On the therapeutic status of McDowell's representationalism

I gave a talk in the Durham Philosophy Deportment which could, I realise in retrospect, have been significantly simplified. The key point, really, was that the move from the representationalism (by which I mean a view of experience as itself a content-laden state) of Mind and World to that of 'Avoiding the myth of the given' and after involved key changes which, rather than merely a matter of degree, change the point and nature of the account.

In the earlier picture, both a partial respect for the coherentism in Davidson's slogan that nothing can count as a reason for a belief except another belief (whilst trying to reject mere frictionless spinning in the void that coherentism might otherwise suggest) and a kind of Sellarsian innocence go hand in hand. The innocence is that, following Sellars' Myth of Jones, it is unproblematic to think of mental states as carrying - because modelled on - the same kind of claims as judgements. In Mind and World, the very same contents can be asserted in judgement, can be carried in the experiences that invite such judgement and can make up part of the world itself, understood as the totality of true Fregean Thoughts. A key passage runs:

In a particular experience in which one is not misled, what one takes in is that things are thus and so. That things are thus and so is the content of the experience, and it can also be the content of a judgement: it becomes the content of a judgement if the subject decides to take the experience at face value. So it is conceptual content. But that things are thus and so is also, if one is not misled, an aspect of the layout of the world: it is how things are. Thus the idea of conceptually structured operations of receptivity puts us in a position to speak of experience as openness to the layout of reality. Experience enables the layout of reality itself to exert a rational influence on what a subject thinks. [McDowell 1994: 26].

The twofold retreat in AMG (and there is a third more recent retreat with the idea that the content of good and bad disjuncts in the underlying disjunctivism is the same; it is merely the way such content is had that differs) runs thus.

I used to assume that to conceive experiences as actualisations of conceptual capacities, we would need to credit experiences with propositional content, the sort of content judgements have. And I used to assume that the content of an experience would need to include everything the experiences enables its subject to know noninferentially.
But both these assumptions now strike me as wrong. [Lindgaard 2008: 3]

This helps illustrate the previous Sellarsian innocence. In response to the question: which concepts structure experience?, the Mind and World McDowell would have replied: all the concepts that the experience non-inferentially warrants. Not any longer.

But, now, there is no longer an easy transition from world, via experience to judgement. The middle step in that chain is mediated by an entirely different kind of content: intuitional rather than propositional. So there is no longer a simple appeal to experience carrying the same kind of content. (This makes his appeal to a Sellarsian thought in his response in the Lindgaard collection to Bill Brewer's objection from the Muller Lyer lines odder, I have to confess.)

Further, the partial role for the Davidsonian slogan is further reduced since experiences no longer share even the form of beliefs. If the world imposes rational friction on judgement in such a way that preserves a common form (since the world is made up of true Thoughts) the connection via two further links - true Thought to intuitional content, and from intuitional content to true Thought - is no longer clear. In the latter case, we are told that bits of intuitional content can be carved up and reassembled as propositional content but since the concepts articulating intuitional content are a subset of those available non-inferentially, many judgments are based on intuitional content only via recognition.

Gosh that is still too long a summary. But my key thought was this. If the assumptions that hold all this in place, that need to be balanced in such a way to reduce felt tensions, can tolerate the double deployment of carving of and recognition from intuitional content, why not cut out the middle man? Carve up and recognise worldly states of affairs. Why, in other words, isn't McDowell Charles Travis? He has already given up the kind of clarity about rational relations that insisting that the relata are conceptual seemed designed to underpin. Who is to say, a priori, therapeutically, how the same concepts in different forms licence rational transitions in the space of reasons?

(No one who wants a theory of experiential content will be tempted by Travis' minimalist work. But McDowell does not explicitly want any such thing. He just wants to reconcile the felt tensions surrounding judgement's responsibility to its subject matter with something like a partial respect for Davidson so that one seeks justificatory links not mere exculpating.)

In questions, Rachael Wiseman suggested that McDowell and Wittgenstein have a different attitude to the first move in philosophy, the one Wittgenstein characterises thus:

How does the philosophical problem about mental processes and states and about behaviourism arise?——The first step is the one that altogether escapes notice. We talk of processes and states and leave their nature undecided. Sometime perhaps we shall know more about them—we think. But that is just what commits us to a particular way of looking at the matter. For we have a definite concept of what it means to learn to know a process better. (The decisive movement in the conjuring trick has been made, and it was the very one that we thought quite innocent.)—And now the analogy which was to make us understand our thoughts falls to pieces. So we have to deny the yet uncomprehended process in the yet unexplored medium. And now it looks as if we had denied mental processes. And naturally we don't want to deny them. [Wittgenstein 1953 §308]

Whilst Wittgenstein doubts that first move and calls us back to the zeroth move, the description of the phenomena, Rachael suggested that McDowell seems much happier to accept philosophical summarising as a kind of first move for subsequent analysis. And hence, she suggested, his less critical deployment of McDowellian metaphors. Such first moves merely call for some holistic pruning. Not all survive. I have not thought in such terms but it seems right. It marks that other curiosity. McDowell thinks that therapeutic dissolution can still - innocently - appeal to the philosophical canon.

(PS: There is a development of some of these thoughts here.)

Why 'In the Space of Reasons'?

In response to a question from someone reading philosophy but not a philosopher (better: a poet).

The phrase comes from a paper (now published as a short book) called ‘Empiricism and the Philosophy of Mind’ by Wilfrid Sellars, the mid C20 American philosopher. One question he addresses is whether knowledge has a foundation. His answer is that it does, it can be grounded in perceptual reports, but that these do not have a property sometimes expected of epistemic foundations: that they can be made independently of holistic considerations. They are not brute data in that sense. That, he suggests, is the ‘Myth of the Given’.

In building to that claim he comments:

The essential point is that in characterizing an episode or a state as that of knowing, we are not giving an empirical description of that episode or state; we are placing it in the logical space of reasons, of justifying and being able to justify what one says. [Sellars EPM §36 italics added]

Sellars claims here that knowledge has an essentially normative status. It belongs to Reason. This contrasts, for example, with those who argue that knowledge is ‘a true belief arrived at by a reliable process’. But it also suggests that holism of conceptual connections which contradicts previous versions of foundationalism (The Myth…). In a little more detail, he attempts to find a middle position between coherence and foundationalism. If something is to count as a perceptual report it must meet 2 conditions:
1) It must be reliable.
2) The subject who makes it must know it is reliable. Otherwise, trained parrots would count as making perceptual reports. But parrots are not justified in their ‘utterances’.

But answering 2) means that a subject must think of her reports as reliable. So she must have a conception of how the world works and how her reports work, under suitable conditions. But if that is the case, this is not a form of the Myth of Given because perceptual reports are now fallible. If her world view is wrong, that will infect her perceptual reports. So neither mere coherence (because perceptual beliefs are not inferred from anything else) but not foundationalism (since perceptual reports are fallible and can be undermined by failures of the world-view).

In Mind and World, John McDowell extends this picture beyond states of knowledge to any state that has empirical content, is about part of the world (whether or not meeting the additional requirements for knowledge). The very idea of mental states having content – intentionality in traditional philosophical vocabulary – requires their rational friction (so not merely causal contact) with the world. But then our only understanding of rational relations requires both relata have conceptual structure. Since experience provides the friction but experience is - for reasons McDowell develops elsewhere - a kind of direct openness to the world (by contrast, eg., with an internal structure in the veil of ideas) then the world itself must have conceptual structure. (McDowell here acknowledges a debt to German Idealism, Kantianism without the ineffable noumenal world, Hegel.) So there is nothing beyond the conceptual realm. There is nothing outside the space of reasons. That space has limits but no externally imposed limitations.

So my blog’s title is supposed to flag the importance of normative connections, of reasons or Reason, whilst at the same time hinting that there’s nothing unavailable to such a perspective. (That said I don’t actually accept the argument to German Idealism. One reason is this.) Given that there is some discussion on it from time to time of psychopathological states, of understanding within mental healthcare, this raises some tensions.

Transcultural psychiatry, cultural formulation and validity in DSM-5

(This is a second draft paper written whilst my mind has been dominated by feelings of dullness and misery. I have at least found trying to concentrate on philosophy some consolation.)

Transcultural psychiatry, cultural formulation and validity in DSM-5

Abstract

DSM-5 puts greater emphasis than previous editions non-Western cultural idioms of mental distress but without making explicit the relation between these and the psychiatric scientific aspirations, such as for their reliability and validity, of the rest of the taxonomy. The first section of this chapter outlines three possible views of the nature of transcultural psychiatric taxonomic concepts: a two factor view of underlying pathology overlain by a cultural shaping presupposed by one view of the role of cultural formulation and two versions of a one factor view, radical and conservative. But I argue in the second section that establishing the correctness of any one is none too easy. Two influential approaches to the nature of the concept of disorder – Wakefield’s harmful dysfunction analysis and Fulford’s failure of ordinary doing – can be pressed with only minor tweaking to support any of the a priori models of transcultural concepts. In the final section I examine one such idiom: khyal cap or wind attacks, a syndrome found among Cambodians. I argue that this does not fit any of the ways of domesticating variation from standard DSM-5 categories and that this suggests that the very idea of transcultural psychiatric diagnostic concepts fits uneasily with the rest of DSM-5.

Introduction: cultural factors in DSM-5

DSM-5 introduces a more explicit treatment of cultural factors in psychiatric diagnosis than previous editions of the DSM. In Section III, there is a discussion of the role of what is called a ‘Cultural Formulation’ including a semi-structured interview to help investigate cultural factors. In the Appendix, there is a ‘Glossary of Cultural Concepts of Distress’ which describes nine common conditions (though see below). In the Introduction, a number of suggestions are made as to how cultural factors might affect diagnosis and prognosis and thus should be investigated in a cultural formulation [APA 2013: 14]. Culture may affect:

·         The boundaries between normality and pathology for different types of behaviour.

·         Vulnerability and suffering (by amplifying fears that maintain panic disorder).

·         The stigma of, or the support for, mental illness.

·         The availability of coping strategies.

·         The acceptance or rejection of a diagnosis and treatments, affecting the course of illness and recovery.

·         The conduct of the clinical encounter itself thus affecting the accuracy of diagnosis, acceptance of treatment, hence prognosis and clinical outcomes.

The Introduction also summarises (in fact at greater length than the later discussion of the cultural formulation in the main text) three distinct ways that culture can impact on diagnoses. The single idea of culture-bound syndromes from DSM-IV is replaced by three notions: cultural syndromes, cultural idioms of distress and cultural explanations (or perceived causes) of illnesses (or symptoms). It is worth quoting the summary in full:

1. Cultural syndrome is a cluster or group of co-occurring, relatively invariant symptoms found in a specific cultural group, community, or context (e.g., ataque de nervios). The syndrome may or may not be recognized as an illness within the culture (e.g., it might be labeled in various ways), but such cultural patterns of distress and features of illness may nevertheless be recognizable by an outside observer.
2. Cultural idiom of distress is a linguistic term, phrase, or way of talking about suffering among individuals of a cultural group (e.g., similar ethnicity and religion) referring to shared concepts of pathology and ways of expressing, communicating, or naming essential features of distress (e.g., kufiingisisa). An idiom of distress need not be associated with specific symptoms, syndromes, or perceived causes. It may be used to convey a wide range of discomfort, including everyday experiences, subclinical conditions, or suffering due to social circumstances rather than mental disorders. For example, most cultures have common bodily idioms of distress used to express a wide range of suffering and concerns.
3. Cultural explanation or perceived cause is a label, attribution, or feature of an explanatory model that provides a culturally conceived etiology or cause for symptoms, illness, or distress (e.g., maladi moun). Causal explanations may be salient features of folk classifications of disease used by laypersons or healers. [ibid: 14]

Although the authors distinguish between these different ideas, they concede that the same elements may play a role in all three categories. For example, in the West, depression is used as an idiom of distress whether of an illness or pathology or of mere normal but significant sadness. But it is also recognised as a mental illness syndrome gathering together a number of symptoms. Finally, it is taken to be the cause of those symptoms. Just as depression can play the role of syndrome, idiom of distress and explanation, so can other concepts local to other cultures. Given this complication, although the ‘Glossary of Cultural Concepts of Distress’ describes nine common culture-bound syndromes, the concepts described may also play a role as idioms of distress and purported explanations or causes of experiences. The cultural concepts described are khyal attacks or khyal cap, ataque de nervios (‘attack of nerves’), dhat (‘semen loss’), kufungisisa (‘thinking too much’ in Shona), maladi moun (‘humanly caused illness’) nervios (‘nerves’), shenjing shuairuo (‘weakness of the nervous system’ in Mandarin Chinese), susto (‘fright’), taijin kyofusho (‘interpersonal fear disorder’ in Japanese). Each is related to similar but different concepts found in other, including Western, cultures. Khyal cap, for example, is linked to panic disorder.

But given that, in the years leading up the publication of DSM-5, much emphasis was placed on the attempt to increase the validity of psychiatric diagnostic categories, by contrast with their reliability which had already been increased by the stress on aetiologically minimal operationalised criteria, what stance does DSM-5 have to the content of the cultural concept? Are they, too, supposed to possess validity, to be genuine descriptions of real features of mental pathology? Or does the cultural sensitivity aimed at in a formulation and the semi-structured interview protocol require the adoption of a kind of anthropological relativism? Can the aim of cultural sensitivity and the articulation of non-Western idioms go hand in hand with the scientific ambitions of twenty-first century western psychiatry? Or does it require a kind of liberal irony towards the rest of the diagnostic manual?

In the next section I will outline three general ways of thinking about the cultural dependence of mental illness categories and hence the possible role of cultural formulations. Then in the subsequent section I will explore the merely loose connection between these and two broad approaches to the concept of mental illness. In the final section, I will return to reconsider the status of the cultural concepts actually listed in DSM-5 through consideration of one of them: khyal cap.

Three models of cultural concepts of distress

A two factor model of cultural variation

One way to understand how culture affects mental illness would be to think of the expression of mental illness as the result of two factors: an invariant endogenous factor and a cultural shaping. On such a view, mental illnesses either are, or are underpinned by, pathologies of some sort of universal substrate such as an essential human nature. This is the first factor.

Perhaps the most obvious candidate for such a substrate is human biological nature. This would fit a common emphasis within mental healthcare on the centrality of biological psychiatry especially for accounts of the aetiology of mental illness. It would also be consistent with Jerome Wakefield’s analysis of the concept of disorder as a harmful dysfunction (although the next section argues against any close connection or implication) [Wakefield 1999]. Setting aside the role of the value term ‘harm’ for the moment, the first factor in a two factor account of culture-bound syndromes might be a biological dysfunction picked out or explained in evolutionary terms.

Whilst biological nature is the most obvious candidate for the first factor, others are also possible. Consider Louis Sass’ account of Schreber’s delusions in Paradoxes of Delusion [Sass 1994]. The main claim of the book is summarised in an early passage thus:

[Schreber’s] mode of experience is strikingly reminiscent of the philosophical doctrine of solipsism, according to which the whole of reality, including the external world and other persons, is but a representation appearing to a single, individual self, namely, the self of the philosopher who holds the doctrine… Many of the details, complexities, and contradictions of Schreber’s delusional world… can be understood in the light of solipsism. [ibid: 8]

But the elucidation or understanding that Sass seeks isn’t merely aimed at one particular delusional experience or even at all of Schreber’s experiences considered as a whole. It is meant to shed light more generally on the nature schizophrenia itself. The reason it can (according to Sass) is that the experiences that characterise schizophrenia derive from a general and abstract feature of rationality:

[Madness] is, to be sure, a self-deceiving condition, but one that is generated from within rationality itself rather than by the loss of rationality. [ibid: 12]

So one might take the first factor of a two factor theory of cultural psychiatry to be an invariant feature of human mindedness whether unified as a biological dysfunction or more generally characterised in mental terms: in Sass’s case as a feature of rationality. It might, in other words, attach to the nature of rational subject-hood however that is (biologically) realised or underpinned. This would form the basis or underpinning of mental illness across cultures and not specific to any one of them.

Cultural variation enters this (two factor) picture only with the second factor. Culturally invariant pathologies of underlying human nature are overlaid by local cultural variation in how they are expressed. ‘Expressed’ could carry either of two meanings. First, it might mean that standing possibilities for biological dysfunction or failings of rational subjectivity might be differently prompted by different social or geographical contexts. This would be akin to akin to variation in heart disease rates and causes in different cultures and hardly merits the label ‘cultural concept’. (I will return to this possibility a little later and will suggest it is better thought of as a one-factor model.)

The more interesting idea is that variation in ‘expression’ picks out the way in which underlying pathologies might be plastic to the different self-interpretations that different people in different cultures come to possess and thus the way the pathologies are experienced and avowed. This would be an example of a cultural idiom of distress in the vocabulary of the DSM-5. But whereas for physical illness, how one understands one’s illness might be thought to be an accidental superficiality compared with the real underlying condition (as understood, perhaps, by the medical profession), one might argue that for mental illness its esse is percipi: how it is perceived at least partly constitutes it. Thus in the case of Sass’ account of schizophrenia, a two factor model would be premissed on the idea that cultural variation might make it difficult to realise that the symptoms reported in different cultures resulted from something like the same failure within rationality. Identifying the common element would require significant interpretative work reflected in a cultural formulation.

I suggested earlier that on a two factor model, mental illnesses either are, or are underpinned by, pathologies of some sort of universal substrate. The difference between these options is the difference between thinking that the alloy of an invariant underlying pathology and a varying cultural overlay itself comprises what we mean by mental illnesses themselves. One might think, for example, that khyal cap and panic disorder have the same underlying biological mechanism but that the characteristic way in which the former carries its own ontology (ie that subjects think of their distress through the conceptual lens of a wind-like substance) is sufficient to mark it off as a different kind of mental illness. Biological dysfunction is then the common cause of two distinct illnesses depending on cultural context. On the other hand, one might think that the real illness is whatever is common to khyal cap and panic disorder. It is merely that the form that that single illness takes can vary.

Whichever view is taken of whether the first factor is the illness or merely the common underpinning of different illnesses, a two factor view of cultural concepts of mental illness suggests a particular view of the aim of a cultural formulation in psychiatric diagnosis. It is a way of reverse engineering, from locally divergent symptoms, the common underlying nature or the underlying causes of mental illness. The aim of sensitivity to cultural difference would be to find a way to penetrate beneath it to a common substrate appropriate for scientific psychiatric research.

This seems to be the view of the ex-president of the World Psychiatric Association Juan Mezzich et al. in their discussion of ‘Cultural formulation guidelines’ when they say:

The cultural formulation of illness aims to summarize how the patient’s illness is enacted and expressed through these representations of his or her social world. [Mezzich et al 2009: 390]

and

Performing a cultural formulation of illness requires of the clinician to translate the patient’s information about self, social situation, health, and illness into a general biopsychosocial framework that the clinician uses to organize diagnostic assessment and therapeutics. In effect, the clinician seeks to map what he or she has learned about the patient’s illness onto the conceptual framework of clinical psychiatry. [ibid: 391]

These passages suggest that there is a division between how an illness is enacted and expressed and the underlying framework set out by Western psychiatry. The former is locally culturally shaped. The latter is invariant. On Mezzich et al’s  account, the only positive role cultural factors can then play is as a source of contingent health promoting resources:

The aim is to summarize how culturally salient themes can be used to enhance care and health promotion strategies (e.g., involvement of the patient’s family, utilization of helpful cultural values). [ibid: 399]

In other words, ‘culturally salient themes’ do not reveal the shape of mental illnesses in themselves but can, contingently, be used to promote health because of their effects on how people understand their own illnesses. All this suggests that this underlying view of the role of a cultural formulation is determined by a two factor view. Such a view is, however, merely one of several possible. I will argue that it is a half way house between two more radical views of the possibilities for cultural psychiatry both of which of versions of a single factor which I will now outline.

Two versions of a one factor model of cultural variation

A two factor model of the nature of transcultural psychiatry requires a distinction between surface appearance and underlying pathology. But it might be that this distinction cannot be drawn. The various ways one might attempt to flesh out the contrast between underlying pathology – for example as biological or some other underpinning notion of universal human nature – and surface appearance might fail.

Consider the two versions of the first factor outlined above: biological dysfunction cashed out in accord with evolutionary theory and a pathology of the structure of rationality as such. In the first case, drawing a distinction between surface form and underlying invariant function or dysfunction might seem unproblematic for physical illnesses. But in the case of mental dysfunctions there may be no principled way of drawing a distinction between ways of thinking, for example, that are problematic within a particular culture and some underlying cognitive function underpinning several different forms. The surface form may simply be the dysfunction. Why? Whilst it is unlikely that there is no such thing as biological human nature and hence some shared biological underpinnings for human mentality it might not, unaided, determine mental pathology because it might not – without education and enculturation, for example – determine the kind of mindedness that mental illness threatens. Mental illness might be a feature of what McDowell calls our ‘second nature’, or, in German, bildung by contrast with biological first nature [McDowell 1994: 183]. Perhaps learning a language is necessary for some, at least, forms of mental illness such as thought disorder. Perhaps there is no principled way to factor conditions like depression into those aspects that require conceptual thought or language and mere biological underpinnings.

A one factor model need not imply that there is any cultural variation of mental illness. It might be that our second nature, or rather that aspect of it relevant for the formation of mental illnesses, is universal. If so, mental illness would be akin to heart disease, varying only in external features such as rates and superficial and unimportant local understandings of it. Any apparent deeper variation would be a mark of our ignorance, our misdiagnosis. So a conservative version of the one factor model likens mental illness to heart disease with no significant space for cultural variation and no need for a cultural formulation to extract or excavate the underlying commonalities because they are open to view.

But it is also possible that, because second nature depends on enculturation and because cultures vary, second nature also varies. If so, the richer notion of human nature, beyond mere biology and sufficient for a conception of mental illness, might not be universal. Cultural variation might go ‘all the way down’. Genuinely different forms of mental illness would emerge from different ways of living in different societies.

To flesh this example out it will be helpful to consider again but in more detail Sass’ account of schizophrenia according to which it is a failure of rationality from within, or driven by, rationality itself rather than a mere absence of rationality. The symptoms of schizophrenia are a kind of lived experience of the philosophical theory or stance of solipsism according to which only the subject of experience – for me: I – exists. Everything else is merely an idea (for me: one of my ideas). Solipsism is thus idealism whose implicit consequences have been explicitly adopted since if everything that exists is merely an idea only the first person subject of thought (for me: I) can have those ideas. But as Wittgenstein argued, solipsism is then strictly nonsensical because it presupposes a contrast between self and other (in the claim that everything is merely an idea) which it cannot consistently draw (since everything is an idea) [Wittgenstein 1929]. Sass, controversially, embraces this further feature of solipsism – that it is nonsense – to shed light on the pathological status of schizophrenia with the hope of solving ‘simultaneously for understanding and strangeness’ in Naomi Eilan’s useful phrase [Eilan 2000: 97]. (For criticism of just this point see [Read 2001; Thornton 2004].)

Such an account can be used to illustrate both the conservative and the radical version of the one factor model of cultural concepts. If one thinks that the history of Western philosophy merely illustrates and unpacks conceptual connections implicit in the rationality of any possible thinker then solipsism is also a standing possibility for any thinker and hence, on Sass’ account, so is schizophrenia as its lived version. That would be a conservative one factor model. Any apparent culturally determined local variation in the experience of schizophrenia, such as the specific contents of delusions by contrast with invariant forms, would be merely superficial, requiring no great cultural sensitivity to detect. (It is the thought that it is merely or trivially superficial which distinguishes this from a two-factor model with its demand for a cultural formulation to penetrate surface features.)

If, on the other hand, one thinks that Western philosophy has been driven not merely by the abstract demands of rationality but by historically contingent assumptions about the nature of subjectivity and the connection of mind and world then the temptation towards solipsism will seem to be a merely local cultural matter. At the risk of being glib, had Descartes not existed, there would have been no such thing as schizophrenia.

This version of the one factor model is more radical than the two factor model even though both agree on the need for some sort of cultural formulation. A radical one factor model of a cultural formulation is more radical because it does not enable one to dig beneath surface difference to find underlying common pathologies but would instead be an articulation of the genuinely different ways people can be ill in different cultures. According to it, there are genuinely different forms of mental illness which need have nothing substantial in common across different cultures.

The loose connections between different models of cultural idioms and rival accounts of mental disorder

Prima facie a two factor model and two versions of a one factor model – according to one of which (the radical version) there is radical cultural variation all the way down and according to the other (the conservative version) there is no cultural variation, akin to heart disease – are all possible approaches to the role of culture in psychiatric diagnosis. Given that the one factor model is coherent, at least, is there any reason to think the two factor model holds good, as Mezzich et al seem to assume in their account of the purpose of a cultural formulation? To repeat, the two factor model postulates a set of invariant underlying pathologies which are overlaid by cultural variation in the way they are expressed. What independent support can be given to that notion?

Given that I have illustrated the two factor approach, above, by appealing to Jerome Wakefield’s harmful dysfunction analysis, it might be hoped that settling on a satisfactory understanding of the concept of illness or disorder will also determine the correct view of transcultural psychiatry. But although there are connections, they are not as simple as they might at first appear. I will explain this by sketching the implications of two rival models of (mental) illness.

According to Wakefield, a disorder is a harmful dysfunction, where a dysfunction is picked out, it turns out, in accordance with evolutionary theory. (I say ‘it turns out’ to mark the fact that function is supposed to be a ‘back box concept’.) Evolutionary theory specifies the biological functions of the traits of the human mind and body. On this picture, there are evolutionary facts about mental function, from which deviations are failures of function, and these facts should hold universally. This approach might thus support the universal substrate necessary for a two factor model.

But, on reflection, it need not. It will only support a two-factor model if the underlying dysfunctions – identified via failure of biological functions – can also be culturally moulded or shaped. Suppose that there is no way to separate an underlying mental function or dysfunction from its surface appearance. One reason for thinking this might be a commitment to a difference of kind between the personal level and the sub-personal combined with the idea that there is no depth dimension to the personal or mental. If so, such a view does not fit a two factor model. How might this be? Consider the following description of a symptom (typical of psychotic conditions) in DSM-5.

Disorganized thinking (formal thought disorder) is typically inferred from the individual's speech. The individual may switch from one topic to another (derailment or loose associations). Answers to questions may be obliquely related or completely unrelated (tangentiality). Rarely, speech may be so severely disorganized that it is nearly incomprehensible and resembles receptive aphasia in its linguistic disorganization (incoherence or "word salad"). Because mildly disorganized speech is common and nonspecific, the symptom must be severe enough to substantially impair effective communication. The severity of the impairment may be difficult to evaluate if the person making the diagnosis comes from a different linguistic background than that of the person being examined. Less severe disorganized thinking or speech may occur during the prodromal and residual periods of schizophrenia. [APA 2013: 88]

Influenced by Wakefield, derailed switching from one thought to another could reasonably be construed as a failure of biological function: a function of the cognitive system. But it is not clear that it is susceptible to any significant cultural shaping. Whilst the subject matter, for example, of jumbled thoughts will depend on the circumstances, including broader cultural circumstances, of the person who experiences them, that seems to be a merely superficial surface colouring of the condition. So in the case of this symptom, the harmful dysfunction model of disorder seems to support a conservative one factor model of cultural psychiatry.

Could the harmful dysfunction analysis of illness be used to support the third option: the radical one factor view according to which mental illnesses are cultural shapings ‘all the way down’? Not, I think, on Wakefield’s own approach which is founded on a historical biological conception of function and dysfunction. But only two changes are necessary to fit that model. First, it requires changing the underlying analysis of function from a historical approach following Wright to a present functioning account following Cummings [Wright 1973, Cummins 1975]. Second, the context of such functions would have to be widened to take account of social functioning. Taken together, such a modification would suggest the possibility of conditions which impaired the local social functioning of individuals. Furthermore, both modifications seem plausible in the light of existing criticism of Wakefield’s precise view [eg Bolton 2008].

The harmful dysfunction account of disorder is consistent with all three models of cultural concept. Thus a defence of harmful dysfunction as a general analysis of mental disorder does not itself determine a view of cultural psychiatry. Wakefield’s position contains a value-free core. There are opposing views that hold that mental illness is evaluative through and through. Can such a ‘values-in’ view determine the correct account of cultural concepts in psychiatry?

On Bill Fulford’s version of a ‘values-in’ theory, illness corresponds to an endogenously caused failure of ordinary doing, an inability to do the sort of things that one should just be able to get on and do [Fulford 1989]. The analysis aims to capture both mental and physical illness. Both are value-laden. Fulford thus challenges an assumption about physical illness shared by both Thomas Szasz and Robert Kendell [Szasz 1960; Kendell 1975]. But he also stresses a contingent difference between mental and physical illness which is relevant here. We typically disagree about the values relevant to mental illness, and hence of the sort of impediments to ordinary doing, whilst we typically and contingently agree in the case of physical illness. We agree about the contribution of a heart to healthy ordinary doing, and hence about heart disease, but not mental flourishing and hence mental illness.

If Fulford is correct about that difference and its significance then his account might seem to undermine the universality of the underlying pathology necessary for a two factor model. (I suspect that that is what Fulford himself would think.) There would be no underlying universal substrate of pathologies because different cultures would have different values and thus quite different ways of failing to be able to act. This would support the radical version of the one factor model.

But that model of cultural psychiatry is not a necessary consequence of a ‘values-in’ view of mental illness. One might think that mental illness and mental health are essentially evaluative notions but that, on a proper view, the values involved are (that is, ought to be) universal. On such a view, there is a rich value-laden notion of the proper way for a human to be, for human flourishing, and deviations from it are value-laden mental illnesses. This might amount to either a two factor or a conservative one factor model. The difference depends on whether one thinks that the kinds of actions one should ordinarily be able to do can be common components within importantly different broader culturally determined contexts (the two factor model) or whether one thinks that the same basic actions, undermined by illness, are visible across different (the conservative version of the one factor model). On the former view there is a need for a cultural formulation to investigate common factors within apparently different practices whilst on the latter the commonalities are clear and hence there is no such need.

My aim so far has been to investigate different possible views of trans-cultural psychiatry. Three different views seem to make sense, each with different implications for the need for a cultural formulation. But there do not seem to be close connections between general accounts of the nature of mental illness and disorder and any one of the three. In other words, there do not seem to be very strong a priori arguments for any particular view. Given this, the only alternative is to look to the actual examples in DSM-5 and examine which model they fit and hence the implicit view of transcultural psychiatry.

The status of Khyal cap

Having sketched some abstract models of ways to understand cultural concepts of mental distress and their merely loose connections to rival accounts of mental disorder in general, I will now turn to one particular example from DSM-5 to see which model applies.

Khal cap is described in the following way. (It will be helpful to quote this one example in full to suggest the kind of description offered in the other cases too).

Khyal cap
‘Khyal attacks’ (khyal cap), or ‘wind attacks,’ is a syndrome found among Cambodians in the United States and Cambodia. Common symptoms include those of panic attacks, such as dizziness, palpitations, shortness of breath, and cold extremities, as well as other symptoms of anxiety and autonomic arousal (e.g., tinnitus and neck soreness). Khyal attacks include catastrophic cognitions centered on the concern that khyal (a windlike substance) may rise in the body—along with blood—and cause a range of serious effects (e.g., compressing the lungs to cause shortness of breath and asphyxia; entering the cranium to cause tinnitus, dizziness, blurry vision, and a fatal syncope). Khyal attacks may occur without warning, but are frequently brought about by triggers such as worrisome thoughts, standing up (i.e., orthostasis), specific odors with negative associations, and agoraphobic type cues like going to crowded spaces or riding in a car. Khyal attacks usually meet panic attack criteria and may shape the experience of other anxiety and trauma- and stress or related disorders. Khyal attacks may be associated with considerable disability.
Related conditions in other cultural contexts: Laos (pen lom), Tibet (srog rlunggi nad), Sri Lanka (vata), and Korea (hwa byung).
Related conditions in DSM-5: Panic attack, panic disorder, generalized anxiety disorder, agoraphobia, posttraumatic stress disorder, illness anxiety disorder. [ibid: 834]

I raised the question at the start of the chapter of whether cultural sensitivity requires the adoption of a kind of anthropological relativism or whether it is consistent with the privileging of a particular cultural standpoint: that of twenty-first century western psychiatry?

In this case, at first sight no such relativism seems necessary. Khal cap can serve as an ‘idiom of distress’: the conception of an experience had by a subject. If someone describes their experience as the rising up of a wind-like substance then that is simply an anthropological fact about the culture. It can serve as a ‘cultural explanation’ because, again, that is a fact about how a culture explains particular experiences without implicit endorsement of that theory of aetiology by the ascriber. But, by the standards of twenty-first century western psychiatry, it can even be described as a ‘cultural syndrome’ since that is defined as ‘a cluster or group of co-occurring, relatively invariant symptoms found in a specific cultural group, community, or context’. If, for whatever reason, the symptoms described co-occur then it is reasonable to call them ‘khal cap’. In other words, the sincere use of ‘khal cap’ by a transcultural psychiatrist need not cause any intellectual difficulty.

But such a reading of the description carries some implications when it comes to understanding the nature of culturally sensitive psychiatry. If the concept of a khyal attack is only ever used within the (intensional) context of what someone from that culture believes – his or her conception of the nature and explanation of their experiences – rather than as an objective description of what is really causing the attack, then that suggests a distinction of kind between cultural concepts (or culture-bound syndromes) and the main elements of DSM-5’s taxonomy.

Consider the question asked from a traditional Western psychiatric standpoint: ‘But from what are they really suffering?’. The description above suggests a ready answer selected from the list of related conditions in DSM-5: ‘Panic attack, panic disorder, generalized anxiety disorder, agoraphobia, posttraumatic stress disorder, illness anxiety disorder’. Such a response suggests that a culturally sensitive psychiatry might be merely a sensitivity to other cultures’ errors: the truthful ascription of a false belief about the causes of abnormal experiences.

With that worry in the background, I can approach the example of khyal attack through the range of options explored above. Recall Mezzich et al’s suggestion that the role of a cultural formulation is to ‘map what he or she has learned about the patient’s illness onto the conceptual framework of clinical psychiatry’. I suggested that this reflected a two factor model. If so, the underlying invariant factor is whatever is picked out by ‘panic attack, panic disorder, generalized anxiety disorder, agoraphobia, posttraumatic stress disorder, [or] illness anxiety disorder’. The varying local cultural shaping is the ‘catastrophic cognitions centered on the concern that khyal (a windlike substance) may rise in the body’.

The example fits the two factor model. But if so there remains an asymmetry between khal cap and panic attack because, from the perspective of the rest of the DSM, the former involves an error about the real aetiology of the condition. Dividing the condition between two factors does nothing to change this perspective.

Nor does it help to adopt the conservative one factor model. That presents a stark choice for any putative newly discovered mental illness. On this conservative view, culture-bound syndromes such as khyal cap can have either of two statuses. They are either really other names for universal conditions also picked out by the vocabulary of Western psychiatry such as ‘panic disorder’. Or they do not exist. For example, if it is an essential part of the theoretical apparatus of khyal cap that it is caused by the rising up of a wind-like substance then given that on our best account of physiology there is no such substance then, equally, there is no such condition. Those who self-report it, or its characteristic symptoms, are in some sense in error about their own conditions.

Could khal cap be understood in accord with the radical one factor model? Again, it seems not. That would require thinking of it as a genuinely different way of being ill resulting from being in a different culture. One way that might come about (influenced by the harmful dysfunction approach to disorder) is if it impedes a social function which has no echo in Western society. Or, drawing on Fulford’s work, it might be that there are local standards for what counts as ordinary doing and hence novel possibilities for endogenously caused failures of such doing. But khal cap does not seem to differ from health in respect of any novel social function or ordinary doing. It does not fit this more radical idea of transcultural psychiatry.

Acknowledgement

This chapter was written whilst a fellow of the Institute for Advanced Study, University of Durham. My thanks both to the IAS, Durham and the University of Central Lancashire for granting me research leave.

Conclusion

One of the criticisms of western psychiatry has been its cultural narrow mindedness, reflecting only a particular socio-cultural perspective [eg Watters 2010]. Thus the idea that DSM-5 contains a wider range of cultural idioms of mental distress than previous editions might suggest progress has been made in addressing this criticism.

Further, there are ways in which the existing DSM diagnostic categories could be augmented by other culturally-specific concepts. On a two factor model, the same underlying illnesses – or the same underlying causes of illness – might be experienced in different ways in different societies. Perhaps the disturbances of self that underpin schizophrenia with its characteristic delusions might be experienced differently in a culture not so influenced by the individualism imparted to the west by Descartes. On a radical one factor model, genuinely distinct conditions might be possibilities in virtue of different conceptions of flourishing, or ordinary actions, or societal functions (depending on the view taken of the concept of disorder). A compendious version of the DSM might chart conditions whether or not they were possible in every society.

However, the conditions set out in DSM-5, exemplified by khyal cap, do not fit either of these possibilities in a way that suggests cultural even handedness. If khyal cap is understood in accord with a two factor approach, the second factor nevertheless involves a kind of mistake. And it does not fit the radical one factor model.

This result is, perhaps, unsurprising. Although these cultural concepts of distress are flagged in the Introduction and discussed in the main body of DSM-5, their articulation and description is restricted to an appendix. They do not form a part of the taxonomy of mental illnesses proper, the taxonomy whose aims included validity. The very idea of transcultural psychiatric diagnostic concepts does not fit easily into DSM-5.

Bibliography

Bolton D. (2008). What is mental disorder? An essay in philosophy, science and values, Oxford: Oxford University Press

Cummins, R. (1975) ‘Functional Analysis’ Journal of Philosophy 72: 741–65

Eilan, N. (2000) ‘On understanding schizophrenia’ In D. Zahavi (Ed.), Exploring the self. Amsterdam: John Benjamins

Fulford, K.W.M. (1989) Moral Theory and Medical Practice, Cambridge: Cambridge University Press

Kendell, R.E. (1975) ‘The concept of disease and its implications for psychiatry’ British Journal of Psychiatry 127: 305-315

Mezzich, J.E., Caracci, G., Fabrega Jr., H. and Kirmayer, L.J. (2009) ‘Cultural formulation guidelines’ Transcultural Psychiatry 46: 383-405

Read, R. (2001) ‘On approaching schizophrenia through Wittgenstein’ Philosophical Psychology 14

Sass, L.A. (1994) The Paradoxes of Delusion, New York: Cornell

Szasz, T. (1960) ‘The myth of mental illness’ American Psychologist, 15: 113-118

Thornton, T. (2004) ‘Wittgenstein and the limits of empathic understanding in psychopathology’ International Review of Psychiatry 16: 216-224

Wakefield, J.C. (1999) ‘Mental disorder as a black box essentialist concept’ Journal of Abnormal Psychology 108: 465-472

Watters, E. (2010) Crazy like us: the globalization of the American psyche, New York : Free Press

Wright, L. (1973) ‘Functions’ Philosophical Review 82: 139–68

Naturalism and dysfunction

(This is a draft paper written whilst my mind has been dominated by feelings of dullness and misery. I have at least found trying to concentrate on philosophy some consolation.)

Introduction: disorder and naturalism

The concepts of illness, disease and disorder all share a prima facie normative character. Even Robert Kendell, who defended a plainly factual account, conceded that appearance in his 1975 paper ‘The concept of disease and its implications for psychiatry’. He writes:

Before we can begin to decide whether mental illnesses are legitimately so called we have first to agree on an adequate definition of illness; to decide if you like what is the defining characteristic or the hallmark of disease… By 1960 the ‘lesion’ concept of disease … had been discredited beyond redemption, but nothing had yet been put in its place. It was clear, though, that its successor would have to be based on a statistical model of the relationship between normality and abnormality… But… [a statistical model] fails to distinguish between deviations from the norm which are harmful, like hypertension, those which are neutral, like great height, and those which are positively beneficial, like superior intelligence. [Kendell 1975: 309]

The normative aspect of disease is suggested in this passage by the dimension spanning harm to benefit. This is a distinction beyond mere degree of difference from a statistical norm. It is normative as opposed to merely (statistically) normal. But normative notions present a challenge for the philosophical programme of placing complex concepts into a conception of nature, or ‘naturalising’ them as that project is usually known especially given the most influential version of philosophical naturalism: reductionism. (In his book on Philosophical Naturalism David Papineau argues that its fundamental characteristic is ‘the thesis that all natural phenomena are, in a sense to be made precise, physical’ [Papineau 1993: 1]. Hence showing how concepts pick out real and natural features of the world involves, ultimately, reducing them to physical concepts.)

The reason that normativity presents a challenge to philosophical naturalism so understood is that, on an influential neo-Humean view, the natural world is not itself the source of normativity: thinking subjects are. As Hamlet says, on this view: ‘there is nothing either good or bad, but thinking makes it so’. Thus normative concepts cannot be thought of describing the natural world but reflecting human subjectivity. This is clearest in cases where the normativity concerned, like Hamlet’s line, takes the form of explicit value judgements.

Here is one such example in the philosophy of disorder. KWM (Bill) Fulford defends an account of illness as an endogenously caused failure of ordinary doing [Fulford 1989]. He argues against both Thomas Szasz, who contrasts mental and physical illness, and Robert Kendell, who assimilates them as value-free, that mental illness and physical illness are both value terms [Szasz 1960]. The idea that illness comprises an internally generated failure of ordinary doing explains its values-ladenness because the concept of failure itself suggests an ineliminable negative value judgement. But Fulford also argues that differences of opinion about the value judgements need not generally imply error because they do not answer to anything objective. Value judgements are projections of a subject’s sentiments onto the world and hence differences of opinions should be explored rather than corrected [eg Fulford 2004]. Hence the class of illnesses does not pick out anything objective. It is a reflection of both worldly facts but also subjective values about which there can be rational disagreement.

Fulford’s account does not fit reductionist naturalism. Illness is not in that sense a ‘natural’ concept but an alloy of worldly fact and human value with the latter underpinning the prima facie normative element of illness. To naturalise illness – at least in accord with the dominant reductionist reading of that term – would require some way to account for the normative dimension in value-free and naturalistic terms. That possibility is the subject matter of this chapter. To investigate its prospects I will discuss Jerome Wakefield’s influential harmful dysfunction model of disorder.

Wakefield’s harmful dysfunction model

To be clear from the start: Wakefield does not attempt to provide a value-free analysis of illness or disease or disorder (unlike others such as Christopher Boorse and Robert Kendell who do.) (Note that although, perhaps influenced by Boorse, Wakefield talks of ‘disorder’ rather than ‘illness’ or ‘disease’, he does not suggest any firm distinctions between them; he comments: ‘some writers draw distinctions among disorder, disease, and illness. Disorder is perhaps the broader term because it covers traumatic injuries as well as disease/illness. I ignore these differences’ [Wakefield 1992: 374] I will follow his lead.) But he suggests that disorder can be analysed as a conjunction of one specific value and a value-free medical science core.

On his account, the normative dimension is divided between two elements. It features in the value ‘harm’ which forms one conjunct and helps encode the practical aims of medicine to intervene in only particular cases, the harmful ones. But it is also present in the concept of a dysfunction which turns out to be ‘anchored in evolutionary theory’ [Wakefield 1999: 465]. The resulting ‘harmful dysfunction analysis’ contrasts with Fulford’s analysis in that, in the latter, facts and values mingle ‘all the way down’. By contrast Wakefield’s approach aims to characterise a purely descriptive core for medical science using the idea of biological functions. In other words, the normative component of any illness is divided into an irreducibly value-laden element of harm and into a deviation from a biological function which is then reduced to, or naturalised via, descriptive biological theory. The class of biological dysfunctions is thus a natural class even if the broader class of disorder is not. The focus here is that narrow class.

The challenge of giving a descriptive, non-normative or non-evaluative account of function to explain this core element of disorder goes hand in hand with giving an account of failure of function. Only if an account can be given of what a divergence of the behaviour of a system from its function comprises has the notion of a function which could be successfully or unsuccessfully executed been substantiated. But if such an account of function can be given in value-free, descriptive terms, then it would ipso facto successfully account for failure of function. Thus it would be a mistake to assume that in characterising disorder partly as a failure of function, Wakefield has already conceded the game to value theorists because ‘failure’ is an evaluative concept as Fulford seems to suggest [Fulford 1999, 2000]. If function can be analysed in descriptive terms then so can ‘failure’ of function: it is any divergence from function. An apparently normative or evaluative concept would be reduced to a value-free descriptive analysis.

It may still seem that, in the case of function, a non-normative descriptive account is a hopeless non-starter precisely because the distinction between success and failure surely cannot be reduced to a purely factual or descriptive vocabulary. But just such descriptivist accounts of natural function have been proposed elsewhere as part of the wider legacy of Darwin. In the philosophy of language and thought, for example, Ruth Garrett Millikan proposes that the intentionality or ‘aboutness’ of thoughts and beliefs can be naturalised using the notion of biological functions [see especially Millikan 1984]. She argues that even conscious human purposes – paradigm instances of genuine teleology – are susceptible to this form of reductionist naturalism [Millikan 1998: 309].

Much has been written on the definition of function. Two broad approaches are perhaps most influential: the views of Cummins and Wright. Rachel Cooper summarises their differences thus:

Of the best known positions, those who adopt Cummins-type views [Cummins 1975] claim that the function of a sub-system is whatever it normally currently does that contributes towards the goals of a larger system. On such an account the function of the heart is to pump blood around the body, as this is what hearts currently normally do that contributes to the organism surviving and reproducing. On the other hand, those who favour Wright-style approaches [Wright 1973] think that the function of a sub-system is fixed by its history. In the biological domain, the Wright-function of a sub-system is whatever it was naturally selected to do. [Cooper 2007: 30-1]

Elsewhere Cooper suggests other options and suggests that there are prima facie difficulties with all of them for the analysis of disorder or disease.

For the function of X to be Z any of the following might be considered necessary:

1.            X was originally selected because it does Z.
2.            In the recent past selection has been responsible for maintaining X because it does Z.
3.            Currently selection is responsible for maintaining X because it does Z.
4.            At all times X has been selected because it does Z.
It is difficult to choose between these options as each is associated with potential problems. [Cooper 2002: 268]
It turns out that the process that explains the prototypical non-accidental benefits is natural selection acting to increase inclusive fitness of the organism. [Wakefield 1999: 471-2]
But the (putative) complexity of the intentional object of a mental state does not, of course, entail the complexity of the mental state itself. It’s here that LOT ventures beyond mere Intentional Realism... LOT claims that mental states - and not just their propositional objects - typically have constituent structure. [Fodor 1987: 136]
One of the problems that Cooper highlights is that if one opts for the original selective advantages of some trait which, as a matter of fact, now also prima facie serves another function then failure of that current prima facie function will not count as disease. But if recent history is taken to be key, then, because human societies and technologies now affect actual reproduction, traits which might seem prima facie to be dysfunctional but which are compensated for through human intervention cannot count as diseases. I will ignore these particular difficulties here.

Both Wakefield and Millikan favour a historical approach (like Wright’s) connecting functions to actual evolutionary selective histories and, as will become clearer, this is most apt for the reductionist project in question. Roughly speaking the biological or proper function of a particular trait of an organism is what explains the evolutionary success and survival value of that trait. (In fact, Millikan defines functions within an account of reproductively established families but the details of her theory will not matter here.)

Crucially, for the purposes of capturing the prima facie normativity of disorder (Walkefield) and intentionality (Millikan), biological functions are distinct from dispositions. The biological function of a trait and its dispositions can diverge. Engineering limitations might cause the actual behavioural dispositions of a trait to diverge from the biological function it thus only partially exemplifies. Further, the divergences might themselves be life threatening and play no positive part in explaining the value of the trait. The best explanation of the survival of that organism and those like it cites the function which helped propagation or predator evasion, for example, and not those aspects of its behavioural dispositions which diverged unhelpfully from it.

This point is sometimes put by saying that what matters is not which traits or dispositions are selected, but what function they are selected for. The distinction between ‘selection of’ and ‘selection for’ can be illustrated by the example of a child’s toy [Sober 1984]. A box allows objects of different shapes to be posted into it through differently shaped slots in the lid. The round slot thus allows the insertion of balls, for example. It may be that the actual balls allowed through or ‘selected’ in one case are all green. But they are selected for their round cross section and not their green colour. Millikan stresses the fact that the biological function of a trait may be displayed in only a minority of actual cases. It is the function of sperm to fertilise an egg but the great majority of sperm fails in this regard [Millikan 1984: 34]. Since biological functions can diverge from mere dispositions, they have extra resources necessary for accounting for the idea of failure of function. The distinction between success and failure of a system, organism or organ can be defined by reference to its functioning in accord with its biological function.

Wakefield’s work on disorder is in the same tradition: providing a reductionist account of a problematic concept by appeal to evolutionary theory. He offers an initially distinct, but eventually similar, account of natural functions. Drawing on essentialist accounts of natural kinds, such as water or gold, he suggests that natural functions likewise have an underlying essence. Thus natural functions are defined as sharing whatever the initially unknown essential process is which explains prototypical non-accidental beneficial effects such as eyes seeing. This is a surprising claim given that what unites natural kinds such as gold or water are first order physical properties. No first order physical properties unite natural functions. But Wakefield goes on to invoke explanation and natural selection in a much more standard way:

A natural function of a biological mechanism is an effect of the mechanism that explains the existence, maintenance or nature of the mechanism via the same essential process (whatever it is) by which prototypical nonaccidental beneficial effects... explain the mechanism which cause them...

Thus, like Millikan, Wakefield relies on an account of natural function drawn from explanation within evolutionary theory to distinguish those dispositions which accord with a system’s naturally selected function from those which do not.

However, despite this connection between the prima facie normativity of the concept of disorder and the normativity, albeit rooted in evolutionary history, of biological functions, there remains an ambiguity between two possible reductionist aims for such an account. In the next section I will clarify this through a detour into the philosophy of thought or mental content. I will then argue, in the following section, that an objection derived from Wittgenstein’s discussion of rules threatens one version but not the other. In the final section I will consider which aim is appropriate to naturalising mental disorder.

What kind of reductionist naturalism?

In this section I will distinguish between two aims for reductionist naturalism which can be compared to two horns of the Euthyphro dilemma. But to make this more concrete, I will characterise the difference using the actual aims of two competing, but reductionist, approaches in the philosophy of content: those of Millikan, already mentioned, and of Jerry Fodor.

In the lengthy appendix to his book Psychosemantics, Jerry Fodor articulates a general argument for reductionism in the philosophy of content.

I suppose that sooner or later the physicists will complete the catalogue they’ve been compiling of the ultimate and irreducible properties of things. When they do, the likes of spin, charm and charge will perhaps appear upon their list. But aboutness surely won’t; intentionality simply doesn’t go that deep. It’s hard to see... how one can be a Realist about intentionality without also being, to some extent or other, a Reductionist. If the semantic and intentional are real properties of things, it must be in virtue of their identity with (or supervenience on?) properties that are neither intentional nor semantic. If aboutness is real, it must be really something else. [Fodor 1987: 97]

The promise of the programme is that intentionality itself will be fitted into a conception of nature – or ‘naturalised’ – by a reduction to properties which are not essentially or intrinsically intentional or semantic. Since the latter are supposed to constitute the former (through identity or supervenience) it is not that they are not intentional or semantic, but they are not essentially so. Thus the concepts in the reduction base can be understood independently of grasp of the concepts to be understood thus serving a project of philosophical naturalism.

The work of the rest of the book looks at first sight to be a contribution to this task through the articulation of what Fodor calls a ‘representational theory of mind’. This comprises a ‘language of thought’ (LOT) to explain the relationships between mental representations construed as internal vehicles of mental content combined with a version of a causal theory of reference (an asymmetric dependence theory) connecting those internal vehicles to the world.

But, on reflection, whilst, if successful, the representational theory of mind would be a step towards a reduction of intentionality, it is not that the actual aim of the representational theory of mind as set out in Psychosemantics is quite as radical as the argument in the appendix. Consider the argument that mental representations must have a structure to map the structure of mental contents.

Practically everybody thinks that the objects of intentional states are in some way complex: for example, that what you believe when you believe that... P & Q is... something composite, whose elements are - as it might be - the proposition that P and the proposition that Q.

The aim of the account seems to be to explain how it is possible for thinkers to think thoughts with the right systematic relations to other thoughts. It is possible if there are inner vehicles of thoughts with an isomorphous structure to the structure of thought and, in turn, if the syntactic properties of those vehicles mirror their semantic relations and are suitably connected to their causal properties. Fodor attempts to show that it is not mysterious – that it is natural – that creatures like us can think the thoughts we can. If I may use the phrase the ‘space of reasons’ to stand for the rational relations between thought contents and between them and the world, it seems that Fodor takes his question to be:

·         Given the space of reasons, how is it possible for creatures like us to respond to it?

His answer is a piece of a priori engineering design. We must be creatures with an innate conceptual repertoire carried by a language of thought. This is still a form of reductionist naturalism. The fact that, medical limitations aside, humans can grasp a potential infinity of thoughts and chart the rational relations between them can seem puzzling and call for philosophical attention. The representational theory of mind is an attempt to make that less mysterious by showing how it would be possible for suitably engineered creatures to have those characteristics. But it is less radical than it might be because, within the main body of the text at least, it takes the conceptual connections themselves for granted.

Millikan, by contrast, aims to do something more ambitious with her evolutionary, or teleosemantic, theory of mental content. As summarised above, she deploys a tool which seems more promising than a causal theory of reference to account for mental content because it is itself an apparently normative notion: biological or proper function. A biological function is normative because it sets a standard against which the actual behaviour or dispositions of a biological trait or subsystem can be compared. She deploys this idea not just aim to explain how possessing mental content is the proper function of some cognitive system. Rather, particular representational contents are supposed to be explained in this way. The contents carried by inner vehicles are specified via the proper functions of those vehicles: that for which they are selected. Hence the selective advantages conferred must be characterisable in non-intentional terms. The meaning or content carried must drop out of the evolutionary theory rather than presupposed in specifying the advantage.

The aim of this analysis is thus more ambitious than Fodor’s project because Millikan aims to naturalise the structure of conceptual connections or ‘the space of reasons’ itself. Assuming that logic charts the rational connections between contents it is significant that Millikan claims that given a teleosemantic account, logic itself will become ‘the first of the natural sciences’ [Millikan 1984: 11]. So her key question is something like:

·         Given our biological natures, how is it possible for creatures like us to respond to what we take to be the space of reasons, whatever it is.

The difference in actual ambition between Fodor and Millikan is akin to the Euthyphro dilemma. Given a suitable theology, the following biconditional would be true:

·         For any act x: x is pious if and only if x is loved by the gods.

The dilemma stems from considering the ‘order of determination’, in Crispin Wright’s phrase, of this biconditional [Wright 1992]. Is the pious loved by the gods because it is pious, or is it pious because it is loved by the gods? Fodor’s and Millikan’s projects take opposing views. In effect, Fodor adopts the first horn and derives a priori engineering constraints on the gods (or thinkers) given that we know that they are able to track piety (or the space of reasons), antecedently understood. Millikan, by contrast, adopts the second horn and aims to explain piety (or the space of reasons) by describing the engineering of the gods (or thinkers) in independent (of piety or the space of reasons) evolutionary terms.

This distinction matters to the force of an objection that can be raised against Millikan’s programme which I will now summarise.

A Wittgensteinian objection to Millikan’s project

There is a familiar objection to Millikan’s programme based on Wittgenstein’s rule following considerations. It is tempting to think that meaning or mental content needs some sort of vehicle such as sign or symbol. Any such sign can, however, be interpreted in an unlimited number of ways and thus needs to be coupled with the correct interpretation. This point is often emphasised by commentators by suggesting interpretations of even extended demonstrations by example of the meaning of words or of mathematical series which are consistent with the examples given deviate or are ‘bent’ in some future application [eg Blackburn 1984]. But if mental content is explained as a mental sign which stands in need of the correct interpretation then the content of the interpretation will also need to be similarly underpinned. And this initiates a vicious infinite regress.

“But how can a rule shew me what I have to do at this point? Whatever I do is, on some interpretation, in accord with the rule.” – That is not what we ought to say, but rather: any interpretation still hangs in the air along with what it interprets, and cannot give it any support. Interpretations by themselves do not determine meaning. [Wittgenstein 1953 §198]

The same goes for accounting for understanding written or spoken signs or symbols. In the absence of any coherent account of a final interpretation which somehow blocks the regress, any account of mental content which depends on an interpretation faces a challenge.

Millikan’s teleosemantic account of mental content is a form of interpretation-based theory. Past behaviour is a set of signs to be interpreted. Like the interpretation of signs, such behaviour is consistent with an unlimited number of possible functions or rules including both continuations that seem natural and logical and an unlimited number of other ‘bent’ rules that deviate in unnatural ways. The normativity of a function implies that not every aspect of the behaviour of a trait or subsystem need match the function: what the trait is for. A trait may fail to match the function for which it was selected. What ensures the determinacy of biological function – what selects just one of the possible rules – is a particular explanation of the persistence of trait over evolutionary time. If the potential rewards of a trait are sufficiently great it may be that actual behavioural dispositions of previous instances of the trait only rarely match the function which explains the trait’s persistence. Hence the potential gap between actual past performance and the appropriate functional explanation. But the lesson from the discussion of ‘bent’ rules is that finite past behaviour could be explained as exemplifying many different or ‘bent’ functions, all of which would have been equally successful in the past but which would diverge in the future.

(Note also that this worry is not merely a kind of Quinean marginal indeterminacy akin to the difference between rabbits and undetached rabbit parts. Competing bent rules might be utterly different in future applications. By what principle is just one selected? I will return to this thought at the end.)

Millikan’s considers and responds to this objection in ‘Truth Rules, Hoverflies, and the Kripke-Wittgenstein Paradox’. Male hoverflies spend their time hovering and waiting for female hover flies to pass by at which point they accelerate in pursuit. ‘[T]he geometry of motion dictates that to intercept the female the male must make a turn that is 1800 away from the target minus about 1/10 of the vector angular velocity (measured in degrees per second) of the target’s image across his retina’ [Millikan 1993: 219]. Millikan calls this the ‘proximal hoverfly rule’ and suggests that male hoverflies are genetically programmed to follow it. That is, they have some internal mechanism ‘of a kind that historically proliferated in part because it was responsible for producing conformity to the proximal hoverfly rule, hence for getting male and female hoverflies together.’ [ibid: 219] So the biological function of the mechanism is to follow that rule.

But the behaviour of actual hoverflies may not accord with just that. One possibility is that hoverflies have some optical blindspots such that a female arriving in the blindspot of a male provokes no reaction. Such a possibility, however, would not be part of what explains the continued existence of the mechanism in the fly population. It would be noise rather than signal. The more worrying possibility is a rule which accords with all past successful fly on fly action but which diverges in the future. What in the evolutionary historical record could rule that out? Millikan dismisses such possibilities as follows:

[The ‘bent’ rule] is not a rule the hoverfly has a biological purpose to follow. For it is not because their behaviour coincided with that rule that the hoverfly’s ancestors managed to catch females, and hence to proliferate. In saying that, I don’t have any particular theory of the nature of explanation up my sleeve. But surely, on any reasonable account, a complexity that can simply be dropped from the explanans without affecting the tightness of the relation of explanans to explanandum is not a functioning part of the explanation. [Millikan 1993: 221]

This is rather a brisk response. A key element of it is that the bent rule contains a complexity which Millikan’s preferred explanation lacks. Her explanation is simpler. But from what perspective is her explanation simpler?

In the case of trying to naturalise mental content, her explanans is the meaning or content of particular vehicles of content. But she cannot invoke our prior grasp of what such contents seem more natural – of what it would be natural to mean – since that is what is supposed to drop out of, rather than being presupposed by, her analysis. And, of course, the content of the proper function is not just a matter of looking to behavioural dispositions but selecting a function which best explains them. But without presupposing the pattern that meaning imposes, the pattern of the space of reasons, what other principle is there to say what makes for a simpler explanation?

Millikan’s response would be legitimate for an attempt to answer the question I have suggested that Fodor attempts (despite his Appendix) to answer:

·         Given the space of reasons, how is it possible for creatures like us to respond to it?

Fodor’s implicit question presupposes, rather than seeking to explain, the pattern of normative liaisons between mental contents. Answering that question, it is not illicit to deploy a notion of simplicity that presupposes a prior grasp of the space of reasons because that is not what the question seeks to answer. But Millikan’s question is more ambitious and hence it is illicit in attempting it to presuppose a notion of simplicity that is based on prior grasp of the space of reasons.

Biological function and illness, disease and disorder

What then is the reductionist aim of appealing to biological functions in the philosophy of mental disorder? Two options can be articulated by translating from Fodor’s and Millikan’s questions in the philosophy of content. I suggested that Fodor’s question was:

·         Given the space of reasons, how is it possible for creatures like us to respond to them?

Translated into the context of naturalising disorder gives something like:

·         Given the concept of disorder, how is it possible for creatures like us to suffer it?

Unlike the parallel question in the philosophy of content, however, this question does not seem worth an a priori answer. I do not mean to presuppose an articulated theory of what is, and isn’t, worth philosophical attention. But there seems no pressing need to articulate a general theory of a failure to meet a normative standard by contrast with the felt need in the parallel semantic case to articulate a general theory of how it might be possible to meet one. Thus this question is not an appropriate way to model reductionism about the concept of mental disorder. What of the other option?

Millikan’s question was:

·         Given our biological natures, how is it possible for creatures like us to respond to what we take to be the space of reasons, whatever it is?

Translated into the philosophy of mental disorder gives question something like:

·         Given our biological natures, how is it possible for creatures like us to suffer what we take to be disorder, whatever it is.

Millikan’s looks the better model question for the philosophy of medicine. It makes questioning the nature of the concept of illness itself central rather than something presupposed.

It may be objected, however, that this cannot apply to Wakefield’s analysis of disorder as harmful dysfunction because he relies on an approach which he describes using the phrase ‘black box’. The notion of a natural function is defined as sharing whatever the initially unknown essential process is which explains prototypical non-accidental beneficial effects such as eyes seeing. It merely transpires – it comes as an a posteriori discovery – that the process that explains the prototypical non-accidental benefits is natural selection acting to increase inclusive fitness of the organism. And hence, the objection might run, this cannot be used to shed light on what is meant by the kind of dysfunction that underpins – when conjoined with harm – the idea of a disorder.

That objection goes too quickly, however. Although it is true that Wakefield does not engage in traditional conceptual analysis to underpin his conception of function, and hence dysfunction, that fact does not rule out the use of the supposedly empirically derived conception to shed light on the nature of function itself. It does not imply that he has to restrict himself merely to the kind of a priori engineering that Fodor attempts.

Furthermore, there is positive reason to think that he does more. One of the virtues that Wakefield claims for his analysis is that it is able to hold contemporary psychiatric taxonomy to account. In his persuasive co-authored book (with Allan Horwitz) The Loss of Sadness, for example, he argues that depression is over diagnosed because ‘normal sadness’ – that is sadness that is in accord with human emotional biological function – is mistaken for genuine, pathological depression [Horwitz & Wakefield 2007]. But since, he argues, mental disorder presupposes an underlying biological dysfunction, however unpleasant grief is, for example, it cannot be a disorder or an illness providing it is serving its (presumed) biological function. Given that the analysis is used to explain the very idea of disorder (via the underlying notion of dysfunction) the form of reductionist naturalism belongs to the more radical second horn.

But if so, because it shares the task of naturalising the normativity of pathology, Wittgenstein’s objection is a serious objection. That is, a biological teleological account cannot rule out wildly divergent accounts of the functions in play, functions which explain the presence of traits. And if so, we need a better version of naturalism for the philosophy of disorder.

A Quinean response?

My attempt to shed light on the nature of, and hence prospects for, reductionism in the philosophy of disorder has turned on an analogy between it and the philosophy of content and the prospects for reductionism about semantics. But it might be objected that there is a key disanalogy between the two cases.

Consider an argument often compared with Wittgenstein’s discussion of rules and meaning: Quine’s argument for the indeterminacy of translation [Quine 1990]. (Since the relation between indeterminacy of translation, inscrutability of reference and holophrastic indeterminacy are subject to interpretation and debate within Quine scholarship, a rough summary of one aspect of the argument will suffice.) Quine approaches the study of meaning via a particular methodological constraint.

In psychology one may or may not be a behaviourist, but in linguistics one has no choice. Each of us learns his language by observing other people’s verbal behaviour and having his own faltering verbal behaviour observed and reinforced and corrected by others… There is nothing in linguistic meaning, then, beyond what is to be gleaned from overt behaviour in overt circumstances [Quine 1990: 5]

This reflects a commitment to the idea that facts about meaning are public and shareable. But Quine goes further in limiting the kind of facts available to fix the facts about meaning to those which fit a particular scientistic world view. The project is constrained by connecting prompted ascent to sentences with environmental stimuli physicalistically described. These facts, however, underdetermine the translation of sentences. Since they are the only relevant facts, this implies that sentence translation is indeterminate.

With this argument in place, one might object that the comparison deployed so far between meaning and the concept of disorder is inappropriate. Whilst Quinean indeterminacy is revisionary of our pre-philosophical concept of meaning, the concept of disorder can more readily tolerate some such slack. So on the assumption that the Wittgensteinian argument outlined in previous sections sufficiently matches the Quinean argument just sketched, and on the assumption that some degree of indeterminacy is no threat to the medical concept of disorder, the claim that the project of reducing the concept of disorder to naturalistic terms introduces an element of indeterminacy is no threat to the project.

Such a response fails, however, because the supposed parallel between Wittgenstein and Quine is misleading. Quine accepts a degree of indeterminacy in his positive account of meaning. He thinks that the evidence that fixes meaning only goes so far. This is, of course, because Quine builds in an assumption that the evidence has to be physicalistically described. Against that background, meaning would be indeterministic. But what, aside from scientism, justifies that restriction?

Wittgenstein, by contrast, does not think that meaning is indeterministic. The contexts which play a role in constraining it are described in intentional terms. The apparent parallel just sketched between Quine and Wittgenstein is not part of the latter’s positive account but rather a reductio ad absurdum of reductionism. More significant, however, is that Wittgenstein’s negative argument does not deliver merely a domesticated indeterminacy but rather no shaping of content in the future at all and hence undermines the very possibility of radical reductionist naturalism in the case of disorder too.

Conclusion

I suggested that if the Wittgensteinian argument against the more radical reductionist aim is successful whilst there is no rational reason to pursue the more modest aim then the philosophy of disorder needs a better conception of philosophical naturalism. Fortunately, there are other approaches which can still justifiably claim to be forms of philosophical naturalism.

One can, for example, sketch the broader context in which apparently puzzling concepts are used in such a way as to make their use clear. One example of this is the way that Daniel Dennett attempts to demystify the mental by describing the ‘intentional stance’ within which mental properties are characterised [Dennett 1991]. Dennett’s aim is, by describing how the stance works and by suggesting that it is merely one of many possible stances for making sense of the world, to show how the properties so ascribed are perfectly natural even though they cannot be reduced to the properties deployed in other stances, such as the physical stance.

Dennett’s approach is one version of naturalism without reductionism. It helps to highlight the key assumption behind Fodor’s argument for reductionism quoted earlier in this chapter. The passage starts with an appeal to the shape of a future, completed physics. That serves as the benchmark of the really real and hence prompts the challenge for puzzling concepts. If they do not appear on the ultimate list then either they mark an unreal property or they must be reducible to concepts on the list. The challenge, however, presupposes without justifying the assumption that the physicists’ list is such a benchmark and that assumption can be contested.

John McDowell, for example, has suggested that nature is not restricted to what can be described using the vocabulary of the physical sciences. Criticising a reductionist construal of naturalism, McDowell, for example, says:

What is at work here is a conception of nature that can seem sheer common sense, though it was not always so; the conception I mean was made available only by a hard-won achievement of human thought at a specific time, the time of the rise of modern science. Modern science understands its subject matter in a way that threatens, at least, to leave it disenchanted...The image marks a contrast between two kinds of intelligibility: the kind that is sought by (as we call it) natural science, and the kind we find in something when we place it in relation to other occupations of ‘the logical space of reasons’, to repeat a suggestive phrase from Wilfrid Sellars. If we identify nature with what natural science aims to make comprehensible, we threaten, at least, to empty it of meaning. By way of compensation, so to speak, we see it as the home of a perhaps inexhaustible supply of intelligibility of the other kind, the kind we find in a phenomenon when we see it as governed by natural law. It was an achievement of modern thought when this second kind of intelligibility was clearly marked off from the first. [McDowell 1994: 70­1].

McDowell commends a different response to the prospects of a failure of reductionism. Rather than regarding this as impugning the reality of the properties or concepts concerned it may merely show that reductionists have started with an impoverished conception of the real or of nature. Central to McDowell’s picture is the possibility of undermining a dualism of normativity and nature. Nature itself may contain norms, it may be ‘fraught with ought’ in Sellars’s phrase. It is not restricted to what fits within the ‘realm of law’ articulated by the physical sciences, but also includes those emergent patterns and properties that have to be fitted within a different pattern of intelligibility: the ‘space of reasons’. This phrase marks in McDowell’s work (following Sellars) the rational pattern of intentional states (broadly construed to include ethical demands; it is thus comparable with but broader than Dennett’s intentional patterns [Dennett 1991]). But an analogous conclusion could be drawn for other, non­intentional, concepts for which naturalists have also attempted philosophical reduction (eg necessity, causality etc). Again the failure of an attempt at reduction in these cases need not undermine their reality.

McDowell’s views are influenced by a reading of Aristotle’s ethical views. He argues, elsewhere, that that both moral values and also secondary qualities form part of the fabric of the world [McDowell 1983]. The suggestion is that there may be features of the world for which one needs a particular kind of mind, perhaps formed partly as the result of training to detect, respond to and even to conceptualise. Thus one needs an appropriate moral education to understand, be sensitive, and resonate, to the demands that, say kindness, makes on one in particular circumstances [McDowell 1979]. But just because these are demands that can be understood only from such a perspective does not undermine their basic reality.

Applied to the concept of disorder, such a conception of philosophical naturalism suggests a different approach to the prima facie normativity of mental illness and disease from reducing it. By contrast with apportioning it either to the irreducibly normative value of harm or anchored, descriptively, in evolutionary theory, a more relaxed conception of nature allows that the normativity may be both more complex but no less part of the natural world. This has an important consequence. A substantial theory of disorder such as Wakefield’s forges a connection between it and dysfunction. On a reductionist interpretation, the latter concept has to be understood independently of, and hence shed independent light on, the former. That connection, however, may be informative even without the reductionism. Our grasp of disorder may contribute to our grasp of function and dysfunction and vice versa. And hence Wakefield’s analysis of the difference between, for example, sadness and depression may remain suggestive and helpful even in a new philosophical setting.

Acknowledgement 
This chapter was written whilst a fellow of the Institute for Advanced Study, University of Durham. My thanks both to the IAS, Durham and the University of Central Lancashire for granting me research leave.

Bibliography

Blackburn, S. (1984) ‘The individual strikes back’ Synthese 58: 281-301

Boorse, C. (1975) ‘On the distinction between disease and illness’ Philosophy and Public Affairs 5: 49-68

Cooper, R. (2002) ‘Disease’ Studies in History and Philosophy of Biological and Biomedical Sciences, 33: 263-282

Cooper, R. (2007) Psychiatry and Philosophy of Science Durham: Acumen

Cummins, R. (1975) ‘Functional Analysis’ Journal of Philosophy 72: 741–65

Dennett, D. (1991) ‘Real patterns’ Journal of Philosophy 88: 27-51

Fodor, J.A. (1987) Psychosemantics. Cambridge, Mass.: MIT Press

Fulford, K.W.M. (1989) Moral Theory and Medical Practice, Cambridge: Cambridge University Press

Fulford, K.W.M. (1999) ‘Nine variations and a coda on the theme of an evolutionary definiction of dysfunction’ Journal of Abnormal Psychology 108: 412-420

Fulford, K.W.M. (2000) ‘Teleology without tears’ Philosophy, Psychiatry and Psychology 7 77-94

Fulford, K.W.M. (2004) ‘Ten Principles of Values-Based Medicine’ in Radden, J. (ed) The Philosophy of Psychiatry: A Companion, New York: Oxford University Press: 205-34

Horwitz, A.V. and Wakefield, J.C. (2007) The Loss of Sadness, Oxford: Oxford University Press

Kendell, R.E. (1975) ‘The concept of disease and its implications for psychiatry’ British Journal of Psychiatry 127: 305-315

McDowell, J. (1979) ‘Virtue and reason’ Monist 62: 331-350.

McDowell, J. (1983) ‘Aesthetic value, objectivity, and the fabric of the world’ In Pleasure Preference and Value, ed. Schaper, E. Cambridge: Cambridge University Press.

McDowell, J. 1994. Mind and world. Cambridge, Mass.: Harvard University Press.

Millikan, R.G. (1984) Language, thought and other biological categories. Cambridge, Mass.: MIT Press

Millikan, R.G. (1993) White queen psychology, Cambridge, Mass.: MIT Press

Millikan, R.G. (1995) ‘A bet with Peacocke’ in Macdonald, C. and Macdonald, G. (eds) (1995) Philosophy of Psychology, Oxford: Blackwell 285-292

Millikan, R.G. (1998) In defence of proper functions. In Nature’s purposes: analyses of function and design in biology, ed. C. Allen and G. Lauder. Cambridge, Mass.: MIT Press

Papineau, D. (1993) Philosophical naturalism, Oxford: Blackwell

Quine, W. V. O. (1960) Word and Object. Cambridge, MA: MIT Press

Sober, E. 1984. The nature of selection. Cambridge, Mass.: MIT Press

Szasz, T. (1960) ‘The myth of mental illness’ American Psychologist, 15: 113-118.

Wakefield, J.C. (1992) ‘The Concept of Mental Disorder: On the Boundary Between Biological Facts and Social Values’ American Psychologist 47: 373-88

Wakefield, J.C. (1999) ‘Mental disorder as a black box essentialist concept’ Journal of Abnormal Psychology 108, no 3: 465-472

Wittgenstein, L. (1953). Philosophical investigations. Oxford: Blackwell

Wright, C. (1992) Truth and objectivity. Cambridge, Mass.: Harvard University Press

Wright, L. (1973) ‘Functions’ Philosophical Review 82: 139–68