Sunday, 5 April 2009

Philosophy of Science for Psychiatry for the Person

I've been asked to draft a short piece for the WPA on philosophy of science for a book on their Program on Psychiatry for the Person. Here is my first stab.

(PS May 2011: The much shorter and altered final published article is here.)


The Institutional Program on Psychiatry for the Person calls for a more comprehensive model of diagnosis in which conventional diagnostic elements are combined with a specifically person centred extra ingredient. The World Psychiatric Association’s call to focus efforts on psychiatry for the person coincides with proposals to revise both DSM and ICD taxonomies which are likely to stress the importance of improving their validity through the epistemic values exemplified in natural science. Thus the Program aims to balance a growing emphasis on the natural scientific underpinnings of psychiatry with an increased focus on the importance and role of the person.

This proposal to balance natural science with the role of the person mirrors some of Karl Jaspers’ aims a century ago. At the turn of the century in Germany, psychiatry was dominated by academic neuroscientists working under the assumption, epitomised by the German psychiatrist Wilhelm Griesinger’s famous aphorism, that ‘Mental illnesses are brain illnesses’ [Jaspers 1997: 459]. Psychiatric researchers such Griesinger, Alzheimer, Nissl, Carl Meynert and Theodor Wernicke were searching for the neuropathological changes to explain the major psychoses motivated by their success with general paralysis. Whilst Jaspers was optimistic about this natural scientific project, he also thought that they might result in a one-sided view of psychiatry arguing ‘These anatomical constructions, however, became quite fantastic (eg. Meynert, Wernicke) and have rightly been called “Brain Mythologies”’ [Jaspers 1997: 18].

Jaspers’ response was to stress the role of understanding in addition to explanation in psychiatry. This reflected the debate, called the Methodenstreit, about the correct methods for psychology in the late nineteenth century. Should the human sciences (the Geisteswissenschaften) attempt to copy the methods of the natural sciences (Naturwissenschaften), or should they follow a distinct method or methods? ‘Positivists’, such as John Stuart Mill, Auguste Comte and Emile Durkheim argued that the human sciences were methodologically no different from the natural sciences. Others, including Heinrich Rickert, Wilhelm Dilthey and Wilhelm Windelband, argued that human and natural sciences had essentially distinct methods.

Jaspers argued for a distinction within psychiatry by both distinguishing between subjective and objective symptoms and by describing two approaches to understanding subjective symptoms [Jaspers (1912) 1968, (1913) 1974]. Static understanding, usually called ‘phenomenology’ by Jaspers aims to articulate what mental states are, or feel, like. Genetic understanding, usually called ‘empathy’, concerns how one kind of mental state ideally and typically gives rise to another.

Setting the development of the Program on Psychiatry for the Person against that background suggests the importance for it also of an understanding of the Methodenstreit.

The modern Methodenstreit

Whilst empathy and phenomenology still have a role in contemporary discussion, Jaspers’ account of the possible distinction between the human and natural sciences is not influential. The most influential set of arguments for a distinction between human and natural sciences were developed in the 1960s drawing on the work of the Cambridge philosopher Ludwig Wittgenstein (1889-1951). One such influential work was Peter Winch’s (1927-98) The Idea of a Social Science [Winch [1958] 1990].

Winch argues that there could be no such thing as a social science modelled on natural sciemce. Human understanding cannot and should not be modelled on the natural sciences because it employs a different form of intelligibility. In the preface to the 1988 edition of the book, Winch suggests that he does not mean by this the distinction between explanation and understanding as developed by the nineteenth century German philosopher Max Weber and incorporated into Jaspers’ psychopathology. Rather he has a deeper point. Only if there is an antecedent or background level of understanding can the sort of deficiency of understanding which an explanation might fill, be intelligible. (Thus ‘social science’ is the label for something impossible. Human understanding is not a social science. With that caveat in place, I will speak more loosely of human understanding and even human science in what follows with the proviso that ‘science’ need not mean a nomological study.)

This presupposed background understanding is ‘expressed in the concepts which constitute the subject matter we are concerned with. These concepts... also express certain aspects of the life characteristic of those who apply them’ [Winch 1988 : x]. One of the key aims of Winch’s book is to chart just what this background understanding is like.

Winch argues that a central element of understanding meaningful behaviour is an understanding of the nature of rules. For this he draws on Wittgenstein’s discussion of rule following in the Philosophical Investigations [Wittgenstein 1953 §§139-239]. Winch argues that:

1. Rules are central to social science because actions are constituted as the actions that they are by the rules that are operating. Thus, to give one of his examples, putting a cross on a piece of paper is an act of voting given the right context of rules.
2. Explaining an action by citing a rule presupposes a grasp of the rule not just by the social scientist but also (to a first approximation) by the agent whose behaviour is being explained.
3. Rule following is grounded in implicit practical knowledge of what actions count as going on in the same way. Rule following cannot rest entirely on explicit linguistically codified knowledge because that explicit knowledge would require further implicit knowledge of how the written prescription is to be interpreted.

Rules have a further important feature. They are normative: they prescribe correct and incorrect behaviour. In the example mentioned above they prescribe the difference between a successful vote and a spoiled ballet paper. Only certain actions count as casting a vote. So if understanding an event involves relating it to a rule, this form of understanding involves a notion of correctness or normativity. It involves understanding what makes it correct or appropriate as a piece of voting behaviour. This is not the same as saying that most votes are cast at a particular time of day or night or by a particular socio-economic proportion of the electorate. That may be discovered by empirical study. But the normative rules that characterise an event as an act of voting are not provided by any such statistical generalisations.

With these claims in place, Winch goes on to argue that the kind of understanding usually thought to make up social science is fundamentally dissimilar to natural science. Because explanation of meaningful action in terms of rules presupposes a grasp of the rules in question by the people whose actions are being explained or understood, human understanding deploys fundamentally different kinds of generalisations to natural sciences. They are not universal true generalisations under which events can be subsumed. They are instead open-textured patterns of behaviour which, by virtue of the normative rules acquired by shared use in the social context, grant actions their meaningful intelligibility.

One of the examples that Winch gives is characterising the behaviour of a cat as writhing. As he says, the very same movements might be plotted out in great detail in a physical vocabulary. But the two statements could not be substituted one for another. They belong to different conceptual frameworks. Now one might think that what is missing from the purely physical description of the movement of the cat is the fact that the cat is a conscious animal. And that may be true. Perhaps the concept of writhing is reserved for conscious beings. But the key point here is more modest: writhing is connected as a matter of its meaning with pain and ascriptions of pain can be used to explain or rationalise certain forms of (subsequent) behaviour. By contrast the purely physical description does not sustain these rational connections. (Note also that in this example, the rationality in play is not rationality for the cat. In this case, a pattern of understanding that has its primary role for rational agents – for persons – is extended to a mere beast.)

Winch followed Wittgenstein who had a profound influence on philosophy in the UK but much less so in the USA. But similar arguments are implicit in the work of US philosophers. Wilfrid Sellars (1912-89), for example, argued for a fundamental distinction between natural scientific and normative conceptions along similar lines. Sellars distinguishes between a natural scientific view of the world (or ‘scientific image’) and the ‘manifest image’ which is defined thus:

The ‘manifest’ image of man-in-the-world …is… the framework in terms of which man came to be aware of himself as man-in-the-world… [A]nything which can properly be called conceptual thinking can occur only within a framework of conceptual thinking in terms of which it can be criticized, supported, refuted, in short, evaluated. To be able to think is to be able to measure one’s thoughts by standards of correctness, of relevance, of evidence… [T]he transition from pre-conceptual patterns of behaviour to conceptual thinking was a holistic one, a jump to a level of awareness which is irreducibly new, a jump which was the coming into being of man. [Sellars 1963: 6]

This summary of the manifest image has three important aspects:
1. The manifest image is closely connected to the ability to exercise conceptual thought.
2. Conceptual thought depends essentially on critical evaluation.
3. There is a fundamental discontinuity between, on the one hand, conceptually structured thought and action and, on the other, preconceptual behaviour.
These three elements mark an important contrast with the elements that make up a scientific image of man.

Sellars is not alone in taking there to be a key distinction between normative person-level descriptions and underlying natural scientific accounts. Building on Sellars’ work, John McDowell, for example, contrasts the logical space of reasons with the realm of law or of natural science [McDowell 1994]. Donald Davidson argues for the ‘constitutive ideal of rationality’ which has ‘no echo in physical theory’ [Davidson 1980: 223, 231]. All three authors share an assumption about the central importance of the normativity of person-level descriptions. Normativity cannot be accounted for in natural scientific descriptions of the world. Thus, on Winch’s assumption that the social or human sciences chart the rules that shape human behaviour, there is a fundamental distinction between them and natural science.


The view that there is a fundamental distinction between the normative concepts that characterise the human sciences and non-normative concepts that underpin natural sciences has its critics. One approach is to attempt to explain the normativity of rules and concepts through phenomena that are, on the one hand, normative (or quasi-normative) but are, on the other, clearly rooted in the natural world and available to natural scientific theory.

In the philosophy of thought and language Ruth Millikan has argued that language is a biological category [Millikan 1989]. Its representational properties are explained through the biological, natural or proper functions of relevant human neurological traits. Such functions can be used to distinguish between behavioural dispositions which are in accord and those which are not in accord with the functions of a trait. Thus, functions set a normative standard (akin to rules) but are, nevertheless, a proper feature of biological and evolutionary theory. Her view has clear influenced that of Wakefield in the philosophy of psychiatry and his attempts to explain mental disorder through a similar idea of failure of biological function.

Also working with the philosophy of psychiatry, Derek Bolton and Jonathan Hill have attempted to reconcile the normative and non-normative. Instead of distinguishing between rational reasons and causes they distinguish between intentional and non-intentional causes. This places the hard physical sciences on one side of the divide and the equally hard biological and behavioural sciences on the other. Everyday psychological explanations – folk psychology – are classified alongside and continuously with sciences that invoke the notion of information alongside that of causality. What was useful and appropriate about the distinction between reasons and causes can be captured by the new distinction without incurring its difficulties. Reason explanation is a form of intentional-causal explanation as exemplified in many respectable sciences. They summarise the flow of their argument as follows.

The first step... is that explanation of action in terms of meaningful states has predictive power; the second is... that such explanation is causal; the third is the assumption... that the brain causally regulates action, all of which can be made compatible on the methodological assumption that the meaning (information) that regulates action is encoded in the brain. [Bolton and Hill 2003: 86]

The idea is as follows. The explanatory power of everyday intentional or ‘folk psychological’ explanation derives from the causal power of reasons. But the historical division between reasons and causes voiced, for example, by Karl Jaspers puts this claim under threat. If, instead, one distinguishes between intentional and non-intentional causation, folk psychological explanation ceases to be exceptional and in need of special philosophical explanation and becomes instead a particular example of a more general phenomenon.

Conclusions for Psychiatry for the Person

This is not the place to attempt to adjudicate the debate about whether there is a fundamental distinction between the natural and the human sciences. Those who argue that there is, stress the central importance of rule-governed or conceptually structured behaviour which, they argue, cannot be captured within a natural science view of subsumption under general or statistical laws. The ‘space of reasons’ is distinct from the ‘realm of law’. Those who argue that there is not a distinction of kind – although there may be a difference of degree – argue either that the normativity of rules can be reduced to behaviour of biological functions which also impose a normative standard of sorts on behaviour. Or, they argue that the distinction between reasons and causes is a mistake and intentional causation is a feature of the broader natural world. Counter moves include arguing that biological functions do not behave sufficiently alike rules, or illicitly presuppose the normativity of rules and challenging the role of meaning in sub-personal or natural features.

But the debate has important consequences for whether the additional elements called for in a comprehensive model of psychiatric diagnosis are genuinely distinct in their underlying logic to the conventional and natural science inspired diagnostic elements. If they are not distinct in underlying form then it seems possible that a comprehensive model of diagnosis might be a unified one, deploying, for example, concepts of intentional causation across the board. But if, as Jaspers assumed, there is a fundamental divide, the new elements may be needed to give a fuller picture of the human subject, a person whose experiences need both understanding in meaning-laden and normative terms as well as explanation.


Bolton, D. and Hill, J. (2003) Mind Meaning and Mental Disorder, Oxford: Oxford University Press
Davidson, D. (1980) Essays on Actions and Events, Oxford: Oxford University Press
Jaspers, K. ([1912] 1968) ‘The phenomenological approach in psychopathology’ British Journal of Psychiatry 114: 1313-1323
Jaspers, K. ([1913] 1974) ‘Causal and “Meaningful” Connections between Life History and Psychosis’, trans. by J.Hoenig, in S.R.Hirsch and M.Shepherd. in Hirsch, S.R., and M. Shepherd, Themes and Variations in European Psychiatry, Bristol: Wright: 80-93
Jaspers, K. ([1913] 1997) General psychopathology, Baltimore: Johns Hopkins University Press
McDowell, J. (1994) Mind and World, Cambridge, Mass.: Harvard University Press.
Millikan, R.G. (1989) ‘Biosemantics’ Journal of Philosophy 86: 281-97
Sellars, W. (1963) ‘Philosophy and the Scientific Image of Man’ in Science, Perception and Reality, London: Routledge & Kegan Paul
Winch, P. ([1958] 1990). The Idea of a Social Science and its Relation to Philosophy, London: Routledge.
Wittgenstein, L. (1953) Philosophical Investigations, Oxford: Blackwell.