinsheimer and Marshall1960). The central tenets of this tradition are, however, most clearly articulated in the writings of the American philosopher Donald Davidson who, partly in response to the perceived inadequacies of Hempel's account, argued that the mental is ‘anomalous’ and not subject to scientific laws, and that as such it can best be understood by reflecting on the nature of interpretation. Since Davidson's thinking has assimilated and progressed beyond the Hempelian neopositivism that informs current scientific psychiatry, his arguments for ‘anomalism’ and ‘interpretationism’ are likely to provide the basis for thinking that the claims of Kendell and Cheng are mistaken.

With this in mind, in his essay ‘Mental events’ Davidson (Reference Davidson and Davidson1970) articulates the view that there is a categorical difference between the mental and the physical and that there cannot, in any useful sense, be a science of the mental because of the impossibility of either strict psychological or strict psychophysical laws. Against psychophysical laws, Davidson argues that mental and physical concepts have different conditions of application or ‘constitutive principles’ and that ‘there cannot be tight connections [psychophysical bridge laws] between the realms if each is to retain allegiance to its proper source of evidence’ (Reference Davidson and DavidsonDavidson, 1970: p. 222). On the nature of the mental, McLaughlin writes:

(Reference McLaughlin, McLaughlin and LeporeMcLaughlin, 1985: p. 356).

Physical theory, on the other hand, ‘promises to provide a comprehensive closed system guaranteed to yield a standardised, unique description of every physical event couched in a vocabulary amenable to law’ (Reference Davidson and DavidsonDavidson, 1970: pp. 223–224). In the light of these different commitments the existence of psychophysical laws must a priori be impossible. If they did exist, such laws would have the intolerable effect of transmitting the conditions of application between the mental and the physical, with the result that either irrational thought and behaviour would violate the laws of physics or the mental would be constrained by physical constitutive principles in a way that would breach the norms of rationality. Davidson concluded that ‘the nomological irreducibility of the psychological means... that the social sciences cannot be expected to develop in ways exactly parallel to the physical sciences’ (Reference Davidson and DavidsonDavidson, 1973a : p. 230).

The argument against psychological laws derives from the view that in the natural sciences the predictive and explanatory power of laws can be improved by more accurately specifying the causal variables. In psychology, on the other hand, there is no hope of improvement in our everyday way of understanding individuals, because, given that psychophysical laws do not exist, there is no independent way of identifying and measuring the mental states of those individuals. Take the putative psychological law ‘if a man wants to eat an acorn, then he will if the opportunity exists and no other desire overrides’ (Reference Davidson and DavidsonDavidson, 1973a ). The difficulty with this law is that (as Evnine puts it),

(Reference EvnineEvnine, 1991: pp. 21–22).

PHILOSOPHICAL ANTHROPOLOGY

The remarks quoted above indicate that a science relying on mental states is implausible. The mention of ‘interpretation’, however, gives an indication of the likely basis of an alternative account. The nature of interpretation can be informed by a consideration of the theoretical difficulties encountered in trying to understand the language and behaviour of individuals from other cultures. In its philosophical form this project amounts to the search for a theory of meaning for a foreign language (L) which will consist of axioms and theorems that together give the semantic properties, the truth-conditions, of the sentences of the language concerned. Each theorem will consist of an English sentence that is true if and only if the foreign language sentence is true: for example, ‘il fait chaud’ is true in language L (in this case French) if and only if it is warm. If a speaker of L does assent to the sentence ‘il fait chaud’ only when it is warm, then one has confirmed the theorem, interpreted the foreign language sentence, and therefore made a start in constructing a theory of meaning for that language. In theory this process is repeated until eventually one comes to understand the whole of the language in question.

In pursuing the theoretical complexities of constructing such a theory, Davidson realised that if the speaker of L assented to the sentence ‘il fait chaud’ then this was not just because ‘il fait chaud’ means ‘it is warm’, but also because the speaker believed it to be warm. Unfortunately, however – and this is the crucial point – if the speaker of L believes it to be warm when it is not warm, because the evidence for the truth of the theorem is purely in the form of what sentences are assented to under which conditions then an interpreter will not know what the speaker means, since the former does not know what the latter believes. Davidson writes:

(Reference Davidson and DavidsonDavidson, 1974: p.142).

Ideally, an interpreter needs to be able to hold an individual's beliefs constant in order to establish the meaning of that individual's words; the difficulty is, however, that beliefs require interpretation as much as do meanings. In order to resolve this difficulty, Davidson embraced the hermeneutical idea that beliefs and meanings are internally related, and introduced the ‘principle of charity’. This principle asserts that when one is interpreting someone from another culture one should assume, without otherwise being able to access an internal circle, that they are psychologically similar to and, therefore, believe the same as oneself. Davidson writes that this assumption

(Reference Davidson and DavidsonDavidson, 1973b : p.137).

The principle of charity is a way of giving substance to the idea that people from different cultures are psychologically similar, and it enables Davidson to resist cultural relativism, albeit by an a priori rather than an empirically based argument.

MENTAL AND PHYSICAL ILLNESS

Informed by the view that a scientific approach to mental states is untenable, Kendell's arguments in favour of identifying mental and physical illness can now be assessed. In terms of aetiology, the claim that mental illness can have physical causes is not enough to sustain Kendell's claim. The physical symptoms of illness are presumably related to underlying physical causes by laws that can be sharpened as science improves our knowledge of the physical universe. In the case of mental symptoms, since there are no laws there is no such hope, and there is no reason whatsoever to think that knowledge of the phenomenology of mental illness will improve beyond its current state. This, in fact, is the main reason why – despite enormous progress in terms of the scientific understanding of physical illness in the last century – Jasper's General Psychopathology remains unsurpassed.

The different constitutive principles of the mental and physical give reason to think that not all mental illness has a physical cause. Physical symptom-concepts have physical conditions of application, whereas mental symptom-concepts, particularly those of minor mental illness and personality disorder, have broadly rational conditions of application; they relate to individuals' other beliefs, desires and actions in a normative way that cannot be captured by a theory that is formulated in a physical vocabulary. For example, the application of the concept of ‘misattribution’, which is likely to prove aetiologically important in somatisation disorder (Reference Butler, Chalder and WesselyButler et al, 2001), requires a prior familiarity with the normative constraints of belief formation. However, as Margolis reminds us,

(Reference Margolis, Sadler and SchwartzMargolis, 1994: pp. 124–125).

The absence of a principled relation between mental symptoms and underlying physical causes coupled with the normative nature of the mental means that the diagnosis of mental illness will, even if it is secondary to an identifiable physical cause or illness, always have to be based on an assessment of the phenomenology. There are, of course, mental illnesses that result from physical causes at a level with sufficiently general effects as to be amenable to physical intervention. Neurologists can, in principle, compete with psychiatry over the management of these disorders in a way Kendell presumably envisages – but only after mental state examinations have provided them with diagnoses. Furthermore, even if it did turn out that all people with a given mental illness such as depression had a given neurophysiological state in common, one could not sanction giving them a physical treatment without deferring to a mental state examination.

INTERPRETATION AND CULTURE

Cheng's views inherit the difficulties just mentioned since, in the absence of an identifiable physical basis, the culture-specificity or otherwise of mental illness becomes that of legitimately identifying cross-cultural psycholinguistic equivalents. The discussion of interpretation implied, however, that to find such equivalents an interpreter interested in mental illness is required to assume precisely what Cheng sets out to prove, namely that mental illness is the same in all cultures. In other words, without making the assumption that ‘underneath’ linguistic and conceptual differences different cultures really are talking about the same thing, how can one ever be sure that one's interpretations are accurate? The matter is, incidentally, complicated significantly by the decision to allow the existence of ‘culture-specific illness behaviour’ (Reference ChengCheng, 2001: p. 1).

In resting on the assumption of something like the principle of charity, the viability of cross-cultural psychiatry, like anthropology in general, comes at a high price. Although the principle ostensibly enables cultural relativism to be avoided, it achieves this at the cost of converting what Cheng would construe as primarily empirical into an interpretative enterprise. The principle of charity, which is essentially an injunction to assume that other individuals are psychologically the same as oneself, can hardly function as a fundamental scientific principle on which to erect a science of cross-cultural psychiatry. Furthermore, if these views are right, the enormous amount of time and effort that goes into ‘finding’ psycholinguistic equivalents should be interpreted, not as evidence of empirical rigor, but rather (as Husserl might say) as reflective of the infinite nature of the task. This is in some ways reassuring, because it carries the implication that the work of psychiatry – unlike that of physical medicine, or indeed science in general – will never be complete.

Having conceded that the project is interpretational, it might be advisable to go the whole way and admit that the distinction between mental illness and the associated illness behaviour is at best vague, and following Margolis, embrace some form of cultural relativism. It is likely, however, that over the coming years cross-cultural psychiatrists will become increasingly reluctant to concede this issue, as they are better able to identify psycholinguistic equivalents. However, if other cultures continue to be exposed to Western ideas at the current rate, then the requirement for the principle of charity will gradually recede as cultures come to resemble each other. If this occurs – and one may reasonably assume that it will – then it will not indicate that cross-cultural psychiatry has come of age, but rather support the implicit claim of this editorial, that psychiatry has more in common with politics than with medicine.