PLEASE NOTE: This blog has now moved.
As of 7/11/15 all new posts will appear on
and comments on this site are closed.

Monday, 27 January 2014

Guest post. The manufacture of madness? Why social construction in psychiatry is not as simple as it seems

Huw Green

Mental health problems: Constructed or discovered?
Anyone who has spent time reading or listening to psychologists recently is likely to have encountered the idea that mental health problems are ‘social constructs’. What is meant by this is that entities such as depression or schizophrenia and personality disorder, which we might ordinarily think of as diseases; are actually descriptions that flow out of our culture and moment in history. There may be good reasons for thinking about mental health in this way. Anybody who claims that there is no social construction involved in the disorders outlined by the American Psychiatric Association (APA) is unaware of the way the manual is written. Prior to 1952 there was no DSM, and every 15 years or so since, a revision has appeared. These updatings are usually chock full of new diagnoses, many of which have been regretted by the very people who helped bring them into existence. Indisputable though this may be, it is a form of description which can stand in the way of understanding the true complexity of such problems. If they are just constructs then why do so many of the people who experience them find the experience so like a disorder; so real? In order to be clearer about this we need to ask exactly what we mean by social construction.

For some commentators, the implication seems to be that if we stopped talking about ‘schizophrenia’ or ‘personality disorder’, then they would more or less disappear. This is the argument which Mary Boyle appears to make in the final chapter of Schizophrenia: A Scientific Delusion? In this line of reasoning, there is much to be gained from demonstrating that life events, social inequality, abuse, and even the mental health system create ordinary, understandable distress, which then gets inaccurately and arbitrarily labelled. It is likely that this depiction is true in a good many cases.

 However, there are at least two meanings we might intend by saying that something is socially constructed and it is a distinction that is easy to fudge. In the first sense, we could be suggesting that social conventions are the only sense in which something exists (as with, for example, The Human Rights Act, The Premier League and The Church of England) and that a change in our verbal behaviour could eliminate it.

A second meaning would be to draw attention to the fact that certain parts of the natural world cannot easily be spoken about without recourse to elaborate, but potentially misleading, metaphors. Thus the space-time continuum gets referred to as a ‘ rubber sheet‘ ; strands of DNA get ‘ hijacked‘  and the hippocampus ‘ stores‘  memories. These are all constructs insofar as they are they are linguistically created mental images which help us imagine what is going on in reality. None of them is straightforwardly untrue, but if we take them too literally (DNA has never been held up at gunpoint by a primordial molecular criminal) they give us a misleading picture of complex processes.

This distinction between constructs that only exist through consensus and constructs that are created around real entities and processes, is drawn by the philosopher of science Ian Hacking who, in his book The Social Construction of What, explores the ways that childhood sexual abuse, mental illness and even rocks are ‘constructed’ in the latter sense of the word. Much of the debate about the reality of DSM constructs is tangled up in this distinction.

There exist, to paraphrase Wittgenstein, states of affairs in the world. Language meanwhile, is the system we have for trying to describe those states of affairs. Although the states of affairs can’t be changed by the words we speak, the way we furnish our world is no trivial matter, for what we call reality consists both of these states of affairs and the way we describe them.

So what of this second meaning of ‘socially constructed’? When we talk about the world we necessarily translate it into words, conjuring up mental images that move us away from the reality of the thing itself. Cancer is a ‘real’ physical event, but it remains true, as Susan Sontag points out in her beautiful Illness and it’s Metaphors that we speak about it in ways that are unhelpful. Take for example, the suggestion that people ought to ‘fight’ their cancer, implying that the sufferer can somehow do something tangible about their illness by sheer will alone, creating unnecessary misery for people with the disease.

What are the implications of this second form of social construction for our attempts to talk about, say, schizophrenia? One important difference is that although we can still say that schizophrenia is socially constructed, we remain nonetheless able to entertain the possibility that there is a distinct neurocognitive ‘disorder’ in the organism towards which this construct is legitimately trying to point. The definition given in the DSM, which has changed in various ways over the years, is very obviously constructed; a definition after all is just a verbal attempt to capture some state of affairs in the world. Meanwhile, the reality (the ‘state of affairs' itself) is something ‘out there’ beyond language and is not ‘constructed’ in the sense we are interested in here.

If this seems arcane, we should take it back to the level of the concrete. Psychiatric diagnosis is in a protracted state of disarray. There exist numerous accounts attesting to the fact that schizophrenia is frequently diagnosed in people who not only recover rapidly from their distress but feel the label itself does them more harm than good (the articulate Eleanor Longden and Peter Bullimore are prominent examples). Alongside these, there exist accounts (those of Elyn Saks or Peter Chadwick stand as good examples) suggesting that a Schizophrenia diagnosis can act in much the same way as the identification of any other disorder, accurately naming a real problem and flagging up a treatment more useful than any other that has been encountered. These two possibilities suggest a diagnosis that is overly inclusive rather than one which can simply be dissolved.

Under this description, the question is not one of rejecting psychiatric diagnoses but of improving them, both in terms of their design and in the rigidity with which they end up being applied in clinical practice. Part of the answer will lie in finding out the nature of different disabling neurocognitive states of affairs, how they manifest themselves, and how they are exacerbated by our social environment. Only when we have done this, can we stop using existing constructs (DSM diagnoses) and start using a system that resembles not so much labelling, but diagnosing proper.

Huw Green is a PhD student and trainee clinical psychologist based in New York. You can follow him on Twitter here. He blogs regularly at

Those keen for further discussion of the social construction and mental health may find this paper by Michael Walker of interest. (Ed).


  1. Good to see you mix up the view on the site. Not sure I agree with Huw though. This is not my understanding of Mary Boyle at all. 'Real' doesn't necessarily mean seeing things in a neurological way.

    1. Hi Ben,

      Thanks for your comment. Can you say more about how I may have misunderstood Boyle? To be clear in advance, I don't think she equates "real" and "neurological" but I do think it is unclear at the end of her book whether she believes there could plausibly be an entity worth talking/studying about in lieu of the DSM construct "Schizophrenia".

  2. My own personal view is that one cannot pick and choose what is socially constructed and what is not. Either it all is, or it all isn't. I'm unsure how much philosophical theory backs me on that, but I'm sure it wouldn't be far off.

    Even the seemingly "unconstructed" illnesses and diseases are constructed, if you choose to use the theory. Cancer, for example, is a word we created to describe the phenomenon of rapidly multiplying cells that formed lumps and eventually kill people. The words "cell" and "kill", again, are created.

    What i believe we DO have in psychiatry, which is different to the rest of medicine, is PRAGMATIC diagnoses instead of SCIENTIFIC ones. People making boxes and labels that seem to make life easier or our system more efficient, but which don't have any nosological validity. This is bad. It stems from our unfortunate lack of objective tests, which is not a failing of psychiatric diagnoses in themselves, but leaves them open to abuse. If this continues, we'll enter a vicious circle of invalidity.

    So in summary when we say "socially constructed", in a way I think we mean "pragmatic and unscientific", but the implication stands, and the practice has to stop.

    1. Short response to your comment Alex, had noticed your tweet. I thought cancer cells were in our bodies naturally and that they rise up, for some reason, in numbers to take over our system, to its detriment. Similar to psychiatry, if you like, which can take over, against our will.

    2. Interesting position but what's your reason for thinking that social construction (of mental illness I assume) must be all or nothing? In physical illness it seems possible that certain illnesses are socially constructed whilst others are not and I don't immediately see a reason why that shouldn't be the case here, particularly once we've introduced some nuance in the definition of social construction.
      Also interested in the implied contrast between pragmatic and scientific concerns. Haven't scientific categories and diagnoses always been shaped, even driven, by pragmatic concerns? If nosological validity is what draws the line between the two, how is that to be cashed out in a non-question-begging way?

  3. I should add, I only feel certain psychiatric diagnoses fulfill this criticism, like the American nonsense 'disruptive mood dysregulation disorder'. Overall, the major categories need only fine tuning (based on science not pragmatism).

  4. Thanks Huw for an interesting blog post with many twists and turns. I agree with your strapline after 40yrs of engaging with psychiatry in every decade since 1970 as a carer, patient, carer again, now a survivor activist and human rights campaigner.

    I've never believed in mental illness, having seen at first hand what the label can mean to me and mine. Forced treatment, locked up, stigmatised, blamed, shamed. And for what reason? Because we all in my family experience altered mind states when traumatised or at life/body transitions. We're sensitive people, creative thinkers, it's normal for us to be vocalise and externalise our distress. Rather than internalise it and get depressed.

    But the psych drugs depressed us anyway so same difference. We've all had various diagnoses and recovered despite them, by not believing a word of it. To improve the system you'll need to hand over the running of it to the people in it. As in, taking over the asylum. Here's to a new order of individual responsibility and civil servants who serve rather than rule.

    1. Hi Chrys,

      Thanks for your comment, it reminds me that although I think it's correct to say that social construction-arguments can't *dissolve* the problems we study in mental health, it doesn't follow from this that the treatments we use are always desirable or effective. Equally, it doesn't follow that they are "illnesses" or "diseases" (see professor Bentall's comments below), which remain words I am unsure how to define.

      What I seek to suggest is that although narratives and constructions are powerful and important to how we experience the world, we need to make *some* concession to an idea of external reality or succumb to a relativism that strikes me as unhelpful.

      I see something in your comment which suggests we agree here-you too describe an "altered mind-state" (a state of affairs in the world) which then gets described (constructed) in particular ways by different people. My main hope in writing here is that we can see that both social-constructionist thinkers and more straightforwardly positivist thinkers can see that their ways of speaking and thinking are not necessarily and hopelessly at odds.

  5. I find the pure social constructionist position as misguided as the pure biomedical position. My view is that people do have unusual experiences and our biology is involved (both genes and brain structure). However, it is psychosocial factors than cause genes to be expressed and brains have plasticity so are open to change without medication. The social construction position best explains how people get trained into the role of being a 'schizophrenic' by people around them such as their family (back to Laing's work) and then by services.

  6. Angela Gilchrist Clinical Psychologist27 January 2014 at 15:13

    It's interesting that some here are taking up an 'either/or' position i.e. that something is either completely socially constructed or it's not. I don't think that's true. Someone's experience of mental distress can be terrifyingly real, but that doesn't mean that there is a biological disease to be found underpinning it. Neither does it mean that biology is not involved in producing the distress. Clearly, biology is profoundly involved in the experience of mental distress, but it doesn't mean that it has caused it. Indeed, it seems more likely that disrupted biology results from the experience of trauma and turmoil. Sadly, the either/or position represents the Cartesian dualism that still largely informs medicine. Mind and body are surely one entity and constantly in interaction with one another, rather than separately functioning entities. It's interesting too, that Alex believes that the major categories of psychiatric diagnosis are settled and that these are based on science. What science, where? We have waited decades for bio markers, but psychiatry has failed to produce them. It blunders on, assuming that they will indeed be found, rather than turning its attention more usefully elsewhere. I suspect that these riddles will only be solved once we have untangled the mysteries of consciousness and discover how it is that emotions produce biological correlates. And even when we do understand that, our perceptions of the processes and their meanings will involve social construction. So, sorry guys, an either/or position on social construction is untenable.

  7. I think it is very helpful to make the distinction between our language of
    description and classification (which is much affected by social forces,
    vested interests and other quirks of human life, and hence which leads to
    imperfect taxonomic judgements) and actual states of affairs (which are
    not). A fundamentalist form of social constructionism fails to make this
    distinction whereas a critical realist account does. But there are three

    1. Finding out states of affairs in the world is not easy. An interesting
    comparison is between the classification of psychiatric disorders and the
    classification of species (which was probably a source of inspiration for
    Emil Kraepelin via his older brother, the distinguished insect taxonomist
    Carl). Species classification is hierarchical (with each species nested in
    a genus, nested within a family, nested within an order…. and so on, all
    the way up to the Animalia, Plantae, Fungi, Protista, Archaea, and
    Bacteria kingdoms) as opposed to the 'flat' system in the DSM. These days,
    research on biological classification is heavily influenced by a strict
    statistical approach applied to both morphological and genetic data (known
    as cladistics; see Gee, 2000 for a popular account).
    But there are still problems, for example in the very definition of a
    species. Since Darwin, it has been recognised that species evolve into
    other species, creating problems for where to draw the line between one
    species and another. Ernst Mayr (1942) famously suggested that the line
    should be drawn in terms of whether mating led to fertile offspring but
    this can be difficult to tell. The domestic dog, previous canis
    familiaris, was reclassified as a subspecies of wolf, canis lupus
    familiaris, in 1990 because it was found that it could interbreed with
    wolves (Nettle, 2009). It used to be thought that tigons (hybrid of a male
    tiger and female lion) and ligers (hybrid of a male lion and a female
    tiger) are infertile but the latest evidence is that they are not.
    Philosopher John Dupré (1993) has argued that there is no unambiguous way
    of defining species but it doesn't mean there aren't patterns in nature.
    He argues for promiscuous realism – choosing different taxonomies for
    different purposes. Similarly we should perhaps not expect that a system
    that works well in, say, public health, would be useful in the practice of
    clinical psychology. More to follow...

  8. 2. In the case of psychopathology, it is entirely possible that
    culture/historical forces influence both 'states of affairs' (how people
    express distress) and also how we describe them. While no one doubts that
    the latter is true, there is at least some reason for thinking that the
    former is also true. Ian Hacking (1995) marshalled considerable historical
    evidence that the presentation of ‘multiple personality disorder’ has
    changed over time, and has argued for a ‘looping effect of human kinds’
    according to which being classified as a certain ‘kind’ leads the person
    classified to change their behaviour in such a way that, over long
    periods, new classifications are required (Hacking, 1996). (I understand
    that he is sceptical about whether this is true for ‘schizophrenia’).
    Even if this is the case, however, it does not follow that there are no
    meaningful or even no universal patterns in experience and behaviour but
    it will make them more difficult to find. We might expect some historical
    and cross-cultural consistency at more general levels of classification (I
    am pretty convinced by Eysenck’s concept of ‘neuroticism’, and by the
    classification of common psychiatric problems into the internalizing and
    externalizing spectra (Kessler et al., 2011; Krueger, 1999) with the later
    addition of a broad psychosis spectrum (Caspi et al., 2013; Kotov et al.,

    3. The issue of whether any particular category of behaviour or
    experience is undesirable or qualifies as a 'disease' is completely
    different to the issue of whether there are patterns in those behaviours
    and experiences. The concept of disease is particularly worth thinking
    about in this context. Various criteria have been suggested for disease,
    including lesions (Virchow), syndromes (Sydenham) and harmful dysfunctions
    (Wakefield, 1992) but, arguably, none apply unambiguously in all cases –
    its rather like the example of species. Furthermore there is also a values
    component – if an inflammation of the appendix caused a doubling of our IQ
    with only mild negative consequences we'd be unlikely to consider it a
    disease – and also always a human interest element - a bacterial
    infection, from the point of the bacterium, is successful reproduction.
    These are, in the end, moral and pragmatic judgments. So, for example, we
    may agree on whether or not there is a cluster of traits that correspond
    to the label 'homosexuality', or about whether there is a meaningful type
    of experience corresponding to ‘hearing voices’, but, even if there is, we
    can still change our minds about whether to think about it as a disease.

    1. Hi Richard,

      Thanks for your comments, they are a valuable supplement to my post. I think Hacking's writings on looping effects and transient disorders are the most sophisticated treatments of this subject that I am aware of. Hacking shows us that classification is no simple matter and (as I am sure any service users are personally aware) has pervasive effects on how people are treated and on reality itself. The question you implicitly raise at the end of your point "2" is the same one I am always asking myself; for any given set of distressing experiences is there a consistency that we can pick out, classify and use as a replacement for DSM-diagnosis? My suspicion is that there is in some cases and that some DSM-categories may cleave closer to a meaningful entity than others.

      As for your point 3, you are of course right, and I have no particular horse in the race as regards adopting or discarding the language of "disease" or "illness". I am struck though by how discrepant people's experiences of those words are. For some they are accurate, for others they are anathema. Is this simply a question of taste? My wonder is whether, with greater care about classification, we can more successfully pick out those experiences that do indeed merit "illness" and those that merit a different characterisation ("reaction" "episode" etc.). An implication that seems to arise in some critiques of the DSM is that *any* bid to diagnose is mere hubris. Although I see the potentially atomising and dehumanising effects of diagnosing done wrong, this is not a view I currently share.

    2. I'm wondering about the spiritual or existential and where it fits in to the science, medical and behaviourist models? The "why are we here" and what's it all about" questions. That essence of humanity that is more than body, mind and brain, which can resist, stand and overcome in the midst of adversity or things happening outwith our control. I'm interested in what makes us individuals and different.

      The problem with psychiatric treatment, in my experience over a lifetime (I'm now 61) is that we are firstly seen as symptoms, presentations, behaviours, problems rather than a person with a story, past and present with potential in the future. Then we're given drugs and managed, if "psychotic", to become clinically depressed like the other patients. Non-compliance and free thinking not allowed. Straitjackets at the ready, chemically and institutionally. Some of us tongue the drugs, others resist and are forced, still others can swallow whatever and retain a personality.

      If the drugs don't work then you're "treatment resistant", if you are non-conformist in real life then you're "non-compliant". They call the drugs medication and therapeutic. Your psychiatric notes will record what the nurses think even done to every minute detail, if there has been an altercation or injury. "Difficult and demanding mother" might be written or "family history of ..." to justify coercion and detention. The mental health safeguards won't always be safe. It depends on the postcode lottery of mental health services.

      Meaningful involvement of "experts by experience" becomes meaningless tokenism, a tick box exercise when targets have to be met and time is short. Or that's the excuse given to high heid yins and uncivil servants.

      But what do I know? The schizo label is written indelibly in my psychiatric notes and it must be true. Even though I completely recovered, getting off the psych drugs under my own steam. (how dare you) I got off the lithium by tapering even though they said I had lifelong mental illness. If you don't believe it then it doesn't have the power over you, is how I see it. Could it be they got it wrong? And if so is it ever right?

    3. This point encapsulates one of the central problems in mental health care. As professionals we should want our interventions to be demonstrably effective and worth public money. As humans we should want want them to be able to engage with the spiritual and existential parts of our lives; the meaning that makes it all so important. I confess to not knowing how to make these fit together in harmony.

      We can and should conduct empirical studies to test our services, but it is easy for "what matters" to get overlooked, even squeezed out in the efficacy/efficiency trade off. Systematic efforts to honour individual meaning at a service wide level can struggle because it is an entirely subjective matter. What seems to one person to be a meaningful exchange looks to another like a trite gesture.

      What seems to happen in practice is that people end up working with professionals who do or do not embody an intelligent, receptive way of working with individual meaning. It is like a kind of spiritual post-code lottery. This seems unsatisfactory and surely there are ways of improving the situation (more money and more time would always help), but I have to agree that we seem to have no decent way of crow-barring in the "big questions" as things stand.

  9. And a few references to finish

    Caspi, A., Houts, R.M., Belsky, D.W., Goldman-Mellor, S.J., Harrington,
    H., Israel, S., . . . Moffitt, T.E. (2013). The p factor: One general
    psychopathology factor in the structure of psychiatric diorders? Clinical
    Psychological Science.
    Dupré, J. (1993). The disorder of things: Metaphysical foundations of the
    disunity of science. Cambridge, MA: Harvard University Press.
    Gee, H. (2000). Deep time: Cladistics, the revolution in evolution.
    London: Fourth Estate.
    Hacking, I. (1995). Rewriting the soul: Multiple personality and the
    sciences of memory. Princeton: Princeton University Press.
    Hacking, I. (1996). The looping effect of human kinds. In D. Sperber, D.
    Premack & A.J. Premack (Eds.), Causal cognition: A multidiciplinary debate
    (pp. 351-383). Oxford: Oxford University Press.
    Kessler, R. C., Ormel, J., Petukhova, M., McLaughlin, K. A., Green, J. G.,
    Russo, L. J., . . . Ustun, T. B. (2011). Development of lifetime
    comorbidity in the World Health Organization world mental health surveys.
    [Research Support, N.I.H., Extramural
    Research Support, Non-U.S. Gov't
    Research Support, U.S. Gov't, P.H.S.]. Archives of General Psychiatry,
    68(1), 90-100. doi: 10.1001/archgenpsychiatry.2010.180
    Kotov, R., Chang, S. W., Fochtmann, L. J., Mojtabai, R., Carlson, G. A.,
    Sedler, M. J., & Bromet, E. J. (2011). Schizophrenia in the
    internalizing-externalizing framework: a third dimension? Schizophr Bull,
    37(6), 1168-1178. doi: 10.1093/schbul/sbq024
    Krueger, R. (1999). The structure of common mental disorders. Archives of
    General Psychiatry, 56, 921-926.
    Mayr, E. (1942). Systematics and the origin of species, from the viewpoint
    of a zoologist. Cambridge, Mass: Harvard University Press.
    Nettle, D. (2009). Evolution and genetics for psychology. Oxford: Oxford
    University Press.
    Wakefield, J.C. (1992). The concept of mental disorder: On the boundary
    between biological facts and social values. American Psychologist, 47,


    1. Hi Richard, I agree with all your points of analysis but would expand on the second. From a critical realist perspective (not sure if this is also 'promiscuous'!) we can talk safely in broad terms about all societies noting and responding in some way or other to misery (we are hardwired as mammals to be frightened and sad in response to threat and loss), madness (a lack of unintelligibility is disruptive to role-rule relationships in any society) and persistent egocentricity (we are an interdependent species and so incorrigible egocentricity evokes anger, fear, exasperation or pity). These are 'demi-regularities' but the particular ways that they are codified or responded to, at the personal and population level, will vary over time and place because of cultural differences in relation to economic organisational forms (feudalism cf capitalism) and cosmological expectations (religious cf secular). Madness was thought of differently in antiquarian Greece than in the USA now but in both settings madness (like misery) was easily spotted-no psychiatric training required....Dave

  10. Interesting points from the commentators, but the post itself is an example of seriously confused thinking.

    Let’s unpick the argument a bit. It hinges on the contention that all diagnoses (and other descriptions of the world) are constructs - as indeed they are. Tumours don’t come neatly labelled as ‘breast cancer.’ Irregularities in insulin production don’t carry a sign saying ‘diabetes.’ Medical researchers come up with a descriptive term and use it as the basis for further research and refinement. This is neither new nor controversial – I doubt that any doctors would disagree with it.

    The crucial question is whether these constructs can be shown to have some meaningful correspondence to patterns or objects in the real world, thus enabling us to make predictions etc. Some constructs (God, ghosts, witches) clearly do not. Some (cancer, diabetes) do. The problem for psychiatry is that with a few exceptions such as dementia, all its ‘diagnoses’ belong to the first type. Moreover, diagnoses claim to represent a particular type of pattern, that is, a disease process. This is even further from being demonstrable - as even the people who invented (I use the term advisedly) them are now admitting.

    Huw Green attempts to sidestep this awkward fact by introducing the red herring of social constructionism. What has this got to do with the idea that diagnoses are constructs? Absolutely nothing at all. But by using the term ‘social constructs’ he somehow manages to conflate the discussion with social constructionism, and with the notion that social constructionists (silly people!) think that by discarding a label, you can get rid of the problems it attempts to represent. In other words, this is a version of the old ‘So you are denying that people are distressed?’ canard. I have never met anyone, whether of a social constructionist persuasion or not, who believed this - certainly Mary Boyle doesn’t. But this allows Huw to go merrily on his question-begging way, using the terms like ‘schizophrenia’ and ‘disorder’ as if they were unproblematic, and arguing that we shouldn’t reject psychiatric diagnoses but improve them. It’s a bit arguing that if we improved our definition of witchcraft we’d be sure to find more witches. You can’t ‘improve’ them because they are not medical diagnoses in any scientific sense of the word in the first place. That’s the whole point. If successive editions of DSM have taught us anything, it is that attempts at ‘improvement’ can only lead to ridicule.

    I can only be grateful that here in Britain, clinical psychologists have cut through this kind of nonsense and taken a clear stance on the issue of classification. Evidently the USA is a long way behind us. Can I suggest that instead of endlessly re-hashing these muddled and specious lines of argument, we devote more time to promoting and developing the many excellent examples of non-diagnostic practice?

    1. *ouch* - thanks for your comment Clare. Can my piece withstand these criticisms? I hope so. Just to quickly introduce myself (the context adds something here) I am actually British and have done my time in the NHS as a mental health worker and assistant psychologist. I moved to the US for personal reasons and have not “gone over to the dark side” as some might see it. I became interested in this debate because I wanted to explore what we could and could not say accurately both for and against psychicatric diagnosis.

      The first thing I would say is that I have not attempted here to launch a defence of the DSM, if it comes over that way then I maybe haven’t written clearly enough. The guiding bet of this post and others I have written is that the DSM is very problematic, but that diagnosis in psychiatry per se may not be as bad a thing as it is sometimes said to be.

      So to the meat. You ask why I refer to “social constructionism”; well, that is the philosophical milieu in which this debate takes place. Academics talk and write about the social construction of mental illness and Hacking’s book directly engages with Mary Boyle in this territory. Contrary to you, I do not see this as a red herring and I think your accusation on that score is unfair to my case.

      What I do suggest is that some (not all) MH problems might reasonably be regarded as “disorders”. I don’t think the DSM is necessarily much of a guide as to which can and which can’t. This is tricky territory. As Richard Bentall points out here, “disorder” connotes a value judgement, so why do I think the word might have some worth some of the time? Because of testimony by people (cited in my post) whose experiences of MH problems seemed to them to be “disorders”. Be assured though, I remain chary about whether and when to talk of disorder; the only reason I can see to use it is that it seems powerfully descriptive of what is going in some cases.

      As for improving diagnosis, I do not mean by this the endless refining of the DSM, which is slow, political and rooted in a particular psychiatric tradition. Although I feel less uncomfortable about the DSM than I used to, I see that it remains a problematic document. What I am in fact referring to in my final paragraph is the work of various researchers (many of them clinical psychologists and some of them British) who are teasing out the cognitive mechanisms and causes involved in various mental health problems. Some of those mechanisms have consistent neuropsychological underpinnings it will be valuable for us to know about. It may not be diagnosis as we know it, but recognising specific processes, which give rise to particular problems and can be addressed by specific interventions seems to be to be something approaching a diagnosis in all but name.

  11. More confused logic in your reply, I’m afraid.

    First, to be clear, I didn’t ask why you referred to social constructionism. I objected to your somewhat devious use of the term to imply that critics of diagnosis are denying the reality of people’s feelings.

    Second, what exactly do you mean by ‘disorder’? People can call their experiences what they like, but when professionals use these terms, they have a responsibility to be clear what they mean. I suspect that, exactly like the authors of DSM and ICD, you are using the word in a deliberately ambiguous way so that you can then turn round and say, ‘Aha! You agreed these are disorders, so obviously we need diagnoses!’ This does not stand up as an argument.

    Third, you seem to be unaware that you are using the term ‘diagnosis’ in two different senses. You argue that we need to ‘diagnose’ in the broad meaning of the word - ie understand what kind of problem we are dealing with, in the same sense that an electrician might ‘diagnose’ a malfunction in someone’s washing machine. You then slip in the assumption that this general kind of ‘diagnosis’ – identifying the nature of a problem - should be of a specific nature, ie a medical diagnosis. But this is a very particular kind of procedure which depends on theoretical links and biological evidence of the kind that is completely absent in psychiatry. Studies into cognitive styles and their neurological correlates etc may well come up with some interesting patterns, but there is no justification at all for assuming that these will be best understood as disease processes, or for making the leap from one sense of ‘diagnosis’ to the other. On the contrary, there are many very good reasons not to do so – not least the fact that we have an overwhelming amount of evidence suggesting a main causal role for life events in all forms of mental distress.

    If you are familiar with Eleanor Longden’s work you will know her motto: ‘Instead of asking “What is wrong with you?” we need to ask “What has happened to you?”’ This is a vitally important call from someone who was about as seriously 'mentally ill' as it is possible to be. There is no need or place for psychiatric diagnosis if we heed her message.

    Sorry, Huw Green, but we can see the holes in your argument even if you can’t.

    1. Hi Clare,

      Just a few brief comments on what you’ve said.

      You're right of course that you didn't ask directly why I used "social construction" but I wanted to explain why I don’t consider it a “red-herring” in this area. I certainly don't claim diagnosis' critics are denying the reality of people's feelings and I think "devious" is unfair.

      What I mean by "disorder" is not central to my case, but I have in mind roughly what I take Jerome Wakefield to mean: that an aspect of our psychology is not functioning in the way we want it too and that this causes us harm. Of course that in turn contains terms that need to be unpacked, notably ‘harm’, and 'function'-but that would need a whole other post. We might expect some disorders to have a primarily biological explanation, but not all, and a psychological causal explanation is still closer to a diagnosis than the sometimes arbitrary labelling of the DSM.

      I’m grateful for the distinction you raise between different senses of diagnosis. Far from being unaware of it, it underlies my argument, and I think that bringing it to light reveals what look like points of contact between our positions.

      To wit we can use your clarification to better articulate what I want to say, and for that reason I am glad you raise Eleanor Longden's work. Her proposal, that some subset of voice hearing is caused by trauma-induced dissociation, is a suitable candidate for what I mean when I say "diagnosing proper" (and what I think you mean by diagnosis in the "broad sense"). This offers us a theory of cause, mechanism and intervention. I absolutely think doctors ought to ask "what's happened to you?" if they are to fully understand the nature of the problem they are trying to help with. This is also true in other fields of medicine.

  12. OK, so let’s see where we’ve got to.

    For the sake of clarity, let’s avoid question-begging medical language such as ‘disorder’ or ‘diagnosis’ and simply talk about ‘difficulties’ and ‘an understanding of the reasons for someone’s difficulties.’

    Translated into these terms, this is what you now seem to be saying. People come along to mental health services with very severe difficulties which they are struggling with. We need to understand the reasons for those difficulties, and how and why they arose. We can do this by asking people about what has happened to them, and developing a psychological account of how this led to the difficulties. Over time, we need to develop more sophisticated accounts in terms of general psychological pathways and their neurobiological correlates.

    So – with the obvious exceptions that we know about, no need for medical diagnosis. And I think psychologists already have something that fits the bill instead. It is a theory of cause (or more appropriately for human beings, ‘reason’), mechanism and intervention. It’s called a formulation.

    In summary, you are calling for psychological formulation as an alternative to psychiatric diagnosis. Just like clinical psychologists in the UK. Yes?

    1. I am happy with (though not intent on) retaining "diagnosis" and "disorder", even if you don't feel I have defined them well enough here. I wrote this to try clear the ground conceptually and am not really "calling for" anything. However, I indeed do hope that the projects of formulation and diagnosis could be mutually complementary rather than pitched against one another.

      Some descriptions of formulation suggest it ought to replace diagnosis on the grounds that the latter is always undesirable. Contrary to this I am optimistic that neuropsychiatric, cognitive and clinical psychological research have the potential to one day yield a diagnostic system that better outlines the causal pathways and functional roots of mental health problems.

  13. Thanks, Huw, for your piece. You say that you think it’s unclear at the end of my book on the concept of schizophrenia whether I believe there could plausibly be an entity worth talking about/studying in lieu of the DSM constructed ‘schizophrenia’. I thought I had made it pretty clear that the answer is no, or at least I haven’t encountered any evidence so far.
    What I do believe is that as a culture we are in thrall to the idea of entities in this area – entities with their roots in the individual – and find it extraordinarily difficult to let go. This thinking seems to powerfully direct our attention to any evidence which seems to support it, however slightly, and away from evidence against it. This is also seen in discussions of classification. Richard B is surely right that we can identify different patterns for different purposes but many discussions on classification are still focused on individual attributes, or supposed attributes (Dave P’s post is a counter to this) rather than on behaviour/emotions etc in context.
    Like others, I find the ‘socially constructed or real?’ arguments a bit wearying and possibly not helped by talking about ‘the social construction of schizophrenia’ or ‘the social construction of mental illness’ when at least in this context it might be better to say ‘the construction of XYZ as schizophrenia’ or’ XYZ as mental illness’. I’m labouring this a bit because you suggest that I imply that if we stop talking about ‘schizophrenia’ or ‘personality disorder’ then they would more or less disappear. But what does ‘they’ refer to in that sentence? The words might disappear but not what they denote. This kind of linguistic slippage makes it more difficult to think outside the ‘entities’ box.
    What concerns me is that the DSM is busily selling a set of assumption- laden concepts as ‘descriptions of mental disorders’ and shaping our thinking in very unhelpful ways. I’m not sure that ‘socially constructed or real’ arguments are much help here. But when concepts supposedly developed within a natural sciences/medical framework can’t even begin to meet the criteria for validity set by that framework, then constructionist questions are urgent: How are these concepts made to seem credible? How do they shape people’s identities and with what consequences? Can they be resisted? What power relations are involved in their use? etc etc. Of course, we also have to try and account for the form and content of distressing/disturbing experiences and behaviour, how they arise, persist, change and are responded to. Here, I believe, various approaches can and do make a valuable contribution. But funnily enough, it’s mostly social constructionists and critical realists who acknowledge the need to make explicit and justify the assumptions underlying their approach. Positivists tend to present themselves as self-evidently, objectively studying reality. Rather ironic considering they often overlook large chunks of the world outside people’s heads.
    This relates loosely to your point about ‘disorder’, that the only reason you can see to use it is that it seems powerfully descriptive of what is going on in some cases. The meaning of ‘use’ seems crucial here – between consenting adults in private, acknowledging that you are talking about a metaphor at least one of you finds useful? Or, at the other extreme in the DSM sense where it claims to be categorising disorders that people have? Attempts to define disorder or dysfunction, including Wakefield’s, usually end in tears (see chapter 7 of my schizophrenia book). We know that psychology and psychiatry are riddled with reified metaphors and at the moment ‘disorder’ is a very powerful one whose use should probably come with a serious health warning.

  14. Hi Mary,

    I had promised myself I would stop answering to this thread but I could't resist the opportunity to speak to such a clear and helpful comment. Thanks for the clarification about your position-that makes things easier.

    I agree with much of what you say and certainly buy into the concerns that motivate it. Society does have a simplifying/reifying view of mental health problems and that does limit our vision when we engage in the project of trying to understand and ameliorate them. Power makes official languages, and their attendant concepts, hard to resist. I also think you're right that I fall victim to a certain linguistic slippage. Perhaps this is because the problems we group together as "Schizophrenia" lack a non question-begging description. You expressed frustration with the same difficulty in your book.

    I suppose I ultimately want to ask how much of our reality would be changed by curtailing "disorder talk", and one way to answer comes from a thought experiment about the future trajectory of our knowledge.

    At some point along the continuum of how concretely we can define problems, their reification as stops striking us as a negative thing. As we come to understand more about mental health, I suspect more problems (though likely never *all*) will fall toward the non-problematic end of this continuum.

    For example, I presently feel comfortable saying "disorder" to refer to "Autism" or "PTSD". As a society we seem to be largely able to see those problems in ways that are sympathetic to individual differences while still accounting for inductive inferences across cases. Meanwhile I am not at all comfortable speaking in the same way about "Oppositional Defiant Disorder" or "ADHD"

    My wonder is whether our increased understanding of more controversial categories (certain forms of psychosis bring paradigmatic examples) will bring us to a point where we can not only draw fruitful analogies between them and cases like Autism/PTSD, but also say "disorder" to refer to them. We seem to be moving to great a appreciation of causal processes-environmental and biological-and are becoming better able to say how particular functions of the mind get disrupted. The most important thing will be if this can lead to improved treatments and patient experience.

    Contrary to me, you might also feel that Autism and PTSD are not to be called "disorders" at all, but If that's the case then we may simply have different notions of appropriate usage. Partly this is a question of taste and will be influenced by how empowering or dis-empowering we have seen the word be for service users.

    But usage is not my main concern here. Like you I am frustrated that we think so much in terms of "real" vs. "socially constructed". I am also keen to stress that--criticism of simplistic and abusive practices aside-- I don't see any contradiction between promoting a rich idiosyncratic understanding of people and a positivist "diagnostic" understanding of the mental health problems they experience. Perhaps where we find most common ground is on the notion that different descriptions serve different social ends. As you point out, the language in such a tricky field should be used advisedly and with a view to revealing constructs and metaphors for what they are.

  15. The map is not the territory; but this is not to say that the territory does (or does not) exist.

    That's got to rank highly in "most pretentious sentences I have ever written", but there you go.

  16. Diagnosis in psychiatry is portrayed as the same type of activity as diagnosis in other areas of medicine. However, the notion that psychiatric conditions are equivalent to physical diseases has been contested for several decades. In this paper, I use the work of Jeff Coulter and David Ingelby to explore the role of diagnosis in routine psychiatric practice. Coulter examined the process of identification of mental disturbance and suggested that it was quite different from the process of identifying a physical disease, as it was dependent on social norms and circumstances. Ingelby pointed out that it was the apparent medical nature of the process that enabled it to act as a justification for the actions that followed. I describe the stories of two patients, which illustrate the themes Ingelby and Coulter identified. In particular they demonstrate that, in contrast to the idea that diagnosis should determine treatment, diagnoses in psychiatry are applied to justify predetermined social responses, designed to control and contain disturbed behaviour and provide care for dependents. Hence psychiatric diagnosis functions as a political device employed to legitimate activities that might otherwise be contested.


Note: only a member of this blog may post a comment.